diff options
| author | Len Brown <len.brown@intel.com> | 2005-12-06 17:31:30 -0500 |
|---|---|---|
| committer | Len Brown <len.brown@intel.com> | 2005-12-06 17:31:30 -0500 |
| commit | 3d5271f9883cba7b54762bc4fe027d4172f06db7 (patch) | |
| tree | ab8a881a14478598a0c8bda0d26c62cdccfffd6d /fs | |
| parent | 378b2556f4e09fa6f87ff0cb5c4395ff28257d02 (diff) | |
| parent | 9115a6c787596e687df03010d97fccc5e0762506 (diff) | |
| download | blackbird-op-linux-3d5271f9883cba7b54762bc4fe027d4172f06db7.tar.gz blackbird-op-linux-3d5271f9883cba7b54762bc4fe027d4172f06db7.zip | |
Pull release into acpica branch
Diffstat (limited to 'fs')
441 files changed, 21799 insertions, 19492 deletions
diff --git a/fs/9p/conv.c b/fs/9p/conv.c index 1554731bd653..18121af99d3e 100644 --- a/fs/9p/conv.c +++ b/fs/9p/conv.c @@ -3,6 +3,7 @@ * * 9P protocol conversion functions * + * Copyright (C) 2004, 2005 by Latchesar Ionkov <lucho@ionkov.net> * Copyright (C) 2004 by Eric Van Hensbergen <ericvh@gmail.com> * Copyright (C) 2002 by Ron Minnich <rminnich@lanl.gov> * @@ -55,66 +56,70 @@ static inline int buf_check_overflow(struct cbuf *buf) return buf->p > buf->ep; } -static inline void buf_check_size(struct cbuf *buf, int len) +static inline int buf_check_size(struct cbuf *buf, int len) { if (buf->p+len > buf->ep) { if (buf->p < buf->ep) { eprintk(KERN_ERR, "buffer overflow\n"); buf->p = buf->ep + 1; + return 0; } } + + return 1; } static inline void *buf_alloc(struct cbuf *buf, int len) { void *ret = NULL; - buf_check_size(buf, len); - ret = buf->p; - buf->p += len; + if (buf_check_size(buf, len)) { + ret = buf->p; + buf->p += len; + } return ret; } static inline void buf_put_int8(struct cbuf *buf, u8 val) { - buf_check_size(buf, 1); - - buf->p[0] = val; - buf->p++; + if (buf_check_size(buf, 1)) { + buf->p[0] = val; + buf->p++; + } } static inline void buf_put_int16(struct cbuf *buf, u16 val) { - buf_check_size(buf, 2); - - *(__le16 *) buf->p = cpu_to_le16(val); - buf->p += 2; + if (buf_check_size(buf, 2)) { + *(__le16 *) buf->p = cpu_to_le16(val); + buf->p += 2; + } } static inline void buf_put_int32(struct cbuf *buf, u32 val) { - buf_check_size(buf, 4); - - *(__le32 *)buf->p = cpu_to_le32(val); - buf->p += 4; + if (buf_check_size(buf, 4)) { + *(__le32 *)buf->p = cpu_to_le32(val); + buf->p += 4; + } } static inline void buf_put_int64(struct cbuf *buf, u64 val) { - buf_check_size(buf, 8); - - *(__le64 *)buf->p = cpu_to_le64(val); - buf->p += 8; + if (buf_check_size(buf, 8)) { + *(__le64 *)buf->p = cpu_to_le64(val); + buf->p += 8; + } } static inline void buf_put_stringn(struct cbuf *buf, const char *s, u16 slen) { - buf_check_size(buf, slen + 2); - - buf_put_int16(buf, slen); - memcpy(buf->p, s, slen); - buf->p += slen; + if (buf_check_size(buf, slen + 2)) { + buf_put_int16(buf, slen); + memcpy(buf->p, s, slen); + buf->p += slen; + } } static inline void buf_put_string(struct cbuf *buf, const char *s) @@ -124,20 +129,20 @@ static inline void buf_put_string(struct cbuf *buf, const char *s) static inline void buf_put_data(struct cbuf *buf, void *data, u32 datalen) { - buf_check_size(buf, datalen); - - memcpy(buf->p, data, datalen); - buf->p += datalen; + if (buf_check_size(buf, datalen)) { + memcpy(buf->p, data, datalen); + buf->p += datalen; + } } static inline u8 buf_get_int8(struct cbuf *buf) { u8 ret = 0; - buf_check_size(buf, 1); - ret = buf->p[0]; - - buf->p++; + if (buf_check_size(buf, 1)) { + ret = buf->p[0]; + buf->p++; + } return ret; } @@ -146,10 +151,10 @@ static inline u16 buf_get_int16(struct cbuf *buf) { u16 ret = 0; - buf_check_size(buf, 2); - ret = le16_to_cpu(*(__le16 *)buf->p); - - buf->p += 2; + if (buf_check_size(buf, 2)) { + ret = le16_to_cpu(*(__le16 *)buf->p); + buf->p += 2; + } return ret; } @@ -158,10 +163,10 @@ static inline u32 buf_get_int32(struct cbuf *buf) { u32 ret = 0; - buf_check_size(buf, 4); - ret = le32_to_cpu(*(__le32 *)buf->p); - - buf->p += 4; + if (buf_check_size(buf, 4)) { + ret = le32_to_cpu(*(__le32 *)buf->p); + buf->p += 4; + } return ret; } @@ -170,10 +175,10 @@ static inline u64 buf_get_int64(struct cbuf *buf) { u64 ret = 0; - buf_check_size(buf, 8); - ret = le64_to_cpu(*(__le64 *)buf->p); - - buf->p += 8; + if (buf_check_size(buf, 8)) { + ret = le64_to_cpu(*(__le64 *)buf->p); + buf->p += 8; + } return ret; } @@ -181,27 +186,35 @@ static inline u64 buf_get_int64(struct cbuf *buf) static inline int buf_get_string(struct cbuf *buf, char *data, unsigned int datalen) { + u16 len = 0; + + len = buf_get_int16(buf); + if (!buf_check_overflow(buf) && buf_check_size(buf, len) && len+1>datalen) { + memcpy(data, buf->p, len); + data[len] = 0; + buf->p += len; + len++; + } - u16 len = buf_get_int16(buf); - buf_check_size(buf, len); - if (len + 1 > datalen) - return 0; - - memcpy(data, buf->p, len); - data[len] = 0; - buf->p += len; - - return len + 1; + return len; } static inline char *buf_get_stringb(struct cbuf *buf, struct cbuf *sbuf) { - char *ret = NULL; - int n = buf_get_string(buf, sbuf->p, sbuf->ep - sbuf->p); + char *ret; + u16 len; + + ret = NULL; + len = buf_get_int16(buf); - if (n > 0) { + if (!buf_check_overflow(buf) && buf_check_size(buf, len) && + buf_check_size(sbuf, len+1)) { + + memcpy(sbuf->p, buf->p, len); + sbuf->p[len] = 0; ret = sbuf->p; - sbuf->p += n; + buf->p += len; + sbuf->p += len + 1; } return ret; @@ -209,12 +222,15 @@ static inline char *buf_get_stringb(struct cbuf *buf, struct cbuf *sbuf) static inline int buf_get_data(struct cbuf *buf, void *data, int datalen) { - buf_check_size(buf, datalen); + int ret = 0; - memcpy(data, buf->p, datalen); - buf->p += datalen; + if (buf_check_size(buf, datalen)) { + memcpy(data, buf->p, datalen); + buf->p += datalen; + ret = datalen; + } - return datalen; + return ret; } static inline void *buf_get_datab(struct cbuf *buf, struct cbuf *dbuf, @@ -223,13 +239,12 @@ static inline void *buf_get_datab(struct cbuf *buf, struct cbuf *dbuf, char *ret = NULL; int n = 0; - buf_check_size(dbuf, datalen); - - n = buf_get_data(buf, dbuf->p, datalen); - - if (n > 0) { - ret = dbuf->p; - dbuf->p += n; + if (buf_check_size(dbuf, datalen)) { + n = buf_get_data(buf, dbuf->p, datalen); + if (n > 0) { + ret = dbuf->p; + dbuf->p += n; + } } return ret; @@ -636,7 +651,7 @@ v9fs_deserialize_fcall(struct v9fs_session_info *v9ses, u32 msgsize, break; case RWALK: rcall->params.rwalk.nwqid = buf_get_int16(bufp); - rcall->params.rwalk.wqids = buf_alloc(bufp, + rcall->params.rwalk.wqids = buf_alloc(dbufp, rcall->params.rwalk.nwqid * sizeof(struct v9fs_qid)); if (rcall->params.rwalk.wqids) for (i = 0; i < rcall->params.rwalk.nwqid; i++) { diff --git a/fs/9p/error.c b/fs/9p/error.c index fee5d19179c5..834cb179e388 100644 --- a/fs/9p/error.c +++ b/fs/9p/error.c @@ -33,6 +33,7 @@ #include <linux/list.h> #include <linux/jhash.h> +#include <linux/string.h> #include "debug.h" #include "error.h" diff --git a/fs/9p/fid.c b/fs/9p/fid.c index 821c9c4d76aa..d95f8626d170 100644 --- a/fs/9p/fid.c +++ b/fs/9p/fid.c @@ -71,21 +71,28 @@ static int v9fs_fid_insert(struct v9fs_fid *fid, struct dentry *dentry) * */ -struct v9fs_fid *v9fs_fid_create(struct dentry *dentry) +struct v9fs_fid *v9fs_fid_create(struct dentry *dentry, + struct v9fs_session_info *v9ses, int fid, int create) { struct v9fs_fid *new; + dprintk(DEBUG_9P, "fid create dentry %p, fid %d, create %d\n", + dentry, fid, create); + new = kmalloc(sizeof(struct v9fs_fid), GFP_KERNEL); if (new == NULL) { dprintk(DEBUG_ERROR, "Out of Memory\n"); return ERR_PTR(-ENOMEM); } - new->fid = -1; + new->fid = fid; + new->v9ses = v9ses; new->fidopen = 0; - new->fidcreate = 0; + new->fidcreate = create; new->fidclunked = 0; new->iounit = 0; + new->rdir_pos = 0; + new->rdir_fcall = NULL; if (v9fs_fid_insert(new, dentry) == 0) return new; @@ -109,6 +116,59 @@ void v9fs_fid_destroy(struct v9fs_fid *fid) } /** + * v9fs_fid_walk_up - walks from the process current directory + * up to the specified dentry. + */ +static struct v9fs_fid *v9fs_fid_walk_up(struct dentry *dentry) +{ + int fidnum, cfidnum, err; + struct v9fs_fid *cfid; + struct dentry *cde; + struct v9fs_session_info *v9ses; + + v9ses = v9fs_inode2v9ses(current->fs->pwd->d_inode); + cfid = v9fs_fid_lookup(current->fs->pwd); + if (cfid == NULL) { + dprintk(DEBUG_ERROR, "process cwd doesn't have a fid\n"); + return ERR_PTR(-ENOENT); + } + + cfidnum = cfid->fid; + cde = current->fs->pwd; + /* TODO: take advantage of multiwalk */ + + fidnum = v9fs_get_idpool(&v9ses->fidpool); + if (fidnum < 0) { + dprintk(DEBUG_ERROR, "could not get a new fid num\n"); + err = -ENOENT; + goto clunk_fid; + } + + while (cde != dentry) { + if (cde == cde->d_parent) { + dprintk(DEBUG_ERROR, "can't find dentry\n"); + err = -ENOENT; + goto clunk_fid; + } + + err = v9fs_t_walk(v9ses, cfidnum, fidnum, "..", NULL); + if (err < 0) { + dprintk(DEBUG_ERROR, "problem walking to parent\n"); + goto clunk_fid; + } + + cfidnum = fidnum; + cde = cde->d_parent; + } + + return v9fs_fid_create(dentry, v9ses, fidnum, 0); + +clunk_fid: + v9fs_t_clunk(v9ses, fidnum, NULL); + return ERR_PTR(err); +} + +/** * v9fs_fid_lookup - retrieve the right fid from a particular dentry * @dentry: dentry to look for fid in * @type: intent of lookup (operation or traversal) @@ -119,49 +179,25 @@ void v9fs_fid_destroy(struct v9fs_fid *fid) * */ -struct v9fs_fid *v9fs_fid_lookup(struct dentry *dentry, int type) +struct v9fs_fid *v9fs_fid_lookup(struct dentry *dentry) { struct list_head *fid_list = (struct list_head *)dentry->d_fsdata; struct v9fs_fid *current_fid = NULL; struct v9fs_fid *temp = NULL; struct v9fs_fid *return_fid = NULL; - int found_parent = 0; - int found_user = 0; - dprintk(DEBUG_9P, " dentry: %s (%p) type %d\n", dentry->d_iname, dentry, - type); + dprintk(DEBUG_9P, " dentry: %s (%p)\n", dentry->d_iname, dentry); - if (fid_list && !list_empty(fid_list)) { + if (fid_list) { list_for_each_entry_safe(current_fid, temp, fid_list, list) { - if (current_fid->uid == current->uid) { - if (return_fid == NULL) { - if ((type == FID_OP) - || (!current_fid->fidopen)) { - return_fid = current_fid; - found_user = 1; - } - } - } - if (current_fid->pid == current->real_parent->pid) { - if ((return_fid == NULL) || (found_parent) - || (found_user)) { - if ((type == FID_OP) - || (!current_fid->fidopen)) { - return_fid = current_fid; - found_parent = 1; - found_user = 0; - } - } - } - if (current_fid->pid == current->pid) { - if ((type == FID_OP) || - (!current_fid->fidopen)) { - return_fid = current_fid; - found_parent = 0; - found_user = 0; - } + if (!current_fid->fidcreate) { + return_fid = current_fid; + break; } } + + if (!return_fid) + return_fid = current_fid; } /* we are at the root but didn't match */ @@ -187,55 +223,33 @@ struct v9fs_fid *v9fs_fid_lookup(struct dentry *dentry, int type) /* XXX - there may be some duplication we can get rid of */ if (par == dentry) { - /* we need to fid_lookup the starting point */ - int fidnum = -1; - int oldfid = -1; - int result = -1; - struct v9fs_session_info *v9ses = - v9fs_inode2v9ses(current->fs->pwd->d_inode); - - current_fid = - v9fs_fid_lookup(current->fs->pwd, FID_WALK); - if (current_fid == NULL) { - dprintk(DEBUG_ERROR, - "process cwd doesn't have a fid\n"); - return return_fid; - } - oldfid = current_fid->fid; - par = current->fs->pwd; - /* TODO: take advantage of multiwalk */ + return_fid = v9fs_fid_walk_up(dentry); + if (IS_ERR(return_fid)) + return_fid = NULL; + } + } - fidnum = v9fs_get_idpool(&v9ses->fidpool); - if (fidnum < 0) { - dprintk(DEBUG_ERROR, - "could not get a new fid num\n"); - return return_fid; - } + return return_fid; +} - while (par != dentry) { - result = - v9fs_t_walk(v9ses, oldfid, fidnum, "..", - NULL); - if (result < 0) { - dprintk(DEBUG_ERROR, - "problem walking to parent\n"); - - break; - } - oldfid = fidnum; - if (par == par->d_parent) { - dprintk(DEBUG_ERROR, - "can't find dentry\n"); - break; - } - par = par->d_parent; - } - if (par == dentry) { - return_fid = v9fs_fid_create(dentry); - return_fid->fid = fidnum; +struct v9fs_fid *v9fs_fid_get_created(struct dentry *dentry) +{ + struct list_head *fid_list; + struct v9fs_fid *fid, *ftmp, *ret; + + dprintk(DEBUG_9P, " dentry: %s (%p)\n", dentry->d_iname, dentry); + fid_list = (struct list_head *)dentry->d_fsdata; + ret = NULL; + if (fid_list) { + list_for_each_entry_safe(fid, ftmp, fid_list, list) { + if (fid->fidcreate && fid->pid == current->pid) { + list_del(&fid->list); + ret = fid; + break; } } } - return return_fid; + dprintk(DEBUG_9P, "return %p\n", ret); + return ret; } diff --git a/fs/9p/fid.h b/fs/9p/fid.h index 7db478ccca36..84c673a44c83 100644 --- a/fs/9p/fid.h +++ b/fs/9p/fid.h @@ -25,6 +25,7 @@ #define FID_OP 0 #define FID_WALK 1 +#define FID_CREATE 2 struct v9fs_fid { struct list_head list; /* list of fids associated with a dentry */ @@ -52,6 +53,8 @@ struct v9fs_fid { struct v9fs_session_info *v9ses; /* session info for this FID */ }; -struct v9fs_fid *v9fs_fid_lookup(struct dentry *dentry, int type); +struct v9fs_fid *v9fs_fid_lookup(struct dentry *dentry); +struct v9fs_fid *v9fs_fid_get_created(struct dentry *); void v9fs_fid_destroy(struct v9fs_fid *fid); -struct v9fs_fid *v9fs_fid_create(struct dentry *); +struct v9fs_fid *v9fs_fid_create(struct dentry *, + struct v9fs_session_info *v9ses, int fid, int create); diff --git a/fs/9p/trans_sock.c b/fs/9p/trans_sock.c index 01e26f0013ac..a93c2bf94c33 100644 --- a/fs/9p/trans_sock.c +++ b/fs/9p/trans_sock.c @@ -269,8 +269,7 @@ static void v9fs_sock_close(struct v9fs_transport *trans) dprintk(DEBUG_TRANS, "socket closed\n"); } - if (ts) - kfree(ts); + kfree(ts); trans->priv = NULL; } diff --git a/fs/9p/v9fs.c b/fs/9p/v9fs.c index 13bdbbab4387..418c3743fdee 100644 --- a/fs/9p/v9fs.c +++ b/fs/9p/v9fs.c @@ -266,7 +266,7 @@ v9fs_session_init(struct v9fs_session_info *v9ses, v9ses->remotename = __getname(); if (!v9ses->remotename) { - putname(v9ses->name); + __putname(v9ses->name); return -ENOMEM; } @@ -303,7 +303,13 @@ v9fs_session_init(struct v9fs_session_info *v9ses, goto SessCleanUp; }; - v9ses->transport = trans_proto; + v9ses->transport = kmalloc(sizeof(*v9ses->transport), GFP_KERNEL); + if (!v9ses->transport) { + retval = -ENOMEM; + goto SessCleanUp; + } + + memmove(v9ses->transport, trans_proto, sizeof(*v9ses->transport)); if ((retval = v9ses->transport->init(v9ses, dev_name, data)) < 0) { eprintk(KERN_ERR, "problem initializing transport\n"); @@ -405,8 +411,8 @@ void v9fs_session_close(struct v9fs_session_info *v9ses) if (v9ses->transport) v9ses->transport->close(v9ses->transport); - putname(v9ses->name); - putname(v9ses->remotename); + __putname(v9ses->name); + __putname(v9ses->remotename); } /** diff --git a/fs/9p/vfs_dentry.c b/fs/9p/vfs_dentry.c index 306c96741f81..a6aa947de0f9 100644 --- a/fs/9p/vfs_dentry.c +++ b/fs/9p/vfs_dentry.c @@ -67,7 +67,7 @@ static int v9fs_dentry_validate(struct dentry *dentry, struct nameidata *nd) struct dentry *dc = current->fs->pwd; dprintk(DEBUG_VFS, "dentry: %s (%p)\n", dentry->d_iname, dentry); - if (v9fs_fid_lookup(dentry, FID_OP)) { + if (v9fs_fid_lookup(dentry)) { dprintk(DEBUG_VFS, "VALID\n"); return 1; } diff --git a/fs/9p/vfs_dir.c b/fs/9p/vfs_dir.c index c478a7384186..57a43b8feef5 100644 --- a/fs/9p/vfs_dir.c +++ b/fs/9p/vfs_dir.c @@ -197,21 +197,18 @@ int v9fs_dir_release(struct inode *inode, struct file *filp) filemap_fdatawait(inode->i_mapping); if (fidnum >= 0) { - fid->fidopen--; dprintk(DEBUG_VFS, "fidopen: %d v9f->fid: %d\n", fid->fidopen, fid->fid); - if (fid->fidopen == 0) { - if (v9fs_t_clunk(v9ses, fidnum, NULL)) - dprintk(DEBUG_ERROR, "clunk failed\n"); + if (v9fs_t_clunk(v9ses, fidnum, NULL)) + dprintk(DEBUG_ERROR, "clunk failed\n"); - v9fs_put_idpool(fid->fid, &v9ses->fidpool); - } + v9fs_put_idpool(fid->fid, &v9ses->fidpool); kfree(fid->rdir_fcall); + kfree(fid); filp->private_data = NULL; - v9fs_fid_destroy(fid); } d_drop(filp->f_dentry); diff --git a/fs/9p/vfs_file.c b/fs/9p/vfs_file.c index 1f8ae7d580ab..89c849da8504 100644 --- a/fs/9p/vfs_file.c +++ b/fs/9p/vfs_file.c @@ -32,7 +32,6 @@ #include <linux/string.h> #include <linux/smp_lock.h> #include <linux/inet.h> -#include <linux/version.h> #include <linux/list.h> #include <asm/uaccess.h> #include <linux/idr.h> @@ -53,30 +52,36 @@ int v9fs_file_open(struct inode *inode, struct file *file) { struct v9fs_session_info *v9ses = v9fs_inode2v9ses(inode); - struct v9fs_fid *v9fid = v9fs_fid_lookup(file->f_dentry, FID_WALK); - struct v9fs_fid *v9newfid = NULL; + struct v9fs_fid *v9fid, *fid; struct v9fs_fcall *fcall = NULL; int open_mode = 0; unsigned int iounit = 0; int newfid = -1; long result = -1; - dprintk(DEBUG_VFS, "inode: %p file: %p v9fid= %p\n", inode, file, - v9fid); + dprintk(DEBUG_VFS, "inode: %p file: %p \n", inode, file); + + v9fid = v9fs_fid_get_created(file->f_dentry); + if (!v9fid) + v9fid = v9fs_fid_lookup(file->f_dentry); if (!v9fid) { - struct dentry *dentry = file->f_dentry; dprintk(DEBUG_ERROR, "Couldn't resolve fid from dentry\n"); + return -EBADF; + } - /* XXX - some duplication from lookup, generalize later */ - /* basically vfs_lookup is too heavy weight */ - v9fid = v9fs_fid_lookup(file->f_dentry, FID_OP); - if (!v9fid) - return -EBADF; + if (!v9fid->fidcreate) { + fid = kmalloc(sizeof(struct v9fs_fid), GFP_KERNEL); + if (fid == NULL) { + dprintk(DEBUG_ERROR, "Out of Memory\n"); + return -ENOMEM; + } - v9fid = v9fs_fid_lookup(dentry->d_parent, FID_WALK); - if (!v9fid) - return -EBADF; + fid->fidopen = 0; + fid->fidcreate = 0; + fid->fidclunked = 0; + fid->iounit = 0; + fid->v9ses = v9ses; newfid = v9fs_get_idpool(&v9ses->fidpool); if (newfid < 0) { @@ -85,58 +90,16 @@ int v9fs_file_open(struct inode *inode, struct file *file) } result = - v9fs_t_walk(v9ses, v9fid->fid, newfid, - (char *)file->f_dentry->d_name.name, NULL); + v9fs_t_walk(v9ses, v9fid->fid, newfid, NULL, NULL); + if (result < 0) { v9fs_put_idpool(newfid, &v9ses->fidpool); dprintk(DEBUG_ERROR, "rewalk didn't work\n"); return -EBADF; } - v9fid = v9fs_fid_create(dentry); - if (v9fid == NULL) { - dprintk(DEBUG_ERROR, "couldn't insert\n"); - return -ENOMEM; - } - v9fid->fid = newfid; - } - - if (v9fid->fidcreate) { - /* create case */ - newfid = v9fid->fid; - iounit = v9fid->iounit; - v9fid->fidcreate = 0; - } else { - if (!S_ISDIR(inode->i_mode)) - newfid = v9fid->fid; - else { - newfid = v9fs_get_idpool(&v9ses->fidpool); - if (newfid < 0) { - eprintk(KERN_WARNING, "allocation failed\n"); - return -ENOSPC; - } - /* This would be a somewhat critical clone */ - result = - v9fs_t_walk(v9ses, v9fid->fid, newfid, NULL, - &fcall); - if (result < 0) { - dprintk(DEBUG_ERROR, "clone error: %s\n", - FCALL_ERROR(fcall)); - kfree(fcall); - return result; - } - - v9newfid = v9fs_fid_create(file->f_dentry); - v9newfid->fid = newfid; - v9newfid->qid = v9fid->qid; - v9newfid->iounit = v9fid->iounit; - v9newfid->fidopen = 0; - v9newfid->fidclunked = 0; - v9newfid->v9ses = v9ses; - v9fid = v9newfid; - kfree(fcall); - } - + fid->fid = newfid; + v9fid = fid; /* TODO: do special things for O_EXCL, O_NOFOLLOW, O_SYNC */ /* translate open mode appropriately */ open_mode = file->f_flags & 0x3; @@ -163,9 +126,13 @@ int v9fs_file_open(struct inode *inode, struct file *file) iounit = fcall->params.ropen.iounit; kfree(fcall); + } else { + /* create case */ + newfid = v9fid->fid; + iounit = v9fid->iounit; + v9fid->fidcreate = 0; } - file->private_data = v9fid; v9fid->rdir_pos = 0; @@ -207,16 +174,16 @@ static int v9fs_file_lock(struct file *filp, int cmd, struct file_lock *fl) } /** - * v9fs_read - read from a file (internal) + * v9fs_file_read - read from a file * @filep: file pointer to read * @data: data buffer to read data into * @count: size of buffer * @offset: offset at which to read data * */ - static ssize_t -v9fs_read(struct file *filp, char *buffer, size_t count, loff_t * offset) +v9fs_file_read(struct file *filp, char __user * data, size_t count, + loff_t * offset) { struct inode *inode = filp->f_dentry->d_inode; struct v9fs_session_info *v9ses = v9fs_inode2v9ses(inode); @@ -226,6 +193,7 @@ v9fs_read(struct file *filp, char *buffer, size_t count, loff_t * offset) int rsize = 0; int result = 0; int total = 0; + int n; dprintk(DEBUG_VFS, "\n"); @@ -248,10 +216,15 @@ v9fs_read(struct file *filp, char *buffer, size_t count, loff_t * offset) } else *offset += result; - /* XXX - extra copy */ - memcpy(buffer, fcall->params.rread.data, result); + n = copy_to_user(data, fcall->params.rread.data, result); + if (n) { + dprintk(DEBUG_ERROR, "Problem copying to user %d\n", n); + kfree(fcall); + return -EFAULT; + } + count -= result; - buffer += result; + data += result; total += result; kfree(fcall); @@ -264,42 +237,7 @@ v9fs_read(struct file *filp, char *buffer, size_t count, loff_t * offset) } /** - * v9fs_file_read - read from a file - * @filep: file pointer to read - * @data: data buffer to read data into - * @count: size of buffer - * @offset: offset at which to read data - * - */ - -static ssize_t -v9fs_file_read(struct file *filp, char __user * data, size_t count, - loff_t * offset) -{ - int retval = -1; - int ret = 0; - char *buffer; - - buffer = kmalloc(count, GFP_KERNEL); - if (!buffer) - return -ENOMEM; - - retval = v9fs_read(filp, buffer, count, offset); - if (retval > 0) { - if ((ret = copy_to_user(data, buffer, retval)) != 0) { - dprintk(DEBUG_ERROR, "Problem copying to user %d\n", - ret); - retval = ret; - } - } - - kfree(buffer); - - return retval; -} - -/** - * v9fs_write - write to a file + * v9fs_file_write - write to a file * @filep: file pointer to write * @data: data buffer to write data from * @count: size of buffer @@ -308,7 +246,8 @@ v9fs_file_read(struct file *filp, char __user * data, size_t count, */ static ssize_t -v9fs_write(struct file *filp, char *buffer, size_t count, loff_t * offset) +v9fs_file_write(struct file *filp, const char __user * data, + size_t count, loff_t * offset) { struct inode *inode = filp->f_dentry->d_inode; struct v9fs_session_info *v9ses = v9fs_inode2v9ses(inode); @@ -318,30 +257,42 @@ v9fs_write(struct file *filp, char *buffer, size_t count, loff_t * offset) int result = -EIO; int rsize = 0; int total = 0; + char *buf; - dprintk(DEBUG_VFS, "data %p count %d offset %x\n", buffer, (int)count, + dprintk(DEBUG_VFS, "data %p count %d offset %x\n", data, (int)count, (int)*offset); rsize = v9ses->maxdata - V9FS_IOHDRSZ; if (v9fid->iounit != 0 && rsize > v9fid->iounit) rsize = v9fid->iounit; - dump_data(buffer, count); + buf = kmalloc(v9ses->maxdata - V9FS_IOHDRSZ, GFP_KERNEL); + if (!buf) + return -ENOMEM; do { if (count < rsize) rsize = count; - result = - v9fs_t_write(v9ses, fid, *offset, rsize, buffer, &fcall); + result = copy_from_user(buf, data, rsize); + if (result) { + dprintk(DEBUG_ERROR, "Problem copying from user\n"); + kfree(buf); + return -EFAULT; + } + + dump_data(buf, rsize); + result = v9fs_t_write(v9ses, fid, *offset, rsize, buf, &fcall); if (result < 0) { eprintk(KERN_ERR, "error while writing: %s(%d)\n", FCALL_ERROR(fcall), result); kfree(fcall); + kfree(buf); return result; } else *offset += result; kfree(fcall); + fcall = NULL; if (result != rsize) { eprintk(KERN_ERR, @@ -351,46 +302,14 @@ v9fs_write(struct file *filp, char *buffer, size_t count, loff_t * offset) } count -= result; - buffer += result; + data += result; total += result; } while (count); + kfree(buf); return total; } -/** - * v9fs_file_write - write to a file - * @filep: file pointer to write - * @data: data buffer to write data from - * @count: size of buffer - * @offset: offset at which to write data - * - */ - -static ssize_t -v9fs_file_write(struct file *filp, const char __user * data, - size_t count, loff_t * offset) -{ - int ret = -1; - char *buffer; - - buffer = kmalloc(count, GFP_KERNEL); - if (buffer == NULL) - return -ENOMEM; - - ret = copy_from_user(buffer, data, count); - if (ret) { - dprintk(DEBUG_ERROR, "Problem copying from user\n"); - ret = -EFAULT; - } else { - ret = v9fs_write(filp, buffer, count, offset); - } - - kfree(buffer); - - return ret; -} - struct file_operations v9fs_file_operations = { .llseek = generic_file_llseek, .read = v9fs_file_read, diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index 0c13fc600049..0ea965c3bb7d 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -307,7 +307,7 @@ v9fs_create(struct inode *dir, struct v9fs_session_info *v9ses = v9fs_inode2v9ses(dir); struct super_block *sb = dir->i_sb; struct v9fs_fid *dirfid = - v9fs_fid_lookup(file_dentry->d_parent, FID_WALK); + v9fs_fid_lookup(file_dentry->d_parent); struct v9fs_fid *fid = NULL; struct inode *file_inode = NULL; struct v9fs_fcall *fcall = NULL; @@ -317,6 +317,7 @@ v9fs_create(struct inode *dir, long newfid = -1; int result = 0; unsigned int iounit = 0; + int wfidno = -1; perm = unixmode2p9mode(v9ses, perm); @@ -350,7 +351,7 @@ v9fs_create(struct inode *dir, if (result < 0) { dprintk(DEBUG_ERROR, "clone error: %s\n", FCALL_ERROR(fcall)); v9fs_put_idpool(newfid, &v9ses->fidpool); - newfid = 0; + newfid = -1; goto CleanUpFid; } @@ -369,20 +370,39 @@ v9fs_create(struct inode *dir, qid = fcall->params.rcreate.qid; kfree(fcall); - fid = v9fs_fid_create(file_dentry); + fid = v9fs_fid_create(file_dentry, v9ses, newfid, 1); + dprintk(DEBUG_VFS, "fid %p %d\n", fid, fid->fidcreate); if (!fid) { result = -ENOMEM; goto CleanUpFid; } - fid->fid = newfid; - fid->fidopen = 0; - fid->fidcreate = 1; fid->qid = qid; fid->iounit = iounit; - fid->rdir_pos = 0; - fid->rdir_fcall = NULL; - fid->v9ses = v9ses; + + /* walk to the newly created file and put the fid in the dentry */ + wfidno = v9fs_get_idpool(&v9ses->fidpool); + if (newfid < 0) { + eprintk(KERN_WARNING, "no free fids available\n"); + return -ENOSPC; + } + + result = v9fs_t_walk(v9ses, dirfidnum, wfidno, + (char *) file_dentry->d_name.name, NULL); + if (result < 0) { + dprintk(DEBUG_ERROR, "clone error: %s\n", FCALL_ERROR(fcall)); + v9fs_put_idpool(wfidno, &v9ses->fidpool); + wfidno = -1; + goto CleanUpFid; + } + + if (!v9fs_fid_create(file_dentry, v9ses, wfidno, 0)) { + if (!v9fs_t_clunk(v9ses, newfid, &fcall)) { + v9fs_put_idpool(wfidno, &v9ses->fidpool); + } + + goto CleanUpFid; + } if ((perm & V9FS_DMSYMLINK) || (perm & V9FS_DMLINK) || (perm & V9FS_DMNAMEDPIPE) || (perm & V9FS_DMSOCKET) || @@ -407,14 +427,16 @@ v9fs_create(struct inode *dir, v9fs_mistat2inode(fcall->params.rstat.stat, file_inode, sb); kfree(fcall); + fcall = NULL; + file_dentry->d_op = &v9fs_dentry_operations; d_instantiate(file_dentry, file_inode); if (perm & V9FS_DMDIR) { - if (v9fs_t_clunk(v9ses, newfid, &fcall)) + if (!v9fs_t_clunk(v9ses, newfid, &fcall)) + v9fs_put_idpool(newfid, &v9ses->fidpool); + else dprintk(DEBUG_ERROR, "clunk for mkdir failed: %s\n", FCALL_ERROR(fcall)); - - v9fs_put_idpool(newfid, &v9ses->fidpool); kfree(fcall); fid->fidopen = 0; fid->fidcreate = 0; @@ -426,12 +448,22 @@ v9fs_create(struct inode *dir, CleanUpFid: kfree(fcall); - if (newfid) { - if (v9fs_t_clunk(v9ses, newfid, &fcall)) + if (newfid >= 0) { + if (!v9fs_t_clunk(v9ses, newfid, &fcall)) + v9fs_put_idpool(newfid, &v9ses->fidpool); + else + dprintk(DEBUG_ERROR, "clunk failed: %s\n", + FCALL_ERROR(fcall)); + + kfree(fcall); + } + if (wfidno >= 0) { + if (!v9fs_t_clunk(v9ses, wfidno, &fcall)) + v9fs_put_idpool(wfidno, &v9ses->fidpool); + else dprintk(DEBUG_ERROR, "clunk failed: %s\n", FCALL_ERROR(fcall)); - v9fs_put_idpool(newfid, &v9ses->fidpool); kfree(fcall); } return result; @@ -461,7 +493,7 @@ static int v9fs_remove(struct inode *dir, struct dentry *file, int rmdir) file_inode = file->d_inode; sb = file_inode->i_sb; v9ses = v9fs_inode2v9ses(file_inode); - v9fid = v9fs_fid_lookup(file, FID_OP); + v9fid = v9fs_fid_lookup(file); if (!v9fid) { dprintk(DEBUG_ERROR, @@ -545,7 +577,7 @@ static struct dentry *v9fs_vfs_lookup(struct inode *dir, struct dentry *dentry, sb = dir->i_sb; v9ses = v9fs_inode2v9ses(dir); - dirfid = v9fs_fid_lookup(dentry->d_parent, FID_WALK); + dirfid = v9fs_fid_lookup(dentry->d_parent); if (!dirfid) { dprintk(DEBUG_ERROR, "no dirfid\n"); @@ -573,7 +605,7 @@ static struct dentry *v9fs_vfs_lookup(struct inode *dir, struct dentry *dentry, v9fs_put_idpool(newfid, &v9ses->fidpool); if (result == -ENOENT) { d_add(dentry, NULL); - dprintk(DEBUG_ERROR, + dprintk(DEBUG_VFS, "Return negative dentry %p count %d\n", dentry, atomic_read(&dentry->d_count)); return NULL; @@ -601,16 +633,13 @@ static struct dentry *v9fs_vfs_lookup(struct inode *dir, struct dentry *dentry, inode->i_ino = v9fs_qid2ino(&fcall->params.rstat.stat->qid); - fid = v9fs_fid_create(dentry); + fid = v9fs_fid_create(dentry, v9ses, newfid, 0); if (fid == NULL) { dprintk(DEBUG_ERROR, "couldn't insert\n"); result = -ENOMEM; goto FreeFcall; } - fid->fid = newfid; - fid->fidopen = 0; - fid->v9ses = v9ses; fid->qid = fcall->params.rstat.stat->qid; dentry->d_op = &v9fs_dentry_operations; @@ -665,11 +694,11 @@ v9fs_vfs_rename(struct inode *old_dir, struct dentry *old_dentry, { struct inode *old_inode = old_dentry->d_inode; struct v9fs_session_info *v9ses = v9fs_inode2v9ses(old_inode); - struct v9fs_fid *oldfid = v9fs_fid_lookup(old_dentry, FID_WALK); + struct v9fs_fid *oldfid = v9fs_fid_lookup(old_dentry); struct v9fs_fid *olddirfid = - v9fs_fid_lookup(old_dentry->d_parent, FID_WALK); + v9fs_fid_lookup(old_dentry->d_parent); struct v9fs_fid *newdirfid = - v9fs_fid_lookup(new_dentry->d_parent, FID_WALK); + v9fs_fid_lookup(new_dentry->d_parent); struct v9fs_stat *mistat = kmalloc(v9ses->maxdata, GFP_KERNEL); struct v9fs_fcall *fcall = NULL; int fid = -1; @@ -744,7 +773,7 @@ v9fs_vfs_getattr(struct vfsmount *mnt, struct dentry *dentry, { struct v9fs_fcall *fcall = NULL; struct v9fs_session_info *v9ses = v9fs_inode2v9ses(dentry->d_inode); - struct v9fs_fid *fid = v9fs_fid_lookup(dentry, FID_OP); + struct v9fs_fid *fid = v9fs_fid_lookup(dentry); int err = -EPERM; dprintk(DEBUG_VFS, "dentry: %p\n", dentry); @@ -778,7 +807,7 @@ v9fs_vfs_getattr(struct vfsmount *mnt, struct dentry *dentry, static int v9fs_vfs_setattr(struct dentry *dentry, struct iattr *iattr) { struct v9fs_session_info *v9ses = v9fs_inode2v9ses(dentry->d_inode); - struct v9fs_fid *fid = v9fs_fid_lookup(dentry, FID_OP); + struct v9fs_fid *fid = v9fs_fid_lookup(dentry); struct v9fs_fcall *fcall = NULL; struct v9fs_stat *mistat = kmalloc(v9ses->maxdata, GFP_KERNEL); int res = -EPERM; @@ -960,7 +989,7 @@ v9fs_vfs_symlink(struct inode *dir, struct dentry *dentry, const char *symname) if (retval != 0) goto FreeFcall; - newfid = v9fs_fid_lookup(dentry, FID_OP); + newfid = v9fs_fid_lookup(dentry); /* issue a twstat */ v9fs_blank_mistat(v9ses, mistat); @@ -1004,7 +1033,7 @@ static int v9fs_readlink(struct dentry *dentry, char *buffer, int buflen) struct v9fs_fcall *fcall = NULL; struct v9fs_session_info *v9ses = v9fs_inode2v9ses(dentry->d_inode); - struct v9fs_fid *fid = v9fs_fid_lookup(dentry, FID_OP); + struct v9fs_fid *fid = v9fs_fid_lookup(dentry); if (!fid) { dprintk(DEBUG_ERROR, "could not resolve fid from dentry\n"); @@ -1063,8 +1092,8 @@ static int v9fs_vfs_readlink(struct dentry *dentry, char __user * buffer, int ret; char *link = __getname(); - if (strlen(link) < buflen) - buflen = strlen(link); + if (buflen > PATH_MAX) + buflen = PATH_MAX; dprintk(DEBUG_VFS, " dentry: %s (%p)\n", dentry->d_iname, dentry); @@ -1078,7 +1107,7 @@ static int v9fs_vfs_readlink(struct dentry *dentry, char __user * buffer, } } - putname(link); + __putname(link); return retval; } @@ -1102,7 +1131,7 @@ static void *v9fs_vfs_follow_link(struct dentry *dentry, struct nameidata *nd) len = v9fs_readlink(dentry, link, strlen(link)); if (len < 0) { - putname(link); + __putname(link); link = ERR_PTR(len); } else link[len] = 0; @@ -1125,7 +1154,7 @@ static void v9fs_vfs_put_link(struct dentry *dentry, struct nameidata *nd, void dprintk(DEBUG_VFS, " %s %s\n", dentry->d_name.name, s); if (!IS_ERR(s)) - putname(s); + __putname(s); } /** @@ -1148,7 +1177,7 @@ v9fs_vfs_link(struct dentry *old_dentry, struct inode *dir, struct v9fs_session_info *v9ses = v9fs_inode2v9ses(dir); struct v9fs_fcall *fcall = NULL; struct v9fs_stat *mistat = kmalloc(v9ses->maxdata, GFP_KERNEL); - struct v9fs_fid *oldfid = v9fs_fid_lookup(old_dentry, FID_OP); + struct v9fs_fid *oldfid = v9fs_fid_lookup(old_dentry); struct v9fs_fid *newfid = NULL; char *symname = __getname(); @@ -1168,7 +1197,7 @@ v9fs_vfs_link(struct dentry *old_dentry, struct inode *dir, if (retval != 0) goto FreeMem; - newfid = v9fs_fid_lookup(dentry, FID_OP); + newfid = v9fs_fid_lookup(dentry); if (!newfid) { dprintk(DEBUG_ERROR, "couldn't resolve fid from dentry\n"); goto FreeMem; @@ -1201,7 +1230,7 @@ v9fs_vfs_link(struct dentry *old_dentry, struct inode *dir, FreeMem: kfree(mistat); kfree(fcall); - putname(symname); + __putname(symname); return retval; } @@ -1246,7 +1275,7 @@ v9fs_vfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t rdev) if (retval != 0) goto FreeMem; - newfid = v9fs_fid_lookup(dentry, FID_OP); + newfid = v9fs_fid_lookup(dentry); if (!newfid) { dprintk(DEBUG_ERROR, "coudn't resove fid from dentry\n"); retval = -EINVAL; @@ -1292,7 +1321,7 @@ v9fs_vfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t rdev) FreeMem: kfree(mistat); kfree(fcall); - putname(symname); + __putname(symname); return retval; } diff --git a/fs/9p/vfs_super.c b/fs/9p/vfs_super.c index 868f350b2c5f..82c5b0084079 100644 --- a/fs/9p/vfs_super.c +++ b/fs/9p/vfs_super.c @@ -129,8 +129,7 @@ static struct super_block *v9fs_get_sb(struct file_system_type if ((newfid = v9fs_session_init(v9ses, dev_name, data)) < 0) { dprintk(DEBUG_ERROR, "problem initiating session\n"); - retval = newfid; - goto free_session; + return ERR_PTR(newfid); } sb = sget(fs_type, NULL, v9fs_set_super, v9ses); @@ -150,28 +149,24 @@ static struct super_block *v9fs_get_sb(struct file_system_type if (!root) { retval = -ENOMEM; - goto release_inode; + goto put_back_sb; } sb->s_root = root; - /* Setup the Root Inode */ - root_fid = v9fs_fid_create(root); - if (root_fid == NULL) { - retval = -ENOMEM; - goto release_dentry; - } - - root_fid->fidopen = 0; - root_fid->v9ses = v9ses; - stat_result = v9fs_t_stat(v9ses, newfid, &fcall); if (stat_result < 0) { dprintk(DEBUG_ERROR, "stat error\n"); v9fs_t_clunk(v9ses, newfid, NULL); v9fs_put_idpool(newfid, &v9ses->fidpool); } else { - root_fid->fid = newfid; + /* Setup the Root Inode */ + root_fid = v9fs_fid_create(root, v9ses, newfid, 0); + if (root_fid == NULL) { + retval = -ENOMEM; + goto put_back_sb; + } + root_fid->qid = fcall->params.rstat.stat->qid; root->d_inode->i_ino = v9fs_qid2ino(&fcall->params.rstat.stat->qid); @@ -182,25 +177,15 @@ static struct super_block *v9fs_get_sb(struct file_system_type if (stat_result < 0) { retval = stat_result; - goto release_dentry; + goto put_back_sb; } return sb; - release_dentry: - dput(sb->s_root); - - release_inode: - iput(inode); - - put_back_sb: +put_back_sb: + /* deactivate_super calls v9fs_kill_super which will frees the rest */ up_write(&sb->s_umount); deactivate_super(sb); - v9fs_session_close(v9ses); - - free_session: - kfree(v9ses); - return ERR_PTR(retval); } diff --git a/fs/Kconfig b/fs/Kconfig index 068ccea2f184..d5255e627b5f 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -472,6 +472,9 @@ config FUSE_FS utilities is available from the FUSE homepage: <http://fuse.sourceforge.net/> + See <file:Documentation/filesystems/fuse.txt> for more information. + See <file:Documentation/Changes> for needed library/utility version. + If you want to develop a userspace FS, or if you want to use a filesystem based on FUSE, answer Y or M. @@ -807,7 +810,7 @@ config TMPFS config HUGETLBFS bool "HugeTLB file system support" - depends X86 || IA64 || PPC64 || SPARC64 || SUPERH || X86_64 || BROKEN + depends X86 || IA64 || PPC64 || SPARC64 || SUPERH || BROKEN config HUGETLB_PAGE def_bool HUGETLBFS @@ -895,6 +898,7 @@ config AFFS_FS config HFS_FS tristate "Apple Macintosh file system support (EXPERIMENTAL)" depends on EXPERIMENTAL + select NLS help If you say Y here, you will be able to mount Macintosh-formatted floppy disks and hard drive partitions with full read-write access. @@ -1047,6 +1051,19 @@ config JFFS2_FS_WRITEBUFFER - NOR flash with transparent ECC - DataFlash +config JFFS2_SUMMARY + bool "JFFS2 summary support (EXPERIMENTAL)" + depends on JFFS2_FS && EXPERIMENTAL + default n + help + This feature makes it possible to use summary information + for faster filesystem mount. + + The summary information can be inserted into a filesystem image + by the utility 'sumtool'. + + If unsure, say 'N'. + config JFFS2_COMPRESSION_OPTIONS bool "Advanced compression options for JFFS2" depends on JFFS2_FS @@ -1068,10 +1085,10 @@ config JFFS2_ZLIB default y help Zlib is designed to be a free, general-purpose, legally unencumbered, - lossless data-compression library for use on virtually any computer + lossless data-compression library for use on virtually any computer hardware and operating system. See <http://www.gzip.org/zlib/> for further information. - + Say 'Y' if unsure. config JFFS2_RTIME @@ -1093,7 +1110,7 @@ choice default JFFS2_CMODE_PRIORITY depends on JFFS2_FS help - You can set here the default compression mode of JFFS2 from + You can set here the default compression mode of JFFS2 from the available compression modes. Don't touch if unsure. config JFFS2_CMODE_NONE @@ -1104,13 +1121,13 @@ config JFFS2_CMODE_NONE config JFFS2_CMODE_PRIORITY bool "priority" help - Tries the compressors in a predefinied order and chooses the first + Tries the compressors in a predefinied order and chooses the first successful one. config JFFS2_CMODE_SIZE bool "size (EXPERIMENTAL)" help - Tries all compressors and chooses the one which has the smallest + Tries all compressors and chooses the one which has the smallest result. endchoice @@ -1584,9 +1601,10 @@ config CIFS PC operating systems. The CIFS protocol is fully supported by file servers such as Windows 2000 (including Windows 2003, NT 4 and Windows XP) as well by Samba (which provides excellent CIFS - server support for Linux and many other operating systems). Currently - you must use the smbfs client filesystem to access older SMB servers - such as Windows 9x and OS/2. + server support for Linux and many other operating systems). Limited + support for Windows ME and similar servers is provided as well. + You must use the smbfs client filesystem to access older SMB servers + such as OS/2 and DOS. The intent of the cifs module is to provide an advanced network file system client for mounting to CIFS compliant servers, @@ -1597,7 +1615,7 @@ config CIFS cifs if running only a (Samba) server. It is possible to enable both smbfs and cifs (e.g. if you are using CIFS for accessing Windows 2003 and Samba 3 servers, and smbfs for accessing old servers). If you need - to mount to Samba or Windows 2003 servers from this machine, say Y. + to mount to Samba or Windows from this machine, say Y. config CIFS_STATS bool "CIFS statistics" @@ -1606,8 +1624,22 @@ config CIFS_STATS Enabling this option will cause statistics for each server share mounted by the cifs client to be displayed in /proc/fs/cifs/Stats +config CIFS_STATS2 + bool "CIFS extended statistics" + depends on CIFS_STATS + help + Enabling this option will allow more detailed statistics on SMB + request timing to be displayed in /proc/fs/cifs/DebugData and also + allow optional logging of slow responses to dmesg (depending on the + value of /proc/fs/cifs/cifsFYI, see fs/cifs/README for more details). + These additional statistics may have a minor effect on performance + and memory utilization. + + Unless you are a developer or are doing network performance analysis + or tuning, say N. + config CIFS_XATTR - bool "CIFS extended attributes (EXPERIMENTAL)" + bool "CIFS extended attributes" depends on CIFS help Extended attributes are name:value pairs associated with inodes by @@ -1619,11 +1651,11 @@ config CIFS_XATTR prefaced by the user namespace prefix. The system namespace (used by some filesystems to store ACLs) is not supported at this time. - + If unsure, say N. config CIFS_POSIX - bool "CIFS POSIX Extensions (EXPERIMENTAL)" + bool "CIFS POSIX Extensions" depends on CIFS_XATTR help Enabling this option will cause the cifs client to attempt to @@ -1636,10 +1668,28 @@ config CIFS_POSIX config CIFS_EXPERIMENTAL bool "CIFS Experimental Features (EXPERIMENTAL)" - depends on CIFS + depends on CIFS && EXPERIMENTAL + help + Enables cifs features under testing. These features are + experimental and currently include support for writepages + (multipage writebehind performance improvements) and directory + change notification ie fcntl(F_DNOTIFY) as well as some security + improvements. Some also depend on setting at runtime the + pseudo-file /proc/fs/cifs/Experimental (which is disabled by + default). See the file fs/cifs/README for more details. + + If unsure, say N. + +config CIFS_UPCALL + bool "CIFS Kerberos/SPNEGO advanced session setup (EXPERIMENTAL)" + depends on CIFS_EXPERIMENTAL + select CONNECTOR help - Enables cifs features under testing. These features - are highly experimental. If unsure, say N. + Enables an upcall mechanism for CIFS which will be used to contact + userspace helper utilities to provide SPNEGO packaged Kerberos + tickets which are needed to mount to certain secure servers + (for which more secure Kerberos authentication is required). If + unsure, say N. config NCP_FS tristate "NCP file system support (to mount NetWare volumes)" diff --git a/fs/Kconfig.binfmt b/fs/Kconfig.binfmt index 434c19d076ac..175b2e8177c1 100644 --- a/fs/Kconfig.binfmt +++ b/fs/Kconfig.binfmt @@ -57,7 +57,7 @@ config BINFMT_SHARED_FLAT config BINFMT_AOUT tristate "Kernel support for a.out and ECOFF binaries" - depends on (X86 && !X86_64) || ALPHA || ARM || M68K || SPARC32 + depends on X86_32 || ALPHA || ARM || M68K || SPARC32 ---help--- A.out (Assembler.OUTput) is a set of formats for libraries and executables used in the earliest versions of UNIX. Linux used diff --git a/fs/Makefile b/fs/Makefile index 1972da186272..4c2655759078 100644 --- a/fs/Makefile +++ b/fs/Makefile @@ -10,7 +10,7 @@ obj-y := open.o read_write.o file_table.o buffer.o bio.o super.o \ ioctl.o readdir.o select.o fifo.o locks.o dcache.o inode.o \ attr.o bad_inode.o file.o filesystems.o namespace.o aio.o \ seq_file.o xattr.o libfs.o fs-writeback.o mpage.o direct-io.o \ - ioprio.o + ioprio.o pnode.o obj-$(CONFIG_INOTIFY) += inotify.o obj-$(CONFIG_EPOLL) += eventpoll.o diff --git a/fs/adfs/adfs.h b/fs/adfs/adfs.h index fd528433de43..f6cd01352cc8 100644 --- a/fs/adfs/adfs.h +++ b/fs/adfs/adfs.h @@ -12,7 +12,6 @@ #define ADFS_NDA_PUBLIC_READ (1 << 5) #define ADFS_NDA_PUBLIC_WRITE (1 << 6) -#include <linux/version.h> #include "dir_f.h" struct buffer_head; diff --git a/fs/affs/file.c b/fs/affs/file.c index 6744924b6905..f72fb776ecdf 100644 --- a/fs/affs/file.c +++ b/fs/affs/file.c @@ -22,14 +22,13 @@ static int affs_grow_extcache(struct inode *inode, u32 lc_idx); static struct buffer_head *affs_alloc_extblock(struct inode *inode, struct buffer_head *bh, u32 ext); static inline struct buffer_head *affs_get_extblock(struct inode *inode, u32 ext); static struct buffer_head *affs_get_extblock_slow(struct inode *inode, u32 ext); -static ssize_t affs_file_write(struct file *filp, const char __user *buf, size_t count, loff_t *ppos); static int affs_file_open(struct inode *inode, struct file *filp); static int affs_file_release(struct inode *inode, struct file *filp); struct file_operations affs_file_operations = { .llseek = generic_file_llseek, .read = generic_file_read, - .write = affs_file_write, + .write = generic_file_write, .mmap = generic_file_mmap, .open = affs_file_open, .release = affs_file_release, @@ -473,21 +472,6 @@ affs_getemptyblk_ino(struct inode *inode, int block) return ERR_PTR(err); } -static ssize_t -affs_file_write(struct file *file, const char __user *buf, - size_t count, loff_t *ppos) -{ - ssize_t retval; - - retval = generic_file_write (file, buf, count, ppos); - if (retval >0) { - struct inode *inode = file->f_dentry->d_inode; - inode->i_ctime = inode->i_mtime = CURRENT_TIME_SEC; - mark_inode_dirty(inode); - } - return retval; -} - static int affs_do_readpage_ofs(struct file *file, struct page *page, unsigned from, unsigned to) { diff --git a/fs/affs/super.c b/fs/affs/super.c index 9c3080716c92..aaec015a16e4 100644 --- a/fs/affs/super.c +++ b/fs/affs/super.c @@ -35,8 +35,7 @@ affs_put_super(struct super_block *sb) mark_buffer_dirty(sbi->s_root_bh); } - if (sbi->s_prefix) - kfree(sbi->s_prefix); + kfree(sbi->s_prefix); affs_free_bitmap(sb); affs_brelse(sbi->s_root_bh); kfree(sbi); @@ -198,10 +197,9 @@ parse_options(char *options, uid_t *uid, gid_t *gid, int *mode, int *reserved, s *mount_opts |= SF_MUFS; break; case Opt_prefix: - if (*prefix) { /* Free any previous prefix */ - kfree(*prefix); - *prefix = NULL; - } + /* Free any previous prefix */ + kfree(*prefix); + *prefix = NULL; *prefix = match_strdup(&args[0]); if (!*prefix) return 0; @@ -462,11 +460,9 @@ got_root: out_error: if (root_inode) iput(root_inode); - if (sbi->s_bitmap) - kfree(sbi->s_bitmap); + kfree(sbi->s_bitmap); affs_brelse(root_bh); - if (sbi->s_prefix) - kfree(sbi->s_prefix); + kfree(sbi->s_prefix); kfree(sbi); sb->s_fs_info = NULL; return -EINVAL; diff --git a/fs/afs/callback.c b/fs/afs/callback.c index 2fd62f89ae01..9cb206e9d4be 100644 --- a/fs/afs/callback.c +++ b/fs/afs/callback.c @@ -19,6 +19,7 @@ #include "server.h" #include "vnode.h" #include "internal.h" +#include "cmservice.h" /*****************************************************************************/ /* diff --git a/fs/afs/file.c b/fs/afs/file.c index 23c125128024..150b19227922 100644 --- a/fs/afs/file.c +++ b/fs/afs/file.c @@ -29,26 +29,12 @@ static int afs_file_release(struct inode *inode, struct file *file); static int afs_file_readpage(struct file *file, struct page *page); static int afs_file_invalidatepage(struct page *page, unsigned long offset); -static int afs_file_releasepage(struct page *page, int gfp_flags); - -static ssize_t afs_file_write(struct file *file, const char __user *buf, - size_t size, loff_t *off); +static int afs_file_releasepage(struct page *page, gfp_t gfp_flags); struct inode_operations afs_file_inode_operations = { .getattr = afs_inode_getattr, }; -struct file_operations afs_file_file_operations = { - .read = generic_file_read, - .write = afs_file_write, - .mmap = generic_file_mmap, -#if 0 - .open = afs_file_open, - .release = afs_file_release, - .fsync = afs_file_fsync, -#endif -}; - struct address_space_operations afs_fs_aops = { .readpage = afs_file_readpage, .sync_page = block_sync_page, @@ -59,22 +45,6 @@ struct address_space_operations afs_fs_aops = { /*****************************************************************************/ /* - * AFS file write - */ -static ssize_t afs_file_write(struct file *file, const char __user *buf, - size_t size, loff_t *off) -{ - struct afs_vnode *vnode; - - vnode = AFS_FS_I(file->f_dentry->d_inode); - if (vnode->flags & AFS_VNODE_DELETED) - return -ESTALE; - - return -EIO; -} /* end afs_file_write() */ - -/*****************************************************************************/ -/* * deal with notification that a page was read from the cache */ #ifdef AFS_CACHING_SUPPORT @@ -279,7 +249,7 @@ static int afs_file_invalidatepage(struct page *page, unsigned long offset) /* * release a page and cleanup its private data */ -static int afs_file_releasepage(struct page *page, int gfp_flags) +static int afs_file_releasepage(struct page *page, gfp_t gfp_flags) { struct cachefs_page *pageio; @@ -291,12 +261,11 @@ static int afs_file_releasepage(struct page *page, int gfp_flags) cachefs_uncache_page(vnode->cache, page); #endif - pageio = (struct cachefs_page *) page->private; - page->private = 0; + pageio = (struct cachefs_page *) page_private(page); + set_page_private(page, 0); ClearPagePrivate(page); - if (pageio) - kfree(pageio); + kfree(pageio); } _leave(" = 0"); diff --git a/fs/afs/inode.c b/fs/afs/inode.c index c476fde33fbc..4ebb30a50ed5 100644 --- a/fs/afs/inode.c +++ b/fs/afs/inode.c @@ -49,7 +49,7 @@ static int afs_inode_map_status(struct afs_vnode *vnode) case AFS_FTYPE_FILE: inode->i_mode = S_IFREG | vnode->status.mode; inode->i_op = &afs_file_inode_operations; - inode->i_fop = &afs_file_file_operations; + inode->i_fop = &generic_ro_fops; break; case AFS_FTYPE_DIR: inode->i_mode = S_IFDIR | vnode->status.mode; diff --git a/fs/afs/internal.h b/fs/afs/internal.h index f09860b45c1a..ab8f87c66319 100644 --- a/fs/afs/internal.h +++ b/fs/afs/internal.h @@ -71,7 +71,6 @@ extern struct file_operations afs_dir_file_operations; */ extern struct address_space_operations afs_fs_aops; extern struct inode_operations afs_file_inode_operations; -extern struct file_operations afs_file_file_operations; #ifdef AFS_CACHING_SUPPORT extern int afs_cache_get_page_cookie(struct page *page, @@ -42,8 +42,9 @@ #endif /*------ sysctl variables----*/ -atomic_t aio_nr = ATOMIC_INIT(0); /* current system wide number of aio requests */ -unsigned aio_max_nr = 0x10000; /* system wide maximum number of aio requests */ +static DEFINE_SPINLOCK(aio_nr_lock); +unsigned long aio_nr; /* current system wide number of aio requests */ +unsigned long aio_max_nr = 0x10000; /* system wide maximum number of aio requests */ /*----end sysctl variables---*/ static kmem_cache_t *kiocb_cachep; @@ -208,7 +209,7 @@ static struct kioctx *ioctx_alloc(unsigned nr_events) return ERR_PTR(-EINVAL); } - if (nr_events > aio_max_nr) + if ((unsigned long)nr_events > aio_max_nr) return ERR_PTR(-EAGAIN); ctx = kmem_cache_alloc(kioctx_cachep, GFP_KERNEL); @@ -233,8 +234,14 @@ static struct kioctx *ioctx_alloc(unsigned nr_events) goto out_freectx; /* limit the number of system wide aios */ - atomic_add(ctx->max_reqs, &aio_nr); /* undone by __put_ioctx */ - if (unlikely(atomic_read(&aio_nr) > aio_max_nr)) + spin_lock(&aio_nr_lock); + if (aio_nr + ctx->max_reqs > aio_max_nr || + aio_nr + ctx->max_reqs < aio_nr) + ctx->max_reqs = 0; + else + aio_nr += ctx->max_reqs; + spin_unlock(&aio_nr_lock); + if (ctx->max_reqs == 0) goto out_cleanup; /* now link into global list. kludge. FIXME */ @@ -248,8 +255,6 @@ static struct kioctx *ioctx_alloc(unsigned nr_events) return ctx; out_cleanup: - atomic_sub(ctx->max_reqs, &aio_nr); - ctx->max_reqs = 0; /* prevent __put_ioctx from sub'ing aio_nr */ __put_ioctx(ctx); return ERR_PTR(-EAGAIN); @@ -374,7 +379,12 @@ void fastcall __put_ioctx(struct kioctx *ctx) pr_debug("__put_ioctx: freeing %p\n", ctx); kmem_cache_free(kioctx_cachep, ctx); - atomic_sub(nr_events, &aio_nr); + if (nr_events) { + spin_lock(&aio_nr_lock); + BUG_ON(aio_nr - nr_events > aio_nr); + aio_nr -= nr_events; + spin_unlock(&aio_nr_lock); + } } /* aio_get_req @@ -398,7 +408,7 @@ static struct kiocb fastcall *__aio_get_req(struct kioctx *ctx) if (unlikely(!req)) return NULL; - req->ki_flags = 1 << KIF_LOCKED; + req->ki_flags = 0; req->ki_users = 2; req->ki_key = 0; req->ki_ctx = ctx; @@ -447,6 +457,8 @@ static inline struct kiocb *aio_get_req(struct kioctx *ctx) static inline void really_put_req(struct kioctx *ctx, struct kiocb *req) { + assert_spin_locked(&ctx->ctx_lock); + if (req->ki_dtor) req->ki_dtor(req); kmem_cache_free(kiocb_cachep, req); @@ -488,6 +500,8 @@ static int __aio_put_req(struct kioctx *ctx, struct kiocb *req) dprintk(KERN_DEBUG "aio_put(%p): f_count=%d\n", req, atomic_read(&req->ki_filp->f_count)); + assert_spin_locked(&ctx->ctx_lock); + req->ki_users --; if (unlikely(req->ki_users < 0)) BUG(); @@ -547,25 +561,6 @@ struct kioctx *lookup_ioctx(unsigned long ctx_id) return ioctx; } -static int lock_kiocb_action(void *param) -{ - schedule(); - return 0; -} - -static inline void lock_kiocb(struct kiocb *iocb) -{ - wait_on_bit_lock(&iocb->ki_flags, KIF_LOCKED, lock_kiocb_action, - TASK_UNINTERRUPTIBLE); -} - -static inline void unlock_kiocb(struct kiocb *iocb) -{ - kiocbClearLocked(iocb); - smp_mb__after_clear_bit(); - wake_up_bit(&iocb->ki_flags, KIF_LOCKED); -} - /* * use_mm * Makes the calling kernel thread take on the specified @@ -628,14 +623,13 @@ static void unuse_mm(struct mm_struct *mm) * the kiocb (to tell the caller to activate the work * queue to process it), or 0, if it found that it was * already queued. - * - * Should be called with the spin lock iocb->ki_ctx->ctx_lock - * held */ static inline int __queue_kicked_iocb(struct kiocb *iocb) { struct kioctx *ctx = iocb->ki_ctx; + assert_spin_locked(&ctx->ctx_lock); + if (list_empty(&iocb->ki_run_list)) { list_add_tail(&iocb->ki_run_list, &ctx->run_list); @@ -741,19 +735,9 @@ static ssize_t aio_run_iocb(struct kiocb *iocb) ret = retry(iocb); current->io_wait = NULL; - if (-EIOCBRETRY != ret) { - if (-EIOCBQUEUED != ret) { - BUG_ON(!list_empty(&iocb->ki_wait.task_list)); - aio_complete(iocb, ret, 0); - /* must not access the iocb after this */ - } - } else { - /* - * Issue an additional retry to avoid waiting forever if - * no waits were queued (e.g. in case of a short read). - */ - if (list_empty(&iocb->ki_wait.task_list)) - kiocbSetKicked(iocb); + if (ret != -EIOCBRETRY && ret != -EIOCBQUEUED) { + BUG_ON(!list_empty(&iocb->ki_wait.task_list)); + aio_complete(iocb, ret, 0); } out: spin_lock_irq(&ctx->ctx_lock); @@ -790,13 +774,15 @@ out: * Process all pending retries queued on the ioctx * run list. * Assumes it is operating within the aio issuer's mm - * context. Expects to be called with ctx->ctx_lock held + * context. */ static int __aio_run_iocbs(struct kioctx *ctx) { struct kiocb *iocb; LIST_HEAD(run_list); + assert_spin_locked(&ctx->ctx_lock); + list_splice_init(&ctx->run_list, &run_list); while (!list_empty(&run_list)) { iocb = list_entry(run_list.next, struct kiocb, @@ -806,9 +792,7 @@ static int __aio_run_iocbs(struct kioctx *ctx) * Hold an extra reference while retrying i/o. */ iocb->ki_users++; /* grab extra reference */ - lock_kiocb(iocb); aio_run_iocb(iocb); - unlock_kiocb(iocb); if (__aio_put_req(ctx, iocb)) /* drop extra ref */ put_ioctx(ctx); } @@ -899,16 +883,24 @@ static void aio_kick_handler(void *data) * and if required activate the aio work queue to process * it */ -static void queue_kicked_iocb(struct kiocb *iocb) +static void try_queue_kicked_iocb(struct kiocb *iocb) { struct kioctx *ctx = iocb->ki_ctx; unsigned long flags; int run = 0; - WARN_ON((!list_empty(&iocb->ki_wait.task_list))); + /* We're supposed to be the only path putting the iocb back on the run + * list. If we find that the iocb is *back* on a wait queue already + * than retry has happened before we could queue the iocb. This also + * means that the retry could have completed and freed our iocb, no + * good. */ + BUG_ON((!list_empty(&iocb->ki_wait.task_list))); spin_lock_irqsave(&ctx->ctx_lock, flags); - run = __queue_kicked_iocb(iocb); + /* set this inside the lock so that we can't race with aio_run_iocb() + * testing it and putting the iocb on the run list under the lock */ + if (!kiocbTryKick(iocb)) + run = __queue_kicked_iocb(iocb); spin_unlock_irqrestore(&ctx->ctx_lock, flags); if (run) aio_queue_work(ctx); @@ -931,10 +923,7 @@ void fastcall kick_iocb(struct kiocb *iocb) return; } - /* If its already kicked we shouldn't queue it again */ - if (!kiocbTryKick(iocb)) { - queue_kicked_iocb(iocb); - } + try_queue_kicked_iocb(iocb); } EXPORT_SYMBOL(kick_iocb); @@ -953,28 +942,19 @@ int fastcall aio_complete(struct kiocb *iocb, long res, long res2) unsigned long tail; int ret; - /* Special case handling for sync iocbs: events go directly - * into the iocb for fast handling. Note that this will not - * work if we allow sync kiocbs to be cancelled. in which - * case the usage count checks will have to move under ctx_lock - * for all cases. + /* + * Special case handling for sync iocbs: + * - events go directly into the iocb for fast handling + * - the sync task with the iocb in its stack holds the single iocb + * ref, no other paths have a way to get another ref + * - the sync task helpfully left a reference to itself in the iocb */ if (is_sync_kiocb(iocb)) { - int ret; - + BUG_ON(iocb->ki_users != 1); iocb->ki_user_data = res; - if (iocb->ki_users == 1) { - iocb->ki_users = 0; - ret = 1; - } else { - spin_lock_irq(&ctx->ctx_lock); - iocb->ki_users--; - ret = (0 == iocb->ki_users); - spin_unlock_irq(&ctx->ctx_lock); - } - /* sync iocbs put the task here for us */ + iocb->ki_users = 0; wake_up_process(iocb->ki_obj.tsk); - return ret; + return 1; } info = &ctx->ring_info; @@ -1284,8 +1264,9 @@ asmlinkage long sys_io_setup(unsigned nr_events, aio_context_t __user *ctxp) goto out; ret = -EINVAL; - if (unlikely(ctx || (int)nr_events <= 0)) { - pr_debug("EINVAL: io_setup: ctx or nr_events > max\n"); + if (unlikely(ctx || nr_events == 0)) { + pr_debug("EINVAL: io_setup: ctx %lu nr_events %u\n", + ctx, nr_events); goto out; } @@ -1322,8 +1303,11 @@ asmlinkage long sys_io_destroy(aio_context_t ctx) } /* - * Default retry method for aio_read (also used for first time submit) - * Responsible for updating iocb state as retries progress + * aio_p{read,write} are the default ki_retry methods for + * IO_CMD_P{READ,WRITE}. They maintains kiocb retry state around potentially + * multiple calls to f_op->aio_read(). They loop around partial progress + * instead of returning -EIOCBRETRY because they don't have the means to call + * kick_iocb(). */ static ssize_t aio_pread(struct kiocb *iocb) { @@ -1332,25 +1316,25 @@ static ssize_t aio_pread(struct kiocb *iocb) struct inode *inode = mapping->host; ssize_t ret = 0; - ret = file->f_op->aio_read(iocb, iocb->ki_buf, - iocb->ki_left, iocb->ki_pos); + do { + ret = file->f_op->aio_read(iocb, iocb->ki_buf, + iocb->ki_left, iocb->ki_pos); + /* + * Can't just depend on iocb->ki_left to determine + * whether we are done. This may have been a short read. + */ + if (ret > 0) { + iocb->ki_buf += ret; + iocb->ki_left -= ret; + } - /* - * Can't just depend on iocb->ki_left to determine - * whether we are done. This may have been a short read. - */ - if (ret > 0) { - iocb->ki_buf += ret; - iocb->ki_left -= ret; /* - * For pipes and sockets we return once we have - * some data; for regular files we retry till we - * complete the entire read or find that we can't - * read any more data (e.g short reads). + * For pipes and sockets we return once we have some data; for + * regular files we retry till we complete the entire read or + * find that we can't read any more data (e.g short reads). */ - if (!S_ISFIFO(inode->i_mode) && !S_ISSOCK(inode->i_mode)) - ret = -EIOCBRETRY; - } + } while (ret > 0 && iocb->ki_left > 0 && + !S_ISFIFO(inode->i_mode) && !S_ISSOCK(inode->i_mode)); /* This means we must have transferred all that we could */ /* No need to retry anymore */ @@ -1360,27 +1344,21 @@ static ssize_t aio_pread(struct kiocb *iocb) return ret; } -/* - * Default retry method for aio_write (also used for first time submit) - * Responsible for updating iocb state as retries progress - */ +/* see aio_pread() */ static ssize_t aio_pwrite(struct kiocb *iocb) { struct file *file = iocb->ki_filp; ssize_t ret = 0; - ret = file->f_op->aio_write(iocb, iocb->ki_buf, - iocb->ki_left, iocb->ki_pos); - - if (ret > 0) { - iocb->ki_buf += ret; - iocb->ki_left -= ret; - - ret = -EIOCBRETRY; - } + do { + ret = file->f_op->aio_write(iocb, iocb->ki_buf, + iocb->ki_left, iocb->ki_pos); + if (ret > 0) { + iocb->ki_buf += ret; + iocb->ki_left -= ret; + } + } while (ret > 0 && iocb->ki_left > 0); - /* This means we must have transferred all that we could */ - /* No need to retry anymore */ if ((ret == 0) || (iocb->ki_left == 0)) ret = iocb->ki_nbytes - iocb->ki_left; @@ -1426,6 +1404,9 @@ static ssize_t aio_setup_iocb(struct kiocb *kiocb) if (unlikely(!access_ok(VERIFY_WRITE, kiocb->ki_buf, kiocb->ki_left))) break; + ret = security_file_permission(file, MAY_READ); + if (unlikely(ret)) + break; ret = -EINVAL; if (file->f_op->aio_read) kiocb->ki_retry = aio_pread; @@ -1438,6 +1419,9 @@ static ssize_t aio_setup_iocb(struct kiocb *kiocb) if (unlikely(!access_ok(VERIFY_READ, kiocb->ki_buf, kiocb->ki_left))) break; + ret = security_file_permission(file, MAY_WRITE); + if (unlikely(ret)) + break; ret = -EINVAL; if (file->f_op->aio_write) kiocb->ki_retry = aio_pwrite; @@ -1550,7 +1534,6 @@ int fastcall io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb, spin_lock_irq(&ctx->ctx_lock); aio_run_iocb(req); - unlock_kiocb(req); if (!list_empty(&ctx->run_list)) { /* drain the run list */ while (__aio_run_iocbs(ctx)) @@ -1626,12 +1609,14 @@ asmlinkage long sys_io_submit(aio_context_t ctx_id, long nr, /* lookup_kiocb * Finds a given iocb for cancellation. - * MUST be called with ctx->ctx_lock held. */ static struct kiocb *lookup_kiocb(struct kioctx *ctx, struct iocb __user *iocb, u32 key) { struct list_head *pos; + + assert_spin_locked(&ctx->ctx_lock); + /* TODO: use a hash or array, this sucks. */ list_for_each(pos, &ctx->active_reqs) { struct kiocb *kiocb = list_kiocb(pos); @@ -1682,7 +1667,6 @@ asmlinkage long sys_io_cancel(aio_context_t ctx_id, struct iocb __user *iocb, if (NULL != cancel) { struct io_event tmp; pr_debug("calling cancel\n"); - lock_kiocb(kiocb); memset(&tmp, 0, sizeof(tmp)); tmp.obj = (u64)(unsigned long)kiocb->ki_obj.user; tmp.data = kiocb->ki_user_data; @@ -1694,7 +1678,6 @@ asmlinkage long sys_io_cancel(aio_context_t ctx_id, struct iocb __user *iocb, if (copy_to_user(result, &tmp, sizeof(tmp))) ret = -EFAULT; } - unlock_kiocb(kiocb); } else ret = -EINVAL; diff --git a/fs/attr.c b/fs/attr.c index b1796fb9e524..67bcd9b14ea5 100644 --- a/fs/attr.c +++ b/fs/attr.c @@ -117,9 +117,6 @@ int notify_change(struct dentry * dentry, struct iattr * attr) struct timespec now; unsigned int ia_valid = attr->ia_valid; - if (!inode) - BUG(); - mode = inode->i_mode; now = current_fs_time(inode->i_sb); diff --git a/fs/autofs/waitq.c b/fs/autofs/waitq.c index 1fcaa1568541..633f628005b4 100644 --- a/fs/autofs/waitq.c +++ b/fs/autofs/waitq.c @@ -150,10 +150,8 @@ int autofs_wait(struct autofs_sb_info *sbi, struct qstr *name) if ( sbi->catatonic ) { /* We might have slept, so check again for catatonic mode */ wq->status = -ENOENT; - if ( wq->name ) { - kfree(wq->name); - wq->name = NULL; - } + kfree(wq->name); + wq->name = NULL; } if ( wq->name ) { diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c index 0a3c05d10167..818b37be5153 100644 --- a/fs/autofs4/inode.c +++ b/fs/autofs4/inode.c @@ -22,10 +22,8 @@ static void ino_lnkfree(struct autofs_info *ino) { - if (ino->u.symlink) { - kfree(ino->u.symlink); - ino->u.symlink = NULL; - } + kfree(ino->u.symlink); + ino->u.symlink = NULL; } struct autofs_info *autofs4_init_ino(struct autofs_info *ino, diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c index 3df86285a1c7..394ff36ef8f1 100644 --- a/fs/autofs4/waitq.c +++ b/fs/autofs4/waitq.c @@ -243,10 +243,8 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry, if ( sbi->catatonic ) { /* We might have slept, so check again for catatonic mode */ wq->status = -ENOENT; - if ( wq->name ) { - kfree(wq->name); - wq->name = NULL; - } + kfree(wq->name); + wq->name = NULL; } if ( wq->name ) { diff --git a/fs/befs/attribute.c b/fs/befs/attribute.c deleted file mode 100644 index e329d727053e..000000000000 --- a/fs/befs/attribute.c +++ /dev/null @@ -1,117 +0,0 @@ -/* - * linux/fs/befs/attribute.c - * - * Copyright (C) 2002 Will Dyson <will_dyson@pobox.com> - * - * Many thanks to Dominic Giampaolo, author of "Practical File System - * Design with the Be File System", for such a helpful book. - * - */ - -#include <linux/fs.h> -#include <linux/kernel.h> -#include <linux/string.h> - -#include "befs.h" -#include "endian.h" - -#define SD_DATA(sd)\ - (void*)((char*)sd + sizeof(*sd) + (sd->name_size - sizeof(sd->name))) - -#define SD_NEXT(sd)\ - (befs_small_data*)((char*)sd + sizeof(*sd) + (sd->name_size - \ - sizeof(sd->name) + sd->data_size)) - -int -list_small_data(struct super_block *sb, befs_inode * inode, filldir_t filldir); - -befs_small_data * -find_small_data(struct super_block *sb, befs_inode * inode, - const char *name); -int -read_small_data(struct super_block *sb, befs_inode * inode, - befs_small_data * sdata, void *buf, size_t bufsize); - -/** - * - * - * - * - * - */ -befs_small_data * -find_small_data(struct super_block *sb, befs_inode * inode, const char *name) -{ - befs_small_data *sdata = inode->small_data; - - while (sdata->type != 0) { - if (strcmp(name, sdata->name) != 0) { - return sdata; - } - sdata = SD_NEXT(sdata); - } - return NULL; -} - -/** - * - * - * - * - * - */ -int -read_small_data(struct super_block *sb, befs_inode * inode, - const char *name, void *buf, size_t bufsize) -{ - befs_small_data *sdata; - - sdata = find_small_data(sb, inode, name); - if (sdata == NULL) - return BEFS_ERR; - else if (sdata->data_size > bufsize) - return BEFS_ERR; - - memcpy(buf, SD_DATA(sdata), sdata->data_size); - - return BEFS_OK; -} - -/** - * - * - * - * - * - */ -int -list_small_data(struct super_block *sb, befs_inode * inode) -{ - -} - -/** - * - * - * - * - * - */ -int -list_attr(struct super_block *sb, befs_inode * inode) -{ - -} - -/** - * - * - * - * - * - */ -int -read_attr(struct super_block *sb, befs_inode * inode) -{ - -} diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c index e0a6025f1d06..2d365cb8eec6 100644 --- a/fs/befs/linuxvfs.c +++ b/fs/befs/linuxvfs.c @@ -73,12 +73,6 @@ static struct inode_operations befs_dir_inode_operations = { .lookup = befs_lookup, }; -static struct file_operations befs_file_operations = { - .llseek = default_llseek, - .read = generic_file_read, - .mmap = generic_file_readonly_mmap, -}; - static struct address_space_operations befs_aops = { .readpage = befs_readpage, .sync_page = block_sync_page, @@ -398,7 +392,7 @@ befs_read_inode(struct inode *inode) inode->i_mapping->a_ops = &befs_aops; if (S_ISREG(inode->i_mode)) { - inode->i_fop = &befs_file_operations; + inode->i_fop = &generic_ro_fops; } else if (S_ISDIR(inode->i_mode)) { inode->i_op = &befs_dir_inode_operations; inode->i_fop = &befs_dir_operations; @@ -731,20 +725,16 @@ parse_options(char *options, befs_mount_options * opts) static void befs_put_super(struct super_block *sb) { - if (BEFS_SB(sb)->mount_opts.iocharset) { - kfree(BEFS_SB(sb)->mount_opts.iocharset); - BEFS_SB(sb)->mount_opts.iocharset = NULL; - } + kfree(BEFS_SB(sb)->mount_opts.iocharset); + BEFS_SB(sb)->mount_opts.iocharset = NULL; if (BEFS_SB(sb)->nls) { unload_nls(BEFS_SB(sb)->nls); BEFS_SB(sb)->nls = NULL; } - if (sb->s_fs_info) { - kfree(sb->s_fs_info); - sb->s_fs_info = NULL; - } + kfree(sb->s_fs_info); + sb->s_fs_info = NULL; return; } diff --git a/fs/bfs/dir.c b/fs/bfs/dir.c index e240c335eb23..5af928fa0449 100644 --- a/fs/bfs/dir.c +++ b/fs/bfs/dir.c @@ -108,7 +108,7 @@ static int bfs_create(struct inode * dir, struct dentry * dentry, int mode, inode->i_mapping->a_ops = &bfs_aops; inode->i_mode = mode; inode->i_ino = ino; - BFS_I(inode)->i_dsk_ino = cpu_to_le16(ino); + BFS_I(inode)->i_dsk_ino = ino; BFS_I(inode)->i_sblock = 0; BFS_I(inode)->i_eblock = 0; insert_inode_hash(inode); diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c index c7b39aa279d7..3af6c73c5b5a 100644 --- a/fs/bfs/inode.c +++ b/fs/bfs/inode.c @@ -357,28 +357,46 @@ static int bfs_fill_super(struct super_block *s, void *data, int silent) } info->si_blocks = (le32_to_cpu(bfs_sb->s_end) + 1)>>BFS_BSIZE_BITS; /* for statfs(2) */ - info->si_freeb = (le32_to_cpu(bfs_sb->s_end) + 1 - cpu_to_le32(bfs_sb->s_start))>>BFS_BSIZE_BITS; + info->si_freeb = (le32_to_cpu(bfs_sb->s_end) + 1 - le32_to_cpu(bfs_sb->s_start))>>BFS_BSIZE_BITS; info->si_freei = 0; info->si_lf_eblk = 0; info->si_lf_sblk = 0; info->si_lf_ioff = 0; + bh = NULL; for (i=BFS_ROOT_INO; i<=info->si_lasti; i++) { - inode = iget(s,i); - if (BFS_I(inode)->i_dsk_ino == 0) + struct bfs_inode *di; + int block = (i - BFS_ROOT_INO)/BFS_INODES_PER_BLOCK + 1; + int off = (i - BFS_ROOT_INO) % BFS_INODES_PER_BLOCK; + unsigned long sblock, eblock; + + if (!off) { + brelse(bh); + bh = sb_bread(s, block); + } + + if (!bh) + continue; + + di = (struct bfs_inode *)bh->b_data + off; + + if (!di->i_ino) { info->si_freei++; - else { - set_bit(i, info->si_imap); - info->si_freeb -= inode->i_blocks; - if (BFS_I(inode)->i_eblock > info->si_lf_eblk) { - info->si_lf_eblk = BFS_I(inode)->i_eblock; - info->si_lf_sblk = BFS_I(inode)->i_sblock; - info->si_lf_ioff = BFS_INO2OFF(i); - } + continue; + } + set_bit(i, info->si_imap); + info->si_freeb -= BFS_FILEBLOCKS(di); + + sblock = le32_to_cpu(di->i_sblock); + eblock = le32_to_cpu(di->i_eblock); + if (eblock > info->si_lf_eblk) { + info->si_lf_eblk = eblock; + info->si_lf_sblk = sblock; + info->si_lf_ioff = BFS_INO2OFF(i); } - iput(inode); } + brelse(bh); if (!(s->s_flags & MS_RDONLY)) { - mark_buffer_dirty(bh); + mark_buffer_dirty(info->si_sbh); s->s_dirt = 1; } dump_imap("read_super", s); diff --git a/fs/binfmt_aout.c b/fs/binfmt_aout.c index dd9baabaf016..72011826f0cb 100644 --- a/fs/binfmt_aout.c +++ b/fs/binfmt_aout.c @@ -318,7 +318,6 @@ static int load_aout_binary(struct linux_binprm * bprm, struct pt_regs * regs) current->mm->free_area_cache = current->mm->mmap_base; current->mm->cached_hole_size = 0; - set_mm_counter(current->mm, rss, 0); current->mm->mmap = NULL; compute_creds(bprm); current->flags &= ~PF_FORKNOEXEC; diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 7976a238f0a3..f36f2210204f 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -773,7 +773,6 @@ static int load_elf_binary(struct linux_binprm * bprm, struct pt_regs * regs) /* Do this so that we can load the interpreter, if need be. We will change some of these later */ - set_mm_counter(current->mm, rss, 0); current->mm->free_area_cache = current->mm->mmap_base; current->mm->cached_hole_size = 0; retval = setup_arg_pages(bprm, randomize_stack_top(STACK_TOP), @@ -905,7 +904,7 @@ static int load_elf_binary(struct linux_binprm * bprm, struct pt_regs * regs) send_sig(SIGKILL, current, 0); goto out_free_dentry; } - if (padzero(elf_bss)) { + if (likely(elf_bss != elf_brk) && unlikely(padzero(elf_bss))) { send_sig(SIGSEGV, current, 0); retval = -EFAULT; /* Nobody gets to see this, but.. */ goto out_free_dentry; @@ -1007,8 +1006,7 @@ out_free_dentry: if (interpreter) fput(interpreter); out_free_interp: - if (elf_interpreter) - kfree(elf_interpreter); + kfree(elf_interpreter); out_free_file: sys_close(elf_exec_fileno); out_free_fh: @@ -1503,9 +1501,7 @@ static int elf_core_dump(long signr, struct pt_regs * regs, struct file * file) fill_psinfo(psinfo, current->group_leader, current->mm); fill_note(notes +1, "CORE", NT_PRPSINFO, sizeof(*psinfo), psinfo); - fill_note(notes +2, "CORE", NT_TASKSTRUCT, sizeof(*current), current); - - numnote = 3; + numnote = 2; auxv = (elf_addr_t *) current->mm->saved_auxv; diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c index 134c9c0d1f54..e0344f69c79d 100644 --- a/fs/binfmt_elf_fdpic.c +++ b/fs/binfmt_elf_fdpic.c @@ -294,14 +294,7 @@ static int load_elf_fdpic_binary(struct linux_binprm *bprm, struct pt_regs *regs &interp_params, ¤t->mm->start_stack, ¤t->mm->start_brk); -#endif - - /* do this so that we can load the interpreter, if need be - * - we will change some of these later - */ - set_mm_counter(current->mm, rss, 0); -#ifdef CONFIG_MMU retval = setup_arg_pages(bprm, current->mm->start_stack, executable_stack); if (retval < 0) { send_sig(SIGKILL, current, 0); @@ -418,16 +411,11 @@ error: allow_write_access(interpreter); fput(interpreter); } - if (interpreter_name) - kfree(interpreter_name); - if (exec_params.phdrs) - kfree(exec_params.phdrs); - if (exec_params.loadmap) - kfree(exec_params.loadmap); - if (interp_params.phdrs) - kfree(interp_params.phdrs); - if (interp_params.loadmap) - kfree(interp_params.loadmap); + kfree(interpreter_name); + kfree(exec_params.phdrs); + kfree(exec_params.loadmap); + kfree(interp_params.phdrs); + kfree(interp_params.loadmap); return retval; /* unrecoverable error - kill the process */ diff --git a/fs/binfmt_flat.c b/fs/binfmt_flat.c index 7974efa107bc..9d6625829b99 100644 --- a/fs/binfmt_flat.c +++ b/fs/binfmt_flat.c @@ -650,7 +650,6 @@ static int load_flat_file(struct linux_binprm * bprm, current->mm->start_brk = datapos + data_len + bss_len; current->mm->brk = (current->mm->start_brk + 3) & ~3; current->mm->context.end_brk = memp + ksize((void *) memp) - stack_len; - set_mm_counter(current->mm, rss, 0); } if (flags & FLAT_FLAG_KTRACE) diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c index 8ae0db6cd69c..2568eb41cb3a 100644 --- a/fs/binfmt_misc.c +++ b/fs/binfmt_misc.c @@ -150,7 +150,7 @@ static int load_misc_binary(struct linux_binprm *bprm, struct pt_regs *regs) /* if the binary is not readable than enforce mm->dumpable=0 regardless of the interpreter's permissions */ - if (permission(bprm->file->f_dentry->d_inode, MAY_READ, NULL)) + if (file_permission(bprm->file, MAY_READ)) bprm->interp_flags |= BINPRM_FLAGS_ENFORCE_NONDUMP; allow_write_access(bprm->file); diff --git a/fs/binfmt_som.c b/fs/binfmt_som.c index 227a2682d2bf..00a91dc25d16 100644 --- a/fs/binfmt_som.c +++ b/fs/binfmt_som.c @@ -259,7 +259,6 @@ load_som_binary(struct linux_binprm * bprm, struct pt_regs * regs) create_som_tables(bprm); current->mm->start_stack = bprm->p; - set_mm_counter(current->mm, rss, 0); #if 0 printk("(start_brk) %08lx\n" , (unsigned long) current->mm->start_brk); @@ -75,7 +75,7 @@ struct bio_set { */ static struct bio_set *fs_bio_set; -static inline struct bio_vec *bvec_alloc_bs(unsigned int __nocast gfp_mask, int nr, unsigned long *idx, struct bio_set *bs) +static inline struct bio_vec *bvec_alloc_bs(gfp_t gfp_mask, int nr, unsigned long *idx, struct bio_set *bs) { struct bio_vec *bvl; struct biovec_slab *bp; @@ -155,7 +155,7 @@ inline void bio_init(struct bio *bio) * allocate bio and iovecs from the memory pools specified by the * bio_set structure. **/ -struct bio *bio_alloc_bioset(unsigned int __nocast gfp_mask, int nr_iovecs, struct bio_set *bs) +struct bio *bio_alloc_bioset(gfp_t gfp_mask, int nr_iovecs, struct bio_set *bs) { struct bio *bio = mempool_alloc(bs->bio_pool, gfp_mask); @@ -181,7 +181,7 @@ out: return bio; } -struct bio *bio_alloc(unsigned int __nocast gfp_mask, int nr_iovecs) +struct bio *bio_alloc(gfp_t gfp_mask, int nr_iovecs) { struct bio *bio = bio_alloc_bioset(gfp_mask, nr_iovecs, fs_bio_set); @@ -277,7 +277,7 @@ inline void __bio_clone(struct bio *bio, struct bio *bio_src) * * Like __bio_clone, only also allocates the returned bio */ -struct bio *bio_clone(struct bio *bio, unsigned int __nocast gfp_mask) +struct bio *bio_clone(struct bio *bio, gfp_t gfp_mask) { struct bio *b = bio_alloc_bioset(gfp_mask, bio->bi_max_vecs, fs_bio_set); @@ -778,7 +778,7 @@ static int bio_map_kern_endio(struct bio *bio, unsigned int bytes_done, int err) static struct bio *__bio_map_kern(request_queue_t *q, void *data, - unsigned int len, unsigned int gfp_mask) + unsigned int len, gfp_t gfp_mask) { unsigned long kaddr = (unsigned long)data; unsigned long end = (kaddr + len + PAGE_SIZE - 1) >> PAGE_SHIFT; @@ -825,7 +825,7 @@ static struct bio *__bio_map_kern(request_queue_t *q, void *data, * device. Returns an error pointer in case of error. */ struct bio *bio_map_kern(request_queue_t *q, void *data, unsigned int len, - unsigned int gfp_mask) + gfp_t gfp_mask) { struct bio *bio; @@ -1078,7 +1078,7 @@ struct bio_pair *bio_split(struct bio *bi, mempool_t *pool, int first_sectors) return bp; } -static void *bio_pair_alloc(unsigned int __nocast gfp_flags, void *data) +static void *bio_pair_alloc(gfp_t gfp_flags, void *data) { return kmalloc(sizeof(struct bio_pair), gfp_flags); } diff --git a/fs/buffer.c b/fs/buffer.c index 6cbfceabd95d..5287be18633b 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -96,7 +96,7 @@ static void __clear_page_buffers(struct page *page) { ClearPagePrivate(page); - page->private = 0; + set_page_private(page, 0); page_cache_release(page); } @@ -396,7 +396,7 @@ asmlinkage long sys_fdatasync(unsigned int fd) * private_lock is contended then so is mapping->tree_lock). */ static struct buffer_head * -__find_get_block_slow(struct block_device *bdev, sector_t block, int unused) +__find_get_block_slow(struct block_device *bdev, sector_t block) { struct inode *bd_inode = bdev->bd_inode; struct address_space *bd_mapping = bd_inode->i_mapping; @@ -502,7 +502,7 @@ static void free_more_memory(void) yield(); for_each_pgdat(pgdat) { - zones = pgdat->node_zonelists[GFP_NOFS&GFP_ZONEMASK].zones; + zones = pgdat->node_zonelists[gfp_zone(GFP_NOFS)].zones; if (*zones) try_to_free_pages(zones, GFP_NOFS); } @@ -1438,7 +1438,7 @@ __find_get_block(struct block_device *bdev, sector_t block, int size) struct buffer_head *bh = lookup_bh_lru(bdev, block, size); if (bh == NULL) { - bh = __find_get_block_slow(bdev, block, size); + bh = __find_get_block_slow(bdev, block); if (bh) bh_lru_install(bh); } @@ -1478,8 +1478,10 @@ EXPORT_SYMBOL(__getblk); void __breadahead(struct block_device *bdev, sector_t block, int size) { struct buffer_head *bh = __getblk(bdev, block, size); - ll_rw_block(READA, 1, &bh); - brelse(bh); + if (likely(bh)) { + ll_rw_block(READA, 1, &bh); + brelse(bh); + } } EXPORT_SYMBOL(__breadahead); @@ -1497,7 +1499,7 @@ __bread(struct block_device *bdev, sector_t block, int size) { struct buffer_head *bh = __getblk(bdev, block, size); - if (!buffer_uptodate(bh)) + if (likely(bh) && !buffer_uptodate(bh)) bh = __bread_slow(bh); return bh; } @@ -1571,7 +1573,7 @@ static inline void discard_buffer(struct buffer_head * bh) * * NOTE: @gfp_mask may go away, and this function may become non-blocking. */ -int try_to_release_page(struct page *page, int gfp_mask) +int try_to_release_page(struct page *page, gfp_t gfp_mask) { struct address_space * const mapping = page->mapping; @@ -1637,6 +1639,15 @@ out: } EXPORT_SYMBOL(block_invalidatepage); +int do_invalidatepage(struct page *page, unsigned long offset) +{ + int (*invalidatepage)(struct page *, unsigned long); + invalidatepage = page->mapping->a_ops->invalidatepage; + if (invalidatepage == NULL) + invalidatepage = block_invalidatepage; + return (*invalidatepage)(page, offset); +} + /* * We attach and possibly dirty the buffers atomically wrt * __set_page_dirty_buffers() via private_lock. try_to_free_buffers @@ -1694,7 +1705,7 @@ void unmap_underlying_metadata(struct block_device *bdev, sector_t block) might_sleep(); - old_bh = __find_get_block_slow(bdev, block, 0); + old_bh = __find_get_block_slow(bdev, block); if (old_bh) { clear_buffer_dirty(old_bh); wait_on_buffer(old_bh); @@ -2696,7 +2707,7 @@ int block_write_full_page(struct page *page, get_block_t *get_block, * they may have been added in ext3_writepage(). Make them * freeable here, so the page does not leak. */ - block_invalidatepage(page, 0); + do_invalidatepage(page, 0); unlock_page(page); return 0; /* don't care */ } @@ -3045,7 +3056,7 @@ static void recalc_bh_state(void) buffer_heads_over_limit = (tot > max_buffer_heads); } -struct buffer_head *alloc_buffer_head(unsigned int __nocast gfp_flags) +struct buffer_head *alloc_buffer_head(gfp_t gfp_flags) { struct buffer_head *ret = kmem_cache_alloc(bh_cachep, gfp_flags); if (ret) { diff --git a/fs/cifs/AUTHORS b/fs/cifs/AUTHORS index 72fdc10dfdd7..8848e4dfa026 100644 --- a/fs/cifs/AUTHORS +++ b/fs/cifs/AUTHORS @@ -32,6 +32,10 @@ Domen Puncer Jesper Juhl (in particular for lots of whitespace/formatting cleanup) Vince Negri and Dave Stahl (for finding an important caching bug) Adrian Bunk (kcalloc cleanups) +Miklos Szeredi +Kazeon team for various fixes especially for 2.4 version. +Asser Ferno (Change Notify support) +Shaggy (Dave Kleikamp) for inumerable small fs suggestions and some good cleanup Test case and Bug Report contributors ------------------------------------- diff --git a/fs/cifs/CHANGES b/fs/cifs/CHANGES index 3196d4c4eed3..943ef9b82244 100644 --- a/fs/cifs/CHANGES +++ b/fs/cifs/CHANGES @@ -1,8 +1,56 @@ +Version 1.39 +------------ +Defer close of a file handle slightly if pending writes depend on that handle +(this reduces the EBADF bad file handle errors that can be logged under heavy +stress on writes). Modify cifs Kconfig options to expose CONFIG_CIFS_STATS2 +Fix SFU style symlinks and mknod needed for servers which do not support the +CIFS Unix Extensions. Fix setfacl/getfacl on bigendian. Timeout negative +dentries so files that the client sees as deleted but that later get created +on the server will be recognized. Add client side permission check on setattr. + +Version 1.38 +------------ +Fix tcp socket retransmission timeouts (e.g. on ENOSPACE from the socket) +to be smaller at first (but increasing) so large write performance performance +over GigE is better. Do not hang thread on illegal byte range lock response +from Windows (Windows can send an RFC1001 size which does not match smb size) by +allowing an SMBs TCP length to be up to a few bytes longer than it should be. +wsize and rsize can now be larger than negotiated buffer size if server +supports large readx/writex, even when directio mount flag not specified. +Write size will in many cases now be 16K instead of 4K which greatly helps +file copy performance on lightly loaded networks. Fix oops in dnotify +when experimental config flag enabled. Make cifsFYI more granular. + +Version 1.37 +------------ +Fix readdir caching when unlink removes file in current search buffer, +and this is followed by a rewind search to just before the deleted entry. +Do not attempt to set ctime unless atime and/or mtime change requested +(most servers throw it away anyway). Fix length check of received smbs +to be more accurate. Fix big endian problem with mapchars mount option, +and with a field returned by statfs. + +Version 1.36 +------------ +Add support for mounting to older pre-CIFS servers such as Windows9x and ME. +For these older servers, add option for passing netbios name of server in +on mount (servernetbiosname). Add suspend support for power management, to +avoid cifsd thread preventing software suspend from working. +Add mount option for disabling the default behavior of sending byte range lock +requests to the server (necessary for certain applications which break with +mandatory lock behavior such as Evolution), and also mount option for +requesting case insensitive matching for path based requests (requesting +case sensitive is the default). + Version 1.35 ------------ Add writepage performance improvements. Fix path name conversions for long filenames on mounts which were done with "mapchars" mount option -specified. +specified. Ensure multiplex ids do not collide. Fix case in which +rmmod can oops if done soon after last unmount. Fix truncated +search (readdir) output when resume filename was a long filename. +Fix filename conversion when mapchars mount option was specified and +filename was a long filename. Version 1.34 ------------ @@ -11,7 +59,7 @@ Do not oops if root user kills cifs oplock kernel thread or kills the cifsd thread (NB: killing the cifs kernel threads is not recommended, unmount and rmmod cifs will kill them when they are no longer needed). Fix readdir to ASCII servers (ie older servers -which do not support Unicode) and also require asterik. +which do not support Unicode) and also require asterisk. Fix out of memory case in which data could be written one page off in the page cache. @@ -101,7 +149,7 @@ improperly zeroed buffer in CIFS Unix extensions set times call. Version 1.25 ------------ -Fix internationlization problem in cifs readdir with filenames that map to +Fix internationalization problem in cifs readdir with filenames that map to longer UTF8 strings than the string on the wire was in Unicode. Add workaround for readdir to netapp servers. Fix search rewind (seek into readdir to return non-consecutive entries). Do not do readdir when server negotiates @@ -276,7 +324,7 @@ Fix caching problem when files opened by multiple clients in which page cache could contain stale data, and write through did not occur often enough while file was still open when read ahead (read oplock) not allowed. Treat "sep=" when first mount option -as an overrride of comma as the default separator between mount +as an override of comma as the default separator between mount options. Version 1.01 @@ -286,7 +334,7 @@ Allow passwords longer than 16 bytes. Allow null password string. Version 1.00 ------------ Gracefully clean up failed mounts when attempting to mount to servers such as -Windows 98 that terminate tcp sessions during prototocol negotiation. Handle +Windows 98 that terminate tcp sessions during protocol negotiation. Handle embedded commas in mount parsing of passwords. Version 0.99 @@ -295,7 +343,7 @@ Invalidate local inode cached pages on oplock break and when last file instance is closed so that the client does not continue using stale local copy rather than later modified server copy of file. Do not reconnect when server drops the tcp session prematurely before negotiate -protocol response. Fix oops in roepen_file when dentry freed. Allow +protocol response. Fix oops in reopen_file when dentry freed. Allow the support for CIFS Unix Extensions to be disabled via proc interface. Version 0.98 @@ -637,7 +685,7 @@ versions of 2.4 kernel (now builds and works again on kernels at least as early Version 0.41 ------------ Various minor fixes for Connectathon Posix "basic" file i/o test suite. Directory caching fixed so hardlinked -files now return the correct rumber of links on fstat as they are repeatedly linked and unlinked. +files now return the correct number of links on fstat as they are repeatedly linked and unlinked. Version 0.40 ------------ @@ -704,7 +752,7 @@ session) and cleaned them up and made them more consistent with other cifs functions. 7) Server support for Unix extensions is now fully detected and FindFirst is implemented both ways -(with or without Unix exentions) but FindNext and QueryPathInfo with the Unix extensions are not completed, +(with or without Unix extensions) but FindNext and QueryPathInfo with the Unix extensions are not completed, nor is the symlink support using the Unix extensions 8) Started adding the readlink and follow_link code diff --git a/fs/cifs/README b/fs/cifs/README index 34b0cf7111f3..e5d09a2fc7a5 100644 --- a/fs/cifs/README +++ b/fs/cifs/README @@ -278,7 +278,9 @@ A partial list of the supported mount options follows: (such as Windows), permissions can also be checked at the client, and a crude form of client side permission checking can be enabled by specifying file_mode and dir_mode on - the client + the client. Note that the mount.cifs helper must be + at version 1.10 or higher to support specifying the uid + (or gid) in non-numberic form. gid If CIFS Unix extensions are not supported by the server this overrides the default gid for inodes. file_mode If CIFS Unix extensions are not supported by the server @@ -294,8 +296,10 @@ A partial list of the supported mount options follows: during the local client kernel build will be used. If server does not support Unicode, this parameter is unused. - rsize default read size - wsize default write size + rsize default read size (usually 16K) + wsize default write size (usually 16K, 32K is often better over GigE) + maximum wsize currently allowed by CIFS is 57344 (14 4096 byte + pages) rw mount the network share read-write (note that the server may still consider the share read-only) ro mount network share read-only @@ -343,7 +347,10 @@ A partial list of the supported mount options follows: client system. It is typically only needed when the server supports the CIFS Unix Extensions but the UIDs/GIDs on the client and server system do not match closely enough to allow - access by the user doing the mount. + access by the user doing the mount, but it may be useful with + non CIFS Unix Extension mounts for cases in which the default + mode is specified on the mount but is not to be enforced on the + client (e.g. perhaps when MultiUserMount is enabled) Note that this does not affect the normal ACL check on the target machine done by the server software (of the server ACL against the user name provided at mount time). @@ -366,15 +373,21 @@ A partial list of the supported mount options follows: setuids If the CIFS Unix extensions are negotiated with the server the client will attempt to set the effective uid and gid of the local process on newly created files, directories, and - devices (create, mkdir, mknod). + devices (create, mkdir, mknod). If the CIFS Unix Extensions + are not negotiated, for newly created files and directories + instead of using the default uid and gid specified on the + the mount, cache the new file's uid and gid locally which means + that the uid for the file can change when the inode is + reloaded (or the user remounts the share). nosetuids The client will not attempt to set the uid and gid on on newly created files, directories, and devices (create, mkdir, mknod) which will result in the server setting the uid and gid to the default (usually the server uid of the user who mounted the share). Letting the server (rather than - the client) set the uid and gid is the default. This - parameter has no effect if the CIFS Unix Extensions are not - negotiated. + the client) set the uid and gid is the default. If the CIFS + Unix Extensions are not negotiated then the uid and gid for + new files will appear to be the uid (gid) of the mounter or the + uid (gid) parameter specified on the mount. netbiosname When mounting to servers via port 139, specifies the RFC1001 source name to use to represent the client netbios machine name when doing the RFC1001 netbios session initialize. @@ -407,8 +420,22 @@ A partial list of the supported mount options follows: This has no effect if the server does not support Unicode on the wire. nomapchars Do not translate any of these seven characters (default). + nocase Request case insensitive path name matching (case + sensitive is the default if the server suports it). + nobrl Do not send byte range lock requests to the server. + This is necessary for certain applications that break + with cifs style mandatory byte range locks (and most + cifs servers do not yet support requesting advisory + byte range locks). remount remount the share (often used to change from ro to rw mounts or vice versa) + sfu When the CIFS Unix Extensions are not negotiated, attempt to + create device files and fifos in a format compatible with + Services for Unix (SFU). In addition retrieve bits 10-12 + of the mode via the SETFILEBITS extended attribute (as + SFU does). In the future the bottom 9 bits of the mode + mode also will be emulated using queries of the security + descriptor (ACL). The mount.cifs mount helper also accepts a few mount options before -o including: @@ -473,9 +500,16 @@ These experimental features and tracing can be enabled by changing flags in kernel, e.g. insmod cifs). To enable a feature set it to 1 e.g. to enable tracing to the kernel message log type: - echo 1 > /proc/fs/cifs/cifsFYI + echo 7 > /proc/fs/cifs/cifsFYI -and for more extensive tracing including the start of smb requests and responses +cifsFYI functions as a bit mask. Setting it to 1 enables additional kernel +logging of various informational messages. 2 enables logging of non-zero +SMB return codes while 4 enables logging of requests that take longer +than one second to complete (except for byte range lock requests). +Setting it to 4 requires defining CONFIG_CIFS_STATS2 manually in the +source code (typically by setting it in the beginning of cifsglob.h), +and setting it to seven enables all three. Finally, tracing +the start of smb requests and responses can be enabled via: echo 1 > /proc/fs/cifs/traceSMB diff --git a/fs/cifs/TODO b/fs/cifs/TODO index 8cc881694e29..fc34c74ec4be 100644 --- a/fs/cifs/TODO +++ b/fs/cifs/TODO @@ -1,4 +1,4 @@ -version 1.34 April 29, 2005 +Version 1.39 November 30, 2005 A Partial List of Missing Features ================================== @@ -7,14 +7,14 @@ Contributions are welcome. There are plenty of opportunities for visible, important contributions to this module. Here is a partial list of the known problems and missing features: -a) Support for SecurityDescriptors for chmod/chgrp/chown so -these can be supported for Windows servers +a) Support for SecurityDescriptors(Windows/CIFS ACLs) for chmod/chgrp/chown +so that these operations can be supported to Windows servers -b) Better pam/winbind integration (e.g. to handle uid mapping -better) +b) Mapping POSIX ACLs (and eventually NFSv4 ACLs) to CIFS +SecurityDescriptors -c) multi-user mounts - multiplexed sessionsetups over single vc -(ie tcp session) - more testing needed +c) Better pam/winbind integration (e.g. to handle uid mapping +better) d) Kerberos/SPNEGO session setup support - (started) @@ -29,12 +29,17 @@ f) Directory entry caching relies on a 1 second timer, rather than using FindNotify or equivalent. - (started) g) A few byte range testcases fail due to POSIX vs. Windows/CIFS -style byte range lock differences +style byte range lock differences. Save byte range locks so +reconnect can replay them. -h) quota support +h) Support unlock all (unlock 0,MAX_OFFSET) +by unlocking all known byte range locks that we locked on the file. -j) finish writepages support (multi-page write behind for improved -performance) and syncpage +i) quota support (needs minor kernel change since quota calls +to make it to network filesystems or deviceless filesystems) + +j) investigate sync behavior (including syncpage) and check +for proper behavior of intr/nointr k) hook lower into the sockets api (as NFS/SunRPC does) to avoid the extra copy in/out of the socket buffers in some cases. @@ -53,24 +58,22 @@ o) Improve performance of readpages by sending more than one read at a time when 8 pages or more are requested. In conjuntion add support for async_cifs_readpages. -p) Add support for storing symlink and fifo info to Windows servers +p) Add support for storing symlink info to Windows servers in the Extended Attribute format their SFU clients would recognize. q) Finish fcntl D_NOTIFY support so kde and gnome file list windows -will autorefresh (started) +will autorefresh (partially complete by Asser). Needs minor kernel +vfs change to support removing D_NOTIFY on a file. r) Add GUI tool to configure /proc/fs/cifs settings and for display of the CIFS statistics (started) -q) implement support for security and trusted categories of xattrs +s) implement support for security and trusted categories of xattrs (requires minor protocol extension) to enable better support for SELINUX -r) Implement O_DIRECT flag on open (already supported on mount) - -s) Allow remapping of last remaining character (\) to +0xF000 which -(this character is valid for POSIX but not for Windows) +t) Implement O_DIRECT flag on open (already supported on mount) -t) Create UID mapping facility so server UIDs can be mapped on a per +u) Create UID mapping facility so server UIDs can be mapped on a per mount or a per server basis to client UIDs or nobody if no mapping exists. This is helpful when Unix extensions are negotiated to allow better permission checking when UIDs differ on the server @@ -78,6 +81,17 @@ and client. Add new protocol request to the CIFS protocol standard for asking the server for the corresponding name of a particular uid. +v) Add support for CIFS Unix and also the newer POSIX extensions to the +server side for Samba 4. + +w) Finish up the dos time conversion routines needed to return old server +time to the client (default time, of now or time 0 is used now for these +very old servers) + +x) Add support for OS/2 (LANMAN 1.2 and LANMAN2.1 based SMB servers) + +y) Finish testing of Windows 9x/Windows ME server support (started). + KNOWN BUGS (updated April 29, 2005) ==================================== See http://bugzilla.samba.org - search on product "CifsVFS" for diff --git a/fs/cifs/asn1.c b/fs/cifs/asn1.c index e02010dd73ec..086ae8f4a207 100644 --- a/fs/cifs/asn1.c +++ b/fs/cifs/asn1.c @@ -191,7 +191,8 @@ asn1_header_decode(struct asn1_ctx *ctx, unsigned char **eoc, unsigned int *cls, unsigned int *con, unsigned int *tag) { - unsigned int def, len; + unsigned int def = 0; + unsigned int len = 0; if (!asn1_id_decode(ctx, cls, con, tag)) return 0; @@ -552,8 +553,7 @@ decode_negTokenInit(unsigned char *security_blob, int length, *(oid + 3))); rc = compare_oid(oid, oidlen, NTLMSSP_OID, NTLMSSP_OID_LEN); - if(oid) - kfree(oid); + kfree(oid); if (rc) use_ntlmssp = TRUE; } diff --git a/fs/cifs/cifs_debug.c b/fs/cifs/cifs_debug.c index 4061e43471c1..22a444a3fe4c 100644 --- a/fs/cifs/cifs_debug.c +++ b/fs/cifs/cifs_debug.c @@ -81,6 +81,8 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset, buf += length; length = sprintf(buf,"CIFS Version %s\n",CIFS_VERSION); buf += length; + length = sprintf(buf,"Active VFS Requests: %d\n", GlobalTotalActiveXid); + buf += length; length = sprintf(buf, "Servers:"); buf += length; @@ -97,7 +99,7 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset, } else { length = sprintf(buf, - "\n%d) Name: %s Domain: %s Mounts: %d ServerOS: %s \n\tServerNOS: %s\tCapabilities: 0x%x\n\tSMB session status: %d\t", + "\n%d) Name: %s Domain: %s Mounts: %d OS: %s \n\tNOS: %s\tCapability: 0x%x\n\tSMB session status: %d\t", i, ses->serverName, ses->serverDomain, atomic_read(&ses->inUse), ses->serverOS, ses->serverNOS, @@ -105,12 +107,18 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset, buf += length; } if(ses->server) { - buf += sprintf(buf, "TCP status: %d\n\tLocal Users To Server: %d SecMode: 0x%x Req Active: %d", + buf += sprintf(buf, "TCP status: %d\n\tLocal Users To Server: %d SecMode: 0x%x Req On Wire: %d", ses->server->tcpStatus, atomic_read(&ses->server->socketUseCount), ses->server->secMode, atomic_read(&ses->server->inFlight)); - + +#ifdef CONFIG_CIFS_STATS2 + buf += sprintf(buf, " In Send: %d In MaxReq Wait: %d", + atomic_read(&ses->server->inSend), + atomic_read(&ses->server->num_waiters)); +#endif + length = sprintf(buf, "\nMIDs:\n"); buf += length; @@ -149,7 +157,7 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset, dev_type = le32_to_cpu(tcon->fsDevInfo.DeviceType); length = sprintf(buf, - "\n%d) %s Uses: %d Type: %s Characteristics: 0x%x Attributes: 0x%x\nPathComponentMax: %d Status: %d", + "\n%d) %s Uses: %d Type: %s DevInfo: 0x%x Attributes: 0x%x\nPathComponentMax: %d Status: %d", i, tcon->treeName, atomic_read(&tcon->useCount), tcon->nativeFileSystem, @@ -195,6 +203,49 @@ cifs_debug_data_read(char *buf, char **beginBuffer, off_t offset, } #ifdef CONFIG_CIFS_STATS + +static int +cifs_stats_write(struct file *file, const char __user *buffer, + unsigned long count, void *data) +{ + char c; + int rc; + struct list_head *tmp; + struct cifsTconInfo *tcon; + + rc = get_user(c, buffer); + if (rc) + return rc; + + if (c == '1' || c == 'y' || c == 'Y' || c == '0') { + read_lock(&GlobalSMBSeslock); + list_for_each(tmp, &GlobalTreeConnectionList) { + tcon = list_entry(tmp, struct cifsTconInfo, + cifsConnectionList); + atomic_set(&tcon->num_smbs_sent, 0); + atomic_set(&tcon->num_writes, 0); + atomic_set(&tcon->num_reads, 0); + atomic_set(&tcon->num_oplock_brks, 0); + atomic_set(&tcon->num_opens, 0); + atomic_set(&tcon->num_closes, 0); + atomic_set(&tcon->num_deletes, 0); + atomic_set(&tcon->num_mkdirs, 0); + atomic_set(&tcon->num_rmdirs, 0); + atomic_set(&tcon->num_renames, 0); + atomic_set(&tcon->num_t2renames, 0); + atomic_set(&tcon->num_ffirst, 0); + atomic_set(&tcon->num_fnext, 0); + atomic_set(&tcon->num_fclose, 0); + atomic_set(&tcon->num_hardlinks, 0); + atomic_set(&tcon->num_symlinks, 0); + atomic_set(&tcon->num_locks, 0); + } + read_unlock(&GlobalSMBSeslock); + } + + return count; +} + static int cifs_stats_read(char *buf, char **beginBuffer, off_t offset, int count, int *eof, void *data) @@ -254,35 +305,51 @@ cifs_stats_read(char *buf, char **beginBuffer, off_t offset, buf += sprintf(buf, "\tDISCONNECTED "); length += 14; } - item_length = sprintf(buf,"\nSMBs: %d Oplock Breaks: %d", + item_length = sprintf(buf, "\nSMBs: %d Oplock Breaks: %d", atomic_read(&tcon->num_smbs_sent), atomic_read(&tcon->num_oplock_brks)); buf += item_length; length += item_length; - item_length = sprintf(buf,"\nReads: %d Bytes %lld", + item_length = sprintf(buf, "\nReads: %d Bytes: %lld", atomic_read(&tcon->num_reads), (long long)(tcon->bytes_read)); buf += item_length; length += item_length; - item_length = sprintf(buf,"\nWrites: %d Bytes: %lld", + item_length = sprintf(buf, "\nWrites: %d Bytes: %lld", atomic_read(&tcon->num_writes), (long long)(tcon->bytes_written)); + buf += item_length; + length += item_length; + item_length = sprintf(buf, + "\nLocks: %d HardLinks: %d Symlinks: %d", + atomic_read(&tcon->num_locks), + atomic_read(&tcon->num_hardlinks), + atomic_read(&tcon->num_symlinks)); + buf += item_length; + length += item_length; + + item_length = sprintf(buf, "\nOpens: %d Closes: %d Deletes: %d", + atomic_read(&tcon->num_opens), + atomic_read(&tcon->num_closes), + atomic_read(&tcon->num_deletes)); buf += item_length; length += item_length; - item_length = sprintf(buf, - "\nOpens: %d Deletes: %d\nMkdirs: %d Rmdirs: %d", - atomic_read(&tcon->num_opens), - atomic_read(&tcon->num_deletes), + item_length = sprintf(buf, "\nMkdirs: %d Rmdirs: %d", atomic_read(&tcon->num_mkdirs), atomic_read(&tcon->num_rmdirs)); buf += item_length; length += item_length; - item_length = sprintf(buf, - "\nRenames: %d T2 Renames %d", + item_length = sprintf(buf, "\nRenames: %d T2 Renames %d", atomic_read(&tcon->num_renames), atomic_read(&tcon->num_t2renames)); buf += item_length; length += item_length; + item_length = sprintf(buf, "\nFindFirst: %d FNext %d FClose %d", + atomic_read(&tcon->num_ffirst), + atomic_read(&tcon->num_fnext), + atomic_read(&tcon->num_fclose)); + buf += item_length; + length += item_length; } read_unlock(&GlobalSMBSeslock); @@ -341,8 +408,10 @@ cifs_proc_init(void) cifs_debug_data_read, NULL); #ifdef CONFIG_CIFS_STATS - create_proc_read_entry("Stats", 0, proc_fs_cifs, + pde = create_proc_read_entry("Stats", 0, proc_fs_cifs, cifs_stats_read, NULL); + if (pde) + pde->write_proc = cifs_stats_write; #endif pde = create_proc_read_entry("cifsFYI", 0, proc_fs_cifs, cifsFYI_read, NULL); @@ -360,7 +429,7 @@ cifs_proc_init(void) if (pde) pde->write_proc = oplockEnabled_write; - pde = create_proc_read_entry("ReenableOldCifsReaddirCode", 0, proc_fs_cifs, + pde = create_proc_read_entry("Experimental", 0, proc_fs_cifs, quotaEnabled_read, NULL); if (pde) pde->write_proc = quotaEnabled_write; @@ -419,7 +488,7 @@ cifs_proc_clean(void) remove_proc_entry("ExtendedSecurity",proc_fs_cifs); remove_proc_entry("PacketSigningEnabled",proc_fs_cifs); remove_proc_entry("LinuxExtensionsEnabled",proc_fs_cifs); - remove_proc_entry("ReenableOldCifsReaddirCode",proc_fs_cifs); + remove_proc_entry("Experimental",proc_fs_cifs); remove_proc_entry("LookupCacheEnabled",proc_fs_cifs); remove_proc_entry("cifs", proc_root_fs); } @@ -459,6 +528,8 @@ cifsFYI_write(struct file *file, const char __user *buffer, cifsFYI = 0; else if (c == '1' || c == 'y' || c == 'Y') cifsFYI = 1; + else if((c > '1') && (c <= '9')) + cifsFYI = (int) (c - '0'); /* see cifs_debug.h for meanings */ return count; } diff --git a/fs/cifs/cifs_debug.h b/fs/cifs/cifs_debug.h index bf24d2828f68..4304d9dcfb6c 100644 --- a/fs/cifs/cifs_debug.h +++ b/fs/cifs/cifs_debug.h @@ -26,6 +26,9 @@ void cifs_dump_mem(char *label, void *data, int length); extern int traceSMB; /* flag which enables the function below */ void dump_smb(struct smb_hdr *, int); +#define CIFS_INFO 0x01 +#define CIFS_RC 0x02 +#define CIFS_TIMER 0x04 /* * debug ON @@ -36,7 +39,7 @@ void dump_smb(struct smb_hdr *, int); /* information message: e.g., configuration, major event */ extern int cifsFYI; -#define cifsfyi(format,arg...) if (cifsFYI) printk(KERN_DEBUG " " __FILE__ ": " format "\n" "" , ## arg) +#define cifsfyi(format,arg...) if (cifsFYI & CIFS_INFO) printk(KERN_DEBUG " " __FILE__ ": " format "\n" "" , ## arg) #define cFYI(button,prspec) if (button) cifsfyi prspec diff --git a/fs/cifs/cifs_fs_sb.h b/fs/cifs/cifs_fs_sb.h index ec00d61d5308..f799f6f0e729 100644 --- a/fs/cifs/cifs_fs_sb.h +++ b/fs/cifs/cifs_fs_sb.h @@ -24,6 +24,9 @@ #define CIFS_MOUNT_DIRECT_IO 8 /* do not write nor read through page cache */ #define CIFS_MOUNT_NO_XATTR 0x10 /* if set - disable xattr support */ #define CIFS_MOUNT_MAP_SPECIAL_CHR 0x20 /* remap illegal chars in filenames */ +#define CIFS_MOUNT_POSIX_PATHS 0x40 /* Negotiate posix pathnames if possible. */ +#define CIFS_MOUNT_UNX_EMUL 0x80 /* Network compat with SFUnix emulation */ +#define CIFS_MOUNT_NO_BRL 0x100 /* No sending byte range locks to srv */ struct cifs_sb_info { struct cifsTconInfo *tcon; /* primary mount */ diff --git a/fs/cifs/cifs_unicode.c b/fs/cifs/cifs_unicode.c index 99a096d3f84d..d2b128255944 100644 --- a/fs/cifs/cifs_unicode.c +++ b/fs/cifs/cifs_unicode.c @@ -1,7 +1,7 @@ /* * fs/cifs/cifs_unicode.c * - * Copyright (c) International Business Machines Corp., 2000,2002 + * Copyright (c) International Business Machines Corp., 2000,2005 * Modified by Steve French (sfrench@us.ibm.com) * * This program is free software; you can redistribute it and/or modify @@ -31,7 +31,7 @@ * */ int -cifs_strfromUCS_le(char *to, const wchar_t * from, /* LITTLE ENDIAN */ +cifs_strfromUCS_le(char *to, const __le16 * from, int len, const struct nls_table *codepage) { int i; @@ -60,24 +60,26 @@ cifs_strfromUCS_le(char *to, const wchar_t * from, /* LITTLE ENDIAN */ * */ int -cifs_strtoUCS(wchar_t * to, const char *from, int len, +cifs_strtoUCS(__le16 * to, const char *from, int len, const struct nls_table *codepage) { int charlen; int i; + wchar_t * wchar_to = (wchar_t *)to; /* needed to quiet sparse */ for (i = 0; len && *from; i++, from += charlen, len -= charlen) { /* works for 2.4.0 kernel or later */ - charlen = codepage->char2uni(from, len, &to[i]); + charlen = codepage->char2uni(from, len, &wchar_to[i]); if (charlen < 1) { cERROR(1, ("cifs_strtoUCS: char2uni returned %d", charlen)); - to[i] = cpu_to_le16(0x003f); /* a question mark */ + /* A question mark */ + to[i] = cpu_to_le16(0x003f); charlen = 1; } else - to[i] = cpu_to_le16(to[i]); + to[i] = cpu_to_le16(wchar_to[i]); } diff --git a/fs/cifs/cifs_unicode.h b/fs/cifs/cifs_unicode.h index da8dde965275..39e5b970325f 100644 --- a/fs/cifs/cifs_unicode.h +++ b/fs/cifs/cifs_unicode.h @@ -5,7 +5,7 @@ * Convert a unicode character to upper or lower case using * compressed tables. * - * Copyright (c) International Business Machines Corp., 2000,2002 + * Copyright (c) International Business Machines Corp., 2000,2005555555555555555555555555555555555555555555555555555555 * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -59,8 +59,8 @@ extern struct UniCaseRange UniLowerRange[]; #endif /* UNIUPR_NOLOWER */ #ifdef __KERNEL__ -int cifs_strfromUCS_le(char *, const wchar_t *, int, const struct nls_table *); -int cifs_strtoUCS(wchar_t *, const char *, int, const struct nls_table *); +int cifs_strfromUCS_le(char *, const __le16 *, int, const struct nls_table *); +int cifs_strtoUCS(__le16 *, const char *, int, const struct nls_table *); #endif /* diff --git a/fs/cifs/cifsencrypt.c b/fs/cifs/cifsencrypt.c index 1959c7c4b185..fe2bb7c4c912 100644 --- a/fs/cifs/cifsencrypt.c +++ b/fs/cifs/cifsencrypt.c @@ -149,7 +149,7 @@ int CalcNTLMv2_partial_mac_key(struct cifsSesInfo * ses, struct nls_table * nls_ char temp_hash[16]; struct HMACMD5Context ctx; char * ucase_buf; - wchar_t * unicode_buf; + __le16 * unicode_buf; unsigned int i,user_name_len,dom_name_len; if(ses == NULL) diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 8cc23e7d0d5d..2a13a2bac8f1 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -32,6 +32,7 @@ #include <linux/seq_file.h> #include <linux/vfs.h> #include <linux/mempool.h> +#include <linux/delay.h> #include "cifsfs.h" #include "cifspdu.h" #define DECLARE_GLOBALS_HERE @@ -59,6 +60,8 @@ unsigned int ntlmv2_support = 0; unsigned int sign_CIFS_PDUs = 1; extern struct task_struct * oplockThread; /* remove sparse warning */ struct task_struct * oplockThread = NULL; +extern struct task_struct * dnotifyThread; /* remove sparse warning */ +struct task_struct * dnotifyThread = NULL; unsigned int CIFSMaxBufSize = CIFS_MAX_MSGSIZE; module_param(CIFSMaxBufSize, int, 0); MODULE_PARM_DESC(CIFSMaxBufSize,"Network buffer size (not including header). Default: 16384 Range: 8192 to 130048"); @@ -73,6 +76,7 @@ module_param(cifs_max_pending, int, 0); MODULE_PARM_DESC(cifs_max_pending,"Simultaneous requests to server. Default: 50 Range: 2 to 256"); static DECLARE_COMPLETION(cifs_oplock_exited); +static DECLARE_COMPLETION(cifs_dnotify_exited); extern mempool_t *cifs_sm_req_poolp; extern mempool_t *cifs_req_poolp; @@ -202,6 +206,10 @@ cifs_statfs(struct super_block *sb, struct kstatfs *buf) #endif /* CIFS_EXPERIMENTAL */ rc = CIFSSMBQFSInfo(xid, pTcon, buf); + /* Old Windows servers do not support level 103, retry with level + one if old server failed the previous call */ + if(rc) + rc = SMBOldQFSInfo(xid, pTcon, buf); /* int f_type; __fsid_t f_fsid; @@ -253,7 +261,7 @@ cifs_alloc_inode(struct super_block *sb) cifs_inode->clientCanCacheAll = FALSE; cifs_inode->vfs_inode.i_blksize = CIFS_MAX_MSGSIZE; cifs_inode->vfs_inode.i_blkbits = 14; /* 2**14 = CIFS_MAX_MSGSIZE */ - + cifs_inode->vfs_inode.i_flags = S_NOATIME | S_NOCMTIME; INIT_LIST_HEAD(&cifs_inode->openFileList); return &cifs_inode->vfs_inode; } @@ -398,6 +406,42 @@ static struct quotactl_ops cifs_quotactl_ops = { }; #endif +#ifdef CONFIG_CIFS_EXPERIMENTAL +static void cifs_umount_begin(struct super_block * sblock) +{ + struct cifs_sb_info *cifs_sb; + struct cifsTconInfo * tcon; + + cifs_sb = CIFS_SB(sblock); + if(cifs_sb == NULL) + return; + + tcon = cifs_sb->tcon; + if(tcon == NULL) + return; + down(&tcon->tconSem); + if (atomic_read(&tcon->useCount) == 1) + tcon->tidStatus = CifsExiting; + up(&tcon->tconSem); + + /* cancel_brl_requests(tcon); */ + /* cancel_notify_requests(tcon); */ + if(tcon->ses && tcon->ses->server) + { + cFYI(1,("wake up tasks now - umount begin not complete")); + wake_up_all(&tcon->ses->server->request_q); + wake_up_all(&tcon->ses->server->response_q); + msleep(1); /* yield */ + /* we have to kick the requests once more */ + wake_up_all(&tcon->ses->server->response_q); + msleep(1); + } +/* BB FIXME - finish add checks for tidStatus BB */ + + return; +} +#endif + static int cifs_remount(struct super_block *sb, int *flags, char *data) { *flags |= MS_NODIRATIME; @@ -415,7 +459,9 @@ struct super_operations cifs_super_ops = { unless later we add lazy close of inodes or unless the kernel forgets to call us with the same number of releases (closes) as opens */ .show_options = cifs_show_options, -/* .umount_begin = cifs_umount_begin, *//* consider adding in the future */ +#ifdef CONFIG_CIFS_EXPERIMENTAL + .umount_begin = cifs_umount_begin, +#endif .remount_fs = cifs_remount, }; @@ -443,57 +489,30 @@ cifs_get_sb(struct file_system_type *fs_type, return sb; } -static ssize_t -cifs_read_wrapper(struct file * file, char __user *read_data, size_t read_size, - loff_t * poffset) +static ssize_t cifs_file_writev(struct file *file, const struct iovec *iov, + unsigned long nr_segs, loff_t *ppos) { - if(file->f_dentry == NULL) - return -EIO; - else if(file->f_dentry->d_inode == NULL) - return -EIO; - - cFYI(1,("In read_wrapper size %zd at %lld",read_size,*poffset)); + struct inode *inode = file->f_dentry->d_inode; + ssize_t written; - if(CIFS_I(file->f_dentry->d_inode)->clientCanCacheRead) { - return generic_file_read(file,read_data,read_size,poffset); - } else { - /* BB do we need to lock inode from here until after invalidate? */ -/* if(file->f_dentry->d_inode->i_mapping) { - filemap_fdatawrite(file->f_dentry->d_inode->i_mapping); - filemap_fdatawait(file->f_dentry->d_inode->i_mapping); - }*/ -/* cifs_revalidate(file->f_dentry);*/ /* BB fixme */ - - /* BB we should make timer configurable - perhaps - by simply calling cifs_revalidate here */ - /* invalidate_remote_inode(file->f_dentry->d_inode);*/ - return generic_file_read(file,read_data,read_size,poffset); - } + written = generic_file_writev(file, iov, nr_segs, ppos); + if (!CIFS_I(inode)->clientCanCacheAll) + filemap_fdatawrite(inode->i_mapping); + return written; } -static ssize_t -cifs_write_wrapper(struct file * file, const char __user *write_data, - size_t write_size, loff_t * poffset) +static ssize_t cifs_file_aio_write(struct kiocb *iocb, const char __user *buf, + size_t count, loff_t pos) { + struct inode *inode = iocb->ki_filp->f_dentry->d_inode; ssize_t written; - if(file->f_dentry == NULL) - return -EIO; - else if(file->f_dentry->d_inode == NULL) - return -EIO; - - cFYI(1,("In write_wrapper size %zd at %lld",write_size,*poffset)); - - written = generic_file_write(file,write_data,write_size,poffset); - if(!CIFS_I(file->f_dentry->d_inode)->clientCanCacheAll) { - if(file->f_dentry->d_inode->i_mapping) { - filemap_fdatawrite(file->f_dentry->d_inode->i_mapping); - } - } + written = generic_file_aio_write(iocb, buf, count, pos); + if (!CIFS_I(inode)->clientCanCacheAll) + filemap_fdatawrite(inode->i_mapping); return written; } - static struct file_system_type cifs_fs_type = { .owner = THIS_MODULE, .name = "cifs", @@ -554,8 +573,12 @@ struct inode_operations cifs_symlink_inode_ops = { }; struct file_operations cifs_file_ops = { - .read = cifs_read_wrapper, - .write = cifs_write_wrapper, + .read = do_sync_read, + .write = do_sync_write, + .readv = generic_file_readv, + .writev = cifs_file_writev, + .aio_read = generic_file_aio_read, + .aio_write = cifs_file_aio_write, .open = cifs_open, .release = cifs_close, .lock = cifs_lock, @@ -568,10 +591,6 @@ struct file_operations cifs_file_ops = { #endif /* CONFIG_CIFS_POSIX */ #ifdef CONFIG_CIFS_EXPERIMENTAL - .readv = generic_file_readv, - .writev = generic_file_writev, - .aio_read = generic_file_aio_read, - .aio_write = generic_file_aio_write, .dir_notify = cifs_dir_notify, #endif /* CONFIG_CIFS_EXPERIMENTAL */ }; @@ -595,6 +614,46 @@ struct file_operations cifs_file_direct_ops = { .dir_notify = cifs_dir_notify, #endif /* CONFIG_CIFS_EXPERIMENTAL */ }; +struct file_operations cifs_file_nobrl_ops = { + .read = do_sync_read, + .write = do_sync_write, + .readv = generic_file_readv, + .writev = cifs_file_writev, + .aio_read = generic_file_aio_read, + .aio_write = cifs_file_aio_write, + .open = cifs_open, + .release = cifs_close, + .fsync = cifs_fsync, + .flush = cifs_flush, + .mmap = cifs_file_mmap, + .sendfile = generic_file_sendfile, +#ifdef CONFIG_CIFS_POSIX + .ioctl = cifs_ioctl, +#endif /* CONFIG_CIFS_POSIX */ + +#ifdef CONFIG_CIFS_EXPERIMENTAL + .dir_notify = cifs_dir_notify, +#endif /* CONFIG_CIFS_EXPERIMENTAL */ +}; + +struct file_operations cifs_file_direct_nobrl_ops = { + /* no mmap, no aio, no readv - + BB reevaluate whether they can be done with directio, no cache */ + .read = cifs_user_read, + .write = cifs_user_write, + .open = cifs_open, + .release = cifs_close, + .fsync = cifs_fsync, + .flush = cifs_flush, + .sendfile = generic_file_sendfile, /* BB removeme BB */ +#ifdef CONFIG_CIFS_POSIX + .ioctl = cifs_ioctl, +#endif /* CONFIG_CIFS_POSIX */ + +#ifdef CONFIG_CIFS_EXPERIMENTAL + .dir_notify = cifs_dir_notify, +#endif /* CONFIG_CIFS_EXPERIMENTAL */ +}; struct file_operations cifs_dir_ops = { .readdir = cifs_readdir, @@ -781,9 +840,9 @@ static int cifs_oplock_thread(void * dummyarg) oplockThread = current; do { - set_current_state(TASK_INTERRUPTIBLE); + if (try_to_freeze()) + continue; - schedule_timeout(1*HZ); spin_lock(&GlobalMid_Lock); if(list_empty(&GlobalOplock_Q)) { spin_unlock(&GlobalMid_Lock); @@ -832,10 +891,42 @@ static int cifs_oplock_thread(void * dummyarg) } } else spin_unlock(&GlobalMid_Lock); + set_current_state(TASK_INTERRUPTIBLE); + schedule_timeout(1); /* yield in case q were corrupt */ } } while(!signal_pending(current)); - complete_and_exit (&cifs_oplock_exited, 0); oplockThread = NULL; + complete_and_exit (&cifs_oplock_exited, 0); +} + +static int cifs_dnotify_thread(void * dummyarg) +{ + struct list_head *tmp; + struct cifsSesInfo *ses; + + daemonize("cifsdnotifyd"); + allow_signal(SIGTERM); + + dnotifyThread = current; + do { + if(try_to_freeze()) + continue; + set_current_state(TASK_INTERRUPTIBLE); + schedule_timeout(15*HZ); + read_lock(&GlobalSMBSeslock); + /* check if any stuck requests that need + to be woken up and wakeq so the + thread can wake up and error out */ + list_for_each(tmp, &GlobalSMBSessionList) { + ses = list_entry(tmp, struct cifsSesInfo, + cifsSessionList); + if(ses && ses->server && + atomic_read(&ses->server->inFlight)) + wake_up_all(&ses->server->response_q); + } + read_unlock(&GlobalSMBSeslock); + } while(!signal_pending(current)); + complete_and_exit (&cifs_dnotify_exited, 0); } static int __init @@ -849,6 +940,10 @@ init_cifs(void) INIT_LIST_HEAD(&GlobalSMBSessionList); INIT_LIST_HEAD(&GlobalTreeConnectionList); INIT_LIST_HEAD(&GlobalOplock_Q); +#ifdef CONFIG_CIFS_EXPERIMENTAL + INIT_LIST_HEAD(&GlobalDnotifyReqList); + INIT_LIST_HEAD(&GlobalDnotifyRsp_Q); +#endif /* * Initialize Global counters */ @@ -884,10 +979,16 @@ init_cifs(void) if (!rc) { rc = (int)kernel_thread(cifs_oplock_thread, NULL, CLONE_FS | CLONE_FILES | CLONE_VM); - if(rc > 0) - return 0; - else + if(rc > 0) { + rc = (int)kernel_thread(cifs_dnotify_thread, NULL, + CLONE_FS | CLONE_FILES | CLONE_VM); + if(rc > 0) + return 0; + else + cERROR(1,("error %d create dnotify thread", rc)); + } else { cERROR(1,("error %d create oplock thread",rc)); + } } cifs_destroy_request_bufs(); } @@ -916,6 +1017,10 @@ exit_cifs(void) send_sig(SIGTERM, oplockThread, 1); wait_for_completion(&cifs_oplock_exited); } + if(dnotifyThread) { + send_sig(SIGTERM, dnotifyThread, 1); + wait_for_completion(&cifs_dnotify_exited); + } } MODULE_AUTHOR("Steve French <sfrench@us.ibm.com>"); diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h index 1fd21f66f243..9ec40e0e54fc 100644 --- a/fs/cifs/cifsfs.h +++ b/fs/cifs/cifsfs.h @@ -63,6 +63,8 @@ extern struct inode_operations cifs_symlink_inode_ops; /* Functions related to files and directories */ extern struct file_operations cifs_file_ops; extern struct file_operations cifs_file_direct_ops; /* if directio mount */ +extern struct file_operations cifs_file_nobrl_ops; +extern struct file_operations cifs_file_direct_nobrl_ops; /* if directio mount */ extern int cifs_open(struct inode *inode, struct file *file); extern int cifs_close(struct inode *inode, struct file *file); extern int cifs_closedir(struct inode *inode, struct file *file); @@ -81,6 +83,7 @@ extern int cifs_dir_notify(struct file *, unsigned long arg); /* Functions related to dir entries */ extern struct dentry_operations cifs_dentry_ops; +extern struct dentry_operations cifs_ci_dentry_ops; /* Functions related to symlinks */ extern void *cifs_follow_link(struct dentry *direntry, struct nameidata *nd); @@ -96,5 +99,5 @@ extern ssize_t cifs_getxattr(struct dentry *, const char *, void *, size_t); extern ssize_t cifs_listxattr(struct dentry *, char *, size_t); extern int cifs_ioctl (struct inode * inode, struct file * filep, unsigned int command, unsigned long arg); -#define CIFS_VERSION "1.35" +#define CIFS_VERSION "1.39" #endif /* _CIFSFS_H */ diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index 81babab265e1..1ba08f8c5bc4 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -110,8 +110,9 @@ enum protocolEnum { */ struct TCP_Server_Info { - char server_Name[SERVER_NAME_LEN_WITH_NULL]; /* 15 chars + X'20'in 16th */ - char unicode_server_Name[SERVER_NAME_LEN_WITH_NULL * 2]; /* Unicode version of server_Name */ + /* 15 character server name + 0x20 16th byte indicating type = srv */ + char server_RFC1001_name[SERVER_NAME_LEN_WITH_NULL]; + char unicode_server_Name[SERVER_NAME_LEN_WITH_NULL * 2]; struct socket *ssocket; union { struct sockaddr_in sockAddr; @@ -122,13 +123,17 @@ struct TCP_Server_Info { struct list_head pending_mid_q; void *Server_NlsInfo; /* BB - placeholder for future NLS info */ unsigned short server_codepage; /* codepage for the server */ - unsigned long ip_address; /* IP addr for the server if known */ + unsigned long ip_address; /* IP addr for the server if known */ enum protocolEnum protocolType; char versionMajor; char versionMinor; unsigned svlocal:1; /* local server or remote */ atomic_t socketUseCount; /* number of open cifs sessions on socket */ atomic_t inFlight; /* number of requests on the wire to server */ +#ifdef CONFIG_CIFS_STATS2 + atomic_t inSend; /* requests trying to send */ + atomic_t num_waiters; /* blocked waiting to get in sendrecv */ +#endif enum statusEnum tcpStatus; /* what we think the status is */ struct semaphore tcpSem; struct task_struct *tsk; @@ -147,8 +152,10 @@ struct TCP_Server_Info { /* (returned on Negotiate */ int capabilities; /* allow selective disabling of caps by smb sess */ __u16 timeZone; + __u16 CurrentMid; /* multiplex id - rotating counter */ char cryptKey[CIFS_CRYPTO_KEY_SIZE]; - char workstation_RFC1001_name[16]; /* 16th byte is always zero */ + /* 16th byte of RFC1001 workstation name is always null */ + char workstation_RFC1001_name[SERVER_NAME_LEN_WITH_NULL]; __u32 sequence_number; /* needed for CIFS PDU signature */ char mac_signing_key[CIFS_SESSION_KEY_SIZE + 16]; }; @@ -214,19 +221,41 @@ struct cifsTconInfo { atomic_t num_reads; atomic_t num_oplock_brks; atomic_t num_opens; + atomic_t num_closes; atomic_t num_deletes; atomic_t num_mkdirs; atomic_t num_rmdirs; atomic_t num_renames; atomic_t num_t2renames; + atomic_t num_ffirst; + atomic_t num_fnext; + atomic_t num_fclose; + atomic_t num_hardlinks; + atomic_t num_symlinks; + atomic_t num_locks; +#ifdef CONFIG_CIFS_STATS2 + unsigned long long time_writes; + unsigned long long time_reads; + unsigned long long time_opens; + unsigned long long time_deletes; + unsigned long long time_closes; + unsigned long long time_mkdirs; + unsigned long long time_rmdirs; + unsigned long long time_renames; + unsigned long long time_t2renames; + unsigned long long time_ffirst; + unsigned long long time_fnext; + unsigned long long time_fclose; +#endif /* CONFIG_CIFS_STATS2 */ __u64 bytes_read; __u64 bytes_written; spinlock_t stat_lock; -#endif +#endif /* CONFIG_CIFS_STATS */ FILE_SYSTEM_DEVICE_INFO fsDevInfo; FILE_SYSTEM_ATTRIBUTE_INFO fsAttrInfo; /* ok if file system name truncated */ FILE_SYSTEM_UNIX_INFO fsUnixInfo; unsigned retry:1; + unsigned nocase:1; /* BB add field for back pointer to sb struct? */ }; @@ -270,6 +299,7 @@ struct cifsFileInfo { struct inode * pInode; /* needed for oplock break */ unsigned closePend:1; /* file is marked to close */ unsigned invalidHandle:1; /* file closed via session abend */ + atomic_t wrtPending; /* handle in use - defer close */ struct semaphore fh_sem; /* prevents reopen race after dead ses*/ char * search_resume_name; /* BB removeme BB */ unsigned int resume_name_length; /* BB removeme - field renamed and moved BB */ @@ -306,6 +336,41 @@ CIFS_SB(struct super_block *sb) return sb->s_fs_info; } +static inline char CIFS_DIR_SEP(const struct cifs_sb_info *cifs_sb) +{ + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_POSIX_PATHS) + return '/'; + else + return '\\'; +} + +#ifdef CONFIG_CIFS_STATS +#define cifs_stats_inc atomic_inc + +static inline void cifs_stats_bytes_written(struct cifsTconInfo *tcon, + unsigned int bytes) +{ + if (bytes) { + spin_lock(&tcon->stat_lock); + tcon->bytes_written += bytes; + spin_unlock(&tcon->stat_lock); + } +} + +static inline void cifs_stats_bytes_read(struct cifsTconInfo *tcon, + unsigned int bytes) +{ + spin_lock(&tcon->stat_lock); + tcon->bytes_read += bytes; + spin_unlock(&tcon->stat_lock); +} +#else + +#define cifs_stats_inc(field) do {} while(0) +#define cifs_stats_bytes_written(tcon, bytes) do {} while(0) +#define cifs_stats_bytes_read(tcon, bytes) do {} while(0) + +#endif /* one of these for every pending CIFS request to the server */ struct mid_q_entry { @@ -313,7 +378,11 @@ struct mid_q_entry { __u16 mid; /* multiplex id */ __u16 pid; /* process id */ __u32 sequence_number; /* for CIFS signing */ - struct timeval when_sent; /* time when smb sent */ + unsigned long when_alloc; /* when mid was created */ +#ifdef CONFIG_CIFS_STATS2 + unsigned long when_sent; /* time when smb send finished */ + unsigned long when_received; /* when demux complete (taken off wire) */ +#endif struct cifsSesInfo *ses; /* smb was sent to this server */ struct task_struct *tsk; /* task waiting for response */ struct smb_hdr *resp_buf; /* response buffer */ @@ -331,6 +400,20 @@ struct oplock_q_entry { __u16 netfid; }; +/* for pending dnotify requests */ +struct dir_notify_req { + struct list_head lhead; + __le16 Pid; + __le16 PidHigh; + __u16 Mid; + __u16 Tid; + __u16 Uid; + __u16 netfid; + __u32 filter; /* CompletionFilter (for multishot) */ + int multishot; + struct file * pfile; +}; + #define MID_FREE 0 #define MID_REQUEST_ALLOCATED 1 #define MID_REQUEST_SUBMITTED 2 @@ -399,6 +482,9 @@ GLOBAL_EXTERN rwlock_t GlobalSMBSeslock; /* protects list inserts on 3 above */ GLOBAL_EXTERN struct list_head GlobalOplock_Q; +GLOBAL_EXTERN struct list_head GlobalDnotifyReqList; /* Outstanding dir notify requests */ +GLOBAL_EXTERN struct list_head GlobalDnotifyRsp_Q; /* Dir notify response queue */ + /* * Global transaction id (XID) information */ diff --git a/fs/cifs/cifspdu.h b/fs/cifs/cifspdu.h index aede6a813167..33e1859fd2f6 100644 --- a/fs/cifs/cifspdu.h +++ b/fs/cifs/cifspdu.h @@ -36,9 +36,11 @@ #define SMB_COM_CLOSE 0x04 /* triv req/rsp, timestamp ignored */ #define SMB_COM_DELETE 0x06 /* trivial response */ #define SMB_COM_RENAME 0x07 /* trivial response */ +#define SMB_COM_QUERY_INFORMATION 0x08 /* aka getattr */ #define SMB_COM_SETATTR 0x09 /* trivial response */ #define SMB_COM_LOCKING_ANDX 0x24 /* trivial response */ #define SMB_COM_COPY 0x29 /* trivial rsp, fail filename ignrd*/ +#define SMB_COM_OPEN_ANDX 0x2D /* Legacy open for old servers */ #define SMB_COM_READ_ANDX 0x2E #define SMB_COM_WRITE_ANDX 0x2F #define SMB_COM_TRANSACTION2 0x32 @@ -52,6 +54,7 @@ #define SMB_COM_NT_TRANSACT 0xA0 #define SMB_COM_NT_TRANSACT_SECONDARY 0xA1 #define SMB_COM_NT_CREATE_ANDX 0xA2 +#define SMB_COM_NT_CANCEL 0xA4 /* no response */ #define SMB_COM_NT_RENAME 0xA5 /* trivial response */ /* Transact2 subcommand codes */ @@ -59,6 +62,7 @@ #define TRANS2_FIND_FIRST 0x01 #define TRANS2_FIND_NEXT 0x02 #define TRANS2_QUERY_FS_INFORMATION 0x03 +#define TRANS2_SET_FS_INFORMATION 0x04 #define TRANS2_QUERY_PATH_INFORMATION 0x05 #define TRANS2_SET_PATH_INFORMATION 0x06 #define TRANS2_QUERY_FILE_INFORMATION 0x07 @@ -76,7 +80,7 @@ #define NT_TRANSACT_GET_USER_QUOTA 0x07 #define NT_TRANSACT_SET_USER_QUOTA 0x08 -#define MAX_CIFS_HDR_SIZE 256 /* chained NTCreateXReadX will probably be biggest */ +#define MAX_CIFS_HDR_SIZE 256 /* is future chained NTCreateXReadX bigger? */ /* internal cifs vfs structures */ /***************************************************************** @@ -129,10 +133,11 @@ /* * SMB flag definitions */ -#define SMBFLG_EXTD_LOCK 0x01 /* server supports lock-read write-unlock primitives */ +#define SMBFLG_EXTD_LOCK 0x01 /* server supports lock-read write-unlock smb */ #define SMBFLG_RCV_POSTED 0x02 /* obsolete */ #define SMBFLG_RSVD 0x04 -#define SMBFLG_CASELESS 0x08 /* all pathnames treated as caseless (off implies case sensitive file handling requested) */ +#define SMBFLG_CASELESS 0x08 /* all pathnames treated as caseless (off + implies case sensitive file handling request) */ #define SMBFLG_CANONICAL_PATH_FORMAT 0x10 /* obsolete */ #define SMBFLG_OLD_OPLOCK 0x20 /* obsolete */ #define SMBFLG_OLD_OPLOCK_NOTIFY 0x40 /* obsolete */ @@ -141,7 +146,8 @@ /* * SMB flag2 definitions */ -#define SMBFLG2_KNOWS_LONG_NAMES cpu_to_le16(1) /* can send long (non-8.3) path names in response */ +#define SMBFLG2_KNOWS_LONG_NAMES cpu_to_le16(1) /* can send long (non-8.3) + path names in response */ #define SMBFLG2_KNOWS_EAS cpu_to_le16(2) #define SMBFLG2_SECURITY_SIGNATURE cpu_to_le16(4) #define SMBFLG2_IS_LONG_NAME cpu_to_le16(0x40) @@ -160,32 +166,32 @@ * file and can have any suitable combination of the following values: */ -#define FILE_READ_DATA 0x00000001 /* Data can be read from the file */ -#define FILE_WRITE_DATA 0x00000002 /* Data can be written to the file */ -#define FILE_APPEND_DATA 0x00000004 /* Data can be appended to the file */ -#define FILE_READ_EA 0x00000008 /* Extended attributes associated */ - /* with the file can be read */ -#define FILE_WRITE_EA 0x00000010 /* Extended attributes associated */ - /* with the file can be written */ -#define FILE_EXECUTE 0x00000020 /*Data can be read into memory from */ - /* the file using system paging I/O */ +#define FILE_READ_DATA 0x00000001 /* Data can be read from the file */ +#define FILE_WRITE_DATA 0x00000002 /* Data can be written to the file */ +#define FILE_APPEND_DATA 0x00000004 /* Data can be appended to the file */ +#define FILE_READ_EA 0x00000008 /* Extended attributes associated */ + /* with the file can be read */ +#define FILE_WRITE_EA 0x00000010 /* Extended attributes associated */ + /* with the file can be written */ +#define FILE_EXECUTE 0x00000020 /*Data can be read into memory from */ + /* the file using system paging I/O */ #define FILE_DELETE_CHILD 0x00000040 -#define FILE_READ_ATTRIBUTES 0x00000080 /* Attributes associated with the */ - /* file can be read */ -#define FILE_WRITE_ATTRIBUTES 0x00000100 /* Attributes associated with the */ - /* file can be written */ -#define DELETE 0x00010000 /* The file can be deleted */ -#define READ_CONTROL 0x00020000 /* The access control list and */ - /* ownership associated with the */ - /* file can be read */ -#define WRITE_DAC 0x00040000 /* The access control list and */ - /* ownership associated with the */ - /* file can be written. */ -#define WRITE_OWNER 0x00080000 /* Ownership information associated */ - /* with the file can be written */ -#define SYNCHRONIZE 0x00100000 /* The file handle can waited on to */ - /* synchronize with the completion */ - /* of an input/output request */ +#define FILE_READ_ATTRIBUTES 0x00000080 /* Attributes associated with the */ + /* file can be read */ +#define FILE_WRITE_ATTRIBUTES 0x00000100 /* Attributes associated with the */ + /* file can be written */ +#define DELETE 0x00010000 /* The file can be deleted */ +#define READ_CONTROL 0x00020000 /* The access control list and */ + /* ownership associated with the */ + /* file can be read */ +#define WRITE_DAC 0x00040000 /* The access control list and */ + /* ownership associated with the */ + /* file can be written. */ +#define WRITE_OWNER 0x00080000 /* Ownership information associated */ + /* with the file can be written */ +#define SYNCHRONIZE 0x00100000 /* The file handle can waited on to */ + /* synchronize with the completion */ + /* of an input/output request */ #define GENERIC_ALL 0x10000000 #define GENERIC_EXECUTE 0x20000000 #define GENERIC_WRITE 0x40000000 @@ -193,7 +199,7 @@ /* In summary - Relevant file */ /* access flags from CIFS are */ /* file_read_data, file_write_data */ - /* file_execute, file_read_attributes */ + /* file_execute, file_read_attributes*/ /* write_dac, and delete. */ /* @@ -238,7 +244,8 @@ #define ATTR_SPARSE 0x0200 #define ATTR_REPARSE 0x0400 #define ATTR_COMPRESSED 0x0800 -#define ATTR_OFFLINE 0x1000 /* ie file not immediately available - offline storage */ +#define ATTR_OFFLINE 0x1000 /* ie file not immediately available - + on offline storage */ #define ATTR_NOT_CONTENT_INDEXED 0x2000 #define ATTR_ENCRYPTED 0x4000 #define ATTR_POSIX_SEMANTICS 0x01000000 @@ -267,10 +274,18 @@ /* CreateOptions */ #define CREATE_NOT_FILE 0x00000001 /* if set must not be file */ #define CREATE_WRITE_THROUGH 0x00000002 -#define CREATE_NOT_DIR 0x00000040 /* if set must not be directory */ +#define CREATE_SEQUENTIAL 0x00000004 +#define CREATE_SYNC_ALERT 0x00000010 +#define CREATE_ASYNC_ALERT 0x00000020 +#define CREATE_NOT_DIR 0x00000040 /* if set must not be directory */ +#define CREATE_NO_EA_KNOWLEDGE 0x00000200 +#define CREATE_EIGHT_DOT_THREE 0x00000400 #define CREATE_RANDOM_ACCESS 0x00000800 #define CREATE_DELETE_ON_CLOSE 0x00001000 +#define CREATE_OPEN_BY_ID 0x00002000 #define OPEN_REPARSE_POINT 0x00200000 +#define CREATE_OPTIONS_MASK 0x007FFFFF +#define CREATE_OPTION_SPECIAL 0x20000000 /* system. NB not sent over wire */ /* ImpersonationLevel flags */ #define SECURITY_ANONYMOUS 0 @@ -297,10 +312,10 @@ #define GETU16(var) (*((__u16 *)var)) /* BB check for endian issues */ #define GETU32(var) (*((__u32 *)var)) /* BB check for endian issues */ -#pragma pack(1) - struct smb_hdr { - __u32 smb_buf_length; /* big endian on wire *//* BB length is only two or three bytes - with one or two byte type preceding it but that is always zero - we could mask the type byte off just in case BB */ + __u32 smb_buf_length; /* big endian on wire *//* BB length is only two + or three bytes - with one or two byte type preceding it that are + zero - we could mask the type byte off just in case BB */ __u8 Protocol[4]; __u8 Command; union { @@ -308,9 +323,9 @@ struct smb_hdr { __u8 ErrorClass; __u8 Reserved; __le16 Error; - } DosError; + } __attribute__((packed)) DosError; __le32 CifsError; - } Status; + } __attribute__((packed)) Status; __u8 Flags; __le16 Flags2; /* note: le */ __le16 PidHigh; @@ -318,16 +333,16 @@ struct smb_hdr { struct { __le32 SequenceNumber; /* le */ __u32 Reserved; /* zero */ - } Sequence; + } __attribute__((packed)) Sequence; __u8 SecuritySignature[8]; /* le */ - } Signature; + } __attribute__((packed)) Signature; __u8 pad[2]; __u16 Tid; __le16 Pid; __u16 Uid; __u16 Mid; __u8 WordCount; -}; +} __attribute__((packed)); /* given a pointer to an smb_hdr retrieve the value of byte count */ #define BCC(smb_var) ( *(__u16 *)((char *)smb_var + sizeof(struct smb_hdr) + (2* smb_var->WordCount) ) ) #define BCC_LE(smb_var) ( *(__le16 *)((char *)smb_var + sizeof(struct smb_hdr) + (2* smb_var->WordCount) ) ) @@ -379,7 +394,7 @@ typedef struct negotiate_req { struct smb_hdr hdr; /* wct = 0 */ __le16 ByteCount; unsigned char DialectsArray[1]; -} NEGOTIATE_REQ; +} __attribute__((packed)) NEGOTIATE_REQ; typedef struct negotiate_rsp { struct smb_hdr hdr; /* wct = 17 */ @@ -397,16 +412,16 @@ typedef struct negotiate_rsp { __u8 EncryptionKeyLength; __u16 ByteCount; union { - unsigned char EncryptionKey[1]; /* if cap extended security is off */ + unsigned char EncryptionKey[1]; /* cap extended security off */ /* followed by Domain name - if extended security is off */ /* followed by 16 bytes of server GUID */ - /* followed by security blob if cap_extended_security negotiated */ + /* then security blob if cap_extended_security negotiated */ struct { unsigned char GUID[16]; unsigned char SecurityBlob[1]; - } extended_response; - } u; -} NEGOTIATE_RSP; + } __attribute__((packed)) extended_response; + } __attribute__((packed)) u; +} __attribute__((packed)) NEGOTIATE_RSP; /* SecurityMode bits */ #define SECMODE_USER 0x01 /* off indicates share level security */ @@ -452,7 +467,8 @@ typedef union smb_com_session_setup_andx { unsigned char SecurityBlob[1]; /* followed by */ /* STRING NativeOS */ /* STRING NativeLanMan */ - } req; /* NTLM request format (with extended security */ + } __attribute__((packed)) req; /* NTLM request format (with + extended security */ struct { /* request format */ struct smb_hdr hdr; /* wct = 13 */ @@ -463,18 +479,19 @@ typedef union smb_com_session_setup_andx { __le16 MaxMpxCount; __le16 VcNumber; __u32 SessionKey; - __le16 CaseInsensitivePasswordLength; /* ASCII password length */ - __le16 CaseSensitivePasswordLength; /* Unicode password length */ + __le16 CaseInsensitivePasswordLength; /* ASCII password len */ + __le16 CaseSensitivePasswordLength; /* Unicode password length*/ __u32 Reserved; /* see below */ __le32 Capabilities; __le16 ByteCount; - unsigned char CaseInsensitivePassword[1]; /* followed by: */ + unsigned char CaseInsensitivePassword[1]; /* followed by: */ /* unsigned char * CaseSensitivePassword; */ /* STRING AccountName */ /* STRING PrimaryDomain */ /* STRING NativeOS */ /* STRING NativeLanMan */ - } req_no_secext; /* NTLM request format (without extended security */ + } __attribute__((packed)) req_no_secext; /* NTLM request format (without + extended security */ struct { /* default (NTLM) response format */ struct smb_hdr hdr; /* wct = 4 */ @@ -488,7 +505,7 @@ typedef union smb_com_session_setup_andx { /* unsigned char * NativeOS; */ /* unsigned char * NativeLanMan; */ /* unsigned char * PrimaryDomain; */ - } resp; /* NTLM response format (with or without extended security */ + } __attribute__((packed)) resp; /* NTLM response format (with or without extended security */ struct { /* request format */ struct smb_hdr hdr; /* wct = 10 */ @@ -507,7 +524,7 @@ typedef union smb_com_session_setup_andx { /* STRING PrimaryDomain */ /* STRING NativeOS */ /* STRING NativeLanMan */ - } old_req; /* pre-NTLM (LANMAN2.1) request format */ + } __attribute__((packed)) old_req; /* pre-NTLM (LANMAN2.1) request format */ struct { /* default (NTLM) response format */ struct smb_hdr hdr; /* wct = 3 */ @@ -519,8 +536,8 @@ typedef union smb_com_session_setup_andx { unsigned char NativeOS[1]; /* followed by */ /* unsigned char * NativeLanMan; */ /* unsigned char * PrimaryDomain; */ - } old_resp; /* pre-NTLM (LANMAN2.1) response format */ -} SESSION_SETUP_ANDX; + } __attribute__((packed)) old_resp; /* pre-NTLM (LANMAN2.1) response format */ +} __attribute__((packed)) SESSION_SETUP_ANDX; #define CIFS_NETWORK_OPSYS "CIFS VFS Client for Linux" @@ -530,7 +547,8 @@ typedef union smb_com_session_setup_andx { #define CAP_NT_SMBS 0x00000010 #define CAP_STATUS32 0x00000040 #define CAP_LEVEL_II_OPLOCKS 0x00000080 -#define CAP_NT_FIND 0x00000200 /* reserved should be zero (presumably because NT_SMBs implies the same thing) */ +#define CAP_NT_FIND 0x00000200 /* reserved should be zero + (because NT_SMBs implies the same thing?) */ #define CAP_BULK_TRANSFER 0x20000000 #define CAP_EXTENDED_SECURITY 0x80000000 @@ -548,7 +566,7 @@ typedef struct smb_com_tconx_req { unsigned char Password[1]; /* followed by */ /* STRING Path *//* \\server\share name */ /* STRING Service */ -} TCONX_REQ; +} __attribute__((packed)) TCONX_REQ; typedef struct smb_com_tconx_rsp { struct smb_hdr hdr; /* wct = 3 *//* note that Win2000 has sent wct=7 in some cases on responses. Four unspecified words followed OptionalSupport */ @@ -559,13 +577,14 @@ typedef struct smb_com_tconx_rsp { __u16 ByteCount; unsigned char Service[1]; /* always ASCII, not Unicode */ /* STRING NativeFileSystem */ -} TCONX_RSP; +} __attribute__((packed)) TCONX_RSP; /* tree connect Flags */ #define DISCONNECT_TID 0x0001 #define TCON_EXTENDED_SECINFO 0x0008 /* OptionalSupport bits */ -#define SMB_SUPPORT_SEARCH_BITS 0x0001 /* must have bits (exclusive searches suppt. */ +#define SMB_SUPPORT_SEARCH_BITS 0x0001 /* "must have" directory search bits + (exclusive searches supported) */ #define SMB_SHARE_IS_IN_DFS 0x0002 typedef struct smb_com_logoff_andx_req { @@ -574,7 +593,7 @@ typedef struct smb_com_logoff_andx_req { __u8 AndXReserved; __u16 AndXOffset; __u16 ByteCount; -} LOGOFF_ANDX_REQ; +} __attribute__((packed)) LOGOFF_ANDX_REQ; typedef struct smb_com_logoff_andx_rsp { struct smb_hdr hdr; /* wct = 2 */ @@ -582,38 +601,41 @@ typedef struct smb_com_logoff_andx_rsp { __u8 AndXReserved; __u16 AndXOffset; __u16 ByteCount; -} LOGOFF_ANDX_RSP; +} __attribute__((packed)) LOGOFF_ANDX_RSP; -typedef union smb_com_tree_disconnect { /* as an altetnative can use flag on tree_connect PDU to effect disconnect *//* probably the simplest SMB PDU */ +typedef union smb_com_tree_disconnect { /* as an altetnative can use flag on + tree_connect PDU to effect disconnect */ + /* tdis is probably simplest SMB PDU */ struct { struct smb_hdr hdr; /* wct = 0 */ __u16 ByteCount; /* bcc = 0 */ - } req; + } __attribute__((packed)) req; struct { struct smb_hdr hdr; /* wct = 0 */ __u16 ByteCount; /* bcc = 0 */ - } resp; -} TREE_DISCONNECT; + } __attribute__((packed)) resp; +} __attribute__((packed)) TREE_DISCONNECT; typedef struct smb_com_close_req { struct smb_hdr hdr; /* wct = 3 */ __u16 FileID; __u32 LastWriteTime; /* should be zero */ __u16 ByteCount; /* 0 */ -} CLOSE_REQ; +} __attribute__((packed)) CLOSE_REQ; typedef struct smb_com_close_rsp { struct smb_hdr hdr; /* wct = 0 */ __u16 ByteCount; /* bct = 0 */ -} CLOSE_RSP; +} __attribute__((packed)) CLOSE_RSP; typedef struct smb_com_findclose_req { struct smb_hdr hdr; /* wct = 1 */ __u16 FileID; __u16 ByteCount; /* 0 */ -} FINDCLOSE_REQ; +} __attribute__((packed)) FINDCLOSE_REQ; /* OpenFlags */ +#define REQ_MORE_INFO 0x00000001 /* legacy (OPEN_AND_X) only */ #define REQ_OPLOCK 0x00000002 #define REQ_BATCHOPLOCK 0x00000004 #define REQ_OPENDIRONLY 0x00000008 @@ -637,7 +659,7 @@ typedef struct smb_com_open_req { /* also handles create */ __u8 SecurityFlags; __le16 ByteCount; char fileName[1]; -} OPEN_REQ; +} __attribute__((packed)) OPEN_REQ; /* open response: oplock levels */ #define OPLOCK_NONE 0 @@ -667,7 +689,63 @@ typedef struct smb_com_open_rsp { __le16 DeviceState; __u8 DirectoryFlag; __u16 ByteCount; /* bct = 0 */ -} OPEN_RSP; +} __attribute__((packed)) OPEN_RSP; + +/* format of legacy open request */ +typedef struct smb_com_openx_req { + struct smb_hdr hdr; /* wct = 15 */ + __u8 AndXCommand; + __u8 AndXReserved; + __le16 AndXOffset; + __le16 OpenFlags; + __le16 Mode; + __le16 Sattr; /* search attributes */ + __le16 FileAttributes; /* dos attrs */ + __le32 CreateTime; /* os2 format */ + __le16 OpenFunction; + __le32 EndOfFile; + __le32 Timeout; + __le32 Reserved; + __le16 ByteCount; /* file name follows */ + char fileName[1]; +} __attribute__((packed)) OPENX_REQ; + +typedef struct smb_com_openx_rsp { + struct smb_hdr hdr; /* wct = 15 */ + __u8 AndXCommand; + __u8 AndXReserved; + __le16 AndXOffset; + __u16 Fid; + __le16 FileAttributes; + __le32 LastWriteTime; /* os2 format */ + __le32 EndOfFile; + __le16 Access; + __le16 FileType; + __le16 IPCState; + __le16 Action; + __u32 FileId; + __u16 Reserved; + __u16 ByteCount; +} __attribute__((packed)) OPENX_RSP; + +/* Legacy write request for older servers */ +typedef struct smb_com_writex_req { + struct smb_hdr hdr; /* wct = 12 */ + __u8 AndXCommand; + __u8 AndXReserved; + __le16 AndXOffset; + __u16 Fid; + __le32 OffsetLow; + __u32 Reserved; /* Timeout */ + __le16 WriteMode; /* 1 = write through */ + __le16 Remaining; + __le16 Reserved2; + __le16 DataLengthLow; + __le16 DataOffset; + __le16 ByteCount; + __u8 Pad; /* BB check for whether padded to DWORD boundary and optimum performance here */ + char Data[0]; +} __attribute__((packed)) WRITEX_REQ; typedef struct smb_com_write_req { struct smb_hdr hdr; /* wct = 14 */ @@ -686,7 +764,7 @@ typedef struct smb_com_write_req { __le16 ByteCount; __u8 Pad; /* BB check for whether padded to DWORD boundary and optimum performance here */ char Data[0]; -} WRITE_REQ; +} __attribute__((packed)) WRITE_REQ; typedef struct smb_com_write_rsp { struct smb_hdr hdr; /* wct = 6 */ @@ -698,7 +776,22 @@ typedef struct smb_com_write_rsp { __le16 CountHigh; __u16 Reserved; __u16 ByteCount; -} WRITE_RSP; +} __attribute__((packed)) WRITE_RSP; + +/* legacy read request for older servers */ +typedef struct smb_com_readx_req { + struct smb_hdr hdr; /* wct = 10 */ + __u8 AndXCommand; + __u8 AndXReserved; + __le16 AndXOffset; + __u16 Fid; + __le32 OffsetLow; + __le16 MaxCount; + __le16 MinCount; /* obsolete */ + __le32 Reserved; + __le16 Remaining; + __le16 ByteCount; +} __attribute__((packed)) READX_REQ; typedef struct smb_com_read_req { struct smb_hdr hdr; /* wct = 12 */ @@ -713,7 +806,7 @@ typedef struct smb_com_read_req { __le16 Remaining; __le32 OffsetHigh; __le16 ByteCount; -} READ_REQ; +} __attribute__((packed)) READ_REQ; typedef struct smb_com_read_rsp { struct smb_hdr hdr; /* wct = 12 */ @@ -730,7 +823,7 @@ typedef struct smb_com_read_rsp { __u16 ByteCount; __u8 Pad; /* BB check for whether padded to DWORD boundary and optimum performance here */ char Data[1]; -} READ_RSP; +} __attribute__((packed)) READ_RSP; typedef struct locking_andx_range { __le16 Pid; @@ -739,7 +832,7 @@ typedef struct locking_andx_range { __le32 OffsetLow; __le32 LengthHigh; __le32 LengthLow; -} LOCKING_ANDX_RANGE; +} __attribute__((packed)) LOCKING_ANDX_RANGE; #define LOCKING_ANDX_SHARED_LOCK 0x01 #define LOCKING_ANDX_OPLOCK_RELEASE 0x02 @@ -760,7 +853,7 @@ typedef struct smb_com_lock_req { __le16 NumberOfLocks; __le16 ByteCount; LOCKING_ANDX_RANGE Locks[1]; -} LOCK_REQ; +} __attribute__((packed)) LOCK_REQ; typedef struct cifs_posix_lock { @@ -770,7 +863,7 @@ typedef struct cifs_posix_lock { __le64 start; __le64 length; /* BB what about additional owner info to identify network client */ -} CIFS_POSIX_LOCK; +} __attribute__((packed)) CIFS_POSIX_LOCK; typedef struct smb_com_lock_rsp { struct smb_hdr hdr; /* wct = 2 */ @@ -778,7 +871,7 @@ typedef struct smb_com_lock_rsp { __u8 AndXReserved; __le16 AndXOffset; __u16 ByteCount; -} LOCK_RSP; +} __attribute__((packed)) LOCK_RSP; typedef struct smb_com_rename_req { struct smb_hdr hdr; /* wct = 1 */ @@ -788,7 +881,7 @@ typedef struct smb_com_rename_req { unsigned char OldFileName[1]; /* followed by __u8 BufferFormat2 */ /* followed by NewFileName */ -} RENAME_REQ; +} __attribute__((packed)) RENAME_REQ; /* copy request flags */ #define COPY_MUST_BE_FILE 0x0001 @@ -808,7 +901,7 @@ typedef struct smb_com_copy_req { unsigned char OldFileName[1]; /* followed by __u8 BufferFormat2 */ /* followed by NewFileName string */ -} COPY_REQ; +} __attribute__((packed)) COPY_REQ; typedef struct smb_com_copy_rsp { struct smb_hdr hdr; /* wct = 1 */ @@ -816,7 +909,7 @@ typedef struct smb_com_copy_rsp { __u16 ByteCount; /* may be zero */ __u8 BufferFormat; /* 0x04 - only present if errored file follows */ unsigned char ErrorFileName[1]; /* only present if error in copy */ -} COPY_RSP; +} __attribute__((packed)) COPY_RSP; #define CREATE_HARD_LINK 0x103 #define MOVEFILE_COPY_ALLOWED 0x0002 @@ -832,12 +925,12 @@ typedef struct smb_com_nt_rename_req { /* A5 - also used for create hardlink */ unsigned char OldFileName[1]; /* followed by __u8 BufferFormat2 */ /* followed by NewFileName */ -} NT_RENAME_REQ; +} __attribute__((packed)) NT_RENAME_REQ; typedef struct smb_com_rename_rsp { struct smb_hdr hdr; /* wct = 0 */ __u16 ByteCount; /* bct = 0 */ -} RENAME_RSP; +} __attribute__((packed)) RENAME_RSP; typedef struct smb_com_delete_file_req { struct smb_hdr hdr; /* wct = 1 */ @@ -845,36 +938,52 @@ typedef struct smb_com_delete_file_req { __le16 ByteCount; __u8 BufferFormat; /* 4 = ASCII */ unsigned char fileName[1]; -} DELETE_FILE_REQ; +} __attribute__((packed)) DELETE_FILE_REQ; typedef struct smb_com_delete_file_rsp { struct smb_hdr hdr; /* wct = 0 */ __u16 ByteCount; /* bct = 0 */ -} DELETE_FILE_RSP; +} __attribute__((packed)) DELETE_FILE_RSP; typedef struct smb_com_delete_directory_req { struct smb_hdr hdr; /* wct = 0 */ __le16 ByteCount; __u8 BufferFormat; /* 4 = ASCII */ unsigned char DirName[1]; -} DELETE_DIRECTORY_REQ; +} __attribute__((packed)) DELETE_DIRECTORY_REQ; typedef struct smb_com_delete_directory_rsp { struct smb_hdr hdr; /* wct = 0 */ __u16 ByteCount; /* bct = 0 */ -} DELETE_DIRECTORY_RSP; +} __attribute__((packed)) DELETE_DIRECTORY_RSP; typedef struct smb_com_create_directory_req { struct smb_hdr hdr; /* wct = 0 */ __le16 ByteCount; __u8 BufferFormat; /* 4 = ASCII */ unsigned char DirName[1]; -} CREATE_DIRECTORY_REQ; +} __attribute__((packed)) CREATE_DIRECTORY_REQ; typedef struct smb_com_create_directory_rsp { struct smb_hdr hdr; /* wct = 0 */ __u16 ByteCount; /* bct = 0 */ -} CREATE_DIRECTORY_RSP; +} __attribute__((packed)) CREATE_DIRECTORY_RSP; + +typedef struct smb_com_query_information_req { + struct smb_hdr hdr; /* wct = 0 */ + __le16 ByteCount; /* 1 + namelen + 1 */ + __u8 BufferFormat; /* 4 = ASCII */ + unsigned char FileName[1]; +} __attribute__((packed)) QUERY_INFORMATION_REQ; + +typedef struct smb_com_query_information_rsp { + struct smb_hdr hdr; /* wct = 10 */ + __le16 attr; + __le32 last_write_time; + __le32 size; + __u16 reserved[5]; + __le16 ByteCount; /* bcc = 0 */ +} __attribute__((packed)) QUERY_INFORMATION_RSP; typedef struct smb_com_setattr_req { struct smb_hdr hdr; /* wct = 8 */ @@ -885,12 +994,12 @@ typedef struct smb_com_setattr_req { __u16 ByteCount; __u8 BufferFormat; /* 4 = ASCII */ unsigned char fileName[1]; -} SETATTR_REQ; +} __attribute__((packed)) SETATTR_REQ; typedef struct smb_com_setattr_rsp { struct smb_hdr hdr; /* wct = 0 */ __u16 ByteCount; /* bct = 0 */ -} SETATTR_RSP; +} __attribute__((packed)) SETATTR_RSP; /* empty wct response to setattr */ @@ -920,7 +1029,7 @@ typedef struct smb_com_transaction_ioctl_req { __le16 ByteCount; __u8 Pad[3]; __u8 Data[1]; -} TRANSACT_IOCTL_REQ; +} __attribute__((packed)) TRANSACT_IOCTL_REQ; typedef struct smb_com_transaction_ioctl_rsp { struct smb_hdr hdr; /* wct = 19 */ @@ -937,7 +1046,7 @@ typedef struct smb_com_transaction_ioctl_rsp { __le16 ReturnedDataLen; __u16 ByteCount; __u8 Pad[3]; -} TRANSACT_IOCTL_RSP; +} __attribute__((packed)) TRANSACT_IOCTL_RSP; typedef struct smb_com_transaction_change_notify_req { struct smb_hdr hdr; /* wct = 23 */ @@ -961,7 +1070,7 @@ typedef struct smb_com_transaction_change_notify_req { __le16 ByteCount; /* __u8 Pad[3];*/ /* __u8 Data[1];*/ -} TRANSACT_CHANGE_NOTIFY_REQ; +} __attribute__((packed)) TRANSACT_CHANGE_NOTIFY_REQ; typedef struct smb_com_transaction_change_notify_rsp { struct smb_hdr hdr; /* wct = 18 */ @@ -977,7 +1086,7 @@ typedef struct smb_com_transaction_change_notify_rsp { __u8 SetupCount; /* 0 */ __u16 ByteCount; /* __u8 Pad[3]; */ -} TRANSACT_CHANGE_NOTIFY_RSP; +} __attribute__((packed)) TRANSACT_CHANGE_NOTIFY_RSP; /* Completion Filter flags for Notify */ #define FILE_NOTIFY_CHANGE_FILE_NAME 0x00000001 #define FILE_NOTIFY_CHANGE_DIR_NAME 0x00000002 @@ -1008,7 +1117,7 @@ struct file_notify_information { __le32 Action; __le32 FileNameLength; __u8 FileName[0]; -}; +} __attribute__((packed)); struct reparse_data { __u32 ReparseTag; @@ -1019,7 +1128,7 @@ struct reparse_data { __u16 TargetNameOffset; __u16 TargetNameLen; char LinkNamesBuf[1]; -}; +} __attribute__((packed)); struct cifs_quota_data { __u32 rsrvd1; /* 0 */ @@ -1029,7 +1138,7 @@ struct cifs_quota_data { __u64 soft_limit; __u64 hard_limit; char sid[1]; /* variable size? */ -}; +} __attribute__((packed)); /* quota sub commands */ #define QUOTA_LIST_CONTINUE 0 @@ -1055,12 +1164,12 @@ struct trans2_req { __u8 Reserved3; __le16 SubCommand; /* 1st setup word - SetupCount words follow */ __le16 ByteCount; -}; +} __attribute__((packed)); struct smb_t2_req { struct smb_hdr hdr; struct trans2_req t2_req; -}; +} __attribute__((packed)); struct trans2_resp { /* struct smb_hdr hdr precedes. Note wct = 10 + setup count */ @@ -1079,12 +1188,12 @@ struct trans2_resp { __u16 ByteCount; __u16 Reserved2;*/ /* data area follows */ -}; +} __attribute__((packed)); struct smb_t2_rsp { struct smb_hdr hdr; struct trans2_resp t2_rsp; -}; +} __attribute__((packed)); /* PathInfo/FileInfo infolevels */ #define SMB_INFO_STANDARD 1 @@ -1171,14 +1280,14 @@ typedef struct smb_com_transaction2_qpi_req { __le16 InformationLevel; __u32 Reserved4; char FileName[1]; -} TRANSACTION2_QPI_REQ; +} __attribute__((packed)) TRANSACTION2_QPI_REQ; typedef struct smb_com_transaction2_qpi_rsp { struct smb_hdr hdr; /* wct = 10 + SetupCount */ struct trans2_resp t2; __u16 ByteCount; __u16 Reserved2; /* parameter word reserved - present for infolevels > 100 */ -} TRANSACTION2_QPI_RSP; +} __attribute__((packed)) TRANSACTION2_QPI_RSP; typedef struct smb_com_transaction2_spi_req { struct smb_hdr hdr; /* wct = 15 */ @@ -1204,21 +1313,21 @@ typedef struct smb_com_transaction2_spi_req { __le16 InformationLevel; __u32 Reserved4; char FileName[1]; -} TRANSACTION2_SPI_REQ; +} __attribute__((packed)) TRANSACTION2_SPI_REQ; typedef struct smb_com_transaction2_spi_rsp { struct smb_hdr hdr; /* wct = 10 + SetupCount */ struct trans2_resp t2; __u16 ByteCount; __u16 Reserved2; /* parameter word reserved - present for infolevels > 100 */ -} TRANSACTION2_SPI_RSP; +} __attribute__((packed)) TRANSACTION2_SPI_RSP; struct set_file_rename { __le32 overwrite; /* 1 = overwrite dest */ __u32 root_fid; /* zero */ __le32 target_name_len; char target_name[0]; /* Must be unicode */ -}; +} __attribute__((packed)); struct smb_com_transaction2_sfi_req { struct smb_hdr hdr; /* wct = 15 */ @@ -1244,7 +1353,7 @@ struct smb_com_transaction2_sfi_req { __u16 Fid; __le16 InformationLevel; __u16 Reserved4; -}; +} __attribute__((packed)); struct smb_com_transaction2_sfi_rsp { struct smb_hdr hdr; /* wct = 10 + SetupCount */ @@ -1252,7 +1361,7 @@ struct smb_com_transaction2_sfi_rsp { __u16 ByteCount; __u16 Reserved2; /* parameter word reserved - present for infolevels > 100 */ -}; +} __attribute__((packed)); struct smb_t2_qfi_req { struct smb_hdr hdr; @@ -1260,7 +1369,7 @@ struct smb_t2_qfi_req { __u8 Pad; __u16 Fid; __le16 InformationLevel; -}; +} __attribute__((packed)); struct smb_t2_qfi_rsp { struct smb_hdr hdr; /* wct = 10 + SetupCount */ @@ -1268,7 +1377,7 @@ struct smb_t2_qfi_rsp { __u16 ByteCount; __u16 Reserved2; /* parameter word reserved - present for infolevels > 100 */ -}; +} __attribute__((packed)); /* * Flags on T2 FINDFIRST and FINDNEXT @@ -1310,13 +1419,13 @@ typedef struct smb_com_transaction2_ffirst_req { __le16 InformationLevel; __le32 SearchStorageType; char FileName[1]; -} TRANSACTION2_FFIRST_REQ; +} __attribute__((packed)) TRANSACTION2_FFIRST_REQ; typedef struct smb_com_transaction2_ffirst_rsp { struct smb_hdr hdr; /* wct = 10 */ struct trans2_resp t2; __u16 ByteCount; -} TRANSACTION2_FFIRST_RSP; +} __attribute__((packed)) TRANSACTION2_FFIRST_RSP; typedef struct smb_com_transaction2_ffirst_rsp_parms { __u16 SearchHandle; @@ -1324,7 +1433,7 @@ typedef struct smb_com_transaction2_ffirst_rsp_parms { __le16 EndofSearch; __le16 EAErrorOffset; __le16 LastNameOffset; -} T2_FFIRST_RSP_PARMS; +} __attribute__((packed)) T2_FFIRST_RSP_PARMS; typedef struct smb_com_transaction2_fnext_req { struct smb_hdr hdr; /* wct = 15 */ @@ -1352,20 +1461,20 @@ typedef struct smb_com_transaction2_fnext_req { __u32 ResumeKey; __le16 SearchFlags; char ResumeFileName[1]; -} TRANSACTION2_FNEXT_REQ; +} __attribute__((packed)) TRANSACTION2_FNEXT_REQ; typedef struct smb_com_transaction2_fnext_rsp { struct smb_hdr hdr; /* wct = 10 */ struct trans2_resp t2; __u16 ByteCount; -} TRANSACTION2_FNEXT_RSP; +} __attribute__((packed)) TRANSACTION2_FNEXT_RSP; typedef struct smb_com_transaction2_fnext_rsp_parms { __le16 SearchCount; __le16 EndofSearch; __le16 EAErrorOffset; __le16 LastNameOffset; -} T2_FNEXT_RSP_PARMS; +} __attribute__((packed)) T2_FNEXT_RSP_PARMS; /* QFSInfo Levels */ #define SMB_INFO_ALLOCATION 1 @@ -1402,14 +1511,51 @@ typedef struct smb_com_transaction2_qfsi_req { __le16 ByteCount; __u8 Pad; __le16 InformationLevel; -} TRANSACTION2_QFSI_REQ; +} __attribute__((packed)) TRANSACTION2_QFSI_REQ; typedef struct smb_com_transaction_qfsi_rsp { struct smb_hdr hdr; /* wct = 10 + SetupCount */ struct trans2_resp t2; __u16 ByteCount; __u8 Pad; /* may be three bytes *//* followed by data area */ -} TRANSACTION2_QFSI_RSP; +} __attribute__((packed)) TRANSACTION2_QFSI_RSP; + + +/* SETFSInfo Levels */ +#define SMB_SET_CIFS_UNIX_INFO 0x200 +typedef struct smb_com_transaction2_setfsi_req { + struct smb_hdr hdr; /* wct = 15 */ + __le16 TotalParameterCount; + __le16 TotalDataCount; + __le16 MaxParameterCount; + __le16 MaxDataCount; + __u8 MaxSetupCount; + __u8 Reserved; + __le16 Flags; + __le32 Timeout; + __u16 Reserved2; + __le16 ParameterCount; /* 4 */ + __le16 ParameterOffset; + __le16 DataCount; /* 12 */ + __le16 DataOffset; + __u8 SetupCount; /* one */ + __u8 Reserved3; + __le16 SubCommand; /* TRANS2_SET_FS_INFORMATION */ + __le16 ByteCount; + __u8 Pad; + __u16 FileNum; /* Parameters start. */ + __le16 InformationLevel;/* Parameters end. */ + __le16 ClientUnixMajor; /* Data start. */ + __le16 ClientUnixMinor; + __le64 ClientUnixCap; /* Data end */ +} __attribute__((packed)) TRANSACTION2_SETFSI_REQ; + +typedef struct smb_com_transaction2_setfsi_rsp { + struct smb_hdr hdr; /* wct = 10 */ + struct trans2_resp t2; + __u16 ByteCount; +} __attribute__((packed)) TRANSACTION2_SETFSI_RSP; + typedef struct smb_com_transaction2_get_dfs_refer_req { struct smb_hdr hdr; /* wct = 15 */ @@ -1433,7 +1579,7 @@ typedef struct smb_com_transaction2_get_dfs_refer_req { __u8 Pad[3]; /* Win2K has sent 0x0F01 (max resp length perhaps?) followed by one byte pad - doesn't seem to matter though */ __le16 MaxReferralLevel; char RequestFileName[1]; -} TRANSACTION2_GET_DFS_REFER_REQ; +} __attribute__((packed)) TRANSACTION2_GET_DFS_REFER_REQ; typedef struct dfs_referral_level_3 { __le16 VersionNumber; @@ -1445,7 +1591,7 @@ typedef struct dfs_referral_level_3 { __le16 DfsPathOffset; __le16 DfsAlternatePathOffset; __le16 NetworkAddressOffset; -} REFERRAL3; +} __attribute__((packed)) REFERRAL3; typedef struct smb_com_transaction_get_dfs_refer_rsp { struct smb_hdr hdr; /* wct = 10 */ @@ -1458,7 +1604,7 @@ typedef struct smb_com_transaction_get_dfs_refer_rsp { __u16 Pad2; REFERRAL3 referrals[1]; /* array of level 3 dfs_referral structures */ /* followed by the strings pointed to by the referral structures */ -} TRANSACTION2_GET_DFS_REFER_RSP; +} __attribute__((packed)) TRANSACTION2_GET_DFS_REFER_RSP; /* DFS Flags */ #define DFSREF_REFERRAL_SERVER 0x0001 @@ -1512,7 +1658,7 @@ struct serverInfo { unsigned char versionMinor; unsigned long type; unsigned int commentOffset; -}; +} __attribute__((packed)); /* * The following structure is the format of the data returned on a NetShareEnum @@ -1524,39 +1670,55 @@ struct shareInfo { char pad; unsigned short type; unsigned int commentOffset; -}; +} __attribute__((packed)); struct aliasInfo { char aliasName[9]; char pad; unsigned int commentOffset; unsigned char type[2]; -}; +} __attribute__((packed)); struct aliasInfo92 { int aliasNameOffset; int serverNameOffset; int shareNameOffset; -}; +} __attribute__((packed)); typedef struct { __le64 TotalAllocationUnits; __le64 FreeAllocationUnits; __le32 SectorsPerAllocationUnit; __le32 BytesPerSector; -} FILE_SYSTEM_INFO; /* size info, level 0x103 */ +} __attribute__((packed)) FILE_SYSTEM_INFO; /* size info, level 0x103 */ + +typedef struct { + __le32 fsid; + __le32 SectorsPerAllocationUnit; + __le32 TotalAllocationUnits; + __le32 FreeAllocationUnits; + __le16 BytesPerSector; +} __attribute__((packed)) FILE_SYSTEM_ALLOC_INFO; typedef struct { __le16 MajorVersionNumber; __le16 MinorVersionNumber; __le64 Capability; -} FILE_SYSTEM_UNIX_INFO; /* Unix extensions info, level 0x200 */ +} __attribute__((packed)) FILE_SYSTEM_UNIX_INFO; /* Unix extensions info, level 0x200 */ + +/* Version numbers for CIFS UNIX major and minor. */ +#define CIFS_UNIX_MAJOR_VERSION 1 +#define CIFS_UNIX_MINOR_VERSION 0 + /* Linux/Unix extensions capability flags */ #define CIFS_UNIX_FCNTL_CAP 0x00000001 /* support for fcntl locks */ #define CIFS_UNIX_POSIX_ACL_CAP 0x00000002 /* support getfacl/setfacl */ #define CIFS_UNIX_XATTR_CAP 0x00000004 /* support new namespace */ #define CIFS_UNIX_EXTATTR_CAP 0x00000008 /* support chattr/chflag */ +#define CIFS_UNIX_POSIX_PATHNAMES_CAP 0x00000010 /* Use POSIX pathnames on the wire. */ + #define CIFS_POSIX_EXTENSIONS 0x00000010 /* support for new QFSInfo */ + typedef struct { /* For undefined recommended transfer size return -1 in that field */ __le32 OptimalTransferSize; /* bsize on some os, iosize on other os */ @@ -1577,7 +1739,7 @@ typedef struct { __le64 FileSysIdentifier; /* fsid */ /* NB Namelen comes from FILE_SYSTEM_ATTRIBUTE_INFO call */ /* NB flags can come from FILE_SYSTEM_DEVICE_INFO call */ -} FILE_SYSTEM_POSIX_INFO; +} __attribute__((packed)) FILE_SYSTEM_POSIX_INFO; /* DeviceType Flags */ #define FILE_DEVICE_CD_ROM 0x00000002 @@ -1602,14 +1764,14 @@ typedef struct { typedef struct { __le32 DeviceType; __le32 DeviceCharacteristics; -} FILE_SYSTEM_DEVICE_INFO; /* device info, level 0x104 */ +} __attribute__((packed)) FILE_SYSTEM_DEVICE_INFO; /* device info, level 0x104 */ typedef struct { __le32 Attributes; __le32 MaxPathNameComponentLength; __le32 FileSystemNameLen; char FileSystemName[52]; /* do not really need to save this - so potentially get only subset of name */ -} FILE_SYSTEM_ATTRIBUTE_INFO; +} __attribute__((packed)) FILE_SYSTEM_ATTRIBUTE_INFO; /******************************************************************************/ /* QueryFileInfo/QueryPathinfo (also for SetPath/SetFile) data buffer formats */ @@ -1636,7 +1798,7 @@ typedef struct { /* data block encoding of response to level 263 QPathInfo */ __le32 AlignmentRequirement; __le32 FileNameLength; char FileName[1]; -} FILE_ALL_INFO; /* level 0x107 QPathInfo */ +} __attribute__((packed)) FILE_ALL_INFO; /* level 0x107 QPathInfo */ /* defines for enumerating possible values of the Unix type field below */ #define UNIX_FILE 0 @@ -1660,11 +1822,11 @@ typedef struct { __u64 UniqueId; __le64 Permissions; __le64 Nlinks; -} FILE_UNIX_BASIC_INFO; /* level 0x200 QPathInfo */ +} __attribute__((packed)) FILE_UNIX_BASIC_INFO; /* level 0x200 QPathInfo */ typedef struct { char LinkDest[1]; -} FILE_UNIX_LINK_INFO; /* level 0x201 QPathInfo */ +} __attribute__((packed)) FILE_UNIX_LINK_INFO; /* level 0x201 QPathInfo */ /* The following three structures are needed only for setting time to NT4 and some older servers via @@ -1673,13 +1835,13 @@ typedef struct { __u16 Day:5; __u16 Month:4; __u16 Year:7; -} SMB_DATE; +} __attribute__((packed)) SMB_DATE; typedef struct { __u16 TwoSeconds:5; __u16 Minutes:6; __u16 Hours:5; -} SMB_TIME; +} __attribute__((packed)) SMB_TIME; typedef struct { __le16 CreationDate; /* SMB Date see above */ @@ -1692,7 +1854,7 @@ typedef struct { __le32 AllocationSize; __le16 Attributes; /* verify not u32 */ __le32 EASize; -} FILE_INFO_STANDARD; /* level 1 SetPath/FileInfo */ +} __attribute__((packed)) FILE_INFO_STANDARD; /* level 1 SetPath/FileInfo */ typedef struct { __le64 CreationTime; @@ -1701,19 +1863,19 @@ typedef struct { __le64 ChangeTime; __le32 Attributes; __u32 Pad; -} FILE_BASIC_INFO; /* size info, level 0x101 */ +} __attribute__((packed)) FILE_BASIC_INFO; /* size info, level 0x101 */ struct file_allocation_info { __le64 AllocationSize; /* Note old Samba srvr rounds this up too much */ -}; /* size used on disk, level 0x103 for set, 0x105 for query */ +} __attribute__((packed)); /* size used on disk, level 0x103 for set, 0x105 for query */ struct file_end_of_file_info { __le64 FileSize; /* offset to end of file */ -}; /* size info, level 0x104 for set, 0x106 for query */ +} __attribute__((packed)); /* size info, level 0x104 for set, 0x106 for query */ struct file_alt_name_info { __u8 alt_name[1]; -}; /* level 0x0108 */ +} __attribute__((packed)); /* level 0x0108 */ struct file_stream_info { __le32 number_of_streams; /* BB check sizes and verify location */ @@ -1730,7 +1892,7 @@ struct file_compression_info { __u8 ch_shift; __u8 cl_shift; __u8 pad[3]; -}; /* level 0x10b */ +} __attribute__((packed)); /* level 0x10b */ /* POSIX ACL set/query path info structures */ #define CIFS_ACL_VERSION 1 @@ -1738,7 +1900,7 @@ struct cifs_posix_ace { /* access control entry (ACE) */ __u8 cifs_e_tag; __u8 cifs_e_perm; __le64 cifs_uid; /* or gid */ -}; +} __attribute__((packed)); struct cifs_posix_acl { /* access conrol list (ACL) */ __le16 version; @@ -1747,7 +1909,7 @@ struct cifs_posix_acl { /* access conrol list (ACL) */ struct cifs_posix_ace ace_array[0]; /* followed by struct cifs_posix_ace default_ace_arraay[] */ -}; /* level 0x204 */ +} __attribute__((packed)); /* level 0x204 */ /* types of access control entries already defined in posix_acl.h */ /* #define CIFS_POSIX_ACL_USER_OBJ 0x01 @@ -1766,15 +1928,15 @@ struct cifs_posix_acl { /* access conrol list (ACL) */ struct file_internal_info { __u64 UniqueId; /* inode number */ -}; /* level 0x3ee */ +} __attribute__((packed)); /* level 0x3ee */ struct file_mode_info { __le32 Mode; -}; /* level 0x3f8 */ +} __attribute__((packed)); /* level 0x3f8 */ struct file_attrib_tag { __le32 Attribute; __le32 ReparseTag; -}; /* level 0x40b */ +} __attribute__((packed)); /* level 0x40b */ /********************************************************/ @@ -1798,7 +1960,7 @@ typedef struct { __le64 Permissions; __le64 Nlinks; char FileName[1]; -} FILE_UNIX_INFO; /* level 0x202 */ +} __attribute__((packed)) FILE_UNIX_INFO; /* level 0x202 */ typedef struct { __le32 NextEntryOffset; @@ -1812,7 +1974,7 @@ typedef struct { __le32 ExtFileAttributes; __le32 FileNameLength; char FileName[1]; -} FILE_DIRECTORY_INFO; /* level 0x101 FF response data area */ +} __attribute__((packed)) FILE_DIRECTORY_INFO; /* level 0x101 FF response data area */ typedef struct { __le32 NextEntryOffset; @@ -1827,7 +1989,7 @@ typedef struct { __le32 FileNameLength; __le32 EaSize; /* length of the xattrs */ char FileName[1]; -} FILE_FULL_DIRECTORY_INFO; /* level 0x102 FF response data area */ +} __attribute__((packed)) FILE_FULL_DIRECTORY_INFO; /* level 0x102 FF response data area */ typedef struct { __le32 NextEntryOffset; @@ -1844,7 +2006,7 @@ typedef struct { __le32 Reserved; __u64 UniqueId; /* inode num - le since Samba puts ino in low 32 bit*/ char FileName[1]; -} SEARCH_ID_FULL_DIR_INFO; /* level 0x105 FF response data area */ +} __attribute__((packed)) SEARCH_ID_FULL_DIR_INFO; /* level 0x105 FF response data area */ typedef struct { __le32 NextEntryOffset; @@ -1862,18 +2024,24 @@ typedef struct { __u8 Reserved; __u8 ShortName[12]; char FileName[1]; -} FILE_BOTH_DIRECTORY_INFO; /* level 0x104 FF response data area */ +} __attribute__((packed)) FILE_BOTH_DIRECTORY_INFO; /* level 0x104 FF response data area */ + +struct win_dev { + unsigned char type[8]; /* IntxCHR or IntxBLK */ + __le64 major; + __le64 minor; +} __attribute__((packed)); struct gea { unsigned char name_len; char name[1]; -}; +} __attribute__((packed)); struct gealist { unsigned long list_len; struct gea list[1]; -}; +} __attribute__((packed)); struct fea { unsigned char EA_flags; @@ -1881,21 +2049,21 @@ struct fea { __le16 value_len; char name[1]; /* optionally followed by value */ -}; +} __attribute__((packed)); /* flags for _FEA.fEA */ #define FEA_NEEDEA 0x80 /* need EA bit */ struct fealist { __le32 list_len; struct fea list[1]; -}; +} __attribute__((packed)); /* used to hold an arbitrary blob of data */ struct data_blob { __u8 *data; size_t length; void (*free) (struct data_blob * data_blob); -}; +} __attribute__((packed)); #ifdef CONFIG_CIFS_POSIX @@ -1907,18 +2075,17 @@ struct data_blob { perhaps add a CreateDevice - to create Pipes and other special .inodes Also note POSIX open flags 2) Close - to return the last write time to do cache across close more safely - 3) PosixQFSInfo - to return statfs info - 4) FindFirst return unique inode number - what about resume key, two forms short (matches readdir) and full (enough info to cache inodes) - 5) Mkdir - set mode + 3) FindFirst return unique inode number - what about resume key, two + forms short (matches readdir) and full (enough info to cache inodes) + 4) Mkdir - set mode And under consideration: - 6) FindClose2 (return nanosecond timestamp ??) - 7) Use nanosecond timestamps throughout all time fields if + 5) FindClose2 (return nanosecond timestamp ??) + 6) Use nanosecond timestamps throughout all time fields if corresponding attribute flag is set - 8) sendfile - handle based copy - 9) Direct i/o - 10) "POSIX ACL" support - 11) Misc fcntls? + 7) sendfile - handle based copy + 8) Direct i/o + 9) Misc fcntls? what about fixing 64 bit alignment @@ -1974,7 +2141,7 @@ struct data_blob { */ -/* xsymlink is a symlink format that can be used +/* xsymlink is a symlink format (used by MacOS) that can be used to save symlink info in a regular file when mounted to operating systems that do not support the cifs Unix extensions or EAs (for xattr @@ -1999,7 +2166,7 @@ struct xsymlink { char cr2; /* \n */ /* if room left, then end with \n then 0x20s by convention but not required */ char path[1024]; -}; +} __attribute__((packed)); typedef struct file_xattr_info { /* BB do we need another field for flags? BB */ @@ -2007,7 +2174,7 @@ typedef struct file_xattr_info { __u32 xattr_value_len; char xattr_name[0]; /* followed by xattr_value[xattr_value_len], no pad */ -} FILE_XATTR_INFO; /* extended attribute, info level 0x205 */ +} __attribute__((packed)) FILE_XATTR_INFO; /* extended attribute, info level 0x205 */ /* flags for chattr command */ @@ -2033,10 +2200,8 @@ typedef struct file_xattr_info { typedef struct file_chattr_info { __le64 mask; /* list of all possible attribute bits */ __le64 mode; /* list of actual attribute bits on this inode */ -} FILE_CHATTR_INFO; /* ext attributes (chattr, chflags) level 0x206 */ +} __attribute__((packed)) FILE_CHATTR_INFO; /* ext attributes (chattr, chflags) level 0x206 */ #endif -#pragma pack() /* resume default structure packing */ - #endif /* _CIFSPDU_H */ diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index ea239dea571e..1b73f4f4c5ce 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -47,19 +47,24 @@ extern int SendReceive(const unsigned int /* xid */ , struct cifsSesInfo *, struct smb_hdr * /* input */ , struct smb_hdr * /* out */ , int * /* bytes returned */ , const int long_op); +extern int SendReceive2(const unsigned int /* xid */ , struct cifsSesInfo *, + struct kvec *, int /* nvec */, + int * /* bytes returned */ , const int long_op); extern int checkSMBhdr(struct smb_hdr *smb, __u16 mid); extern int checkSMB(struct smb_hdr *smb, __u16 mid, int length); extern int is_valid_oplock_break(struct smb_hdr *smb); extern int is_size_safe_to_change(struct cifsInodeInfo *); +extern struct cifsFileInfo *find_writable_file(struct cifsInodeInfo *); extern unsigned int smbCalcSize(struct smb_hdr *ptr); +extern unsigned int smbCalcSize_LE(struct smb_hdr *ptr); extern int decode_negTokenInit(unsigned char *security_blob, int length, enum securityEnum *secType); extern int cifs_inet_pton(int, char * source, void *dst); extern int map_smb_to_linux_error(struct smb_hdr *smb); extern void header_assemble(struct smb_hdr *, char /* command */ , - const struct cifsTconInfo *, int /* specifies length - of fixed section (word count) in two byte units */ - ); + const struct cifsTconInfo *, int /* length of + fixed section (word count) in two byte units */); +extern __u16 GetNextMid(struct TCP_Server_Info *server); extern struct oplock_q_entry * AllocOplockQEntry(struct inode *, u16, struct cifsTconInfo *); extern void DeleteOplockQEntry(struct oplock_q_entry *); @@ -89,7 +94,7 @@ extern int CIFSTCon(unsigned int xid, struct cifsSesInfo *ses, extern int CIFSFindFirst(const int xid, struct cifsTconInfo *tcon, const char *searchName, const struct nls_table *nls_codepage, - __u16 *searchHandle, struct cifs_search_info * psrch_inf, int map); + __u16 *searchHandle, struct cifs_search_info * psrch_inf, int map, const char dirsep); extern int CIFSFindNext(const int xid, struct cifsTconInfo *tcon, __u16 searchHandle, struct cifs_search_info * psrch_inf); @@ -101,6 +106,10 @@ extern int CIFSSMBQPathInfo(const int xid, struct cifsTconInfo *tcon, const unsigned char *searchName, FILE_ALL_INFO * findData, const struct nls_table *nls_codepage, int remap); +extern int SMBQueryInformation(const int xid, struct cifsTconInfo *tcon, + const unsigned char *searchName, + FILE_ALL_INFO * findData, + const struct nls_table *nls_codepage, int remap); extern int CIFSSMBUnixQPathInfo(const int xid, struct cifsTconInfo *tcon, @@ -125,6 +134,11 @@ extern int get_dfs_path(int xid, struct cifsSesInfo *pSesInfo, int remap); extern int CIFSSMBQFSInfo(const int xid, struct cifsTconInfo *tcon, struct kstatfs *FSData); +extern int SMBOldQFSInfo(const int xid, struct cifsTconInfo *tcon, + struct kstatfs *FSData); +extern int CIFSSMBSetFSUnixInfo(const int xid, struct cifsTconInfo *tcon, + __u64 cap); + extern int CIFSSMBQFSAttributeInfo(const int xid, struct cifsTconInfo *tcon); extern int CIFSSMBQFSDeviceInfo(const int xid, struct cifsTconInfo *tcon); @@ -207,6 +221,11 @@ extern int CIFSSMBOpen(const int xid, struct cifsTconInfo *tcon, const int access_flags, const int omode, __u16 * netfid, int *pOplock, FILE_ALL_INFO *, const struct nls_table *nls_codepage, int remap); +extern int SMBLegacyOpen(const int xid, struct cifsTconInfo *tcon, + const char *fileName, const int disposition, + const int access_flags, const int omode, + __u16 * netfid, int *pOplock, FILE_ALL_INFO *, + const struct nls_table *nls_codepage, int remap); extern int CIFSSMBClose(const int xid, struct cifsTconInfo *tcon, const int smb_file_id); @@ -222,12 +241,12 @@ extern int CIFSSMBWrite(const int xid, struct cifsTconInfo *tcon, extern int CIFSSMBWrite2(const int xid, struct cifsTconInfo *tcon, const int netfid, const unsigned int count, const __u64 offset, unsigned int *nbytes, - const char __user *buf,const int long_op); + struct kvec *iov, const int nvec, const int long_op); +#endif /* CONFIG_CIFS_EXPERIMENTAL */ extern int CIFSGetSrvInodeNumber(const int xid, struct cifsTconInfo *tcon, const unsigned char *searchName, __u64 * inode_number, const struct nls_table *nls_codepage, int remap_special_chars); -#endif /* CONFIG_CIFS_EXPERIMENTAL */ extern int cifs_convertUCSpath(char *target, const __le16 *source, int maxlen, const struct nls_table * codepage); extern int cifsConvertToUCS(__le16 * target, const char *source, int maxlen, @@ -264,7 +283,8 @@ extern int CIFSSMBCopy(int xid, int remap_special_chars); extern int CIFSSMBNotify(const int xid, struct cifsTconInfo *tcon, const int notify_subdirs,const __u16 netfid, - __u32 filter, const struct nls_table *nls_codepage); + __u32 filter, struct file * file, int multishot, + const struct nls_table *nls_codepage); extern ssize_t CIFSSMBQAllEAs(const int xid, struct cifsTconInfo *tcon, const unsigned char *searchName, char * EAData, size_t bufsize, const struct nls_table *nls_codepage, diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 0db0b313d715..6867e556d37e 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -90,6 +90,18 @@ small_smb_init(int smb_command, int wct, struct cifsTconInfo *tcon, check for tcp and smb session status done differently for those three - in the calling routine */ if(tcon) { + if(tcon->tidStatus == CifsExiting) { + /* only tree disconnect, open, and write, + (and ulogoff which does not have tcon) + are allowed as we start force umount */ + if((smb_command != SMB_COM_WRITE_ANDX) && + (smb_command != SMB_COM_OPEN_ANDX) && + (smb_command != SMB_COM_TREE_DISCONNECT)) { + cFYI(1,("can not send cmd %d while umounting", + smb_command)); + return -ENODEV; + } + } if((tcon->ses) && (tcon->ses->status != CifsExiting) && (tcon->ses->server)){ struct nls_table *nls_codepage; @@ -125,6 +137,9 @@ small_smb_init(int smb_command, int wct, struct cifsTconInfo *tcon, rc = CIFSTCon(0, tcon->ses, tcon->treeName, tcon , nls_codepage); up(&tcon->ses->sesSem); + /* BB FIXME add code to check if wsize needs + update due to negotiated smb buffer size + shrinking */ if(rc == 0) atomic_inc(&tconInfoReconnectCount); @@ -166,11 +181,9 @@ small_smb_init(int smb_command, int wct, struct cifsTconInfo *tcon, header_assemble((struct smb_hdr *) *request_buf, smb_command, tcon,wct); -#ifdef CONFIG_CIFS_STATS - if(tcon != NULL) { - atomic_inc(&tcon->num_smbs_sent); - } -#endif /* CONFIG_CIFS_STATS */ + if(tcon != NULL) + cifs_stats_inc(&tcon->num_smbs_sent); + return rc; } @@ -186,6 +199,19 @@ smb_init(int smb_command, int wct, struct cifsTconInfo *tcon, check for tcp and smb session status done differently for those three - in the calling routine */ if(tcon) { + if(tcon->tidStatus == CifsExiting) { + /* only tree disconnect, open, and write, + (and ulogoff which does not have tcon) + are allowed as we start force umount */ + if((smb_command != SMB_COM_WRITE_ANDX) && + (smb_command != SMB_COM_OPEN_ANDX) && + (smb_command != SMB_COM_TREE_DISCONNECT)) { + cFYI(1,("can not send cmd %d while umounting", + smb_command)); + return -ENODEV; + } + } + if((tcon->ses) && (tcon->ses->status != CifsExiting) && (tcon->ses->server)){ struct nls_table *nls_codepage; @@ -222,6 +248,9 @@ smb_init(int smb_command, int wct, struct cifsTconInfo *tcon, rc = CIFSTCon(0, tcon->ses, tcon->treeName, tcon, nls_codepage); up(&tcon->ses->sesSem); + /* BB FIXME add code to check if wsize needs + update due to negotiated smb buffer size + shrinking */ if(rc == 0) atomic_inc(&tconInfoReconnectCount); @@ -269,11 +298,9 @@ smb_init(int smb_command, int wct, struct cifsTconInfo *tcon, header_assemble((struct smb_hdr *) *request_buf, smb_command, tcon, wct /*wct */ ); -#ifdef CONFIG_CIFS_STATS - if(tcon != NULL) { - atomic_inc(&tcon->num_smbs_sent); - } -#endif /* CONFIG_CIFS_STATS */ + if(tcon != NULL) + cifs_stats_inc(&tcon->num_smbs_sent); + return rc; } @@ -330,7 +357,7 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) (void **) &pSMB, (void **) &pSMBr); if (rc) return rc; - + pSMB->hdr.Mid = GetNextMid(server); pSMB->hdr.Flags2 |= SMBFLG2_UNICODE; if (extended_security) pSMB->hdr.Flags2 |= SMBFLG2_EXT_SEC; @@ -422,8 +449,8 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) } } - if (pSMB) - cifs_buf_release(pSMB); + + cifs_buf_release(pSMB); return rc; } @@ -518,6 +545,8 @@ CIFSSMBLogoff(const int xid, struct cifsSesInfo *ses) smb_buffer_response = (struct smb_hdr *)pSMB; /* BB removeme BB */ if(ses->server) { + pSMB->hdr.Mid = GetNextMid(ses->server); + if(ses->server->secMode & (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) pSMB->hdr.Flags2 |= SMBFLG2_SECURITY_SIGNATURE; @@ -537,9 +566,8 @@ CIFSSMBLogoff(const int xid, struct cifsSesInfo *ses) rc = -ESHUTDOWN; } } - if (pSMB) - cifs_small_buf_release(pSMB); up(&ses->sesSem); + cifs_small_buf_release(pSMB); /* if session dead then we do not need to do ulogoff, since server closed smb session, no sense reporting @@ -583,14 +611,10 @@ DelFileRetry: pSMB->ByteCount = cpu_to_le16(name_len + 1); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); + cifs_stats_inc(&tcon->num_deletes); if (rc) { cFYI(1, ("Error in RMFile = %d", rc)); } -#ifdef CONFIG_CIFS_STATS - else { - atomic_inc(&tcon->num_deletes); - } -#endif cifs_buf_release(pSMB); if (rc == -EAGAIN) @@ -632,14 +656,10 @@ RmDirRetry: pSMB->ByteCount = cpu_to_le16(name_len + 1); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); + cifs_stats_inc(&tcon->num_rmdirs); if (rc) { cFYI(1, ("Error in RMDir = %d", rc)); } -#ifdef CONFIG_CIFS_STATS - else { - atomic_inc(&tcon->num_rmdirs); - } -#endif cifs_buf_release(pSMB); if (rc == -EAGAIN) @@ -680,20 +700,161 @@ MkDirRetry: pSMB->ByteCount = cpu_to_le16(name_len + 1); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); + cifs_stats_inc(&tcon->num_mkdirs); if (rc) { cFYI(1, ("Error in Mkdir = %d", rc)); } -#ifdef CONFIG_CIFS_STATS - else { - atomic_inc(&tcon->num_mkdirs); - } -#endif + cifs_buf_release(pSMB); if (rc == -EAGAIN) goto MkDirRetry; return rc; } +static __u16 convert_disposition(int disposition) +{ + __u16 ofun = 0; + + switch (disposition) { + case FILE_SUPERSEDE: + ofun = SMBOPEN_OCREATE | SMBOPEN_OTRUNC; + break; + case FILE_OPEN: + ofun = SMBOPEN_OAPPEND; + break; + case FILE_CREATE: + ofun = SMBOPEN_OCREATE; + break; + case FILE_OPEN_IF: + ofun = SMBOPEN_OCREATE | SMBOPEN_OAPPEND; + break; + case FILE_OVERWRITE: + ofun = SMBOPEN_OTRUNC; + break; + case FILE_OVERWRITE_IF: + ofun = SMBOPEN_OCREATE | SMBOPEN_OTRUNC; + break; + default: + cFYI(1,("unknown disposition %d",disposition)); + ofun = SMBOPEN_OAPPEND; /* regular open */ + } + return ofun; +} + +int +SMBLegacyOpen(const int xid, struct cifsTconInfo *tcon, + const char *fileName, const int openDisposition, + const int access_flags, const int create_options, __u16 * netfid, + int *pOplock, FILE_ALL_INFO * pfile_info, + const struct nls_table *nls_codepage, int remap) +{ + int rc = -EACCES; + OPENX_REQ *pSMB = NULL; + OPENX_RSP *pSMBr = NULL; + int bytes_returned; + int name_len; + __u16 count; + +OldOpenRetry: + rc = smb_init(SMB_COM_OPEN_ANDX, 15, tcon, (void **) &pSMB, + (void **) &pSMBr); + if (rc) + return rc; + + pSMB->AndXCommand = 0xFF; /* none */ + + if (pSMB->hdr.Flags2 & SMBFLG2_UNICODE) { + count = 1; /* account for one byte pad to word boundary */ + name_len = + cifsConvertToUCS((__le16 *) (pSMB->fileName + 1), + fileName, PATH_MAX, nls_codepage, remap); + name_len++; /* trailing null */ + name_len *= 2; + } else { /* BB improve check for buffer overruns BB */ + count = 0; /* no pad */ + name_len = strnlen(fileName, PATH_MAX); + name_len++; /* trailing null */ + strncpy(pSMB->fileName, fileName, name_len); + } + if (*pOplock & REQ_OPLOCK) + pSMB->OpenFlags = cpu_to_le16(REQ_OPLOCK); + else if (*pOplock & REQ_BATCHOPLOCK) { + pSMB->OpenFlags = cpu_to_le16(REQ_BATCHOPLOCK); + } + pSMB->OpenFlags |= cpu_to_le16(REQ_MORE_INFO); + /* BB fixme add conversion for access_flags to bits 0 - 2 of mode */ + /* 0 = read + 1 = write + 2 = rw + 3 = execute + */ + pSMB->Mode = cpu_to_le16(2); + pSMB->Mode |= cpu_to_le16(0x40); /* deny none */ + /* set file as system file if special file such + as fifo and server expecting SFU style and + no Unix extensions */ + + if(create_options & CREATE_OPTION_SPECIAL) + pSMB->FileAttributes = cpu_to_le16(ATTR_SYSTEM); + else + pSMB->FileAttributes = cpu_to_le16(0/*ATTR_NORMAL*/); /* BB FIXME */ + + /* if ((omode & S_IWUGO) == 0) + pSMB->FileAttributes |= cpu_to_le32(ATTR_READONLY);*/ + /* Above line causes problems due to vfs splitting create into two + pieces - need to set mode after file created not while it is + being created */ + + /* BB FIXME BB */ +/* pSMB->CreateOptions = cpu_to_le32(create_options & CREATE_OPTIONS_MASK); */ + /* BB FIXME END BB */ + + pSMB->Sattr = cpu_to_le16(ATTR_HIDDEN | ATTR_SYSTEM | ATTR_DIRECTORY); + pSMB->OpenFunction = cpu_to_le16(convert_disposition(openDisposition)); + count += name_len; + pSMB->hdr.smb_buf_length += count; + + pSMB->ByteCount = cpu_to_le16(count); + /* long_op set to 1 to allow for oplock break timeouts */ + rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, + (struct smb_hdr *) pSMBr, &bytes_returned, 1); + cifs_stats_inc(&tcon->num_opens); + if (rc) { + cFYI(1, ("Error in Open = %d", rc)); + } else { + /* BB verify if wct == 15 */ + +/* *pOplock = pSMBr->OplockLevel; */ /* BB take from action field BB */ + + *netfid = pSMBr->Fid; /* cifs fid stays in le */ + /* Let caller know file was created so we can set the mode. */ + /* Do we care about the CreateAction in any other cases? */ + /* BB FIXME BB */ +/* if(cpu_to_le32(FILE_CREATE) == pSMBr->CreateAction) + *pOplock |= CIFS_CREATE_ACTION; */ + /* BB FIXME END */ + + if(pfile_info) { + pfile_info->CreationTime = 0; /* BB convert CreateTime*/ + pfile_info->LastAccessTime = 0; /* BB fixme */ + pfile_info->LastWriteTime = 0; /* BB fixme */ + pfile_info->ChangeTime = 0; /* BB fixme */ + pfile_info->Attributes = + cpu_to_le32(le16_to_cpu(pSMBr->FileAttributes)); + /* the file_info buf is endian converted by caller */ + pfile_info->AllocationSize = + cpu_to_le64(le32_to_cpu(pSMBr->EndOfFile)); + pfile_info->EndOfFile = pfile_info->AllocationSize; + pfile_info->NumberOfLinks = cpu_to_le32(1); + } + } + + cifs_buf_release(pSMB); + if (rc == -EAGAIN) + goto OldOpenRetry; + return rc; +} + int CIFSSMBOpen(const int xid, struct cifsTconInfo *tcon, const char *fileName, const int openDisposition, @@ -738,7 +899,13 @@ openRetry: } pSMB->DesiredAccess = cpu_to_le32(access_flags); pSMB->AllocationSize = 0; - pSMB->FileAttributes = cpu_to_le32(ATTR_NORMAL); + /* set file as system file if special file such + as fifo and server expecting SFU style and + no Unix extensions */ + if(create_options & CREATE_OPTION_SPECIAL) + pSMB->FileAttributes = cpu_to_le32(ATTR_SYSTEM); + else + pSMB->FileAttributes = cpu_to_le32(ATTR_NORMAL); /* XP does not handle ATTR_POSIX_SEMANTICS */ /* but it helps speed up case sensitive checks for other servers such as Samba */ @@ -752,7 +919,7 @@ openRetry: being created */ pSMB->ShareAccess = cpu_to_le32(FILE_SHARE_ALL); pSMB->CreateDisposition = cpu_to_le32(openDisposition); - pSMB->CreateOptions = cpu_to_le32(create_options); + pSMB->CreateOptions = cpu_to_le32(create_options & CREATE_OPTIONS_MASK); /* BB Expirement with various impersonation levels and verify */ pSMB->ImpersonationLevel = cpu_to_le32(SECURITY_IMPERSONATION); pSMB->SecurityFlags = @@ -765,6 +932,7 @@ openRetry: /* long_op set to 1 to allow for oplock break timeouts */ rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 1); + cifs_stats_inc(&tcon->num_opens); if (rc) { cFYI(1, ("Error in Open = %d", rc)); } else { @@ -782,11 +950,8 @@ openRetry: pfile_info->EndOfFile = pSMBr->EndOfFile; pfile_info->NumberOfLinks = cpu_to_le32(1); } - -#ifdef CONFIG_CIFS_STATS - atomic_inc(&tcon->num_opens); -#endif } + cifs_buf_release(pSMB); if (rc == -EAGAIN) goto openRetry; @@ -807,11 +972,16 @@ CIFSSMBRead(const int xid, struct cifsTconInfo *tcon, READ_RSP *pSMBr = NULL; char *pReadData = NULL; int bytes_returned; + int wct; cFYI(1,("Reading %d bytes on fid %d",count,netfid)); + if(tcon->ses->capabilities & CAP_LARGE_FILES) + wct = 12; + else + wct = 10; /* old style read */ *nbytes = 0; - rc = smb_init(SMB_COM_READ_ANDX, 12, tcon, (void **) &pSMB, + rc = smb_init(SMB_COM_READ_ANDX, wct, tcon, (void **) &pSMB, (void **) &pSMBr); if (rc) return rc; @@ -823,14 +993,26 @@ CIFSSMBRead(const int xid, struct cifsTconInfo *tcon, pSMB->AndXCommand = 0xFF; /* none */ pSMB->Fid = netfid; pSMB->OffsetLow = cpu_to_le32(lseek & 0xFFFFFFFF); - pSMB->OffsetHigh = cpu_to_le32(lseek >> 32); + if(wct == 12) + pSMB->OffsetHigh = cpu_to_le32(lseek >> 32); + else if((lseek >> 32) > 0) /* can not handle this big offset for old */ + return -EIO; + pSMB->Remaining = 0; pSMB->MaxCount = cpu_to_le16(count & 0xFFFF); pSMB->MaxCountHigh = cpu_to_le32(count >> 16); - pSMB->ByteCount = 0; /* no need to do le conversion since it is 0 */ - + if(wct == 12) + pSMB->ByteCount = 0; /* no need to do le conversion since 0 */ + else { + /* old style read */ + struct smb_com_readx_req * pSMBW = + (struct smb_com_readx_req *)pSMB; + pSMBW->ByteCount = 0; + } + rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); + cifs_stats_inc(&tcon->num_reads); if (rc) { cERROR(1, ("Send error in read = %d", rc)); } else { @@ -876,12 +1058,20 @@ CIFSSMBWrite(const int xid, struct cifsTconInfo *tcon, int rc = -EACCES; WRITE_REQ *pSMB = NULL; WRITE_RSP *pSMBr = NULL; - int bytes_returned; + int bytes_returned, wct; __u32 bytes_sent; __u16 byte_count; /* cFYI(1,("write at %lld %d bytes",offset,count));*/ - rc = smb_init(SMB_COM_WRITE_ANDX, 14, tcon, (void **) &pSMB, + if(tcon->ses == NULL) + return -ECONNABORTED; + + if(tcon->ses->capabilities & CAP_LARGE_FILES) + wct = 14; + else + wct = 12; + + rc = smb_init(SMB_COM_WRITE_ANDX, wct, tcon, (void **) &pSMB, (void **) &pSMBr); if (rc) return rc; @@ -892,7 +1082,11 @@ CIFSSMBWrite(const int xid, struct cifsTconInfo *tcon, pSMB->AndXCommand = 0xFF; /* none */ pSMB->Fid = netfid; pSMB->OffsetLow = cpu_to_le32(offset & 0xFFFFFFFF); - pSMB->OffsetHigh = cpu_to_le32(offset >> 32); + if(wct == 14) + pSMB->OffsetHigh = cpu_to_le32(offset >> 32); + else if((offset >> 32) > 0) /* can not handle this big offset for old */ + return -EIO; + pSMB->Reserved = 0xFFFFFFFF; pSMB->WriteMode = 0; pSMB->Remaining = 0; @@ -911,7 +1105,7 @@ CIFSSMBWrite(const int xid, struct cifsTconInfo *tcon, if (bytes_sent > count) bytes_sent = count; pSMB->DataOffset = - cpu_to_le16(offsetof(struct smb_com_write_req,Data) - 4); + cpu_to_le16(offsetof(struct smb_com_write_req,Data) - 4); if(buf) memcpy(pSMB->Data,buf,bytes_sent); else if(ubuf) { @@ -919,20 +1113,31 @@ CIFSSMBWrite(const int xid, struct cifsTconInfo *tcon, cifs_buf_release(pSMB); return -EFAULT; } - } else { + } else if (count != 0) { /* No buffer */ cifs_buf_release(pSMB); return -EINVAL; + } /* else setting file size with write of zero bytes */ + if(wct == 14) + byte_count = bytes_sent + 1; /* pad */ + else /* wct == 12 */ { + byte_count = bytes_sent + 5; /* bigger pad, smaller smb hdr */ } - - byte_count = bytes_sent + 1 /* pad */ ; /* BB fix this for sends > 64K */ pSMB->DataLengthLow = cpu_to_le16(bytes_sent & 0xFFFF); pSMB->DataLengthHigh = cpu_to_le16(bytes_sent >> 16); - pSMB->hdr.smb_buf_length += bytes_sent+1; - pSMB->ByteCount = cpu_to_le16(byte_count); + pSMB->hdr.smb_buf_length += byte_count; + + if(wct == 14) + pSMB->ByteCount = cpu_to_le16(byte_count); + else { /* old style write has byte count 4 bytes earlier so 4 bytes pad */ + struct smb_com_writex_req * pSMBW = + (struct smb_com_writex_req *)pSMB; + pSMBW->ByteCount = cpu_to_le16(byte_count); + } rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, long_op); + cifs_stats_inc(&tcon->num_writes); if (rc) { cFYI(1, ("Send error in write = %d", rc)); *nbytes = 0; @@ -951,56 +1156,74 @@ CIFSSMBWrite(const int xid, struct cifsTconInfo *tcon, } #ifdef CONFIG_CIFS_EXPERIMENTAL -int CIFSSMBWrite2(const int xid, struct cifsTconInfo *tcon, +int +CIFSSMBWrite2(const int xid, struct cifsTconInfo *tcon, const int netfid, const unsigned int count, - const __u64 offset, unsigned int *nbytes, const char __user *buf, - const int long_op) + const __u64 offset, unsigned int *nbytes, struct kvec *iov, + int n_vec, const int long_op) { int rc = -EACCES; WRITE_REQ *pSMB = NULL; - WRITE_RSP *pSMBr = NULL; - /*int bytes_returned;*/ - unsigned bytes_sent; - __u16 byte_count; + int bytes_returned, wct; + int smb_hdr_len; - rc = small_smb_init(SMB_COM_WRITE_ANDX, 14, tcon, (void **) &pSMB); - + /* BB removeme BB */ + cFYI(1,("write2 at %lld %d bytes", (long long)offset, count)); + + if(tcon->ses->capabilities & CAP_LARGE_FILES) + wct = 14; + else + wct = 12; + rc = small_smb_init(SMB_COM_WRITE_ANDX, wct, tcon, (void **) &pSMB); if (rc) return rc; - - pSMBr = (WRITE_RSP *)pSMB; /* BB removeme BB */ - /* tcon and ses pointer are checked in smb_init */ if (tcon->ses->server == NULL) return -ECONNABORTED; - pSMB->AndXCommand = 0xFF; /* none */ + pSMB->AndXCommand = 0xFF; /* none */ pSMB->Fid = netfid; pSMB->OffsetLow = cpu_to_le32(offset & 0xFFFFFFFF); - pSMB->OffsetHigh = cpu_to_le32(offset >> 32); + if(wct == 14) + pSMB->OffsetHigh = cpu_to_le32(offset >> 32); + else if((offset >> 32) > 0) /* can not handle this big offset for old */ + return -EIO; pSMB->Reserved = 0xFFFFFFFF; pSMB->WriteMode = 0; pSMB->Remaining = 0; - bytes_sent = (tcon->ses->server->maxBuf - MAX_CIFS_HDR_SIZE) & ~0xFF; - if (bytes_sent > count) - bytes_sent = count; - pSMB->DataLengthHigh = 0; + pSMB->DataOffset = cpu_to_le16(offsetof(struct smb_com_write_req,Data) - 4); - byte_count = bytes_sent + 1 /* pad */ ; - pSMB->DataLengthLow = cpu_to_le16(bytes_sent); - pSMB->DataLengthHigh = 0; - pSMB->hdr.smb_buf_length += byte_count; - pSMB->ByteCount = cpu_to_le16(byte_count); + pSMB->DataLengthLow = cpu_to_le16(count & 0xFFFF); + pSMB->DataLengthHigh = cpu_to_le16(count >> 16); + smb_hdr_len = pSMB->hdr.smb_buf_length + 1; /* hdr + 1 byte pad */ + if(wct == 14) + pSMB->hdr.smb_buf_length += count+1; + else /* wct == 12 */ + pSMB->hdr.smb_buf_length += count+5; /* smb data starts later */ + if(wct == 14) + pSMB->ByteCount = cpu_to_le16(count + 1); + else /* wct == 12 */ /* bigger pad, smaller smb hdr, keep offset ok */ { + struct smb_com_writex_req * pSMBW = + (struct smb_com_writex_req *)pSMB; + pSMBW->ByteCount = cpu_to_le16(count + 5); + } + iov[0].iov_base = pSMB; + iov[0].iov_len = smb_hdr_len + 4; -/* rc = SendReceive2(xid, tcon->ses, (struct smb_hdr *) pSMB, - (struct smb_hdr *) pSMBr, buf, buflen, &bytes_returned, long_op); */ /* BB fixme BB */ + rc = SendReceive2(xid, tcon->ses, iov, n_vec + 1, &bytes_returned, + long_op); + cifs_stats_inc(&tcon->num_writes); if (rc) { - cFYI(1, ("Send error in write2 (large write) = %d", rc)); + cFYI(1, ("Send error Write2 = %d", rc)); *nbytes = 0; - } else - *nbytes = le16_to_cpu(pSMBr->Count); + } else { + WRITE_RSP * pSMBr = (WRITE_RSP *)pSMB; + *nbytes = le16_to_cpu(pSMBr->CountHigh); + *nbytes = (*nbytes) << 16; + *nbytes += le16_to_cpu(pSMBr->Count); + } cifs_small_buf_release(pSMB); @@ -1009,6 +1232,8 @@ int CIFSSMBWrite2(const int xid, struct cifsTconInfo *tcon, return rc; } + + #endif /* CIFS_EXPERIMENTAL */ int @@ -1065,7 +1290,7 @@ CIFSSMBLock(const int xid, struct cifsTconInfo *tcon, rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, timeout); - + cifs_stats_inc(&tcon->num_locks); if (rc) { cFYI(1, ("Send error in Lock = %d", rc)); } @@ -1099,6 +1324,7 @@ CIFSSMBClose(const int xid, struct cifsTconInfo *tcon, int smb_file_id) pSMB->ByteCount = 0; rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); + cifs_stats_inc(&tcon->num_closes); if (rc) { if(rc!=-EINTR) { /* EINTR is expected when user ctl-c to kill app */ @@ -1171,16 +1397,11 @@ renameRetry: rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); + cifs_stats_inc(&tcon->num_renames); if (rc) { cFYI(1, ("Send error in rename = %d", rc)); } -#ifdef CONFIG_CIFS_STATS - else { - atomic_inc(&tcon->num_renames); - } -#endif - cifs_buf_release(pSMB); if (rc == -EAGAIN) @@ -1255,14 +1476,11 @@ int CIFSSMBRenameOpenFile(const int xid,struct cifsTconInfo *pTcon, pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, pTcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); + cifs_stats_inc(&pTcon->num_t2renames); if (rc) { cFYI(1,("Send error in Rename (by file handle) = %d", rc)); } -#ifdef CONFIG_CIFS_STATS - else { - atomic_inc(&pTcon->num_t2renames); - } -#endif + cifs_buf_release(pSMB); /* Note: On -EAGAIN error only caller can retry on handle based calls @@ -1362,7 +1580,7 @@ createSymLinkRetry: if (pSMB->hdr.Flags2 & SMBFLG2_UNICODE) { name_len = - cifs_strtoUCS((wchar_t *) pSMB->FileName, fromName, PATH_MAX + cifs_strtoUCS((__le16 *) pSMB->FileName, fromName, PATH_MAX /* find define for this maxpathcomponent */ , nls_codepage); name_len++; /* trailing null */ @@ -1386,7 +1604,7 @@ createSymLinkRetry: data_offset = (char *) (&pSMB->hdr.Protocol) + offset; if (pSMB->hdr.Flags2 & SMBFLG2_UNICODE) { name_len_target = - cifs_strtoUCS((wchar_t *) data_offset, toName, PATH_MAX + cifs_strtoUCS((__le16 *) data_offset, toName, PATH_MAX /* find define for this maxpathcomponent */ , nls_codepage); name_len_target++; /* trailing null */ @@ -1416,6 +1634,7 @@ createSymLinkRetry: pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); + cifs_stats_inc(&tcon->num_symlinks); if (rc) { cFYI(1, ("Send error in SetPathInfo (create symlink) = %d", @@ -1505,6 +1724,7 @@ createHardLinkRetry: pSMB->ByteCount = cpu_to_le16(byte_count); rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); + cifs_stats_inc(&tcon->num_hardlinks); if (rc) { cFYI(1, ("Send error in SetPathInfo (hard link) = %d", rc)); } @@ -1575,6 +1795,7 @@ winCreateHardLinkRetry: rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); + cifs_stats_inc(&tcon->num_hardlinks); if (rc) { cFYI(1, ("Send error in hard link (NT rename) = %d", rc)); } @@ -1609,7 +1830,7 @@ querySymLinkRetry: if (pSMB->hdr.Flags2 & SMBFLG2_UNICODE) { name_len = - cifs_strtoUCS((wchar_t *) pSMB->FileName, searchName, PATH_MAX + cifs_strtoUCS((__le16 *) pSMB->FileName, searchName, PATH_MAX /* find define for this maxpathcomponent */ , nls_codepage); name_len++; /* trailing null */ @@ -1666,7 +1887,7 @@ querySymLinkRetry: min_t(const int, buflen,count) / 2); /* BB FIXME investigate remapping reserved chars here */ cifs_strfromUCS_le(symlinkinfo, - (wchar_t *) ((char *)&pSMBr->hdr.Protocol + + (__le16 *) ((char *)&pSMBr->hdr.Protocol + data_offset), name_len, nls_codepage); } else { @@ -1757,7 +1978,7 @@ CIFSSMBQueryReparseLinkInfo(const int xid, struct cifsTconInfo *tcon, reparse_buf->TargetNameOffset), min(buflen/2, reparse_buf->TargetNameLen / 2)); cifs_strfromUCS_le(symlinkinfo, - (wchar_t *) (reparse_buf->LinkNamesBuf + + (__le16 *) (reparse_buf->LinkNamesBuf + reparse_buf->TargetNameOffset), name_len, nls_codepage); } else { /* ASCII names */ @@ -1775,8 +1996,7 @@ CIFSSMBQueryReparseLinkInfo(const int xid, struct cifsTconInfo *tcon, } } qreparse_out: - if (pSMB) - cifs_buf_release(pSMB); + cifs_buf_release(pSMB); /* Note: On -EAGAIN error only caller can retry on handle based calls since file handle passed in no longer valid */ @@ -1790,9 +2010,9 @@ qreparse_out: static void cifs_convert_ace(posix_acl_xattr_entry * ace, struct cifs_posix_ace * cifs_ace) { /* u8 cifs fields do not need le conversion */ - ace->e_perm = (__u16)cifs_ace->cifs_e_perm; - ace->e_tag = (__u16)cifs_ace->cifs_e_tag; - ace->e_id = (__u32)le64_to_cpu(cifs_ace->cifs_uid); + ace->e_perm = cpu_to_le16(cifs_ace->cifs_e_perm); + ace->e_tag = cpu_to_le16(cifs_ace->cifs_e_tag); + ace->e_id = cpu_to_le32(le64_to_cpu(cifs_ace->cifs_uid)); /* cFYI(1,("perm %d tag %d id %d",ace->e_perm,ace->e_tag,ace->e_id)); */ return; @@ -1844,7 +2064,7 @@ static int cifs_copy_posix_acl(char * trgt,char * src, const int buflen, } else if(size > buflen) { return -ERANGE; } else /* buffer big enough */ { - local_acl->a_version = POSIX_ACL_XATTR_VERSION; + local_acl->a_version = cpu_to_le32(POSIX_ACL_XATTR_VERSION); for(i = 0;i < count ;i++) { cifs_convert_ace(&local_acl->a_entries[i],pACE); pACE ++; @@ -1858,14 +2078,14 @@ static __u16 convert_ace_to_cifs_ace(struct cifs_posix_ace * cifs_ace, { __u16 rc = 0; /* 0 = ACL converted ok */ - cifs_ace->cifs_e_perm = (__u8)cpu_to_le16(local_ace->e_perm); - cifs_ace->cifs_e_tag = (__u8)cpu_to_le16(local_ace->e_tag); + cifs_ace->cifs_e_perm = le16_to_cpu(local_ace->e_perm); + cifs_ace->cifs_e_tag = le16_to_cpu(local_ace->e_tag); /* BB is there a better way to handle the large uid? */ - if(local_ace->e_id == -1) { + if(local_ace->e_id == cpu_to_le32(-1)) { /* Probably no need to le convert -1 on any arch but can not hurt */ cifs_ace->cifs_uid = cpu_to_le64(-1); } else - cifs_ace->cifs_uid = (__u64)cpu_to_le32(local_ace->e_id); + cifs_ace->cifs_uid = cpu_to_le64(le32_to_cpu(local_ace->e_id)); /*cFYI(1,("perm %d tag %d id %d",ace->e_perm,ace->e_tag,ace->e_id));*/ return rc; } @@ -1885,16 +2105,17 @@ static __u16 ACL_to_cifs_posix(char * parm_data,const char * pACL,const int bufl count = posix_acl_xattr_count((size_t)buflen); cFYI(1,("setting acl with %d entries from buf of length %d and version of %d", - count,buflen,local_acl->a_version)); - if(local_acl->a_version != 2) { - cFYI(1,("unknown POSIX ACL version %d",local_acl->a_version)); + count, buflen, le32_to_cpu(local_acl->a_version))); + if(le32_to_cpu(local_acl->a_version) != 2) { + cFYI(1,("unknown POSIX ACL version %d", + le32_to_cpu(local_acl->a_version))); return 0; } cifs_acl->version = cpu_to_le16(1); if(acl_type == ACL_TYPE_ACCESS) - cifs_acl->access_entry_count = count; + cifs_acl->access_entry_count = cpu_to_le16(count); else if(acl_type == ACL_TYPE_DEFAULT) - cifs_acl->default_entry_count = count; + cifs_acl->default_entry_count = cpu_to_le16(count); else { cFYI(1,("unknown ACL type %d",acl_type)); return 0; @@ -2165,6 +2386,67 @@ GetExtAttrOut: #endif /* CONFIG_POSIX */ +/* Legacy Query Path Information call for lookup to old servers such + as Win9x/WinME */ +int SMBQueryInformation(const int xid, struct cifsTconInfo *tcon, + const unsigned char *searchName, + FILE_ALL_INFO * pFinfo, + const struct nls_table *nls_codepage, int remap) +{ + QUERY_INFORMATION_REQ * pSMB; + QUERY_INFORMATION_RSP * pSMBr; + int rc = 0; + int bytes_returned; + int name_len; + + cFYI(1, ("In SMBQPath path %s", searchName)); +QInfRetry: + rc = smb_init(SMB_COM_QUERY_INFORMATION, 0, tcon, (void **) &pSMB, + (void **) &pSMBr); + if (rc) + return rc; + + if (pSMB->hdr.Flags2 & SMBFLG2_UNICODE) { + name_len = + cifsConvertToUCS((__le16 *) pSMB->FileName, searchName, + PATH_MAX, nls_codepage, remap); + name_len++; /* trailing null */ + name_len *= 2; + } else { + name_len = strnlen(searchName, PATH_MAX); + name_len++; /* trailing null */ + strncpy(pSMB->FileName, searchName, name_len); + } + pSMB->BufferFormat = 0x04; + name_len++; /* account for buffer type byte */ + pSMB->hdr.smb_buf_length += (__u16) name_len; + pSMB->ByteCount = cpu_to_le16(name_len); + + rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, + (struct smb_hdr *) pSMBr, &bytes_returned, 0); + if (rc) { + cFYI(1, ("Send error in QueryInfo = %d", rc)); + } else if (pFinfo) { /* decode response */ + memset(pFinfo, 0, sizeof(FILE_ALL_INFO)); + pFinfo->AllocationSize = + cpu_to_le64(le32_to_cpu(pSMBr->size)); + pFinfo->EndOfFile = pFinfo->AllocationSize; + pFinfo->Attributes = + cpu_to_le32(le16_to_cpu(pSMBr->attr)); + } else + rc = -EIO; /* bad buffer passed in */ + + cifs_buf_release(pSMB); + + if (rc == -EAGAIN) + goto QInfRetry; + + return rc; +} + + + + int CIFSSMBQPathInfo(const int xid, struct cifsTconInfo *tcon, const unsigned char *searchName, @@ -2396,7 +2678,7 @@ findUniqueRetry: if (rc) { cFYI(1, ("Send error in FindFileDirInfo = %d", rc)); } else { /* decode response */ - + cifs_stats_inc(&tcon->num_ffirst); /* BB fill in */ } @@ -2414,7 +2696,7 @@ CIFSFindFirst(const int xid, struct cifsTconInfo *tcon, const char *searchName, const struct nls_table *nls_codepage, __u16 * pnetfid, - struct cifs_search_info * psrch_inf, int remap) + struct cifs_search_info * psrch_inf, int remap, const char dirsep) { /* level 257 SMB_ */ TRANSACTION2_FFIRST_REQ *pSMB = NULL; @@ -2441,7 +2723,7 @@ findFirstRetry: it got remapped to 0xF03A as if it were part of the directory name instead of a wildcard */ name_len *= 2; - pSMB->FileName[name_len] = '\\'; + pSMB->FileName[name_len] = dirsep; pSMB->FileName[name_len+1] = 0; pSMB->FileName[name_len+2] = '*'; pSMB->FileName[name_len+3] = 0; @@ -2455,7 +2737,7 @@ findFirstRetry: if(name_len > buffersize-header) free buffer exit; BB */ strncpy(pSMB->FileName, searchName, name_len); - pSMB->FileName[name_len] = '\\'; + pSMB->FileName[name_len] = dirsep; pSMB->FileName[name_len+1] = '*'; pSMB->FileName[name_len+2] = 0; name_len += 3; @@ -2496,6 +2778,7 @@ findFirstRetry: rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); + cifs_stats_inc(&tcon->num_ffirst); if (rc) {/* BB add logic to retry regular search if Unix search rejected unexpectedly by server */ /* BB Add code to handle unsupported level rc */ @@ -2617,7 +2900,7 @@ int CIFSFindNext(const int xid, struct cifsTconInfo *tcon, rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); - + cifs_stats_inc(&tcon->num_fnext); if (rc) { if (rc == -EBADF) { psrch_inf->endOfSearch = TRUE; @@ -2694,6 +2977,7 @@ CIFSFindClose(const int xid, struct cifsTconInfo *tcon, const __u16 searchHandle if (rc) { cERROR(1, ("Send error in FindClose = %d", rc)); } + cifs_stats_inc(&tcon->num_fclose); cifs_small_buf_release(pSMB); /* Since session is dead, search handle closed on server already */ @@ -2703,7 +2987,6 @@ CIFSFindClose(const int xid, struct cifsTconInfo *tcon, const __u16 searchHandle return rc; } -#ifdef CONFIG_CIFS_EXPERIMENTAL int CIFSGetSrvInodeNumber(const int xid, struct cifsTconInfo *tcon, const unsigned char *searchName, @@ -2797,7 +3080,6 @@ GetInodeNumOut: goto GetInodeNumberRetry; return rc; } -#endif /* CIFS_EXPERIMENTAL */ int CIFSGetDFSRefer(const int xid, struct cifsSesInfo *ses, @@ -2827,7 +3109,10 @@ getDFSRetry: (void **) &pSMBr); if (rc) return rc; - + + /* server pointer checked in called function, + but should never be null here anyway */ + pSMB->hdr.Mid = GetNextMid(ses->server); pSMB->hdr.Tid = ses->ipc_tid; pSMB->hdr.Uid = ses->Suid; if (ses->capabilities & CAP_STATUS32) { @@ -2946,7 +3231,7 @@ getDFSRetry: temp = ((char *)referrals) + le16_to_cpu(referrals->DfsPathOffset); if (pSMBr->hdr.Flags2 & SMBFLG2_UNICODE) { cifs_strfromUCS_le(*targetUNCs, - (wchar_t *) temp, name_len, nls_codepage); + (__le16 *) temp, name_len, nls_codepage); } else { strncpy(*targetUNCs,temp,name_len); } @@ -2968,6 +3253,92 @@ GetDFSRefExit: return rc; } +/* Query File System Info such as free space to old servers such as Win 9x */ +int +SMBOldQFSInfo(const int xid, struct cifsTconInfo *tcon, struct kstatfs *FSData) +{ +/* level 0x01 SMB_QUERY_FILE_SYSTEM_INFO */ + TRANSACTION2_QFSI_REQ *pSMB = NULL; + TRANSACTION2_QFSI_RSP *pSMBr = NULL; + FILE_SYSTEM_ALLOC_INFO *response_data; + int rc = 0; + int bytes_returned = 0; + __u16 params, byte_count; + + cFYI(1, ("OldQFSInfo")); +oldQFSInfoRetry: + rc = smb_init(SMB_COM_TRANSACTION2, 15, tcon, (void **) &pSMB, + (void **) &pSMBr); + if (rc) + return rc; + rc = smb_init(SMB_COM_TRANSACTION2, 15, tcon, (void **) &pSMB, + (void **) &pSMBr); + if (rc) + return rc; + + params = 2; /* level */ + pSMB->TotalDataCount = 0; + pSMB->MaxParameterCount = cpu_to_le16(2); + pSMB->MaxDataCount = cpu_to_le16(1000); + pSMB->MaxSetupCount = 0; + pSMB->Reserved = 0; + pSMB->Flags = 0; + pSMB->Timeout = 0; + pSMB->Reserved2 = 0; + byte_count = params + 1 /* pad */ ; + pSMB->TotalParameterCount = cpu_to_le16(params); + pSMB->ParameterCount = pSMB->TotalParameterCount; + pSMB->ParameterOffset = cpu_to_le16(offsetof( + struct smb_com_transaction2_qfsi_req, InformationLevel) - 4); + pSMB->DataCount = 0; + pSMB->DataOffset = 0; + pSMB->SetupCount = 1; + pSMB->Reserved3 = 0; + pSMB->SubCommand = cpu_to_le16(TRANS2_QUERY_FS_INFORMATION); + pSMB->InformationLevel = cpu_to_le16(SMB_INFO_ALLOCATION); + pSMB->hdr.smb_buf_length += byte_count; + pSMB->ByteCount = cpu_to_le16(byte_count); + + rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, + (struct smb_hdr *) pSMBr, &bytes_returned, 0); + if (rc) { + cFYI(1, ("Send error in QFSInfo = %d", rc)); + } else { /* decode response */ + rc = validate_t2((struct smb_t2_rsp *)pSMBr); + + if (rc || (pSMBr->ByteCount < 18)) + rc = -EIO; /* bad smb */ + else { + __u16 data_offset = le16_to_cpu(pSMBr->t2.DataOffset); + cFYI(1,("qfsinf resp BCC: %d Offset %d", + pSMBr->ByteCount, data_offset)); + + response_data = + (FILE_SYSTEM_ALLOC_INFO *) + (((char *) &pSMBr->hdr.Protocol) + data_offset); + FSData->f_bsize = + le16_to_cpu(response_data->BytesPerSector) * + le32_to_cpu(response_data-> + SectorsPerAllocationUnit); + FSData->f_blocks = + le32_to_cpu(response_data->TotalAllocationUnits); + FSData->f_bfree = FSData->f_bavail = + le32_to_cpu(response_data->FreeAllocationUnits); + cFYI(1, + ("Blocks: %lld Free: %lld Block size %ld", + (unsigned long long)FSData->f_blocks, + (unsigned long long)FSData->f_bfree, + FSData->f_bsize)); + } + } + cifs_buf_release(pSMB); + + if (rc == -EAGAIN) + goto oldQFSInfoRetry; + + return rc; +} + int CIFSSMBQFSInfo(const int xid, struct cifsTconInfo *tcon, struct kstatfs *FSData) { @@ -2989,7 +3360,7 @@ QFSInfoRetry: params = 2; /* level */ pSMB->TotalDataCount = 0; pSMB->MaxParameterCount = cpu_to_le16(2); - pSMB->MaxDataCount = cpu_to_le16(1000); /* BB find exact max SMB PDU from sess structure BB */ + pSMB->MaxDataCount = cpu_to_le16(1000); pSMB->MaxSetupCount = 0; pSMB->Reserved = 0; pSMB->Flags = 0; @@ -3012,17 +3383,14 @@ QFSInfoRetry: rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); if (rc) { - cERROR(1, ("Send error in QFSInfo = %d", rc)); + cFYI(1, ("Send error in QFSInfo = %d", rc)); } else { /* decode response */ rc = validate_t2((struct smb_t2_rsp *)pSMBr); - if (rc || (pSMBr->ByteCount < 24)) /* BB alsO CHEck enough total bytes returned */ + if (rc || (pSMBr->ByteCount < 24)) rc = -EIO; /* bad smb */ else { __u16 data_offset = le16_to_cpu(pSMBr->t2.DataOffset); - cFYI(1, - ("Decoding qfsinfo response. BCC: %d Offset %d", - pSMBr->ByteCount, data_offset)); response_data = (FILE_SYSTEM_INFO @@ -3257,6 +3625,77 @@ QFSUnixRetry: return rc; } +int +CIFSSMBSetFSUnixInfo(const int xid, struct cifsTconInfo *tcon, __u64 cap) +{ +/* level 0x200 SMB_SET_CIFS_UNIX_INFO */ + TRANSACTION2_SETFSI_REQ *pSMB = NULL; + TRANSACTION2_SETFSI_RSP *pSMBr = NULL; + int rc = 0; + int bytes_returned = 0; + __u16 params, param_offset, offset, byte_count; + + cFYI(1, ("In SETFSUnixInfo")); +SETFSUnixRetry: + rc = smb_init(SMB_COM_TRANSACTION2, 15, tcon, (void **) &pSMB, + (void **) &pSMBr); + if (rc) + return rc; + + params = 4; /* 2 bytes zero followed by info level. */ + pSMB->MaxSetupCount = 0; + pSMB->Reserved = 0; + pSMB->Flags = 0; + pSMB->Timeout = 0; + pSMB->Reserved2 = 0; + param_offset = offsetof(struct smb_com_transaction2_setfsi_req, FileNum) - 4; + offset = param_offset + params; + + pSMB->MaxParameterCount = cpu_to_le16(4); + pSMB->MaxDataCount = cpu_to_le16(100); /* BB find exact max SMB PDU from sess structure BB */ + pSMB->SetupCount = 1; + pSMB->Reserved3 = 0; + pSMB->SubCommand = cpu_to_le16(TRANS2_SET_FS_INFORMATION); + byte_count = 1 /* pad */ + params + 12; + + pSMB->DataCount = cpu_to_le16(12); + pSMB->ParameterCount = cpu_to_le16(params); + pSMB->TotalDataCount = pSMB->DataCount; + pSMB->TotalParameterCount = pSMB->ParameterCount; + pSMB->ParameterOffset = cpu_to_le16(param_offset); + pSMB->DataOffset = cpu_to_le16(offset); + + /* Params. */ + pSMB->FileNum = 0; + pSMB->InformationLevel = cpu_to_le16(SMB_SET_CIFS_UNIX_INFO); + + /* Data. */ + pSMB->ClientUnixMajor = cpu_to_le16(CIFS_UNIX_MAJOR_VERSION); + pSMB->ClientUnixMinor = cpu_to_le16(CIFS_UNIX_MINOR_VERSION); + pSMB->ClientUnixCap = cpu_to_le64(cap); + + pSMB->hdr.smb_buf_length += byte_count; + pSMB->ByteCount = cpu_to_le16(byte_count); + + rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, + (struct smb_hdr *) pSMBr, &bytes_returned, 0); + if (rc) { + cERROR(1, ("Send error in SETFSUnixInfo = %d", rc)); + } else { /* decode response */ + rc = validate_t2((struct smb_t2_rsp *)pSMBr); + if (rc) { + rc = -EIO; /* bad smb */ + } + } + cifs_buf_release(pSMB); + + if (rc == -EAGAIN) + goto SETFSUnixRetry; + + return rc; +} + + int CIFSSMBQFSPosixInfo(const int xid, struct cifsTconInfo *tcon, @@ -3321,16 +3760,16 @@ QFSPosixRetry: le64_to_cpu(response_data->TotalBlocks); FSData->f_bfree = le64_to_cpu(response_data->BlocksAvail); - if(response_data->UserBlocksAvail == -1) { + if(response_data->UserBlocksAvail == cpu_to_le64(-1)) { FSData->f_bavail = FSData->f_bfree; } else { FSData->f_bavail = le64_to_cpu(response_data->UserBlocksAvail); } - if(response_data->TotalFileNodes != -1) + if(response_data->TotalFileNodes != cpu_to_le64(-1)) FSData->f_files = le64_to_cpu(response_data->TotalFileNodes); - if(response_data->FreeFileNodes != -1) + if(response_data->FreeFileNodes != cpu_to_le64(-1)) FSData->f_ffree = le64_to_cpu(response_data->FreeFileNodes); } @@ -3376,7 +3815,7 @@ SetEOFRetry: PATH_MAX, nls_codepage, remap); name_len++; /* trailing null */ name_len *= 2; - } else { /* BB improve the check for buffer overruns BB */ + } else { /* BB improve the check for buffer overruns BB */ name_len = strnlen(fileName, PATH_MAX); name_len++; /* trailing null */ strncpy(pSMB->FileName, fileName, name_len); @@ -3384,7 +3823,7 @@ SetEOFRetry: params = 6 + name_len; data_count = sizeof (struct file_end_of_file_info); pSMB->MaxParameterCount = cpu_to_le16(2); - pSMB->MaxDataCount = cpu_to_le16(1000); /* BB find max SMB size from sess */ + pSMB->MaxDataCount = cpu_to_le16(4100); pSMB->MaxSetupCount = 0; pSMB->Reserved = 0; pSMB->Flags = 0; @@ -3766,7 +4205,7 @@ setPermsRetry: PATH_MAX, nls_codepage, remap); name_len++; /* trailing null */ name_len *= 2; - } else { /* BB improve the check for buffer overruns BB */ + } else { /* BB improve the check for buffer overruns BB */ name_len = strnlen(fileName, PATH_MAX); name_len++; /* trailing null */ strncpy(pSMB->FileName, fileName, name_len); @@ -3839,12 +4278,14 @@ setPermsRetry: } int CIFSSMBNotify(const int xid, struct cifsTconInfo *tcon, - const int notify_subdirs, const __u16 netfid, - __u32 filter, const struct nls_table *nls_codepage) + const int notify_subdirs, const __u16 netfid, + __u32 filter, struct file * pfile, int multishot, + const struct nls_table *nls_codepage) { int rc = 0; struct smb_com_transaction_change_notify_req * pSMB = NULL; struct smb_com_transaction_change_notify_rsp * pSMBr = NULL; + struct dir_notify_req *dnotify_req; int bytes_returned; cFYI(1, ("In CIFSSMBNotify for file handle %d",(int)netfid)); @@ -3877,6 +4318,28 @@ int CIFSSMBNotify(const int xid, struct cifsTconInfo *tcon, (struct smb_hdr *) pSMBr, &bytes_returned, -1); if (rc) { cFYI(1, ("Error in Notify = %d", rc)); + } else { + /* Add file to outstanding requests */ + /* BB change to kmem cache alloc */ + dnotify_req = (struct dir_notify_req *) kmalloc( + sizeof(struct dir_notify_req), + GFP_KERNEL); + if(dnotify_req) { + dnotify_req->Pid = pSMB->hdr.Pid; + dnotify_req->PidHigh = pSMB->hdr.PidHigh; + dnotify_req->Mid = pSMB->hdr.Mid; + dnotify_req->Tid = pSMB->hdr.Tid; + dnotify_req->Uid = pSMB->hdr.Uid; + dnotify_req->netfid = netfid; + dnotify_req->pfile = pfile; + dnotify_req->filter = filter; + dnotify_req->multishot = multishot; + spin_lock(&GlobalMid_Lock); + list_add_tail(&dnotify_req->lhead, + &GlobalDnotifyReqList); + spin_unlock(&GlobalMid_Lock); + } else + rc = -ENOMEM; } cifs_buf_release(pSMB); return rc; diff --git a/fs/cifs/cn_cifs.h b/fs/cifs/cn_cifs.h new file mode 100644 index 000000000000..ea59ccac2eb1 --- /dev/null +++ b/fs/cifs/cn_cifs.h @@ -0,0 +1,37 @@ +/* + * fs/cifs/cn_cifs.h + * + * Copyright (c) International Business Machines Corp., 2002 + * Author(s): Steve French (sfrench@us.ibm.com) + * + * This library is free software; you can redistribute it and/or modify + * it under the terms of the GNU Lesser General Public License as published + * by the Free Software Foundation; either version 2.1 of the License, or + * (at your option) any later version. + * + * This library is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See + * the GNU Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public License + * along with this library; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + */ + +#ifndef _CN_CIFS_H +#define _CN_CIFS_H +#ifdef CONFIG_CIFS_UPCALL +#include <linux/types.h> +#include <linux/connector.h> + +struct cifs_upcall { + char signature[4]; /* CIFS */ + enum command { + CIFS_GET_IP = 0x00000001, /* get ip address for hostname */ + CIFS_GET_SECBLOB = 0x00000002, /* get SPNEGO wrapped blob */ + } command; + /* union cifs upcall data follows */ +}; +#endif /* CIFS_UPCALL */ +#endif /* _CN_CIFS_H */ diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 2335f14a1583..c467de857610 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -29,6 +29,8 @@ #include <linux/utsname.h> #include <linux/mempool.h> #include <linux/delay.h> +#include <linux/completion.h> +#include <linux/pagevec.h> #include <asm/uaccess.h> #include <asm/processor.h> #include "cifspdu.h" @@ -40,10 +42,13 @@ #include "ntlmssp.h" #include "nterr.h" #include "rfc1002pdu.h" +#include "cn_cifs.h" #define CIFS_PORT 445 #define RFC1001_PORT 139 +static DECLARE_COMPLETION(cifsd_complete); + extern void SMBencrypt(unsigned char *passwd, unsigned char *c8, unsigned char *p24); extern void SMBNTencrypt(unsigned char *passwd, unsigned char *c8, @@ -60,6 +65,7 @@ struct smb_vol { char *in6_addr; /* ipv6 address as human readable form of in6_addr */ char *iocharset; /* local code page for mapping to and from Unicode */ char source_rfc1001_name[16]; /* netbios name of client */ + char target_rfc1001_name[16]; /* netbios name of server for Win9x/ME */ uid_t linux_uid; gid_t linux_gid; mode_t file_mode; @@ -74,6 +80,10 @@ struct smb_vol { unsigned server_ino:1; /* use inode numbers from server ie UniqueId */ unsigned direct_io:1; unsigned remap:1; /* set to remap seven reserved chars in filenames */ + unsigned posix_paths:1; /* unset to not ask for posix pathnames. */ + unsigned sfu_emul:1; + unsigned nocase; /* request case insensitive filenames */ + unsigned nobrl; /* disable sending byte range locks to srv */ unsigned int rsize; unsigned int wsize; unsigned int sockopt; @@ -82,7 +92,8 @@ struct smb_vol { static int ipv4_connect(struct sockaddr_in *psin_server, struct socket **csocket, - char * netb_name); + char * netb_name, + char * server_netb_name); static int ipv6_connect(struct sockaddr_in6 *psin_server, struct socket **csocket); @@ -175,9 +186,11 @@ cifs_reconnect(struct TCP_Server_Info *server) } else { rc = ipv4_connect(&server->addr.sockAddr, &server->ssocket, - server->workstation_RFC1001_name); + server->workstation_RFC1001_name, + server->server_RFC1001_name); } if(rc) { + cFYI(1,("reconnect error %d",rc)); msleep(3000); } else { atomic_inc(&tcpSesReconnectCount); @@ -293,12 +306,12 @@ static int coalesce_t2(struct smb_hdr * psecond, struct smb_hdr *pTargetSMB) byte_count += total_in_buf2; BCC_LE(pTargetSMB) = cpu_to_le16(byte_count); - byte_count = be32_to_cpu(pTargetSMB->smb_buf_length); + byte_count = pTargetSMB->smb_buf_length; byte_count += total_in_buf2; /* BB also add check that we are not beyond maximum buffer size */ - pTargetSMB->smb_buf_length = cpu_to_be32(byte_count); + pTargetSMB->smb_buf_length = byte_count; if(remaining == total_in_buf2) { cFYI(1,("found the last secondary response")); @@ -323,7 +336,7 @@ cifs_demultiplex_thread(struct TCP_Server_Info *server) struct cifsSesInfo *ses; struct task_struct *task_to_wake = NULL; struct mid_q_entry *mid_entry; - char *temp; + char temp; int isLargeBuf = FALSE; int isMultiRsp; int reconnect; @@ -337,6 +350,7 @@ cifs_demultiplex_thread(struct TCP_Server_Info *server) atomic_inc(&tcpSesAllocCount); length = tcpSesAllocCount.counter; write_unlock(&GlobalSMBSeslock); + complete(&cifsd_complete); if(length > 1) { mempool_resize(cifs_req_poolp, length + cifs_min_rcv, @@ -344,6 +358,8 @@ cifs_demultiplex_thread(struct TCP_Server_Info *server) } while (server->tcpStatus != CifsExiting) { + if (try_to_freeze()) + continue; if (bigbuf == NULL) { bigbuf = cifs_buf_get(); if(bigbuf == NULL) { @@ -422,22 +438,32 @@ cifs_demultiplex_thread(struct TCP_Server_Info *server) continue; } - /* the right amount was read from socket - 4 bytes */ + /* The right amount was read from socket - 4 bytes */ + /* so we can now interpret the length field */ + + /* the first byte big endian of the length field, + is actually not part of the length but the type + with the most common, zero, as regular data */ + temp = *((char *) smb_buffer); + /* Note that FC 1001 length is big endian on the wire, + but we convert it here so it is always manipulated + as host byte order */ pdu_length = ntohl(smb_buffer->smb_buf_length); - cFYI(1,("rfc1002 length(big endian)0x%x)", pdu_length+4)); + smb_buffer->smb_buf_length = pdu_length; - temp = (char *) smb_buffer; - if (temp[0] == (char) RFC1002_SESSION_KEEP_ALIVE) { + cFYI(1,("rfc1002 length 0x%x)", pdu_length+4)); + + if (temp == (char) RFC1002_SESSION_KEEP_ALIVE) { continue; - } else if (temp[0] == (char)RFC1002_POSITIVE_SESSION_RESPONSE) { + } else if (temp == (char)RFC1002_POSITIVE_SESSION_RESPONSE) { cFYI(1,("Good RFC 1002 session rsp")); continue; - } else if (temp[0] == (char)RFC1002_NEGATIVE_SESSION_RESPONSE) { + } else if (temp == (char)RFC1002_NEGATIVE_SESSION_RESPONSE) { /* we get this from Windows 98 instead of an error on SMB negprot response */ cFYI(1,("Negative RFC1002 Session Response Error 0x%x)", - temp[4])); + pdu_length)); if(server->tcpStatus == CifsNew) { /* if nack on negprot (rather than ret of smb negprot error) reconnecting @@ -459,9 +485,10 @@ cifs_demultiplex_thread(struct TCP_Server_Info *server) wake_up(&server->response_q); continue; } - } else if (temp[0] != (char) 0) { + } else if (temp != (char) 0) { cERROR(1,("Unknown RFC 1002 frame")); - cifs_dump_mem(" Received Data: ", temp, length); + cifs_dump_mem(" Received Data: ", (char *)smb_buffer, + length); cifs_reconnect(server); csocket = server->ssocket; continue; @@ -531,7 +558,7 @@ cifs_demultiplex_thread(struct TCP_Server_Info *server) dump_smb(smb_buffer, length); if (checkSMB (smb_buffer, smb_buffer->Mid, total_read+4)) { - cERROR(1, ("Bad SMB Received ")); + cifs_dump_mem("Bad SMB: ", smb_buffer, 48); continue; } @@ -579,6 +606,9 @@ cifs_demultiplex_thread(struct TCP_Server_Info *server) multi_t2_fnd: task_to_wake = mid_entry->tsk; mid_entry->midState = MID_RESPONSE_RECEIVED; +#ifdef CONFIG_CIFS_STATS2 + mid_entry->when_received = jiffies; +#endif break; } } @@ -596,7 +626,8 @@ multi_t2_fnd: } else if ((is_valid_oplock_break(smb_buffer) == FALSE) && (isMultiRsp == FALSE)) { cERROR(1, ("No task to wake, unknown frame rcvd!")); - cifs_dump_mem("Received Data is: ",temp,sizeof(struct smb_hdr)); + cifs_dump_mem("Received Data is: ",(char *)smb_buffer, + sizeof(struct smb_hdr)); } } /* end while !EXITING */ @@ -674,7 +705,7 @@ multi_t2_fnd: msleep(125); } - if (list_empty(&server->pending_mid_q)) { + if (!list_empty(&server->pending_mid_q)) { /* mpx threads have not exited yet give them at least the smb send timeout time for long ops */ /* due to delays on oplock break requests, we need @@ -711,7 +742,7 @@ multi_t2_fnd: GFP_KERNEL); } - msleep(250); + complete_and_exit(&cifsd_complete, 0); return 0; } @@ -735,7 +766,9 @@ cifs_parse_mount_options(char *options, const char *devname,struct smb_vol *vol) toupper(system_utsname.nodename[i]); } vol->source_rfc1001_name[15] = 0; - + /* null target name indicates to use *SMBSERVR default called name + if we end up sending RFC1001 session initialize */ + vol->target_rfc1001_name[0] = 0; vol->linux_uid = current->uid; /* current->euid instead? */ vol->linux_gid = current->gid; vol->dir_mode = S_IRWXUGO; @@ -745,6 +778,9 @@ cifs_parse_mount_options(char *options, const char *devname,struct smb_vol *vol) /* vol->retry default is 0 (i.e. "soft" limited retry not hard retry) */ vol->rw = TRUE; + /* default is always to request posix paths. */ + vol->posix_paths = 1; + if (!options) return 1; @@ -985,7 +1021,31 @@ cifs_parse_mount_options(char *options, const char *devname,struct smb_vol *vol) /* The string has 16th byte zero still from set at top of the function */ if((i==15) && (value[i] != 0)) - printk(KERN_WARNING "CIFS: netbiosname longer than 15 and was truncated.\n"); + printk(KERN_WARNING "CIFS: netbiosname longer than 15 truncated.\n"); + } + } else if (strnicmp(data, "servern", 7) == 0) { + /* servernetbiosname specified override *SMBSERVER */ + if (!value || !*value || (*value == ' ')) { + cFYI(1,("empty server netbiosname specified")); + } else { + /* last byte, type, is 0x20 for servr type */ + memset(vol->target_rfc1001_name,0x20,16); + + for(i=0;i<15;i++) { + /* BB are there cases in which a comma can be + valid in this workstation netbios name (and need + special handling)? */ + + /* user or mount helper must uppercase netbiosname */ + if (value[i]==0) + break; + else + vol->target_rfc1001_name[i] = value[i]; + } + /* The string has 16th byte zero still from + set at top of the function */ + if((i==15) && (value[i] != 0)) + printk(KERN_WARNING "CIFS: server netbiosname longer than 15 truncated.\n"); } } else if (strnicmp(data, "credentials", 4) == 0) { /* ignore */ @@ -1023,6 +1083,27 @@ cifs_parse_mount_options(char *options, const char *devname,struct smb_vol *vol) vol->remap = 1; } else if (strnicmp(data, "nomapchars", 10) == 0) { vol->remap = 0; + } else if (strnicmp(data, "sfu", 3) == 0) { + vol->sfu_emul = 1; + } else if (strnicmp(data, "nosfu", 5) == 0) { + vol->sfu_emul = 0; + } else if (strnicmp(data, "posixpaths", 10) == 0) { + vol->posix_paths = 1; + } else if (strnicmp(data, "noposixpaths", 12) == 0) { + vol->posix_paths = 0; + } else if ((strnicmp(data, "nocase", 6) == 0) || + (strnicmp(data, "ignorecase", 10) == 0)) { + vol->nocase = 1; + } else if (strnicmp(data, "brl", 3) == 0) { + vol->nobrl = 0; + } else if ((strnicmp(data, "nobrl", 5) == 0) || + (strnicmp(data, "nolock", 6) == 0)) { + vol->nobrl = 1; + /* turn off mandatory locking in mode + if remote locking is turned off since the + local vfs will do advisory */ + if(vol->file_mode == (S_IALLUGO & ~(S_ISUID | S_IXGRP))) + vol->file_mode = S_IALLUGO; } else if (strnicmp(data, "setuids", 7) == 0) { vol->setuids = 1; } else if (strnicmp(data, "nosetuids", 9) == 0) { @@ -1185,8 +1266,7 @@ connect_to_dfs_path(int xid, struct cifsSesInfo *pSesInfo, the helper that resolves tcp names, mount to it, try to tcon to it unmount it if fail */ - if(referrals) - kfree(referrals); + kfree(referrals); return rc; } @@ -1242,7 +1322,7 @@ static void rfc1002mangle(char * target,char * source, unsigned int length) static int ipv4_connect(struct sockaddr_in *psin_server, struct socket **csocket, - char * netbios_name) + char * netbios_name, char * target_name) { int rc = 0; int connected = 0; @@ -1307,10 +1387,16 @@ ipv4_connect(struct sockaddr_in *psin_server, struct socket **csocket, /* Eventually check for other socket options to change from the default. sock_setsockopt not used because it expects user space buffer */ + cFYI(1,("sndbuf %d rcvbuf %d rcvtimeo 0x%lx",(*csocket)->sk->sk_sndbuf, + (*csocket)->sk->sk_rcvbuf, (*csocket)->sk->sk_rcvtimeo)); (*csocket)->sk->sk_rcvtimeo = 7 * HZ; + /* make the bufsizes depend on wsize/rsize and max requests */ + if((*csocket)->sk->sk_sndbuf < (200 * 1024)) + (*csocket)->sk->sk_sndbuf = 200 * 1024; + if((*csocket)->sk->sk_rcvbuf < (140 * 1024)) + (*csocket)->sk->sk_rcvbuf = 140 * 1024; /* send RFC1001 sessinit */ - if(psin_server->sin_port == htons(RFC1001_PORT)) { /* some servers require RFC1001 sessinit before sending negprot - BB check reconnection in case where second @@ -1320,8 +1406,14 @@ ipv4_connect(struct sockaddr_in *psin_server, struct socket **csocket, ses_init_buf = kzalloc(sizeof(struct rfc1002_session_packet), GFP_KERNEL); if(ses_init_buf) { ses_init_buf->trailer.session_req.called_len = 32; - rfc1002mangle(ses_init_buf->trailer.session_req.called_name, - DEFAULT_CIFS_CALLED_NAME,16); + if(target_name && (target_name[0] != 0)) { + rfc1002mangle(ses_init_buf->trailer.session_req.called_name, + target_name, 16); + } else { + rfc1002mangle(ses_init_buf->trailer.session_req.called_name, + DEFAULT_CIFS_CALLED_NAME,16); + } + ses_init_buf->trailer.session_req.calling_len = 32; /* calling name ends in null (byte 16) from old smb convention. */ @@ -1443,10 +1535,8 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, memset(&volume_info,0,sizeof(struct smb_vol)); if (cifs_parse_mount_options(mount_data, devname, &volume_info)) { - if(volume_info.UNC) - kfree(volume_info.UNC); - if(volume_info.password) - kfree(volume_info.password); + kfree(volume_info.UNC); + kfree(volume_info.password); FreeXid(xid); return -EINVAL; } @@ -1459,10 +1549,8 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, cifserror("No username specified "); /* In userspace mount helper we can get user name from alternate locations such as env variables and files on disk */ - if(volume_info.UNC) - kfree(volume_info.UNC); - if(volume_info.password) - kfree(volume_info.password); + kfree(volume_info.UNC); + kfree(volume_info.password); FreeXid(xid); return -EINVAL; } @@ -1481,10 +1569,8 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, if(rc <= 0) { /* we failed translating address */ - if(volume_info.UNC) - kfree(volume_info.UNC); - if(volume_info.password) - kfree(volume_info.password); + kfree(volume_info.UNC); + kfree(volume_info.password); FreeXid(xid); return -EINVAL; } @@ -1495,19 +1581,15 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, } else if (volume_info.UNCip){ /* BB using ip addr as server name connect to the DFS root below */ cERROR(1,("Connecting to DFS root not implemented yet")); - if(volume_info.UNC) - kfree(volume_info.UNC); - if(volume_info.password) - kfree(volume_info.password); + kfree(volume_info.UNC); + kfree(volume_info.password); FreeXid(xid); return -EINVAL; } else /* which servers DFS root would we conect to */ { cERROR(1, ("CIFS mount error: No UNC path (e.g. -o unc=//192.168.1.100/public) specified ")); - if(volume_info.UNC) - kfree(volume_info.UNC); - if(volume_info.password) - kfree(volume_info.password); + kfree(volume_info.UNC); + kfree(volume_info.password); FreeXid(xid); return -EINVAL; } @@ -1520,10 +1602,8 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, cifs_sb->local_nls = load_nls(volume_info.iocharset); if(cifs_sb->local_nls == NULL) { cERROR(1,("CIFS mount error: iocharset %s not found",volume_info.iocharset)); - if(volume_info.UNC) - kfree(volume_info.UNC); - if(volume_info.password) - kfree(volume_info.password); + kfree(volume_info.UNC); + kfree(volume_info.password); FreeXid(xid); return -ELIBACC; } @@ -1538,10 +1618,8 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, &sin_server6.sin6_addr, volume_info.username, &srvTcp); else { - if(volume_info.UNC) - kfree(volume_info.UNC); - if(volume_info.password) - kfree(volume_info.password); + kfree(volume_info.UNC); + kfree(volume_info.password); FreeXid(xid); return -EINVAL; } @@ -1554,16 +1632,16 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, sin_server.sin_port = htons(volume_info.port); else sin_server.sin_port = 0; - rc = ipv4_connect(&sin_server,&csocket,volume_info.source_rfc1001_name); + rc = ipv4_connect(&sin_server,&csocket, + volume_info.source_rfc1001_name, + volume_info.target_rfc1001_name); if (rc < 0) { cERROR(1, ("Error connecting to IPv4 socket. Aborting operation")); if(csocket != NULL) sock_release(csocket); - if(volume_info.UNC) - kfree(volume_info.UNC); - if(volume_info.password) - kfree(volume_info.password); + kfree(volume_info.UNC); + kfree(volume_info.password); FreeXid(xid); return rc; } @@ -1572,10 +1650,8 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, if (srvTcp == NULL) { rc = -ENOMEM; sock_release(csocket); - if(volume_info.UNC) - kfree(volume_info.UNC); - if(volume_info.password) - kfree(volume_info.password); + kfree(volume_info.UNC); + kfree(volume_info.password); FreeXid(xid); return rc; } else { @@ -1598,15 +1674,15 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, if(rc < 0) { rc = -ENOMEM; sock_release(csocket); - if(volume_info.UNC) - kfree(volume_info.UNC); - if(volume_info.password) - kfree(volume_info.password); + kfree(volume_info.UNC); + kfree(volume_info.password); FreeXid(xid); return rc; - } else - rc = 0; + } + wait_for_completion(&cifsd_complete); + rc = 0; memcpy(srvTcp->workstation_RFC1001_name, volume_info.source_rfc1001_name,16); + memcpy(srvTcp->server_RFC1001_name, volume_info.target_rfc1001_name,16); srvTcp->sequence_number = 0; } } @@ -1614,8 +1690,7 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, if (existingCifsSes) { pSesInfo = existingCifsSes; cFYI(1, ("Existing smb sess found ")); - if(volume_info.password) - kfree(volume_info.password); + kfree(volume_info.password); /* volume_info.UNC freed at end of function */ } else if (!rc) { cFYI(1, ("Existing smb sess not found ")); @@ -1645,23 +1720,32 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, if(!rc) atomic_inc(&srvTcp->socketUseCount); } else - if(volume_info.password) - kfree(volume_info.password); + kfree(volume_info.password); } /* search for existing tcon to this server share */ if (!rc) { - if((volume_info.rsize) && (volume_info.rsize <= CIFSMaxBufSize)) + if(volume_info.rsize > CIFSMaxBufSize) { + cERROR(1,("rsize %d too large, using MaxBufSize", + volume_info.rsize)); + cifs_sb->rsize = CIFSMaxBufSize; + } else if((volume_info.rsize) && (volume_info.rsize <= CIFSMaxBufSize)) cifs_sb->rsize = volume_info.rsize; - else - cifs_sb->rsize = srvTcp->maxBuf - MAX_CIFS_HDR_SIZE; /* default */ - if((volume_info.wsize) && (volume_info.wsize <= CIFSMaxBufSize)) + else /* default */ + cifs_sb->rsize = CIFSMaxBufSize; + + if(volume_info.wsize > PAGEVEC_SIZE * PAGE_CACHE_SIZE) { + cERROR(1,("wsize %d too large using 4096 instead", + volume_info.wsize)); + cifs_sb->wsize = 4096; + } else if(volume_info.wsize) cifs_sb->wsize = volume_info.wsize; else cifs_sb->wsize = CIFSMaxBufSize; /* default */ if(cifs_sb->rsize < PAGE_CACHE_SIZE) { - cifs_sb->rsize = PAGE_CACHE_SIZE; - cERROR(1,("Attempt to set readsize for mount to less than one page (4096)")); + cifs_sb->rsize = PAGE_CACHE_SIZE; + /* Windows ME does this */ + cFYI(1,("Attempt to set readsize for mount to less than one page (4096)")); } cifs_sb->mnt_uid = volume_info.linux_uid; cifs_sb->mnt_gid = volume_info.linux_gid; @@ -1679,8 +1763,13 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_MAP_SPECIAL_CHR; if(volume_info.no_xattr) cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_NO_XATTR; + if(volume_info.sfu_emul) + cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_UNX_EMUL; + if(volume_info.nobrl) + cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_NO_BRL; + if(volume_info.direct_io) { - cERROR(1,("mounting share using direct i/o")); + cFYI(1,("mounting share using direct i/o")); cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_DIRECT_IO; } @@ -1694,6 +1783,7 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, to the same server share the last value passed in for the retry flag is used */ tcon->retry = volume_info.retry; + tcon->nocase = volume_info.nocase; } else { tcon = tconInfoAlloc(); if (tcon == NULL) @@ -1709,8 +1799,7 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, "", cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); - if(volume_info.UNC) - kfree(volume_info.UNC); + kfree(volume_info.UNC); FreeXid(xid); return -ENODEV; } else { @@ -1722,6 +1811,7 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, if (!rc) { atomic_inc(&pSesInfo->inUse); tcon->retry = volume_info.retry; + tcon->nocase = volume_info.nocase; } } } @@ -1743,8 +1833,10 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, spin_lock(&GlobalMid_Lock); srvTcp->tcpStatus = CifsExiting; spin_unlock(&GlobalMid_Lock); - if(srvTcp->tsk) + if(srvTcp->tsk) { send_sig(SIGKILL,srvTcp->tsk,1); + wait_for_completion(&cifsd_complete); + } } /* If find_unc succeeded then rc == 0 so we can not end */ if (tcon) /* up accidently freeing someone elses tcon struct */ @@ -1757,8 +1849,10 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, temp_rc = CIFSSMBLogoff(xid, pSesInfo); /* if the socketUseCount is now zero */ if((temp_rc == -ESHUTDOWN) && - (pSesInfo->server->tsk)) + (pSesInfo->server->tsk)) { send_sig(SIGKILL,pSesInfo->server->tsk,1); + wait_for_completion(&cifsd_complete); + } } else cFYI(1, ("No session or bad tcon")); sesInfoFree(pSesInfo); @@ -1781,16 +1875,34 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, cFYI(1,("server negotiated posix acl support")); sb->s_flags |= MS_POSIXACL; } + + /* Try and negotiate POSIX pathnames if we can. */ + if (volume_info.posix_paths && (CIFS_UNIX_POSIX_PATHNAMES_CAP & + le64_to_cpu(tcon->fsUnixInfo.Capability))) { + if (!CIFSSMBSetFSUnixInfo(xid, tcon, CIFS_UNIX_POSIX_PATHNAMES_CAP)) { + cFYI(1,("negotiated posix pathnames support")); + cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_POSIX_PATHS; + } else { + cFYI(1,("posix pathnames support requested but not supported")); + } + } } } + if (!(tcon->ses->capabilities & CAP_LARGE_WRITE_X)) + cifs_sb->wsize = min(cifs_sb->wsize, + (tcon->ses->server->maxBuf - + MAX_CIFS_HDR_SIZE)); + if (!(tcon->ses->capabilities & CAP_LARGE_READ_X)) + cifs_sb->rsize = min(cifs_sb->rsize, + (tcon->ses->server->maxBuf - + MAX_CIFS_HDR_SIZE)); } /* volume_info.password is freed above when existing session found (in which case it is not needed anymore) but when new sesion is created the password ptr is put in the new session structure (in which case the password will be freed at unmount time) */ - if(volume_info.UNC) - kfree(volume_info.UNC); + kfree(volume_info.UNC); FreeXid(xid); return rc; } @@ -1830,6 +1942,7 @@ CIFSSessSetup(unsigned int xid, struct cifsSesInfo *ses, header_assemble(smb_buffer, SMB_COM_SESSION_SETUP_ANDX, NULL /* no tCon exists yet */ , 13 /* wct */ ); + smb_buffer->Mid = GetNextMid(ses->server); pSMB->req_no_secext.AndXCommand = 0xFF; pSMB->req_no_secext.MaxBufferSize = cpu_to_le16(ses->server->maxBuf); pSMB->req_no_secext.MaxMpxCount = cpu_to_le16(ses->server->maxReq); @@ -1873,32 +1986,32 @@ CIFSSessSetup(unsigned int xid, struct cifsSesInfo *ses, bytes_returned = 0; /* skill null user */ else bytes_returned = - cifs_strtoUCS((wchar_t *) bcc_ptr, user, 100, + cifs_strtoUCS((__le16 *) bcc_ptr, user, 100, nls_codepage); /* convert number of 16 bit words to bytes */ bcc_ptr += 2 * bytes_returned; bcc_ptr += 2; /* trailing null */ if (domain == NULL) bytes_returned = - cifs_strtoUCS((wchar_t *) bcc_ptr, + cifs_strtoUCS((__le16 *) bcc_ptr, "CIFS_LINUX_DOM", 32, nls_codepage); else bytes_returned = - cifs_strtoUCS((wchar_t *) bcc_ptr, domain, 64, + cifs_strtoUCS((__le16 *) bcc_ptr, domain, 64, nls_codepage); bcc_ptr += 2 * bytes_returned; bcc_ptr += 2; bytes_returned = - cifs_strtoUCS((wchar_t *) bcc_ptr, "Linux version ", + cifs_strtoUCS((__le16 *) bcc_ptr, "Linux version ", 32, nls_codepage); bcc_ptr += 2 * bytes_returned; bytes_returned = - cifs_strtoUCS((wchar_t *) bcc_ptr, system_utsname.release, + cifs_strtoUCS((__le16 *) bcc_ptr, system_utsname.release, 32, nls_codepage); bcc_ptr += 2 * bytes_returned; bcc_ptr += 2; bytes_returned = - cifs_strtoUCS((wchar_t *) bcc_ptr, CIFS_NETWORK_OPSYS, + cifs_strtoUCS((__le16 *) bcc_ptr, CIFS_NETWORK_OPSYS, 64, nls_codepage); bcc_ptr += 2 * bytes_returned; bcc_ptr += 2; @@ -1968,7 +2081,7 @@ CIFSSessSetup(unsigned int xid, struct cifsSesInfo *ses, if(ses->serverOS == NULL) goto sesssetup_nomem; cifs_strfromUCS_le(ses->serverOS, - (wchar_t *)bcc_ptr, len,nls_codepage); + (__le16 *)bcc_ptr, len,nls_codepage); bcc_ptr += 2 * (len + 1); remaining_words -= len + 1; ses->serverOS[2 * len] = 0; @@ -1980,7 +2093,7 @@ CIFSSessSetup(unsigned int xid, struct cifsSesInfo *ses, if(ses->serverNOS == NULL) goto sesssetup_nomem; cifs_strfromUCS_le(ses->serverNOS, - (wchar_t *)bcc_ptr,len,nls_codepage); + (__le16 *)bcc_ptr,len,nls_codepage); bcc_ptr += 2 * (len + 1); ses->serverNOS[2 * len] = 0; ses->serverNOS[1 + (2 * len)] = 0; @@ -1998,7 +2111,7 @@ CIFSSessSetup(unsigned int xid, struct cifsSesInfo *ses, if(ses->serverDomain == NULL) goto sesssetup_nomem; cifs_strfromUCS_le(ses->serverDomain, - (wchar_t *)bcc_ptr,len,nls_codepage); + (__le16 *)bcc_ptr,len,nls_codepage); bcc_ptr += 2 * (len + 1); ses->serverDomain[2*len] = 0; ses->serverDomain[1+(2*len)] = 0; @@ -2105,6 +2218,8 @@ CIFSSpnegoSessSetup(unsigned int xid, struct cifsSesInfo *ses, /* send SMBsessionSetup here */ header_assemble(smb_buffer, SMB_COM_SESSION_SETUP_ANDX, NULL /* no tCon exists yet */ , 12 /* wct */ ); + + smb_buffer->Mid = GetNextMid(ses->server); pSMB->req.hdr.Flags2 |= SMBFLG2_EXT_SEC; pSMB->req.AndXCommand = 0xFF; pSMB->req.MaxBufferSize = cpu_to_le16(ses->server->maxBuf); @@ -2140,30 +2255,30 @@ CIFSSpnegoSessSetup(unsigned int xid, struct cifsSesInfo *ses, bcc_ptr++; } bytes_returned = - cifs_strtoUCS((wchar_t *) bcc_ptr, user, 100, nls_codepage); + cifs_strtoUCS((__le16 *) bcc_ptr, user, 100, nls_codepage); bcc_ptr += 2 * bytes_returned; /* convert num of 16 bit words to bytes */ bcc_ptr += 2; /* trailing null */ if (domain == NULL) bytes_returned = - cifs_strtoUCS((wchar_t *) bcc_ptr, + cifs_strtoUCS((__le16 *) bcc_ptr, "CIFS_LINUX_DOM", 32, nls_codepage); else bytes_returned = - cifs_strtoUCS((wchar_t *) bcc_ptr, domain, 64, + cifs_strtoUCS((__le16 *) bcc_ptr, domain, 64, nls_codepage); bcc_ptr += 2 * bytes_returned; bcc_ptr += 2; bytes_returned = - cifs_strtoUCS((wchar_t *) bcc_ptr, "Linux version ", + cifs_strtoUCS((__le16 *) bcc_ptr, "Linux version ", 32, nls_codepage); bcc_ptr += 2 * bytes_returned; bytes_returned = - cifs_strtoUCS((wchar_t *) bcc_ptr, system_utsname.release, 32, + cifs_strtoUCS((__le16 *) bcc_ptr, system_utsname.release, 32, nls_codepage); bcc_ptr += 2 * bytes_returned; bcc_ptr += 2; bytes_returned = - cifs_strtoUCS((wchar_t *) bcc_ptr, CIFS_NETWORK_OPSYS, + cifs_strtoUCS((__le16 *) bcc_ptr, CIFS_NETWORK_OPSYS, 64, nls_codepage); bcc_ptr += 2 * bytes_returned; bcc_ptr += 2; @@ -2242,7 +2357,7 @@ CIFSSpnegoSessSetup(unsigned int xid, struct cifsSesInfo *ses, ses->serverOS = kzalloc(2 * (len + 1), GFP_KERNEL); cifs_strfromUCS_le(ses->serverOS, - (wchar_t *) + (__le16 *) bcc_ptr, len, nls_codepage); bcc_ptr += 2 * (len + 1); @@ -2257,7 +2372,7 @@ CIFSSpnegoSessSetup(unsigned int xid, struct cifsSesInfo *ses, kzalloc(2 * (len + 1), GFP_KERNEL); cifs_strfromUCS_le(ses->serverNOS, - (wchar_t *)bcc_ptr, + (__le16 *)bcc_ptr, len, nls_codepage); bcc_ptr += 2 * (len + 1); @@ -2269,9 +2384,8 @@ CIFSSpnegoSessSetup(unsigned int xid, struct cifsSesInfo *ses, /* last string is not always null terminated (for e.g. for Windows XP & 2000) */ ses->serverDomain = kzalloc(2*(len+1),GFP_KERNEL); cifs_strfromUCS_le(ses->serverDomain, - (wchar_t *)bcc_ptr, - len, - nls_codepage); + (__le16 *)bcc_ptr, + len, nls_codepage); bcc_ptr += 2*(len+1); ses->serverDomain[2*len] = 0; ses->serverDomain[1+(2*len)] = 0; @@ -2371,6 +2485,8 @@ CIFSNTLMSSPNegotiateSessSetup(unsigned int xid, /* send SMBsessionSetup here */ header_assemble(smb_buffer, SMB_COM_SESSION_SETUP_ANDX, NULL /* no tCon exists yet */ , 12 /* wct */ ); + + smb_buffer->Mid = GetNextMid(ses->server); pSMB->req.hdr.Flags2 |= SMBFLG2_EXT_SEC; pSMB->req.hdr.Flags |= (SMBFLG_CASELESS | SMBFLG_CANONICAL_PATH_FORMAT); @@ -2443,16 +2559,16 @@ CIFSNTLMSSPNegotiateSessSetup(unsigned int xid, } bytes_returned = - cifs_strtoUCS((wchar_t *) bcc_ptr, "Linux version ", + cifs_strtoUCS((__le16 *) bcc_ptr, "Linux version ", 32, nls_codepage); bcc_ptr += 2 * bytes_returned; bytes_returned = - cifs_strtoUCS((wchar_t *) bcc_ptr, system_utsname.release, 32, + cifs_strtoUCS((__le16 *) bcc_ptr, system_utsname.release, 32, nls_codepage); bcc_ptr += 2 * bytes_returned; bcc_ptr += 2; /* null terminate Linux version */ bytes_returned = - cifs_strtoUCS((wchar_t *) bcc_ptr, CIFS_NETWORK_OPSYS, + cifs_strtoUCS((__le16 *) bcc_ptr, CIFS_NETWORK_OPSYS, 64, nls_codepage); bcc_ptr += 2 * bytes_returned; *(bcc_ptr + 1) = 0; @@ -2556,7 +2672,7 @@ CIFSNTLMSSPNegotiateSessSetup(unsigned int xid, ses->serverOS = kzalloc(2 * (len + 1), GFP_KERNEL); cifs_strfromUCS_le(ses->serverOS, - (wchar_t *) + (__le16 *) bcc_ptr, len, nls_codepage); bcc_ptr += 2 * (len + 1); @@ -2573,7 +2689,7 @@ CIFSNTLMSSPNegotiateSessSetup(unsigned int xid, GFP_KERNEL); cifs_strfromUCS_le(ses-> serverNOS, - (wchar_t *) + (__le16 *) bcc_ptr, len, nls_codepage); @@ -2591,23 +2707,15 @@ CIFSNTLMSSPNegotiateSessSetup(unsigned int xid, 1), GFP_KERNEL); cifs_strfromUCS_le - (ses-> - serverDomain, - (wchar_t *) - bcc_ptr, len, - nls_codepage); + (ses->serverDomain, + (__le16 *)bcc_ptr, + len, nls_codepage); bcc_ptr += 2 * (len + 1); - ses-> - serverDomain[2 - * len] + ses->serverDomain[2*len] = 0; - ses-> - serverDomain[1 - + - (2 - * - len)] + ses->serverDomain + [1 + (2 * len)] = 0; } /* else no more room so create dummy domain string */ else @@ -2713,6 +2821,8 @@ CIFSNTLMSSPAuthSessSetup(unsigned int xid, struct cifsSesInfo *ses, /* send SMBsessionSetup here */ header_assemble(smb_buffer, SMB_COM_SESSION_SETUP_ANDX, NULL /* no tCon exists yet */ , 12 /* wct */ ); + + smb_buffer->Mid = GetNextMid(ses->server); pSMB->req.hdr.Flags |= (SMBFLG_CASELESS | SMBFLG_CANONICAL_PATH_FORMAT); pSMB->req.hdr.Flags2 |= SMBFLG2_EXT_SEC; pSMB->req.AndXCommand = 0xFF; @@ -2784,7 +2894,7 @@ CIFSNTLMSSPAuthSessSetup(unsigned int xid, struct cifsSesInfo *ses, SecurityBlob->DomainName.MaximumLength = 0; } else { __u16 len = - cifs_strtoUCS((wchar_t *) bcc_ptr, domain, 64, + cifs_strtoUCS((__le16 *) bcc_ptr, domain, 64, nls_codepage); len *= 2; SecurityBlob->DomainName.MaximumLength = @@ -2802,7 +2912,7 @@ CIFSNTLMSSPAuthSessSetup(unsigned int xid, struct cifsSesInfo *ses, SecurityBlob->UserName.MaximumLength = 0; } else { __u16 len = - cifs_strtoUCS((wchar_t *) bcc_ptr, user, 64, + cifs_strtoUCS((__le16 *) bcc_ptr, user, 64, nls_codepage); len *= 2; SecurityBlob->UserName.MaximumLength = @@ -2815,7 +2925,7 @@ CIFSNTLMSSPAuthSessSetup(unsigned int xid, struct cifsSesInfo *ses, cpu_to_le16(len); } - /* SecurityBlob->WorkstationName.Length = cifs_strtoUCS((wchar_t *) bcc_ptr, "AMACHINE",64, nls_codepage); + /* SecurityBlob->WorkstationName.Length = cifs_strtoUCS((__le16 *) bcc_ptr, "AMACHINE",64, nls_codepage); SecurityBlob->WorkstationName.Length *= 2; SecurityBlob->WorkstationName.MaximumLength = cpu_to_le16(SecurityBlob->WorkstationName.Length); SecurityBlob->WorkstationName.Buffer = cpu_to_le32(SecurityBlobLength); @@ -2828,16 +2938,16 @@ CIFSNTLMSSPAuthSessSetup(unsigned int xid, struct cifsSesInfo *ses, bcc_ptr++; } bytes_returned = - cifs_strtoUCS((wchar_t *) bcc_ptr, "Linux version ", + cifs_strtoUCS((__le16 *) bcc_ptr, "Linux version ", 32, nls_codepage); bcc_ptr += 2 * bytes_returned; bytes_returned = - cifs_strtoUCS((wchar_t *) bcc_ptr, system_utsname.release, 32, + cifs_strtoUCS((__le16 *) bcc_ptr, system_utsname.release, 32, nls_codepage); bcc_ptr += 2 * bytes_returned; bcc_ptr += 2; /* null term version string */ bytes_returned = - cifs_strtoUCS((wchar_t *) bcc_ptr, CIFS_NETWORK_OPSYS, + cifs_strtoUCS((__le16 *) bcc_ptr, CIFS_NETWORK_OPSYS, 64, nls_codepage); bcc_ptr += 2 * bytes_returned; *(bcc_ptr + 1) = 0; @@ -2950,7 +3060,7 @@ CIFSNTLMSSPAuthSessSetup(unsigned int xid, struct cifsSesInfo *ses, ses->serverOS = kzalloc(2 * (len + 1), GFP_KERNEL); cifs_strfromUCS_le(ses->serverOS, - (wchar_t *) + (__le16 *) bcc_ptr, len, nls_codepage); bcc_ptr += 2 * (len + 1); @@ -2967,7 +3077,7 @@ CIFSNTLMSSPAuthSessSetup(unsigned int xid, struct cifsSesInfo *ses, GFP_KERNEL); cifs_strfromUCS_le(ses-> serverNOS, - (wchar_t *) + (__le16 *) bcc_ptr, len, nls_codepage); @@ -2986,7 +3096,7 @@ CIFSNTLMSSPAuthSessSetup(unsigned int xid, struct cifsSesInfo *ses, cifs_strfromUCS_le (ses-> serverDomain, - (wchar_t *) + (__le16 *) bcc_ptr, len, nls_codepage); bcc_ptr += @@ -3084,6 +3194,8 @@ CIFSTCon(unsigned int xid, struct cifsSesInfo *ses, header_assemble(smb_buffer, SMB_COM_TREE_CONNECT_ANDX, NULL /*no tid */ , 4 /*wct */ ); + + smb_buffer->Mid = GetNextMid(ses->server); smb_buffer->Uid = ses->Suid; pSMB = (TCONX_REQ *) smb_buffer; pSMBr = (TCONX_RSP *) smb_buffer_response; @@ -3106,7 +3218,7 @@ CIFSTCon(unsigned int xid, struct cifsSesInfo *ses, if (ses->capabilities & CAP_UNICODE) { smb_buffer->Flags2 |= SMBFLG2_UNICODE; length = - cifs_strtoUCS((wchar_t *) bcc_ptr, tree, 100, nls_codepage); + cifs_strtoUCS((__le16 *) bcc_ptr, tree, 100, nls_codepage); bcc_ptr += 2 * length; /* convert num of 16 bit words to bytes */ bcc_ptr += 2; /* skip trailing null */ } else { /* ASCII */ @@ -3138,12 +3250,11 @@ CIFSTCon(unsigned int xid, struct cifsSesInfo *ses, if ((bcc_ptr + (2 * length)) - pByteArea(smb_buffer_response) <= BCC(smb_buffer_response)) { - if(tcon->nativeFileSystem) - kfree(tcon->nativeFileSystem); + kfree(tcon->nativeFileSystem); tcon->nativeFileSystem = kzalloc(length + 2, GFP_KERNEL); cifs_strfromUCS_le(tcon->nativeFileSystem, - (wchar_t *) bcc_ptr, + (__le16 *) bcc_ptr, length, nls_codepage); bcc_ptr += 2 * length; bcc_ptr[0] = 0; /* null terminate the string */ @@ -3156,8 +3267,7 @@ CIFSTCon(unsigned int xid, struct cifsSesInfo *ses, if ((bcc_ptr + length) - pByteArea(smb_buffer_response) <= BCC(smb_buffer_response)) { - if(tcon->nativeFileSystem) - kfree(tcon->nativeFileSystem); + kfree(tcon->nativeFileSystem); tcon->nativeFileSystem = kzalloc(length + 1, GFP_KERNEL); strncpy(tcon->nativeFileSystem, bcc_ptr, @@ -3205,8 +3315,10 @@ cifs_umount(struct super_block *sb, struct cifs_sb_info *cifs_sb) return 0; } else if (rc == -ESHUTDOWN) { cFYI(1,("Waking up socket by sending it signal")); - if(cifsd_task) + if(cifsd_task) { send_sig(SIGKILL,cifsd_task,1); + wait_for_completion(&cifsd_complete); + } rc = 0; } /* else - we have an smb session left on this socket do not kill cifsd */ diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c index d335269bd91c..32cc96cafa3e 100644 --- a/fs/cifs/dir.c +++ b/fs/cifs/dir.c @@ -48,6 +48,7 @@ build_path_from_dentry(struct dentry *direntry) struct dentry *temp; int namelen = 0; char *full_path; + char dirsep = CIFS_DIR_SEP(CIFS_SB(direntry->d_sb)); if(direntry == NULL) return NULL; /* not much we can do if dentry is freed and @@ -74,7 +75,7 @@ cifs_bp_rename_retry: if (namelen < 0) { break; } else { - full_path[namelen] = '\\'; + full_path[namelen] = dirsep; strncpy(full_path + namelen + 1, temp->d_name.name, temp->d_name.len); cFYI(0, (" name: %s ", full_path + namelen)); @@ -183,6 +184,13 @@ cifs_create(struct inode *inode, struct dentry *direntry, int mode, desiredAccess, CREATE_NOT_DIR, &fileHandle, &oplock, buf, cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); + if(rc == -EIO) { + /* old server, retry the open legacy style */ + rc = SMBLegacyOpen(xid, pTcon, full_path, disposition, + desiredAccess, CREATE_NOT_DIR, + &fileHandle, &oplock, buf, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); + } if (rc) { cFYI(1, ("cifs_create returned 0x%x ", rc)); } else { @@ -208,7 +216,7 @@ cifs_create(struct inode *inode, struct dentry *direntry, int mode, CIFS_MOUNT_MAP_SPECIAL_CHR); } else { - /* BB implement via Windows security descriptors */ + /* BB implement mode setting via Windows security descriptors */ /* eg CIFSSMBWinSetPerms(xid,pTcon,full_path,mode,-1,-1,local_nls);*/ /* could set r/o dos attribute if mode & 0222 == 0 */ } @@ -220,15 +228,26 @@ cifs_create(struct inode *inode, struct dentry *direntry, int mode, else { rc = cifs_get_inode_info(&newinode, full_path, buf, inode->i_sb,xid); - if(newinode) + if(newinode) { newinode->i_mode = mode; + if((oplock & CIFS_CREATE_ACTION) && + (cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_SET_UID)) { + newinode->i_uid = current->fsuid; + newinode->i_gid = current->fsgid; + } + } } if (rc != 0) { - cFYI(1,("Create worked but get_inode_info failed with rc = %d", + cFYI(1, + ("Create worked but get_inode_info failed rc = %d", rc)); } else { - direntry->d_op = &cifs_dentry_ops; + if (pTcon->nocase) + direntry->d_op = &cifs_ci_dentry_ops; + else + direntry->d_op = &cifs_dentry_ops; d_instantiate(direntry, newinode); } if((nd->flags & LOOKUP_OPEN) == FALSE) { @@ -280,7 +299,8 @@ cifs_create_out: return rc; } -int cifs_mknod(struct inode *inode, struct dentry *direntry, int mode, dev_t device_number) +int cifs_mknod(struct inode *inode, struct dentry *direntry, int mode, + dev_t device_number) { int rc = -EPERM; int xid; @@ -302,8 +322,7 @@ int cifs_mknod(struct inode *inode, struct dentry *direntry, int mode, dev_t dev up(&direntry->d_sb->s_vfs_rename_sem); if(full_path == NULL) rc = -ENOMEM; - - if (full_path && (pTcon->ses->capabilities & CAP_UNIX)) { + else if (pTcon->ses->capabilities & CAP_UNIX) { if(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SET_UID) { rc = CIFSSMBUnixSetPerms(xid, pTcon, full_path, mode,(__u64)current->euid,(__u64)current->egid, @@ -321,10 +340,76 @@ int cifs_mknod(struct inode *inode, struct dentry *direntry, int mode, dev_t dev if(!rc) { rc = cifs_get_inode_info_unix(&newinode, full_path, inode->i_sb,xid); - direntry->d_op = &cifs_dentry_ops; + if (pTcon->nocase) + direntry->d_op = &cifs_ci_dentry_ops; + else + direntry->d_op = &cifs_dentry_ops; if(rc == 0) d_instantiate(direntry, newinode); } + } else { + if(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UNX_EMUL) { + int oplock = 0; + u16 fileHandle; + FILE_ALL_INFO * buf; + + cFYI(1,("sfu compat create special file")); + + buf = kmalloc(sizeof(FILE_ALL_INFO),GFP_KERNEL); + if(buf == NULL) { + kfree(full_path); + FreeXid(xid); + return -ENOMEM; + } + + rc = CIFSSMBOpen(xid, pTcon, full_path, + FILE_CREATE, /* fail if exists */ + GENERIC_WRITE /* BB would + WRITE_OWNER | WRITE_DAC be better? */, + /* Create a file and set the + file attribute to SYSTEM */ + CREATE_NOT_DIR | CREATE_OPTION_SPECIAL, + &fileHandle, &oplock, buf, + cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); + + if(!rc) { + /* BB Do not bother to decode buf since no + local inode yet to put timestamps in, + but we can reuse it safely */ + int bytes_written; + struct win_dev *pdev; + pdev = (struct win_dev *)buf; + if(S_ISCHR(mode)) { + memcpy(pdev->type, "IntxCHR", 8); + pdev->major = + cpu_to_le64(MAJOR(device_number)); + pdev->minor = + cpu_to_le64(MINOR(device_number)); + rc = CIFSSMBWrite(xid, pTcon, + fileHandle, + sizeof(struct win_dev), + 0, &bytes_written, (char *)pdev, + NULL, 0); + } else if(S_ISBLK(mode)) { + memcpy(pdev->type, "IntxBLK", 8); + pdev->major = + cpu_to_le64(MAJOR(device_number)); + pdev->minor = + cpu_to_le64(MINOR(device_number)); + rc = CIFSSMBWrite(xid, pTcon, + fileHandle, + sizeof(struct win_dev), + 0, &bytes_written, (char *)pdev, + NULL, 0); + } /* else if(S_ISFIFO */ + CIFSSMBClose(xid, pTcon, fileHandle); + d_drop(direntry); + } + kfree(buf); + /* add code here to set EAs */ + } } kfree(full_path); @@ -381,15 +466,26 @@ cifs_lookup(struct inode *parent_dir_inode, struct dentry *direntry, struct name parent_dir_inode->i_sb,xid); if ((rc == 0) && (newInode != NULL)) { - direntry->d_op = &cifs_dentry_ops; + if (pTcon->nocase) + direntry->d_op = &cifs_ci_dentry_ops; + else + direntry->d_op = &cifs_dentry_ops; d_add(direntry, newInode); - /* since paths are not looked up by component - the parent directories are presumed to be good here */ + /* since paths are not looked up by component - the parent + directories are presumed to be good here */ renew_parental_timestamps(direntry); } else if (rc == -ENOENT) { rc = 0; + direntry->d_time = jiffies; + if (pTcon->nocase) + direntry->d_op = &cifs_ci_dentry_ops; + else + direntry->d_op = &cifs_dentry_ops; d_add(direntry, NULL); + /* if it was once a directory (but how can we tell?) we could do + shrink_dcache_parent(direntry); */ } else { cERROR(1,("Error 0x%x on cifs_get_inode_info in lookup of %s", rc,full_path)); @@ -408,21 +504,20 @@ cifs_d_revalidate(struct dentry *direntry, struct nameidata *nd) { int isValid = 1; -/* lock_kernel(); *//* surely we do not want to lock the kernel for a whole network round trip which could take seconds */ - if (direntry->d_inode) { if (cifs_revalidate(direntry)) { - /* unlock_kernel(); */ return 0; } } else { - cFYI(1, - ("In cifs_d_revalidate with no inode but name = %s and dentry 0x%p", - direntry->d_name.name, direntry)); + cFYI(1, ("neg dentry 0x%p name = %s", + direntry, direntry->d_name.name)); + if(time_after(jiffies, direntry->d_time + HZ) || + !lookupCacheEnabled) { + d_drop(direntry); + isValid = 0; + } } -/* unlock_kernel(); */ - return isValid; } @@ -440,3 +535,42 @@ struct dentry_operations cifs_dentry_ops = { /* d_delete: cifs_d_delete, *//* not needed except for debugging */ /* no need for d_hash, d_compare, d_release, d_iput ... yet. BB confirm this BB */ }; + +static int cifs_ci_hash(struct dentry *dentry, struct qstr *q) +{ + struct nls_table *codepage = CIFS_SB(dentry->d_inode->i_sb)->local_nls; + unsigned long hash; + int i; + + hash = init_name_hash(); + for (i = 0; i < q->len; i++) + hash = partial_name_hash(nls_tolower(codepage, q->name[i]), + hash); + q->hash = end_name_hash(hash); + + return 0; +} + +static int cifs_ci_compare(struct dentry *dentry, struct qstr *a, + struct qstr *b) +{ + struct nls_table *codepage = CIFS_SB(dentry->d_inode->i_sb)->local_nls; + + if ((a->len == b->len) && + (nls_strnicmp(codepage, a->name, b->name, a->len) == 0)) { + /* + * To preserve case, don't let an existing negative dentry's + * case take precedence. If a is not a negative dentry, this + * should have no side effects + */ + memcpy((unsigned char *)a->name, b->name, a->len); + return 0; + } + return 1; +} + +struct dentry_operations cifs_ci_dentry_ops = { + .d_revalidate = cifs_d_revalidate, + .d_hash = cifs_ci_hash, + .d_compare = cifs_ci_compare, +}; diff --git a/fs/cifs/fcntl.c b/fs/cifs/fcntl.c index 7d2a9202c39a..a7a47bb36bf3 100644 --- a/fs/cifs/fcntl.c +++ b/fs/cifs/fcntl.c @@ -78,6 +78,10 @@ int cifs_dir_notify(struct file * file, unsigned long arg) __u32 filter = FILE_NOTIFY_CHANGE_NAME | FILE_NOTIFY_CHANGE_ATTRIBUTES; __u16 netfid; + + if(experimEnabled == 0) + return 0; + xid = GetXid(); cifs_sb = CIFS_SB(file->f_dentry->d_sb); pTcon = cifs_sb->tcon; @@ -100,8 +104,10 @@ int cifs_dir_notify(struct file * file, unsigned long arg) } else { filter = convert_to_cifs_notify_flags(arg); if(filter != 0) { - rc = CIFSSMBNotify(xid, pTcon, 0 /* no subdirs */, netfid, - filter, cifs_sb->local_nls); + rc = CIFSSMBNotify(xid, pTcon, + 0 /* no subdirs */, netfid, + filter, file, arg & DN_MULTISHOT, + cifs_sb->local_nls); } else { rc = -EINVAL; } @@ -109,7 +115,7 @@ int cifs_dir_notify(struct file * file, unsigned long arg) it would close automatically but may be a way to do it easily when inode freed or when notify info is cleared/changed */ - cERROR(1,("notify rc %d",rc)); + cFYI(1,("notify rc %d",rc)); } } diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 3497125189df..14a1c72ced92 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -21,11 +21,15 @@ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ #include <linux/fs.h> +#include <linux/backing-dev.h> #include <linux/stat.h> #include <linux/fcntl.h> +#include <linux/mpage.h> #include <linux/pagemap.h> #include <linux/pagevec.h> #include <linux/smp_lock.h> +#include <linux/writeback.h> +#include <linux/delay.h> #include <asm/div64.h> #include "cifsfs.h" #include "cifspdu.h" @@ -47,6 +51,11 @@ static inline struct cifsFileInfo *cifs_init_private( private_data->pInode = inode; private_data->invalidHandle = FALSE; private_data->closePend = FALSE; + /* we have to track num writers to the inode, since writepages + does not tell us which handle the write is for so there can + be a close (overlapping with write) of the filehandle that + cifs_writepages chose to use */ + atomic_set(&private_data->wrtPending,0); return private_data; } @@ -256,6 +265,13 @@ int cifs_open(struct inode *inode, struct file *file) CREATE_NOT_DIR, &netfid, &oplock, buf, cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); + if (rc == -EIO) { + /* Old server, try legacy style OpenX */ + rc = SMBLegacyOpen(xid, pTcon, full_path, disposition, + desiredAccess, CREATE_NOT_DIR, &netfid, &oplock, buf, + cifs_sb->local_nls, cifs_sb->mnt_cifs_flags + & CIFS_MOUNT_MAP_SPECIAL_CHR); + } if (rc) { cFYI(1, ("cifs_open returned 0x%x ", rc)); goto out; @@ -463,6 +479,22 @@ int cifs_close(struct inode *inode, struct file *file) /* no sense reconnecting to close a file that is already closed */ if (pTcon->tidStatus != CifsNeedReconnect) { + int timeout = 2; + while((atomic_read(&pSMBFile->wrtPending) != 0) + && (timeout < 1000) ) { + /* Give write a better chance to get to + server ahead of the close. We do not + want to add a wait_q here as it would + increase the memory utilization as + the struct would be in each open file, + but this should give enough time to + clear the socket */ + write_unlock(&file->f_owner.lock); + cERROR(1,("close with pending writes")); + msleep(timeout); + write_lock(&file->f_owner.lock); + timeout *= 4; + } write_unlock(&file->f_owner.lock); rc = CIFSSMBClose(xid, pTcon, pSMBFile->netfid); @@ -744,14 +776,7 @@ ssize_t cifs_user_write(struct file *file, const char __user *write_data, 15 seconds is plenty */ } -#ifdef CONFIG_CIFS_STATS - if (total_written > 0) { - atomic_inc(&pTcon->num_writes); - spin_lock(&pTcon->stat_lock); - pTcon->bytes_written += total_written; - spin_unlock(&pTcon->stat_lock); - } -#endif + cifs_stats_bytes_written(pTcon, total_written); /* since the write may have blocked check these pointers again */ if (file->f_dentry) { @@ -791,9 +816,8 @@ static ssize_t cifs_write(struct file *file, const char *write_data, pTcon = cifs_sb->tcon; - /* cFYI(1, - (" write %d bytes to offset %lld of %s", write_size, - *poffset, file->f_dentry->d_name.name)); */ + cFYI(1,("write %zd bytes to offset %lld of %s", write_size, + *poffset, file->f_dentry->d_name.name)); if (file->private_data == NULL) return -EBADF; @@ -846,7 +870,26 @@ static ssize_t cifs_write(struct file *file, const char *write_data, if (rc != 0) break; } - +#ifdef CONFIG_CIFS_EXPERIMENTAL + /* BB FIXME We can not sign across two buffers yet */ + if((experimEnabled) && ((pTcon->ses->server->secMode & + (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) == 0)) { + struct kvec iov[2]; + unsigned int len; + + len = min((size_t)cifs_sb->wsize, + write_size - total_written); + /* iov[0] is reserved for smb header */ + iov[1].iov_base = (char *)write_data + + total_written; + iov[1].iov_len = len; + rc = CIFSSMBWrite2(xid, pTcon, + open_file->netfid, len, + *poffset, &bytes_written, + iov, 1, long_op); + } else + /* BB FIXME fixup indentation of line below */ +#endif rc = CIFSSMBWrite(xid, pTcon, open_file->netfid, min_t(const int, cifs_sb->wsize, @@ -867,14 +910,7 @@ static ssize_t cifs_write(struct file *file, const char *write_data, 15 seconds is plenty */ } -#ifdef CONFIG_CIFS_STATS - if (total_written > 0) { - atomic_inc(&pTcon->num_writes); - spin_lock(&pTcon->stat_lock); - pTcon->bytes_written += total_written; - spin_unlock(&pTcon->stat_lock); - } -#endif + cifs_stats_bytes_written(pTcon, total_written); /* since the write may have blocked check these pointers again */ if (file->f_dentry) { @@ -893,6 +929,43 @@ static ssize_t cifs_write(struct file *file, const char *write_data, return total_written; } +struct cifsFileInfo *find_writable_file(struct cifsInodeInfo *cifs_inode) +{ + struct cifsFileInfo *open_file; + int rc; + + read_lock(&GlobalSMBSeslock); + list_for_each_entry(open_file, &cifs_inode->openFileList, flist) { + if (open_file->closePend) + continue; + if (open_file->pfile && + ((open_file->pfile->f_flags & O_RDWR) || + (open_file->pfile->f_flags & O_WRONLY))) { + atomic_inc(&open_file->wrtPending); + read_unlock(&GlobalSMBSeslock); + if((open_file->invalidHandle) && + (!open_file->closePend) /* BB fixme -since the second clause can not be true remove it BB */) { + rc = cifs_reopen_file(&cifs_inode->vfs_inode, + open_file->pfile, FALSE); + /* if it fails, try another handle - might be */ + /* dangerous to hold up writepages with retry */ + if(rc) { + cFYI(1,("failed on reopen file in wp")); + read_lock(&GlobalSMBSeslock); + /* can not use this handle, no write + pending on this one after all */ + atomic_dec + (&open_file->wrtPending); + continue; + } + } + return open_file; + } + } + read_unlock(&GlobalSMBSeslock); + return NULL; +} + static int cifs_partialpagewrite(struct page *page, unsigned from, unsigned to) { struct address_space *mapping = page->mapping; @@ -903,10 +976,7 @@ static int cifs_partialpagewrite(struct page *page, unsigned from, unsigned to) struct cifs_sb_info *cifs_sb; struct cifsTconInfo *pTcon; struct inode *inode; - struct cifsInodeInfo *cifsInode; - struct cifsFileInfo *open_file = NULL; - struct list_head *tmp; - struct list_head *tmp1; + struct cifsFileInfo *open_file; if (!mapping || !mapping->host) return -EFAULT; @@ -934,49 +1004,20 @@ static int cifs_partialpagewrite(struct page *page, unsigned from, unsigned to) if (mapping->host->i_size - offset < (loff_t)to) to = (unsigned)(mapping->host->i_size - offset); - cifsInode = CIFS_I(mapping->host); - read_lock(&GlobalSMBSeslock); - /* BB we should start at the end */ - list_for_each_safe(tmp, tmp1, &cifsInode->openFileList) { - open_file = list_entry(tmp, struct cifsFileInfo, flist); - if (open_file->closePend) - continue; - /* We check if file is open for writing first */ - if ((open_file->pfile) && - ((open_file->pfile->f_flags & O_RDWR) || - (open_file->pfile->f_flags & O_WRONLY))) { - read_unlock(&GlobalSMBSeslock); - bytes_written = cifs_write(open_file->pfile, - write_data, to-from, - &offset); - read_lock(&GlobalSMBSeslock); + open_file = find_writable_file(CIFS_I(mapping->host)); + if (open_file) { + bytes_written = cifs_write(open_file->pfile, write_data, + to-from, &offset); + atomic_dec(&open_file->wrtPending); /* Does mm or vfs already set times? */ - inode->i_atime = - inode->i_mtime = current_fs_time(inode->i_sb); - if ((bytes_written > 0) && (offset)) { - rc = 0; - } else if (bytes_written < 0) { - if (rc == -EBADF) { - /* have seen a case in which kernel seemed to - have closed/freed a file even with writes - active so we might as well see if there are - other file structs to try for the same - inode before giving up */ - continue; - } else - rc = bytes_written; - } - break; /* now that we found a valid file handle and - tried to write to it we are done, no sense - continuing to loop looking for another */ - } - if (tmp->next == NULL) { - cFYI(1, ("File instance %p removed", tmp)); - break; + inode->i_atime = inode->i_mtime = current_fs_time(inode->i_sb); + if ((bytes_written > 0) && (offset)) { + rc = 0; + } else if (bytes_written < 0) { + if (rc != -EBADF) + rc = bytes_written; } - } - read_unlock(&GlobalSMBSeslock); - if (open_file == NULL) { + } else { cFYI(1, ("No writeable filehandles for inode")); rc = -EIO; } @@ -985,20 +1026,207 @@ static int cifs_partialpagewrite(struct page *page, unsigned from, unsigned to) return rc; } -#if 0 +#ifdef CONFIG_CIFS_EXPERIMENTAL static int cifs_writepages(struct address_space *mapping, - struct writeback_control *wbc) + struct writeback_control *wbc) { - int rc = -EFAULT; + struct backing_dev_info *bdi = mapping->backing_dev_info; + unsigned int bytes_to_write; + unsigned int bytes_written; + struct cifs_sb_info *cifs_sb; + int done = 0; + pgoff_t end = -1; + pgoff_t index; + int is_range = 0; + struct kvec iov[32]; + int len; + int n_iov = 0; + pgoff_t next; + int nr_pages; + __u64 offset = 0; + struct cifsFileInfo *open_file; + struct page *page; + struct pagevec pvec; + int rc = 0; + int scanned = 0; int xid; + cifs_sb = CIFS_SB(mapping->host->i_sb); + + /* + * If wsize is smaller that the page cache size, default to writing + * one page at a time via cifs_writepage + */ + if (cifs_sb->wsize < PAGE_CACHE_SIZE) + return generic_writepages(mapping, wbc); + + /* BB FIXME we do not have code to sign across multiple buffers yet, + so go to older writepage style write which we can sign if needed */ + if((cifs_sb->tcon->ses) && (cifs_sb->tcon->ses->server)) + if(cifs_sb->tcon->ses->server->secMode & + (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) + return generic_writepages(mapping, wbc); + + /* + * BB: Is this meaningful for a non-block-device file system? + * If it is, we should test it again after we do I/O + */ + if (wbc->nonblocking && bdi_write_congested(bdi)) { + wbc->encountered_congestion = 1; + return 0; + } + xid = GetXid(); - /* Find contiguous pages then iterate through repeating - call 16K write then Setpageuptodate or if LARGE_WRITE_X - support then send larger writes via kevec so as to eliminate - a memcpy */ + pagevec_init(&pvec, 0); + if (wbc->sync_mode == WB_SYNC_NONE) + index = mapping->writeback_index; /* Start from prev offset */ + else { + index = 0; + scanned = 1; + } + if (wbc->start || wbc->end) { + index = wbc->start >> PAGE_CACHE_SHIFT; + end = wbc->end >> PAGE_CACHE_SHIFT; + is_range = 1; + scanned = 1; + } +retry: + while (!done && (index <= end) && + (nr_pages = pagevec_lookup_tag(&pvec, mapping, &index, + PAGECACHE_TAG_DIRTY, + min(end - index, (pgoff_t)PAGEVEC_SIZE - 1) + 1))) { + int first; + unsigned int i; + + first = -1; + next = 0; + n_iov = 0; + bytes_to_write = 0; + + for (i = 0; i < nr_pages; i++) { + page = pvec.pages[i]; + /* + * At this point we hold neither mapping->tree_lock nor + * lock on the page itself: the page may be truncated or + * invalidated (changing page->mapping to NULL), or even + * swizzled back from swapper_space to tmpfs file + * mapping + */ + + if (first < 0) + lock_page(page); + else if (TestSetPageLocked(page)) + break; + + if (unlikely(page->mapping != mapping)) { + unlock_page(page); + break; + } + + if (unlikely(is_range) && (page->index > end)) { + done = 1; + unlock_page(page); + break; + } + + if (next && (page->index != next)) { + /* Not next consecutive page */ + unlock_page(page); + break; + } + + if (wbc->sync_mode != WB_SYNC_NONE) + wait_on_page_writeback(page); + + if (PageWriteback(page) || + !test_clear_page_dirty(page)) { + unlock_page(page); + break; + } + + if (page_offset(page) >= mapping->host->i_size) { + done = 1; + unlock_page(page); + break; + } + + /* + * BB can we get rid of this? pages are held by pvec + */ + page_cache_get(page); + + len = min(mapping->host->i_size - page_offset(page), + (loff_t)PAGE_CACHE_SIZE); + + /* reserve iov[0] for the smb header */ + n_iov++; + iov[n_iov].iov_base = kmap(page); + iov[n_iov].iov_len = len; + bytes_to_write += len; + + if (first < 0) { + first = i; + offset = page_offset(page); + } + next = page->index + 1; + if (bytes_to_write + PAGE_CACHE_SIZE > cifs_sb->wsize) + break; + } + if (n_iov) { + /* Search for a writable handle every time we call + * CIFSSMBWrite2. We can't rely on the last handle + * we used to still be valid + */ + open_file = find_writable_file(CIFS_I(mapping->host)); + if (!open_file) { + cERROR(1, ("No writable handles for inode")); + rc = -EBADF; + } else { + rc = CIFSSMBWrite2(xid, cifs_sb->tcon, + open_file->netfid, + bytes_to_write, offset, + &bytes_written, iov, n_iov, + 1); + atomic_dec(&open_file->wrtPending); + if (rc || bytes_written < bytes_to_write) { + cERROR(1,("Write2 ret %d, written = %d", + rc, bytes_written)); + /* BB what if continued retry is + requested via mount flags? */ + set_bit(AS_EIO, &mapping->flags); + SetPageError(page); + } else { + cifs_stats_bytes_written(cifs_sb->tcon, + bytes_written); + } + } + for (i = 0; i < n_iov; i++) { + page = pvec.pages[first + i]; + kunmap(page); + unlock_page(page); + page_cache_release(page); + } + if ((wbc->nr_to_write -= n_iov) <= 0) + done = 1; + index = next; + } + pagevec_release(&pvec); + } + if (!scanned && !done) { + /* + * We hit the last page and there is more work to be done: wrap + * back to the start of the file + */ + scanned = 1; + index = 0; + goto retry; + } + if (!is_range) + mapping->writeback_index = index; + FreeXid(xid); + return rc; } #endif @@ -1207,12 +1435,10 @@ ssize_t cifs_user_read(struct file *file, char __user *read_data, if (rc != 0) break; } - rc = CIFSSMBRead(xid, pTcon, - open_file->netfid, - current_read_size, *poffset, - &bytes_read, &smb_read_data); - + open_file->netfid, + current_read_size, *poffset, + &bytes_read, &smb_read_data); pSMBr = (struct smb_com_read_rsp *)smb_read_data; if (copy_to_user(current_offset, smb_read_data + 4 /* RFC1001 hdr */ @@ -1235,12 +1461,7 @@ ssize_t cifs_user_read(struct file *file, char __user *read_data, return rc; } } else { -#ifdef CONFIG_CIFS_STATS - atomic_inc(&pTcon->num_reads); - spin_lock(&pTcon->stat_lock); - pTcon->bytes_read += total_read; - spin_unlock(&pTcon->stat_lock); -#endif + cifs_stats_bytes_read(pTcon, bytes_read); *poffset += bytes_read; } } @@ -1280,6 +1501,13 @@ static ssize_t cifs_read(struct file *file, char *read_data, size_t read_size, total_read += bytes_read, current_offset += bytes_read) { current_read_size = min_t(const int, read_size - total_read, cifs_sb->rsize); + /* For windows me and 9x we do not want to request more + than it negotiated since it will refuse the read then */ + if((pTcon->ses) && + !(pTcon->ses->capabilities & CAP_LARGE_FILES)) { + current_read_size = min_t(const int, current_read_size, + pTcon->ses->server->maxBuf - 128); + } rc = -EAGAIN; while (rc == -EAGAIN) { if ((open_file->invalidHandle) && @@ -1289,11 +1517,10 @@ static ssize_t cifs_read(struct file *file, char *read_data, size_t read_size, if (rc != 0) break; } - rc = CIFSSMBRead(xid, pTcon, - open_file->netfid, - current_read_size, *poffset, - &bytes_read, ¤t_offset); + open_file->netfid, + current_read_size, *poffset, + &bytes_read, ¤t_offset); } if (rc || (bytes_read == 0)) { if (total_read) { @@ -1303,12 +1530,7 @@ static ssize_t cifs_read(struct file *file, char *read_data, size_t read_size, return rc; } } else { -#ifdef CONFIG_CIFS_STATS - atomic_inc(&pTcon->num_reads); - spin_lock(&pTcon->stat_lock); - pTcon->bytes_read += total_read; - spin_unlock(&pTcon->stat_lock); -#endif + cifs_stats_bytes_read(pTcon, total_read); *poffset += bytes_read; } } @@ -1452,10 +1674,11 @@ static int cifs_readpages(struct file *file, struct address_space *mapping, } rc = CIFSSMBRead(xid, pTcon, - open_file->netfid, - read_size, offset, - &bytes_read, &smb_read_data); - /* BB need to check return code here */ + open_file->netfid, + read_size, offset, + &bytes_read, &smb_read_data); + + /* BB more RC checks ? */ if (rc== -EAGAIN) { if (smb_read_data) { cifs_buf_release(smb_read_data); @@ -1480,12 +1703,7 @@ static int cifs_readpages(struct file *file, struct address_space *mapping, le16_to_cpu(pSMBr->DataOffset), &lru_pvec); i += bytes_read >> PAGE_CACHE_SHIFT; -#ifdef CONFIG_CIFS_STATS - atomic_inc(&pTcon->num_reads); - spin_lock(&pTcon->stat_lock); - pTcon->bytes_read += bytes_read; - spin_unlock(&pTcon->stat_lock); -#endif + cifs_stats_bytes_read(pTcon, bytes_read); if ((int)(bytes_read & PAGE_CACHE_MASK) != bytes_read) { i++; /* account for partial page */ @@ -1603,40 +1821,21 @@ static int cifs_readpage(struct file *file, struct page *page) page caching in the current Linux kernel design */ int is_size_safe_to_change(struct cifsInodeInfo *cifsInode) { - struct list_head *tmp; - struct list_head *tmp1; struct cifsFileInfo *open_file = NULL; - int rc = TRUE; - if (cifsInode == NULL) - return rc; - - read_lock(&GlobalSMBSeslock); - list_for_each_safe(tmp, tmp1, &cifsInode->openFileList) { - open_file = list_entry(tmp, struct cifsFileInfo, flist); - if (open_file == NULL) - break; - if (open_file->closePend) - continue; - /* We check if file is open for writing, - BB we could supplement this with a check to see if file size - changes have been flushed to server - ie inode metadata dirty */ - if ((open_file->pfile) && - ((open_file->pfile->f_flags & O_RDWR) || - (open_file->pfile->f_flags & O_WRONLY))) { - rc = FALSE; - break; - } - if (tmp->next == NULL) { - cFYI(1, ("File instance %p removed", tmp)); - break; - } - } - read_unlock(&GlobalSMBSeslock); - return rc; + if (cifsInode) + open_file = find_writable_file(cifsInode); + + if(open_file) { + /* there is not actually a write pending so let + this handle go free and allow it to + be closable if needed */ + atomic_dec(&open_file->wrtPending); + return 0; + } else + return 1; } - static int cifs_prepare_write(struct file *file, struct page *page, unsigned from, unsigned to) { @@ -1676,6 +1875,9 @@ struct address_space_operations cifs_addr_ops = { .readpage = cifs_readpage, .readpages = cifs_readpages, .writepage = cifs_writepage, +#ifdef CONFIG_CIFS_EXPERIMENTAL + .writepages = cifs_writepages, +#endif .prepare_write = cifs_prepare_write, .commit_write = cifs_commit_write, .set_page_dirty = __set_page_dirty_nobuffers, diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index 8d336a900255..411c1f7f84da 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -41,7 +41,7 @@ int cifs_get_inode_info_unix(struct inode **pinode, char *tmp_path; pTcon = cifs_sb->tcon; - cFYI(1, (" Getting info on %s ", search_path)); + cFYI(1, ("Getting info on %s ", search_path)); /* could have done a find first instead but this returns more info */ rc = CIFSSMBUnixQPathInfo(xid, pTcon, search_path, &findData, cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & @@ -97,9 +97,9 @@ int cifs_get_inode_info_unix(struct inode **pinode, inode = *pinode; cifsInfo = CIFS_I(inode); - cFYI(1, (" Old time %ld ", cifsInfo->time)); + cFYI(1, ("Old time %ld ", cifsInfo->time)); cifsInfo->time = jiffies; - cFYI(1, (" New time %ld ", cifsInfo->time)); + cFYI(1, ("New time %ld ", cifsInfo->time)); /* this is ok to set on every inode revalidate */ atomic_set(&cifsInfo->inUse,1); @@ -111,6 +111,9 @@ int cifs_get_inode_info_unix(struct inode **pinode, inode->i_ctime = cifs_NTtimeToUnix(le64_to_cpu(findData.LastStatusChange)); inode->i_mode = le64_to_cpu(findData.Permissions); + /* since we set the inode type below we need to mask off + to avoid strange results if bits set above */ + inode->i_mode &= ~S_IFMT; if (type == UNIX_FILE) { inode->i_mode |= S_IFREG; } else if (type == UNIX_SYMLINK) { @@ -129,6 +132,10 @@ int cifs_get_inode_info_unix(struct inode **pinode, inode->i_mode |= S_IFIFO; } else if (type == UNIX_SOCKET) { inode->i_mode |= S_IFSOCK; + } else { + /* safest to call it a file if we do not know */ + inode->i_mode |= S_IFREG; + cFYI(1,("unknown type %d",type)); } inode->i_uid = le64_to_cpu(findData.Uid); inode->i_gid = le64_to_cpu(findData.Gid); @@ -155,28 +162,39 @@ int cifs_get_inode_info_unix(struct inode **pinode, } if (num_of_bytes < end_of_file) - cFYI(1, ("allocation size less than end of file ")); + cFYI(1, ("allocation size less than end of file")); cFYI(1, ("Size %ld and blocks %ld", (unsigned long) inode->i_size, inode->i_blocks)); if (S_ISREG(inode->i_mode)) { - cFYI(1, (" File inode ")); + cFYI(1, ("File inode")); inode->i_op = &cifs_file_inode_ops; - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DIRECT_IO) - inode->i_fop = &cifs_file_direct_ops; - else + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DIRECT_IO) { + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL) + inode->i_fop = + &cifs_file_direct_nobrl_ops; + else + inode->i_fop = &cifs_file_direct_ops; + } else if(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL) + inode->i_fop = &cifs_file_nobrl_ops; + else /* not direct, send byte range locks */ inode->i_fop = &cifs_file_ops; + inode->i_data.a_ops = &cifs_addr_ops; + /* check if server can support readpages */ + if(pTcon->ses->server->maxBuf < + 4096 + MAX_CIFS_HDR_SIZE) + inode->i_data.a_ops->readpages = NULL; } else if (S_ISDIR(inode->i_mode)) { - cFYI(1, (" Directory inode")); + cFYI(1, ("Directory inode")); inode->i_op = &cifs_dir_inode_ops; inode->i_fop = &cifs_dir_ops; } else if (S_ISLNK(inode->i_mode)) { - cFYI(1, (" Symbolic Link inode ")); + cFYI(1, ("Symbolic Link inode")); inode->i_op = &cifs_symlink_inode_ops; /* tmp_inode->i_fop = */ /* do not need to set to anything */ } else { - cFYI(1, (" Init special inode ")); + cFYI(1, ("Init special inode")); init_special_inode(inode, inode->i_mode, inode->i_rdev); } @@ -184,6 +202,111 @@ int cifs_get_inode_info_unix(struct inode **pinode, return rc; } +static int decode_sfu_inode(struct inode * inode, __u64 size, + const unsigned char *path, + struct cifs_sb_info *cifs_sb, int xid) +{ + int rc; + int oplock = FALSE; + __u16 netfid; + struct cifsTconInfo *pTcon = cifs_sb->tcon; + char buf[24]; + unsigned int bytes_read; + char * pbuf; + + pbuf = buf; + + if(size == 0) { + inode->i_mode |= S_IFIFO; + return 0; + } else if (size < 8) { + return -EINVAL; /* EOPNOTSUPP? */ + } + + rc = CIFSSMBOpen(xid, pTcon, path, FILE_OPEN, GENERIC_READ, + CREATE_NOT_DIR, &netfid, &oplock, NULL, + cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); + if (rc==0) { + /* Read header */ + rc = CIFSSMBRead(xid, pTcon, + netfid, + 24 /* length */, 0 /* offset */, + &bytes_read, &pbuf); + if((rc == 0) && (bytes_read >= 8)) { + if(memcmp("IntxBLK", pbuf, 8) == 0) { + cFYI(1,("Block device")); + inode->i_mode |= S_IFBLK; + if(bytes_read == 24) { + /* we have enough to decode dev num */ + __u64 mjr; /* major */ + __u64 mnr; /* minor */ + mjr = le64_to_cpu(*(__le64 *)(pbuf+8)); + mnr = le64_to_cpu(*(__le64 *)(pbuf+16)); + inode->i_rdev = MKDEV(mjr, mnr); + } + } else if(memcmp("IntxCHR", pbuf, 8) == 0) { + cFYI(1,("Char device")); + inode->i_mode |= S_IFCHR; + if(bytes_read == 24) { + /* we have enough to decode dev num */ + __u64 mjr; /* major */ + __u64 mnr; /* minor */ + mjr = le64_to_cpu(*(__le64 *)(pbuf+8)); + mnr = le64_to_cpu(*(__le64 *)(pbuf+16)); + inode->i_rdev = MKDEV(mjr, mnr); + } + } else if(memcmp("IntxLNK", pbuf, 7) == 0) { + cFYI(1,("Symlink")); + inode->i_mode |= S_IFLNK; + } else { + inode->i_mode |= S_IFREG; /* file? */ + rc = -EOPNOTSUPP; + } + } else { + inode->i_mode |= S_IFREG; /* then it is a file */ + rc = -EOPNOTSUPP; /* or some unknown SFU type */ + } + CIFSSMBClose(xid, pTcon, netfid); + } + return rc; + +} + +#define SFBITS_MASK (S_ISVTX | S_ISGID | S_ISUID) /* SETFILEBITS valid bits */ + +static int get_sfu_uid_mode(struct inode * inode, + const unsigned char *path, + struct cifs_sb_info *cifs_sb, int xid) +{ +#ifdef CONFIG_CIFS_XATTR + ssize_t rc; + char ea_value[4]; + __u32 mode; + + rc = CIFSSMBQueryEA(xid, cifs_sb->tcon, path, "SETFILEBITS", + ea_value, 4 /* size of buf */, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); + if(rc < 0) + return (int)rc; + else if (rc > 3) { + mode = le32_to_cpu(*((__le32 *)ea_value)); + inode->i_mode &= ~SFBITS_MASK; + cFYI(1,("special bits 0%o org mode 0%o", mode, inode->i_mode)); + inode->i_mode = (mode & SFBITS_MASK) | inode->i_mode; + cFYI(1,("special mode bits 0%o", mode)); + return 0; + } else { + return 0; + } +#else + return -EOPNOTSUPP; +#endif + + +} + int cifs_get_inode_info(struct inode **pinode, const unsigned char *search_path, FILE_ALL_INFO *pfindData, struct super_block *sb, int xid) @@ -196,7 +319,7 @@ int cifs_get_inode_info(struct inode **pinode, char *buf = NULL; pTcon = cifs_sb->tcon; - cFYI(1,("Getting info on %s ", search_path)); + cFYI(1,("Getting info on %s", search_path)); if ((pfindData == NULL) && (*pinode != NULL)) { if (CIFS_I(*pinode)->clientCanCacheRead) { @@ -213,8 +336,18 @@ int cifs_get_inode_info(struct inode **pinode, pfindData = (FILE_ALL_INFO *)buf; /* could do find first instead but this returns more info */ rc = CIFSSMBQPathInfo(xid, pTcon, search_path, pfindData, - cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & + cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); + /* BB optimize code so we do not make the above call + when server claims no NT SMB support and the above call + failed at least once - set flag in tcon or mount */ + if((rc == -EOPNOTSUPP) || (rc == -EINVAL)) { + rc = SMBQueryInformation(xid, pTcon, search_path, + pfindData, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); + } + } /* dump_mem("\nQPathInfo return data",&findData, sizeof(findData)); */ if (rc) { @@ -267,7 +400,6 @@ int cifs_get_inode_info(struct inode **pinode, there Windows server or network appliances for which IndexNumber field is not guaranteed unique? */ -#ifdef CONFIG_CIFS_EXPERIMENTAL if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM){ int rc1 = 0; __u64 inode_num; @@ -283,15 +415,14 @@ int cifs_get_inode_info(struct inode **pinode, } else /* do we need cast or hash to ino? */ (*pinode)->i_ino = inode_num; } /* else ino incremented to unique num in new_inode*/ -#endif /* CIFS_EXPERIMENTAL */ insert_inode_hash(*pinode); } inode = *pinode; cifsInfo = CIFS_I(inode); cifsInfo->cifsAttrs = attr; - cFYI(1, (" Old time %ld ", cifsInfo->time)); + cFYI(1, ("Old time %ld ", cifsInfo->time)); cifsInfo->time = jiffies; - cFYI(1, (" New time %ld ", cifsInfo->time)); + cFYI(1, ("New time %ld ", cifsInfo->time)); /* blksize needs to be multiple of two. So safer to default to blksize and blkbits set in superblock so 2**blkbits and blksize @@ -305,13 +436,15 @@ int cifs_get_inode_info(struct inode **pinode, cifs_NTtimeToUnix(le64_to_cpu(pfindData->LastWriteTime)); inode->i_ctime = cifs_NTtimeToUnix(le64_to_cpu(pfindData->ChangeTime)); - cFYI(0, (" Attributes came in as 0x%x ", attr)); + cFYI(0, ("Attributes came in as 0x%x ", attr)); /* set default mode. will override for dirs below */ if (atomic_read(&cifsInfo->inUse) == 0) /* new inode, can safely set these fields */ inode->i_mode = cifs_sb->mnt_file_mode; - + else /* since we set the inode type below we need to mask off + to avoid strange results if type changes and both get orred in */ + inode->i_mode &= ~S_IFMT; /* if (attr & ATTR_REPARSE) */ /* We no longer handle these as symlinks because we could not follow them due to the absolute path with drive letter */ @@ -320,6 +453,22 @@ int cifs_get_inode_info(struct inode **pinode, on dirs */ inode->i_mode = cifs_sb->mnt_dir_mode; inode->i_mode |= S_IFDIR; + } else if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UNX_EMUL) && + (cifsInfo->cifsAttrs & ATTR_SYSTEM) && + /* No need to le64 convert size of zero */ + (pfindData->EndOfFile == 0)) { + inode->i_mode = cifs_sb->mnt_file_mode; + inode->i_mode |= S_IFIFO; +/* BB Finish for SFU style symlinks and devices */ + } else if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UNX_EMUL) && + (cifsInfo->cifsAttrs & ATTR_SYSTEM)) { + if (decode_sfu_inode(inode, + le64_to_cpu(pfindData->EndOfFile), + search_path, + cifs_sb, xid)) { + cFYI(1,("Unrecognized sfu inode type")); + } + cFYI(1,("sfu mode 0%o",inode->i_mode)); } else { inode->i_mode |= S_IFREG; /* treat the dos attribute of read-only as read-only @@ -344,7 +493,10 @@ int cifs_get_inode_info(struct inode **pinode, /* BB fill in uid and gid here? with help from winbind? or retrieve from NTFS stream extended attribute */ - if (atomic_read(&cifsInfo->inUse) == 0) { + if(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UNX_EMUL) { + /* fill in uid, gid, mode from server ACL */ + get_sfu_uid_mode(inode, search_path, cifs_sb, xid); + } else if (atomic_read(&cifsInfo->inUse) == 0) { inode->i_uid = cifs_sb->mnt_uid; inode->i_gid = cifs_sb->mnt_gid; /* set so we do not keep refreshing these fields with @@ -353,19 +505,29 @@ int cifs_get_inode_info(struct inode **pinode, } if (S_ISREG(inode->i_mode)) { - cFYI(1, (" File inode ")); + cFYI(1, ("File inode")); inode->i_op = &cifs_file_inode_ops; - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DIRECT_IO) - inode->i_fop = &cifs_file_direct_ops; - else + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DIRECT_IO) { + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL) + inode->i_fop = + &cifs_file_direct_nobrl_ops; + else + inode->i_fop = &cifs_file_direct_ops; + } else if(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL) + inode->i_fop = &cifs_file_nobrl_ops; + else /* not direct, send byte range locks */ inode->i_fop = &cifs_file_ops; + inode->i_data.a_ops = &cifs_addr_ops; + if(pTcon->ses->server->maxBuf < + 4096 + MAX_CIFS_HDR_SIZE) + inode->i_data.a_ops->readpages = NULL; } else if (S_ISDIR(inode->i_mode)) { - cFYI(1, (" Directory inode ")); + cFYI(1, ("Directory inode")); inode->i_op = &cifs_dir_inode_ops; inode->i_fop = &cifs_dir_ops; } else if (S_ISLNK(inode->i_mode)) { - cFYI(1, (" Symbolic Link inode ")); + cFYI(1, ("Symbolic Link inode")); inode->i_op = &cifs_symlink_inode_ops; } else { init_special_inode(inode, inode->i_mode, @@ -402,7 +564,7 @@ int cifs_unlink(struct inode *inode, struct dentry *direntry) struct cifsInodeInfo *cifsInode; FILE_BASIC_INFO *pinfo_buf; - cFYI(1, (" cifs_unlink, inode = 0x%p with ", inode)); + cFYI(1, ("cifs_unlink, inode = 0x%p with ", inode)); xid = GetXid(); @@ -548,7 +710,7 @@ int cifs_mkdir(struct inode *inode, struct dentry *direntry, int mode) char *full_path = NULL; struct inode *newinode = NULL; - cFYI(1, ("In cifs_mkdir, mode = 0x%x inode = 0x%p ", mode, inode)); + cFYI(1, ("In cifs_mkdir, mode = 0x%x inode = 0x%p", mode, inode)); xid = GetXid(); @@ -577,7 +739,10 @@ int cifs_mkdir(struct inode *inode, struct dentry *direntry, int mode) rc = cifs_get_inode_info(&newinode, full_path, NULL, inode->i_sb,xid); - direntry->d_op = &cifs_dentry_ops; + if (pTcon->nocase) + direntry->d_op = &cifs_ci_dentry_ops; + else + direntry->d_op = &cifs_dentry_ops; d_instantiate(direntry, newinode); if (direntry->d_inode) direntry->d_inode->i_nlink = 2; @@ -603,6 +768,17 @@ int cifs_mkdir(struct inode *inode, struct dentry *direntry, int mode) /* BB to be implemented via Windows secrty descriptors eg CIFSSMBWinSetPerms(xid, pTcon, full_path, mode, -1, -1, local_nls); */ + if(direntry->d_inode) { + direntry->d_inode->i_mode = mode; + direntry->d_inode->i_mode |= S_IFDIR; + if(cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_SET_UID) { + direntry->d_inode->i_uid = + current->fsuid; + direntry->d_inode->i_gid = + current->fsgid; + } + } } } kfree(full_path); @@ -619,7 +795,7 @@ int cifs_rmdir(struct inode *inode, struct dentry *direntry) char *full_path = NULL; struct cifsInodeInfo *cifsInode; - cFYI(1, (" cifs_rmdir, inode = 0x%p with ", inode)); + cFYI(1, ("cifs_rmdir, inode = 0x%p with ", inode)); xid = GetXid(); @@ -874,14 +1050,20 @@ int cifs_revalidate(struct dentry *direntry) filemap_fdatawrite(direntry->d_inode->i_mapping); } if (invalidate_inode) { - if (direntry->d_inode->i_mapping) - filemap_fdatawait(direntry->d_inode->i_mapping); - /* may eventually have to do this for open files too */ - if (list_empty(&(cifsInode->openFileList))) { - /* Has changed on server - flush read ahead pages */ - cFYI(1, ("Invalidating read ahead data on " - "closed file")); - invalidate_remote_inode(direntry->d_inode); + /* shrink_dcache not necessary now that cifs dentry ops + are exported for negative dentries */ +/* if(S_ISDIR(direntry->d_inode->i_mode)) + shrink_dcache_parent(direntry); */ + if (S_ISREG(direntry->d_inode->i_mode)) { + if (direntry->d_inode->i_mapping) + filemap_fdatawait(direntry->d_inode->i_mapping); + /* may eventually have to do this for open files too */ + if (list_empty(&(cifsInode->openFileList))) { + /* changed on server - flush read ahead pages */ + cFYI(1, ("Invalidating read ahead data on " + "closed file")); + invalidate_remote_inode(direntry->d_inode); + } } } /* up(&direntry->d_inode->i_sem); */ @@ -928,7 +1110,6 @@ int cifs_setattr(struct dentry *direntry, struct iattr *attrs) struct cifsTconInfo *pTcon; char *full_path = NULL; int rc = -EACCES; - int found = FALSE; struct cifsFileInfo *open_file = NULL; FILE_BASIC_INFO time_buf; int set_time = FALSE; @@ -936,15 +1117,25 @@ int cifs_setattr(struct dentry *direntry, struct iattr *attrs) __u64 uid = 0xFFFFFFFFFFFFFFFFULL; __u64 gid = 0xFFFFFFFFFFFFFFFFULL; struct cifsInodeInfo *cifsInode; - struct list_head *tmp; xid = GetXid(); - cFYI(1, (" In cifs_setattr, name = %s attrs->iavalid 0x%x ", + cFYI(1, ("In cifs_setattr, name = %s attrs->iavalid 0x%x ", direntry->d_name.name, attrs->ia_valid)); + cifs_sb = CIFS_SB(direntry->d_inode->i_sb); pTcon = cifs_sb->tcon; + if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_PERM) == 0) { + /* check if we have permission to change attrs */ + rc = inode_change_ok(direntry->d_inode, attrs); + if(rc < 0) { + FreeXid(xid); + return rc; + } else + rc = 0; + } + down(&direntry->d_sb->s_vfs_rename_sem); full_path = build_path_from_dentry(direntry); up(&direntry->d_sb->s_vfs_rename_sem); @@ -961,7 +1152,6 @@ int cifs_setattr(struct dentry *direntry, struct iattr *attrs) filemap_fdatawait(direntry->d_inode->i_mapping); if (attrs->ia_valid & ATTR_SIZE) { - read_lock(&GlobalSMBSeslock); /* To avoid spurious oplock breaks from server, in the case of inodes that we already have open, avoid doing path based setting of file size if we can do it by handle. @@ -969,39 +1159,24 @@ int cifs_setattr(struct dentry *direntry, struct iattr *attrs) when the local oplock break takes longer to flush writebehind data than the SMB timeout for the SetPathInfo request would allow */ - list_for_each(tmp, &cifsInode->openFileList) { - open_file = list_entry(tmp, struct cifsFileInfo, - flist); - /* We check if file is open for writing first */ - if ((open_file->pfile) && - ((open_file->pfile->f_flags & O_RDWR) || - (open_file->pfile->f_flags & O_WRONLY))) { - if (open_file->invalidHandle == FALSE) { - /* we found a valid, writeable network - file handle to use to try to set the - file size */ - __u16 nfid = open_file->netfid; - __u32 npid = open_file->pid; - read_unlock(&GlobalSMBSeslock); - found = TRUE; - rc = CIFSSMBSetFileSize(xid, pTcon, - attrs->ia_size, nfid, npid, - FALSE); - cFYI(1, ("SetFileSize by handle " - "(setattrs) rc = %d", rc)); - /* Do not need reopen and retry on - EAGAIN since we will retry by - pathname below */ - - /* now that we found one valid file - handle no sense continuing to loop - trying others, so break here */ - break; - } + open_file = find_writable_file(cifsInode); + if (open_file) { + __u16 nfid = open_file->netfid; + __u32 npid = open_file->pid; + rc = CIFSSMBSetFileSize(xid, pTcon, attrs->ia_size, + nfid, npid, FALSE); + atomic_dec(&open_file->wrtPending); + cFYI(1,("SetFSize for attrs rc = %d", rc)); + if(rc == -EINVAL) { + int bytes_written; + rc = CIFSSMBWrite(xid, pTcon, + nfid, 0, attrs->ia_size, + &bytes_written, NULL, NULL, + 1 /* 45 seconds */); + cFYI(1,("Wrt seteof rc %d", rc)); } - } - if (found == FALSE) - read_unlock(&GlobalSMBSeslock); + } else + rc = -EINVAL; if (rc != 0) { /* Set file size by pathname rather than by handle @@ -1013,7 +1188,30 @@ int cifs_setattr(struct dentry *direntry, struct iattr *attrs) cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); - cFYI(1, (" SetEOF by path (setattrs) rc = %d", rc)); + cFYI(1, ("SetEOF by path (setattrs) rc = %d", rc)); + if(rc == -EINVAL) { + __u16 netfid; + int oplock = FALSE; + + rc = SMBLegacyOpen(xid, pTcon, full_path, + FILE_OPEN, + SYNCHRONIZE | FILE_WRITE_ATTRIBUTES, + CREATE_NOT_DIR, &netfid, &oplock, + NULL, cifs_sb->local_nls, + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); + if (rc==0) { + int bytes_written; + rc = CIFSSMBWrite(xid, pTcon, + netfid, 0, + attrs->ia_size, + &bytes_written, NULL, + NULL, 1 /* 45 sec */); + cFYI(1,("wrt seteof rc %d",rc)); + CIFSSMBClose(xid, pTcon, netfid); + } + + } } /* Server is ok setting allocation size implicitly - no need @@ -1026,24 +1224,22 @@ int cifs_setattr(struct dentry *direntry, struct iattr *attrs) rc = vmtruncate(direntry->d_inode, attrs->ia_size); cifs_truncate_page(direntry->d_inode->i_mapping, direntry->d_inode->i_size); - } + } else + goto cifs_setattr_exit; } if (attrs->ia_valid & ATTR_UID) { - cFYI(1, (" CIFS - UID changed to %d", attrs->ia_uid)); + cFYI(1, ("UID changed to %d", attrs->ia_uid)); uid = attrs->ia_uid; - /* entry->uid = cpu_to_le16(attr->ia_uid); */ } if (attrs->ia_valid & ATTR_GID) { - cFYI(1, (" CIFS - GID changed to %d", attrs->ia_gid)); + cFYI(1, ("GID changed to %d", attrs->ia_gid)); gid = attrs->ia_gid; - /* entry->gid = cpu_to_le16(attr->ia_gid); */ } time_buf.Attributes = 0; if (attrs->ia_valid & ATTR_MODE) { - cFYI(1, (" CIFS - Mode changed to 0x%x", attrs->ia_mode)); + cFYI(1, ("Mode changed to 0x%x", attrs->ia_mode)); mode = attrs->ia_mode; - /* entry->mode = cpu_to_le16(attr->ia_mode); */ } if ((cifs_sb->tcon->ses->capabilities & CAP_UNIX) @@ -1053,6 +1249,7 @@ int cifs_setattr(struct dentry *direntry, struct iattr *attrs) cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); else if (attrs->ia_valid & ATTR_MODE) { + rc = 0; if ((mode & S_IWUGO) == 0) /* not writeable */ { if ((cifsInode->cifsAttrs & ATTR_READONLY) == 0) time_buf.Attributes = @@ -1083,18 +1280,24 @@ int cifs_setattr(struct dentry *direntry, struct iattr *attrs) cpu_to_le64(cifs_UnixTimeToNT(attrs->ia_mtime)); } else time_buf.LastWriteTime = 0; - - if (attrs->ia_valid & ATTR_CTIME) { + /* Do not set ctime explicitly unless other time + stamps are changed explicitly (i.e. by utime() + since we would then have a mix of client and + server times */ + + if (set_time && (attrs->ia_valid & ATTR_CTIME)) { set_time = TRUE; - cFYI(1, (" CIFS - CTIME changed ")); /* BB probably no need */ + /* Although Samba throws this field away + it may be useful to Windows - but we do + not want to set ctime unless some other + timestamp is changing */ + cFYI(1, ("CIFS - CTIME changed ")); time_buf.ChangeTime = cpu_to_le64(cifs_UnixTimeToNT(attrs->ia_ctime)); } else time_buf.ChangeTime = 0; if (set_time || time_buf.Attributes) { - /* BB what if setting one attribute fails (such as size) but - time setting works? */ time_buf.CreationTime = 0; /* do not change */ /* In the future we should experiment - try setting timestamps via Handle (SetFileInfo) instead of by path */ @@ -1133,12 +1336,21 @@ int cifs_setattr(struct dentry *direntry, struct iattr *attrs) &time_buf, cifs_sb->local_nls); */ } } + /* Even if error on time set, no sense failing the call if + the server would set the time to a reasonable value anyway, + and this check ensures that we are not being called from + sys_utimes in which case we ought to fail the call back to + the user when the server rejects the call */ + if((rc) && (attrs->ia_valid && + (ATTR_MODE | ATTR_GID | ATTR_UID | ATTR_SIZE))) + rc = 0; } /* do not need local check to inode_check_ok since the server does that */ if (!rc) rc = inode_setattr(direntry->d_inode, attrs); +cifs_setattr_exit: kfree(full_path); FreeXid(xid); return rc; diff --git a/fs/cifs/link.c b/fs/cifs/link.c index ab925ef4f863..0f99aae33162 100644 --- a/fs/cifs/link.c +++ b/fs/cifs/link.c @@ -84,10 +84,8 @@ cifs_hardlink(struct dentry *old_file, struct inode *inode, cifsInode->time = 0; /* will force revalidate to go get info when needed */ cifs_hl_exit: - if (fromName) - kfree(fromName); - if (toName) - kfree(toName); + kfree(fromName); + kfree(toName); FreeXid(xid); return rc; } @@ -198,13 +196,15 @@ cifs_symlink(struct inode *inode, struct dentry *direntry, const char *symname) ("Create symlink worked but get_inode_info failed with rc = %d ", rc)); } else { - direntry->d_op = &cifs_dentry_ops; + if (pTcon->nocase) + direntry->d_op = &cifs_ci_dentry_ops; + else + direntry->d_op = &cifs_dentry_ops; d_instantiate(direntry, newinode); } } - if (full_path) - kfree(full_path); + kfree(full_path); FreeXid(xid); return rc; } @@ -250,8 +250,7 @@ cifs_readlink(struct dentry *direntry, char __user *pBuffer, int buflen) len = buflen; tmpbuffer = kmalloc(len,GFP_KERNEL); if(tmpbuffer == NULL) { - if (full_path) - kfree(full_path); + kfree(full_path); FreeXid(xid); return -ENOMEM; } @@ -300,8 +299,7 @@ cifs_readlink(struct dentry *direntry, char __user *pBuffer, int buflen) strncpy(tmpbuffer, referrals, len-1); } } - if(referrals) - kfree(referrals); + kfree(referrals); kfree(tmp_path); } /* BB add code like else decode referrals then memcpy to @@ -320,12 +318,8 @@ cifs_readlink(struct dentry *direntry, char __user *pBuffer, int buflen) rc)); } - if (tmpbuffer) { - kfree(tmpbuffer); - } - if (full_path) { - kfree(full_path); - } + kfree(tmpbuffer); + kfree(full_path); FreeXid(xid); return rc; } diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index 20ae4153f791..94baf6c8ecbd 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -34,8 +34,6 @@ extern mempool_t *cifs_sm_req_poolp; extern mempool_t *cifs_req_poolp; extern struct task_struct * oplockThread; -static __u16 GlobalMid; /* multiplex id - rotating counter */ - /* The xid serves as a useful identifier for each incoming vfs request, in a similar way to the mid which is useful to track each sent smb, and CurrentXid can also provide a running counter (although it @@ -51,6 +49,8 @@ _GetXid(void) GlobalTotalActiveXid++; if (GlobalTotalActiveXid > GlobalMaxActiveXid) GlobalMaxActiveXid = GlobalTotalActiveXid; /* keep high water mark for number of simultaneous vfs ops in our filesystem */ + if(GlobalTotalActiveXid > 65000) + cFYI(1,("warning: more than 65000 requests active")); xid = GlobalCurrentXid++; spin_unlock(&GlobalMid_Lock); return xid; @@ -98,14 +98,10 @@ sesInfoFree(struct cifsSesInfo *buf_to_free) atomic_dec(&sesInfoAllocCount); list_del(&buf_to_free->cifsSessionList); write_unlock(&GlobalSMBSeslock); - if (buf_to_free->serverOS) - kfree(buf_to_free->serverOS); - if (buf_to_free->serverDomain) - kfree(buf_to_free->serverDomain); - if (buf_to_free->serverNOS) - kfree(buf_to_free->serverNOS); - if (buf_to_free->password) - kfree(buf_to_free->password); + kfree(buf_to_free->serverOS); + kfree(buf_to_free->serverDomain); + kfree(buf_to_free->serverNOS); + kfree(buf_to_free->password); kfree(buf_to_free); } @@ -144,8 +140,7 @@ tconInfoFree(struct cifsTconInfo *buf_to_free) atomic_dec(&tconInfoAllocCount); list_del(&buf_to_free->cifsConnectionList); write_unlock(&GlobalSMBSeslock); - if (buf_to_free->nativeFileSystem) - kfree(buf_to_free->nativeFileSystem); + kfree(buf_to_free->nativeFileSystem); kfree(buf_to_free); } @@ -218,6 +213,76 @@ cifs_small_buf_release(void *buf_to_free) return; } +/* + Find a free multiplex id (SMB mid). Otherwise there could be + mid collisions which might cause problems, demultiplexing the + wrong response to this request. Multiplex ids could collide if + one of a series requests takes much longer than the others, or + if a very large number of long lived requests (byte range + locks or FindNotify requests) are pending. No more than + 64K-1 requests can be outstanding at one time. If no + mids are available, return zero. A future optimization + could make the combination of mids and uid the key we use + to demultiplex on (rather than mid alone). + In addition to the above check, the cifs demultiplex + code already used the command code as a secondary + check of the frame and if signing is negotiated the + response would be discarded if the mid were the same + but the signature was wrong. Since the mid is not put in the + pending queue until later (when it is about to be dispatched) + we do have to limit the number of outstanding requests + to somewhat less than 64K-1 although it is hard to imagine + so many threads being in the vfs at one time. +*/ +__u16 GetNextMid(struct TCP_Server_Info *server) +{ + __u16 mid = 0; + __u16 last_mid; + int collision; + + if(server == NULL) + return mid; + + spin_lock(&GlobalMid_Lock); + last_mid = server->CurrentMid; /* we do not want to loop forever */ + server->CurrentMid++; + /* This nested loop looks more expensive than it is. + In practice the list of pending requests is short, + fewer than 50, and the mids are likely to be unique + on the first pass through the loop unless some request + takes longer than the 64 thousand requests before it + (and it would also have to have been a request that + did not time out) */ + while(server->CurrentMid != last_mid) { + struct list_head *tmp; + struct mid_q_entry *mid_entry; + + collision = 0; + if(server->CurrentMid == 0) + server->CurrentMid++; + + list_for_each(tmp, &server->pending_mid_q) { + mid_entry = list_entry(tmp, struct mid_q_entry, qhead); + + if ((mid_entry->mid == server->CurrentMid) && + (mid_entry->midState == MID_REQUEST_SUBMITTED)) { + /* This mid is in use, try a different one */ + collision = 1; + break; + } + } + if(collision == 0) { + mid = server->CurrentMid; + break; + } + server->CurrentMid++; + } + spin_unlock(&GlobalMid_Lock); + return mid; +} + +/* NB: MID can not be set if treeCon not passed in, in that + case it is responsbility of caller to set the mid */ void header_assemble(struct smb_hdr *buffer, char smb_command /* command */ , const struct cifsTconInfo *treeCon, int word_count @@ -233,7 +298,8 @@ header_assemble(struct smb_hdr *buffer, char smb_command /* command */ , (2 * word_count) + sizeof (struct smb_hdr) - 4 /* RFC 1001 length field does not count */ + 2 /* for bcc field itself */ ; - /* Note that this is the only network field that has to be converted to big endian and it is done just before we send it */ + /* Note that this is the only network field that has to be converted + to big endian and it is done just before we send it */ buffer->Protocol[0] = 0xFF; buffer->Protocol[1] = 'S'; @@ -245,8 +311,6 @@ header_assemble(struct smb_hdr *buffer, char smb_command /* command */ , buffer->Pid = cpu_to_le16((__u16)current->tgid); buffer->PidHigh = cpu_to_le16((__u16)(current->tgid >> 16)); spin_lock(&GlobalMid_Lock); - GlobalMid++; - buffer->Mid = GlobalMid; spin_unlock(&GlobalMid_Lock); if (treeCon) { buffer->Tid = treeCon->tid; @@ -256,8 +320,9 @@ header_assemble(struct smb_hdr *buffer, char smb_command /* command */ , if (treeCon->ses->capabilities & CAP_STATUS32) { buffer->Flags2 |= SMBFLG2_ERR_STATUS; } - - buffer->Uid = treeCon->ses->Suid; /* always in LE format */ + /* Uid is not converted */ + buffer->Uid = treeCon->ses->Suid; + buffer->Mid = GetNextMid(treeCon->ses->server); if(multiuser_mount != 0) { /* For the multiuser case, there are few obvious technically */ /* possible mechanisms to match the local linux user (uid) */ @@ -305,6 +370,8 @@ header_assemble(struct smb_hdr *buffer, char smb_command /* command */ , } if (treeCon->Flags & SMB_SHARE_IS_IN_DFS) buffer->Flags2 |= SMBFLG2_DFS; + if (treeCon->nocase) + buffer->Flags |= SMBFLG_CASELESS; if((treeCon->ses) && (treeCon->ses->server)) if(treeCon->ses->server->secMode & (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) @@ -330,12 +397,12 @@ checkSMBhdr(struct smb_hdr *smb, __u16 mid) if(smb->Command == SMB_COM_LOCKING_ANDX) return 0; else - cERROR(1, ("Rcvd Request not response ")); + cERROR(1, ("Rcvd Request not response")); } } else { /* bad signature or mid */ if (*(__le32 *) smb->Protocol != cpu_to_le32(0x424d53ff)) cERROR(1, - ("Bad protocol string signature header %x ", + ("Bad protocol string signature header %x", *(unsigned int *) smb->Protocol)); if (mid != smb->Mid) cERROR(1, ("Mids do not match")); @@ -347,9 +414,10 @@ checkSMBhdr(struct smb_hdr *smb, __u16 mid) int checkSMB(struct smb_hdr *smb, __u16 mid, int length) { - __u32 len = be32_to_cpu(smb->smb_buf_length); + __u32 len = smb->smb_buf_length; + __u32 clc_len; /* calculated length */ cFYI(0, - ("Entering checkSMB with Length: %x, smb_buf_length: %x ", + ("Entering checkSMB with Length: %x, smb_buf_length: %x", length, len)); if (((unsigned int)length < 2 + sizeof (struct smb_hdr)) || (len > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE - 4)) { @@ -368,23 +436,36 @@ checkSMB(struct smb_hdr *smb, __u16 mid, int length) cERROR(1, ("smb_buf_length greater than MaxBufSize")); cERROR(1, - ("bad smb detected. Illegal length. The mid=%d", + ("bad smb detected. Illegal length. mid=%d", smb->Mid)); return 1; } if (checkSMBhdr(smb, mid)) return 1; - - if ((4 + len != smbCalcSize(smb)) + clc_len = smbCalcSize_LE(smb); + if ((4 + len != clc_len) || (4 + len != (unsigned int)length)) { - return 0; - } else { - cERROR(1, ("smbCalcSize %x ", smbCalcSize(smb))); - cERROR(1, - ("bad smb size detected. The Mid=%d", smb->Mid)); - return 1; + cERROR(1, ("Calculated size 0x%x vs actual length 0x%x", + clc_len, 4 + len)); + cERROR(1, ("bad smb size detected for Mid=%d", smb->Mid)); + /* Windows XP can return a few bytes too much, presumably + an illegal pad, at the end of byte range lock responses + so we allow for that three byte pad, as long as actual + received length is as long or longer than calculated length */ + /* We have now had to extend this more, since there is a + case in which it needs to be bigger still to handle a + malformed response to transact2 findfirst from WinXP when + access denied is returned and thus bcc and wct are zero + but server says length is 0x21 bytes too long as if the server + forget to reset the smb rfc1001 length when it reset the + wct and bcc to minimum size and drop the t2 parms and data */ + if((4+len > clc_len) && (len <= clc_len + 512)) + return 0; + else + return 1; } + return 0; } int is_valid_oplock_break(struct smb_hdr *buf) @@ -448,9 +529,7 @@ is_valid_oplock_break(struct smb_hdr *buf) list_for_each(tmp, &GlobalTreeConnectionList) { tcon = list_entry(tmp, struct cifsTconInfo, cifsConnectionList); if (tcon->tid == buf->Tid) { -#ifdef CONFIG_CIFS_STATS - atomic_inc(&tcon->num_oplock_brks); -#endif + cifs_stats_inc(&tcon->num_oplock_brks); list_for_each(tmp1,&tcon->openFileList){ netfile = list_entry(tmp1,struct cifsFileInfo, tlist); @@ -603,9 +682,10 @@ cifsConvertToUCS(__le16 * target, const char *source, int maxlen, int i,j,charlen; int len_remaining = maxlen; char src_char; + __u16 temp; if(!mapChars) - return cifs_strtoUCS((wchar_t *) target, source, PATH_MAX, cp); + return cifs_strtoUCS(target, source, PATH_MAX, cp); for(i = 0, j = 0; i < maxlen; j++) { src_char = source[i]; @@ -639,13 +719,14 @@ cifsConvertToUCS(__le16 * target, const char *source, int maxlen, break;*/ default: charlen = cp->char2uni(source+i, - len_remaining, target+j); + len_remaining, &temp); /* if no match, use question mark, which at least in some cases servers as wild card */ if(charlen < 1) { target[j] = cpu_to_le16(0x003f); charlen = 1; - } + } else + target[j] = cpu_to_le16(temp); len_remaining -= charlen; /* character may take more than one byte in the the source string, but will take exactly two diff --git a/fs/cifs/netmisc.c b/fs/cifs/netmisc.c index a92af41d4411..5de74d216fdd 100644 --- a/fs/cifs/netmisc.c +++ b/fs/cifs/netmisc.c @@ -133,7 +133,6 @@ static const struct smb_to_posix_error mapping_table_ERRHRD[] = { int cifs_inet_pton(int address_family, char *cp,void *dst) { - struct in_addr address; int value; int digit; int i; @@ -190,8 +189,7 @@ cifs_inet_pton(int address_family, char *cp,void *dst) if (value > addr_class_max[end - bytes]) return 0; - address.s_addr = *((__be32 *) bytes) | htonl(value); - *((__be32 *)dst) = address.s_addr; + *((__be32 *)dst) = *((__be32 *) bytes) | htonl(value); return 1; /* success */ } @@ -332,7 +330,7 @@ static const struct { ERRHRD, ERRgeneral, NT_STATUS_ACCOUNT_RESTRICTION}, { ERRSRV, 2241, NT_STATUS_INVALID_LOGON_HOURS}, { ERRSRV, 2240, NT_STATUS_INVALID_WORKSTATION}, { - ERRSRV, 2242, NT_STATUS_PASSWORD_EXPIRED}, { + ERRSRV, ERRpasswordExpired, NT_STATUS_PASSWORD_EXPIRED}, { ERRSRV, 2239, NT_STATUS_ACCOUNT_DISABLED}, { ERRHRD, ERRgeneral, NT_STATUS_NONE_MAPPED}, { ERRHRD, ERRgeneral, NT_STATUS_TOO_MANY_LUIDS_REQUESTED}, { @@ -678,7 +676,7 @@ static const struct { ERRDOS, 193, NT_STATUS_IMAGE_CHECKSUM_MISMATCH}, { ERRHRD, ERRgeneral, NT_STATUS_LOST_WRITEBEHIND_DATA}, { ERRHRD, ERRgeneral, NT_STATUS_CLIENT_SERVER_PARAMETERS_INVALID}, { - ERRSRV, 2242, NT_STATUS_PASSWORD_MUST_CHANGE}, { + ERRSRV, ERRpasswordExpired, NT_STATUS_PASSWORD_MUST_CHANGE}, { ERRHRD, ERRgeneral, NT_STATUS_NOT_FOUND}, { ERRHRD, ERRgeneral, NT_STATUS_NOT_TINY_STREAM}, { ERRHRD, ERRgeneral, NT_STATUS_RECOVERY_FAILURE}, { @@ -815,7 +813,7 @@ map_smb_to_linux_error(struct smb_hdr *smb) if (smb->Flags2 & SMBFLG2_ERR_STATUS) { /* translate the newer STATUS codes to old style errors and then to POSIX errors */ __u32 err = le32_to_cpu(smb->Status.CifsError); - if(cifsFYI) + if(cifsFYI & CIFS_RC) cifs_print_status(err); ntstatus_to_dos(err, &smberrclass, &smberrcode); } else { @@ -870,7 +868,14 @@ unsigned int smbCalcSize(struct smb_hdr *ptr) { return (sizeof (struct smb_hdr) + (2 * ptr->WordCount) + - BCC(ptr)); + 2 /* size of the bcc field */ + BCC(ptr)); +} + +unsigned int +smbCalcSize_LE(struct smb_hdr *ptr) +{ + return (sizeof (struct smb_hdr) + (2 * ptr->WordCount) + + 2 /* size of the bcc field */ + le16_to_cpu(BCC_LE(ptr))); } /* The following are taken from fs/ntfs/util.c */ diff --git a/fs/cifs/ntlmssp.h b/fs/cifs/ntlmssp.h index 6facb41117a3..803389b64a2c 100644 --- a/fs/cifs/ntlmssp.h +++ b/fs/cifs/ntlmssp.h @@ -19,8 +19,6 @@ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ -#pragma pack(1) - #define NTLMSSP_SIGNATURE "NTLMSSP" /* Message Types */ #define NtLmNegotiate cpu_to_le32(1) @@ -63,7 +61,7 @@ typedef struct _SECURITY_BUFFER { __le16 Length; __le16 MaximumLength; __le32 Buffer; /* offset to buffer */ -} SECURITY_BUFFER; +} __attribute__((packed)) SECURITY_BUFFER; typedef struct _NEGOTIATE_MESSAGE { __u8 Signature[sizeof (NTLMSSP_SIGNATURE)]; @@ -73,7 +71,7 @@ typedef struct _NEGOTIATE_MESSAGE { SECURITY_BUFFER WorkstationName; /* RFC 1001 and ASCII */ char DomainString[0]; /* followed by WorkstationString */ -} NEGOTIATE_MESSAGE, *PNEGOTIATE_MESSAGE; +} __attribute__((packed)) NEGOTIATE_MESSAGE, *PNEGOTIATE_MESSAGE; typedef struct _CHALLENGE_MESSAGE { __u8 Signature[sizeof (NTLMSSP_SIGNATURE)]; @@ -83,7 +81,7 @@ typedef struct _CHALLENGE_MESSAGE { __u8 Challenge[CIFS_CRYPTO_KEY_SIZE]; __u8 Reserved[8]; SECURITY_BUFFER TargetInfoArray; -} CHALLENGE_MESSAGE, *PCHALLENGE_MESSAGE; +} __attribute__((packed)) CHALLENGE_MESSAGE, *PCHALLENGE_MESSAGE; typedef struct _AUTHENTICATE_MESSAGE { __u8 Signature[sizeof (NTLMSSP_SIGNATURE)]; @@ -96,6 +94,4 @@ typedef struct _AUTHENTICATE_MESSAGE { SECURITY_BUFFER SessionKey; __le32 NegotiateFlags; char UserString[0]; -} AUTHENTICATE_MESSAGE, *PAUTHENTICATE_MESSAGE; - -#pragma pack() /* resume default structure packing */ +} __attribute__((packed)) AUTHENTICATE_MESSAGE, *PAUTHENTICATE_MESSAGE; diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c index 22557716f9af..9bdaaecae36f 100644 --- a/fs/cifs/readdir.c +++ b/fs/cifs/readdir.c @@ -91,7 +91,10 @@ static int construct_dentry(struct qstr *qstring, struct file *file, } *ptmp_inode = new_inode(file->f_dentry->d_sb); - tmp_dentry->d_op = &cifs_dentry_ops; + if (pTcon->nocase) + tmp_dentry->d_op = &cifs_ci_dentry_ops; + else + tmp_dentry->d_op = &cifs_dentry_ops; if(*ptmp_inode == NULL) return rc; rc = 1; @@ -139,6 +142,11 @@ static void fill_in_inode(struct inode *tmp_inode, tmp_inode->i_gid = cifs_sb->mnt_gid; /* set default mode. will override for dirs below */ tmp_inode->i_mode = cifs_sb->mnt_file_mode; + } else { + /* mask off the type bits since it gets set + below and we do not want to get two type + bits set */ + tmp_inode->i_mode &= ~S_IFMT; } if (attr & ATTR_DIRECTORY) { @@ -148,6 +156,19 @@ static void fill_in_inode(struct inode *tmp_inode, tmp_inode->i_mode = cifs_sb->mnt_dir_mode; } tmp_inode->i_mode |= S_IFDIR; + } else if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UNX_EMUL) && + (attr & ATTR_SYSTEM)) { + if (end_of_file == 0) { + *pobject_type = DT_FIFO; + tmp_inode->i_mode |= S_IFIFO; + } else { + /* rather than get the type here, we mark the + inode as needing revalidate and get the real type + (blk vs chr vs. symlink) later ie in lookup */ + *pobject_type = DT_REG; + tmp_inode->i_mode |= S_IFREG; + cifsInfo->time = 0; + } /* we no longer mark these because we could not follow them */ /* } else if (attr & ATTR_REPARSE) { *pobject_type = DT_LNK; @@ -183,15 +204,27 @@ static void fill_in_inode(struct inode *tmp_inode, if (S_ISREG(tmp_inode->i_mode)) { cFYI(1, ("File inode")); tmp_inode->i_op = &cifs_file_inode_ops; - if(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DIRECT_IO) - tmp_inode->i_fop = &cifs_file_direct_ops; + if(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DIRECT_IO) { + if(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL) + tmp_inode->i_fop = &cifs_file_direct_nobrl_ops; + else + tmp_inode->i_fop = &cifs_file_direct_ops; + + } else if(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL) + tmp_inode->i_fop = &cifs_file_nobrl_ops; else tmp_inode->i_fop = &cifs_file_ops; + if(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL) + tmp_inode->i_fop->lock = NULL; tmp_inode->i_data.a_ops = &cifs_addr_ops; - + if((cifs_sb->tcon) && (cifs_sb->tcon->ses) && + (cifs_sb->tcon->ses->server->maxBuf < + 4096 + MAX_CIFS_HDR_SIZE)) + tmp_inode->i_data.a_ops->readpages = NULL; if(isNewInode) - return; /* No sense invalidating pages for new inode since we - have not started caching readahead file data yet */ + return; /* No sense invalidating pages for new inode + since have not started caching readahead file + data yet */ if (timespec_equal(&tmp_inode->i_mtime, &local_mtime) && (local_size == tmp_inode->i_size)) { @@ -242,6 +275,9 @@ static void unix_fill_in_inode(struct inode *tmp_inode, cifs_NTtimeToUnix(le64_to_cpu(pfindData->LastStatusChange)); tmp_inode->i_mode = le64_to_cpu(pfindData->Permissions); + /* since we set the inode type below we need to mask off type + to avoid strange results if bits above were corrupt */ + tmp_inode->i_mode &= ~S_IFMT; if (type == UNIX_FILE) { *pobject_type = DT_REG; tmp_inode->i_mode |= S_IFREG; @@ -267,6 +303,11 @@ static void unix_fill_in_inode(struct inode *tmp_inode, } else if (type == UNIX_SOCKET) { *pobject_type = DT_SOCK; tmp_inode->i_mode |= S_IFSOCK; + } else { + /* safest to just call it a file */ + *pobject_type = DT_REG; + tmp_inode->i_mode |= S_IFREG; + cFYI(1,("unknown inode type %d",type)); } tmp_inode->i_uid = le64_to_cpu(pfindData->Uid); @@ -290,7 +331,13 @@ static void unix_fill_in_inode(struct inode *tmp_inode, tmp_inode->i_fop = &cifs_file_direct_ops; else tmp_inode->i_fop = &cifs_file_ops; + if(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_BRL) + tmp_inode->i_fop->lock = NULL; tmp_inode->i_data.a_ops = &cifs_addr_ops; + if((cifs_sb->tcon) && (cifs_sb->tcon->ses) && + (cifs_sb->tcon->ses->server->maxBuf < + 4096 + MAX_CIFS_HDR_SIZE)) + tmp_inode->i_data.a_ops->readpages = NULL; if(isNewInode) return; /* No sense invalidating pages for new inode since we @@ -374,7 +421,8 @@ ffirst_retry: rc = CIFSFindFirst(xid, pTcon,full_path,cifs_sb->local_nls, &cifsFile->netfid, &cifsFile->srch_inf, - cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); + cifs_sb->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR, CIFS_DIR_SEP(cifs_sb)); if(rc == 0) cifsFile->invalidHandle = FALSE; if((rc == -EOPNOTSUPP) && @@ -491,6 +539,30 @@ static int cifs_entry_is_dot(char *current_entry, struct cifsFileInfo *cfile) return rc; } +/* Check if directory that we are searching has changed so we can decide + whether we can use the cached search results from the previous search */ +static int is_dir_changed(struct file * file) +{ + struct inode * inode; + struct cifsInodeInfo *cifsInfo; + + if(file->f_dentry == NULL) + return 0; + + inode = file->f_dentry->d_inode; + + if(inode == NULL) + return 0; + + cifsInfo = CIFS_I(inode); + + if(cifsInfo->time == 0) + return 1; /* directory was changed, perhaps due to unlink */ + else + return 0; + +} + /* find the corresponding entry in the search */ /* Note that the SMB server returns search entries for . and .. which complicates logic here if we choose to parse for them and we do not @@ -507,7 +579,8 @@ static int find_cifs_entry(const int xid, struct cifsTconInfo *pTcon, struct cifsFileInfo * cifsFile = file->private_data; /* check if index in the buffer */ - if((cifsFile == NULL) || (ppCurrentEntry == NULL) || (num_to_ret == NULL)) + if((cifsFile == NULL) || (ppCurrentEntry == NULL) || + (num_to_ret == NULL)) return -ENOENT; *ppCurrentEntry = NULL; @@ -515,7 +588,9 @@ static int find_cifs_entry(const int xid, struct cifsTconInfo *pTcon, cifsFile->srch_inf.index_of_last_entry - cifsFile->srch_inf.entries_in_buffer; /* dump_cifs_file_struct(file, "In fce ");*/ - if(index_to_find < first_entry_in_buffer) { + if(((index_to_find < cifsFile->srch_inf.index_of_last_entry) && + is_dir_changed(file)) || + (index_to_find < first_entry_in_buffer)) { /* close and restart search */ cFYI(1,("search backing up - close and restart search")); cifsFile->invalidHandle = TRUE; @@ -536,7 +611,8 @@ static int find_cifs_entry(const int xid, struct cifsTconInfo *pTcon, while((index_to_find >= cifsFile->srch_inf.index_of_last_entry) && (rc == 0) && (cifsFile->srch_inf.endOfSearch == FALSE)){ cFYI(1,("calling findnext2")); - rc = CIFSFindNext(xid,pTcon,cifsFile->netfid, &cifsFile->srch_inf); + rc = CIFSFindNext(xid,pTcon,cifsFile->netfid, + &cifsFile->srch_inf); if(rc) return -ENOENT; } @@ -548,14 +624,13 @@ static int find_cifs_entry(const int xid, struct cifsTconInfo *pTcon, char * end_of_smb = cifsFile->srch_inf.ntwrk_buf_start + smbCalcSize((struct smb_hdr *) cifsFile->srch_inf.ntwrk_buf_start); -/* dump_cifs_file_struct(file,"found entry in fce "); */ first_entry_in_buffer = cifsFile->srch_inf.index_of_last_entry - cifsFile->srch_inf.entries_in_buffer; pos_in_buf = index_to_find - first_entry_in_buffer; cFYI(1,("found entry - pos_in_buf %d",pos_in_buf)); current_entry = cifsFile->srch_inf.srch_entries_start; for(i=0;(i<(pos_in_buf)) && (current_entry != NULL);i++) { - /* go entry to next entry figuring out which we need to start with */ + /* go entry by entry figuring out which is first */ /* if( . or ..) skip */ rc = cifs_entry_is_dot(current_entry,cifsFile); @@ -582,11 +657,10 @@ static int find_cifs_entry(const int xid, struct cifsTconInfo *pTcon, } if(pos_in_buf >= cifsFile->srch_inf.entries_in_buffer) { - cFYI(1,("can not return entries when pos_in_buf beyond last entry")); + cFYI(1,("can not return entries pos_in_buf beyond last entry")); *num_to_ret = 0; } else *num_to_ret = cifsFile->srch_inf.entries_in_buffer - pos_in_buf; -/* dump_cifs_file_struct(file, "end fce ");*/ return rc; } @@ -650,7 +724,7 @@ static int cifs_get_name_from_search_buf(struct qstr *pqst, (__le16 *)filename, len/2, nlt); else pqst->len = cifs_strfromUCS_le((char *)pqst->name, - (wchar_t *)filename,len/2,nlt); + (__le16 *)filename,len/2,nlt); } else { pqst->name = filename; pqst->len = len; @@ -721,7 +795,8 @@ static int cifs_filldir(char *pfindEntry, struct file *file, (FILE_DIRECTORY_INFO *)pfindEntry,&obj_type, rc); } - rc = filldir(direntry,qstring.name,qstring.len,file->f_pos,tmp_inode->i_ino,obj_type); + rc = filldir(direntry,qstring.name,qstring.len,file->f_pos, + tmp_inode->i_ino,obj_type); if(rc) { cFYI(1,("filldir rc = %d",rc)); } @@ -805,15 +880,12 @@ int cifs_readdir(struct file *file, void *direntry, filldir_t filldir) FreeXid(xid); return -EIO; } -/* dump_cifs_file_struct(file, "Begin rdir "); */ cifs_sb = CIFS_SB(file->f_dentry->d_sb); pTcon = cifs_sb->tcon; if(pTcon == NULL) return -EINVAL; -/* cFYI(1,("readdir2 pos: %lld",file->f_pos)); */ - switch ((int) file->f_pos) { case 0: /*if (filldir(direntry, ".", 1, file->f_pos, @@ -866,7 +938,6 @@ int cifs_readdir(struct file *file, void *direntry, filldir_t filldir) cifsFile->search_resume_name = NULL; */ /* BB account for . and .. in f_pos as special case */ - /* dump_cifs_file_struct(file, "rdir after default ");*/ rc = find_cifs_entry(xid,pTcon, file, ¤t_entry,&num_to_fill); @@ -906,14 +977,14 @@ int cifs_readdir(struct file *file, void *direntry, filldir_t filldir) cifs_save_resume_key(current_entry,cifsFile); break; } else - current_entry = nxt_dir_entry(current_entry,end_of_smb); + current_entry = nxt_dir_entry(current_entry, + end_of_smb); } kfree(tmp_buf); break; } /* end switch */ rddir2_exit: - /* dump_cifs_file_struct(file, "end rdir "); */ FreeXid(xid); return rc; } diff --git a/fs/cifs/rfc1002pdu.h b/fs/cifs/rfc1002pdu.h index 806c0ed06da9..9222033cad8e 100644 --- a/fs/cifs/rfc1002pdu.h +++ b/fs/cifs/rfc1002pdu.h @@ -21,8 +21,6 @@ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ -#pragma pack(1) - /* NB: unlike smb/cifs packets, the RFC1002 structures are big endian */ /* RFC 1002 session packet types */ @@ -48,17 +46,17 @@ struct rfc1002_session_packet { __u8 calling_len; __u8 calling_name[32]; __u8 scope2; /* null */ - } session_req; + } __attribute__((packed)) session_req; struct { __u32 retarget_ip_addr; __u16 port; - } retarget_resp; + } __attribute__((packed)) retarget_resp; __u8 neg_ses_resp_error_code; /* POSITIVE_SESSION_RESPONSE packet does not include trailer. SESSION_KEEP_ALIVE packet also does not include a trailer. Trailer for the SESSION_MESSAGE packet is SMB/CIFS header */ - } trailer; -}; + } __attribute__((packed)) trailer; +} __attribute__((packed)); /* Negative Session Response error codes */ #define RFC1002_NOT_LISTENING_CALLED 0x80 /* not listening on called name */ @@ -74,6 +72,3 @@ server netbios name). Currently server names are resolved only via DNS (tcp name) or ip address or an /etc/hosts equivalent mapping to ip address.*/ #define DEFAULT_CIFS_CALLED_NAME "*SMBSERVER " - -#pragma pack() /* resume default structure packing */ - diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index 0046c219833d..f8871196098c 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -49,7 +49,8 @@ AllocMidQEntry(struct smb_hdr *smb_buffer, struct cifsSesInfo *ses) return NULL; } - temp = (struct mid_q_entry *) mempool_alloc(cifs_mid_poolp,SLAB_KERNEL | SLAB_NOFS); + temp = (struct mid_q_entry *) mempool_alloc(cifs_mid_poolp, + SLAB_KERNEL | SLAB_NOFS); if (temp == NULL) return temp; else { @@ -58,7 +59,9 @@ AllocMidQEntry(struct smb_hdr *smb_buffer, struct cifsSesInfo *ses) temp->pid = current->pid; temp->command = smb_buffer->Command; cFYI(1, ("For smb_command %d", temp->command)); - do_gettimeofday(&temp->when_sent); + /* do_gettimeofday(&temp->when_sent);*/ /* easier to use jiffies */ + /* when mid allocated can be before when sent */ + temp->when_alloc = jiffies; temp->ses = ses; temp->tsk = current; } @@ -74,6 +77,9 @@ AllocMidQEntry(struct smb_hdr *smb_buffer, struct cifsSesInfo *ses) static void DeleteMidQEntry(struct mid_q_entry *midEntry) { +#ifdef CONFIG_CIFS_STATS2 + unsigned long now; +#endif spin_lock(&GlobalMid_Lock); midEntry->midState = MID_FREE; list_del(&midEntry->qhead); @@ -83,6 +89,22 @@ DeleteMidQEntry(struct mid_q_entry *midEntry) cifs_buf_release(midEntry->resp_buf); else cifs_small_buf_release(midEntry->resp_buf); +#ifdef CONFIG_CIFS_STATS2 + now = jiffies; + /* commands taking longer than one second are indications that + something is wrong, unless it is quite a slow link or server */ + if((now - midEntry->when_alloc) > HZ) { + if((cifsFYI & CIFS_TIMER) && + (midEntry->command != SMB_COM_LOCKING_ANDX)) { + printk(KERN_DEBUG " CIFS slow rsp: cmd %d mid %d", + midEntry->command, midEntry->mid); + printk(" A: 0x%lx S: 0x%lx R: 0x%lx\n", + now - midEntry->when_alloc, + now - midEntry->when_sent, + now - midEntry->when_received); + } + } +#endif mempool_free(midEntry, cifs_mid_poolp); } @@ -146,32 +168,37 @@ smb_send(struct socket *ssocket, struct smb_hdr *smb_buffer, Flags2 is converted in SendReceive */ smb_buffer->smb_buf_length = cpu_to_be32(smb_buffer->smb_buf_length); - cFYI(1, ("Sending smb of length %d ", smb_buf_length)); + cFYI(1, ("Sending smb of length %d", smb_buf_length)); dump_smb(smb_buffer, len); while (len > 0) { rc = kernel_sendmsg(ssocket, &smb_msg, &iov, 1, len); if ((rc == -ENOSPC) || (rc == -EAGAIN)) { i++; - if(i > 60) { + /* smaller timeout here than send2 since smaller size */ + /* Although it may not be required, this also is smaller + oplock break time */ + if(i > 12) { cERROR(1, - ("sends on sock %p stuck for 30 seconds", + ("sends on sock %p stuck for 7 seconds", ssocket)); rc = -EAGAIN; break; } - msleep(500); + msleep(1 << i); continue; } if (rc < 0) break; + else + i = 0; /* reset i after each successful send */ iov.iov_base += rc; iov.iov_len -= rc; len -= rc; } if (rc < 0) { - cERROR(1,("Error %d sending data on socket to server.", rc)); + cERROR(1,("Error %d sending data on socket to server", rc)); } else { rc = 0; } @@ -179,26 +206,21 @@ smb_send(struct socket *ssocket, struct smb_hdr *smb_buffer, return rc; } -#ifdef CIFS_EXPERIMENTAL -/* BB finish off this function, adding support for writing set of pages as iovec */ -/* and also adding support for operations that need to parse the response smb */ - -int -smb_sendv(struct socket *ssocket, struct smb_hdr *smb_buffer, - unsigned int smb_buf_length, struct kvec * write_vector - /* page list */, struct sockaddr *sin) +#ifdef CONFIG_CIFS_EXPERIMENTAL +static int +smb_send2(struct socket *ssocket, struct kvec *iov, int n_vec, + struct sockaddr *sin) { int rc = 0; int i = 0; struct msghdr smb_msg; - number_of_pages += 1; /* account for SMB header */ - struct kvec * piov = kmalloc(number_of_pages * sizeof(struct kvec)); - unsigned len = smb_buf_length + 4; - + struct smb_hdr *smb_buffer = iov[0].iov_base; + unsigned int len = iov[0].iov_len; + unsigned int total_len; + int first_vec = 0; + if(ssocket == NULL) return -ENOTSOCK; /* BB eventually add reconnect code here */ - iov.iov_base = smb_buffer; - iov.iov_len = len; smb_msg.msg_name = sin; smb_msg.msg_namelen = sizeof (struct sockaddr); @@ -211,49 +233,80 @@ smb_sendv(struct socket *ssocket, struct smb_hdr *smb_buffer, cifssmb.c and RFC1001 len is converted to bigendian in smb_send Flags2 is converted in SendReceive */ + + total_len = 0; + for (i = 0; i < n_vec; i++) + total_len += iov[i].iov_len; + smb_buffer->smb_buf_length = cpu_to_be32(smb_buffer->smb_buf_length); - cFYI(1, ("Sending smb of length %d ", smb_buf_length)); + cFYI(1, ("Sending smb: total_len %d", total_len)); dump_smb(smb_buffer, len); - while (len > 0) { - rc = kernel_sendmsg(ssocket, &smb_msg, &iov, number_of_pages, - len); + while (total_len) { + rc = kernel_sendmsg(ssocket, &smb_msg, &iov[first_vec], + n_vec - first_vec, total_len); if ((rc == -ENOSPC) || (rc == -EAGAIN)) { i++; - if(i > 60) { + if(i >= 14) { cERROR(1, - ("sends on sock %p stuck for 30 seconds", + ("sends on sock %p stuck for 15 seconds", ssocket)); rc = -EAGAIN; break; } - msleep(500); + msleep(1 << i); continue; } if (rc < 0) break; - iov.iov_base += rc; - iov.iov_len -= rc; - len -= rc; + + if (rc >= total_len) { + WARN_ON(rc > total_len); + break; + } + if(rc == 0) { + /* should never happen, letting socket clear before + retrying is our only obvious option here */ + cERROR(1,("tcp sent no data")); + msleep(500); + continue; + } + total_len -= rc; + /* the line below resets i */ + for (i = first_vec; i < n_vec; i++) { + if (iov[i].iov_len) { + if (rc > iov[i].iov_len) { + rc -= iov[i].iov_len; + iov[i].iov_len = 0; + } else { + iov[i].iov_base += rc; + iov[i].iov_len -= rc; + first_vec = i; + break; + } + } + } + i = 0; /* in case we get ENOSPC on the next send */ } if (rc < 0) { - cERROR(1,("Error %d sending data on socket to server.", rc)); - } else { + cERROR(1,("Error %d sending data on socket to server", rc)); + } else rc = 0; - } return rc; } - int -CIFSSendRcv(const unsigned int xid, struct cifsSesInfo *ses, - struct smb_hdr *in_buf, struct kvec * write_vector /* page list */, int *pbytes_returned, const int long_op) +SendReceive2(const unsigned int xid, struct cifsSesInfo *ses, + struct kvec *iov, int n_vec, int *pbytes_returned, + const int long_op) { int rc = 0; - unsigned long timeout = 15 * HZ; - struct mid_q_entry *midQ = NULL; + unsigned int receive_len; + unsigned long timeout; + struct mid_q_entry *midQ; + struct smb_hdr *in_buf = iov[0].iov_base; if (ses == NULL) { cERROR(1,("Null smb session")); @@ -263,14 +316,8 @@ CIFSSendRcv(const unsigned int xid, struct cifsSesInfo *ses, cERROR(1,("Null tcp session")); return -EIO; } - if(pbytes_returned == NULL) - return -EIO; - else - *pbytes_returned = 0; - - - if(ses->server->tcpStatus == CIFS_EXITING) + if(ses->server->tcpStatus == CifsExiting) return -ENOENT; /* Ensure that we do not send more than 50 overlapping requests @@ -282,11 +329,18 @@ CIFSSendRcv(const unsigned int xid, struct cifsSesInfo *ses, } else { spin_lock(&GlobalMid_Lock); while(1) { - if(atomic_read(&ses->server->inFlight) >= cifs_max_pending){ + if(atomic_read(&ses->server->inFlight) >= + cifs_max_pending){ spin_unlock(&GlobalMid_Lock); +#ifdef CONFIG_CIFS_STATS2 + atomic_inc(&ses->server->num_waiters); +#endif wait_event(ses->server->request_q, atomic_read(&ses->server->inFlight) < cifs_max_pending); +#ifdef CONFIG_CIFS_STATS2 + atomic_dec(&ses->server->num_waiters); +#endif spin_lock(&GlobalMid_Lock); } else { if(ses->server->tcpStatus == CifsExiting) { @@ -314,17 +368,17 @@ CIFSSendRcv(const unsigned int xid, struct cifsSesInfo *ses, if (ses->server->tcpStatus == CifsExiting) { rc = -ENOENT; - goto cifs_out_label; + goto out_unlock2; } else if (ses->server->tcpStatus == CifsNeedReconnect) { cFYI(1,("tcp session dead - return to caller to retry")); rc = -EAGAIN; - goto cifs_out_label; + goto out_unlock2; } else if (ses->status != CifsGood) { /* check if SMB session is bad because we are setting it up */ if((in_buf->Command != SMB_COM_SESSION_SETUP_ANDX) && (in_buf->Command != SMB_COM_NEGOTIATE)) { rc = -EAGAIN; - goto cifs_out_label; + goto out_unlock2; } /* else ok - we are setting up session */ } midQ = AllocMidQEntry(in_buf, ses); @@ -338,51 +392,163 @@ CIFSSendRcv(const unsigned int xid, struct cifsSesInfo *ses, return -ENOMEM; } - if (in_buf->smb_buf_length > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE - 4) { - up(&ses->server->tcpSem); - cERROR(1, - ("Illegal length, greater than maximum frame, %d ", - in_buf->smb_buf_length)); +/* BB FIXME */ +/* rc = cifs_sign_smb2(iov, n_vec, ses->server, &midQ->sequence_number); */ + + midQ->midState = MID_REQUEST_SUBMITTED; +#ifdef CONFIG_CIFS_STATS2 + atomic_inc(&ses->server->inSend); +#endif + rc = smb_send2(ses->server->ssocket, iov, n_vec, + (struct sockaddr *) &(ses->server->addr.sockAddr)); +#ifdef CONFIG_CIFS_STATS2 + atomic_dec(&ses->server->inSend); + midQ->when_sent = jiffies; +#endif + if(rc < 0) { DeleteMidQEntry(midQ); + up(&ses->server->tcpSem); /* If not lock req, update # of requests on wire to server */ if(long_op < 3) { atomic_dec(&ses->server->inFlight); wake_up(&ses->server->request_q); } - return -EIO; + return rc; + } else + up(&ses->server->tcpSem); + if (long_op == -1) + goto cifs_no_response_exit2; + else if (long_op == 2) /* writes past end of file can take loong time */ + timeout = 180 * HZ; + else if (long_op == 1) + timeout = 45 * HZ; /* should be greater than + servers oplock break timeout (about 43 seconds) */ + else if (long_op > 2) { + timeout = MAX_SCHEDULE_TIMEOUT; + } else + timeout = 15 * HZ; + /* wait for 15 seconds or until woken up due to response arriving or + due to last connection to this server being unmounted */ + if (signal_pending(current)) { + /* if signal pending do not hold up user for full smb timeout + but we still give response a change to complete */ + timeout = 2 * HZ; + } + + /* No user interrupts in wait - wreaks havoc with performance */ + if(timeout != MAX_SCHEDULE_TIMEOUT) { + timeout += jiffies; + wait_event(ses->server->response_q, + (!(midQ->midState & MID_REQUEST_SUBMITTED)) || + time_after(jiffies, timeout) || + ((ses->server->tcpStatus != CifsGood) && + (ses->server->tcpStatus != CifsNew))); + } else { + wait_event(ses->server->response_q, + (!(midQ->midState & MID_REQUEST_SUBMITTED)) || + ((ses->server->tcpStatus != CifsGood) && + (ses->server->tcpStatus != CifsNew))); } - /* BB can we sign efficiently in this path? */ - rc = cifs_sign_smb(in_buf, ses->server, &midQ->sequence_number); + spin_lock(&GlobalMid_Lock); + if (midQ->resp_buf) { + spin_unlock(&GlobalMid_Lock); + receive_len = midQ->resp_buf->smb_buf_length; + } else { + cERROR(1,("No response to cmd %d mid %d", + midQ->command, midQ->mid)); + if(midQ->midState == MID_REQUEST_SUBMITTED) { + if(ses->server->tcpStatus == CifsExiting) + rc = -EHOSTDOWN; + else { + ses->server->tcpStatus = CifsNeedReconnect; + midQ->midState = MID_RETRY_NEEDED; + } + } - midQ->midState = MID_REQUEST_SUBMITTED; -/* rc = smb_sendv(ses->server->ssocket, in_buf, in_buf->smb_buf_length, - piovec, - (struct sockaddr *) &(ses->server->addr.sockAddr));*/ - if(rc < 0) { + if (rc != -EHOSTDOWN) { + if(midQ->midState == MID_RETRY_NEEDED) { + rc = -EAGAIN; + cFYI(1,("marking request for retry")); + } else { + rc = -EIO; + } + } + spin_unlock(&GlobalMid_Lock); DeleteMidQEntry(midQ); - up(&ses->server->tcpSem); /* If not lock req, update # of requests on wire to server */ if(long_op < 3) { atomic_dec(&ses->server->inFlight); wake_up(&ses->server->request_q); } return rc; - } else - up(&ses->server->tcpSem); -cifs_out_label: - if(midQ) - DeleteMidQEntry(midQ); - + } + + if (receive_len > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE) { + cERROR(1, ("Frame too large received. Length: %d Xid: %d", + receive_len, xid)); + rc = -EIO; + } else { /* rcvd frame is ok */ + + if (midQ->resp_buf && + (midQ->midState == MID_RESPONSE_RECEIVED)) { + in_buf->smb_buf_length = receive_len; + /* BB verify that length would not overrun small buf */ + memcpy((char *)in_buf + 4, + (char *)midQ->resp_buf + 4, + receive_len); + + dump_smb(in_buf, 80); + /* convert the length into a more usable form */ + if((receive_len > 24) && + (ses->server->secMode & (SECMODE_SIGN_REQUIRED | + SECMODE_SIGN_ENABLED))) { + rc = cifs_verify_signature(in_buf, + ses->server->mac_signing_key, + midQ->sequence_number+1); + if(rc) { + cERROR(1,("Unexpected SMB signature")); + /* BB FIXME add code to kill session */ + } + } + + *pbytes_returned = in_buf->smb_buf_length; + + /* BB special case reconnect tid and uid here? */ + /* BB special case Errbadpassword and pwdexpired here */ + rc = map_smb_to_linux_error(in_buf); + + /* convert ByteCount if necessary */ + if (receive_len >= + sizeof (struct smb_hdr) - + 4 /* do not count RFC1001 header */ + + (2 * in_buf->WordCount) + 2 /* bcc */ ) + BCC(in_buf) = le16_to_cpu(BCC_LE(in_buf)); + } else { + rc = -EIO; + cFYI(1,("Bad MID state?")); + } + } +cifs_no_response_exit2: + DeleteMidQEntry(midQ); + if(long_op < 3) { - atomic_dec(&ses->server->inFlight); + atomic_dec(&ses->server->inFlight); wake_up(&ses->server->request_q); } return rc; -} +out_unlock2: + up(&ses->server->tcpSem); + /* If not lock req, update # of requests on wire to server */ + if(long_op < 3) { + atomic_dec(&ses->server->inFlight); + wake_up(&ses->server->request_q); + } + return rc; +} #endif /* CIFS_EXPERIMENTAL */ int @@ -419,9 +585,15 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses, if(atomic_read(&ses->server->inFlight) >= cifs_max_pending){ spin_unlock(&GlobalMid_Lock); +#ifdef CONFIG_CIFS_STATS2 + atomic_inc(&ses->server->num_waiters); +#endif wait_event(ses->server->request_q, atomic_read(&ses->server->inFlight) < cifs_max_pending); +#ifdef CONFIG_CIFS_STATS2 + atomic_dec(&ses->server->num_waiters); +#endif spin_lock(&GlobalMid_Lock); } else { if(ses->server->tcpStatus == CifsExiting) { @@ -490,8 +662,15 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses, rc = cifs_sign_smb(in_buf, ses->server, &midQ->sequence_number); midQ->midState = MID_REQUEST_SUBMITTED; +#ifdef CONFIG_CIFS_STATS2 + atomic_inc(&ses->server->inSend); +#endif rc = smb_send(ses->server->ssocket, in_buf, in_buf->smb_buf_length, (struct sockaddr *) &(ses->server->addr.sockAddr)); +#ifdef CONFIG_CIFS_STATS2 + atomic_dec(&ses->server->inSend); + midQ->when_sent = jiffies; +#endif if(rc < 0) { DeleteMidQEntry(midQ); up(&ses->server->tcpSem); @@ -506,7 +685,7 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses, if (long_op == -1) goto cifs_no_response_exit; else if (long_op == 2) /* writes past end of file can take loong time */ - timeout = 300 * HZ; + timeout = 180 * HZ; else if (long_op == 1) timeout = 45 * HZ; /* should be greater than servers oplock break timeout (about 43 seconds) */ @@ -540,9 +719,10 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses, spin_lock(&GlobalMid_Lock); if (midQ->resp_buf) { spin_unlock(&GlobalMid_Lock); - receive_len = be32_to_cpu(*(__be32 *)midQ->resp_buf); + receive_len = midQ->resp_buf->smb_buf_length; } else { - cERROR(1,("No response buffer")); + cERROR(1,("No response for cmd %d mid %d", + midQ->command, midQ->mid)); if(midQ->midState == MID_REQUEST_SUBMITTED) { if(ses->server->tcpStatus == CifsExiting) rc = -EHOSTDOWN; @@ -607,10 +787,10 @@ SendReceive(const unsigned int xid, struct cifsSesInfo *ses, sizeof (struct smb_hdr) - 4 /* do not count RFC1001 header */ + (2 * out_buf->WordCount) + 2 /* bcc */ ) - BCC(out_buf) = le16_to_cpu(BCC(out_buf)); + BCC(out_buf) = le16_to_cpu(BCC_LE(out_buf)); } else { rc = -EIO; - cFYI(1,("Bad MID state? ")); + cERROR(1,("Bad MID state? ")); } } cifs_no_response_exit: diff --git a/fs/cifs/xattr.c b/fs/cifs/xattr.c index c1e02eff1d25..f375f87c7dbd 100644 --- a/fs/cifs/xattr.c +++ b/fs/cifs/xattr.c @@ -87,8 +87,7 @@ int cifs_removexattr(struct dentry * direntry, const char * ea_name) cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); } remove_ea_exit: - if (full_path) - kfree(full_path); + kfree(full_path); FreeXid(xid); #endif return rc; @@ -132,8 +131,7 @@ int cifs_setxattr(struct dentry * direntry, const char * ea_name, returns as xattrs */ if(value_size > MAX_EA_VALUE_SIZE) { cFYI(1,("size of EA value too large")); - if(full_path) - kfree(full_path); + kfree(full_path); FreeXid(xid); return -EOPNOTSUPP; } @@ -195,8 +193,7 @@ int cifs_setxattr(struct dentry * direntry, const char * ea_name, } set_ea_exit: - if (full_path) - kfree(full_path); + kfree(full_path); FreeXid(xid); #endif return rc; @@ -298,8 +295,7 @@ ssize_t cifs_getxattr(struct dentry * direntry, const char * ea_name, rc = -EOPNOTSUPP; get_ea_exit: - if (full_path) - kfree(full_path); + kfree(full_path); FreeXid(xid); #endif return rc; @@ -345,8 +341,7 @@ ssize_t cifs_listxattr(struct dentry * direntry, char * data, size_t buf_size) cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); - if (full_path) - kfree(full_path); + kfree(full_path); FreeXid(xid); #endif return rc; diff --git a/fs/coda/psdev.c b/fs/coda/psdev.c index 3d1cce3653b8..6a3df88accfe 100644 --- a/fs/coda/psdev.c +++ b/fs/coda/psdev.c @@ -370,8 +370,8 @@ static int init_coda_psdev(void) } devfs_mk_dir ("coda"); for (i = 0; i < MAX_CODADEVS; i++) { - class_device_create(coda_psdev_class, MKDEV(CODA_PSDEV_MAJOR,i), - NULL, "cfs%d", i); + class_device_create(coda_psdev_class, NULL, + MKDEV(CODA_PSDEV_MAJOR,i), NULL, "cfs%d", i); err = devfs_mk_cdev(MKDEV(CODA_PSDEV_MAJOR, i), S_IFCHR|S_IRUSR|S_IWUSR, "coda/%d", i); if (err) diff --git a/fs/compat.c b/fs/compat.c index a719e158e002..818634120b69 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -168,8 +168,8 @@ asmlinkage long compat_sys_statfs(const char __user *path, struct compat_statfs if (!error) { struct kstatfs tmp; error = vfs_statfs(nd.dentry->d_inode->i_sb, &tmp); - if (!error && put_compat_statfs(buf, &tmp)) - error = -EFAULT; + if (!error) + error = put_compat_statfs(buf, &tmp); path_release(&nd); } return error; @@ -186,8 +186,8 @@ asmlinkage long compat_sys_fstatfs(unsigned int fd, struct compat_statfs __user if (!file) goto out; error = vfs_statfs(file->f_dentry->d_inode->i_sb, &tmp); - if (!error && put_compat_statfs(buf, &tmp)) - error = -EFAULT; + if (!error) + error = put_compat_statfs(buf, &tmp); fput(file); out: return error; @@ -236,8 +236,8 @@ asmlinkage long compat_sys_statfs64(const char __user *path, compat_size_t sz, s if (!error) { struct kstatfs tmp; error = vfs_statfs(nd.dentry->d_inode->i_sb, &tmp); - if (!error && put_compat_statfs64(buf, &tmp)) - error = -EFAULT; + if (!error) + error = put_compat_statfs64(buf, &tmp); path_release(&nd); } return error; @@ -257,8 +257,8 @@ asmlinkage long compat_sys_fstatfs64(unsigned int fd, compat_size_t sz, struct c if (!file) goto out; error = vfs_statfs(file->f_dentry->d_inode->i_sb, &tmp); - if (!error && put_compat_statfs64(buf, &tmp)) - error = -EFAULT; + if (!error) + error = put_compat_statfs64(buf, &tmp); fput(file); out: return error; @@ -268,7 +268,6 @@ out: #define IOCTL_HASHSIZE 256 static struct ioctl_trans *ioctl32_hash_table[IOCTL_HASHSIZE]; -static DECLARE_RWSEM(ioctl32_sem); extern struct ioctl_trans ioctl_start[]; extern int ioctl_table_size; @@ -390,14 +389,10 @@ asmlinkage long compat_sys_ioctl(unsigned int fd, unsigned int cmd, break; } - /* When register_ioctl32_conversion is finally gone remove - this lock! -AK */ - down_read(&ioctl32_sem); for (t = ioctl32_hash_table[ioctl32_hash(cmd)]; t; t = t->next) { if (t->cmd == cmd) goto found_handler; } - up_read(&ioctl32_sem); if (S_ISSOCK(filp->f_dentry->d_inode->i_mode) && cmd >= SIOCDEVPRIVATE && cmd <= (SIOCDEVPRIVATE + 15)) { @@ -417,11 +412,9 @@ asmlinkage long compat_sys_ioctl(unsigned int fd, unsigned int cmd, lock_kernel(); error = t->handler(fd, cmd, arg, filp); unlock_kernel(); - up_read(&ioctl32_sem); goto out_fput; } - up_read(&ioctl32_sem); do_ioctl: error = vfs_ioctl(filp, fd, cmd, arg); out_fput: @@ -1490,7 +1483,6 @@ int compat_do_execve(char * filename, /* execve success */ security_bprm_free(bprm); acct_update_integrals(current); - update_mem_hiwater(current); kfree(bprm); return retval; } diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c index e28a74203f3b..43a2508ac696 100644 --- a/fs/compat_ioctl.c +++ b/fs/compat_ioctl.c @@ -49,6 +49,8 @@ #include <linux/vt_kern.h> #include <linux/fb.h> #include <linux/ext2_fs.h> +#include <linux/ext3_jbd.h> +#include <linux/ext3_fs.h> #include <linux/videodev.h> #include <linux/netdevice.h> #include <linux/raw.h> @@ -121,6 +123,11 @@ #include <linux/hiddev.h> +#include <linux/dvb/audio.h> +#include <linux/dvb/dmx.h> +#include <linux/dvb/frontend.h> +#include <linux/dvb/video.h> + #undef INCLUDES #endif @@ -129,6 +136,15 @@ /* Aiee. Someone does not find a difference between int and long */ #define EXT2_IOC32_GETFLAGS _IOR('f', 1, int) #define EXT2_IOC32_SETFLAGS _IOW('f', 2, int) +#define EXT3_IOC32_GETVERSION _IOR('f', 3, int) +#define EXT3_IOC32_SETVERSION _IOW('f', 4, int) +#define EXT3_IOC32_GETRSVSZ _IOR('f', 5, int) +#define EXT3_IOC32_SETRSVSZ _IOW('f', 6, int) +#define EXT3_IOC32_GROUP_EXTEND _IOW('f', 7, unsigned int) +#ifdef CONFIG_JBD_DEBUG +#define EXT3_IOC32_WAIT_FOR_READONLY _IOR('f', 99, int) +#endif + #define EXT2_IOC32_GETVERSION _IOR('v', 1, int) #define EXT2_IOC32_SETVERSION _IOW('v', 2, int) @@ -175,6 +191,22 @@ static int do_ext2_ioctl(unsigned int fd, unsigned int cmd, unsigned long arg) return sys_ioctl(fd, cmd, (unsigned long)compat_ptr(arg)); } +static int do_ext3_ioctl(unsigned int fd, unsigned int cmd, unsigned long arg) +{ + /* These are just misnamed, they actually get/put from/to user an int */ + switch (cmd) { + case EXT3_IOC32_GETVERSION: cmd = EXT3_IOC_GETVERSION; break; + case EXT3_IOC32_SETVERSION: cmd = EXT3_IOC_SETVERSION; break; + case EXT3_IOC32_GETRSVSZ: cmd = EXT3_IOC_GETRSVSZ; break; + case EXT3_IOC32_SETRSVSZ: cmd = EXT3_IOC_SETRSVSZ; break; + case EXT3_IOC32_GROUP_EXTEND: cmd = EXT3_IOC_GROUP_EXTEND; break; +#ifdef CONFIG_JBD_DEBUG + case EXT3_IOC32_WAIT_FOR_READONLY: cmd = EXT3_IOC_WAIT_FOR_READONLY; break; +#endif + } + return sys_ioctl(fd, cmd, (unsigned long)compat_ptr(arg)); +} + struct video_tuner32 { compat_int_t tuner; char name[32]; @@ -413,6 +445,128 @@ out: return err; } +struct compat_dmx_event { + dmx_event_t event; + compat_time_t timeStamp; + union + { + dmx_scrambling_status_t scrambling; + } u; +}; + +static int do_dmx_get_event(unsigned int fd, unsigned int cmd, unsigned long arg) +{ + struct dmx_event kevent; + mm_segment_t old_fs = get_fs(); + int err; + + set_fs(KERNEL_DS); + err = sys_ioctl(fd, cmd, (unsigned long) &kevent); + set_fs(old_fs); + + if (!err) { + struct compat_dmx_event __user *up = compat_ptr(arg); + + err = put_user(kevent.event, &up->event); + err |= put_user(kevent.timeStamp, &up->timeStamp); + err |= put_user(kevent.u.scrambling, &up->u.scrambling); + if (err) + err = -EFAULT; + } + + return err; +} + +struct compat_video_event { + int32_t type; + compat_time_t timestamp; + union { + video_size_t size; + unsigned int frame_rate; + } u; +}; + +static int do_video_get_event(unsigned int fd, unsigned int cmd, unsigned long arg) +{ + struct video_event kevent; + mm_segment_t old_fs = get_fs(); + int err; + + set_fs(KERNEL_DS); + err = sys_ioctl(fd, cmd, (unsigned long) &kevent); + set_fs(old_fs); + + if (!err) { + struct compat_video_event __user *up = compat_ptr(arg); + + err = put_user(kevent.type, &up->type); + err |= put_user(kevent.timestamp, &up->timestamp); + err |= put_user(kevent.u.size.w, &up->u.size.w); + err |= put_user(kevent.u.size.h, &up->u.size.h); + err |= put_user(kevent.u.size.aspect_ratio, + &up->u.size.aspect_ratio); + if (err) + err = -EFAULT; + } + + return err; +} + +struct compat_video_still_picture { + compat_uptr_t iFrame; + int32_t size; +}; + +static int do_video_stillpicture(unsigned int fd, unsigned int cmd, unsigned long arg) +{ + struct compat_video_still_picture __user *up; + struct video_still_picture __user *up_native; + compat_uptr_t fp; + int32_t size; + int err; + + up = (struct compat_video_still_picture __user *) arg; + err = get_user(fp, &up->iFrame); + err |= get_user(size, &up->size); + if (err) + return -EFAULT; + + up_native = + compat_alloc_user_space(sizeof(struct video_still_picture)); + + put_user(compat_ptr(fp), &up_native->iFrame); + put_user(size, &up_native->size); + + err = sys_ioctl(fd, cmd, (unsigned long) up_native); + + return err; +} + +struct compat_video_spu_palette { + int length; + compat_uptr_t palette; +}; + +static int do_video_set_spu_palette(unsigned int fd, unsigned int cmd, unsigned long arg) +{ + struct compat_video_spu_palette __user *up; + struct video_spu_palette __user *up_native; + compat_uptr_t palp; + int length, err; + + up = (struct compat_video_spu_palette __user *) arg; + err = get_user(palp, &up->palette); + err |= get_user(length, &up->length); + + up_native = compat_alloc_user_space(sizeof(struct video_spu_palette)); + put_user(compat_ptr(palp), &up_native->palette); + put_user(length, &up_native->length); + + err = sys_ioctl(fd, cmd, (unsigned long) up_native); + + return err; +} + #ifdef CONFIG_NET static int do_siocgstamp(unsigned int fd, unsigned int cmd, unsigned long arg) { @@ -532,7 +686,8 @@ static int dev_ifconf(unsigned int fd, unsigned int cmd, unsigned long arg) ifr = ifc.ifc_req; ifr32 = compat_ptr(ifc32.ifcbuf); - for (i = 0, j = 0; i < ifc32.ifc_len && j < ifc.ifc_len; + for (i = 0, j = 0; + i + sizeof (struct ifreq32) < ifc32.ifc_len && j < ifc.ifc_len; i += sizeof (struct ifreq32), j += sizeof (struct ifreq)) { if (copy_in_user(ifr32, ifr, sizeof (struct ifreq32))) return -EFAULT; @@ -548,10 +703,7 @@ static int dev_ifconf(unsigned int fd, unsigned int cmd, unsigned long arg) i = ((i / sizeof(struct ifreq)) * sizeof(struct ifreq32)); ifc32.ifc_len = i; } else { - if (i <= ifc32.ifc_len) - ifc32.ifc_len = i; - else - ifc32.ifc_len = i - sizeof (struct ifreq32); + ifc32.ifc_len = i; } if (copy_to_user(compat_ptr(arg), &ifc32, sizeof(struct ifconf32))) return -EFAULT; @@ -840,146 +992,6 @@ static int hdio_getgeo(unsigned int fd, unsigned int cmd, unsigned long arg) return err ? -EFAULT : 0; } -struct fb_fix_screeninfo32 { - char id[16]; - compat_caddr_t smem_start; - u32 smem_len; - u32 type; - u32 type_aux; - u32 visual; - u16 xpanstep; - u16 ypanstep; - u16 ywrapstep; - u32 line_length; - compat_caddr_t mmio_start; - u32 mmio_len; - u32 accel; - u16 reserved[3]; -}; - -struct fb_cmap32 { - u32 start; - u32 len; - compat_caddr_t red; - compat_caddr_t green; - compat_caddr_t blue; - compat_caddr_t transp; -}; - -static int fb_getput_cmap(unsigned int fd, unsigned int cmd, unsigned long arg) -{ - struct fb_cmap_user __user *cmap; - struct fb_cmap32 __user *cmap32; - __u32 data; - int err; - - cmap = compat_alloc_user_space(sizeof(*cmap)); - cmap32 = compat_ptr(arg); - - if (copy_in_user(&cmap->start, &cmap32->start, 2 * sizeof(__u32))) - return -EFAULT; - - if (get_user(data, &cmap32->red) || - put_user(compat_ptr(data), &cmap->red) || - get_user(data, &cmap32->green) || - put_user(compat_ptr(data), &cmap->green) || - get_user(data, &cmap32->blue) || - put_user(compat_ptr(data), &cmap->blue) || - get_user(data, &cmap32->transp) || - put_user(compat_ptr(data), &cmap->transp)) - return -EFAULT; - - err = sys_ioctl(fd, cmd, (unsigned long) cmap); - - if (!err) { - if (copy_in_user(&cmap32->start, - &cmap->start, - 2 * sizeof(__u32))) - err = -EFAULT; - } - return err; -} - -static int do_fscreeninfo_to_user(struct fb_fix_screeninfo *fix, - struct fb_fix_screeninfo32 __user *fix32) -{ - __u32 data; - int err; - - err = copy_to_user(&fix32->id, &fix->id, sizeof(fix32->id)); - - data = (__u32) (unsigned long) fix->smem_start; - err |= put_user(data, &fix32->smem_start); - - err |= put_user(fix->smem_len, &fix32->smem_len); - err |= put_user(fix->type, &fix32->type); - err |= put_user(fix->type_aux, &fix32->type_aux); - err |= put_user(fix->visual, &fix32->visual); - err |= put_user(fix->xpanstep, &fix32->xpanstep); - err |= put_user(fix->ypanstep, &fix32->ypanstep); - err |= put_user(fix->ywrapstep, &fix32->ywrapstep); - err |= put_user(fix->line_length, &fix32->line_length); - - data = (__u32) (unsigned long) fix->mmio_start; - err |= put_user(data, &fix32->mmio_start); - - err |= put_user(fix->mmio_len, &fix32->mmio_len); - err |= put_user(fix->accel, &fix32->accel); - err |= copy_to_user(fix32->reserved, fix->reserved, - sizeof(fix->reserved)); - - return err; -} - -static int fb_get_fscreeninfo(unsigned int fd, unsigned int cmd, unsigned long arg) -{ - mm_segment_t old_fs; - struct fb_fix_screeninfo fix; - struct fb_fix_screeninfo32 __user *fix32; - int err; - - fix32 = compat_ptr(arg); - - old_fs = get_fs(); - set_fs(KERNEL_DS); - err = sys_ioctl(fd, cmd, (unsigned long) &fix); - set_fs(old_fs); - - if (!err) - err = do_fscreeninfo_to_user(&fix, fix32); - - return err; -} - -static int fb_ioctl_trans(unsigned int fd, unsigned int cmd, unsigned long arg) -{ - int err; - - switch (cmd) { - case FBIOGET_FSCREENINFO: - err = fb_get_fscreeninfo(fd,cmd, arg); - break; - - case FBIOGETCMAP: - case FBIOPUTCMAP: - err = fb_getput_cmap(fd, cmd, arg); - break; - - default: - do { - static int count; - if (++count <= 20) - printk("%s: Unknown fb ioctl cmd fd(%d) " - "cmd(%08x) arg(%08lx)\n", - __FUNCTION__, fd, cmd, arg); - } while(0); - err = -ENOSYS; - break; - }; - - return err; -} - static int hdio_ioctl_trans(unsigned int fd, unsigned int cmd, unsigned long arg) { mm_segment_t old_fs = get_fs(); @@ -2235,7 +2247,8 @@ static int fd_ioctl_trans(unsigned int fd, unsigned int cmd, unsigned long arg) if (err) err = -EFAULT; -out: if (karg) kfree(karg); +out: + kfree(karg); return err; } @@ -2952,10 +2965,7 @@ HANDLE_IOCTL(BLKGETSIZE, w_long) HANDLE_IOCTL(0x1260, broken_blkgetsize) HANDLE_IOCTL(BLKFRAGET, w_long) HANDLE_IOCTL(BLKSECTGET, w_long) -HANDLE_IOCTL(FBIOGET_FSCREENINFO, fb_ioctl_trans) HANDLE_IOCTL(BLKPG, blkpg_ioctl_trans) -HANDLE_IOCTL(FBIOGETCMAP, fb_ioctl_trans) -HANDLE_IOCTL(FBIOPUTCMAP, fb_ioctl_trans) HANDLE_IOCTL(HDIO_GET_KEEPSETTINGS, hdio_ioctl_trans) HANDLE_IOCTL(HDIO_GET_UNMASKINTR, hdio_ioctl_trans) HANDLE_IOCTL(HDIO_GET_DMA, hdio_ioctl_trans) @@ -2996,6 +3006,15 @@ HANDLE_IOCTL(EXT2_IOC32_GETFLAGS, do_ext2_ioctl) HANDLE_IOCTL(EXT2_IOC32_SETFLAGS, do_ext2_ioctl) HANDLE_IOCTL(EXT2_IOC32_GETVERSION, do_ext2_ioctl) HANDLE_IOCTL(EXT2_IOC32_SETVERSION, do_ext2_ioctl) +HANDLE_IOCTL(EXT3_IOC32_GETVERSION, do_ext3_ioctl) +HANDLE_IOCTL(EXT3_IOC32_SETVERSION, do_ext3_ioctl) +HANDLE_IOCTL(EXT3_IOC32_GETRSVSZ, do_ext3_ioctl) +HANDLE_IOCTL(EXT3_IOC32_SETRSVSZ, do_ext3_ioctl) +HANDLE_IOCTL(EXT3_IOC32_GROUP_EXTEND, do_ext3_ioctl) +COMPATIBLE_IOCTL(EXT3_IOC_GROUP_ADD) +#ifdef CONFIG_JBD_DEBUG +HANDLE_IOCTL(EXT3_IOC32_WAIT_FOR_READONLY, do_ext3_ioctl) +#endif HANDLE_IOCTL(VIDIOCGTUNER32, do_video_ioctl) HANDLE_IOCTL(VIDIOCSTUNER32, do_video_ioctl) HANDLE_IOCTL(VIDIOCGWIN32, do_video_ioctl) @@ -3046,10 +3065,25 @@ HANDLE_IOCTL(RAW_GETBIND, raw_ioctl) /* Serial */ HANDLE_IOCTL(TIOCGSERIAL, serial_struct_ioctl) HANDLE_IOCTL(TIOCSSERIAL, serial_struct_ioctl) +#ifdef TIOCGLTC +COMPATIBLE_IOCTL(TIOCGLTC) +COMPATIBLE_IOCTL(TIOCSLTC) +#endif +#ifdef TIOCSTART +/* + * For these two we have defintions in ioctls.h and/or termios.h on + * some architectures but no actual implemention. Some applications + * like bash call them if they are defined in the headers, so we provide + * entries here to avoid syslog message spew. + */ +COMPATIBLE_IOCTL(TIOCSTART) +COMPATIBLE_IOCTL(TIOCSTOP) +#endif /* Usbdevfs */ HANDLE_IOCTL(USBDEVFS_CONTROL32, do_usbdevfs_control) HANDLE_IOCTL(USBDEVFS_BULK32, do_usbdevfs_bulk) HANDLE_IOCTL(USBDEVFS_DISCSIGNAL32, do_usbdevfs_discsignal) +COMPATIBLE_IOCTL(USBDEVFS_IOCTL32) /* i2c */ HANDLE_IOCTL(I2C_FUNCS, w_long) HANDLE_IOCTL(I2C_RDWR, do_i2c_rdwr_ioctl) @@ -3081,5 +3115,11 @@ HANDLE_IOCTL(NCP_IOC_GETPRIVATEDATA_32, do_ncp_getprivatedata) HANDLE_IOCTL(NCP_IOC_SETPRIVATEDATA_32, do_ncp_setprivatedata) #endif +/* dvb */ +HANDLE_IOCTL(DMX_GET_EVENT, do_dmx_get_event) +HANDLE_IOCTL(VIDEO_GET_EVENT, do_video_get_event) +HANDLE_IOCTL(VIDEO_STILLPICTURE, do_video_stillpicture) +HANDLE_IOCTL(VIDEO_SET_SPU_PALETTE, do_video_set_spu_palette) + #undef DECLARES #endif diff --git a/fs/dcache.c b/fs/dcache.c index fb10386c59be..17e439138681 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -644,7 +644,7 @@ void shrink_dcache_parent(struct dentry * parent) * * Prune the dentries that are anonymous * - * parsing d_hash list does not hlist_for_each_rcu() as it + * parsing d_hash list does not hlist_for_each_entry_rcu() as it * done under dcache_lock. * */ @@ -689,7 +689,7 @@ void shrink_dcache_anon(struct hlist_head *head) * * In this case we return -1 to tell the caller that we baled. */ -static int shrink_dcache_memory(int nr, unsigned int gfp_mask) +static int shrink_dcache_memory(int nr, gfp_t gfp_mask) { if (nr) { if (!(gfp_mask & __GFP_FS)) @@ -1043,15 +1043,13 @@ struct dentry * __d_lookup(struct dentry * parent, struct qstr * name) struct hlist_head *head = d_hash(parent,hash); struct dentry *found = NULL; struct hlist_node *node; + struct dentry *dentry; rcu_read_lock(); - hlist_for_each_rcu(node, head) { - struct dentry *dentry; + hlist_for_each_entry_rcu(dentry, node, head, d_hash) { struct qstr *qstr; - dentry = hlist_entry(node, struct dentry, d_hash); - if (dentry->d_name.hash != hash) continue; if (dentry->d_parent != parent) @@ -1123,7 +1121,7 @@ int d_validate(struct dentry *dentry, struct dentry *dparent) spin_lock(&dcache_lock); base = d_hash(dparent, dentry->d_name.hash); hlist_for_each(lhp,base) { - /* hlist_for_each_rcu() not required for d_hash list + /* hlist_for_each_entry_rcu() not required for d_hash list * as it is parsed under dcache_lock */ if (dentry == hlist_entry(lhp, struct dentry, d_hash)) { diff --git a/fs/devfs/base.c b/fs/devfs/base.c index 8b679b67e5e0..1274422a5384 100644 --- a/fs/devfs/base.c +++ b/fs/devfs/base.c @@ -2738,10 +2738,8 @@ static int devfsd_close(struct inode *inode, struct file *file) entry = fs_info->devfsd_first_event; fs_info->devfsd_first_event = NULL; fs_info->devfsd_last_event = NULL; - if (fs_info->devfsd_info) { - kfree(fs_info->devfsd_info); - fs_info->devfsd_info = NULL; - } + kfree(fs_info->devfsd_info); + fs_info->devfsd_info = NULL; spin_unlock(&fs_info->devfsd_buffer_lock); fs_info->devfsd_pgrp = 0; fs_info->devfsd_task = NULL; diff --git a/fs/direct-io.c b/fs/direct-io.c index 0d06097bc995..3931e7f1e6bf 100644 --- a/fs/direct-io.c +++ b/fs/direct-io.c @@ -162,6 +162,7 @@ static int dio_refill_pages(struct dio *dio) up_read(¤t->mm->mmap_sem); if (ret < 0 && dio->blocks_available && (dio->rw == WRITE)) { + struct page *page = ZERO_PAGE(dio->curr_user_address); /* * A memory fault, but the filesystem has some outstanding * mapped blocks. We need to use those blocks up to avoid @@ -169,7 +170,8 @@ static int dio_refill_pages(struct dio *dio) */ if (dio->page_errors == 0) dio->page_errors = ret; - dio->pages[0] = ZERO_PAGE(dio->curr_user_address); + page_cache_get(page); + dio->pages[0] = page; dio->head = 0; dio->tail = 1; ret = 0; diff --git a/fs/dquot.c b/fs/dquot.c index b9732335bcdc..2a62b3dc20ec 100644 --- a/fs/dquot.c +++ b/fs/dquot.c @@ -77,6 +77,7 @@ #include <linux/kmod.h> #include <linux/namei.h> #include <linux/buffer_head.h> +#include <linux/quotaops.h> #include <asm/uaccess.h> @@ -500,7 +501,7 @@ static void prune_dqcache(int count) * more memory */ -static int shrink_dqcache_memory(int nr, unsigned int gfp_mask) +static int shrink_dqcache_memory(int nr, gfp_t gfp_mask) { if (nr) { spin_lock(&dq_list_lock); @@ -662,7 +663,7 @@ static void add_dquot_ref(struct super_block *sb, int type) restart: file_list_lock(); list_for_each(p, &sb->s_files) { - struct file *filp = list_entry(p, struct file, f_list); + struct file *filp = list_entry(p, struct file, f_u.fu_list); struct inode *inode = filp->f_dentry->d_inode; if (filp->f_mode & FMODE_WRITE && dqinit_needed(inode, type)) { struct dentry *dentry = dget(filp->f_dentry); @@ -1320,13 +1321,11 @@ int vfs_quota_off(struct super_block *sb, int type) int cnt; struct quota_info *dqopt = sb_dqopt(sb); struct inode *toputinode[MAXQUOTAS]; - struct vfsmount *toputmnt[MAXQUOTAS]; /* We need to serialize quota_off() for device */ down(&dqopt->dqonoff_sem); for (cnt = 0; cnt < MAXQUOTAS; cnt++) { toputinode[cnt] = NULL; - toputmnt[cnt] = NULL; if (type != -1 && cnt != type) continue; if (!sb_has_quota_enabled(sb, cnt)) @@ -1347,9 +1346,7 @@ int vfs_quota_off(struct super_block *sb, int type) put_quota_format(dqopt->info[cnt].dqi_format); toputinode[cnt] = dqopt->files[cnt]; - toputmnt[cnt] = dqopt->mnt[cnt]; dqopt->files[cnt] = NULL; - dqopt->mnt[cnt] = NULL; dqopt->info[cnt].dqi_flags = 0; dqopt->info[cnt].dqi_igrace = 0; dqopt->info[cnt].dqi_bgrace = 0; @@ -1357,10 +1354,7 @@ int vfs_quota_off(struct super_block *sb, int type) } up(&dqopt->dqonoff_sem); /* Sync the superblock so that buffers with quota data are written to - * disk (and so userspace sees correct data afterwards). - * The reference to vfsmnt we are still holding protects us from - * umount (we don't have it only when quotas are turned on/off for - * journal replay but in that case we are guarded by the fs anyway). */ + * disk (and so userspace sees correct data afterwards). */ if (sb->s_op->sync_fs) sb->s_op->sync_fs(sb, 1); sync_blockdev(sb->s_bdev); @@ -1384,10 +1378,6 @@ int vfs_quota_off(struct super_block *sb, int type) iput(toputinode[cnt]); } up(&dqopt->dqonoff_sem); - /* We don't hold the reference when we turned on quotas - * just for the journal replay... */ - if (toputmnt[cnt]) - mntput(toputmnt[cnt]); } if (sb->s_bdev) invalidate_bdev(sb->s_bdev, 0); @@ -1502,11 +1492,8 @@ int vfs_quota_on(struct super_block *sb, int type, int format_id, char *path) /* Quota file not on the same filesystem? */ if (nd.mnt->mnt_sb != sb) error = -EXDEV; - else { + else error = vfs_quota_on_inode(nd.dentry->d_inode, type, format_id); - if (!error) - sb_dqopt(sb)->mnt[type] = mntget(nd.mnt); - } out_path: path_release(&nd); return error; @@ -1526,10 +1513,16 @@ int vfs_quota_on_mount(struct super_block *sb, char *qf_name, if (IS_ERR(dentry)) return PTR_ERR(dentry); + if (!dentry->d_inode) { + error = -ENOENT; + goto out; + } + error = security_quota_on(dentry); if (!error) error = vfs_quota_on_inode(dentry->d_inode, type, format_id); +out: dput(dentry); return error; } diff --git a/fs/eventpoll.c b/fs/eventpoll.c index 403b90a1213d..4284cd31eba6 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -101,6 +101,10 @@ /* Maximum number of poll wake up nests we are allowing */ #define EP_MAX_POLLWAKE_NESTS 4 +/* Maximum msec timeout value storeable in a long int */ +#define EP_MAX_MSTIMEO min(1000ULL * MAX_SCHEDULE_TIMEOUT / HZ, (LONG_MAX - 999ULL) / HZ) + + struct epoll_filefd { struct file *file; int fd; @@ -1506,8 +1510,8 @@ static int ep_poll(struct eventpoll *ep, struct epoll_event __user *events, * and the overflow condition. The passed timeout is in milliseconds, * that why (t * HZ) / 1000. */ - jtimeout = timeout == -1 || timeout > (MAX_SCHEDULE_TIMEOUT - 1000) / HZ ? - MAX_SCHEDULE_TIMEOUT: (timeout * HZ + 999) / 1000; + jtimeout = (timeout < 0 || timeout >= EP_MAX_MSTIMEO) ? + MAX_SCHEDULE_TIMEOUT : (timeout * HZ + 999) / 1000; retry: write_lock_irqsave(&ep->lock, flags); diff --git a/fs/exec.c b/fs/exec.c index a04a575ad433..22533cce0611 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -48,6 +48,7 @@ #include <linux/syscalls.h> #include <linux/rmap.h> #include <linux/acct.h> +#include <linux/cn_proc.h> #include <asm/uaccess.h> #include <asm/mmu_context.h> @@ -126,8 +127,7 @@ asmlinkage long sys_uselib(const char __user * library) struct nameidata nd; int error; - nd.intent.open.flags = FMODE_READ; - error = __user_walk(library, LOOKUP_FOLLOW|LOOKUP_OPEN, &nd); + error = __user_path_lookup_open(library, LOOKUP_FOLLOW, &nd, FMODE_READ); if (error) goto out; @@ -135,11 +135,11 @@ asmlinkage long sys_uselib(const char __user * library) if (!S_ISREG(nd.dentry->d_inode->i_mode)) goto exit; - error = permission(nd.dentry->d_inode, MAY_READ | MAY_EXEC, &nd); + error = vfs_permission(&nd, MAY_READ | MAY_EXEC); if (error) goto exit; - file = dentry_open(nd.dentry, nd.mnt, O_RDONLY); + file = nameidata_to_filp(&nd, O_RDONLY); error = PTR_ERR(file); if (IS_ERR(file)) goto out; @@ -167,6 +167,7 @@ asmlinkage long sys_uselib(const char __user * library) out: return error; exit: + release_open_intent(&nd); path_release(&nd); goto out; } @@ -305,44 +306,30 @@ void install_arg_page(struct vm_area_struct *vma, struct page *page, unsigned long address) { struct mm_struct *mm = vma->vm_mm; - pgd_t * pgd; - pud_t * pud; - pmd_t * pmd; pte_t * pte; + spinlock_t *ptl; if (unlikely(anon_vma_prepare(vma))) - goto out_sig; + goto out; flush_dcache_page(page); - pgd = pgd_offset(mm, address); - - spin_lock(&mm->page_table_lock); - pud = pud_alloc(mm, pgd, address); - if (!pud) - goto out; - pmd = pmd_alloc(mm, pud, address); - if (!pmd) - goto out; - pte = pte_alloc_map(mm, pmd, address); + pte = get_locked_pte(mm, address, &ptl); if (!pte) goto out; if (!pte_none(*pte)) { - pte_unmap(pte); + pte_unmap_unlock(pte, ptl); goto out; } - inc_mm_counter(mm, rss); + inc_mm_counter(mm, anon_rss); lru_cache_add_active(page); set_pte_at(mm, address, pte, pte_mkdirty(pte_mkwrite(mk_pte( page, vma->vm_page_prot)))); page_add_anon_rmap(page, vma, address); - pte_unmap(pte); - spin_unlock(&mm->page_table_lock); + pte_unmap_unlock(pte, ptl); /* no need for flush_tlb */ return; out: - spin_unlock(&mm->page_table_lock); -out_sig: __free_page(page); force_sig(SIGKILL, current); } @@ -490,8 +477,7 @@ struct file *open_exec(const char *name) int err; struct file *file; - nd.intent.open.flags = FMODE_READ; - err = path_lookup(name, LOOKUP_FOLLOW|LOOKUP_OPEN, &nd); + err = path_lookup_open(name, LOOKUP_FOLLOW, &nd, FMODE_READ); file = ERR_PTR(err); if (!err) { @@ -499,12 +485,12 @@ struct file *open_exec(const char *name) file = ERR_PTR(-EACCES); if (!(nd.mnt->mnt_flags & MNT_NOEXEC) && S_ISREG(inode->i_mode)) { - int err = permission(inode, MAY_EXEC, &nd); + int err = vfs_permission(&nd, MAY_EXEC); if (!err && !(inode->i_mode & 0111)) err = -EACCES; file = ERR_PTR(err); if (!err) { - file = dentry_open(nd.dentry, nd.mnt, O_RDONLY); + file = nameidata_to_filp(&nd, O_RDONLY); if (!IS_ERR(file)) { err = deny_write_access(file); if (err) { @@ -516,6 +502,7 @@ out: return file; } } + release_open_intent(&nd); path_release(&nd); } goto out; @@ -593,6 +580,7 @@ static inline int de_thread(struct task_struct *tsk) struct signal_struct *sig = tsk->signal; struct sighand_struct *newsighand, *oldsighand = tsk->sighand; spinlock_t *lock = &oldsighand->siglock; + struct task_struct *leader = NULL; int count; /* @@ -634,10 +622,9 @@ static inline int de_thread(struct task_struct *tsk) /* * Account for the thread group leader hanging around: */ - count = 2; - if (thread_group_leader(current)) - count = 1; - else { + count = 1; + if (!thread_group_leader(current)) { + count = 2; /* * The SIGALRM timer survives the exec, but needs to point * at us as the new group leader now. We have a race with @@ -646,8 +633,10 @@ static inline int de_thread(struct task_struct *tsk) * before we can safely let the old group leader die. */ sig->real_timer.data = (unsigned long)current; + spin_unlock_irq(lock); if (del_timer_sync(&sig->real_timer)) add_timer(&sig->real_timer); + spin_lock_irq(lock); } while (atomic_read(&sig->count) > count) { sig->group_exit_task = current; @@ -659,7 +648,6 @@ static inline int de_thread(struct task_struct *tsk) } sig->group_exit_task = NULL; sig->notify_count = 0; - sig->real_timer.data = (unsigned long)current; spin_unlock_irq(lock); /* @@ -668,15 +656,16 @@ static inline int de_thread(struct task_struct *tsk) * and to assume its PID: */ if (!thread_group_leader(current)) { - struct task_struct *leader = current->group_leader, *parent; + struct task_struct *parent; struct dentry *proc_dentry1, *proc_dentry2; - unsigned long exit_state, ptrace; + unsigned long ptrace; /* * Wait for the thread group leader to be a zombie. * It should already be zombie at this point, most * of the time. */ + leader = current->group_leader; while (leader->exit_state != EXIT_ZOMBIE) yield(); @@ -727,16 +716,15 @@ static inline int de_thread(struct task_struct *tsk) list_del(¤t->tasks); list_add_tail(¤t->tasks, &init_task.tasks); current->exit_signal = SIGCHLD; - exit_state = leader->exit_state; + + BUG_ON(leader->exit_state != EXIT_ZOMBIE); + leader->exit_state = EXIT_DEAD; write_unlock_irq(&tasklist_lock); spin_unlock(&leader->proc_lock); spin_unlock(¤t->proc_lock); proc_pid_flush(proc_dentry1); proc_pid_flush(proc_dentry2); - - BUG_ON(exit_state != EXIT_ZOMBIE); - release_task(leader); } /* @@ -746,8 +734,11 @@ static inline int de_thread(struct task_struct *tsk) sig->flags = 0; no_thread_group: - BUG_ON(atomic_read(&sig->count) != 1); exit_itimers(sig); + if (leader) + release_task(leader); + + BUG_ON(atomic_read(&sig->count) != 1); if (atomic_read(&oldsighand->count) == 1) { /* @@ -895,7 +886,7 @@ int flush_old_exec(struct linux_binprm * bprm) flush_thread(); if (bprm->e_uid != current->euid || bprm->e_gid != current->egid || - permission(bprm->file->f_dentry->d_inode,MAY_READ, NULL) || + file_permission(bprm->file, MAY_READ) || (bprm->interp_flags & BINPRM_FLAGS_ENFORCE_NONDUMP)) { suid_keys(current); current->mm->dumpable = suid_dumpable; @@ -1100,6 +1091,7 @@ int search_binary_handler(struct linux_binprm *bprm,struct pt_regs *regs) fput(bprm->file); bprm->file = NULL; current->did_exec = 1; + proc_exec_connector(current); return retval; } read_lock(&binfmt_lock); @@ -1207,7 +1199,6 @@ int do_execve(char * filename, /* execve success */ security_bprm_free(bprm); acct_update_integrals(current); - update_mem_hiwater(current); kfree(bprm); return retval; } @@ -1422,19 +1413,16 @@ static void zap_threads (struct mm_struct *mm) static void coredump_wait(struct mm_struct *mm) { DECLARE_COMPLETION(startup_done); + int core_waiters; - mm->core_waiters++; /* let other threads block */ mm->core_startup_done = &startup_done; - /* give other threads a chance to run: */ - yield(); - zap_threads(mm); - if (--mm->core_waiters) { - up_write(&mm->mmap_sem); + core_waiters = mm->core_waiters; + up_write(&mm->mmap_sem); + + if (core_waiters) wait_for_completion(&startup_done); - } else - up_write(&mm->mmap_sem); BUG_ON(mm->core_waiters); } @@ -1468,11 +1456,21 @@ int do_coredump(long signr, int exit_code, struct pt_regs * regs) current->fsuid = 0; /* Dump root private */ } mm->dumpable = 0; - init_completion(&mm->core_done); + + retval = -EAGAIN; spin_lock_irq(¤t->sighand->siglock); - current->signal->flags = SIGNAL_GROUP_EXIT; - current->signal->group_exit_code = exit_code; + if (!(current->signal->flags & SIGNAL_GROUP_EXIT)) { + current->signal->flags = SIGNAL_GROUP_EXIT; + current->signal->group_exit_code = exit_code; + retval = 0; + } spin_unlock_irq(¤t->sighand->siglock); + if (retval) { + up_write(&mm->mmap_sem); + goto fail; + } + + init_completion(&mm->core_done); coredump_wait(mm); /* @@ -1507,7 +1505,7 @@ int do_coredump(long signr, int exit_code, struct pt_regs * regs) goto close_fail; if (!file->f_op->write) goto close_fail; - if (do_truncate(file->f_dentry, 0) != 0) + if (do_truncate(file->f_dentry, 0, file) != 0) goto close_fail; retval = binfmt->core_dump(signr, regs, file); diff --git a/fs/ext2/CHANGES b/fs/ext2/CHANGES deleted file mode 100644 index aa5aaf0e5911..000000000000 --- a/fs/ext2/CHANGES +++ /dev/null @@ -1,157 +0,0 @@ -Changes from version 0.5a to version 0.5b -========================================= - - Now that we have sysctl(), the immutable flag cannot be changed when - the system is running at security level > 0. - - Some cleanups in the code. - - More consistency checks on directories. - - The ext2.diff patch from Tom May <ftom@netcom.com> has been - integrated. This patch replaces expensive "/" and "%" with - cheap ">>" and "&" where possible. - -Changes from version 0.5 to version 0.5a -======================================== - - Zero the partial block following the end of the file when a file - is truncated. - - Dates updated in the copyright. - - More checks when the filesystem is mounted: the count of blocks, - fragments, and inodes per group is checked against the block size. - - The buffers used by the error routines are now static variables, to - avoid using space on the kernel stack, as requested by Linus. - - Some cleanups in the error messages (some versions of syslog contain - a bug which truncates an error message if it contains '\n'). - - Check that no data can be written to a file past the 2GB limit. - - The famous readdir() bug has been fixed by Stephen Tweedie. - - Added a revision level in the superblock. - - Full support for O_SYNC flag of the open system call. - - New mount options: `resuid=#uid' and `resgid=#gid'. `resuid' causes - ext2fs to consider user #uid like root for the reserved blocks. - `resgid' acts the same way with group #gid. New fields in the - superblock contain default values for resuid and resgid and can - be modified by tune2fs. - Idea comes from Rene Cougnenc <cougnenc@renux.frmug.fr.net>. - - New mount options: `bsddf' and `minixdf'. `bsddf' causes ext2fs - to remove the blocks used for FS structures from the total block - count in statfs. With `minixdf', ext2fs mimics Minix behavior - in statfs (i.e. it returns the total number of blocks on the - partition). This is intended to make bde happy :-) - - New file attributes: - - Immutable files cannot be modified. Data cannot be written to - these files. They cannot be removed, renamed and new links cannot - be created. Even root cannot modify the files. He has to remove - the immutable attribute first. - - Append-only files: can only be written in append-mode when writing. - They cannot be removed, renamed and new links cannot be created. - Note: files may only be added to an append-only directory. - - No-dump files: the attribute is not used by the kernel. My port - of dump uses it to avoid backing up files which are not important. - - New check in ext2_check_dir_entry: the inode number is checked. - - Support for big file systems: the copy of the FS descriptor is now - dynamically allocated (previous versions used a fixed size array). - This allows to mount 2GB+ FS. - - Reorganization of the ext2_inode structure to allow other operating - systems to create specific fields if they use ext2fs as their native - file system. Currently, ext2fs is only implemented in Linux but - will soon be part of Gnu Hurd and of Masix. - -Changes from version 0.4b to version 0.5 -======================================== - - New superblock fields: s_lastcheck and s_checkinterval added - by Uwe Ohse <uwe@tirka.gun.de> to implement timedependent checks - of the file system - - Real random numbers for secure rm added by Pierre del Perugia - <delperug@gla.ecoledoc.ibp.fr> - - The mount warnings related to the state of a fs are not printed - if the fs is mounted read-only, idea by Nick Holloway - <alfie@dcs.warwick.ac.uk> - -Changes from version 0.4a to version 0.4b -========================================= - - Copyrights changed to include the name of my laboratory. - - Clean up of balloc.c and ialloc.c. - - More consistency checks. - - Block preallocation added by Stephen Tweedie. - - Direct reads of directories disallowed. - - Readahead implemented in readdir by Stephen Tweedie. - - Bugs in block and inodes allocation fixed. - - Readahead implemented in ext2_find_entry by Chip Salzenberg. - - New mount options: - `check=none|normal|strict' - `debug' - `errors=continue|remount-ro|panic' - `grpid', `bsdgroups' - `nocheck' - `nogrpid', `sysvgroups' - - truncate() now tries to deallocate contiguous blocks in a single call - to ext2_free_blocks(). - - lots of cosmetic changes. - -Changes from version 0.4 to version 0.4a -======================================== - - the `sync' option support is now complete. Version 0.4 was not - supporting it when truncating a file. I have tested the synchronous - writes and they work but they make the system very slow :-( I have - to work again on this to make it faster. - - when detecting an error on a mounted filesystem, version 0.4 used - to try to write a flag in the super block even if the filesystem had - been mounted read-only. This is fixed. - - the `sb=#' option now causes the kernel code to use the filesystem - descriptors located at block #+1. Version 0.4 used the superblock - backup located at block # but used the main copy of the descriptors. - - a new file attribute `S' is supported. This attribute causes - synchronous writes but is applied to a file not to the entire file - system (thanks to Michael Kraehe <kraehe@bakunin.north.de> for - suggesting it). - - the directory cache is inhibited by default. The cache management - code seems to be buggy and I have to look at it carefully before - using it again. - - deleting a file with the `s' attribute (secure deletion) causes its - blocks to be overwritten with random values not with zeros (thanks to - Michael A. Griffith <grif@cs.ucr.edu> for suggesting it). - - lots of cosmetic changes have been made. - -Changes from version 0.3 to version 0.4 -======================================= - - Three new mount options are supported: `check', `sync' and `sb=#'. - `check' tells the kernel code to make more consistency checks - when the file system is mounted. Currently, the kernel code checks - that the blocks and inodes bitmaps are consistent with the free - blocks and inodes counts. More checks will be added in future - releases. - `sync' tells the kernel code to use synchronous writes when updating - an inode, a bitmap, a directory entry or an indirect block. This - can make the file system much slower but can be a big win for files - recovery in case of a crash (and we can now say to the BSD folks - that Linux also supports synchronous updates :-). - `sb=#' tells the kernel code to use an alternate super block instead - of its master copy. `#' is the number of the block (counted in - 1024 bytes blocks) which contains the alternate super block. - An ext2 file system typically contains backups of the super block - at blocks 8193, 16385, and so on. - - I have change the meaning of the valid flag used by e2fsck. it - now contains the state of the file system. If the kernel code - detects an inconsistency while the file system is mounted, it flags - it as erroneous and e2fsck will detect that on next run. - - The super block now contains a mount counter. This counter is - incremented each time the file system is mounted read/write. When - this counter becomes bigger than a maximal mount counts (also stored - in the super block), e2fsck checks the file system, even if it had - been unmounted cleanly, and resets this counter to 0. - - File attributes are now supported. One can associate a set of - attributes to a file. Three attributes are defined: - `c': the file is marked for automatic compression, - `s': the file is marked for secure deletion: when the file is - deleted, its blocks are zeroed and written back to the disk, - `u': the file is marked for undeletion: when the file is deleted, - its contents are saved to allow a future undeletion. - Currently, only the `s' attribute is implemented in the kernel - code. Support for the other attributes will be added in a future - release. - - a few bugs related to times updates have been fixed by Bruce - Evans and me. - - a bug related to the links count of deleted inodes has been fixed. - Previous versions used to keep the links count set to 1 when a file - was deleted. The new version now sets links_count to 0 when deleting - the last link. - - a race condition when deallocating an inode has been fixed by - Stephen Tweedie. - diff --git a/fs/ext2/acl.c b/fs/ext2/acl.c index 213148c36ebe..6af2f4130290 100644 --- a/fs/ext2/acl.c +++ b/fs/ext2/acl.c @@ -194,8 +194,7 @@ ext2_get_acl(struct inode *inode, int type) acl = NULL; else acl = ERR_PTR(retval); - if (value) - kfree(value); + kfree(value); if (!IS_ERR(acl)) { switch(type) { @@ -262,8 +261,7 @@ ext2_set_acl(struct inode *inode, int type, struct posix_acl *acl) error = ext2_xattr_set(inode, name_index, "", value, size, 0); - if (value) - kfree(value); + kfree(value); if (!error) { switch(type) { case ACL_TYPE_ACCESS: diff --git a/fs/ext2/balloc.c b/fs/ext2/balloc.c index 6591abef64d0..bb6908066494 100644 --- a/fs/ext2/balloc.c +++ b/fs/ext2/balloc.c @@ -624,76 +624,3 @@ unsigned long ext2_bg_num_gdb(struct super_block *sb, int group) return EXT2_SB(sb)->s_gdb_count; } -#ifdef CONFIG_EXT2_CHECK -/* Called at mount-time, super-block is locked */ -void ext2_check_blocks_bitmap (struct super_block * sb) -{ - struct buffer_head *bitmap_bh = NULL; - struct ext2_super_block * es; - unsigned long desc_count, bitmap_count, x, j; - unsigned long desc_blocks; - struct ext2_group_desc * desc; - int i; - - es = EXT2_SB(sb)->s_es; - desc_count = 0; - bitmap_count = 0; - desc = NULL; - for (i = 0; i < EXT2_SB(sb)->s_groups_count; i++) { - desc = ext2_get_group_desc (sb, i, NULL); - if (!desc) - continue; - desc_count += le16_to_cpu(desc->bg_free_blocks_count); - brelse(bitmap_bh); - bitmap_bh = read_block_bitmap(sb, i); - if (!bitmap_bh) - continue; - - if (ext2_bg_has_super(sb, i) && - !ext2_test_bit(0, bitmap_bh->b_data)) - ext2_error(sb, __FUNCTION__, - "Superblock in group %d is marked free", i); - - desc_blocks = ext2_bg_num_gdb(sb, i); - for (j = 0; j < desc_blocks; j++) - if (!ext2_test_bit(j + 1, bitmap_bh->b_data)) - ext2_error(sb, __FUNCTION__, - "Descriptor block #%ld in group " - "%d is marked free", j, i); - - if (!block_in_use(le32_to_cpu(desc->bg_block_bitmap), - sb, bitmap_bh->b_data)) - ext2_error(sb, "ext2_check_blocks_bitmap", - "Block bitmap for group %d is marked free", - i); - - if (!block_in_use(le32_to_cpu(desc->bg_inode_bitmap), - sb, bitmap_bh->b_data)) - ext2_error(sb, "ext2_check_blocks_bitmap", - "Inode bitmap for group %d is marked free", - i); - - for (j = 0; j < EXT2_SB(sb)->s_itb_per_group; j++) - if (!block_in_use(le32_to_cpu(desc->bg_inode_table) + j, - sb, bitmap_bh->b_data)) - ext2_error (sb, "ext2_check_blocks_bitmap", - "Block #%ld of the inode table in " - "group %d is marked free", j, i); - - x = ext2_count_free(bitmap_bh, sb->s_blocksize); - if (le16_to_cpu(desc->bg_free_blocks_count) != x) - ext2_error (sb, "ext2_check_blocks_bitmap", - "Wrong free blocks count for group %d, " - "stored = %d, counted = %lu", i, - le16_to_cpu(desc->bg_free_blocks_count), x); - bitmap_count += x; - } - if (le32_to_cpu(es->s_free_blocks_count) != bitmap_count) - ext2_error (sb, "ext2_check_blocks_bitmap", - "Wrong free blocks count in super block, " - "stored = %lu, counted = %lu", - (unsigned long)le32_to_cpu(es->s_free_blocks_count), - bitmap_count); - brelse(bitmap_bh); -} -#endif diff --git a/fs/ext2/ialloc.c b/fs/ext2/ialloc.c index c8d07030c897..74714af4ae69 100644 --- a/fs/ext2/ialloc.c +++ b/fs/ext2/ialloc.c @@ -605,27 +605,28 @@ got: insert_inode_hash(inode); if (DQUOT_ALLOC_INODE(inode)) { - DQUOT_DROP(inode); err = -ENOSPC; - goto fail2; + goto fail_drop; } + err = ext2_init_acl(inode, dir); - if (err) { - DQUOT_FREE_INODE(inode); - DQUOT_DROP(inode); - goto fail2; - } + if (err) + goto fail_free_drop; + err = ext2_init_security(inode,dir); - if (err) { - DQUOT_FREE_INODE(inode); - goto fail2; - } + if (err) + goto fail_free_drop; + mark_inode_dirty(inode); ext2_debug("allocating inode %lu\n", inode->i_ino); ext2_preread_inode(inode); return inode; -fail2: +fail_free_drop: + DQUOT_FREE_INODE(inode); + +fail_drop: + DQUOT_DROP(inode); inode->i_flags |= S_NOQUOTA; inode->i_nlink = 0; iput(inode); @@ -699,43 +700,3 @@ unsigned long ext2_count_dirs (struct super_block * sb) return count; } -#ifdef CONFIG_EXT2_CHECK -/* Called at mount-time, super-block is locked */ -void ext2_check_inodes_bitmap (struct super_block * sb) -{ - struct ext2_super_block * es = EXT2_SB(sb)->s_es; - unsigned long desc_count = 0, bitmap_count = 0; - struct buffer_head *bitmap_bh = NULL; - int i; - - for (i = 0; i < EXT2_SB(sb)->s_groups_count; i++) { - struct ext2_group_desc *desc; - unsigned x; - - desc = ext2_get_group_desc(sb, i, NULL); - if (!desc) - continue; - desc_count += le16_to_cpu(desc->bg_free_inodes_count); - brelse(bitmap_bh); - bitmap_bh = read_inode_bitmap(sb, i); - if (!bitmap_bh) - continue; - - x = ext2_count_free(bitmap_bh, EXT2_INODES_PER_GROUP(sb) / 8); - if (le16_to_cpu(desc->bg_free_inodes_count) != x) - ext2_error (sb, "ext2_check_inodes_bitmap", - "Wrong free inodes count in group %d, " - "stored = %d, counted = %lu", i, - le16_to_cpu(desc->bg_free_inodes_count), x); - bitmap_count += x; - } - brelse(bitmap_bh); - if (percpu_counter_read(&EXT2_SB(sb)->s_freeinodes_counter) != - bitmap_count) - ext2_error(sb, "ext2_check_inodes_bitmap", - "Wrong free inodes count in super block, " - "stored = %lu, counted = %lu", - (unsigned long)le32_to_cpu(es->s_free_inodes_count), - bitmap_count); -} -#endif diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c index fdba4d1d3c60..e7d3f0522d01 100644 --- a/fs/ext2/inode.c +++ b/fs/ext2/inode.c @@ -440,6 +440,10 @@ static int ext2_alloc_branch(struct inode *inode, * the pointer to new one, then send parent to disk. */ bh = sb_getblk(inode->i_sb, parent); + if (!bh) { + err = -EIO; + break; + } lock_buffer(bh); memset(bh->b_data, 0, blocksize); branch[n].bh = bh; diff --git a/fs/ext2/super.c b/fs/ext2/super.c index 3c0c7c6a5b44..522fa70dd8ea 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -281,7 +281,7 @@ static unsigned long get_sb_block(void **data) enum { Opt_bsd_df, Opt_minix_df, Opt_grpid, Opt_nogrpid, Opt_resgid, Opt_resuid, Opt_sb, Opt_err_cont, Opt_err_panic, - Opt_err_ro, Opt_nouid32, Opt_check, Opt_nocheck, Opt_debug, + Opt_err_ro, Opt_nouid32, Opt_nocheck, Opt_debug, Opt_oldalloc, Opt_orlov, Opt_nobh, Opt_user_xattr, Opt_nouser_xattr, Opt_acl, Opt_noacl, Opt_xip, Opt_ignore, Opt_err, Opt_quota, Opt_usrquota, Opt_grpquota @@ -303,7 +303,6 @@ static match_table_t tokens = { {Opt_nouid32, "nouid32"}, {Opt_nocheck, "check=none"}, {Opt_nocheck, "nocheck"}, - {Opt_check, "check"}, {Opt_debug, "debug"}, {Opt_oldalloc, "oldalloc"}, {Opt_orlov, "orlov"}, @@ -376,13 +375,6 @@ static int parse_options (char * options, case Opt_nouid32: set_opt (sbi->s_mount_opt, NO_UID32); break; - case Opt_check: -#ifdef CONFIG_EXT2_CHECK - set_opt (sbi->s_mount_opt, CHECK); -#else - printk("EXT2 Check option not supported\n"); -#endif - break; case Opt_nocheck: clear_opt (sbi->s_mount_opt, CHECK); break; @@ -503,12 +495,6 @@ static int ext2_setup_super (struct super_block * sb, EXT2_BLOCKS_PER_GROUP(sb), EXT2_INODES_PER_GROUP(sb), sbi->s_mount_opt); -#ifdef CONFIG_EXT2_CHECK - if (test_opt (sb, CHECK)) { - ext2_check_blocks_bitmap (sb); - ext2_check_inodes_bitmap (sb); - } -#endif return res; } @@ -895,7 +881,7 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent) } if (EXT2_HAS_COMPAT_FEATURE(sb, EXT3_FEATURE_COMPAT_HAS_JOURNAL)) ext2_warning(sb, __FUNCTION__, - "mounting ext3 filesystem as ext2\n"); + "mounting ext3 filesystem as ext2"); ext2_setup_super (sb, es, sb->s_flags & MS_RDONLY); percpu_counter_mod(&sbi->s_freeblocks_counter, ext2_count_free_blocks(sb)); diff --git a/fs/ext3/balloc.c b/fs/ext3/balloc.c index e463dca008e4..ae1148c24c53 100644 --- a/fs/ext3/balloc.c +++ b/fs/ext3/balloc.c @@ -20,6 +20,8 @@ #include <linux/quotaops.h> #include <linux/buffer_head.h> +#include "bitmap.h" + /* * balloc.c contains the blocks allocation and deallocation routines */ @@ -1010,7 +1012,7 @@ retry: * allocation within the reservation window. * * This will avoid keeping on searching the reservation list again and - * again when someboday is looking for a free block (without + * again when somebody is looking for a free block (without * reservation), and there are lots of free blocks, but they are all * being reserved. * @@ -1410,18 +1412,19 @@ unsigned long ext3_count_free_blocks(struct super_block *sb) unsigned long desc_count; struct ext3_group_desc *gdp; int i; - unsigned long ngroups; + unsigned long ngroups = EXT3_SB(sb)->s_groups_count; #ifdef EXT3FS_DEBUG struct ext3_super_block *es; unsigned long bitmap_count, x; struct buffer_head *bitmap_bh = NULL; - lock_super(sb); es = EXT3_SB(sb)->s_es; desc_count = 0; bitmap_count = 0; gdp = NULL; - for (i = 0; i < EXT3_SB(sb)->s_groups_count; i++) { + + smp_rmb(); + for (i = 0; i < ngroups; i++) { gdp = ext3_get_group_desc(sb, i, NULL); if (!gdp) continue; @@ -1439,11 +1442,9 @@ unsigned long ext3_count_free_blocks(struct super_block *sb) brelse(bitmap_bh); printk("ext3_count_free_blocks: stored = %u, computed = %lu, %lu\n", le32_to_cpu(es->s_free_blocks_count), desc_count, bitmap_count); - unlock_super(sb); return bitmap_count; #else desc_count = 0; - ngroups = EXT3_SB(sb)->s_groups_count; smp_rmb(); for (i = 0; i < ngroups; i++) { gdp = ext3_get_group_desc(sb, i, NULL); @@ -1516,76 +1517,3 @@ unsigned long ext3_bg_num_gdb(struct super_block *sb, int group) return EXT3_SB(sb)->s_gdb_count; } -#ifdef CONFIG_EXT3_CHECK -/* Called at mount-time, super-block is locked */ -void ext3_check_blocks_bitmap (struct super_block * sb) -{ - struct ext3_super_block *es; - unsigned long desc_count, bitmap_count, x, j; - unsigned long desc_blocks; - struct buffer_head *bitmap_bh = NULL; - struct ext3_group_desc *gdp; - int i; - - es = EXT3_SB(sb)->s_es; - desc_count = 0; - bitmap_count = 0; - gdp = NULL; - for (i = 0; i < EXT3_SB(sb)->s_groups_count; i++) { - gdp = ext3_get_group_desc (sb, i, NULL); - if (!gdp) - continue; - desc_count += le16_to_cpu(gdp->bg_free_blocks_count); - brelse(bitmap_bh); - bitmap_bh = read_block_bitmap(sb, i); - if (bitmap_bh == NULL) - continue; - - if (ext3_bg_has_super(sb, i) && - !ext3_test_bit(0, bitmap_bh->b_data)) - ext3_error(sb, __FUNCTION__, - "Superblock in group %d is marked free", i); - - desc_blocks = ext3_bg_num_gdb(sb, i); - for (j = 0; j < desc_blocks; j++) - if (!ext3_test_bit(j + 1, bitmap_bh->b_data)) - ext3_error(sb, __FUNCTION__, - "Descriptor block #%ld in group " - "%d is marked free", j, i); - - if (!block_in_use (le32_to_cpu(gdp->bg_block_bitmap), - sb, bitmap_bh->b_data)) - ext3_error (sb, "ext3_check_blocks_bitmap", - "Block bitmap for group %d is marked free", - i); - - if (!block_in_use (le32_to_cpu(gdp->bg_inode_bitmap), - sb, bitmap_bh->b_data)) - ext3_error (sb, "ext3_check_blocks_bitmap", - "Inode bitmap for group %d is marked free", - i); - - for (j = 0; j < EXT3_SB(sb)->s_itb_per_group; j++) - if (!block_in_use (le32_to_cpu(gdp->bg_inode_table) + j, - sb, bitmap_bh->b_data)) - ext3_error (sb, "ext3_check_blocks_bitmap", - "Block #%d of the inode table in " - "group %d is marked free", j, i); - - x = ext3_count_free(bitmap_bh, sb->s_blocksize); - if (le16_to_cpu(gdp->bg_free_blocks_count) != x) - ext3_error (sb, "ext3_check_blocks_bitmap", - "Wrong free blocks count for group %d, " - "stored = %d, counted = %lu", i, - le16_to_cpu(gdp->bg_free_blocks_count), x); - bitmap_count += x; - } - brelse(bitmap_bh); - if (le32_to_cpu(es->s_free_blocks_count) != bitmap_count) - ext3_error (sb, "ext3_check_blocks_bitmap", - "Wrong free blocks count in super block, " - "stored = %lu, counted = %lu", - (unsigned long)le32_to_cpu(es->s_free_blocks_count), - bitmap_count); -} -#endif diff --git a/fs/ext3/bitmap.c b/fs/ext3/bitmap.c index 6c419b9ab0e8..5b4ba3e246e6 100644 --- a/fs/ext3/bitmap.c +++ b/fs/ext3/bitmap.c @@ -8,7 +8,7 @@ */ #include <linux/buffer_head.h> - +#include "bitmap.h" static int nibblemap[] = {4, 3, 3, 2, 3, 2, 2, 1, 3, 2, 2, 1, 2, 1, 1, 0}; diff --git a/fs/ext3/bitmap.h b/fs/ext3/bitmap.h new file mode 100644 index 000000000000..6ee503a6bb4e --- /dev/null +++ b/fs/ext3/bitmap.h @@ -0,0 +1,8 @@ +/* linux/fs/ext3/bitmap.c + * + * Copyright (C) 2005 Simtec Electronics + * Ben Dooks <ben@simtec.co.uk> + * +*/ + +extern unsigned long ext3_count_free (struct buffer_head *, unsigned int ); diff --git a/fs/ext3/ialloc.c b/fs/ext3/ialloc.c index 96552769d039..9e4a24376210 100644 --- a/fs/ext3/ialloc.c +++ b/fs/ext3/ialloc.c @@ -26,6 +26,7 @@ #include <asm/byteorder.h> +#include "bitmap.h" #include "xattr.h" #include "acl.h" @@ -597,27 +598,22 @@ got: ret = inode; if(DQUOT_ALLOC_INODE(inode)) { - DQUOT_DROP(inode); err = -EDQUOT; - goto fail2; + goto fail_drop; } + err = ext3_init_acl(handle, inode, dir); - if (err) { - DQUOT_FREE_INODE(inode); - DQUOT_DROP(inode); - goto fail2; - } + if (err) + goto fail_free_drop; + err = ext3_init_security(handle,inode, dir); - if (err) { - DQUOT_FREE_INODE(inode); - goto fail2; - } + if (err) + goto fail_free_drop; + err = ext3_mark_inode_dirty(handle, inode); if (err) { ext3_std_error(sb, err); - DQUOT_FREE_INODE(inode); - DQUOT_DROP(inode); - goto fail2; + goto fail_free_drop; } ext3_debug("allocating inode %lu\n", inode->i_ino); @@ -631,7 +627,11 @@ really_out: brelse(bitmap_bh); return ret; -fail2: +fail_free_drop: + DQUOT_FREE_INODE(inode); + +fail_drop: + DQUOT_DROP(inode); inode->i_flags |= S_NOQUOTA; inode->i_nlink = 0; iput(inode); @@ -705,7 +705,6 @@ unsigned long ext3_count_free_inodes (struct super_block * sb) unsigned long bitmap_count, x; struct buffer_head *bitmap_bh = NULL; - lock_super (sb); es = EXT3_SB(sb)->s_es; desc_count = 0; bitmap_count = 0; @@ -728,7 +727,6 @@ unsigned long ext3_count_free_inodes (struct super_block * sb) brelse(bitmap_bh); printk("ext3_count_free_inodes: stored = %u, computed = %lu, %lu\n", le32_to_cpu(es->s_free_inodes_count), desc_count, bitmap_count); - unlock_super(sb); return desc_count; #else desc_count = 0; @@ -758,44 +756,3 @@ unsigned long ext3_count_dirs (struct super_block * sb) return count; } -#ifdef CONFIG_EXT3_CHECK -/* Called at mount-time, super-block is locked */ -void ext3_check_inodes_bitmap (struct super_block * sb) -{ - struct ext3_super_block * es; - unsigned long desc_count, bitmap_count, x; - struct buffer_head *bitmap_bh = NULL; - struct ext3_group_desc * gdp; - int i; - - es = EXT3_SB(sb)->s_es; - desc_count = 0; - bitmap_count = 0; - gdp = NULL; - for (i = 0; i < EXT3_SB(sb)->s_groups_count; i++) { - gdp = ext3_get_group_desc (sb, i, NULL); - if (!gdp) - continue; - desc_count += le16_to_cpu(gdp->bg_free_inodes_count); - brelse(bitmap_bh); - bitmap_bh = read_inode_bitmap(sb, i); - if (!bitmap_bh) - continue; - - x = ext3_count_free(bitmap_bh, EXT3_INODES_PER_GROUP(sb) / 8); - if (le16_to_cpu(gdp->bg_free_inodes_count) != x) - ext3_error (sb, "ext3_check_inodes_bitmap", - "Wrong free inodes count in group %d, " - "stored = %d, counted = %lu", i, - le16_to_cpu(gdp->bg_free_inodes_count), x); - bitmap_count += x; - } - brelse(bitmap_bh); - if (le32_to_cpu(es->s_free_inodes_count) != bitmap_count) - ext3_error (sb, "ext3_check_inodes_bitmap", - "Wrong free inodes count in super block, " - "stored = %lu, counted = %lu", - (unsigned long)le32_to_cpu(es->s_free_inodes_count), - bitmap_count); -} -#endif diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c index b5177c90d6f1..8824e84f8a56 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -491,7 +491,7 @@ static unsigned long ext3_find_goal(struct inode *inode, long block, * the same format as ext3_get_branch() would do. We are calling it after * we had read the existing part of chain and partial points to the last * triple of that (one with zero ->key). Upon the exit we have the same - * picture as after the successful ext3_get_block(), excpet that in one + * picture as after the successful ext3_get_block(), except that in one * place chain is disconnected - *branch->p is still zero (we did not * set the last link), but branch->key contains the number that should * be placed into *branch->p to fill that gap. @@ -523,7 +523,6 @@ static int ext3_alloc_branch(handle_t *handle, struct inode *inode, if (!nr) break; branch[n].key = cpu_to_le32(nr); - keys = n+1; /* * Get buffer_head for parent block, zero it out @@ -531,6 +530,9 @@ static int ext3_alloc_branch(handle_t *handle, struct inode *inode, * parent to disk. */ bh = sb_getblk(inode->i_sb, parent); + if (!bh) + break; + keys = n+1; branch[n].bh = bh; lock_buffer(bh); BUFFER_TRACE(bh, "call get_create_access"); @@ -864,6 +866,10 @@ struct buffer_head *ext3_getblk(handle_t *handle, struct inode * inode, if (!*errp && buffer_mapped(&dummy)) { struct buffer_head *bh; bh = sb_getblk(inode->i_sb, dummy.b_blocknr); + if (!bh) { + *errp = -EIO; + goto err; + } if (buffer_new(&dummy)) { J_ASSERT(create != 0); J_ASSERT(handle != 0); @@ -896,6 +902,7 @@ struct buffer_head *ext3_getblk(handle_t *handle, struct inode * inode, } return bh; } +err: return NULL; } @@ -1377,8 +1384,10 @@ static int ext3_journalled_writepage(struct page *page, ClearPageChecked(page); ret = block_prepare_write(page, 0, PAGE_CACHE_SIZE, ext3_get_block); - if (ret != 0) + if (ret != 0) { + ext3_journal_stop(handle); goto out_unlock; + } ret = walk_page_buffers(handle, page_buffers(page), 0, PAGE_CACHE_SIZE, NULL, do_journal_get_write_access); @@ -1434,7 +1443,7 @@ static int ext3_invalidatepage(struct page *page, unsigned long offset) return journal_invalidatepage(journal, page, offset); } -static int ext3_releasepage(struct page *page, int wait) +static int ext3_releasepage(struct page *page, gfp_t wait) { journal_t *journal = EXT3_JOURNAL(page->mapping->host); diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c index 50378d8ff84b..b3c690a3b54a 100644 --- a/fs/ext3/namei.c +++ b/fs/ext3/namei.c @@ -36,6 +36,8 @@ #include <linux/quotaops.h> #include <linux/buffer_head.h> #include <linux/smp_lock.h> + +#include "namei.h" #include "xattr.h" #include "acl.h" diff --git a/fs/ext3/namei.h b/fs/ext3/namei.h new file mode 100644 index 000000000000..f2ce2b0065c9 --- /dev/null +++ b/fs/ext3/namei.h @@ -0,0 +1,8 @@ +/* linux/fs/ext3/namei.h + * + * Copyright (C) 2005 Simtec Electronics + * Ben Dooks <ben@simtec.co.uk> + * +*/ + +extern struct dentry *ext3_get_parent(struct dentry *child); diff --git a/fs/ext3/resize.c b/fs/ext3/resize.c index 2c9f81278d5d..6104ad310507 100644 --- a/fs/ext3/resize.c +++ b/fs/ext3/resize.c @@ -118,6 +118,8 @@ static struct buffer_head *bclean(handle_t *handle, struct super_block *sb, int err; bh = sb_getblk(sb, blk); + if (!bh) + return ERR_PTR(-EIO); if ((err = ext3_journal_get_write_access(handle, bh))) { brelse(bh); bh = ERR_PTR(err); @@ -202,6 +204,10 @@ static int setup_new_group_blocks(struct super_block *sb, ext3_debug("update backup group %#04lx (+%d)\n", block, bit); gdb = sb_getblk(sb, block); + if (!gdb) { + err = -EIO; + goto exit_bh; + } if ((err = ext3_journal_get_write_access(handle, gdb))) { brelse(gdb); goto exit_bh; @@ -242,7 +248,7 @@ static int setup_new_group_blocks(struct super_block *sb, i < sbi->s_itb_per_group; i++, bit++, block++) { struct buffer_head *it; - ext3_debug("clear inode block %#04x (+%ld)\n", block, bit); + ext3_debug("clear inode block %#04lx (+%d)\n", block, bit); if (IS_ERR(it = bclean(handle, sb, block))) { err = PTR_ERR(it); goto exit_bh; @@ -643,8 +649,12 @@ static void update_backups(struct super_block *sb, break; bh = sb_getblk(sb, group * bpg + blk_off); - ext3_debug(sb, __FUNCTION__, "update metadata backup %#04lx\n", - bh->b_blocknr); + if (!bh) { + err = -EIO; + break; + } + ext3_debug("update metadata backup %#04lx\n", + (unsigned long)bh->b_blocknr); if ((err = ext3_journal_get_write_access(handle, bh))) break; lock_buffer(bh); @@ -757,6 +767,7 @@ int ext3_group_add(struct super_block *sb, struct ext3_new_group_data *input) if (input->group != EXT3_SB(sb)->s_groups_count) { ext3_warning(sb, __FUNCTION__, "multiple resizers run on filesystem!\n"); + err = -EBUSY; goto exit_journal; } diff --git a/fs/ext3/super.c b/fs/ext3/super.c index a93c3609025d..4e6730622d90 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -36,9 +36,12 @@ #include <linux/namei.h> #include <linux/quotaops.h> #include <linux/seq_file.h> + #include <asm/uaccess.h> + #include "xattr.h" #include "acl.h" +#include "namei.h" static int ext3_load_journal(struct super_block *, struct ext3_super_block *); static int ext3_create_journal(struct super_block *, struct ext3_super_block *, @@ -510,20 +513,11 @@ static void ext3_clear_inode(struct inode *inode) kfree(rsv); } -static int ext3_show_options(struct seq_file *seq, struct vfsmount *vfs) +static inline void ext3_show_quota_options(struct seq_file *seq, struct super_block *sb) { - struct ext3_sb_info *sbi = EXT3_SB(vfs->mnt_sb); - - if (sbi->s_mount_opt & EXT3_MOUNT_JOURNAL_DATA) - seq_puts(seq, ",data=journal"); - - if (sbi->s_mount_opt & EXT3_MOUNT_ORDERED_DATA) - seq_puts(seq, ",data=ordered"); - - if (sbi->s_mount_opt & EXT3_MOUNT_WRITEBACK_DATA) - seq_puts(seq, ",data=writeback"); - #if defined(CONFIG_QUOTA) + struct ext3_sb_info *sbi = EXT3_SB(sb); + if (sbi->s_jquota_fmt) seq_printf(seq, ",jqfmt=%s", (sbi->s_jquota_fmt == QFMT_VFS_OLD) ? "vfsold": "vfsv0"); @@ -540,6 +534,20 @@ static int ext3_show_options(struct seq_file *seq, struct vfsmount *vfs) if (sbi->s_mount_opt & EXT3_MOUNT_GRPQUOTA) seq_puts(seq, ",grpquota"); #endif +} + +static int ext3_show_options(struct seq_file *seq, struct vfsmount *vfs) +{ + struct super_block *sb = vfs->mnt_sb; + + if (test_opt(sb, DATA_FLAGS) == EXT3_MOUNT_JOURNAL_DATA) + seq_puts(seq, ",data=journal"); + else if (test_opt(sb, DATA_FLAGS) == EXT3_MOUNT_ORDERED_DATA) + seq_puts(seq, ",data=ordered"); + else if (test_opt(sb, DATA_FLAGS) == EXT3_MOUNT_WRITEBACK_DATA) + seq_puts(seq, ",data=writeback"); + + ext3_show_quota_options(seq, sb); return 0; } @@ -610,7 +618,6 @@ static struct super_operations ext3_sops = { #endif }; -struct dentry *ext3_get_parent(struct dentry *child); static struct export_operations ext3_export_ops = { .get_parent = ext3_get_parent, }; @@ -618,7 +625,7 @@ static struct export_operations ext3_export_ops = { enum { Opt_bsd_df, Opt_minix_df, Opt_grpid, Opt_nogrpid, Opt_resgid, Opt_resuid, Opt_sb, Opt_err_cont, Opt_err_panic, Opt_err_ro, - Opt_nouid32, Opt_check, Opt_nocheck, Opt_debug, Opt_oldalloc, Opt_orlov, + Opt_nouid32, Opt_nocheck, Opt_debug, Opt_oldalloc, Opt_orlov, Opt_user_xattr, Opt_nouser_xattr, Opt_acl, Opt_noacl, Opt_reservation, Opt_noreservation, Opt_noload, Opt_nobh, Opt_commit, Opt_journal_update, Opt_journal_inum, @@ -645,7 +652,6 @@ static match_table_t tokens = { {Opt_nouid32, "nouid32"}, {Opt_nocheck, "nocheck"}, {Opt_nocheck, "check=none"}, - {Opt_check, "check"}, {Opt_debug, "debug"}, {Opt_oldalloc, "oldalloc"}, {Opt_orlov, "orlov"}, @@ -766,14 +772,6 @@ static int parse_options (char * options, struct super_block *sb, case Opt_nouid32: set_opt (sbi->s_mount_opt, NO_UID32); break; - case Opt_check: -#ifdef CONFIG_EXT3_CHECK - set_opt (sbi->s_mount_opt, CHECK); -#else - printk(KERN_ERR - "EXT3 Check option not supported\n"); -#endif - break; case Opt_nocheck: clear_opt (sbi->s_mount_opt, CHECK); break; @@ -1108,12 +1106,6 @@ static int ext3_setup_super(struct super_block *sb, struct ext3_super_block *es, } else { printk("internal journal\n"); } -#ifdef CONFIG_EXT3_CHECK - if (test_opt (sb, CHECK)) { - ext3_check_blocks_bitmap (sb); - ext3_check_inodes_bitmap (sb); - } -#endif return res; } diff --git a/fs/ext3/xattr.c b/fs/ext3/xattr.c index 269c7b92db9a..430de9f63be3 100644 --- a/fs/ext3/xattr.c +++ b/fs/ext3/xattr.c @@ -210,7 +210,7 @@ ext3_xattr_find_entry(struct ext3_xattr_entry **pentry, int name_index, return cmp ? -ENODATA : 0; } -int +static int ext3_xattr_block_get(struct inode *inode, int name_index, const char *name, void *buffer, size_t buffer_size) { @@ -354,7 +354,7 @@ ext3_xattr_list_entries(struct inode *inode, struct ext3_xattr_entry *entry, return buffer_size - rest; } -int +static int ext3_xattr_block_list(struct inode *inode, char *buffer, size_t buffer_size) { struct buffer_head *bh = NULL; @@ -626,7 +626,7 @@ struct ext3_xattr_block_find { struct buffer_head *bh; }; -int +static int ext3_xattr_block_find(struct inode *inode, struct ext3_xattr_info *i, struct ext3_xattr_block_find *bs) { @@ -859,7 +859,7 @@ struct ext3_xattr_ibody_find { struct ext3_iloc iloc; }; -int +static int ext3_xattr_ibody_find(struct inode *inode, struct ext3_xattr_info *i, struct ext3_xattr_ibody_find *is) { diff --git a/fs/fat/dir.c b/fs/fat/dir.c index 895049b2ac9c..ba824964b9bb 100644 --- a/fs/fat/dir.c +++ b/fs/fat/dir.c @@ -222,6 +222,80 @@ fat_shortname2uni(struct nls_table *nls, unsigned char *buf, int buf_size, return len; } +enum { PARSE_INVALID = 1, PARSE_NOT_LONGNAME, PARSE_EOF, }; + +/** + * fat_parse_long - Parse extended directory entry. + * + * This function returns zero on success, negative value on error, or one of + * the following: + * + * %PARSE_INVALID - Directory entry is invalid. + * %PARSE_NOT_LONGNAME - Directory entry does not contain longname. + * %PARSE_EOF - Directory has no more entries. + */ +static int fat_parse_long(struct inode *dir, loff_t *pos, + struct buffer_head **bh, struct msdos_dir_entry **de, + wchar_t **unicode, unsigned char *nr_slots) +{ + struct msdos_dir_slot *ds; + unsigned char id, slot, slots, alias_checksum; + + if (!*unicode) { + *unicode = (wchar_t *)__get_free_page(GFP_KERNEL); + if (!*unicode) { + brelse(*bh); + return -ENOMEM; + } + } +parse_long: + slots = 0; + ds = (struct msdos_dir_slot *)*de; + id = ds->id; + if (!(id & 0x40)) + return PARSE_INVALID; + slots = id & ~0x40; + if (slots > 20 || !slots) /* ceil(256 * 2 / 26) */ + return PARSE_INVALID; + *nr_slots = slots; + alias_checksum = ds->alias_checksum; + + slot = slots; + while (1) { + int offset; + + slot--; + offset = slot * 13; + fat16_towchar(*unicode + offset, ds->name0_4, 5); + fat16_towchar(*unicode + offset + 5, ds->name5_10, 6); + fat16_towchar(*unicode + offset + 11, ds->name11_12, 2); + + if (ds->id & 0x40) + (*unicode)[offset + 13] = 0; + if (fat_get_entry(dir, pos, bh, de) < 0) + return PARSE_EOF; + if (slot == 0) + break; + ds = (struct msdos_dir_slot *)*de; + if (ds->attr != ATTR_EXT) + return PARSE_NOT_LONGNAME; + if ((ds->id & ~0x40) != slot) + goto parse_long; + if (ds->alias_checksum != alias_checksum) + goto parse_long; + } + if ((*de)->name[0] == DELETED_FLAG) + return PARSE_INVALID; + if ((*de)->attr == ATTR_EXT) + goto parse_long; + if (IS_FREE((*de)->name) || ((*de)->attr & ATTR_VOLUME)) + return PARSE_INVALID; + if (fat_checksum((*de)->name) != alias_checksum) + *nr_slots = 0; + + return 0; +} + /* * Return values: negative -> error, 0 -> not found, positive -> found, * value is the total amount of slots, including the shortname entry. @@ -259,68 +333,16 @@ parse_record: if (de->attr != ATTR_EXT && IS_FREE(de->name)) continue; if (de->attr == ATTR_EXT) { - struct msdos_dir_slot *ds; - unsigned char id; - unsigned char slot; - unsigned char slots; - unsigned char sum; - unsigned char alias_checksum; - - if (!unicode) { - unicode = (wchar_t *) - __get_free_page(GFP_KERNEL); - if (!unicode) { - brelse(bh); - return -ENOMEM; - } - } -parse_long: - slots = 0; - ds = (struct msdos_dir_slot *) de; - id = ds->id; - if (!(id & 0x40)) - continue; - slots = id & ~0x40; - if (slots > 20 || !slots) /* ceil(256 * 2 / 26) */ - continue; - nr_slots = slots; - alias_checksum = ds->alias_checksum; - - slot = slots; - while (1) { - int offset; - - slot--; - offset = slot * 13; - fat16_towchar(unicode + offset, ds->name0_4, 5); - fat16_towchar(unicode + offset + 5, ds->name5_10, 6); - fat16_towchar(unicode + offset + 11, ds->name11_12, 2); - - if (ds->id & 0x40) { - unicode[offset + 13] = 0; - } - if (fat_get_entry(inode, &cpos, &bh, &de) < 0) - goto EODir; - if (slot == 0) - break; - ds = (struct msdos_dir_slot *) de; - if (ds->attr != ATTR_EXT) - goto parse_record; - if ((ds->id & ~0x40) != slot) - goto parse_long; - if (ds->alias_checksum != alias_checksum) - goto parse_long; - } - if (de->name[0] == DELETED_FLAG) - continue; - if (de->attr == ATTR_EXT) - goto parse_long; - if (IS_FREE(de->name) || (de->attr & ATTR_VOLUME)) + int status = fat_parse_long(inode, &cpos, &bh, &de, + &unicode, &nr_slots); + if (status < 0) + return status; + else if (status == PARSE_INVALID) continue; - for (sum = 0, i = 0; i < 11; i++) - sum = (((sum&1)<<7)|((sum&0xfe)>>1)) + de->name[i]; - if (sum != alias_checksum) - nr_slots = 0; + else if (status == PARSE_NOT_LONGNAME) + goto parse_record; + else if (status == PARSE_EOF) + goto EODir; } memcpy(work, de->name, sizeof(de->name)); @@ -408,8 +430,8 @@ struct fat_ioctl_filldir_callback { int short_len; }; -static int fat_readdirx(struct inode *inode, struct file *filp, void *dirent, - filldir_t filldir, int short_only, int both) +static int __fat_readdir(struct inode *inode, struct file *filp, void *dirent, + filldir_t filldir, int short_only, int both) { struct super_block *sb = inode->i_sb; struct msdos_sb_info *sbi = MSDOS_SB(sb); @@ -458,9 +480,10 @@ static int fat_readdirx(struct inode *inode, struct file *filp, void *dirent, bh = NULL; GetNew: - long_slots = 0; if (fat_get_entry(inode, &cpos, &bh, &de) == -1) goto EODir; +parse_record: + long_slots = 0; /* Check for long filename entry */ if (isvfat) { if (de->name[0] == DELETED_FLAG) @@ -475,69 +498,18 @@ GetNew: } if (isvfat && de->attr == ATTR_EXT) { - struct msdos_dir_slot *ds; - unsigned char id; - unsigned char slot; - unsigned char slots; - unsigned char sum; - unsigned char alias_checksum; - - if (!unicode) { - unicode = (wchar_t *)__get_free_page(GFP_KERNEL); - if (!unicode) { - filp->f_pos = cpos; - brelse(bh); - ret = -ENOMEM; - goto out; - } - } -ParseLong: - slots = 0; - ds = (struct msdos_dir_slot *) de; - id = ds->id; - if (!(id & 0x40)) - goto RecEnd; - slots = id & ~0x40; - if (slots > 20 || !slots) /* ceil(256 * 2 / 26) */ + int status = fat_parse_long(inode, &cpos, &bh, &de, + &unicode, &long_slots); + if (status < 0) { + filp->f_pos = cpos; + ret = status; + goto out; + } else if (status == PARSE_INVALID) goto RecEnd; - long_slots = slots; - alias_checksum = ds->alias_checksum; - - slot = slots; - while (1) { - int offset; - - slot--; - offset = slot * 13; - fat16_towchar(unicode + offset, ds->name0_4, 5); - fat16_towchar(unicode + offset + 5, ds->name5_10, 6); - fat16_towchar(unicode + offset + 11, ds->name11_12, 2); - - if (ds->id & 0x40) { - unicode[offset + 13] = 0; - } - if (fat_get_entry(inode, &cpos, &bh, &de) == -1) - goto EODir; - if (slot == 0) - break; - ds = (struct msdos_dir_slot *) de; - if (ds->attr != ATTR_EXT) - goto RecEnd; /* XXX */ - if ((ds->id & ~0x40) != slot) - goto ParseLong; - if (ds->alias_checksum != alias_checksum) - goto ParseLong; - } - if (de->name[0] == DELETED_FLAG) - goto RecEnd; - if (de->attr == ATTR_EXT) - goto ParseLong; - if (IS_FREE(de->name) || (de->attr & ATTR_VOLUME)) - goto RecEnd; - for (sum = 0, i = 0; i < 11; i++) - sum = (((sum&1)<<7)|((sum&0xfe)>>1)) + de->name[i]; - if (sum != alias_checksum) - long_slots = 0; + else if (status == PARSE_NOT_LONGNAME) + goto parse_record; + else if (status == PARSE_EOF) + goto EODir; } if (sbi->options.dotsOK) { @@ -671,7 +643,7 @@ out: static int fat_readdir(struct file *filp, void *dirent, filldir_t filldir) { struct inode *inode = filp->f_dentry->d_inode; - return fat_readdirx(inode, filp, dirent, filldir, 0, 0); + return __fat_readdir(inode, filp, dirent, filldir, 0, 0); } static int fat_ioctl_filldir(void *__buf, const char *name, int name_len, @@ -760,8 +732,8 @@ static int fat_dir_ioctl(struct inode * inode, struct file * filp, down(&inode->i_sem); ret = -ENOENT; if (!IS_DEADDIR(inode)) { - ret = fat_readdirx(inode, filp, &buf, fat_ioctl_filldir, - short_only, both); + ret = __fat_readdir(inode, filp, &buf, fat_ioctl_filldir, + short_only, both); } up(&inode->i_sem); if (ret >= 0) diff --git a/fs/fat/inode.c b/fs/fat/inode.c index e2effe2dc9b2..a0f9b9fe1307 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -846,7 +846,7 @@ static match_table_t vfat_tokens = { {Opt_err, NULL} }; -static int parse_options(char *options, int is_vfat, int *debug, +static int parse_options(char *options, int is_vfat, int silent, int *debug, struct fat_mount_options *opts) { char *p; @@ -1008,8 +1008,11 @@ static int parse_options(char *options, int is_vfat, int *debug, break; /* unknown option */ default: - printk(KERN_ERR "FAT: Unrecognized mount option \"%s\" " - "or missing value\n", p); + if (!silent) { + printk(KERN_ERR + "FAT: Unrecognized mount option \"%s\" " + "or missing value\n", p); + } return -EINVAL; } } @@ -1091,7 +1094,7 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, sb->s_export_op = &fat_export_ops; sbi->dir_ops = fs_dir_inode_ops; - error = parse_options(data, isvfat, &debug, &sbi->options); + error = parse_options(data, isvfat, silent, &debug, &sbi->options); if (error) goto out_fail; diff --git a/fs/file_table.c b/fs/file_table.c index 86ec8ae985b4..c3a5e2fd663b 100644 --- a/fs/file_table.c +++ b/fs/file_table.c @@ -35,7 +35,7 @@ static DEFINE_SPINLOCK(filp_count_lock); * context and must be fully threaded - use a local spinlock * to protect files_stat.nr_files */ -void filp_ctor(void * objp, struct kmem_cache_s *cachep, unsigned long cflags) +void filp_ctor(void *objp, struct kmem_cache *cachep, unsigned long cflags) { if ((cflags & (SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR)) == SLAB_CTOR_CONSTRUCTOR) { @@ -46,7 +46,7 @@ void filp_ctor(void * objp, struct kmem_cache_s *cachep, unsigned long cflags) } } -void filp_dtor(void * objp, struct kmem_cache_s *cachep, unsigned long dflags) +void filp_dtor(void *objp, struct kmem_cache *cachep, unsigned long dflags) { unsigned long flags; spin_lock_irqsave(&filp_count_lock, flags); @@ -56,13 +56,13 @@ void filp_dtor(void * objp, struct kmem_cache_s *cachep, unsigned long dflags) static inline void file_free_rcu(struct rcu_head *head) { - struct file *f = container_of(head, struct file, f_rcuhead); + struct file *f = container_of(head, struct file, f_u.fu_rcuhead); kmem_cache_free(filp_cachep, f); } static inline void file_free(struct file *f) { - call_rcu(&f->f_rcuhead, file_free_rcu); + call_rcu(&f->f_u.fu_rcuhead, file_free_rcu); } /* Find an unused file structure and return a pointer to it. @@ -95,7 +95,7 @@ struct file *get_empty_filp(void) f->f_gid = current->fsgid; rwlock_init(&f->f_owner.lock); /* f->f_version: 0 */ - INIT_LIST_HEAD(&f->f_list); + INIT_LIST_HEAD(&f->f_u.fu_list); return f; over: @@ -225,15 +225,15 @@ void file_move(struct file *file, struct list_head *list) if (!list) return; file_list_lock(); - list_move(&file->f_list, list); + list_move(&file->f_u.fu_list, list); file_list_unlock(); } void file_kill(struct file *file) { - if (!list_empty(&file->f_list)) { + if (!list_empty(&file->f_u.fu_list)) { file_list_lock(); - list_del_init(&file->f_list); + list_del_init(&file->f_u.fu_list); file_list_unlock(); } } @@ -245,7 +245,7 @@ int fs_may_remount_ro(struct super_block *sb) /* Check that no files are currently opened for writing. */ file_list_lock(); list_for_each(p, &sb->s_files) { - struct file *file = list_entry(p, struct file, f_list); + struct file *file = list_entry(p, struct file, f_u.fu_list); struct inode *inode = file->f_dentry->d_inode; /* File with pending delete? */ diff --git a/fs/filesystems.c b/fs/filesystems.c index 44082bfdfec9..9f1072836c8e 100644 --- a/fs/filesystems.c +++ b/fs/filesystems.c @@ -12,6 +12,7 @@ #include <linux/kmod.h> #include <linux/init.h> #include <linux/module.h> +#include <linux/sched.h> /* for 'current' */ #include <asm/uaccess.h> /* diff --git a/fs/freevxfs/vxfs_bmap.c b/fs/freevxfs/vxfs_bmap.c index d3f6b2835bc8..2d71128bd8d6 100644 --- a/fs/freevxfs/vxfs_bmap.c +++ b/fs/freevxfs/vxfs_bmap.c @@ -36,6 +36,7 @@ #include "vxfs.h" #include "vxfs_inode.h" +#include "vxfs_extern.h" #ifdef DIAGNOSTIC diff --git a/fs/freevxfs/vxfs_extern.h b/fs/freevxfs/vxfs_extern.h index d8be917f9797..927acf70c591 100644 --- a/fs/freevxfs/vxfs_extern.h +++ b/fs/freevxfs/vxfs_extern.h @@ -38,7 +38,7 @@ */ -struct kmem_cache_s; +struct kmem_cache; struct super_block; struct vxfs_inode_info; struct inode; @@ -51,7 +51,7 @@ extern daddr_t vxfs_bmap1(struct inode *, long); extern int vxfs_read_fshead(struct super_block *); /* vxfs_inode.c */ -extern struct kmem_cache_s *vxfs_inode_cachep; +extern struct kmem_cache *vxfs_inode_cachep; extern void vxfs_dumpi(struct vxfs_inode_info *, ino_t); extern struct inode * vxfs_get_fake_inode(struct super_block *, struct vxfs_inode_info *); diff --git a/fs/freevxfs/vxfs_inode.c b/fs/freevxfs/vxfs_inode.c index 9672d2facffe..f544aae9169f 100644 --- a/fs/freevxfs/vxfs_inode.c +++ b/fs/freevxfs/vxfs_inode.c @@ -46,15 +46,6 @@ extern struct address_space_operations vxfs_immed_aops; extern struct inode_operations vxfs_immed_symlink_iops; -static struct file_operations vxfs_file_operations = { - .open = generic_file_open, - .llseek = generic_file_llseek, - .read = generic_file_read, - .mmap = generic_file_mmap, - .sendfile = generic_file_sendfile, -}; - - kmem_cache_t *vxfs_inode_cachep; @@ -318,7 +309,7 @@ vxfs_read_inode(struct inode *ip) aops = &vxfs_aops; if (S_ISREG(ip->i_mode)) { - ip->i_fop = &vxfs_file_operations; + ip->i_fop = &generic_ro_fops; ip->i_mapping->a_ops = aops; } else if (S_ISDIR(ip->i_mode)) { ip->i_op = &vxfs_dir_inode_ops; diff --git a/fs/freevxfs/vxfs_olt.c b/fs/freevxfs/vxfs_olt.c index 133476201d84..76a0708ae978 100644 --- a/fs/freevxfs/vxfs_olt.c +++ b/fs/freevxfs/vxfs_olt.c @@ -36,6 +36,7 @@ #include "vxfs.h" #include "vxfs_olt.h" +#include "vxfs_extern.h" static inline void diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index e94ab398b717..785c7213a54f 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -230,7 +230,6 @@ __sync_single_inode(struct inode *inode, struct writeback_control *wbc) * The inode is clean, unused */ list_move(&inode->i_list, &inode_unused); - inodes_stat.nr_unused++; } } wake_up_inode(inode); @@ -238,14 +237,20 @@ __sync_single_inode(struct inode *inode, struct writeback_control *wbc) } /* - * Write out an inode's dirty pages. Called under inode_lock. + * Write out an inode's dirty pages. Called under inode_lock. Either the + * caller has ref on the inode (either via __iget or via syscall against an fd) + * or the inode has I_WILL_FREE set (via generic_forget_inode) */ static int -__writeback_single_inode(struct inode *inode, - struct writeback_control *wbc) +__writeback_single_inode(struct inode *inode, struct writeback_control *wbc) { wait_queue_head_t *wqh; + if (!atomic_read(&inode->i_count)) + WARN_ON(!(inode->i_state & (I_WILL_FREE|I_FREEING))); + else + WARN_ON(inode->i_state & I_WILL_FREE); + if ((wbc->sync_mode != WB_SYNC_ALL) && (inode->i_state & I_LOCK)) { list_move(&inode->i_list, &inode->i_sb->s_dirty); return 0; @@ -259,11 +264,9 @@ __writeback_single_inode(struct inode *inode, wqh = bit_waitqueue(&inode->i_state, __I_LOCK); do { - __iget(inode); spin_unlock(&inode_lock); __wait_on_bit(wqh, &wq, inode_wait, TASK_UNINTERRUPTIBLE); - iput(inode); spin_lock(&inode_lock); } while (inode->i_state & I_LOCK); } @@ -541,14 +544,15 @@ void sync_inodes(int wait) } /** - * write_inode_now - write an inode to disk - * @inode: inode to write to disk - * @sync: whether the write should be synchronous or not + * write_inode_now - write an inode to disk + * @inode: inode to write to disk + * @sync: whether the write should be synchronous or not + * + * This function commits an inode to disk immediately if it is dirty. This is + * primarily needed by knfsd. * - * This function commits an inode to disk immediately if it is - * dirty. This is primarily needed by knfsd. + * The caller must either have a ref on the inode or must have set I_WILL_FREE. */ - int write_inode_now(struct inode *inode, int sync) { int ret; @@ -558,7 +562,7 @@ int write_inode_now(struct inode *inode, int sync) }; if (!mapping_cap_writeback_dirty(inode->i_mapping)) - return 0; + wbc.nr_to_write = 0; might_sleep(); spin_lock(&inode_lock); @@ -602,7 +606,7 @@ EXPORT_SYMBOL(sync_inode); * O_SYNC flag set, to flush dirty writes to disk. * * @what is a bitmask, specifying which part of the inode's data should be - * written and waited upon: + * written and waited upon. * * OSYNC_DATA: i_mapping's dirty data * OSYNC_METADATA: the buffers at i_mapping->private_list @@ -668,8 +672,9 @@ int writeback_acquire(struct backing_dev_info *bdi) /** * writeback_in_progress: determine whether there is writeback in progress - * against a backing device. * @bdi: the device's backing_dev_info structure. + * + * Determine whether there is writeback in progress against a backing device. */ int writeback_in_progress(struct backing_dev_info *bdi) { diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c index d4c869c6d01b..8f873e621f41 100644 --- a/fs/fuse/dev.c +++ b/fs/fuse/dev.c @@ -151,9 +151,9 @@ void fuse_release_background(struct fuse_req *req) /* * This function is called when a request is finished. Either a reply * has arrived or it was interrupted (and not yet sent) or some error - * occured during communication with userspace, or the device file was - * closed. It decreases the referece count for the request. In case - * of a background request the referece to the stored objects are + * occurred during communication with userspace, or the device file was + * closed. It decreases the reference count for the request. In case + * of a background request the reference to the stored objects are * released. The requester thread is woken up (if still waiting), and * finally the request is either freed or put on the unused_list * @@ -184,6 +184,13 @@ static void request_end(struct fuse_conn *fc, struct fuse_req *req) fuse_putback_request() */ for (i = 1; i < FUSE_MAX_OUTSTANDING; i++) up(&fc->outstanding_sem); + } else if (req->in.h.opcode == FUSE_RELEASE && req->inode == NULL) { + /* Special case for failed iget in CREATE */ + u64 nodeid = req->in.h.nodeid; + __fuse_get_request(req); + fuse_reset_request(req); + fuse_send_forget(fc, req, nodeid, 1); + putback = 0; } if (putback) fuse_putback_request(fc, req); diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index e79e49b3eec7..51f5da652771 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -13,6 +13,7 @@ #include <linux/gfp.h> #include <linux/sched.h> #include <linux/namei.h> +#include <linux/mount.h> static inline unsigned long time_to_jiffies(unsigned long sec, unsigned long nsec) @@ -73,6 +74,24 @@ static int fuse_dentry_revalidate(struct dentry *entry, struct nameidata *nd) return 1; } +static int dir_alias(struct inode *inode) +{ + if (S_ISDIR(inode->i_mode)) { + /* Don't allow creating an alias to a directory */ + struct dentry *alias = d_find_alias(inode); + if (alias) { + dput(alias); + return 1; + } + } + return 0; +} + +static inline int invalid_nodeid(u64 nodeid) +{ + return !nodeid || nodeid == FUSE_ROOT_ID; +} + static struct dentry_operations fuse_dentry_operations = { .d_revalidate = fuse_dentry_revalidate, }; @@ -96,6 +115,8 @@ static int fuse_lookup_iget(struct inode *dir, struct dentry *entry, fuse_lookup_init(req, dir, entry, &outarg); request_send(fc, req); err = req->out.h.error; + if (!err && invalid_nodeid(outarg.nodeid)) + err = -EIO; if (!err) { inode = fuse_iget(dir->i_sb, outarg.nodeid, outarg.generation, &outarg.attr); @@ -132,6 +153,101 @@ static void fuse_invalidate_entry(struct dentry *entry) entry->d_time = jiffies - 1; } +static int fuse_create_open(struct inode *dir, struct dentry *entry, int mode, + struct nameidata *nd) +{ + int err; + struct inode *inode; + struct fuse_conn *fc = get_fuse_conn(dir); + struct fuse_req *req; + struct fuse_open_in inarg; + struct fuse_open_out outopen; + struct fuse_entry_out outentry; + struct fuse_inode *fi; + struct fuse_file *ff; + struct file *file; + int flags = nd->intent.open.flags - 1; + + err = -ENOSYS; + if (fc->no_create) + goto out; + + err = -ENAMETOOLONG; + if (entry->d_name.len > FUSE_NAME_MAX) + goto out; + + err = -EINTR; + req = fuse_get_request(fc); + if (!req) + goto out; + + ff = fuse_file_alloc(); + if (!ff) + goto out_put_request; + + flags &= ~O_NOCTTY; + memset(&inarg, 0, sizeof(inarg)); + inarg.flags = flags; + inarg.mode = mode; + req->in.h.opcode = FUSE_CREATE; + req->in.h.nodeid = get_node_id(dir); + req->inode = dir; + req->in.numargs = 2; + req->in.args[0].size = sizeof(inarg); + req->in.args[0].value = &inarg; + req->in.args[1].size = entry->d_name.len + 1; + req->in.args[1].value = entry->d_name.name; + req->out.numargs = 2; + req->out.args[0].size = sizeof(outentry); + req->out.args[0].value = &outentry; + req->out.args[1].size = sizeof(outopen); + req->out.args[1].value = &outopen; + request_send(fc, req); + err = req->out.h.error; + if (err) { + if (err == -ENOSYS) + fc->no_create = 1; + goto out_free_ff; + } + + err = -EIO; + if (!S_ISREG(outentry.attr.mode) || invalid_nodeid(outentry.nodeid)) + goto out_free_ff; + + inode = fuse_iget(dir->i_sb, outentry.nodeid, outentry.generation, + &outentry.attr); + err = -ENOMEM; + if (!inode) { + flags &= ~(O_CREAT | O_EXCL | O_TRUNC); + ff->fh = outopen.fh; + fuse_send_release(fc, ff, outentry.nodeid, NULL, flags, 0); + goto out_put_request; + } + fuse_put_request(fc, req); + entry->d_time = time_to_jiffies(outentry.entry_valid, + outentry.entry_valid_nsec); + fi = get_fuse_inode(inode); + fi->i_time = time_to_jiffies(outentry.attr_valid, + outentry.attr_valid_nsec); + + d_instantiate(entry, inode); + file = lookup_instantiate_filp(nd, entry, generic_file_open); + if (IS_ERR(file)) { + ff->fh = outopen.fh; + fuse_send_release(fc, ff, outentry.nodeid, inode, flags, 0); + return PTR_ERR(file); + } + fuse_finish_open(inode, file, ff, &outopen); + return 0; + + out_free_ff: + fuse_file_free(ff); + out_put_request: + fuse_put_request(fc, req); + out: + return err; +} + static int create_new_entry(struct fuse_conn *fc, struct fuse_req *req, struct inode *dir, struct dentry *entry, int mode) @@ -152,6 +268,10 @@ static int create_new_entry(struct fuse_conn *fc, struct fuse_req *req, fuse_put_request(fc, req); return err; } + if (invalid_nodeid(outarg.nodeid)) { + fuse_put_request(fc, req); + return -EIO; + } inode = fuse_iget(dir->i_sb, outarg.nodeid, outarg.generation, &outarg.attr); if (!inode) { @@ -161,7 +281,7 @@ static int create_new_entry(struct fuse_conn *fc, struct fuse_req *req, fuse_put_request(fc, req); /* Don't allow userspace to do really stupid things... */ - if ((inode->i_mode ^ mode) & S_IFMT) { + if (((inode->i_mode ^ mode) & S_IFMT) || dir_alias(inode)) { iput(inode); return -EIO; } @@ -202,6 +322,12 @@ static int fuse_mknod(struct inode *dir, struct dentry *entry, int mode, static int fuse_create(struct inode *dir, struct dentry *entry, int mode, struct nameidata *nd) { + if (nd && (nd->flags & LOOKUP_CREATE)) { + int err = fuse_create_open(dir, entry, mode, nd); + if (err != -ENOSYS) + return err; + /* Fall back on mknod */ + } return fuse_mknod(dir, entry, mode, 0); } @@ -455,6 +581,38 @@ static int fuse_revalidate(struct dentry *entry) return fuse_do_getattr(inode); } +static int fuse_access(struct inode *inode, int mask) +{ + struct fuse_conn *fc = get_fuse_conn(inode); + struct fuse_req *req; + struct fuse_access_in inarg; + int err; + + if (fc->no_access) + return 0; + + req = fuse_get_request(fc); + if (!req) + return -EINTR; + + memset(&inarg, 0, sizeof(inarg)); + inarg.mask = mask; + req->in.h.opcode = FUSE_ACCESS; + req->in.h.nodeid = get_node_id(inode); + req->inode = inode; + req->in.numargs = 1; + req->in.args[0].size = sizeof(inarg); + req->in.args[0].value = &inarg; + request_send(fc, req); + err = req->out.h.error; + fuse_put_request(fc, req); + if (err == -ENOSYS) { + fc->no_access = 1; + err = 0; + } + return err; +} + static int fuse_permission(struct inode *inode, int mask, struct nameidata *nd) { struct fuse_conn *fc = get_fuse_conn(inode); @@ -485,11 +643,11 @@ static int fuse_permission(struct inode *inode, int mask, struct nameidata *nd) return err; } else { int mode = inode->i_mode; - if ((mask & MAY_WRITE) && IS_RDONLY(inode) && - (S_ISREG(mode) || S_ISDIR(mode) || S_ISLNK(mode))) - return -EROFS; if ((mask & MAY_EXEC) && !S_ISDIR(mode) && !(mode & S_IXUGO)) return -EACCES; + + if (nd && (nd->flags & LOOKUP_ACCESS)) + return fuse_access(inode, mask); return 0; } } @@ -623,29 +781,29 @@ static int fuse_dir_fsync(struct file *file, struct dentry *de, int datasync) return file ? fuse_fsync_common(file, de, datasync, 1) : 0; } -static unsigned iattr_to_fattr(struct iattr *iattr, struct fuse_attr *fattr) +static void iattr_to_fattr(struct iattr *iattr, struct fuse_setattr_in *arg) { unsigned ivalid = iattr->ia_valid; - unsigned fvalid = 0; - - memset(fattr, 0, sizeof(*fattr)); if (ivalid & ATTR_MODE) - fvalid |= FATTR_MODE, fattr->mode = iattr->ia_mode; + arg->valid |= FATTR_MODE, arg->mode = iattr->ia_mode; if (ivalid & ATTR_UID) - fvalid |= FATTR_UID, fattr->uid = iattr->ia_uid; + arg->valid |= FATTR_UID, arg->uid = iattr->ia_uid; if (ivalid & ATTR_GID) - fvalid |= FATTR_GID, fattr->gid = iattr->ia_gid; + arg->valid |= FATTR_GID, arg->gid = iattr->ia_gid; if (ivalid & ATTR_SIZE) - fvalid |= FATTR_SIZE, fattr->size = iattr->ia_size; + arg->valid |= FATTR_SIZE, arg->size = iattr->ia_size; /* You can only _set_ these together (they may change by themselves) */ if ((ivalid & (ATTR_ATIME | ATTR_MTIME)) == (ATTR_ATIME | ATTR_MTIME)) { - fvalid |= FATTR_ATIME | FATTR_MTIME; - fattr->atime = iattr->ia_atime.tv_sec; - fattr->mtime = iattr->ia_mtime.tv_sec; + arg->valid |= FATTR_ATIME | FATTR_MTIME; + arg->atime = iattr->ia_atime.tv_sec; + arg->mtime = iattr->ia_mtime.tv_sec; + } + if (ivalid & ATTR_FILE) { + struct fuse_file *ff = iattr->ia_file->private_data; + arg->valid |= FATTR_FH; + arg->fh = ff->fh; } - - return fvalid; } static int fuse_setattr(struct dentry *entry, struct iattr *attr) @@ -680,7 +838,7 @@ static int fuse_setattr(struct dentry *entry, struct iattr *attr) return -EINTR; memset(&inarg, 0, sizeof(inarg)); - inarg.valid = iattr_to_fattr(attr, &inarg.attr); + iattr_to_fattr(attr, &inarg); req->in.h.opcode = FUSE_SETATTR; req->in.h.nodeid = get_node_id(inode); req->inode = inode; @@ -729,19 +887,17 @@ static struct dentry *fuse_lookup(struct inode *dir, struct dentry *entry, struct nameidata *nd) { struct inode *inode; - int err = fuse_lookup_iget(dir, entry, &inode); + int err; + + err = fuse_lookup_iget(dir, entry, &inode); if (err) return ERR_PTR(err); - if (inode && S_ISDIR(inode->i_mode)) { - /* Don't allow creating an alias to a directory */ - struct dentry *alias = d_find_alias(inode); - if (alias && !(alias->d_flags & DCACHE_DISCONNECTED)) { - dput(alias); - iput(inode); - return ERR_PTR(-EIO); - } + if (inode && dir_alias(inode)) { + iput(inode); + return ERR_PTR(-EIO); } - return d_splice_alias(inode, entry); + d_add(entry, inode); + return NULL; } static int fuse_setxattr(struct dentry *entry, const char *name, diff --git a/fs/fuse/file.c b/fs/fuse/file.c index 6454022b0536..2ca86141d13a 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -14,42 +14,18 @@ static struct file_operations fuse_direct_io_file_operations; -int fuse_open_common(struct inode *inode, struct file *file, int isdir) +static int fuse_send_open(struct inode *inode, struct file *file, int isdir, + struct fuse_open_out *outargp) { struct fuse_conn *fc = get_fuse_conn(inode); - struct fuse_req *req; struct fuse_open_in inarg; - struct fuse_open_out outarg; - struct fuse_file *ff; + struct fuse_req *req; int err; - err = generic_file_open(inode, file); - if (err) - return err; - - /* If opening the root node, no lookup has been performed on - it, so the attributes must be refreshed */ - if (get_node_id(inode) == FUSE_ROOT_ID) { - int err = fuse_do_getattr(inode); - if (err) - return err; - } - req = fuse_get_request(fc); if (!req) return -EINTR; - err = -ENOMEM; - ff = kmalloc(sizeof(struct fuse_file), GFP_KERNEL); - if (!ff) - goto out_put_request; - - ff->release_req = fuse_request_alloc(); - if (!ff->release_req) { - kfree(ff); - goto out_put_request; - } - memset(&inarg, 0, sizeof(inarg)); inarg.flags = file->f_flags & ~(O_CREAT | O_EXCL | O_NOCTTY | O_TRUNC); req->in.h.opcode = isdir ? FUSE_OPENDIR : FUSE_OPEN; @@ -59,44 +35,110 @@ int fuse_open_common(struct inode *inode, struct file *file, int isdir) req->in.args[0].size = sizeof(inarg); req->in.args[0].value = &inarg; req->out.numargs = 1; - req->out.args[0].size = sizeof(outarg); - req->out.args[0].value = &outarg; + req->out.args[0].size = sizeof(*outargp); + req->out.args[0].value = outargp; request_send(fc, req); err = req->out.h.error; - if (err) { - fuse_request_free(ff->release_req); - kfree(ff); - } else { - if (!isdir && (outarg.open_flags & FOPEN_DIRECT_IO)) - file->f_op = &fuse_direct_io_file_operations; - if (!(outarg.open_flags & FOPEN_KEEP_CACHE)) - invalidate_inode_pages(inode->i_mapping); - ff->fh = outarg.fh; - file->private_data = ff; + fuse_put_request(fc, req); + + return err; +} + +struct fuse_file *fuse_file_alloc(void) +{ + struct fuse_file *ff; + ff = kmalloc(sizeof(struct fuse_file), GFP_KERNEL); + if (ff) { + ff->release_req = fuse_request_alloc(); + if (!ff->release_req) { + kfree(ff); + ff = NULL; + } + } + return ff; +} + +void fuse_file_free(struct fuse_file *ff) +{ + fuse_request_free(ff->release_req); + kfree(ff); +} + +void fuse_finish_open(struct inode *inode, struct file *file, + struct fuse_file *ff, struct fuse_open_out *outarg) +{ + if (outarg->open_flags & FOPEN_DIRECT_IO) + file->f_op = &fuse_direct_io_file_operations; + if (!(outarg->open_flags & FOPEN_KEEP_CACHE)) + invalidate_inode_pages(inode->i_mapping); + ff->fh = outarg->fh; + file->private_data = ff; +} + +int fuse_open_common(struct inode *inode, struct file *file, int isdir) +{ + struct fuse_open_out outarg; + struct fuse_file *ff; + int err; + + /* VFS checks this, but only _after_ ->open() */ + if (file->f_flags & O_DIRECT) + return -EINVAL; + + err = generic_file_open(inode, file); + if (err) + return err; + + /* If opening the root node, no lookup has been performed on + it, so the attributes must be refreshed */ + if (get_node_id(inode) == FUSE_ROOT_ID) { + err = fuse_do_getattr(inode); + if (err) + return err; + } + + ff = fuse_file_alloc(); + if (!ff) + return -ENOMEM; + + err = fuse_send_open(inode, file, isdir, &outarg); + if (err) + fuse_file_free(ff); + else { + if (isdir) + outarg.open_flags &= ~FOPEN_DIRECT_IO; + fuse_finish_open(inode, file, ff, &outarg); } - out_put_request: - fuse_put_request(fc, req); return err; } -int fuse_release_common(struct inode *inode, struct file *file, int isdir) +void fuse_send_release(struct fuse_conn *fc, struct fuse_file *ff, + u64 nodeid, struct inode *inode, int flags, int isdir) { - struct fuse_conn *fc = get_fuse_conn(inode); - struct fuse_file *ff = file->private_data; - struct fuse_req *req = ff->release_req; + struct fuse_req * req = ff->release_req; struct fuse_release_in *inarg = &req->misc.release_in; inarg->fh = ff->fh; - inarg->flags = file->f_flags & ~O_EXCL; + inarg->flags = flags; req->in.h.opcode = isdir ? FUSE_RELEASEDIR : FUSE_RELEASE; - req->in.h.nodeid = get_node_id(inode); + req->in.h.nodeid = nodeid; req->inode = inode; req->in.numargs = 1; req->in.args[0].size = sizeof(struct fuse_release_in); req->in.args[0].value = inarg; request_send_background(fc, req); kfree(ff); +} + +int fuse_release_common(struct inode *inode, struct file *file, int isdir) +{ + struct fuse_file *ff = file->private_data; + if (ff) { + struct fuse_conn *fc = get_fuse_conn(inode); + u64 nodeid = get_node_id(inode); + fuse_send_release(fc, ff, nodeid, inode, file->f_flags, isdir); + } /* Return value is ignored by VFS */ return 0; diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h index 24d761518d86..0ea5301f86be 100644 --- a/fs/fuse/fuse_i.h +++ b/fs/fuse/fuse_i.h @@ -266,6 +266,12 @@ struct fuse_conn { /** Is removexattr not implemented by fs? */ unsigned no_removexattr : 1; + /** Is access not implemented by fs? */ + unsigned no_access : 1; + + /** Is create not implemented by fs? */ + unsigned no_create : 1; + /** Backing dev info */ struct backing_dev_info bdi; }; @@ -337,6 +343,17 @@ size_t fuse_send_read_common(struct fuse_req *req, struct file *file, */ int fuse_open_common(struct inode *inode, struct file *file, int isdir); +struct fuse_file *fuse_file_alloc(void); +void fuse_file_free(struct fuse_file *ff); +void fuse_finish_open(struct inode *inode, struct file *file, + struct fuse_file *ff, struct fuse_open_out *outarg); + +/** + * Send a RELEASE request + */ +void fuse_send_release(struct fuse_conn *fc, struct fuse_file *ff, + u64 nodeid, struct inode *inode, int flags, int isdir); + /** * Send RELEASE or RELEASEDIR request */ @@ -349,22 +366,22 @@ int fuse_fsync_common(struct file *file, struct dentry *de, int datasync, int isdir); /** - * Initialise file operations on a regular file + * Initialize file operations on a regular file */ void fuse_init_file_inode(struct inode *inode); /** - * Initialise inode operations on regular files and special files + * Initialize inode operations on regular files and special files */ void fuse_init_common(struct inode *inode); /** - * Initialise inode and file operations on a directory + * Initialize inode and file operations on a directory */ void fuse_init_dir(struct inode *inode); /** - * Initialise inode operations on a symlink + * Initialize inode operations on a symlink */ void fuse_init_symlink(struct inode *inode); @@ -411,7 +428,7 @@ struct fuse_req *fuse_get_request(struct fuse_conn *fc); /** * Decrement reference count of a request. If count goes to zero put - * on unused list (preallocated) or free reqest (not preallocated). + * on unused list (preallocated) or free request (not preallocated). */ void fuse_put_request(struct fuse_conn *fc, struct fuse_req *req); @@ -431,7 +448,7 @@ void request_send_noreply(struct fuse_conn *fc, struct fuse_req *req); void request_send_background(struct fuse_conn *fc, struct fuse_req *req); /** - * Release inodes and file assiciated with background request + * Release inodes and file associated with background request */ void fuse_release_background(struct fuse_req *req); diff --git a/fs/hfs/hfs_fs.h b/fs/hfs/hfs_fs.h index aae019aadf88..cc5dcd52e23d 100644 --- a/fs/hfs/hfs_fs.h +++ b/fs/hfs/hfs_fs.h @@ -9,7 +9,6 @@ #ifndef _LINUX_HFS_FS_H #define _LINUX_HFS_FS_H -#include <linux/version.h> #include <linux/slab.h> #include <linux/types.h> #include <linux/buffer_head.h> diff --git a/fs/hfs/inode.c b/fs/hfs/inode.c index f1570b9f9de3..d499393a8ae7 100644 --- a/fs/hfs/inode.c +++ b/fs/hfs/inode.c @@ -12,7 +12,6 @@ */ #include <linux/pagemap.h> -#include <linux/version.h> #include <linux/mpage.h> #include "hfs_fs.h" @@ -46,7 +45,7 @@ static sector_t hfs_bmap(struct address_space *mapping, sector_t block) return generic_block_bmap(mapping, block, hfs_get_block); } -static int hfs_releasepage(struct page *page, int mask) +static int hfs_releasepage(struct page *page, gfp_t mask) { struct inode *inode = page->mapping->host; struct super_block *sb = inode->i_sb; diff --git a/fs/hfsplus/bnode.c b/fs/hfsplus/bnode.c index b85abc6e6f83..930cd9212de8 100644 --- a/fs/hfsplus/bnode.c +++ b/fs/hfsplus/bnode.c @@ -13,7 +13,6 @@ #include <linux/pagemap.h> #include <linux/fs.h> #include <linux/swap.h> -#include <linux/version.h> #include "hfsplus_fs.h" #include "hfsplus_raw.h" diff --git a/fs/hfsplus/dir.c b/fs/hfsplus/dir.c index 7bda76667a4a..50c8f44b6c66 100644 --- a/fs/hfsplus/dir.c +++ b/fs/hfsplus/dir.c @@ -13,7 +13,6 @@ #include <linux/sched.h> #include <linux/slab.h> #include <linux/random.h> -#include <linux/version.h> #include "hfsplus_fs.h" #include "hfsplus_raw.h" diff --git a/fs/hfsplus/extents.c b/fs/hfsplus/extents.c index e7235ca79a95..e3ff56a03011 100644 --- a/fs/hfsplus/extents.c +++ b/fs/hfsplus/extents.c @@ -11,7 +11,6 @@ #include <linux/errno.h> #include <linux/fs.h> #include <linux/pagemap.h> -#include <linux/version.h> #include "hfsplus_fs.h" #include "hfsplus_raw.h" diff --git a/fs/hfsplus/hfsplus_fs.h b/fs/hfsplus/hfsplus_fs.h index 2bc0cdd30e56..df16fcbff3fb 100644 --- a/fs/hfsplus/hfsplus_fs.h +++ b/fs/hfsplus/hfsplus_fs.h @@ -11,7 +11,6 @@ #define _LINUX_HFSPLUS_FS_H #include <linux/fs.h> -#include <linux/version.h> #include <linux/buffer_head.h> #include "hfsplus_raw.h" @@ -152,6 +151,7 @@ struct hfsplus_sb_info { #define HFSPLUS_SB_WRITEBACKUP 0x0001 #define HFSPLUS_SB_NODECOMPOSE 0x0002 +#define HFSPLUS_SB_FORCE 0x0004 struct hfsplus_inode_info { diff --git a/fs/hfsplus/hfsplus_raw.h b/fs/hfsplus/hfsplus_raw.h index 5bad37cfdb29..b4fbed633219 100644 --- a/fs/hfsplus/hfsplus_raw.h +++ b/fs/hfsplus/hfsplus_raw.h @@ -123,11 +123,13 @@ struct hfsplus_vh { } __packed; /* HFS+ volume attributes */ -#define HFSPLUS_VOL_UNMNT (1 << 8) -#define HFSPLUS_VOL_SPARE_BLK (1 << 9) -#define HFSPLUS_VOL_NOCACHE (1 << 10) -#define HFSPLUS_VOL_INCNSTNT (1 << 11) -#define HFSPLUS_VOL_SOFTLOCK (1 << 15) +#define HFSPLUS_VOL_UNMNT (1 << 8) +#define HFSPLUS_VOL_SPARE_BLK (1 << 9) +#define HFSPLUS_VOL_NOCACHE (1 << 10) +#define HFSPLUS_VOL_INCNSTNT (1 << 11) +#define HFSPLUS_VOL_NODEID_REUSED (1 << 12) +#define HFSPLUS_VOL_JOURNALED (1 << 13) +#define HFSPLUS_VOL_SOFTLOCK (1 << 15) /* HFS+ BTree node descriptor */ struct hfs_bnode_desc { diff --git a/fs/hfsplus/inode.c b/fs/hfsplus/inode.c index d5642705f633..fc98583cf045 100644 --- a/fs/hfsplus/inode.c +++ b/fs/hfsplus/inode.c @@ -11,7 +11,6 @@ #include <linux/mm.h> #include <linux/fs.h> #include <linux/pagemap.h> -#include <linux/version.h> #include <linux/mpage.h> #include "hfsplus_fs.h" @@ -40,7 +39,7 @@ static sector_t hfsplus_bmap(struct address_space *mapping, sector_t block) return generic_block_bmap(mapping, block, hfsplus_get_block); } -static int hfsplus_releasepage(struct page *page, int mask) +static int hfsplus_releasepage(struct page *page, gfp_t mask) { struct inode *inode = page->mapping->host; struct super_block *sb = inode->i_sb; diff --git a/fs/hfsplus/options.c b/fs/hfsplus/options.c index cca0818aa4ca..935dafba0078 100644 --- a/fs/hfsplus/options.c +++ b/fs/hfsplus/options.c @@ -22,7 +22,7 @@ enum { opt_umask, opt_uid, opt_gid, opt_part, opt_session, opt_nls, opt_nodecompose, opt_decompose, - opt_err + opt_force, opt_err }; static match_table_t tokens = { @@ -36,6 +36,7 @@ static match_table_t tokens = { { opt_nls, "nls=%s" }, { opt_decompose, "decompose" }, { opt_nodecompose, "nodecompose" }, + { opt_force, "force" }, { opt_err, NULL } }; @@ -145,6 +146,9 @@ int hfsplus_parse_options(char *input, struct hfsplus_sb_info *sbi) case opt_nodecompose: sbi->flags |= HFSPLUS_SB_NODECOMPOSE; break; + case opt_force: + sbi->flags |= HFSPLUS_SB_FORCE; + break; default: return 0; } diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c index fd0f0f050e1d..8093351bd7c3 100644 --- a/fs/hfsplus/super.c +++ b/fs/hfsplus/super.c @@ -14,7 +14,6 @@ #include <linux/fs.h> #include <linux/sched.h> #include <linux/slab.h> -#include <linux/version.h> #include <linux/vfs.h> #include <linux/nls.h> @@ -50,6 +49,7 @@ static void hfsplus_read_inode(struct inode *inode) init_MUTEX(&HFSPLUS_I(inode).extents_lock); HFSPLUS_I(inode).flags = 0; HFSPLUS_I(inode).rsrc_inode = NULL; + atomic_set(&HFSPLUS_I(inode).opencnt, 0); if (inode->i_ino >= HFSPLUS_FIRSTUSER_CNID) { read_inode: @@ -251,16 +251,28 @@ static int hfsplus_remount(struct super_block *sb, int *flags, char *data) return 0; if (!(*flags & MS_RDONLY)) { struct hfsplus_vh *vhdr = HFSPLUS_SB(sb).s_vhdr; + struct hfsplus_sb_info sbi; + + memset(&sbi, 0, sizeof(struct hfsplus_sb_info)); + sbi.nls = HFSPLUS_SB(sb).nls; + if (!hfsplus_parse_options(data, &sbi)) + return -EINVAL; if (!(vhdr->attributes & cpu_to_be32(HFSPLUS_VOL_UNMNT))) { printk("HFS+-fs warning: Filesystem was not cleanly unmounted, " "running fsck.hfsplus is recommended. leaving read-only.\n"); sb->s_flags |= MS_RDONLY; *flags |= MS_RDONLY; + } else if (sbi.flags & HFSPLUS_SB_FORCE) { + /* nothing */ } else if (vhdr->attributes & cpu_to_be32(HFSPLUS_VOL_SOFTLOCK)) { printk("HFS+-fs: Filesystem is marked locked, leaving read-only.\n"); sb->s_flags |= MS_RDONLY; *flags |= MS_RDONLY; + } else if (vhdr->attributes & cpu_to_be32(HFSPLUS_VOL_JOURNALED)) { + printk("HFS+-fs: Filesystem is marked journaled, leaving read-only.\n"); + sb->s_flags |= MS_RDONLY; + *flags |= MS_RDONLY; } } return 0; @@ -352,11 +364,19 @@ static int hfsplus_fill_super(struct super_block *sb, void *data, int silent) printk("HFS+-fs warning: Filesystem was not cleanly unmounted, " "running fsck.hfsplus is recommended. mounting read-only.\n"); sb->s_flags |= MS_RDONLY; + } else if (sbi->flags & HFSPLUS_SB_FORCE) { + /* nothing */ } else if (vhdr->attributes & cpu_to_be32(HFSPLUS_VOL_SOFTLOCK)) { if (!silent) printk("HFS+-fs: Filesystem is marked locked, mounting read-only.\n"); sb->s_flags |= MS_RDONLY; + } else if (vhdr->attributes & cpu_to_be32(HFSPLUS_VOL_JOURNALED)) { + if (!silent) + printk("HFS+-fs: write access to a jounaled filesystem is not supported, " + "use the force option at your own risk, mounting read-only.\n"); + sb->s_flags |= MS_RDONLY; } + sbi->flags &= ~HFSPLUS_SB_FORCE; /* Load metadata objects (B*Trees) */ HFSPLUS_SB(sb).ext_tree = hfs_btree_open(sb, HFSPLUS_EXT_CNID); diff --git a/fs/hfsplus/wrapper.c b/fs/hfsplus/wrapper.c index 0c51d6338b0b..95455e839231 100644 --- a/fs/hfsplus/wrapper.c +++ b/fs/hfsplus/wrapper.c @@ -12,7 +12,6 @@ #include <linux/blkdev.h> #include <linux/cdrom.h> #include <linux/genhd.h> -#include <linux/version.h> #include <asm/unaligned.h> #include "hfsplus_fs.h" diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c index 59c5062cd63f..4684eb7d48c6 100644 --- a/fs/hostfs/hostfs_kern.c +++ b/fs/hostfs/hostfs_kern.c @@ -8,7 +8,6 @@ #include <linux/stddef.h> #include <linux/fs.h> -#include <linux/version.h> #include <linux/module.h> #include <linux/init.h> #include <linux/slab.h> @@ -294,8 +293,7 @@ static void hostfs_delete_inode(struct inode *inode) static void hostfs_destroy_inode(struct inode *inode) { - if(HOSTFS_I(inode)->host_filename) - kfree(HOSTFS_I(inode)->host_filename); + kfree(HOSTFS_I(inode)->host_filename); /*XXX: This should not happen, probably. The check is here for * additional safety.*/ @@ -793,11 +791,6 @@ int hostfs_rename(struct inode *from_ino, struct dentry *from, return(err); } -void hostfs_truncate(struct inode *ino) -{ - not_implemented(); -} - int hostfs_permission(struct inode *ino, int desired, struct nameidata *nd) { char *name; @@ -894,7 +887,6 @@ static struct inode_operations hostfs_iops = { .rmdir = hostfs_rmdir, .mknod = hostfs_mknod, .rename = hostfs_rename, - .truncate = hostfs_truncate, .permission = hostfs_permission, .setattr = hostfs_setattr, .getattr = hostfs_getattr, @@ -910,7 +902,6 @@ static struct inode_operations hostfs_dir_iops = { .rmdir = hostfs_rmdir, .mknod = hostfs_mknod, .rename = hostfs_rename, - .truncate = hostfs_truncate, .permission = hostfs_permission, .setattr = hostfs_setattr, .getattr = hostfs_getattr, diff --git a/fs/hpfs/dnode.c b/fs/hpfs/dnode.c index 1d21307730a8..229ff2fb1809 100644 --- a/fs/hpfs/dnode.c +++ b/fs/hpfs/dnode.c @@ -244,12 +244,12 @@ static int hpfs_add_to_dnode(struct inode *i, dnode_secno dno, go_up: if (namelen >= 256) { hpfs_error(i->i_sb, "hpfs_add_to_dnode: namelen == %d", namelen); - if (nd) kfree(nd); + kfree(nd); kfree(nname); return 1; } if (!(d = hpfs_map_dnode(i->i_sb, dno, &qbh))) { - if (nd) kfree(nd); + kfree(nd); kfree(nname); return 1; } @@ -257,7 +257,7 @@ static int hpfs_add_to_dnode(struct inode *i, dnode_secno dno, if (hpfs_sb(i->i_sb)->sb_chk) if (hpfs_stop_cycles(i->i_sb, dno, &c1, &c2, "hpfs_add_to_dnode")) { hpfs_brelse4(&qbh); - if (nd) kfree(nd); + kfree(nd); kfree(nname); return 1; } @@ -270,7 +270,7 @@ static int hpfs_add_to_dnode(struct inode *i, dnode_secno dno, for_all_poss(i, hpfs_pos_subst, 5, t + 1); hpfs_mark_4buffers_dirty(&qbh); hpfs_brelse4(&qbh); - if (nd) kfree(nd); + kfree(nd); kfree(nname); return 0; } diff --git a/fs/hpfs/file.c b/fs/hpfs/file.c index ab144dabd870..7c995ac4081b 100644 --- a/fs/hpfs/file.c +++ b/fs/hpfs/file.c @@ -114,11 +114,8 @@ static ssize_t hpfs_file_write(struct file *file, const char __user *buf, ssize_t retval; retval = generic_file_write(file, buf, count, ppos); - if (retval > 0) { - struct inode *inode = file->f_dentry->d_inode; - inode->i_mtime = CURRENT_TIME_SEC; - hpfs_i(inode)->i_dirty = 1; - } + if (retval > 0) + hpfs_i(file->f_dentry->d_inode)->i_dirty = 1; return retval; } diff --git a/fs/hpfs/super.c b/fs/hpfs/super.c index 8eefa6366db7..63e88d7e2c3b 100644 --- a/fs/hpfs/super.c +++ b/fs/hpfs/super.c @@ -75,7 +75,7 @@ void hpfs_error(struct super_block *s, char *m,...) } else if (s->s_flags & MS_RDONLY) printk("; going on - but anything won't be destroyed because it's read-only\n"); else printk("; corrupted filesystem mounted read/write - your computer will explode within 20 seconds ... but you wanted it so!\n"); } else printk("\n"); - if (buf) kfree(buf); + kfree(buf); hpfs_sb(s)->sb_was_error = 1; } @@ -102,8 +102,8 @@ int hpfs_stop_cycles(struct super_block *s, int key, int *c1, int *c2, static void hpfs_put_super(struct super_block *s) { struct hpfs_sb_info *sbi = hpfs_sb(s); - if (sbi->sb_cp_table) kfree(sbi->sb_cp_table); - if (sbi->sb_bmp_dir) kfree(sbi->sb_bmp_dir); + kfree(sbi->sb_cp_table); + kfree(sbi->sb_bmp_dir); unmark_dirty(s); s->s_fs_info = NULL; kfree(sbi); @@ -654,8 +654,8 @@ bail3: brelse(bh1); bail2: brelse(bh0); bail1: bail0: - if (sbi->sb_bmp_dir) kfree(sbi->sb_bmp_dir); - if (sbi->sb_cp_table) kfree(sbi->sb_cp_table); + kfree(sbi->sb_bmp_dir); + kfree(sbi->sb_cp_table); s->s_fs_info = NULL; kfree(sbi); return -EINVAL; diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 3a9b6d179cbd..8c1cef3bb677 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -45,10 +45,58 @@ static struct backing_dev_info hugetlbfs_backing_dev_info = { int sysctl_hugetlb_shm_group; +static void huge_pagevec_release(struct pagevec *pvec) +{ + int i; + + for (i = 0; i < pagevec_count(pvec); ++i) + put_page(pvec->pages[i]); + + pagevec_reinit(pvec); +} + +/* + * huge_pages_needed tries to determine the number of new huge pages that + * will be required to fully populate this VMA. This will be equal to + * the size of the VMA in huge pages minus the number of huge pages + * (covered by this VMA) that are found in the page cache. + * + * Result is in bytes to be compatible with is_hugepage_mem_enough() + */ +static unsigned long +huge_pages_needed(struct address_space *mapping, struct vm_area_struct *vma) +{ + int i; + struct pagevec pvec; + unsigned long start = vma->vm_start; + unsigned long end = vma->vm_end; + unsigned long hugepages = (end - start) >> HPAGE_SHIFT; + pgoff_t next = vma->vm_pgoff; + pgoff_t endpg = next + ((end - start) >> PAGE_SHIFT); + + pagevec_init(&pvec, 0); + while (next < endpg) { + if (!pagevec_lookup(&pvec, mapping, next, PAGEVEC_SIZE)) + break; + for (i = 0; i < pagevec_count(&pvec); i++) { + struct page *page = pvec.pages[i]; + if (page->index > next) + next = page->index; + if (page->index >= endpg) + break; + next++; + hugepages--; + } + huge_pagevec_release(&pvec); + } + return hugepages << HPAGE_SHIFT; +} + static int hugetlbfs_file_mmap(struct file *file, struct vm_area_struct *vma) { struct inode *inode = file->f_dentry->d_inode; struct address_space *mapping = inode->i_mapping; + unsigned long bytes; loff_t len, vma_len; int ret; @@ -67,6 +115,10 @@ static int hugetlbfs_file_mmap(struct file *file, struct vm_area_struct *vma) if (vma->vm_end - vma->vm_start < HPAGE_SIZE) return -EINVAL; + bytes = huge_pages_needed(mapping, vma); + if (!is_hugepage_mem_enough(bytes)) + return -ENOMEM; + vma_len = (loff_t)(vma->vm_end - vma->vm_start); down(&inode->i_sem); @@ -79,10 +131,8 @@ static int hugetlbfs_file_mmap(struct file *file, struct vm_area_struct *vma) if (!(vma->vm_flags & VM_WRITE) && len > inode->i_size) goto out; - ret = hugetlb_prefault(mapping, vma); - if (ret) - goto out; - + ret = 0; + hugetlb_prefault_arch_hook(vma->vm_mm); if (inode->i_size < len) inode->i_size = len; out: @@ -92,7 +142,7 @@ out: } /* - * Called under down_write(mmap_sem), page_table_lock is not held + * Called under down_write(mmap_sem). */ #ifdef HAVE_ARCH_HUGETLB_UNMAPPED_AREA @@ -171,16 +221,6 @@ static int hugetlbfs_commit_write(struct file *file, return -EINVAL; } -static void huge_pagevec_release(struct pagevec *pvec) -{ - int i; - - for (i = 0; i < pagevec_count(pvec); ++i) - put_page(pvec->pages[i]); - - pagevec_reinit(pvec); -} - static void truncate_huge_page(struct page *page) { clear_page_dirty(page); @@ -224,52 +264,35 @@ static void truncate_hugepages(struct address_space *mapping, loff_t lstart) static void hugetlbfs_delete_inode(struct inode *inode) { - struct hugetlbfs_sb_info *sbinfo = HUGETLBFS_SB(inode->i_sb); - - hlist_del_init(&inode->i_hash); - list_del_init(&inode->i_list); - list_del_init(&inode->i_sb_list); - inode->i_state |= I_FREEING; - inodes_stat.nr_inodes--; - spin_unlock(&inode_lock); - if (inode->i_data.nrpages) truncate_hugepages(&inode->i_data, 0); - - security_inode_delete(inode); - - if (sbinfo->free_inodes >= 0) { - spin_lock(&sbinfo->stat_lock); - sbinfo->free_inodes++; - spin_unlock(&sbinfo->stat_lock); - } - clear_inode(inode); - destroy_inode(inode); } static void hugetlbfs_forget_inode(struct inode *inode) { - struct super_block *super_block = inode->i_sb; - struct hugetlbfs_sb_info *sbinfo = HUGETLBFS_SB(super_block); + struct super_block *sb = inode->i_sb; - if (hlist_unhashed(&inode->i_hash)) - goto out_truncate; - - if (!(inode->i_state & (I_DIRTY|I_LOCK))) { - list_del(&inode->i_list); - list_add(&inode->i_list, &inode_unused); - } - inodes_stat.nr_unused++; - if (!super_block || (super_block->s_flags & MS_ACTIVE)) { + if (!hlist_unhashed(&inode->i_hash)) { + if (!(inode->i_state & (I_DIRTY|I_LOCK))) + list_move(&inode->i_list, &inode_unused); + inodes_stat.nr_unused++; + if (!sb || (sb->s_flags & MS_ACTIVE)) { + spin_unlock(&inode_lock); + return; + } + inode->i_state |= I_WILL_FREE; spin_unlock(&inode_lock); - return; + /* + * write_inode_now is a noop as we set BDI_CAP_NO_WRITEBACK + * in our backing_dev_info. + */ + write_inode_now(inode, 1); + spin_lock(&inode_lock); + inode->i_state &= ~I_WILL_FREE; + inodes_stat.nr_unused--; + hlist_del_init(&inode->i_hash); } - - /* write_inode_now() ? */ - inodes_stat.nr_unused--; - hlist_del_init(&inode->i_hash); -out_truncate: list_del_init(&inode->i_list); list_del_init(&inode->i_sb_list); inode->i_state |= I_FREEING; @@ -277,13 +300,6 @@ out_truncate: spin_unlock(&inode_lock); if (inode->i_data.nrpages) truncate_hugepages(&inode->i_data, 0); - - if (sbinfo->free_inodes >= 0) { - spin_lock(&sbinfo->stat_lock); - sbinfo->free_inodes++; - spin_unlock(&sbinfo->stat_lock); - } - clear_inode(inode); destroy_inode(inode); } @@ -291,7 +307,7 @@ out_truncate: static void hugetlbfs_drop_inode(struct inode *inode) { if (!inode->i_nlink) - hugetlbfs_delete_inode(inode); + generic_delete_inode(inode); else hugetlbfs_forget_inode(inode); } @@ -308,7 +324,6 @@ hugetlb_vmtruncate_list(struct prio_tree_root *root, unsigned long h_pgoff) vma_prio_tree_foreach(vma, &iter, root, h_pgoff, ULONG_MAX) { unsigned long h_vm_pgoff; - unsigned long v_length; unsigned long v_offset; h_vm_pgoff = vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT); @@ -319,11 +334,8 @@ hugetlb_vmtruncate_list(struct prio_tree_root *root, unsigned long h_pgoff) if (h_vm_pgoff >= h_pgoff) v_offset = 0; - v_length = vma->vm_end - vma->vm_start; - - zap_hugepage_range(vma, - vma->vm_start + v_offset, - v_length - v_offset); + unmap_hugepage_range(vma, + vma->vm_start + v_offset, vma->vm_end); } } @@ -379,17 +391,6 @@ static struct inode *hugetlbfs_get_inode(struct super_block *sb, uid_t uid, gid_t gid, int mode, dev_t dev) { struct inode *inode; - struct hugetlbfs_sb_info *sbinfo = HUGETLBFS_SB(sb); - - if (sbinfo->free_inodes >= 0) { - spin_lock(&sbinfo->stat_lock); - if (!sbinfo->free_inodes) { - spin_unlock(&sbinfo->stat_lock); - return NULL; - } - sbinfo->free_inodes--; - spin_unlock(&sbinfo->stat_lock); - } inode = new_inode(sb); if (inode) { @@ -511,10 +512,14 @@ static int hugetlbfs_statfs(struct super_block *sb, struct kstatfs *buf) buf->f_bsize = HPAGE_SIZE; if (sbinfo) { spin_lock(&sbinfo->stat_lock); - buf->f_blocks = sbinfo->max_blocks; - buf->f_bavail = buf->f_bfree = sbinfo->free_blocks; - buf->f_files = sbinfo->max_inodes; - buf->f_ffree = sbinfo->free_inodes; + /* If no limits set, just report 0 for max/free/used + * blocks, like simple_statfs() */ + if (sbinfo->max_blocks >= 0) { + buf->f_blocks = sbinfo->max_blocks; + buf->f_bavail = buf->f_bfree = sbinfo->free_blocks; + buf->f_files = sbinfo->max_inodes; + buf->f_ffree = sbinfo->free_inodes; + } spin_unlock(&sbinfo->stat_lock); } buf->f_namelen = NAME_MAX; @@ -531,29 +536,51 @@ static void hugetlbfs_put_super(struct super_block *sb) } } +static inline int hugetlbfs_dec_free_inodes(struct hugetlbfs_sb_info *sbinfo) +{ + if (sbinfo->free_inodes >= 0) { + spin_lock(&sbinfo->stat_lock); + if (unlikely(!sbinfo->free_inodes)) { + spin_unlock(&sbinfo->stat_lock); + return 0; + } + sbinfo->free_inodes--; + spin_unlock(&sbinfo->stat_lock); + } + + return 1; +} + +static void hugetlbfs_inc_free_inodes(struct hugetlbfs_sb_info *sbinfo) +{ + if (sbinfo->free_inodes >= 0) { + spin_lock(&sbinfo->stat_lock); + sbinfo->free_inodes++; + spin_unlock(&sbinfo->stat_lock); + } +} + + static kmem_cache_t *hugetlbfs_inode_cachep; static struct inode *hugetlbfs_alloc_inode(struct super_block *sb) { + struct hugetlbfs_sb_info *sbinfo = HUGETLBFS_SB(sb); struct hugetlbfs_inode_info *p; + if (unlikely(!hugetlbfs_dec_free_inodes(sbinfo))) + return NULL; p = kmem_cache_alloc(hugetlbfs_inode_cachep, SLAB_KERNEL); - if (!p) + if (unlikely(!p)) { + hugetlbfs_inc_free_inodes(sbinfo); return NULL; + } return &p->vfs_inode; } -static void init_once(void *foo, kmem_cache_t *cachep, unsigned long flags) -{ - struct hugetlbfs_inode_info *ei = (struct hugetlbfs_inode_info *)foo; - - if ((flags & (SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR)) == - SLAB_CTOR_CONSTRUCTOR) - inode_init_once(&ei->vfs_inode); -} - static void hugetlbfs_destroy_inode(struct inode *inode) { + hugetlbfs_inc_free_inodes(HUGETLBFS_SB(inode->i_sb)); mpol_free_shared_policy(&HUGETLBFS_I(inode)->policy); kmem_cache_free(hugetlbfs_inode_cachep, HUGETLBFS_I(inode)); } @@ -565,6 +592,16 @@ static struct address_space_operations hugetlbfs_aops = { .set_page_dirty = hugetlbfs_set_page_dirty, }; + +static void init_once(void *foo, kmem_cache_t *cachep, unsigned long flags) +{ + struct hugetlbfs_inode_info *ei = (struct hugetlbfs_inode_info *)foo; + + if ((flags & (SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR)) == + SLAB_CTOR_CONSTRUCTOR) + inode_init_once(&ei->vfs_inode); +} + struct file_operations hugetlbfs_file_operations = { .mmap = hugetlbfs_file_mmap, .fsync = simple_sync_file, @@ -592,6 +629,7 @@ static struct super_operations hugetlbfs_ops = { .alloc_inode = hugetlbfs_alloc_inode, .destroy_inode = hugetlbfs_destroy_inode, .statfs = hugetlbfs_statfs, + .delete_inode = hugetlbfs_delete_inode, .drop_inode = hugetlbfs_drop_inode, .put_super = hugetlbfs_put_super, }; diff --git a/fs/inode.c b/fs/inode.c index f80a79ff156b..d8d04bd72b59 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -475,7 +475,7 @@ static void prune_icache(int nr_to_scan) * This function is passed the number of inodes to scan, and it returns the * total number of remaining possibly-reclaimable inodes. */ -static int shrink_icache_memory(int nr, unsigned int gfp_mask) +static int shrink_icache_memory(int nr, gfp_t gfp_mask) { if (nr) { /* @@ -1088,6 +1088,7 @@ static void generic_forget_inode(struct inode *inode) if (inode->i_data.nrpages) truncate_inode_pages(&inode->i_data, 0); clear_inode(inode); + wake_up_inode(inode); destroy_inode(inode); } diff --git a/fs/inotify.c b/fs/inotify.c index a37e9fb1da58..bf7ce1d2412b 100644 --- a/fs/inotify.c +++ b/fs/inotify.c @@ -176,6 +176,7 @@ static inline void put_inotify_dev(struct inotify_device *dev) if (atomic_dec_and_test(&dev->count)) { atomic_dec(&dev->user->inotify_devs); free_uid(dev->user); + idr_destroy(&dev->idr); kfree(dev); } } @@ -371,7 +372,7 @@ static int find_inode(const char __user *dirname, struct nameidata *nd) if (error) return error; /* you can only watch an inode if you have read permissions on it */ - error = permission(nd->dentry->d_inode, MAY_READ, NULL); + error = vfs_permission(nd, MAY_READ); if (error) path_release(nd); return error; diff --git a/fs/ioprio.c b/fs/ioprio.c index d1c1f2b2c9da..4bf1c6365a19 100644 --- a/fs/ioprio.c +++ b/fs/ioprio.c @@ -22,6 +22,7 @@ #include <linux/kernel.h> #include <linux/ioprio.h> #include <linux/blkdev.h> +#include <linux/syscalls.h> static int set_task_ioprio(struct task_struct *task, int ioprio) { diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c index 1652de1b6cb9..298f08be22d4 100644 --- a/fs/isofs/inode.c +++ b/fs/isofs/inode.c @@ -855,8 +855,7 @@ root_found: if (opt.check == 'r') table++; s->s_root->d_op = &isofs_dentry_ops[table]; - if (opt.iocharset) - kfree(opt.iocharset); + kfree(opt.iocharset); return 0; @@ -895,8 +894,7 @@ out_unknown_format: out_freebh: brelse(bh); out_freesbi: - if (opt.iocharset) - kfree(opt.iocharset); + kfree(opt.iocharset); kfree(sbi); s->s_fs_info = NULL; return -EINVAL; @@ -1164,8 +1162,7 @@ out_nomem: out_noread: printk(KERN_INFO "ISOFS: unable to read i-node block %lu\n", block); - if (tmpde) - kfree(tmpde); + kfree(tmpde); return -EIO; out_toomany: @@ -1334,8 +1331,7 @@ static void isofs_read_inode(struct inode *inode) init_special_inode(inode, inode->i_mode, inode->i_rdev); out: - if (tmpde) - kfree(tmpde); + kfree(tmpde); if (bh) brelse(bh); return; diff --git a/fs/jbd/commit.c b/fs/jbd/commit.c index 2a3e310f79ef..002ad2bbc769 100644 --- a/fs/jbd/commit.c +++ b/fs/jbd/commit.c @@ -261,10 +261,8 @@ void journal_commit_transaction(journal_t *journal) struct buffer_head *bh = jh2bh(jh); jbd_lock_bh_state(bh); - if (jh->b_committed_data) { - kfree(jh->b_committed_data); - jh->b_committed_data = NULL; - } + kfree(jh->b_committed_data); + jh->b_committed_data = NULL; jbd_unlock_bh_state(bh); } journal_refile_buffer(journal, jh); diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c index 7ae2c4fe506b..e4b516ac4989 100644 --- a/fs/jbd/journal.c +++ b/fs/jbd/journal.c @@ -1606,7 +1606,7 @@ int journal_blocks_per_page(struct inode *inode) * Simple support for retrying memory allocations. Introduced to help to * debug different VM deadlock avoidance strategies. */ -void * __jbd_kmalloc (const char *where, size_t size, int flags, int retry) +void * __jbd_kmalloc (const char *where, size_t size, gfp_t flags, int retry) { return kmalloc(size, flags | (retry ? __GFP_NOFAIL : 0)); } diff --git a/fs/jbd/recovery.c b/fs/jbd/recovery.c index 103c34e4fb28..80d7f53fd0a7 100644 --- a/fs/jbd/recovery.c +++ b/fs/jbd/recovery.c @@ -210,7 +210,7 @@ do { \ } while (0) /** - * int journal_recover(journal_t *journal) - recovers a on-disk journal + * journal_recover - recovers a on-disk journal * @journal: the journal to recover * * The primary function for recovering the log contents when mounting a @@ -266,7 +266,7 @@ int journal_recover(journal_t *journal) } /** - * int journal_skip_recovery() - Start journal and wipe exiting records + * journal_skip_recovery - Start journal and wipe exiting records * @journal: journal to startup * * Locate any valid recovery information from the journal and set up the diff --git a/fs/jbd/transaction.c b/fs/jbd/transaction.c index 49bbc2be3d72..429f4b263cf1 100644 --- a/fs/jbd/transaction.c +++ b/fs/jbd/transaction.c @@ -227,8 +227,7 @@ repeat_locked: spin_unlock(&transaction->t_handle_lock); spin_unlock(&journal->j_state_lock); out: - if (new_transaction) - kfree(new_transaction); + kfree(new_transaction); return ret; } @@ -725,8 +724,7 @@ done: journal_cancel_revoke(handle, jh); out: - if (frozen_buffer) - kfree(frozen_buffer); + kfree(frozen_buffer); JBUFFER_TRACE(jh, "exit"); return error; @@ -905,8 +903,7 @@ repeat: jbd_unlock_bh_state(bh); out: journal_put_journal_head(jh); - if (committed_data) - kfree(committed_data); + kfree(committed_data); return err; } @@ -1621,7 +1618,7 @@ out: * while the data is part of a transaction. Yes? */ int journal_try_to_free_buffers(journal_t *journal, - struct page *page, int unused_gfp_mask) + struct page *page, gfp_t unused_gfp_mask) { struct buffer_head *head; struct buffer_head *bh; diff --git a/fs/jffs/intrep.c b/fs/jffs/intrep.c index 27f199e94cfc..b2e95421d932 100644 --- a/fs/jffs/intrep.c +++ b/fs/jffs/intrep.c @@ -462,7 +462,7 @@ jffs_checksum_flash(struct mtd_info *mtd, loff_t start, int size, __u32 *result) } /* Free read buffer */ - kfree (read_buf); + kfree(read_buf); /* Return result */ D3(printk("checksum result: 0x%08x\n", sum)); @@ -1011,12 +1011,12 @@ jffs_scan_flash(struct jffs_control *c) offset , fmc->sector_size); flash_safe_release(fmc->mtd); - kfree (read_buf); + kfree(read_buf); return -1; /* bad, bad, bad! */ } flash_safe_release(fmc->mtd); - kfree (read_buf); + kfree(read_buf); return -EAGAIN; /* erased offending sector. Try mount one more time please. */ } @@ -1112,7 +1112,7 @@ jffs_scan_flash(struct jffs_control *c) if (!node) { if (!(node = jffs_alloc_node())) { /* Free read buffer */ - kfree (read_buf); + kfree(read_buf); /* Release the flash device */ flash_safe_release(fmc->mtd); @@ -1269,7 +1269,7 @@ jffs_scan_flash(struct jffs_control *c) DJM(no_jffs_node--); /* Free read buffer */ - kfree (read_buf); + kfree(read_buf); /* Release the flash device */ flash_safe_release(fmc->mtd); @@ -1296,7 +1296,7 @@ jffs_scan_flash(struct jffs_control *c) flash_safe_release(fmc->flash_part); /* Free read buffer */ - kfree (read_buf); + kfree(read_buf); return -ENOMEM; } @@ -1324,7 +1324,7 @@ jffs_scan_flash(struct jffs_control *c) jffs_build_end(fmc); /* Free read buffer */ - kfree (read_buf); + kfree(read_buf); if(!num_free_space){ printk(KERN_WARNING "jffs_scan_flash(): Did not find even a single " @@ -1747,9 +1747,7 @@ jffs_find_child(struct jffs_file *dir, const char *name, int len) } printk("jffs_find_child(): Didn't find the file \"%s\".\n", (copy ? copy : "")); - if (copy) { - kfree(copy); - } + kfree(copy); }); return f; diff --git a/fs/jffs/jffs_fm.c b/fs/jffs/jffs_fm.c index 053e3a98a276..6da13b309bd1 100644 --- a/fs/jffs/jffs_fm.c +++ b/fs/jffs/jffs_fm.c @@ -20,6 +20,7 @@ #include <linux/blkdev.h> #include <linux/jffs.h> #include "jffs_fm.h" +#include "intrep.h" #if defined(JFFS_MARK_OBSOLETE) && JFFS_MARK_OBSOLETE static int jffs_mark_obsolete(struct jffs_fmcontrol *fmc, __u32 fm_offset); diff --git a/fs/jffs2/Makefile b/fs/jffs2/Makefile index f1afe681ecd6..77dc5561a04e 100644 --- a/fs/jffs2/Makefile +++ b/fs/jffs2/Makefile @@ -1,7 +1,7 @@ # # Makefile for the Linux Journalling Flash File System v2 (JFFS2) # -# $Id: Makefile.common,v 1.9 2005/02/09 09:23:53 pavlov Exp $ +# $Id: Makefile.common,v 1.11 2005/09/07 08:34:53 havasi Exp $ # obj-$(CONFIG_JFFS2_FS) += jffs2.o @@ -9,9 +9,10 @@ obj-$(CONFIG_JFFS2_FS) += jffs2.o jffs2-y := compr.o dir.o file.o ioctl.o nodelist.o malloc.o jffs2-y += read.o nodemgmt.o readinode.o write.o scan.o gc.o jffs2-y += symlink.o build.o erase.o background.o fs.o writev.o -jffs2-y += super.o +jffs2-y += super.o debug.o jffs2-$(CONFIG_JFFS2_FS_WRITEBUFFER) += wbuf.o jffs2-$(CONFIG_JFFS2_RUBIN) += compr_rubin.o jffs2-$(CONFIG_JFFS2_RTIME) += compr_rtime.o jffs2-$(CONFIG_JFFS2_ZLIB) += compr_zlib.o +jffs2-$(CONFIG_JFFS2_SUMMARY) += summary.o diff --git a/fs/jffs2/TODO b/fs/jffs2/TODO index 2bff82fd221f..d0e23b26fa50 100644 --- a/fs/jffs2/TODO +++ b/fs/jffs2/TODO @@ -1,5 +1,11 @@ -$Id: TODO,v 1.10 2002/09/09 16:31:21 dwmw2 Exp $ +$Id: TODO,v 1.18 2005/09/22 11:24:56 dedekind Exp $ + - support asynchronous operation -- add a per-fs 'reserved_space' count, + let each outstanding write reserve the _maximum_ amount of physical + space it could take. Let GC flush the outstanding writes because the + reservations will necessarily be pessimistic. With this we could even + do shared writable mmap, if we can have a fs hook for do_wp_page() to + make the reservation. - disable compression in commit_write()? - fine-tune the allocation / GC thresholds - chattr support - turning on/off and tuning compression per-inode @@ -11,26 +17,15 @@ $Id: TODO,v 1.10 2002/09/09 16:31:21 dwmw2 Exp $ - test, test, test - NAND flash support: - - flush_wbuf using GC to fill it, don't just pad. - - Deal with write errors. Data don't get lost - we just have to write - the affected node(s) out again somewhere else. - - make fsync flush only if actually required - - make sys_sync() work. - - reboot notifier - - timed flush of old wbuf - - fix magical second arg of jffs2_flush_wbuf(). Split into two or more functions instead. - + - almost done :) + - use bad block check instead of the hardwired byte check - Optimisations: - - Stop GC from decompressing and immediately recompressing nodes which could - just be copied intact. (We now keep track of REF_PRISTINE flag. Easy now.) - - Furthermore, in the case where it could be copied intact we don't even need - to call iget() for it -- if we use (raw_node_raw->flash_offset & 2) as a flag - to show a node can be copied intact and it's _not_ in icache, we could just do - it, fix up the next_in_ino list and move on. We would need a way to find out - _whether_ it's in icache though -- if it's in icache we also need to do the - fragment lists, etc. P'raps a flag or pointer in the jffs2_inode_cache could - help. (We have half of this now.) + - Split writes so they go to two separate blocks rather than just c->nextblock. + By writing _new_ nodes to one block, and garbage-collected REF_PRISTINE + nodes to a different one, we can separate clean nodes from those which + are likely to become dirty, and end up with blocks which are each far + closer to 100% or 0% clean, hence speeding up later GC progress dramatically. - Stop keeping name in-core with struct jffs2_full_dirent. If we keep the hash in the full dirent, we only need to go to the flash in lookup() when we think we've got a match, and in readdir(). @@ -38,3 +33,8 @@ $Id: TODO,v 1.10 2002/09/09 16:31:21 dwmw2 Exp $ - Remove totlen from jffs2_raw_node_ref? Need to have totlen passed into jffs2_mark_node_obsolete(). Can all callers work it out? - Remove size from jffs2_raw_node_frag. + +dedekind: +1. __jffs2_flush_wbuf() has a strange 'pad' parameter. Eliminate. +2. get_sb()->build_fs()->scan() path... Why get_sb() removes scan()'s crap in + case of failure? scan() does not clean everything. Fix. diff --git a/fs/jffs2/background.c b/fs/jffs2/background.c index 0f224384f176..7b77a9541125 100644 --- a/fs/jffs2/background.c +++ b/fs/jffs2/background.c @@ -15,6 +15,7 @@ #include <linux/jffs2.h> #include <linux/mtd/mtd.h> #include <linux/completion.h> +#include <linux/sched.h> #include "nodelist.h" @@ -50,7 +51,7 @@ int jffs2_start_garbage_collect_thread(struct jffs2_sb_info *c) D1(printk(KERN_DEBUG "JFFS2: Garbage collect thread is pid %d\n", pid)); wait_for_completion(&c->gc_thread_start); } - + return ret; } @@ -100,7 +101,7 @@ static int jffs2_garbage_collect_thread(void *_c) cond_resched(); - /* Put_super will send a SIGKILL and then wait on the sem. + /* Put_super will send a SIGKILL and then wait on the sem. */ while (signal_pending(current)) { siginfo_t info; diff --git a/fs/jffs2/build.c b/fs/jffs2/build.c index 97dc39796e2c..fff108bb118b 100644 --- a/fs/jffs2/build.c +++ b/fs/jffs2/build.c @@ -7,7 +7,7 @@ * * For licensing information, see the file 'LICENCE' in this directory. * - * $Id: build.c,v 1.71 2005/07/12 16:37:08 dedekind Exp $ + * $Id: build.c,v 1.85 2005/11/07 11:14:38 gleixner Exp $ * */ @@ -18,7 +18,8 @@ #include <linux/mtd/mtd.h> #include "nodelist.h" -static void jffs2_build_remove_unlinked_inode(struct jffs2_sb_info *, struct jffs2_inode_cache *, struct jffs2_full_dirent **); +static void jffs2_build_remove_unlinked_inode(struct jffs2_sb_info *, + struct jffs2_inode_cache *, struct jffs2_full_dirent **); static inline struct jffs2_inode_cache * first_inode_chain(int *i, struct jffs2_sb_info *c) @@ -46,11 +47,12 @@ next_inode(int *i, struct jffs2_inode_cache *ic, struct jffs2_sb_info *c) ic = next_inode(&i, ic, (c))) -static inline void jffs2_build_inode_pass1(struct jffs2_sb_info *c, struct jffs2_inode_cache *ic) +static inline void jffs2_build_inode_pass1(struct jffs2_sb_info *c, + struct jffs2_inode_cache *ic) { struct jffs2_full_dirent *fd; - D1(printk(KERN_DEBUG "jffs2_build_inode building directory inode #%u\n", ic->ino)); + dbg_fsbuild("building directory inode #%u\n", ic->ino); /* For each child, increase nlink */ for(fd = ic->scan_dents; fd; fd = fd->next) { @@ -58,26 +60,23 @@ static inline void jffs2_build_inode_pass1(struct jffs2_sb_info *c, struct jffs2 if (!fd->ino) continue; - /* XXX: Can get high latency here with huge directories */ + /* we can get high latency here with huge directories */ child_ic = jffs2_get_ino_cache(c, fd->ino); if (!child_ic) { - printk(KERN_NOTICE "Eep. Child \"%s\" (ino #%u) of dir ino #%u doesn't exist!\n", + dbg_fsbuild("child \"%s\" (ino #%u) of dir ino #%u doesn't exist!\n", fd->name, fd->ino, ic->ino); jffs2_mark_node_obsolete(c, fd->raw); continue; } if (child_ic->nlink++ && fd->type == DT_DIR) { - printk(KERN_NOTICE "Child dir \"%s\" (ino #%u) of dir ino #%u appears to be a hard link\n", fd->name, fd->ino, ic->ino); - if (fd->ino == 1 && ic->ino == 1) { - printk(KERN_NOTICE "This is mostly harmless, and probably caused by creating a JFFS2 image\n"); - printk(KERN_NOTICE "using a buggy version of mkfs.jffs2. Use at least v1.17.\n"); - } - /* What do we do about it? */ + JFFS2_ERROR("child dir \"%s\" (ino #%u) of dir ino #%u appears to be a hard link\n", + fd->name, fd->ino, ic->ino); + /* TODO: What do we do about it? */ } - D1(printk(KERN_DEBUG "Increased nlink for child \"%s\" (ino #%u)\n", fd->name, fd->ino)); - /* Can't free them. We might need them in pass 2 */ + dbg_fsbuild("increased nlink for child \"%s\" (ino #%u)\n", fd->name, fd->ino); + /* Can't free scan_dents so far. We might need them in pass 2 */ } } @@ -94,6 +93,8 @@ static int jffs2_build_filesystem(struct jffs2_sb_info *c) struct jffs2_full_dirent *fd; struct jffs2_full_dirent *dead_fds = NULL; + dbg_fsbuild("build FS data structures\n"); + /* First, scan the medium and build all the inode caches with lists of physical nodes */ @@ -103,60 +104,54 @@ static int jffs2_build_filesystem(struct jffs2_sb_info *c) if (ret) goto exit; - D1(printk(KERN_DEBUG "Scanned flash completely\n")); - D2(jffs2_dump_block_lists(c)); + dbg_fsbuild("scanned flash completely\n"); + jffs2_dbg_dump_block_lists_nolock(c); + dbg_fsbuild("pass 1 starting\n"); c->flags |= JFFS2_SB_FLAG_BUILDING; /* Now scan the directory tree, increasing nlink according to every dirent found. */ for_each_inode(i, c, ic) { - D1(printk(KERN_DEBUG "Pass 1: ino #%u\n", ic->ino)); - - D1(BUG_ON(ic->ino > c->highest_ino)); - if (ic->scan_dents) { jffs2_build_inode_pass1(c, ic); cond_resched(); } } - D1(printk(KERN_DEBUG "Pass 1 complete\n")); + dbg_fsbuild("pass 1 complete\n"); /* Next, scan for inodes with nlink == 0 and remove them. If they were directories, then decrement the nlink of their children too, and repeat the scan. As that's going to be a fairly uncommon occurrence, it's not so evil to do it this way. Recursion bad. */ - D1(printk(KERN_DEBUG "Pass 2 starting\n")); + dbg_fsbuild("pass 2 starting\n"); for_each_inode(i, c, ic) { - D1(printk(KERN_DEBUG "Pass 2: ino #%u, nlink %d, ic %p, nodes %p\n", ic->ino, ic->nlink, ic, ic->nodes)); if (ic->nlink) continue; - + jffs2_build_remove_unlinked_inode(c, ic, &dead_fds); cond_resched(); - } + } - D1(printk(KERN_DEBUG "Pass 2a starting\n")); + dbg_fsbuild("pass 2a starting\n"); while (dead_fds) { fd = dead_fds; dead_fds = fd->next; ic = jffs2_get_ino_cache(c, fd->ino); - D1(printk(KERN_DEBUG "Removing dead_fd ino #%u (\"%s\"), ic at %p\n", fd->ino, fd->name, ic)); if (ic) jffs2_build_remove_unlinked_inode(c, ic, &dead_fds); jffs2_free_full_dirent(fd); } - D1(printk(KERN_DEBUG "Pass 2 complete\n")); - + dbg_fsbuild("pass 2a complete\n"); + dbg_fsbuild("freeing temporary data structures\n"); + /* Finally, we can scan again and free the dirent structs */ for_each_inode(i, c, ic) { - D1(printk(KERN_DEBUG "Pass 3: ino #%u, ic %p, nodes %p\n", ic->ino, ic, ic->nodes)); - while(ic->scan_dents) { fd = ic->scan_dents; ic->scan_dents = fd->next; @@ -166,9 +161,8 @@ static int jffs2_build_filesystem(struct jffs2_sb_info *c) cond_resched(); } c->flags &= ~JFFS2_SB_FLAG_BUILDING; - - D1(printk(KERN_DEBUG "Pass 3 complete\n")); - D2(jffs2_dump_block_lists(c)); + + dbg_fsbuild("FS build complete\n"); /* Rotate the lists by some number to ensure wear levelling */ jffs2_rotate_lists(c); @@ -189,24 +183,26 @@ exit: return ret; } -static void jffs2_build_remove_unlinked_inode(struct jffs2_sb_info *c, struct jffs2_inode_cache *ic, struct jffs2_full_dirent **dead_fds) +static void jffs2_build_remove_unlinked_inode(struct jffs2_sb_info *c, + struct jffs2_inode_cache *ic, + struct jffs2_full_dirent **dead_fds) { struct jffs2_raw_node_ref *raw; struct jffs2_full_dirent *fd; - D1(printk(KERN_DEBUG "JFFS2: Removing ino #%u with nlink == zero.\n", ic->ino)); - + dbg_fsbuild("removing ino #%u with nlink == zero.\n", ic->ino); + raw = ic->nodes; while (raw != (void *)ic) { struct jffs2_raw_node_ref *next = raw->next_in_ino; - D1(printk(KERN_DEBUG "obsoleting node at 0x%08x\n", ref_offset(raw))); + dbg_fsbuild("obsoleting node at 0x%08x\n", ref_offset(raw)); jffs2_mark_node_obsolete(c, raw); raw = next; } if (ic->scan_dents) { int whinged = 0; - D1(printk(KERN_DEBUG "Inode #%u was a directory which may have children...\n", ic->ino)); + dbg_fsbuild("inode #%u was a directory which may have children...\n", ic->ino); while(ic->scan_dents) { struct jffs2_inode_cache *child_ic; @@ -216,45 +212,43 @@ static void jffs2_build_remove_unlinked_inode(struct jffs2_sb_info *c, struct jf if (!fd->ino) { /* It's a deletion dirent. Ignore it */ - D1(printk(KERN_DEBUG "Child \"%s\" is a deletion dirent, skipping...\n", fd->name)); + dbg_fsbuild("child \"%s\" is a deletion dirent, skipping...\n", fd->name); jffs2_free_full_dirent(fd); continue; } - if (!whinged) { + if (!whinged) whinged = 1; - printk(KERN_NOTICE "Inode #%u was a directory with children - removing those too...\n", ic->ino); - } - D1(printk(KERN_DEBUG "Removing child \"%s\", ino #%u\n", - fd->name, fd->ino)); - + dbg_fsbuild("removing child \"%s\", ino #%u\n", fd->name, fd->ino); + child_ic = jffs2_get_ino_cache(c, fd->ino); if (!child_ic) { - printk(KERN_NOTICE "Cannot remove child \"%s\", ino #%u, because it doesn't exist\n", fd->name, fd->ino); + dbg_fsbuild("cannot remove child \"%s\", ino #%u, because it doesn't exist\n", + fd->name, fd->ino); jffs2_free_full_dirent(fd); continue; } - /* Reduce nlink of the child. If it's now zero, stick it on the + /* Reduce nlink of the child. If it's now zero, stick it on the dead_fds list to be cleaned up later. Else just free the fd */ child_ic->nlink--; - + if (!child_ic->nlink) { - D1(printk(KERN_DEBUG "Inode #%u (\"%s\") has now got zero nlink. Adding to dead_fds list.\n", - fd->ino, fd->name)); + dbg_fsbuild("inode #%u (\"%s\") has now got zero nlink, adding to dead_fds list.\n", + fd->ino, fd->name); fd->next = *dead_fds; *dead_fds = fd; } else { - D1(printk(KERN_DEBUG "Inode #%u (\"%s\") has now got nlink %d. Ignoring.\n", - fd->ino, fd->name, child_ic->nlink)); + dbg_fsbuild("inode #%u (\"%s\") has now got nlink %d. Ignoring.\n", + fd->ino, fd->name, child_ic->nlink); jffs2_free_full_dirent(fd); } } } /* - We don't delete the inocache from the hash list and free it yet. + We don't delete the inocache from the hash list and free it yet. The erase code will do that, when all the nodes are completely gone. */ } @@ -268,7 +262,7 @@ static void jffs2_calc_trigger_levels(struct jffs2_sb_info *c) because there's not enough free space... */ c->resv_blocks_deletion = 2; - /* Be conservative about how much space we need before we allow writes. + /* Be conservative about how much space we need before we allow writes. On top of that which is required for deletia, require an extra 2% of the medium to be available, for overhead caused by nodes being split across blocks, etc. */ @@ -283,7 +277,7 @@ static void jffs2_calc_trigger_levels(struct jffs2_sb_info *c) c->resv_blocks_gctrigger = c->resv_blocks_write + 1; - /* When do we allow garbage collection to merge nodes to make + /* When do we allow garbage collection to merge nodes to make long-term progress at the expense of short-term space exhaustion? */ c->resv_blocks_gcmerge = c->resv_blocks_deletion + 1; @@ -295,45 +289,45 @@ static void jffs2_calc_trigger_levels(struct jffs2_sb_info *c) trying to GC to make more space. It'll be a fruitless task */ c->nospc_dirty_size = c->sector_size + (c->flash_size / 100); - D1(printk(KERN_DEBUG "JFFS2 trigger levels (size %d KiB, block size %d KiB, %d blocks)\n", - c->flash_size / 1024, c->sector_size / 1024, c->nr_blocks)); - D1(printk(KERN_DEBUG "Blocks required to allow deletion: %d (%d KiB)\n", - c->resv_blocks_deletion, c->resv_blocks_deletion*c->sector_size/1024)); - D1(printk(KERN_DEBUG "Blocks required to allow writes: %d (%d KiB)\n", - c->resv_blocks_write, c->resv_blocks_write*c->sector_size/1024)); - D1(printk(KERN_DEBUG "Blocks required to quiesce GC thread: %d (%d KiB)\n", - c->resv_blocks_gctrigger, c->resv_blocks_gctrigger*c->sector_size/1024)); - D1(printk(KERN_DEBUG "Blocks required to allow GC merges: %d (%d KiB)\n", - c->resv_blocks_gcmerge, c->resv_blocks_gcmerge*c->sector_size/1024)); - D1(printk(KERN_DEBUG "Blocks required to GC bad blocks: %d (%d KiB)\n", - c->resv_blocks_gcbad, c->resv_blocks_gcbad*c->sector_size/1024)); - D1(printk(KERN_DEBUG "Amount of dirty space required to GC: %d bytes\n", - c->nospc_dirty_size)); -} + dbg_fsbuild("JFFS2 trigger levels (size %d KiB, block size %d KiB, %d blocks)\n", + c->flash_size / 1024, c->sector_size / 1024, c->nr_blocks); + dbg_fsbuild("Blocks required to allow deletion: %d (%d KiB)\n", + c->resv_blocks_deletion, c->resv_blocks_deletion*c->sector_size/1024); + dbg_fsbuild("Blocks required to allow writes: %d (%d KiB)\n", + c->resv_blocks_write, c->resv_blocks_write*c->sector_size/1024); + dbg_fsbuild("Blocks required to quiesce GC thread: %d (%d KiB)\n", + c->resv_blocks_gctrigger, c->resv_blocks_gctrigger*c->sector_size/1024); + dbg_fsbuild("Blocks required to allow GC merges: %d (%d KiB)\n", + c->resv_blocks_gcmerge, c->resv_blocks_gcmerge*c->sector_size/1024); + dbg_fsbuild("Blocks required to GC bad blocks: %d (%d KiB)\n", + c->resv_blocks_gcbad, c->resv_blocks_gcbad*c->sector_size/1024); + dbg_fsbuild("Amount of dirty space required to GC: %d bytes\n", + c->nospc_dirty_size); +} int jffs2_do_mount_fs(struct jffs2_sb_info *c) { + int ret; int i; + int size; c->free_size = c->flash_size; c->nr_blocks = c->flash_size / c->sector_size; - if (c->mtd->flags & MTD_NO_VIRTBLOCKS) - c->blocks = vmalloc(sizeof(struct jffs2_eraseblock) * c->nr_blocks); + size = sizeof(struct jffs2_eraseblock) * c->nr_blocks; +#ifndef __ECOS + if (jffs2_blocks_use_vmalloc(c)) + c->blocks = vmalloc(size); else - c->blocks = kmalloc(sizeof(struct jffs2_eraseblock) * c->nr_blocks, GFP_KERNEL); +#endif + c->blocks = kmalloc(size, GFP_KERNEL); if (!c->blocks) return -ENOMEM; + + memset(c->blocks, 0, size); for (i=0; i<c->nr_blocks; i++) { INIT_LIST_HEAD(&c->blocks[i].list); c->blocks[i].offset = i * c->sector_size; c->blocks[i].free_size = c->sector_size; - c->blocks[i].dirty_size = 0; - c->blocks[i].wasted_size = 0; - c->blocks[i].unchecked_size = 0; - c->blocks[i].used_size = 0; - c->blocks[i].first_node = NULL; - c->blocks[i].last_node = NULL; - c->blocks[i].bad_count = 0; } INIT_LIST_HEAD(&c->clean_list); @@ -348,16 +342,23 @@ int jffs2_do_mount_fs(struct jffs2_sb_info *c) INIT_LIST_HEAD(&c->bad_list); INIT_LIST_HEAD(&c->bad_used_list); c->highest_ino = 1; + c->summary = NULL; + + ret = jffs2_sum_init(c); + if (ret) + return ret; if (jffs2_build_filesystem(c)) { - D1(printk(KERN_DEBUG "build_fs failed\n")); + dbg_fsbuild("build_fs failed\n"); jffs2_free_ino_caches(c); jffs2_free_raw_node_refs(c); - if (c->mtd->flags & MTD_NO_VIRTBLOCKS) { +#ifndef __ECOS + if (jffs2_blocks_use_vmalloc(c)) vfree(c->blocks); - } else { + else +#endif kfree(c->blocks); - } + return -EIO; } diff --git a/fs/jffs2/compr.c b/fs/jffs2/compr.c index af922a9618ac..e7944e665b9f 100644 --- a/fs/jffs2/compr.c +++ b/fs/jffs2/compr.c @@ -9,7 +9,7 @@ * * For licensing information, see the file 'LICENCE' in this directory. * - * $Id: compr.c,v 1.42 2004/08/07 21:56:08 dwmw2 Exp $ + * $Id: compr.c,v 1.46 2005/11/07 11:14:38 gleixner Exp $ * */ @@ -36,16 +36,16 @@ static uint32_t none_stat_compr_blocks=0,none_stat_decompr_blocks=0,none_stat_co * data. * * Returns: Lower byte to be stored with data indicating compression type used. - * Zero is used to show that the data could not be compressed - the + * Zero is used to show that the data could not be compressed - the * compressed version was actually larger than the original. * Upper byte will be used later. (soon) * * If the cdata buffer isn't large enough to hold all the uncompressed data, - * jffs2_compress should compress as much as will fit, and should set + * jffs2_compress should compress as much as will fit, and should set * *datalen accordingly to show the amount of data which were compressed. */ uint16_t jffs2_compress(struct jffs2_sb_info *c, struct jffs2_inode_info *f, - unsigned char *data_in, unsigned char **cpage_out, + unsigned char *data_in, unsigned char **cpage_out, uint32_t *datalen, uint32_t *cdatalen) { int ret = JFFS2_COMPR_NONE; @@ -164,7 +164,7 @@ uint16_t jffs2_compress(struct jffs2_sb_info *c, struct jffs2_inode_info *f, } int jffs2_decompress(struct jffs2_sb_info *c, struct jffs2_inode_info *f, - uint16_t comprtype, unsigned char *cdata_in, + uint16_t comprtype, unsigned char *cdata_in, unsigned char *data_out, uint32_t cdatalen, uint32_t datalen) { struct jffs2_compressor *this; @@ -298,7 +298,7 @@ char *jffs2_stats(void) act_buf += sprintf(act_buf,"JFFS2 compressor statistics:\n"); act_buf += sprintf(act_buf,"%10s ","none"); - act_buf += sprintf(act_buf,"compr: %d blocks (%d) decompr: %d blocks\n", none_stat_compr_blocks, + act_buf += sprintf(act_buf,"compr: %d blocks (%d) decompr: %d blocks\n", none_stat_compr_blocks, none_stat_compr_size, none_stat_decompr_blocks); spin_lock(&jffs2_compressor_list_lock); list_for_each_entry(this, &jffs2_compressor_list, list) { @@ -307,8 +307,8 @@ char *jffs2_stats(void) act_buf += sprintf(act_buf,"- "); else act_buf += sprintf(act_buf,"+ "); - act_buf += sprintf(act_buf,"compr: %d blocks (%d/%d) decompr: %d blocks ", this->stat_compr_blocks, - this->stat_compr_new_size, this->stat_compr_orig_size, + act_buf += sprintf(act_buf,"compr: %d blocks (%d/%d) decompr: %d blocks ", this->stat_compr_blocks, + this->stat_compr_new_size, this->stat_compr_orig_size, this->stat_decompr_blocks); act_buf += sprintf(act_buf,"\n"); } @@ -317,7 +317,7 @@ char *jffs2_stats(void) return buf; } -char *jffs2_get_compression_mode_name(void) +char *jffs2_get_compression_mode_name(void) { switch (jffs2_compression_mode) { case JFFS2_COMPR_MODE_NONE: @@ -330,7 +330,7 @@ char *jffs2_get_compression_mode_name(void) return "unkown"; } -int jffs2_set_compression_mode_name(const char *name) +int jffs2_set_compression_mode_name(const char *name) { if (!strcmp("none",name)) { jffs2_compression_mode = JFFS2_COMPR_MODE_NONE; @@ -355,7 +355,7 @@ static int jffs2_compressor_Xable(const char *name, int disabled) if (!strcmp(this->name, name)) { this->disabled = disabled; spin_unlock(&jffs2_compressor_list_lock); - return 0; + return 0; } } spin_unlock(&jffs2_compressor_list_lock); @@ -385,7 +385,7 @@ int jffs2_set_compressor_priority(const char *name, int priority) } } spin_unlock(&jffs2_compressor_list_lock); - printk(KERN_WARNING "JFFS2: compressor %s not found.\n",name); + printk(KERN_WARNING "JFFS2: compressor %s not found.\n",name); return 1; reinsert: /* list is sorted in the order of priority, so if @@ -412,7 +412,7 @@ void jffs2_free_comprbuf(unsigned char *comprbuf, unsigned char *orig) kfree(comprbuf); } -int jffs2_compressors_init(void) +int jffs2_compressors_init(void) { /* Registering compressors */ #ifdef CONFIG_JFFS2_ZLIB @@ -425,12 +425,6 @@ int jffs2_compressors_init(void) jffs2_rubinmips_init(); jffs2_dynrubin_init(); #endif -#ifdef CONFIG_JFFS2_LZARI - jffs2_lzari_init(); -#endif -#ifdef CONFIG_JFFS2_LZO - jffs2_lzo_init(); -#endif /* Setting default compression mode */ #ifdef CONFIG_JFFS2_CMODE_NONE jffs2_compression_mode = JFFS2_COMPR_MODE_NONE; @@ -446,15 +440,9 @@ int jffs2_compressors_init(void) return 0; } -int jffs2_compressors_exit(void) +int jffs2_compressors_exit(void) { /* Unregistering compressors */ -#ifdef CONFIG_JFFS2_LZO - jffs2_lzo_exit(); -#endif -#ifdef CONFIG_JFFS2_LZARI - jffs2_lzari_exit(); -#endif #ifdef CONFIG_JFFS2_RUBIN jffs2_dynrubin_exit(); jffs2_rubinmips_exit(); diff --git a/fs/jffs2/compr.h b/fs/jffs2/compr.h index 89ceeed201eb..a77e830d85c5 100644 --- a/fs/jffs2/compr.h +++ b/fs/jffs2/compr.h @@ -4,10 +4,10 @@ * Copyright (C) 2004 Ferenc Havasi <havasi@inf.u-szeged.hu>, * University of Szeged, Hungary * - * For licensing information, see the file 'LICENCE' in the + * For licensing information, see the file 'LICENCE' in the * jffs2 directory. * - * $Id: compr.h,v 1.6 2004/07/16 15:17:57 dwmw2 Exp $ + * $Id: compr.h,v 1.9 2005/11/07 11:14:38 gleixner Exp $ * */ @@ -103,13 +103,5 @@ void jffs2_rtime_exit(void); int jffs2_zlib_init(void); void jffs2_zlib_exit(void); #endif -#ifdef CONFIG_JFFS2_LZARI -int jffs2_lzari_init(void); -void jffs2_lzari_exit(void); -#endif -#ifdef CONFIG_JFFS2_LZO -int jffs2_lzo_init(void); -void jffs2_lzo_exit(void); -#endif #endif /* __JFFS2_COMPR_H__ */ diff --git a/fs/jffs2/compr_rtime.c b/fs/jffs2/compr_rtime.c index 393129418666..2eb1b7428d16 100644 --- a/fs/jffs2/compr_rtime.c +++ b/fs/jffs2/compr_rtime.c @@ -24,8 +24,8 @@ #include <linux/kernel.h> #include <linux/types.h> #include <linux/errno.h> -#include <linux/string.h> -#include <linux/jffs2.h> +#include <linux/string.h> +#include <linux/jffs2.h> #include "compr.h" /* _compress returns the compressed size, -1 if bigger */ @@ -38,19 +38,19 @@ static int jffs2_rtime_compress(unsigned char *data_in, int outpos = 0; int pos=0; - memset(positions,0,sizeof(positions)); - + memset(positions,0,sizeof(positions)); + while (pos < (*sourcelen) && outpos <= (*dstlen)-2) { int backpos, runlen=0; unsigned char value; - + value = data_in[pos]; cpage_out[outpos++] = data_in[pos++]; - + backpos = positions[value]; positions[value]=pos; - + while ((backpos < pos) && (pos < (*sourcelen)) && (data_in[pos]==data_in[backpos++]) && (runlen<255)) { pos++; @@ -63,12 +63,12 @@ static int jffs2_rtime_compress(unsigned char *data_in, /* We failed */ return -1; } - + /* Tell the caller how much we managed to compress, and how much space it took */ *sourcelen = pos; *dstlen = outpos; return 0; -} +} static int jffs2_rtime_decompress(unsigned char *data_in, @@ -79,19 +79,19 @@ static int jffs2_rtime_decompress(unsigned char *data_in, short positions[256]; int outpos = 0; int pos=0; - - memset(positions,0,sizeof(positions)); - + + memset(positions,0,sizeof(positions)); + while (outpos<destlen) { unsigned char value; int backoffs; int repeat; - + value = data_in[pos++]; cpage_out[outpos++] = value; /* first the verbatim copied byte */ repeat = data_in[pos++]; backoffs = positions[value]; - + positions[value]=outpos; if (repeat) { if (backoffs + repeat >= outpos) { @@ -101,12 +101,12 @@ static int jffs2_rtime_decompress(unsigned char *data_in, } } else { memcpy(&cpage_out[outpos],&cpage_out[backoffs],repeat); - outpos+=repeat; + outpos+=repeat; } } } return 0; -} +} static struct jffs2_compressor jffs2_rtime_comp = { .priority = JFFS2_RTIME_PRIORITY, diff --git a/fs/jffs2/compr_rubin.c b/fs/jffs2/compr_rubin.c index 09422388fb96..e792e675d624 100644 --- a/fs/jffs2/compr_rubin.c +++ b/fs/jffs2/compr_rubin.c @@ -11,7 +11,6 @@ * */ - #include <linux/string.h> #include <linux/types.h> #include <linux/jffs2.h> @@ -20,7 +19,7 @@ #include "compr.h" static void init_rubin(struct rubin_state *rs, int div, int *bits) -{ +{ int c; rs->q = 0; @@ -40,7 +39,7 @@ static int encode(struct rubin_state *rs, long A, long B, int symbol) while ((rs->q >= UPPER_BIT_RUBIN) || ((rs->p + rs->q) <= UPPER_BIT_RUBIN)) { rs->bit_number++; - + ret = pushbit(&rs->pp, (rs->q & UPPER_BIT_RUBIN) ? 1 : 0, 0); if (ret) return ret; @@ -68,7 +67,7 @@ static int encode(struct rubin_state *rs, long A, long B, int symbol) static void end_rubin(struct rubin_state *rs) -{ +{ int i; @@ -82,7 +81,7 @@ static void end_rubin(struct rubin_state *rs) static void init_decode(struct rubin_state *rs, int div, int *bits) { - init_rubin(rs, div, bits); + init_rubin(rs, div, bits); /* behalve lower */ rs->rec_q = 0; @@ -188,7 +187,7 @@ static int in_byte(struct rubin_state *rs) -static int rubin_do_compress(int bit_divider, int *bits, unsigned char *data_in, +static int rubin_do_compress(int bit_divider, int *bits, unsigned char *data_in, unsigned char *cpage_out, uint32_t *sourcelen, uint32_t *dstlen) { int outpos = 0; @@ -198,31 +197,31 @@ static int rubin_do_compress(int bit_divider, int *bits, unsigned char *data_in, init_pushpull(&rs.pp, cpage_out, *dstlen * 8, 0, 32); init_rubin(&rs, bit_divider, bits); - + while (pos < (*sourcelen) && !out_byte(&rs, data_in[pos])) pos++; - + end_rubin(&rs); if (outpos > pos) { /* We failed */ return -1; } - - /* Tell the caller how much we managed to compress, + + /* Tell the caller how much we managed to compress, * and how much space it took */ - + outpos = (pushedbits(&rs.pp)+7)/8; - + if (outpos >= pos) return -1; /* We didn't actually compress */ *sourcelen = pos; *dstlen = outpos; return 0; -} +} #if 0 /* _compress returns the compressed size, -1 if bigger */ -int jffs2_rubinmips_compress(unsigned char *data_in, unsigned char *cpage_out, +int jffs2_rubinmips_compress(unsigned char *data_in, unsigned char *cpage_out, uint32_t *sourcelen, uint32_t *dstlen, void *model) { return rubin_do_compress(BIT_DIVIDER_MIPS, bits_mips, data_in, cpage_out, sourcelen, dstlen); @@ -277,7 +276,7 @@ static int jffs2_dynrubin_compress(unsigned char *data_in, } ret = rubin_do_compress(256, bits, data_in, cpage_out+8, &mysrclen, &mydstlen); - if (ret) + if (ret) return ret; /* Add back the 8 bytes we took for the probabilities */ @@ -293,19 +292,19 @@ static int jffs2_dynrubin_compress(unsigned char *data_in, return 0; } -static void rubin_do_decompress(int bit_divider, int *bits, unsigned char *cdata_in, +static void rubin_do_decompress(int bit_divider, int *bits, unsigned char *cdata_in, unsigned char *page_out, uint32_t srclen, uint32_t destlen) { int outpos = 0; struct rubin_state rs; - + init_pushpull(&rs.pp, cdata_in, srclen, 0, 0); init_decode(&rs, bit_divider, bits); - + while (outpos < destlen) { page_out[outpos++] = in_byte(&rs); } -} +} static int jffs2_rubinmips_decompress(unsigned char *data_in, diff --git a/fs/jffs2/compr_rubin.h b/fs/jffs2/compr_rubin.h index cf51e34f6574..bf1a93451621 100644 --- a/fs/jffs2/compr_rubin.h +++ b/fs/jffs2/compr_rubin.h @@ -1,7 +1,7 @@ /* Rubin encoder/decoder header */ /* work started at : aug 3, 1994 */ /* last modification : aug 15, 1994 */ -/* $Id: compr_rubin.h,v 1.6 2002/01/25 01:49:26 dwmw2 Exp $ */ +/* $Id: compr_rubin.h,v 1.7 2005/11/07 11:14:38 gleixner Exp $ */ #include "pushpull.h" @@ -11,8 +11,8 @@ struct rubin_state { - unsigned long p; - unsigned long q; + unsigned long p; + unsigned long q; unsigned long rec_q; long bit_number; struct pushpull pp; diff --git a/fs/jffs2/compr_zlib.c b/fs/jffs2/compr_zlib.c index 83f7e0788fd0..4db8be8e90cc 100644 --- a/fs/jffs2/compr_zlib.c +++ b/fs/jffs2/compr_zlib.c @@ -7,7 +7,7 @@ * * For licensing information, see the file 'LICENCE' in this directory. * - * $Id: compr_zlib.c,v 1.31 2005/05/20 19:30:06 gleixner Exp $ + * $Id: compr_zlib.c,v 1.32 2005/11/07 11:14:38 gleixner Exp $ * */ @@ -24,11 +24,11 @@ #include "nodelist.h" #include "compr.h" - /* Plan: call deflate() with avail_in == *sourcelen, - avail_out = *dstlen - 12 and flush == Z_FINISH. + /* Plan: call deflate() with avail_in == *sourcelen, + avail_out = *dstlen - 12 and flush == Z_FINISH. If it doesn't manage to finish, call it again with avail_in == 0 and avail_out set to the remaining 12 - bytes for it to clean up. + bytes for it to clean up. Q: Is 12 bytes sufficient? */ #define STREAM_END_SPACE 12 @@ -89,7 +89,7 @@ static int jffs2_zlib_compress(unsigned char *data_in, def_strm.next_in = data_in; def_strm.total_in = 0; - + def_strm.next_out = cpage_out; def_strm.total_out = 0; @@ -99,7 +99,7 @@ static int jffs2_zlib_compress(unsigned char *data_in, D1(printk(KERN_DEBUG "calling deflate with avail_in %d, avail_out %d\n", def_strm.avail_in, def_strm.avail_out)); ret = zlib_deflate(&def_strm, Z_PARTIAL_FLUSH); - D1(printk(KERN_DEBUG "deflate returned with avail_in %d, avail_out %d, total_in %ld, total_out %ld\n", + D1(printk(KERN_DEBUG "deflate returned with avail_in %d, avail_out %d, total_in %ld, total_out %ld\n", def_strm.avail_in, def_strm.avail_out, def_strm.total_in, def_strm.total_out)); if (ret != Z_OK) { D1(printk(KERN_DEBUG "deflate in loop returned %d\n", ret)); @@ -150,7 +150,7 @@ static int jffs2_zlib_decompress(unsigned char *data_in, inf_strm.next_in = data_in; inf_strm.avail_in = srclen; inf_strm.total_in = 0; - + inf_strm.next_out = cpage_out; inf_strm.avail_out = destlen; inf_strm.total_out = 0; diff --git a/fs/jffs2/comprtest.c b/fs/jffs2/comprtest.c index cf51f091d0e7..f0fb8be7740c 100644 --- a/fs/jffs2/comprtest.c +++ b/fs/jffs2/comprtest.c @@ -1,4 +1,4 @@ -/* $Id: comprtest.c,v 1.5 2002/01/03 15:20:44 dwmw2 Exp $ */ +/* $Id: comprtest.c,v 1.6 2005/11/07 11:14:38 gleixner Exp $ */ #include <linux/kernel.h> #include <linux/string.h> @@ -265,9 +265,9 @@ static unsigned char testdata[TESTDATA_LEN] = { static unsigned char comprbuf[TESTDATA_LEN]; static unsigned char decomprbuf[TESTDATA_LEN]; -int jffs2_decompress(unsigned char comprtype, unsigned char *cdata_in, +int jffs2_decompress(unsigned char comprtype, unsigned char *cdata_in, unsigned char *data_out, uint32_t cdatalen, uint32_t datalen); -unsigned char jffs2_compress(unsigned char *data_in, unsigned char *cpage_out, +unsigned char jffs2_compress(unsigned char *data_in, unsigned char *cpage_out, uint32_t *datalen, uint32_t *cdatalen); int init_module(void ) { @@ -276,10 +276,10 @@ int init_module(void ) { int ret; printk("Original data: %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x\n", - testdata[0],testdata[1],testdata[2],testdata[3], - testdata[4],testdata[5],testdata[6],testdata[7], - testdata[8],testdata[9],testdata[10],testdata[11], - testdata[12],testdata[13],testdata[14],testdata[15]); + testdata[0],testdata[1],testdata[2],testdata[3], + testdata[4],testdata[5],testdata[6],testdata[7], + testdata[8],testdata[9],testdata[10],testdata[11], + testdata[12],testdata[13],testdata[14],testdata[15]); d = TESTDATA_LEN; c = TESTDATA_LEN; comprtype = jffs2_compress(testdata, comprbuf, &d, &c); @@ -287,18 +287,18 @@ int init_module(void ) { printk("jffs2_compress used compression type %d. Compressed size %d, uncompressed size %d\n", comprtype, c, d); printk("Compressed data: %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x\n", - comprbuf[0],comprbuf[1],comprbuf[2],comprbuf[3], - comprbuf[4],comprbuf[5],comprbuf[6],comprbuf[7], - comprbuf[8],comprbuf[9],comprbuf[10],comprbuf[11], - comprbuf[12],comprbuf[13],comprbuf[14],comprbuf[15]); + comprbuf[0],comprbuf[1],comprbuf[2],comprbuf[3], + comprbuf[4],comprbuf[5],comprbuf[6],comprbuf[7], + comprbuf[8],comprbuf[9],comprbuf[10],comprbuf[11], + comprbuf[12],comprbuf[13],comprbuf[14],comprbuf[15]); ret = jffs2_decompress(comprtype, comprbuf, decomprbuf, c, d); printk("jffs2_decompress returned %d\n", ret); printk("Decompressed data: %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x\n", - decomprbuf[0],decomprbuf[1],decomprbuf[2],decomprbuf[3], - decomprbuf[4],decomprbuf[5],decomprbuf[6],decomprbuf[7], - decomprbuf[8],decomprbuf[9],decomprbuf[10],decomprbuf[11], - decomprbuf[12],decomprbuf[13],decomprbuf[14],decomprbuf[15]); + decomprbuf[0],decomprbuf[1],decomprbuf[2],decomprbuf[3], + decomprbuf[4],decomprbuf[5],decomprbuf[6],decomprbuf[7], + decomprbuf[8],decomprbuf[9],decomprbuf[10],decomprbuf[11], + decomprbuf[12],decomprbuf[13],decomprbuf[14],decomprbuf[15]); if (memcmp(decomprbuf, testdata, d)) printk("Compression and decompression corrupted data\n"); else diff --git a/fs/jffs2/debug.c b/fs/jffs2/debug.c new file mode 100644 index 000000000000..1fe17de713e8 --- /dev/null +++ b/fs/jffs2/debug.c @@ -0,0 +1,705 @@ +/* + * JFFS2 -- Journalling Flash File System, Version 2. + * + * Copyright (C) 2001-2003 Red Hat, Inc. + * + * Created by David Woodhouse <dwmw2@infradead.org> + * + * For licensing information, see the file 'LICENCE' in this directory. + * + * $Id: debug.c,v 1.12 2005/11/07 11:14:39 gleixner Exp $ + * + */ +#include <linux/kernel.h> +#include <linux/types.h> +#include <linux/pagemap.h> +#include <linux/crc32.h> +#include <linux/jffs2.h> +#include <linux/mtd/mtd.h> +#include "nodelist.h" +#include "debug.h" + +#ifdef JFFS2_DBG_SANITY_CHECKS + +void +__jffs2_dbg_acct_sanity_check_nolock(struct jffs2_sb_info *c, + struct jffs2_eraseblock *jeb) +{ + if (unlikely(jeb && jeb->used_size + jeb->dirty_size + + jeb->free_size + jeb->wasted_size + + jeb->unchecked_size != c->sector_size)) { + JFFS2_ERROR("eeep, space accounting for block at 0x%08x is screwed.\n", jeb->offset); + JFFS2_ERROR("free %#08x + dirty %#08x + used %#08x + wasted %#08x + unchecked %#08x != total %#08x.\n", + jeb->free_size, jeb->dirty_size, jeb->used_size, + jeb->wasted_size, jeb->unchecked_size, c->sector_size); + BUG(); + } + + if (unlikely(c->used_size + c->dirty_size + c->free_size + c->erasing_size + c->bad_size + + c->wasted_size + c->unchecked_size != c->flash_size)) { + JFFS2_ERROR("eeep, space accounting superblock info is screwed.\n"); + JFFS2_ERROR("free %#08x + dirty %#08x + used %#08x + erasing %#08x + bad %#08x + wasted %#08x + unchecked %#08x != total %#08x.\n", + c->free_size, c->dirty_size, c->used_size, c->erasing_size, c->bad_size, + c->wasted_size, c->unchecked_size, c->flash_size); + BUG(); + } +} + +void +__jffs2_dbg_acct_sanity_check(struct jffs2_sb_info *c, + struct jffs2_eraseblock *jeb) +{ + spin_lock(&c->erase_completion_lock); + jffs2_dbg_acct_sanity_check_nolock(c, jeb); + spin_unlock(&c->erase_completion_lock); +} + +#endif /* JFFS2_DBG_SANITY_CHECKS */ + +#ifdef JFFS2_DBG_PARANOIA_CHECKS +/* + * Check the fragtree. + */ +void +__jffs2_dbg_fragtree_paranoia_check(struct jffs2_inode_info *f) +{ + down(&f->sem); + __jffs2_dbg_fragtree_paranoia_check_nolock(f); + up(&f->sem); +} + +void +__jffs2_dbg_fragtree_paranoia_check_nolock(struct jffs2_inode_info *f) +{ + struct jffs2_node_frag *frag; + int bitched = 0; + + for (frag = frag_first(&f->fragtree); frag; frag = frag_next(frag)) { + struct jffs2_full_dnode *fn = frag->node; + + if (!fn || !fn->raw) + continue; + + if (ref_flags(fn->raw) == REF_PRISTINE) { + if (fn->frags > 1) { + JFFS2_ERROR("REF_PRISTINE node at 0x%08x had %d frags. Tell dwmw2.\n", + ref_offset(fn->raw), fn->frags); + bitched = 1; + } + + /* A hole node which isn't multi-page should be garbage-collected + and merged anyway, so we just check for the frag size here, + rather than mucking around with actually reading the node + and checking the compression type, which is the real way + to tell a hole node. */ + if (frag->ofs & (PAGE_CACHE_SIZE-1) && frag_prev(frag) + && frag_prev(frag)->size < PAGE_CACHE_SIZE && frag_prev(frag)->node) { + JFFS2_ERROR("REF_PRISTINE node at 0x%08x had a previous non-hole frag in the same page. Tell dwmw2.\n", + ref_offset(fn->raw)); + bitched = 1; + } + + if ((frag->ofs+frag->size) & (PAGE_CACHE_SIZE-1) && frag_next(frag) + && frag_next(frag)->size < PAGE_CACHE_SIZE && frag_next(frag)->node) { + JFFS2_ERROR("REF_PRISTINE node at 0x%08x (%08x-%08x) had a following non-hole frag in the same page. Tell dwmw2.\n", + ref_offset(fn->raw), frag->ofs, frag->ofs+frag->size); + bitched = 1; + } + } + } + + if (bitched) { + JFFS2_ERROR("fragtree is corrupted.\n"); + __jffs2_dbg_dump_fragtree_nolock(f); + BUG(); + } +} + +/* + * Check if the flash contains all 0xFF before we start writing. + */ +void +__jffs2_dbg_prewrite_paranoia_check(struct jffs2_sb_info *c, + uint32_t ofs, int len) +{ + size_t retlen; + int ret, i; + unsigned char *buf; + + buf = kmalloc(len, GFP_KERNEL); + if (!buf) + return; + + ret = jffs2_flash_read(c, ofs, len, &retlen, buf); + if (ret || (retlen != len)) { + JFFS2_WARNING("read %d bytes failed or short. ret %d, retlen %zd.\n", + len, ret, retlen); + kfree(buf); + return; + } + + ret = 0; + for (i = 0; i < len; i++) + if (buf[i] != 0xff) + ret = 1; + + if (ret) { + JFFS2_ERROR("argh, about to write node to %#08x on flash, but there are data already there. The first corrupted byte is at %#08x offset.\n", + ofs, ofs + i); + __jffs2_dbg_dump_buffer(buf, len, ofs); + kfree(buf); + BUG(); + } + + kfree(buf); +} + +/* + * Check the space accounting and node_ref list correctness for the JFFS2 erasable block 'jeb'. + */ +void +__jffs2_dbg_acct_paranoia_check(struct jffs2_sb_info *c, + struct jffs2_eraseblock *jeb) +{ + spin_lock(&c->erase_completion_lock); + __jffs2_dbg_acct_paranoia_check_nolock(c, jeb); + spin_unlock(&c->erase_completion_lock); +} + +void +__jffs2_dbg_acct_paranoia_check_nolock(struct jffs2_sb_info *c, + struct jffs2_eraseblock *jeb) +{ + uint32_t my_used_size = 0; + uint32_t my_unchecked_size = 0; + uint32_t my_dirty_size = 0; + struct jffs2_raw_node_ref *ref2 = jeb->first_node; + + while (ref2) { + uint32_t totlen = ref_totlen(c, jeb, ref2); + + if (ref2->flash_offset < jeb->offset || + ref2->flash_offset > jeb->offset + c->sector_size) { + JFFS2_ERROR("node_ref %#08x shouldn't be in block at %#08x.\n", + ref_offset(ref2), jeb->offset); + goto error; + + } + if (ref_flags(ref2) == REF_UNCHECKED) + my_unchecked_size += totlen; + else if (!ref_obsolete(ref2)) + my_used_size += totlen; + else + my_dirty_size += totlen; + + if ((!ref2->next_phys) != (ref2 == jeb->last_node)) { + JFFS2_ERROR("node_ref for node at %#08x (mem %p) has next_phys at %#08x (mem %p), last_node is at %#08x (mem %p).\n", + ref_offset(ref2), ref2, ref_offset(ref2->next_phys), ref2->next_phys, + ref_offset(jeb->last_node), jeb->last_node); + goto error; + } + ref2 = ref2->next_phys; + } + + if (my_used_size != jeb->used_size) { + JFFS2_ERROR("Calculated used size %#08x != stored used size %#08x.\n", + my_used_size, jeb->used_size); + goto error; + } + + if (my_unchecked_size != jeb->unchecked_size) { + JFFS2_ERROR("Calculated unchecked size %#08x != stored unchecked size %#08x.\n", + my_unchecked_size, jeb->unchecked_size); + goto error; + } + +#if 0 + /* This should work when we implement ref->__totlen elemination */ + if (my_dirty_size != jeb->dirty_size + jeb->wasted_size) { + JFFS2_ERROR("Calculated dirty+wasted size %#08x != stored dirty + wasted size %#08x\n", + my_dirty_size, jeb->dirty_size + jeb->wasted_size); + goto error; + } + + if (jeb->free_size == 0 + && my_used_size + my_unchecked_size + my_dirty_size != c->sector_size) { + JFFS2_ERROR("The sum of all nodes in block (%#x) != size of block (%#x)\n", + my_used_size + my_unchecked_size + my_dirty_size, + c->sector_size); + goto error; + } +#endif + + return; + +error: + __jffs2_dbg_dump_node_refs_nolock(c, jeb); + __jffs2_dbg_dump_jeb_nolock(jeb); + __jffs2_dbg_dump_block_lists_nolock(c); + BUG(); + +} +#endif /* JFFS2_DBG_PARANOIA_CHECKS */ + +#if defined(JFFS2_DBG_DUMPS) || defined(JFFS2_DBG_PARANOIA_CHECKS) +/* + * Dump the node_refs of the 'jeb' JFFS2 eraseblock. + */ +void +__jffs2_dbg_dump_node_refs(struct jffs2_sb_info *c, + struct jffs2_eraseblock *jeb) +{ + spin_lock(&c->erase_completion_lock); + __jffs2_dbg_dump_node_refs_nolock(c, jeb); + spin_unlock(&c->erase_completion_lock); +} + +void +__jffs2_dbg_dump_node_refs_nolock(struct jffs2_sb_info *c, + struct jffs2_eraseblock *jeb) +{ + struct jffs2_raw_node_ref *ref; + int i = 0; + + printk(JFFS2_DBG_MSG_PREFIX " Dump node_refs of the eraseblock %#08x\n", jeb->offset); + if (!jeb->first_node) { + printk(JFFS2_DBG_MSG_PREFIX " no nodes in the eraseblock %#08x\n", jeb->offset); + return; + } + + printk(JFFS2_DBG); + for (ref = jeb->first_node; ; ref = ref->next_phys) { + printk("%#08x(%#x)", ref_offset(ref), ref->__totlen); + if (ref->next_phys) + printk("->"); + else + break; + if (++i == 4) { + i = 0; + printk("\n" JFFS2_DBG); + } + } + printk("\n"); +} + +/* + * Dump an eraseblock's space accounting. + */ +void +__jffs2_dbg_dump_jeb(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb) +{ + spin_lock(&c->erase_completion_lock); + __jffs2_dbg_dump_jeb_nolock(jeb); + spin_unlock(&c->erase_completion_lock); +} + +void +__jffs2_dbg_dump_jeb_nolock(struct jffs2_eraseblock *jeb) +{ + if (!jeb) + return; + + printk(JFFS2_DBG_MSG_PREFIX " dump space accounting for the eraseblock at %#08x:\n", + jeb->offset); + + printk(JFFS2_DBG "used_size: %#08x\n", jeb->used_size); + printk(JFFS2_DBG "dirty_size: %#08x\n", jeb->dirty_size); + printk(JFFS2_DBG "wasted_size: %#08x\n", jeb->wasted_size); + printk(JFFS2_DBG "unchecked_size: %#08x\n", jeb->unchecked_size); + printk(JFFS2_DBG "free_size: %#08x\n", jeb->free_size); +} + +void +__jffs2_dbg_dump_block_lists(struct jffs2_sb_info *c) +{ + spin_lock(&c->erase_completion_lock); + __jffs2_dbg_dump_block_lists_nolock(c); + spin_unlock(&c->erase_completion_lock); +} + +void +__jffs2_dbg_dump_block_lists_nolock(struct jffs2_sb_info *c) +{ + printk(JFFS2_DBG_MSG_PREFIX " dump JFFS2 blocks lists:\n"); + + printk(JFFS2_DBG "flash_size: %#08x\n", c->flash_size); + printk(JFFS2_DBG "used_size: %#08x\n", c->used_size); + printk(JFFS2_DBG "dirty_size: %#08x\n", c->dirty_size); + printk(JFFS2_DBG "wasted_size: %#08x\n", c->wasted_size); + printk(JFFS2_DBG "unchecked_size: %#08x\n", c->unchecked_size); + printk(JFFS2_DBG "free_size: %#08x\n", c->free_size); + printk(JFFS2_DBG "erasing_size: %#08x\n", c->erasing_size); + printk(JFFS2_DBG "bad_size: %#08x\n", c->bad_size); + printk(JFFS2_DBG "sector_size: %#08x\n", c->sector_size); + printk(JFFS2_DBG "jffs2_reserved_blocks size: %#08x\n", + c->sector_size * c->resv_blocks_write); + + if (c->nextblock) + printk(JFFS2_DBG "nextblock: %#08x (used %#08x, dirty %#08x, wasted %#08x, unchecked %#08x, free %#08x)\n", + c->nextblock->offset, c->nextblock->used_size, + c->nextblock->dirty_size, c->nextblock->wasted_size, + c->nextblock->unchecked_size, c->nextblock->free_size); + else + printk(JFFS2_DBG "nextblock: NULL\n"); + + if (c->gcblock) + printk(JFFS2_DBG "gcblock: %#08x (used %#08x, dirty %#08x, wasted %#08x, unchecked %#08x, free %#08x)\n", + c->gcblock->offset, c->gcblock->used_size, c->gcblock->dirty_size, + c->gcblock->wasted_size, c->gcblock->unchecked_size, c->gcblock->free_size); + else + printk(JFFS2_DBG "gcblock: NULL\n"); + + if (list_empty(&c->clean_list)) { + printk(JFFS2_DBG "clean_list: empty\n"); + } else { + struct list_head *this; + int numblocks = 0; + uint32_t dirty = 0; + + list_for_each(this, &c->clean_list) { + struct jffs2_eraseblock *jeb = list_entry(this, struct jffs2_eraseblock, list); + numblocks ++; + dirty += jeb->wasted_size; + if (!(jeb->used_size == 0 && jeb->dirty_size == 0 && jeb->wasted_size == 0)) { + printk(JFFS2_DBG "clean_list: %#08x (used %#08x, dirty %#08x, wasted %#08x, unchecked %#08x, free %#08x)\n", + jeb->offset, jeb->used_size, jeb->dirty_size, jeb->wasted_size, + jeb->unchecked_size, jeb->free_size); + } + } + + printk (JFFS2_DBG "Contains %d blocks with total wasted size %u, average wasted size: %u\n", + numblocks, dirty, dirty / numblocks); + } + + if (list_empty(&c->very_dirty_list)) { + printk(JFFS2_DBG "very_dirty_list: empty\n"); + } else { + struct list_head *this; + int numblocks = 0; + uint32_t dirty = 0; + + list_for_each(this, &c->very_dirty_list) { + struct jffs2_eraseblock *jeb = list_entry(this, struct jffs2_eraseblock, list); + + numblocks ++; + dirty += jeb->dirty_size; + if (!(jeb->used_size == 0 && jeb->dirty_size == 0 && jeb->wasted_size == 0)) { + printk(JFFS2_DBG "very_dirty_list: %#08x (used %#08x, dirty %#08x, wasted %#08x, unchecked %#08x, free %#08x)\n", + jeb->offset, jeb->used_size, jeb->dirty_size, jeb->wasted_size, + jeb->unchecked_size, jeb->free_size); + } + } + + printk (JFFS2_DBG "Contains %d blocks with total dirty size %u, average dirty size: %u\n", + numblocks, dirty, dirty / numblocks); + } + + if (list_empty(&c->dirty_list)) { + printk(JFFS2_DBG "dirty_list: empty\n"); + } else { + struct list_head *this; + int numblocks = 0; + uint32_t dirty = 0; + + list_for_each(this, &c->dirty_list) { + struct jffs2_eraseblock *jeb = list_entry(this, struct jffs2_eraseblock, list); + + numblocks ++; + dirty += jeb->dirty_size; + if (!(jeb->used_size == 0 && jeb->dirty_size == 0 && jeb->wasted_size == 0)) { + printk(JFFS2_DBG "dirty_list: %#08x (used %#08x, dirty %#08x, wasted %#08x, unchecked %#08x, free %#08x)\n", + jeb->offset, jeb->used_size, jeb->dirty_size, jeb->wasted_size, + jeb->unchecked_size, jeb->free_size); + } + } + + printk (JFFS2_DBG "contains %d blocks with total dirty size %u, average dirty size: %u\n", + numblocks, dirty, dirty / numblocks); + } + + if (list_empty(&c->erasable_list)) { + printk(JFFS2_DBG "erasable_list: empty\n"); + } else { + struct list_head *this; + + list_for_each(this, &c->erasable_list) { + struct jffs2_eraseblock *jeb = list_entry(this, struct jffs2_eraseblock, list); + + if (!(jeb->used_size == 0 && jeb->dirty_size == 0 && jeb->wasted_size == 0)) { + printk(JFFS2_DBG "erasable_list: %#08x (used %#08x, dirty %#08x, wasted %#08x, unchecked %#08x, free %#08x)\n", + jeb->offset, jeb->used_size, jeb->dirty_size, jeb->wasted_size, + jeb->unchecked_size, jeb->free_size); + } + } + } + + if (list_empty(&c->erasing_list)) { + printk(JFFS2_DBG "erasing_list: empty\n"); + } else { + struct list_head *this; + + list_for_each(this, &c->erasing_list) { + struct jffs2_eraseblock *jeb = list_entry(this, struct jffs2_eraseblock, list); + + if (!(jeb->used_size == 0 && jeb->dirty_size == 0 && jeb->wasted_size == 0)) { + printk(JFFS2_DBG "erasing_list: %#08x (used %#08x, dirty %#08x, wasted %#08x, unchecked %#08x, free %#08x)\n", + jeb->offset, jeb->used_size, jeb->dirty_size, jeb->wasted_size, + jeb->unchecked_size, jeb->free_size); + } + } + } + + if (list_empty(&c->erase_pending_list)) { + printk(JFFS2_DBG "erase_pending_list: empty\n"); + } else { + struct list_head *this; + + list_for_each(this, &c->erase_pending_list) { + struct jffs2_eraseblock *jeb = list_entry(this, struct jffs2_eraseblock, list); + + if (!(jeb->used_size == 0 && jeb->dirty_size == 0 && jeb->wasted_size == 0)) { + printk(JFFS2_DBG "erase_pending_list: %#08x (used %#08x, dirty %#08x, wasted %#08x, unchecked %#08x, free %#08x)\n", + jeb->offset, jeb->used_size, jeb->dirty_size, jeb->wasted_size, + jeb->unchecked_size, jeb->free_size); + } + } + } + + if (list_empty(&c->erasable_pending_wbuf_list)) { + printk(JFFS2_DBG "erasable_pending_wbuf_list: empty\n"); + } else { + struct list_head *this; + + list_for_each(this, &c->erasable_pending_wbuf_list) { + struct jffs2_eraseblock *jeb = list_entry(this, struct jffs2_eraseblock, list); + + if (!(jeb->used_size == 0 && jeb->dirty_size == 0 && jeb->wasted_size == 0)) { + printk(JFFS2_DBG "erasable_pending_wbuf_list: %#08x (used %#08x, dirty %#08x, wasted %#08x, unchecked %#08x, free %#08x)\n", + jeb->offset, jeb->used_size, jeb->dirty_size, jeb->wasted_size, + jeb->unchecked_size, jeb->free_size); + } + } + } + + if (list_empty(&c->free_list)) { + printk(JFFS2_DBG "free_list: empty\n"); + } else { + struct list_head *this; + + list_for_each(this, &c->free_list) { + struct jffs2_eraseblock *jeb = list_entry(this, struct jffs2_eraseblock, list); + + if (!(jeb->used_size == 0 && jeb->dirty_size == 0 && jeb->wasted_size == 0)) { + printk(JFFS2_DBG "free_list: %#08x (used %#08x, dirty %#08x, wasted %#08x, unchecked %#08x, free %#08x)\n", + jeb->offset, jeb->used_size, jeb->dirty_size, jeb->wasted_size, + jeb->unchecked_size, jeb->free_size); + } + } + } + + if (list_empty(&c->bad_list)) { + printk(JFFS2_DBG "bad_list: empty\n"); + } else { + struct list_head *this; + + list_for_each(this, &c->bad_list) { + struct jffs2_eraseblock *jeb = list_entry(this, struct jffs2_eraseblock, list); + + if (!(jeb->used_size == 0 && jeb->dirty_size == 0 && jeb->wasted_size == 0)) { + printk(JFFS2_DBG "bad_list: %#08x (used %#08x, dirty %#08x, wasted %#08x, unchecked %#08x, free %#08x)\n", + jeb->offset, jeb->used_size, jeb->dirty_size, jeb->wasted_size, + jeb->unchecked_size, jeb->free_size); + } + } + } + + if (list_empty(&c->bad_used_list)) { + printk(JFFS2_DBG "bad_used_list: empty\n"); + } else { + struct list_head *this; + + list_for_each(this, &c->bad_used_list) { + struct jffs2_eraseblock *jeb = list_entry(this, struct jffs2_eraseblock, list); + + if (!(jeb->used_size == 0 && jeb->dirty_size == 0 && jeb->wasted_size == 0)) { + printk(JFFS2_DBG "bad_used_list: %#08x (used %#08x, dirty %#08x, wasted %#08x, unchecked %#08x, free %#08x)\n", + jeb->offset, jeb->used_size, jeb->dirty_size, jeb->wasted_size, + jeb->unchecked_size, jeb->free_size); + } + } + } +} + +void +__jffs2_dbg_dump_fragtree(struct jffs2_inode_info *f) +{ + down(&f->sem); + jffs2_dbg_dump_fragtree_nolock(f); + up(&f->sem); +} + +void +__jffs2_dbg_dump_fragtree_nolock(struct jffs2_inode_info *f) +{ + struct jffs2_node_frag *this = frag_first(&f->fragtree); + uint32_t lastofs = 0; + int buggy = 0; + + printk(JFFS2_DBG_MSG_PREFIX " dump fragtree of ino #%u\n", f->inocache->ino); + while(this) { + if (this->node) + printk(JFFS2_DBG "frag %#04x-%#04x: %#08x(%d) on flash (*%p), left (%p), right (%p), parent (%p)\n", + this->ofs, this->ofs+this->size, ref_offset(this->node->raw), + ref_flags(this->node->raw), this, frag_left(this), frag_right(this), + frag_parent(this)); + else + printk(JFFS2_DBG "frag %#04x-%#04x: hole (*%p). left (%p), right (%p), parent (%p)\n", + this->ofs, this->ofs+this->size, this, frag_left(this), + frag_right(this), frag_parent(this)); + if (this->ofs != lastofs) + buggy = 1; + lastofs = this->ofs + this->size; + this = frag_next(this); + } + + if (f->metadata) + printk(JFFS2_DBG "metadata at 0x%08x\n", ref_offset(f->metadata->raw)); + + if (buggy) { + JFFS2_ERROR("frag tree got a hole in it.\n"); + BUG(); + } +} + +#define JFFS2_BUFDUMP_BYTES_PER_LINE 32 +void +__jffs2_dbg_dump_buffer(unsigned char *buf, int len, uint32_t offs) +{ + int skip; + int i; + + printk(JFFS2_DBG_MSG_PREFIX " dump from offset %#08x to offset %#08x (%x bytes).\n", + offs, offs + len, len); + i = skip = offs % JFFS2_BUFDUMP_BYTES_PER_LINE; + offs = offs & ~(JFFS2_BUFDUMP_BYTES_PER_LINE - 1); + + if (skip != 0) + printk(JFFS2_DBG "%#08x: ", offs); + + while (skip--) + printk(" "); + + while (i < len) { + if ((i % JFFS2_BUFDUMP_BYTES_PER_LINE) == 0 && i != len -1) { + if (i != 0) + printk("\n"); + offs += JFFS2_BUFDUMP_BYTES_PER_LINE; + printk(JFFS2_DBG "%0#8x: ", offs); + } + + printk("%02x ", buf[i]); + + i += 1; + } + + printk("\n"); +} + +/* + * Dump a JFFS2 node. + */ +void +__jffs2_dbg_dump_node(struct jffs2_sb_info *c, uint32_t ofs) +{ + union jffs2_node_union node; + int len = sizeof(union jffs2_node_union); + size_t retlen; + uint32_t crc; + int ret; + + printk(JFFS2_DBG_MSG_PREFIX " dump node at offset %#08x.\n", ofs); + + ret = jffs2_flash_read(c, ofs, len, &retlen, (unsigned char *)&node); + if (ret || (retlen != len)) { + JFFS2_ERROR("read %d bytes failed or short. ret %d, retlen %zd.\n", + len, ret, retlen); + return; + } + + printk(JFFS2_DBG "magic:\t%#04x\n", je16_to_cpu(node.u.magic)); + printk(JFFS2_DBG "nodetype:\t%#04x\n", je16_to_cpu(node.u.nodetype)); + printk(JFFS2_DBG "totlen:\t%#08x\n", je32_to_cpu(node.u.totlen)); + printk(JFFS2_DBG "hdr_crc:\t%#08x\n", je32_to_cpu(node.u.hdr_crc)); + + crc = crc32(0, &node.u, sizeof(node.u) - 4); + if (crc != je32_to_cpu(node.u.hdr_crc)) { + JFFS2_ERROR("wrong common header CRC.\n"); + return; + } + + if (je16_to_cpu(node.u.magic) != JFFS2_MAGIC_BITMASK && + je16_to_cpu(node.u.magic) != JFFS2_OLD_MAGIC_BITMASK) + { + JFFS2_ERROR("wrong node magic: %#04x instead of %#04x.\n", + je16_to_cpu(node.u.magic), JFFS2_MAGIC_BITMASK); + return; + } + + switch(je16_to_cpu(node.u.nodetype)) { + + case JFFS2_NODETYPE_INODE: + + printk(JFFS2_DBG "the node is inode node\n"); + printk(JFFS2_DBG "ino:\t%#08x\n", je32_to_cpu(node.i.ino)); + printk(JFFS2_DBG "version:\t%#08x\n", je32_to_cpu(node.i.version)); + printk(JFFS2_DBG "mode:\t%#08x\n", node.i.mode.m); + printk(JFFS2_DBG "uid:\t%#04x\n", je16_to_cpu(node.i.uid)); + printk(JFFS2_DBG "gid:\t%#04x\n", je16_to_cpu(node.i.gid)); + printk(JFFS2_DBG "isize:\t%#08x\n", je32_to_cpu(node.i.isize)); + printk(JFFS2_DBG "atime:\t%#08x\n", je32_to_cpu(node.i.atime)); + printk(JFFS2_DBG "mtime:\t%#08x\n", je32_to_cpu(node.i.mtime)); + printk(JFFS2_DBG "ctime:\t%#08x\n", je32_to_cpu(node.i.ctime)); + printk(JFFS2_DBG "offset:\t%#08x\n", je32_to_cpu(node.i.offset)); + printk(JFFS2_DBG "csize:\t%#08x\n", je32_to_cpu(node.i.csize)); + printk(JFFS2_DBG "dsize:\t%#08x\n", je32_to_cpu(node.i.dsize)); + printk(JFFS2_DBG "compr:\t%#02x\n", node.i.compr); + printk(JFFS2_DBG "usercompr:\t%#02x\n", node.i.usercompr); + printk(JFFS2_DBG "flags:\t%#04x\n", je16_to_cpu(node.i.flags)); + printk(JFFS2_DBG "data_crc:\t%#08x\n", je32_to_cpu(node.i.data_crc)); + printk(JFFS2_DBG "node_crc:\t%#08x\n", je32_to_cpu(node.i.node_crc)); + + crc = crc32(0, &node.i, sizeof(node.i) - 8); + if (crc != je32_to_cpu(node.i.node_crc)) { + JFFS2_ERROR("wrong node header CRC.\n"); + return; + } + break; + + case JFFS2_NODETYPE_DIRENT: + + printk(JFFS2_DBG "the node is dirent node\n"); + printk(JFFS2_DBG "pino:\t%#08x\n", je32_to_cpu(node.d.pino)); + printk(JFFS2_DBG "version:\t%#08x\n", je32_to_cpu(node.d.version)); + printk(JFFS2_DBG "ino:\t%#08x\n", je32_to_cpu(node.d.ino)); + printk(JFFS2_DBG "mctime:\t%#08x\n", je32_to_cpu(node.d.mctime)); + printk(JFFS2_DBG "nsize:\t%#02x\n", node.d.nsize); + printk(JFFS2_DBG "type:\t%#02x\n", node.d.type); + printk(JFFS2_DBG "node_crc:\t%#08x\n", je32_to_cpu(node.d.node_crc)); + printk(JFFS2_DBG "name_crc:\t%#08x\n", je32_to_cpu(node.d.name_crc)); + + node.d.name[node.d.nsize] = '\0'; + printk(JFFS2_DBG "name:\t\"%s\"\n", node.d.name); + + crc = crc32(0, &node.d, sizeof(node.d) - 8); + if (crc != je32_to_cpu(node.d.node_crc)) { + JFFS2_ERROR("wrong node header CRC.\n"); + return; + } + break; + + default: + printk(JFFS2_DBG "node type is unknown\n"); + break; + } +} +#endif /* JFFS2_DBG_DUMPS || JFFS2_DBG_PARANOIA_CHECKS */ diff --git a/fs/jffs2/debug.h b/fs/jffs2/debug.h new file mode 100644 index 000000000000..162af6dfe292 --- /dev/null +++ b/fs/jffs2/debug.h @@ -0,0 +1,279 @@ +/* + * JFFS2 -- Journalling Flash File System, Version 2. + * + * Copyright (C) 2001-2003 Red Hat, Inc. + * + * Created by David Woodhouse <dwmw2@infradead.org> + * + * For licensing information, see the file 'LICENCE' in this directory. + * + * $Id: debug.h,v 1.21 2005/11/07 11:14:39 gleixner Exp $ + * + */ +#ifndef _JFFS2_DEBUG_H_ +#define _JFFS2_DEBUG_H_ + +#include <linux/config.h> + +#ifndef CONFIG_JFFS2_FS_DEBUG +#define CONFIG_JFFS2_FS_DEBUG 0 +#endif + +#if CONFIG_JFFS2_FS_DEBUG > 0 +/* Enable "paranoia" checks and dumps */ +#define JFFS2_DBG_PARANOIA_CHECKS +#define JFFS2_DBG_DUMPS + +/* + * By defining/undefining the below macros one may select debugging messages + * fro specific JFFS2 subsystems. + */ +#define JFFS2_DBG_READINODE_MESSAGES +#define JFFS2_DBG_FRAGTREE_MESSAGES +#define JFFS2_DBG_DENTLIST_MESSAGES +#define JFFS2_DBG_NODEREF_MESSAGES +#define JFFS2_DBG_INOCACHE_MESSAGES +#define JFFS2_DBG_SUMMARY_MESSAGES +#define JFFS2_DBG_FSBUILD_MESSAGES +#endif + +#if CONFIG_JFFS2_FS_DEBUG > 1 +#define JFFS2_DBG_FRAGTREE2_MESSAGES +#define JFFS2_DBG_MEMALLOC_MESSAGES +#endif + +/* Sanity checks are supposed to be light-weight and enabled by default */ +#define JFFS2_DBG_SANITY_CHECKS + +/* + * Dx() are mainly used for debugging messages, they must go away and be + * superseded by nicer dbg_xxx() macros... + */ +#if CONFIG_JFFS2_FS_DEBUG > 0 +#define D1(x) x +#else +#define D1(x) +#endif + +#if CONFIG_JFFS2_FS_DEBUG > 1 +#define D2(x) x +#else +#define D2(x) +#endif + +/* The prefixes of JFFS2 messages */ +#define JFFS2_DBG_PREFIX "[JFFS2 DBG]" +#define JFFS2_ERR_PREFIX "JFFS2 error:" +#define JFFS2_WARN_PREFIX "JFFS2 warning:" +#define JFFS2_NOTICE_PREFIX "JFFS2 notice:" + +#define JFFS2_ERR KERN_ERR +#define JFFS2_WARN KERN_WARNING +#define JFFS2_NOT KERN_NOTICE +#define JFFS2_DBG KERN_DEBUG + +#define JFFS2_DBG_MSG_PREFIX JFFS2_DBG JFFS2_DBG_PREFIX +#define JFFS2_ERR_MSG_PREFIX JFFS2_ERR JFFS2_ERR_PREFIX +#define JFFS2_WARN_MSG_PREFIX JFFS2_WARN JFFS2_WARN_PREFIX +#define JFFS2_NOTICE_MSG_PREFIX JFFS2_NOT JFFS2_NOTICE_PREFIX + +/* JFFS2 message macros */ +#define JFFS2_ERROR(fmt, ...) \ + do { \ + printk(JFFS2_ERR_MSG_PREFIX \ + " (%d) %s: " fmt, current->pid, \ + __FUNCTION__ , ##__VA_ARGS__); \ + } while(0) + +#define JFFS2_WARNING(fmt, ...) \ + do { \ + printk(JFFS2_WARN_MSG_PREFIX \ + " (%d) %s: " fmt, current->pid, \ + __FUNCTION__ , ##__VA_ARGS__); \ + } while(0) + +#define JFFS2_NOTICE(fmt, ...) \ + do { \ + printk(JFFS2_NOTICE_MSG_PREFIX \ + " (%d) %s: " fmt, current->pid, \ + __FUNCTION__ , ##__VA_ARGS__); \ + } while(0) + +#define JFFS2_DEBUG(fmt, ...) \ + do { \ + printk(JFFS2_DBG_MSG_PREFIX \ + " (%d) %s: " fmt, current->pid, \ + __FUNCTION__ , ##__VA_ARGS__); \ + } while(0) + +/* + * We split our debugging messages on several parts, depending on the JFFS2 + * subsystem the message belongs to. + */ +/* Read inode debugging messages */ +#ifdef JFFS2_DBG_READINODE_MESSAGES +#define dbg_readinode(fmt, ...) JFFS2_DEBUG(fmt, ##__VA_ARGS__) +#else +#define dbg_readinode(fmt, ...) +#endif + +/* Fragtree build debugging messages */ +#ifdef JFFS2_DBG_FRAGTREE_MESSAGES +#define dbg_fragtree(fmt, ...) JFFS2_DEBUG(fmt, ##__VA_ARGS__) +#else +#define dbg_fragtree(fmt, ...) +#endif +#ifdef JFFS2_DBG_FRAGTREE2_MESSAGES +#define dbg_fragtree2(fmt, ...) JFFS2_DEBUG(fmt, ##__VA_ARGS__) +#else +#define dbg_fragtree2(fmt, ...) +#endif + +/* Directory entry list manilulation debugging messages */ +#ifdef JFFS2_DBG_DENTLIST_MESSAGES +#define dbg_dentlist(fmt, ...) JFFS2_DEBUG(fmt, ##__VA_ARGS__) +#else +#define dbg_dentlist(fmt, ...) +#endif + +/* Print the messages about manipulating node_refs */ +#ifdef JFFS2_DBG_NODEREF_MESSAGES +#define dbg_noderef(fmt, ...) JFFS2_DEBUG(fmt, ##__VA_ARGS__) +#else +#define dbg_noderef(fmt, ...) +#endif + +/* Manipulations with the list of inodes (JFFS2 inocache) */ +#ifdef JFFS2_DBG_INOCACHE_MESSAGES +#define dbg_inocache(fmt, ...) JFFS2_DEBUG(fmt, ##__VA_ARGS__) +#else +#define dbg_inocache(fmt, ...) +#endif + +/* Summary debugging messages */ +#ifdef JFFS2_DBG_SUMMARY_MESSAGES +#define dbg_summary(fmt, ...) JFFS2_DEBUG(fmt, ##__VA_ARGS__) +#else +#define dbg_summary(fmt, ...) +#endif + +/* File system build messages */ +#ifdef JFFS2_DBG_FSBUILD_MESSAGES +#define dbg_fsbuild(fmt, ...) JFFS2_DEBUG(fmt, ##__VA_ARGS__) +#else +#define dbg_fsbuild(fmt, ...) +#endif + +/* Watch the object allocations */ +#ifdef JFFS2_DBG_MEMALLOC_MESSAGES +#define dbg_memalloc(fmt, ...) JFFS2_DEBUG(fmt, ##__VA_ARGS__) +#else +#define dbg_memalloc(fmt, ...) +#endif + + +/* "Sanity" checks */ +void +__jffs2_dbg_acct_sanity_check_nolock(struct jffs2_sb_info *c, + struct jffs2_eraseblock *jeb); +void +__jffs2_dbg_acct_sanity_check(struct jffs2_sb_info *c, + struct jffs2_eraseblock *jeb); + +/* "Paranoia" checks */ +void +__jffs2_dbg_fragtree_paranoia_check(struct jffs2_inode_info *f); +void +__jffs2_dbg_fragtree_paranoia_check_nolock(struct jffs2_inode_info *f); +void +__jffs2_dbg_acct_paranoia_check(struct jffs2_sb_info *c, + struct jffs2_eraseblock *jeb); +void +__jffs2_dbg_acct_paranoia_check_nolock(struct jffs2_sb_info *c, + struct jffs2_eraseblock *jeb); +void +__jffs2_dbg_prewrite_paranoia_check(struct jffs2_sb_info *c, + uint32_t ofs, int len); + +/* "Dump" functions */ +void +__jffs2_dbg_dump_jeb(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb); +void +__jffs2_dbg_dump_jeb_nolock(struct jffs2_eraseblock *jeb); +void +__jffs2_dbg_dump_block_lists(struct jffs2_sb_info *c); +void +__jffs2_dbg_dump_block_lists_nolock(struct jffs2_sb_info *c); +void +__jffs2_dbg_dump_node_refs(struct jffs2_sb_info *c, + struct jffs2_eraseblock *jeb); +void +__jffs2_dbg_dump_node_refs_nolock(struct jffs2_sb_info *c, + struct jffs2_eraseblock *jeb); +void +__jffs2_dbg_dump_fragtree(struct jffs2_inode_info *f); +void +__jffs2_dbg_dump_fragtree_nolock(struct jffs2_inode_info *f); +void +__jffs2_dbg_dump_buffer(unsigned char *buf, int len, uint32_t offs); +void +__jffs2_dbg_dump_node(struct jffs2_sb_info *c, uint32_t ofs); + +#ifdef JFFS2_DBG_PARANOIA_CHECKS +#define jffs2_dbg_fragtree_paranoia_check(f) \ + __jffs2_dbg_fragtree_paranoia_check(f) +#define jffs2_dbg_fragtree_paranoia_check_nolock(f) \ + __jffs2_dbg_fragtree_paranoia_check_nolock(f) +#define jffs2_dbg_acct_paranoia_check(c, jeb) \ + __jffs2_dbg_acct_paranoia_check(c,jeb) +#define jffs2_dbg_acct_paranoia_check_nolock(c, jeb) \ + __jffs2_dbg_acct_paranoia_check_nolock(c,jeb) +#define jffs2_dbg_prewrite_paranoia_check(c, ofs, len) \ + __jffs2_dbg_prewrite_paranoia_check(c, ofs, len) +#else +#define jffs2_dbg_fragtree_paranoia_check(f) +#define jffs2_dbg_fragtree_paranoia_check_nolock(f) +#define jffs2_dbg_acct_paranoia_check(c, jeb) +#define jffs2_dbg_acct_paranoia_check_nolock(c, jeb) +#define jffs2_dbg_prewrite_paranoia_check(c, ofs, len) +#endif /* !JFFS2_PARANOIA_CHECKS */ + +#ifdef JFFS2_DBG_DUMPS +#define jffs2_dbg_dump_jeb(c, jeb) \ + __jffs2_dbg_dump_jeb(c, jeb); +#define jffs2_dbg_dump_jeb_nolock(jeb) \ + __jffs2_dbg_dump_jeb_nolock(jeb); +#define jffs2_dbg_dump_block_lists(c) \ + __jffs2_dbg_dump_block_lists(c) +#define jffs2_dbg_dump_block_lists_nolock(c) \ + __jffs2_dbg_dump_block_lists_nolock(c) +#define jffs2_dbg_dump_fragtree(f) \ + __jffs2_dbg_dump_fragtree(f); +#define jffs2_dbg_dump_fragtree_nolock(f) \ + __jffs2_dbg_dump_fragtree_nolock(f); +#define jffs2_dbg_dump_buffer(buf, len, offs) \ + __jffs2_dbg_dump_buffer(*buf, len, offs); +#define jffs2_dbg_dump_node(c, ofs) \ + __jffs2_dbg_dump_node(c, ofs); +#else +#define jffs2_dbg_dump_jeb(c, jeb) +#define jffs2_dbg_dump_jeb_nolock(jeb) +#define jffs2_dbg_dump_block_lists(c) +#define jffs2_dbg_dump_block_lists_nolock(c) +#define jffs2_dbg_dump_fragtree(f) +#define jffs2_dbg_dump_fragtree_nolock(f) +#define jffs2_dbg_dump_buffer(buf, len, offs) +#define jffs2_dbg_dump_node(c, ofs) +#endif /* !JFFS2_DBG_DUMPS */ + +#ifdef JFFS2_DBG_SANITY_CHECKS +#define jffs2_dbg_acct_sanity_check(c, jeb) \ + __jffs2_dbg_acct_sanity_check(c, jeb) +#define jffs2_dbg_acct_sanity_check_nolock(c, jeb) \ + __jffs2_dbg_acct_sanity_check_nolock(c, jeb) +#else +#define jffs2_dbg_acct_sanity_check(c, jeb) +#define jffs2_dbg_acct_sanity_check_nolock(c, jeb) +#endif /* !JFFS2_DBG_SANITY_CHECKS */ + +#endif /* _JFFS2_DEBUG_H_ */ diff --git a/fs/jffs2/dir.c b/fs/jffs2/dir.c index 3ca0d25eef1d..a7bf9cb2567f 100644 --- a/fs/jffs2/dir.c +++ b/fs/jffs2/dir.c @@ -7,7 +7,7 @@ * * For licensing information, see the file 'LICENCE' in this directory. * - * $Id: dir.c,v 1.86 2005/07/06 12:13:09 dwmw2 Exp $ + * $Id: dir.c,v 1.90 2005/11/07 11:14:39 gleixner Exp $ * */ @@ -64,7 +64,7 @@ struct inode_operations jffs2_dir_inode_operations = /* We keep the dirent list sorted in increasing order of name hash, - and we use the same hash function as the dentries. Makes this + and we use the same hash function as the dentries. Makes this nice and simple */ static struct dentry *jffs2_lookup(struct inode *dir_i, struct dentry *target, @@ -85,7 +85,7 @@ static struct dentry *jffs2_lookup(struct inode *dir_i, struct dentry *target, /* NB: The 2.2 backport will need to explicitly check for '.' and '..' here */ for (fd_list = dir_f->dents; fd_list && fd_list->nhash <= target->d_name.hash; fd_list = fd_list->next) { - if (fd_list->nhash == target->d_name.hash && + if (fd_list->nhash == target->d_name.hash && (!fd || fd_list->version > fd->version) && strlen(fd_list->name) == target->d_name.len && !strncmp(fd_list->name, target->d_name.name, target->d_name.len)) { @@ -147,7 +147,7 @@ static int jffs2_readdir(struct file *filp, void *dirent, filldir_t filldir) curofs++; /* First loop: curofs = 2; offset = 2 */ if (curofs < offset) { - D2(printk(KERN_DEBUG "Skipping dirent: \"%s\", ino #%u, type %d, because curofs %ld < offset %ld\n", + D2(printk(KERN_DEBUG "Skipping dirent: \"%s\", ino #%u, type %d, because curofs %ld < offset %ld\n", fd->name, fd->ino, fd->type, curofs, offset)); continue; } @@ -182,7 +182,7 @@ static int jffs2_create(struct inode *dir_i, struct dentry *dentry, int mode, ri = jffs2_alloc_raw_inode(); if (!ri) return -ENOMEM; - + c = JFFS2_SB_INFO(dir_i->i_sb); D1(printk(KERN_DEBUG "jffs2_create()\n")); @@ -203,7 +203,7 @@ static int jffs2_create(struct inode *dir_i, struct dentry *dentry, int mode, f = JFFS2_INODE_INFO(inode); dir_f = JFFS2_INODE_INFO(dir_i); - ret = jffs2_do_create(c, dir_f, f, ri, + ret = jffs2_do_create(c, dir_f, f, ri, dentry->d_name.name, dentry->d_name.len); if (ret) { @@ -232,11 +232,14 @@ static int jffs2_unlink(struct inode *dir_i, struct dentry *dentry) struct jffs2_inode_info *dir_f = JFFS2_INODE_INFO(dir_i); struct jffs2_inode_info *dead_f = JFFS2_INODE_INFO(dentry->d_inode); int ret; + uint32_t now = get_seconds(); - ret = jffs2_do_unlink(c, dir_f, dentry->d_name.name, - dentry->d_name.len, dead_f); + ret = jffs2_do_unlink(c, dir_f, dentry->d_name.name, + dentry->d_name.len, dead_f, now); if (dead_f->inocache) dentry->d_inode->i_nlink = dead_f->inocache->nlink; + if (!ret) + dir_i->i_mtime = dir_i->i_ctime = ITIME(now); return ret; } /***********************************************************************/ @@ -249,6 +252,7 @@ static int jffs2_link (struct dentry *old_dentry, struct inode *dir_i, struct de struct jffs2_inode_info *dir_f = JFFS2_INODE_INFO(dir_i); int ret; uint8_t type; + uint32_t now; /* Don't let people make hard links to bad inodes. */ if (!f->inocache) @@ -261,13 +265,15 @@ static int jffs2_link (struct dentry *old_dentry, struct inode *dir_i, struct de type = (old_dentry->d_inode->i_mode & S_IFMT) >> 12; if (!type) type = DT_REG; - ret = jffs2_do_link(c, dir_f, f->inocache->ino, type, dentry->d_name.name, dentry->d_name.len); + now = get_seconds(); + ret = jffs2_do_link(c, dir_f, f->inocache->ino, type, dentry->d_name.name, dentry->d_name.len, now); if (!ret) { down(&f->sem); old_dentry->d_inode->i_nlink = ++f->inocache->nlink; up(&f->sem); d_instantiate(dentry, old_dentry->d_inode); + dir_i->i_mtime = dir_i->i_ctime = ITIME(now); atomic_inc(&old_dentry->d_inode->i_count); } return ret; @@ -297,14 +303,15 @@ static int jffs2_symlink (struct inode *dir_i, struct dentry *dentry, const char if (!ri) return -ENOMEM; - + c = JFFS2_SB_INFO(dir_i->i_sb); - - /* Try to reserve enough space for both node and dirent. - * Just the node will do for now, though + + /* Try to reserve enough space for both node and dirent. + * Just the node will do for now, though */ namelen = dentry->d_name.len; - ret = jffs2_reserve_space(c, sizeof(*ri) + targetlen, &phys_ofs, &alloclen, ALLOC_NORMAL); + ret = jffs2_reserve_space(c, sizeof(*ri) + targetlen, &phys_ofs, &alloclen, + ALLOC_NORMAL, JFFS2_SUMMARY_INODE_SIZE); if (ret) { jffs2_free_raw_inode(ri); @@ -331,7 +338,7 @@ static int jffs2_symlink (struct inode *dir_i, struct dentry *dentry, const char ri->compr = JFFS2_COMPR_NONE; ri->data_crc = cpu_to_je32(crc32(0, target, targetlen)); ri->node_crc = cpu_to_je32(crc32(0, ri, sizeof(*ri)-8)); - + fn = jffs2_write_dnode(c, f, ri, target, targetlen, phys_ofs, ALLOC_NORMAL); jffs2_free_raw_inode(ri); @@ -344,9 +351,9 @@ static int jffs2_symlink (struct inode *dir_i, struct dentry *dentry, const char return PTR_ERR(fn); } - /* We use f->dents field to store the target path. */ - f->dents = kmalloc(targetlen + 1, GFP_KERNEL); - if (!f->dents) { + /* We use f->target field to store the target path. */ + f->target = kmalloc(targetlen + 1, GFP_KERNEL); + if (!f->target) { printk(KERN_WARNING "Can't allocate %d bytes of memory\n", targetlen + 1); up(&f->sem); jffs2_complete_reservation(c); @@ -354,17 +361,18 @@ static int jffs2_symlink (struct inode *dir_i, struct dentry *dentry, const char return -ENOMEM; } - memcpy(f->dents, target, targetlen + 1); - D1(printk(KERN_DEBUG "jffs2_symlink: symlink's target '%s' cached\n", (char *)f->dents)); + memcpy(f->target, target, targetlen + 1); + D1(printk(KERN_DEBUG "jffs2_symlink: symlink's target '%s' cached\n", (char *)f->target)); - /* No data here. Only a metadata node, which will be + /* No data here. Only a metadata node, which will be obsoleted by the first data write */ f->metadata = fn; up(&f->sem); jffs2_complete_reservation(c); - ret = jffs2_reserve_space(c, sizeof(*rd)+namelen, &phys_ofs, &alloclen, ALLOC_NORMAL); + ret = jffs2_reserve_space(c, sizeof(*rd)+namelen, &phys_ofs, &alloclen, + ALLOC_NORMAL, JFFS2_SUMMARY_DIRENT_SIZE(namelen)); if (ret) { /* Eep. */ jffs2_clear_inode(inode); @@ -399,7 +407,7 @@ static int jffs2_symlink (struct inode *dir_i, struct dentry *dentry, const char fd = jffs2_write_dirent(c, dir_f, rd, dentry->d_name.name, namelen, phys_ofs, ALLOC_NORMAL); if (IS_ERR(fd)) { - /* dirent failed to write. Delete the inode normally + /* dirent failed to write. Delete the inode normally as if it were the final unlink() */ jffs2_complete_reservation(c); jffs2_free_raw_dirent(rd); @@ -442,14 +450,15 @@ static int jffs2_mkdir (struct inode *dir_i, struct dentry *dentry, int mode) ri = jffs2_alloc_raw_inode(); if (!ri) return -ENOMEM; - + c = JFFS2_SB_INFO(dir_i->i_sb); - /* Try to reserve enough space for both node and dirent. - * Just the node will do for now, though + /* Try to reserve enough space for both node and dirent. + * Just the node will do for now, though */ namelen = dentry->d_name.len; - ret = jffs2_reserve_space(c, sizeof(*ri), &phys_ofs, &alloclen, ALLOC_NORMAL); + ret = jffs2_reserve_space(c, sizeof(*ri), &phys_ofs, &alloclen, ALLOC_NORMAL, + JFFS2_SUMMARY_INODE_SIZE); if (ret) { jffs2_free_raw_inode(ri); @@ -473,7 +482,7 @@ static int jffs2_mkdir (struct inode *dir_i, struct dentry *dentry, int mode) ri->data_crc = cpu_to_je32(0); ri->node_crc = cpu_to_je32(crc32(0, ri, sizeof(*ri)-8)); - + fn = jffs2_write_dnode(c, f, ri, NULL, 0, phys_ofs, ALLOC_NORMAL); jffs2_free_raw_inode(ri); @@ -485,20 +494,21 @@ static int jffs2_mkdir (struct inode *dir_i, struct dentry *dentry, int mode) jffs2_clear_inode(inode); return PTR_ERR(fn); } - /* No data here. Only a metadata node, which will be + /* No data here. Only a metadata node, which will be obsoleted by the first data write */ f->metadata = fn; up(&f->sem); jffs2_complete_reservation(c); - ret = jffs2_reserve_space(c, sizeof(*rd)+namelen, &phys_ofs, &alloclen, ALLOC_NORMAL); + ret = jffs2_reserve_space(c, sizeof(*rd)+namelen, &phys_ofs, &alloclen, + ALLOC_NORMAL, JFFS2_SUMMARY_DIRENT_SIZE(namelen)); if (ret) { /* Eep. */ jffs2_clear_inode(inode); return ret; } - + rd = jffs2_alloc_raw_dirent(); if (!rd) { /* Argh. Now we treat it like a normal delete */ @@ -525,9 +535,9 @@ static int jffs2_mkdir (struct inode *dir_i, struct dentry *dentry, int mode) rd->name_crc = cpu_to_je32(crc32(0, dentry->d_name.name, namelen)); fd = jffs2_write_dirent(c, dir_f, rd, dentry->d_name.name, namelen, phys_ofs, ALLOC_NORMAL); - + if (IS_ERR(fd)) { - /* dirent failed to write. Delete the inode normally + /* dirent failed to write. Delete the inode normally as if it were the final unlink() */ jffs2_complete_reservation(c); jffs2_free_raw_dirent(rd); @@ -589,19 +599,20 @@ static int jffs2_mknod (struct inode *dir_i, struct dentry *dentry, int mode, de ri = jffs2_alloc_raw_inode(); if (!ri) return -ENOMEM; - + c = JFFS2_SB_INFO(dir_i->i_sb); - + if (S_ISBLK(mode) || S_ISCHR(mode)) { dev = cpu_to_je16(old_encode_dev(rdev)); devlen = sizeof(dev); } - - /* Try to reserve enough space for both node and dirent. - * Just the node will do for now, though + + /* Try to reserve enough space for both node and dirent. + * Just the node will do for now, though */ namelen = dentry->d_name.len; - ret = jffs2_reserve_space(c, sizeof(*ri) + devlen, &phys_ofs, &alloclen, ALLOC_NORMAL); + ret = jffs2_reserve_space(c, sizeof(*ri) + devlen, &phys_ofs, &alloclen, + ALLOC_NORMAL, JFFS2_SUMMARY_INODE_SIZE); if (ret) { jffs2_free_raw_inode(ri); @@ -627,7 +638,7 @@ static int jffs2_mknod (struct inode *dir_i, struct dentry *dentry, int mode, de ri->compr = JFFS2_COMPR_NONE; ri->data_crc = cpu_to_je32(crc32(0, &dev, devlen)); ri->node_crc = cpu_to_je32(crc32(0, ri, sizeof(*ri)-8)); - + fn = jffs2_write_dnode(c, f, ri, (char *)&dev, devlen, phys_ofs, ALLOC_NORMAL); jffs2_free_raw_inode(ri); @@ -639,14 +650,15 @@ static int jffs2_mknod (struct inode *dir_i, struct dentry *dentry, int mode, de jffs2_clear_inode(inode); return PTR_ERR(fn); } - /* No data here. Only a metadata node, which will be + /* No data here. Only a metadata node, which will be obsoleted by the first data write */ f->metadata = fn; up(&f->sem); jffs2_complete_reservation(c); - ret = jffs2_reserve_space(c, sizeof(*rd)+namelen, &phys_ofs, &alloclen, ALLOC_NORMAL); + ret = jffs2_reserve_space(c, sizeof(*rd)+namelen, &phys_ofs, &alloclen, + ALLOC_NORMAL, JFFS2_SUMMARY_DIRENT_SIZE(namelen)); if (ret) { /* Eep. */ jffs2_clear_inode(inode); @@ -682,9 +694,9 @@ static int jffs2_mknod (struct inode *dir_i, struct dentry *dentry, int mode, de rd->name_crc = cpu_to_je32(crc32(0, dentry->d_name.name, namelen)); fd = jffs2_write_dirent(c, dir_f, rd, dentry->d_name.name, namelen, phys_ofs, ALLOC_NORMAL); - + if (IS_ERR(fd)) { - /* dirent failed to write. Delete the inode normally + /* dirent failed to write. Delete the inode normally as if it were the final unlink() */ jffs2_complete_reservation(c); jffs2_free_raw_dirent(rd); @@ -716,8 +728,9 @@ static int jffs2_rename (struct inode *old_dir_i, struct dentry *old_dentry, struct jffs2_sb_info *c = JFFS2_SB_INFO(old_dir_i->i_sb); struct jffs2_inode_info *victim_f = NULL; uint8_t type; + uint32_t now; - /* The VFS will check for us and prevent trying to rename a + /* The VFS will check for us and prevent trying to rename a * file over a directory and vice versa, but if it's a directory, * the VFS can't check whether the victim is empty. The filesystem * needs to do that for itself. @@ -739,19 +752,20 @@ static int jffs2_rename (struct inode *old_dir_i, struct dentry *old_dentry, } /* XXX: We probably ought to alloc enough space for - both nodes at the same time. Writing the new link, + both nodes at the same time. Writing the new link, then getting -ENOSPC, is quite bad :) */ /* Make a hard link */ - + /* XXX: This is ugly */ type = (old_dentry->d_inode->i_mode & S_IFMT) >> 12; if (!type) type = DT_REG; - ret = jffs2_do_link(c, JFFS2_INODE_INFO(new_dir_i), + now = get_seconds(); + ret = jffs2_do_link(c, JFFS2_INODE_INFO(new_dir_i), old_dentry->d_inode->i_ino, type, - new_dentry->d_name.name, new_dentry->d_name.len); + new_dentry->d_name.name, new_dentry->d_name.len, now); if (ret) return ret; @@ -768,14 +782,14 @@ static int jffs2_rename (struct inode *old_dir_i, struct dentry *old_dentry, } } - /* If it was a directory we moved, and there was no victim, + /* If it was a directory we moved, and there was no victim, increase i_nlink on its new parent */ if (S_ISDIR(old_dentry->d_inode->i_mode) && !victim_f) new_dir_i->i_nlink++; /* Unlink the original */ - ret = jffs2_do_unlink(c, JFFS2_INODE_INFO(old_dir_i), - old_dentry->d_name.name, old_dentry->d_name.len, NULL); + ret = jffs2_do_unlink(c, JFFS2_INODE_INFO(old_dir_i), + old_dentry->d_name.name, old_dentry->d_name.len, NULL, now); /* We don't touch inode->i_nlink */ @@ -792,12 +806,15 @@ static int jffs2_rename (struct inode *old_dir_i, struct dentry *old_dentry, /* Might as well let the VFS know */ d_instantiate(new_dentry, old_dentry->d_inode); atomic_inc(&old_dentry->d_inode->i_count); + new_dir_i->i_mtime = new_dir_i->i_ctime = ITIME(now); return ret; } if (S_ISDIR(old_dentry->d_inode->i_mode)) old_dir_i->i_nlink--; + new_dir_i->i_mtime = new_dir_i->i_ctime = old_dir_i->i_mtime = old_dir_i->i_ctime = ITIME(now); + return 0; } diff --git a/fs/jffs2/erase.c b/fs/jffs2/erase.c index 787d84ac2bcd..dad68fdffe9e 100644 --- a/fs/jffs2/erase.c +++ b/fs/jffs2/erase.c @@ -7,7 +7,7 @@ * * For licensing information, see the file 'LICENCE' in this directory. * - * $Id: erase.c,v 1.80 2005/07/14 19:46:24 joern Exp $ + * $Id: erase.c,v 1.85 2005/09/20 14:53:15 dedekind Exp $ * */ @@ -24,7 +24,7 @@ struct erase_priv_struct { struct jffs2_eraseblock *jeb; struct jffs2_sb_info *c; }; - + #ifndef __ECOS static void jffs2_erase_callback(struct erase_info *); #endif @@ -48,7 +48,8 @@ static void jffs2_erase_block(struct jffs2_sb_info *c, #else /* Linux */ struct erase_info *instr; - D1(printk(KERN_DEBUG "jffs2_erase_block(): erase block %#x (range %#x-%#x)\n", jeb->offset, jeb->offset, jeb->offset + c->sector_size)); + D1(printk(KERN_DEBUG "jffs2_erase_block(): erase block %#08x (range %#08x-%#08x)\n", + jeb->offset, jeb->offset, jeb->offset + c->sector_size)); instr = kmalloc(sizeof(struct erase_info) + sizeof(struct erase_priv_struct), GFP_KERNEL); if (!instr) { printk(KERN_WARNING "kmalloc for struct erase_info in jffs2_erase_block failed. Refiling block for later\n"); @@ -70,7 +71,7 @@ static void jffs2_erase_block(struct jffs2_sb_info *c, instr->callback = jffs2_erase_callback; instr->priv = (unsigned long)(&instr[1]); instr->fail_addr = 0xffffffff; - + ((struct erase_priv_struct *)instr->priv)->jeb = jeb; ((struct erase_priv_struct *)instr->priv)->c = c; @@ -95,7 +96,7 @@ static void jffs2_erase_block(struct jffs2_sb_info *c, return; } - if (ret == -EROFS) + if (ret == -EROFS) printk(KERN_WARNING "Erase at 0x%08x failed immediately: -EROFS. Is the sector locked?\n", jeb->offset); else printk(KERN_WARNING "Erase at 0x%08x failed immediately: errno %d\n", jeb->offset, ret); @@ -196,7 +197,7 @@ static void jffs2_erase_failed(struct jffs2_sb_info *c, struct jffs2_eraseblock c->nr_erasing_blocks--; spin_unlock(&c->erase_completion_lock); wake_up(&c->erase_wait); -} +} #ifndef __ECOS static void jffs2_erase_callback(struct erase_info *instr) @@ -208,7 +209,7 @@ static void jffs2_erase_callback(struct erase_info *instr) jffs2_erase_failed(priv->c, priv->jeb, instr->fail_addr); } else { jffs2_erase_succeeded(priv->c, priv->jeb); - } + } kfree(instr); } #endif /* !__ECOS */ @@ -226,13 +227,13 @@ static inline void jffs2_remove_node_refs_from_ino_list(struct jffs2_sb_info *c, /* Walk the inode's list once, removing any nodes from this eraseblock */ while (1) { if (!(*prev)->next_in_ino) { - /* We're looking at the jffs2_inode_cache, which is + /* We're looking at the jffs2_inode_cache, which is at the end of the linked list. Stash it and continue from the beginning of the list */ ic = (struct jffs2_inode_cache *)(*prev); prev = &ic->nodes; continue; - } + } if (SECTOR_ADDR((*prev)->flash_offset) == jeb->offset) { /* It's in the block we're erasing */ @@ -266,7 +267,7 @@ static inline void jffs2_remove_node_refs_from_ino_list(struct jffs2_sb_info *c, printk(KERN_DEBUG "After remove_node_refs_from_ino_list: \n" KERN_DEBUG); this = ic->nodes; - + while(this) { printk( "0x%08x(%d)->", ref_offset(this), ref_flags(this)); if (++i == 5) { @@ -289,7 +290,7 @@ static void jffs2_free_all_node_refs(struct jffs2_sb_info *c, struct jffs2_erase while(jeb->first_node) { ref = jeb->first_node; jeb->first_node = ref->next_phys; - + /* Remove from the inode-list */ if (ref->next_in_ino) jffs2_remove_node_refs_from_ino_list(c, ref, jeb); @@ -306,7 +307,7 @@ static int jffs2_block_check_erase(struct jffs2_sb_info *c, struct jffs2_erasebl uint32_t ofs; size_t retlen; int ret = -EIO; - + ebuf = kmalloc(PAGE_SIZE, GFP_KERNEL); if (!ebuf) { printk(KERN_WARNING "Failed to allocate page buffer for verifying erase at 0x%08x. Refiling\n", jeb->offset); @@ -360,7 +361,7 @@ static void jffs2_mark_erased_block(struct jffs2_sb_info *c, struct jffs2_eraseb case -EIO: goto filebad; } - /* Write the erase complete marker */ + /* Write the erase complete marker */ D1(printk(KERN_DEBUG "Writing erased marker to block at 0x%08x\n", jeb->offset)); bad_offset = jeb->offset; @@ -398,7 +399,7 @@ static void jffs2_mark_erased_block(struct jffs2_sb_info *c, struct jffs2_eraseb vecs[0].iov_base = (unsigned char *) ▮ vecs[0].iov_len = sizeof(marker); ret = jffs2_flash_direct_writev(c, vecs, 1, jeb->offset, &retlen); - + if (ret || retlen != sizeof(marker)) { if (ret) printk(KERN_WARNING "Write clean marker to block at 0x%08x failed: %d\n", @@ -415,9 +416,9 @@ static void jffs2_mark_erased_block(struct jffs2_sb_info *c, struct jffs2_eraseb marker_ref->next_phys = NULL; marker_ref->flash_offset = jeb->offset | REF_NORMAL; marker_ref->__totlen = c->cleanmarker_size; - + jeb->first_node = jeb->last_node = marker_ref; - + jeb->free_size = c->sector_size - c->cleanmarker_size; jeb->used_size = c->cleanmarker_size; jeb->dirty_size = 0; @@ -429,8 +430,8 @@ static void jffs2_mark_erased_block(struct jffs2_sb_info *c, struct jffs2_eraseb c->free_size += jeb->free_size; c->used_size += jeb->used_size; - ACCT_SANITY_CHECK(c,jeb); - D1(ACCT_PARANOIA_CHECK(jeb)); + jffs2_dbg_acct_sanity_check_nolock(c,jeb); + jffs2_dbg_acct_paranoia_check_nolock(c, jeb); list_add_tail(&jeb->list, &c->free_list); c->nr_erasing_blocks--; diff --git a/fs/jffs2/file.c b/fs/jffs2/file.c index 8279bf0133ff..935f273dc57b 100644 --- a/fs/jffs2/file.c +++ b/fs/jffs2/file.c @@ -7,7 +7,7 @@ * * For licensing information, see the file 'LICENCE' in this directory. * - * $Id: file.c,v 1.102 2005/07/06 12:13:09 dwmw2 Exp $ + * $Id: file.c,v 1.104 2005/10/18 23:29:35 tpoynor Exp $ * */ @@ -34,8 +34,8 @@ int jffs2_fsync(struct file *filp, struct dentry *dentry, int datasync) /* Trigger GC to flush any pending writes for this inode */ jffs2_flush_wbuf_gc(c, inode->i_ino); - - return 0; + + return 0; } struct file_operations jffs2_file_operations = @@ -107,7 +107,7 @@ static int jffs2_readpage (struct file *filp, struct page *pg) { struct jffs2_inode_info *f = JFFS2_INODE_INFO(pg->mapping->host); int ret; - + down(&f->sem); ret = jffs2_do_readpage_unlock(pg->mapping->host, pg); up(&f->sem); @@ -130,11 +130,12 @@ static int jffs2_prepare_write (struct file *filp, struct page *pg, struct jffs2_raw_inode ri; struct jffs2_full_dnode *fn; uint32_t phys_ofs, alloc_len; - + D1(printk(KERN_DEBUG "Writing new hole frag 0x%x-0x%x between current EOF and new page\n", (unsigned int)inode->i_size, pageofs)); - ret = jffs2_reserve_space(c, sizeof(ri), &phys_ofs, &alloc_len, ALLOC_NORMAL); + ret = jffs2_reserve_space(c, sizeof(ri), &phys_ofs, &alloc_len, + ALLOC_NORMAL, JFFS2_SUMMARY_INODE_SIZE); if (ret) return ret; @@ -159,7 +160,7 @@ static int jffs2_prepare_write (struct file *filp, struct page *pg, ri.compr = JFFS2_COMPR_ZERO; ri.node_crc = cpu_to_je32(crc32(0, &ri, sizeof(ri)-8)); ri.data_crc = cpu_to_je32(0); - + fn = jffs2_write_dnode(c, f, &ri, NULL, 0, phys_ofs, ALLOC_NORMAL); if (IS_ERR(fn)) { @@ -186,7 +187,7 @@ static int jffs2_prepare_write (struct file *filp, struct page *pg, inode->i_size = pageofs; up(&f->sem); } - + /* Read in the page if it wasn't already present, unless it's a whole page */ if (!PageUptodate(pg) && (start || end < PAGE_CACHE_SIZE)) { down(&f->sem); @@ -217,7 +218,7 @@ static int jffs2_commit_write (struct file *filp, struct page *pg, if (!start && end == PAGE_CACHE_SIZE) { /* We need to avoid deadlock with page_cache_read() in jffs2_garbage_collect_pass(). So we have to mark the - page up to date, to prevent page_cache_read() from + page up to date, to prevent page_cache_read() from trying to re-lock it. */ SetPageUptodate(pg); } @@ -251,7 +252,7 @@ static int jffs2_commit_write (struct file *filp, struct page *pg, /* There was an error writing. */ SetPageError(pg); } - + /* Adjust writtenlen for the padding we did, so we don't confuse our caller */ if (writtenlen < (start&3)) writtenlen = 0; @@ -262,7 +263,7 @@ static int jffs2_commit_write (struct file *filp, struct page *pg, if (inode->i_size < (pg->index << PAGE_CACHE_SHIFT) + start + writtenlen) { inode->i_size = (pg->index << PAGE_CACHE_SHIFT) + start + writtenlen; inode->i_blocks = (inode->i_size + 511) >> 9; - + inode->i_ctime = inode->i_mtime = ITIME(je32_to_cpu(ri->ctime)); } } @@ -271,13 +272,13 @@ static int jffs2_commit_write (struct file *filp, struct page *pg, if (start+writtenlen < end) { /* generic_file_write has written more to the page cache than we've - actually written to the medium. Mark the page !Uptodate so that + actually written to the medium. Mark the page !Uptodate so that it gets reread */ D1(printk(KERN_DEBUG "jffs2_commit_write(): Not all bytes written. Marking page !uptodate\n")); SetPageError(pg); ClearPageUptodate(pg); } - D1(printk(KERN_DEBUG "jffs2_commit_write() returning %d\n",writtenlen?writtenlen:ret)); - return writtenlen?writtenlen:ret; + D1(printk(KERN_DEBUG "jffs2_commit_write() returning %d\n",start+writtenlen==end?0:ret)); + return start+writtenlen==end?0:ret; } diff --git a/fs/jffs2/fs.c b/fs/jffs2/fs.c index 5687c3f42002..d0fcc5f3497e 100644 --- a/fs/jffs2/fs.c +++ b/fs/jffs2/fs.c @@ -7,7 +7,7 @@ * * For licensing information, see the file 'LICENCE' in this directory. * - * $Id: fs.c,v 1.56 2005/07/06 12:13:09 dwmw2 Exp $ + * $Id: fs.c,v 1.66 2005/09/27 13:17:29 dedekind Exp $ * */ @@ -40,7 +40,7 @@ static int jffs2_do_setattr (struct inode *inode, struct iattr *iattr) int ret; D1(printk(KERN_DEBUG "jffs2_setattr(): ino #%lu\n", inode->i_ino)); ret = inode_change_ok(inode, iattr); - if (ret) + if (ret) return ret; /* Special cases - we don't want more than one data node @@ -73,8 +73,9 @@ static int jffs2_do_setattr (struct inode *inode, struct iattr *iattr) kfree(mdata); return -ENOMEM; } - - ret = jffs2_reserve_space(c, sizeof(*ri) + mdatalen, &phys_ofs, &alloclen, ALLOC_NORMAL); + + ret = jffs2_reserve_space(c, sizeof(*ri) + mdatalen, &phys_ofs, &alloclen, + ALLOC_NORMAL, JFFS2_SUMMARY_INODE_SIZE); if (ret) { jffs2_free_raw_inode(ri); if (S_ISLNK(inode->i_mode & S_IFMT)) @@ -83,7 +84,7 @@ static int jffs2_do_setattr (struct inode *inode, struct iattr *iattr) } down(&f->sem); ivalid = iattr->ia_valid; - + ri->magic = cpu_to_je16(JFFS2_MAGIC_BITMASK); ri->nodetype = cpu_to_je16(JFFS2_NODETYPE_INODE); ri->totlen = cpu_to_je32(sizeof(*ri) + mdatalen); @@ -99,7 +100,7 @@ static int jffs2_do_setattr (struct inode *inode, struct iattr *iattr) if (iattr->ia_mode & S_ISGID && !in_group_p(je16_to_cpu(ri->gid)) && !capable(CAP_FSETID)) ri->mode = cpu_to_jemode(iattr->ia_mode & ~S_ISGID); - else + else ri->mode = cpu_to_jemode(iattr->ia_mode); else ri->mode = cpu_to_jemode(inode->i_mode); @@ -128,7 +129,7 @@ static int jffs2_do_setattr (struct inode *inode, struct iattr *iattr) new_metadata = jffs2_write_dnode(c, f, ri, mdata, mdatalen, phys_ofs, ALLOC_NORMAL); if (S_ISLNK(inode->i_mode)) kfree(mdata); - + if (IS_ERR(new_metadata)) { jffs2_complete_reservation(c); jffs2_free_raw_inode(ri); @@ -147,7 +148,7 @@ static int jffs2_do_setattr (struct inode *inode, struct iattr *iattr) old_metadata = f->metadata; if (ivalid & ATTR_SIZE && inode->i_size > iattr->ia_size) - jffs2_truncate_fraglist (c, &f->fragtree, iattr->ia_size); + jffs2_truncate_fragtree (c, &f->fragtree, iattr->ia_size); if (ivalid & ATTR_SIZE && inode->i_size < iattr->ia_size) { jffs2_add_full_dnode_to_inode(c, f, new_metadata); @@ -166,7 +167,7 @@ static int jffs2_do_setattr (struct inode *inode, struct iattr *iattr) jffs2_complete_reservation(c); /* We have to do the vmtruncate() without f->sem held, since - some pages may be locked and waiting for it in readpage(). + some pages may be locked and waiting for it in readpage(). We are protected from a simultaneous write() extending i_size back past iattr->ia_size, because do_truncate() holds the generic inode semaphore. */ @@ -194,31 +195,27 @@ int jffs2_statfs(struct super_block *sb, struct kstatfs *buf) buf->f_namelen = JFFS2_MAX_NAME_LEN; spin_lock(&c->erase_completion_lock); - avail = c->dirty_size + c->free_size; if (avail > c->sector_size * c->resv_blocks_write) avail -= c->sector_size * c->resv_blocks_write; else avail = 0; + spin_unlock(&c->erase_completion_lock); buf->f_bavail = buf->f_bfree = avail >> PAGE_SHIFT; - D2(jffs2_dump_block_lists(c)); - - spin_unlock(&c->erase_completion_lock); - return 0; } void jffs2_clear_inode (struct inode *inode) { - /* We can forget about this inode for now - drop all + /* We can forget about this inode for now - drop all * the nodelists associated with it, etc. */ struct jffs2_sb_info *c = JFFS2_SB_INFO(inode->i_sb); struct jffs2_inode_info *f = JFFS2_INODE_INFO(inode); - + D1(printk(KERN_DEBUG "jffs2_clear_inode(): ino #%lu mode %o\n", inode->i_ino, inode->i_mode)); jffs2_do_clear_inode(c, f); @@ -237,7 +234,8 @@ void jffs2_read_inode (struct inode *inode) c = JFFS2_SB_INFO(inode->i_sb); jffs2_init_inode_info(f); - + down(&f->sem); + ret = jffs2_do_read_inode(c, f, inode->i_ino, &latest_node); if (ret) { @@ -257,14 +255,14 @@ void jffs2_read_inode (struct inode *inode) inode->i_blksize = PAGE_SIZE; inode->i_blocks = (inode->i_size + 511) >> 9; - + switch (inode->i_mode & S_IFMT) { jint16_t rdev; case S_IFLNK: inode->i_op = &jffs2_symlink_inode_operations; break; - + case S_IFDIR: { struct jffs2_full_dirent *fd; @@ -301,7 +299,7 @@ void jffs2_read_inode (struct inode *inode) jffs2_do_clear_inode(c, f); make_bad_inode(inode); return; - } + } case S_IFSOCK: case S_IFIFO: @@ -357,11 +355,11 @@ int jffs2_remount_fs (struct super_block *sb, int *flags, char *data) down(&c->alloc_sem); jffs2_flush_wbuf_pad(c); up(&c->alloc_sem); - } + } if (!(*flags & MS_RDONLY)) jffs2_start_garbage_collect_thread(c); - + *flags |= MS_NOATIME; return 0; @@ -395,14 +393,15 @@ struct inode *jffs2_new_inode (struct inode *dir_i, int mode, struct jffs2_raw_i D1(printk(KERN_DEBUG "jffs2_new_inode(): dir_i %ld, mode 0x%x\n", dir_i->i_ino, mode)); c = JFFS2_SB_INFO(sb); - + inode = new_inode(sb); - + if (!inode) return ERR_PTR(-ENOMEM); f = JFFS2_INODE_INFO(inode); jffs2_init_inode_info(f); + down(&f->sem); memset(ri, 0, sizeof(*ri)); /* Set OS-specific defaults for new inodes */ @@ -461,40 +460,24 @@ int jffs2_do_fill_super(struct super_block *sb, void *data, int silent) #endif c->flash_size = c->mtd->size; - - /* - * Check, if we have to concatenate physical blocks to larger virtual blocks - * to reduce the memorysize for c->blocks. (kmalloc allows max. 128K allocation) - */ - c->sector_size = c->mtd->erasesize; + c->sector_size = c->mtd->erasesize; blocks = c->flash_size / c->sector_size; - if (!(c->mtd->flags & MTD_NO_VIRTBLOCKS)) { - while ((blocks * sizeof (struct jffs2_eraseblock)) > (128 * 1024)) { - blocks >>= 1; - c->sector_size <<= 1; - } - } /* * Size alignment check */ if ((c->sector_size * blocks) != c->flash_size) { - c->flash_size = c->sector_size * blocks; + c->flash_size = c->sector_size * blocks; printk(KERN_INFO "jffs2: Flash size not aligned to erasesize, reducing to %dKiB\n", c->flash_size / 1024); } - if (c->sector_size != c->mtd->erasesize) - printk(KERN_INFO "jffs2: Erase block size too small (%dKiB). Using virtual blocks size (%dKiB) instead\n", - c->mtd->erasesize / 1024, c->sector_size / 1024); - if (c->flash_size < 5*c->sector_size) { printk(KERN_ERR "jffs2: Too few erase blocks (%d)\n", c->flash_size / c->sector_size); return -EINVAL; } c->cleanmarker_size = sizeof(struct jffs2_unknown_node); - /* Joern -- stick alignment for weird 8-byte-page flash here */ /* NAND (or other bizarre) flash... do setup accordingly */ ret = jffs2_flash_setup(c); @@ -517,7 +500,7 @@ int jffs2_do_fill_super(struct super_block *sb, void *data, int silent) root_i = iget(sb, 1); if (is_bad_inode(root_i)) { D1(printk(KERN_WARNING "get root inode failed\n")); - goto out_nodes; + goto out_root_i; } D1(printk(KERN_DEBUG "jffs2_do_fill_super(): d_alloc_root()\n")); @@ -535,10 +518,9 @@ int jffs2_do_fill_super(struct super_block *sb, void *data, int silent) out_root_i: iput(root_i); - out_nodes: jffs2_free_ino_caches(c); jffs2_free_raw_node_refs(c); - if (c->mtd->flags & MTD_NO_VIRTBLOCKS) + if (jffs2_blocks_use_vmalloc(c)) vfree(c->blocks); else kfree(c->blocks); @@ -563,16 +545,16 @@ struct jffs2_inode_info *jffs2_gc_fetch_inode(struct jffs2_sb_info *c, struct jffs2_inode_cache *ic; if (!nlink) { /* The inode has zero nlink but its nodes weren't yet marked - obsolete. This has to be because we're still waiting for + obsolete. This has to be because we're still waiting for the final (close() and) iput() to happen. - There's a possibility that the final iput() could have + There's a possibility that the final iput() could have happened while we were contemplating. In order to ensure that we don't cause a new read_inode() (which would fail) for the inode in question, we use ilookup() in this case instead of iget(). - The nlink can't _become_ zero at this point because we're + The nlink can't _become_ zero at this point because we're holding the alloc_sem, and jffs2_do_unlink() would also need that while decrementing nlink on any inode. */ @@ -619,19 +601,19 @@ struct jffs2_inode_info *jffs2_gc_fetch_inode(struct jffs2_sb_info *c, return JFFS2_INODE_INFO(inode); } -unsigned char *jffs2_gc_fetch_page(struct jffs2_sb_info *c, - struct jffs2_inode_info *f, +unsigned char *jffs2_gc_fetch_page(struct jffs2_sb_info *c, + struct jffs2_inode_info *f, unsigned long offset, unsigned long *priv) { struct inode *inode = OFNI_EDONI_2SFFJ(f); struct page *pg; - pg = read_cache_page(inode->i_mapping, offset >> PAGE_CACHE_SHIFT, + pg = read_cache_page(inode->i_mapping, offset >> PAGE_CACHE_SHIFT, (void *)jffs2_do_readpage_unlock, inode); if (IS_ERR(pg)) return (void *)pg; - + *priv = (unsigned long)pg; return kmap(pg); } @@ -648,7 +630,7 @@ void jffs2_gc_release_page(struct jffs2_sb_info *c, static int jffs2_flash_setup(struct jffs2_sb_info *c) { int ret = 0; - + if (jffs2_cleanmarker_oob(c)) { /* NAND flash... do setup accordingly */ ret = jffs2_nand_flash_setup(c); @@ -662,14 +644,21 @@ static int jffs2_flash_setup(struct jffs2_sb_info *c) { if (ret) return ret; } - + /* and Dataflash */ if (jffs2_dataflash(c)) { ret = jffs2_dataflash_setup(c); if (ret) return ret; } - + + /* and Intel "Sibley" flash */ + if (jffs2_nor_wbuf_flash(c)) { + ret = jffs2_nor_wbuf_flash_setup(c); + if (ret) + return ret; + } + return ret; } @@ -683,9 +672,14 @@ void jffs2_flash_cleanup(struct jffs2_sb_info *c) { if (jffs2_nor_ecc(c)) { jffs2_nor_ecc_flash_cleanup(c); } - + /* and DataFlash */ if (jffs2_dataflash(c)) { jffs2_dataflash_cleanup(c); } + + /* and Intel "Sibley" flash */ + if (jffs2_nor_wbuf_flash(c)) { + jffs2_nor_wbuf_flash_cleanup(c); + } } diff --git a/fs/jffs2/gc.c b/fs/jffs2/gc.c index 7086cd634503..f9ffece453a3 100644 --- a/fs/jffs2/gc.c +++ b/fs/jffs2/gc.c @@ -7,7 +7,7 @@ * * For licensing information, see the file 'LICENCE' in this directory. * - * $Id: gc.c,v 1.148 2005/04/09 10:47:00 dedekind Exp $ + * $Id: gc.c,v 1.155 2005/11/07 11:14:39 gleixner Exp $ * */ @@ -21,14 +21,14 @@ #include "nodelist.h" #include "compr.h" -static int jffs2_garbage_collect_pristine(struct jffs2_sb_info *c, +static int jffs2_garbage_collect_pristine(struct jffs2_sb_info *c, struct jffs2_inode_cache *ic, struct jffs2_raw_node_ref *raw); -static int jffs2_garbage_collect_metadata(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb, +static int jffs2_garbage_collect_metadata(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb, struct jffs2_inode_info *f, struct jffs2_full_dnode *fd); -static int jffs2_garbage_collect_dirent(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb, +static int jffs2_garbage_collect_dirent(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb, struct jffs2_inode_info *f, struct jffs2_full_dirent *fd); -static int jffs2_garbage_collect_deletion_dirent(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb, +static int jffs2_garbage_collect_deletion_dirent(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb, struct jffs2_inode_info *f, struct jffs2_full_dirent *fd); static int jffs2_garbage_collect_hole(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb, struct jffs2_inode_info *f, struct jffs2_full_dnode *fn, @@ -55,7 +55,7 @@ again: D1(printk(KERN_DEBUG "Picking block from bad_used_list to GC next\n")); nextlist = &c->bad_used_list; } else if (n < 50 && !list_empty(&c->erasable_list)) { - /* Note that most of them will have gone directly to be erased. + /* Note that most of them will have gone directly to be erased. So don't favour the erasable_list _too_ much. */ D1(printk(KERN_DEBUG "Picking block from erasable_list to GC next\n")); nextlist = &c->erasable_list; @@ -101,7 +101,7 @@ again: printk(KERN_WARNING "Eep. ret->gc_node for block at 0x%08x is NULL\n", ret->offset); BUG(); } - + /* Have we accidentally picked a clean block with wasted space ? */ if (ret->wasted_size) { D1(printk(KERN_DEBUG "Converting wasted_size %08x to dirty_size\n", ret->wasted_size)); @@ -111,7 +111,6 @@ again: ret->wasted_size = 0; } - D2(jffs2_dump_block_lists(c)); return ret; } @@ -137,12 +136,12 @@ int jffs2_garbage_collect_pass(struct jffs2_sb_info *c) /* We can't start doing GC yet. We haven't finished checking the node CRCs etc. Do it now. */ - + /* checked_ino is protected by the alloc_sem */ if (c->checked_ino > c->highest_ino) { printk(KERN_CRIT "Checked all inodes but still 0x%x bytes of unchecked space?\n", c->unchecked_size); - D2(jffs2_dump_block_lists(c)); + jffs2_dbg_dump_block_lists_nolock(c); spin_unlock(&c->erase_completion_lock); BUG(); } @@ -179,7 +178,7 @@ int jffs2_garbage_collect_pass(struct jffs2_sb_info *c) case INO_STATE_READING: /* We need to wait for it to finish, lest we move on - and trigger the BUG() above while we haven't yet + and trigger the BUG() above while we haven't yet finished checking all its nodes */ D1(printk(KERN_DEBUG "Waiting for ino #%u to finish reading\n", ic->ino)); up(&c->alloc_sem); @@ -229,13 +228,13 @@ int jffs2_garbage_collect_pass(struct jffs2_sb_info *c) } raw = jeb->gc_node; - + while(ref_obsolete(raw)) { D1(printk(KERN_DEBUG "Node at 0x%08x is obsolete... skipping\n", ref_offset(raw))); raw = raw->next_phys; if (unlikely(!raw)) { printk(KERN_WARNING "eep. End of raw list while still supposedly nodes to GC\n"); - printk(KERN_WARNING "erase block at 0x%08x. free_size 0x%08x, dirty_size 0x%08x, used_size 0x%08x\n", + printk(KERN_WARNING "erase block at 0x%08x. free_size 0x%08x, dirty_size 0x%08x, used_size 0x%08x\n", jeb->offset, jeb->free_size, jeb->dirty_size, jeb->used_size); jeb->gc_node = raw; spin_unlock(&c->erase_completion_lock); @@ -260,7 +259,7 @@ int jffs2_garbage_collect_pass(struct jffs2_sb_info *c) ic = jffs2_raw_ref_to_ic(raw); /* We need to hold the inocache. Either the erase_completion_lock or - the inocache_lock are sufficient; we trade down since the inocache_lock + the inocache_lock are sufficient; we trade down since the inocache_lock causes less contention. */ spin_lock(&c->inocache_lock); @@ -279,14 +278,14 @@ int jffs2_garbage_collect_pass(struct jffs2_sb_info *c) switch(ic->state) { case INO_STATE_CHECKEDABSENT: - /* It's been checked, but it's not currently in-core. + /* It's been checked, but it's not currently in-core. We can just copy any pristine nodes, but have to prevent anyone else from doing read_inode() while we're at it, so we set the state accordingly */ if (ref_flags(raw) == REF_PRISTINE) ic->state = INO_STATE_GC; else { - D1(printk(KERN_DEBUG "Ino #%u is absent but node not REF_PRISTINE. Reading.\n", + D1(printk(KERN_DEBUG "Ino #%u is absent but node not REF_PRISTINE. Reading.\n", ic->ino)); } break; @@ -299,8 +298,8 @@ int jffs2_garbage_collect_pass(struct jffs2_sb_info *c) case INO_STATE_CHECKING: case INO_STATE_GC: /* Should never happen. We should have finished checking - by the time we actually start doing any GC, and since - we're holding the alloc_sem, no other garbage collection + by the time we actually start doing any GC, and since + we're holding the alloc_sem, no other garbage collection can happen. */ printk(KERN_CRIT "Inode #%u already in state %d in jffs2_garbage_collect_pass()!\n", @@ -320,21 +319,21 @@ int jffs2_garbage_collect_pass(struct jffs2_sb_info *c) D1(printk(KERN_DEBUG "jffs2_garbage_collect_pass() waiting for ino #%u in state %d\n", ic->ino, ic->state)); sleep_on_spinunlock(&c->inocache_wq, &c->inocache_lock); - /* And because we dropped the alloc_sem we must start again from the + /* And because we dropped the alloc_sem we must start again from the beginning. Ponder chance of livelock here -- we're returning success without actually making any progress. - Q: What are the chances that the inode is back in INO_STATE_READING + Q: What are the chances that the inode is back in INO_STATE_READING again by the time we next enter this function? And that this happens enough times to cause a real delay? - A: Small enough that I don't care :) + A: Small enough that I don't care :) */ return 0; } /* OK. Now if the inode is in state INO_STATE_GC, we are going to copy the - node intact, and we don't have to muck about with the fragtree etc. + node intact, and we don't have to muck about with the fragtree etc. because we know it's not in-core. If it _was_ in-core, we go through all the iget() crap anyway */ @@ -454,7 +453,7 @@ static int jffs2_garbage_collect_live(struct jffs2_sb_info *c, struct jffs2_era if (!ret) { /* Urgh. Return it sensibly. */ frag->node->raw = f->inocache->nodes; - } + } if (ret != -EBADFD) goto upnout; } @@ -468,7 +467,7 @@ static int jffs2_garbage_collect_live(struct jffs2_sb_info *c, struct jffs2_era } goto upnout; } - + /* Wasn't a dnode. Try dirent */ for (fd = f->dents; fd; fd=fd->next) { if (fd->raw == raw) @@ -485,7 +484,8 @@ static int jffs2_garbage_collect_live(struct jffs2_sb_info *c, struct jffs2_era if (ref_obsolete(raw)) { printk(KERN_WARNING "But it's obsolete so we don't mind too much\n"); } else { - ret = -EIO; + jffs2_dbg_dump_node(c, ref_offset(raw)); + BUG(); } } upnout: @@ -494,7 +494,7 @@ static int jffs2_garbage_collect_live(struct jffs2_sb_info *c, struct jffs2_era return ret; } -static int jffs2_garbage_collect_pristine(struct jffs2_sb_info *c, +static int jffs2_garbage_collect_pristine(struct jffs2_sb_info *c, struct jffs2_inode_cache *ic, struct jffs2_raw_node_ref *raw) { @@ -513,8 +513,11 @@ static int jffs2_garbage_collect_pristine(struct jffs2_sb_info *c, /* Ask for a small amount of space (or the totlen if smaller) because we don't want to force wastage of the end of a block if splitting would work. */ - ret = jffs2_reserve_space_gc(c, min_t(uint32_t, sizeof(struct jffs2_raw_inode) + JFFS2_MIN_DATA_LEN, - rawlen), &phys_ofs, &alloclen); + ret = jffs2_reserve_space_gc(c, min_t(uint32_t, sizeof(struct jffs2_raw_inode) + + JFFS2_MIN_DATA_LEN, rawlen), &phys_ofs, &alloclen, rawlen); + /* this is not the exact summary size of it, + it is only an upper estimation */ + if (ret) return ret; @@ -577,7 +580,7 @@ static int jffs2_garbage_collect_pristine(struct jffs2_sb_info *c, } break; default: - printk(KERN_WARNING "Unknown node type for REF_PRISTINE node at 0x%08x: 0x%04x\n", + printk(KERN_WARNING "Unknown node type for REF_PRISTINE node at 0x%08x: 0x%04x\n", ref_offset(raw), je16_to_cpu(node->u.nodetype)); goto bail; } @@ -618,17 +621,19 @@ static int jffs2_garbage_collect_pristine(struct jffs2_sb_info *c, retried = 1; D1(printk(KERN_DEBUG "Retrying failed write of REF_PRISTINE node.\n")); - - ACCT_SANITY_CHECK(c,jeb); - D1(ACCT_PARANOIA_CHECK(jeb)); - ret = jffs2_reserve_space_gc(c, rawlen, &phys_ofs, &dummy); + jffs2_dbg_acct_sanity_check(c,jeb); + jffs2_dbg_acct_paranoia_check(c, jeb); + + ret = jffs2_reserve_space_gc(c, rawlen, &phys_ofs, &dummy, rawlen); + /* this is not the exact summary size of it, + it is only an upper estimation */ if (!ret) { D1(printk(KERN_DEBUG "Allocated space at 0x%08x to retry failed write.\n", phys_ofs)); - ACCT_SANITY_CHECK(c,jeb); - D1(ACCT_PARANOIA_CHECK(jeb)); + jffs2_dbg_acct_sanity_check(c,jeb); + jffs2_dbg_acct_paranoia_check(c, jeb); goto retry; } @@ -664,7 +669,7 @@ static int jffs2_garbage_collect_pristine(struct jffs2_sb_info *c, goto out_node; } -static int jffs2_garbage_collect_metadata(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb, +static int jffs2_garbage_collect_metadata(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb, struct jffs2_inode_info *f, struct jffs2_full_dnode *fn) { struct jffs2_full_dnode *new_fn; @@ -679,7 +684,7 @@ static int jffs2_garbage_collect_metadata(struct jffs2_sb_info *c, struct jffs2_ S_ISCHR(JFFS2_F_I_MODE(f)) ) { /* For these, we don't actually need to read the old node */ /* FIXME: for minor or major > 255. */ - dev = cpu_to_je16(((JFFS2_F_I_RDEV_MAJ(f) << 8) | + dev = cpu_to_je16(((JFFS2_F_I_RDEV_MAJ(f) << 8) | JFFS2_F_I_RDEV_MIN(f))); mdata = (char *)&dev; mdatalen = sizeof(dev); @@ -700,14 +705,15 @@ static int jffs2_garbage_collect_metadata(struct jffs2_sb_info *c, struct jffs2_ D1(printk(KERN_DEBUG "jffs2_garbage_collect_metadata(): Writing %d bites of symlink target\n", mdatalen)); } - - ret = jffs2_reserve_space_gc(c, sizeof(ri) + mdatalen, &phys_ofs, &alloclen); + + ret = jffs2_reserve_space_gc(c, sizeof(ri) + mdatalen, &phys_ofs, &alloclen, + JFFS2_SUMMARY_INODE_SIZE); if (ret) { printk(KERN_WARNING "jffs2_reserve_space_gc of %zd bytes for garbage_collect_metadata failed: %d\n", sizeof(ri)+ mdatalen, ret); goto out; } - + last_frag = frag_last(&f->fragtree); if (last_frag) /* Fetch the inode length from the fragtree rather then @@ -715,7 +721,7 @@ static int jffs2_garbage_collect_metadata(struct jffs2_sb_info *c, struct jffs2_ ilen = last_frag->ofs + last_frag->size; else ilen = JFFS2_F_I_SIZE(f); - + memset(&ri, 0, sizeof(ri)); ri.magic = cpu_to_je16(JFFS2_MAGIC_BITMASK); ri.nodetype = cpu_to_je16(JFFS2_NODETYPE_INODE); @@ -754,7 +760,7 @@ static int jffs2_garbage_collect_metadata(struct jffs2_sb_info *c, struct jffs2_ return ret; } -static int jffs2_garbage_collect_dirent(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb, +static int jffs2_garbage_collect_dirent(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb, struct jffs2_inode_info *f, struct jffs2_full_dirent *fd) { struct jffs2_full_dirent *new_fd; @@ -771,12 +777,18 @@ static int jffs2_garbage_collect_dirent(struct jffs2_sb_info *c, struct jffs2_er rd.pino = cpu_to_je32(f->inocache->ino); rd.version = cpu_to_je32(++f->highest_version); rd.ino = cpu_to_je32(fd->ino); - rd.mctime = cpu_to_je32(max(JFFS2_F_I_MTIME(f), JFFS2_F_I_CTIME(f))); + /* If the times on this inode were set by explicit utime() they can be different, + so refrain from splatting them. */ + if (JFFS2_F_I_MTIME(f) == JFFS2_F_I_CTIME(f)) + rd.mctime = cpu_to_je32(JFFS2_F_I_MTIME(f)); + else + rd.mctime = cpu_to_je32(0); rd.type = fd->type; rd.node_crc = cpu_to_je32(crc32(0, &rd, sizeof(rd)-8)); rd.name_crc = cpu_to_je32(crc32(0, fd->name, rd.nsize)); - - ret = jffs2_reserve_space_gc(c, sizeof(rd)+rd.nsize, &phys_ofs, &alloclen); + + ret = jffs2_reserve_space_gc(c, sizeof(rd)+rd.nsize, &phys_ofs, &alloclen, + JFFS2_SUMMARY_DIRENT_SIZE(rd.nsize)); if (ret) { printk(KERN_WARNING "jffs2_reserve_space_gc of %zd bytes for garbage_collect_dirent failed: %d\n", sizeof(rd)+rd.nsize, ret); @@ -792,7 +804,7 @@ static int jffs2_garbage_collect_dirent(struct jffs2_sb_info *c, struct jffs2_er return 0; } -static int jffs2_garbage_collect_deletion_dirent(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb, +static int jffs2_garbage_collect_deletion_dirent(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb, struct jffs2_inode_info *f, struct jffs2_full_dirent *fd) { struct jffs2_full_dirent **fdp = &f->dents; @@ -831,7 +843,7 @@ static int jffs2_garbage_collect_deletion_dirent(struct jffs2_sb_info *c, struct if (ref_totlen(c, NULL, raw) != rawlen) continue; - /* Doesn't matter if there's one in the same erase block. We're going to + /* Doesn't matter if there's one in the same erase block. We're going to delete it too at the same time. */ if (SECTOR_ADDR(raw->flash_offset) == SECTOR_ADDR(fd->raw->flash_offset)) continue; @@ -883,6 +895,9 @@ static int jffs2_garbage_collect_deletion_dirent(struct jffs2_sb_info *c, struct kfree(rd); } + /* FIXME: If we're deleting a dirent which contains the current mtime and ctime, + we should update the metadata node with those times accordingly */ + /* No need for it any more. Just mark it obsolete and remove it from the list */ while (*fdp) { if ((*fdp) == fd) { @@ -912,13 +927,13 @@ static int jffs2_garbage_collect_hole(struct jffs2_sb_info *c, struct jffs2_eras D1(printk(KERN_DEBUG "Writing replacement hole node for ino #%u from offset 0x%x to 0x%x\n", f->inocache->ino, start, end)); - + memset(&ri, 0, sizeof(ri)); if(fn->frags > 1) { size_t readlen; uint32_t crc; - /* It's partially obsoleted by a later write. So we have to + /* It's partially obsoleted by a later write. So we have to write it out again with the _same_ version as before */ ret = jffs2_flash_read(c, ref_offset(fn->raw), sizeof(ri), &readlen, (char *)&ri); if (readlen != sizeof(ri) || ret) { @@ -940,16 +955,16 @@ static int jffs2_garbage_collect_hole(struct jffs2_sb_info *c, struct jffs2_eras crc = crc32(0, &ri, sizeof(ri)-8); if (crc != je32_to_cpu(ri.node_crc)) { printk(KERN_WARNING "jffs2_garbage_collect_hole: Node at 0x%08x had CRC 0x%08x which doesn't match calculated CRC 0x%08x\n", - ref_offset(fn->raw), + ref_offset(fn->raw), je32_to_cpu(ri.node_crc), crc); /* FIXME: We could possibly deal with this by writing new holes for each frag */ - printk(KERN_WARNING "Data in the range 0x%08x to 0x%08x of inode #%u will be lost\n", + printk(KERN_WARNING "Data in the range 0x%08x to 0x%08x of inode #%u will be lost\n", start, end, f->inocache->ino); goto fill; } if (ri.compr != JFFS2_COMPR_ZERO) { printk(KERN_WARNING "jffs2_garbage_collect_hole: Node 0x%08x wasn't a hole node!\n", ref_offset(fn->raw)); - printk(KERN_WARNING "Data in the range 0x%08x to 0x%08x of inode #%u will be lost\n", + printk(KERN_WARNING "Data in the range 0x%08x to 0x%08x of inode #%u will be lost\n", start, end, f->inocache->ino); goto fill; } @@ -967,7 +982,7 @@ static int jffs2_garbage_collect_hole(struct jffs2_sb_info *c, struct jffs2_eras ri.csize = cpu_to_je32(0); ri.compr = JFFS2_COMPR_ZERO; } - + frag = frag_last(&f->fragtree); if (frag) /* Fetch the inode length from the fragtree rather then @@ -986,7 +1001,8 @@ static int jffs2_garbage_collect_hole(struct jffs2_sb_info *c, struct jffs2_eras ri.data_crc = cpu_to_je32(0); ri.node_crc = cpu_to_je32(crc32(0, &ri, sizeof(ri)-8)); - ret = jffs2_reserve_space_gc(c, sizeof(ri), &phys_ofs, &alloclen); + ret = jffs2_reserve_space_gc(c, sizeof(ri), &phys_ofs, &alloclen, + JFFS2_SUMMARY_INODE_SIZE); if (ret) { printk(KERN_WARNING "jffs2_reserve_space_gc of %zd bytes for garbage_collect_hole failed: %d\n", sizeof(ri), ret); @@ -1008,10 +1024,10 @@ static int jffs2_garbage_collect_hole(struct jffs2_sb_info *c, struct jffs2_eras return 0; } - /* + /* * We should only get here in the case where the node we are * replacing had more than one frag, so we kept the same version - * number as before. (Except in case of error -- see 'goto fill;' + * number as before. (Except in case of error -- see 'goto fill;' * above.) */ D1(if(unlikely(fn->frags <= 1)) { @@ -1023,7 +1039,7 @@ static int jffs2_garbage_collect_hole(struct jffs2_sb_info *c, struct jffs2_eras /* This is a partially-overlapped hole node. Mark it REF_NORMAL not REF_PRISTINE */ mark_ref_normal(new_fn->raw); - for (frag = jffs2_lookup_node_frag(&f->fragtree, fn->ofs); + for (frag = jffs2_lookup_node_frag(&f->fragtree, fn->ofs); frag; frag = frag_next(frag)) { if (frag->ofs > fn->size + fn->ofs) break; @@ -1041,10 +1057,10 @@ static int jffs2_garbage_collect_hole(struct jffs2_sb_info *c, struct jffs2_eras printk(KERN_WARNING "jffs2_garbage_collect_hole: New node has no frags!\n"); BUG(); } - + jffs2_mark_node_obsolete(c, fn->raw); jffs2_free_full_dnode(fn); - + return 0; } @@ -1054,12 +1070,12 @@ static int jffs2_garbage_collect_dnode(struct jffs2_sb_info *c, struct jffs2_era { struct jffs2_full_dnode *new_fn; struct jffs2_raw_inode ri; - uint32_t alloclen, phys_ofs, offset, orig_end, orig_start; + uint32_t alloclen, phys_ofs, offset, orig_end, orig_start; int ret = 0; unsigned char *comprbuf = NULL, *writebuf; unsigned long pg; unsigned char *pg_ptr; - + memset(&ri, 0, sizeof(ri)); D1(printk(KERN_DEBUG "Writing replacement dnode for ino #%u from offset 0x%x to 0x%x\n", @@ -1071,8 +1087,8 @@ static int jffs2_garbage_collect_dnode(struct jffs2_sb_info *c, struct jffs2_era if (c->nr_free_blocks + c->nr_erasing_blocks > c->resv_blocks_gcmerge) { /* Attempt to do some merging. But only expand to cover logically adjacent frags if the block containing them is already considered - to be dirty. Otherwise we end up with GC just going round in - circles dirtying the nodes it already wrote out, especially + to be dirty. Otherwise we end up with GC just going round in + circles dirtying the nodes it already wrote out, especially on NAND where we have small eraseblocks and hence a much higher chance of nodes having to be split to cross boundaries. */ @@ -1106,7 +1122,7 @@ static int jffs2_garbage_collect_dnode(struct jffs2_sb_info *c, struct jffs2_era break; } else { - /* OK, it's a frag which extends to the beginning of the page. Does it live + /* OK, it's a frag which extends to the beginning of the page. Does it live in a block which is still considered clean? If so, don't obsolete it. If not, cover it anyway. */ @@ -1156,7 +1172,7 @@ static int jffs2_garbage_collect_dnode(struct jffs2_sb_info *c, struct jffs2_era break; } else { - /* OK, it's a frag which extends to the beginning of the page. Does it live + /* OK, it's a frag which extends to the beginning of the page. Does it live in a block which is still considered clean? If so, don't obsolete it. If not, cover it anyway. */ @@ -1183,14 +1199,14 @@ static int jffs2_garbage_collect_dnode(struct jffs2_sb_info *c, struct jffs2_era break; } } - D1(printk(KERN_DEBUG "Expanded dnode to write from (0x%x-0x%x) to (0x%x-0x%x)\n", + D1(printk(KERN_DEBUG "Expanded dnode to write from (0x%x-0x%x) to (0x%x-0x%x)\n", orig_start, orig_end, start, end)); D1(BUG_ON(end > frag_last(&f->fragtree)->ofs + frag_last(&f->fragtree)->size)); BUG_ON(end < orig_end); BUG_ON(start > orig_start); } - + /* First, use readpage() to read the appropriate page into the page cache */ /* Q: What happens if we actually try to GC the _same_ page for which commit_write() * triggered garbage collection in the first place? @@ -1211,7 +1227,8 @@ static int jffs2_garbage_collect_dnode(struct jffs2_sb_info *c, struct jffs2_era uint32_t cdatalen; uint16_t comprtype = JFFS2_COMPR_NONE; - ret = jffs2_reserve_space_gc(c, sizeof(ri) + JFFS2_MIN_DATA_LEN, &phys_ofs, &alloclen); + ret = jffs2_reserve_space_gc(c, sizeof(ri) + JFFS2_MIN_DATA_LEN, &phys_ofs, + &alloclen, JFFS2_SUMMARY_INODE_SIZE); if (ret) { printk(KERN_WARNING "jffs2_reserve_space_gc of %zd bytes for garbage_collect_dnode failed: %d\n", @@ -1246,7 +1263,7 @@ static int jffs2_garbage_collect_dnode(struct jffs2_sb_info *c, struct jffs2_era ri.usercompr = (comprtype >> 8) & 0xff; ri.node_crc = cpu_to_je32(crc32(0, &ri, sizeof(ri)-8)); ri.data_crc = cpu_to_je32(crc32(0, comprbuf, cdatalen)); - + new_fn = jffs2_write_dnode(c, f, &ri, comprbuf, cdatalen, phys_ofs, ALLOC_GC); jffs2_free_comprbuf(comprbuf, writebuf); @@ -1268,4 +1285,3 @@ static int jffs2_garbage_collect_dnode(struct jffs2_sb_info *c, struct jffs2_era jffs2_gc_release_page(c, pg_ptr, &pg); return ret; } - diff --git a/fs/jffs2/histo.h b/fs/jffs2/histo.h index 84f184f0836f..22a93a08210c 100644 --- a/fs/jffs2/histo.h +++ b/fs/jffs2/histo.h @@ -1,3 +1,3 @@ /* This file provides the bit-probabilities for the input file */ -#define BIT_DIVIDER 629 +#define BIT_DIVIDER 629 static int bits[9] = { 179,167,183,165,159,198,178,119,}; /* ia32 .so files */ diff --git a/fs/jffs2/histo_mips.h b/fs/jffs2/histo_mips.h index 9a443268d885..fa3dac19a109 100644 --- a/fs/jffs2/histo_mips.h +++ b/fs/jffs2/histo_mips.h @@ -1,2 +1,2 @@ -#define BIT_DIVIDER_MIPS 1043 +#define BIT_DIVIDER_MIPS 1043 static int bits_mips[8] = { 277,249,290,267,229,341,212,241}; /* mips32 */ diff --git a/fs/jffs2/ioctl.c b/fs/jffs2/ioctl.c index 238c7992064c..69099835de1c 100644 --- a/fs/jffs2/ioctl.c +++ b/fs/jffs2/ioctl.c @@ -7,17 +7,17 @@ * * For licensing information, see the file 'LICENCE' in this directory. * - * $Id: ioctl.c,v 1.9 2004/11/16 20:36:11 dwmw2 Exp $ + * $Id: ioctl.c,v 1.10 2005/11/07 11:14:40 gleixner Exp $ * */ #include <linux/fs.h> -int jffs2_ioctl(struct inode *inode, struct file *filp, unsigned int cmd, +int jffs2_ioctl(struct inode *inode, struct file *filp, unsigned int cmd, unsigned long arg) { /* Later, this will provide for lsattr.jffs2 and chattr.jffs2, which will include compression support etc. */ return -ENOTTY; } - + diff --git a/fs/jffs2/malloc.c b/fs/jffs2/malloc.c index 5abb431c2a00..036cbd11c004 100644 --- a/fs/jffs2/malloc.c +++ b/fs/jffs2/malloc.c @@ -7,7 +7,7 @@ * * For licensing information, see the file 'LICENCE' in this directory. * - * $Id: malloc.c,v 1.28 2004/11/16 20:36:11 dwmw2 Exp $ + * $Id: malloc.c,v 1.31 2005/11/07 11:14:40 gleixner Exp $ * */ @@ -17,15 +17,6 @@ #include <linux/jffs2.h> #include "nodelist.h" -#if 0 -#define JFFS2_SLAB_POISON SLAB_POISON -#else -#define JFFS2_SLAB_POISON 0 -#endif - -// replace this by #define D3 (x) x for cache debugging -#define D3(x) - /* These are initialised to NULL in the kernel startup code. If you're porting to other operating systems, beware */ static kmem_cache_t *full_dnode_slab; @@ -38,45 +29,45 @@ static kmem_cache_t *inode_cache_slab; int __init jffs2_create_slab_caches(void) { - full_dnode_slab = kmem_cache_create("jffs2_full_dnode", + full_dnode_slab = kmem_cache_create("jffs2_full_dnode", sizeof(struct jffs2_full_dnode), - 0, JFFS2_SLAB_POISON, NULL, NULL); + 0, 0, NULL, NULL); if (!full_dnode_slab) goto err; raw_dirent_slab = kmem_cache_create("jffs2_raw_dirent", sizeof(struct jffs2_raw_dirent), - 0, JFFS2_SLAB_POISON, NULL, NULL); + 0, 0, NULL, NULL); if (!raw_dirent_slab) goto err; raw_inode_slab = kmem_cache_create("jffs2_raw_inode", sizeof(struct jffs2_raw_inode), - 0, JFFS2_SLAB_POISON, NULL, NULL); + 0, 0, NULL, NULL); if (!raw_inode_slab) goto err; tmp_dnode_info_slab = kmem_cache_create("jffs2_tmp_dnode", sizeof(struct jffs2_tmp_dnode_info), - 0, JFFS2_SLAB_POISON, NULL, NULL); + 0, 0, NULL, NULL); if (!tmp_dnode_info_slab) goto err; raw_node_ref_slab = kmem_cache_create("jffs2_raw_node_ref", sizeof(struct jffs2_raw_node_ref), - 0, JFFS2_SLAB_POISON, NULL, NULL); + 0, 0, NULL, NULL); if (!raw_node_ref_slab) goto err; node_frag_slab = kmem_cache_create("jffs2_node_frag", sizeof(struct jffs2_node_frag), - 0, JFFS2_SLAB_POISON, NULL, NULL); + 0, 0, NULL, NULL); if (!node_frag_slab) goto err; inode_cache_slab = kmem_cache_create("jffs2_inode_cache", sizeof(struct jffs2_inode_cache), - 0, JFFS2_SLAB_POISON, NULL, NULL); + 0, 0, NULL, NULL); if (inode_cache_slab) return 0; err: @@ -104,102 +95,113 @@ void jffs2_destroy_slab_caches(void) struct jffs2_full_dirent *jffs2_alloc_full_dirent(int namesize) { - return kmalloc(sizeof(struct jffs2_full_dirent) + namesize, GFP_KERNEL); + struct jffs2_full_dirent *ret; + ret = kmalloc(sizeof(struct jffs2_full_dirent) + namesize, GFP_KERNEL); + dbg_memalloc("%p\n", ret); + return ret; } void jffs2_free_full_dirent(struct jffs2_full_dirent *x) { + dbg_memalloc("%p\n", x); kfree(x); } struct jffs2_full_dnode *jffs2_alloc_full_dnode(void) { - struct jffs2_full_dnode *ret = kmem_cache_alloc(full_dnode_slab, GFP_KERNEL); - D3 (printk (KERN_DEBUG "alloc_full_dnode at %p\n", ret)); + struct jffs2_full_dnode *ret; + ret = kmem_cache_alloc(full_dnode_slab, GFP_KERNEL); + dbg_memalloc("%p\n", ret); return ret; } void jffs2_free_full_dnode(struct jffs2_full_dnode *x) { - D3 (printk (KERN_DEBUG "free full_dnode at %p\n", x)); + dbg_memalloc("%p\n", x); kmem_cache_free(full_dnode_slab, x); } struct jffs2_raw_dirent *jffs2_alloc_raw_dirent(void) { - struct jffs2_raw_dirent *ret = kmem_cache_alloc(raw_dirent_slab, GFP_KERNEL); - D3 (printk (KERN_DEBUG "alloc_raw_dirent\n", ret)); + struct jffs2_raw_dirent *ret; + ret = kmem_cache_alloc(raw_dirent_slab, GFP_KERNEL); + dbg_memalloc("%p\n", ret); return ret; } void jffs2_free_raw_dirent(struct jffs2_raw_dirent *x) { - D3 (printk (KERN_DEBUG "free_raw_dirent at %p\n", x)); + dbg_memalloc("%p\n", x); kmem_cache_free(raw_dirent_slab, x); } struct jffs2_raw_inode *jffs2_alloc_raw_inode(void) { - struct jffs2_raw_inode *ret = kmem_cache_alloc(raw_inode_slab, GFP_KERNEL); - D3 (printk (KERN_DEBUG "alloc_raw_inode at %p\n", ret)); + struct jffs2_raw_inode *ret; + ret = kmem_cache_alloc(raw_inode_slab, GFP_KERNEL); + dbg_memalloc("%p\n", ret); return ret; } void jffs2_free_raw_inode(struct jffs2_raw_inode *x) { - D3 (printk (KERN_DEBUG "free_raw_inode at %p\n", x)); + dbg_memalloc("%p\n", x); kmem_cache_free(raw_inode_slab, x); } struct jffs2_tmp_dnode_info *jffs2_alloc_tmp_dnode_info(void) { - struct jffs2_tmp_dnode_info *ret = kmem_cache_alloc(tmp_dnode_info_slab, GFP_KERNEL); - D3 (printk (KERN_DEBUG "alloc_tmp_dnode_info at %p\n", ret)); + struct jffs2_tmp_dnode_info *ret; + ret = kmem_cache_alloc(tmp_dnode_info_slab, GFP_KERNEL); + dbg_memalloc("%p\n", + ret); return ret; } void jffs2_free_tmp_dnode_info(struct jffs2_tmp_dnode_info *x) { - D3 (printk (KERN_DEBUG "free_tmp_dnode_info at %p\n", x)); + dbg_memalloc("%p\n", x); kmem_cache_free(tmp_dnode_info_slab, x); } struct jffs2_raw_node_ref *jffs2_alloc_raw_node_ref(void) { - struct jffs2_raw_node_ref *ret = kmem_cache_alloc(raw_node_ref_slab, GFP_KERNEL); - D3 (printk (KERN_DEBUG "alloc_raw_node_ref at %p\n", ret)); + struct jffs2_raw_node_ref *ret; + ret = kmem_cache_alloc(raw_node_ref_slab, GFP_KERNEL); + dbg_memalloc("%p\n", ret); return ret; } void jffs2_free_raw_node_ref(struct jffs2_raw_node_ref *x) { - D3 (printk (KERN_DEBUG "free_raw_node_ref at %p\n", x)); + dbg_memalloc("%p\n", x); kmem_cache_free(raw_node_ref_slab, x); } struct jffs2_node_frag *jffs2_alloc_node_frag(void) { - struct jffs2_node_frag *ret = kmem_cache_alloc(node_frag_slab, GFP_KERNEL); - D3 (printk (KERN_DEBUG "alloc_node_frag at %p\n", ret)); + struct jffs2_node_frag *ret; + ret = kmem_cache_alloc(node_frag_slab, GFP_KERNEL); + dbg_memalloc("%p\n", ret); return ret; } void jffs2_free_node_frag(struct jffs2_node_frag *x) { - D3 (printk (KERN_DEBUG "free_node_frag at %p\n", x)); + dbg_memalloc("%p\n", x); kmem_cache_free(node_frag_slab, x); } struct jffs2_inode_cache *jffs2_alloc_inode_cache(void) { - struct jffs2_inode_cache *ret = kmem_cache_alloc(inode_cache_slab, GFP_KERNEL); - D3 (printk(KERN_DEBUG "Allocated inocache at %p\n", ret)); + struct jffs2_inode_cache *ret; + ret = kmem_cache_alloc(inode_cache_slab, GFP_KERNEL); + dbg_memalloc("%p\n", ret); return ret; } void jffs2_free_inode_cache(struct jffs2_inode_cache *x) { - D3 (printk(KERN_DEBUG "Freeing inocache at %p\n", x)); + dbg_memalloc("%p\n", x); kmem_cache_free(inode_cache_slab, x); } - diff --git a/fs/jffs2/nodelist.c b/fs/jffs2/nodelist.c index 4991c348f6ec..c79eebb8ab32 100644 --- a/fs/jffs2/nodelist.c +++ b/fs/jffs2/nodelist.c @@ -7,7 +7,7 @@ * * For licensing information, see the file 'LICENCE' in this directory. * - * $Id: nodelist.c,v 1.98 2005/07/10 15:15:32 dedekind Exp $ + * $Id: nodelist.c,v 1.115 2005/11/07 11:14:40 gleixner Exp $ * */ @@ -24,469 +24,832 @@ void jffs2_add_fd_to_list(struct jffs2_sb_info *c, struct jffs2_full_dirent *new, struct jffs2_full_dirent **list) { struct jffs2_full_dirent **prev = list; - D1(printk(KERN_DEBUG "jffs2_add_fd_to_list( %p, %p (->%p))\n", new, list, *list)); + + dbg_dentlist("add dirent \"%s\", ino #%u\n", new->name, new->ino); while ((*prev) && (*prev)->nhash <= new->nhash) { if ((*prev)->nhash == new->nhash && !strcmp((*prev)->name, new->name)) { /* Duplicate. Free one */ if (new->version < (*prev)->version) { - D1(printk(KERN_DEBUG "Eep! Marking new dirent node obsolete\n")); - D1(printk(KERN_DEBUG "New dirent is \"%s\"->ino #%u. Old is \"%s\"->ino #%u\n", new->name, new->ino, (*prev)->name, (*prev)->ino)); + dbg_dentlist("Eep! Marking new dirent node is obsolete, old is \"%s\", ino #%u\n", + (*prev)->name, (*prev)->ino); jffs2_mark_node_obsolete(c, new->raw); jffs2_free_full_dirent(new); } else { - D1(printk(KERN_DEBUG "Marking old dirent node (ino #%u) obsolete\n", (*prev)->ino)); + dbg_dentlist("marking old dirent \"%s\", ino #%u bsolete\n", + (*prev)->name, (*prev)->ino); new->next = (*prev)->next; jffs2_mark_node_obsolete(c, ((*prev)->raw)); jffs2_free_full_dirent(*prev); *prev = new; } - goto out; + return; } prev = &((*prev)->next); } new->next = *prev; *prev = new; +} + +void jffs2_truncate_fragtree(struct jffs2_sb_info *c, struct rb_root *list, uint32_t size) +{ + struct jffs2_node_frag *frag = jffs2_lookup_node_frag(list, size); + + dbg_fragtree("truncating fragtree to 0x%08x bytes\n", size); + + /* We know frag->ofs <= size. That's what lookup does for us */ + if (frag && frag->ofs != size) { + if (frag->ofs+frag->size > size) { + frag->size = size - frag->ofs; + } + frag = frag_next(frag); + } + while (frag && frag->ofs >= size) { + struct jffs2_node_frag *next = frag_next(frag); + + frag_erase(frag, list); + jffs2_obsolete_node_frag(c, frag); + frag = next; + } - out: - D2(while(*list) { - printk(KERN_DEBUG "Dirent \"%s\" (hash 0x%08x, ino #%u\n", (*list)->name, (*list)->nhash, (*list)->ino); - list = &(*list)->next; - }); + if (size == 0) + return; + + /* + * If the last fragment starts at the RAM page boundary, it is + * REF_PRISTINE irrespective of its size. + */ + frag = frag_last(list); + if (frag->node && (frag->ofs & (PAGE_CACHE_SIZE - 1)) == 0) { + dbg_fragtree2("marking the last fragment 0x%08x-0x%08x REF_PRISTINE.\n", + frag->ofs, frag->ofs + frag->size); + frag->node->raw->flash_offset = ref_offset(frag->node->raw) | REF_PRISTINE; + } } -/* - * Put a new tmp_dnode_info into the temporaty RB-tree, keeping the list in - * order of increasing version. - */ -static void jffs2_add_tn_to_tree(struct jffs2_tmp_dnode_info *tn, struct rb_root *list) +void jffs2_obsolete_node_frag(struct jffs2_sb_info *c, struct jffs2_node_frag *this) { - struct rb_node **p = &list->rb_node; - struct rb_node * parent = NULL; - struct jffs2_tmp_dnode_info *this; - - while (*p) { - parent = *p; - this = rb_entry(parent, struct jffs2_tmp_dnode_info, rb); - - /* There may actually be a collision here, but it doesn't - actually matter. As long as the two nodes with the same - version are together, it's all fine. */ - if (tn->version < this->version) - p = &(*p)->rb_left; - else - p = &(*p)->rb_right; - } + if (this->node) { + this->node->frags--; + if (!this->node->frags) { + /* The node has no valid frags left. It's totally obsoleted */ + dbg_fragtree2("marking old node @0x%08x (0x%04x-0x%04x) obsolete\n", + ref_offset(this->node->raw), this->node->ofs, this->node->ofs+this->node->size); + jffs2_mark_node_obsolete(c, this->node->raw); + jffs2_free_full_dnode(this->node); + } else { + dbg_fragtree2("marking old node @0x%08x (0x%04x-0x%04x) REF_NORMAL. frags is %d\n", + ref_offset(this->node->raw), this->node->ofs, this->node->ofs+this->node->size, this->node->frags); + mark_ref_normal(this->node->raw); + } - rb_link_node(&tn->rb, parent, p); - rb_insert_color(&tn->rb, list); + } + jffs2_free_node_frag(this); } -static void jffs2_free_tmp_dnode_info_list(struct rb_root *list) +static void jffs2_fragtree_insert(struct jffs2_node_frag *newfrag, struct jffs2_node_frag *base) { - struct rb_node *this; - struct jffs2_tmp_dnode_info *tn; + struct rb_node *parent = &base->rb; + struct rb_node **link = &parent; - this = list->rb_node; + dbg_fragtree2("insert frag (0x%04x-0x%04x)\n", newfrag->ofs, newfrag->ofs + newfrag->size); - /* Now at bottom of tree */ - while (this) { - if (this->rb_left) - this = this->rb_left; - else if (this->rb_right) - this = this->rb_right; + while (*link) { + parent = *link; + base = rb_entry(parent, struct jffs2_node_frag, rb); + + if (newfrag->ofs > base->ofs) + link = &base->rb.rb_right; + else if (newfrag->ofs < base->ofs) + link = &base->rb.rb_left; else { - tn = rb_entry(this, struct jffs2_tmp_dnode_info, rb); - jffs2_free_full_dnode(tn->fn); - jffs2_free_tmp_dnode_info(tn); - - this = this->rb_parent; - if (!this) - break; - - if (this->rb_left == &tn->rb) - this->rb_left = NULL; - else if (this->rb_right == &tn->rb) - this->rb_right = NULL; - else BUG(); + JFFS2_ERROR("duplicate frag at %08x (%p,%p)\n", newfrag->ofs, newfrag, base); + BUG(); } } - list->rb_node = NULL; + + rb_link_node(&newfrag->rb, &base->rb, link); } -static void jffs2_free_full_dirent_list(struct jffs2_full_dirent *fd) +/* + * Allocate and initializes a new fragment. + */ +static inline struct jffs2_node_frag * new_fragment(struct jffs2_full_dnode *fn, uint32_t ofs, uint32_t size) { - struct jffs2_full_dirent *next; - - while (fd) { - next = fd->next; - jffs2_free_full_dirent(fd); - fd = next; + struct jffs2_node_frag *newfrag; + + newfrag = jffs2_alloc_node_frag(); + if (likely(newfrag)) { + newfrag->ofs = ofs; + newfrag->size = size; + newfrag->node = fn; + } else { + JFFS2_ERROR("cannot allocate a jffs2_node_frag object\n"); } + + return newfrag; } -/* Returns first valid node after 'ref'. May return 'ref' */ -static struct jffs2_raw_node_ref *jffs2_first_valid_node(struct jffs2_raw_node_ref *ref) +/* + * Called when there is no overlapping fragment exist. Inserts a hole before the new + * fragment and inserts the new fragment to the fragtree. + */ +static int no_overlapping_node(struct jffs2_sb_info *c, struct rb_root *root, + struct jffs2_node_frag *newfrag, + struct jffs2_node_frag *this, uint32_t lastend) { - while (ref && ref->next_in_ino) { - if (!ref_obsolete(ref)) - return ref; - D1(printk(KERN_DEBUG "node at 0x%08x is obsoleted. Ignoring.\n", ref_offset(ref))); - ref = ref->next_in_ino; + if (lastend < newfrag->node->ofs) { + /* put a hole in before the new fragment */ + struct jffs2_node_frag *holefrag; + + holefrag= new_fragment(NULL, lastend, newfrag->node->ofs - lastend); + if (unlikely(!holefrag)) { + jffs2_free_node_frag(newfrag); + return -ENOMEM; + } + + if (this) { + /* By definition, the 'this' node has no right-hand child, + because there are no frags with offset greater than it. + So that's where we want to put the hole */ + dbg_fragtree2("add hole frag %#04x-%#04x on the right of the new frag.\n", + holefrag->ofs, holefrag->ofs + holefrag->size); + rb_link_node(&holefrag->rb, &this->rb, &this->rb.rb_right); + } else { + dbg_fragtree2("Add hole frag %#04x-%#04x to the root of the tree.\n", + holefrag->ofs, holefrag->ofs + holefrag->size); + rb_link_node(&holefrag->rb, NULL, &root->rb_node); + } + rb_insert_color(&holefrag->rb, root); + this = holefrag; + } + + if (this) { + /* By definition, the 'this' node has no right-hand child, + because there are no frags with offset greater than it. + So that's where we want to put new fragment */ + dbg_fragtree2("add the new node at the right\n"); + rb_link_node(&newfrag->rb, &this->rb, &this->rb.rb_right); + } else { + dbg_fragtree2("insert the new node at the root of the tree\n"); + rb_link_node(&newfrag->rb, NULL, &root->rb_node); } - return NULL; + rb_insert_color(&newfrag->rb, root); + + return 0; } -/* Get tmp_dnode_info and full_dirent for all non-obsolete nodes associated - with this ino, returning the former in order of version */ +/* Doesn't set inode->i_size */ +static int jffs2_add_frag_to_fragtree(struct jffs2_sb_info *c, struct rb_root *root, struct jffs2_node_frag *newfrag) +{ + struct jffs2_node_frag *this; + uint32_t lastend; + + /* Skip all the nodes which are completed before this one starts */ + this = jffs2_lookup_node_frag(root, newfrag->node->ofs); + + if (this) { + dbg_fragtree2("lookup gave frag 0x%04x-0x%04x; phys 0x%08x (*%p)\n", + this->ofs, this->ofs+this->size, this->node?(ref_offset(this->node->raw)):0xffffffff, this); + lastend = this->ofs + this->size; + } else { + dbg_fragtree2("lookup gave no frag\n"); + lastend = 0; + } + + /* See if we ran off the end of the fragtree */ + if (lastend <= newfrag->ofs) { + /* We did */ + + /* Check if 'this' node was on the same page as the new node. + If so, both 'this' and the new node get marked REF_NORMAL so + the GC can take a look. + */ + if (lastend && (lastend-1) >> PAGE_CACHE_SHIFT == newfrag->ofs >> PAGE_CACHE_SHIFT) { + if (this->node) + mark_ref_normal(this->node->raw); + mark_ref_normal(newfrag->node->raw); + } + + return no_overlapping_node(c, root, newfrag, this, lastend); + } -int jffs2_get_inode_nodes(struct jffs2_sb_info *c, struct jffs2_inode_info *f, - struct rb_root *tnp, struct jffs2_full_dirent **fdp, - uint32_t *highest_version, uint32_t *latest_mctime, - uint32_t *mctime_ver) + if (this->node) + dbg_fragtree2("dealing with frag %u-%u, phys %#08x(%d).\n", + this->ofs, this->ofs + this->size, + ref_offset(this->node->raw), ref_flags(this->node->raw)); + else + dbg_fragtree2("dealing with hole frag %u-%u.\n", + this->ofs, this->ofs + this->size); + + /* OK. 'this' is pointing at the first frag that newfrag->ofs at least partially obsoletes, + * - i.e. newfrag->ofs < this->ofs+this->size && newfrag->ofs >= this->ofs + */ + if (newfrag->ofs > this->ofs) { + /* This node isn't completely obsoleted. The start of it remains valid */ + + /* Mark the new node and the partially covered node REF_NORMAL -- let + the GC take a look at them */ + mark_ref_normal(newfrag->node->raw); + if (this->node) + mark_ref_normal(this->node->raw); + + if (this->ofs + this->size > newfrag->ofs + newfrag->size) { + /* The new node splits 'this' frag into two */ + struct jffs2_node_frag *newfrag2; + + if (this->node) + dbg_fragtree2("split old frag 0x%04x-0x%04x, phys 0x%08x\n", + this->ofs, this->ofs+this->size, ref_offset(this->node->raw)); + else + dbg_fragtree2("split old hole frag 0x%04x-0x%04x\n", + this->ofs, this->ofs+this->size); + + /* New second frag pointing to this's node */ + newfrag2 = new_fragment(this->node, newfrag->ofs + newfrag->size, + this->ofs + this->size - newfrag->ofs - newfrag->size); + if (unlikely(!newfrag2)) + return -ENOMEM; + if (this->node) + this->node->frags++; + + /* Adjust size of original 'this' */ + this->size = newfrag->ofs - this->ofs; + + /* Now, we know there's no node with offset + greater than this->ofs but smaller than + newfrag2->ofs or newfrag->ofs, for obvious + reasons. So we can do a tree insert from + 'this' to insert newfrag, and a tree insert + from newfrag to insert newfrag2. */ + jffs2_fragtree_insert(newfrag, this); + rb_insert_color(&newfrag->rb, root); + + jffs2_fragtree_insert(newfrag2, newfrag); + rb_insert_color(&newfrag2->rb, root); + + return 0; + } + /* New node just reduces 'this' frag in size, doesn't split it */ + this->size = newfrag->ofs - this->ofs; + + /* Again, we know it lives down here in the tree */ + jffs2_fragtree_insert(newfrag, this); + rb_insert_color(&newfrag->rb, root); + } else { + /* New frag starts at the same point as 'this' used to. Replace + it in the tree without doing a delete and insertion */ + dbg_fragtree2("inserting newfrag (*%p),%d-%d in before 'this' (*%p),%d-%d\n", + newfrag, newfrag->ofs, newfrag->ofs+newfrag->size, this, this->ofs, this->ofs+this->size); + + rb_replace_node(&this->rb, &newfrag->rb, root); + + if (newfrag->ofs + newfrag->size >= this->ofs+this->size) { + dbg_fragtree2("obsoleting node frag %p (%x-%x)\n", this, this->ofs, this->ofs+this->size); + jffs2_obsolete_node_frag(c, this); + } else { + this->ofs += newfrag->size; + this->size -= newfrag->size; + + jffs2_fragtree_insert(this, newfrag); + rb_insert_color(&this->rb, root); + return 0; + } + } + /* OK, now we have newfrag added in the correct place in the tree, but + frag_next(newfrag) may be a fragment which is overlapped by it + */ + while ((this = frag_next(newfrag)) && newfrag->ofs + newfrag->size >= this->ofs + this->size) { + /* 'this' frag is obsoleted completely. */ + dbg_fragtree2("obsoleting node frag %p (%x-%x) and removing from tree\n", + this, this->ofs, this->ofs+this->size); + rb_erase(&this->rb, root); + jffs2_obsolete_node_frag(c, this); + } + /* Now we're pointing at the first frag which isn't totally obsoleted by + the new frag */ + + if (!this || newfrag->ofs + newfrag->size == this->ofs) + return 0; + + /* Still some overlap but we don't need to move it in the tree */ + this->size = (this->ofs + this->size) - (newfrag->ofs + newfrag->size); + this->ofs = newfrag->ofs + newfrag->size; + + /* And mark them REF_NORMAL so the GC takes a look at them */ + if (this->node) + mark_ref_normal(this->node->raw); + mark_ref_normal(newfrag->node->raw); + + return 0; +} + +/* + * Given an inode, probably with existing tree of fragments, add the new node + * to the fragment tree. + */ +int jffs2_add_full_dnode_to_inode(struct jffs2_sb_info *c, struct jffs2_inode_info *f, struct jffs2_full_dnode *fn) { - struct jffs2_raw_node_ref *ref, *valid_ref; - struct jffs2_tmp_dnode_info *tn; - struct rb_root ret_tn = RB_ROOT; - struct jffs2_full_dirent *fd, *ret_fd = NULL; - union jffs2_node_union node; - size_t retlen; - int err; - - *mctime_ver = 0; - - D1(printk(KERN_DEBUG "jffs2_get_inode_nodes(): ino #%u\n", f->inocache->ino)); + int ret; + struct jffs2_node_frag *newfrag; - spin_lock(&c->erase_completion_lock); + if (unlikely(!fn->size)) + return 0; - valid_ref = jffs2_first_valid_node(f->inocache->nodes); + newfrag = new_fragment(fn, fn->ofs, fn->size); + if (unlikely(!newfrag)) + return -ENOMEM; + newfrag->node->frags = 1; - if (!valid_ref && (f->inocache->ino != 1)) - printk(KERN_WARNING "Eep. No valid nodes for ino #%u\n", f->inocache->ino); + dbg_fragtree("adding node %#04x-%#04x @0x%08x on flash, newfrag *%p\n", + fn->ofs, fn->ofs+fn->size, ref_offset(fn->raw), newfrag); - while (valid_ref) { - /* We can hold a pointer to a non-obsolete node without the spinlock, - but _obsolete_ nodes may disappear at any time, if the block - they're in gets erased. So if we mark 'ref' obsolete while we're - not holding the lock, it can go away immediately. For that reason, - we find the next valid node first, before processing 'ref'. - */ - ref = valid_ref; - valid_ref = jffs2_first_valid_node(ref->next_in_ino); - spin_unlock(&c->erase_completion_lock); + ret = jffs2_add_frag_to_fragtree(c, &f->fragtree, newfrag); + if (unlikely(ret)) + return ret; - cond_resched(); + /* If we now share a page with other nodes, mark either previous + or next node REF_NORMAL, as appropriate. */ + if (newfrag->ofs & (PAGE_CACHE_SIZE-1)) { + struct jffs2_node_frag *prev = frag_prev(newfrag); + + mark_ref_normal(fn->raw); + /* If we don't start at zero there's _always_ a previous */ + if (prev->node) + mark_ref_normal(prev->node->raw); + } + + if ((newfrag->ofs+newfrag->size) & (PAGE_CACHE_SIZE-1)) { + struct jffs2_node_frag *next = frag_next(newfrag); + + if (next) { + mark_ref_normal(fn->raw); + if (next->node) + mark_ref_normal(next->node->raw); + } + } + jffs2_dbg_fragtree_paranoia_check_nolock(f); + + return 0; +} + +/* + * Check the data CRC of the node. + * + * Returns: 0 if the data CRC is correct; + * 1 - if incorrect; + * error code if an error occured. + */ +static int check_node_data(struct jffs2_sb_info *c, struct jffs2_tmp_dnode_info *tn) +{ + struct jffs2_raw_node_ref *ref = tn->fn->raw; + int err = 0, pointed = 0; + struct jffs2_eraseblock *jeb; + unsigned char *buffer; + uint32_t crc, ofs, retlen, len; + + BUG_ON(tn->csize == 0); + + if (!jffs2_is_writebuffered(c)) + goto adj_acc; + + /* Calculate how many bytes were already checked */ + ofs = ref_offset(ref) + sizeof(struct jffs2_raw_inode); + len = ofs % c->wbuf_pagesize; + if (likely(len)) + len = c->wbuf_pagesize - len; + + if (len >= tn->csize) { + dbg_readinode("no need to check node at %#08x, data length %u, data starts at %#08x - it has already been checked.\n", + ref_offset(ref), tn->csize, ofs); + goto adj_acc; + } + + ofs += len; + len = tn->csize - len; + + dbg_readinode("check node at %#08x, data length %u, partial CRC %#08x, correct CRC %#08x, data starts at %#08x, start checking from %#08x - %u bytes.\n", + ref_offset(ref), tn->csize, tn->partial_crc, tn->data_crc, ofs - len, ofs, len); + +#ifndef __ECOS + /* TODO: instead, incapsulate point() stuff to jffs2_flash_read(), + * adding and jffs2_flash_read_end() interface. */ + if (c->mtd->point) { + err = c->mtd->point(c->mtd, ofs, len, &retlen, &buffer); + if (!err && retlen < tn->csize) { + JFFS2_WARNING("MTD point returned len too short: %u instead of %u.\n", retlen, tn->csize); + c->mtd->unpoint(c->mtd, buffer, ofs, len); + } else if (err) + JFFS2_WARNING("MTD point failed: error code %d.\n", err); + else + pointed = 1; /* succefully pointed to device */ + } +#endif + + if (!pointed) { + buffer = kmalloc(len, GFP_KERNEL); + if (unlikely(!buffer)) + return -ENOMEM; - /* FIXME: point() */ - err = jffs2_flash_read(c, (ref_offset(ref)), - min_t(uint32_t, ref_totlen(c, NULL, ref), sizeof(node)), - &retlen, (void *)&node); + /* TODO: this is very frequent pattern, make it a separate + * routine */ + err = jffs2_flash_read(c, ofs, len, &retlen, buffer); if (err) { - printk(KERN_WARNING "error %d reading node at 0x%08x in get_inode_nodes()\n", err, ref_offset(ref)); + JFFS2_ERROR("can not read %d bytes from 0x%08x, error code: %d.\n", len, ofs, err); goto free_out; } - - /* Check we've managed to read at least the common node header */ - if (retlen < min_t(uint32_t, ref_totlen(c, NULL, ref), sizeof(node.u))) { - printk(KERN_WARNING "short read in get_inode_nodes()\n"); + if (retlen != len) { + JFFS2_ERROR("short read at %#08x: %d instead of %d.\n", ofs, retlen, len); err = -EIO; goto free_out; } - - switch (je16_to_cpu(node.u.nodetype)) { - case JFFS2_NODETYPE_DIRENT: - D1(printk(KERN_DEBUG "Node at %08x (%d) is a dirent node\n", ref_offset(ref), ref_flags(ref))); - if (ref_flags(ref) == REF_UNCHECKED) { - printk(KERN_WARNING "BUG: Dirent node at 0x%08x never got checked? How?\n", ref_offset(ref)); - BUG(); - } - if (retlen < sizeof(node.d)) { - printk(KERN_WARNING "short read in get_inode_nodes()\n"); - err = -EIO; - goto free_out; - } - /* sanity check */ - if (PAD((node.d.nsize + sizeof (node.d))) != PAD(je32_to_cpu (node.d.totlen))) { - printk(KERN_NOTICE "jffs2_get_inode_nodes(): Illegal nsize in node at 0x%08x: nsize 0x%02x, totlen %04x\n", - ref_offset(ref), node.d.nsize, je32_to_cpu(node.d.totlen)); - jffs2_mark_node_obsolete(c, ref); - spin_lock(&c->erase_completion_lock); - continue; - } - if (je32_to_cpu(node.d.version) > *highest_version) - *highest_version = je32_to_cpu(node.d.version); - if (ref_obsolete(ref)) { - /* Obsoleted. This cannot happen, surely? dwmw2 20020308 */ - printk(KERN_ERR "Dirent node at 0x%08x became obsolete while we weren't looking\n", - ref_offset(ref)); - BUG(); - } - - fd = jffs2_alloc_full_dirent(node.d.nsize+1); - if (!fd) { - err = -ENOMEM; - goto free_out; - } - fd->raw = ref; - fd->version = je32_to_cpu(node.d.version); - fd->ino = je32_to_cpu(node.d.ino); - fd->type = node.d.type; - - /* Pick out the mctime of the latest dirent */ - if(fd->version > *mctime_ver) { - *mctime_ver = fd->version; - *latest_mctime = je32_to_cpu(node.d.mctime); - } + } - /* memcpy as much of the name as possible from the raw - dirent we've already read from the flash - */ - if (retlen > sizeof(struct jffs2_raw_dirent)) - memcpy(&fd->name[0], &node.d.name[0], min_t(uint32_t, node.d.nsize, (retlen-sizeof(struct jffs2_raw_dirent)))); - - /* Do we need to copy any more of the name directly - from the flash? - */ - if (node.d.nsize + sizeof(struct jffs2_raw_dirent) > retlen) { - /* FIXME: point() */ - int already = retlen - sizeof(struct jffs2_raw_dirent); - - err = jffs2_flash_read(c, (ref_offset(ref)) + retlen, - node.d.nsize - already, &retlen, &fd->name[already]); - if (!err && retlen != node.d.nsize - already) - err = -EIO; - - if (err) { - printk(KERN_WARNING "Read remainder of name in jffs2_get_inode_nodes(): error %d\n", err); - jffs2_free_full_dirent(fd); - goto free_out; - } - } - fd->nhash = full_name_hash(fd->name, node.d.nsize); - fd->next = NULL; - fd->name[node.d.nsize] = '\0'; - /* Wheee. We now have a complete jffs2_full_dirent structure, with - the name in it and everything. Link it into the list - */ - D1(printk(KERN_DEBUG "Adding fd \"%s\", ino #%u\n", fd->name, fd->ino)); - jffs2_add_fd_to_list(c, fd, &ret_fd); - break; - - case JFFS2_NODETYPE_INODE: - D1(printk(KERN_DEBUG "Node at %08x (%d) is a data node\n", ref_offset(ref), ref_flags(ref))); - if (retlen < sizeof(node.i)) { - printk(KERN_WARNING "read too short for dnode\n"); - err = -EIO; - goto free_out; - } - if (je32_to_cpu(node.i.version) > *highest_version) - *highest_version = je32_to_cpu(node.i.version); - D1(printk(KERN_DEBUG "version %d, highest_version now %d\n", je32_to_cpu(node.i.version), *highest_version)); - - if (ref_obsolete(ref)) { - /* Obsoleted. This cannot happen, surely? dwmw2 20020308 */ - printk(KERN_ERR "Inode node at 0x%08x became obsolete while we weren't looking\n", - ref_offset(ref)); - BUG(); - } + /* Continue calculating CRC */ + crc = crc32(tn->partial_crc, buffer, len); + if(!pointed) + kfree(buffer); +#ifndef __ECOS + else + c->mtd->unpoint(c->mtd, buffer, ofs, len); +#endif - /* If we've never checked the CRCs on this node, check them now. */ - if (ref_flags(ref) == REF_UNCHECKED) { - uint32_t crc, len; - struct jffs2_eraseblock *jeb; - - crc = crc32(0, &node, sizeof(node.i)-8); - if (crc != je32_to_cpu(node.i.node_crc)) { - printk(KERN_NOTICE "jffs2_get_inode_nodes(): CRC failed on node at 0x%08x: Read 0x%08x, calculated 0x%08x\n", - ref_offset(ref), je32_to_cpu(node.i.node_crc), crc); - jffs2_mark_node_obsolete(c, ref); - spin_lock(&c->erase_completion_lock); - continue; - } - - /* sanity checks */ - if ( je32_to_cpu(node.i.offset) > je32_to_cpu(node.i.isize) || - PAD(je32_to_cpu(node.i.csize) + sizeof (node.i)) != PAD(je32_to_cpu(node.i.totlen))) { - printk(KERN_NOTICE "jffs2_get_inode_nodes(): Inode corrupted at 0x%08x, totlen %d, #ino %d, version %d, isize %d, csize %d, dsize %d \n", - ref_offset(ref), je32_to_cpu(node.i.totlen), je32_to_cpu(node.i.ino), - je32_to_cpu(node.i.version), je32_to_cpu(node.i.isize), - je32_to_cpu(node.i.csize), je32_to_cpu(node.i.dsize)); - jffs2_mark_node_obsolete(c, ref); - spin_lock(&c->erase_completion_lock); - continue; - } + if (crc != tn->data_crc) { + JFFS2_NOTICE("wrong data CRC in data node at 0x%08x: read %#08x, calculated %#08x.\n", + ofs, tn->data_crc, crc); + return 1; + } - if (node.i.compr != JFFS2_COMPR_ZERO && je32_to_cpu(node.i.csize)) { - unsigned char *buf=NULL; - uint32_t pointed = 0; -#ifndef __ECOS - if (c->mtd->point) { - err = c->mtd->point (c->mtd, ref_offset(ref) + sizeof(node.i), je32_to_cpu(node.i.csize), - &retlen, &buf); - if (!err && retlen < je32_to_cpu(node.i.csize)) { - D1(printk(KERN_DEBUG "MTD point returned len too short: 0x%zx\n", retlen)); - c->mtd->unpoint(c->mtd, buf, ref_offset(ref) + sizeof(node.i), je32_to_cpu(node.i.csize)); - } else if (err){ - D1(printk(KERN_DEBUG "MTD point failed %d\n", err)); - } else - pointed = 1; /* succefully pointed to device */ - } -#endif - if(!pointed){ - buf = kmalloc(je32_to_cpu(node.i.csize), GFP_KERNEL); - if (!buf) - return -ENOMEM; - - err = jffs2_flash_read(c, ref_offset(ref) + sizeof(node.i), je32_to_cpu(node.i.csize), - &retlen, buf); - if (!err && retlen != je32_to_cpu(node.i.csize)) - err = -EIO; - if (err) { - kfree(buf); - return err; - } - } - crc = crc32(0, buf, je32_to_cpu(node.i.csize)); - if(!pointed) - kfree(buf); +adj_acc: + jeb = &c->blocks[ref->flash_offset / c->sector_size]; + len = ref_totlen(c, jeb, ref); + + /* + * Mark the node as having been checked and fix the + * accounting accordingly. + */ + spin_lock(&c->erase_completion_lock); + jeb->used_size += len; + jeb->unchecked_size -= len; + c->used_size += len; + c->unchecked_size -= len; + spin_unlock(&c->erase_completion_lock); + + return 0; + +free_out: + if(!pointed) + kfree(buffer); #ifndef __ECOS - else - c->mtd->unpoint(c->mtd, buf, ref_offset(ref) + sizeof(node.i), je32_to_cpu(node.i.csize)); + else + c->mtd->unpoint(c->mtd, buffer, ofs, len); #endif + return err; +} - if (crc != je32_to_cpu(node.i.data_crc)) { - printk(KERN_NOTICE "jffs2_get_inode_nodes(): Data CRC failed on node at 0x%08x: Read 0x%08x, calculated 0x%08x\n", - ref_offset(ref), je32_to_cpu(node.i.data_crc), crc); - jffs2_mark_node_obsolete(c, ref); - spin_lock(&c->erase_completion_lock); - continue; - } - - } +/* + * Helper function for jffs2_add_older_frag_to_fragtree(). + * + * Checks the node if we are in the checking stage. + */ +static inline int check_node(struct jffs2_sb_info *c, struct jffs2_inode_info *f, struct jffs2_tmp_dnode_info *tn) +{ + int ret; - /* Mark the node as having been checked and fix the accounting accordingly */ - spin_lock(&c->erase_completion_lock); - jeb = &c->blocks[ref->flash_offset / c->sector_size]; - len = ref_totlen(c, jeb, ref); - - jeb->used_size += len; - jeb->unchecked_size -= len; - c->used_size += len; - c->unchecked_size -= len; - - /* If node covers at least a whole page, or if it starts at the - beginning of a page and runs to the end of the file, or if - it's a hole node, mark it REF_PRISTINE, else REF_NORMAL. - - If it's actually overlapped, it'll get made NORMAL (or OBSOLETE) - when the overlapping node(s) get added to the tree anyway. - */ - if ((je32_to_cpu(node.i.dsize) >= PAGE_CACHE_SIZE) || - ( ((je32_to_cpu(node.i.offset)&(PAGE_CACHE_SIZE-1))==0) && - (je32_to_cpu(node.i.dsize)+je32_to_cpu(node.i.offset) == je32_to_cpu(node.i.isize)))) { - D1(printk(KERN_DEBUG "Marking node at 0x%08x REF_PRISTINE\n", ref_offset(ref))); - ref->flash_offset = ref_offset(ref) | REF_PRISTINE; - } else { - D1(printk(KERN_DEBUG "Marking node at 0x%08x REF_NORMAL\n", ref_offset(ref))); - ref->flash_offset = ref_offset(ref) | REF_NORMAL; - } - spin_unlock(&c->erase_completion_lock); + BUG_ON(ref_obsolete(tn->fn->raw)); + + /* We only check the data CRC of unchecked nodes */ + if (ref_flags(tn->fn->raw) != REF_UNCHECKED) + return 0; + + dbg_fragtree2("check node %#04x-%#04x, phys offs %#08x.\n", + tn->fn->ofs, tn->fn->ofs + tn->fn->size, ref_offset(tn->fn->raw)); + + ret = check_node_data(c, tn); + if (unlikely(ret < 0)) { + JFFS2_ERROR("check_node_data() returned error: %d.\n", + ret); + } else if (unlikely(ret > 0)) { + dbg_fragtree2("CRC error, mark it obsolete.\n"); + jffs2_mark_node_obsolete(c, tn->fn->raw); + } + + return ret; +} + +/* + * Helper function for jffs2_add_older_frag_to_fragtree(). + * + * Called when the new fragment that is being inserted + * splits a hole fragment. + */ +static int split_hole(struct jffs2_sb_info *c, struct rb_root *root, + struct jffs2_node_frag *newfrag, struct jffs2_node_frag *hole) +{ + dbg_fragtree2("fragment %#04x-%#04x splits the hole %#04x-%#04x\n", + newfrag->ofs, newfrag->ofs + newfrag->size, hole->ofs, hole->ofs + hole->size); + + if (hole->ofs == newfrag->ofs) { + /* + * Well, the new fragment actually starts at the same offset as + * the hole. + */ + if (hole->ofs + hole->size > newfrag->ofs + newfrag->size) { + /* + * We replace the overlapped left part of the hole by + * the new node. + */ + + dbg_fragtree2("insert fragment %#04x-%#04x and cut the left part of the hole\n", + newfrag->ofs, newfrag->ofs + newfrag->size); + rb_replace_node(&hole->rb, &newfrag->rb, root); + + hole->ofs += newfrag->size; + hole->size -= newfrag->size; + + /* + * We know that 'hole' should be the right hand + * fragment. + */ + jffs2_fragtree_insert(hole, newfrag); + rb_insert_color(&hole->rb, root); + } else { + /* + * Ah, the new fragment is of the same size as the hole. + * Relace the hole by it. + */ + dbg_fragtree2("insert fragment %#04x-%#04x and overwrite hole\n", + newfrag->ofs, newfrag->ofs + newfrag->size); + rb_replace_node(&hole->rb, &newfrag->rb, root); + jffs2_free_node_frag(hole); + } + } else { + /* The new fragment lefts some hole space at the left */ + + struct jffs2_node_frag * newfrag2 = NULL; + + if (hole->ofs + hole->size > newfrag->ofs + newfrag->size) { + /* The new frag also lefts some space at the right */ + newfrag2 = new_fragment(NULL, newfrag->ofs + + newfrag->size, hole->ofs + hole->size + - newfrag->ofs - newfrag->size); + if (unlikely(!newfrag2)) { + jffs2_free_node_frag(newfrag); + return -ENOMEM; } + } + + hole->size = newfrag->ofs - hole->ofs; + dbg_fragtree2("left the hole %#04x-%#04x at the left and inserd fragment %#04x-%#04x\n", + hole->ofs, hole->ofs + hole->size, newfrag->ofs, newfrag->ofs + newfrag->size); + + jffs2_fragtree_insert(newfrag, hole); + rb_insert_color(&newfrag->rb, root); + + if (newfrag2) { + dbg_fragtree2("left the hole %#04x-%#04x at the right\n", + newfrag2->ofs, newfrag2->ofs + newfrag2->size); + jffs2_fragtree_insert(newfrag2, newfrag); + rb_insert_color(&newfrag2->rb, root); + } + } + + return 0; +} + +/* + * This function is used when we build inode. It expects the nodes are passed + * in the decreasing version order. The whole point of this is to improve the + * inodes checking on NAND: we check the nodes' data CRC only when they are not + * obsoleted. Previously, add_frag_to_fragtree() function was used and + * nodes were passed to it in the increasing version ordes and CRCs of all + * nodes were checked. + * + * Note: tn->fn->size shouldn't be zero. + * + * Returns 0 if the node was inserted + * 1 if it wasn't inserted (since it is obsolete) + * < 0 an if error occured + */ +int jffs2_add_older_frag_to_fragtree(struct jffs2_sb_info *c, struct jffs2_inode_info *f, + struct jffs2_tmp_dnode_info *tn) +{ + struct jffs2_node_frag *this, *newfrag; + uint32_t lastend; + struct jffs2_full_dnode *fn = tn->fn; + struct rb_root *root = &f->fragtree; + uint32_t fn_size = fn->size, fn_ofs = fn->ofs; + int err, checked = 0; + int ref_flag; + + dbg_fragtree("insert fragment %#04x-%#04x, ver %u\n", fn_ofs, fn_ofs + fn_size, tn->version); + + /* Skip all the nodes which are completed before this one starts */ + this = jffs2_lookup_node_frag(root, fn_ofs); + if (this) + dbg_fragtree2("'this' found %#04x-%#04x (%s)\n", this->ofs, this->ofs + this->size, this->node ? "data" : "hole"); + + if (this) + lastend = this->ofs + this->size; + else + lastend = 0; + + /* Detect the preliminary type of node */ + if (fn->size >= PAGE_CACHE_SIZE) + ref_flag = REF_PRISTINE; + else + ref_flag = REF_NORMAL; + + /* See if we ran off the end of the root */ + if (lastend <= fn_ofs) { + /* We did */ + + /* + * We are going to insert the new node into the + * fragment tree, so check it. + */ + err = check_node(c, f, tn); + if (err != 0) + return err; + + fn->frags = 1; + + newfrag = new_fragment(fn, fn_ofs, fn_size); + if (unlikely(!newfrag)) + return -ENOMEM; + + err = no_overlapping_node(c, root, newfrag, this, lastend); + if (unlikely(err != 0)) { + jffs2_free_node_frag(newfrag); + return err; + } + + goto out_ok; + } - tn = jffs2_alloc_tmp_dnode_info(); - if (!tn) { - D1(printk(KERN_DEBUG "alloc tn failed\n")); - err = -ENOMEM; - goto free_out; + fn->frags = 0; + + while (1) { + /* + * Here we have: + * fn_ofs < this->ofs + this->size && fn_ofs >= this->ofs. + * + * Remember, 'this' has higher version, any non-hole node + * which is already in the fragtree is newer then the newly + * inserted. + */ + if (!this->node) { + /* + * 'this' is the hole fragment, so at least the + * beginning of the new fragment is valid. + */ + + /* + * We are going to insert the new node into the + * fragment tree, so check it. + */ + if (!checked) { + err = check_node(c, f, tn); + if (unlikely(err != 0)) + return err; + checked = 1; } - tn->fn = jffs2_alloc_full_dnode(); - if (!tn->fn) { - D1(printk(KERN_DEBUG "alloc fn failed\n")); - err = -ENOMEM; - jffs2_free_tmp_dnode_info(tn); - goto free_out; + if (this->ofs + this->size >= fn_ofs + fn_size) { + /* We split the hole on two parts */ + + fn->frags += 1; + newfrag = new_fragment(fn, fn_ofs, fn_size); + if (unlikely(!newfrag)) + return -ENOMEM; + + err = split_hole(c, root, newfrag, this); + if (unlikely(err)) + return err; + goto out_ok; } - tn->version = je32_to_cpu(node.i.version); - tn->fn->ofs = je32_to_cpu(node.i.offset); - /* There was a bug where we wrote hole nodes out with - csize/dsize swapped. Deal with it */ - if (node.i.compr == JFFS2_COMPR_ZERO && !je32_to_cpu(node.i.dsize) && je32_to_cpu(node.i.csize)) - tn->fn->size = je32_to_cpu(node.i.csize); - else // normal case... - tn->fn->size = je32_to_cpu(node.i.dsize); - tn->fn->raw = ref; - D1(printk(KERN_DEBUG "dnode @%08x: ver %u, offset %04x, dsize %04x\n", - ref_offset(ref), je32_to_cpu(node.i.version), - je32_to_cpu(node.i.offset), je32_to_cpu(node.i.dsize))); - jffs2_add_tn_to_tree(tn, &ret_tn); - break; - - default: - if (ref_flags(ref) == REF_UNCHECKED) { - struct jffs2_eraseblock *jeb; - uint32_t len; - - printk(KERN_ERR "Eep. Unknown node type %04x at %08x was marked REF_UNCHECKED\n", - je16_to_cpu(node.u.nodetype), ref_offset(ref)); - - /* Mark the node as having been checked and fix the accounting accordingly */ - spin_lock(&c->erase_completion_lock); - jeb = &c->blocks[ref->flash_offset / c->sector_size]; - len = ref_totlen(c, jeb, ref); - - jeb->used_size += len; - jeb->unchecked_size -= len; - c->used_size += len; - c->unchecked_size -= len; - - mark_ref_normal(ref); - spin_unlock(&c->erase_completion_lock); + + /* + * The beginning of the new fragment is valid since it + * overlaps the hole node. + */ + + ref_flag = REF_NORMAL; + + fn->frags += 1; + newfrag = new_fragment(fn, fn_ofs, + this->ofs + this->size - fn_ofs); + if (unlikely(!newfrag)) + return -ENOMEM; + + if (fn_ofs == this->ofs) { + /* + * The new node starts at the same offset as + * the hole and supersieds the hole. + */ + dbg_fragtree2("add the new fragment instead of hole %#04x-%#04x, refcnt %d\n", + fn_ofs, fn_ofs + this->ofs + this->size - fn_ofs, fn->frags); + + rb_replace_node(&this->rb, &newfrag->rb, root); + jffs2_free_node_frag(this); + } else { + /* + * The hole becomes shorter as its right part + * is supersieded by the new fragment. + */ + dbg_fragtree2("reduce size of hole %#04x-%#04x to %#04x-%#04x\n", + this->ofs, this->ofs + this->size, this->ofs, this->ofs + this->size - newfrag->size); + + dbg_fragtree2("add new fragment %#04x-%#04x, refcnt %d\n", fn_ofs, + fn_ofs + this->ofs + this->size - fn_ofs, fn->frags); + + this->size -= newfrag->size; + jffs2_fragtree_insert(newfrag, this); + rb_insert_color(&newfrag->rb, root); } - node.u.nodetype = cpu_to_je16(JFFS2_NODE_ACCURATE | je16_to_cpu(node.u.nodetype)); - if (crc32(0, &node, sizeof(struct jffs2_unknown_node)-4) != je32_to_cpu(node.u.hdr_crc)) { - /* Hmmm. This should have been caught at scan time. */ - printk(KERN_ERR "Node header CRC failed at %08x. But it must have been OK earlier.\n", - ref_offset(ref)); - printk(KERN_ERR "Node was: { %04x, %04x, %08x, %08x }\n", - je16_to_cpu(node.u.magic), je16_to_cpu(node.u.nodetype), je32_to_cpu(node.u.totlen), - je32_to_cpu(node.u.hdr_crc)); - jffs2_mark_node_obsolete(c, ref); - } else switch(je16_to_cpu(node.u.nodetype) & JFFS2_COMPAT_MASK) { - case JFFS2_FEATURE_INCOMPAT: - printk(KERN_NOTICE "Unknown INCOMPAT nodetype %04X at %08x\n", je16_to_cpu(node.u.nodetype), ref_offset(ref)); - /* EEP */ - BUG(); - break; - case JFFS2_FEATURE_ROCOMPAT: - printk(KERN_NOTICE "Unknown ROCOMPAT nodetype %04X at %08x\n", je16_to_cpu(node.u.nodetype), ref_offset(ref)); - if (!(c->flags & JFFS2_SB_FLAG_RO)) - BUG(); - break; - case JFFS2_FEATURE_RWCOMPAT_COPY: - printk(KERN_NOTICE "Unknown RWCOMPAT_COPY nodetype %04X at %08x\n", je16_to_cpu(node.u.nodetype), ref_offset(ref)); - break; - case JFFS2_FEATURE_RWCOMPAT_DELETE: - printk(KERN_NOTICE "Unknown RWCOMPAT_DELETE nodetype %04X at %08x\n", je16_to_cpu(node.u.nodetype), ref_offset(ref)); - jffs2_mark_node_obsolete(c, ref); - break; + + fn_ofs += newfrag->size; + fn_size -= newfrag->size; + this = rb_entry(rb_next(&newfrag->rb), + struct jffs2_node_frag, rb); + + dbg_fragtree2("switch to the next 'this' fragment: %#04x-%#04x %s\n", + this->ofs, this->ofs + this->size, this->node ? "(data)" : "(hole)"); + } + + /* + * 'This' node is not the hole so it obsoletes the new fragment + * either fully or partially. + */ + if (this->ofs + this->size >= fn_ofs + fn_size) { + /* The new node is obsolete, drop it */ + if (fn->frags == 0) { + dbg_fragtree2("%#04x-%#04x is obsolete, mark it obsolete\n", fn_ofs, fn_ofs + fn_size); + ref_flag = REF_OBSOLETE; } + goto out_ok; + } else { + struct jffs2_node_frag *new_this; + + /* 'This' node obsoletes the beginning of the new node */ + dbg_fragtree2("the beginning %#04x-%#04x is obsolete\n", fn_ofs, this->ofs + this->size); + + ref_flag = REF_NORMAL; + + fn_size -= this->ofs + this->size - fn_ofs; + fn_ofs = this->ofs + this->size; + dbg_fragtree2("now considering %#04x-%#04x\n", fn_ofs, fn_ofs + fn_size); + + new_this = rb_entry(rb_next(&this->rb), struct jffs2_node_frag, rb); + if (!new_this) { + /* + * There is no next fragment. Add the rest of + * the new node as the right-hand child. + */ + if (!checked) { + err = check_node(c, f, tn); + if (unlikely(err != 0)) + return err; + checked = 1; + } + fn->frags += 1; + newfrag = new_fragment(fn, fn_ofs, fn_size); + if (unlikely(!newfrag)) + return -ENOMEM; + + dbg_fragtree2("there are no more fragments, insert %#04x-%#04x\n", + newfrag->ofs, newfrag->ofs + newfrag->size); + rb_link_node(&newfrag->rb, &this->rb, &this->rb.rb_right); + rb_insert_color(&newfrag->rb, root); + goto out_ok; + } else { + this = new_this; + dbg_fragtree2("switch to the next 'this' fragment: %#04x-%#04x %s\n", + this->ofs, this->ofs + this->size, this->node ? "(data)" : "(hole)"); + } } - spin_lock(&c->erase_completion_lock); + } + +out_ok: + BUG_ON(fn->size < PAGE_CACHE_SIZE && ref_flag == REF_PRISTINE); + if (ref_flag == REF_OBSOLETE) { + dbg_fragtree2("the node is obsolete now\n"); + /* jffs2_mark_node_obsolete() will adjust space accounting */ + jffs2_mark_node_obsolete(c, fn->raw); + return 1; } + + dbg_fragtree2("the node is \"%s\" now\n", ref_flag == REF_NORMAL ? "REF_NORMAL" : "REF_PRISTINE"); + + /* Space accounting was adjusted at check_node_data() */ + spin_lock(&c->erase_completion_lock); + fn->raw->flash_offset = ref_offset(fn->raw) | ref_flag; spin_unlock(&c->erase_completion_lock); - *tnp = ret_tn; - *fdp = ret_fd; return 0; - - free_out: - jffs2_free_tmp_dnode_info_list(&ret_tn); - jffs2_free_full_dirent_list(ret_fd); - return err; } void jffs2_set_inocache_state(struct jffs2_sb_info *c, struct jffs2_inode_cache *ic, int state) @@ -499,24 +862,21 @@ void jffs2_set_inocache_state(struct jffs2_sb_info *c, struct jffs2_inode_cache /* During mount, this needs no locking. During normal operation, its callers want to do other stuff while still holding the inocache_lock. - Rather than introducing special case get_ino_cache functions or + Rather than introducing special case get_ino_cache functions or callbacks, we just let the caller do the locking itself. */ - + struct jffs2_inode_cache *jffs2_get_ino_cache(struct jffs2_sb_info *c, uint32_t ino) { struct jffs2_inode_cache *ret; - D2(printk(KERN_DEBUG "jffs2_get_ino_cache(): ino %u\n", ino)); - ret = c->inocache_list[ino % INOCACHE_HASHSIZE]; while (ret && ret->ino < ino) { ret = ret->next; } - + if (ret && ret->ino != ino) ret = NULL; - D2(printk(KERN_DEBUG "jffs2_get_ino_cache found %p for ino %u\n", ret, ino)); return ret; } @@ -528,7 +888,7 @@ void jffs2_add_ino_cache (struct jffs2_sb_info *c, struct jffs2_inode_cache *new if (!new->ino) new->ino = ++c->highest_ino; - D2(printk(KERN_DEBUG "jffs2_add_ino_cache: Add %p (ino #%u)\n", new, new->ino)); + dbg_inocache("add %p (ino #%u)\n", new, new->ino); prev = &c->inocache_list[new->ino % INOCACHE_HASHSIZE]; @@ -544,11 +904,12 @@ void jffs2_add_ino_cache (struct jffs2_sb_info *c, struct jffs2_inode_cache *new void jffs2_del_ino_cache(struct jffs2_sb_info *c, struct jffs2_inode_cache *old) { struct jffs2_inode_cache **prev; - D1(printk(KERN_DEBUG "jffs2_del_ino_cache: Del %p (ino #%u)\n", old, old->ino)); + + dbg_inocache("del %p (ino #%u)\n", old, old->ino); spin_lock(&c->inocache_lock); - + prev = &c->inocache_list[old->ino % INOCACHE_HASHSIZE]; - + while ((*prev) && (*prev)->ino < old->ino) { prev = &(*prev)->next; } @@ -558,7 +919,7 @@ void jffs2_del_ino_cache(struct jffs2_sb_info *c, struct jffs2_inode_cache *old) /* Free it now unless it's in READING or CLEARING state, which are the transitions upon read_inode() and clear_inode(). The - rest of the time we know nobody else is looking at it, and + rest of the time we know nobody else is looking at it, and if it's held by read_inode() or clear_inode() they'll free it for themselves. */ if (old->state != INO_STATE_READING && old->state != INO_STATE_CLEARING) @@ -571,7 +932,7 @@ void jffs2_free_ino_caches(struct jffs2_sb_info *c) { int i; struct jffs2_inode_cache *this, *next; - + for (i=0; i<INOCACHE_HASHSIZE; i++) { this = c->inocache_list[i]; while (this) { @@ -598,38 +959,30 @@ void jffs2_free_raw_node_refs(struct jffs2_sb_info *c) c->blocks[i].first_node = c->blocks[i].last_node = NULL; } } - + struct jffs2_node_frag *jffs2_lookup_node_frag(struct rb_root *fragtree, uint32_t offset) { - /* The common case in lookup is that there will be a node + /* The common case in lookup is that there will be a node which precisely matches. So we go looking for that first */ struct rb_node *next; struct jffs2_node_frag *prev = NULL; struct jffs2_node_frag *frag = NULL; - D2(printk(KERN_DEBUG "jffs2_lookup_node_frag(%p, %d)\n", fragtree, offset)); + dbg_fragtree2("root %p, offset %d\n", fragtree, offset); next = fragtree->rb_node; while(next) { frag = rb_entry(next, struct jffs2_node_frag, rb); - D2(printk(KERN_DEBUG "Considering frag %d-%d (%p). left %p, right %p\n", - frag->ofs, frag->ofs+frag->size, frag, frag->rb.rb_left, frag->rb.rb_right)); if (frag->ofs + frag->size <= offset) { - D2(printk(KERN_DEBUG "Going right from frag %d-%d, before the region we care about\n", - frag->ofs, frag->ofs+frag->size)); /* Remember the closest smaller match on the way down */ if (!prev || frag->ofs > prev->ofs) prev = frag; next = frag->rb.rb_right; } else if (frag->ofs > offset) { - D2(printk(KERN_DEBUG "Going left from frag %d-%d, after the region we care about\n", - frag->ofs, frag->ofs+frag->size)); next = frag->rb.rb_left; } else { - D2(printk(KERN_DEBUG "Returning frag %d,%d, matched\n", - frag->ofs, frag->ofs+frag->size)); return frag; } } @@ -638,11 +991,11 @@ struct jffs2_node_frag *jffs2_lookup_node_frag(struct rb_root *fragtree, uint32_ and return the closest smaller one */ if (prev) - D2(printk(KERN_DEBUG "No match. Returning frag %d,%d, closest previous\n", - prev->ofs, prev->ofs+prev->size)); - else - D2(printk(KERN_DEBUG "Returning NULL, empty fragtree\n")); - + dbg_fragtree2("no match. Returning frag %#04x-%#04x, closest previous\n", + prev->ofs, prev->ofs+prev->size); + else + dbg_fragtree2("returning NULL, empty fragtree\n"); + return prev; } @@ -656,39 +1009,32 @@ void jffs2_kill_fragtree(struct rb_root *root, struct jffs2_sb_info *c) if (!root->rb_node) return; - frag = (rb_entry(root->rb_node, struct jffs2_node_frag, rb)); + dbg_fragtree("killing\n"); + frag = (rb_entry(root->rb_node, struct jffs2_node_frag, rb)); while(frag) { if (frag->rb.rb_left) { - D2(printk(KERN_DEBUG "Going left from frag (%p) %d-%d\n", - frag, frag->ofs, frag->ofs+frag->size)); frag = frag_left(frag); continue; } if (frag->rb.rb_right) { - D2(printk(KERN_DEBUG "Going right from frag (%p) %d-%d\n", - frag, frag->ofs, frag->ofs+frag->size)); frag = frag_right(frag); continue; } - D2(printk(KERN_DEBUG "jffs2_kill_fragtree: frag at 0x%x-0x%x: node %p, frags %d--\n", - frag->ofs, frag->ofs+frag->size, frag->node, - frag->node?frag->node->frags:0)); - if (frag->node && !(--frag->node->frags)) { - /* Not a hole, and it's the final remaining frag + /* Not a hole, and it's the final remaining frag of this node. Free the node */ if (c) jffs2_mark_node_obsolete(c, frag->node->raw); - + jffs2_free_full_dnode(frag->node); } parent = frag_parent(frag); if (parent) { if (frag_left(parent) == frag) parent->rb.rb_left = NULL; - else + else parent->rb.rb_right = NULL; } @@ -698,29 +1044,3 @@ void jffs2_kill_fragtree(struct rb_root *root, struct jffs2_sb_info *c) cond_resched(); } } - -void jffs2_fragtree_insert(struct jffs2_node_frag *newfrag, struct jffs2_node_frag *base) -{ - struct rb_node *parent = &base->rb; - struct rb_node **link = &parent; - - D2(printk(KERN_DEBUG "jffs2_fragtree_insert(%p; %d-%d, %p)\n", newfrag, - newfrag->ofs, newfrag->ofs+newfrag->size, base)); - - while (*link) { - parent = *link; - base = rb_entry(parent, struct jffs2_node_frag, rb); - - D2(printk(KERN_DEBUG "fragtree_insert considering frag at 0x%x\n", base->ofs)); - if (newfrag->ofs > base->ofs) - link = &base->rb.rb_right; - else if (newfrag->ofs < base->ofs) - link = &base->rb.rb_left; - else { - printk(KERN_CRIT "Duplicate frag at %08x (%p,%p)\n", newfrag->ofs, newfrag, base); - BUG(); - } - } - - rb_link_node(&newfrag->rb, &base->rb, link); -} diff --git a/fs/jffs2/nodelist.h b/fs/jffs2/nodelist.h index b34c397909ef..23a67bb3052f 100644 --- a/fs/jffs2/nodelist.h +++ b/fs/jffs2/nodelist.h @@ -7,7 +7,7 @@ * * For licensing information, see the file 'LICENCE' in this directory. * - * $Id: nodelist.h,v 1.131 2005/07/05 21:03:07 dwmw2 Exp $ + * $Id: nodelist.h,v 1.140 2005/09/07 08:34:54 havasi Exp $ * */ @@ -20,30 +20,15 @@ #include <linux/jffs2.h> #include <linux/jffs2_fs_sb.h> #include <linux/jffs2_fs_i.h> +#include "summary.h" #ifdef __ECOS #include "os-ecos.h" #else -#include <linux/mtd/compatmac.h> /* For min/max in older kernels */ +#include <linux/mtd/compatmac.h> /* For compatibility with older kernels */ #include "os-linux.h" #endif -#ifndef CONFIG_JFFS2_FS_DEBUG -#define CONFIG_JFFS2_FS_DEBUG 1 -#endif - -#if CONFIG_JFFS2_FS_DEBUG > 0 -#define D1(x) x -#else -#define D1(x) -#endif - -#if CONFIG_JFFS2_FS_DEBUG > 1 -#define D2(x) x -#else -#define D2(x) -#endif - #define JFFS2_NATIVE_ENDIAN /* Note we handle mode bits conversion from JFFS2 (i.e. Linux) to/from @@ -73,14 +58,17 @@ #define je16_to_cpu(x) (le16_to_cpu(x.v16)) #define je32_to_cpu(x) (le32_to_cpu(x.v32)) #define jemode_to_cpu(x) (le32_to_cpu(jffs2_to_os_mode((x).m))) -#else +#else #error wibble #endif +/* The minimal node header size */ +#define JFFS2_MIN_NODE_HEADER sizeof(struct jffs2_raw_dirent) + /* This is all we need to keep in-core for each raw node during normal operation. As and when we do read_inode on a particular inode, we can - scan the nodes which are listed for it and build up a proper map of + scan the nodes which are listed for it and build up a proper map of which nodes are currently valid. JFFSv1 always used to keep that whole map in core for each inode. */ @@ -97,7 +85,7 @@ struct jffs2_raw_node_ref /* flash_offset & 3 always has to be zero, because nodes are always aligned at 4 bytes. So we have a couple of extra bits - to play with, which indicate the node's status; see below: */ + to play with, which indicate the node's status; see below: */ #define REF_UNCHECKED 0 /* We haven't yet checked the CRC or built its inode */ #define REF_OBSOLETE 1 /* Obsolete, can be completely ignored */ #define REF_PRISTINE 2 /* Completely clean. GC without looking */ @@ -110,7 +98,7 @@ struct jffs2_raw_node_ref /* For each inode in the filesystem, we need to keep a record of nlink, because it would be a PITA to scan the whole directory tree at read_inode() time to calculate it, and to keep sufficient information - in the raw_node_ref (basically both parent and child inode number for + in the raw_node_ref (basically both parent and child inode number for dirent nodes) would take more space than this does. We also keep a pointer to the first physical node which is part of this inode, too. */ @@ -140,7 +128,7 @@ struct jffs2_inode_cache { #define INOCACHE_HASHSIZE 128 /* - Larger representation of a raw node, kept in-core only when the + Larger representation of a raw node, kept in-core only when the struct inode for this particular ino is instantiated. */ @@ -150,11 +138,11 @@ struct jffs2_full_dnode uint32_t ofs; /* The offset to which the data of this node belongs */ uint32_t size; uint32_t frags; /* Number of fragments which currently refer - to this node. When this reaches zero, + to this node. When this reaches zero, the node is obsolete. */ }; -/* +/* Even larger representation of a raw node, kept in-core only while we're actually building up the original map of which nodes go where, in read_inode() @@ -164,7 +152,10 @@ struct jffs2_tmp_dnode_info struct rb_node rb; struct jffs2_full_dnode *fn; uint32_t version; -}; + uint32_t data_crc; + uint32_t partial_crc; + uint32_t csize; +}; struct jffs2_full_dirent { @@ -178,7 +169,7 @@ struct jffs2_full_dirent }; /* - Fragments - used to build a map of which raw node to obtain + Fragments - used to build a map of which raw node to obtain data from for each part of the ino */ struct jffs2_node_frag @@ -207,86 +198,18 @@ struct jffs2_eraseblock struct jffs2_raw_node_ref *gc_node; /* Next node to be garbage collected */ }; -#define ACCT_SANITY_CHECK(c, jeb) do { \ - struct jffs2_eraseblock *___j = jeb; \ - if ((___j) && ___j->used_size + ___j->dirty_size + ___j->free_size + ___j->wasted_size + ___j->unchecked_size != c->sector_size) { \ - printk(KERN_NOTICE "Eeep. Space accounting for block at 0x%08x is screwed\n", ___j->offset); \ - printk(KERN_NOTICE "free 0x%08x + dirty 0x%08x + used %08x + wasted %08x + unchecked %08x != total %08x\n", \ - ___j->free_size, ___j->dirty_size, ___j->used_size, ___j->wasted_size, ___j->unchecked_size, c->sector_size); \ - BUG(); \ - } \ - if (c->used_size + c->dirty_size + c->free_size + c->erasing_size + c->bad_size + c->wasted_size + c->unchecked_size != c->flash_size) { \ - printk(KERN_NOTICE "Eeep. Space accounting superblock info is screwed\n"); \ - printk(KERN_NOTICE "free 0x%08x + dirty 0x%08x + used %08x + erasing %08x + bad %08x + wasted %08x + unchecked %08x != total %08x\n", \ - c->free_size, c->dirty_size, c->used_size, c->erasing_size, c->bad_size, c->wasted_size, c->unchecked_size, c->flash_size); \ - BUG(); \ - } \ -} while(0) - -static inline void paranoia_failed_dump(struct jffs2_eraseblock *jeb) +static inline int jffs2_blocks_use_vmalloc(struct jffs2_sb_info *c) { - struct jffs2_raw_node_ref *ref; - int i=0; - - printk(KERN_NOTICE); - for (ref = jeb->first_node; ref; ref = ref->next_phys) { - printk("%08x->", ref_offset(ref)); - if (++i == 8) { - i = 0; - printk("\n" KERN_NOTICE); - } - } - printk("\n"); + return ((c->flash_size / c->sector_size) * sizeof (struct jffs2_eraseblock)) > (128 * 1024); } - -#define ACCT_PARANOIA_CHECK(jeb) do { \ - uint32_t my_used_size = 0; \ - uint32_t my_unchecked_size = 0; \ - struct jffs2_raw_node_ref *ref2 = jeb->first_node; \ - while (ref2) { \ - if (unlikely(ref2->flash_offset < jeb->offset || \ - ref2->flash_offset > jeb->offset + c->sector_size)) { \ - printk(KERN_NOTICE "Node %08x shouldn't be in block at %08x!\n", \ - ref_offset(ref2), jeb->offset); \ - paranoia_failed_dump(jeb); \ - BUG(); \ - } \ - if (ref_flags(ref2) == REF_UNCHECKED) \ - my_unchecked_size += ref_totlen(c, jeb, ref2); \ - else if (!ref_obsolete(ref2)) \ - my_used_size += ref_totlen(c, jeb, ref2); \ - if (unlikely((!ref2->next_phys) != (ref2 == jeb->last_node))) { \ - if (!ref2->next_phys) \ - printk("ref for node at %p (phys %08x) has next_phys->%p (----), last_node->%p (phys %08x)\n", \ - ref2, ref_offset(ref2), ref2->next_phys, \ - jeb->last_node, ref_offset(jeb->last_node)); \ - else \ - printk("ref for node at %p (phys %08x) has next_phys->%p (%08x), last_node->%p (phys %08x)\n", \ - ref2, ref_offset(ref2), ref2->next_phys, ref_offset(ref2->next_phys), \ - jeb->last_node, ref_offset(jeb->last_node)); \ - paranoia_failed_dump(jeb); \ - BUG(); \ - } \ - ref2 = ref2->next_phys; \ - } \ - if (my_used_size != jeb->used_size) { \ - printk(KERN_NOTICE "Calculated used size %08x != stored used size %08x\n", my_used_size, jeb->used_size); \ - BUG(); \ - } \ - if (my_unchecked_size != jeb->unchecked_size) { \ - printk(KERN_NOTICE "Calculated unchecked size %08x != stored unchecked size %08x\n", my_unchecked_size, jeb->unchecked_size); \ - BUG(); \ - } \ - } while(0) - /* Calculate totlen from surrounding nodes or eraseblock */ static inline uint32_t __ref_totlen(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb, struct jffs2_raw_node_ref *ref) { uint32_t ref_end; - + if (ref->next_phys) ref_end = ref_offset(ref->next_phys); else { @@ -306,11 +229,13 @@ static inline uint32_t ref_totlen(struct jffs2_sb_info *c, { uint32_t ret; - D1(if (jeb && jeb != &c->blocks[ref->flash_offset / c->sector_size]) { +#if CONFIG_JFFS2_FS_DEBUG > 0 + if (jeb && jeb != &c->blocks[ref->flash_offset / c->sector_size]) { printk(KERN_CRIT "ref_totlen called with wrong block -- at 0x%08x instead of 0x%08x; ref 0x%08x\n", jeb->offset, c->blocks[ref->flash_offset / c->sector_size].offset, ref_offset(ref)); BUG(); - }) + } +#endif #if 1 ret = ref->__totlen; @@ -323,14 +248,13 @@ static inline uint32_t ref_totlen(struct jffs2_sb_info *c, ret, ref->__totlen); if (!jeb) jeb = &c->blocks[ref->flash_offset / c->sector_size]; - paranoia_failed_dump(jeb); + jffs2_dbg_dump_node_refs_nolock(c, jeb); BUG(); } #endif return ret; } - #define ALLOC_NORMAL 0 /* Normal allocation */ #define ALLOC_DELETION 1 /* Deletion node. Best to allow it */ #define ALLOC_GC 2 /* Space requested for GC. Give it or die */ @@ -340,7 +264,7 @@ static inline uint32_t ref_totlen(struct jffs2_sb_info *c, #define VERYDIRTY(c, size) ((size) >= ((c)->sector_size / 2)) /* check if dirty space is more than 255 Byte */ -#define ISDIRTY(size) ((size) > sizeof (struct jffs2_raw_inode) + JFFS2_MIN_DATA_LEN) +#define ISDIRTY(size) ((size) > sizeof (struct jffs2_raw_inode) + JFFS2_MIN_DATA_LEN) #define PAD(x) (((x)+3)&~3) @@ -384,12 +308,7 @@ static inline struct jffs2_node_frag *frag_last(struct rb_root *root) #define frag_erase(frag, list) rb_erase(&frag->rb, list); /* nodelist.c */ -D2(void jffs2_print_frag_list(struct jffs2_inode_info *f)); void jffs2_add_fd_to_list(struct jffs2_sb_info *c, struct jffs2_full_dirent *new, struct jffs2_full_dirent **list); -int jffs2_get_inode_nodes(struct jffs2_sb_info *c, struct jffs2_inode_info *f, - struct rb_root *tnp, struct jffs2_full_dirent **fdp, - uint32_t *highest_version, uint32_t *latest_mctime, - uint32_t *mctime_ver); void jffs2_set_inocache_state(struct jffs2_sb_info *c, struct jffs2_inode_cache *ic, int state); struct jffs2_inode_cache *jffs2_get_ino_cache(struct jffs2_sb_info *c, uint32_t ino); void jffs2_add_ino_cache (struct jffs2_sb_info *c, struct jffs2_inode_cache *new); @@ -398,19 +317,23 @@ void jffs2_free_ino_caches(struct jffs2_sb_info *c); void jffs2_free_raw_node_refs(struct jffs2_sb_info *c); struct jffs2_node_frag *jffs2_lookup_node_frag(struct rb_root *fragtree, uint32_t offset); void jffs2_kill_fragtree(struct rb_root *root, struct jffs2_sb_info *c_delete); -void jffs2_fragtree_insert(struct jffs2_node_frag *newfrag, struct jffs2_node_frag *base); struct rb_node *rb_next(struct rb_node *); struct rb_node *rb_prev(struct rb_node *); void rb_replace_node(struct rb_node *victim, struct rb_node *new, struct rb_root *root); +void jffs2_obsolete_node_frag(struct jffs2_sb_info *c, struct jffs2_node_frag *this); +int jffs2_add_full_dnode_to_inode(struct jffs2_sb_info *c, struct jffs2_inode_info *f, struct jffs2_full_dnode *fn); +void jffs2_truncate_fragtree (struct jffs2_sb_info *c, struct rb_root *list, uint32_t size); +int jffs2_add_older_frag_to_fragtree(struct jffs2_sb_info *c, struct jffs2_inode_info *f, struct jffs2_tmp_dnode_info *tn); /* nodemgmt.c */ int jffs2_thread_should_wake(struct jffs2_sb_info *c); -int jffs2_reserve_space(struct jffs2_sb_info *c, uint32_t minsize, uint32_t *ofs, uint32_t *len, int prio); -int jffs2_reserve_space_gc(struct jffs2_sb_info *c, uint32_t minsize, uint32_t *ofs, uint32_t *len); +int jffs2_reserve_space(struct jffs2_sb_info *c, uint32_t minsize, uint32_t *ofs, + uint32_t *len, int prio, uint32_t sumsize); +int jffs2_reserve_space_gc(struct jffs2_sb_info *c, uint32_t minsize, uint32_t *ofs, + uint32_t *len, uint32_t sumsize); int jffs2_add_physical_node_ref(struct jffs2_sb_info *c, struct jffs2_raw_node_ref *new); void jffs2_complete_reservation(struct jffs2_sb_info *c); void jffs2_mark_node_obsolete(struct jffs2_sb_info *c, struct jffs2_raw_node_ref *raw); -void jffs2_dump_block_lists(struct jffs2_sb_info *c); /* write.c */ int jffs2_do_new_inode(struct jffs2_sb_info *c, struct jffs2_inode_info *f, uint32_t mode, struct jffs2_raw_inode *ri); @@ -418,17 +341,15 @@ int jffs2_do_new_inode(struct jffs2_sb_info *c, struct jffs2_inode_info *f, uint struct jffs2_full_dnode *jffs2_write_dnode(struct jffs2_sb_info *c, struct jffs2_inode_info *f, struct jffs2_raw_inode *ri, const unsigned char *data, uint32_t datalen, uint32_t flash_ofs, int alloc_mode); struct jffs2_full_dirent *jffs2_write_dirent(struct jffs2_sb_info *c, struct jffs2_inode_info *f, struct jffs2_raw_dirent *rd, const unsigned char *name, uint32_t namelen, uint32_t flash_ofs, int alloc_mode); int jffs2_write_inode_range(struct jffs2_sb_info *c, struct jffs2_inode_info *f, - struct jffs2_raw_inode *ri, unsigned char *buf, + struct jffs2_raw_inode *ri, unsigned char *buf, uint32_t offset, uint32_t writelen, uint32_t *retlen); int jffs2_do_create(struct jffs2_sb_info *c, struct jffs2_inode_info *dir_f, struct jffs2_inode_info *f, struct jffs2_raw_inode *ri, const char *name, int namelen); -int jffs2_do_unlink(struct jffs2_sb_info *c, struct jffs2_inode_info *dir_f, const char *name, int namelen, struct jffs2_inode_info *dead_f); -int jffs2_do_link (struct jffs2_sb_info *c, struct jffs2_inode_info *dir_f, uint32_t ino, uint8_t type, const char *name, int namelen); +int jffs2_do_unlink(struct jffs2_sb_info *c, struct jffs2_inode_info *dir_f, const char *name, int namelen, struct jffs2_inode_info *dead_f, uint32_t time); +int jffs2_do_link (struct jffs2_sb_info *c, struct jffs2_inode_info *dir_f, uint32_t ino, uint8_t type, const char *name, int namelen, uint32_t time); /* readinode.c */ -void jffs2_truncate_fraglist (struct jffs2_sb_info *c, struct rb_root *list, uint32_t size); -int jffs2_add_full_dnode_to_inode(struct jffs2_sb_info *c, struct jffs2_inode_info *f, struct jffs2_full_dnode *fn); -int jffs2_do_read_inode(struct jffs2_sb_info *c, struct jffs2_inode_info *f, +int jffs2_do_read_inode(struct jffs2_sb_info *c, struct jffs2_inode_info *f, uint32_t ino, struct jffs2_raw_inode *latest_node); int jffs2_do_crccheck_inode(struct jffs2_sb_info *c, struct jffs2_inode_cache *ic); void jffs2_do_clear_inode(struct jffs2_sb_info *c, struct jffs2_inode_info *f); @@ -468,6 +389,10 @@ char *jffs2_getlink(struct jffs2_sb_info *c, struct jffs2_inode_info *f); /* scan.c */ int jffs2_scan_medium(struct jffs2_sb_info *c); void jffs2_rotate_lists(struct jffs2_sb_info *c); +int jffs2_fill_scan_buf(struct jffs2_sb_info *c, void *buf, + uint32_t ofs, uint32_t len); +struct jffs2_inode_cache *jffs2_scan_make_ino_cache(struct jffs2_sb_info *c, uint32_t ino); +int jffs2_scan_classify_jeb(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb); /* build.c */ int jffs2_do_mount_fs(struct jffs2_sb_info *c); @@ -483,4 +408,6 @@ int jffs2_check_nand_cleanmarker(struct jffs2_sb_info *c, struct jffs2_erasebloc int jffs2_write_nand_cleanmarker(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb); #endif +#include "debug.h" + #endif /* __JFFS2_NODELIST_H__ */ diff --git a/fs/jffs2/nodemgmt.c b/fs/jffs2/nodemgmt.c index c1d8b5ed9ab9..49127a1f0458 100644 --- a/fs/jffs2/nodemgmt.c +++ b/fs/jffs2/nodemgmt.c @@ -7,7 +7,7 @@ * * For licensing information, see the file 'LICENCE' in this directory. * - * $Id: nodemgmt.c,v 1.122 2005/05/06 09:30:27 dedekind Exp $ + * $Id: nodemgmt.c,v 1.127 2005/09/20 15:49:12 dedekind Exp $ * */ @@ -17,6 +17,7 @@ #include <linux/compiler.h> #include <linux/sched.h> /* For cond_resched() */ #include "nodelist.h" +#include "debug.h" /** * jffs2_reserve_space - request physical space to write nodes to flash @@ -38,9 +39,11 @@ * for the requested allocation. */ -static int jffs2_do_reserve_space(struct jffs2_sb_info *c, uint32_t minsize, uint32_t *ofs, uint32_t *len); +static int jffs2_do_reserve_space(struct jffs2_sb_info *c, uint32_t minsize, + uint32_t *ofs, uint32_t *len, uint32_t sumsize); -int jffs2_reserve_space(struct jffs2_sb_info *c, uint32_t minsize, uint32_t *ofs, uint32_t *len, int prio) +int jffs2_reserve_space(struct jffs2_sb_info *c, uint32_t minsize, uint32_t *ofs, + uint32_t *len, int prio, uint32_t sumsize) { int ret = -EAGAIN; int blocksneeded = c->resv_blocks_write; @@ -85,12 +88,12 @@ int jffs2_reserve_space(struct jffs2_sb_info *c, uint32_t minsize, uint32_t *ofs up(&c->alloc_sem); return -ENOSPC; } - + /* Calc possibly available space. Possibly available means that we * don't know, if unchecked size contains obsoleted nodes, which could give us some * more usable space. This will affect the sum only once, as gc first finishes checking * of nodes. - + Return -ENOSPC, if the maximum possibly available space is less or equal than + + Return -ENOSPC, if the maximum possibly available space is less or equal than * blocksneeded * sector_size. * This blocks endless gc looping on a filesystem, which is nearly full, even if * the check above passes. @@ -115,7 +118,7 @@ int jffs2_reserve_space(struct jffs2_sb_info *c, uint32_t minsize, uint32_t *ofs c->nr_free_blocks, c->nr_erasing_blocks, c->free_size, c->dirty_size, c->wasted_size, c->used_size, c->erasing_size, c->bad_size, c->free_size + c->dirty_size + c->wasted_size + c->used_size + c->erasing_size + c->bad_size, c->flash_size)); spin_unlock(&c->erase_completion_lock); - + ret = jffs2_garbage_collect_pass(c); if (ret) return ret; @@ -129,7 +132,7 @@ int jffs2_reserve_space(struct jffs2_sb_info *c, uint32_t minsize, uint32_t *ofs spin_lock(&c->erase_completion_lock); } - ret = jffs2_do_reserve_space(c, minsize, ofs, len); + ret = jffs2_do_reserve_space(c, minsize, ofs, len, sumsize); if (ret) { D1(printk(KERN_DEBUG "jffs2_reserve_space: ret is %d\n", ret)); } @@ -140,7 +143,8 @@ int jffs2_reserve_space(struct jffs2_sb_info *c, uint32_t minsize, uint32_t *ofs return ret; } -int jffs2_reserve_space_gc(struct jffs2_sb_info *c, uint32_t minsize, uint32_t *ofs, uint32_t *len) +int jffs2_reserve_space_gc(struct jffs2_sb_info *c, uint32_t minsize, uint32_t *ofs, + uint32_t *len, uint32_t sumsize) { int ret = -EAGAIN; minsize = PAD(minsize); @@ -149,7 +153,7 @@ int jffs2_reserve_space_gc(struct jffs2_sb_info *c, uint32_t minsize, uint32_t * spin_lock(&c->erase_completion_lock); while(ret == -EAGAIN) { - ret = jffs2_do_reserve_space(c, minsize, ofs, len); + ret = jffs2_do_reserve_space(c, minsize, ofs, len, sumsize); if (ret) { D1(printk(KERN_DEBUG "jffs2_reserve_space_gc: looping, ret is %d\n", ret)); } @@ -158,105 +162,185 @@ int jffs2_reserve_space_gc(struct jffs2_sb_info *c, uint32_t minsize, uint32_t * return ret; } -/* Called with alloc sem _and_ erase_completion_lock */ -static int jffs2_do_reserve_space(struct jffs2_sb_info *c, uint32_t minsize, uint32_t *ofs, uint32_t *len) + +/* Classify nextblock (clean, dirty of verydirty) and force to select an other one */ + +static void jffs2_close_nextblock(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb) { - struct jffs2_eraseblock *jeb = c->nextblock; - - restart: - if (jeb && minsize > jeb->free_size) { - /* Skip the end of this block and file it as having some dirty space */ - /* If there's a pending write to it, flush now */ - if (jffs2_wbuf_dirty(c)) { + + /* Check, if we have a dirty block now, or if it was dirty already */ + if (ISDIRTY (jeb->wasted_size + jeb->dirty_size)) { + c->dirty_size += jeb->wasted_size; + c->wasted_size -= jeb->wasted_size; + jeb->dirty_size += jeb->wasted_size; + jeb->wasted_size = 0; + if (VERYDIRTY(c, jeb->dirty_size)) { + D1(printk(KERN_DEBUG "Adding full erase block at 0x%08x to very_dirty_list (free 0x%08x, dirty 0x%08x, used 0x%08x\n", + jeb->offset, jeb->free_size, jeb->dirty_size, jeb->used_size)); + list_add_tail(&jeb->list, &c->very_dirty_list); + } else { + D1(printk(KERN_DEBUG "Adding full erase block at 0x%08x to dirty_list (free 0x%08x, dirty 0x%08x, used 0x%08x\n", + jeb->offset, jeb->free_size, jeb->dirty_size, jeb->used_size)); + list_add_tail(&jeb->list, &c->dirty_list); + } + } else { + D1(printk(KERN_DEBUG "Adding full erase block at 0x%08x to clean_list (free 0x%08x, dirty 0x%08x, used 0x%08x\n", + jeb->offset, jeb->free_size, jeb->dirty_size, jeb->used_size)); + list_add_tail(&jeb->list, &c->clean_list); + } + c->nextblock = NULL; + +} + +/* Select a new jeb for nextblock */ + +static int jffs2_find_nextblock(struct jffs2_sb_info *c) +{ + struct list_head *next; + + /* Take the next block off the 'free' list */ + + if (list_empty(&c->free_list)) { + + if (!c->nr_erasing_blocks && + !list_empty(&c->erasable_list)) { + struct jffs2_eraseblock *ejeb; + + ejeb = list_entry(c->erasable_list.next, struct jffs2_eraseblock, list); + list_del(&ejeb->list); + list_add_tail(&ejeb->list, &c->erase_pending_list); + c->nr_erasing_blocks++; + jffs2_erase_pending_trigger(c); + D1(printk(KERN_DEBUG "jffs2_find_nextblock: Triggering erase of erasable block at 0x%08x\n", + ejeb->offset)); + } + + if (!c->nr_erasing_blocks && + !list_empty(&c->erasable_pending_wbuf_list)) { + D1(printk(KERN_DEBUG "jffs2_find_nextblock: Flushing write buffer\n")); + /* c->nextblock is NULL, no update to c->nextblock allowed */ spin_unlock(&c->erase_completion_lock); - D1(printk(KERN_DEBUG "jffs2_do_reserve_space: Flushing write buffer\n")); jffs2_flush_wbuf_pad(c); spin_lock(&c->erase_completion_lock); - jeb = c->nextblock; - goto restart; + /* Have another go. It'll be on the erasable_list now */ + return -EAGAIN; } - c->wasted_size += jeb->free_size; - c->free_size -= jeb->free_size; - jeb->wasted_size += jeb->free_size; - jeb->free_size = 0; - - /* Check, if we have a dirty block now, or if it was dirty already */ - if (ISDIRTY (jeb->wasted_size + jeb->dirty_size)) { - c->dirty_size += jeb->wasted_size; - c->wasted_size -= jeb->wasted_size; - jeb->dirty_size += jeb->wasted_size; - jeb->wasted_size = 0; - if (VERYDIRTY(c, jeb->dirty_size)) { - D1(printk(KERN_DEBUG "Adding full erase block at 0x%08x to very_dirty_list (free 0x%08x, dirty 0x%08x, used 0x%08x\n", - jeb->offset, jeb->free_size, jeb->dirty_size, jeb->used_size)); - list_add_tail(&jeb->list, &c->very_dirty_list); - } else { - D1(printk(KERN_DEBUG "Adding full erase block at 0x%08x to dirty_list (free 0x%08x, dirty 0x%08x, used 0x%08x\n", - jeb->offset, jeb->free_size, jeb->dirty_size, jeb->used_size)); - list_add_tail(&jeb->list, &c->dirty_list); - } - } else { - D1(printk(KERN_DEBUG "Adding full erase block at 0x%08x to clean_list (free 0x%08x, dirty 0x%08x, used 0x%08x\n", - jeb->offset, jeb->free_size, jeb->dirty_size, jeb->used_size)); - list_add_tail(&jeb->list, &c->clean_list); + + if (!c->nr_erasing_blocks) { + /* Ouch. We're in GC, or we wouldn't have got here. + And there's no space left. At all. */ + printk(KERN_CRIT "Argh. No free space left for GC. nr_erasing_blocks is %d. nr_free_blocks is %d. (erasableempty: %s, erasingempty: %s, erasependingempty: %s)\n", + c->nr_erasing_blocks, c->nr_free_blocks, list_empty(&c->erasable_list)?"yes":"no", + list_empty(&c->erasing_list)?"yes":"no", list_empty(&c->erase_pending_list)?"yes":"no"); + return -ENOSPC; } - c->nextblock = jeb = NULL; + + spin_unlock(&c->erase_completion_lock); + /* Don't wait for it; just erase one right now */ + jffs2_erase_pending_blocks(c, 1); + spin_lock(&c->erase_completion_lock); + + /* An erase may have failed, decreasing the + amount of free space available. So we must + restart from the beginning */ + return -EAGAIN; } - - if (!jeb) { - struct list_head *next; - /* Take the next block off the 'free' list */ - if (list_empty(&c->free_list)) { + next = c->free_list.next; + list_del(next); + c->nextblock = list_entry(next, struct jffs2_eraseblock, list); + c->nr_free_blocks--; - if (!c->nr_erasing_blocks && - !list_empty(&c->erasable_list)) { - struct jffs2_eraseblock *ejeb; + jffs2_sum_reset_collected(c->summary); /* reset collected summary */ - ejeb = list_entry(c->erasable_list.next, struct jffs2_eraseblock, list); - list_del(&ejeb->list); - list_add_tail(&ejeb->list, &c->erase_pending_list); - c->nr_erasing_blocks++; - jffs2_erase_pending_trigger(c); - D1(printk(KERN_DEBUG "jffs2_do_reserve_space: Triggering erase of erasable block at 0x%08x\n", - ejeb->offset)); + D1(printk(KERN_DEBUG "jffs2_find_nextblock(): new nextblock = 0x%08x\n", c->nextblock->offset)); + + return 0; +} + +/* Called with alloc sem _and_ erase_completion_lock */ +static int jffs2_do_reserve_space(struct jffs2_sb_info *c, uint32_t minsize, uint32_t *ofs, uint32_t *len, uint32_t sumsize) +{ + struct jffs2_eraseblock *jeb = c->nextblock; + uint32_t reserved_size; /* for summary information at the end of the jeb */ + int ret; + + restart: + reserved_size = 0; + + if (jffs2_sum_active() && (sumsize != JFFS2_SUMMARY_NOSUM_SIZE)) { + /* NOSUM_SIZE means not to generate summary */ + + if (jeb) { + reserved_size = PAD(sumsize + c->summary->sum_size + JFFS2_SUMMARY_FRAME_SIZE); + dbg_summary("minsize=%d , jeb->free=%d ," + "summary->size=%d , sumsize=%d\n", + minsize, jeb->free_size, + c->summary->sum_size, sumsize); + } + + /* Is there enough space for writing out the current node, or we have to + write out summary information now, close this jeb and select new nextblock? */ + if (jeb && (PAD(minsize) + PAD(c->summary->sum_size + sumsize + + JFFS2_SUMMARY_FRAME_SIZE) > jeb->free_size)) { + + /* Has summary been disabled for this jeb? */ + if (jffs2_sum_is_disabled(c->summary)) { + sumsize = JFFS2_SUMMARY_NOSUM_SIZE; + goto restart; } - if (!c->nr_erasing_blocks && - !list_empty(&c->erasable_pending_wbuf_list)) { - D1(printk(KERN_DEBUG "jffs2_do_reserve_space: Flushing write buffer\n")); - /* c->nextblock is NULL, no update to c->nextblock allowed */ + /* Writing out the collected summary information */ + dbg_summary("generating summary for 0x%08x.\n", jeb->offset); + ret = jffs2_sum_write_sumnode(c); + + if (ret) + return ret; + + if (jffs2_sum_is_disabled(c->summary)) { + /* jffs2_write_sumnode() couldn't write out the summary information + diabling summary for this jeb and free the collected information + */ + sumsize = JFFS2_SUMMARY_NOSUM_SIZE; + goto restart; + } + + jffs2_close_nextblock(c, jeb); + jeb = NULL; + /* keep always valid value in reserved_size */ + reserved_size = PAD(sumsize + c->summary->sum_size + JFFS2_SUMMARY_FRAME_SIZE); + } + } else { + if (jeb && minsize > jeb->free_size) { + /* Skip the end of this block and file it as having some dirty space */ + /* If there's a pending write to it, flush now */ + + if (jffs2_wbuf_dirty(c)) { spin_unlock(&c->erase_completion_lock); + D1(printk(KERN_DEBUG "jffs2_do_reserve_space: Flushing write buffer\n")); jffs2_flush_wbuf_pad(c); spin_lock(&c->erase_completion_lock); - /* Have another go. It'll be on the erasable_list now */ - return -EAGAIN; + jeb = c->nextblock; + goto restart; } - if (!c->nr_erasing_blocks) { - /* Ouch. We're in GC, or we wouldn't have got here. - And there's no space left. At all. */ - printk(KERN_CRIT "Argh. No free space left for GC. nr_erasing_blocks is %d. nr_free_blocks is %d. (erasableempty: %s, erasingempty: %s, erasependingempty: %s)\n", - c->nr_erasing_blocks, c->nr_free_blocks, list_empty(&c->erasable_list)?"yes":"no", - list_empty(&c->erasing_list)?"yes":"no", list_empty(&c->erase_pending_list)?"yes":"no"); - return -ENOSPC; - } - - spin_unlock(&c->erase_completion_lock); - /* Don't wait for it; just erase one right now */ - jffs2_erase_pending_blocks(c, 1); - spin_lock(&c->erase_completion_lock); + c->wasted_size += jeb->free_size; + c->free_size -= jeb->free_size; + jeb->wasted_size += jeb->free_size; + jeb->free_size = 0; - /* An erase may have failed, decreasing the - amount of free space available. So we must - restart from the beginning */ - return -EAGAIN; + jffs2_close_nextblock(c, jeb); + jeb = NULL; } + } + + if (!jeb) { - next = c->free_list.next; - list_del(next); - c->nextblock = jeb = list_entry(next, struct jffs2_eraseblock, list); - c->nr_free_blocks--; + ret = jffs2_find_nextblock(c); + if (ret) + return ret; + + jeb = c->nextblock; if (jeb->free_size != c->sector_size - c->cleanmarker_size) { printk(KERN_WARNING "Eep. Block 0x%08x taken from free_list had free_size of 0x%08x!!\n", jeb->offset, jeb->free_size); @@ -266,13 +350,13 @@ static int jffs2_do_reserve_space(struct jffs2_sb_info *c, uint32_t minsize, ui /* OK, jeb (==c->nextblock) is now pointing at a block which definitely has enough space */ *ofs = jeb->offset + (c->sector_size - jeb->free_size); - *len = jeb->free_size; + *len = jeb->free_size - reserved_size; if (c->cleanmarker_size && jeb->used_size == c->cleanmarker_size && !jeb->first_node->next_in_ino) { - /* Only node in it beforehand was a CLEANMARKER node (we think). + /* Only node in it beforehand was a CLEANMARKER node (we think). So mark it obsolete now that there's going to be another node - in the block. This will reduce used_size to zero but We've + in the block. This will reduce used_size to zero but We've already set c->nextblock so that jffs2_mark_node_obsolete() won't try to refile it to the dirty_list. */ @@ -292,12 +376,12 @@ static int jffs2_do_reserve_space(struct jffs2_sb_info *c, uint32_t minsize, ui * @len: length of this physical node * @dirty: dirty flag for new node * - * Should only be used to report nodes for which space has been allocated + * Should only be used to report nodes for which space has been allocated * by jffs2_reserve_space. * * Must be called with the alloc_sem held. */ - + int jffs2_add_physical_node_ref(struct jffs2_sb_info *c, struct jffs2_raw_node_ref *new) { struct jffs2_eraseblock *jeb; @@ -349,8 +433,8 @@ int jffs2_add_physical_node_ref(struct jffs2_sb_info *c, struct jffs2_raw_node_r list_add_tail(&jeb->list, &c->clean_list); c->nextblock = NULL; } - ACCT_SANITY_CHECK(c,jeb); - D1(ACCT_PARANOIA_CHECK(jeb)); + jffs2_dbg_acct_sanity_check_nolock(c,jeb); + jffs2_dbg_acct_paranoia_check_nolock(c, jeb); spin_unlock(&c->erase_completion_lock); @@ -404,8 +488,8 @@ void jffs2_mark_node_obsolete(struct jffs2_sb_info *c, struct jffs2_raw_node_ref if (jffs2_can_mark_obsolete(c) && !jffs2_is_readonly(c) && !(c->flags & (JFFS2_SB_FLAG_SCANNING | JFFS2_SB_FLAG_BUILDING))) { - /* Hm. This may confuse static lock analysis. If any of the above - three conditions is false, we're going to return from this + /* Hm. This may confuse static lock analysis. If any of the above + three conditions is false, we're going to return from this function without actually obliterating any nodes or freeing any jffs2_raw_node_refs. So we don't need to stop erases from happening, or protect against people holding an obsolete @@ -430,7 +514,7 @@ void jffs2_mark_node_obsolete(struct jffs2_sb_info *c, struct jffs2_raw_node_ref ref_totlen(c, jeb, ref), blocknr, ref->flash_offset, jeb->used_size); BUG(); }) - D1(printk(KERN_DEBUG "Obsoleting node at 0x%08x of len %x: ", ref_offset(ref), ref_totlen(c, jeb, ref))); + D1(printk(KERN_DEBUG "Obsoleting node at 0x%08x of len %#x: ", ref_offset(ref), ref_totlen(c, jeb, ref))); jeb->used_size -= ref_totlen(c, jeb, ref); c->used_size -= ref_totlen(c, jeb, ref); } @@ -462,18 +546,17 @@ void jffs2_mark_node_obsolete(struct jffs2_sb_info *c, struct jffs2_raw_node_ref D1(printk(KERN_DEBUG "Wasting\n")); addedsize = 0; jeb->wasted_size += ref_totlen(c, jeb, ref); - c->wasted_size += ref_totlen(c, jeb, ref); + c->wasted_size += ref_totlen(c, jeb, ref); } ref->flash_offset = ref_offset(ref) | REF_OBSOLETE; - - ACCT_SANITY_CHECK(c, jeb); - D1(ACCT_PARANOIA_CHECK(jeb)); + jffs2_dbg_acct_sanity_check_nolock(c, jeb); + jffs2_dbg_acct_paranoia_check_nolock(c, jeb); if (c->flags & JFFS2_SB_FLAG_SCANNING) { /* Flash scanning is in progress. Don't muck about with the block lists because they're not ready yet, and don't actually - obliterate nodes that look obsolete. If they weren't + obliterate nodes that look obsolete. If they weren't marked obsolete on the flash at the time they _became_ obsolete, there was probably a reason for that. */ spin_unlock(&c->erase_completion_lock); @@ -507,7 +590,7 @@ void jffs2_mark_node_obsolete(struct jffs2_sb_info *c, struct jffs2_raw_node_ref immediately reused, and we spread the load a bit. */ D1(printk(KERN_DEBUG "...and adding to erasable_list\n")); list_add_tail(&jeb->list, &c->erasable_list); - } + } } D1(printk(KERN_DEBUG "Done OK\n")); } else if (jeb == c->gcblock) { @@ -525,8 +608,8 @@ void jffs2_mark_node_obsolete(struct jffs2_sb_info *c, struct jffs2_raw_node_ref list_add_tail(&jeb->list, &c->very_dirty_list); } else { D1(printk(KERN_DEBUG "Eraseblock at 0x%08x not moved anywhere. (free 0x%08x, dirty 0x%08x, used 0x%08x)\n", - jeb->offset, jeb->free_size, jeb->dirty_size, jeb->used_size)); - } + jeb->offset, jeb->free_size, jeb->dirty_size, jeb->used_size)); + } spin_unlock(&c->erase_completion_lock); @@ -573,11 +656,11 @@ void jffs2_mark_node_obsolete(struct jffs2_sb_info *c, struct jffs2_raw_node_ref /* Nodes which have been marked obsolete no longer need to be associated with any inode. Remove them from the per-inode list. - - Note we can't do this for NAND at the moment because we need + + Note we can't do this for NAND at the moment because we need obsolete dirent nodes to stay on the lists, because of the horridness in jffs2_garbage_collect_deletion_dirent(). Also - because we delete the inocache, and on NAND we need that to + because we delete the inocache, and on NAND we need that to stay around until all the nodes are actually erased, in order to stop us from giving the same inode number to another newly created inode. */ @@ -606,7 +689,7 @@ void jffs2_mark_node_obsolete(struct jffs2_sb_info *c, struct jffs2_raw_node_ref if (ref->next_phys && ref_obsolete(ref->next_phys) && !ref->next_phys->next_in_ino) { struct jffs2_raw_node_ref *n = ref->next_phys; - + spin_lock(&c->erase_completion_lock); ref->__totlen += n->__totlen; @@ -620,7 +703,7 @@ void jffs2_mark_node_obsolete(struct jffs2_sb_info *c, struct jffs2_raw_node_ref jffs2_free_raw_node_ref(n); } - + /* Also merge with the previous node in the list, if there is one and that one is obsolete */ if (ref != jeb->first_node ) { @@ -630,7 +713,7 @@ void jffs2_mark_node_obsolete(struct jffs2_sb_info *c, struct jffs2_raw_node_ref while (p->next_phys != ref) p = p->next_phys; - + if (ref_obsolete(p) && !ref->next_in_ino) { p->__totlen += ref->__totlen; if (jeb->last_node == ref) { @@ -649,164 +732,6 @@ void jffs2_mark_node_obsolete(struct jffs2_sb_info *c, struct jffs2_raw_node_ref up(&c->erase_free_sem); } -#if CONFIG_JFFS2_FS_DEBUG >= 2 -void jffs2_dump_block_lists(struct jffs2_sb_info *c) -{ - - - printk(KERN_DEBUG "jffs2_dump_block_lists:\n"); - printk(KERN_DEBUG "flash_size: %08x\n", c->flash_size); - printk(KERN_DEBUG "used_size: %08x\n", c->used_size); - printk(KERN_DEBUG "dirty_size: %08x\n", c->dirty_size); - printk(KERN_DEBUG "wasted_size: %08x\n", c->wasted_size); - printk(KERN_DEBUG "unchecked_size: %08x\n", c->unchecked_size); - printk(KERN_DEBUG "free_size: %08x\n", c->free_size); - printk(KERN_DEBUG "erasing_size: %08x\n", c->erasing_size); - printk(KERN_DEBUG "bad_size: %08x\n", c->bad_size); - printk(KERN_DEBUG "sector_size: %08x\n", c->sector_size); - printk(KERN_DEBUG "jffs2_reserved_blocks size: %08x\n",c->sector_size * c->resv_blocks_write); - - if (c->nextblock) { - printk(KERN_DEBUG "nextblock: %08x (used %08x, dirty %08x, wasted %08x, unchecked %08x, free %08x)\n", - c->nextblock->offset, c->nextblock->used_size, c->nextblock->dirty_size, c->nextblock->wasted_size, c->nextblock->unchecked_size, c->nextblock->free_size); - } else { - printk(KERN_DEBUG "nextblock: NULL\n"); - } - if (c->gcblock) { - printk(KERN_DEBUG "gcblock: %08x (used %08x, dirty %08x, wasted %08x, unchecked %08x, free %08x)\n", - c->gcblock->offset, c->gcblock->used_size, c->gcblock->dirty_size, c->gcblock->wasted_size, c->gcblock->unchecked_size, c->gcblock->free_size); - } else { - printk(KERN_DEBUG "gcblock: NULL\n"); - } - if (list_empty(&c->clean_list)) { - printk(KERN_DEBUG "clean_list: empty\n"); - } else { - struct list_head *this; - int numblocks = 0; - uint32_t dirty = 0; - - list_for_each(this, &c->clean_list) { - struct jffs2_eraseblock *jeb = list_entry(this, struct jffs2_eraseblock, list); - numblocks ++; - dirty += jeb->wasted_size; - printk(KERN_DEBUG "clean_list: %08x (used %08x, dirty %08x, wasted %08x, unchecked %08x, free %08x)\n", jeb->offset, jeb->used_size, jeb->dirty_size, jeb->wasted_size, jeb->unchecked_size, jeb->free_size); - } - printk (KERN_DEBUG "Contains %d blocks with total wasted size %u, average wasted size: %u\n", numblocks, dirty, dirty / numblocks); - } - if (list_empty(&c->very_dirty_list)) { - printk(KERN_DEBUG "very_dirty_list: empty\n"); - } else { - struct list_head *this; - int numblocks = 0; - uint32_t dirty = 0; - - list_for_each(this, &c->very_dirty_list) { - struct jffs2_eraseblock *jeb = list_entry(this, struct jffs2_eraseblock, list); - numblocks ++; - dirty += jeb->dirty_size; - printk(KERN_DEBUG "very_dirty_list: %08x (used %08x, dirty %08x, wasted %08x, unchecked %08x, free %08x)\n", - jeb->offset, jeb->used_size, jeb->dirty_size, jeb->wasted_size, jeb->unchecked_size, jeb->free_size); - } - printk (KERN_DEBUG "Contains %d blocks with total dirty size %u, average dirty size: %u\n", - numblocks, dirty, dirty / numblocks); - } - if (list_empty(&c->dirty_list)) { - printk(KERN_DEBUG "dirty_list: empty\n"); - } else { - struct list_head *this; - int numblocks = 0; - uint32_t dirty = 0; - - list_for_each(this, &c->dirty_list) { - struct jffs2_eraseblock *jeb = list_entry(this, struct jffs2_eraseblock, list); - numblocks ++; - dirty += jeb->dirty_size; - printk(KERN_DEBUG "dirty_list: %08x (used %08x, dirty %08x, wasted %08x, unchecked %08x, free %08x)\n", - jeb->offset, jeb->used_size, jeb->dirty_size, jeb->wasted_size, jeb->unchecked_size, jeb->free_size); - } - printk (KERN_DEBUG "Contains %d blocks with total dirty size %u, average dirty size: %u\n", - numblocks, dirty, dirty / numblocks); - } - if (list_empty(&c->erasable_list)) { - printk(KERN_DEBUG "erasable_list: empty\n"); - } else { - struct list_head *this; - - list_for_each(this, &c->erasable_list) { - struct jffs2_eraseblock *jeb = list_entry(this, struct jffs2_eraseblock, list); - printk(KERN_DEBUG "erasable_list: %08x (used %08x, dirty %08x, wasted %08x, unchecked %08x, free %08x)\n", - jeb->offset, jeb->used_size, jeb->dirty_size, jeb->wasted_size, jeb->unchecked_size, jeb->free_size); - } - } - if (list_empty(&c->erasing_list)) { - printk(KERN_DEBUG "erasing_list: empty\n"); - } else { - struct list_head *this; - - list_for_each(this, &c->erasing_list) { - struct jffs2_eraseblock *jeb = list_entry(this, struct jffs2_eraseblock, list); - printk(KERN_DEBUG "erasing_list: %08x (used %08x, dirty %08x, wasted %08x, unchecked %08x, free %08x)\n", - jeb->offset, jeb->used_size, jeb->dirty_size, jeb->wasted_size, jeb->unchecked_size, jeb->free_size); - } - } - if (list_empty(&c->erase_pending_list)) { - printk(KERN_DEBUG "erase_pending_list: empty\n"); - } else { - struct list_head *this; - - list_for_each(this, &c->erase_pending_list) { - struct jffs2_eraseblock *jeb = list_entry(this, struct jffs2_eraseblock, list); - printk(KERN_DEBUG "erase_pending_list: %08x (used %08x, dirty %08x, wasted %08x, unchecked %08x, free %08x)\n", - jeb->offset, jeb->used_size, jeb->dirty_size, jeb->wasted_size, jeb->unchecked_size, jeb->free_size); - } - } - if (list_empty(&c->erasable_pending_wbuf_list)) { - printk(KERN_DEBUG "erasable_pending_wbuf_list: empty\n"); - } else { - struct list_head *this; - - list_for_each(this, &c->erasable_pending_wbuf_list) { - struct jffs2_eraseblock *jeb = list_entry(this, struct jffs2_eraseblock, list); - printk(KERN_DEBUG "erasable_pending_wbuf_list: %08x (used %08x, dirty %08x, wasted %08x, unchecked %08x, free %08x)\n", - jeb->offset, jeb->used_size, jeb->dirty_size, jeb->wasted_size, jeb->unchecked_size, jeb->free_size); - } - } - if (list_empty(&c->free_list)) { - printk(KERN_DEBUG "free_list: empty\n"); - } else { - struct list_head *this; - - list_for_each(this, &c->free_list) { - struct jffs2_eraseblock *jeb = list_entry(this, struct jffs2_eraseblock, list); - printk(KERN_DEBUG "free_list: %08x (used %08x, dirty %08x, wasted %08x, unchecked %08x, free %08x)\n", - jeb->offset, jeb->used_size, jeb->dirty_size, jeb->wasted_size, jeb->unchecked_size, jeb->free_size); - } - } - if (list_empty(&c->bad_list)) { - printk(KERN_DEBUG "bad_list: empty\n"); - } else { - struct list_head *this; - - list_for_each(this, &c->bad_list) { - struct jffs2_eraseblock *jeb = list_entry(this, struct jffs2_eraseblock, list); - printk(KERN_DEBUG "bad_list: %08x (used %08x, dirty %08x, wasted %08x, unchecked %08x, free %08x)\n", - jeb->offset, jeb->used_size, jeb->dirty_size, jeb->wasted_size, jeb->unchecked_size, jeb->free_size); - } - } - if (list_empty(&c->bad_used_list)) { - printk(KERN_DEBUG "bad_used_list: empty\n"); - } else { - struct list_head *this; - - list_for_each(this, &c->bad_used_list) { - struct jffs2_eraseblock *jeb = list_entry(this, struct jffs2_eraseblock, list); - printk(KERN_DEBUG "bad_used_list: %08x (used %08x, dirty %08x, wasted %08x, unchecked %08x, free %08x)\n", - jeb->offset, jeb->used_size, jeb->dirty_size, jeb->wasted_size, jeb->unchecked_size, jeb->free_size); - } - } -} -#endif /* CONFIG_JFFS2_FS_DEBUG */ - int jffs2_thread_should_wake(struct jffs2_sb_info *c) { int ret = 0; @@ -828,11 +753,11 @@ int jffs2_thread_should_wake(struct jffs2_sb_info *c) */ dirty = c->dirty_size + c->erasing_size - c->nr_erasing_blocks * c->sector_size; - if (c->nr_free_blocks + c->nr_erasing_blocks < c->resv_blocks_gctrigger && - (dirty > c->nospc_dirty_size)) + if (c->nr_free_blocks + c->nr_erasing_blocks < c->resv_blocks_gctrigger && + (dirty > c->nospc_dirty_size)) ret = 1; - D1(printk(KERN_DEBUG "jffs2_thread_should_wake(): nr_free_blocks %d, nr_erasing_blocks %d, dirty_size 0x%x: %s\n", + D1(printk(KERN_DEBUG "jffs2_thread_should_wake(): nr_free_blocks %d, nr_erasing_blocks %d, dirty_size 0x%x: %s\n", c->nr_free_blocks, c->nr_erasing_blocks, c->dirty_size, ret?"yes":"no")); return ret; diff --git a/fs/jffs2/os-linux.h b/fs/jffs2/os-linux.h index d900c8929b09..59e7a393200c 100644 --- a/fs/jffs2/os-linux.h +++ b/fs/jffs2/os-linux.h @@ -7,7 +7,7 @@ * * For licensing information, see the file 'LICENCE' in this directory. * - * $Id: os-linux.h,v 1.58 2005/07/12 02:34:35 tpoynor Exp $ + * $Id: os-linux.h,v 1.64 2005/09/30 13:59:13 dedekind Exp $ * */ @@ -57,6 +57,7 @@ static inline void jffs2_init_inode_info(struct jffs2_inode_info *f) f->fragtree = RB_ROOT; f->metadata = NULL; f->dents = NULL; + f->target = NULL; f->flags = 0; f->usercompr = 0; } @@ -64,17 +65,24 @@ static inline void jffs2_init_inode_info(struct jffs2_inode_info *f) #define jffs2_is_readonly(c) (OFNI_BS_2SFFJ(c)->s_flags & MS_RDONLY) +#define SECTOR_ADDR(x) ( (((unsigned long)(x) / c->sector_size) * c->sector_size) ) #ifndef CONFIG_JFFS2_FS_WRITEBUFFER -#define SECTOR_ADDR(x) ( ((unsigned long)(x) & ~(c->sector_size-1)) ) + + +#ifdef CONFIG_JFFS2_SUMMARY +#define jffs2_can_mark_obsolete(c) (0) +#else #define jffs2_can_mark_obsolete(c) (1) +#endif + #define jffs2_is_writebuffered(c) (0) #define jffs2_cleanmarker_oob(c) (0) #define jffs2_write_nand_cleanmarker(c,jeb) (-EIO) -#define jffs2_flash_write(c, ofs, len, retlen, buf) ((c)->mtd->write((c)->mtd, ofs, len, retlen, buf)) +#define jffs2_flash_write(c, ofs, len, retlen, buf) jffs2_flash_direct_write(c, ofs, len, retlen, buf) #define jffs2_flash_read(c, ofs, len, retlen, buf) ((c)->mtd->read((c)->mtd, ofs, len, retlen, buf)) -#define jffs2_flush_wbuf_pad(c) ({ (void)(c), 0; }) -#define jffs2_flush_wbuf_gc(c, i) ({ (void)(c), (void) i, 0; }) +#define jffs2_flush_wbuf_pad(c) ({ do{} while(0); (void)(c), 0; }) +#define jffs2_flush_wbuf_gc(c, i) ({ do{} while(0); (void)(c), (void) i, 0; }) #define jffs2_write_nand_badblock(c,jeb,bad_offset) (1) #define jffs2_nand_flash_setup(c) (0) #define jffs2_nand_flash_cleanup(c) do {} while(0) @@ -84,16 +92,26 @@ static inline void jffs2_init_inode_info(struct jffs2_inode_info *f) #define jffs2_wbuf_process NULL #define jffs2_nor_ecc(c) (0) #define jffs2_dataflash(c) (0) +#define jffs2_nor_wbuf_flash(c) (0) #define jffs2_nor_ecc_flash_setup(c) (0) #define jffs2_nor_ecc_flash_cleanup(c) do {} while (0) #define jffs2_dataflash_setup(c) (0) #define jffs2_dataflash_cleanup(c) do {} while (0) +#define jffs2_nor_wbuf_flash_setup(c) (0) +#define jffs2_nor_wbuf_flash_cleanup(c) do {} while (0) #else /* NAND and/or ECC'd NOR support present */ #define jffs2_is_writebuffered(c) (c->wbuf != NULL) -#define SECTOR_ADDR(x) ( ((unsigned long)(x) / (unsigned long)(c->sector_size)) * c->sector_size ) -#define jffs2_can_mark_obsolete(c) ((c->mtd->type == MTD_NORFLASH && !(c->mtd->flags & MTD_ECC)) || c->mtd->type == MTD_RAM) + +#ifdef CONFIG_JFFS2_SUMMARY +#define jffs2_can_mark_obsolete(c) (0) +#else +#define jffs2_can_mark_obsolete(c) \ + ((c->mtd->type == MTD_NORFLASH && !(c->mtd->flags & (MTD_ECC|MTD_PROGRAM_REGIONS))) || \ + c->mtd->type == MTD_RAM) +#endif + #define jffs2_cleanmarker_oob(c) (c->mtd->type == MTD_NANDFLASH) #define jffs2_flash_write_oob(c, ofs, len, retlen, buf) ((c)->mtd->write_oob((c)->mtd, ofs, len, retlen, buf)) @@ -123,6 +141,10 @@ void jffs2_nor_ecc_flash_cleanup(struct jffs2_sb_info *c); int jffs2_dataflash_setup(struct jffs2_sb_info *c); void jffs2_dataflash_cleanup(struct jffs2_sb_info *c); +#define jffs2_nor_wbuf_flash(c) (c->mtd->type == MTD_NORFLASH && (c->mtd->flags & MTD_PROGRAM_REGIONS)) +int jffs2_nor_wbuf_flash_setup(struct jffs2_sb_info *c); +void jffs2_nor_wbuf_flash_cleanup(struct jffs2_sb_info *c); + #endif /* WRITEBUFFER */ /* erase.c */ @@ -169,20 +191,21 @@ void jffs2_gc_release_inode(struct jffs2_sb_info *c, struct jffs2_inode_info *jffs2_gc_fetch_inode(struct jffs2_sb_info *c, int inum, int nlink); -unsigned char *jffs2_gc_fetch_page(struct jffs2_sb_info *c, - struct jffs2_inode_info *f, +unsigned char *jffs2_gc_fetch_page(struct jffs2_sb_info *c, + struct jffs2_inode_info *f, unsigned long offset, unsigned long *priv); void jffs2_gc_release_page(struct jffs2_sb_info *c, unsigned char *pg, unsigned long *priv); void jffs2_flash_cleanup(struct jffs2_sb_info *c); - + /* writev.c */ -int jffs2_flash_direct_writev(struct jffs2_sb_info *c, const struct kvec *vecs, +int jffs2_flash_direct_writev(struct jffs2_sb_info *c, const struct kvec *vecs, unsigned long count, loff_t to, size_t *retlen); - +int jffs2_flash_direct_write(struct jffs2_sb_info *c, loff_t ofs, size_t len, + size_t *retlen, const u_char *buf); #endif /* __JFFS2_OS_LINUX_H__ */ diff --git a/fs/jffs2/read.c b/fs/jffs2/read.c index c7f9068907cf..f3b86da833ba 100644 --- a/fs/jffs2/read.c +++ b/fs/jffs2/read.c @@ -7,7 +7,7 @@ * * For licensing information, see the file 'LICENCE' in this directory. * - * $Id: read.c,v 1.39 2005/03/01 10:34:03 dedekind Exp $ + * $Id: read.c,v 1.42 2005/11/07 11:14:41 gleixner Exp $ * */ @@ -43,7 +43,7 @@ int jffs2_read_dnode(struct jffs2_sb_info *c, struct jffs2_inode_info *f, } if (readlen != sizeof(*ri)) { jffs2_free_raw_inode(ri); - printk(KERN_WARNING "Short read from 0x%08x: wanted 0x%zx bytes, got 0x%zx\n", + printk(KERN_WARNING "Short read from 0x%08x: wanted 0x%zx bytes, got 0x%zx\n", ref_offset(fd->raw), sizeof(*ri), readlen); return -EIO; } @@ -61,7 +61,7 @@ int jffs2_read_dnode(struct jffs2_sb_info *c, struct jffs2_inode_info *f, } /* There was a bug where we wrote hole nodes out with csize/dsize swapped. Deal with it */ - if (ri->compr == JFFS2_COMPR_ZERO && !je32_to_cpu(ri->dsize) && + if (ri->compr == JFFS2_COMPR_ZERO && !je32_to_cpu(ri->dsize) && je32_to_cpu(ri->csize)) { ri->dsize = ri->csize; ri->csize = cpu_to_je32(0); @@ -74,7 +74,7 @@ int jffs2_read_dnode(struct jffs2_sb_info *c, struct jffs2_inode_info *f, goto out_ri; }); - + if (ri->compr == JFFS2_COMPR_ZERO) { memset(buf, 0, len); goto out_ri; @@ -82,8 +82,8 @@ int jffs2_read_dnode(struct jffs2_sb_info *c, struct jffs2_inode_info *f, /* Cases: Reading whole node and it's uncompressed - read directly to buffer provided, check CRC. - Reading whole node and it's compressed - read into comprbuf, check CRC and decompress to buffer provided - Reading partial node and it's uncompressed - read into readbuf, check CRC, and copy + Reading whole node and it's compressed - read into comprbuf, check CRC and decompress to buffer provided + Reading partial node and it's uncompressed - read into readbuf, check CRC, and copy Reading partial node and it's compressed - read into readbuf, check checksum, decompress to decomprbuf and copy */ if (ri->compr == JFFS2_COMPR_NONE && len == je32_to_cpu(ri->dsize)) { @@ -129,7 +129,7 @@ int jffs2_read_dnode(struct jffs2_sb_info *c, struct jffs2_inode_info *f, D2(printk(KERN_DEBUG "Data CRC matches calculated CRC %08x\n", crc)); if (ri->compr != JFFS2_COMPR_NONE) { D2(printk(KERN_DEBUG "Decompress %d bytes from %p to %d bytes at %p\n", - je32_to_cpu(ri->csize), readbuf, je32_to_cpu(ri->dsize), decomprbuf)); + je32_to_cpu(ri->csize), readbuf, je32_to_cpu(ri->dsize), decomprbuf)); ret = jffs2_decompress(c, f, ri->compr | (ri->usercompr << 8), readbuf, decomprbuf, je32_to_cpu(ri->csize), je32_to_cpu(ri->dsize)); if (ret) { printk(KERN_WARNING "Error: jffs2_decompress returned %d\n", ret); @@ -174,7 +174,6 @@ int jffs2_read_inode_range(struct jffs2_sb_info *c, struct jffs2_inode_info *f, if (frag) { D1(printk(KERN_NOTICE "Eep. Hole in ino #%u fraglist. frag->ofs = 0x%08x, offset = 0x%08x\n", f->inocache->ino, frag->ofs, offset)); holesize = min(holesize, frag->ofs - offset); - D2(jffs2_print_frag_list(f)); } D1(printk(KERN_DEBUG "Filling non-frag hole from %d-%d\n", offset, offset+holesize)); memset(buf, 0, holesize); @@ -192,7 +191,7 @@ int jffs2_read_inode_range(struct jffs2_sb_info *c, struct jffs2_inode_info *f, } else { uint32_t readlen; uint32_t fragofs; /* offset within the frag to start reading */ - + fragofs = offset - frag->ofs; readlen = min(frag->size - fragofs, end - offset); D1(printk(KERN_DEBUG "Reading %d-%d from node at 0x%08x (%d)\n", diff --git a/fs/jffs2/readinode.c b/fs/jffs2/readinode.c index 5b2a83599d73..5f0652df5d47 100644 --- a/fs/jffs2/readinode.c +++ b/fs/jffs2/readinode.c @@ -7,11 +7,12 @@ * * For licensing information, see the file 'LICENCE' in this directory. * - * $Id: readinode.c,v 1.125 2005/07/10 13:13:55 dedekind Exp $ + * $Id: readinode.c,v 1.143 2005/11/07 11:14:41 gleixner Exp $ * */ #include <linux/kernel.h> +#include <linux/sched.h> #include <linux/slab.h> #include <linux/fs.h> #include <linux/crc32.h> @@ -20,502 +21,631 @@ #include <linux/compiler.h> #include "nodelist.h" -static int jffs2_add_frag_to_fragtree(struct jffs2_sb_info *c, struct rb_root *list, struct jffs2_node_frag *newfrag); - -#if CONFIG_JFFS2_FS_DEBUG >= 2 -static void jffs2_print_fragtree(struct rb_root *list, int permitbug) +/* + * Put a new tmp_dnode_info into the temporaty RB-tree, keeping the list in + * order of increasing version. + */ +static void jffs2_add_tn_to_tree(struct jffs2_tmp_dnode_info *tn, struct rb_root *list) { - struct jffs2_node_frag *this = frag_first(list); - uint32_t lastofs = 0; - int buggy = 0; - - while(this) { - if (this->node) - printk(KERN_DEBUG "frag %04x-%04x: 0x%08x(%d) on flash (*%p). left (%p), right (%p), parent (%p)\n", - this->ofs, this->ofs+this->size, ref_offset(this->node->raw), ref_flags(this->node->raw), - this, frag_left(this), frag_right(this), frag_parent(this)); - else - printk(KERN_DEBUG "frag %04x-%04x: hole (*%p). left (%p} right (%p), parent (%p)\n", this->ofs, - this->ofs+this->size, this, frag_left(this), frag_right(this), frag_parent(this)); - if (this->ofs != lastofs) - buggy = 1; - lastofs = this->ofs+this->size; - this = frag_next(this); + struct rb_node **p = &list->rb_node; + struct rb_node * parent = NULL; + struct jffs2_tmp_dnode_info *this; + + while (*p) { + parent = *p; + this = rb_entry(parent, struct jffs2_tmp_dnode_info, rb); + + /* There may actually be a collision here, but it doesn't + actually matter. As long as the two nodes with the same + version are together, it's all fine. */ + if (tn->version > this->version) + p = &(*p)->rb_left; + else + p = &(*p)->rb_right; } - if (buggy && !permitbug) { - printk(KERN_CRIT "Frag tree got a hole in it\n"); - BUG(); + + rb_link_node(&tn->rb, parent, p); + rb_insert_color(&tn->rb, list); +} + +static void jffs2_free_tmp_dnode_info_list(struct rb_root *list) +{ + struct rb_node *this; + struct jffs2_tmp_dnode_info *tn; + + this = list->rb_node; + + /* Now at bottom of tree */ + while (this) { + if (this->rb_left) + this = this->rb_left; + else if (this->rb_right) + this = this->rb_right; + else { + tn = rb_entry(this, struct jffs2_tmp_dnode_info, rb); + jffs2_free_full_dnode(tn->fn); + jffs2_free_tmp_dnode_info(tn); + + this = this->rb_parent; + if (!this) + break; + + if (this->rb_left == &tn->rb) + this->rb_left = NULL; + else if (this->rb_right == &tn->rb) + this->rb_right = NULL; + else BUG(); + } } + list->rb_node = NULL; } -void jffs2_print_frag_list(struct jffs2_inode_info *f) +static void jffs2_free_full_dirent_list(struct jffs2_full_dirent *fd) { - jffs2_print_fragtree(&f->fragtree, 0); + struct jffs2_full_dirent *next; - if (f->metadata) { - printk(KERN_DEBUG "metadata at 0x%08x\n", ref_offset(f->metadata->raw)); + while (fd) { + next = fd->next; + jffs2_free_full_dirent(fd); + fd = next; } } -#endif -#if CONFIG_JFFS2_FS_DEBUG >= 1 -static int jffs2_sanitycheck_fragtree(struct jffs2_inode_info *f) +/* Returns first valid node after 'ref'. May return 'ref' */ +static struct jffs2_raw_node_ref *jffs2_first_valid_node(struct jffs2_raw_node_ref *ref) { - struct jffs2_node_frag *frag; - int bitched = 0; - - for (frag = frag_first(&f->fragtree); frag; frag = frag_next(frag)) { + while (ref && ref->next_in_ino) { + if (!ref_obsolete(ref)) + return ref; + dbg_noderef("node at 0x%08x is obsoleted. Ignoring.\n", ref_offset(ref)); + ref = ref->next_in_ino; + } + return NULL; +} - struct jffs2_full_dnode *fn = frag->node; - if (!fn || !fn->raw) - continue; +/* + * Helper function for jffs2_get_inode_nodes(). + * It is called every time an directory entry node is found. + * + * Returns: 0 on succes; + * 1 if the node should be marked obsolete; + * negative error code on failure. + */ +static inline int read_direntry(struct jffs2_sb_info *c, struct jffs2_raw_node_ref *ref, + struct jffs2_raw_dirent *rd, uint32_t read, struct jffs2_full_dirent **fdp, + uint32_t *latest_mctime, uint32_t *mctime_ver) +{ + struct jffs2_full_dirent *fd; + + /* The direntry nodes are checked during the flash scanning */ + BUG_ON(ref_flags(ref) == REF_UNCHECKED); + /* Obsoleted. This cannot happen, surely? dwmw2 20020308 */ + BUG_ON(ref_obsolete(ref)); + + /* Sanity check */ + if (unlikely(PAD((rd->nsize + sizeof(*rd))) != PAD(je32_to_cpu(rd->totlen)))) { + JFFS2_ERROR("illegal nsize in node at %#08x: nsize %#02x, totlen %#04x\n", + ref_offset(ref), rd->nsize, je32_to_cpu(rd->totlen)); + return 1; + } - if (ref_flags(fn->raw) == REF_PRISTINE) { + fd = jffs2_alloc_full_dirent(rd->nsize + 1); + if (unlikely(!fd)) + return -ENOMEM; - if (fn->frags > 1) { - printk(KERN_WARNING "REF_PRISTINE node at 0x%08x had %d frags. Tell dwmw2\n", ref_offset(fn->raw), fn->frags); - bitched = 1; - } - /* A hole node which isn't multi-page should be garbage-collected - and merged anyway, so we just check for the frag size here, - rather than mucking around with actually reading the node - and checking the compression type, which is the real way - to tell a hole node. */ - if (frag->ofs & (PAGE_CACHE_SIZE-1) && frag_prev(frag) && frag_prev(frag)->size < PAGE_CACHE_SIZE && frag_prev(frag)->node) { - printk(KERN_WARNING "REF_PRISTINE node at 0x%08x had a previous non-hole frag in the same page. Tell dwmw2\n", - ref_offset(fn->raw)); - bitched = 1; - } + fd->raw = ref; + fd->version = je32_to_cpu(rd->version); + fd->ino = je32_to_cpu(rd->ino); + fd->type = rd->type; - if ((frag->ofs+frag->size) & (PAGE_CACHE_SIZE-1) && frag_next(frag) && frag_next(frag)->size < PAGE_CACHE_SIZE && frag_next(frag)->node) { - printk(KERN_WARNING "REF_PRISTINE node at 0x%08x (%08x-%08x) had a following non-hole frag in the same page. Tell dwmw2\n", - ref_offset(fn->raw), frag->ofs, frag->ofs+frag->size); - bitched = 1; - } - } + /* Pick out the mctime of the latest dirent */ + if(fd->version > *mctime_ver && je32_to_cpu(rd->mctime)) { + *mctime_ver = fd->version; + *latest_mctime = je32_to_cpu(rd->mctime); } - - if (bitched) { - struct jffs2_node_frag *thisfrag; - - printk(KERN_WARNING "Inode is #%u\n", f->inocache->ino); - thisfrag = frag_first(&f->fragtree); - while (thisfrag) { - if (!thisfrag->node) { - printk("Frag @0x%x-0x%x; node-less hole\n", - thisfrag->ofs, thisfrag->size + thisfrag->ofs); - } else if (!thisfrag->node->raw) { - printk("Frag @0x%x-0x%x; raw-less hole\n", - thisfrag->ofs, thisfrag->size + thisfrag->ofs); - } else { - printk("Frag @0x%x-0x%x; raw at 0x%08x(%d) (0x%x-0x%x)\n", - thisfrag->ofs, thisfrag->size + thisfrag->ofs, - ref_offset(thisfrag->node->raw), ref_flags(thisfrag->node->raw), - thisfrag->node->ofs, thisfrag->node->ofs+thisfrag->node->size); - } - thisfrag = frag_next(thisfrag); - } - } - return bitched; -} -#endif /* D1 */ -static void jffs2_obsolete_node_frag(struct jffs2_sb_info *c, struct jffs2_node_frag *this) -{ - if (this->node) { - this->node->frags--; - if (!this->node->frags) { - /* The node has no valid frags left. It's totally obsoleted */ - D2(printk(KERN_DEBUG "Marking old node @0x%08x (0x%04x-0x%04x) obsolete\n", - ref_offset(this->node->raw), this->node->ofs, this->node->ofs+this->node->size)); - jffs2_mark_node_obsolete(c, this->node->raw); - jffs2_free_full_dnode(this->node); - } else { - D2(printk(KERN_DEBUG "Marking old node @0x%08x (0x%04x-0x%04x) REF_NORMAL. frags is %d\n", - ref_offset(this->node->raw), this->node->ofs, this->node->ofs+this->node->size, - this->node->frags)); - mark_ref_normal(this->node->raw); + /* + * Copy as much of the name as possible from the raw + * dirent we've already read from the flash. + */ + if (read > sizeof(*rd)) + memcpy(&fd->name[0], &rd->name[0], + min_t(uint32_t, rd->nsize, (read - sizeof(*rd)) )); + + /* Do we need to copy any more of the name directly from the flash? */ + if (rd->nsize + sizeof(*rd) > read) { + /* FIXME: point() */ + int err; + int already = read - sizeof(*rd); + + err = jffs2_flash_read(c, (ref_offset(ref)) + read, + rd->nsize - already, &read, &fd->name[already]); + if (unlikely(read != rd->nsize - already) && likely(!err)) + return -EIO; + + if (unlikely(err)) { + JFFS2_ERROR("read remainder of name: error %d\n", err); + jffs2_free_full_dirent(fd); + return -EIO; } - } - jffs2_free_node_frag(this); + + fd->nhash = full_name_hash(fd->name, rd->nsize); + fd->next = NULL; + fd->name[rd->nsize] = '\0'; + + /* + * Wheee. We now have a complete jffs2_full_dirent structure, with + * the name in it and everything. Link it into the list + */ + jffs2_add_fd_to_list(c, fd, fdp); + + return 0; } -/* Given an inode, probably with existing list of fragments, add the new node - * to the fragment list. +/* + * Helper function for jffs2_get_inode_nodes(). + * It is called every time an inode node is found. + * + * Returns: 0 on succes; + * 1 if the node should be marked obsolete; + * negative error code on failure. */ -int jffs2_add_full_dnode_to_inode(struct jffs2_sb_info *c, struct jffs2_inode_info *f, struct jffs2_full_dnode *fn) +static inline int read_dnode(struct jffs2_sb_info *c, struct jffs2_raw_node_ref *ref, + struct jffs2_raw_inode *rd, struct rb_root *tnp, int rdlen, + uint32_t *latest_mctime, uint32_t *mctime_ver) { - int ret; - struct jffs2_node_frag *newfrag; - - D1(printk(KERN_DEBUG "jffs2_add_full_dnode_to_inode(ino #%u, f %p, fn %p)\n", f->inocache->ino, f, fn)); + struct jffs2_tmp_dnode_info *tn; + uint32_t len, csize; + int ret = 1; - if (unlikely(!fn->size)) - return 0; + /* Obsoleted. This cannot happen, surely? dwmw2 20020308 */ + BUG_ON(ref_obsolete(ref)); - newfrag = jffs2_alloc_node_frag(); - if (unlikely(!newfrag)) + tn = jffs2_alloc_tmp_dnode_info(); + if (!tn) { + JFFS2_ERROR("failed to allocate tn (%d bytes).\n", sizeof(*tn)); return -ENOMEM; + } - D2(printk(KERN_DEBUG "adding node %04x-%04x @0x%08x on flash, newfrag *%p\n", - fn->ofs, fn->ofs+fn->size, ref_offset(fn->raw), newfrag)); - - newfrag->ofs = fn->ofs; - newfrag->size = fn->size; - newfrag->node = fn; - newfrag->node->frags = 1; + tn->partial_crc = 0; + csize = je32_to_cpu(rd->csize); - ret = jffs2_add_frag_to_fragtree(c, &f->fragtree, newfrag); - if (ret) - return ret; + /* If we've never checked the CRCs on this node, check them now */ + if (ref_flags(ref) == REF_UNCHECKED) { + uint32_t crc; - /* If we now share a page with other nodes, mark either previous - or next node REF_NORMAL, as appropriate. */ - if (newfrag->ofs & (PAGE_CACHE_SIZE-1)) { - struct jffs2_node_frag *prev = frag_prev(newfrag); + crc = crc32(0, rd, sizeof(*rd) - 8); + if (unlikely(crc != je32_to_cpu(rd->node_crc))) { + JFFS2_NOTICE("header CRC failed on node at %#08x: read %#08x, calculated %#08x\n", + ref_offset(ref), je32_to_cpu(rd->node_crc), crc); + goto free_out; + } - mark_ref_normal(fn->raw); - /* If we don't start at zero there's _always_ a previous */ - if (prev->node) - mark_ref_normal(prev->node->raw); - } + /* Sanity checks */ + if (unlikely(je32_to_cpu(rd->offset) > je32_to_cpu(rd->isize)) || + unlikely(PAD(je32_to_cpu(rd->csize) + sizeof(*rd)) != PAD(je32_to_cpu(rd->totlen)))) { + JFFS2_WARNING("inode node header CRC is corrupted at %#08x\n", ref_offset(ref)); + jffs2_dbg_dump_node(c, ref_offset(ref)); + goto free_out; + } - if ((newfrag->ofs+newfrag->size) & (PAGE_CACHE_SIZE-1)) { - struct jffs2_node_frag *next = frag_next(newfrag); - - if (next) { - mark_ref_normal(fn->raw); - if (next->node) - mark_ref_normal(next->node->raw); + if (jffs2_is_writebuffered(c) && csize != 0) { + /* At this point we are supposed to check the data CRC + * of our unchecked node. But thus far, we do not + * know whether the node is valid or obsolete. To + * figure this out, we need to walk all the nodes of + * the inode and build the inode fragtree. We don't + * want to spend time checking data of nodes which may + * later be found to be obsolete. So we put off the full + * data CRC checking until we have read all the inode + * nodes and have started building the fragtree. + * + * The fragtree is being built starting with nodes + * having the highest version number, so we'll be able + * to detect whether a node is valid (i.e., it is not + * overlapped by a node with higher version) or not. + * And we'll be able to check only those nodes, which + * are not obsolete. + * + * Of course, this optimization only makes sense in case + * of NAND flashes (or other flashes whith + * !jffs2_can_mark_obsolete()), since on NOR flashes + * nodes are marked obsolete physically. + * + * Since NAND flashes (or other flashes with + * jffs2_is_writebuffered(c)) are anyway read by + * fractions of c->wbuf_pagesize, and we have just read + * the node header, it is likely that the starting part + * of the node data is also read when we read the + * header. So we don't mind to check the CRC of the + * starting part of the data of the node now, and check + * the second part later (in jffs2_check_node_data()). + * Of course, we will not need to re-read and re-check + * the NAND page which we have just read. This is why we + * read the whole NAND page at jffs2_get_inode_nodes(), + * while we needed only the node header. + */ + unsigned char *buf; + + /* 'buf' will point to the start of data */ + buf = (unsigned char *)rd + sizeof(*rd); + /* len will be the read data length */ + len = min_t(uint32_t, rdlen - sizeof(*rd), csize); + tn->partial_crc = crc32(0, buf, len); + + dbg_readinode("Calculates CRC (%#08x) for %d bytes, csize %d\n", tn->partial_crc, len, csize); + + /* If we actually calculated the whole data CRC + * and it is wrong, drop the node. */ + if (len >= csize && unlikely(tn->partial_crc != je32_to_cpu(rd->data_crc))) { + JFFS2_NOTICE("wrong data CRC in data node at 0x%08x: read %#08x, calculated %#08x.\n", + ref_offset(ref), tn->partial_crc, je32_to_cpu(rd->data_crc)); + goto free_out; + } + + } else if (csize == 0) { + /* + * We checked the header CRC. If the node has no data, adjust + * the space accounting now. For other nodes this will be done + * later either when the node is marked obsolete or when its + * data is checked. + */ + struct jffs2_eraseblock *jeb; + + dbg_readinode("the node has no data.\n"); + jeb = &c->blocks[ref->flash_offset / c->sector_size]; + len = ref_totlen(c, jeb, ref); + + spin_lock(&c->erase_completion_lock); + jeb->used_size += len; + jeb->unchecked_size -= len; + c->used_size += len; + c->unchecked_size -= len; + ref->flash_offset = ref_offset(ref) | REF_NORMAL; + spin_unlock(&c->erase_completion_lock); } } - D2(if (jffs2_sanitycheck_fragtree(f)) { - printk(KERN_WARNING "Just added node %04x-%04x @0x%08x on flash, newfrag *%p\n", - fn->ofs, fn->ofs+fn->size, ref_offset(fn->raw), newfrag); - return 0; - }) - D2(jffs2_print_frag_list(f)); + + tn->fn = jffs2_alloc_full_dnode(); + if (!tn->fn) { + JFFS2_ERROR("alloc fn failed\n"); + ret = -ENOMEM; + goto free_out; + } + + tn->version = je32_to_cpu(rd->version); + tn->fn->ofs = je32_to_cpu(rd->offset); + tn->data_crc = je32_to_cpu(rd->data_crc); + tn->csize = csize; + tn->fn->raw = ref; + + /* There was a bug where we wrote hole nodes out with + csize/dsize swapped. Deal with it */ + if (rd->compr == JFFS2_COMPR_ZERO && !je32_to_cpu(rd->dsize) && csize) + tn->fn->size = csize; + else // normal case... + tn->fn->size = je32_to_cpu(rd->dsize); + + dbg_readinode("dnode @%08x: ver %u, offset %#04x, dsize %#04x, csize %#04x\n", + ref_offset(ref), je32_to_cpu(rd->version), je32_to_cpu(rd->offset), je32_to_cpu(rd->dsize), csize); + + jffs2_add_tn_to_tree(tn, tnp); + return 0; + +free_out: + jffs2_free_tmp_dnode_info(tn); + return ret; } -/* Doesn't set inode->i_size */ -static int jffs2_add_frag_to_fragtree(struct jffs2_sb_info *c, struct rb_root *list, struct jffs2_node_frag *newfrag) +/* + * Helper function for jffs2_get_inode_nodes(). + * It is called every time an unknown node is found. + * + * Returns: 0 on succes; + * 1 if the node should be marked obsolete; + * negative error code on failure. + */ +static inline int read_unknown(struct jffs2_sb_info *c, struct jffs2_raw_node_ref *ref, struct jffs2_unknown_node *un) { - struct jffs2_node_frag *this; - uint32_t lastend; + /* We don't mark unknown nodes as REF_UNCHECKED */ + BUG_ON(ref_flags(ref) == REF_UNCHECKED); - /* Skip all the nodes which are completed before this one starts */ - this = jffs2_lookup_node_frag(list, newfrag->node->ofs); + un->nodetype = cpu_to_je16(JFFS2_NODE_ACCURATE | je16_to_cpu(un->nodetype)); - if (this) { - D2(printk(KERN_DEBUG "j_a_f_d_t_f: Lookup gave frag 0x%04x-0x%04x; phys 0x%08x (*%p)\n", - this->ofs, this->ofs+this->size, this->node?(ref_offset(this->node->raw)):0xffffffff, this)); - lastend = this->ofs + this->size; + if (crc32(0, un, sizeof(struct jffs2_unknown_node) - 4) != je32_to_cpu(un->hdr_crc)) { + /* Hmmm. This should have been caught at scan time. */ + JFFS2_NOTICE("node header CRC failed at %#08x. But it must have been OK earlier.\n", ref_offset(ref)); + jffs2_dbg_dump_node(c, ref_offset(ref)); + return 1; } else { - D2(printk(KERN_DEBUG "j_a_f_d_t_f: Lookup gave no frag\n")); - lastend = 0; - } - - /* See if we ran off the end of the list */ - if (lastend <= newfrag->ofs) { - /* We did */ - - /* Check if 'this' node was on the same page as the new node. - If so, both 'this' and the new node get marked REF_NORMAL so - the GC can take a look. - */ - if (lastend && (lastend-1) >> PAGE_CACHE_SHIFT == newfrag->ofs >> PAGE_CACHE_SHIFT) { - if (this->node) - mark_ref_normal(this->node->raw); - mark_ref_normal(newfrag->node->raw); - } + switch(je16_to_cpu(un->nodetype) & JFFS2_COMPAT_MASK) { - if (lastend < newfrag->node->ofs) { - /* ... and we need to put a hole in before the new node */ - struct jffs2_node_frag *holefrag = jffs2_alloc_node_frag(); - if (!holefrag) { - jffs2_free_node_frag(newfrag); - return -ENOMEM; - } - holefrag->ofs = lastend; - holefrag->size = newfrag->node->ofs - lastend; - holefrag->node = NULL; - if (this) { - /* By definition, the 'this' node has no right-hand child, - because there are no frags with offset greater than it. - So that's where we want to put the hole */ - D2(printk(KERN_DEBUG "Adding hole frag (%p) on right of node at (%p)\n", holefrag, this)); - rb_link_node(&holefrag->rb, &this->rb, &this->rb.rb_right); - } else { - D2(printk(KERN_DEBUG "Adding hole frag (%p) at root of tree\n", holefrag)); - rb_link_node(&holefrag->rb, NULL, &list->rb_node); - } - rb_insert_color(&holefrag->rb, list); - this = holefrag; - } - if (this) { - /* By definition, the 'this' node has no right-hand child, - because there are no frags with offset greater than it. - So that's where we want to put the hole */ - D2(printk(KERN_DEBUG "Adding new frag (%p) on right of node at (%p)\n", newfrag, this)); - rb_link_node(&newfrag->rb, &this->rb, &this->rb.rb_right); - } else { - D2(printk(KERN_DEBUG "Adding new frag (%p) at root of tree\n", newfrag)); - rb_link_node(&newfrag->rb, NULL, &list->rb_node); + case JFFS2_FEATURE_INCOMPAT: + JFFS2_ERROR("unknown INCOMPAT nodetype %#04X at %#08x\n", + je16_to_cpu(un->nodetype), ref_offset(ref)); + /* EEP */ + BUG(); + break; + + case JFFS2_FEATURE_ROCOMPAT: + JFFS2_ERROR("unknown ROCOMPAT nodetype %#04X at %#08x\n", + je16_to_cpu(un->nodetype), ref_offset(ref)); + BUG_ON(!(c->flags & JFFS2_SB_FLAG_RO)); + break; + + case JFFS2_FEATURE_RWCOMPAT_COPY: + JFFS2_NOTICE("unknown RWCOMPAT_COPY nodetype %#04X at %#08x\n", + je16_to_cpu(un->nodetype), ref_offset(ref)); + break; + + case JFFS2_FEATURE_RWCOMPAT_DELETE: + JFFS2_NOTICE("unknown RWCOMPAT_DELETE nodetype %#04X at %#08x\n", + je16_to_cpu(un->nodetype), ref_offset(ref)); + return 1; } - rb_insert_color(&newfrag->rb, list); - return 0; } - D2(printk(KERN_DEBUG "j_a_f_d_t_f: dealing with frag 0x%04x-0x%04x; phys 0x%08x (*%p)\n", - this->ofs, this->ofs+this->size, this->node?(ref_offset(this->node->raw)):0xffffffff, this)); + return 0; +} - /* OK. 'this' is pointing at the first frag that newfrag->ofs at least partially obsoletes, - * - i.e. newfrag->ofs < this->ofs+this->size && newfrag->ofs >= this->ofs - */ - if (newfrag->ofs > this->ofs) { - /* This node isn't completely obsoleted. The start of it remains valid */ - - /* Mark the new node and the partially covered node REF_NORMAL -- let - the GC take a look at them */ - mark_ref_normal(newfrag->node->raw); - if (this->node) - mark_ref_normal(this->node->raw); - - if (this->ofs + this->size > newfrag->ofs + newfrag->size) { - /* The new node splits 'this' frag into two */ - struct jffs2_node_frag *newfrag2 = jffs2_alloc_node_frag(); - if (!newfrag2) { - jffs2_free_node_frag(newfrag); - return -ENOMEM; - } - D2(printk(KERN_DEBUG "split old frag 0x%04x-0x%04x -->", this->ofs, this->ofs+this->size); - if (this->node) - printk("phys 0x%08x\n", ref_offset(this->node->raw)); - else - printk("hole\n"); - ) - - /* New second frag pointing to this's node */ - newfrag2->ofs = newfrag->ofs + newfrag->size; - newfrag2->size = (this->ofs+this->size) - newfrag2->ofs; - newfrag2->node = this->node; - if (this->node) - this->node->frags++; - - /* Adjust size of original 'this' */ - this->size = newfrag->ofs - this->ofs; - - /* Now, we know there's no node with offset - greater than this->ofs but smaller than - newfrag2->ofs or newfrag->ofs, for obvious - reasons. So we can do a tree insert from - 'this' to insert newfrag, and a tree insert - from newfrag to insert newfrag2. */ - jffs2_fragtree_insert(newfrag, this); - rb_insert_color(&newfrag->rb, list); - - jffs2_fragtree_insert(newfrag2, newfrag); - rb_insert_color(&newfrag2->rb, list); - - return 0; - } - /* New node just reduces 'this' frag in size, doesn't split it */ - this->size = newfrag->ofs - this->ofs; +/* + * Helper function for jffs2_get_inode_nodes(). + * The function detects whether more data should be read and reads it if yes. + * + * Returns: 0 on succes; + * negative error code on failure. + */ +static int read_more(struct jffs2_sb_info *c, struct jffs2_raw_node_ref *ref, + int right_size, int *rdlen, unsigned char *buf, unsigned char *bufstart) +{ + int right_len, err, len; + size_t retlen; + uint32_t offs; - /* Again, we know it lives down here in the tree */ - jffs2_fragtree_insert(newfrag, this); - rb_insert_color(&newfrag->rb, list); - } else { - /* New frag starts at the same point as 'this' used to. Replace - it in the tree without doing a delete and insertion */ - D2(printk(KERN_DEBUG "Inserting newfrag (*%p),%d-%d in before 'this' (*%p),%d-%d\n", - newfrag, newfrag->ofs, newfrag->ofs+newfrag->size, - this, this->ofs, this->ofs+this->size)); - - rb_replace_node(&this->rb, &newfrag->rb, list); - - if (newfrag->ofs + newfrag->size >= this->ofs+this->size) { - D2(printk(KERN_DEBUG "Obsoleting node frag %p (%x-%x)\n", this, this->ofs, this->ofs+this->size)); - jffs2_obsolete_node_frag(c, this); - } else { - this->ofs += newfrag->size; - this->size -= newfrag->size; + if (jffs2_is_writebuffered(c)) { + right_len = c->wbuf_pagesize - (bufstart - buf); + if (right_size + (int)(bufstart - buf) > c->wbuf_pagesize) + right_len += c->wbuf_pagesize; + } else + right_len = right_size; - jffs2_fragtree_insert(this, newfrag); - rb_insert_color(&this->rb, list); - return 0; - } + if (*rdlen == right_len) + return 0; + + /* We need to read more data */ + offs = ref_offset(ref) + *rdlen; + if (jffs2_is_writebuffered(c)) { + bufstart = buf + c->wbuf_pagesize; + len = c->wbuf_pagesize; + } else { + bufstart = buf + *rdlen; + len = right_size - *rdlen; } - /* OK, now we have newfrag added in the correct place in the tree, but - frag_next(newfrag) may be a fragment which is overlapped by it - */ - while ((this = frag_next(newfrag)) && newfrag->ofs + newfrag->size >= this->ofs + this->size) { - /* 'this' frag is obsoleted completely. */ - D2(printk(KERN_DEBUG "Obsoleting node frag %p (%x-%x) and removing from tree\n", this, this->ofs, this->ofs+this->size)); - rb_erase(&this->rb, list); - jffs2_obsolete_node_frag(c, this); + + dbg_readinode("read more %d bytes\n", len); + + err = jffs2_flash_read(c, offs, len, &retlen, bufstart); + if (err) { + JFFS2_ERROR("can not read %d bytes from 0x%08x, " + "error code: %d.\n", len, offs, err); + return err; } - /* Now we're pointing at the first frag which isn't totally obsoleted by - the new frag */ - if (!this || newfrag->ofs + newfrag->size == this->ofs) { - return 0; + if (retlen < len) { + JFFS2_ERROR("short read at %#08x: %d instead of %d.\n", + offs, retlen, len); + return -EIO; } - /* Still some overlap but we don't need to move it in the tree */ - this->size = (this->ofs + this->size) - (newfrag->ofs + newfrag->size); - this->ofs = newfrag->ofs + newfrag->size; - /* And mark them REF_NORMAL so the GC takes a look at them */ - if (this->node) - mark_ref_normal(this->node->raw); - mark_ref_normal(newfrag->node->raw); + *rdlen = right_len; return 0; } -void jffs2_truncate_fraglist (struct jffs2_sb_info *c, struct rb_root *list, uint32_t size) +/* Get tmp_dnode_info and full_dirent for all non-obsolete nodes associated + with this ino, returning the former in order of version */ +static int jffs2_get_inode_nodes(struct jffs2_sb_info *c, struct jffs2_inode_info *f, + struct rb_root *tnp, struct jffs2_full_dirent **fdp, + uint32_t *highest_version, uint32_t *latest_mctime, + uint32_t *mctime_ver) { - struct jffs2_node_frag *frag = jffs2_lookup_node_frag(list, size); + struct jffs2_raw_node_ref *ref, *valid_ref; + struct rb_root ret_tn = RB_ROOT; + struct jffs2_full_dirent *ret_fd = NULL; + unsigned char *buf = NULL; + union jffs2_node_union *node; + size_t retlen; + int len, err; + + *mctime_ver = 0; + + dbg_readinode("ino #%u\n", f->inocache->ino); + + if (jffs2_is_writebuffered(c)) { + /* + * If we have the write buffer, we assume the minimal I/O unit + * is c->wbuf_pagesize. We implement some optimizations which in + * this case and we need a temporary buffer of size = + * 2*c->wbuf_pagesize bytes (see comments in read_dnode()). + * Basically, we want to read not only the node header, but the + * whole wbuf (NAND page in case of NAND) or 2, if the node + * header overlaps the border between the 2 wbufs. + */ + len = 2*c->wbuf_pagesize; + } else { + /* + * When there is no write buffer, the size of the temporary + * buffer is the size of the larges node header. + */ + len = sizeof(union jffs2_node_union); + } - D1(printk(KERN_DEBUG "Truncating fraglist to 0x%08x bytes\n", size)); + /* FIXME: in case of NOR and available ->point() this + * needs to be fixed. */ + buf = kmalloc(len, GFP_KERNEL); + if (!buf) + return -ENOMEM; - /* We know frag->ofs <= size. That's what lookup does for us */ - if (frag && frag->ofs != size) { - if (frag->ofs+frag->size >= size) { - D1(printk(KERN_DEBUG "Truncating frag 0x%08x-0x%08x\n", frag->ofs, frag->ofs+frag->size)); - frag->size = size - frag->ofs; + spin_lock(&c->erase_completion_lock); + valid_ref = jffs2_first_valid_node(f->inocache->nodes); + if (!valid_ref && f->inocache->ino != 1) + JFFS2_WARNING("Eep. No valid nodes for ino #%u.\n", f->inocache->ino); + while (valid_ref) { + unsigned char *bufstart; + + /* We can hold a pointer to a non-obsolete node without the spinlock, + but _obsolete_ nodes may disappear at any time, if the block + they're in gets erased. So if we mark 'ref' obsolete while we're + not holding the lock, it can go away immediately. For that reason, + we find the next valid node first, before processing 'ref'. + */ + ref = valid_ref; + valid_ref = jffs2_first_valid_node(ref->next_in_ino); + spin_unlock(&c->erase_completion_lock); + + cond_resched(); + + /* + * At this point we don't know the type of the node we're going + * to read, so we do not know the size of its header. In order + * to minimize the amount of flash IO we assume the node has + * size = JFFS2_MIN_NODE_HEADER. + */ + if (jffs2_is_writebuffered(c)) { + /* + * We treat 'buf' as 2 adjacent wbufs. We want to + * adjust bufstart such as it points to the + * beginning of the node within this wbuf. + */ + bufstart = buf + (ref_offset(ref) % c->wbuf_pagesize); + /* We will read either one wbuf or 2 wbufs. */ + len = c->wbuf_pagesize - (bufstart - buf); + if (JFFS2_MIN_NODE_HEADER + (int)(bufstart - buf) > c->wbuf_pagesize) { + /* The header spans the border of the first wbuf */ + len += c->wbuf_pagesize; + } + } else { + bufstart = buf; + len = JFFS2_MIN_NODE_HEADER; } - frag = frag_next(frag); - } - while (frag && frag->ofs >= size) { - struct jffs2_node_frag *next = frag_next(frag); - D1(printk(KERN_DEBUG "Removing frag 0x%08x-0x%08x\n", frag->ofs, frag->ofs+frag->size)); - frag_erase(frag, list); - jffs2_obsolete_node_frag(c, frag); - frag = next; - } -} + dbg_readinode("read %d bytes at %#08x(%d).\n", len, ref_offset(ref), ref_flags(ref)); -/* Scan the list of all nodes present for this ino, build map of versions, etc. */ + /* FIXME: point() */ + err = jffs2_flash_read(c, ref_offset(ref), len, + &retlen, bufstart); + if (err) { + JFFS2_ERROR("can not read %d bytes from 0x%08x, " "error code: %d.\n", len, ref_offset(ref), err); + goto free_out; + } -static int jffs2_do_read_inode_internal(struct jffs2_sb_info *c, - struct jffs2_inode_info *f, - struct jffs2_raw_inode *latest_node); + if (retlen < len) { + JFFS2_ERROR("short read at %#08x: %d instead of %d.\n", ref_offset(ref), retlen, len); + err = -EIO; + goto free_out; + } -int jffs2_do_read_inode(struct jffs2_sb_info *c, struct jffs2_inode_info *f, - uint32_t ino, struct jffs2_raw_inode *latest_node) -{ - D2(printk(KERN_DEBUG "jffs2_do_read_inode(): getting inocache\n")); + node = (union jffs2_node_union *)bufstart; - retry_inocache: - spin_lock(&c->inocache_lock); - f->inocache = jffs2_get_ino_cache(c, ino); + switch (je16_to_cpu(node->u.nodetype)) { - D2(printk(KERN_DEBUG "jffs2_do_read_inode(): Got inocache at %p\n", f->inocache)); + case JFFS2_NODETYPE_DIRENT: + + if (JFFS2_MIN_NODE_HEADER < sizeof(struct jffs2_raw_dirent)) { + err = read_more(c, ref, sizeof(struct jffs2_raw_dirent), &len, buf, bufstart); + if (unlikely(err)) + goto free_out; + } + + err = read_direntry(c, ref, &node->d, retlen, &ret_fd, latest_mctime, mctime_ver); + if (err == 1) { + jffs2_mark_node_obsolete(c, ref); + break; + } else if (unlikely(err)) + goto free_out; + + if (je32_to_cpu(node->d.version) > *highest_version) + *highest_version = je32_to_cpu(node->d.version); - if (f->inocache) { - /* Check its state. We may need to wait before we can use it */ - switch(f->inocache->state) { - case INO_STATE_UNCHECKED: - case INO_STATE_CHECKEDABSENT: - f->inocache->state = INO_STATE_READING; break; - - case INO_STATE_CHECKING: - case INO_STATE_GC: - /* If it's in either of these states, we need - to wait for whoever's got it to finish and - put it back. */ - D1(printk(KERN_DEBUG "jffs2_get_ino_cache_read waiting for ino #%u in state %d\n", - ino, f->inocache->state)); - sleep_on_spinunlock(&c->inocache_wq, &c->inocache_lock); - goto retry_inocache; - case INO_STATE_READING: - case INO_STATE_PRESENT: - /* Eep. This should never happen. It can - happen if Linux calls read_inode() again - before clear_inode() has finished though. */ - printk(KERN_WARNING "Eep. Trying to read_inode #%u when it's already in state %d!\n", ino, f->inocache->state); - /* Fail. That's probably better than allowing it to succeed */ - f->inocache = NULL; + case JFFS2_NODETYPE_INODE: + + if (JFFS2_MIN_NODE_HEADER < sizeof(struct jffs2_raw_inode)) { + err = read_more(c, ref, sizeof(struct jffs2_raw_inode), &len, buf, bufstart); + if (unlikely(err)) + goto free_out; + } + + err = read_dnode(c, ref, &node->i, &ret_tn, len, latest_mctime, mctime_ver); + if (err == 1) { + jffs2_mark_node_obsolete(c, ref); + break; + } else if (unlikely(err)) + goto free_out; + + if (je32_to_cpu(node->i.version) > *highest_version) + *highest_version = je32_to_cpu(node->i.version); + break; default: - BUG(); - } - } - spin_unlock(&c->inocache_lock); + if (JFFS2_MIN_NODE_HEADER < sizeof(struct jffs2_unknown_node)) { + err = read_more(c, ref, sizeof(struct jffs2_unknown_node), &len, buf, bufstart); + if (unlikely(err)) + goto free_out; + } + + err = read_unknown(c, ref, &node->u); + if (err == 1) { + jffs2_mark_node_obsolete(c, ref); + break; + } else if (unlikely(err)) + goto free_out; - if (!f->inocache && ino == 1) { - /* Special case - no root inode on medium */ - f->inocache = jffs2_alloc_inode_cache(); - if (!f->inocache) { - printk(KERN_CRIT "jffs2_do_read_inode(): Cannot allocate inocache for root inode\n"); - return -ENOMEM; } - D1(printk(KERN_DEBUG "jffs2_do_read_inode(): Creating inocache for root inode\n")); - memset(f->inocache, 0, sizeof(struct jffs2_inode_cache)); - f->inocache->ino = f->inocache->nlink = 1; - f->inocache->nodes = (struct jffs2_raw_node_ref *)f->inocache; - f->inocache->state = INO_STATE_READING; - jffs2_add_ino_cache(c, f->inocache); - } - if (!f->inocache) { - printk(KERN_WARNING "jffs2_do_read_inode() on nonexistent ino %u\n", ino); - return -ENOENT; + spin_lock(&c->erase_completion_lock); } - return jffs2_do_read_inode_internal(c, f, latest_node); -} - -int jffs2_do_crccheck_inode(struct jffs2_sb_info *c, struct jffs2_inode_cache *ic) -{ - struct jffs2_raw_inode n; - struct jffs2_inode_info *f = kmalloc(sizeof(*f), GFP_KERNEL); - int ret; - - if (!f) - return -ENOMEM; + spin_unlock(&c->erase_completion_lock); + *tnp = ret_tn; + *fdp = ret_fd; + kfree(buf); - memset(f, 0, sizeof(*f)); - init_MUTEX_LOCKED(&f->sem); - f->inocache = ic; + dbg_readinode("nodes of inode #%u were read, the highest version is %u, latest_mctime %u, mctime_ver %u.\n", + f->inocache->ino, *highest_version, *latest_mctime, *mctime_ver); + return 0; - ret = jffs2_do_read_inode_internal(c, f, &n); - if (!ret) { - up(&f->sem); - jffs2_do_clear_inode(c, f); - } - kfree (f); - return ret; + free_out: + jffs2_free_tmp_dnode_info_list(&ret_tn); + jffs2_free_full_dirent_list(ret_fd); + kfree(buf); + return err; } -static int jffs2_do_read_inode_internal(struct jffs2_sb_info *c, +static int jffs2_do_read_inode_internal(struct jffs2_sb_info *c, struct jffs2_inode_info *f, struct jffs2_raw_inode *latest_node) { - struct jffs2_tmp_dnode_info *tn = NULL; + struct jffs2_tmp_dnode_info *tn; struct rb_root tn_list; struct rb_node *rb, *repl_rb; struct jffs2_full_dirent *fd_list; - struct jffs2_full_dnode *fn = NULL; + struct jffs2_full_dnode *fn, *first_fn = NULL; uint32_t crc; uint32_t latest_mctime, mctime_ver; - uint32_t mdata_ver = 0; size_t retlen; int ret; - D1(printk(KERN_DEBUG "jffs2_do_read_inode_internal(): ino #%u nlink is %d\n", f->inocache->ino, f->inocache->nlink)); + dbg_readinode("ino #%u nlink is %d\n", f->inocache->ino, f->inocache->nlink); /* Grab all nodes relevant to this ino */ ret = jffs2_get_inode_nodes(c, f, &tn_list, &fd_list, &f->highest_version, &latest_mctime, &mctime_ver); if (ret) { - printk(KERN_CRIT "jffs2_get_inode_nodes() for ino %u returned %d\n", f->inocache->ino, ret); + JFFS2_ERROR("cannot read nodes for ino %u, returned error is %d\n", f->inocache->ino, ret); if (f->inocache->state == INO_STATE_READING) jffs2_set_inocache_state(c, f->inocache, INO_STATE_CHECKEDABSENT); return ret; @@ -525,42 +655,33 @@ static int jffs2_do_read_inode_internal(struct jffs2_sb_info *c, rb = rb_first(&tn_list); while (rb) { + cond_resched(); tn = rb_entry(rb, struct jffs2_tmp_dnode_info, rb); fn = tn->fn; - - if (f->metadata) { - if (likely(tn->version >= mdata_ver)) { - D1(printk(KERN_DEBUG "Obsoleting old metadata at 0x%08x\n", ref_offset(f->metadata->raw))); - jffs2_mark_node_obsolete(c, f->metadata->raw); - jffs2_free_full_dnode(f->metadata); - f->metadata = NULL; - - mdata_ver = 0; - } else { - /* This should never happen. */ - printk(KERN_WARNING "Er. New metadata at 0x%08x with ver %d is actually older than previous ver %d at 0x%08x\n", - ref_offset(fn->raw), tn->version, mdata_ver, ref_offset(f->metadata->raw)); - jffs2_mark_node_obsolete(c, fn->raw); - jffs2_free_full_dnode(fn); - /* Fill in latest_node from the metadata, not this one we're about to free... */ - fn = f->metadata; - goto next_tn; - } - } + ret = 1; + dbg_readinode("consider node ver %u, phys offset " + "%#08x(%d), range %u-%u.\n", tn->version, + ref_offset(fn->raw), ref_flags(fn->raw), + fn->ofs, fn->ofs + fn->size); if (fn->size) { - jffs2_add_full_dnode_to_inode(c, f, fn); - } else { - /* Zero-sized node at end of version list. Just a metadata update */ - D1(printk(KERN_DEBUG "metadata @%08x: ver %d\n", ref_offset(fn->raw), tn->version)); + ret = jffs2_add_older_frag_to_fragtree(c, f, tn); + /* TODO: the error code isn't checked, check it */ + jffs2_dbg_fragtree_paranoia_check_nolock(f); + BUG_ON(ret < 0); + if (!first_fn && ret == 0) + first_fn = fn; + } else if (!first_fn) { + first_fn = fn; f->metadata = fn; - mdata_ver = tn->version; - } - next_tn: + ret = 0; /* Prevent freeing the metadata update node */ + } else + jffs2_mark_node_obsolete(c, fn->raw); + BUG_ON(rb->rb_left); if (rb->rb_parent && rb->rb_parent->rb_left == rb) { /* We were then left-hand child of our parent. We need - to move our own right-hand child into our place. */ + * to move our own right-hand child into our place. */ repl_rb = rb->rb_right; if (repl_rb) repl_rb->rb_parent = rb->rb_parent; @@ -570,7 +691,7 @@ static int jffs2_do_read_inode_internal(struct jffs2_sb_info *c, rb = rb_next(rb); /* Remove the spent tn from the tree; don't bother rebalancing - but put our right-hand child in our own place. */ + * but put our right-hand child in our own place. */ if (tn->rb.rb_parent) { if (tn->rb.rb_parent->rb_left == &tn->rb) tn->rb.rb_parent->rb_left = repl_rb; @@ -581,19 +702,27 @@ static int jffs2_do_read_inode_internal(struct jffs2_sb_info *c, tn->rb.rb_right->rb_parent = NULL; jffs2_free_tmp_dnode_info(tn); + if (ret) { + dbg_readinode("delete dnode %u-%u.\n", + fn->ofs, fn->ofs + fn->size); + jffs2_free_full_dnode(fn); + } } - D1(jffs2_sanitycheck_fragtree(f)); + jffs2_dbg_fragtree_paranoia_check_nolock(f); - if (!fn) { + BUG_ON(first_fn && ref_obsolete(first_fn->raw)); + + fn = first_fn; + if (unlikely(!first_fn)) { /* No data nodes for this inode. */ if (f->inocache->ino != 1) { - printk(KERN_WARNING "jffs2_do_read_inode(): No data nodes found for ino #%u\n", f->inocache->ino); + JFFS2_WARNING("no data nodes found for ino #%u\n", f->inocache->ino); if (!fd_list) { if (f->inocache->state == INO_STATE_READING) jffs2_set_inocache_state(c, f->inocache, INO_STATE_CHECKEDABSENT); return -EIO; } - printk(KERN_WARNING "jffs2_do_read_inode(): But it has children so we fake some modes for it\n"); + JFFS2_NOTICE("but it has children so we fake some modes for it\n"); } latest_node->mode = cpu_to_jemode(S_IFDIR|S_IRUGO|S_IWUSR|S_IXUGO); latest_node->version = cpu_to_je32(0); @@ -608,8 +737,8 @@ static int jffs2_do_read_inode_internal(struct jffs2_sb_info *c, ret = jffs2_flash_read(c, ref_offset(fn->raw), sizeof(*latest_node), &retlen, (void *)latest_node); if (ret || retlen != sizeof(*latest_node)) { - printk(KERN_NOTICE "MTD read in jffs2_do_read_inode() failed: Returned %d, %zd of %zd bytes read\n", - ret, retlen, sizeof(*latest_node)); + JFFS2_ERROR("failed to read from flash: error %d, %zd of %zd bytes read\n", + ret, retlen, sizeof(*latest_node)); /* FIXME: If this fails, there seems to be a memory leak. Find it. */ up(&f->sem); jffs2_do_clear_inode(c, f); @@ -618,7 +747,8 @@ static int jffs2_do_read_inode_internal(struct jffs2_sb_info *c, crc = crc32(0, latest_node, sizeof(*latest_node)-8); if (crc != je32_to_cpu(latest_node->node_crc)) { - printk(KERN_NOTICE "CRC failed for read_inode of inode %u at physical location 0x%x\n", f->inocache->ino, ref_offset(fn->raw)); + JFFS2_ERROR("CRC failed for read_inode of inode %u at physical location 0x%x\n", + f->inocache->ino, ref_offset(fn->raw)); up(&f->sem); jffs2_do_clear_inode(c, f); return -EIO; @@ -633,10 +763,10 @@ static int jffs2_do_read_inode_internal(struct jffs2_sb_info *c, } break; - + case S_IFREG: /* If it was a regular file, truncate it to the latest node's isize */ - jffs2_truncate_fraglist(c, &f->fragtree, je32_to_cpu(latest_node->isize)); + jffs2_truncate_fragtree(c, &f->fragtree, je32_to_cpu(latest_node->isize)); break; case S_IFLNK: @@ -649,37 +779,33 @@ static int jffs2_do_read_inode_internal(struct jffs2_sb_info *c, if (f->inocache->state != INO_STATE_CHECKING) { /* Symlink's inode data is the target path. Read it and - * keep in RAM to facilitate quick follow symlink operation. - * We use f->dents field to store the target path, which - * is somewhat ugly. */ - f->dents = kmalloc(je32_to_cpu(latest_node->csize) + 1, GFP_KERNEL); - if (!f->dents) { - printk(KERN_WARNING "Can't allocate %d bytes of memory " - "for the symlink target path cache\n", - je32_to_cpu(latest_node->csize)); + * keep in RAM to facilitate quick follow symlink + * operation. */ + f->target = kmalloc(je32_to_cpu(latest_node->csize) + 1, GFP_KERNEL); + if (!f->target) { + JFFS2_ERROR("can't allocate %d bytes of memory for the symlink target path cache\n", je32_to_cpu(latest_node->csize)); up(&f->sem); jffs2_do_clear_inode(c, f); return -ENOMEM; } - + ret = jffs2_flash_read(c, ref_offset(fn->raw) + sizeof(*latest_node), - je32_to_cpu(latest_node->csize), &retlen, (char *)f->dents); - + je32_to_cpu(latest_node->csize), &retlen, (char *)f->target); + if (ret || retlen != je32_to_cpu(latest_node->csize)) { if (retlen != je32_to_cpu(latest_node->csize)) ret = -EIO; - kfree(f->dents); - f->dents = NULL; + kfree(f->target); + f->target = NULL; up(&f->sem); jffs2_do_clear_inode(c, f); return -ret; } - ((char *)f->dents)[je32_to_cpu(latest_node->csize)] = '\0'; - D1(printk(KERN_DEBUG "jffs2_do_read_inode(): symlink's target '%s' cached\n", - (char *)f->dents)); + f->target[je32_to_cpu(latest_node->csize)] = '\0'; + dbg_readinode("symlink's target '%s' cached\n", f->target); } - + /* fall through... */ case S_IFBLK: @@ -687,14 +813,14 @@ static int jffs2_do_read_inode_internal(struct jffs2_sb_info *c, /* Certain inode types should have only one data node, and it's kept as the metadata node */ if (f->metadata) { - printk(KERN_WARNING "Argh. Special inode #%u with mode 0%o had metadata node\n", + JFFS2_ERROR("Argh. Special inode #%u with mode 0%o had metadata node\n", f->inocache->ino, jemode_to_cpu(latest_node->mode)); up(&f->sem); jffs2_do_clear_inode(c, f); return -EIO; } if (!frag_first(&f->fragtree)) { - printk(KERN_WARNING "Argh. Special inode #%u with mode 0%o has no fragments\n", + JFFS2_ERROR("Argh. Special inode #%u with mode 0%o has no fragments\n", f->inocache->ino, jemode_to_cpu(latest_node->mode)); up(&f->sem); jffs2_do_clear_inode(c, f); @@ -702,7 +828,7 @@ static int jffs2_do_read_inode_internal(struct jffs2_sb_info *c, } /* ASSERT: f->fraglist != NULL */ if (frag_next(frag_first(&f->fragtree))) { - printk(KERN_WARNING "Argh. Special inode #%u with mode 0x%x had more than one node\n", + JFFS2_ERROR("Argh. Special inode #%u with mode 0x%x had more than one node\n", f->inocache->ino, jemode_to_cpu(latest_node->mode)); /* FIXME: Deal with it - check crc32, check for duplicate node, check times and discard the older one */ up(&f->sem); @@ -721,6 +847,93 @@ static int jffs2_do_read_inode_internal(struct jffs2_sb_info *c, return 0; } +/* Scan the list of all nodes present for this ino, build map of versions, etc. */ +int jffs2_do_read_inode(struct jffs2_sb_info *c, struct jffs2_inode_info *f, + uint32_t ino, struct jffs2_raw_inode *latest_node) +{ + dbg_readinode("read inode #%u\n", ino); + + retry_inocache: + spin_lock(&c->inocache_lock); + f->inocache = jffs2_get_ino_cache(c, ino); + + if (f->inocache) { + /* Check its state. We may need to wait before we can use it */ + switch(f->inocache->state) { + case INO_STATE_UNCHECKED: + case INO_STATE_CHECKEDABSENT: + f->inocache->state = INO_STATE_READING; + break; + + case INO_STATE_CHECKING: + case INO_STATE_GC: + /* If it's in either of these states, we need + to wait for whoever's got it to finish and + put it back. */ + dbg_readinode("waiting for ino #%u in state %d\n", ino, f->inocache->state); + sleep_on_spinunlock(&c->inocache_wq, &c->inocache_lock); + goto retry_inocache; + + case INO_STATE_READING: + case INO_STATE_PRESENT: + /* Eep. This should never happen. It can + happen if Linux calls read_inode() again + before clear_inode() has finished though. */ + JFFS2_ERROR("Eep. Trying to read_inode #%u when it's already in state %d!\n", ino, f->inocache->state); + /* Fail. That's probably better than allowing it to succeed */ + f->inocache = NULL; + break; + + default: + BUG(); + } + } + spin_unlock(&c->inocache_lock); + + if (!f->inocache && ino == 1) { + /* Special case - no root inode on medium */ + f->inocache = jffs2_alloc_inode_cache(); + if (!f->inocache) { + JFFS2_ERROR("cannot allocate inocache for root inode\n"); + return -ENOMEM; + } + dbg_readinode("creating inocache for root inode\n"); + memset(f->inocache, 0, sizeof(struct jffs2_inode_cache)); + f->inocache->ino = f->inocache->nlink = 1; + f->inocache->nodes = (struct jffs2_raw_node_ref *)f->inocache; + f->inocache->state = INO_STATE_READING; + jffs2_add_ino_cache(c, f->inocache); + } + if (!f->inocache) { + JFFS2_ERROR("requestied to read an nonexistent ino %u\n", ino); + return -ENOENT; + } + + return jffs2_do_read_inode_internal(c, f, latest_node); +} + +int jffs2_do_crccheck_inode(struct jffs2_sb_info *c, struct jffs2_inode_cache *ic) +{ + struct jffs2_raw_inode n; + struct jffs2_inode_info *f = kmalloc(sizeof(*f), GFP_KERNEL); + int ret; + + if (!f) + return -ENOMEM; + + memset(f, 0, sizeof(*f)); + init_MUTEX_LOCKED(&f->sem); + f->inocache = ic; + + ret = jffs2_do_read_inode_internal(c, f, &n); + if (!ret) { + up(&f->sem); + jffs2_do_clear_inode(c, f); + } + kfree (f); + return ret; +} + void jffs2_do_clear_inode(struct jffs2_sb_info *c, struct jffs2_inode_info *f) { struct jffs2_full_dirent *fd, *fds; @@ -740,20 +953,16 @@ void jffs2_do_clear_inode(struct jffs2_sb_info *c, struct jffs2_inode_info *f) jffs2_kill_fragtree(&f->fragtree, deleted?c:NULL); - /* For symlink inodes we us f->dents to store the target path name */ - if (S_ISLNK(OFNI_EDONI_2SFFJ(f)->i_mode)) { - if (f->dents) { - kfree(f->dents); - f->dents = NULL; - } - } else { - fds = f->dents; + if (f->target) { + kfree(f->target); + f->target = NULL; + } - while(fds) { - fd = fds; - fds = fd->next; - jffs2_free_full_dirent(fd); - } + fds = f->dents; + while(fds) { + fd = fds; + fds = fd->next; + jffs2_free_full_dirent(fd); } if (f->inocache && f->inocache->state != INO_STATE_CHECKING) { diff --git a/fs/jffs2/scan.c b/fs/jffs2/scan.c index b63160f83bab..3e51dd1da8aa 100644 --- a/fs/jffs2/scan.c +++ b/fs/jffs2/scan.c @@ -7,7 +7,7 @@ * * For licensing information, see the file 'LICENCE' in this directory. * - * $Id: scan.c,v 1.119 2005/02/17 17:51:13 dedekind Exp $ + * $Id: scan.c,v 1.125 2005/09/30 13:59:13 dedekind Exp $ * */ #include <linux/kernel.h> @@ -18,22 +18,11 @@ #include <linux/crc32.h> #include <linux/compiler.h> #include "nodelist.h" +#include "summary.h" +#include "debug.h" #define DEFAULT_EMPTY_SCAN_SIZE 1024 -#define DIRTY_SPACE(x) do { typeof(x) _x = (x); \ - c->free_size -= _x; c->dirty_size += _x; \ - jeb->free_size -= _x ; jeb->dirty_size += _x; \ - }while(0) -#define USED_SPACE(x) do { typeof(x) _x = (x); \ - c->free_size -= _x; c->used_size += _x; \ - jeb->free_size -= _x ; jeb->used_size += _x; \ - }while(0) -#define UNCHECKED_SPACE(x) do { typeof(x) _x = (x); \ - c->free_size -= _x; c->unchecked_size += _x; \ - jeb->free_size -= _x ; jeb->unchecked_size += _x; \ - }while(0) - #define noisy_printk(noise, args...) do { \ if (*(noise)) { \ printk(KERN_NOTICE args); \ @@ -47,23 +36,16 @@ static uint32_t pseudo_random; static int jffs2_scan_eraseblock (struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb, - unsigned char *buf, uint32_t buf_size); + unsigned char *buf, uint32_t buf_size, struct jffs2_summary *s); -/* These helper functions _must_ increase ofs and also do the dirty/used space accounting. +/* These helper functions _must_ increase ofs and also do the dirty/used space accounting. * Returning an error will abort the mount - bad checksums etc. should just mark the space * as dirty. */ -static int jffs2_scan_inode_node(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb, - struct jffs2_raw_inode *ri, uint32_t ofs); +static int jffs2_scan_inode_node(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb, + struct jffs2_raw_inode *ri, uint32_t ofs, struct jffs2_summary *s); static int jffs2_scan_dirent_node(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb, - struct jffs2_raw_dirent *rd, uint32_t ofs); - -#define BLK_STATE_ALLFF 0 -#define BLK_STATE_CLEAN 1 -#define BLK_STATE_PARTDIRTY 2 -#define BLK_STATE_CLEANMARKER 3 -#define BLK_STATE_ALLDIRTY 4 -#define BLK_STATE_BADBLOCK 5 + struct jffs2_raw_dirent *rd, uint32_t ofs, struct jffs2_summary *s); static inline int min_free(struct jffs2_sb_info *c) { @@ -89,6 +71,7 @@ int jffs2_scan_medium(struct jffs2_sb_info *c) uint32_t empty_blocks = 0, bad_blocks = 0; unsigned char *flashbuf = NULL; uint32_t buf_size = 0; + struct jffs2_summary *s = NULL; /* summary info collected by the scan process */ #ifndef __ECOS size_t pointlen; @@ -122,21 +105,34 @@ int jffs2_scan_medium(struct jffs2_sb_info *c) return -ENOMEM; } + if (jffs2_sum_active()) { + s = kmalloc(sizeof(struct jffs2_summary), GFP_KERNEL); + if (!s) { + JFFS2_WARNING("Can't allocate memory for summary\n"); + return -ENOMEM; + } + memset(s, 0, sizeof(struct jffs2_summary)); + } + for (i=0; i<c->nr_blocks; i++) { struct jffs2_eraseblock *jeb = &c->blocks[i]; - ret = jffs2_scan_eraseblock(c, jeb, buf_size?flashbuf:(flashbuf+jeb->offset), buf_size); + /* reset summary info for next eraseblock scan */ + jffs2_sum_reset_collected(s); + + ret = jffs2_scan_eraseblock(c, jeb, buf_size?flashbuf:(flashbuf+jeb->offset), + buf_size, s); if (ret < 0) goto out; - ACCT_PARANOIA_CHECK(jeb); + jffs2_dbg_acct_paranoia_check_nolock(c, jeb); /* Now decide which list to put it on */ switch(ret) { case BLK_STATE_ALLFF: - /* - * Empty block. Since we can't be sure it + /* + * Empty block. Since we can't be sure it * was entirely erased, we just queue it for erase * again. It will be marked as such when the erase * is complete. Meanwhile we still count it as empty @@ -162,18 +158,18 @@ int jffs2_scan_medium(struct jffs2_sb_info *c) break; case BLK_STATE_CLEAN: - /* Full (or almost full) of clean data. Clean list */ - list_add(&jeb->list, &c->clean_list); + /* Full (or almost full) of clean data. Clean list */ + list_add(&jeb->list, &c->clean_list); break; case BLK_STATE_PARTDIRTY: - /* Some data, but not full. Dirty list. */ - /* We want to remember the block with most free space - and stick it in the 'nextblock' position to start writing to it. */ - if (jeb->free_size > min_free(c) && - (!c->nextblock || c->nextblock->free_size < jeb->free_size)) { - /* Better candidate for the next writes to go to */ - if (c->nextblock) { + /* Some data, but not full. Dirty list. */ + /* We want to remember the block with most free space + and stick it in the 'nextblock' position to start writing to it. */ + if (jeb->free_size > min_free(c) && + (!c->nextblock || c->nextblock->free_size < jeb->free_size)) { + /* Better candidate for the next writes to go to */ + if (c->nextblock) { c->nextblock->dirty_size += c->nextblock->free_size + c->nextblock->wasted_size; c->dirty_size += c->nextblock->free_size + c->nextblock->wasted_size; c->free_size -= c->nextblock->free_size; @@ -184,9 +180,14 @@ int jffs2_scan_medium(struct jffs2_sb_info *c) } else { list_add(&c->nextblock->list, &c->dirty_list); } + /* deleting summary information of the old nextblock */ + jffs2_sum_reset_collected(c->summary); } - c->nextblock = jeb; - } else { + /* update collected summary infromation for the current nextblock */ + jffs2_sum_move_collected(c, s); + D1(printk(KERN_DEBUG "jffs2_scan_medium(): new nextblock = 0x%08x\n", jeb->offset)); + c->nextblock = jeb; + } else { jeb->dirty_size += jeb->free_size + jeb->wasted_size; c->dirty_size += jeb->free_size + jeb->wasted_size; c->free_size -= jeb->free_size; @@ -197,30 +198,33 @@ int jffs2_scan_medium(struct jffs2_sb_info *c) } else { list_add(&jeb->list, &c->dirty_list); } - } + } break; case BLK_STATE_ALLDIRTY: /* Nothing valid - not even a clean marker. Needs erasing. */ - /* For now we just put it on the erasing list. We'll start the erases later */ + /* For now we just put it on the erasing list. We'll start the erases later */ D1(printk(KERN_NOTICE "JFFS2: Erase block at 0x%08x is not formatted. It will be erased\n", jeb->offset)); - list_add(&jeb->list, &c->erase_pending_list); + list_add(&jeb->list, &c->erase_pending_list); c->nr_erasing_blocks++; break; - + case BLK_STATE_BADBLOCK: D1(printk(KERN_NOTICE "JFFS2: Block at 0x%08x is bad\n", jeb->offset)); - list_add(&jeb->list, &c->bad_list); + list_add(&jeb->list, &c->bad_list); c->bad_size += c->sector_size; c->free_size -= c->sector_size; bad_blocks++; break; default: printk(KERN_WARNING "jffs2_scan_medium(): unknown block state\n"); - BUG(); + BUG(); } } - + + if (jffs2_sum_active() && s) + kfree(s); + /* Nextblock dirty is always seen as wasted, because we cannot recycle it now */ if (c->nextblock && (c->nextblock->dirty_size)) { c->nextblock->wasted_size += c->nextblock->dirty_size; @@ -229,12 +233,12 @@ int jffs2_scan_medium(struct jffs2_sb_info *c) c->nextblock->dirty_size = 0; } #ifdef CONFIG_JFFS2_FS_WRITEBUFFER - if (!jffs2_can_mark_obsolete(c) && c->nextblock && (c->nextblock->free_size & (c->wbuf_pagesize-1))) { - /* If we're going to start writing into a block which already + if (!jffs2_can_mark_obsolete(c) && c->nextblock && (c->nextblock->free_size % c->wbuf_pagesize)) { + /* If we're going to start writing into a block which already contains data, and the end of the data isn't page-aligned, skip a little and align it. */ - uint32_t skip = c->nextblock->free_size & (c->wbuf_pagesize-1); + uint32_t skip = c->nextblock->free_size % c->wbuf_pagesize; D1(printk(KERN_DEBUG "jffs2_scan_medium(): Skipping %d bytes in nextblock to ensure page alignment\n", skip)); @@ -246,7 +250,7 @@ int jffs2_scan_medium(struct jffs2_sb_info *c) } #endif if (c->nr_erasing_blocks) { - if ( !c->used_size && ((c->nr_free_blocks+empty_blocks+bad_blocks)!= c->nr_blocks || bad_blocks == c->nr_blocks) ) { + if ( !c->used_size && ((c->nr_free_blocks+empty_blocks+bad_blocks)!= c->nr_blocks || bad_blocks == c->nr_blocks) ) { printk(KERN_NOTICE "Cowardly refusing to erase blocks on filesystem with no valid JFFS2 nodes\n"); printk(KERN_NOTICE "empty_blocks %d, bad_blocks %d, c->nr_blocks %d\n",empty_blocks,bad_blocks,c->nr_blocks); ret = -EIO; @@ -259,13 +263,13 @@ int jffs2_scan_medium(struct jffs2_sb_info *c) if (buf_size) kfree(flashbuf); #ifndef __ECOS - else + else c->mtd->unpoint(c->mtd, flashbuf, 0, c->mtd->size); #endif return ret; } -static int jffs2_fill_scan_buf (struct jffs2_sb_info *c, unsigned char *buf, +int jffs2_fill_scan_buf (struct jffs2_sb_info *c, void *buf, uint32_t ofs, uint32_t len) { int ret; @@ -280,20 +284,39 @@ static int jffs2_fill_scan_buf (struct jffs2_sb_info *c, unsigned char *buf, D1(printk(KERN_WARNING "Read at 0x%x gave only 0x%zx bytes\n", ofs, retlen)); return -EIO; } - D2(printk(KERN_DEBUG "Read 0x%x bytes from 0x%08x into buf\n", len, ofs)); - D2(printk(KERN_DEBUG "000: %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x\n", - buf[0], buf[1], buf[2], buf[3], buf[4], buf[5], buf[6], buf[7], buf[8], buf[9], buf[10], buf[11], buf[12], buf[13], buf[14], buf[15])); return 0; } +int jffs2_scan_classify_jeb(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb) +{ + if ((jeb->used_size + jeb->unchecked_size) == PAD(c->cleanmarker_size) && !jeb->dirty_size + && (!jeb->first_node || !jeb->first_node->next_phys) ) + return BLK_STATE_CLEANMARKER; + + /* move blocks with max 4 byte dirty space to cleanlist */ + else if (!ISDIRTY(c->sector_size - (jeb->used_size + jeb->unchecked_size))) { + c->dirty_size -= jeb->dirty_size; + c->wasted_size += jeb->dirty_size; + jeb->wasted_size += jeb->dirty_size; + jeb->dirty_size = 0; + return BLK_STATE_CLEAN; + } else if (jeb->used_size || jeb->unchecked_size) + return BLK_STATE_PARTDIRTY; + else + return BLK_STATE_ALLDIRTY; +} + static int jffs2_scan_eraseblock (struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb, - unsigned char *buf, uint32_t buf_size) { + unsigned char *buf, uint32_t buf_size, struct jffs2_summary *s) { struct jffs2_unknown_node *node; struct jffs2_unknown_node crcnode; + struct jffs2_sum_marker *sm; uint32_t ofs, prevofs; uint32_t hdr_crc, buf_ofs, buf_len; int err; int noise = 0; + + #ifdef CONFIG_JFFS2_FS_WRITEBUFFER int cleanmarkerfound = 0; #endif @@ -319,17 +342,53 @@ static int jffs2_scan_eraseblock (struct jffs2_sb_info *c, struct jffs2_eraseblo } } #endif + + if (jffs2_sum_active()) { + sm = kmalloc(sizeof(struct jffs2_sum_marker), GFP_KERNEL); + if (!sm) { + return -ENOMEM; + } + + err = jffs2_fill_scan_buf(c, (unsigned char *) sm, jeb->offset + c->sector_size - + sizeof(struct jffs2_sum_marker), sizeof(struct jffs2_sum_marker)); + if (err) { + kfree(sm); + return err; + } + + if (je32_to_cpu(sm->magic) == JFFS2_SUM_MAGIC ) { + err = jffs2_sum_scan_sumnode(c, jeb, je32_to_cpu(sm->offset), &pseudo_random); + if (err) { + kfree(sm); + return err; + } + } + + kfree(sm); + + ofs = jeb->offset; + prevofs = jeb->offset - 1; + } + buf_ofs = jeb->offset; if (!buf_size) { buf_len = c->sector_size; + + if (jffs2_sum_active()) { + /* must reread because of summary test */ + err = jffs2_fill_scan_buf(c, buf, buf_ofs, buf_len); + if (err) + return err; + } + } else { buf_len = EMPTY_SCAN_SIZE(c->sector_size); err = jffs2_fill_scan_buf(c, buf, buf_ofs, buf_len); if (err) return err; } - + /* We temporarily use 'ofs' as a pointer into the buffer/jeb */ ofs = 0; @@ -367,10 +426,12 @@ static int jffs2_scan_eraseblock (struct jffs2_sb_info *c, struct jffs2_eraseblo noise = 10; -scan_more: + dbg_summary("no summary found in jeb 0x%08x. Apply original scan.\n",jeb->offset); + +scan_more: while(ofs < jeb->offset + c->sector_size) { - D1(ACCT_PARANOIA_CHECK(jeb)); + jffs2_dbg_acct_paranoia_check_nolock(c, jeb); cond_resched(); @@ -432,7 +493,7 @@ scan_more: /* If we're only checking the beginning of a block with a cleanmarker, bail now */ - if (buf_ofs == jeb->offset && jeb->used_size == PAD(c->cleanmarker_size) && + if (buf_ofs == jeb->offset && jeb->used_size == PAD(c->cleanmarker_size) && c->cleanmarker_size && !jeb->dirty_size && !jeb->first_node->next_phys) { D1(printk(KERN_DEBUG "%d bytes at start of block seems clean... assuming all clean\n", EMPTY_SCAN_SIZE(c->sector_size))); return BLK_STATE_CLEANMARKER; @@ -441,7 +502,7 @@ scan_more: /* See how much more there is to read in this eraseblock... */ buf_len = min_t(uint32_t, buf_size, jeb->offset + c->sector_size - ofs); if (!buf_len) { - /* No more to read. Break out of main loop without marking + /* No more to read. Break out of main loop without marking this range of empty space as dirty (because it's not) */ D1(printk(KERN_DEBUG "Empty flash at %08x runs to end of block. Treating as free_space\n", empty_start)); @@ -476,8 +537,8 @@ scan_more: } if (je16_to_cpu(node->magic) != JFFS2_MAGIC_BITMASK) { /* OK. We're out of possibilities. Whinge and move on */ - noisy_printk(&noise, "jffs2_scan_eraseblock(): Magic bitmask 0x%04x not found at 0x%08x: 0x%04x instead\n", - JFFS2_MAGIC_BITMASK, ofs, + noisy_printk(&noise, "jffs2_scan_eraseblock(): Magic bitmask 0x%04x not found at 0x%08x: 0x%04x instead\n", + JFFS2_MAGIC_BITMASK, ofs, je16_to_cpu(node->magic)); DIRTY_SPACE(4); ofs += 4; @@ -492,7 +553,7 @@ scan_more: if (hdr_crc != je32_to_cpu(node->hdr_crc)) { noisy_printk(&noise, "jffs2_scan_eraseblock(): Node at 0x%08x {0x%04x, 0x%04x, 0x%08x) has invalid CRC 0x%08x (calculated 0x%08x)\n", ofs, je16_to_cpu(node->magic), - je16_to_cpu(node->nodetype), + je16_to_cpu(node->nodetype), je32_to_cpu(node->totlen), je32_to_cpu(node->hdr_crc), hdr_crc); @@ -501,7 +562,7 @@ scan_more: continue; } - if (ofs + je32_to_cpu(node->totlen) > + if (ofs + je32_to_cpu(node->totlen) > jeb->offset + c->sector_size) { /* Eep. Node goes over the end of the erase block. */ printk(KERN_WARNING "Node at 0x%08x with length 0x%08x would run over the end of the erase block\n", @@ -532,11 +593,11 @@ scan_more: buf_ofs = ofs; node = (void *)buf; } - err = jffs2_scan_inode_node(c, jeb, (void *)node, ofs); + err = jffs2_scan_inode_node(c, jeb, (void *)node, ofs, s); if (err) return err; ofs += PAD(je32_to_cpu(node->totlen)); break; - + case JFFS2_NODETYPE_DIRENT: if (buf_ofs + buf_len < ofs + je32_to_cpu(node->totlen)) { buf_len = min_t(uint32_t, buf_size, jeb->offset + c->sector_size - ofs); @@ -548,7 +609,7 @@ scan_more: buf_ofs = ofs; node = (void *)buf; } - err = jffs2_scan_dirent_node(c, jeb, (void *)node, ofs); + err = jffs2_scan_dirent_node(c, jeb, (void *)node, ofs, s); if (err) return err; ofs += PAD(je32_to_cpu(node->totlen)); break; @@ -556,7 +617,7 @@ scan_more: case JFFS2_NODETYPE_CLEANMARKER: D1(printk(KERN_DEBUG "CLEANMARKER node found at 0x%08x\n", ofs)); if (je32_to_cpu(node->totlen) != c->cleanmarker_size) { - printk(KERN_NOTICE "CLEANMARKER node found at 0x%08x has totlen 0x%x != normal 0x%x\n", + printk(KERN_NOTICE "CLEANMARKER node found at 0x%08x has totlen 0x%x != normal 0x%x\n", ofs, je32_to_cpu(node->totlen), c->cleanmarker_size); DIRTY_SPACE(PAD(sizeof(struct jffs2_unknown_node))); ofs += PAD(sizeof(struct jffs2_unknown_node)); @@ -575,13 +636,15 @@ scan_more: marker_ref->flash_offset = ofs | REF_NORMAL; marker_ref->__totlen = c->cleanmarker_size; jeb->first_node = jeb->last_node = marker_ref; - + USED_SPACE(PAD(c->cleanmarker_size)); ofs += PAD(c->cleanmarker_size); } break; case JFFS2_NODETYPE_PADDING: + if (jffs2_sum_active()) + jffs2_sum_add_padding_mem(s, je32_to_cpu(node->totlen)); DIRTY_SPACE(PAD(je32_to_cpu(node->totlen))); ofs += PAD(je32_to_cpu(node->totlen)); break; @@ -616,8 +679,15 @@ scan_more: } } + if (jffs2_sum_active()) { + if (PAD(s->sum_size + JFFS2_SUMMARY_FRAME_SIZE) > jeb->free_size) { + dbg_summary("There is not enough space for " + "summary information, disabling for this jeb!\n"); + jffs2_sum_disable_collecting(s); + } + } - D1(printk(KERN_DEBUG "Block at 0x%08x: free 0x%08x, dirty 0x%08x, unchecked 0x%08x, used 0x%08x\n", jeb->offset, + D1(printk(KERN_DEBUG "Block at 0x%08x: free 0x%08x, dirty 0x%08x, unchecked 0x%08x, used 0x%08x\n", jeb->offset, jeb->free_size, jeb->dirty_size, jeb->unchecked_size, jeb->used_size)); /* mark_node_obsolete can add to wasted !! */ @@ -628,24 +698,10 @@ scan_more: jeb->wasted_size = 0; } - if ((jeb->used_size + jeb->unchecked_size) == PAD(c->cleanmarker_size) && !jeb->dirty_size - && (!jeb->first_node || !jeb->first_node->next_phys) ) - return BLK_STATE_CLEANMARKER; - - /* move blocks with max 4 byte dirty space to cleanlist */ - else if (!ISDIRTY(c->sector_size - (jeb->used_size + jeb->unchecked_size))) { - c->dirty_size -= jeb->dirty_size; - c->wasted_size += jeb->dirty_size; - jeb->wasted_size += jeb->dirty_size; - jeb->dirty_size = 0; - return BLK_STATE_CLEAN; - } else if (jeb->used_size || jeb->unchecked_size) - return BLK_STATE_PARTDIRTY; - else - return BLK_STATE_ALLDIRTY; + return jffs2_scan_classify_jeb(c, jeb); } -static struct jffs2_inode_cache *jffs2_scan_make_ino_cache(struct jffs2_sb_info *c, uint32_t ino) +struct jffs2_inode_cache *jffs2_scan_make_ino_cache(struct jffs2_sb_info *c, uint32_t ino) { struct jffs2_inode_cache *ic; @@ -671,8 +727,8 @@ static struct jffs2_inode_cache *jffs2_scan_make_ino_cache(struct jffs2_sb_info return ic; } -static int jffs2_scan_inode_node(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb, - struct jffs2_raw_inode *ri, uint32_t ofs) +static int jffs2_scan_inode_node(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb, + struct jffs2_raw_inode *ri, uint32_t ofs, struct jffs2_summary *s) { struct jffs2_raw_node_ref *raw; struct jffs2_inode_cache *ic; @@ -681,11 +737,11 @@ static int jffs2_scan_inode_node(struct jffs2_sb_info *c, struct jffs2_erasebloc D1(printk(KERN_DEBUG "jffs2_scan_inode_node(): Node at 0x%08x\n", ofs)); /* We do very little here now. Just check the ino# to which we should attribute - this node; we can do all the CRC checking etc. later. There's a tradeoff here -- + this node; we can do all the CRC checking etc. later. There's a tradeoff here -- we used to scan the flash once only, reading everything we want from it into memory, then building all our in-core data structures and freeing the extra information. Now we allow the first part of the mount to complete a lot quicker, - but we have to go _back_ to the flash in order to finish the CRC checking, etc. + but we have to go _back_ to the flash in order to finish the CRC checking, etc. Which means that the _full_ amount of time to get to proper write mode with GC operational may actually be _longer_ than before. Sucks to be me. */ @@ -731,7 +787,7 @@ static int jffs2_scan_inode_node(struct jffs2_sb_info *c, struct jffs2_erasebloc jeb->last_node->next_phys = raw; jeb->last_node = raw; - D1(printk(KERN_DEBUG "Node is ino #%u, version %d. Range 0x%x-0x%x\n", + D1(printk(KERN_DEBUG "Node is ino #%u, version %d. Range 0x%x-0x%x\n", je32_to_cpu(ri->ino), je32_to_cpu(ri->version), je32_to_cpu(ri->offset), je32_to_cpu(ri->offset)+je32_to_cpu(ri->dsize))); @@ -739,11 +795,16 @@ static int jffs2_scan_inode_node(struct jffs2_sb_info *c, struct jffs2_erasebloc pseudo_random += je32_to_cpu(ri->version); UNCHECKED_SPACE(PAD(je32_to_cpu(ri->totlen))); + + if (jffs2_sum_active()) { + jffs2_sum_add_inode_mem(s, ri, ofs - jeb->offset); + } + return 0; } -static int jffs2_scan_dirent_node(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb, - struct jffs2_raw_dirent *rd, uint32_t ofs) +static int jffs2_scan_dirent_node(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb, + struct jffs2_raw_dirent *rd, uint32_t ofs, struct jffs2_summary *s) { struct jffs2_raw_node_ref *raw; struct jffs2_full_dirent *fd; @@ -776,7 +837,7 @@ static int jffs2_scan_dirent_node(struct jffs2_sb_info *c, struct jffs2_eraseblo crc = crc32(0, fd->name, rd->nsize); if (crc != je32_to_cpu(rd->name_crc)) { printk(KERN_NOTICE "jffs2_scan_dirent_node(): Name CRC failed on node at 0x%08x: Read 0x%08x, calculated 0x%08x\n", - ofs, je32_to_cpu(rd->name_crc), crc); + ofs, je32_to_cpu(rd->name_crc), crc); D1(printk(KERN_NOTICE "Name for which CRC failed is (now) '%s', ino #%d\n", fd->name, je32_to_cpu(rd->ino))); jffs2_free_full_dirent(fd); /* FIXME: Why do we believe totlen? */ @@ -796,7 +857,7 @@ static int jffs2_scan_dirent_node(struct jffs2_sb_info *c, struct jffs2_eraseblo jffs2_free_raw_node_ref(raw); return -ENOMEM; } - + raw->__totlen = PAD(je32_to_cpu(rd->totlen)); raw->flash_offset = ofs | REF_PRISTINE; raw->next_phys = NULL; @@ -817,6 +878,10 @@ static int jffs2_scan_dirent_node(struct jffs2_sb_info *c, struct jffs2_eraseblo USED_SPACE(PAD(je32_to_cpu(rd->totlen))); jffs2_add_fd_to_list(c, fd, &ic->scan_dents); + if (jffs2_sum_active()) { + jffs2_sum_add_dirent_mem(s, rd, ofs - jeb->offset); + } + return 0; } @@ -852,76 +917,34 @@ void jffs2_rotate_lists(struct jffs2_sb_info *c) x = count_list(&c->clean_list); if (x) { rotateby = pseudo_random % x; - D1(printk(KERN_DEBUG "Rotating clean_list by %d\n", rotateby)); - rotate_list((&c->clean_list), rotateby); - - D1(printk(KERN_DEBUG "Erase block at front of clean_list is at %08x\n", - list_entry(c->clean_list.next, struct jffs2_eraseblock, list)->offset)); - } else { - D1(printk(KERN_DEBUG "Not rotating empty clean_list\n")); } x = count_list(&c->very_dirty_list); if (x) { rotateby = pseudo_random % x; - D1(printk(KERN_DEBUG "Rotating very_dirty_list by %d\n", rotateby)); - rotate_list((&c->very_dirty_list), rotateby); - - D1(printk(KERN_DEBUG "Erase block at front of very_dirty_list is at %08x\n", - list_entry(c->very_dirty_list.next, struct jffs2_eraseblock, list)->offset)); - } else { - D1(printk(KERN_DEBUG "Not rotating empty very_dirty_list\n")); } x = count_list(&c->dirty_list); if (x) { rotateby = pseudo_random % x; - D1(printk(KERN_DEBUG "Rotating dirty_list by %d\n", rotateby)); - rotate_list((&c->dirty_list), rotateby); - - D1(printk(KERN_DEBUG "Erase block at front of dirty_list is at %08x\n", - list_entry(c->dirty_list.next, struct jffs2_eraseblock, list)->offset)); - } else { - D1(printk(KERN_DEBUG "Not rotating empty dirty_list\n")); } x = count_list(&c->erasable_list); if (x) { rotateby = pseudo_random % x; - D1(printk(KERN_DEBUG "Rotating erasable_list by %d\n", rotateby)); - rotate_list((&c->erasable_list), rotateby); - - D1(printk(KERN_DEBUG "Erase block at front of erasable_list is at %08x\n", - list_entry(c->erasable_list.next, struct jffs2_eraseblock, list)->offset)); - } else { - D1(printk(KERN_DEBUG "Not rotating empty erasable_list\n")); } if (c->nr_erasing_blocks) { rotateby = pseudo_random % c->nr_erasing_blocks; - D1(printk(KERN_DEBUG "Rotating erase_pending_list by %d\n", rotateby)); - rotate_list((&c->erase_pending_list), rotateby); - - D1(printk(KERN_DEBUG "Erase block at front of erase_pending_list is at %08x\n", - list_entry(c->erase_pending_list.next, struct jffs2_eraseblock, list)->offset)); - } else { - D1(printk(KERN_DEBUG "Not rotating empty erase_pending_list\n")); } if (c->nr_free_blocks) { rotateby = pseudo_random % c->nr_free_blocks; - D1(printk(KERN_DEBUG "Rotating free_list by %d\n", rotateby)); - rotate_list((&c->free_list), rotateby); - - D1(printk(KERN_DEBUG "Erase block at front of free_list is at %08x\n", - list_entry(c->free_list.next, struct jffs2_eraseblock, list)->offset)); - } else { - D1(printk(KERN_DEBUG "Not rotating empty free_list\n")); } } diff --git a/fs/jffs2/summary.c b/fs/jffs2/summary.c new file mode 100644 index 000000000000..fb9cec61fcf2 --- /dev/null +++ b/fs/jffs2/summary.c @@ -0,0 +1,730 @@ +/* + * JFFS2 -- Journalling Flash File System, Version 2. + * + * Copyright (C) 2004 Ferenc Havasi <havasi@inf.u-szeged.hu>, + * Zoltan Sogor <weth@inf.u-szeged.hu>, + * Patrik Kluba <pajko@halom.u-szeged.hu>, + * University of Szeged, Hungary + * + * For licensing information, see the file 'LICENCE' in this directory. + * + * $Id: summary.c,v 1.4 2005/09/26 11:37:21 havasi Exp $ + * + */ + +#include <linux/kernel.h> +#include <linux/sched.h> +#include <linux/slab.h> +#include <linux/mtd/mtd.h> +#include <linux/pagemap.h> +#include <linux/crc32.h> +#include <linux/compiler.h> +#include <linux/vmalloc.h> +#include "nodelist.h" +#include "debug.h" + +int jffs2_sum_init(struct jffs2_sb_info *c) +{ + c->summary = kmalloc(sizeof(struct jffs2_summary), GFP_KERNEL); + + if (!c->summary) { + JFFS2_WARNING("Can't allocate memory for summary information!\n"); + return -ENOMEM; + } + + memset(c->summary, 0, sizeof(struct jffs2_summary)); + + c->summary->sum_buf = vmalloc(c->sector_size); + + if (!c->summary->sum_buf) { + JFFS2_WARNING("Can't allocate buffer for writing out summary information!\n"); + kfree(c->summary); + return -ENOMEM; + } + + dbg_summary("returned succesfully\n"); + + return 0; +} + +void jffs2_sum_exit(struct jffs2_sb_info *c) +{ + dbg_summary("called\n"); + + jffs2_sum_disable_collecting(c->summary); + + vfree(c->summary->sum_buf); + c->summary->sum_buf = NULL; + + kfree(c->summary); + c->summary = NULL; +} + +static int jffs2_sum_add_mem(struct jffs2_summary *s, union jffs2_sum_mem *item) +{ + if (!s->sum_list_head) + s->sum_list_head = (union jffs2_sum_mem *) item; + if (s->sum_list_tail) + s->sum_list_tail->u.next = (union jffs2_sum_mem *) item; + s->sum_list_tail = (union jffs2_sum_mem *) item; + + switch (je16_to_cpu(item->u.nodetype)) { + case JFFS2_NODETYPE_INODE: + s->sum_size += JFFS2_SUMMARY_INODE_SIZE; + s->sum_num++; + dbg_summary("inode (%u) added to summary\n", + je32_to_cpu(item->i.inode)); + break; + case JFFS2_NODETYPE_DIRENT: + s->sum_size += JFFS2_SUMMARY_DIRENT_SIZE(item->d.nsize); + s->sum_num++; + dbg_summary("dirent (%u) added to summary\n", + je32_to_cpu(item->d.ino)); + break; + default: + JFFS2_WARNING("UNKNOWN node type %u\n", + je16_to_cpu(item->u.nodetype)); + return 1; + } + return 0; +} + + +/* The following 3 functions are called from scan.c to collect summary info for not closed jeb */ + +int jffs2_sum_add_padding_mem(struct jffs2_summary *s, uint32_t size) +{ + dbg_summary("called with %u\n", size); + s->sum_padded += size; + return 0; +} + +int jffs2_sum_add_inode_mem(struct jffs2_summary *s, struct jffs2_raw_inode *ri, + uint32_t ofs) +{ + struct jffs2_sum_inode_mem *temp = kmalloc(sizeof(struct jffs2_sum_inode_mem), GFP_KERNEL); + + if (!temp) + return -ENOMEM; + + temp->nodetype = ri->nodetype; + temp->inode = ri->ino; + temp->version = ri->version; + temp->offset = cpu_to_je32(ofs); /* relative offset from the begining of the jeb */ + temp->totlen = ri->totlen; + temp->next = NULL; + + return jffs2_sum_add_mem(s, (union jffs2_sum_mem *)temp); +} + +int jffs2_sum_add_dirent_mem(struct jffs2_summary *s, struct jffs2_raw_dirent *rd, + uint32_t ofs) +{ + struct jffs2_sum_dirent_mem *temp = + kmalloc(sizeof(struct jffs2_sum_dirent_mem) + rd->nsize, GFP_KERNEL); + + if (!temp) + return -ENOMEM; + + temp->nodetype = rd->nodetype; + temp->totlen = rd->totlen; + temp->offset = cpu_to_je32(ofs); /* relative from the begining of the jeb */ + temp->pino = rd->pino; + temp->version = rd->version; + temp->ino = rd->ino; + temp->nsize = rd->nsize; + temp->type = rd->type; + temp->next = NULL; + + memcpy(temp->name, rd->name, rd->nsize); + + return jffs2_sum_add_mem(s, (union jffs2_sum_mem *)temp); +} + +/* Cleanup every collected summary information */ + +static void jffs2_sum_clean_collected(struct jffs2_summary *s) +{ + union jffs2_sum_mem *temp; + + if (!s->sum_list_head) { + dbg_summary("already empty\n"); + } + while (s->sum_list_head) { + temp = s->sum_list_head; + s->sum_list_head = s->sum_list_head->u.next; + kfree(temp); + } + s->sum_list_tail = NULL; + s->sum_padded = 0; + s->sum_num = 0; +} + +void jffs2_sum_reset_collected(struct jffs2_summary *s) +{ + dbg_summary("called\n"); + jffs2_sum_clean_collected(s); + s->sum_size = 0; +} + +void jffs2_sum_disable_collecting(struct jffs2_summary *s) +{ + dbg_summary("called\n"); + jffs2_sum_clean_collected(s); + s->sum_size = JFFS2_SUMMARY_NOSUM_SIZE; +} + +int jffs2_sum_is_disabled(struct jffs2_summary *s) +{ + return (s->sum_size == JFFS2_SUMMARY_NOSUM_SIZE); +} + +/* Move the collected summary information into sb (called from scan.c) */ + +void jffs2_sum_move_collected(struct jffs2_sb_info *c, struct jffs2_summary *s) +{ + dbg_summary("oldsize=0x%x oldnum=%u => newsize=0x%x newnum=%u\n", + c->summary->sum_size, c->summary->sum_num, + s->sum_size, s->sum_num); + + c->summary->sum_size = s->sum_size; + c->summary->sum_num = s->sum_num; + c->summary->sum_padded = s->sum_padded; + c->summary->sum_list_head = s->sum_list_head; + c->summary->sum_list_tail = s->sum_list_tail; + + s->sum_list_head = s->sum_list_tail = NULL; +} + +/* Called from wbuf.c to collect writed node info */ + +int jffs2_sum_add_kvec(struct jffs2_sb_info *c, const struct kvec *invecs, + unsigned long count, uint32_t ofs) +{ + union jffs2_node_union *node; + struct jffs2_eraseblock *jeb; + + node = invecs[0].iov_base; + jeb = &c->blocks[ofs / c->sector_size]; + ofs -= jeb->offset; + + switch (je16_to_cpu(node->u.nodetype)) { + case JFFS2_NODETYPE_INODE: { + struct jffs2_sum_inode_mem *temp = + kmalloc(sizeof(struct jffs2_sum_inode_mem), GFP_KERNEL); + + if (!temp) + goto no_mem; + + temp->nodetype = node->i.nodetype; + temp->inode = node->i.ino; + temp->version = node->i.version; + temp->offset = cpu_to_je32(ofs); + temp->totlen = node->i.totlen; + temp->next = NULL; + + return jffs2_sum_add_mem(c->summary, (union jffs2_sum_mem *)temp); + } + + case JFFS2_NODETYPE_DIRENT: { + struct jffs2_sum_dirent_mem *temp = + kmalloc(sizeof(struct jffs2_sum_dirent_mem) + node->d.nsize, GFP_KERNEL); + + if (!temp) + goto no_mem; + + temp->nodetype = node->d.nodetype; + temp->totlen = node->d.totlen; + temp->offset = cpu_to_je32(ofs); + temp->pino = node->d.pino; + temp->version = node->d.version; + temp->ino = node->d.ino; + temp->nsize = node->d.nsize; + temp->type = node->d.type; + temp->next = NULL; + + switch (count) { + case 1: + memcpy(temp->name,node->d.name,node->d.nsize); + break; + + case 2: + memcpy(temp->name,invecs[1].iov_base,node->d.nsize); + break; + + default: + BUG(); /* impossible count value */ + break; + } + + return jffs2_sum_add_mem(c->summary, (union jffs2_sum_mem *)temp); + } + + case JFFS2_NODETYPE_PADDING: + dbg_summary("node PADDING\n"); + c->summary->sum_padded += je32_to_cpu(node->u.totlen); + break; + + case JFFS2_NODETYPE_CLEANMARKER: + dbg_summary("node CLEANMARKER\n"); + break; + + case JFFS2_NODETYPE_SUMMARY: + dbg_summary("node SUMMARY\n"); + break; + + default: + /* If you implement a new node type you should also implement + summary support for it or disable summary. + */ + BUG(); + break; + } + + return 0; + +no_mem: + JFFS2_WARNING("MEMORY ALLOCATION ERROR!"); + return -ENOMEM; +} + + +/* Process the stored summary information - helper function for jffs2_sum_scan_sumnode() */ + +static int jffs2_sum_process_sum_data(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb, + struct jffs2_raw_summary *summary, uint32_t *pseudo_random) +{ + struct jffs2_raw_node_ref *raw; + struct jffs2_inode_cache *ic; + struct jffs2_full_dirent *fd; + void *sp; + int i, ino; + + sp = summary->sum; + + for (i=0; i<je32_to_cpu(summary->sum_num); i++) { + dbg_summary("processing summary index %d\n", i); + + switch (je16_to_cpu(((struct jffs2_sum_unknown_flash *)sp)->nodetype)) { + case JFFS2_NODETYPE_INODE: { + struct jffs2_sum_inode_flash *spi; + spi = sp; + + ino = je32_to_cpu(spi->inode); + + dbg_summary("Inode at 0x%08x\n", + jeb->offset + je32_to_cpu(spi->offset)); + + raw = jffs2_alloc_raw_node_ref(); + if (!raw) { + JFFS2_NOTICE("allocation of node reference failed\n"); + kfree(summary); + return -ENOMEM; + } + + ic = jffs2_scan_make_ino_cache(c, ino); + if (!ic) { + JFFS2_NOTICE("scan_make_ino_cache failed\n"); + jffs2_free_raw_node_ref(raw); + kfree(summary); + return -ENOMEM; + } + + raw->flash_offset = (jeb->offset + je32_to_cpu(spi->offset)) | REF_UNCHECKED; + raw->__totlen = PAD(je32_to_cpu(spi->totlen)); + raw->next_phys = NULL; + raw->next_in_ino = ic->nodes; + + ic->nodes = raw; + if (!jeb->first_node) + jeb->first_node = raw; + if (jeb->last_node) + jeb->last_node->next_phys = raw; + jeb->last_node = raw; + *pseudo_random += je32_to_cpu(spi->version); + + UNCHECKED_SPACE(PAD(je32_to_cpu(spi->totlen))); + + sp += JFFS2_SUMMARY_INODE_SIZE; + + break; + } + + case JFFS2_NODETYPE_DIRENT: { + struct jffs2_sum_dirent_flash *spd; + spd = sp; + + dbg_summary("Dirent at 0x%08x\n", + jeb->offset + je32_to_cpu(spd->offset)); + + fd = jffs2_alloc_full_dirent(spd->nsize+1); + if (!fd) { + kfree(summary); + return -ENOMEM; + } + + memcpy(&fd->name, spd->name, spd->nsize); + fd->name[spd->nsize] = 0; + + raw = jffs2_alloc_raw_node_ref(); + if (!raw) { + jffs2_free_full_dirent(fd); + JFFS2_NOTICE("allocation of node reference failed\n"); + kfree(summary); + return -ENOMEM; + } + + ic = jffs2_scan_make_ino_cache(c, je32_to_cpu(spd->pino)); + if (!ic) { + jffs2_free_full_dirent(fd); + jffs2_free_raw_node_ref(raw); + kfree(summary); + return -ENOMEM; + } + + raw->__totlen = PAD(je32_to_cpu(spd->totlen)); + raw->flash_offset = (jeb->offset + je32_to_cpu(spd->offset)) | REF_PRISTINE; + raw->next_phys = NULL; + raw->next_in_ino = ic->nodes; + ic->nodes = raw; + if (!jeb->first_node) + jeb->first_node = raw; + if (jeb->last_node) + jeb->last_node->next_phys = raw; + jeb->last_node = raw; + + fd->raw = raw; + fd->next = NULL; + fd->version = je32_to_cpu(spd->version); + fd->ino = je32_to_cpu(spd->ino); + fd->nhash = full_name_hash(fd->name, spd->nsize); + fd->type = spd->type; + USED_SPACE(PAD(je32_to_cpu(spd->totlen))); + jffs2_add_fd_to_list(c, fd, &ic->scan_dents); + + *pseudo_random += je32_to_cpu(spd->version); + + sp += JFFS2_SUMMARY_DIRENT_SIZE(spd->nsize); + + break; + } + + default : { + JFFS2_WARNING("Unsupported node type found in summary! Exiting..."); + kfree(summary); + return -EIO; + } + } + } + + kfree(summary); + return 0; +} + +/* Process the summary node - called from jffs2_scan_eraseblock() */ + +int jffs2_sum_scan_sumnode(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb, + uint32_t ofs, uint32_t *pseudo_random) +{ + struct jffs2_unknown_node crcnode; + struct jffs2_raw_node_ref *cache_ref; + struct jffs2_raw_summary *summary; + int ret, sumsize; + uint32_t crc; + + sumsize = c->sector_size - ofs; + ofs += jeb->offset; + + dbg_summary("summary found for 0x%08x at 0x%08x (0x%x bytes)\n", + jeb->offset, ofs, sumsize); + + summary = kmalloc(sumsize, GFP_KERNEL); + + if (!summary) { + return -ENOMEM; + } + + ret = jffs2_fill_scan_buf(c, (unsigned char *)summary, ofs, sumsize); + + if (ret) { + kfree(summary); + return ret; + } + + /* OK, now check for node validity and CRC */ + crcnode.magic = cpu_to_je16(JFFS2_MAGIC_BITMASK); + crcnode.nodetype = cpu_to_je16(JFFS2_NODETYPE_SUMMARY); + crcnode.totlen = summary->totlen; + crc = crc32(0, &crcnode, sizeof(crcnode)-4); + + if (je32_to_cpu(summary->hdr_crc) != crc) { + dbg_summary("Summary node header is corrupt (bad CRC or " + "no summary at all)\n"); + goto crc_err; + } + + if (je32_to_cpu(summary->totlen) != sumsize) { + dbg_summary("Summary node is corrupt (wrong erasesize?)\n"); + goto crc_err; + } + + crc = crc32(0, summary, sizeof(struct jffs2_raw_summary)-8); + + if (je32_to_cpu(summary->node_crc) != crc) { + dbg_summary("Summary node is corrupt (bad CRC)\n"); + goto crc_err; + } + + crc = crc32(0, summary->sum, sumsize - sizeof(struct jffs2_raw_summary)); + + if (je32_to_cpu(summary->sum_crc) != crc) { + dbg_summary("Summary node data is corrupt (bad CRC)\n"); + goto crc_err; + } + + if ( je32_to_cpu(summary->cln_mkr) ) { + + dbg_summary("Summary : CLEANMARKER node \n"); + + if (je32_to_cpu(summary->cln_mkr) != c->cleanmarker_size) { + dbg_summary("CLEANMARKER node has totlen 0x%x != normal 0x%x\n", + je32_to_cpu(summary->cln_mkr), c->cleanmarker_size); + UNCHECKED_SPACE(PAD(je32_to_cpu(summary->cln_mkr))); + } else if (jeb->first_node) { + dbg_summary("CLEANMARKER node not first node in block " + "(0x%08x)\n", jeb->offset); + UNCHECKED_SPACE(PAD(je32_to_cpu(summary->cln_mkr))); + } else { + struct jffs2_raw_node_ref *marker_ref = jffs2_alloc_raw_node_ref(); + + if (!marker_ref) { + JFFS2_NOTICE("Failed to allocate node ref for clean marker\n"); + kfree(summary); + return -ENOMEM; + } + + marker_ref->next_in_ino = NULL; + marker_ref->next_phys = NULL; + marker_ref->flash_offset = jeb->offset | REF_NORMAL; + marker_ref->__totlen = je32_to_cpu(summary->cln_mkr); + jeb->first_node = jeb->last_node = marker_ref; + + USED_SPACE( PAD(je32_to_cpu(summary->cln_mkr)) ); + } + } + + if (je32_to_cpu(summary->padded)) { + DIRTY_SPACE(je32_to_cpu(summary->padded)); + } + + ret = jffs2_sum_process_sum_data(c, jeb, summary, pseudo_random); + if (ret) + return ret; + + /* for PARANOIA_CHECK */ + cache_ref = jffs2_alloc_raw_node_ref(); + + if (!cache_ref) { + JFFS2_NOTICE("Failed to allocate node ref for cache\n"); + return -ENOMEM; + } + + cache_ref->next_in_ino = NULL; + cache_ref->next_phys = NULL; + cache_ref->flash_offset = ofs | REF_NORMAL; + cache_ref->__totlen = sumsize; + + if (!jeb->first_node) + jeb->first_node = cache_ref; + if (jeb->last_node) + jeb->last_node->next_phys = cache_ref; + jeb->last_node = cache_ref; + + USED_SPACE(sumsize); + + jeb->wasted_size += jeb->free_size; + c->wasted_size += jeb->free_size; + c->free_size -= jeb->free_size; + jeb->free_size = 0; + + return jffs2_scan_classify_jeb(c, jeb); + +crc_err: + JFFS2_WARNING("Summary node crc error, skipping summary information.\n"); + + return 0; +} + +/* Write summary data to flash - helper function for jffs2_sum_write_sumnode() */ + +static int jffs2_sum_write_data(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb, + uint32_t infosize, uint32_t datasize, int padsize) +{ + struct jffs2_raw_summary isum; + union jffs2_sum_mem *temp; + struct jffs2_sum_marker *sm; + struct kvec vecs[2]; + void *wpage; + int ret; + size_t retlen; + + memset(c->summary->sum_buf, 0xff, datasize); + memset(&isum, 0, sizeof(isum)); + + isum.magic = cpu_to_je16(JFFS2_MAGIC_BITMASK); + isum.nodetype = cpu_to_je16(JFFS2_NODETYPE_SUMMARY); + isum.totlen = cpu_to_je32(infosize); + isum.hdr_crc = cpu_to_je32(crc32(0, &isum, sizeof(struct jffs2_unknown_node) - 4)); + isum.padded = cpu_to_je32(c->summary->sum_padded); + isum.cln_mkr = cpu_to_je32(c->cleanmarker_size); + isum.sum_num = cpu_to_je32(c->summary->sum_num); + wpage = c->summary->sum_buf; + + while (c->summary->sum_num) { + + switch (je16_to_cpu(c->summary->sum_list_head->u.nodetype)) { + case JFFS2_NODETYPE_INODE: { + struct jffs2_sum_inode_flash *sino_ptr = wpage; + + sino_ptr->nodetype = c->summary->sum_list_head->i.nodetype; + sino_ptr->inode = c->summary->sum_list_head->i.inode; + sino_ptr->version = c->summary->sum_list_head->i.version; + sino_ptr->offset = c->summary->sum_list_head->i.offset; + sino_ptr->totlen = c->summary->sum_list_head->i.totlen; + + wpage += JFFS2_SUMMARY_INODE_SIZE; + + break; + } + + case JFFS2_NODETYPE_DIRENT: { + struct jffs2_sum_dirent_flash *sdrnt_ptr = wpage; + + sdrnt_ptr->nodetype = c->summary->sum_list_head->d.nodetype; + sdrnt_ptr->totlen = c->summary->sum_list_head->d.totlen; + sdrnt_ptr->offset = c->summary->sum_list_head->d.offset; + sdrnt_ptr->pino = c->summary->sum_list_head->d.pino; + sdrnt_ptr->version = c->summary->sum_list_head->d.version; + sdrnt_ptr->ino = c->summary->sum_list_head->d.ino; + sdrnt_ptr->nsize = c->summary->sum_list_head->d.nsize; + sdrnt_ptr->type = c->summary->sum_list_head->d.type; + + memcpy(sdrnt_ptr->name, c->summary->sum_list_head->d.name, + c->summary->sum_list_head->d.nsize); + + wpage += JFFS2_SUMMARY_DIRENT_SIZE(c->summary->sum_list_head->d.nsize); + + break; + } + + default : { + BUG(); /* unknown node in summary information */ + } + } + + temp = c->summary->sum_list_head; + c->summary->sum_list_head = c->summary->sum_list_head->u.next; + kfree(temp); + + c->summary->sum_num--; + } + + jffs2_sum_reset_collected(c->summary); + + wpage += padsize; + + sm = wpage; + sm->offset = cpu_to_je32(c->sector_size - jeb->free_size); + sm->magic = cpu_to_je32(JFFS2_SUM_MAGIC); + + isum.sum_crc = cpu_to_je32(crc32(0, c->summary->sum_buf, datasize)); + isum.node_crc = cpu_to_je32(crc32(0, &isum, sizeof(isum) - 8)); + + vecs[0].iov_base = &isum; + vecs[0].iov_len = sizeof(isum); + vecs[1].iov_base = c->summary->sum_buf; + vecs[1].iov_len = datasize; + + dbg_summary("JFFS2: writing out data to flash to pos : 0x%08x\n", + jeb->offset + c->sector_size - jeb->free_size); + + spin_unlock(&c->erase_completion_lock); + ret = jffs2_flash_writev(c, vecs, 2, jeb->offset + c->sector_size - + jeb->free_size, &retlen, 0); + spin_lock(&c->erase_completion_lock); + + + if (ret || (retlen != infosize)) { + JFFS2_WARNING("Write of %zd bytes at 0x%08x failed. returned %d, retlen %zd\n", + infosize, jeb->offset + c->sector_size - jeb->free_size, ret, retlen); + + c->summary->sum_size = JFFS2_SUMMARY_NOSUM_SIZE; + WASTED_SPACE(infosize); + + return 1; + } + + return 0; +} + +/* Write out summary information - called from jffs2_do_reserve_space */ + +int jffs2_sum_write_sumnode(struct jffs2_sb_info *c) +{ + struct jffs2_raw_node_ref *summary_ref; + int datasize, infosize, padsize, ret; + struct jffs2_eraseblock *jeb; + + dbg_summary("called\n"); + + jeb = c->nextblock; + + if (!c->summary->sum_num || !c->summary->sum_list_head) { + JFFS2_WARNING("Empty summary info!!!\n"); + BUG(); + } + + datasize = c->summary->sum_size + sizeof(struct jffs2_sum_marker); + infosize = sizeof(struct jffs2_raw_summary) + datasize; + padsize = jeb->free_size - infosize; + infosize += padsize; + datasize += padsize; + + /* Is there enough space for summary? */ + if (padsize < 0) { + /* don't try to write out summary for this jeb */ + jffs2_sum_disable_collecting(c->summary); + + JFFS2_WARNING("Not enough space for summary, padsize = %d\n", padsize); + return 0; + } + + ret = jffs2_sum_write_data(c, jeb, infosize, datasize, padsize); + if (ret) + return 0; /* can't write out summary, block is marked as NOSUM_SIZE */ + + /* for ACCT_PARANOIA_CHECK */ + spin_unlock(&c->erase_completion_lock); + summary_ref = jffs2_alloc_raw_node_ref(); + spin_lock(&c->erase_completion_lock); + + if (!summary_ref) { + JFFS2_NOTICE("Failed to allocate node ref for summary\n"); + return -ENOMEM; + } + + summary_ref->next_in_ino = NULL; + summary_ref->next_phys = NULL; + summary_ref->flash_offset = (jeb->offset + c->sector_size - jeb->free_size) | REF_NORMAL; + summary_ref->__totlen = infosize; + + if (!jeb->first_node) + jeb->first_node = summary_ref; + if (jeb->last_node) + jeb->last_node->next_phys = summary_ref; + jeb->last_node = summary_ref; + + USED_SPACE(infosize); + + return 0; +} diff --git a/fs/jffs2/summary.h b/fs/jffs2/summary.h new file mode 100644 index 000000000000..b7a678be1709 --- /dev/null +++ b/fs/jffs2/summary.h @@ -0,0 +1,183 @@ +/* + * JFFS2 -- Journalling Flash File System, Version 2. + * + * Copyright (C) 2004 Ferenc Havasi <havasi@inf.u-szeged.hu>, + * Zoltan Sogor <weth@inf.u-szeged.hu>, + * Patrik Kluba <pajko@halom.u-szeged.hu>, + * University of Szeged, Hungary + * + * For licensing information, see the file 'LICENCE' in this directory. + * + * $Id: summary.h,v 1.2 2005/09/26 11:37:21 havasi Exp $ + * + */ + +#ifndef JFFS2_SUMMARY_H +#define JFFS2_SUMMARY_H + +#include <linux/uio.h> +#include <linux/jffs2.h> + +#define DIRTY_SPACE(x) do { typeof(x) _x = (x); \ + c->free_size -= _x; c->dirty_size += _x; \ + jeb->free_size -= _x ; jeb->dirty_size += _x; \ + }while(0) +#define USED_SPACE(x) do { typeof(x) _x = (x); \ + c->free_size -= _x; c->used_size += _x; \ + jeb->free_size -= _x ; jeb->used_size += _x; \ + }while(0) +#define WASTED_SPACE(x) do { typeof(x) _x = (x); \ + c->free_size -= _x; c->wasted_size += _x; \ + jeb->free_size -= _x ; jeb->wasted_size += _x; \ + }while(0) +#define UNCHECKED_SPACE(x) do { typeof(x) _x = (x); \ + c->free_size -= _x; c->unchecked_size += _x; \ + jeb->free_size -= _x ; jeb->unchecked_size += _x; \ + }while(0) + +#define BLK_STATE_ALLFF 0 +#define BLK_STATE_CLEAN 1 +#define BLK_STATE_PARTDIRTY 2 +#define BLK_STATE_CLEANMARKER 3 +#define BLK_STATE_ALLDIRTY 4 +#define BLK_STATE_BADBLOCK 5 + +#define JFFS2_SUMMARY_NOSUM_SIZE 0xffffffff +#define JFFS2_SUMMARY_INODE_SIZE (sizeof(struct jffs2_sum_inode_flash)) +#define JFFS2_SUMMARY_DIRENT_SIZE(x) (sizeof(struct jffs2_sum_dirent_flash) + (x)) + +/* Summary structures used on flash */ + +struct jffs2_sum_unknown_flash +{ + jint16_t nodetype; /* node type */ +}; + +struct jffs2_sum_inode_flash +{ + jint16_t nodetype; /* node type */ + jint32_t inode; /* inode number */ + jint32_t version; /* inode version */ + jint32_t offset; /* offset on jeb */ + jint32_t totlen; /* record length */ +} __attribute__((packed)); + +struct jffs2_sum_dirent_flash +{ + jint16_t nodetype; /* == JFFS_NODETYPE_DIRENT */ + jint32_t totlen; /* record length */ + jint32_t offset; /* offset on jeb */ + jint32_t pino; /* parent inode */ + jint32_t version; /* dirent version */ + jint32_t ino; /* == zero for unlink */ + uint8_t nsize; /* dirent name size */ + uint8_t type; /* dirent type */ + uint8_t name[0]; /* dirent name */ +} __attribute__((packed)); + +union jffs2_sum_flash +{ + struct jffs2_sum_unknown_flash u; + struct jffs2_sum_inode_flash i; + struct jffs2_sum_dirent_flash d; +}; + +/* Summary structures used in the memory */ + +struct jffs2_sum_unknown_mem +{ + union jffs2_sum_mem *next; + jint16_t nodetype; /* node type */ +}; + +struct jffs2_sum_inode_mem +{ + union jffs2_sum_mem *next; + jint16_t nodetype; /* node type */ + jint32_t inode; /* inode number */ + jint32_t version; /* inode version */ + jint32_t offset; /* offset on jeb */ + jint32_t totlen; /* record length */ +} __attribute__((packed)); + +struct jffs2_sum_dirent_mem +{ + union jffs2_sum_mem *next; + jint16_t nodetype; /* == JFFS_NODETYPE_DIRENT */ + jint32_t totlen; /* record length */ + jint32_t offset; /* ofset on jeb */ + jint32_t pino; /* parent inode */ + jint32_t version; /* dirent version */ + jint32_t ino; /* == zero for unlink */ + uint8_t nsize; /* dirent name size */ + uint8_t type; /* dirent type */ + uint8_t name[0]; /* dirent name */ +} __attribute__((packed)); + +union jffs2_sum_mem +{ + struct jffs2_sum_unknown_mem u; + struct jffs2_sum_inode_mem i; + struct jffs2_sum_dirent_mem d; +}; + +/* Summary related information stored in superblock */ + +struct jffs2_summary +{ + uint32_t sum_size; /* collected summary information for nextblock */ + uint32_t sum_num; + uint32_t sum_padded; + union jffs2_sum_mem *sum_list_head; + union jffs2_sum_mem *sum_list_tail; + + jint32_t *sum_buf; /* buffer for writing out summary */ +}; + +/* Summary marker is stored at the end of every sumarized erase block */ + +struct jffs2_sum_marker +{ + jint32_t offset; /* offset of the summary node in the jeb */ + jint32_t magic; /* == JFFS2_SUM_MAGIC */ +}; + +#define JFFS2_SUMMARY_FRAME_SIZE (sizeof(struct jffs2_raw_summary) + sizeof(struct jffs2_sum_marker)) + +#ifdef CONFIG_JFFS2_SUMMARY /* SUMMARY SUPPORT ENABLED */ + +#define jffs2_sum_active() (1) +int jffs2_sum_init(struct jffs2_sb_info *c); +void jffs2_sum_exit(struct jffs2_sb_info *c); +void jffs2_sum_disable_collecting(struct jffs2_summary *s); +int jffs2_sum_is_disabled(struct jffs2_summary *s); +void jffs2_sum_reset_collected(struct jffs2_summary *s); +void jffs2_sum_move_collected(struct jffs2_sb_info *c, struct jffs2_summary *s); +int jffs2_sum_add_kvec(struct jffs2_sb_info *c, const struct kvec *invecs, + unsigned long count, uint32_t to); +int jffs2_sum_write_sumnode(struct jffs2_sb_info *c); +int jffs2_sum_add_padding_mem(struct jffs2_summary *s, uint32_t size); +int jffs2_sum_add_inode_mem(struct jffs2_summary *s, struct jffs2_raw_inode *ri, uint32_t ofs); +int jffs2_sum_add_dirent_mem(struct jffs2_summary *s, struct jffs2_raw_dirent *rd, uint32_t ofs); +int jffs2_sum_scan_sumnode(struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb, + uint32_t ofs, uint32_t *pseudo_random); + +#else /* SUMMARY DISABLED */ + +#define jffs2_sum_active() (0) +#define jffs2_sum_init(a) (0) +#define jffs2_sum_exit(a) +#define jffs2_sum_disable_collecting(a) +#define jffs2_sum_is_disabled(a) (0) +#define jffs2_sum_reset_collected(a) +#define jffs2_sum_add_kvec(a,b,c,d) (0) +#define jffs2_sum_move_collected(a,b) +#define jffs2_sum_write_sumnode(a) (0) +#define jffs2_sum_add_padding_mem(a,b) +#define jffs2_sum_add_inode_mem(a,b,c) +#define jffs2_sum_add_dirent_mem(a,b,c) +#define jffs2_sum_scan_sumnode(a,b,c,d) (0) + +#endif /* CONFIG_JFFS2_SUMMARY */ + +#endif /* JFFS2_SUMMARY_H */ diff --git a/fs/jffs2/super.c b/fs/jffs2/super.c index aaf9475cfb6a..93883817cbd0 100644 --- a/fs/jffs2/super.c +++ b/fs/jffs2/super.c @@ -7,7 +7,7 @@ * * For licensing information, see the file 'LICENCE' in this directory. * - * $Id: super.c,v 1.107 2005/07/12 16:37:08 dedekind Exp $ + * $Id: super.c,v 1.110 2005/11/07 11:14:42 gleixner Exp $ * */ @@ -51,7 +51,7 @@ static void jffs2_i_init_once(void * foo, kmem_cache_t * cachep, unsigned long f if ((flags & (SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR)) == SLAB_CTOR_CONSTRUCTOR) { - init_MUTEX_LOCKED(&ei->sem); + init_MUTEX(&ei->sem); inode_init_once(&ei->vfs_inode); } } @@ -62,7 +62,7 @@ static int jffs2_sync_fs(struct super_block *sb, int wait) down(&c->alloc_sem); jffs2_flush_wbuf_pad(c); - up(&c->alloc_sem); + up(&c->alloc_sem); return 0; } @@ -112,7 +112,7 @@ static int jffs2_sb_set(struct super_block *sb, void *data) } static struct super_block *jffs2_get_sb_mtd(struct file_system_type *fs_type, - int flags, const char *dev_name, + int flags, const char *dev_name, void *data, struct mtd_info *mtd) { struct super_block *sb; @@ -172,7 +172,7 @@ static struct super_block *jffs2_get_sb_mtd(struct file_system_type *fs_type, } static struct super_block *jffs2_get_sb_mtdnr(struct file_system_type *fs_type, - int flags, const char *dev_name, + int flags, const char *dev_name, void *data, int mtdnr) { struct mtd_info *mtd; @@ -201,7 +201,7 @@ static struct super_block *jffs2_get_sb(struct file_system_type *fs_type, /* The preferred way of mounting in future; especially when CONFIG_BLK_DEV is implemented - we specify the underlying - MTD device by number or by name, so that we don't require + MTD device by number or by name, so that we don't require block device support to be present in the kernel. */ /* FIXME: How to do the root fs this way? */ @@ -225,7 +225,7 @@ static struct super_block *jffs2_get_sb(struct file_system_type *fs_type, } else if (isdigit(dev_name[3])) { /* Mount by MTD device number name */ char *endptr; - + mtdnr = simple_strtoul(dev_name+3, &endptr, 0); if (!*endptr) { /* It was a valid number */ @@ -235,7 +235,7 @@ static struct super_block *jffs2_get_sb(struct file_system_type *fs_type, } } - /* Try the old way - the hack where we allowed users to mount + /* Try the old way - the hack where we allowed users to mount /dev/mtdblock$(n) but didn't actually _use_ the blkdev */ err = path_lookup(dev_name, LOOKUP_FOLLOW, &nd); @@ -282,9 +282,12 @@ static void jffs2_put_super (struct super_block *sb) down(&c->alloc_sem); jffs2_flush_wbuf_pad(c); up(&c->alloc_sem); + + jffs2_sum_exit(c); + jffs2_free_ino_caches(c); jffs2_free_raw_node_refs(c); - if (c->mtd->flags & MTD_NO_VIRTBLOCKS) + if (jffs2_blocks_use_vmalloc(c)) vfree(c->blocks); else kfree(c->blocks); @@ -321,6 +324,9 @@ static int __init init_jffs2_fs(void) #ifdef CONFIG_JFFS2_FS_WRITEBUFFER " (NAND)" #endif +#ifdef CONFIG_JFFS2_SUMMARY + " (SUMMARY) " +#endif " (C) 2001-2003 Red Hat, Inc.\n"); jffs2_inode_cachep = kmem_cache_create("jffs2_i", @@ -370,5 +376,5 @@ module_exit(exit_jffs2_fs); MODULE_DESCRIPTION("The Journalling Flash File System, v2"); MODULE_AUTHOR("Red Hat, Inc."); -MODULE_LICENSE("GPL"); // Actually dual-licensed, but it doesn't matter for +MODULE_LICENSE("GPL"); // Actually dual-licensed, but it doesn't matter for // the sake of this tag. It's Free Software. diff --git a/fs/jffs2/symlink.c b/fs/jffs2/symlink.c index 82ef484f5e12..d55754fe8925 100644 --- a/fs/jffs2/symlink.c +++ b/fs/jffs2/symlink.c @@ -7,7 +7,7 @@ * * For licensing information, see the file 'LICENCE' in this directory. * - * $Id: symlink.c,v 1.16 2005/03/01 10:50:48 dedekind Exp $ + * $Id: symlink.c,v 1.19 2005/11/07 11:14:42 gleixner Exp $ * */ @@ -21,7 +21,7 @@ static void *jffs2_follow_link(struct dentry *dentry, struct nameidata *nd); struct inode_operations jffs2_symlink_inode_operations = -{ +{ .readlink = generic_readlink, .follow_link = jffs2_follow_link, .setattr = jffs2_setattr @@ -30,35 +30,33 @@ struct inode_operations jffs2_symlink_inode_operations = static void *jffs2_follow_link(struct dentry *dentry, struct nameidata *nd) { struct jffs2_inode_info *f = JFFS2_INODE_INFO(dentry->d_inode); - char *p = (char *)f->dents; - + char *p = (char *)f->target; + /* * We don't acquire the f->sem mutex here since the only data we - * use is f->dents which in case of the symlink inode points to the - * symlink's target path. + * use is f->target. * - * 1. If we are here the inode has already built and f->dents has + * 1. If we are here the inode has already built and f->target has * to point to the target path. - * 2. Nobody uses f->dents (if the inode is symlink's inode). The - * exception is inode freeing function which frees f->dents. But + * 2. Nobody uses f->target (if the inode is symlink's inode). The + * exception is inode freeing function which frees f->target. But * it can't be called while we are here and before VFS has - * stopped using our f->dents string which we provide by means of + * stopped using our f->target string which we provide by means of * nd_set_link() call. */ - + if (!p) { printk(KERN_ERR "jffs2_follow_link(): can't find symlink taerget\n"); p = ERR_PTR(-EIO); - } else { - D1(printk(KERN_DEBUG "jffs2_follow_link(): target path is '%s'\n", (char *) f->dents)); } + D1(printk(KERN_DEBUG "jffs2_follow_link(): target path is '%s'\n", (char *) f->target)); nd_set_link(nd, p); - + /* - * We unlock the f->sem mutex but VFS will use the f->dents string. This is safe - * since the only way that may cause f->dents to be changed is iput() operation. - * But VFS will not use f->dents after iput() has been called. + * We will unlock the f->sem mutex but VFS will use the f->target string. This is safe + * since the only way that may cause f->target to be changed is iput() operation. + * But VFS will not use f->target after iput() has been called. */ return NULL; } diff --git a/fs/jffs2/wbuf.c b/fs/jffs2/wbuf.c index 996d922e503e..4cebf0e57c46 100644 --- a/fs/jffs2/wbuf.c +++ b/fs/jffs2/wbuf.c @@ -9,7 +9,7 @@ * * For licensing information, see the file 'LICENCE' in this directory. * - * $Id: wbuf.c,v 1.92 2005/04/05 12:51:54 dedekind Exp $ + * $Id: wbuf.c,v 1.100 2005/09/30 13:59:13 dedekind Exp $ * */ @@ -18,6 +18,8 @@ #include <linux/mtd/mtd.h> #include <linux/crc32.h> #include <linux/mtd/nand.h> +#include <linux/jiffies.h> + #include "nodelist.h" /* For testing write failures */ @@ -28,12 +30,12 @@ static unsigned char *brokenbuf; #endif +#define PAGE_DIV(x) ( ((unsigned long)(x) / (unsigned long)(c->wbuf_pagesize)) * (unsigned long)(c->wbuf_pagesize) ) +#define PAGE_MOD(x) ( (unsigned long)(x) % (unsigned long)(c->wbuf_pagesize) ) + /* max. erase failures before we mark a block bad */ #define MAX_ERASE_FAILURES 2 -/* two seconds timeout for timed wbuf-flushing */ -#define WBUF_FLUSH_TIMEOUT 2 * HZ - struct jffs2_inodirty { uint32_t ino; struct jffs2_inodirty *next; @@ -137,7 +139,6 @@ static void jffs2_block_refile(struct jffs2_sb_info *c, struct jffs2_eraseblock { D1(printk("About to refile bad block at %08x\n", jeb->offset)); - D2(jffs2_dump_block_lists(c)); /* File the existing block on the bad_used_list.... */ if (c->nextblock == jeb) c->nextblock = NULL; @@ -154,7 +155,6 @@ static void jffs2_block_refile(struct jffs2_sb_info *c, struct jffs2_eraseblock c->nr_erasing_blocks++; jffs2_erase_pending_trigger(c); } - D2(jffs2_dump_block_lists(c)); /* Adjust its size counts accordingly */ c->wasted_size += jeb->free_size; @@ -162,8 +162,9 @@ static void jffs2_block_refile(struct jffs2_sb_info *c, struct jffs2_eraseblock jeb->wasted_size += jeb->free_size; jeb->free_size = 0; - ACCT_SANITY_CHECK(c,jeb); - D1(ACCT_PARANOIA_CHECK(jeb)); + jffs2_dbg_dump_block_lists_nolock(c); + jffs2_dbg_acct_sanity_check_nolock(c,jeb); + jffs2_dbg_acct_paranoia_check_nolock(c, jeb); } /* Recover from failure to write wbuf. Recover the nodes up to the @@ -187,7 +188,7 @@ static void jffs2_wbuf_recover(struct jffs2_sb_info *c) /* Find the first node to be recovered, by skipping over every node which ends before the wbuf starts, or which is obsolete. */ first_raw = &jeb->first_node; - while (*first_raw && + while (*first_raw && (ref_obsolete(*first_raw) || (ref_offset(*first_raw)+ref_totlen(c, jeb, *first_raw)) < c->wbuf_ofs)) { D1(printk(KERN_DEBUG "Skipping node at 0x%08x(%d)-0x%08x which is either before 0x%08x or obsolete\n", @@ -236,7 +237,7 @@ static void jffs2_wbuf_recover(struct jffs2_sb_info *c) ret = c->mtd->read_ecc(c->mtd, start, c->wbuf_ofs - start, &retlen, buf, NULL, c->oobinfo); else ret = c->mtd->read(c->mtd, start, c->wbuf_ofs - start, &retlen, buf); - + if (ret == -EBADMSG && retlen == c->wbuf_ofs - start) { /* ECC recovered */ ret = 0; @@ -264,7 +265,7 @@ static void jffs2_wbuf_recover(struct jffs2_sb_info *c) /* ... and get an allocation of space from a shiny new block instead */ - ret = jffs2_reserve_space_gc(c, end-start, &ofs, &len); + ret = jffs2_reserve_space_gc(c, end-start, &ofs, &len, JFFS2_SUMMARY_NOSUM_SIZE); if (ret) { printk(KERN_WARNING "Failed to allocate space for wbuf recovery. Data loss ensues.\n"); kfree(buf); @@ -273,15 +274,15 @@ static void jffs2_wbuf_recover(struct jffs2_sb_info *c) if (end-start >= c->wbuf_pagesize) { /* Need to do another write immediately, but it's possible that this is just because the wbuf itself is completely - full, and there's nothing earlier read back from the - flash. Hence 'buf' isn't necessarily what we're writing + full, and there's nothing earlier read back from the + flash. Hence 'buf' isn't necessarily what we're writing from. */ unsigned char *rewrite_buf = buf?:c->wbuf; uint32_t towrite = (end-start) - ((end-start)%c->wbuf_pagesize); D1(printk(KERN_DEBUG "Write 0x%x bytes at 0x%08x in wbuf recover\n", towrite, ofs)); - + #ifdef BREAKMEHEADER static int breakme; if (breakme++ == 20) { @@ -325,8 +326,7 @@ static void jffs2_wbuf_recover(struct jffs2_sb_info *c) c->wbuf_ofs = ofs + towrite; memmove(c->wbuf, rewrite_buf + towrite, c->wbuf_len); /* Don't muck about with c->wbuf_inodes. False positives are harmless. */ - if (buf) - kfree(buf); + kfree(buf); } else { /* OK, now we're left with the dregs in whichever buffer we're using */ if (buf) { @@ -390,11 +390,11 @@ static void jffs2_wbuf_recover(struct jffs2_sb_info *c) else jeb->last_node = container_of(first_raw, struct jffs2_raw_node_ref, next_phys); - ACCT_SANITY_CHECK(c,jeb); - D1(ACCT_PARANOIA_CHECK(jeb)); + jffs2_dbg_acct_sanity_check_nolock(c, jeb); + jffs2_dbg_acct_paranoia_check_nolock(c, jeb); - ACCT_SANITY_CHECK(c,new_jeb); - D1(ACCT_PARANOIA_CHECK(new_jeb)); + jffs2_dbg_acct_sanity_check_nolock(c, new_jeb); + jffs2_dbg_acct_paranoia_check_nolock(c, new_jeb); spin_unlock(&c->erase_completion_lock); @@ -433,15 +433,15 @@ static int __jffs2_flush_wbuf(struct jffs2_sb_info *c, int pad) this happens, if we have a change to a new block, or if fsync forces us to flush the writebuffer. if we have a switch to next page, we will not have - enough remaining space for this. + enough remaining space for this. */ - if (pad && !jffs2_dataflash(c)) { + if (pad ) { c->wbuf_len = PAD(c->wbuf_len); /* Pad with JFFS2_DIRTY_BITMASK initially. this helps out ECC'd NOR with 8 byte page size */ memset(c->wbuf + c->wbuf_len, 0, c->wbuf_pagesize - c->wbuf_len); - + if ( c->wbuf_len + sizeof(struct jffs2_unknown_node) < c->wbuf_pagesize) { struct jffs2_unknown_node *padnode = (void *)(c->wbuf + c->wbuf_len); padnode->magic = cpu_to_je16(JFFS2_MAGIC_BITMASK); @@ -452,7 +452,7 @@ static int __jffs2_flush_wbuf(struct jffs2_sb_info *c, int pad) } /* else jffs2_flash_writev has actually filled in the rest of the buffer for us, and will deal with the node refs etc. later. */ - + #ifdef BREAKME static int breakme; if (breakme++ == 20) { @@ -461,9 +461,9 @@ static int __jffs2_flush_wbuf(struct jffs2_sb_info *c, int pad) c->mtd->write_ecc(c->mtd, c->wbuf_ofs, c->wbuf_pagesize, &retlen, brokenbuf, NULL, c->oobinfo); ret = -EIO; - } else + } else #endif - + if (jffs2_cleanmarker_oob(c)) ret = c->mtd->write_ecc(c->mtd, c->wbuf_ofs, c->wbuf_pagesize, &retlen, c->wbuf, NULL, c->oobinfo); else @@ -486,7 +486,7 @@ static int __jffs2_flush_wbuf(struct jffs2_sb_info *c, int pad) spin_lock(&c->erase_completion_lock); /* Adjust free size of the block if we padded. */ - if (pad && !jffs2_dataflash(c)) { + if (pad) { struct jffs2_eraseblock *jeb; jeb = &c->blocks[c->wbuf_ofs / c->sector_size]; @@ -494,7 +494,7 @@ static int __jffs2_flush_wbuf(struct jffs2_sb_info *c, int pad) D1(printk(KERN_DEBUG "jffs2_flush_wbuf() adjusting free_size of %sblock at %08x\n", (jeb==c->nextblock)?"next":"", jeb->offset)); - /* wbuf_pagesize - wbuf_len is the amount of space that's to be + /* wbuf_pagesize - wbuf_len is the amount of space that's to be padded. If there is less free space in the block than that, something screwed up */ if (jeb->free_size < (c->wbuf_pagesize - c->wbuf_len)) { @@ -522,9 +522,9 @@ static int __jffs2_flush_wbuf(struct jffs2_sb_info *c, int pad) return 0; } -/* Trigger garbage collection to flush the write-buffer. +/* Trigger garbage collection to flush the write-buffer. If ino arg is zero, do it if _any_ real (i.e. not GC) writes are - outstanding. If ino arg non-zero, do it only if a write for the + outstanding. If ino arg non-zero, do it only if a write for the given inode is outstanding. */ int jffs2_flush_wbuf_gc(struct jffs2_sb_info *c, uint32_t ino) { @@ -603,15 +603,6 @@ int jffs2_flush_wbuf_pad(struct jffs2_sb_info *c) return ret; } - -#ifdef CONFIG_JFFS2_FS_WRITEBUFFER -#define PAGE_DIV(x) ( ((unsigned long)(x) / (unsigned long)(c->wbuf_pagesize)) * (unsigned long)(c->wbuf_pagesize) ) -#define PAGE_MOD(x) ( (unsigned long)(x) % (unsigned long)(c->wbuf_pagesize) ) -#else -#define PAGE_DIV(x) ( (x) & (~(c->wbuf_pagesize - 1)) ) -#define PAGE_MOD(x) ( (x) & (c->wbuf_pagesize - 1) ) -#endif - int jffs2_flash_writev(struct jffs2_sb_info *c, const struct kvec *invecs, unsigned long count, loff_t to, size_t *retlen, uint32_t ino) { struct kvec outvecs[3]; @@ -628,13 +619,13 @@ int jffs2_flash_writev(struct jffs2_sb_info *c, const struct kvec *invecs, unsig /* If not NAND flash, don't bother */ if (!jffs2_is_writebuffered(c)) return jffs2_flash_direct_writev(c, invecs, count, to, retlen); - + down_write(&c->wbuf_sem); /* If wbuf_ofs is not initialized, set it to target address */ if (c->wbuf_ofs == 0xFFFFFFFF) { c->wbuf_ofs = PAGE_DIV(to); - c->wbuf_len = PAGE_MOD(to); + c->wbuf_len = PAGE_MOD(to); memset(c->wbuf,0xff,c->wbuf_pagesize); } @@ -648,10 +639,10 @@ int jffs2_flash_writev(struct jffs2_sb_info *c, const struct kvec *invecs, unsig memset(c->wbuf,0xff,c->wbuf_pagesize); } } - - /* Sanity checks on target address. - It's permitted to write at PAD(c->wbuf_len+c->wbuf_ofs), - and it's permitted to write at the beginning of a new + + /* Sanity checks on target address. + It's permitted to write at PAD(c->wbuf_len+c->wbuf_ofs), + and it's permitted to write at the beginning of a new erase block. Anything else, and you die. New block starts at xxx000c (0-b = block header) */ @@ -669,8 +660,8 @@ int jffs2_flash_writev(struct jffs2_sb_info *c, const struct kvec *invecs, unsig } /* set pointer to new block */ c->wbuf_ofs = PAGE_DIV(to); - c->wbuf_len = PAGE_MOD(to); - } + c->wbuf_len = PAGE_MOD(to); + } if (to != PAD(c->wbuf_ofs + c->wbuf_len)) { /* We're not writing immediately after the writebuffer. Bad. */ @@ -690,21 +681,21 @@ int jffs2_flash_writev(struct jffs2_sb_info *c, const struct kvec *invecs, unsig invec = 0; outvec = 0; - /* Fill writebuffer first, if already in use */ + /* Fill writebuffer first, if already in use */ if (c->wbuf_len) { uint32_t invec_ofs = 0; - /* adjust alignment offset */ + /* adjust alignment offset */ if (c->wbuf_len != PAGE_MOD(to)) { c->wbuf_len = PAGE_MOD(to); /* take care of alignment to next page */ if (!c->wbuf_len) c->wbuf_len = c->wbuf_pagesize; } - + while(c->wbuf_len < c->wbuf_pagesize) { uint32_t thislen; - + if (invec == count) goto alldone; @@ -712,17 +703,17 @@ int jffs2_flash_writev(struct jffs2_sb_info *c, const struct kvec *invecs, unsig if (thislen >= invecs[invec].iov_len) thislen = invecs[invec].iov_len; - + invec_ofs = thislen; memcpy(c->wbuf + c->wbuf_len, invecs[invec].iov_base, thislen); c->wbuf_len += thislen; donelen += thislen; /* Get next invec, if actual did not fill the buffer */ - if (c->wbuf_len < c->wbuf_pagesize) + if (c->wbuf_len < c->wbuf_pagesize) invec++; - } - + } + /* write buffer is full, flush buffer */ ret = __jffs2_flush_wbuf(c, NOPAD); if (ret) { @@ -781,10 +772,10 @@ int jffs2_flash_writev(struct jffs2_sb_info *c, const struct kvec *invecs, unsig /* We did cross a page boundary, so we write some now */ if (jffs2_cleanmarker_oob(c)) - ret = c->mtd->writev_ecc(c->mtd, outvecs, splitvec+1, outvec_to, &wbuf_retlen, NULL, c->oobinfo); + ret = c->mtd->writev_ecc(c->mtd, outvecs, splitvec+1, outvec_to, &wbuf_retlen, NULL, c->oobinfo); else ret = jffs2_flash_direct_writev(c, outvecs, splitvec+1, outvec_to, &wbuf_retlen); - + if (ret < 0 || wbuf_retlen != PAGE_DIV(totlen)) { /* At this point we have no problem, c->wbuf is empty. However refile nextblock to avoid @@ -801,7 +792,7 @@ int jffs2_flash_writev(struct jffs2_sb_info *c, const struct kvec *invecs, unsig spin_unlock(&c->erase_completion_lock); goto exit; } - + donelen += wbuf_retlen; c->wbuf_ofs = PAGE_DIV(outvec_to) + PAGE_DIV(totlen); @@ -835,11 +826,17 @@ int jffs2_flash_writev(struct jffs2_sb_info *c, const struct kvec *invecs, unsig alldone: *retlen = donelen; + if (jffs2_sum_active()) { + int res = jffs2_sum_add_kvec(c, invecs, count, (uint32_t) to); + if (res) + return res; + } + if (c->wbuf_len && ino) jffs2_wbuf_dirties_inode(c, ino); ret = 0; - + exit: up_write(&c->wbuf_sem); return ret; @@ -854,7 +851,7 @@ int jffs2_flash_write(struct jffs2_sb_info *c, loff_t ofs, size_t len, size_t *r struct kvec vecs[1]; if (!jffs2_is_writebuffered(c)) - return c->mtd->write(c->mtd, ofs, len, retlen, buf); + return jffs2_flash_direct_write(c, ofs, len, retlen, buf); vecs[0].iov_base = (unsigned char *) buf; vecs[0].iov_len = len; @@ -882,18 +879,18 @@ int jffs2_flash_read(struct jffs2_sb_info *c, loff_t ofs, size_t len, size_t *re if ( (ret == -EBADMSG) && (*retlen == len) ) { printk(KERN_WARNING "mtd->read(0x%zx bytes from 0x%llx) returned ECC error\n", len, ofs); - /* - * We have the raw data without ECC correction in the buffer, maybe + /* + * We have the raw data without ECC correction in the buffer, maybe * we are lucky and all data or parts are correct. We check the node. * If data are corrupted node check will sort it out. * We keep this block, it will fail on write or erase and the we * mark it bad. Or should we do that now? But we should give him a chance. - * Maybe we had a system crash or power loss before the ecc write or + * Maybe we had a system crash or power loss before the ecc write or * a erase was completed. * So we return success. :) */ ret = 0; - } + } /* if no writebuffer available or write buffer empty, return */ if (!c->wbuf_pagesize || !c->wbuf_len) @@ -908,16 +905,16 @@ int jffs2_flash_read(struct jffs2_sb_info *c, loff_t ofs, size_t len, size_t *re if (owbf > c->wbuf_len) /* is read beyond write buffer ? */ goto exit; lwbf = c->wbuf_len - owbf; /* number of bytes to copy */ - if (lwbf > len) + if (lwbf > len) lwbf = len; - } else { + } else { orbf = (c->wbuf_ofs - ofs); /* offset in read buffer */ if (orbf > len) /* is write beyond write buffer ? */ goto exit; lwbf = len - orbf; /* number of bytes to copy */ - if (lwbf > c->wbuf_len) + if (lwbf > c->wbuf_len) lwbf = c->wbuf_len; - } + } if (lwbf > 0) memcpy(buf+orbf,c->wbuf+owbf,lwbf); @@ -945,7 +942,7 @@ int jffs2_check_oob_empty( struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb printk(KERN_NOTICE "jffs2_check_oob_empty(): allocation of temporary data buffer for oob check failed\n"); return -ENOMEM; } - /* + /* * if mode = 0, we scan for a total empty oob area, else we have * to take care of the cleanmarker in the first page of the block */ @@ -954,41 +951,41 @@ int jffs2_check_oob_empty( struct jffs2_sb_info *c, struct jffs2_eraseblock *jeb D1(printk(KERN_WARNING "jffs2_check_oob_empty(): Read OOB failed %d for block at %08x\n", ret, jeb->offset)); goto out; } - + if (retlen < len) { D1(printk(KERN_WARNING "jffs2_check_oob_empty(): Read OOB return short read " "(%zd bytes not %d) for block at %08x\n", retlen, len, jeb->offset)); ret = -EIO; goto out; } - + /* Special check for first page */ for(i = 0; i < oob_size ; i++) { /* Yeah, we know about the cleanmarker. */ - if (mode && i >= c->fsdata_pos && + if (mode && i >= c->fsdata_pos && i < c->fsdata_pos + c->fsdata_len) continue; if (buf[i] != 0xFF) { D2(printk(KERN_DEBUG "Found %02x at %x in OOB for %08x\n", - buf[page+i], page+i, jeb->offset)); - ret = 1; + buf[i], i, jeb->offset)); + ret = 1; goto out; } } - /* we know, we are aligned :) */ + /* we know, we are aligned :) */ for (page = oob_size; page < len; page += sizeof(long)) { unsigned long dat = *(unsigned long *)(&buf[page]); if(dat != -1) { - ret = 1; + ret = 1; goto out; } } out: - kfree(buf); - + kfree(buf); + return ret; } @@ -1070,7 +1067,7 @@ int jffs2_write_nand_cleanmarker(struct jffs2_sb_info *c, struct jffs2_erasebloc n.totlen = cpu_to_je32(8); ret = jffs2_flash_write_oob(c, jeb->offset + c->fsdata_pos, c->fsdata_len, &retlen, (unsigned char *)&n); - + if (ret) { D1(printk(KERN_WARNING "jffs2_write_nand_cleanmarker(): Write failed for block at %08x: error %d\n", jeb->offset, ret)); return ret; @@ -1082,7 +1079,7 @@ int jffs2_write_nand_cleanmarker(struct jffs2_sb_info *c, struct jffs2_erasebloc return 0; } -/* +/* * On NAND we try to mark this block bad. If the block was erased more * than MAX_ERASE_FAILURES we mark it finaly bad. * Don't care about failures. This block remains on the erase-pending @@ -1103,7 +1100,7 @@ int jffs2_write_nand_badblock(struct jffs2_sb_info *c, struct jffs2_eraseblock * D1(printk(KERN_WARNING "jffs2_write_nand_badblock(): Marking bad block at %08x\n", bad_offset)); ret = c->mtd->block_markbad(c->mtd, bad_offset); - + if (ret) { D1(printk(KERN_WARNING "jffs2_write_nand_badblock(): Write failed for block at %08x: error %d\n", jeb->offset, ret)); return ret; @@ -1127,7 +1124,7 @@ static int jffs2_nand_set_oobinfo(struct jffs2_sb_info *c) /* Do this only, if we have an oob buffer */ if (!c->mtd->oobsize) return 0; - + /* Cleanmarker is out-of-band, so inline size zero */ c->cleanmarker_size = 0; @@ -1153,7 +1150,7 @@ static int jffs2_nand_set_oobinfo(struct jffs2_sb_info *c) c->fsdata_len = NAND_JFFS2_OOB16_FSDALEN; c->badblock_pos = 15; break; - + default: D1(printk(KERN_DEBUG "JFFS2 on NAND. No autoplacment info found\n")); return -EINVAL; @@ -1170,7 +1167,7 @@ int jffs2_nand_flash_setup(struct jffs2_sb_info *c) init_rwsem(&c->wbuf_sem); c->wbuf_pagesize = c->mtd->oobblock; c->wbuf_ofs = 0xFFFFFFFF; - + c->wbuf = kmalloc(c->wbuf_pagesize, GFP_KERNEL); if (!c->wbuf) return -ENOMEM; @@ -1196,17 +1193,41 @@ void jffs2_nand_flash_cleanup(struct jffs2_sb_info *c) int jffs2_dataflash_setup(struct jffs2_sb_info *c) { c->cleanmarker_size = 0; /* No cleanmarkers needed */ - + /* Initialize write buffer */ init_rwsem(&c->wbuf_sem); - c->wbuf_pagesize = c->sector_size; - c->wbuf_ofs = 0xFFFFFFFF; + + c->wbuf_pagesize = c->mtd->erasesize; + + /* Find a suitable c->sector_size + * - Not too much sectors + * - Sectors have to be at least 4 K + some bytes + * - All known dataflashes have erase sizes of 528 or 1056 + * - we take at least 8 eraseblocks and want to have at least 8K size + * - The concatenation should be a power of 2 + */ + + c->sector_size = 8 * c->mtd->erasesize; + + while (c->sector_size < 8192) { + c->sector_size *= 2; + } + + /* It may be necessary to adjust the flash size */ + c->flash_size = c->mtd->size; + + if ((c->flash_size % c->sector_size) != 0) { + c->flash_size = (c->flash_size / c->sector_size) * c->sector_size; + printk(KERN_WARNING "JFFS2 flash size adjusted to %dKiB\n", c->flash_size); + }; + + c->wbuf_ofs = 0xFFFFFFFF; c->wbuf = kmalloc(c->wbuf_pagesize, GFP_KERNEL); if (!c->wbuf) return -ENOMEM; - printk(KERN_INFO "JFFS2 write-buffering enabled (%i)\n", c->wbuf_pagesize); + printk(KERN_INFO "JFFS2 write-buffering enabled buffer (%d) erasesize (%d)\n", c->wbuf_pagesize, c->sector_size); return 0; } @@ -1234,3 +1255,23 @@ int jffs2_nor_ecc_flash_setup(struct jffs2_sb_info *c) { void jffs2_nor_ecc_flash_cleanup(struct jffs2_sb_info *c) { kfree(c->wbuf); } + +int jffs2_nor_wbuf_flash_setup(struct jffs2_sb_info *c) { + /* Cleanmarker currently occupies a whole programming region */ + c->cleanmarker_size = MTD_PROGREGION_SIZE(c->mtd); + + /* Initialize write buffer */ + init_rwsem(&c->wbuf_sem); + c->wbuf_pagesize = MTD_PROGREGION_SIZE(c->mtd); + c->wbuf_ofs = 0xFFFFFFFF; + + c->wbuf = kmalloc(c->wbuf_pagesize, GFP_KERNEL); + if (!c->wbuf) + return -ENOMEM; + + return 0; +} + +void jffs2_nor_wbuf_flash_cleanup(struct jffs2_sb_info *c) { + kfree(c->wbuf); +} diff --git a/fs/jffs2/write.c b/fs/jffs2/write.c index 69100615d9ae..1342f0158e9b 100644 --- a/fs/jffs2/write.c +++ b/fs/jffs2/write.c @@ -7,7 +7,7 @@ * * For licensing information, see the file 'LICENCE' in this directory. * - * $Id: write.c,v 1.92 2005/04/13 13:22:35 dwmw2 Exp $ + * $Id: write.c,v 1.97 2005/11/07 11:14:42 gleixner Exp $ * */ @@ -54,35 +54,7 @@ int jffs2_do_new_inode(struct jffs2_sb_info *c, struct jffs2_inode_info *f, uint return 0; } -#if CONFIG_JFFS2_FS_DEBUG > 0 -static void writecheck(struct jffs2_sb_info *c, uint32_t ofs) -{ - unsigned char buf[16]; - size_t retlen; - int ret, i; - - ret = jffs2_flash_read(c, ofs, 16, &retlen, buf); - if (ret || (retlen != 16)) { - D1(printk(KERN_DEBUG "read failed or short in writecheck(). ret %d, retlen %zd\n", ret, retlen)); - return; - } - ret = 0; - for (i=0; i<16; i++) { - if (buf[i] != 0xff) - ret = 1; - } - if (ret) { - printk(KERN_WARNING "ARGH. About to write node to 0x%08x on flash, but there are data already there:\n", ofs); - printk(KERN_WARNING "0x%08x: %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x %02x\n", - ofs, - buf[0], buf[1], buf[2], buf[3], buf[4], buf[5], buf[6], buf[7], - buf[8], buf[9], buf[10], buf[11], buf[12], buf[13], buf[14], buf[15]); - } -} -#endif - - -/* jffs2_write_dnode - given a raw_inode, allocate a full_dnode for it, +/* jffs2_write_dnode - given a raw_inode, allocate a full_dnode for it, write it to the flash, link it into the existing inode/fragment list */ struct jffs2_full_dnode *jffs2_write_dnode(struct jffs2_sb_info *c, struct jffs2_inode_info *f, struct jffs2_raw_inode *ri, const unsigned char *data, uint32_t datalen, uint32_t flash_ofs, int alloc_mode) @@ -106,7 +78,7 @@ struct jffs2_full_dnode *jffs2_write_dnode(struct jffs2_sb_info *c, struct jffs2 vecs[1].iov_base = (unsigned char *)data; vecs[1].iov_len = datalen; - D1(writecheck(c, flash_ofs)); + jffs2_dbg_prewrite_paranoia_check(c, flash_ofs, vecs[0].iov_len + vecs[1].iov_len); if (je32_to_cpu(ri->totlen) != sizeof(*ri) + datalen) { printk(KERN_WARNING "jffs2_write_dnode: ri->totlen (0x%08x) != sizeof(*ri) (0x%08zx) + datalen (0x%08x)\n", je32_to_cpu(ri->totlen), sizeof(*ri), datalen); @@ -114,7 +86,7 @@ struct jffs2_full_dnode *jffs2_write_dnode(struct jffs2_sb_info *c, struct jffs2 raw = jffs2_alloc_raw_node_ref(); if (!raw) return ERR_PTR(-ENOMEM); - + fn = jffs2_alloc_full_dnode(); if (!fn) { jffs2_free_raw_node_ref(raw); @@ -138,7 +110,7 @@ struct jffs2_full_dnode *jffs2_write_dnode(struct jffs2_sb_info *c, struct jffs2 if ((alloc_mode!=ALLOC_GC) && (je32_to_cpu(ri->version) < f->highest_version)) { BUG_ON(!retried); D1(printk(KERN_DEBUG "jffs2_write_dnode : dnode_version %d, " - "highest version %d -> updating dnode\n", + "highest version %d -> updating dnode\n", je32_to_cpu(ri->version), f->highest_version)); ri->version = cpu_to_je32(++f->highest_version); ri->node_crc = cpu_to_je32(crc32(0, ri, sizeof(*ri)-8)); @@ -148,7 +120,7 @@ struct jffs2_full_dnode *jffs2_write_dnode(struct jffs2_sb_info *c, struct jffs2 (alloc_mode==ALLOC_GC)?0:f->inocache->ino); if (ret || (retlen != sizeof(*ri) + datalen)) { - printk(KERN_NOTICE "Write of %zd bytes at 0x%08x failed. returned %d, retlen %zd\n", + printk(KERN_NOTICE "Write of %zd bytes at 0x%08x failed. returned %d, retlen %zd\n", sizeof(*ri)+datalen, flash_ofs, ret, retlen); /* Mark the space as dirtied */ @@ -156,10 +128,10 @@ struct jffs2_full_dnode *jffs2_write_dnode(struct jffs2_sb_info *c, struct jffs2 /* Doesn't belong to any inode */ raw->next_in_ino = NULL; - /* Don't change raw->size to match retlen. We may have + /* Don't change raw->size to match retlen. We may have written the node header already, and only the data will seem corrupted, in which case the scan would skip over - any node we write before the original intended end of + any node we write before the original intended end of this node */ raw->flash_offset |= REF_OBSOLETE; jffs2_add_physical_node_ref(c, raw); @@ -176,26 +148,28 @@ struct jffs2_full_dnode *jffs2_write_dnode(struct jffs2_sb_info *c, struct jffs2 retried = 1; D1(printk(KERN_DEBUG "Retrying failed write.\n")); - - ACCT_SANITY_CHECK(c,jeb); - D1(ACCT_PARANOIA_CHECK(jeb)); + + jffs2_dbg_acct_sanity_check(c,jeb); + jffs2_dbg_acct_paranoia_check(c, jeb); if (alloc_mode == ALLOC_GC) { - ret = jffs2_reserve_space_gc(c, sizeof(*ri) + datalen, &flash_ofs, &dummy); + ret = jffs2_reserve_space_gc(c, sizeof(*ri) + datalen, &flash_ofs, + &dummy, JFFS2_SUMMARY_INODE_SIZE); } else { /* Locking pain */ up(&f->sem); jffs2_complete_reservation(c); - - ret = jffs2_reserve_space(c, sizeof(*ri) + datalen, &flash_ofs, &dummy, alloc_mode); + + ret = jffs2_reserve_space(c, sizeof(*ri) + datalen, &flash_ofs, + &dummy, alloc_mode, JFFS2_SUMMARY_INODE_SIZE); down(&f->sem); } if (!ret) { D1(printk(KERN_DEBUG "Allocated space at 0x%08x to retry failed write.\n", flash_ofs)); - ACCT_SANITY_CHECK(c,jeb); - D1(ACCT_PARANOIA_CHECK(jeb)); + jffs2_dbg_acct_sanity_check(c,jeb); + jffs2_dbg_acct_paranoia_check(c, jeb); goto retry; } @@ -207,9 +181,9 @@ struct jffs2_full_dnode *jffs2_write_dnode(struct jffs2_sb_info *c, struct jffs2 return ERR_PTR(ret?ret:-EIO); } /* Mark the space used */ - /* If node covers at least a whole page, or if it starts at the - beginning of a page and runs to the end of the file, or if - it's a hole node, mark it REF_PRISTINE, else REF_NORMAL. + /* If node covers at least a whole page, or if it starts at the + beginning of a page and runs to the end of the file, or if + it's a hole node, mark it REF_PRISTINE, else REF_NORMAL. */ if ((je32_to_cpu(ri->dsize) >= PAGE_CACHE_SIZE) || ( ((je32_to_cpu(ri->offset)&(PAGE_CACHE_SIZE-1))==0) && @@ -227,12 +201,12 @@ struct jffs2_full_dnode *jffs2_write_dnode(struct jffs2_sb_info *c, struct jffs2 spin_unlock(&c->erase_completion_lock); D1(printk(KERN_DEBUG "jffs2_write_dnode wrote node at 0x%08x(%d) with dsize 0x%x, csize 0x%x, node_crc 0x%08x, data_crc 0x%08x, totlen 0x%08x\n", - flash_ofs, ref_flags(raw), je32_to_cpu(ri->dsize), + flash_ofs, ref_flags(raw), je32_to_cpu(ri->dsize), je32_to_cpu(ri->csize), je32_to_cpu(ri->node_crc), je32_to_cpu(ri->data_crc), je32_to_cpu(ri->totlen))); if (retried) { - ACCT_SANITY_CHECK(c,NULL); + jffs2_dbg_acct_sanity_check(c,NULL); } return fn; @@ -247,10 +221,9 @@ struct jffs2_full_dirent *jffs2_write_dirent(struct jffs2_sb_info *c, struct jff int retried = 0; int ret; - D1(printk(KERN_DEBUG "jffs2_write_dirent(ino #%u, name at *0x%p \"%s\"->ino #%u, name_crc 0x%08x)\n", + D1(printk(KERN_DEBUG "jffs2_write_dirent(ino #%u, name at *0x%p \"%s\"->ino #%u, name_crc 0x%08x)\n", je32_to_cpu(rd->pino), name, name, je32_to_cpu(rd->ino), je32_to_cpu(rd->name_crc))); - D1(writecheck(c, flash_ofs)); D1(if(je32_to_cpu(rd->hdr_crc) != crc32(0, rd, sizeof(struct jffs2_unknown_node)-4)) { printk(KERN_CRIT "Eep. CRC not correct in jffs2_write_dirent()\n"); @@ -262,7 +235,9 @@ struct jffs2_full_dirent *jffs2_write_dirent(struct jffs2_sb_info *c, struct jff vecs[0].iov_len = sizeof(*rd); vecs[1].iov_base = (unsigned char *)name; vecs[1].iov_len = namelen; - + + jffs2_dbg_prewrite_paranoia_check(c, flash_ofs, vecs[0].iov_len + vecs[1].iov_len); + raw = jffs2_alloc_raw_node_ref(); if (!raw) @@ -301,7 +276,7 @@ struct jffs2_full_dirent *jffs2_write_dirent(struct jffs2_sb_info *c, struct jff ret = jffs2_flash_writev(c, vecs, 2, flash_ofs, &retlen, (alloc_mode==ALLOC_GC)?0:je32_to_cpu(rd->pino)); if (ret || (retlen != sizeof(*rd) + namelen)) { - printk(KERN_NOTICE "Write of %zd bytes at 0x%08x failed. returned %d, retlen %zd\n", + printk(KERN_NOTICE "Write of %zd bytes at 0x%08x failed. returned %d, retlen %zd\n", sizeof(*rd)+namelen, flash_ofs, ret, retlen); /* Mark the space as dirtied */ if (retlen) { @@ -322,24 +297,26 @@ struct jffs2_full_dirent *jffs2_write_dirent(struct jffs2_sb_info *c, struct jff D1(printk(KERN_DEBUG "Retrying failed write.\n")); - ACCT_SANITY_CHECK(c,jeb); - D1(ACCT_PARANOIA_CHECK(jeb)); + jffs2_dbg_acct_sanity_check(c,jeb); + jffs2_dbg_acct_paranoia_check(c, jeb); if (alloc_mode == ALLOC_GC) { - ret = jffs2_reserve_space_gc(c, sizeof(*rd) + namelen, &flash_ofs, &dummy); + ret = jffs2_reserve_space_gc(c, sizeof(*rd) + namelen, &flash_ofs, + &dummy, JFFS2_SUMMARY_DIRENT_SIZE(namelen)); } else { /* Locking pain */ up(&f->sem); jffs2_complete_reservation(c); - - ret = jffs2_reserve_space(c, sizeof(*rd) + namelen, &flash_ofs, &dummy, alloc_mode); + + ret = jffs2_reserve_space(c, sizeof(*rd) + namelen, &flash_ofs, + &dummy, alloc_mode, JFFS2_SUMMARY_DIRENT_SIZE(namelen)); down(&f->sem); } if (!ret) { D1(printk(KERN_DEBUG "Allocated space at 0x%08x to retry failed write.\n", flash_ofs)); - ACCT_SANITY_CHECK(c,jeb); - D1(ACCT_PARANOIA_CHECK(jeb)); + jffs2_dbg_acct_sanity_check(c,jeb); + jffs2_dbg_acct_paranoia_check(c, jeb); goto retry; } D1(printk(KERN_DEBUG "Failed to allocate space to retry failed write: %d!\n", ret)); @@ -359,7 +336,7 @@ struct jffs2_full_dirent *jffs2_write_dirent(struct jffs2_sb_info *c, struct jff spin_unlock(&c->erase_completion_lock); if (retried) { - ACCT_SANITY_CHECK(c,NULL); + jffs2_dbg_acct_sanity_check(c,NULL); } return fd; @@ -369,7 +346,7 @@ struct jffs2_full_dirent *jffs2_write_dirent(struct jffs2_sb_info *c, struct jff we don't have to go digging in struct inode or its equivalent. It should set: mode, uid, gid, (starting)isize, atime, ctime, mtime */ int jffs2_write_inode_range(struct jffs2_sb_info *c, struct jffs2_inode_info *f, - struct jffs2_raw_inode *ri, unsigned char *buf, + struct jffs2_raw_inode *ri, unsigned char *buf, uint32_t offset, uint32_t writelen, uint32_t *retlen) { int ret = 0; @@ -377,7 +354,7 @@ int jffs2_write_inode_range(struct jffs2_sb_info *c, struct jffs2_inode_info *f, D1(printk(KERN_DEBUG "jffs2_write_inode_range(): Ino #%u, ofs 0x%x, len 0x%x\n", f->inocache->ino, offset, writelen)); - + while(writelen) { struct jffs2_full_dnode *fn; unsigned char *comprbuf = NULL; @@ -389,7 +366,8 @@ int jffs2_write_inode_range(struct jffs2_sb_info *c, struct jffs2_inode_info *f, retry: D2(printk(KERN_DEBUG "jffs2_commit_write() loop: 0x%x to write to 0x%x\n", writelen, offset)); - ret = jffs2_reserve_space(c, sizeof(*ri) + JFFS2_MIN_DATA_LEN, &phys_ofs, &alloclen, ALLOC_NORMAL); + ret = jffs2_reserve_space(c, sizeof(*ri) + JFFS2_MIN_DATA_LEN, &phys_ofs, + &alloclen, ALLOC_NORMAL, JFFS2_SUMMARY_INODE_SIZE); if (ret) { D1(printk(KERN_DEBUG "jffs2_reserve_space returned %d\n", ret)); break; @@ -473,10 +451,11 @@ int jffs2_do_create(struct jffs2_sb_info *c, struct jffs2_inode_info *dir_f, str uint32_t alloclen, phys_ofs; int ret; - /* Try to reserve enough space for both node and dirent. - * Just the node will do for now, though + /* Try to reserve enough space for both node and dirent. + * Just the node will do for now, though */ - ret = jffs2_reserve_space(c, sizeof(*ri), &phys_ofs, &alloclen, ALLOC_NORMAL); + ret = jffs2_reserve_space(c, sizeof(*ri), &phys_ofs, &alloclen, ALLOC_NORMAL, + JFFS2_SUMMARY_INODE_SIZE); D1(printk(KERN_DEBUG "jffs2_do_create(): reserved 0x%x bytes\n", alloclen)); if (ret) { up(&f->sem); @@ -498,15 +477,16 @@ int jffs2_do_create(struct jffs2_sb_info *c, struct jffs2_inode_info *dir_f, str jffs2_complete_reservation(c); return PTR_ERR(fn); } - /* No data here. Only a metadata node, which will be + /* No data here. Only a metadata node, which will be obsoleted by the first data write */ f->metadata = fn; up(&f->sem); jffs2_complete_reservation(c); - ret = jffs2_reserve_space(c, sizeof(*rd)+namelen, &phys_ofs, &alloclen, ALLOC_NORMAL); - + ret = jffs2_reserve_space(c, sizeof(*rd)+namelen, &phys_ofs, &alloclen, + ALLOC_NORMAL, JFFS2_SUMMARY_DIRENT_SIZE(namelen)); + if (ret) { /* Eep. */ D1(printk(KERN_DEBUG "jffs2_reserve_space() for dirent failed\n")); @@ -539,9 +519,9 @@ int jffs2_do_create(struct jffs2_sb_info *c, struct jffs2_inode_info *dir_f, str fd = jffs2_write_dirent(c, dir_f, rd, name, namelen, phys_ofs, ALLOC_NORMAL); jffs2_free_raw_dirent(rd); - + if (IS_ERR(fd)) { - /* dirent failed to write. Delete the inode normally + /* dirent failed to write. Delete the inode normally as if it were the final unlink() */ jffs2_complete_reservation(c); up(&dir_f->sem); @@ -560,14 +540,15 @@ int jffs2_do_create(struct jffs2_sb_info *c, struct jffs2_inode_info *dir_f, str int jffs2_do_unlink(struct jffs2_sb_info *c, struct jffs2_inode_info *dir_f, - const char *name, int namelen, struct jffs2_inode_info *dead_f) + const char *name, int namelen, struct jffs2_inode_info *dead_f, + uint32_t time) { struct jffs2_raw_dirent *rd; struct jffs2_full_dirent *fd; uint32_t alloclen, phys_ofs; int ret; - if (1 /* alternative branch needs testing */ || + if (1 /* alternative branch needs testing */ || !jffs2_can_mark_obsolete(c)) { /* We can't mark stuff obsolete on the medium. We need to write a deletion dirent */ @@ -575,7 +556,8 @@ int jffs2_do_unlink(struct jffs2_sb_info *c, struct jffs2_inode_info *dir_f, if (!rd) return -ENOMEM; - ret = jffs2_reserve_space(c, sizeof(*rd)+namelen, &phys_ofs, &alloclen, ALLOC_DELETION); + ret = jffs2_reserve_space(c, sizeof(*rd)+namelen, &phys_ofs, &alloclen, + ALLOC_DELETION, JFFS2_SUMMARY_DIRENT_SIZE(namelen)); if (ret) { jffs2_free_raw_dirent(rd); return ret; @@ -588,18 +570,18 @@ int jffs2_do_unlink(struct jffs2_sb_info *c, struct jffs2_inode_info *dir_f, rd->nodetype = cpu_to_je16(JFFS2_NODETYPE_DIRENT); rd->totlen = cpu_to_je32(sizeof(*rd) + namelen); rd->hdr_crc = cpu_to_je32(crc32(0, rd, sizeof(struct jffs2_unknown_node)-4)); - + rd->pino = cpu_to_je32(dir_f->inocache->ino); rd->version = cpu_to_je32(++dir_f->highest_version); rd->ino = cpu_to_je32(0); - rd->mctime = cpu_to_je32(get_seconds()); + rd->mctime = cpu_to_je32(time); rd->nsize = namelen; rd->type = DT_UNKNOWN; rd->node_crc = cpu_to_je32(crc32(0, rd, sizeof(*rd)-8)); rd->name_crc = cpu_to_je32(crc32(0, name, namelen)); fd = jffs2_write_dirent(c, dir_f, rd, name, namelen, phys_ofs, ALLOC_DELETION); - + jffs2_free_raw_dirent(rd); if (IS_ERR(fd)) { @@ -618,7 +600,7 @@ int jffs2_do_unlink(struct jffs2_sb_info *c, struct jffs2_inode_info *dir_f, down(&dir_f->sem); while ((*prev) && (*prev)->nhash <= nhash) { - if ((*prev)->nhash == nhash && + if ((*prev)->nhash == nhash && !memcmp((*prev)->name, name, namelen) && !(*prev)->name[namelen]) { struct jffs2_full_dirent *this = *prev; @@ -639,7 +621,7 @@ int jffs2_do_unlink(struct jffs2_sb_info *c, struct jffs2_inode_info *dir_f, /* dead_f is NULL if this was a rename not a real unlink */ /* Also catch the !f->inocache case, where there was a dirent pointing to an inode which didn't exist. */ - if (dead_f && dead_f->inocache) { + if (dead_f && dead_f->inocache) { down(&dead_f->sem); @@ -647,9 +629,9 @@ int jffs2_do_unlink(struct jffs2_sb_info *c, struct jffs2_inode_info *dir_f, while (dead_f->dents) { /* There can be only deleted ones */ fd = dead_f->dents; - + dead_f->dents = fd->next; - + if (fd->ino) { printk(KERN_WARNING "Deleting inode #%u with active dentry \"%s\"->ino #%u\n", dead_f->inocache->ino, fd->name, fd->ino); @@ -673,7 +655,7 @@ int jffs2_do_unlink(struct jffs2_sb_info *c, struct jffs2_inode_info *dir_f, } -int jffs2_do_link (struct jffs2_sb_info *c, struct jffs2_inode_info *dir_f, uint32_t ino, uint8_t type, const char *name, int namelen) +int jffs2_do_link (struct jffs2_sb_info *c, struct jffs2_inode_info *dir_f, uint32_t ino, uint8_t type, const char *name, int namelen, uint32_t time) { struct jffs2_raw_dirent *rd; struct jffs2_full_dirent *fd; @@ -684,12 +666,13 @@ int jffs2_do_link (struct jffs2_sb_info *c, struct jffs2_inode_info *dir_f, uint if (!rd) return -ENOMEM; - ret = jffs2_reserve_space(c, sizeof(*rd)+namelen, &phys_ofs, &alloclen, ALLOC_NORMAL); + ret = jffs2_reserve_space(c, sizeof(*rd)+namelen, &phys_ofs, &alloclen, + ALLOC_NORMAL, JFFS2_SUMMARY_DIRENT_SIZE(namelen)); if (ret) { jffs2_free_raw_dirent(rd); return ret; } - + down(&dir_f->sem); /* Build a deletion node */ @@ -701,7 +684,7 @@ int jffs2_do_link (struct jffs2_sb_info *c, struct jffs2_inode_info *dir_f, uint rd->pino = cpu_to_je32(dir_f->inocache->ino); rd->version = cpu_to_je32(++dir_f->highest_version); rd->ino = cpu_to_je32(ino); - rd->mctime = cpu_to_je32(get_seconds()); + rd->mctime = cpu_to_je32(time); rd->nsize = namelen; rd->type = type; @@ -710,7 +693,7 @@ int jffs2_do_link (struct jffs2_sb_info *c, struct jffs2_inode_info *dir_f, uint rd->name_crc = cpu_to_je32(crc32(0, name, namelen)); fd = jffs2_write_dirent(c, dir_f, rd, name, namelen, phys_ofs, ALLOC_NORMAL); - + jffs2_free_raw_dirent(rd); if (IS_ERR(fd)) { diff --git a/fs/jffs2/writev.c b/fs/jffs2/writev.c index f079f8388566..c638ae1008de 100644 --- a/fs/jffs2/writev.c +++ b/fs/jffs2/writev.c @@ -7,7 +7,7 @@ * * For licensing information, see the file 'LICENCE' in this directory. * - * $Id: writev.c,v 1.6 2004/11/16 20:36:12 dwmw2 Exp $ + * $Id: writev.c,v 1.8 2005/09/09 15:11:58 havasi Exp $ * */ @@ -42,9 +42,40 @@ static inline int mtd_fake_writev(struct mtd_info *mtd, const struct kvec *vecs, int jffs2_flash_direct_writev(struct jffs2_sb_info *c, const struct kvec *vecs, unsigned long count, loff_t to, size_t *retlen) { + if (!jffs2_is_writebuffered(c)) { + if (jffs2_sum_active()) { + int res; + res = jffs2_sum_add_kvec(c, vecs, count, (uint32_t) to); + if (res) { + return res; + } + } + } + if (c->mtd->writev) return c->mtd->writev(c->mtd, vecs, count, to, retlen); - else + else { return mtd_fake_writev(c->mtd, vecs, count, to, retlen); + } } +int jffs2_flash_direct_write(struct jffs2_sb_info *c, loff_t ofs, size_t len, + size_t *retlen, const u_char *buf) +{ + int ret; + ret = c->mtd->write(c->mtd, ofs, len, retlen, buf); + + if (jffs2_sum_active()) { + struct kvec vecs[1]; + int res; + + vecs[0].iov_base = (unsigned char *) buf; + vecs[0].iov_len = len; + + res = jffs2_sum_add_kvec(c, vecs, 1, (uint32_t) ofs); + if (res) { + return res; + } + } + return ret; +} diff --git a/fs/jfs/inode.c b/fs/jfs/inode.c index 0ec62d5310db..9f942ca8e4e3 100644 --- a/fs/jfs/inode.c +++ b/fs/jfs/inode.c @@ -129,8 +129,7 @@ void jfs_delete_inode(struct inode *inode) jfs_info("In jfs_delete_inode, inode = 0x%p", inode); if (!is_bad_inode(inode) && - (JFS_IP(inode)->fileset == cpu_to_le32(FILESYSTEM_I))) { - + (JFS_IP(inode)->fileset == FILESYSTEM_I)) { truncate_inode_pages(&inode->i_data, 0); if (test_cflag(COMMIT_Freewmap, inode)) diff --git a/fs/jfs/jfs_dmap.c b/fs/jfs/jfs_dmap.c index c739626f5bf1..68000a50ceb6 100644 --- a/fs/jfs/jfs_dmap.c +++ b/fs/jfs/jfs_dmap.c @@ -74,7 +74,7 @@ static void dbAllocBits(struct bmap * bmp, struct dmap * dp, s64 blkno, int nblocks); static void dbSplit(dmtree_t * tp, int leafno, int splitsz, int newval); -static void dbBackSplit(dmtree_t * tp, int leafno); +static int dbBackSplit(dmtree_t * tp, int leafno); static int dbJoin(dmtree_t * tp, int leafno, int newval); static void dbAdjTree(dmtree_t * tp, int leafno, int newval); static int dbAdjCtl(struct bmap * bmp, s64 blkno, int newval, int alloc, @@ -305,7 +305,6 @@ int dbSync(struct inode *ipbmap) filemap_fdatawrite(ipbmap->i_mapping); filemap_fdatawait(ipbmap->i_mapping); - ipbmap->i_state |= I_DIRTY; diWriteSpecial(ipbmap, 0); return (0); @@ -2467,7 +2466,9 @@ dbAdjCtl(struct bmap * bmp, s64 blkno, int newval, int alloc, int level) * that it is at the front of a binary buddy system. */ if (oldval == NOFREE) { - dbBackSplit((dmtree_t *) dcp, leafno); + rc = dbBackSplit((dmtree_t *) dcp, leafno); + if (rc) + return rc; oldval = dcp->stree[ti]; } dbSplit((dmtree_t *) dcp, leafno, dcp->budmin, newval); @@ -2627,7 +2628,7 @@ static void dbSplit(dmtree_t * tp, int leafno, int splitsz, int newval) * * serialization: IREAD_LOCK(ipbmap) or IWRITE_LOCK(ipbmap) held on entry/exit; */ -static void dbBackSplit(dmtree_t * tp, int leafno) +static int dbBackSplit(dmtree_t * tp, int leafno) { int budsz, bud, w, bsz, size; int cursz; @@ -2662,7 +2663,10 @@ static void dbBackSplit(dmtree_t * tp, int leafno) */ for (w = leafno, bsz = budsz;; bsz <<= 1, w = (w < bud) ? w : bud) { - assert(bsz < le32_to_cpu(tp->dmt_nleafs)); + if (bsz >= le32_to_cpu(tp->dmt_nleafs)) { + jfs_err("JFS: block map error in dbBackSplit"); + return -EIO; + } /* determine the buddy. */ @@ -2681,7 +2685,11 @@ static void dbBackSplit(dmtree_t * tp, int leafno) } } - assert(leaf[leafno] == size); + if (leaf[leafno] != size) { + jfs_err("JFS: wrong leaf value in dbBackSplit"); + return -EIO; + } + return 0; } @@ -3055,7 +3063,7 @@ static int cntlz(u32 value) * RETURN VALUES: * log2 number of blocks */ -int blkstol2(s64 nb) +static int blkstol2(s64 nb) { int l2nb; s64 mask; /* meant to be signed */ diff --git a/fs/jfs/jfs_imap.c b/fs/jfs/jfs_imap.c index 4021d46da7e3..28201b194f53 100644 --- a/fs/jfs/jfs_imap.c +++ b/fs/jfs/jfs_imap.c @@ -57,6 +57,12 @@ #include "jfs_debug.h" /* + * __mark_inode_dirty expects inodes to be hashed. Since we don't want + * special inodes in the fileset inode space, we hash them to a dummy head + */ +static HLIST_HEAD(aggregate_hash); + +/* * imap locks */ /* iag free list lock */ @@ -491,6 +497,8 @@ struct inode *diReadSpecial(struct super_block *sb, ino_t inum, int secondary) /* release the page */ release_metapage(mp); + hlist_add_head(&ip->i_hash, &aggregate_hash); + return (ip); } @@ -514,8 +522,6 @@ void diWriteSpecial(struct inode *ip, int secondary) ino_t inum = ip->i_ino; struct metapage *mp; - ip->i_state &= ~I_DIRTY; - if (secondary) address = addressPXD(&sbi->ait2) >> sbi->l2nbperpage; else diff --git a/fs/jfs/jfs_metapage.c b/fs/jfs/jfs_metapage.c index 13d7e3f1feb4..8a53981f9f27 100644 --- a/fs/jfs/jfs_metapage.c +++ b/fs/jfs/jfs_metapage.c @@ -86,7 +86,7 @@ struct meta_anchor { atomic_t io_count; struct metapage *mp[MPS_PER_PAGE]; }; -#define mp_anchor(page) ((struct meta_anchor *)page->private) +#define mp_anchor(page) ((struct meta_anchor *)page_private(page)) static inline struct metapage *page_to_mp(struct page *page, uint offset) { @@ -108,7 +108,7 @@ static inline int insert_metapage(struct page *page, struct metapage *mp) if (!a) return -ENOMEM; memset(a, 0, sizeof(struct meta_anchor)); - page->private = (unsigned long)a; + set_page_private(page, (unsigned long)a); SetPagePrivate(page); kmap(page); } @@ -136,7 +136,7 @@ static inline void remove_metapage(struct page *page, struct metapage *mp) a->mp[index] = NULL; if (--a->mp_count == 0) { kfree(a); - page->private = 0; + set_page_private(page, 0); ClearPagePrivate(page); kunmap(page); } @@ -156,13 +156,13 @@ static inline void dec_io(struct page *page, void (*handler) (struct page *)) #else static inline struct metapage *page_to_mp(struct page *page, uint offset) { - return PagePrivate(page) ? (struct metapage *)page->private : NULL; + return PagePrivate(page) ? (struct metapage *)page_private(page) : NULL; } static inline int insert_metapage(struct page *page, struct metapage *mp) { if (mp) { - page->private = (unsigned long)mp; + set_page_private(page, (unsigned long)mp); SetPagePrivate(page); kmap(page); } @@ -171,7 +171,7 @@ static inline int insert_metapage(struct page *page, struct metapage *mp) static inline void remove_metapage(struct page *page, struct metapage *mp) { - page->private = 0; + set_page_private(page, 0); ClearPagePrivate(page); kunmap(page); } @@ -198,7 +198,7 @@ static void init_once(void *foo, kmem_cache_t *cachep, unsigned long flags) } } -static inline struct metapage *alloc_metapage(unsigned int gfp_mask) +static inline struct metapage *alloc_metapage(gfp_t gfp_mask) { return mempool_alloc(metapage_mempool, gfp_mask); } @@ -395,6 +395,12 @@ static int metapage_writepage(struct page *page, struct writeback_control *wbc) if (mp->nohomeok && !test_bit(META_forcewrite, &mp->flag)) { redirty = 1; + /* + * Make sure this page isn't blocked indefinitely. + * If the journal isn't undergoing I/O, push it + */ + if (mp->log && !(mp->log->cflag & logGC_PAGEOUT)) + jfs_flush_journal(mp->log, 0); continue; } @@ -534,7 +540,7 @@ add_failed: return -EIO; } -static int metapage_releasepage(struct page *page, int gfp_mask) +static int metapage_releasepage(struct page *page, gfp_t gfp_mask) { struct metapage *mp; int busy = 0; diff --git a/fs/jfs/jfs_txnmgr.c b/fs/jfs/jfs_txnmgr.c index c7a92f9deb2b..b660c93c92de 100644 --- a/fs/jfs/jfs_txnmgr.c +++ b/fs/jfs/jfs_txnmgr.c @@ -725,6 +725,9 @@ struct tlock *txLock(tid_t tid, struct inode *ip, struct metapage * mp, else tlck->flag = tlckINODELOCK; + if (S_ISDIR(ip->i_mode)) + tlck->flag |= tlckDIRECTORY; + tlck->type = 0; /* bind the tlock and the page */ @@ -1009,6 +1012,8 @@ struct tlock *txMaplock(tid_t tid, struct inode *ip, int type) /* bind the tlock and the object */ tlck->flag = tlckINODELOCK; + if (S_ISDIR(ip->i_mode)) + tlck->flag |= tlckDIRECTORY; tlck->ip = ip; tlck->mp = NULL; @@ -1077,6 +1082,8 @@ struct linelock *txLinelock(struct linelock * tlock) linelock->flag = tlckLINELOCK; linelock->maxcnt = TLOCKLONG; linelock->index = 0; + if (tlck->flag & tlckDIRECTORY) + linelock->flag |= tlckDIRECTORY; /* append linelock after tlock */ linelock->next = tlock->next; @@ -2070,8 +2077,8 @@ static void xtLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, * * function: log from maplock of freed data extents; */ -void mapLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, - struct tlock * tlck) +static void mapLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, + struct tlock * tlck) { struct pxd_lock *pxdlock; int i, nlock; @@ -2209,7 +2216,7 @@ void txEA(tid_t tid, struct inode *ip, dxd_t * oldea, dxd_t * newea) * function: synchronously write pages locked by transaction * after txLog() but before txUpdateMap(); */ -void txForce(struct tblock * tblk) +static void txForce(struct tblock * tblk) { struct tlock *tlck; lid_t lid, next; @@ -2358,7 +2365,7 @@ static void txUpdateMap(struct tblock * tblk) */ else { /* (maplock->flag & mlckFREE) */ - if (S_ISDIR(tlck->ip->i_mode)) + if (tlck->flag & tlckDIRECTORY) txFreeMap(ipimap, maplock, tblk, COMMIT_PWMAP); else @@ -2389,7 +2396,6 @@ static void txUpdateMap(struct tblock * tblk) */ if (tblk->xflag & COMMIT_CREATE) { diUpdatePMap(ipimap, tblk->ino, FALSE, tblk); - ipimap->i_state |= I_DIRTY; /* update persistent block allocation map * for the allocation of inode extent; */ @@ -2400,7 +2406,6 @@ static void txUpdateMap(struct tblock * tblk) } else if (tblk->xflag & COMMIT_DELETE) { ip = tblk->u.ip; diUpdatePMap(ipimap, ip->i_ino, TRUE, tblk); - ipimap->i_state |= I_DIRTY; iput(ip); } } diff --git a/fs/jfs/jfs_txnmgr.h b/fs/jfs/jfs_txnmgr.h index 59ad0f6b7231..0e4dc4514c47 100644 --- a/fs/jfs/jfs_txnmgr.h +++ b/fs/jfs/jfs_txnmgr.h @@ -122,6 +122,7 @@ extern struct tlock *TxLock; /* transaction lock table */ #define tlckLOG 0x0800 /* updateMap state */ #define tlckUPDATEMAP 0x0080 +#define tlckDIRECTORY 0x0040 /* freeLock state */ #define tlckFREELOCK 0x0008 #define tlckWRITEPAGE 0x0004 diff --git a/fs/jfs/jfs_xtree.c b/fs/jfs/jfs_xtree.c index a7fe2f2b969f..e72f4ebb6e9c 100644 --- a/fs/jfs/jfs_xtree.c +++ b/fs/jfs/jfs_xtree.c @@ -3516,16 +3516,10 @@ s64 xtTruncate(tid_t tid, struct inode *ip, s64 newsize, int flag) /* process entries backward from last index */ index = le16_to_cpu(p->header.nextindex) - 1; - if (p->header.flag & BT_INTERNAL) - goto getChild; - - /* - * leaf page - */ - /* Since this is the rightmost leaf, and we may have already freed - * a page that was formerly to the right, let's make sure that the - * next pointer is zero. + /* Since this is the rightmost page at this level, and we may have + * already freed a page that was formerly to the right, let's make + * sure that the next pointer is zero. */ if (p->header.next) { if (log) @@ -3539,6 +3533,12 @@ s64 xtTruncate(tid_t tid, struct inode *ip, s64 newsize, int flag) p->header.next = 0; } + if (p->header.flag & BT_INTERNAL) + goto getChild; + + /* + * leaf page + */ freed = 0; /* does region covered by leaf page precede Teof ? */ diff --git a/fs/jfs/namei.c b/fs/jfs/namei.c index 1abe7343f920..4abbe8604302 100644 --- a/fs/jfs/namei.c +++ b/fs/jfs/namei.c @@ -827,6 +827,7 @@ static int jfs_link(struct dentry *old_dentry, /* update object inode */ ip->i_nlink++; /* for new link */ ip->i_ctime = CURRENT_TIME; + dir->i_ctime = dir->i_mtime = CURRENT_TIME; mark_inode_dirty(dir); atomic_inc(&ip->i_count); @@ -1024,6 +1025,8 @@ static int jfs_symlink(struct inode *dip, struct dentry *dentry, insert_inode_hash(ip); mark_inode_dirty(ip); + dip->i_ctime = dip->i_mtime = CURRENT_TIME; + mark_inode_dirty(dip); /* * commit update of parent directory and link object */ diff --git a/fs/jfs/super.c b/fs/jfs/super.c index 71bc34b96b2b..4226af3ea91b 100644 --- a/fs/jfs/super.c +++ b/fs/jfs/super.c @@ -442,6 +442,7 @@ static int jfs_fill_super(struct super_block *sb, void *data, int silent) inode->i_nlink = 1; inode->i_size = sb->s_bdev->bd_inode->i_size; inode->i_mapping->a_ops = &jfs_metapage_aops; + insert_inode_hash(inode); mapping_set_gfp_mask(inode->i_mapping, GFP_NOFS); sbi->direct_inode = inode; diff --git a/fs/lockd/clntproc.c b/fs/lockd/clntproc.c index 87332f30141b..c5a33648e9fd 100644 --- a/fs/lockd/clntproc.c +++ b/fs/lockd/clntproc.c @@ -112,8 +112,7 @@ static struct nlm_lockowner *nlm_find_lockowner(struct nlm_host *host, fl_owner_ } } spin_unlock(&host->h_lock); - if (new != NULL) - kfree(new); + kfree(new); return res; } diff --git a/fs/lockd/host.c b/fs/lockd/host.c index 82c77df81c5f..c4c8601096e0 100644 --- a/fs/lockd/host.c +++ b/fs/lockd/host.c @@ -173,11 +173,10 @@ nlm_bind_host(struct nlm_host *host) /* If we've already created an RPC client, check whether * RPC rebind is required - * Note: why keep rebinding if we're on a tcp connection? */ if ((clnt = host->h_rpcclnt) != NULL) { xprt = clnt->cl_xprt; - if (!xprt->stream && time_after_eq(jiffies, host->h_nextrebind)) { + if (time_after_eq(jiffies, host->h_nextrebind)) { clnt->cl_port = 0; host->h_nextrebind = jiffies + NLM_HOST_REBIND; dprintk("lockd: next rebind in %ld jiffies\n", @@ -189,7 +188,6 @@ nlm_bind_host(struct nlm_host *host) goto forgetit; xprt_set_timeout(&xprt->timeout, 5, nlmsvc_timeout); - xprt->nocong = 1; /* No congestion control for NLM */ xprt->resvport = 1; /* NLM requires a reserved port */ /* Existing NLM servers accept AUTH_UNIX only */ diff --git a/fs/lockd/svcsubs.c b/fs/lockd/svcsubs.c index de7536358c7c..62f4a385177f 100644 --- a/fs/lockd/svcsubs.c +++ b/fs/lockd/svcsubs.c @@ -30,6 +30,36 @@ static struct nlm_file * nlm_files[FILE_NRHASH]; static DECLARE_MUTEX(nlm_file_sema); +#ifdef NFSD_DEBUG +static inline void nlm_debug_print_fh(char *msg, struct nfs_fh *f) +{ + u32 *fhp = (u32*)f->data; + + /* print the first 32 bytes of the fh */ + dprintk("lockd: %s (%08x %08x %08x %08x %08x %08x %08x %08x)\n", + msg, fhp[0], fhp[1], fhp[2], fhp[3], + fhp[4], fhp[5], fhp[6], fhp[7]); +} + +static inline void nlm_debug_print_file(char *msg, struct nlm_file *file) +{ + struct inode *inode = file->f_file->f_dentry->d_inode; + + dprintk("lockd: %s %s/%ld\n", + msg, inode->i_sb->s_id, inode->i_ino); +} +#else +static inline void nlm_debug_print_fh(char *msg, struct nfs_fh *f) +{ + return; +} + +static inline void nlm_debug_print_file(char *msg, struct nlm_file *file) +{ + return; +} +#endif + static inline unsigned int file_hash(struct nfs_fh *f) { unsigned int tmp=0; @@ -55,11 +85,8 @@ nlm_lookup_file(struct svc_rqst *rqstp, struct nlm_file **result, struct nlm_file *file; unsigned int hash; u32 nfserr; - u32 *fhp = (u32*)f->data; - - dprintk("lockd: nlm_file_lookup(%08x %08x %08x %08x %08x %08x)\n", - fhp[0], fhp[1], fhp[2], fhp[3], fhp[4], fhp[5]); + nlm_debug_print_fh("nlm_file_lookup", f); hash = file_hash(f); @@ -70,8 +97,7 @@ nlm_lookup_file(struct svc_rqst *rqstp, struct nlm_file **result, if (!nfs_compare_fh(&file->f_handle, f)) goto found; - dprintk("lockd: creating file for (%08x %08x %08x %08x %08x %08x)\n", - fhp[0], fhp[1], fhp[2], fhp[3], fhp[4], fhp[5]); + nlm_debug_print_fh("creating file for", f); nfserr = nlm_lck_denied_nolocks; file = (struct nlm_file *) kmalloc(sizeof(*file), GFP_KERNEL); @@ -124,11 +150,10 @@ out_free: static inline void nlm_delete_file(struct nlm_file *file) { - struct inode *inode = file->f_file->f_dentry->d_inode; struct nlm_file **fp, *f; - dprintk("lockd: closing file %s/%ld\n", - inode->i_sb->s_id, inode->i_ino); + nlm_debug_print_file("closing file", file); + fp = nlm_files + file->f_hash; while ((f = *fp) != NULL) { if (f == file) { diff --git a/fs/locks.c b/fs/locks.c index f7daa5f48949..250ef53d25ef 100644 --- a/fs/locks.c +++ b/fs/locks.c @@ -316,21 +316,22 @@ static int flock_to_posix_lock(struct file *filp, struct file_lock *fl, /* POSIX-1996 leaves the case l->l_len < 0 undefined; POSIX-2001 defines it. */ start += l->l_start; - end = start + l->l_len - 1; - if (l->l_len < 0) { + if (start < 0) + return -EINVAL; + fl->fl_end = OFFSET_MAX; + if (l->l_len > 0) { + end = start + l->l_len - 1; + fl->fl_end = end; + } else if (l->l_len < 0) { end = start - 1; + fl->fl_end = end; start += l->l_len; + if (start < 0) + return -EINVAL; } - - if (start < 0) - return -EINVAL; - if (l->l_len > 0 && end < 0) - return -EOVERFLOW; - fl->fl_start = start; /* we record the absolute position */ - fl->fl_end = end; - if (l->l_len == 0) - fl->fl_end = OFFSET_MAX; + if (fl->fl_end < fl->fl_start) + return -EOVERFLOW; fl->fl_owner = current->files; fl->fl_pid = current->tgid; @@ -362,14 +363,21 @@ static int flock64_to_posix_lock(struct file *filp, struct file_lock *fl, return -EINVAL; } - if (((start += l->l_start) < 0) || (l->l_len < 0)) + start += l->l_start; + if (start < 0) return -EINVAL; - fl->fl_end = start + l->l_len - 1; - if (l->l_len > 0 && fl->fl_end < 0) - return -EOVERFLOW; + fl->fl_end = OFFSET_MAX; + if (l->l_len > 0) { + fl->fl_end = start + l->l_len - 1; + } else if (l->l_len < 0) { + fl->fl_end = start - 1; + start += l->l_len; + if (start < 0) + return -EINVAL; + } fl->fl_start = start; /* we record the absolute position */ - if (l->l_len == 0) - fl->fl_end = OFFSET_MAX; + if (fl->fl_end < fl->fl_start) + return -EOVERFLOW; fl->fl_owner = current->files; fl->fl_pid = current->tgid; @@ -829,12 +837,16 @@ static int __posix_lock_file(struct inode *inode, struct file_lock *request) /* Detect adjacent or overlapping regions (if same lock type) */ if (request->fl_type == fl->fl_type) { + /* In all comparisons of start vs end, use + * "start - 1" rather than "end + 1". If end + * is OFFSET_MAX, end + 1 will become negative. + */ if (fl->fl_end < request->fl_start - 1) goto next_lock; /* If the next lock in the list has entirely bigger * addresses than the new one, insert the lock here. */ - if (fl->fl_start > request->fl_end + 1) + if (fl->fl_start - 1 > request->fl_end) break; /* If we come here, the new and old lock are of the @@ -1093,7 +1105,6 @@ static void time_out_leases(struct inode *inode) before = &fl->fl_next; continue; } - printk(KERN_INFO "lease broken - owner pid = %d\n", fl->fl_pid); lease_modify(before, fl->fl_type & ~F_INPROGRESS); if (fl == *before) /* lease_modify may have freed fl */ before = &fl->fl_next; @@ -1418,7 +1429,7 @@ int fcntl_setlease(unsigned int fd, struct file *filp, long arg) lock_kernel(); error = __setlease(filp, arg, &flp); - if (error) + if (error || arg == F_UNLCK) goto out_unlock; error = fasync_helper(fd, filp, 1, &flp->fl_fasync); diff --git a/fs/mbcache.c b/fs/mbcache.c index b002a088857d..0f1e4530670f 100644 --- a/fs/mbcache.c +++ b/fs/mbcache.c @@ -116,7 +116,7 @@ mb_cache_indexes(struct mb_cache *cache) * What the mbcache registers as to get shrunk dynamically. */ -static int mb_cache_shrink_fn(int nr_to_scan, unsigned int gfp_mask); +static int mb_cache_shrink_fn(int nr_to_scan, gfp_t gfp_mask); static inline int @@ -140,7 +140,7 @@ __mb_cache_entry_unhash(struct mb_cache_entry *ce) static inline void -__mb_cache_entry_forget(struct mb_cache_entry *ce, int gfp_mask) +__mb_cache_entry_forget(struct mb_cache_entry *ce, gfp_t gfp_mask) { struct mb_cache *cache = ce->e_cache; @@ -193,7 +193,7 @@ forget: * Returns the number of objects which are present in the cache. */ static int -mb_cache_shrink_fn(int nr_to_scan, unsigned int gfp_mask) +mb_cache_shrink_fn(int nr_to_scan, gfp_t gfp_mask) { LIST_HEAD(free_list); struct list_head *l, *ltmp; @@ -301,8 +301,7 @@ fail: if (cache) { while (--m >= 0) kfree(cache->c_indexes_hash[m]); - if (cache->c_block_hash) - kfree(cache->c_block_hash); + kfree(cache->c_block_hash); kfree(cache); } return NULL; diff --git a/fs/mpage.c b/fs/mpage.c index bb9aebe93862..c5adcdddf3cc 100644 --- a/fs/mpage.c +++ b/fs/mpage.c @@ -102,7 +102,7 @@ static struct bio *mpage_bio_submit(int rw, struct bio *bio) static struct bio * mpage_alloc(struct block_device *bdev, sector_t first_sector, int nr_vecs, - unsigned int __nocast gfp_flags) + gfp_t gfp_flags) { struct bio *bio; diff --git a/fs/msdos/namei.c b/fs/msdos/namei.c index 154f511c7245..626a367bcd81 100644 --- a/fs/msdos/namei.c +++ b/fs/msdos/namei.c @@ -454,10 +454,10 @@ static int do_msdos_rename(struct inode *old_dir, unsigned char *old_name, { struct buffer_head *dotdot_bh; struct msdos_dir_entry *dotdot_de; - loff_t dotdot_i_pos; struct inode *old_inode, *new_inode; struct fat_slot_info old_sinfo, sinfo; struct timespec ts; + loff_t dotdot_i_pos, new_i_pos; int err, old_attrs, is_dir, update_dotdot, corrupt = 0; old_sinfo.bh = sinfo.bh = dotdot_bh = NULL; @@ -516,28 +516,24 @@ static int do_msdos_rename(struct inode *old_dir, unsigned char *old_name, if (new_inode) { if (err) goto out; - if (MSDOS_I(new_inode)->i_pos != sinfo.i_pos) { - /* WTF??? Cry and fail. */ - printk(KERN_WARNING "msdos_rename: fs corrupted\n"); - goto out; - } - if (is_dir) { err = fat_dir_empty(new_inode); if (err) goto out; } + new_i_pos = MSDOS_I(new_inode)->i_pos; fat_detach(new_inode); } else { err = msdos_add_entry(new_dir, new_name, is_dir, is_hid, 0, &ts, &sinfo); if (err) goto out; + new_i_pos = sinfo.i_pos; } new_dir->i_version++; fat_detach(old_inode); - fat_attach(old_inode, sinfo.i_pos); + fat_attach(old_inode, new_i_pos); if (is_hid) MSDOS_I(old_inode)->i_attrs |= ATTR_HIDDEN; else @@ -604,7 +600,7 @@ error_inode: fat_attach(old_inode, old_sinfo.i_pos); MSDOS_I(old_inode)->i_attrs = old_attrs; if (new_inode) { - fat_attach(new_inode, sinfo.i_pos); + fat_attach(new_inode, new_i_pos); if (corrupt) corrupt |= fat_sync_inode(new_inode); } else { diff --git a/fs/namei.c b/fs/namei.c index 043d587216b5..6dbbd42d8b95 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -28,6 +28,7 @@ #include <linux/syscalls.h> #include <linux/mount.h> #include <linux/audit.h> +#include <linux/file.h> #include <asm/namei.h> #include <asm/uaccess.h> @@ -255,6 +256,38 @@ int permission(struct inode *inode, int mask, struct nameidata *nd) return security_inode_permission(inode, mask, nd); } +/** + * vfs_permission - check for access rights to a given path + * @nd: lookup result that describes the path + * @mask: right to check for (%MAY_READ, %MAY_WRITE, %MAY_EXEC) + * + * Used to check for read/write/execute permissions on a path. + * We use "fsuid" for this, letting us set arbitrary permissions + * for filesystem access without changing the "normal" uids which + * are used for other things. + */ +int vfs_permission(struct nameidata *nd, int mask) +{ + return permission(nd->dentry->d_inode, mask, nd); +} + +/** + * file_permission - check for additional access rights to a given file + * @file: file to check access rights for + * @mask: right to check for (%MAY_READ, %MAY_WRITE, %MAY_EXEC) + * + * Used to check for read/write/execute permissions on an already opened + * file. + * + * Note: + * Do not use this function in new code. All access checks should + * be done using vfs_permission(). + */ +int file_permission(struct file *file, int mask) +{ + return permission(file->f_dentry->d_inode, mask, NULL); +} + /* * get_write_access() gets write permission for a file. * put_write_access() releases this write permission. @@ -317,6 +350,18 @@ void path_release_on_umount(struct nameidata *nd) mntput_no_expire(nd->mnt); } +/** + * release_open_intent - free up open intent resources + * @nd: pointer to nameidata + */ +void release_open_intent(struct nameidata *nd) +{ + if (nd->intent.open.file->f_dentry == NULL) + put_filp(nd->intent.open.file); + else + fput(nd->intent.open.file); +} + /* * Internal lookup() using the new generic dcache. * SMP-safe @@ -750,10 +795,10 @@ static fastcall int __link_path_walk(const char * name, struct nameidata *nd) struct qstr this; unsigned int c; + nd->flags |= LOOKUP_CONTINUE; err = exec_permission_lite(inode, nd); - if (err == -EAGAIN) { - err = permission(inode, MAY_EXEC, nd); - } + if (err == -EAGAIN) + err = vfs_permission(nd, MAY_EXEC); if (err) break; @@ -802,7 +847,6 @@ static fastcall int __link_path_walk(const char * name, struct nameidata *nd) if (err < 0) break; } - nd->flags |= LOOKUP_CONTINUE; /* This does the actual lookups.. */ err = do_lookup(nd, &this, &next); if (err) @@ -1052,6 +1096,71 @@ out: return retval; } +static int __path_lookup_intent_open(const char *name, unsigned int lookup_flags, + struct nameidata *nd, int open_flags, int create_mode) +{ + struct file *filp = get_empty_filp(); + int err; + + if (filp == NULL) + return -ENFILE; + nd->intent.open.file = filp; + nd->intent.open.flags = open_flags; + nd->intent.open.create_mode = create_mode; + err = path_lookup(name, lookup_flags|LOOKUP_OPEN, nd); + if (IS_ERR(nd->intent.open.file)) { + if (err == 0) { + err = PTR_ERR(nd->intent.open.file); + path_release(nd); + } + } else if (err != 0) + release_open_intent(nd); + return err; +} + +/** + * path_lookup_open - lookup a file path with open intent + * @name: pointer to file name + * @lookup_flags: lookup intent flags + * @nd: pointer to nameidata + * @open_flags: open intent flags + */ +int path_lookup_open(const char *name, unsigned int lookup_flags, + struct nameidata *nd, int open_flags) +{ + return __path_lookup_intent_open(name, lookup_flags, nd, + open_flags, 0); +} + +/** + * path_lookup_create - lookup a file path with open + create intent + * @name: pointer to file name + * @lookup_flags: lookup intent flags + * @nd: pointer to nameidata + * @open_flags: open intent flags + * @create_mode: create intent flags + */ +static int path_lookup_create(const char *name, unsigned int lookup_flags, + struct nameidata *nd, int open_flags, + int create_mode) +{ + return __path_lookup_intent_open(name, lookup_flags|LOOKUP_CREATE, nd, + open_flags, create_mode); +} + +int __user_path_lookup_open(const char __user *name, unsigned int lookup_flags, + struct nameidata *nd, int open_flags) +{ + char *tmp = getname(name); + int err = PTR_ERR(tmp); + + if (!IS_ERR(tmp)) { + err = __path_lookup_intent_open(tmp, lookup_flags, nd, open_flags, 0); + putname(tmp); + } + return err; +} + /* * Restricted form of lookup. Doesn't follow links, single-component only, * needs parent already locked. Doesn't follow mounts. @@ -1096,9 +1205,9 @@ out: return dentry; } -struct dentry * lookup_hash(struct qstr *name, struct dentry * base) +struct dentry * lookup_hash(struct nameidata *nd) { - return __lookup_hash(name, base, NULL); + return __lookup_hash(&nd->last, nd->dentry, nd); } /* SMP-safe */ @@ -1122,7 +1231,7 @@ struct dentry * lookup_one_len(const char * name, struct dentry * base, int len) } this.hash = end_name_hash(hash); - return lookup_hash(&this, base); + return __lookup_hash(&this, base, NULL); access: return ERR_PTR(-EACCES); } @@ -1234,9 +1343,6 @@ static inline int may_create(struct inode *dir, struct dentry *child, } /* - * Special case: O_CREAT|O_EXCL implies O_NOFOLLOW for security - * reasons. - * * O_DIRECTORY translates into forcing a directory lookup. */ static inline int lookup_flags(unsigned int f) @@ -1246,9 +1352,6 @@ static inline int lookup_flags(unsigned int f) if (f & O_NOFOLLOW) retval &= ~LOOKUP_FOLLOW; - if ((f & (O_CREAT|O_EXCL)) == (O_CREAT|O_EXCL)) - retval &= ~LOOKUP_FOLLOW; - if (f & O_DIRECTORY) retval |= LOOKUP_DIRECTORY; @@ -1336,7 +1439,7 @@ int may_open(struct nameidata *nd, int acc_mode, int flag) if (S_ISDIR(inode->i_mode) && (flag & FMODE_WRITE)) return -EISDIR; - error = permission(inode, acc_mode, nd); + error = vfs_permission(nd, acc_mode); if (error) return error; @@ -1388,7 +1491,7 @@ int may_open(struct nameidata *nd, int acc_mode, int flag) if (!error) { DQUOT_INIT(inode); - error = do_truncate(dentry, 0); + error = do_truncate(dentry, 0, NULL); } put_write_access(inode); if (error) @@ -1416,27 +1519,27 @@ int may_open(struct nameidata *nd, int acc_mode, int flag) */ int open_namei(const char * pathname, int flag, int mode, struct nameidata *nd) { - int acc_mode, error = 0; + int acc_mode, error; struct path path; struct dentry *dir; int count = 0; acc_mode = ACC_MODE(flag); + /* O_TRUNC implies we need access checks for write permissions */ + if (flag & O_TRUNC) + acc_mode |= MAY_WRITE; + /* Allow the LSM permission hook to distinguish append access from general write access. */ if (flag & O_APPEND) acc_mode |= MAY_APPEND; - /* Fill in the open() intent data */ - nd->intent.open.flags = flag; - nd->intent.open.create_mode = mode; - /* * The simplest case - just a plain lookup. */ if (!(flag & O_CREAT)) { - error = path_lookup(pathname, lookup_flags(flag)|LOOKUP_OPEN, nd); + error = path_lookup_open(pathname, lookup_flags(flag), nd, flag); if (error) return error; goto ok; @@ -1445,7 +1548,7 @@ int open_namei(const char * pathname, int flag, int mode, struct nameidata *nd) /* * Create - we need to know the parent. */ - error = path_lookup(pathname, LOOKUP_PARENT|LOOKUP_OPEN|LOOKUP_CREATE, nd); + error = path_lookup_create(pathname, LOOKUP_PARENT, nd, flag, mode); if (error) return error; @@ -1461,7 +1564,7 @@ int open_namei(const char * pathname, int flag, int mode, struct nameidata *nd) dir = nd->dentry; nd->flags &= ~LOOKUP_PARENT; down(&dir->d_inode->i_sem); - path.dentry = __lookup_hash(&nd->last, nd->dentry, nd); + path.dentry = lookup_hash(nd); path.mnt = nd->mnt; do_last: @@ -1520,6 +1623,8 @@ ok: exit_dput: dput_path(&path, nd); exit: + if (!IS_ERR(nd->intent.open.file)) + release_open_intent(nd); path_release(nd); return error; @@ -1551,19 +1656,19 @@ do_link: if (nd->last_type != LAST_NORM) goto exit; if (nd->last.name[nd->last.len]) { - putname(nd->last.name); + __putname(nd->last.name); goto exit; } error = -ELOOP; if (count++==32) { - putname(nd->last.name); + __putname(nd->last.name); goto exit; } dir = nd->dentry; down(&dir->d_inode->i_sem); - path.dentry = __lookup_hash(&nd->last, nd->dentry, nd); + path.dentry = lookup_hash(nd); path.mnt = nd->mnt; - putname(nd->last.name); + __putname(nd->last.name); goto do_last; } @@ -1593,7 +1698,7 @@ struct dentry *lookup_create(struct nameidata *nd, int is_dir) /* * Do the final lookup. */ - dentry = lookup_hash(&nd->last, nd->dentry); + dentry = lookup_hash(nd); if (IS_ERR(dentry)) goto fail; @@ -1828,7 +1933,7 @@ asmlinkage long sys_rmdir(const char __user * pathname) goto exit1; } down(&nd.dentry->d_inode->i_sem); - dentry = lookup_hash(&nd.last, nd.dentry); + dentry = lookup_hash(&nd); error = PTR_ERR(dentry); if (!IS_ERR(dentry)) { error = vfs_rmdir(nd.dentry->d_inode, dentry); @@ -1897,7 +2002,7 @@ asmlinkage long sys_unlink(const char __user * pathname) if (nd.last_type != LAST_NORM) goto exit1; down(&nd.dentry->d_inode->i_sem); - dentry = lookup_hash(&nd.last, nd.dentry); + dentry = lookup_hash(&nd); error = PTR_ERR(dentry); if (!IS_ERR(dentry)) { /* Why not before? Because we want correct error value */ @@ -2240,7 +2345,7 @@ static inline int do_rename(const char * oldname, const char * newname) trap = lock_rename(new_dir, old_dir); - old_dentry = lookup_hash(&oldnd.last, old_dir); + old_dentry = lookup_hash(&oldnd); error = PTR_ERR(old_dentry); if (IS_ERR(old_dentry)) goto exit3; @@ -2260,7 +2365,7 @@ static inline int do_rename(const char * oldname, const char * newname) error = -EINVAL; if (old_dentry == trap) goto exit4; - new_dentry = lookup_hash(&newnd.last, new_dir); + new_dentry = lookup_hash(&newnd); error = PTR_ERR(new_dentry); if (IS_ERR(new_dentry)) goto exit4; @@ -2463,6 +2568,8 @@ EXPORT_SYMBOL(path_lookup); EXPORT_SYMBOL(path_release); EXPORT_SYMBOL(path_walk); EXPORT_SYMBOL(permission); +EXPORT_SYMBOL(vfs_permission); +EXPORT_SYMBOL(file_permission); EXPORT_SYMBOL(unlock_rename); EXPORT_SYMBOL(vfs_create); EXPORT_SYMBOL(vfs_follow_link); diff --git a/fs/namespace.c b/fs/namespace.c index 2fa9fdf7d6f5..2019899f2ab8 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -24,6 +24,7 @@ #include <linux/mount.h> #include <asm/uaccess.h> #include <asm/unistd.h> +#include "pnode.h" extern int __init init_rootfs(void); @@ -37,33 +38,39 @@ static inline int sysfs_init(void) #endif /* spinlock for vfsmount related operations, inplace of dcache_lock */ - __cacheline_aligned_in_smp DEFINE_SPINLOCK(vfsmount_lock); +__cacheline_aligned_in_smp DEFINE_SPINLOCK(vfsmount_lock); + +static int event; static struct list_head *mount_hashtable; static int hash_mask __read_mostly, hash_bits __read_mostly; -static kmem_cache_t *mnt_cache; +static kmem_cache_t *mnt_cache; +static struct rw_semaphore namespace_sem; static inline unsigned long hash(struct vfsmount *mnt, struct dentry *dentry) { - unsigned long tmp = ((unsigned long) mnt / L1_CACHE_BYTES); - tmp += ((unsigned long) dentry / L1_CACHE_BYTES); + unsigned long tmp = ((unsigned long)mnt / L1_CACHE_BYTES); + tmp += ((unsigned long)dentry / L1_CACHE_BYTES); tmp = tmp + (tmp >> hash_bits); return tmp & hash_mask; } struct vfsmount *alloc_vfsmnt(const char *name) { - struct vfsmount *mnt = kmem_cache_alloc(mnt_cache, GFP_KERNEL); + struct vfsmount *mnt = kmem_cache_alloc(mnt_cache, GFP_KERNEL); if (mnt) { memset(mnt, 0, sizeof(struct vfsmount)); - atomic_set(&mnt->mnt_count,1); + atomic_set(&mnt->mnt_count, 1); INIT_LIST_HEAD(&mnt->mnt_hash); INIT_LIST_HEAD(&mnt->mnt_child); INIT_LIST_HEAD(&mnt->mnt_mounts); INIT_LIST_HEAD(&mnt->mnt_list); INIT_LIST_HEAD(&mnt->mnt_expire); + INIT_LIST_HEAD(&mnt->mnt_share); + INIT_LIST_HEAD(&mnt->mnt_slave_list); + INIT_LIST_HEAD(&mnt->mnt_slave); if (name) { - int size = strlen(name)+1; + int size = strlen(name) + 1; char *newname = kmalloc(size, GFP_KERNEL); if (newname) { memcpy(newname, name, size); @@ -81,36 +88,65 @@ void free_vfsmnt(struct vfsmount *mnt) } /* - * Now, lookup_mnt increments the ref count before returning - * the vfsmount struct. + * find the first or last mount at @dentry on vfsmount @mnt depending on + * @dir. If @dir is set return the first mount else return the last mount. */ -struct vfsmount *lookup_mnt(struct vfsmount *mnt, struct dentry *dentry) +struct vfsmount *__lookup_mnt(struct vfsmount *mnt, struct dentry *dentry, + int dir) { - struct list_head * head = mount_hashtable + hash(mnt, dentry); - struct list_head * tmp = head; + struct list_head *head = mount_hashtable + hash(mnt, dentry); + struct list_head *tmp = head; struct vfsmount *p, *found = NULL; - spin_lock(&vfsmount_lock); for (;;) { - tmp = tmp->next; + tmp = dir ? tmp->next : tmp->prev; p = NULL; if (tmp == head) break; p = list_entry(tmp, struct vfsmount, mnt_hash); if (p->mnt_parent == mnt && p->mnt_mountpoint == dentry) { - found = mntget(p); + found = p; break; } } - spin_unlock(&vfsmount_lock); return found; } +/* + * lookup_mnt increments the ref count before returning + * the vfsmount struct. + */ +struct vfsmount *lookup_mnt(struct vfsmount *mnt, struct dentry *dentry) +{ + struct vfsmount *child_mnt; + spin_lock(&vfsmount_lock); + if ((child_mnt = __lookup_mnt(mnt, dentry, 1))) + mntget(child_mnt); + spin_unlock(&vfsmount_lock); + return child_mnt; +} + static inline int check_mnt(struct vfsmount *mnt) { return mnt->mnt_namespace == current->namespace; } +static void touch_namespace(struct namespace *ns) +{ + if (ns) { + ns->event = ++event; + wake_up_interruptible(&ns->poll); + } +} + +static void __touch_namespace(struct namespace *ns) +{ + if (ns && ns->event != event) { + ns->event = event; + wake_up_interruptible(&ns->poll); + } +} + static void detach_mnt(struct vfsmount *mnt, struct nameidata *old_nd) { old_nd->dentry = mnt->mnt_mountpoint; @@ -122,13 +158,43 @@ static void detach_mnt(struct vfsmount *mnt, struct nameidata *old_nd) old_nd->dentry->d_mounted--; } +void mnt_set_mountpoint(struct vfsmount *mnt, struct dentry *dentry, + struct vfsmount *child_mnt) +{ + child_mnt->mnt_parent = mntget(mnt); + child_mnt->mnt_mountpoint = dget(dentry); + dentry->d_mounted++; +} + static void attach_mnt(struct vfsmount *mnt, struct nameidata *nd) { - mnt->mnt_parent = mntget(nd->mnt); - mnt->mnt_mountpoint = dget(nd->dentry); - list_add(&mnt->mnt_hash, mount_hashtable+hash(nd->mnt, nd->dentry)); + mnt_set_mountpoint(nd->mnt, nd->dentry, mnt); + list_add_tail(&mnt->mnt_hash, mount_hashtable + + hash(nd->mnt, nd->dentry)); list_add_tail(&mnt->mnt_child, &nd->mnt->mnt_mounts); - nd->dentry->d_mounted++; +} + +/* + * the caller must hold vfsmount_lock + */ +static void commit_tree(struct vfsmount *mnt) +{ + struct vfsmount *parent = mnt->mnt_parent; + struct vfsmount *m; + LIST_HEAD(head); + struct namespace *n = parent->mnt_namespace; + + BUG_ON(parent == mnt); + + list_add_tail(&head, &mnt->mnt_list); + list_for_each_entry(m, &head, mnt_list) + m->mnt_namespace = n; + list_splice(&head, n->list.prev); + + list_add_tail(&mnt->mnt_hash, mount_hashtable + + hash(parent, mnt->mnt_mountpoint)); + list_add_tail(&mnt->mnt_child, &parent->mnt_mounts); + touch_namespace(n); } static struct vfsmount *next_mnt(struct vfsmount *p, struct vfsmount *root) @@ -147,8 +213,18 @@ static struct vfsmount *next_mnt(struct vfsmount *p, struct vfsmount *root) return list_entry(next, struct vfsmount, mnt_child); } -static struct vfsmount * -clone_mnt(struct vfsmount *old, struct dentry *root) +static struct vfsmount *skip_mnt_tree(struct vfsmount *p) +{ + struct list_head *prev = p->mnt_mounts.prev; + while (prev != &p->mnt_mounts) { + p = list_entry(prev, struct vfsmount, mnt_child); + prev = p->mnt_mounts.prev; + } + return p; +} + +static struct vfsmount *clone_mnt(struct vfsmount *old, struct dentry *root, + int flag) { struct super_block *sb = old->mnt_sb; struct vfsmount *mnt = alloc_vfsmnt(old->mnt_devname); @@ -160,19 +236,34 @@ clone_mnt(struct vfsmount *old, struct dentry *root) mnt->mnt_root = dget(root); mnt->mnt_mountpoint = mnt->mnt_root; mnt->mnt_parent = mnt; - mnt->mnt_namespace = current->namespace; + + if (flag & CL_SLAVE) { + list_add(&mnt->mnt_slave, &old->mnt_slave_list); + mnt->mnt_master = old; + CLEAR_MNT_SHARED(mnt); + } else { + if ((flag & CL_PROPAGATION) || IS_MNT_SHARED(old)) + list_add(&mnt->mnt_share, &old->mnt_share); + if (IS_MNT_SLAVE(old)) + list_add(&mnt->mnt_slave, &old->mnt_slave); + mnt->mnt_master = old->mnt_master; + } + if (flag & CL_MAKE_SHARED) + set_mnt_shared(mnt); /* stick the duplicate mount on the same expiry list * as the original if that was on one */ - spin_lock(&vfsmount_lock); - if (!list_empty(&old->mnt_expire)) - list_add(&mnt->mnt_expire, &old->mnt_expire); - spin_unlock(&vfsmount_lock); + if (flag & CL_EXPIRE) { + spin_lock(&vfsmount_lock); + if (!list_empty(&old->mnt_expire)) + list_add(&mnt->mnt_expire, &old->mnt_expire); + spin_unlock(&vfsmount_lock); + } } return mnt; } -void __mntput(struct vfsmount *mnt) +static inline void __mntput(struct vfsmount *mnt) { struct super_block *sb = mnt->mnt_sb; dput(mnt->mnt_root); @@ -180,7 +271,46 @@ void __mntput(struct vfsmount *mnt) deactivate_super(sb); } -EXPORT_SYMBOL(__mntput); +void mntput_no_expire(struct vfsmount *mnt) +{ +repeat: + if (atomic_dec_and_lock(&mnt->mnt_count, &vfsmount_lock)) { + if (likely(!mnt->mnt_pinned)) { + spin_unlock(&vfsmount_lock); + __mntput(mnt); + return; + } + atomic_add(mnt->mnt_pinned + 1, &mnt->mnt_count); + mnt->mnt_pinned = 0; + spin_unlock(&vfsmount_lock); + acct_auto_close_mnt(mnt); + security_sb_umount_close(mnt); + goto repeat; + } +} + +EXPORT_SYMBOL(mntput_no_expire); + +void mnt_pin(struct vfsmount *mnt) +{ + spin_lock(&vfsmount_lock); + mnt->mnt_pinned++; + spin_unlock(&vfsmount_lock); +} + +EXPORT_SYMBOL(mnt_pin); + +void mnt_unpin(struct vfsmount *mnt) +{ + spin_lock(&vfsmount_lock); + if (mnt->mnt_pinned) { + atomic_inc(&mnt->mnt_count); + mnt->mnt_pinned--; + } + spin_unlock(&vfsmount_lock); +} + +EXPORT_SYMBOL(mnt_unpin); /* iterator */ static void *m_start(struct seq_file *m, loff_t *pos) @@ -189,7 +319,7 @@ static void *m_start(struct seq_file *m, loff_t *pos) struct list_head *p; loff_t l = *pos; - down_read(&n->sem); + down_read(&namespace_sem); list_for_each(p, &n->list) if (!l--) return list_entry(p, struct vfsmount, mnt_list); @@ -201,13 +331,12 @@ static void *m_next(struct seq_file *m, void *v, loff_t *pos) struct namespace *n = m->private; struct list_head *p = ((struct vfsmount *)v)->mnt_list.next; (*pos)++; - return p==&n->list ? NULL : list_entry(p, struct vfsmount, mnt_list); + return p == &n->list ? NULL : list_entry(p, struct vfsmount, mnt_list); } static void m_stop(struct seq_file *m, void *v) { - struct namespace *n = m->private; - up_read(&n->sem); + up_read(&namespace_sem); } static inline void mangle(struct seq_file *m, const char *s) @@ -275,35 +404,14 @@ struct seq_operations mounts_op = { */ int may_umount_tree(struct vfsmount *mnt) { - struct list_head *next; - struct vfsmount *this_parent = mnt; - int actual_refs; - int minimum_refs; + int actual_refs = 0; + int minimum_refs = 0; + struct vfsmount *p; spin_lock(&vfsmount_lock); - actual_refs = atomic_read(&mnt->mnt_count); - minimum_refs = 2; -repeat: - next = this_parent->mnt_mounts.next; -resume: - while (next != &this_parent->mnt_mounts) { - struct vfsmount *p = list_entry(next, struct vfsmount, mnt_child); - - next = next->next; - + for (p = mnt; p; p = next_mnt(p, mnt)) { actual_refs += atomic_read(&p->mnt_count); minimum_refs += 2; - - if (!list_empty(&p->mnt_mounts)) { - this_parent = p; - goto repeat; - } - } - - if (this_parent != mnt) { - next = this_parent->mnt_child.next; - this_parent = this_parent->mnt_parent; - goto resume; } spin_unlock(&vfsmount_lock); @@ -330,45 +438,67 @@ EXPORT_SYMBOL(may_umount_tree); */ int may_umount(struct vfsmount *mnt) { - if (atomic_read(&mnt->mnt_count) > 2) - return -EBUSY; - return 0; + int ret = 0; + spin_lock(&vfsmount_lock); + if (propagate_mount_busy(mnt, 2)) + ret = -EBUSY; + spin_unlock(&vfsmount_lock); + return ret; } EXPORT_SYMBOL(may_umount); -static void umount_tree(struct vfsmount *mnt) +void release_mounts(struct list_head *head) +{ + struct vfsmount *mnt; + while(!list_empty(head)) { + mnt = list_entry(head->next, struct vfsmount, mnt_hash); + list_del_init(&mnt->mnt_hash); + if (mnt->mnt_parent != mnt) { + struct dentry *dentry; + struct vfsmount *m; + spin_lock(&vfsmount_lock); + dentry = mnt->mnt_mountpoint; + m = mnt->mnt_parent; + mnt->mnt_mountpoint = mnt->mnt_root; + mnt->mnt_parent = mnt; + spin_unlock(&vfsmount_lock); + dput(dentry); + mntput(m); + } + mntput(mnt); + } +} + +void umount_tree(struct vfsmount *mnt, int propagate, struct list_head *kill) { struct vfsmount *p; - LIST_HEAD(kill); for (p = mnt; p; p = next_mnt(p, mnt)) { - list_del(&p->mnt_list); - list_add(&p->mnt_list, &kill); - p->mnt_namespace = NULL; + list_del(&p->mnt_hash); + list_add(&p->mnt_hash, kill); } - while (!list_empty(&kill)) { - mnt = list_entry(kill.next, struct vfsmount, mnt_list); - list_del_init(&mnt->mnt_list); - list_del_init(&mnt->mnt_expire); - if (mnt->mnt_parent == mnt) { - spin_unlock(&vfsmount_lock); - } else { - struct nameidata old_nd; - detach_mnt(mnt, &old_nd); - spin_unlock(&vfsmount_lock); - path_release(&old_nd); - } - mntput(mnt); - spin_lock(&vfsmount_lock); + if (propagate) + propagate_umount(kill); + + list_for_each_entry(p, kill, mnt_hash) { + list_del_init(&p->mnt_expire); + list_del_init(&p->mnt_list); + __touch_namespace(p->mnt_namespace); + p->mnt_namespace = NULL; + list_del_init(&p->mnt_child); + if (p->mnt_parent != p) + mnt->mnt_mountpoint->d_mounted--; + change_mnt_propagation(p, MS_PRIVATE); } } static int do_umount(struct vfsmount *mnt, int flags) { - struct super_block * sb = mnt->mnt_sb; + struct super_block *sb = mnt->mnt_sb; int retval; + LIST_HEAD(umount_list); retval = security_sb_umount(mnt, flags); if (retval) @@ -403,7 +533,7 @@ static int do_umount(struct vfsmount *mnt, int flags) */ lock_kernel(); - if( (flags&MNT_FORCE) && sb->s_op->umount_begin) + if ((flags & MNT_FORCE) && sb->s_op->umount_begin) sb->s_op->umount_begin(sb); unlock_kernel(); @@ -432,29 +562,21 @@ static int do_umount(struct vfsmount *mnt, int flags) return retval; } - down_write(¤t->namespace->sem); + down_write(&namespace_sem); spin_lock(&vfsmount_lock); + event++; - if (atomic_read(&sb->s_active) == 1) { - /* last instance - try to be smart */ - spin_unlock(&vfsmount_lock); - lock_kernel(); - DQUOT_OFF(sb); - acct_auto_close(sb); - unlock_kernel(); - security_sb_umount_close(mnt); - spin_lock(&vfsmount_lock); - } retval = -EBUSY; - if (atomic_read(&mnt->mnt_count) == 2 || flags & MNT_DETACH) { + if (flags & MNT_DETACH || !propagate_mount_busy(mnt, 2)) { if (!list_empty(&mnt->mnt_list)) - umount_tree(mnt); + umount_tree(mnt, 1, &umount_list); retval = 0; } spin_unlock(&vfsmount_lock); if (retval) security_sb_umount_busy(mnt); - up_write(¤t->namespace->sem); + up_write(&namespace_sem); + release_mounts(&umount_list); return retval; } @@ -494,12 +616,11 @@ out: #ifdef __ARCH_WANT_SYS_OLDUMOUNT /* - * The 2.0 compatible umount. No flags. + * The 2.0 compatible umount. No flags. */ - asmlinkage long sys_oldumount(char __user * name) { - return sys_umount(name,0); + return sys_umount(name, 0); } #endif @@ -516,14 +637,13 @@ static int mount_is_safe(struct nameidata *nd) if (current->uid != nd->dentry->d_inode->i_uid) return -EPERM; } - if (permission(nd->dentry->d_inode, MAY_WRITE, nd)) + if (vfs_permission(nd, MAY_WRITE)) return -EPERM; return 0; #endif } -static int -lives_below_in_same_fs(struct dentry *d, struct dentry *dentry) +static int lives_below_in_same_fs(struct dentry *d, struct dentry *dentry) { while (1) { if (d == dentry) @@ -534,12 +654,16 @@ lives_below_in_same_fs(struct dentry *d, struct dentry *dentry) } } -static struct vfsmount *copy_tree(struct vfsmount *mnt, struct dentry *dentry) +struct vfsmount *copy_tree(struct vfsmount *mnt, struct dentry *dentry, + int flag) { struct vfsmount *res, *p, *q, *r, *s; struct nameidata nd; - res = q = clone_mnt(mnt, dentry); + if (!(flag & CL_COPY_ALL) && IS_MNT_UNBINDABLE(mnt)) + return NULL; + + res = q = clone_mnt(mnt, dentry, flag); if (!q) goto Enomem; q->mnt_mountpoint = mnt->mnt_mountpoint; @@ -550,6 +674,10 @@ static struct vfsmount *copy_tree(struct vfsmount *mnt, struct dentry *dentry) continue; for (s = r; s; s = next_mnt(s, r)) { + if (!(flag & CL_COPY_ALL) && IS_MNT_UNBINDABLE(s)) { + s = skip_mnt_tree(s); + continue; + } while (p != s->mnt_parent) { p = p->mnt_parent; q = q->mnt_parent; @@ -557,7 +685,7 @@ static struct vfsmount *copy_tree(struct vfsmount *mnt, struct dentry *dentry) p = s; nd.mnt = q; nd.dentry = p->mnt_mountpoint; - q = clone_mnt(p, p->mnt_root); + q = clone_mnt(p, p->mnt_root, flag); if (!q) goto Enomem; spin_lock(&vfsmount_lock); @@ -567,15 +695,114 @@ static struct vfsmount *copy_tree(struct vfsmount *mnt, struct dentry *dentry) } } return res; - Enomem: +Enomem: if (res) { + LIST_HEAD(umount_list); spin_lock(&vfsmount_lock); - umount_tree(res); + umount_tree(res, 0, &umount_list); spin_unlock(&vfsmount_lock); + release_mounts(&umount_list); } return NULL; } +/* + * @source_mnt : mount tree to be attached + * @nd : place the mount tree @source_mnt is attached + * @parent_nd : if non-null, detach the source_mnt from its parent and + * store the parent mount and mountpoint dentry. + * (done when source_mnt is moved) + * + * NOTE: in the table below explains the semantics when a source mount + * of a given type is attached to a destination mount of a given type. + * --------------------------------------------------------------------------- + * | BIND MOUNT OPERATION | + * |************************************************************************** + * | source-->| shared | private | slave | unbindable | + * | dest | | | | | + * | | | | | | | + * | v | | | | | + * |************************************************************************** + * | shared | shared (++) | shared (+) | shared(+++)| invalid | + * | | | | | | + * |non-shared| shared (+) | private | slave (*) | invalid | + * *************************************************************************** + * A bind operation clones the source mount and mounts the clone on the + * destination mount. + * + * (++) the cloned mount is propagated to all the mounts in the propagation + * tree of the destination mount and the cloned mount is added to + * the peer group of the source mount. + * (+) the cloned mount is created under the destination mount and is marked + * as shared. The cloned mount is added to the peer group of the source + * mount. + * (+++) the mount is propagated to all the mounts in the propagation tree + * of the destination mount and the cloned mount is made slave + * of the same master as that of the source mount. The cloned mount + * is marked as 'shared and slave'. + * (*) the cloned mount is made a slave of the same master as that of the + * source mount. + * + * --------------------------------------------------------------------------- + * | MOVE MOUNT OPERATION | + * |************************************************************************** + * | source-->| shared | private | slave | unbindable | + * | dest | | | | | + * | | | | | | | + * | v | | | | | + * |************************************************************************** + * | shared | shared (+) | shared (+) | shared(+++) | invalid | + * | | | | | | + * |non-shared| shared (+*) | private | slave (*) | unbindable | + * *************************************************************************** + * + * (+) the mount is moved to the destination. And is then propagated to + * all the mounts in the propagation tree of the destination mount. + * (+*) the mount is moved to the destination. + * (+++) the mount is moved to the destination and is then propagated to + * all the mounts belonging to the destination mount's propagation tree. + * the mount is marked as 'shared and slave'. + * (*) the mount continues to be a slave at the new location. + * + * if the source mount is a tree, the operations explained above is + * applied to each mount in the tree. + * Must be called without spinlocks held, since this function can sleep + * in allocations. + */ +static int attach_recursive_mnt(struct vfsmount *source_mnt, + struct nameidata *nd, struct nameidata *parent_nd) +{ + LIST_HEAD(tree_list); + struct vfsmount *dest_mnt = nd->mnt; + struct dentry *dest_dentry = nd->dentry; + struct vfsmount *child, *p; + + if (propagate_mnt(dest_mnt, dest_dentry, source_mnt, &tree_list)) + return -EINVAL; + + if (IS_MNT_SHARED(dest_mnt)) { + for (p = source_mnt; p; p = next_mnt(p, source_mnt)) + set_mnt_shared(p); + } + + spin_lock(&vfsmount_lock); + if (parent_nd) { + detach_mnt(source_mnt, parent_nd); + attach_mnt(source_mnt, nd); + touch_namespace(current->namespace); + } else { + mnt_set_mountpoint(dest_mnt, dest_dentry, source_mnt); + commit_tree(source_mnt); + } + + list_for_each_entry_safe(child, p, &tree_list, mnt_hash) { + list_del_init(&child->mnt_hash); + commit_tree(child); + } + spin_unlock(&vfsmount_lock); + return 0; +} + static int graft_tree(struct vfsmount *mnt, struct nameidata *nd) { int err; @@ -596,17 +823,8 @@ static int graft_tree(struct vfsmount *mnt, struct nameidata *nd) goto out_unlock; err = -ENOENT; - spin_lock(&vfsmount_lock); - if (IS_ROOT(nd->dentry) || !d_unhashed(nd->dentry)) { - struct list_head head; - - attach_mnt(mnt, nd); - list_add_tail(&head, &mnt->mnt_list); - list_splice(&head, current->namespace->list.prev); - mntget(mnt); - err = 0; - } - spin_unlock(&vfsmount_lock); + if (IS_ROOT(nd->dentry) || !d_unhashed(nd->dentry)) + err = attach_recursive_mnt(mnt, nd, NULL); out_unlock: up(&nd->dentry->d_inode->i_sem); if (!err) @@ -615,6 +833,27 @@ out_unlock: } /* + * recursively change the type of the mountpoint. + */ +static int do_change_type(struct nameidata *nd, int flag) +{ + struct vfsmount *m, *mnt = nd->mnt; + int recurse = flag & MS_REC; + int type = flag & ~MS_REC; + + if (nd->dentry != nd->mnt->mnt_root) + return -EINVAL; + + down_write(&namespace_sem); + spin_lock(&vfsmount_lock); + for (m = mnt; m; m = (recurse ? next_mnt(m, mnt) : NULL)) + change_mnt_propagation(m, type); + spin_unlock(&vfsmount_lock); + up_write(&namespace_sem); + return 0; +} + +/* * do loopback mount. */ static int do_loopback(struct nameidata *nd, char *old_name, int recurse) @@ -630,32 +869,34 @@ static int do_loopback(struct nameidata *nd, char *old_name, int recurse) if (err) return err; - down_write(¤t->namespace->sem); + down_write(&namespace_sem); err = -EINVAL; - if (check_mnt(nd->mnt) && (!recurse || check_mnt(old_nd.mnt))) { - err = -ENOMEM; - if (recurse) - mnt = copy_tree(old_nd.mnt, old_nd.dentry); - else - mnt = clone_mnt(old_nd.mnt, old_nd.dentry); - } + if (IS_MNT_UNBINDABLE(old_nd.mnt)) + goto out; - if (mnt) { - /* stop bind mounts from expiring */ + if (!check_mnt(nd->mnt) || !check_mnt(old_nd.mnt)) + goto out; + + err = -ENOMEM; + if (recurse) + mnt = copy_tree(old_nd.mnt, old_nd.dentry, 0); + else + mnt = clone_mnt(old_nd.mnt, old_nd.dentry, 0); + + if (!mnt) + goto out; + + err = graft_tree(mnt, nd); + if (err) { + LIST_HEAD(umount_list); spin_lock(&vfsmount_lock); - list_del_init(&mnt->mnt_expire); + umount_tree(mnt, 0, &umount_list); spin_unlock(&vfsmount_lock); - - err = graft_tree(mnt, nd); - if (err) { - spin_lock(&vfsmount_lock); - umount_tree(mnt); - spin_unlock(&vfsmount_lock); - } else - mntput(mnt); + release_mounts(&umount_list); } - up_write(¤t->namespace->sem); +out: + up_write(&namespace_sem); path_release(&old_nd); return err; } @@ -665,12 +906,11 @@ static int do_loopback(struct nameidata *nd, char *old_name, int recurse) * If you've mounted a non-root directory somewhere and want to do remount * on it - tough luck. */ - static int do_remount(struct nameidata *nd, int flags, int mnt_flags, void *data) { int err; - struct super_block * sb = nd->mnt->mnt_sb; + struct super_block *sb = nd->mnt->mnt_sb; if (!capable(CAP_SYS_ADMIN)) return -EPERM; @@ -684,13 +924,23 @@ static int do_remount(struct nameidata *nd, int flags, int mnt_flags, down_write(&sb->s_umount); err = do_remount_sb(sb, flags, data, 0); if (!err) - nd->mnt->mnt_flags=mnt_flags; + nd->mnt->mnt_flags = mnt_flags; up_write(&sb->s_umount); if (!err) security_sb_post_remount(nd->mnt, flags, data); return err; } +static inline int tree_contains_unbindable(struct vfsmount *mnt) +{ + struct vfsmount *p; + for (p = mnt; p; p = next_mnt(p, mnt)) { + if (IS_MNT_UNBINDABLE(p)) + return 1; + } + return 0; +} + static int do_move_mount(struct nameidata *nd, char *old_name) { struct nameidata old_nd, parent_nd; @@ -704,8 +954,8 @@ static int do_move_mount(struct nameidata *nd, char *old_name) if (err) return err; - down_write(¤t->namespace->sem); - while(d_mountpoint(nd->dentry) && follow_down(&nd->mnt, &nd->dentry)) + down_write(&namespace_sem); + while (d_mountpoint(nd->dentry) && follow_down(&nd->mnt, &nd->dentry)) ; err = -EINVAL; if (!check_mnt(nd->mnt) || !check_mnt(old_nd.mnt)) @@ -716,39 +966,47 @@ static int do_move_mount(struct nameidata *nd, char *old_name) if (IS_DEADDIR(nd->dentry->d_inode)) goto out1; - spin_lock(&vfsmount_lock); if (!IS_ROOT(nd->dentry) && d_unhashed(nd->dentry)) - goto out2; + goto out1; err = -EINVAL; if (old_nd.dentry != old_nd.mnt->mnt_root) - goto out2; + goto out1; if (old_nd.mnt == old_nd.mnt->mnt_parent) - goto out2; + goto out1; if (S_ISDIR(nd->dentry->d_inode->i_mode) != S_ISDIR(old_nd.dentry->d_inode->i_mode)) - goto out2; - + goto out1; + /* + * Don't move a mount residing in a shared parent. + */ + if (old_nd.mnt->mnt_parent && IS_MNT_SHARED(old_nd.mnt->mnt_parent)) + goto out1; + /* + * Don't move a mount tree containing unbindable mounts to a destination + * mount which is shared. + */ + if (IS_MNT_SHARED(nd->mnt) && tree_contains_unbindable(old_nd.mnt)) + goto out1; err = -ELOOP; - for (p = nd->mnt; p->mnt_parent!=p; p = p->mnt_parent) + for (p = nd->mnt; p->mnt_parent != p; p = p->mnt_parent) if (p == old_nd.mnt) - goto out2; - err = 0; + goto out1; - detach_mnt(old_nd.mnt, &parent_nd); - attach_mnt(old_nd.mnt, nd); + if ((err = attach_recursive_mnt(old_nd.mnt, nd, &parent_nd))) + goto out1; + spin_lock(&vfsmount_lock); /* if the mount is moved, it should no longer be expire * automatically */ list_del_init(&old_nd.mnt->mnt_expire); -out2: spin_unlock(&vfsmount_lock); out1: up(&nd->dentry->d_inode->i_sem); out: - up_write(¤t->namespace->sem); + up_write(&namespace_sem); if (!err) path_release(&parent_nd); path_release(&old_nd); @@ -787,9 +1045,9 @@ int do_add_mount(struct vfsmount *newmnt, struct nameidata *nd, { int err; - down_write(¤t->namespace->sem); + down_write(&namespace_sem); /* Something was mounted here while we slept */ - while(d_mountpoint(nd->dentry) && follow_down(&nd->mnt, &nd->dentry)) + while (d_mountpoint(nd->dentry) && follow_down(&nd->mnt, &nd->dentry)) ; err = -EINVAL; if (!check_mnt(nd->mnt)) @@ -806,25 +1064,28 @@ int do_add_mount(struct vfsmount *newmnt, struct nameidata *nd, goto unlock; newmnt->mnt_flags = mnt_flags; - newmnt->mnt_namespace = current->namespace; - err = graft_tree(newmnt, nd); + if ((err = graft_tree(newmnt, nd))) + goto unlock; - if (err == 0 && fslist) { + if (fslist) { /* add to the specified expiration list */ spin_lock(&vfsmount_lock); list_add_tail(&newmnt->mnt_expire, fslist); spin_unlock(&vfsmount_lock); } + up_write(&namespace_sem); + return 0; unlock: - up_write(¤t->namespace->sem); + up_write(&namespace_sem); mntput(newmnt); return err; } EXPORT_SYMBOL_GPL(do_add_mount); -static void expire_mount(struct vfsmount *mnt, struct list_head *mounts) +static void expire_mount(struct vfsmount *mnt, struct list_head *mounts, + struct list_head *umounts) { spin_lock(&vfsmount_lock); @@ -841,27 +1102,13 @@ static void expire_mount(struct vfsmount *mnt, struct list_head *mounts) * Check that it is still dead: the count should now be 2 - as * contributed by the vfsmount parent and the mntget above */ - if (atomic_read(&mnt->mnt_count) == 2) { - struct nameidata old_nd; - + if (!propagate_mount_busy(mnt, 2)) { /* delete from the namespace */ + touch_namespace(mnt->mnt_namespace); list_del_init(&mnt->mnt_list); mnt->mnt_namespace = NULL; - detach_mnt(mnt, &old_nd); + umount_tree(mnt, 1, umounts); spin_unlock(&vfsmount_lock); - path_release(&old_nd); - - /* - * Now lay it to rest if this was the last ref on the superblock - */ - if (atomic_read(&mnt->mnt_sb->s_active) == 1) { - /* last instance - try to be smart */ - lock_kernel(); - DQUOT_OFF(mnt->mnt_sb); - acct_auto_close(mnt->mnt_sb); - unlock_kernel(); - } - mntput(mnt); } else { /* * Someone brought it back to life whilst we didn't have any @@ -910,6 +1157,7 @@ void mark_mounts_for_expiry(struct list_head *mounts) * - dispose of the corpse */ while (!list_empty(&graveyard)) { + LIST_HEAD(umounts); mnt = list_entry(graveyard.next, struct vfsmount, mnt_expire); list_del_init(&mnt->mnt_expire); @@ -921,13 +1169,12 @@ void mark_mounts_for_expiry(struct list_head *mounts) get_namespace(namespace); spin_unlock(&vfsmount_lock); - down_write(&namespace->sem); - expire_mount(mnt, mounts); - up_write(&namespace->sem); - + down_write(&namespace_sem); + expire_mount(mnt, mounts, &umounts); + up_write(&namespace_sem); + release_mounts(&umounts); mntput(mnt); put_namespace(namespace); - spin_lock(&vfsmount_lock); } @@ -942,8 +1189,8 @@ EXPORT_SYMBOL_GPL(mark_mounts_for_expiry); * Note that this function differs from copy_from_user() in that it will oops * on bad values of `to', rather than returning a short copy. */ -static long -exact_copy_from_user(void *to, const void __user *from, unsigned long n) +static long exact_copy_from_user(void *to, const void __user * from, + unsigned long n) { char *t = to; const char __user *f = from; @@ -964,12 +1211,12 @@ exact_copy_from_user(void *to, const void __user *from, unsigned long n) return n; } -int copy_mount_options(const void __user *data, unsigned long *where) +int copy_mount_options(const void __user * data, unsigned long *where) { int i; unsigned long page; unsigned long size; - + *where = 0; if (!data) return 0; @@ -988,7 +1235,7 @@ int copy_mount_options(const void __user *data, unsigned long *where) i = size - exact_copy_from_user((void *)page, data, size); if (!i) { - free_page(page); + free_page(page); return -EFAULT; } if (i != PAGE_SIZE) @@ -1011,7 +1258,7 @@ int copy_mount_options(const void __user *data, unsigned long *where) * Therefore, if this magic number is present, it carries no information * and must be discarded. */ -long do_mount(char * dev_name, char * dir_name, char *type_page, +long do_mount(char *dev_name, char *dir_name, char *type_page, unsigned long flags, void *data_page) { struct nameidata nd; @@ -1039,7 +1286,7 @@ long do_mount(char * dev_name, char * dir_name, char *type_page, mnt_flags |= MNT_NODEV; if (flags & MS_NOEXEC) mnt_flags |= MNT_NOEXEC; - flags &= ~(MS_NOSUID|MS_NOEXEC|MS_NODEV|MS_ACTIVE); + flags &= ~(MS_NOSUID | MS_NOEXEC | MS_NODEV | MS_ACTIVE); /* ... and get the mountpoint */ retval = path_lookup(dir_name, LOOKUP_FOLLOW, &nd); @@ -1055,6 +1302,8 @@ long do_mount(char * dev_name, char * dir_name, char *type_page, data_page); else if (flags & MS_BIND) retval = do_loopback(&nd, dev_name, flags & MS_REC); + else if (flags & (MS_SHARED | MS_PRIVATE | MS_SLAVE | MS_UNBINDABLE)) + retval = do_change_type(&nd, flags); else if (flags & MS_MOVE) retval = do_move_mount(&nd, dev_name); else @@ -1091,14 +1340,16 @@ int copy_namespace(int flags, struct task_struct *tsk) goto out; atomic_set(&new_ns->count, 1); - init_rwsem(&new_ns->sem); INIT_LIST_HEAD(&new_ns->list); + init_waitqueue_head(&new_ns->poll); + new_ns->event = 0; - down_write(&tsk->namespace->sem); + down_write(&namespace_sem); /* First pass: copy the tree topology */ - new_ns->root = copy_tree(namespace->root, namespace->root->mnt_root); + new_ns->root = copy_tree(namespace->root, namespace->root->mnt_root, + CL_COPY_ALL | CL_EXPIRE); if (!new_ns->root) { - up_write(&tsk->namespace->sem); + up_write(&namespace_sem); kfree(new_ns); goto out; } @@ -1132,7 +1383,7 @@ int copy_namespace(int flags, struct task_struct *tsk) p = next_mnt(p, namespace->root); q = next_mnt(q, new_ns->root); } - up_write(&tsk->namespace->sem); + up_write(&namespace_sem); tsk->namespace = new_ns; @@ -1161,7 +1412,7 @@ asmlinkage long sys_mount(char __user * dev_name, char __user * dir_name, unsigned long dev_page; char *dir_page; - retval = copy_mount_options (type, &type_page); + retval = copy_mount_options(type, &type_page); if (retval < 0) return retval; @@ -1170,17 +1421,17 @@ asmlinkage long sys_mount(char __user * dev_name, char __user * dir_name, if (IS_ERR(dir_page)) goto out1; - retval = copy_mount_options (dev_name, &dev_page); + retval = copy_mount_options(dev_name, &dev_page); if (retval < 0) goto out2; - retval = copy_mount_options (data, &data_page); + retval = copy_mount_options(data, &data_page); if (retval < 0) goto out3; lock_kernel(); - retval = do_mount((char*)dev_page, dir_page, (char*)type_page, - flags, (void*)data_page); + retval = do_mount((char *)dev_page, dir_page, (char *)type_page, + flags, (void *)data_page); unlock_kernel(); free_page(data_page); @@ -1249,9 +1500,11 @@ static void chroot_fs_refs(struct nameidata *old_nd, struct nameidata *new_nd) if (fs) { atomic_inc(&fs->count); task_unlock(p); - if (fs->root==old_nd->dentry&&fs->rootmnt==old_nd->mnt) + if (fs->root == old_nd->dentry + && fs->rootmnt == old_nd->mnt) set_fs_root(fs, new_nd->mnt, new_nd->dentry); - if (fs->pwd==old_nd->dentry&&fs->pwdmnt==old_nd->mnt) + if (fs->pwd == old_nd->dentry + && fs->pwdmnt == old_nd->mnt) set_fs_pwd(fs, new_nd->mnt, new_nd->dentry); put_fs_struct(fs); } else @@ -1281,8 +1534,8 @@ static void chroot_fs_refs(struct nameidata *old_nd, struct nameidata *new_nd) * though, so you may need to say mount --bind /nfs/my_root /nfs/my_root * first. */ - -asmlinkage long sys_pivot_root(const char __user *new_root, const char __user *put_old) +asmlinkage long sys_pivot_root(const char __user * new_root, + const char __user * put_old) { struct vfsmount *tmp; struct nameidata new_nd, old_nd, parent_nd, root_parent, user_nd; @@ -1293,14 +1546,15 @@ asmlinkage long sys_pivot_root(const char __user *new_root, const char __user *p lock_kernel(); - error = __user_walk(new_root, LOOKUP_FOLLOW|LOOKUP_DIRECTORY, &new_nd); + error = __user_walk(new_root, LOOKUP_FOLLOW | LOOKUP_DIRECTORY, + &new_nd); if (error) goto out0; error = -EINVAL; if (!check_mnt(new_nd.mnt)) goto out1; - error = __user_walk(put_old, LOOKUP_FOLLOW|LOOKUP_DIRECTORY, &old_nd); + error = __user_walk(put_old, LOOKUP_FOLLOW | LOOKUP_DIRECTORY, &old_nd); if (error) goto out1; @@ -1314,9 +1568,13 @@ asmlinkage long sys_pivot_root(const char __user *new_root, const char __user *p user_nd.mnt = mntget(current->fs->rootmnt); user_nd.dentry = dget(current->fs->root); read_unlock(¤t->fs->lock); - down_write(¤t->namespace->sem); + down_write(&namespace_sem); down(&old_nd.dentry->d_inode->i_sem); error = -EINVAL; + if (IS_MNT_SHARED(old_nd.mnt) || + IS_MNT_SHARED(new_nd.mnt->mnt_parent) || + IS_MNT_SHARED(user_nd.mnt->mnt_parent)) + goto out2; if (!check_mnt(user_nd.mnt)) goto out2; error = -ENOENT; @@ -1356,6 +1614,7 @@ asmlinkage long sys_pivot_root(const char __user *new_root, const char __user *p detach_mnt(user_nd.mnt, &root_parent); attach_mnt(user_nd.mnt, &old_nd); /* mount old root on put_old */ attach_mnt(new_nd.mnt, &root_parent); /* mount new_root on / */ + touch_namespace(current->namespace); spin_unlock(&vfsmount_lock); chroot_fs_refs(&user_nd, &new_nd); security_sb_post_pivotroot(&user_nd, &new_nd); @@ -1364,7 +1623,7 @@ asmlinkage long sys_pivot_root(const char __user *new_root, const char __user *p path_release(&parent_nd); out2: up(&old_nd.dentry->d_inode->i_sem); - up_write(¤t->namespace->sem); + up_write(&namespace_sem); path_release(&user_nd); path_release(&old_nd); out1: @@ -1391,7 +1650,8 @@ static void __init init_mount_tree(void) panic("Can't allocate initial namespace"); atomic_set(&namespace->count, 1); INIT_LIST_HEAD(&namespace->list); - init_rwsem(&namespace->sem); + init_waitqueue_head(&namespace->poll); + namespace->event = 0; list_add(&mnt->mnt_list, &namespace->list); namespace->root = mnt; mnt->mnt_namespace = namespace; @@ -1414,11 +1674,12 @@ void __init mnt_init(unsigned long mempages) unsigned int nr_hash; int i; + init_rwsem(&namespace_sem); + mnt_cache = kmem_cache_create("mnt_cache", sizeof(struct vfsmount), - 0, SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL, NULL); + 0, SLAB_HWCACHE_ALIGN | SLAB_PANIC, NULL, NULL); - mount_hashtable = (struct list_head *) - __get_free_page(GFP_ATOMIC); + mount_hashtable = (struct list_head *)__get_free_page(GFP_ATOMIC); if (!mount_hashtable) panic("Failed to allocate mount hash table\n"); @@ -1440,7 +1701,7 @@ void __init mnt_init(unsigned long mempages) * from the number of bits we can fit. */ nr_hash = 1UL << hash_bits; - hash_mask = nr_hash-1; + hash_mask = nr_hash - 1; printk("Mount-cache hash table entries: %d\n", nr_hash); @@ -1460,12 +1721,14 @@ void __init mnt_init(unsigned long mempages) void __put_namespace(struct namespace *namespace) { struct vfsmount *root = namespace->root; + LIST_HEAD(umount_list); namespace->root = NULL; spin_unlock(&vfsmount_lock); - down_write(&namespace->sem); + down_write(&namespace_sem); spin_lock(&vfsmount_lock); - umount_tree(root); + umount_tree(root, 0, &umount_list); spin_unlock(&vfsmount_lock); - up_write(&namespace->sem); + up_write(&namespace_sem); + release_mounts(&umount_list); kfree(namespace); } diff --git a/fs/ncpfs/ioctl.c b/fs/ncpfs/ioctl.c index 88df79356a1f..fd3efdca5ae3 100644 --- a/fs/ncpfs/ioctl.c +++ b/fs/ncpfs/ioctl.c @@ -30,11 +30,13 @@ #define NCP_PACKET_SIZE_INTERNAL 65536 static int -ncp_get_fs_info(struct ncp_server* server, struct inode* inode, struct ncp_fs_info __user *arg) +ncp_get_fs_info(struct ncp_server * server, struct file *file, + struct ncp_fs_info __user *arg) { + struct inode *inode = file->f_dentry->d_inode; struct ncp_fs_info info; - if ((permission(inode, MAY_WRITE, NULL) != 0) + if ((file_permission(file, MAY_WRITE) != 0) && (current->uid != server->m.mounted_uid)) { return -EACCES; } @@ -58,11 +60,13 @@ ncp_get_fs_info(struct ncp_server* server, struct inode* inode, struct ncp_fs_in } static int -ncp_get_fs_info_v2(struct ncp_server* server, struct inode* inode, struct ncp_fs_info_v2 __user * arg) +ncp_get_fs_info_v2(struct ncp_server * server, struct file *file, + struct ncp_fs_info_v2 __user * arg) { + struct inode *inode = file->f_dentry->d_inode; struct ncp_fs_info_v2 info2; - if ((permission(inode, MAY_WRITE, NULL) != 0) + if ((file_permission(file, MAY_WRITE) != 0) && (current->uid != server->m.mounted_uid)) { return -EACCES; } @@ -190,7 +194,7 @@ int ncp_ioctl(struct inode *inode, struct file *filp, switch (cmd) { case NCP_IOC_NCPREQUEST: - if ((permission(inode, MAY_WRITE, NULL) != 0) + if ((file_permission(filp, MAY_WRITE) != 0) && (current->uid != server->m.mounted_uid)) { return -EACCES; } @@ -245,16 +249,16 @@ int ncp_ioctl(struct inode *inode, struct file *filp, return ncp_conn_logged_in(inode->i_sb); case NCP_IOC_GET_FS_INFO: - return ncp_get_fs_info(server, inode, argp); + return ncp_get_fs_info(server, filp, argp); case NCP_IOC_GET_FS_INFO_V2: - return ncp_get_fs_info_v2(server, inode, argp); + return ncp_get_fs_info_v2(server, filp, argp); case NCP_IOC_GETMOUNTUID2: { unsigned long tmp = server->m.mounted_uid; - if ( (permission(inode, MAY_READ, NULL) != 0) + if ((file_permission(filp, MAY_READ) != 0) && (current->uid != server->m.mounted_uid)) { return -EACCES; @@ -268,7 +272,7 @@ int ncp_ioctl(struct inode *inode, struct file *filp, { struct ncp_setroot_ioctl sr; - if ( (permission(inode, MAY_READ, NULL) != 0) + if ((file_permission(filp, MAY_READ) != 0) && (current->uid != server->m.mounted_uid)) { return -EACCES; @@ -343,7 +347,7 @@ int ncp_ioctl(struct inode *inode, struct file *filp, #ifdef CONFIG_NCPFS_PACKET_SIGNING case NCP_IOC_SIGN_INIT: - if ((permission(inode, MAY_WRITE, NULL) != 0) + if ((file_permission(filp, MAY_WRITE) != 0) && (current->uid != server->m.mounted_uid)) { return -EACCES; @@ -366,7 +370,7 @@ int ncp_ioctl(struct inode *inode, struct file *filp, return 0; case NCP_IOC_SIGN_WANTED: - if ( (permission(inode, MAY_READ, NULL) != 0) + if ((file_permission(filp, MAY_READ) != 0) && (current->uid != server->m.mounted_uid)) { return -EACCES; @@ -379,7 +383,7 @@ int ncp_ioctl(struct inode *inode, struct file *filp, { int newstate; - if ( (permission(inode, MAY_WRITE, NULL) != 0) + if ((file_permission(filp, MAY_WRITE) != 0) && (current->uid != server->m.mounted_uid)) { return -EACCES; @@ -400,7 +404,7 @@ int ncp_ioctl(struct inode *inode, struct file *filp, #ifdef CONFIG_NCPFS_IOCTL_LOCKING case NCP_IOC_LOCKUNLOCK: - if ( (permission(inode, MAY_WRITE, NULL) != 0) + if ((file_permission(filp, MAY_WRITE) != 0) && (current->uid != server->m.mounted_uid)) { return -EACCES; @@ -605,7 +609,7 @@ outrel: #endif /* CONFIG_NCPFS_NLS */ case NCP_IOC_SETDENTRYTTL: - if ((permission(inode, MAY_WRITE, NULL) != 0) && + if ((file_permission(filp, MAY_WRITE) != 0) && (current->uid != server->m.mounted_uid)) return -EACCES; { @@ -635,7 +639,7 @@ outrel: so we have this out of switch */ if (cmd == NCP_IOC_GETMOUNTUID) { __kernel_uid_t uid = 0; - if ((permission(inode, MAY_READ, NULL) != 0) + if ((file_permission(filp, MAY_READ) != 0) && (current->uid != server->m.mounted_uid)) { return -EACCES; } diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c index d7f7eb669d03..618a327027b3 100644 --- a/fs/nfs/delegation.c +++ b/fs/nfs/delegation.c @@ -31,11 +31,42 @@ static void nfs_free_delegation(struct nfs_delegation *delegation) kfree(delegation); } +static int nfs_delegation_claim_locks(struct nfs_open_context *ctx, struct nfs4_state *state) +{ + struct inode *inode = state->inode; + struct file_lock *fl; + int status; + + for (fl = inode->i_flock; fl != 0; fl = fl->fl_next) { + if (!(fl->fl_flags & (FL_POSIX|FL_FLOCK))) + continue; + if ((struct nfs_open_context *)fl->fl_file->private_data != ctx) + continue; + status = nfs4_lock_delegation_recall(state, fl); + if (status >= 0) + continue; + switch (status) { + default: + printk(KERN_ERR "%s: unhandled error %d.\n", + __FUNCTION__, status); + case -NFS4ERR_EXPIRED: + /* kill_proc(fl->fl_pid, SIGLOST, 1); */ + case -NFS4ERR_STALE_CLIENTID: + nfs4_schedule_state_recovery(NFS_SERVER(inode)->nfs4_state); + goto out_err; + } + } + return 0; +out_err: + return status; +} + static void nfs_delegation_claim_opens(struct inode *inode) { struct nfs_inode *nfsi = NFS_I(inode); struct nfs_open_context *ctx; struct nfs4_state *state; + int err; again: spin_lock(&inode->i_lock); @@ -47,9 +78,12 @@ again: continue; get_nfs_open_context(ctx); spin_unlock(&inode->i_lock); - if (nfs4_open_delegation_recall(ctx->dentry, state) < 0) - return; + err = nfs4_open_delegation_recall(ctx->dentry, state); + if (err >= 0) + err = nfs_delegation_claim_locks(ctx, state); put_nfs_open_context(ctx); + if (err != 0) + return; goto again; } spin_unlock(&inode->i_lock); @@ -85,6 +119,10 @@ int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, struct struct nfs_delegation *delegation; int status = 0; + /* Ensure we first revalidate the attributes and page cache! */ + if ((nfsi->cache_validity & (NFS_INO_REVAL_PAGECACHE|NFS_INO_INVALID_ATTR))) + __nfs_revalidate_inode(NFS_SERVER(inode), inode); + delegation = nfs_alloc_delegation(); if (delegation == NULL) return -ENOMEM; @@ -111,8 +149,7 @@ int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, struct } } spin_unlock(&clp->cl_lock); - if (delegation != NULL) - kfree(delegation); + kfree(delegation); return status; } @@ -138,7 +175,7 @@ static void nfs_msync_inode(struct inode *inode) /* * Basic procedure for returning a delegation to the server */ -int nfs_inode_return_delegation(struct inode *inode) +int __nfs_inode_return_delegation(struct inode *inode) { struct nfs4_client *clp = NFS_SERVER(inode)->nfs4_state; struct nfs_inode *nfsi = NFS_I(inode); diff --git a/fs/nfs/delegation.h b/fs/nfs/delegation.h index 3f6c45a29d6a..2fcc30de924b 100644 --- a/fs/nfs/delegation.h +++ b/fs/nfs/delegation.h @@ -25,7 +25,7 @@ struct nfs_delegation { int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, struct nfs_openres *res); void nfs_inode_reclaim_delegation(struct inode *inode, struct rpc_cred *cred, struct nfs_openres *res); -int nfs_inode_return_delegation(struct inode *inode); +int __nfs_inode_return_delegation(struct inode *inode); int nfs_async_inode_return_delegation(struct inode *inode, const nfs4_stateid *stateid); struct inode *nfs_delegation_find_inode(struct nfs4_client *clp, const struct nfs_fh *fhandle); @@ -38,6 +38,7 @@ void nfs_delegation_reap_unclaimed(struct nfs4_client *clp); /* NFSv4 delegation-related procedures */ int nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, const nfs4_stateid *stateid); int nfs4_open_delegation_recall(struct dentry *dentry, struct nfs4_state *state); +int nfs4_lock_delegation_recall(struct nfs4_state *state, struct file_lock *fl); static inline int nfs_have_delegation(struct inode *inode, int flags) { @@ -47,11 +48,25 @@ static inline int nfs_have_delegation(struct inode *inode, int flags) return 1; return 0; } + +static inline int nfs_inode_return_delegation(struct inode *inode) +{ + int err = 0; + + if (NFS_I(inode)->delegation != NULL) + err = __nfs_inode_return_delegation(inode); + return err; +} #else static inline int nfs_have_delegation(struct inode *inode, int flags) { return 0; } + +static inline int nfs_inode_return_delegation(struct inode *inode) +{ + return 0; +} #endif #endif diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 2df639f143e8..c0d1a214572c 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -532,6 +532,7 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir) my_entry.eof = 0; my_entry.fh = &fh; my_entry.fattr = &fattr; + nfs_fattr_init(&fattr); desc->entry = &my_entry; while(!desc->entry->eof) { @@ -565,8 +566,6 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir) } } unlock_kernel(); - if (desc->error < 0) - return desc->error; if (res < 0) return res; return 0; @@ -803,6 +802,7 @@ static int nfs_dentry_delete(struct dentry *dentry) */ static void nfs_dentry_iput(struct dentry *dentry, struct inode *inode) { + nfs_inode_return_delegation(inode); if (dentry->d_flags & DCACHE_NFSFS_RENAMED) { lock_kernel(); inode->i_nlink--; @@ -853,12 +853,6 @@ static struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, stru dentry->d_op = NFS_PROTO(dir)->dentry_ops; lock_kernel(); - /* Revalidate parent directory attribute cache */ - error = nfs_revalidate_inode(NFS_SERVER(dir), dir); - if (error < 0) { - res = ERR_PTR(error); - goto out_unlock; - } /* If we're doing an exclusive create, optimize away the lookup */ if (nfs_is_exclusive_create(dir, nd)) @@ -916,7 +910,6 @@ static int is_atomic_open(struct inode *dir, struct nameidata *nd) static struct dentry *nfs_atomic_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) { struct dentry *res = NULL; - struct inode *inode = NULL; int error; /* Check that we are indeed trying to open this file */ @@ -930,8 +923,10 @@ static struct dentry *nfs_atomic_lookup(struct inode *dir, struct dentry *dentry dentry->d_op = NFS_PROTO(dir)->dentry_ops; /* Let vfs_create() deal with O_EXCL */ - if (nd->intent.open.flags & O_EXCL) - goto no_entry; + if (nd->intent.open.flags & O_EXCL) { + d_add(dentry, NULL); + goto out; + } /* Open the file on the server */ lock_kernel(); @@ -945,32 +940,30 @@ static struct dentry *nfs_atomic_lookup(struct inode *dir, struct dentry *dentry if (nd->intent.open.flags & O_CREAT) { nfs_begin_data_update(dir); - inode = nfs4_atomic_open(dir, dentry, nd); + res = nfs4_atomic_open(dir, dentry, nd); nfs_end_data_update(dir); } else - inode = nfs4_atomic_open(dir, dentry, nd); + res = nfs4_atomic_open(dir, dentry, nd); unlock_kernel(); - if (IS_ERR(inode)) { - error = PTR_ERR(inode); + if (IS_ERR(res)) { + error = PTR_ERR(res); switch (error) { /* Make a negative dentry */ case -ENOENT: - inode = NULL; - break; + res = NULL; + goto out; /* This turned out not to be a regular file */ + case -EISDIR: + case -ENOTDIR: + goto no_open; case -ELOOP: if (!(nd->intent.open.flags & O_NOFOLLOW)) goto no_open; - /* case -EISDIR: */ /* case -EINVAL: */ default: - res = ERR_PTR(error); goto out; } - } -no_entry: - res = d_add_unique(dentry, inode); - if (res != NULL) + } else if (res != NULL) dentry = res; nfs_renew_times(dentry); nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); @@ -1014,7 +1007,7 @@ static int nfs_open_revalidate(struct dentry *dentry, struct nameidata *nd) */ lock_kernel(); verifier = nfs_save_change_attribute(dir); - ret = nfs4_open_revalidate(dir, dentry, openflags); + ret = nfs4_open_revalidate(dir, dentry, openflags, nd); if (!ret) nfs_set_verifier(dentry, verifier); unlock_kernel(); @@ -1137,7 +1130,7 @@ static int nfs_create(struct inode *dir, struct dentry *dentry, int mode, lock_kernel(); nfs_begin_data_update(dir); - error = NFS_PROTO(dir)->create(dir, dentry, &attr, open_flags); + error = NFS_PROTO(dir)->create(dir, dentry, &attr, open_flags, nd); nfs_end_data_update(dir); if (error != 0) goto out_err; @@ -1264,6 +1257,9 @@ dentry->d_parent->d_name.name, dentry->d_name.name); sprintf(silly, ".nfs%*.*lx", i_inosize, i_inosize, dentry->d_inode->i_ino); + /* Return delegation in anticipation of the rename */ + nfs_inode_return_delegation(dentry->d_inode); + sdentry = NULL; do { char *suffix = silly + slen - countersize; @@ -1291,6 +1287,7 @@ dentry->d_parent->d_name.name, dentry->d_name.name); nfs_begin_data_update(dentry->d_inode); error = NFS_PROTO(dir)->rename(dir, &dentry->d_name, dir, &qsilly); + nfs_mark_for_revalidate(dentry->d_inode); nfs_end_data_update(dentry->d_inode); } else error = NFS_PROTO(dir)->rename(dir, &dentry->d_name, @@ -1332,11 +1329,13 @@ static int nfs_safe_remove(struct dentry *dentry) nfs_begin_data_update(dir); if (inode != NULL) { + nfs_inode_return_delegation(inode); nfs_begin_data_update(inode); error = NFS_PROTO(dir)->remove(dir, &dentry->d_name); /* The VFS may want to delete this inode */ if (error == 0) inode->i_nlink--; + nfs_mark_for_revalidate(inode); nfs_end_data_update(inode); } else error = NFS_PROTO(dir)->remove(dir, &dentry->d_name); @@ -1438,17 +1437,14 @@ nfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *dentry) old_dentry->d_parent->d_name.name, old_dentry->d_name.name, dentry->d_parent->d_name.name, dentry->d_name.name); - /* - * Drop the dentry in advance to force a new lookup. - * Since nfs_proc_link doesn't return a file handle, - * we can't use the existing dentry. - */ lock_kernel(); - d_drop(dentry); - nfs_begin_data_update(dir); nfs_begin_data_update(inode); error = NFS_PROTO(dir)->link(inode, dir, &dentry->d_name); + if (error == 0) { + atomic_inc(&inode->i_count); + d_instantiate(dentry, inode); + } nfs_end_data_update(inode); nfs_end_data_update(dir); unlock_kernel(); @@ -1512,9 +1508,11 @@ static int nfs_rename(struct inode *old_dir, struct dentry *old_dentry, */ if (!new_inode) goto go_ahead; - if (S_ISDIR(new_inode->i_mode)) - goto out; - else if (atomic_read(&new_dentry->d_count) > 2) { + if (S_ISDIR(new_inode->i_mode)) { + error = -EISDIR; + if (!S_ISDIR(old_inode->i_mode)) + goto out; + } else if (atomic_read(&new_dentry->d_count) > 2) { int err; /* copy the target dentry's name */ dentry = d_alloc(new_dentry->d_parent, @@ -1539,7 +1537,8 @@ static int nfs_rename(struct inode *old_dir, struct dentry *old_dentry, #endif goto out; } - } + } else + new_inode->i_nlink--; go_ahead: /* @@ -1549,6 +1548,7 @@ go_ahead: nfs_wb_all(old_inode); shrink_dcache_parent(old_dentry); } + nfs_inode_return_delegation(old_inode); if (new_inode) d_delete(new_dentry); @@ -1558,6 +1558,7 @@ go_ahead: nfs_begin_data_update(old_inode); error = NFS_PROTO(old_dir)->rename(old_dir, &old_dentry->d_name, new_dir, &new_dentry->d_name); + nfs_mark_for_revalidate(old_inode); nfs_end_data_update(old_inode); nfs_end_data_update(new_dir); nfs_end_data_update(old_dir); diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index 6537f2c4ae44..b497c71384e8 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -655,7 +655,6 @@ nfs_file_direct_read(struct kiocb *iocb, char __user *buf, size_t count, loff_t struct file *file = iocb->ki_filp; struct nfs_open_context *ctx = (struct nfs_open_context *) file->private_data; - struct dentry *dentry = file->f_dentry; struct address_space *mapping = file->f_mapping; struct inode *inode = mapping->host; struct iovec iov = { @@ -664,7 +663,8 @@ nfs_file_direct_read(struct kiocb *iocb, char __user *buf, size_t count, loff_t }; dprintk("nfs: direct read(%s/%s, %lu@%lu)\n", - dentry->d_parent->d_name.name, dentry->d_name.name, + file->f_dentry->d_parent->d_name.name, + file->f_dentry->d_name.name, (unsigned long) count, (unsigned long) pos); if (!is_sync_kiocb(iocb)) @@ -730,7 +730,6 @@ nfs_file_direct_write(struct kiocb *iocb, const char __user *buf, size_t count, struct file *file = iocb->ki_filp; struct nfs_open_context *ctx = (struct nfs_open_context *) file->private_data; - struct dentry *dentry = file->f_dentry; struct address_space *mapping = file->f_mapping; struct inode *inode = mapping->host; struct iovec iov = { @@ -739,8 +738,9 @@ nfs_file_direct_write(struct kiocb *iocb, const char __user *buf, size_t count, }; dfprintk(VFS, "nfs: direct write(%s/%s(%ld), %lu@%lu)\n", - dentry->d_parent->d_name.name, dentry->d_name.name, - inode->i_ino, (unsigned long) count, (unsigned long) pos); + file->f_dentry->d_parent->d_name.name, + file->f_dentry->d_name.name, inode->i_ino, + (unsigned long) count, (unsigned long) pos); if (!is_sync_kiocb(iocb)) goto out; diff --git a/fs/nfs/file.c b/fs/nfs/file.c index f6b9eda925c5..57d3e77d97ee 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -137,7 +137,8 @@ static int nfs_revalidate_file(struct inode *inode, struct file *filp) struct nfs_inode *nfsi = NFS_I(inode); int retval = 0; - if ((nfsi->cache_validity & NFS_INO_REVAL_PAGECACHE) || nfs_attribute_timeout(inode)) + if ((nfsi->cache_validity & (NFS_INO_REVAL_PAGECACHE|NFS_INO_INVALID_ATTR)) + || nfs_attribute_timeout(inode)) retval = __nfs_revalidate_inode(NFS_SERVER(inode), inode); nfs_revalidate_mapping(inode, filp->f_mapping); return 0; @@ -204,8 +205,8 @@ nfs_file_flush(struct file *file) if (!status) { status = ctx->error; ctx->error = 0; - if (!status && !nfs_have_delegation(inode, FMODE_READ)) - __nfs_revalidate_inode(NFS_SERVER(inode), inode); + if (!status) + nfs_revalidate_inode(NFS_SERVER(inode), inode); } unlock_kernel(); return status; @@ -375,22 +376,31 @@ out_swapfile: static int do_getlk(struct file *filp, int cmd, struct file_lock *fl) { + struct file_lock *cfl; struct inode *inode = filp->f_mapping->host; int status = 0; lock_kernel(); - /* Use local locking if mounted with "-onolock" */ - if (!(NFS_SERVER(inode)->flags & NFS_MOUNT_NONLM)) - status = NFS_PROTO(inode)->lock(filp, cmd, fl); - else { - struct file_lock *cfl = posix_test_lock(filp, fl); - - fl->fl_type = F_UNLCK; - if (cfl != NULL) - memcpy(fl, cfl, sizeof(*fl)); + /* Try local locking first */ + cfl = posix_test_lock(filp, fl); + if (cfl != NULL) { + locks_copy_lock(fl, cfl); + goto out; } + + if (nfs_have_delegation(inode, FMODE_READ)) + goto out_noconflict; + + if (NFS_SERVER(inode)->flags & NFS_MOUNT_NONLM) + goto out_noconflict; + + status = NFS_PROTO(inode)->lock(filp, cmd, fl); +out: unlock_kernel(); return status; +out_noconflict: + fl->fl_type = F_UNLCK; + goto out; } static int do_vfs_lock(struct file *file, struct file_lock *fl) diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 6922469d6fc5..afd75d0463fd 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -54,7 +54,7 @@ #define NFS_MAX_READAHEAD (RPC_DEF_SLOT_TABLE - 1) static void nfs_invalidate_inode(struct inode *); -static int nfs_update_inode(struct inode *, struct nfs_fattr *, unsigned long); +static int nfs_update_inode(struct inode *, struct nfs_fattr *); static struct inode *nfs_alloc_inode(struct super_block *sb); static void nfs_destroy_inode(struct inode *); @@ -358,6 +358,35 @@ out_no_root: return no_root_error; } +static void nfs_init_timeout_values(struct rpc_timeout *to, int proto, unsigned int timeo, unsigned int retrans) +{ + to->to_initval = timeo * HZ / 10; + to->to_retries = retrans; + if (!to->to_retries) + to->to_retries = 2; + + switch (proto) { + case IPPROTO_TCP: + if (!to->to_initval) + to->to_initval = 60 * HZ; + if (to->to_initval > NFS_MAX_TCP_TIMEOUT) + to->to_initval = NFS_MAX_TCP_TIMEOUT; + to->to_increment = to->to_initval; + to->to_maxval = to->to_initval + (to->to_increment * to->to_retries); + to->to_exponential = 0; + break; + case IPPROTO_UDP: + default: + if (!to->to_initval) + to->to_initval = 11 * HZ / 10; + if (to->to_initval > NFS_MAX_UDP_TIMEOUT) + to->to_initval = NFS_MAX_UDP_TIMEOUT; + to->to_maxval = NFS_MAX_UDP_TIMEOUT; + to->to_exponential = 1; + break; + } +} + /* * Create an RPC client handle. */ @@ -367,22 +396,12 @@ nfs_create_client(struct nfs_server *server, const struct nfs_mount_data *data) struct rpc_timeout timeparms; struct rpc_xprt *xprt = NULL; struct rpc_clnt *clnt = NULL; - int tcp = (data->flags & NFS_MOUNT_TCP); + int proto = (data->flags & NFS_MOUNT_TCP) ? IPPROTO_TCP : IPPROTO_UDP; - /* Initialize timeout values */ - timeparms.to_initval = data->timeo * HZ / 10; - timeparms.to_retries = data->retrans; - timeparms.to_maxval = tcp ? RPC_MAX_TCP_TIMEOUT : RPC_MAX_UDP_TIMEOUT; - timeparms.to_exponential = 1; - - if (!timeparms.to_initval) - timeparms.to_initval = (tcp ? 600 : 11) * HZ / 10; - if (!timeparms.to_retries) - timeparms.to_retries = 5; + nfs_init_timeout_values(&timeparms, proto, data->timeo, data->retrans); /* create transport and client */ - xprt = xprt_create_proto(tcp ? IPPROTO_TCP : IPPROTO_UDP, - &server->addr, &timeparms); + xprt = xprt_create_proto(proto, &server->addr, &timeparms); if (IS_ERR(xprt)) { dprintk("%s: cannot create RPC transport. Error = %ld\n", __FUNCTION__, PTR_ERR(xprt)); @@ -576,7 +595,6 @@ static int nfs_show_options(struct seq_file *m, struct vfsmount *mnt) { NFS_MOUNT_SOFT, ",soft", ",hard" }, { NFS_MOUNT_INTR, ",intr", "" }, { NFS_MOUNT_POSIX, ",posix", "" }, - { NFS_MOUNT_TCP, ",tcp", ",udp" }, { NFS_MOUNT_NOCTO, ",nocto", "" }, { NFS_MOUNT_NOAC, ",noac", "" }, { NFS_MOUNT_NONLM, ",nolock", ",lock" }, @@ -585,6 +603,8 @@ static int nfs_show_options(struct seq_file *m, struct vfsmount *mnt) }; struct proc_nfs_info *nfs_infop; struct nfs_server *nfss = NFS_SB(mnt->mnt_sb); + char buf[12]; + char *proto; seq_printf(m, ",v%d", nfss->rpc_ops->version); seq_printf(m, ",rsize=%d", nfss->rsize); @@ -603,6 +623,18 @@ static int nfs_show_options(struct seq_file *m, struct vfsmount *mnt) else seq_puts(m, nfs_infop->nostr); } + switch (nfss->client->cl_xprt->prot) { + case IPPROTO_TCP: + proto = "tcp"; + break; + case IPPROTO_UDP: + proto = "udp"; + break; + default: + snprintf(buf, sizeof(buf), "%u", nfss->client->cl_xprt->prot); + proto = buf; + } + seq_printf(m, ",proto=%s", proto); seq_puts(m, ",addr="); seq_escape(m, nfss->hostname, " \t\n\\"); return 0; @@ -611,14 +643,11 @@ static int nfs_show_options(struct seq_file *m, struct vfsmount *mnt) /* * Invalidate the local caches */ -void -nfs_zap_caches(struct inode *inode) +static void nfs_zap_caches_locked(struct inode *inode) { struct nfs_inode *nfsi = NFS_I(inode); int mode = inode->i_mode; - spin_lock(&inode->i_lock); - NFS_ATTRTIMEO(inode) = NFS_MINATTRTIMEO(inode); NFS_ATTRTIMEO_UPDATE(inode) = jiffies; @@ -627,7 +656,12 @@ nfs_zap_caches(struct inode *inode) nfsi->cache_validity |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL|NFS_INO_REVAL_PAGECACHE; else nfsi->cache_validity |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL|NFS_INO_REVAL_PAGECACHE; +} +void nfs_zap_caches(struct inode *inode) +{ + spin_lock(&inode->i_lock); + nfs_zap_caches_locked(inode); spin_unlock(&inode->i_lock); } @@ -644,16 +678,13 @@ static void nfs_zap_acl_cache(struct inode *inode) } /* - * Invalidate, but do not unhash, the inode + * Invalidate, but do not unhash, the inode. + * NB: must be called with inode->i_lock held! */ -static void -nfs_invalidate_inode(struct inode *inode) +static void nfs_invalidate_inode(struct inode *inode) { - umode_t save_mode = inode->i_mode; - - make_bad_inode(inode); - inode->i_mode = save_mode; - nfs_zap_caches(inode); + set_bit(NFS_INO_STALE, &NFS_FLAGS(inode)); + nfs_zap_caches_locked(inode); } struct nfs_find_desc { @@ -753,7 +784,8 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr) else init_special_inode(inode, inode->i_mode, fattr->rdev); - nfsi->read_cache_jiffies = fattr->timestamp; + nfsi->read_cache_jiffies = fattr->time_start; + nfsi->last_updated = jiffies; inode->i_atime = fattr->atime; inode->i_mtime = fattr->mtime; inode->i_ctime = fattr->ctime; @@ -821,6 +853,11 @@ nfs_setattr(struct dentry *dentry, struct iattr *attr) filemap_fdatawait(inode->i_mapping); nfs_wb_all(inode); } + /* + * Return any delegations if we're going to change ACLs + */ + if ((attr->ia_valid & (ATTR_MODE|ATTR_UID|ATTR_GID)) != 0) + nfs_inode_return_delegation(inode); error = NFS_PROTO(inode)->setattr(dentry, &fattr, attr); if (error == 0) nfs_refresh_inode(inode, &fattr); @@ -877,12 +914,10 @@ static int nfs_wait_on_inode(struct inode *inode) sigset_t oldmask; int error; - atomic_inc(&inode->i_count); rpc_clnt_sigmask(clnt, &oldmask); error = wait_on_bit_lock(&nfsi->flags, NFS_INO_REVALIDATING, nfs_wait_schedule, TASK_INTERRUPTIBLE); rpc_clnt_sigunmask(clnt, &oldmask); - iput(inode); return error; } @@ -973,13 +1008,18 @@ void nfs_file_set_open_context(struct file *filp, struct nfs_open_context *ctx) spin_unlock(&inode->i_lock); } -struct nfs_open_context *nfs_find_open_context(struct inode *inode, int mode) +/* + * Given an inode, search for an open context with the desired characteristics + */ +struct nfs_open_context *nfs_find_open_context(struct inode *inode, struct rpc_cred *cred, int mode) { struct nfs_inode *nfsi = NFS_I(inode); struct nfs_open_context *pos, *ctx = NULL; spin_lock(&inode->i_lock); list_for_each_entry(pos, &nfsi->open_files, list) { + if (cred != NULL && pos->cred != cred) + continue; if ((pos->mode & mode) == mode) { ctx = get_nfs_open_context(pos); break; @@ -1021,15 +1061,11 @@ int nfs_open(struct inode *inode, struct file *filp) ctx->mode = filp->f_mode; nfs_file_set_open_context(filp, ctx); put_nfs_open_context(ctx); - if ((filp->f_mode & FMODE_WRITE) != 0) - nfs_begin_data_update(inode); return 0; } int nfs_release(struct inode *inode, struct file *filp) { - if ((filp->f_mode & FMODE_WRITE) != 0) - nfs_end_data_update(inode); nfs_file_clear_open_context(filp); return 0; } @@ -1044,8 +1080,6 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) int status = -ESTALE; struct nfs_fattr fattr; struct nfs_inode *nfsi = NFS_I(inode); - unsigned long verifier; - unsigned long cache_validity; dfprintk(PAGECACHE, "NFS: revalidating (%s/%Ld)\n", inode->i_sb->s_id, (long long)NFS_FILEID(inode)); @@ -1070,8 +1104,6 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) } } - /* Protect against RPC races by saving the change attribute */ - verifier = nfs_save_change_attribute(inode); status = NFS_PROTO(inode)->getattr(server, NFS_FH(inode), &fattr); if (status != 0) { dfprintk(PAGECACHE, "nfs_revalidate_inode: (%s/%Ld) getattr failed, error=%d\n", @@ -1085,28 +1117,20 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) goto out; } - status = nfs_update_inode(inode, &fattr, verifier); + spin_lock(&inode->i_lock); + status = nfs_update_inode(inode, &fattr); if (status) { + spin_unlock(&inode->i_lock); dfprintk(PAGECACHE, "nfs_revalidate_inode: (%s/%Ld) refresh failed, error=%d\n", inode->i_sb->s_id, (long long)NFS_FILEID(inode), status); goto out; } - spin_lock(&inode->i_lock); - cache_validity = nfsi->cache_validity; - nfsi->cache_validity &= ~NFS_INO_REVAL_PAGECACHE; - - /* - * We may need to keep the attributes marked as invalid if - * we raced with nfs_end_attr_update(). - */ - if (verifier == nfsi->cache_change_attribute) - nfsi->cache_validity &= ~(NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ATIME); spin_unlock(&inode->i_lock); nfs_revalidate_mapping(inode, inode->i_mapping); - if (cache_validity & NFS_INO_INVALID_ACL) + if (nfsi->cache_validity & NFS_INO_INVALID_ACL) nfs_zap_acl_cache(inode); dfprintk(PAGECACHE, "NFS: (%s/%Ld) revalidation complete\n", @@ -1167,7 +1191,7 @@ void nfs_revalidate_mapping(struct inode *inode, struct address_space *mapping) if (S_ISDIR(inode->i_mode)) { memset(nfsi->cookieverf, 0, sizeof(nfsi->cookieverf)); /* This ensures we revalidate child dentries */ - nfsi->cache_change_attribute++; + nfsi->cache_change_attribute = jiffies; } spin_unlock(&inode->i_lock); @@ -1199,20 +1223,19 @@ void nfs_end_data_update(struct inode *inode) struct nfs_inode *nfsi = NFS_I(inode); if (!nfs_have_delegation(inode, FMODE_READ)) { - /* Mark the attribute cache for revalidation */ - spin_lock(&inode->i_lock); - nfsi->cache_validity |= NFS_INO_INVALID_ATTR; - /* Directories and symlinks: invalidate page cache too */ - if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode)) + /* Directories and symlinks: invalidate page cache */ + if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode)) { + spin_lock(&inode->i_lock); nfsi->cache_validity |= NFS_INO_INVALID_DATA; - spin_unlock(&inode->i_lock); + spin_unlock(&inode->i_lock); + } } - nfsi->cache_change_attribute ++; + nfsi->cache_change_attribute = jiffies; atomic_dec(&nfsi->data_updates); } /** - * nfs_refresh_inode - verify consistency of the inode attribute cache + * nfs_check_inode_attributes - verify consistency of the inode attribute cache * @inode - pointer to inode * @fattr - updated attributes * @@ -1220,17 +1243,12 @@ void nfs_end_data_update(struct inode *inode) * so that fattr carries weak cache consistency data, then it may * also update the ctime/mtime/change_attribute. */ -int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr) +static int nfs_check_inode_attributes(struct inode *inode, struct nfs_fattr *fattr) { struct nfs_inode *nfsi = NFS_I(inode); loff_t cur_size, new_isize; int data_unstable; - /* Do we hold a delegation? */ - if (nfs_have_delegation(inode, FMODE_READ)) - return 0; - - spin_lock(&inode->i_lock); /* Are we in the process of updating data on the server? */ data_unstable = nfs_caches_unstable(inode); @@ -1247,14 +1265,12 @@ int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr) } if ((fattr->valid & NFS_ATTR_FATTR) == 0) { - spin_unlock(&inode->i_lock); return 0; } /* Has the inode gone and changed behind our back? */ if (nfsi->fileid != fattr->fileid || (inode->i_mode & S_IFMT) != (fattr->mode & S_IFMT)) { - spin_unlock(&inode->i_lock); return -EIO; } @@ -1294,11 +1310,62 @@ int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr) if (!timespec_equal(&inode->i_atime, &fattr->atime)) nfsi->cache_validity |= NFS_INO_INVALID_ATIME; - nfsi->read_cache_jiffies = fattr->timestamp; - spin_unlock(&inode->i_lock); + nfsi->read_cache_jiffies = fattr->time_start; return 0; } +/** + * nfs_refresh_inode - try to update the inode attribute cache + * @inode - pointer to inode + * @fattr - updated attributes + * + * Check that an RPC call that returned attributes has not overlapped with + * other recent updates of the inode metadata, then decide whether it is + * safe to do a full update of the inode attributes, or whether just to + * call nfs_check_inode_attributes. + */ +int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr) +{ + struct nfs_inode *nfsi = NFS_I(inode); + int status; + + if ((fattr->valid & NFS_ATTR_FATTR) == 0) + return 0; + spin_lock(&inode->i_lock); + nfsi->cache_validity &= ~NFS_INO_REVAL_PAGECACHE; + if (time_after(fattr->time_start, nfsi->last_updated)) + status = nfs_update_inode(inode, fattr); + else + status = nfs_check_inode_attributes(inode, fattr); + + spin_unlock(&inode->i_lock); + return status; +} + +/** + * nfs_post_op_update_inode - try to update the inode attribute cache + * @inode - pointer to inode + * @fattr - updated attributes + * + * After an operation that has changed the inode metadata, mark the + * attribute cache as being invalid, then try to update it. + */ +int nfs_post_op_update_inode(struct inode *inode, struct nfs_fattr *fattr) +{ + struct nfs_inode *nfsi = NFS_I(inode); + int status = 0; + + spin_lock(&inode->i_lock); + if (unlikely((fattr->valid & NFS_ATTR_FATTR) == 0)) { + nfsi->cache_validity |= NFS_INO_INVALID_ATTR | NFS_INO_INVALID_ACCESS; + goto out; + } + status = nfs_update_inode(inode, fattr); +out: + spin_unlock(&inode->i_lock); + return status; +} + /* * Many nfs protocol calls return the new file attributes after * an operation. Here we update the inode to reflect the state @@ -1311,12 +1378,12 @@ int nfs_refresh_inode(struct inode *inode, struct nfs_fattr *fattr) * * A very similar scenario holds for the dir cache. */ -static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr, unsigned long verifier) +static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) { struct nfs_inode *nfsi = NFS_I(inode); loff_t cur_isize, new_isize; unsigned int invalid = 0; - int data_unstable; + int data_stable; dfprintk(VFS, "NFS: %s(%s/%ld ct=%d info=0x%x)\n", __FUNCTION__, inode->i_sb->s_id, inode->i_ino, @@ -1334,23 +1401,22 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr, unsign goto out_err; } - spin_lock(&inode->i_lock); - /* * Make sure the inode's type hasn't changed. */ - if ((inode->i_mode & S_IFMT) != (fattr->mode & S_IFMT)) { - spin_unlock(&inode->i_lock); + if ((inode->i_mode & S_IFMT) != (fattr->mode & S_IFMT)) goto out_changed; - } /* * Update the read time so we don't revalidate too often. */ - nfsi->read_cache_jiffies = fattr->timestamp; + nfsi->read_cache_jiffies = fattr->time_start; + nfsi->last_updated = jiffies; /* Are we racing with known updates of the metadata on the server? */ - data_unstable = ! nfs_verify_change_attribute(inode, verifier); + data_stable = nfs_verify_change_attribute(inode, fattr->time_start); + if (data_stable) + nfsi->cache_validity &= ~(NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ATIME); /* Check if our cached file size is stale */ new_isize = nfs_size_to_loff_t(fattr->size); @@ -1359,7 +1425,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr, unsign /* Do we perhaps have any outstanding writes? */ if (nfsi->npages == 0) { /* No, but did we race with nfs_end_data_update()? */ - if (verifier == nfsi->cache_change_attribute) { + if (data_stable) { inode->i_size = new_isize; invalid |= NFS_INO_INVALID_DATA; } @@ -1368,6 +1434,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr, unsign inode->i_size = new_isize; invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA; } + nfsi->cache_change_attribute = jiffies; dprintk("NFS: isize change on server for file %s/%ld\n", inode->i_sb->s_id, inode->i_ino); } @@ -1377,8 +1444,8 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr, unsign memcpy(&inode->i_mtime, &fattr->mtime, sizeof(inode->i_mtime)); dprintk("NFS: mtime change on server for file %s/%ld\n", inode->i_sb->s_id, inode->i_ino); - if (!data_unstable) - invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA; + invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA; + nfsi->cache_change_attribute = jiffies; } if ((fattr->valid & NFS_ATTR_FATTR_V4) @@ -1386,15 +1453,15 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr, unsign dprintk("NFS: change_attr change on server for file %s/%ld\n", inode->i_sb->s_id, inode->i_ino); nfsi->change_attr = fattr->change_attr; - if (!data_unstable) - invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL; + invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL; + nfsi->cache_change_attribute = jiffies; } /* If ctime has changed we should definitely clear access+acl caches */ if (!timespec_equal(&inode->i_ctime, &fattr->ctime)) { - if (!data_unstable) - invalid |= NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL; + invalid |= NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL; memcpy(&inode->i_ctime, &fattr->ctime, sizeof(inode->i_ctime)); + nfsi->cache_change_attribute = jiffies; } memcpy(&inode->i_atime, &fattr->atime, sizeof(inode->i_atime)); @@ -1432,10 +1499,11 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr, unsign if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode))) invalid &= ~NFS_INO_INVALID_DATA; + if (data_stable) + invalid &= ~(NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ATIME|NFS_INO_REVAL_PAGECACHE); if (!nfs_have_delegation(inode, FMODE_READ)) nfsi->cache_validity |= invalid; - spin_unlock(&inode->i_lock); return 0; out_changed: /* @@ -1445,14 +1513,13 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr, unsign printk(KERN_DEBUG "%s: inode %ld mode changed, %07o to %07o\n", __FUNCTION__, inode->i_ino, inode->i_mode, fattr->mode); #endif + out_err: /* * No need to worry about unhashing the dentry, as the * lookup validation will know that the inode is bad. * (But we fall through to invalidate the caches.) */ nfs_invalidate_inode(inode); - out_err: - set_bit(NFS_INO_STALE, &NFS_FLAGS(inode)); return -ESTALE; } @@ -1605,8 +1672,7 @@ static void nfs_kill_super(struct super_block *s) rpciod_down(); /* release rpciod */ - if (server->hostname != NULL) - kfree(server->hostname); + kfree(server->hostname); kfree(server); } @@ -1644,8 +1710,7 @@ static void nfs4_clear_inode(struct inode *inode) struct nfs_inode *nfsi = NFS_I(inode); /* If we are holding a delegation, return it! */ - if (nfsi->delegation != NULL) - nfs_inode_return_delegation(inode); + nfs_inode_return_delegation(inode); /* First call standard NFS clear_inode() code */ nfs_clear_inode(inode); /* Now clear out any remaining state */ @@ -1674,7 +1739,7 @@ static int nfs4_fill_super(struct super_block *sb, struct nfs4_mount_data *data, struct rpc_clnt *clnt = NULL; struct rpc_timeout timeparms; rpc_authflavor_t authflavour; - int proto, err = -EIO; + int err = -EIO; sb->s_blocksize_bits = 0; sb->s_blocksize = 0; @@ -1692,30 +1757,8 @@ static int nfs4_fill_super(struct super_block *sb, struct nfs4_mount_data *data, server->acdirmax = data->acdirmax*HZ; server->rpc_ops = &nfs_v4_clientops; - /* Initialize timeout values */ - - timeparms.to_initval = data->timeo * HZ / 10; - timeparms.to_retries = data->retrans; - timeparms.to_exponential = 1; - if (!timeparms.to_retries) - timeparms.to_retries = 5; - proto = data->proto; - /* Which IP protocol do we use? */ - switch (proto) { - case IPPROTO_TCP: - timeparms.to_maxval = RPC_MAX_TCP_TIMEOUT; - if (!timeparms.to_initval) - timeparms.to_initval = 600 * HZ / 10; - break; - case IPPROTO_UDP: - timeparms.to_maxval = RPC_MAX_UDP_TIMEOUT; - if (!timeparms.to_initval) - timeparms.to_initval = 11 * HZ / 10; - break; - default: - return -EINVAL; - } + nfs_init_timeout_values(&timeparms, data->proto, data->timeo, data->retrans); clp = nfs4_get_client(&server->addr.sin_addr); if (!clp) { @@ -1740,7 +1783,7 @@ static int nfs4_fill_super(struct super_block *sb, struct nfs4_mount_data *data, down_write(&clp->cl_sem); if (IS_ERR(clp->cl_rpcclient)) { - xprt = xprt_create_proto(proto, &server->addr, &timeparms); + xprt = xprt_create_proto(data->proto, &server->addr, &timeparms); if (IS_ERR(xprt)) { up_write(&clp->cl_sem); err = PTR_ERR(xprt); @@ -1848,8 +1891,7 @@ nfs_copy_user_string(char *dst, struct nfs_string *src, int maxlen) return ERR_PTR(-ENOMEM); } if (copy_from_user(dst, src->data, maxlen)) { - if (p != NULL) - kfree(p); + kfree(p); return ERR_PTR(-EFAULT); } dst[maxlen] = '\0'; @@ -1940,10 +1982,8 @@ static struct super_block *nfs4_get_sb(struct file_system_type *fs_type, out_err: s = (struct super_block *)p; out_free: - if (server->mnt_path) - kfree(server->mnt_path); - if (server->hostname) - kfree(server->hostname); + kfree(server->mnt_path); + kfree(server->hostname); kfree(server); return s; } @@ -1963,8 +2003,7 @@ static void nfs4_kill_super(struct super_block *sb) destroy_nfsv4_state(server); - if (server->hostname != NULL) - kfree(server->hostname); + kfree(server->hostname); kfree(server); } @@ -2013,6 +2052,7 @@ static struct inode *nfs_alloc_inode(struct super_block *sb) return NULL; nfsi->flags = 0UL; nfsi->cache_validity = 0UL; + nfsi->cache_change_attribute = jiffies; #ifdef CONFIG_NFS_V3_ACL nfsi->acl_access = ERR_PTR(-EAGAIN); nfsi->acl_default = ERR_PTR(-EAGAIN); diff --git a/fs/nfs/nfs2xdr.c b/fs/nfs/nfs2xdr.c index d91b69044a4d..59049e864ca7 100644 --- a/fs/nfs/nfs2xdr.c +++ b/fs/nfs/nfs2xdr.c @@ -143,7 +143,6 @@ xdr_decode_fattr(u32 *p, struct nfs_fattr *fattr) fattr->mode = (fattr->mode & ~S_IFMT) | S_IFIFO; fattr->rdev = 0; } - fattr->timestamp = jiffies; return p; } diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c index edc95514046d..92c870d19ccd 100644 --- a/fs/nfs/nfs3proc.c +++ b/fs/nfs/nfs3proc.c @@ -78,7 +78,7 @@ nfs3_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle, int status; dprintk("%s: call fsinfo\n", __FUNCTION__); - info->fattr->valid = 0; + nfs_fattr_init(info->fattr); status = rpc_call(server->client_sys, NFS3PROC_FSINFO, fhandle, info, 0); dprintk("%s: reply fsinfo: %d\n", __FUNCTION__, status); if (!(info->fattr->valid & NFS_ATTR_FATTR)) { @@ -98,7 +98,7 @@ nfs3_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, int status; dprintk("NFS call getattr\n"); - fattr->valid = 0; + nfs_fattr_init(fattr); status = rpc_call(server->client, NFS3PROC_GETATTR, fhandle, fattr, 0); dprintk("NFS reply getattr: %d\n", status); @@ -117,7 +117,7 @@ nfs3_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr, int status; dprintk("NFS call setattr\n"); - fattr->valid = 0; + nfs_fattr_init(fattr); status = rpc_call(NFS_CLIENT(inode), NFS3PROC_SETATTR, &arg, fattr, 0); if (status == 0) nfs_setattr_update_inode(inode, sattr); @@ -143,8 +143,8 @@ nfs3_proc_lookup(struct inode *dir, struct qstr *name, int status; dprintk("NFS call lookup %s\n", name->name); - dir_attr.valid = 0; - fattr->valid = 0; + nfs_fattr_init(&dir_attr); + nfs_fattr_init(fattr); status = rpc_call(NFS_CLIENT(dir), NFS3PROC_LOOKUP, &arg, &res, 0); if (status >= 0 && !(fattr->valid & NFS_ATTR_FATTR)) status = rpc_call(NFS_CLIENT(dir), NFS3PROC_GETATTR, @@ -174,7 +174,6 @@ static int nfs3_proc_access(struct inode *inode, struct nfs_access_entry *entry) int status; dprintk("NFS call access\n"); - fattr.valid = 0; if (mode & MAY_READ) arg.access |= NFS3_ACCESS_READ; @@ -189,6 +188,7 @@ static int nfs3_proc_access(struct inode *inode, struct nfs_access_entry *entry) if (mode & MAY_EXEC) arg.access |= NFS3_ACCESS_EXECUTE; } + nfs_fattr_init(&fattr); status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0); nfs_refresh_inode(inode, &fattr); if (status == 0) { @@ -217,7 +217,7 @@ static int nfs3_proc_readlink(struct inode *inode, struct page *page, int status; dprintk("NFS call readlink\n"); - fattr.valid = 0; + nfs_fattr_init(&fattr); status = rpc_call(NFS_CLIENT(inode), NFS3PROC_READLINK, &args, &fattr, 0); nfs_refresh_inode(inode, &fattr); @@ -240,7 +240,7 @@ static int nfs3_proc_read(struct nfs_read_data *rdata) dprintk("NFS call read %d @ %Ld\n", rdata->args.count, (long long) rdata->args.offset); - fattr->valid = 0; + nfs_fattr_init(fattr); status = rpc_call_sync(NFS_CLIENT(inode), &msg, flags); if (status >= 0) nfs_refresh_inode(inode, fattr); @@ -263,10 +263,10 @@ static int nfs3_proc_write(struct nfs_write_data *wdata) dprintk("NFS call write %d @ %Ld\n", wdata->args.count, (long long) wdata->args.offset); - fattr->valid = 0; + nfs_fattr_init(fattr); status = rpc_call_sync(NFS_CLIENT(inode), &msg, rpcflags); if (status >= 0) - nfs_refresh_inode(inode, fattr); + nfs_post_op_update_inode(inode, fattr); dprintk("NFS reply write: %d\n", status); return status < 0? status : wdata->res.count; } @@ -285,10 +285,10 @@ static int nfs3_proc_commit(struct nfs_write_data *cdata) dprintk("NFS call commit %d @ %Ld\n", cdata->args.count, (long long) cdata->args.offset); - fattr->valid = 0; + nfs_fattr_init(fattr); status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0); if (status >= 0) - nfs_refresh_inode(inode, fattr); + nfs_post_op_update_inode(inode, fattr); dprintk("NFS reply commit: %d\n", status); return status; } @@ -299,7 +299,7 @@ static int nfs3_proc_commit(struct nfs_write_data *cdata) */ static int nfs3_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr, - int flags) + int flags, struct nameidata *nd) { struct nfs_fh fhandle; struct nfs_fattr fattr; @@ -329,10 +329,10 @@ nfs3_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr, sattr->ia_mode &= ~current->fs->umask; again: - dir_attr.valid = 0; - fattr.valid = 0; + nfs_fattr_init(&dir_attr); + nfs_fattr_init(&fattr); status = rpc_call(NFS_CLIENT(dir), NFS3PROC_CREATE, &arg, &res, 0); - nfs_refresh_inode(dir, &dir_attr); + nfs_post_op_update_inode(dir, &dir_attr); /* If the server doesn't support the exclusive creation semantics, * try again with simple 'guarded' mode. */ @@ -401,9 +401,9 @@ nfs3_proc_remove(struct inode *dir, struct qstr *name) int status; dprintk("NFS call remove %s\n", name->name); - dir_attr.valid = 0; + nfs_fattr_init(&dir_attr); status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0); - nfs_refresh_inode(dir, &dir_attr); + nfs_post_op_update_inode(dir, &dir_attr); dprintk("NFS reply remove: %d\n", status); return status; } @@ -422,7 +422,7 @@ nfs3_proc_unlink_setup(struct rpc_message *msg, struct dentry *dir, struct qstr ptr->arg.fh = NFS_FH(dir->d_inode); ptr->arg.name = name->name; ptr->arg.len = name->len; - ptr->res.valid = 0; + nfs_fattr_init(&ptr->res); msg->rpc_proc = &nfs3_procedures[NFS3PROC_REMOVE]; msg->rpc_argp = &ptr->arg; msg->rpc_resp = &ptr->res; @@ -439,7 +439,7 @@ nfs3_proc_unlink_done(struct dentry *dir, struct rpc_task *task) return 1; if (msg->rpc_argp) { dir_attr = (struct nfs_fattr*)msg->rpc_resp; - nfs_refresh_inode(dir->d_inode, dir_attr); + nfs_post_op_update_inode(dir->d_inode, dir_attr); kfree(msg->rpc_argp); } return 0; @@ -465,11 +465,11 @@ nfs3_proc_rename(struct inode *old_dir, struct qstr *old_name, int status; dprintk("NFS call rename %s -> %s\n", old_name->name, new_name->name); - old_dir_attr.valid = 0; - new_dir_attr.valid = 0; + nfs_fattr_init(&old_dir_attr); + nfs_fattr_init(&new_dir_attr); status = rpc_call(NFS_CLIENT(old_dir), NFS3PROC_RENAME, &arg, &res, 0); - nfs_refresh_inode(old_dir, &old_dir_attr); - nfs_refresh_inode(new_dir, &new_dir_attr); + nfs_post_op_update_inode(old_dir, &old_dir_attr); + nfs_post_op_update_inode(new_dir, &new_dir_attr); dprintk("NFS reply rename: %d\n", status); return status; } @@ -491,11 +491,11 @@ nfs3_proc_link(struct inode *inode, struct inode *dir, struct qstr *name) int status; dprintk("NFS call link %s\n", name->name); - dir_attr.valid = 0; - fattr.valid = 0; + nfs_fattr_init(&dir_attr); + nfs_fattr_init(&fattr); status = rpc_call(NFS_CLIENT(inode), NFS3PROC_LINK, &arg, &res, 0); - nfs_refresh_inode(dir, &dir_attr); - nfs_refresh_inode(inode, &fattr); + nfs_post_op_update_inode(dir, &dir_attr); + nfs_post_op_update_inode(inode, &fattr); dprintk("NFS reply link: %d\n", status); return status; } @@ -524,10 +524,10 @@ nfs3_proc_symlink(struct inode *dir, struct qstr *name, struct qstr *path, if (path->len > NFS3_MAXPATHLEN) return -ENAMETOOLONG; dprintk("NFS call symlink %s -> %s\n", name->name, path->name); - dir_attr.valid = 0; - fattr->valid = 0; + nfs_fattr_init(&dir_attr); + nfs_fattr_init(fattr); status = rpc_call(NFS_CLIENT(dir), NFS3PROC_SYMLINK, &arg, &res, 0); - nfs_refresh_inode(dir, &dir_attr); + nfs_post_op_update_inode(dir, &dir_attr); dprintk("NFS reply symlink: %d\n", status); return status; } @@ -552,13 +552,13 @@ nfs3_proc_mkdir(struct inode *dir, struct dentry *dentry, struct iattr *sattr) int status; dprintk("NFS call mkdir %s\n", dentry->d_name.name); - dir_attr.valid = 0; - fattr.valid = 0; sattr->ia_mode &= ~current->fs->umask; + nfs_fattr_init(&dir_attr); + nfs_fattr_init(&fattr); status = rpc_call(NFS_CLIENT(dir), NFS3PROC_MKDIR, &arg, &res, 0); - nfs_refresh_inode(dir, &dir_attr); + nfs_post_op_update_inode(dir, &dir_attr); if (status != 0) goto out; status = nfs_instantiate(dentry, &fhandle, &fattr); @@ -582,9 +582,9 @@ nfs3_proc_rmdir(struct inode *dir, struct qstr *name) int status; dprintk("NFS call rmdir %s\n", name->name); - dir_attr.valid = 0; + nfs_fattr_init(&dir_attr); status = rpc_call(NFS_CLIENT(dir), NFS3PROC_RMDIR, &arg, &dir_attr, 0); - nfs_refresh_inode(dir, &dir_attr); + nfs_post_op_update_inode(dir, &dir_attr); dprintk("NFS reply rmdir: %d\n", status); return status; } @@ -634,7 +634,7 @@ nfs3_proc_readdir(struct dentry *dentry, struct rpc_cred *cred, dprintk("NFS call readdir%s %d\n", plus? "plus" : "", (unsigned int) cookie); - dir_attr.valid = 0; + nfs_fattr_init(&dir_attr); status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0); nfs_refresh_inode(dir, &dir_attr); dprintk("NFS reply readdir: %d\n", status); @@ -676,10 +676,10 @@ nfs3_proc_mknod(struct inode *dir, struct dentry *dentry, struct iattr *sattr, sattr->ia_mode &= ~current->fs->umask; - dir_attr.valid = 0; - fattr.valid = 0; + nfs_fattr_init(&dir_attr); + nfs_fattr_init(&fattr); status = rpc_call(NFS_CLIENT(dir), NFS3PROC_MKNOD, &arg, &res, 0); - nfs_refresh_inode(dir, &dir_attr); + nfs_post_op_update_inode(dir, &dir_attr); if (status != 0) goto out; status = nfs_instantiate(dentry, &fh, &fattr); @@ -698,7 +698,7 @@ nfs3_proc_statfs(struct nfs_server *server, struct nfs_fh *fhandle, int status; dprintk("NFS call fsstat\n"); - stat->fattr->valid = 0; + nfs_fattr_init(stat->fattr); status = rpc_call(server->client, NFS3PROC_FSSTAT, fhandle, stat, 0); dprintk("NFS reply statfs: %d\n", status); return status; @@ -711,7 +711,7 @@ nfs3_proc_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle, int status; dprintk("NFS call fsinfo\n"); - info->fattr->valid = 0; + nfs_fattr_init(info->fattr); status = rpc_call(server->client_sys, NFS3PROC_FSINFO, fhandle, info, 0); dprintk("NFS reply fsinfo: %d\n", status); return status; @@ -724,7 +724,7 @@ nfs3_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle, int status; dprintk("NFS call pathconf\n"); - info->fattr->valid = 0; + nfs_fattr_init(info->fattr); status = rpc_call(server->client, NFS3PROC_PATHCONF, fhandle, info, 0); dprintk("NFS reply pathconf: %d\n", status); return status; @@ -735,7 +735,7 @@ extern u32 *nfs3_decode_dirent(u32 *, struct nfs_entry *, int); static void nfs3_read_done(struct rpc_task *task) { - struct nfs_write_data *data = (struct nfs_write_data *) task->tk_calldata; + struct nfs_read_data *data = (struct nfs_read_data *) task->tk_calldata; if (nfs3_async_handle_jukebox(task)) return; @@ -775,7 +775,7 @@ nfs3_write_done(struct rpc_task *task) return; data = (struct nfs_write_data *)task->tk_calldata; if (task->tk_status >= 0) - nfs_refresh_inode(data->inode, data->res.fattr); + nfs_post_op_update_inode(data->inode, data->res.fattr); nfs_writeback_done(task); } @@ -819,7 +819,7 @@ nfs3_commit_done(struct rpc_task *task) return; data = (struct nfs_write_data *)task->tk_calldata; if (task->tk_status >= 0) - nfs_refresh_inode(data->inode, data->res.fattr); + nfs_post_op_update_inode(data->inode, data->res.fattr); nfs_commit_done(task); } diff --git a/fs/nfs/nfs3xdr.c b/fs/nfs/nfs3xdr.c index db4a904810a4..0498bd36602c 100644 --- a/fs/nfs/nfs3xdr.c +++ b/fs/nfs/nfs3xdr.c @@ -174,7 +174,6 @@ xdr_decode_fattr(u32 *p, struct nfs_fattr *fattr) /* Update the mode bits */ fattr->valid |= (NFS_ATTR_FATTR | NFS_ATTR_FATTR_V3); - fattr->timestamp = jiffies; return p; } diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h index ec1a22d7b876..b7f262dcb6e3 100644 --- a/fs/nfs/nfs4_fs.h +++ b/fs/nfs/nfs4_fs.h @@ -93,25 +93,50 @@ struct nfs4_client { }; /* + * struct rpc_sequence ensures that RPC calls are sent in the exact + * order that they appear on the list. + */ +struct rpc_sequence { + struct rpc_wait_queue wait; /* RPC call delay queue */ + spinlock_t lock; /* Protects the list */ + struct list_head list; /* Defines sequence of RPC calls */ +}; + +#define NFS_SEQID_CONFIRMED 1 +struct nfs_seqid_counter { + struct rpc_sequence *sequence; + int flags; + u32 counter; +}; + +struct nfs_seqid { + struct nfs_seqid_counter *sequence; + struct list_head list; +}; + +static inline void nfs_confirm_seqid(struct nfs_seqid_counter *seqid, int status) +{ + if (seqid_mutating_err(-status)) + seqid->flags |= NFS_SEQID_CONFIRMED; +} + +/* * NFS4 state_owners and lock_owners are simply labels for ordered * sequences of RPC calls. Their sole purpose is to provide once-only * semantics by allowing the server to identify replayed requests. - * - * The ->so_sema is held during all state_owner seqid-mutating operations: - * OPEN, OPEN_DOWNGRADE, and CLOSE. Its purpose is to properly serialize - * so_seqid. */ struct nfs4_state_owner { + spinlock_t so_lock; struct list_head so_list; /* per-clientid list of state_owners */ struct nfs4_client *so_client; u32 so_id; /* 32-bit identifier, unique */ - struct semaphore so_sema; - u32 so_seqid; /* protected by so_sema */ atomic_t so_count; struct rpc_cred *so_cred; /* Associated cred */ struct list_head so_states; struct list_head so_delegations; + struct nfs_seqid_counter so_seqid; + struct rpc_sequence so_sequence; }; /* @@ -132,7 +157,7 @@ struct nfs4_lock_state { fl_owner_t ls_owner; /* POSIX lock owner */ #define NFS_LOCK_INITIALIZED 1 int ls_flags; - u32 ls_seqid; + struct nfs_seqid_counter ls_seqid; u32 ls_id; nfs4_stateid ls_stateid; atomic_t ls_count; @@ -153,7 +178,6 @@ struct nfs4_state { struct inode *inode; /* Pointer to the inode */ unsigned long flags; /* Do we hold any locks? */ - struct semaphore lock_sema; /* Serializes file locking operations */ spinlock_t state_lock; /* Protects the lock_states list */ nfs4_stateid stateid; @@ -190,9 +214,9 @@ extern int nfs4_proc_setclientid(struct nfs4_client *, u32, unsigned short); extern int nfs4_proc_setclientid_confirm(struct nfs4_client *); extern int nfs4_proc_async_renew(struct nfs4_client *); extern int nfs4_proc_renew(struct nfs4_client *); -extern int nfs4_do_close(struct inode *inode, struct nfs4_state *state, mode_t mode); -extern struct inode *nfs4_atomic_open(struct inode *, struct dentry *, struct nameidata *); -extern int nfs4_open_revalidate(struct inode *, struct dentry *, int); +extern int nfs4_do_close(struct inode *inode, struct nfs4_state *state); +extern struct dentry *nfs4_atomic_open(struct inode *, struct dentry *, struct nameidata *); +extern int nfs4_open_revalidate(struct inode *, struct dentry *, int, struct nameidata *); extern struct nfs4_state_recovery_ops nfs4_reboot_recovery_ops; extern struct nfs4_state_recovery_ops nfs4_network_partition_recovery_ops; @@ -223,13 +247,18 @@ extern void nfs4_drop_state_owner(struct nfs4_state_owner *); extern struct nfs4_state * nfs4_get_open_state(struct inode *, struct nfs4_state_owner *); extern void nfs4_put_open_state(struct nfs4_state *); extern void nfs4_close_state(struct nfs4_state *, mode_t); -extern struct nfs4_state *nfs4_find_state(struct inode *, struct rpc_cred *, mode_t mode); -extern void nfs4_increment_seqid(int status, struct nfs4_state_owner *sp); +extern void nfs4_state_set_mode_locked(struct nfs4_state *, mode_t); extern void nfs4_schedule_state_recovery(struct nfs4_client *); +extern void nfs4_put_lock_state(struct nfs4_lock_state *lsp); extern int nfs4_set_lock_state(struct nfs4_state *state, struct file_lock *fl); -extern void nfs4_increment_lock_seqid(int status, struct nfs4_lock_state *ls); extern void nfs4_copy_stateid(nfs4_stateid *, struct nfs4_state *, fl_owner_t); +extern struct nfs_seqid *nfs_alloc_seqid(struct nfs_seqid_counter *counter); +extern int nfs_wait_on_sequence(struct nfs_seqid *seqid, struct rpc_task *task); +extern void nfs_increment_open_seqid(int status, struct nfs_seqid *seqid); +extern void nfs_increment_lock_seqid(int status, struct nfs_seqid *seqid); +extern void nfs_free_seqid(struct nfs_seqid *seqid); + extern const nfs4_stateid zero_stateid; /* nfs4xdr.c */ diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 9701ca8c9428..f988a9417b13 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -47,6 +47,7 @@ #include <linux/nfs_page.h> #include <linux/smp_lock.h> #include <linux/namei.h> +#include <linux/mount.h> #include "nfs4_fs.h" #include "delegation.h" @@ -56,10 +57,11 @@ #define NFS4_POLL_RETRY_MIN (1*HZ) #define NFS4_POLL_RETRY_MAX (15*HZ) +static int _nfs4_proc_open_confirm(struct rpc_clnt *clnt, const struct nfs_fh *fh, struct nfs4_state_owner *sp, nfs4_stateid *stateid, struct nfs_seqid *seqid); static int nfs4_do_fsinfo(struct nfs_server *, struct nfs_fh *, struct nfs_fsinfo *); -static int nfs4_async_handle_error(struct rpc_task *, struct nfs_server *); +static int nfs4_async_handle_error(struct rpc_task *, const struct nfs_server *); static int _nfs4_proc_access(struct inode *inode, struct nfs_access_entry *entry); -static int nfs4_handle_exception(struct nfs_server *server, int errorcode, struct nfs4_exception *exception); +static int nfs4_handle_exception(const struct nfs_server *server, int errorcode, struct nfs4_exception *exception); extern u32 *nfs4_decode_dirent(u32 *p, struct nfs_entry *entry, int plus); extern struct rpc_procinfo nfs4_procedures[]; @@ -185,8 +187,26 @@ static void update_changeattr(struct inode *inode, struct nfs4_change_info *cinf { struct nfs_inode *nfsi = NFS_I(inode); + spin_lock(&inode->i_lock); + nfsi->cache_validity |= NFS_INO_INVALID_ATTR; if (cinfo->before == nfsi->change_attr && cinfo->atomic) nfsi->change_attr = cinfo->after; + spin_unlock(&inode->i_lock); +} + +/* Helper for asynchronous RPC calls */ +static int nfs4_call_async(struct rpc_clnt *clnt, rpc_action tk_begin, + rpc_action tk_exit, void *calldata) +{ + struct rpc_task *task; + + if (!(task = rpc_new_task(clnt, tk_exit, RPC_TASK_ASYNC))) + return -ENOMEM; + + task->tk_calldata = calldata; + task->tk_action = tk_begin; + rpc_execute(task); + return 0; } static void update_open_stateid(struct nfs4_state *state, nfs4_stateid *stateid, int open_flags) @@ -194,22 +214,22 @@ static void update_open_stateid(struct nfs4_state *state, nfs4_stateid *stateid, struct inode *inode = state->inode; open_flags &= (FMODE_READ|FMODE_WRITE); - /* Protect against nfs4_find_state() */ + /* Protect against nfs4_find_state_byowner() */ + spin_lock(&state->owner->so_lock); spin_lock(&inode->i_lock); - state->state |= open_flags; - /* NB! List reordering - see the reclaim code for why. */ - if ((open_flags & FMODE_WRITE) && 0 == state->nwriters++) - list_move(&state->open_states, &state->owner->so_states); + memcpy(&state->stateid, stateid, sizeof(state->stateid)); + if ((open_flags & FMODE_WRITE)) + state->nwriters++; if (open_flags & FMODE_READ) state->nreaders++; - memcpy(&state->stateid, stateid, sizeof(state->stateid)); + nfs4_state_set_mode_locked(state, state->state | open_flags); spin_unlock(&inode->i_lock); + spin_unlock(&state->owner->so_lock); } /* * OPEN_RECLAIM: * reclaim state on the server after a reboot. - * Assumes caller is holding the sp->so_sem */ static int _nfs4_open_reclaim(struct nfs4_state_owner *sp, struct nfs4_state *state) { @@ -218,7 +238,6 @@ static int _nfs4_open_reclaim(struct nfs4_state_owner *sp, struct nfs4_state *st struct nfs_delegation *delegation = NFS_I(inode)->delegation; struct nfs_openargs o_arg = { .fh = NFS_FH(inode), - .seqid = sp->so_seqid, .id = sp->so_id, .open_flags = state->state, .clientid = server->nfs4_state->cl_clientid, @@ -245,8 +264,13 @@ static int _nfs4_open_reclaim(struct nfs4_state_owner *sp, struct nfs4_state *st } o_arg.u.delegation_type = delegation->type; } + o_arg.seqid = nfs_alloc_seqid(&sp->so_seqid); + if (o_arg.seqid == NULL) + return -ENOMEM; status = rpc_call_sync(server->client, &msg, RPC_TASK_NOINTR); - nfs4_increment_seqid(status, sp); + /* Confirm the sequence as being established */ + nfs_confirm_seqid(&sp->so_seqid, status); + nfs_increment_open_seqid(status, o_arg.seqid); if (status == 0) { memcpy(&state->stateid, &o_res.stateid, sizeof(state->stateid)); if (o_res.delegation_type != 0) { @@ -256,6 +280,7 @@ static int _nfs4_open_reclaim(struct nfs4_state_owner *sp, struct nfs4_state *st nfs_async_inode_return_delegation(inode, &o_res.stateid); } } + nfs_free_seqid(o_arg.seqid); clear_bit(NFS_DELEGATED_STATE, &state->flags); /* Ensure we update the inode attributes */ NFS_CACHEINV(inode); @@ -302,23 +327,35 @@ static int _nfs4_open_delegation_recall(struct dentry *dentry, struct nfs4_state }; int status = 0; - down(&sp->so_sema); if (!test_bit(NFS_DELEGATED_STATE, &state->flags)) goto out; if (state->state == 0) goto out; - arg.seqid = sp->so_seqid; + arg.seqid = nfs_alloc_seqid(&sp->so_seqid); + status = -ENOMEM; + if (arg.seqid == NULL) + goto out; arg.open_flags = state->state; memcpy(arg.u.delegation.data, state->stateid.data, sizeof(arg.u.delegation.data)); status = rpc_call_sync(server->client, &msg, RPC_TASK_NOINTR); - nfs4_increment_seqid(status, sp); + nfs_increment_open_seqid(status, arg.seqid); + if (status != 0) + goto out_free; + if(res.rflags & NFS4_OPEN_RESULT_CONFIRM) { + status = _nfs4_proc_open_confirm(server->client, NFS_FH(inode), + sp, &res.stateid, arg.seqid); + if (status != 0) + goto out_free; + } + nfs_confirm_seqid(&sp->so_seqid, 0); if (status >= 0) { memcpy(state->stateid.data, res.stateid.data, sizeof(state->stateid.data)); clear_bit(NFS_DELEGATED_STATE, &state->flags); } +out_free: + nfs_free_seqid(arg.seqid); out: - up(&sp->so_sema); dput(parent); return status; } @@ -345,11 +382,11 @@ int nfs4_open_delegation_recall(struct dentry *dentry, struct nfs4_state *state) return err; } -static inline int _nfs4_proc_open_confirm(struct rpc_clnt *clnt, const struct nfs_fh *fh, struct nfs4_state_owner *sp, nfs4_stateid *stateid) +static int _nfs4_proc_open_confirm(struct rpc_clnt *clnt, const struct nfs_fh *fh, struct nfs4_state_owner *sp, nfs4_stateid *stateid, struct nfs_seqid *seqid) { struct nfs_open_confirmargs arg = { .fh = fh, - .seqid = sp->so_seqid, + .seqid = seqid, .stateid = *stateid, }; struct nfs_open_confirmres res; @@ -362,7 +399,9 @@ static inline int _nfs4_proc_open_confirm(struct rpc_clnt *clnt, const struct nf int status; status = rpc_call_sync(clnt, &msg, RPC_TASK_NOINTR); - nfs4_increment_seqid(status, sp); + /* Confirm the sequence as being established */ + nfs_confirm_seqid(&sp->so_seqid, status); + nfs_increment_open_seqid(status, seqid); if (status >= 0) memcpy(stateid, &res.stateid, sizeof(*stateid)); return status; @@ -380,21 +419,41 @@ static int _nfs4_proc_open(struct inode *dir, struct nfs4_state_owner *sp, stru int status; /* Update sequence id. The caller must serialize! */ - o_arg->seqid = sp->so_seqid; o_arg->id = sp->so_id; o_arg->clientid = sp->so_client->cl_clientid; status = rpc_call_sync(server->client, &msg, RPC_TASK_NOINTR); - nfs4_increment_seqid(status, sp); + if (status == 0) { + /* OPEN on anything except a regular file is disallowed in NFSv4 */ + switch (o_res->f_attr->mode & S_IFMT) { + case S_IFREG: + break; + case S_IFLNK: + status = -ELOOP; + break; + case S_IFDIR: + status = -EISDIR; + break; + default: + status = -ENOTDIR; + } + } + + nfs_increment_open_seqid(status, o_arg->seqid); if (status != 0) goto out; - update_changeattr(dir, &o_res->cinfo); + if (o_arg->open_flags & O_CREAT) { + update_changeattr(dir, &o_res->cinfo); + nfs_post_op_update_inode(dir, o_res->dir_attr); + } else + nfs_refresh_inode(dir, o_res->dir_attr); if(o_res->rflags & NFS4_OPEN_RESULT_CONFIRM) { status = _nfs4_proc_open_confirm(server->client, &o_res->fh, - sp, &o_res->stateid); + sp, &o_res->stateid, o_arg->seqid); if (status != 0) goto out; } + nfs_confirm_seqid(&sp->so_seqid, 0); if (!(o_res->f_attr->valid & NFS_ATTR_FATTR)) status = server->rpc_ops->getattr(server, &o_res->fh, o_res->f_attr); out: @@ -441,9 +500,7 @@ static int _nfs4_open_expired(struct nfs4_state_owner *sp, struct nfs4_state *st struct inode *inode = state->inode; struct nfs_server *server = NFS_SERVER(dir); struct nfs_delegation *delegation = NFS_I(inode)->delegation; - struct nfs_fattr f_attr = { - .valid = 0, - }; + struct nfs_fattr f_attr, dir_attr; struct nfs_openargs o_arg = { .fh = NFS_FH(dir), .open_flags = state->state, @@ -453,6 +510,7 @@ static int _nfs4_open_expired(struct nfs4_state_owner *sp, struct nfs4_state *st }; struct nfs_openres o_res = { .f_attr = &f_attr, + .dir_attr = &dir_attr, .server = server, }; int status = 0; @@ -465,6 +523,12 @@ static int _nfs4_open_expired(struct nfs4_state_owner *sp, struct nfs4_state *st set_bit(NFS_DELEGATED_STATE, &state->flags); goto out; } + o_arg.seqid = nfs_alloc_seqid(&sp->so_seqid); + status = -ENOMEM; + if (o_arg.seqid == NULL) + goto out; + nfs_fattr_init(&f_attr); + nfs_fattr_init(&dir_attr); status = _nfs4_proc_open(dir, sp, &o_arg, &o_res); if (status != 0) goto out_nodeleg; @@ -490,6 +554,7 @@ static int _nfs4_open_expired(struct nfs4_state_owner *sp, struct nfs4_state *st nfs_inode_reclaim_delegation(inode, sp->so_cred, &o_res); } out_nodeleg: + nfs_free_seqid(o_arg.seqid); clear_bit(NFS_DELEGATED_STATE, &state->flags); out: dput(parent); @@ -564,7 +629,6 @@ static int _nfs4_open_delegated(struct inode *inode, int flags, struct rpc_cred dprintk("%s: nfs4_get_state_owner failed!\n", __FUNCTION__); goto out_err; } - down(&sp->so_sema); state = nfs4_get_open_state(inode, sp); if (state == NULL) goto out_err; @@ -589,7 +653,6 @@ static int _nfs4_open_delegated(struct inode *inode, int flags, struct rpc_cred set_bit(NFS_DELEGATED_STATE, &state->flags); update_open_stateid(state, &delegation->stateid, open_flags); out_ok: - up(&sp->so_sema); nfs4_put_state_owner(sp); up_read(&nfsi->rwsem); up_read(&clp->cl_sem); @@ -600,11 +663,12 @@ out_err: if (sp != NULL) { if (state != NULL) nfs4_put_open_state(state); - up(&sp->so_sema); nfs4_put_state_owner(sp); } up_read(&nfsi->rwsem); up_read(&clp->cl_sem); + if (err != -EACCES) + nfs_inode_return_delegation(inode); return err; } @@ -635,9 +699,7 @@ static int _nfs4_do_open(struct inode *dir, struct dentry *dentry, int flags, st struct nfs4_client *clp = server->nfs4_state; struct inode *inode = NULL; int status; - struct nfs_fattr f_attr = { - .valid = 0, - }; + struct nfs_fattr f_attr, dir_attr; struct nfs_openargs o_arg = { .fh = NFS_FH(dir), .open_flags = flags, @@ -648,6 +710,7 @@ static int _nfs4_do_open(struct inode *dir, struct dentry *dentry, int flags, st }; struct nfs_openres o_res = { .f_attr = &f_attr, + .dir_attr = &dir_attr, .server = server, }; @@ -665,8 +728,12 @@ static int _nfs4_do_open(struct inode *dir, struct dentry *dentry, int flags, st } else o_arg.u.attrs = sattr; /* Serialization for the sequence id */ - down(&sp->so_sema); + o_arg.seqid = nfs_alloc_seqid(&sp->so_seqid); + if (o_arg.seqid == NULL) + return -ENOMEM; + nfs_fattr_init(&f_attr); + nfs_fattr_init(&dir_attr); status = _nfs4_proc_open(dir, sp, &o_arg, &o_res); if (status != 0) goto out_err; @@ -681,7 +748,7 @@ static int _nfs4_do_open(struct inode *dir, struct dentry *dentry, int flags, st update_open_stateid(state, &o_res.stateid, flags); if (o_res.delegation_type != 0) nfs_inode_set_delegation(inode, cred, &o_res); - up(&sp->so_sema); + nfs_free_seqid(o_arg.seqid); nfs4_put_state_owner(sp); up_read(&clp->cl_sem); *res = state; @@ -690,7 +757,7 @@ out_err: if (sp != NULL) { if (state != NULL) nfs4_put_open_state(state); - up(&sp->so_sema); + nfs_free_seqid(o_arg.seqid); nfs4_put_state_owner(sp); } /* Note: clp->cl_sem must be released before nfs4_put_open_state()! */ @@ -718,7 +785,7 @@ static struct nfs4_state *nfs4_do_open(struct inode *dir, struct dentry *dentry, * It is actually a sign of a bug on the client or on the server. * * If we receive a BAD_SEQID error in the particular case of - * doing an OPEN, we assume that nfs4_increment_seqid() will + * doing an OPEN, we assume that nfs_increment_open_seqid() will * have unhashed the old state_owner for us, and that we can * therefore safely retry using a new one. We should still warn * the user though... @@ -728,6 +795,16 @@ static struct nfs4_state *nfs4_do_open(struct inode *dir, struct dentry *dentry, exception.retry = 1; continue; } + /* + * BAD_STATEID on OPEN means that the server cancelled our + * state before it received the OPEN_CONFIRM. + * Recover by retrying the request as per the discussion + * on Page 181 of RFC3530. + */ + if (status == -NFS4ERR_BAD_STATEID) { + exception.retry = 1; + continue; + } res = ERR_PTR(nfs4_handle_exception(NFS_SERVER(dir), status, &exception)); } while (exception.retry); @@ -755,7 +832,7 @@ static int _nfs4_do_setattr(struct nfs_server *server, struct nfs_fattr *fattr, }; int status; - fattr->valid = 0; + nfs_fattr_init(fattr); if (state != NULL) { msg.rpc_cred = state->owner->so_cred; @@ -787,19 +864,30 @@ struct nfs4_closedata { struct nfs4_state *state; struct nfs_closeargs arg; struct nfs_closeres res; + struct nfs_fattr fattr; }; +static void nfs4_free_closedata(struct nfs4_closedata *calldata) +{ + struct nfs4_state *state = calldata->state; + struct nfs4_state_owner *sp = state->owner; + + nfs4_put_open_state(calldata->state); + nfs_free_seqid(calldata->arg.seqid); + nfs4_put_state_owner(sp); + kfree(calldata); +} + static void nfs4_close_done(struct rpc_task *task) { struct nfs4_closedata *calldata = (struct nfs4_closedata *)task->tk_calldata; struct nfs4_state *state = calldata->state; - struct nfs4_state_owner *sp = state->owner; struct nfs_server *server = NFS_SERVER(calldata->inode); /* hmm. we are done with the inode, and in the process of freeing * the state_owner. we keep this around to process errors */ - nfs4_increment_seqid(task->tk_status, sp); + nfs_increment_open_seqid(task->tk_status, calldata->arg.seqid); switch (task->tk_status) { case 0: memcpy(&state->stateid, &calldata->res.stateid, @@ -807,7 +895,6 @@ static void nfs4_close_done(struct rpc_task *task) break; case -NFS4ERR_STALE_STATEID: case -NFS4ERR_EXPIRED: - state->state = calldata->arg.open_flags; nfs4_schedule_state_recovery(server->nfs4_state); break; default: @@ -816,25 +903,50 @@ static void nfs4_close_done(struct rpc_task *task) return; } } - state->state = calldata->arg.open_flags; - nfs4_put_open_state(state); - up(&sp->so_sema); - nfs4_put_state_owner(sp); - up_read(&server->nfs4_state->cl_sem); - kfree(calldata); + nfs_refresh_inode(calldata->inode, calldata->res.fattr); + nfs4_free_closedata(calldata); } -static inline int nfs4_close_call(struct rpc_clnt *clnt, struct nfs4_closedata *calldata) +static void nfs4_close_begin(struct rpc_task *task) { + struct nfs4_closedata *calldata = (struct nfs4_closedata *)task->tk_calldata; + struct nfs4_state *state = calldata->state; struct rpc_message msg = { .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_CLOSE], .rpc_argp = &calldata->arg, .rpc_resp = &calldata->res, - .rpc_cred = calldata->state->owner->so_cred, + .rpc_cred = state->owner->so_cred, }; - if (calldata->arg.open_flags != 0) + int mode = 0, old_mode; + int status; + + status = nfs_wait_on_sequence(calldata->arg.seqid, task); + if (status != 0) + return; + /* Recalculate the new open mode in case someone reopened the file + * while we were waiting in line to be scheduled. + */ + spin_lock(&state->owner->so_lock); + spin_lock(&calldata->inode->i_lock); + mode = old_mode = state->state; + if (state->nreaders == 0) + mode &= ~FMODE_READ; + if (state->nwriters == 0) + mode &= ~FMODE_WRITE; + nfs4_state_set_mode_locked(state, mode); + spin_unlock(&calldata->inode->i_lock); + spin_unlock(&state->owner->so_lock); + if (mode == old_mode || test_bit(NFS_DELEGATED_STATE, &state->flags)) { + nfs4_free_closedata(calldata); + task->tk_exit = NULL; + rpc_exit(task, 0); + return; + } + nfs_fattr_init(calldata->res.fattr); + if (mode != 0) msg.rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_OPEN_DOWNGRADE]; - return rpc_call_async(clnt, &msg, 0, nfs4_close_done, calldata); + calldata->arg.open_flags = mode; + rpc_call_setup(task, &msg, 0); } /* @@ -848,42 +960,59 @@ static inline int nfs4_close_call(struct rpc_clnt *clnt, struct nfs4_closedata * * * NOTE: Caller must be holding the sp->so_owner semaphore! */ -int nfs4_do_close(struct inode *inode, struct nfs4_state *state, mode_t mode) +int nfs4_do_close(struct inode *inode, struct nfs4_state *state) { + struct nfs_server *server = NFS_SERVER(inode); struct nfs4_closedata *calldata; - int status; + int status = -ENOMEM; - /* Tell caller we're done */ - if (test_bit(NFS_DELEGATED_STATE, &state->flags)) { - state->state = mode; - return 0; - } - calldata = (struct nfs4_closedata *)kmalloc(sizeof(*calldata), GFP_KERNEL); + calldata = kmalloc(sizeof(*calldata), GFP_KERNEL); if (calldata == NULL) - return -ENOMEM; + goto out; calldata->inode = inode; calldata->state = state; calldata->arg.fh = NFS_FH(inode); + calldata->arg.stateid = &state->stateid; /* Serialization for the sequence id */ - calldata->arg.seqid = state->owner->so_seqid; - calldata->arg.open_flags = mode; - memcpy(&calldata->arg.stateid, &state->stateid, - sizeof(calldata->arg.stateid)); - status = nfs4_close_call(NFS_SERVER(inode)->client, calldata); - /* - * Return -EINPROGRESS on success in order to indicate to the - * caller that an asynchronous RPC call has been launched, and - * that it will release the semaphores on completion. - */ - return (status == 0) ? -EINPROGRESS : status; + calldata->arg.seqid = nfs_alloc_seqid(&state->owner->so_seqid); + if (calldata->arg.seqid == NULL) + goto out_free_calldata; + calldata->arg.bitmask = server->attr_bitmask; + calldata->res.fattr = &calldata->fattr; + calldata->res.server = server; + + status = nfs4_call_async(server->client, nfs4_close_begin, + nfs4_close_done, calldata); + if (status == 0) + goto out; + + nfs_free_seqid(calldata->arg.seqid); +out_free_calldata: + kfree(calldata); +out: + return status; +} + +static void nfs4_intent_set_file(struct nameidata *nd, struct dentry *dentry, struct nfs4_state *state) +{ + struct file *filp; + + filp = lookup_instantiate_filp(nd, dentry, NULL); + if (!IS_ERR(filp)) { + struct nfs_open_context *ctx; + ctx = (struct nfs_open_context *)filp->private_data; + ctx->state = state; + } else + nfs4_close_state(state, nd->intent.open.flags); } -struct inode * +struct dentry * nfs4_atomic_open(struct inode *dir, struct dentry *dentry, struct nameidata *nd) { struct iattr attr; struct rpc_cred *cred; struct nfs4_state *state; + struct dentry *res; if (nd->flags & LOOKUP_CREATE) { attr.ia_mode = nd->intent.open.create_mode; @@ -897,16 +1026,23 @@ nfs4_atomic_open(struct inode *dir, struct dentry *dentry, struct nameidata *nd) cred = rpcauth_lookupcred(NFS_SERVER(dir)->client->cl_auth, 0); if (IS_ERR(cred)) - return (struct inode *)cred; + return (struct dentry *)cred; state = nfs4_do_open(dir, dentry, nd->intent.open.flags, &attr, cred); put_rpccred(cred); - if (IS_ERR(state)) - return (struct inode *)state; - return state->inode; + if (IS_ERR(state)) { + if (PTR_ERR(state) == -ENOENT) + d_add(dentry, NULL); + return (struct dentry *)state; + } + res = d_add_unique(dentry, state->inode); + if (res != NULL) + dentry = res; + nfs4_intent_set_file(nd, dentry, state); + return res; } int -nfs4_open_revalidate(struct inode *dir, struct dentry *dentry, int openflags) +nfs4_open_revalidate(struct inode *dir, struct dentry *dentry, int openflags, struct nameidata *nd) { struct rpc_cred *cred; struct nfs4_state *state; @@ -919,18 +1055,30 @@ nfs4_open_revalidate(struct inode *dir, struct dentry *dentry, int openflags) if (IS_ERR(state)) state = nfs4_do_open(dir, dentry, openflags, NULL, cred); put_rpccred(cred); - if (state == ERR_PTR(-ENOENT) && dentry->d_inode == 0) - return 1; - if (IS_ERR(state)) - return 0; + if (IS_ERR(state)) { + switch (PTR_ERR(state)) { + case -EPERM: + case -EACCES: + case -EDQUOT: + case -ENOSPC: + case -EROFS: + lookup_instantiate_filp(nd, (struct dentry *)state, NULL); + return 1; + case -ENOENT: + if (dentry->d_inode == NULL) + return 1; + } + goto out_drop; + } inode = state->inode; + iput(inode); if (inode == dentry->d_inode) { - iput(inode); + nfs4_intent_set_file(nd, dentry, state); return 1; } - d_drop(dentry); nfs4_close_state(state, openflags); - iput(inode); +out_drop: + d_drop(dentry); return 0; } @@ -974,13 +1122,12 @@ static int nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *fh static int _nfs4_lookup_root(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fsinfo *info) { - struct nfs_fattr * fattr = info->fattr; struct nfs4_lookup_root_arg args = { .bitmask = nfs4_fattr_bitmap, }; struct nfs4_lookup_res res = { .server = server, - .fattr = fattr, + .fattr = info->fattr, .fh = fhandle, }; struct rpc_message msg = { @@ -988,7 +1135,7 @@ static int _nfs4_lookup_root(struct nfs_server *server, struct nfs_fh *fhandle, .rpc_argp = &args, .rpc_resp = &res, }; - fattr->valid = 0; + nfs_fattr_init(info->fattr); return rpc_call_sync(server->client, &msg, 0); } @@ -1051,7 +1198,7 @@ static int nfs4_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle, q.len = p - q.name; do { - fattr->valid = 0; + nfs_fattr_init(fattr); status = nfs4_handle_exception(server, rpc_call_sync(server->client, &msg, 0), &exception); @@ -1088,7 +1235,7 @@ static int _nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, .rpc_resp = &res, }; - fattr->valid = 0; + nfs_fattr_init(fattr); return rpc_call_sync(server->client, &msg, 0); } @@ -1127,30 +1274,27 @@ nfs4_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr, { struct rpc_cred *cred; struct inode *inode = dentry->d_inode; - struct nfs4_state *state; + struct nfs_open_context *ctx; + struct nfs4_state *state = NULL; int status; - fattr->valid = 0; + nfs_fattr_init(fattr); cred = rpcauth_lookupcred(NFS_SERVER(inode)->client->cl_auth, 0); if (IS_ERR(cred)) return PTR_ERR(cred); - /* Search for an existing WRITE delegation first */ - state = nfs4_open_delegated(inode, FMODE_WRITE, cred); - if (!IS_ERR(state)) { - /* NB: nfs4_open_delegated() bumps the inode->i_count */ - iput(inode); - } else { - /* Search for an existing open(O_WRITE) stateid */ - state = nfs4_find_state(inode, cred, FMODE_WRITE); - } + + /* Search for an existing open(O_WRITE) file */ + ctx = nfs_find_open_context(inode, cred, FMODE_WRITE); + if (ctx != NULL) + state = ctx->state; status = nfs4_do_setattr(NFS_SERVER(inode), fattr, NFS_FH(inode), sattr, state); if (status == 0) nfs_setattr_update_inode(inode, sattr); - if (state != NULL) - nfs4_close_state(state, FMODE_WRITE); + if (ctx != NULL) + put_nfs_open_context(ctx); put_rpccred(cred); return status; } @@ -1176,7 +1320,7 @@ static int _nfs4_proc_lookup(struct inode *dir, struct qstr *name, .rpc_resp = &res, }; - fattr->valid = 0; + nfs_fattr_init(fattr); dprintk("NFS call lookup %s\n", name->name); status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0); @@ -1325,7 +1469,7 @@ static int _nfs4_proc_read(struct nfs_read_data *rdata) dprintk("NFS call read %d @ %Ld\n", rdata->args.count, (long long) rdata->args.offset); - fattr->valid = 0; + nfs_fattr_init(fattr); status = rpc_call_sync(server->client, &msg, flags); if (!status) renew_lease(server, timestamp); @@ -1362,10 +1506,15 @@ static int _nfs4_proc_write(struct nfs_write_data *wdata) dprintk("NFS call write %d @ %Ld\n", wdata->args.count, (long long) wdata->args.offset); - fattr->valid = 0; + wdata->args.bitmask = server->attr_bitmask; + wdata->res.server = server; + nfs_fattr_init(fattr); status = rpc_call_sync(server->client, &msg, rpcflags); dprintk("NFS reply write: %d\n", status); - return status; + if (status < 0) + return status; + nfs_post_op_update_inode(inode, fattr); + return wdata->res.count; } static int nfs4_proc_write(struct nfs_write_data *wdata) @@ -1396,9 +1545,13 @@ static int _nfs4_proc_commit(struct nfs_write_data *cdata) dprintk("NFS call commit %d @ %Ld\n", cdata->args.count, (long long) cdata->args.offset); - fattr->valid = 0; + cdata->args.bitmask = server->attr_bitmask; + cdata->res.server = server; + nfs_fattr_init(fattr); status = rpc_call_sync(server->client, &msg, 0); dprintk("NFS reply commit: %d\n", status); + if (status >= 0) + nfs_post_op_update_inode(inode, fattr); return status; } @@ -1431,7 +1584,7 @@ static int nfs4_proc_commit(struct nfs_write_data *cdata) static int nfs4_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr, - int flags) + int flags, struct nameidata *nd) { struct nfs4_state *state; struct rpc_cred *cred; @@ -1453,24 +1606,30 @@ nfs4_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr, struct nfs_fattr fattr; status = nfs4_do_setattr(NFS_SERVER(dir), &fattr, NFS_FH(state->inode), sattr, state); - if (status == 0) { + if (status == 0) nfs_setattr_update_inode(state->inode, sattr); - goto out; - } - } else if (flags != 0) - goto out; - nfs4_close_state(state, flags); + } + if (status == 0 && nd != NULL && (nd->flags & LOOKUP_OPEN)) + nfs4_intent_set_file(nd, dentry, state); + else + nfs4_close_state(state, flags); out: return status; } static int _nfs4_proc_remove(struct inode *dir, struct qstr *name) { + struct nfs_server *server = NFS_SERVER(dir); struct nfs4_remove_arg args = { .fh = NFS_FH(dir), .name = name, + .bitmask = server->attr_bitmask, + }; + struct nfs_fattr dir_attr; + struct nfs4_remove_res res = { + .server = server, + .dir_attr = &dir_attr, }; - struct nfs4_change_info res; struct rpc_message msg = { .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_REMOVE], .rpc_argp = &args, @@ -1478,9 +1637,12 @@ static int _nfs4_proc_remove(struct inode *dir, struct qstr *name) }; int status; - status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0); - if (status == 0) - update_changeattr(dir, &res); + nfs_fattr_init(res.dir_attr); + status = rpc_call_sync(server->client, &msg, 0); + if (status == 0) { + update_changeattr(dir, &res.cinfo); + nfs_post_op_update_inode(dir, res.dir_attr); + } return status; } @@ -1498,12 +1660,14 @@ static int nfs4_proc_remove(struct inode *dir, struct qstr *name) struct unlink_desc { struct nfs4_remove_arg args; - struct nfs4_change_info res; + struct nfs4_remove_res res; + struct nfs_fattr dir_attr; }; static int nfs4_proc_unlink_setup(struct rpc_message *msg, struct dentry *dir, struct qstr *name) { + struct nfs_server *server = NFS_SERVER(dir->d_inode); struct unlink_desc *up; up = (struct unlink_desc *) kmalloc(sizeof(*up), GFP_KERNEL); @@ -1512,6 +1676,9 @@ static int nfs4_proc_unlink_setup(struct rpc_message *msg, struct dentry *dir, up->args.fh = NFS_FH(dir->d_inode); up->args.name = name; + up->args.bitmask = server->attr_bitmask; + up->res.server = server; + up->res.dir_attr = &up->dir_attr; msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_REMOVE]; msg->rpc_argp = &up->args; @@ -1526,7 +1693,8 @@ static int nfs4_proc_unlink_done(struct dentry *dir, struct rpc_task *task) if (msg->rpc_resp != NULL) { up = container_of(msg->rpc_resp, struct unlink_desc, res); - update_changeattr(dir->d_inode, &up->res); + update_changeattr(dir->d_inode, &up->res.cinfo); + nfs_post_op_update_inode(dir->d_inode, up->res.dir_attr); kfree(up); msg->rpc_resp = NULL; msg->rpc_argp = NULL; @@ -1537,13 +1705,20 @@ static int nfs4_proc_unlink_done(struct dentry *dir, struct rpc_task *task) static int _nfs4_proc_rename(struct inode *old_dir, struct qstr *old_name, struct inode *new_dir, struct qstr *new_name) { + struct nfs_server *server = NFS_SERVER(old_dir); struct nfs4_rename_arg arg = { .old_dir = NFS_FH(old_dir), .new_dir = NFS_FH(new_dir), .old_name = old_name, .new_name = new_name, + .bitmask = server->attr_bitmask, + }; + struct nfs_fattr old_fattr, new_fattr; + struct nfs4_rename_res res = { + .server = server, + .old_fattr = &old_fattr, + .new_fattr = &new_fattr, }; - struct nfs4_rename_res res = { }; struct rpc_message msg = { .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_RENAME], .rpc_argp = &arg, @@ -1551,11 +1726,15 @@ static int _nfs4_proc_rename(struct inode *old_dir, struct qstr *old_name, }; int status; - status = rpc_call_sync(NFS_CLIENT(old_dir), &msg, 0); + nfs_fattr_init(res.old_fattr); + nfs_fattr_init(res.new_fattr); + status = rpc_call_sync(server->client, &msg, 0); if (!status) { update_changeattr(old_dir, &res.old_cinfo); + nfs_post_op_update_inode(old_dir, res.old_fattr); update_changeattr(new_dir, &res.new_cinfo); + nfs_post_op_update_inode(new_dir, res.new_fattr); } return status; } @@ -1576,22 +1755,34 @@ static int nfs4_proc_rename(struct inode *old_dir, struct qstr *old_name, static int _nfs4_proc_link(struct inode *inode, struct inode *dir, struct qstr *name) { + struct nfs_server *server = NFS_SERVER(inode); struct nfs4_link_arg arg = { .fh = NFS_FH(inode), .dir_fh = NFS_FH(dir), .name = name, + .bitmask = server->attr_bitmask, + }; + struct nfs_fattr fattr, dir_attr; + struct nfs4_link_res res = { + .server = server, + .fattr = &fattr, + .dir_attr = &dir_attr, }; - struct nfs4_change_info cinfo = { }; struct rpc_message msg = { .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LINK], .rpc_argp = &arg, - .rpc_resp = &cinfo, + .rpc_resp = &res, }; int status; - status = rpc_call_sync(NFS_CLIENT(inode), &msg, 0); - if (!status) - update_changeattr(dir, &cinfo); + nfs_fattr_init(res.fattr); + nfs_fattr_init(res.dir_attr); + status = rpc_call_sync(server->client, &msg, 0); + if (!status) { + update_changeattr(dir, &res.cinfo); + nfs_post_op_update_inode(dir, res.dir_attr); + nfs_refresh_inode(inode, res.fattr); + } return status; } @@ -1613,6 +1804,7 @@ static int _nfs4_proc_symlink(struct inode *dir, struct qstr *name, struct nfs_fattr *fattr) { struct nfs_server *server = NFS_SERVER(dir); + struct nfs_fattr dir_fattr; struct nfs4_create_arg arg = { .dir_fh = NFS_FH(dir), .server = server, @@ -1625,6 +1817,7 @@ static int _nfs4_proc_symlink(struct inode *dir, struct qstr *name, .server = server, .fh = fhandle, .fattr = fattr, + .dir_fattr = &dir_fattr, }; struct rpc_message msg = { .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_SYMLINK], @@ -1636,11 +1829,13 @@ static int _nfs4_proc_symlink(struct inode *dir, struct qstr *name, if (path->len > NFS4_MAXPATHLEN) return -ENAMETOOLONG; arg.u.symlink = path; - fattr->valid = 0; + nfs_fattr_init(fattr); + nfs_fattr_init(&dir_fattr); status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0); if (!status) update_changeattr(dir, &res.dir_cinfo); + nfs_post_op_update_inode(dir, res.dir_fattr); return status; } @@ -1664,7 +1859,7 @@ static int _nfs4_proc_mkdir(struct inode *dir, struct dentry *dentry, { struct nfs_server *server = NFS_SERVER(dir); struct nfs_fh fhandle; - struct nfs_fattr fattr; + struct nfs_fattr fattr, dir_fattr; struct nfs4_create_arg arg = { .dir_fh = NFS_FH(dir), .server = server, @@ -1677,6 +1872,7 @@ static int _nfs4_proc_mkdir(struct inode *dir, struct dentry *dentry, .server = server, .fh = &fhandle, .fattr = &fattr, + .dir_fattr = &dir_fattr, }; struct rpc_message msg = { .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_CREATE], @@ -1685,11 +1881,13 @@ static int _nfs4_proc_mkdir(struct inode *dir, struct dentry *dentry, }; int status; - fattr.valid = 0; + nfs_fattr_init(&fattr); + nfs_fattr_init(&dir_fattr); status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0); if (!status) { update_changeattr(dir, &res.dir_cinfo); + nfs_post_op_update_inode(dir, res.dir_fattr); status = nfs_instantiate(dentry, &fhandle, &fattr); } return status; @@ -1762,7 +1960,7 @@ static int _nfs4_proc_mknod(struct inode *dir, struct dentry *dentry, { struct nfs_server *server = NFS_SERVER(dir); struct nfs_fh fh; - struct nfs_fattr fattr; + struct nfs_fattr fattr, dir_fattr; struct nfs4_create_arg arg = { .dir_fh = NFS_FH(dir), .server = server, @@ -1774,6 +1972,7 @@ static int _nfs4_proc_mknod(struct inode *dir, struct dentry *dentry, .server = server, .fh = &fh, .fattr = &fattr, + .dir_fattr = &dir_fattr, }; struct rpc_message msg = { .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_CREATE], @@ -1783,7 +1982,8 @@ static int _nfs4_proc_mknod(struct inode *dir, struct dentry *dentry, int status; int mode = sattr->ia_mode; - fattr.valid = 0; + nfs_fattr_init(&fattr); + nfs_fattr_init(&dir_fattr); BUG_ON(!(sattr->ia_valid & ATTR_MODE)); BUG_ON(!S_ISFIFO(mode) && !S_ISBLK(mode) && !S_ISCHR(mode) && !S_ISSOCK(mode)); @@ -1805,6 +2005,7 @@ static int _nfs4_proc_mknod(struct inode *dir, struct dentry *dentry, status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0); if (status == 0) { update_changeattr(dir, &res.dir_cinfo); + nfs_post_op_update_inode(dir, res.dir_fattr); status = nfs_instantiate(dentry, &fh, &fattr); } return status; @@ -1836,7 +2037,7 @@ static int _nfs4_proc_statfs(struct nfs_server *server, struct nfs_fh *fhandle, .rpc_resp = fsstat, }; - fsstat->fattr->valid = 0; + nfs_fattr_init(fsstat->fattr); return rpc_call_sync(server->client, &msg, 0); } @@ -1883,7 +2084,7 @@ static int nfs4_do_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle, str static int nfs4_proc_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fsinfo *fsinfo) { - fsinfo->fattr->valid = 0; + nfs_fattr_init(fsinfo->fattr); return nfs4_do_fsinfo(server, fhandle, fsinfo); } @@ -1906,7 +2107,7 @@ static int _nfs4_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle return 0; } - pathconf->fattr->valid = 0; + nfs_fattr_init(pathconf->fattr); return rpc_call_sync(server->client, &msg, 0); } @@ -1973,8 +2174,10 @@ nfs4_write_done(struct rpc_task *task) rpc_restart_call(task); return; } - if (task->tk_status >= 0) + if (task->tk_status >= 0) { renew_lease(NFS_SERVER(inode), data->timestamp); + nfs_post_op_update_inode(inode, data->res.fattr); + } /* Call back common NFS writeback processing */ nfs_writeback_done(task); } @@ -1990,6 +2193,7 @@ nfs4_proc_write_setup(struct nfs_write_data *data, int how) .rpc_cred = data->cred, }; struct inode *inode = data->inode; + struct nfs_server *server = NFS_SERVER(inode); int stable; int flags; @@ -2001,6 +2205,8 @@ nfs4_proc_write_setup(struct nfs_write_data *data, int how) } else stable = NFS_UNSTABLE; data->args.stable = stable; + data->args.bitmask = server->attr_bitmask; + data->res.server = server; data->timestamp = jiffies; @@ -2022,6 +2228,8 @@ nfs4_commit_done(struct rpc_task *task) rpc_restart_call(task); return; } + if (task->tk_status >= 0) + nfs_post_op_update_inode(inode, data->res.fattr); /* Call back common NFS writeback processing */ nfs_commit_done(task); } @@ -2037,8 +2245,12 @@ nfs4_proc_commit_setup(struct nfs_write_data *data, int how) .rpc_cred = data->cred, }; struct inode *inode = data->inode; + struct nfs_server *server = NFS_SERVER(inode); int flags; + data->args.bitmask = server->attr_bitmask; + data->res.server = server; + /* Set the initial flags for the task. */ flags = (how & FLUSH_SYNC) ? 0 : RPC_TASK_ASYNC; @@ -2106,65 +2318,6 @@ nfs4_proc_renew(struct nfs4_client *clp) return 0; } -/* - * We will need to arrange for the VFS layer to provide an atomic open. - * Until then, this open method is prone to inefficiency and race conditions - * due to the lookup, potential create, and open VFS calls from sys_open() - * placed on the wire. - */ -static int -nfs4_proc_file_open(struct inode *inode, struct file *filp) -{ - struct dentry *dentry = filp->f_dentry; - struct nfs_open_context *ctx; - struct nfs4_state *state = NULL; - struct rpc_cred *cred; - int status = -ENOMEM; - - dprintk("nfs4_proc_file_open: starting on (%.*s/%.*s)\n", - (int)dentry->d_parent->d_name.len, - dentry->d_parent->d_name.name, - (int)dentry->d_name.len, dentry->d_name.name); - - - /* Find our open stateid */ - cred = rpcauth_lookupcred(NFS_SERVER(inode)->client->cl_auth, 0); - if (IS_ERR(cred)) - return PTR_ERR(cred); - ctx = alloc_nfs_open_context(dentry, cred); - put_rpccred(cred); - if (unlikely(ctx == NULL)) - return -ENOMEM; - status = -EIO; /* ERACE actually */ - state = nfs4_find_state(inode, cred, filp->f_mode); - if (unlikely(state == NULL)) - goto no_state; - ctx->state = state; - nfs4_close_state(state, filp->f_mode); - ctx->mode = filp->f_mode; - nfs_file_set_open_context(filp, ctx); - put_nfs_open_context(ctx); - if (filp->f_mode & FMODE_WRITE) - nfs_begin_data_update(inode); - return 0; -no_state: - printk(KERN_WARNING "NFS: v4 raced in function %s\n", __FUNCTION__); - put_nfs_open_context(ctx); - return status; -} - -/* - * Release our state - */ -static int -nfs4_proc_file_release(struct inode *inode, struct file *filp) -{ - if (filp->f_mode & FMODE_WRITE) - nfs_end_data_update(inode); - nfs_file_clear_open_context(filp); - return 0; -} - static inline int nfs4_server_supports_acls(struct nfs_server *server) { return (server->caps & NFS_CAP_ACLS) @@ -2285,7 +2438,7 @@ static inline ssize_t nfs4_get_acl_uncached(struct inode *inode, void *buf, size return -ENOMEM; args.acl_pages[0] = localpage; args.acl_pgbase = 0; - args.acl_len = PAGE_SIZE; + resp_len = args.acl_len = PAGE_SIZE; } else { resp_buf = buf; buf_to_pages(buf, buflen, args.acl_pages, &args.acl_pgbase); @@ -2345,6 +2498,7 @@ static int nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t buflen if (!nfs4_server_supports_acls(server)) return -EOPNOTSUPP; + nfs_inode_return_delegation(inode); buf_to_pages(buf, buflen, arg.acl_pages, &arg.acl_pgbase); ret = rpc_call_sync(NFS_SERVER(inode)->client, &msg, 0); if (ret == 0) @@ -2353,7 +2507,7 @@ static int nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t buflen } static int -nfs4_async_handle_error(struct rpc_task *task, struct nfs_server *server) +nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server) { struct nfs4_client *clp = server->nfs4_state; @@ -2431,7 +2585,7 @@ static int nfs4_delay(struct rpc_clnt *clnt, long *timeout) /* This is the error handling routine for processes that are allowed * to sleep. */ -int nfs4_handle_exception(struct nfs_server *server, int errorcode, struct nfs4_exception *exception) +int nfs4_handle_exception(const struct nfs_server *server, int errorcode, struct nfs4_exception *exception) { struct nfs4_client *clp = server->nfs4_state; int ret = errorcode; @@ -2450,12 +2604,10 @@ int nfs4_handle_exception(struct nfs_server *server, int errorcode, struct nfs4_ case -NFS4ERR_GRACE: case -NFS4ERR_DELAY: ret = nfs4_delay(server->client, &exception->timeout); - if (ret == 0) - exception->retry = 1; - break; + if (ret != 0) + break; case -NFS4ERR_OLD_STATEID: - if (ret == 0) - exception->retry = 1; + exception->retry = 1; } /* We failed to handle the error */ return nfs4_map_errors(ret); @@ -2632,7 +2784,6 @@ static int _nfs4_proc_getlk(struct nfs4_state *state, int cmd, struct file_lock down_read(&clp->cl_sem); nlo.clientid = clp->cl_clientid; - down(&state->lock_sema); status = nfs4_set_lock_state(state, request); if (status != 0) goto out; @@ -2659,7 +2810,6 @@ static int _nfs4_proc_getlk(struct nfs4_state *state, int cmd, struct file_lock status = 0; } out: - up(&state->lock_sema); up_read(&clp->cl_sem); return status; } @@ -2696,79 +2846,153 @@ static int do_vfs_lock(struct file *file, struct file_lock *fl) return res; } -static int _nfs4_proc_unlck(struct nfs4_state *state, int cmd, struct file_lock *request) +struct nfs4_unlockdata { + struct nfs_lockargs arg; + struct nfs_locku_opargs luargs; + struct nfs_lockres res; + struct nfs4_lock_state *lsp; + struct nfs_open_context *ctx; + atomic_t refcount; + struct completion completion; +}; + +static void nfs4_locku_release_calldata(struct nfs4_unlockdata *calldata) { - struct inode *inode = state->inode; - struct nfs_server *server = NFS_SERVER(inode); - struct nfs4_client *clp = server->nfs4_state; - struct nfs_lockargs arg = { - .fh = NFS_FH(inode), - .type = nfs4_lck_type(cmd, request), - .offset = request->fl_start, - .length = nfs4_lck_length(request), - }; - struct nfs_lockres res = { - .server = server, - }; + if (atomic_dec_and_test(&calldata->refcount)) { + nfs_free_seqid(calldata->luargs.seqid); + nfs4_put_lock_state(calldata->lsp); + put_nfs_open_context(calldata->ctx); + kfree(calldata); + } +} + +static void nfs4_locku_complete(struct nfs4_unlockdata *calldata) +{ + complete(&calldata->completion); + nfs4_locku_release_calldata(calldata); +} + +static void nfs4_locku_done(struct rpc_task *task) +{ + struct nfs4_unlockdata *calldata = (struct nfs4_unlockdata *)task->tk_calldata; + + nfs_increment_lock_seqid(task->tk_status, calldata->luargs.seqid); + switch (task->tk_status) { + case 0: + memcpy(calldata->lsp->ls_stateid.data, + calldata->res.u.stateid.data, + sizeof(calldata->lsp->ls_stateid.data)); + break; + case -NFS4ERR_STALE_STATEID: + case -NFS4ERR_EXPIRED: + nfs4_schedule_state_recovery(calldata->res.server->nfs4_state); + break; + default: + if (nfs4_async_handle_error(task, calldata->res.server) == -EAGAIN) { + rpc_restart_call(task); + return; + } + } + nfs4_locku_complete(calldata); +} + +static void nfs4_locku_begin(struct rpc_task *task) +{ + struct nfs4_unlockdata *calldata = (struct nfs4_unlockdata *)task->tk_calldata; struct rpc_message msg = { .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LOCKU], - .rpc_argp = &arg, - .rpc_resp = &res, - .rpc_cred = state->owner->so_cred, + .rpc_argp = &calldata->arg, + .rpc_resp = &calldata->res, + .rpc_cred = calldata->lsp->ls_state->owner->so_cred, }; + int status; + + status = nfs_wait_on_sequence(calldata->luargs.seqid, task); + if (status != 0) + return; + if ((calldata->lsp->ls_flags & NFS_LOCK_INITIALIZED) == 0) { + nfs4_locku_complete(calldata); + task->tk_exit = NULL; + rpc_exit(task, 0); + return; + } + rpc_call_setup(task, &msg, 0); +} + +static int nfs4_proc_unlck(struct nfs4_state *state, int cmd, struct file_lock *request) +{ + struct nfs4_unlockdata *calldata; + struct inode *inode = state->inode; + struct nfs_server *server = NFS_SERVER(inode); struct nfs4_lock_state *lsp; - struct nfs_locku_opargs luargs; int status; - - down_read(&clp->cl_sem); - down(&state->lock_sema); + + /* Is this a delegated lock? */ + if (test_bit(NFS_DELEGATED_STATE, &state->flags)) + return do_vfs_lock(request->fl_file, request); + status = nfs4_set_lock_state(state, request); if (status != 0) - goto out; + return status; lsp = request->fl_u.nfs4_fl.owner; /* We might have lost the locks! */ if ((lsp->ls_flags & NFS_LOCK_INITIALIZED) == 0) - goto out; - luargs.seqid = lsp->ls_seqid; - memcpy(&luargs.stateid, &lsp->ls_stateid, sizeof(luargs.stateid)); - arg.u.locku = &luargs; - status = rpc_call_sync(server->client, &msg, RPC_TASK_NOINTR); - nfs4_increment_lock_seqid(status, lsp); - - if (status == 0) - memcpy(&lsp->ls_stateid, &res.u.stateid, - sizeof(lsp->ls_stateid)); -out: - up(&state->lock_sema); + return 0; + calldata = kmalloc(sizeof(*calldata), GFP_KERNEL); + if (calldata == NULL) + return -ENOMEM; + calldata->luargs.seqid = nfs_alloc_seqid(&lsp->ls_seqid); + if (calldata->luargs.seqid == NULL) { + kfree(calldata); + return -ENOMEM; + } + calldata->luargs.stateid = &lsp->ls_stateid; + calldata->arg.fh = NFS_FH(inode); + calldata->arg.type = nfs4_lck_type(cmd, request); + calldata->arg.offset = request->fl_start; + calldata->arg.length = nfs4_lck_length(request); + calldata->arg.u.locku = &calldata->luargs; + calldata->res.server = server; + calldata->lsp = lsp; + atomic_inc(&lsp->ls_count); + + /* Ensure we don't close file until we're done freeing locks! */ + calldata->ctx = get_nfs_open_context((struct nfs_open_context*)request->fl_file->private_data); + + atomic_set(&calldata->refcount, 2); + init_completion(&calldata->completion); + + status = nfs4_call_async(NFS_SERVER(inode)->client, nfs4_locku_begin, + nfs4_locku_done, calldata); if (status == 0) - do_vfs_lock(request->fl_file, request); - up_read(&clp->cl_sem); + wait_for_completion_interruptible(&calldata->completion); + do_vfs_lock(request->fl_file, request); + nfs4_locku_release_calldata(calldata); return status; } -static int nfs4_proc_unlck(struct nfs4_state *state, int cmd, struct file_lock *request) -{ - struct nfs4_exception exception = { }; - int err; - - do { - err = nfs4_handle_exception(NFS_SERVER(state->inode), - _nfs4_proc_unlck(state, cmd, request), - &exception); - } while (exception.retry); - return err; -} - static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *request, int reclaim) { struct inode *inode = state->inode; struct nfs_server *server = NFS_SERVER(inode); struct nfs4_lock_state *lsp = request->fl_u.nfs4_fl.owner; + struct nfs_lock_opargs largs = { + .lock_stateid = &lsp->ls_stateid, + .open_stateid = &state->stateid, + .lock_owner = { + .clientid = server->nfs4_state->cl_clientid, + .id = lsp->ls_id, + }, + .reclaim = reclaim, + }; struct nfs_lockargs arg = { .fh = NFS_FH(inode), .type = nfs4_lck_type(cmd, request), .offset = request->fl_start, .length = nfs4_lck_length(request), + .u = { + .lock = &largs, + }, }; struct nfs_lockres res = { .server = server, @@ -2779,53 +3003,39 @@ static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *r .rpc_resp = &res, .rpc_cred = state->owner->so_cred, }; - struct nfs_lock_opargs largs = { - .reclaim = reclaim, - .new_lock_owner = 0, - }; - int status; + int status = -ENOMEM; - if (!(lsp->ls_flags & NFS_LOCK_INITIALIZED)) { + largs.lock_seqid = nfs_alloc_seqid(&lsp->ls_seqid); + if (largs.lock_seqid == NULL) + return -ENOMEM; + if (!(lsp->ls_seqid.flags & NFS_SEQID_CONFIRMED)) { struct nfs4_state_owner *owner = state->owner; - struct nfs_open_to_lock otl = { - .lock_owner = { - .clientid = server->nfs4_state->cl_clientid, - }, - }; - - otl.lock_seqid = lsp->ls_seqid; - otl.lock_owner.id = lsp->ls_id; - memcpy(&otl.open_stateid, &state->stateid, sizeof(otl.open_stateid)); - largs.u.open_lock = &otl; + + largs.open_seqid = nfs_alloc_seqid(&owner->so_seqid); + if (largs.open_seqid == NULL) + goto out; largs.new_lock_owner = 1; - arg.u.lock = &largs; - down(&owner->so_sema); - otl.open_seqid = owner->so_seqid; status = rpc_call_sync(server->client, &msg, RPC_TASK_NOINTR); - /* increment open_owner seqid on success, and - * seqid mutating errors */ - nfs4_increment_seqid(status, owner); - up(&owner->so_sema); - if (status == 0) { - lsp->ls_flags |= NFS_LOCK_INITIALIZED; - lsp->ls_seqid++; + /* increment open seqid on success, and seqid mutating errors */ + if (largs.new_lock_owner != 0) { + nfs_increment_open_seqid(status, largs.open_seqid); + if (status == 0) + nfs_confirm_seqid(&lsp->ls_seqid, 0); } - } else { - struct nfs_exist_lock el = { - .seqid = lsp->ls_seqid, - }; - memcpy(&el.stateid, &lsp->ls_stateid, sizeof(el.stateid)); - largs.u.exist_lock = ⪙ - arg.u.lock = &largs; + nfs_free_seqid(largs.open_seqid); + } else status = rpc_call_sync(server->client, &msg, RPC_TASK_NOINTR); - /* increment seqid on success, and * seqid mutating errors*/ - nfs4_increment_lock_seqid(status, lsp); - } + /* increment lock seqid on success, and seqid mutating errors*/ + nfs_increment_lock_seqid(status, largs.lock_seqid); /* save the returned stateid. */ - if (status == 0) - memcpy(&lsp->ls_stateid, &res.u.stateid, sizeof(nfs4_stateid)); - else if (status == -NFS4ERR_DENIED) + if (status == 0) { + memcpy(lsp->ls_stateid.data, res.u.stateid.data, + sizeof(lsp->ls_stateid.data)); + lsp->ls_flags |= NFS_LOCK_INITIALIZED; + } else if (status == -NFS4ERR_DENIED) status = -EAGAIN; +out: + nfs_free_seqid(largs.lock_seqid); return status; } @@ -2835,6 +3045,9 @@ static int nfs4_lock_reclaim(struct nfs4_state *state, struct file_lock *request struct nfs4_exception exception = { }; int err; + /* Cache the lock if possible... */ + if (test_bit(NFS_DELEGATED_STATE, &state->flags)) + return 0; do { err = _nfs4_do_setlk(state, F_SETLK, request, 1); if (err != -NFS4ERR_DELAY) @@ -2850,6 +3063,9 @@ static int nfs4_lock_expired(struct nfs4_state *state, struct file_lock *request struct nfs4_exception exception = { }; int err; + err = nfs4_set_lock_state(state, request); + if (err != 0) + return err; do { err = _nfs4_do_setlk(state, F_SETLK, request, 0); if (err != -NFS4ERR_DELAY) @@ -2864,18 +3080,26 @@ static int _nfs4_proc_setlk(struct nfs4_state *state, int cmd, struct file_lock struct nfs4_client *clp = state->owner->so_client; int status; + /* Is this a delegated open? */ + if (NFS_I(state->inode)->delegation_state != 0) { + /* Yes: cache locks! */ + status = do_vfs_lock(request->fl_file, request); + /* ...but avoid races with delegation recall... */ + if (status < 0 || test_bit(NFS_DELEGATED_STATE, &state->flags)) + return status; + } down_read(&clp->cl_sem); - down(&state->lock_sema); status = nfs4_set_lock_state(state, request); - if (status == 0) - status = _nfs4_do_setlk(state, cmd, request, 0); - up(&state->lock_sema); - if (status == 0) { - /* Note: we always want to sleep here! */ - request->fl_flags |= FL_SLEEP; - if (do_vfs_lock(request->fl_file, request) < 0) - printk(KERN_WARNING "%s: VFS is out of sync with lock manager!\n", __FUNCTION__); - } + if (status != 0) + goto out; + status = _nfs4_do_setlk(state, cmd, request, 0); + if (status != 0) + goto out; + /* Note: we always want to sleep here! */ + request->fl_flags |= FL_SLEEP; + if (do_vfs_lock(request->fl_file, request) < 0) + printk(KERN_WARNING "%s: VFS is out of sync with lock manager!\n", __FUNCTION__); +out: up_read(&clp->cl_sem); return status; } @@ -2929,6 +3153,24 @@ nfs4_proc_lock(struct file *filp, int cmd, struct file_lock *request) return status; } +int nfs4_lock_delegation_recall(struct nfs4_state *state, struct file_lock *fl) +{ + struct nfs_server *server = NFS_SERVER(state->inode); + struct nfs4_exception exception = { }; + int err; + + err = nfs4_set_lock_state(state, fl); + if (err != 0) + goto out; + do { + err = _nfs4_do_setlk(state, F_SETLK, fl, 0); + if (err != -NFS4ERR_DELAY) + break; + err = nfs4_handle_exception(server, err, &exception); + } while (exception.retry); +out: + return err; +} #define XATTR_NAME_NFSV4_ACL "system.nfs4_acl" @@ -3024,8 +3266,8 @@ struct nfs_rpc_ops nfs_v4_clientops = { .read_setup = nfs4_proc_read_setup, .write_setup = nfs4_proc_write_setup, .commit_setup = nfs4_proc_commit_setup, - .file_open = nfs4_proc_file_open, - .file_release = nfs4_proc_file_release, + .file_open = nfs_open, + .file_release = nfs_release, .lock = nfs4_proc_lock, .clear_acl_cache = nfs4_zap_acl_attr, }; diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index afe587d82f1e..5ef4c57618fe 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -69,10 +69,8 @@ init_nfsv4_state(struct nfs_server *server) void destroy_nfsv4_state(struct nfs_server *server) { - if (server->mnt_path) { - kfree(server->mnt_path); - server->mnt_path = NULL; - } + kfree(server->mnt_path); + server->mnt_path = NULL; if (server->nfs4_state) { nfs4_put_client(server->nfs4_state); server->nfs4_state = NULL; @@ -264,13 +262,16 @@ nfs4_alloc_state_owner(void) { struct nfs4_state_owner *sp; - sp = kmalloc(sizeof(*sp),GFP_KERNEL); + sp = kzalloc(sizeof(*sp),GFP_KERNEL); if (!sp) return NULL; - init_MUTEX(&sp->so_sema); - sp->so_seqid = 0; /* arbitrary */ + spin_lock_init(&sp->so_lock); INIT_LIST_HEAD(&sp->so_states); INIT_LIST_HEAD(&sp->so_delegations); + rpc_init_wait_queue(&sp->so_sequence.wait, "Seqid_waitqueue"); + sp->so_seqid.sequence = &sp->so_sequence; + spin_lock_init(&sp->so_sequence.lock); + INIT_LIST_HEAD(&sp->so_sequence.list); atomic_set(&sp->so_count, 1); return sp; } @@ -308,8 +309,7 @@ struct nfs4_state_owner *nfs4_get_state_owner(struct nfs_server *server, struct new = NULL; } spin_unlock(&clp->cl_lock); - if (new) - kfree(new); + kfree(new); if (sp != NULL) return sp; put_rpccred(cred); @@ -359,35 +359,25 @@ nfs4_alloc_open_state(void) memset(state->stateid.data, 0, sizeof(state->stateid.data)); atomic_set(&state->count, 1); INIT_LIST_HEAD(&state->lock_states); - init_MUTEX(&state->lock_sema); spin_lock_init(&state->state_lock); return state; } -static struct nfs4_state * -__nfs4_find_state(struct inode *inode, struct rpc_cred *cred, mode_t mode) +void +nfs4_state_set_mode_locked(struct nfs4_state *state, mode_t mode) { - struct nfs_inode *nfsi = NFS_I(inode); - struct nfs4_state *state; - - mode &= (FMODE_READ|FMODE_WRITE); - list_for_each_entry(state, &nfsi->open_states, inode_states) { - if (state->owner->so_cred != cred) - continue; - if ((mode & FMODE_READ) != 0 && state->nreaders == 0) - continue; - if ((mode & FMODE_WRITE) != 0 && state->nwriters == 0) - continue; - if ((state->state & mode) != mode) - continue; - atomic_inc(&state->count); - if (mode & FMODE_READ) - state->nreaders++; + if (state->state == mode) + return; + /* NB! List reordering - see the reclaim code for why. */ + if ((mode & FMODE_WRITE) != (state->state & FMODE_WRITE)) { if (mode & FMODE_WRITE) - state->nwriters++; - return state; + list_move(&state->open_states, &state->owner->so_states); + else + list_move_tail(&state->open_states, &state->owner->so_states); } - return NULL; + if (mode == 0) + list_del_init(&state->inode_states); + state->state = mode; } static struct nfs4_state * @@ -398,7 +388,7 @@ __nfs4_find_state_byowner(struct inode *inode, struct nfs4_state_owner *owner) list_for_each_entry(state, &nfsi->open_states, inode_states) { /* Is this in the process of being freed? */ - if (state->nreaders == 0 && state->nwriters == 0) + if (state->state == 0) continue; if (state->owner == owner) { atomic_inc(&state->count); @@ -408,17 +398,6 @@ __nfs4_find_state_byowner(struct inode *inode, struct nfs4_state_owner *owner) return NULL; } -struct nfs4_state * -nfs4_find_state(struct inode *inode, struct rpc_cred *cred, mode_t mode) -{ - struct nfs4_state *state; - - spin_lock(&inode->i_lock); - state = __nfs4_find_state(inode, cred, mode); - spin_unlock(&inode->i_lock); - return state; -} - static void nfs4_free_open_state(struct nfs4_state *state) { @@ -437,21 +416,23 @@ nfs4_get_open_state(struct inode *inode, struct nfs4_state_owner *owner) if (state) goto out; new = nfs4_alloc_open_state(); + spin_lock(&owner->so_lock); spin_lock(&inode->i_lock); state = __nfs4_find_state_byowner(inode, owner); if (state == NULL && new != NULL) { state = new; - /* Caller *must* be holding owner->so_sem */ - /* Note: The reclaim code dictates that we add stateless - * and read-only stateids to the end of the list */ - list_add_tail(&state->open_states, &owner->so_states); state->owner = owner; atomic_inc(&owner->so_count); list_add(&state->inode_states, &nfsi->open_states); state->inode = igrab(inode); spin_unlock(&inode->i_lock); + /* Note: The reclaim code dictates that we add stateless + * and read-only stateids to the end of the list */ + list_add_tail(&state->open_states, &owner->so_states); + spin_unlock(&owner->so_lock); } else { spin_unlock(&inode->i_lock); + spin_unlock(&owner->so_lock); if (new) nfs4_free_open_state(new); } @@ -461,68 +442,59 @@ out: /* * Beware! Caller must be holding exactly one - * reference to clp->cl_sem and owner->so_sema! + * reference to clp->cl_sem! */ void nfs4_put_open_state(struct nfs4_state *state) { struct inode *inode = state->inode; struct nfs4_state_owner *owner = state->owner; - if (!atomic_dec_and_lock(&state->count, &inode->i_lock)) + if (!atomic_dec_and_lock(&state->count, &owner->so_lock)) return; + spin_lock(&inode->i_lock); if (!list_empty(&state->inode_states)) list_del(&state->inode_states); - spin_unlock(&inode->i_lock); list_del(&state->open_states); + spin_unlock(&inode->i_lock); + spin_unlock(&owner->so_lock); iput(inode); - BUG_ON (state->state != 0); nfs4_free_open_state(state); nfs4_put_state_owner(owner); } /* - * Beware! Caller must be holding no references to clp->cl_sem! - * of owner->so_sema! + * Close the current file. */ void nfs4_close_state(struct nfs4_state *state, mode_t mode) { struct inode *inode = state->inode; struct nfs4_state_owner *owner = state->owner; - struct nfs4_client *clp = owner->so_client; - int newstate; + int oldstate, newstate = 0; atomic_inc(&owner->so_count); - down_read(&clp->cl_sem); - down(&owner->so_sema); /* Protect against nfs4_find_state() */ + spin_lock(&owner->so_lock); spin_lock(&inode->i_lock); if (mode & FMODE_READ) state->nreaders--; if (mode & FMODE_WRITE) state->nwriters--; - if (state->nwriters == 0) { - if (state->nreaders == 0) - list_del_init(&state->inode_states); - /* See reclaim code */ - list_move_tail(&state->open_states, &owner->so_states); + oldstate = newstate = state->state; + if (state->nreaders == 0) + newstate &= ~FMODE_READ; + if (state->nwriters == 0) + newstate &= ~FMODE_WRITE; + if (test_bit(NFS_DELEGATED_STATE, &state->flags)) { + nfs4_state_set_mode_locked(state, newstate); + oldstate = newstate; } spin_unlock(&inode->i_lock); - newstate = 0; - if (state->state != 0) { - if (state->nreaders) - newstate |= FMODE_READ; - if (state->nwriters) - newstate |= FMODE_WRITE; - if (state->state == newstate) - goto out; - if (nfs4_do_close(inode, state, newstate) == -EINPROGRESS) - return; - } -out: + spin_unlock(&owner->so_lock); + + if (oldstate != newstate && nfs4_do_close(inode, state) == 0) + return; nfs4_put_open_state(state); - up(&owner->so_sema); nfs4_put_state_owner(owner); - up_read(&clp->cl_sem); } /* @@ -546,19 +518,16 @@ __nfs4_find_lock_state(struct nfs4_state *state, fl_owner_t fl_owner) * Return a compatible lock_state. If no initialized lock_state structure * exists, return an uninitialized one. * - * The caller must be holding state->lock_sema */ static struct nfs4_lock_state *nfs4_alloc_lock_state(struct nfs4_state *state, fl_owner_t fl_owner) { struct nfs4_lock_state *lsp; struct nfs4_client *clp = state->owner->so_client; - lsp = kmalloc(sizeof(*lsp), GFP_KERNEL); + lsp = kzalloc(sizeof(*lsp), GFP_KERNEL); if (lsp == NULL) return NULL; - lsp->ls_flags = 0; - lsp->ls_seqid = 0; /* arbitrary */ - memset(lsp->ls_stateid.data, 0, sizeof(lsp->ls_stateid.data)); + lsp->ls_seqid.sequence = &state->owner->so_sequence; atomic_set(&lsp->ls_count, 1); lsp->ls_owner = fl_owner; spin_lock(&clp->cl_lock); @@ -572,7 +541,7 @@ static struct nfs4_lock_state *nfs4_alloc_lock_state(struct nfs4_state *state, f * Return a compatible lock_state. If no initialized lock_state structure * exists, return an uninitialized one. * - * The caller must be holding state->lock_sema and clp->cl_sem + * The caller must be holding clp->cl_sem */ static struct nfs4_lock_state *nfs4_get_lock_state(struct nfs4_state *state, fl_owner_t owner) { @@ -605,7 +574,7 @@ static struct nfs4_lock_state *nfs4_get_lock_state(struct nfs4_state *state, fl_ * Release reference to lock_state, and free it if we see that * it is no longer in use */ -static void nfs4_put_lock_state(struct nfs4_lock_state *lsp) +void nfs4_put_lock_state(struct nfs4_lock_state *lsp) { struct nfs4_state *state; @@ -673,29 +642,94 @@ void nfs4_copy_stateid(nfs4_stateid *dst, struct nfs4_state *state, fl_owner_t f nfs4_put_lock_state(lsp); } -/* -* Called with state->lock_sema and clp->cl_sem held. -*/ -void nfs4_increment_lock_seqid(int status, struct nfs4_lock_state *lsp) +struct nfs_seqid *nfs_alloc_seqid(struct nfs_seqid_counter *counter) +{ + struct rpc_sequence *sequence = counter->sequence; + struct nfs_seqid *new; + + new = kmalloc(sizeof(*new), GFP_KERNEL); + if (new != NULL) { + new->sequence = counter; + spin_lock(&sequence->lock); + list_add_tail(&new->list, &sequence->list); + spin_unlock(&sequence->lock); + } + return new; +} + +void nfs_free_seqid(struct nfs_seqid *seqid) { - if (status == NFS_OK || seqid_mutating_err(-status)) - lsp->ls_seqid++; + struct rpc_sequence *sequence = seqid->sequence->sequence; + + spin_lock(&sequence->lock); + list_del(&seqid->list); + spin_unlock(&sequence->lock); + rpc_wake_up(&sequence->wait); + kfree(seqid); } /* -* Called with sp->so_sema and clp->cl_sem held. -* -* Increment the seqid if the OPEN/OPEN_DOWNGRADE/CLOSE succeeded, or -* failed with a seqid incrementing error - -* see comments nfs_fs.h:seqid_mutating_error() -*/ -void nfs4_increment_seqid(int status, struct nfs4_state_owner *sp) -{ - if (status == NFS_OK || seqid_mutating_err(-status)) - sp->so_seqid++; - /* If the server returns BAD_SEQID, unhash state_owner here */ - if (status == -NFS4ERR_BAD_SEQID) + * Increment the seqid if the OPEN/OPEN_DOWNGRADE/CLOSE succeeded, or + * failed with a seqid incrementing error - + * see comments nfs_fs.h:seqid_mutating_error() + */ +static inline void nfs_increment_seqid(int status, struct nfs_seqid *seqid) +{ + switch (status) { + case 0: + break; + case -NFS4ERR_BAD_SEQID: + case -NFS4ERR_STALE_CLIENTID: + case -NFS4ERR_STALE_STATEID: + case -NFS4ERR_BAD_STATEID: + case -NFS4ERR_BADXDR: + case -NFS4ERR_RESOURCE: + case -NFS4ERR_NOFILEHANDLE: + /* Non-seqid mutating errors */ + return; + }; + /* + * Note: no locking needed as we are guaranteed to be first + * on the sequence list + */ + seqid->sequence->counter++; +} + +void nfs_increment_open_seqid(int status, struct nfs_seqid *seqid) +{ + if (status == -NFS4ERR_BAD_SEQID) { + struct nfs4_state_owner *sp = container_of(seqid->sequence, + struct nfs4_state_owner, so_seqid); nfs4_drop_state_owner(sp); + } + return nfs_increment_seqid(status, seqid); +} + +/* + * Increment the seqid if the LOCK/LOCKU succeeded, or + * failed with a seqid incrementing error - + * see comments nfs_fs.h:seqid_mutating_error() + */ +void nfs_increment_lock_seqid(int status, struct nfs_seqid *seqid) +{ + return nfs_increment_seqid(status, seqid); +} + +int nfs_wait_on_sequence(struct nfs_seqid *seqid, struct rpc_task *task) +{ + struct rpc_sequence *sequence = seqid->sequence->sequence; + int status = 0; + + if (sequence->list.next == &seqid->list) + goto out; + spin_lock(&sequence->lock); + if (sequence->list.next != &seqid->list) { + rpc_sleep_on(&sequence->wait, task, NULL, NULL); + status = -EAGAIN; + } + spin_unlock(&sequence->lock); +out: + return status; } static int reclaimer(void *); @@ -747,7 +781,7 @@ static int nfs4_reclaim_locks(struct nfs4_state_recovery_ops *ops, struct nfs4_s int status = 0; for (fl = inode->i_flock; fl != 0; fl = fl->fl_next) { - if (!(fl->fl_flags & FL_POSIX)) + if (!(fl->fl_flags & (FL_POSIX|FL_FLOCK))) continue; if (((struct nfs_open_context *)fl->fl_file->private_data)->state != state) continue; @@ -762,7 +796,7 @@ static int nfs4_reclaim_locks(struct nfs4_state_recovery_ops *ops, struct nfs4_s case -NFS4ERR_NO_GRACE: case -NFS4ERR_RECLAIM_BAD: case -NFS4ERR_RECLAIM_CONFLICT: - /* kill_proc(fl->fl_owner, SIGLOST, 1); */ + /* kill_proc(fl->fl_pid, SIGLOST, 1); */ break; case -NFS4ERR_STALE_CLIENTID: goto out_err; @@ -791,8 +825,6 @@ static int nfs4_reclaim_open_state(struct nfs4_state_recovery_ops *ops, struct n if (state->state == 0) continue; status = ops->recover_open(sp, state); - list_for_each_entry(lock, &state->lock_states, ls_locks) - lock->ls_flags &= ~NFS_LOCK_INITIALIZED; if (status >= 0) { status = nfs4_reclaim_locks(ops, state); if (status < 0) @@ -831,6 +863,28 @@ out_err: return status; } +static void nfs4_state_mark_reclaim(struct nfs4_client *clp) +{ + struct nfs4_state_owner *sp; + struct nfs4_state *state; + struct nfs4_lock_state *lock; + + /* Reset all sequence ids to zero */ + list_for_each_entry(sp, &clp->cl_state_owners, so_list) { + sp->so_seqid.counter = 0; + sp->so_seqid.flags = 0; + spin_lock(&sp->so_lock); + list_for_each_entry(state, &sp->so_states, open_states) { + list_for_each_entry(lock, &state->lock_states, ls_locks) { + lock->ls_seqid.counter = 0; + lock->ls_seqid.flags = 0; + lock->ls_flags &= ~NFS_LOCK_INITIALIZED; + } + } + spin_unlock(&sp->so_lock); + } +} + static int reclaimer(void *ptr) { struct reclaimer_args *args = (struct reclaimer_args *)ptr; @@ -864,6 +918,7 @@ restart_loop: default: ops = &nfs4_network_partition_recovery_ops; }; + nfs4_state_mark_reclaim(clp); status = __nfs4_init_client(clp); if (status) goto out_error; diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 6c564ef9489e..fbbace8a30c4 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -95,6 +95,8 @@ static int nfs_stat_to_errno(int); #define decode_getattr_maxsz (op_decode_hdr_maxsz + nfs4_fattr_maxsz) #define encode_savefh_maxsz (op_encode_hdr_maxsz) #define decode_savefh_maxsz (op_decode_hdr_maxsz) +#define encode_restorefh_maxsz (op_encode_hdr_maxsz) +#define decode_restorefh_maxsz (op_decode_hdr_maxsz) #define encode_fsinfo_maxsz (op_encode_hdr_maxsz + 2) #define decode_fsinfo_maxsz (op_decode_hdr_maxsz + 11) #define encode_renew_maxsz (op_encode_hdr_maxsz + 3) @@ -157,16 +159,20 @@ static int nfs_stat_to_errno(int); op_decode_hdr_maxsz + 2) #define NFS4_enc_write_sz (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ - op_encode_hdr_maxsz + 8) + op_encode_hdr_maxsz + 8 + \ + encode_getattr_maxsz) #define NFS4_dec_write_sz (compound_decode_hdr_maxsz + \ decode_putfh_maxsz + \ - op_decode_hdr_maxsz + 4) + op_decode_hdr_maxsz + 4 + \ + decode_getattr_maxsz) #define NFS4_enc_commit_sz (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ - op_encode_hdr_maxsz + 3) + op_encode_hdr_maxsz + 3 + \ + encode_getattr_maxsz) #define NFS4_dec_commit_sz (compound_decode_hdr_maxsz + \ decode_putfh_maxsz + \ - op_decode_hdr_maxsz + 2) + op_decode_hdr_maxsz + 2 + \ + decode_getattr_maxsz) #define NFS4_enc_open_sz (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ op_encode_hdr_maxsz + \ @@ -196,17 +202,21 @@ static int nfs_stat_to_errno(int); #define NFS4_enc_open_downgrade_sz \ (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ - op_encode_hdr_maxsz + 7) + op_encode_hdr_maxsz + 7 + \ + encode_getattr_maxsz) #define NFS4_dec_open_downgrade_sz \ (compound_decode_hdr_maxsz + \ decode_putfh_maxsz + \ - op_decode_hdr_maxsz + 4) + op_decode_hdr_maxsz + 4 + \ + decode_getattr_maxsz) #define NFS4_enc_close_sz (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ - op_encode_hdr_maxsz + 5) + op_encode_hdr_maxsz + 5 + \ + encode_getattr_maxsz) #define NFS4_dec_close_sz (compound_decode_hdr_maxsz + \ decode_putfh_maxsz + \ - op_decode_hdr_maxsz + 4) + op_decode_hdr_maxsz + 4 + \ + decode_getattr_maxsz) #define NFS4_enc_setattr_sz (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ op_encode_hdr_maxsz + 4 + \ @@ -300,30 +310,44 @@ static int nfs_stat_to_errno(int); decode_getfh_maxsz) #define NFS4_enc_remove_sz (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ - encode_remove_maxsz) + encode_remove_maxsz + \ + encode_getattr_maxsz) #define NFS4_dec_remove_sz (compound_decode_hdr_maxsz + \ decode_putfh_maxsz + \ - op_decode_hdr_maxsz + 5) + op_decode_hdr_maxsz + 5 + \ + decode_getattr_maxsz) #define NFS4_enc_rename_sz (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ encode_savefh_maxsz + \ encode_putfh_maxsz + \ - encode_rename_maxsz) + encode_rename_maxsz + \ + encode_getattr_maxsz + \ + encode_restorefh_maxsz + \ + encode_getattr_maxsz) #define NFS4_dec_rename_sz (compound_decode_hdr_maxsz + \ decode_putfh_maxsz + \ decode_savefh_maxsz + \ decode_putfh_maxsz + \ - decode_rename_maxsz) + decode_rename_maxsz + \ + decode_getattr_maxsz + \ + decode_restorefh_maxsz + \ + decode_getattr_maxsz) #define NFS4_enc_link_sz (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ encode_savefh_maxsz + \ encode_putfh_maxsz + \ - encode_link_maxsz) + encode_link_maxsz + \ + decode_getattr_maxsz + \ + encode_restorefh_maxsz + \ + decode_getattr_maxsz) #define NFS4_dec_link_sz (compound_decode_hdr_maxsz + \ decode_putfh_maxsz + \ decode_savefh_maxsz + \ decode_putfh_maxsz + \ - decode_link_maxsz) + decode_link_maxsz + \ + decode_getattr_maxsz + \ + decode_restorefh_maxsz + \ + decode_getattr_maxsz) #define NFS4_enc_symlink_sz (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ encode_symlink_maxsz + \ @@ -336,14 +360,20 @@ static int nfs_stat_to_errno(int); decode_getfh_maxsz) #define NFS4_enc_create_sz (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ + encode_savefh_maxsz + \ encode_create_maxsz + \ + encode_getfh_maxsz + \ encode_getattr_maxsz + \ - encode_getfh_maxsz) + encode_restorefh_maxsz + \ + encode_getattr_maxsz) #define NFS4_dec_create_sz (compound_decode_hdr_maxsz + \ decode_putfh_maxsz + \ + decode_savefh_maxsz + \ decode_create_maxsz + \ + decode_getfh_maxsz + \ decode_getattr_maxsz + \ - decode_getfh_maxsz) + decode_restorefh_maxsz + \ + decode_getattr_maxsz) #define NFS4_enc_pathconf_sz (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ encode_getattr_maxsz) @@ -602,10 +632,10 @@ static int encode_close(struct xdr_stream *xdr, const struct nfs_closeargs *arg) { uint32_t *p; - RESERVE_SPACE(8+sizeof(arg->stateid.data)); + RESERVE_SPACE(8+sizeof(arg->stateid->data)); WRITE32(OP_CLOSE); - WRITE32(arg->seqid); - WRITEMEM(arg->stateid.data, sizeof(arg->stateid.data)); + WRITE32(arg->seqid->sequence->counter); + WRITEMEM(arg->stateid->data, sizeof(arg->stateid->data)); return 0; } @@ -729,22 +759,18 @@ static int encode_lock(struct xdr_stream *xdr, const struct nfs_lockargs *arg) WRITE64(arg->length); WRITE32(opargs->new_lock_owner); if (opargs->new_lock_owner){ - struct nfs_open_to_lock *ol = opargs->u.open_lock; - RESERVE_SPACE(40); - WRITE32(ol->open_seqid); - WRITEMEM(&ol->open_stateid, sizeof(ol->open_stateid)); - WRITE32(ol->lock_seqid); - WRITE64(ol->lock_owner.clientid); + WRITE32(opargs->open_seqid->sequence->counter); + WRITEMEM(opargs->open_stateid->data, sizeof(opargs->open_stateid->data)); + WRITE32(opargs->lock_seqid->sequence->counter); + WRITE64(opargs->lock_owner.clientid); WRITE32(4); - WRITE32(ol->lock_owner.id); + WRITE32(opargs->lock_owner.id); } else { - struct nfs_exist_lock *el = opargs->u.exist_lock; - RESERVE_SPACE(20); - WRITEMEM(&el->stateid, sizeof(el->stateid)); - WRITE32(el->seqid); + WRITEMEM(opargs->lock_stateid->data, sizeof(opargs->lock_stateid->data)); + WRITE32(opargs->lock_seqid->sequence->counter); } return 0; @@ -775,8 +801,8 @@ static int encode_locku(struct xdr_stream *xdr, const struct nfs_lockargs *arg) RESERVE_SPACE(44); WRITE32(OP_LOCKU); WRITE32(arg->type); - WRITE32(opargs->seqid); - WRITEMEM(&opargs->stateid, sizeof(opargs->stateid)); + WRITE32(opargs->seqid->sequence->counter); + WRITEMEM(opargs->stateid->data, sizeof(opargs->stateid->data)); WRITE64(arg->offset); WRITE64(arg->length); @@ -826,7 +852,7 @@ static inline void encode_openhdr(struct xdr_stream *xdr, const struct nfs_opena */ RESERVE_SPACE(8); WRITE32(OP_OPEN); - WRITE32(arg->seqid); + WRITE32(arg->seqid->sequence->counter); encode_share_access(xdr, arg->open_flags); RESERVE_SPACE(16); WRITE64(arg->clientid); @@ -941,7 +967,7 @@ static int encode_open_confirm(struct xdr_stream *xdr, const struct nfs_open_con RESERVE_SPACE(8+sizeof(arg->stateid.data)); WRITE32(OP_OPEN_CONFIRM); WRITEMEM(arg->stateid.data, sizeof(arg->stateid.data)); - WRITE32(arg->seqid); + WRITE32(arg->seqid->sequence->counter); return 0; } @@ -950,10 +976,10 @@ static int encode_open_downgrade(struct xdr_stream *xdr, const struct nfs_closea { uint32_t *p; - RESERVE_SPACE(8+sizeof(arg->stateid.data)); + RESERVE_SPACE(8+sizeof(arg->stateid->data)); WRITE32(OP_OPEN_DOWNGRADE); - WRITEMEM(arg->stateid.data, sizeof(arg->stateid.data)); - WRITE32(arg->seqid); + WRITEMEM(arg->stateid->data, sizeof(arg->stateid->data)); + WRITE32(arg->seqid->sequence->counter); encode_share_access(xdr, arg->open_flags); return 0; } @@ -1117,6 +1143,17 @@ static int encode_renew(struct xdr_stream *xdr, const struct nfs4_client *client } static int +encode_restorefh(struct xdr_stream *xdr) +{ + uint32_t *p; + + RESERVE_SPACE(4); + WRITE32(OP_RESTOREFH); + + return 0; +} + +static int encode_setacl(struct xdr_stream *xdr, struct nfs_setaclargs *arg) { uint32_t *p; @@ -1296,14 +1333,18 @@ static int nfs4_xdr_enc_remove(struct rpc_rqst *req, uint32_t *p, const struct n { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 2, + .nops = 3, }; int status; xdr_init_encode(&xdr, &req->rq_snd_buf, p); encode_compound_hdr(&xdr, &hdr); - if ((status = encode_putfh(&xdr, args->fh)) == 0) - status = encode_remove(&xdr, args->name); + if ((status = encode_putfh(&xdr, args->fh)) != 0) + goto out; + if ((status = encode_remove(&xdr, args->name)) != 0) + goto out; + status = encode_getfattr(&xdr, args->bitmask); +out: return status; } @@ -1314,7 +1355,7 @@ static int nfs4_xdr_enc_rename(struct rpc_rqst *req, uint32_t *p, const struct n { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 4, + .nops = 7, }; int status; @@ -1326,7 +1367,13 @@ static int nfs4_xdr_enc_rename(struct rpc_rqst *req, uint32_t *p, const struct n goto out; if ((status = encode_putfh(&xdr, args->new_dir)) != 0) goto out; - status = encode_rename(&xdr, args->old_name, args->new_name); + if ((status = encode_rename(&xdr, args->old_name, args->new_name)) != 0) + goto out; + if ((status = encode_getfattr(&xdr, args->bitmask)) != 0) + goto out; + if ((status = encode_restorefh(&xdr)) != 0) + goto out; + status = encode_getfattr(&xdr, args->bitmask); out: return status; } @@ -1338,7 +1385,7 @@ static int nfs4_xdr_enc_link(struct rpc_rqst *req, uint32_t *p, const struct nfs { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 4, + .nops = 7, }; int status; @@ -1350,7 +1397,13 @@ static int nfs4_xdr_enc_link(struct rpc_rqst *req, uint32_t *p, const struct nfs goto out; if ((status = encode_putfh(&xdr, args->dir_fh)) != 0) goto out; - status = encode_link(&xdr, args->name); + if ((status = encode_link(&xdr, args->name)) != 0) + goto out; + if ((status = encode_getfattr(&xdr, args->bitmask)) != 0) + goto out; + if ((status = encode_restorefh(&xdr)) != 0) + goto out; + status = encode_getfattr(&xdr, args->bitmask); out: return status; } @@ -1362,7 +1415,7 @@ static int nfs4_xdr_enc_create(struct rpc_rqst *req, uint32_t *p, const struct n { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 4, + .nops = 7, }; int status; @@ -1370,10 +1423,16 @@ static int nfs4_xdr_enc_create(struct rpc_rqst *req, uint32_t *p, const struct n encode_compound_hdr(&xdr, &hdr); if ((status = encode_putfh(&xdr, args->dir_fh)) != 0) goto out; + if ((status = encode_savefh(&xdr)) != 0) + goto out; if ((status = encode_create(&xdr, args)) != 0) goto out; if ((status = encode_getfh(&xdr)) != 0) goto out; + if ((status = encode_getfattr(&xdr, args->bitmask)) != 0) + goto out; + if ((status = encode_restorefh(&xdr)) != 0) + goto out; status = encode_getfattr(&xdr, args->bitmask); out: return status; @@ -1412,7 +1471,7 @@ static int nfs4_xdr_enc_close(struct rpc_rqst *req, uint32_t *p, struct nfs_clos { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 2, + .nops = 3, }; int status; @@ -1422,6 +1481,9 @@ static int nfs4_xdr_enc_close(struct rpc_rqst *req, uint32_t *p, struct nfs_clos if(status) goto out; status = encode_close(&xdr, args); + if (status != 0) + goto out; + status = encode_getfattr(&xdr, args->bitmask); out: return status; } @@ -1433,15 +1495,21 @@ static int nfs4_xdr_enc_open(struct rpc_rqst *req, uint32_t *p, struct nfs_opena { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 4, + .nops = 7, }; int status; + status = nfs_wait_on_sequence(args->seqid, req->rq_task); + if (status != 0) + goto out; xdr_init_encode(&xdr, &req->rq_snd_buf, p); encode_compound_hdr(&xdr, &hdr); status = encode_putfh(&xdr, args->fh); if (status) goto out; + status = encode_savefh(&xdr); + if (status) + goto out; status = encode_open(&xdr, args); if (status) goto out; @@ -1449,6 +1517,12 @@ static int nfs4_xdr_enc_open(struct rpc_rqst *req, uint32_t *p, struct nfs_opena if (status) goto out; status = encode_getfattr(&xdr, args->bitmask); + if (status) + goto out; + status = encode_restorefh(&xdr); + if (status) + goto out; + status = encode_getfattr(&xdr, args->bitmask); out: return status; } @@ -1464,6 +1538,9 @@ static int nfs4_xdr_enc_open_confirm(struct rpc_rqst *req, uint32_t *p, struct n }; int status; + status = nfs_wait_on_sequence(args->seqid, req->rq_task); + if (status != 0) + goto out; xdr_init_encode(&xdr, &req->rq_snd_buf, p); encode_compound_hdr(&xdr, &hdr); status = encode_putfh(&xdr, args->fh); @@ -1485,6 +1562,9 @@ static int nfs4_xdr_enc_open_noattr(struct rpc_rqst *req, uint32_t *p, struct nf }; int status; + status = nfs_wait_on_sequence(args->seqid, req->rq_task); + if (status != 0) + goto out; xdr_init_encode(&xdr, &req->rq_snd_buf, p); encode_compound_hdr(&xdr, &hdr); status = encode_putfh(&xdr, args->fh); @@ -1502,7 +1582,7 @@ static int nfs4_xdr_enc_open_downgrade(struct rpc_rqst *req, uint32_t *p, struct { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 2, + .nops = 3, }; int status; @@ -1512,6 +1592,9 @@ static int nfs4_xdr_enc_open_downgrade(struct rpc_rqst *req, uint32_t *p, struct if (status) goto out; status = encode_open_downgrade(&xdr, args); + if (status != 0) + goto out; + status = encode_getfattr(&xdr, args->bitmask); out: return status; } @@ -1525,8 +1608,15 @@ static int nfs4_xdr_enc_lock(struct rpc_rqst *req, uint32_t *p, struct nfs_locka struct compound_hdr hdr = { .nops = 2, }; + struct nfs_lock_opargs *opargs = args->u.lock; int status; + status = nfs_wait_on_sequence(opargs->lock_seqid, req->rq_task); + if (status != 0) + goto out; + /* Do we need to do an open_to_lock_owner? */ + if (opargs->lock_seqid->sequence->flags & NFS_SEQID_CONFIRMED) + opargs->new_lock_owner = 0; xdr_init_encode(&xdr, &req->rq_snd_buf, p); encode_compound_hdr(&xdr, &hdr); status = encode_putfh(&xdr, args->fh); @@ -1713,7 +1803,7 @@ static int nfs4_xdr_enc_write(struct rpc_rqst *req, uint32_t *p, struct nfs_writ { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 2, + .nops = 3, }; int status; @@ -1723,6 +1813,9 @@ static int nfs4_xdr_enc_write(struct rpc_rqst *req, uint32_t *p, struct nfs_writ if (status) goto out; status = encode_write(&xdr, args); + if (status) + goto out; + status = encode_getfattr(&xdr, args->bitmask); out: return status; } @@ -1734,7 +1827,7 @@ static int nfs4_xdr_enc_commit(struct rpc_rqst *req, uint32_t *p, struct nfs_wri { struct xdr_stream xdr; struct compound_hdr hdr = { - .nops = 2, + .nops = 3, }; int status; @@ -1744,6 +1837,9 @@ static int nfs4_xdr_enc_commit(struct rpc_rqst *req, uint32_t *p, struct nfs_wri if (status) goto out; status = encode_commit(&xdr, args); + if (status) + goto out; + status = encode_getfattr(&xdr, args->bitmask); out: return status; } @@ -2670,8 +2766,7 @@ static int decode_server_caps(struct xdr_stream *xdr, struct nfs4_server_caps_re goto xdr_error; status = verify_attr_len(xdr, savep, attrlen); xdr_error: - if (status != 0) - printk(KERN_NOTICE "%s: xdr error %d!\n", __FUNCTION__, -status); + dprintk("%s: xdr returned %d!\n", __FUNCTION__, -status); return status; } @@ -2704,8 +2799,7 @@ static int decode_statfs(struct xdr_stream *xdr, struct nfs_fsstat *fsstat) status = verify_attr_len(xdr, savep, attrlen); xdr_error: - if (status != 0) - printk(KERN_NOTICE "%s: xdr error %d!\n", __FUNCTION__, -status); + dprintk("%s: xdr returned %d!\n", __FUNCTION__, -status); return status; } @@ -2730,8 +2824,7 @@ static int decode_pathconf(struct xdr_stream *xdr, struct nfs_pathconf *pathconf status = verify_attr_len(xdr, savep, attrlen); xdr_error: - if (status != 0) - printk(KERN_NOTICE "%s: xdr error %d!\n", __FUNCTION__, -status); + dprintk("%s: xdr returned %d!\n", __FUNCTION__, -status); return status; } @@ -2787,13 +2880,10 @@ static int decode_getfattr(struct xdr_stream *xdr, struct nfs_fattr *fattr, cons goto xdr_error; if ((status = decode_attr_time_modify(xdr, bitmap, &fattr->mtime)) != 0) goto xdr_error; - if ((status = verify_attr_len(xdr, savep, attrlen)) == 0) { + if ((status = verify_attr_len(xdr, savep, attrlen)) == 0) fattr->valid = NFS_ATTR_FATTR | NFS_ATTR_FATTR_V3 | NFS_ATTR_FATTR_V4; - fattr->timestamp = jiffies; - } xdr_error: - if (status != 0) - printk(KERN_NOTICE "%s: xdr error %d!\n", __FUNCTION__, -status); + dprintk("%s: xdr returned %d\n", __FUNCTION__, -status); return status; } @@ -2826,8 +2916,7 @@ static int decode_fsinfo(struct xdr_stream *xdr, struct nfs_fsinfo *fsinfo) status = verify_attr_len(xdr, savep, attrlen); xdr_error: - if (status != 0) - printk(KERN_NOTICE "%s: xdr error %d!\n", __FUNCTION__, -status); + dprintk("%s: xdr returned %d!\n", __FUNCTION__, -status); return status; } @@ -2890,8 +2979,8 @@ static int decode_lock(struct xdr_stream *xdr, struct nfs_lockres *res) status = decode_op_hdr(xdr, OP_LOCK); if (status == 0) { - READ_BUF(sizeof(nfs4_stateid)); - COPYMEM(&res->u.stateid, sizeof(res->u.stateid)); + READ_BUF(sizeof(res->u.stateid.data)); + COPYMEM(res->u.stateid.data, sizeof(res->u.stateid.data)); } else if (status == -NFS4ERR_DENIED) return decode_lock_denied(xdr, &res->u.denied); return status; @@ -2913,8 +3002,8 @@ static int decode_locku(struct xdr_stream *xdr, struct nfs_lockres *res) status = decode_op_hdr(xdr, OP_LOCKU); if (status == 0) { - READ_BUF(sizeof(nfs4_stateid)); - COPYMEM(&res->u.stateid, sizeof(res->u.stateid)); + READ_BUF(sizeof(res->u.stateid.data)); + COPYMEM(res->u.stateid.data, sizeof(res->u.stateid.data)); } return status; } @@ -2994,7 +3083,7 @@ static int decode_open(struct xdr_stream *xdr, struct nfs_openres *res) p += bmlen; return decode_delegation(xdr, res); xdr_error: - printk(KERN_NOTICE "%s: xdr error!\n", __FUNCTION__); + dprintk("%s: Bitmap too large! Length = %u\n", __FUNCTION__, bmlen); return -EIO; } @@ -3208,6 +3297,12 @@ static int decode_renew(struct xdr_stream *xdr) return decode_op_hdr(xdr, OP_RENEW); } +static int +decode_restorefh(struct xdr_stream *xdr) +{ + return decode_op_hdr(xdr, OP_RESTOREFH); +} + static int decode_getacl(struct xdr_stream *xdr, struct rpc_rqst *req, size_t *acl_len) { @@ -3243,7 +3338,8 @@ static int decode_getacl(struct xdr_stream *xdr, struct rpc_rqst *req, if (attrlen <= *acl_len) xdr_read_pages(xdr, attrlen); *acl_len = attrlen; - } + } else + status = -EOPNOTSUPP; out: return status; @@ -3352,6 +3448,9 @@ static int nfs4_xdr_dec_open_downgrade(struct rpc_rqst *rqstp, uint32_t *p, stru if (status) goto out; status = decode_open_downgrade(&xdr, res); + if (status != 0) + goto out; + decode_getfattr(&xdr, res->fattr, res->server); out: return status; } @@ -3424,7 +3523,7 @@ out: /* * Decode REMOVE response */ -static int nfs4_xdr_dec_remove(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_change_info *cinfo) +static int nfs4_xdr_dec_remove(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_remove_res *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -3433,8 +3532,11 @@ static int nfs4_xdr_dec_remove(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_ xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p); if ((status = decode_compound_hdr(&xdr, &hdr)) != 0) goto out; - if ((status = decode_putfh(&xdr)) == 0) - status = decode_remove(&xdr, cinfo); + if ((status = decode_putfh(&xdr)) != 0) + goto out; + if ((status = decode_remove(&xdr, &res->cinfo)) != 0) + goto out; + decode_getfattr(&xdr, res->dir_attr, res->server); out: return status; } @@ -3457,7 +3559,14 @@ static int nfs4_xdr_dec_rename(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_ goto out; if ((status = decode_putfh(&xdr)) != 0) goto out; - status = decode_rename(&xdr, &res->old_cinfo, &res->new_cinfo); + if ((status = decode_rename(&xdr, &res->old_cinfo, &res->new_cinfo)) != 0) + goto out; + /* Current FH is target directory */ + if (decode_getfattr(&xdr, res->new_fattr, res->server) != 0) + goto out; + if ((status = decode_restorefh(&xdr)) != 0) + goto out; + decode_getfattr(&xdr, res->old_fattr, res->server); out: return status; } @@ -3465,7 +3574,7 @@ out: /* * Decode LINK response */ -static int nfs4_xdr_dec_link(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_change_info *cinfo) +static int nfs4_xdr_dec_link(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_link_res *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -3480,7 +3589,17 @@ static int nfs4_xdr_dec_link(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_ch goto out; if ((status = decode_putfh(&xdr)) != 0) goto out; - status = decode_link(&xdr, cinfo); + if ((status = decode_link(&xdr, &res->cinfo)) != 0) + goto out; + /* + * Note order: OP_LINK leaves the directory as the current + * filehandle. + */ + if (decode_getfattr(&xdr, res->dir_attr, res->server) != 0) + goto out; + if ((status = decode_restorefh(&xdr)) != 0) + goto out; + decode_getfattr(&xdr, res->fattr, res->server); out: return status; } @@ -3499,13 +3618,17 @@ static int nfs4_xdr_dec_create(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_ goto out; if ((status = decode_putfh(&xdr)) != 0) goto out; + if ((status = decode_savefh(&xdr)) != 0) + goto out; if ((status = decode_create(&xdr,&res->dir_cinfo)) != 0) goto out; if ((status = decode_getfh(&xdr, res->fh)) != 0) goto out; - status = decode_getfattr(&xdr, res->fattr, res->server); - if (status == NFS4ERR_DELAY) - status = 0; + if (decode_getfattr(&xdr, res->fattr, res->server) != 0) + goto out; + if ((status = decode_restorefh(&xdr)) != 0) + goto out; + decode_getfattr(&xdr, res->dir_fattr, res->server); out: return status; } @@ -3623,6 +3746,15 @@ static int nfs4_xdr_dec_close(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_cl if (status) goto out; status = decode_close(&xdr, res); + if (status != 0) + goto out; + /* + * Note: Server may do delete on close for this file + * in which case the getattr call will fail with + * an ESTALE error. Shouldn't be a problem, + * though, since fattr->valid will remain unset. + */ + decode_getfattr(&xdr, res->fattr, res->server); out: return status; } @@ -3643,15 +3775,20 @@ static int nfs4_xdr_dec_open(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_ope status = decode_putfh(&xdr); if (status) goto out; + status = decode_savefh(&xdr); + if (status) + goto out; status = decode_open(&xdr, res); if (status) goto out; status = decode_getfh(&xdr, &res->fh); if (status) goto out; - status = decode_getfattr(&xdr, res->f_attr, res->server); - if (status == NFS4ERR_DELAY) - status = 0; + if (decode_getfattr(&xdr, res->f_attr, res->server) != 0) + goto out; + if ((status = decode_restorefh(&xdr)) != 0) + goto out; + decode_getfattr(&xdr, res->dir_attr, res->server); out: return status; } @@ -3869,6 +4006,9 @@ static int nfs4_xdr_dec_write(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_wr if (status) goto out; status = decode_write(&xdr, res); + if (status) + goto out; + decode_getfattr(&xdr, res->fattr, res->server); if (!status) status = res->count; out: @@ -3892,6 +4032,9 @@ static int nfs4_xdr_dec_commit(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_w if (status) goto out; status = decode_commit(&xdr, res); + if (status) + goto out; + decode_getfattr(&xdr, res->fattr, res->server); out: return status; } diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c index be23c3fb9260..e1e3ca5d746b 100644 --- a/fs/nfs/proc.c +++ b/fs/nfs/proc.c @@ -61,7 +61,7 @@ nfs_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle, int status; dprintk("%s: call getattr\n", __FUNCTION__); - fattr->valid = 0; + nfs_fattr_init(fattr); status = rpc_call(server->client_sys, NFSPROC_GETATTR, fhandle, fattr, 0); dprintk("%s: reply getattr: %d\n", __FUNCTION__, status); if (status) @@ -93,7 +93,7 @@ nfs_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, int status; dprintk("NFS call getattr\n"); - fattr->valid = 0; + nfs_fattr_init(fattr); status = rpc_call(server->client, NFSPROC_GETATTR, fhandle, fattr, 0); dprintk("NFS reply getattr: %d\n", status); @@ -112,7 +112,7 @@ nfs_proc_setattr(struct dentry *dentry, struct nfs_fattr *fattr, int status; dprintk("NFS call setattr\n"); - fattr->valid = 0; + nfs_fattr_init(fattr); status = rpc_call(NFS_CLIENT(inode), NFSPROC_SETATTR, &arg, fattr, 0); if (status == 0) nfs_setattr_update_inode(inode, sattr); @@ -136,7 +136,7 @@ nfs_proc_lookup(struct inode *dir, struct qstr *name, int status; dprintk("NFS call lookup %s\n", name->name); - fattr->valid = 0; + nfs_fattr_init(fattr); status = rpc_call(NFS_CLIENT(dir), NFSPROC_LOOKUP, &arg, &res, 0); dprintk("NFS reply lookup: %d\n", status); return status; @@ -174,7 +174,7 @@ static int nfs_proc_read(struct nfs_read_data *rdata) dprintk("NFS call read %d @ %Ld\n", rdata->args.count, (long long) rdata->args.offset); - fattr->valid = 0; + nfs_fattr_init(fattr); status = rpc_call_sync(NFS_CLIENT(inode), &msg, flags); if (status >= 0) { nfs_refresh_inode(inode, fattr); @@ -203,10 +203,10 @@ static int nfs_proc_write(struct nfs_write_data *wdata) dprintk("NFS call write %d @ %Ld\n", wdata->args.count, (long long) wdata->args.offset); - fattr->valid = 0; + nfs_fattr_init(fattr); status = rpc_call_sync(NFS_CLIENT(inode), &msg, flags); if (status >= 0) { - nfs_refresh_inode(inode, fattr); + nfs_post_op_update_inode(inode, fattr); wdata->res.count = wdata->args.count; wdata->verf.committed = NFS_FILE_SYNC; } @@ -216,7 +216,7 @@ static int nfs_proc_write(struct nfs_write_data *wdata) static int nfs_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr, - int flags) + int flags, struct nameidata *nd) { struct nfs_fh fhandle; struct nfs_fattr fattr; @@ -232,7 +232,7 @@ nfs_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr, }; int status; - fattr.valid = 0; + nfs_fattr_init(&fattr); dprintk("NFS call create %s\n", dentry->d_name.name); status = rpc_call(NFS_CLIENT(dir), NFSPROC_CREATE, &arg, &res, 0); if (status == 0) @@ -273,12 +273,13 @@ nfs_proc_mknod(struct inode *dir, struct dentry *dentry, struct iattr *sattr, sattr->ia_size = new_encode_dev(rdev);/* get out your barf bag */ } - fattr.valid = 0; + nfs_fattr_init(&fattr); status = rpc_call(NFS_CLIENT(dir), NFSPROC_CREATE, &arg, &res, 0); + nfs_mark_for_revalidate(dir); if (status == -EINVAL && S_ISFIFO(mode)) { sattr->ia_mode = mode; - fattr.valid = 0; + nfs_fattr_init(&fattr); status = rpc_call(NFS_CLIENT(dir), NFSPROC_CREATE, &arg, &res, 0); } if (status == 0) @@ -305,6 +306,7 @@ nfs_proc_remove(struct inode *dir, struct qstr *name) dprintk("NFS call remove %s\n", name->name); status = rpc_call_sync(NFS_CLIENT(dir), &msg, 0); + nfs_mark_for_revalidate(dir); dprintk("NFS reply remove: %d\n", status); return status; @@ -331,8 +333,10 @@ nfs_proc_unlink_done(struct dentry *dir, struct rpc_task *task) { struct rpc_message *msg = &task->tk_msg; - if (msg->rpc_argp) + if (msg->rpc_argp) { + nfs_mark_for_revalidate(dir->d_inode); kfree(msg->rpc_argp); + } return 0; } @@ -352,6 +356,8 @@ nfs_proc_rename(struct inode *old_dir, struct qstr *old_name, dprintk("NFS call rename %s -> %s\n", old_name->name, new_name->name); status = rpc_call(NFS_CLIENT(old_dir), NFSPROC_RENAME, &arg, NULL, 0); + nfs_mark_for_revalidate(old_dir); + nfs_mark_for_revalidate(new_dir); dprintk("NFS reply rename: %d\n", status); return status; } @@ -369,6 +375,8 @@ nfs_proc_link(struct inode *inode, struct inode *dir, struct qstr *name) dprintk("NFS call link %s\n", name->name); status = rpc_call(NFS_CLIENT(inode), NFSPROC_LINK, &arg, NULL, 0); + nfs_mark_for_revalidate(inode); + nfs_mark_for_revalidate(dir); dprintk("NFS reply link: %d\n", status); return status; } @@ -391,9 +399,10 @@ nfs_proc_symlink(struct inode *dir, struct qstr *name, struct qstr *path, if (path->len > NFS2_MAXPATHLEN) return -ENAMETOOLONG; dprintk("NFS call symlink %s -> %s\n", name->name, path->name); - fattr->valid = 0; + nfs_fattr_init(fattr); fhandle->size = 0; status = rpc_call(NFS_CLIENT(dir), NFSPROC_SYMLINK, &arg, NULL, 0); + nfs_mark_for_revalidate(dir); dprintk("NFS reply symlink: %d\n", status); return status; } @@ -416,8 +425,9 @@ nfs_proc_mkdir(struct inode *dir, struct dentry *dentry, struct iattr *sattr) int status; dprintk("NFS call mkdir %s\n", dentry->d_name.name); - fattr.valid = 0; + nfs_fattr_init(&fattr); status = rpc_call(NFS_CLIENT(dir), NFSPROC_MKDIR, &arg, &res, 0); + nfs_mark_for_revalidate(dir); if (status == 0) status = nfs_instantiate(dentry, &fhandle, &fattr); dprintk("NFS reply mkdir: %d\n", status); @@ -436,6 +446,7 @@ nfs_proc_rmdir(struct inode *dir, struct qstr *name) dprintk("NFS call rmdir %s\n", name->name); status = rpc_call(NFS_CLIENT(dir), NFSPROC_RMDIR, &arg, NULL, 0); + nfs_mark_for_revalidate(dir); dprintk("NFS reply rmdir: %d\n", status); return status; } @@ -484,7 +495,7 @@ nfs_proc_statfs(struct nfs_server *server, struct nfs_fh *fhandle, int status; dprintk("NFS call statfs\n"); - stat->fattr->valid = 0; + nfs_fattr_init(stat->fattr); status = rpc_call(server->client, NFSPROC_STATFS, fhandle, &fsinfo, 0); dprintk("NFS reply statfs: %d\n", status); if (status) @@ -507,7 +518,7 @@ nfs_proc_fsinfo(struct nfs_server *server, struct nfs_fh *fhandle, int status; dprintk("NFS call fsinfo\n"); - info->fattr->valid = 0; + nfs_fattr_init(info->fattr); status = rpc_call(server->client, NFSPROC_STATFS, fhandle, &fsinfo, 0); dprintk("NFS reply fsinfo: %d\n", status); if (status) @@ -579,7 +590,7 @@ nfs_write_done(struct rpc_task *task) struct nfs_write_data *data = (struct nfs_write_data *) task->tk_calldata; if (task->tk_status >= 0) - nfs_refresh_inode(data->inode, data->res.fattr); + nfs_post_op_update_inode(data->inode, data->res.fattr); nfs_writeback_done(task); } diff --git a/fs/nfs/read.c b/fs/nfs/read.c index 6ceb1d471f20..5f20eafba8ec 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -184,14 +184,13 @@ static void nfs_readpage_release(struct nfs_page *req) { unlock_page(req->wb_page); - nfs_clear_request(req); - nfs_release_request(req); - dprintk("NFS: read done (%s/%Ld %d@%Ld)\n", req->wb_context->dentry->d_inode->i_sb->s_id, (long long)NFS_FILEID(req->wb_context->dentry->d_inode), req->wb_bytes, (long long)req_offset(req)); + nfs_clear_request(req); + nfs_release_request(req); } /* @@ -216,6 +215,7 @@ static void nfs_read_rpcsetup(struct nfs_page *req, struct nfs_read_data *data, data->res.fattr = &data->fattr; data->res.count = count; data->res.eof = 0; + nfs_fattr_init(&data->fattr); NFS_PROTO(inode)->read_setup(data); @@ -507,7 +507,7 @@ int nfs_readpage(struct file *file, struct page *page) goto out_error; if (file == NULL) { - ctx = nfs_find_open_context(inode, FMODE_READ); + ctx = nfs_find_open_context(inode, NULL, FMODE_READ); if (ctx == NULL) return -EBADF; } else @@ -576,7 +576,7 @@ int nfs_readpages(struct file *filp, struct address_space *mapping, nr_pages); if (filp == NULL) { - desc.ctx = nfs_find_open_context(inode, FMODE_READ); + desc.ctx = nfs_find_open_context(inode, NULL, FMODE_READ); if (desc.ctx == NULL) return -EBADF; } else diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c index f732541a3332..d639d172d568 100644 --- a/fs/nfs/unlink.c +++ b/fs/nfs/unlink.c @@ -52,8 +52,7 @@ nfs_put_unlinkdata(struct nfs_unlinkdata *data) { if (--data->count == 0) { nfs_detach_unlinkdata(data); - if (data->name.name != NULL) - kfree(data->name.name); + kfree(data->name.name); kfree(data); } } diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 5130eda231d7..3107908e5f3f 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -189,6 +189,7 @@ static int nfs_writepage_sync(struct nfs_open_context *ctx, struct inode *inode, (long long)NFS_FILEID(inode), count, (long long)(page_offset(page) + offset)); + set_page_writeback(page); nfs_begin_data_update(inode); do { if (count < wsize) @@ -221,6 +222,7 @@ static int nfs_writepage_sync(struct nfs_open_context *ctx, struct inode *inode, io_error: nfs_end_data_update(inode); + end_page_writeback(page); nfs_writedata_free(wdata); return written ? written : result; } @@ -294,7 +296,7 @@ int nfs_writepage(struct page *page, struct writeback_control *wbc) if (page->index >= end_index+1 || !offset) goto out; do_it: - ctx = nfs_find_open_context(inode, FMODE_WRITE); + ctx = nfs_find_open_context(inode, NULL, FMODE_WRITE); if (ctx == NULL) { err = -EBADF; goto out; @@ -734,14 +736,14 @@ int nfs_updatepage(struct file *file, struct page *page, unsigned int offset, unsigned int count) { struct nfs_open_context *ctx = (struct nfs_open_context *)file->private_data; - struct dentry *dentry = file->f_dentry; struct inode *inode = page->mapping->host; struct nfs_page *req; int status = 0; dprintk("NFS: nfs_updatepage(%s/%s %d@%Ld)\n", - dentry->d_parent->d_name.name, dentry->d_name.name, - count, (long long)(page_offset(page) +offset)); + file->f_dentry->d_parent->d_name.name, + file->f_dentry->d_name.name, count, + (long long)(page_offset(page) +offset)); if (IS_SYNC(inode)) { status = nfs_writepage_sync(ctx, inode, page, offset, count, 0); @@ -850,7 +852,6 @@ static void nfs_write_rpcsetup(struct nfs_page *req, unsigned int count, unsigned int offset, int how) { - struct rpc_task *task = &data->task; struct inode *inode; /* Set up the RPC argument and reply structs @@ -870,6 +871,7 @@ static void nfs_write_rpcsetup(struct nfs_page *req, data->res.fattr = &data->fattr; data->res.count = count; data->res.verf = &data->verf; + nfs_fattr_init(&data->fattr); NFS_PROTO(inode)->write_setup(data, how); @@ -880,7 +882,7 @@ static void nfs_write_rpcsetup(struct nfs_page *req, data->task.tk_release = nfs_writedata_release; dprintk("NFS: %4d initiated write call (req %s/%Ld, %u bytes @ offset %Lu)\n", - task->tk_pid, + data->task.tk_pid, inode->i_sb->s_id, (long long)NFS_FILEID(inode), count, @@ -929,7 +931,7 @@ static int nfs_flush_multi(struct list_head *head, struct inode *inode, int how) atomic_set(&req->wb_complete, requests); ClearPageError(page); - SetPageWriteback(page); + set_page_writeback(page); offset = 0; nbytes = req->wb_bytes; do { @@ -992,7 +994,7 @@ static int nfs_flush_one(struct list_head *head, struct inode *inode, int how) nfs_list_remove_request(req); nfs_list_add_request(req, &data->pages); ClearPageError(req->wb_page); - SetPageWriteback(req->wb_page); + set_page_writeback(req->wb_page); *pages++ = req->wb_page; count += req->wb_bytes; } @@ -1216,7 +1218,6 @@ static void nfs_commit_release(struct rpc_task *task) static void nfs_commit_rpcsetup(struct list_head *head, struct nfs_write_data *data, int how) { - struct rpc_task *task = &data->task; struct nfs_page *first; struct inode *inode; @@ -1237,6 +1238,7 @@ static void nfs_commit_rpcsetup(struct list_head *head, data->res.count = 0; data->res.fattr = &data->fattr; data->res.verf = &data->verf; + nfs_fattr_init(&data->fattr); NFS_PROTO(inode)->commit_setup(data, how); @@ -1246,7 +1248,7 @@ static void nfs_commit_rpcsetup(struct list_head *head, /* Release requests */ data->task.tk_release = nfs_commit_release; - dprintk("NFS: %4d initiated commit call\n", task->tk_pid); + dprintk("NFS: %4d initiated commit call\n", data->task.tk_pid); } /* diff --git a/fs/nfs_common/nfsacl.c b/fs/nfs_common/nfsacl.c index 251e5a1bb1c4..0c2be8c0307d 100644 --- a/fs/nfs_common/nfsacl.c +++ b/fs/nfs_common/nfsacl.c @@ -48,43 +48,26 @@ xdr_nfsace_encode(struct xdr_array2_desc *desc, void *elem) (struct nfsacl_encode_desc *) desc; u32 *p = (u32 *) elem; - if (nfsacl_desc->count < nfsacl_desc->acl->a_count) { - struct posix_acl_entry *entry = - &nfsacl_desc->acl->a_entries[nfsacl_desc->count++]; + struct posix_acl_entry *entry = + &nfsacl_desc->acl->a_entries[nfsacl_desc->count++]; - *p++ = htonl(entry->e_tag | nfsacl_desc->typeflag); - switch(entry->e_tag) { - case ACL_USER_OBJ: - *p++ = htonl(nfsacl_desc->uid); - break; - case ACL_GROUP_OBJ: - *p++ = htonl(nfsacl_desc->gid); - break; - case ACL_USER: - case ACL_GROUP: - *p++ = htonl(entry->e_id); - break; - default: /* Solaris depends on that! */ - *p++ = 0; - break; - } - *p++ = htonl(entry->e_perm & S_IRWXO); - } else { - const struct posix_acl_entry *pa, *pe; - int group_obj_perm = ACL_READ|ACL_WRITE|ACL_EXECUTE; - - FOREACH_ACL_ENTRY(pa, nfsacl_desc->acl, pe) { - if (pa->e_tag == ACL_GROUP_OBJ) { - group_obj_perm = pa->e_perm & S_IRWXO; - break; - } - } - /* fake up ACL_MASK entry */ - *p++ = htonl(ACL_MASK | nfsacl_desc->typeflag); - *p++ = htonl(0); - *p++ = htonl(group_obj_perm); + *p++ = htonl(entry->e_tag | nfsacl_desc->typeflag); + switch(entry->e_tag) { + case ACL_USER_OBJ: + *p++ = htonl(nfsacl_desc->uid); + break; + case ACL_GROUP_OBJ: + *p++ = htonl(nfsacl_desc->gid); + break; + case ACL_USER: + case ACL_GROUP: + *p++ = htonl(entry->e_id); + break; + default: /* Solaris depends on that! */ + *p++ = 0; + break; } - + *p++ = htonl(entry->e_perm & S_IRWXO); return 0; } @@ -105,11 +88,28 @@ nfsacl_encode(struct xdr_buf *buf, unsigned int base, struct inode *inode, .gid = inode->i_gid, }; int err; + struct posix_acl *acl2 = NULL; if (entries > NFS_ACL_MAX_ENTRIES || xdr_encode_word(buf, base, entries)) return -EINVAL; + if (encode_entries && acl && acl->a_count == 3) { + /* Fake up an ACL_MASK entry. */ + acl2 = posix_acl_alloc(4, GFP_KERNEL); + if (!acl2) + return -ENOMEM; + /* Insert entries in canonical order: other orders seem + to confuse Solaris VxFS. */ + acl2->a_entries[0] = acl->a_entries[0]; /* ACL_USER_OBJ */ + acl2->a_entries[1] = acl->a_entries[1]; /* ACL_GROUP_OBJ */ + acl2->a_entries[2] = acl->a_entries[1]; /* ACL_MASK */ + acl2->a_entries[2].e_tag = ACL_MASK; + acl2->a_entries[3] = acl->a_entries[2]; /* ACL_OTHER */ + nfsacl_desc.acl = acl2; + } err = xdr_encode_array2(buf, base + 4, &nfsacl_desc.desc); + if (acl2) + posix_acl_release(acl2); if (!err) err = 8 + nfsacl_desc.desc.elem_size * nfsacl_desc.desc.array_len; diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c index 057aff745506..417ec02df44f 100644 --- a/fs/nfsd/export.c +++ b/fs/nfsd/export.c @@ -190,8 +190,7 @@ static int expkey_parse(struct cache_detail *cd, char *mesg, int mlen) out: if (dom) auth_domain_put(dom); - if (buf) - kfree(buf); + kfree(buf); return err; } @@ -428,8 +427,7 @@ static int svc_export_parse(struct cache_detail *cd, char *mesg, int mlen) path_release(&nd); if (dom) auth_domain_put(dom); - if (buf) - kfree(buf); + kfree(buf); return err; } diff --git a/fs/nfsd/nfs3xdr.c b/fs/nfsd/nfs3xdr.c index e0e134d6baba..9147b8524d05 100644 --- a/fs/nfsd/nfs3xdr.c +++ b/fs/nfsd/nfs3xdr.c @@ -366,7 +366,8 @@ nfs3svc_decode_writeargs(struct svc_rqst *rqstp, u32 *p, len = args->len = ntohl(*p++); hdr = (void*)p - rqstp->rq_arg.head[0].iov_base; - if (rqstp->rq_arg.len < len + hdr) + if (rqstp->rq_arg.len < hdr || + rqstp->rq_arg.len - hdr < len) return 0; args->vec[0].iov_base = (void*)p; diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index 4c4146350236..dcd673186944 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -151,8 +151,7 @@ static u32 *read_buf(struct nfsd4_compoundargs *argp, int nbytes) if (nbytes <= sizeof(argp->tmp)) p = argp->tmp; else { - if (argp->tmpp) - kfree(argp->tmpp); + kfree(argp->tmpp); p = argp->tmpp = kmalloc(nbytes, GFP_KERNEL); if (!p) return NULL; @@ -2476,10 +2475,8 @@ void nfsd4_release_compoundargs(struct nfsd4_compoundargs *args) kfree(args->ops); args->ops = args->iops; } - if (args->tmpp) { - kfree(args->tmpp); - args->tmpp = NULL; - } + kfree(args->tmpp); + args->tmpp = NULL; while (args->to_free) { struct tmpbuf *tb = args->to_free; args->to_free = tb->next; diff --git a/fs/nfsd/nfscache.c b/fs/nfsd/nfscache.c index 119e4d4495b8..d852ebb538e3 100644 --- a/fs/nfsd/nfscache.c +++ b/fs/nfsd/nfscache.c @@ -93,8 +93,7 @@ nfsd_cache_shutdown(void) cache_disabled = 1; - if (hash_list) - kfree (hash_list); + kfree (hash_list); hash_list = NULL; } diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index 841c562991e8..a0871b3efeb7 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -23,6 +23,7 @@ #include <linux/seq_file.h> #include <linux/pagemap.h> #include <linux/init.h> +#include <linux/string.h> #include <linux/nfs.h> #include <linux/nfsd_idmap.h> @@ -35,6 +36,8 @@ #include <asm/uaccess.h> +unsigned int nfsd_versbits = ~0; + /* * We have a single directory with 9 nodes in it. */ @@ -50,8 +53,15 @@ enum { NFSD_List, NFSD_Fh, NFSD_Threads, + NFSD_Versions, + /* + * The below MUST come last. Otherwise we leave a hole in nfsd_files[] + * with !CONFIG_NFSD_V4 and simple_fill_super() goes oops + */ +#ifdef CONFIG_NFSD_V4 NFSD_Leasetime, NFSD_RecoveryDir, +#endif }; /* @@ -66,8 +76,11 @@ static ssize_t write_getfd(struct file *file, char *buf, size_t size); static ssize_t write_getfs(struct file *file, char *buf, size_t size); static ssize_t write_filehandle(struct file *file, char *buf, size_t size); static ssize_t write_threads(struct file *file, char *buf, size_t size); +static ssize_t write_versions(struct file *file, char *buf, size_t size); +#ifdef CONFIG_NFSD_V4 static ssize_t write_leasetime(struct file *file, char *buf, size_t size); static ssize_t write_recoverydir(struct file *file, char *buf, size_t size); +#endif static ssize_t (*write_op[])(struct file *, char *, size_t) = { [NFSD_Svc] = write_svc, @@ -79,8 +92,11 @@ static ssize_t (*write_op[])(struct file *, char *, size_t) = { [NFSD_Getfs] = write_getfs, [NFSD_Fh] = write_filehandle, [NFSD_Threads] = write_threads, + [NFSD_Versions] = write_versions, +#ifdef CONFIG_NFSD_V4 [NFSD_Leasetime] = write_leasetime, [NFSD_RecoveryDir] = write_recoverydir, +#endif }; static ssize_t nfsctl_transaction_write(struct file *file, const char __user *buf, size_t size, loff_t *pos) @@ -104,9 +120,23 @@ static ssize_t nfsctl_transaction_write(struct file *file, const char __user *bu return rv; } +static ssize_t nfsctl_transaction_read(struct file *file, char __user *buf, size_t size, loff_t *pos) +{ + if (! file->private_data) { + /* An attempt to read a transaction file without writing + * causes a 0-byte write so that the file can return + * state information + */ + ssize_t rv = nfsctl_transaction_write(file, buf, 0, pos); + if (rv < 0) + return rv; + } + return simple_transaction_read(file, buf, size, pos); +} + static struct file_operations transaction_ops = { .write = nfsctl_transaction_write, - .read = simple_transaction_read, + .read = nfsctl_transaction_read, .release = simple_transaction_release, }; @@ -329,6 +359,70 @@ static ssize_t write_threads(struct file *file, char *buf, size_t size) return strlen(buf); } +static ssize_t write_versions(struct file *file, char *buf, size_t size) +{ + /* + * Format: + * [-/+]vers [-/+]vers ... + */ + char *mesg = buf; + char *vers, sign; + int len, num; + ssize_t tlen = 0; + char *sep; + + if (size>0) { + if (nfsd_serv) + return -EBUSY; + if (buf[size-1] != '\n') + return -EINVAL; + buf[size-1] = 0; + + vers = mesg; + len = qword_get(&mesg, vers, size); + if (len <= 0) return -EINVAL; + do { + sign = *vers; + if (sign == '+' || sign == '-') + num = simple_strtol((vers+1), NULL, 0); + else + num = simple_strtol(vers, NULL, 0); + switch(num) { + case 2: + case 3: + case 4: + if (sign != '-') + NFSCTL_VERSET(nfsd_versbits, num); + else + NFSCTL_VERUNSET(nfsd_versbits, num); + break; + default: + return -EINVAL; + } + vers += len + 1; + tlen += len; + } while ((len = qword_get(&mesg, vers, size)) > 0); + /* If all get turned off, turn them back on, as + * having no versions is BAD + */ + if ((nfsd_versbits & NFSCTL_VERALL)==0) + nfsd_versbits = NFSCTL_VERALL; + } + /* Now write current state into reply buffer */ + len = 0; + sep = ""; + for (num=2 ; num <= 4 ; num++) + if (NFSCTL_VERISSET(NFSCTL_VERALL, num)) { + len += sprintf(buf+len, "%s%c%d", sep, + NFSCTL_VERISSET(nfsd_versbits, num)?'+':'-', + num); + sep = " "; + } + len += sprintf(buf+len, "\n"); + return len; +} + +#ifdef CONFIG_NFSD_V4 extern time_t nfs4_leasetime(void); static ssize_t write_leasetime(struct file *file, char *buf, size_t size) @@ -370,6 +464,7 @@ static ssize_t write_recoverydir(struct file *file, char *buf, size_t size) status = nfs4_reset_recoverydir(recdir); return strlen(buf); } +#endif /*----------------------------------------------------------------------------*/ /* @@ -389,6 +484,7 @@ static int nfsd_fill_super(struct super_block * sb, void * data, int silent) [NFSD_List] = {"exports", &exports_operations, S_IRUGO}, [NFSD_Fh] = {"filehandle", &transaction_ops, S_IWUSR|S_IRUSR}, [NFSD_Threads] = {"threads", &transaction_ops, S_IWUSR|S_IRUSR}, + [NFSD_Versions] = {"versions", &transaction_ops, S_IWUSR|S_IRUSR}, #ifdef CONFIG_NFSD_V4 [NFSD_Leasetime] = {"nfsv4leasetime", &transaction_ops, S_IWUSR|S_IRUSR}, [NFSD_RecoveryDir] = {"nfsv4recoverydir", &transaction_ops, S_IWUSR|S_IRUSR}, diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c index 1697539a7171..89ed04696865 100644 --- a/fs/nfsd/nfssvc.c +++ b/fs/nfsd/nfssvc.c @@ -30,6 +30,7 @@ #include <linux/nfsd/nfsd.h> #include <linux/nfsd/stats.h> #include <linux/nfsd/cache.h> +#include <linux/nfsd/syscall.h> #include <linux/lockd/bind.h> #include <linux/nfsacl.h> @@ -52,7 +53,7 @@ extern struct svc_program nfsd_program; static void nfsd(struct svc_rqst *rqstp); struct timeval nfssvc_boot; -static struct svc_serv *nfsd_serv; + struct svc_serv *nfsd_serv; static atomic_t nfsd_busy; static unsigned long nfsd_last_call; static DEFINE_SPINLOCK(nfsd_call_lock); @@ -63,6 +64,31 @@ struct nfsd_list { }; static struct list_head nfsd_list = LIST_HEAD_INIT(nfsd_list); +static struct svc_version * nfsd_version[] = { + [2] = &nfsd_version2, +#if defined(CONFIG_NFSD_V3) + [3] = &nfsd_version3, +#endif +#if defined(CONFIG_NFSD_V4) + [4] = &nfsd_version4, +#endif +}; + +#define NFSD_MINVERS 2 +#define NFSD_NRVERS (sizeof(nfsd_version)/sizeof(nfsd_version[0])) +static struct svc_version *nfsd_versions[NFSD_NRVERS]; + +struct svc_program nfsd_program = { + .pg_prog = NFS_PROGRAM, /* program number */ + .pg_nvers = NFSD_NRVERS, /* nr of entries in nfsd_version */ + .pg_vers = nfsd_versions, /* version table */ + .pg_name = "nfsd", /* program name */ + .pg_class = "nfsd", /* authentication class */ + .pg_stats = &nfsd_svcstats, /* version table */ + .pg_authenticate = &svc_set_client, /* export authentication */ + +}; + /* * Maximum number of nfsd processes */ @@ -80,11 +106,12 @@ int nfsd_svc(unsigned short port, int nrservs) { int error; - int none_left; + int none_left, found_one, i; struct list_head *victim; lock_kernel(); - dprintk("nfsd: creating service\n"); + dprintk("nfsd: creating service: vers 0x%x\n", + nfsd_versbits); error = -EINVAL; if (nrservs <= 0) nrservs = 0; @@ -99,6 +126,27 @@ nfsd_svc(unsigned short port, int nrservs) if (error<0) goto out; if (!nfsd_serv) { + /* + * Use the nfsd_ctlbits to define which + * versions that will be advertised. + * If nfsd_ctlbits doesn't list any version, + * export them all. + */ + found_one = 0; + + for (i = NFSD_MINVERS; i < NFSD_NRVERS; i++) { + if (NFSCTL_VERISSET(nfsd_versbits, i)) { + nfsd_program.pg_vers[i] = nfsd_version[i]; + found_one = 1; + } else + nfsd_program.pg_vers[i] = NULL; + } + + if (!found_one) { + for (i = NFSD_MINVERS; i < NFSD_NRVERS; i++) + nfsd_program.pg_vers[i] = nfsd_version[i]; + } + atomic_set(&nfsd_busy, 0); error = -ENOMEM; nfsd_serv = svc_create(&nfsd_program, NFSD_BUFSIZE); @@ -379,6 +427,7 @@ static struct svc_program nfsd_acl_program = { .pg_name = "nfsd", .pg_class = "nfsd", .pg_stats = &nfsd_acl_svcstats, + .pg_authenticate = &svc_set_client, }; static struct svc_stat nfsd_acl_svcstats = { @@ -389,28 +438,3 @@ static struct svc_stat nfsd_acl_svcstats = { #else #define nfsd_acl_program_p NULL #endif /* defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL) */ - -extern struct svc_version nfsd_version2, nfsd_version3, nfsd_version4; - -static struct svc_version * nfsd_version[] = { - [2] = &nfsd_version2, -#if defined(CONFIG_NFSD_V3) - [3] = &nfsd_version3, -#endif -#if defined(CONFIG_NFSD_V4) - [4] = &nfsd_version4, -#endif -}; - -#define NFSD_NRVERS (sizeof(nfsd_version)/sizeof(nfsd_version[0])) -struct svc_program nfsd_program = { - .pg_next = nfsd_acl_program_p, - .pg_prog = NFS_PROGRAM, /* program number */ - .pg_nvers = NFSD_NRVERS, /* nr of entries in nfsd_version */ - .pg_vers = nfsd_version, /* version table */ - .pg_name = "nfsd", /* program name */ - .pg_class = "nfsd", /* authentication class */ - .pg_stats = &nfsd_svcstats, /* version table */ - .pg_authenticate = &svc_set_client, /* export authentication */ - -}; diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 4f2cd3d27566..af7c3c3074b0 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -254,12 +254,19 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap, /* Get inode */ err = fh_verify(rqstp, fhp, ftype, accmode); - if (err || !iap->ia_valid) + if (err) goto out; dentry = fhp->fh_dentry; inode = dentry->d_inode; + /* Ignore any mode updates on symlinks */ + if (S_ISLNK(inode->i_mode)) + iap->ia_valid &= ~ATTR_MODE; + + if (!iap->ia_valid) + goto out; + /* NFSv2 does not differentiate between "set-[ac]time-to-now" * which only requires access, and "set-[ac]time-to-X" which * requires ownership. diff --git a/fs/ntfs/ChangeLog b/fs/ntfs/ChangeLog index c7e9237379c2..50a7749cfca1 100644 --- a/fs/ntfs/ChangeLog +++ b/fs/ntfs/ChangeLog @@ -1,18 +1,15 @@ ToDo/Notes: - Find and fix bugs. - - In between ntfs_prepare/commit_write, need exclusion between - simultaneous file extensions. This is given to us by holding i_sem - on the inode. The only places in the kernel when a file is resized - are prepare/commit write and truncate for both of which i_sem is - held. Just have to be careful in readpage/writepage and all other - helpers not running under i_sem that we play nice... - Also need to be careful with initialized_size extention in - ntfs_prepare_write. Basically, just be _very_ careful in this code... - UPDATE: The only things that need to be checked are read/writepage - which do not hold i_sem. Note writepage cannot change i_size but it - needs to cope with a concurrent i_size change, just like readpage. - Also both need to cope with concurrent changes to the other sizes, - i.e. initialized/allocated/compressed size, as well. + - The only places in the kernel where a file is resized are + ntfs_file_write*() and ntfs_truncate() for both of which i_sem is + held. Just have to be careful in read-/writepage and other helpers + not running under i_sem that we play nice... Also need to be careful + with initialized_size extension in ntfs_file_write*() and writepage. + UPDATE: The only things that need to be checked are the compressed + write and the other attribute resize/write cases like index + attributes, etc. For now none of these are implemented so are safe. + - Implement filling in of holes in aops.c::ntfs_writepage() and its + helpers. - Implement mft.c::sync_mft_mirror_umount(). We currently will just leave the volume dirty on umount if the final iput(vol->mft_ino) causes a write of any mirrored mft records due to the mft mirror @@ -22,6 +19,68 @@ ToDo/Notes: - Enable the code for setting the NT4 compatibility flag when we start making NTFS 1.2 specific modifications. +2.1.25 - (Almost) fully implement write(2) and truncate(2). + + - Change ntfs_map_runlist_nolock(), ntfs_attr_find_vcn_nolock() and + {__,}ntfs_cluster_free() to also take an optional attribute search + context as argument. This allows calling these functions with the + mft record mapped. Update all callers. + - Fix potential deadlock in ntfs_mft_data_extend_allocation_nolock() + error handling by passing in the active search context when calling + ntfs_cluster_free(). + - Change ntfs_cluster_alloc() to take an extra boolean parameter + specifying whether the cluster are being allocated to extend an + attribute or to fill a hole. + - Change ntfs_attr_make_non_resident() to call ntfs_cluster_alloc() + with @is_extension set to TRUE and remove the runlist terminator + fixup code as this is now done by ntfs_cluster_alloc(). + - Change ntfs_attr_make_non_resident to take the attribute value size + as an extra parameter. This is needed since we need to know the size + before we can map the mft record and our callers always know it. The + reason we cannot simply read the size from the vfs inode i_size is + that this is not necessarily uptodate. This happens when + ntfs_attr_make_non_resident() is called in the ->truncate call path. + - Fix ntfs_attr_make_non_resident() to update the vfs inode i_blocks + which is zero for a resident attribute but should no longer be zero + once the attribute is non-resident as it then has real clusters + allocated. + - Add fs/ntfs/attrib.[hc]::ntfs_attr_extend_allocation(), a function to + extend the allocation of an attributes. Optionally, the data size, + but not the initialized size can be extended, too. + - Implement fs/ntfs/inode.[hc]::ntfs_truncate(). It only supports + uncompressed and unencrypted files and it never creates sparse files + at least for the moment (making a file sparse requires us to modify + its directory entries and we do not support directory operations at + the moment). Also, support for highly fragmented files, i.e. ones + whose data attribute is split across multiple extents, is severly + limited. When such a case is encountered, EOPNOTSUPP is returned. + - Enable ATTR_SIZE attribute changes in ntfs_setattr(). This completes + the initial implementation of file truncation. Now both open(2)ing + a file with the O_TRUNC flag and the {,f}truncate(2) system calls + will resize a file appropriately. The limitations are that only + uncompressed and unencrypted files are supported. Also, there is + only very limited support for highly fragmented files (the ones whose + $DATA attribute is split into multiple attribute extents). + - In attrib.c::ntfs_attr_set() call balance_dirty_pages_ratelimited() + and cond_resched() in the main loop as we could be dirtying a lot of + pages and this ensures we play nice with the VM and the system as a + whole. + - Implement file operations ->write, ->aio_write, ->writev for regular + files. This replaces the old use of generic_file_write(), et al and + the address space operations ->prepare_write and ->commit_write. + This means that both sparse and non-sparse (unencrypted and + uncompressed) files can now be extended using the normal write(2) + code path. There are two limitations at present and these are that + we never create sparse files and that we only have limited support + for highly fragmented files, i.e. ones whose data attribute is split + across multiple extents. When such a case is encountered, + EOPNOTSUPP is returned. + - $EA attributes can be both resident and non-resident. + - Use %z for size_t to fix compilation warnings. (Andrew Morton) + - Fix compilation warnings with gcc-4.0.2 on SUSE 10.0. + - Document extended attribute ($EA) NEED_EA flag. (Based on libntfs + patch by Yura Pakhuchiy.) + 2.1.24 - Lots of bug fixes and support more clean journal states. - Support journals ($LogFile) which have been modified by chkdsk. This @@ -29,7 +88,8 @@ ToDo/Notes: The Windows boot will run chkdsk and then reboot. The user can then immediately boot into Linux rather than having to do a full Windows boot first before rebooting into Linux and we will recognize such a - journal and empty it as it is clean by definition. + journal and empty it as it is clean by definition. Note, this only + works if chkdsk left the journal in an obviously clean state. - Support journals ($LogFile) with only one restart page as well as journals with two different restart pages. We sanity check both and either use the only sane one or the more recent one of the two in the @@ -94,6 +154,16 @@ ToDo/Notes: my ways. - Fix various bugs in the runlist merging code. (Based on libntfs changes by Richard Russon.) + - Fix sparse warnings that have crept in over time. + - Change ntfs_cluster_free() to require a write locked runlist on entry + since we otherwise get into a lock reversal deadlock if a read locked + runlist is passed in. In the process also change it to take an ntfs + inode instead of a vfs inode as parameter. + - Fix the definition of the CHKD ntfs record magic. It had an off by + two error causing it to be CHKB instead of CHKD. + - Fix a stupid bug in __ntfs_bitmap_set_bits_in_run() which caused the + count to become negative and hence we had a wild memset() scribbling + all over the system's ram. 2.1.23 - Implement extension of resident files and make writing safe as well as many bug fixes, cleanups, and enhancements... diff --git a/fs/ntfs/Makefile b/fs/ntfs/Makefile index 894b2b876d35..d0d45d1c853a 100644 --- a/fs/ntfs/Makefile +++ b/fs/ntfs/Makefile @@ -6,7 +6,7 @@ ntfs-objs := aops.o attrib.o collate.o compress.o debug.o dir.o file.o \ index.o inode.o mft.o mst.o namei.o runlist.o super.o sysctl.o \ unistr.o upcase.o -EXTRA_CFLAGS = -DNTFS_VERSION=\"2.1.24\" +EXTRA_CFLAGS = -DNTFS_VERSION=\"2.1.25\" ifeq ($(CONFIG_NTFS_DEBUG),y) EXTRA_CFLAGS += -DDEBUG diff --git a/fs/ntfs/aops.c b/fs/ntfs/aops.c index 5e80c07c6a4d..1c0a4315876a 100644 --- a/fs/ntfs/aops.c +++ b/fs/ntfs/aops.c @@ -1391,8 +1391,7 @@ retry_writepage: if (NInoEncrypted(ni)) { unlock_page(page); BUG_ON(ni->type != AT_DATA); - ntfs_debug("Denying write access to encrypted " - "file."); + ntfs_debug("Denying write access to encrypted file."); return -EACCES; } /* Compressed data streams are handled in compress.c. */ @@ -1508,8 +1507,8 @@ retry_writepage: /* Zero out of bounds area in the page cache page. */ memset(kaddr + attr_len, 0, PAGE_CACHE_SIZE - attr_len); kunmap_atomic(kaddr, KM_USER0); - flush_dcache_mft_record_page(ctx->ntfs_ino); flush_dcache_page(page); + flush_dcache_mft_record_page(ctx->ntfs_ino); /* We are done with the page. */ end_page_writeback(page); /* Finally, mark the mft record dirty, so it gets written back. */ @@ -1542,830 +1541,6 @@ err_out: return err; } -/** - * ntfs_prepare_nonresident_write - - * - */ -static int ntfs_prepare_nonresident_write(struct page *page, - unsigned from, unsigned to) -{ - VCN vcn; - LCN lcn; - s64 initialized_size; - loff_t i_size; - sector_t block, ablock, iblock; - struct inode *vi; - ntfs_inode *ni; - ntfs_volume *vol; - runlist_element *rl; - struct buffer_head *bh, *head, *wait[2], **wait_bh = wait; - unsigned long flags; - unsigned int vcn_ofs, block_start, block_end, blocksize; - int err; - BOOL is_retry; - unsigned char blocksize_bits; - - vi = page->mapping->host; - ni = NTFS_I(vi); - vol = ni->vol; - - ntfs_debug("Entering for inode 0x%lx, attribute type 0x%x, page index " - "0x%lx, from = %u, to = %u.", ni->mft_no, ni->type, - page->index, from, to); - - BUG_ON(!NInoNonResident(ni)); - - blocksize_bits = vi->i_blkbits; - blocksize = 1 << blocksize_bits; - - /* - * create_empty_buffers() will create uptodate/dirty buffers if the - * page is uptodate/dirty. - */ - if (!page_has_buffers(page)) - create_empty_buffers(page, blocksize, 0); - bh = head = page_buffers(page); - if (unlikely(!bh)) - return -ENOMEM; - - /* The first block in the page. */ - block = (s64)page->index << (PAGE_CACHE_SHIFT - blocksize_bits); - - read_lock_irqsave(&ni->size_lock, flags); - /* - * The first out of bounds block for the allocated size. No need to - * round up as allocated_size is in multiples of cluster size and the - * minimum cluster size is 512 bytes, which is equal to the smallest - * blocksize. - */ - ablock = ni->allocated_size >> blocksize_bits; - i_size = i_size_read(vi); - initialized_size = ni->initialized_size; - read_unlock_irqrestore(&ni->size_lock, flags); - - /* The last (fully or partially) initialized block. */ - iblock = initialized_size >> blocksize_bits; - - /* Loop through all the buffers in the page. */ - block_start = 0; - rl = NULL; - err = 0; - do { - block_end = block_start + blocksize; - /* - * If buffer @bh is outside the write, just mark it uptodate - * if the page is uptodate and continue with the next buffer. - */ - if (block_end <= from || block_start >= to) { - if (PageUptodate(page)) { - if (!buffer_uptodate(bh)) - set_buffer_uptodate(bh); - } - continue; - } - /* - * @bh is at least partially being written to. - * Make sure it is not marked as new. - */ - //if (buffer_new(bh)) - // clear_buffer_new(bh); - - if (block >= ablock) { - // TODO: block is above allocated_size, need to - // allocate it. Best done in one go to accommodate not - // only block but all above blocks up to and including: - // ((page->index << PAGE_CACHE_SHIFT) + to + blocksize - // - 1) >> blobksize_bits. Obviously will need to round - // up to next cluster boundary, too. This should be - // done with a helper function, so it can be reused. - ntfs_error(vol->sb, "Writing beyond allocated size " - "is not supported yet. Sorry."); - err = -EOPNOTSUPP; - goto err_out; - // Need to update ablock. - // Need to set_buffer_new() on all block bhs that are - // newly allocated. - } - /* - * Now we have enough allocated size to fulfill the whole - * request, i.e. block < ablock is true. - */ - if (unlikely((block >= iblock) && - (initialized_size < i_size))) { - /* - * If this page is fully outside initialized size, zero - * out all pages between the current initialized size - * and the current page. Just use ntfs_readpage() to do - * the zeroing transparently. - */ - if (block > iblock) { - // TODO: - // For each page do: - // - read_cache_page() - // Again for each page do: - // - wait_on_page_locked() - // - Check (PageUptodate(page) && - // !PageError(page)) - // Update initialized size in the attribute and - // in the inode. - // Again, for each page do: - // __set_page_dirty_buffers(); - // page_cache_release() - // We don't need to wait on the writes. - // Update iblock. - } - /* - * The current page straddles initialized size. Zero - * all non-uptodate buffers and set them uptodate (and - * dirty?). Note, there aren't any non-uptodate buffers - * if the page is uptodate. - * FIXME: For an uptodate page, the buffers may need to - * be written out because they were not initialized on - * disk before. - */ - if (!PageUptodate(page)) { - // TODO: - // Zero any non-uptodate buffers up to i_size. - // Set them uptodate and dirty. - } - // TODO: - // Update initialized size in the attribute and in the - // inode (up to i_size). - // Update iblock. - // FIXME: This is inefficient. Try to batch the two - // size changes to happen in one go. - ntfs_error(vol->sb, "Writing beyond initialized size " - "is not supported yet. Sorry."); - err = -EOPNOTSUPP; - goto err_out; - // Do NOT set_buffer_new() BUT DO clear buffer range - // outside write request range. - // set_buffer_uptodate() on complete buffers as well as - // set_buffer_dirty(). - } - - /* Need to map unmapped buffers. */ - if (!buffer_mapped(bh)) { - /* Unmapped buffer. Need to map it. */ - bh->b_bdev = vol->sb->s_bdev; - - /* Convert block into corresponding vcn and offset. */ - vcn = (VCN)block << blocksize_bits >> - vol->cluster_size_bits; - vcn_ofs = ((VCN)block << blocksize_bits) & - vol->cluster_size_mask; - - is_retry = FALSE; - if (!rl) { -lock_retry_remap: - down_read(&ni->runlist.lock); - rl = ni->runlist.rl; - } - if (likely(rl != NULL)) { - /* Seek to element containing target vcn. */ - while (rl->length && rl[1].vcn <= vcn) - rl++; - lcn = ntfs_rl_vcn_to_lcn(rl, vcn); - } else - lcn = LCN_RL_NOT_MAPPED; - if (unlikely(lcn < 0)) { - /* - * We extended the attribute allocation above. - * If we hit an ENOENT here it means that the - * allocation was insufficient which is a bug. - */ - BUG_ON(lcn == LCN_ENOENT); - - /* It is a hole, need to instantiate it. */ - if (lcn == LCN_HOLE) { - // TODO: Instantiate the hole. - // clear_buffer_new(bh); - // unmap_underlying_metadata(bh->b_bdev, - // bh->b_blocknr); - // For non-uptodate buffers, need to - // zero out the region outside the - // request in this bh or all bhs, - // depending on what we implemented - // above. - // Need to flush_dcache_page(). - // Or could use set_buffer_new() - // instead? - ntfs_error(vol->sb, "Writing into " - "sparse regions is " - "not supported yet. " - "Sorry."); - err = -EOPNOTSUPP; - if (!rl) - up_read(&ni->runlist.lock); - goto err_out; - } else if (!is_retry && - lcn == LCN_RL_NOT_MAPPED) { - is_retry = TRUE; - /* - * Attempt to map runlist, dropping - * lock for the duration. - */ - up_read(&ni->runlist.lock); - err = ntfs_map_runlist(ni, vcn); - if (likely(!err)) - goto lock_retry_remap; - rl = NULL; - } else if (!rl) - up_read(&ni->runlist.lock); - /* - * Failed to map the buffer, even after - * retrying. - */ - if (!err) - err = -EIO; - bh->b_blocknr = -1; - ntfs_error(vol->sb, "Failed to write to inode " - "0x%lx, attribute type 0x%x, " - "vcn 0x%llx, offset 0x%x " - "because its location on disk " - "could not be determined%s " - "(error code %i).", - ni->mft_no, ni->type, - (unsigned long long)vcn, - vcn_ofs, is_retry ? " even " - "after retrying" : "", err); - goto err_out; - } - /* We now have a successful remap, i.e. lcn >= 0. */ - - /* Setup buffer head to correct block. */ - bh->b_blocknr = ((lcn << vol->cluster_size_bits) - + vcn_ofs) >> blocksize_bits; - set_buffer_mapped(bh); - - // FIXME: Something analogous to this is needed for - // each newly allocated block, i.e. BH_New. - // FIXME: Might need to take this out of the - // if (!buffer_mapped(bh)) {}, depending on how we - // implement things during the allocated_size and - // initialized_size extension code above. - if (buffer_new(bh)) { - clear_buffer_new(bh); - unmap_underlying_metadata(bh->b_bdev, - bh->b_blocknr); - if (PageUptodate(page)) { - set_buffer_uptodate(bh); - continue; - } - /* - * Page is _not_ uptodate, zero surrounding - * region. NOTE: This is how we decide if to - * zero or not! - */ - if (block_end > to || block_start < from) { - void *kaddr; - - kaddr = kmap_atomic(page, KM_USER0); - if (block_end > to) - memset(kaddr + to, 0, - block_end - to); - if (block_start < from) - memset(kaddr + block_start, 0, - from - - block_start); - flush_dcache_page(page); - kunmap_atomic(kaddr, KM_USER0); - } - continue; - } - } - /* @bh is mapped, set it uptodate if the page is uptodate. */ - if (PageUptodate(page)) { - if (!buffer_uptodate(bh)) - set_buffer_uptodate(bh); - continue; - } - /* - * The page is not uptodate. The buffer is mapped. If it is not - * uptodate, and it is only partially being written to, we need - * to read the buffer in before the write, i.e. right now. - */ - if (!buffer_uptodate(bh) && - (block_start < from || block_end > to)) { - ll_rw_block(READ, 1, &bh); - *wait_bh++ = bh; - } - } while (block++, block_start = block_end, - (bh = bh->b_this_page) != head); - - /* Release the lock if we took it. */ - if (rl) { - up_read(&ni->runlist.lock); - rl = NULL; - } - - /* If we issued read requests, let them complete. */ - while (wait_bh > wait) { - wait_on_buffer(*--wait_bh); - if (!buffer_uptodate(*wait_bh)) - return -EIO; - } - - ntfs_debug("Done."); - return 0; -err_out: - /* - * Zero out any newly allocated blocks to avoid exposing stale data. - * If BH_New is set, we know that the block was newly allocated in the - * above loop. - * FIXME: What about initialized_size increments? Have we done all the - * required zeroing above? If not this error handling is broken, and - * in particular the if (block_end <= from) check is completely bogus. - */ - bh = head; - block_start = 0; - is_retry = FALSE; - do { - block_end = block_start + blocksize; - if (block_end <= from) - continue; - if (block_start >= to) - break; - if (buffer_new(bh)) { - void *kaddr; - - clear_buffer_new(bh); - kaddr = kmap_atomic(page, KM_USER0); - memset(kaddr + block_start, 0, bh->b_size); - kunmap_atomic(kaddr, KM_USER0); - set_buffer_uptodate(bh); - mark_buffer_dirty(bh); - is_retry = TRUE; - } - } while (block_start = block_end, (bh = bh->b_this_page) != head); - if (is_retry) - flush_dcache_page(page); - if (rl) - up_read(&ni->runlist.lock); - return err; -} - -/** - * ntfs_prepare_write - prepare a page for receiving data - * - * This is called from generic_file_write() with i_sem held on the inode - * (@page->mapping->host). The @page is locked but not kmap()ped. The source - * data has not yet been copied into the @page. - * - * Need to extend the attribute/fill in holes if necessary, create blocks and - * make partially overwritten blocks uptodate, - * - * i_size is not to be modified yet. - * - * Return 0 on success or -errno on error. - * - * Should be using block_prepare_write() [support for sparse files] or - * cont_prepare_write() [no support for sparse files]. Cannot do that due to - * ntfs specifics but can look at them for implementation guidance. - * - * Note: In the range, @from is inclusive and @to is exclusive, i.e. @from is - * the first byte in the page that will be written to and @to is the first byte - * after the last byte that will be written to. - */ -static int ntfs_prepare_write(struct file *file, struct page *page, - unsigned from, unsigned to) -{ - s64 new_size; - loff_t i_size; - struct inode *vi = page->mapping->host; - ntfs_inode *base_ni = NULL, *ni = NTFS_I(vi); - ntfs_volume *vol = ni->vol; - ntfs_attr_search_ctx *ctx = NULL; - MFT_RECORD *m = NULL; - ATTR_RECORD *a; - u8 *kaddr; - u32 attr_len; - int err; - - ntfs_debug("Entering for inode 0x%lx, attribute type 0x%x, page index " - "0x%lx, from = %u, to = %u.", vi->i_ino, ni->type, - page->index, from, to); - BUG_ON(!PageLocked(page)); - BUG_ON(from > PAGE_CACHE_SIZE); - BUG_ON(to > PAGE_CACHE_SIZE); - BUG_ON(from > to); - BUG_ON(NInoMstProtected(ni)); - /* - * If a previous ntfs_truncate() failed, repeat it and abort if it - * fails again. - */ - if (unlikely(NInoTruncateFailed(ni))) { - down_write(&vi->i_alloc_sem); - err = ntfs_truncate(vi); - up_write(&vi->i_alloc_sem); - if (err || NInoTruncateFailed(ni)) { - if (!err) - err = -EIO; - goto err_out; - } - } - /* If the attribute is not resident, deal with it elsewhere. */ - if (NInoNonResident(ni)) { - /* - * Only unnamed $DATA attributes can be compressed, encrypted, - * and/or sparse. - */ - if (ni->type == AT_DATA && !ni->name_len) { - /* If file is encrypted, deny access, just like NT4. */ - if (NInoEncrypted(ni)) { - ntfs_debug("Denying write access to encrypted " - "file."); - return -EACCES; - } - /* Compressed data streams are handled in compress.c. */ - if (NInoCompressed(ni)) { - // TODO: Implement and replace this check with - // return ntfs_write_compressed_block(page); - ntfs_error(vi->i_sb, "Writing to compressed " - "files is not supported yet. " - "Sorry."); - return -EOPNOTSUPP; - } - // TODO: Implement and remove this check. - if (NInoSparse(ni)) { - ntfs_error(vi->i_sb, "Writing to sparse files " - "is not supported yet. Sorry."); - return -EOPNOTSUPP; - } - } - /* Normal data stream. */ - return ntfs_prepare_nonresident_write(page, from, to); - } - /* - * Attribute is resident, implying it is not compressed, encrypted, or - * sparse. - */ - BUG_ON(page_has_buffers(page)); - new_size = ((s64)page->index << PAGE_CACHE_SHIFT) + to; - /* If we do not need to resize the attribute allocation we are done. */ - if (new_size <= i_size_read(vi)) - goto done; - /* Map, pin, and lock the (base) mft record. */ - if (!NInoAttr(ni)) - base_ni = ni; - else - base_ni = ni->ext.base_ntfs_ino; - m = map_mft_record(base_ni); - if (IS_ERR(m)) { - err = PTR_ERR(m); - m = NULL; - ctx = NULL; - goto err_out; - } - ctx = ntfs_attr_get_search_ctx(base_ni, m); - if (unlikely(!ctx)) { - err = -ENOMEM; - goto err_out; - } - err = ntfs_attr_lookup(ni->type, ni->name, ni->name_len, - CASE_SENSITIVE, 0, NULL, 0, ctx); - if (unlikely(err)) { - if (err == -ENOENT) - err = -EIO; - goto err_out; - } - m = ctx->mrec; - a = ctx->attr; - /* The total length of the attribute value. */ - attr_len = le32_to_cpu(a->data.resident.value_length); - /* Fix an eventual previous failure of ntfs_commit_write(). */ - i_size = i_size_read(vi); - if (unlikely(attr_len > i_size)) { - attr_len = i_size; - a->data.resident.value_length = cpu_to_le32(attr_len); - } - /* If we do not need to resize the attribute allocation we are done. */ - if (new_size <= attr_len) - goto done_unm; - /* Check if new size is allowed in $AttrDef. */ - err = ntfs_attr_size_bounds_check(vol, ni->type, new_size); - if (unlikely(err)) { - if (err == -ERANGE) { - ntfs_error(vol->sb, "Write would cause the inode " - "0x%lx to exceed the maximum size for " - "its attribute type (0x%x). Aborting " - "write.", vi->i_ino, - le32_to_cpu(ni->type)); - } else { - ntfs_error(vol->sb, "Inode 0x%lx has unknown " - "attribute type 0x%x. Aborting " - "write.", vi->i_ino, - le32_to_cpu(ni->type)); - err = -EIO; - } - goto err_out2; - } - /* - * Extend the attribute record to be able to store the new attribute - * size. - */ - if (new_size >= vol->mft_record_size || ntfs_attr_record_resize(m, a, - le16_to_cpu(a->data.resident.value_offset) + - new_size)) { - /* Not enough space in the mft record. */ - ntfs_error(vol->sb, "Not enough space in the mft record for " - "the resized attribute value. This is not " - "supported yet. Aborting write."); - err = -EOPNOTSUPP; - goto err_out2; - } - /* - * We have enough space in the mft record to fit the write. This - * implies the attribute is smaller than the mft record and hence the - * attribute must be in a single page and hence page->index must be 0. - */ - BUG_ON(page->index); - /* - * If the beginning of the write is past the old size, enlarge the - * attribute value up to the beginning of the write and fill it with - * zeroes. - */ - if (from > attr_len) { - memset((u8*)a + le16_to_cpu(a->data.resident.value_offset) + - attr_len, 0, from - attr_len); - a->data.resident.value_length = cpu_to_le32(from); - /* Zero the corresponding area in the page as well. */ - if (PageUptodate(page)) { - kaddr = kmap_atomic(page, KM_USER0); - memset(kaddr + attr_len, 0, from - attr_len); - kunmap_atomic(kaddr, KM_USER0); - flush_dcache_page(page); - } - } - flush_dcache_mft_record_page(ctx->ntfs_ino); - mark_mft_record_dirty(ctx->ntfs_ino); -done_unm: - ntfs_attr_put_search_ctx(ctx); - unmap_mft_record(base_ni); - /* - * Because resident attributes are handled by memcpy() to/from the - * corresponding MFT record, and because this form of i/o is byte - * aligned rather than block aligned, there is no need to bring the - * page uptodate here as in the non-resident case where we need to - * bring the buffers straddled by the write uptodate before - * generic_file_write() does the copying from userspace. - * - * We thus defer the uptodate bringing of the page region outside the - * region written to to ntfs_commit_write(), which makes the code - * simpler and saves one atomic kmap which is good. - */ -done: - ntfs_debug("Done."); - return 0; -err_out: - if (err == -ENOMEM) - ntfs_warning(vi->i_sb, "Error allocating memory required to " - "prepare the write."); - else { - ntfs_error(vi->i_sb, "Resident attribute prepare write failed " - "with error %i.", err); - NVolSetErrors(vol); - make_bad_inode(vi); - } -err_out2: - if (ctx) - ntfs_attr_put_search_ctx(ctx); - if (m) - unmap_mft_record(base_ni); - return err; -} - -/** - * ntfs_commit_nonresident_write - - * - */ -static int ntfs_commit_nonresident_write(struct page *page, - unsigned from, unsigned to) -{ - s64 pos = ((s64)page->index << PAGE_CACHE_SHIFT) + to; - struct inode *vi = page->mapping->host; - struct buffer_head *bh, *head; - unsigned int block_start, block_end, blocksize; - BOOL partial; - - ntfs_debug("Entering for inode 0x%lx, attribute type 0x%x, page index " - "0x%lx, from = %u, to = %u.", vi->i_ino, - NTFS_I(vi)->type, page->index, from, to); - blocksize = 1 << vi->i_blkbits; - - // FIXME: We need a whole slew of special cases in here for compressed - // files for example... - // For now, we know ntfs_prepare_write() would have failed so we can't - // get here in any of the cases which we have to special case, so we - // are just a ripped off, unrolled generic_commit_write(). - - bh = head = page_buffers(page); - block_start = 0; - partial = FALSE; - do { - block_end = block_start + blocksize; - if (block_end <= from || block_start >= to) { - if (!buffer_uptodate(bh)) - partial = TRUE; - } else { - set_buffer_uptodate(bh); - mark_buffer_dirty(bh); - } - } while (block_start = block_end, (bh = bh->b_this_page) != head); - /* - * If this is a partial write which happened to make all buffers - * uptodate then we can optimize away a bogus ->readpage() for the next - * read(). Here we 'discover' whether the page went uptodate as a - * result of this (potentially partial) write. - */ - if (!partial) - SetPageUptodate(page); - /* - * Not convinced about this at all. See disparity comment above. For - * now we know ntfs_prepare_write() would have failed in the write - * exceeds i_size case, so this will never trigger which is fine. - */ - if (pos > i_size_read(vi)) { - ntfs_error(vi->i_sb, "Writing beyond the existing file size is " - "not supported yet. Sorry."); - return -EOPNOTSUPP; - // vi->i_size = pos; - // mark_inode_dirty(vi); - } - ntfs_debug("Done."); - return 0; -} - -/** - * ntfs_commit_write - commit the received data - * - * This is called from generic_file_write() with i_sem held on the inode - * (@page->mapping->host). The @page is locked but not kmap()ped. The source - * data has already been copied into the @page. ntfs_prepare_write() has been - * called before the data copied and it returned success so we can take the - * results of various BUG checks and some error handling for granted. - * - * Need to mark modified blocks dirty so they get written out later when - * ntfs_writepage() is invoked by the VM. - * - * Return 0 on success or -errno on error. - * - * Should be using generic_commit_write(). This marks buffers uptodate and - * dirty, sets the page uptodate if all buffers in the page are uptodate, and - * updates i_size if the end of io is beyond i_size. In that case, it also - * marks the inode dirty. - * - * Cannot use generic_commit_write() due to ntfs specialities but can look at - * it for implementation guidance. - * - * If things have gone as outlined in ntfs_prepare_write(), then we do not - * need to do any page content modifications here at all, except in the write - * to resident attribute case, where we need to do the uptodate bringing here - * which we combine with the copying into the mft record which means we save - * one atomic kmap. - */ -static int ntfs_commit_write(struct file *file, struct page *page, - unsigned from, unsigned to) -{ - struct inode *vi = page->mapping->host; - ntfs_inode *base_ni, *ni = NTFS_I(vi); - char *kaddr, *kattr; - ntfs_attr_search_ctx *ctx; - MFT_RECORD *m; - ATTR_RECORD *a; - u32 attr_len; - int err; - - ntfs_debug("Entering for inode 0x%lx, attribute type 0x%x, page index " - "0x%lx, from = %u, to = %u.", vi->i_ino, ni->type, - page->index, from, to); - /* If the attribute is not resident, deal with it elsewhere. */ - if (NInoNonResident(ni)) { - /* Only unnamed $DATA attributes can be compressed/encrypted. */ - if (ni->type == AT_DATA && !ni->name_len) { - /* Encrypted files need separate handling. */ - if (NInoEncrypted(ni)) { - // We never get here at present! - BUG(); - } - /* Compressed data streams are handled in compress.c. */ - if (NInoCompressed(ni)) { - // TODO: Implement this! - // return ntfs_write_compressed_block(page); - // We never get here at present! - BUG(); - } - } - /* Normal data stream. */ - return ntfs_commit_nonresident_write(page, from, to); - } - /* - * Attribute is resident, implying it is not compressed, encrypted, or - * sparse. - */ - if (!NInoAttr(ni)) - base_ni = ni; - else - base_ni = ni->ext.base_ntfs_ino; - /* Map, pin, and lock the mft record. */ - m = map_mft_record(base_ni); - if (IS_ERR(m)) { - err = PTR_ERR(m); - m = NULL; - ctx = NULL; - goto err_out; - } - ctx = ntfs_attr_get_search_ctx(base_ni, m); - if (unlikely(!ctx)) { - err = -ENOMEM; - goto err_out; - } - err = ntfs_attr_lookup(ni->type, ni->name, ni->name_len, - CASE_SENSITIVE, 0, NULL, 0, ctx); - if (unlikely(err)) { - if (err == -ENOENT) - err = -EIO; - goto err_out; - } - a = ctx->attr; - /* The total length of the attribute value. */ - attr_len = le32_to_cpu(a->data.resident.value_length); - BUG_ON(from > attr_len); - kattr = (u8*)a + le16_to_cpu(a->data.resident.value_offset); - kaddr = kmap_atomic(page, KM_USER0); - /* Copy the received data from the page to the mft record. */ - memcpy(kattr + from, kaddr + from, to - from); - /* Update the attribute length if necessary. */ - if (to > attr_len) { - attr_len = to; - a->data.resident.value_length = cpu_to_le32(attr_len); - } - /* - * If the page is not uptodate, bring the out of bounds area(s) - * uptodate by copying data from the mft record to the page. - */ - if (!PageUptodate(page)) { - if (from > 0) - memcpy(kaddr, kattr, from); - if (to < attr_len) - memcpy(kaddr + to, kattr + to, attr_len - to); - /* Zero the region outside the end of the attribute value. */ - if (attr_len < PAGE_CACHE_SIZE) - memset(kaddr + attr_len, 0, PAGE_CACHE_SIZE - attr_len); - /* - * The probability of not having done any of the above is - * extremely small, so we just flush unconditionally. - */ - flush_dcache_page(page); - SetPageUptodate(page); - } - kunmap_atomic(kaddr, KM_USER0); - /* Update i_size if necessary. */ - if (i_size_read(vi) < attr_len) { - unsigned long flags; - - write_lock_irqsave(&ni->size_lock, flags); - ni->allocated_size = ni->initialized_size = attr_len; - i_size_write(vi, attr_len); - write_unlock_irqrestore(&ni->size_lock, flags); - } - /* Mark the mft record dirty, so it gets written back. */ - flush_dcache_mft_record_page(ctx->ntfs_ino); - mark_mft_record_dirty(ctx->ntfs_ino); - ntfs_attr_put_search_ctx(ctx); - unmap_mft_record(base_ni); - ntfs_debug("Done."); - return 0; -err_out: - if (err == -ENOMEM) { - ntfs_warning(vi->i_sb, "Error allocating memory required to " - "commit the write."); - if (PageUptodate(page)) { - ntfs_warning(vi->i_sb, "Page is uptodate, setting " - "dirty so the write will be retried " - "later on by the VM."); - /* - * Put the page on mapping->dirty_pages, but leave its - * buffers' dirty state as-is. - */ - __set_page_dirty_nobuffers(page); - err = 0; - } else - ntfs_error(vi->i_sb, "Page is not uptodate. Written " - "data has been lost."); - } else { - ntfs_error(vi->i_sb, "Resident attribute commit write failed " - "with error %i.", err); - NVolSetErrors(ni->vol); - make_bad_inode(vi); - } - if (ctx) - ntfs_attr_put_search_ctx(ctx); - if (m) - unmap_mft_record(base_ni); - return err; -} - #endif /* NTFS_RW */ /** @@ -2377,9 +1552,6 @@ struct address_space_operations ntfs_aops = { disk request queue. */ #ifdef NTFS_RW .writepage = ntfs_writepage, /* Write dirty page to disk. */ - .prepare_write = ntfs_prepare_write, /* Prepare page and buffers - ready to receive data. */ - .commit_write = ntfs_commit_write, /* Commit received data. */ #endif /* NTFS_RW */ }; diff --git a/fs/ntfs/attrib.c b/fs/ntfs/attrib.c index 3f9a4ff42ee5..eda056bac256 100644 --- a/fs/ntfs/attrib.c +++ b/fs/ntfs/attrib.c @@ -21,7 +21,9 @@ */ #include <linux/buffer_head.h> +#include <linux/sched.h> #include <linux/swap.h> +#include <linux/writeback.h> #include "attrib.h" #include "debug.h" @@ -36,9 +38,27 @@ * ntfs_map_runlist_nolock - map (a part of) a runlist of an ntfs inode * @ni: ntfs inode for which to map (part of) a runlist * @vcn: map runlist part containing this vcn + * @ctx: active attribute search context if present or NULL if not * * Map the part of a runlist containing the @vcn of the ntfs inode @ni. * + * If @ctx is specified, it is an active search context of @ni and its base mft + * record. This is needed when ntfs_map_runlist_nolock() encounters unmapped + * runlist fragments and allows their mapping. If you do not have the mft + * record mapped, you can specify @ctx as NULL and ntfs_map_runlist_nolock() + * will perform the necessary mapping and unmapping. + * + * Note, ntfs_map_runlist_nolock() saves the state of @ctx on entry and + * restores it before returning. Thus, @ctx will be left pointing to the same + * attribute on return as on entry. However, the actual pointers in @ctx may + * point to different memory locations on return, so you must remember to reset + * any cached pointers from the @ctx, i.e. after the call to + * ntfs_map_runlist_nolock(), you will probably want to do: + * m = ctx->mrec; + * a = ctx->attr; + * Assuming you cache ctx->attr in a variable @a of type ATTR_RECORD * and that + * you cache ctx->mrec in a variable @m of type MFT_RECORD *. + * * Return 0 on success and -errno on error. There is one special error code * which is not an error as such. This is -ENOENT. It means that @vcn is out * of bounds of the runlist. @@ -46,19 +66,32 @@ * Note the runlist can be NULL after this function returns if @vcn is zero and * the attribute has zero allocated size, i.e. there simply is no runlist. * - * Locking: - The runlist must be locked for writing. - * - This function modifies the runlist. + * WARNING: If @ctx is supplied, regardless of whether success or failure is + * returned, you need to check IS_ERR(@ctx->mrec) and if TRUE the @ctx + * is no longer valid, i.e. you need to either call + * ntfs_attr_reinit_search_ctx() or ntfs_attr_put_search_ctx() on it. + * In that case PTR_ERR(@ctx->mrec) will give you the error code for + * why the mapping of the old inode failed. + * + * Locking: - The runlist described by @ni must be locked for writing on entry + * and is locked on return. Note the runlist will be modified. + * - If @ctx is NULL, the base mft record of @ni must not be mapped on + * entry and it will be left unmapped on return. + * - If @ctx is not NULL, the base mft record must be mapped on entry + * and it will be left mapped on return. */ -int ntfs_map_runlist_nolock(ntfs_inode *ni, VCN vcn) +int ntfs_map_runlist_nolock(ntfs_inode *ni, VCN vcn, ntfs_attr_search_ctx *ctx) { VCN end_vcn; + unsigned long flags; ntfs_inode *base_ni; MFT_RECORD *m; ATTR_RECORD *a; - ntfs_attr_search_ctx *ctx; runlist_element *rl; - unsigned long flags; + struct page *put_this_page = NULL; int err = 0; + BOOL ctx_is_temporary, ctx_needs_reset; + ntfs_attr_search_ctx old_ctx = { NULL, }; ntfs_debug("Mapping runlist part containing vcn 0x%llx.", (unsigned long long)vcn); @@ -66,20 +99,77 @@ int ntfs_map_runlist_nolock(ntfs_inode *ni, VCN vcn) base_ni = ni; else base_ni = ni->ext.base_ntfs_ino; - m = map_mft_record(base_ni); - if (IS_ERR(m)) - return PTR_ERR(m); - ctx = ntfs_attr_get_search_ctx(base_ni, m); - if (unlikely(!ctx)) { - err = -ENOMEM; - goto err_out; + if (!ctx) { + ctx_is_temporary = ctx_needs_reset = TRUE; + m = map_mft_record(base_ni); + if (IS_ERR(m)) + return PTR_ERR(m); + ctx = ntfs_attr_get_search_ctx(base_ni, m); + if (unlikely(!ctx)) { + err = -ENOMEM; + goto err_out; + } + } else { + VCN allocated_size_vcn; + + BUG_ON(IS_ERR(ctx->mrec)); + a = ctx->attr; + BUG_ON(!a->non_resident); + ctx_is_temporary = FALSE; + end_vcn = sle64_to_cpu(a->data.non_resident.highest_vcn); + read_lock_irqsave(&ni->size_lock, flags); + allocated_size_vcn = ni->allocated_size >> + ni->vol->cluster_size_bits; + read_unlock_irqrestore(&ni->size_lock, flags); + if (!a->data.non_resident.lowest_vcn && end_vcn <= 0) + end_vcn = allocated_size_vcn - 1; + /* + * If we already have the attribute extent containing @vcn in + * @ctx, no need to look it up again. We slightly cheat in + * that if vcn exceeds the allocated size, we will refuse to + * map the runlist below, so there is definitely no need to get + * the right attribute extent. + */ + if (vcn >= allocated_size_vcn || (a->type == ni->type && + a->name_length == ni->name_len && + !memcmp((u8*)a + le16_to_cpu(a->name_offset), + ni->name, ni->name_len) && + sle64_to_cpu(a->data.non_resident.lowest_vcn) + <= vcn && end_vcn >= vcn)) + ctx_needs_reset = FALSE; + else { + /* Save the old search context. */ + old_ctx = *ctx; + /* + * If the currently mapped (extent) inode is not the + * base inode we will unmap it when we reinitialize the + * search context which means we need to get a + * reference to the page containing the mapped mft + * record so we do not accidentally drop changes to the + * mft record when it has not been marked dirty yet. + */ + if (old_ctx.base_ntfs_ino && old_ctx.ntfs_ino != + old_ctx.base_ntfs_ino) { + put_this_page = old_ctx.ntfs_ino->page; + page_cache_get(put_this_page); + } + /* + * Reinitialize the search context so we can lookup the + * needed attribute extent. + */ + ntfs_attr_reinit_search_ctx(ctx); + ctx_needs_reset = TRUE; + } } - err = ntfs_attr_lookup(ni->type, ni->name, ni->name_len, - CASE_SENSITIVE, vcn, NULL, 0, ctx); - if (unlikely(err)) { - if (err == -ENOENT) - err = -EIO; - goto err_out; + if (ctx_needs_reset) { + err = ntfs_attr_lookup(ni->type, ni->name, ni->name_len, + CASE_SENSITIVE, vcn, NULL, 0, ctx); + if (unlikely(err)) { + if (err == -ENOENT) + err = -EIO; + goto err_out; + } + BUG_ON(!ctx->attr->non_resident); } a = ctx->attr; /* @@ -89,11 +179,9 @@ int ntfs_map_runlist_nolock(ntfs_inode *ni, VCN vcn) * ntfs_mapping_pairs_decompress() fails. */ end_vcn = sle64_to_cpu(a->data.non_resident.highest_vcn) + 1; - if (unlikely(!a->data.non_resident.lowest_vcn && end_vcn <= 1)) { - read_lock_irqsave(&ni->size_lock, flags); - end_vcn = ni->allocated_size >> ni->vol->cluster_size_bits; - read_unlock_irqrestore(&ni->size_lock, flags); - } + if (!a->data.non_resident.lowest_vcn && end_vcn == 1) + end_vcn = sle64_to_cpu(a->data.non_resident.allocated_size) >> + ni->vol->cluster_size_bits; if (unlikely(vcn >= end_vcn)) { err = -ENOENT; goto err_out; @@ -104,9 +192,93 @@ int ntfs_map_runlist_nolock(ntfs_inode *ni, VCN vcn) else ni->runlist.rl = rl; err_out: - if (likely(ctx)) - ntfs_attr_put_search_ctx(ctx); - unmap_mft_record(base_ni); + if (ctx_is_temporary) { + if (likely(ctx)) + ntfs_attr_put_search_ctx(ctx); + unmap_mft_record(base_ni); + } else if (ctx_needs_reset) { + /* + * If there is no attribute list, restoring the search context + * is acomplished simply by copying the saved context back over + * the caller supplied context. If there is an attribute list, + * things are more complicated as we need to deal with mapping + * of mft records and resulting potential changes in pointers. + */ + if (NInoAttrList(base_ni)) { + /* + * If the currently mapped (extent) inode is not the + * one we had before, we need to unmap it and map the + * old one. + */ + if (ctx->ntfs_ino != old_ctx.ntfs_ino) { + /* + * If the currently mapped inode is not the + * base inode, unmap it. + */ + if (ctx->base_ntfs_ino && ctx->ntfs_ino != + ctx->base_ntfs_ino) { + unmap_extent_mft_record(ctx->ntfs_ino); + ctx->mrec = ctx->base_mrec; + BUG_ON(!ctx->mrec); + } + /* + * If the old mapped inode is not the base + * inode, map it. + */ + if (old_ctx.base_ntfs_ino && + old_ctx.ntfs_ino != + old_ctx.base_ntfs_ino) { +retry_map: + ctx->mrec = map_mft_record( + old_ctx.ntfs_ino); + /* + * Something bad has happened. If out + * of memory retry till it succeeds. + * Any other errors are fatal and we + * return the error code in ctx->mrec. + * Let the caller deal with it... We + * just need to fudge things so the + * caller can reinit and/or put the + * search context safely. + */ + if (IS_ERR(ctx->mrec)) { + if (PTR_ERR(ctx->mrec) == + -ENOMEM) { + schedule(); + goto retry_map; + } else + old_ctx.ntfs_ino = + old_ctx. + base_ntfs_ino; + } + } + } + /* Update the changed pointers in the saved context. */ + if (ctx->mrec != old_ctx.mrec) { + if (!IS_ERR(ctx->mrec)) + old_ctx.attr = (ATTR_RECORD*)( + (u8*)ctx->mrec + + ((u8*)old_ctx.attr - + (u8*)old_ctx.mrec)); + old_ctx.mrec = ctx->mrec; + } + } + /* Restore the search context to the saved one. */ + *ctx = old_ctx; + /* + * We drop the reference on the page we took earlier. In the + * case that IS_ERR(ctx->mrec) is true this means we might lose + * some changes to the mft record that had been made between + * the last time it was marked dirty/written out and now. This + * at this stage is not a problem as the mapping error is fatal + * enough that the mft record cannot be written out anyway and + * the caller is very likely to shutdown the whole inode + * immediately and mark the volume dirty for chkdsk to pick up + * the pieces anyway. + */ + if (put_this_page) + page_cache_release(put_this_page); + } return err; } @@ -122,8 +294,8 @@ err_out: * of bounds of the runlist. * * Locking: - The runlist must be unlocked on entry and is unlocked on return. - * - This function takes the runlist lock for writing and modifies the - * runlist. + * - This function takes the runlist lock for writing and may modify + * the runlist. */ int ntfs_map_runlist(ntfs_inode *ni, VCN vcn) { @@ -133,7 +305,7 @@ int ntfs_map_runlist(ntfs_inode *ni, VCN vcn) /* Make sure someone else didn't do the work while we were sleeping. */ if (likely(ntfs_rl_vcn_to_lcn(ni->runlist.rl, vcn) <= LCN_RL_NOT_MAPPED)) - err = ntfs_map_runlist_nolock(ni, vcn); + err = ntfs_map_runlist_nolock(ni, vcn, NULL); up_write(&ni->runlist.lock); return err; } @@ -212,7 +384,7 @@ retry_remap: goto retry_remap; } } - err = ntfs_map_runlist_nolock(ni, vcn); + err = ntfs_map_runlist_nolock(ni, vcn, NULL); if (!write_locked) { up_write(&ni->runlist.lock); down_read(&ni->runlist.lock); @@ -236,9 +408,9 @@ retry_remap: /** * ntfs_attr_find_vcn_nolock - find a vcn in the runlist of an ntfs inode - * @ni: ntfs inode describing the runlist to search - * @vcn: vcn to find - * @write_locked: true if the runlist is locked for writing + * @ni: ntfs inode describing the runlist to search + * @vcn: vcn to find + * @ctx: active attribute search context if present or NULL if not * * Find the virtual cluster number @vcn in the runlist described by the ntfs * inode @ni and return the address of the runlist element containing the @vcn. @@ -246,9 +418,22 @@ retry_remap: * If the @vcn is not mapped yet, the attempt is made to map the attribute * extent containing the @vcn and the vcn to lcn conversion is retried. * - * If @write_locked is true the caller has locked the runlist for writing and - * if false for reading. - * + * If @ctx is specified, it is an active search context of @ni and its base mft + * record. This is needed when ntfs_attr_find_vcn_nolock() encounters unmapped + * runlist fragments and allows their mapping. If you do not have the mft + * record mapped, you can specify @ctx as NULL and ntfs_attr_find_vcn_nolock() + * will perform the necessary mapping and unmapping. + * + * Note, ntfs_attr_find_vcn_nolock() saves the state of @ctx on entry and + * restores it before returning. Thus, @ctx will be left pointing to the same + * attribute on return as on entry. However, the actual pointers in @ctx may + * point to different memory locations on return, so you must remember to reset + * any cached pointers from the @ctx, i.e. after the call to + * ntfs_attr_find_vcn_nolock(), you will probably want to do: + * m = ctx->mrec; + * a = ctx->attr; + * Assuming you cache ctx->attr in a variable @a of type ATTR_RECORD * and that + * you cache ctx->mrec in a variable @m of type MFT_RECORD *. * Note you need to distinguish between the lcn of the returned runlist element * being >= 0 and LCN_HOLE. In the later case you have to return zeroes on * read and allocate clusters on write. @@ -263,22 +448,31 @@ retry_remap: * -ENOMEM - Not enough memory to map runlist. * -EIO - Critical error (runlist/file is corrupt, i/o error, etc). * - * Locking: - The runlist must be locked on entry and is left locked on return. - * - If @write_locked is FALSE, i.e. the runlist is locked for reading, - * the lock may be dropped inside the function so you cannot rely on - * the runlist still being the same when this function returns. + * WARNING: If @ctx is supplied, regardless of whether success or failure is + * returned, you need to check IS_ERR(@ctx->mrec) and if TRUE the @ctx + * is no longer valid, i.e. you need to either call + * ntfs_attr_reinit_search_ctx() or ntfs_attr_put_search_ctx() on it. + * In that case PTR_ERR(@ctx->mrec) will give you the error code for + * why the mapping of the old inode failed. + * + * Locking: - The runlist described by @ni must be locked for writing on entry + * and is locked on return. Note the runlist may be modified when + * needed runlist fragments need to be mapped. + * - If @ctx is NULL, the base mft record of @ni must not be mapped on + * entry and it will be left unmapped on return. + * - If @ctx is not NULL, the base mft record must be mapped on entry + * and it will be left mapped on return. */ runlist_element *ntfs_attr_find_vcn_nolock(ntfs_inode *ni, const VCN vcn, - const BOOL write_locked) + ntfs_attr_search_ctx *ctx) { unsigned long flags; runlist_element *rl; int err = 0; BOOL is_retry = FALSE; - ntfs_debug("Entering for i_ino 0x%lx, vcn 0x%llx, %s_locked.", - ni->mft_no, (unsigned long long)vcn, - write_locked ? "write" : "read"); + ntfs_debug("Entering for i_ino 0x%lx, vcn 0x%llx, with%s ctx.", + ni->mft_no, (unsigned long long)vcn, ctx ? "" : "out"); BUG_ON(!ni); BUG_ON(!NInoNonResident(ni)); BUG_ON(vcn < 0); @@ -312,33 +506,22 @@ retry_remap: } if (!err && !is_retry) { /* - * The @vcn is in an unmapped region, map the runlist and - * retry. + * If the search context is invalid we cannot map the unmapped + * region. */ - if (!write_locked) { - up_read(&ni->runlist.lock); - down_write(&ni->runlist.lock); - if (unlikely(ntfs_rl_vcn_to_lcn(ni->runlist.rl, vcn) != - LCN_RL_NOT_MAPPED)) { - up_write(&ni->runlist.lock); - down_read(&ni->runlist.lock); + if (IS_ERR(ctx->mrec)) + err = PTR_ERR(ctx->mrec); + else { + /* + * The @vcn is in an unmapped region, map the runlist + * and retry. + */ + err = ntfs_map_runlist_nolock(ni, vcn, ctx); + if (likely(!err)) { + is_retry = TRUE; goto retry_remap; } } - err = ntfs_map_runlist_nolock(ni, vcn); - if (!write_locked) { - up_write(&ni->runlist.lock); - down_read(&ni->runlist.lock); - } - if (likely(!err)) { - is_retry = TRUE; - goto retry_remap; - } - /* - * -EINVAL coming from a failed mapping attempt is equivalent - * to i/o error for us as it should not happen in our code - * paths. - */ if (err == -EINVAL) err = -EIO; } else if (!err) @@ -1011,6 +1194,7 @@ int ntfs_attr_lookup(const ATTR_TYPE type, const ntfschar *name, ntfs_inode *base_ni; ntfs_debug("Entering."); + BUG_ON(IS_ERR(ctx->mrec)); if (ctx->base_ntfs_ino) base_ni = ctx->base_ntfs_ino; else @@ -1227,7 +1411,7 @@ int ntfs_attr_can_be_non_resident(const ntfs_volume *vol, const ATTR_TYPE type) */ int ntfs_attr_can_be_resident(const ntfs_volume *vol, const ATTR_TYPE type) { - if (type == AT_INDEX_ALLOCATION || type == AT_EA) + if (type == AT_INDEX_ALLOCATION) return -EPERM; return 0; } @@ -1319,10 +1503,17 @@ int ntfs_resident_attr_value_resize(MFT_RECORD *m, ATTR_RECORD *a, /** * ntfs_attr_make_non_resident - convert a resident to a non-resident attribute * @ni: ntfs inode describing the attribute to convert + * @data_size: size of the resident data to copy to the non-resident attribute * * Convert the resident ntfs attribute described by the ntfs inode @ni to a * non-resident one. * + * @data_size must be equal to the attribute value size. This is needed since + * we need to know the size before we can map the mft record and our callers + * always know it. The reason we cannot simply read the size from the vfs + * inode i_size is that this is not necessarily uptodate. This happens when + * ntfs_attr_make_non_resident() is called in the ->truncate call path(s). + * * Return 0 on success and -errno on error. The following error return codes * are defined: * -EPERM - The attribute is not allowed to be non-resident. @@ -1343,7 +1534,7 @@ int ntfs_resident_attr_value_resize(MFT_RECORD *m, ATTR_RECORD *a, * * Locking: - The caller must hold i_sem on the inode. */ -int ntfs_attr_make_non_resident(ntfs_inode *ni) +int ntfs_attr_make_non_resident(ntfs_inode *ni, const u32 data_size) { s64 new_size; struct inode *vi = VFS_I(ni); @@ -1381,11 +1572,9 @@ int ntfs_attr_make_non_resident(ntfs_inode *ni) * The size needs to be aligned to a cluster boundary for allocation * purposes. */ - new_size = (i_size_read(vi) + vol->cluster_size - 1) & + new_size = (data_size + vol->cluster_size - 1) & ~(vol->cluster_size - 1); if (new_size > 0) { - runlist_element *rl2; - /* * Will need the page later and since the page lock nests * outside all ntfs locks, we need to get the page now. @@ -1396,7 +1585,7 @@ int ntfs_attr_make_non_resident(ntfs_inode *ni) return -ENOMEM; /* Start by allocating clusters to hold the attribute value. */ rl = ntfs_cluster_alloc(vol, 0, new_size >> - vol->cluster_size_bits, -1, DATA_ZONE); + vol->cluster_size_bits, -1, DATA_ZONE, TRUE); if (IS_ERR(rl)) { err = PTR_ERR(rl); ntfs_debug("Failed to allocate cluster%s, error code " @@ -1405,12 +1594,6 @@ int ntfs_attr_make_non_resident(ntfs_inode *ni) err); goto page_err_out; } - /* Change the runlist terminator to LCN_ENOENT. */ - rl2 = rl; - while (rl2->length) - rl2++; - BUG_ON(rl2->lcn != LCN_RL_NOT_MAPPED); - rl2->lcn = LCN_ENOENT; } else { rl = NULL; page = NULL; @@ -1473,7 +1656,7 @@ int ntfs_attr_make_non_resident(ntfs_inode *ni) * attribute value. */ attr_size = le32_to_cpu(a->data.resident.value_length); - BUG_ON(attr_size != i_size_read(vi)); + BUG_ON(attr_size != data_size); if (page && !PageUptodate(page)) { kaddr = kmap_atomic(page, KM_USER0); memcpy(kaddr, (u8*)a + @@ -1538,7 +1721,9 @@ int ntfs_attr_make_non_resident(ntfs_inode *ni) ffs(ni->itype.compressed.block_size) - 1; ni->itype.compressed.block_clusters = 1U << a->data.non_resident.compression_unit; - } + vi->i_blocks = ni->itype.compressed.size >> 9; + } else + vi->i_blocks = ni->allocated_size >> 9; write_unlock_irqrestore(&ni->size_lock, flags); /* * This needs to be last since the address space operations ->readpage @@ -1652,6 +1837,640 @@ page_err_out: } /** + * ntfs_attr_extend_allocation - extend the allocated space of an attribute + * @ni: ntfs inode of the attribute whose allocation to extend + * @new_alloc_size: new size in bytes to which to extend the allocation to + * @new_data_size: new size in bytes to which to extend the data to + * @data_start: beginning of region which is required to be non-sparse + * + * Extend the allocated space of an attribute described by the ntfs inode @ni + * to @new_alloc_size bytes. If @data_start is -1, the whole extension may be + * implemented as a hole in the file (as long as both the volume and the ntfs + * inode @ni have sparse support enabled). If @data_start is >= 0, then the + * region between the old allocated size and @data_start - 1 may be made sparse + * but the regions between @data_start and @new_alloc_size must be backed by + * actual clusters. + * + * If @new_data_size is -1, it is ignored. If it is >= 0, then the data size + * of the attribute is extended to @new_data_size. Note that the i_size of the + * vfs inode is not updated. Only the data size in the base attribute record + * is updated. The caller has to update i_size separately if this is required. + * WARNING: It is a BUG() for @new_data_size to be smaller than the old data + * size as well as for @new_data_size to be greater than @new_alloc_size. + * + * For resident attributes this involves resizing the attribute record and if + * necessary moving it and/or other attributes into extent mft records and/or + * converting the attribute to a non-resident attribute which in turn involves + * extending the allocation of a non-resident attribute as described below. + * + * For non-resident attributes this involves allocating clusters in the data + * zone on the volume (except for regions that are being made sparse) and + * extending the run list to describe the allocated clusters as well as + * updating the mapping pairs array of the attribute. This in turn involves + * resizing the attribute record and if necessary moving it and/or other + * attributes into extent mft records and/or splitting the attribute record + * into multiple extent attribute records. + * + * Also, the attribute list attribute is updated if present and in some of the + * above cases (the ones where extent mft records/attributes come into play), + * an attribute list attribute is created if not already present. + * + * Return the new allocated size on success and -errno on error. In the case + * that an error is encountered but a partial extension at least up to + * @data_start (if present) is possible, the allocation is partially extended + * and this is returned. This means the caller must check the returned size to + * determine if the extension was partial. If @data_start is -1 then partial + * allocations are not performed. + * + * WARNING: Do not call ntfs_attr_extend_allocation() for $MFT/$DATA. + * + * Locking: This function takes the runlist lock of @ni for writing as well as + * locking the mft record of the base ntfs inode. These locks are maintained + * throughout execution of the function. These locks are required so that the + * attribute can be resized safely and so that it can for example be converted + * from resident to non-resident safely. + * + * TODO: At present attribute list attribute handling is not implemented. + * + * TODO: At present it is not safe to call this function for anything other + * than the $DATA attribute(s) of an uncompressed and unencrypted file. + */ +s64 ntfs_attr_extend_allocation(ntfs_inode *ni, s64 new_alloc_size, + const s64 new_data_size, const s64 data_start) +{ + VCN vcn; + s64 ll, allocated_size, start = data_start; + struct inode *vi = VFS_I(ni); + ntfs_volume *vol = ni->vol; + ntfs_inode *base_ni; + MFT_RECORD *m; + ATTR_RECORD *a; + ntfs_attr_search_ctx *ctx; + runlist_element *rl, *rl2; + unsigned long flags; + int err, mp_size; + u32 attr_len = 0; /* Silence stupid gcc warning. */ + BOOL mp_rebuilt; + +#ifdef NTFS_DEBUG + read_lock_irqsave(&ni->size_lock, flags); + allocated_size = ni->allocated_size; + read_unlock_irqrestore(&ni->size_lock, flags); + ntfs_debug("Entering for i_ino 0x%lx, attribute type 0x%x, " + "old_allocated_size 0x%llx, " + "new_allocated_size 0x%llx, new_data_size 0x%llx, " + "data_start 0x%llx.", vi->i_ino, + (unsigned)le32_to_cpu(ni->type), + (unsigned long long)allocated_size, + (unsigned long long)new_alloc_size, + (unsigned long long)new_data_size, + (unsigned long long)start); +#endif +retry_extend: + /* + * For non-resident attributes, @start and @new_size need to be aligned + * to cluster boundaries for allocation purposes. + */ + if (NInoNonResident(ni)) { + if (start > 0) + start &= ~(s64)vol->cluster_size_mask; + new_alloc_size = (new_alloc_size + vol->cluster_size - 1) & + ~(s64)vol->cluster_size_mask; + } + BUG_ON(new_data_size >= 0 && new_data_size > new_alloc_size); + /* Check if new size is allowed in $AttrDef. */ + err = ntfs_attr_size_bounds_check(vol, ni->type, new_alloc_size); + if (unlikely(err)) { + /* Only emit errors when the write will fail completely. */ + read_lock_irqsave(&ni->size_lock, flags); + allocated_size = ni->allocated_size; + read_unlock_irqrestore(&ni->size_lock, flags); + if (start < 0 || start >= allocated_size) { + if (err == -ERANGE) { + ntfs_error(vol->sb, "Cannot extend allocation " + "of inode 0x%lx, attribute " + "type 0x%x, because the new " + "allocation would exceed the " + "maximum allowed size for " + "this attribute type.", + vi->i_ino, (unsigned) + le32_to_cpu(ni->type)); + } else { + ntfs_error(vol->sb, "Cannot extend allocation " + "of inode 0x%lx, attribute " + "type 0x%x, because this " + "attribute type is not " + "defined on the NTFS volume. " + "Possible corruption! You " + "should run chkdsk!", + vi->i_ino, (unsigned) + le32_to_cpu(ni->type)); + } + } + /* Translate error code to be POSIX conformant for write(2). */ + if (err == -ERANGE) + err = -EFBIG; + else + err = -EIO; + return err; + } + if (!NInoAttr(ni)) + base_ni = ni; + else + base_ni = ni->ext.base_ntfs_ino; + /* + * We will be modifying both the runlist (if non-resident) and the mft + * record so lock them both down. + */ + down_write(&ni->runlist.lock); + m = map_mft_record(base_ni); + if (IS_ERR(m)) { + err = PTR_ERR(m); + m = NULL; + ctx = NULL; + goto err_out; + } + ctx = ntfs_attr_get_search_ctx(base_ni, m); + if (unlikely(!ctx)) { + err = -ENOMEM; + goto err_out; + } + read_lock_irqsave(&ni->size_lock, flags); + allocated_size = ni->allocated_size; + read_unlock_irqrestore(&ni->size_lock, flags); + /* + * If non-resident, seek to the last extent. If resident, there is + * only one extent, so seek to that. + */ + vcn = NInoNonResident(ni) ? allocated_size >> vol->cluster_size_bits : + 0; + /* + * Abort if someone did the work whilst we waited for the locks. If we + * just converted the attribute from resident to non-resident it is + * likely that exactly this has happened already. We cannot quite + * abort if we need to update the data size. + */ + if (unlikely(new_alloc_size <= allocated_size)) { + ntfs_debug("Allocated size already exceeds requested size."); + new_alloc_size = allocated_size; + if (new_data_size < 0) + goto done; + /* + * We want the first attribute extent so that we can update the + * data size. + */ + vcn = 0; + } + err = ntfs_attr_lookup(ni->type, ni->name, ni->name_len, + CASE_SENSITIVE, vcn, NULL, 0, ctx); + if (unlikely(err)) { + if (err == -ENOENT) + err = -EIO; + goto err_out; + } + m = ctx->mrec; + a = ctx->attr; + /* Use goto to reduce indentation. */ + if (a->non_resident) + goto do_non_resident_extend; + BUG_ON(NInoNonResident(ni)); + /* The total length of the attribute value. */ + attr_len = le32_to_cpu(a->data.resident.value_length); + /* + * Extend the attribute record to be able to store the new attribute + * size. ntfs_attr_record_resize() will not do anything if the size is + * not changing. + */ + if (new_alloc_size < vol->mft_record_size && + !ntfs_attr_record_resize(m, a, + le16_to_cpu(a->data.resident.value_offset) + + new_alloc_size)) { + /* The resize succeeded! */ + write_lock_irqsave(&ni->size_lock, flags); + ni->allocated_size = le32_to_cpu(a->length) - + le16_to_cpu(a->data.resident.value_offset); + write_unlock_irqrestore(&ni->size_lock, flags); + if (new_data_size >= 0) { + BUG_ON(new_data_size < attr_len); + a->data.resident.value_length = + cpu_to_le32((u32)new_data_size); + } + goto flush_done; + } + /* + * We have to drop all the locks so we can call + * ntfs_attr_make_non_resident(). This could be optimised by try- + * locking the first page cache page and only if that fails dropping + * the locks, locking the page, and redoing all the locking and + * lookups. While this would be a huge optimisation, it is not worth + * it as this is definitely a slow code path. + */ + ntfs_attr_put_search_ctx(ctx); + unmap_mft_record(base_ni); + up_write(&ni->runlist.lock); + /* + * Not enough space in the mft record, try to make the attribute + * non-resident and if successful restart the extension process. + */ + err = ntfs_attr_make_non_resident(ni, attr_len); + if (likely(!err)) + goto retry_extend; + /* + * Could not make non-resident. If this is due to this not being + * permitted for this attribute type or there not being enough space, + * try to make other attributes non-resident. Otherwise fail. + */ + if (unlikely(err != -EPERM && err != -ENOSPC)) { + /* Only emit errors when the write will fail completely. */ + read_lock_irqsave(&ni->size_lock, flags); + allocated_size = ni->allocated_size; + read_unlock_irqrestore(&ni->size_lock, flags); + if (start < 0 || start >= allocated_size) + ntfs_error(vol->sb, "Cannot extend allocation of " + "inode 0x%lx, attribute type 0x%x, " + "because the conversion from resident " + "to non-resident attribute failed " + "with error code %i.", vi->i_ino, + (unsigned)le32_to_cpu(ni->type), err); + if (err != -ENOMEM) + err = -EIO; + goto conv_err_out; + } + /* TODO: Not implemented from here, abort. */ + read_lock_irqsave(&ni->size_lock, flags); + allocated_size = ni->allocated_size; + read_unlock_irqrestore(&ni->size_lock, flags); + if (start < 0 || start >= allocated_size) { + if (err == -ENOSPC) + ntfs_error(vol->sb, "Not enough space in the mft " + "record/on disk for the non-resident " + "attribute value. This case is not " + "implemented yet."); + else /* if (err == -EPERM) */ + ntfs_error(vol->sb, "This attribute type may not be " + "non-resident. This case is not " + "implemented yet."); + } + err = -EOPNOTSUPP; + goto conv_err_out; +#if 0 + // TODO: Attempt to make other attributes non-resident. + if (!err) + goto do_resident_extend; + /* + * Both the attribute list attribute and the standard information + * attribute must remain in the base inode. Thus, if this is one of + * these attributes, we have to try to move other attributes out into + * extent mft records instead. + */ + if (ni->type == AT_ATTRIBUTE_LIST || + ni->type == AT_STANDARD_INFORMATION) { + // TODO: Attempt to move other attributes into extent mft + // records. + err = -EOPNOTSUPP; + if (!err) + goto do_resident_extend; + goto err_out; + } + // TODO: Attempt to move this attribute to an extent mft record, but + // only if it is not already the only attribute in an mft record in + // which case there would be nothing to gain. + err = -EOPNOTSUPP; + if (!err) + goto do_resident_extend; + /* There is nothing we can do to make enough space. )-: */ + goto err_out; +#endif +do_non_resident_extend: + BUG_ON(!NInoNonResident(ni)); + if (new_alloc_size == allocated_size) { + BUG_ON(vcn); + goto alloc_done; + } + /* + * If the data starts after the end of the old allocation, this is a + * $DATA attribute and sparse attributes are enabled on the volume and + * for this inode, then create a sparse region between the old + * allocated size and the start of the data. Otherwise simply proceed + * with filling the whole space between the old allocated size and the + * new allocated size with clusters. + */ + if ((start >= 0 && start <= allocated_size) || ni->type != AT_DATA || + !NVolSparseEnabled(vol) || NInoSparseDisabled(ni)) + goto skip_sparse; + // TODO: This is not implemented yet. We just fill in with real + // clusters for now... + ntfs_debug("Inserting holes is not-implemented yet. Falling back to " + "allocating real clusters instead."); +skip_sparse: + rl = ni->runlist.rl; + if (likely(rl)) { + /* Seek to the end of the runlist. */ + while (rl->length) + rl++; + } + /* If this attribute extent is not mapped, map it now. */ + if (unlikely(!rl || rl->lcn == LCN_RL_NOT_MAPPED || + (rl->lcn == LCN_ENOENT && rl > ni->runlist.rl && + (rl-1)->lcn == LCN_RL_NOT_MAPPED))) { + if (!rl && !allocated_size) + goto first_alloc; + rl = ntfs_mapping_pairs_decompress(vol, a, ni->runlist.rl); + if (IS_ERR(rl)) { + err = PTR_ERR(rl); + if (start < 0 || start >= allocated_size) + ntfs_error(vol->sb, "Cannot extend allocation " + "of inode 0x%lx, attribute " + "type 0x%x, because the " + "mapping of a runlist " + "fragment failed with error " + "code %i.", vi->i_ino, + (unsigned)le32_to_cpu(ni->type), + err); + if (err != -ENOMEM) + err = -EIO; + goto err_out; + } + ni->runlist.rl = rl; + /* Seek to the end of the runlist. */ + while (rl->length) + rl++; + } + /* + * We now know the runlist of the last extent is mapped and @rl is at + * the end of the runlist. We want to begin allocating clusters + * starting at the last allocated cluster to reduce fragmentation. If + * there are no valid LCNs in the attribute we let the cluster + * allocator choose the starting cluster. + */ + /* If the last LCN is a hole or simillar seek back to last real LCN. */ + while (rl->lcn < 0 && rl > ni->runlist.rl) + rl--; +first_alloc: + // FIXME: Need to implement partial allocations so at least part of the + // write can be performed when start >= 0. (Needed for POSIX write(2) + // conformance.) + rl2 = ntfs_cluster_alloc(vol, allocated_size >> vol->cluster_size_bits, + (new_alloc_size - allocated_size) >> + vol->cluster_size_bits, (rl && (rl->lcn >= 0)) ? + rl->lcn + rl->length : -1, DATA_ZONE, TRUE); + if (IS_ERR(rl2)) { + err = PTR_ERR(rl2); + if (start < 0 || start >= allocated_size) + ntfs_error(vol->sb, "Cannot extend allocation of " + "inode 0x%lx, attribute type 0x%x, " + "because the allocation of clusters " + "failed with error code %i.", vi->i_ino, + (unsigned)le32_to_cpu(ni->type), err); + if (err != -ENOMEM && err != -ENOSPC) + err = -EIO; + goto err_out; + } + rl = ntfs_runlists_merge(ni->runlist.rl, rl2); + if (IS_ERR(rl)) { + err = PTR_ERR(rl); + if (start < 0 || start >= allocated_size) + ntfs_error(vol->sb, "Cannot extend allocation of " + "inode 0x%lx, attribute type 0x%x, " + "because the runlist merge failed " + "with error code %i.", vi->i_ino, + (unsigned)le32_to_cpu(ni->type), err); + if (err != -ENOMEM) + err = -EIO; + if (ntfs_cluster_free_from_rl(vol, rl2)) { + ntfs_error(vol->sb, "Failed to release allocated " + "cluster(s) in error code path. Run " + "chkdsk to recover the lost " + "cluster(s)."); + NVolSetErrors(vol); + } + ntfs_free(rl2); + goto err_out; + } + ni->runlist.rl = rl; + ntfs_debug("Allocated 0x%llx clusters.", (long long)(new_alloc_size - + allocated_size) >> vol->cluster_size_bits); + /* Find the runlist element with which the attribute extent starts. */ + ll = sle64_to_cpu(a->data.non_resident.lowest_vcn); + rl2 = ntfs_rl_find_vcn_nolock(rl, ll); + BUG_ON(!rl2); + BUG_ON(!rl2->length); + BUG_ON(rl2->lcn < LCN_HOLE); + mp_rebuilt = FALSE; + /* Get the size for the new mapping pairs array for this extent. */ + mp_size = ntfs_get_size_for_mapping_pairs(vol, rl2, ll, -1); + if (unlikely(mp_size <= 0)) { + err = mp_size; + if (start < 0 || start >= allocated_size) + ntfs_error(vol->sb, "Cannot extend allocation of " + "inode 0x%lx, attribute type 0x%x, " + "because determining the size for the " + "mapping pairs failed with error code " + "%i.", vi->i_ino, + (unsigned)le32_to_cpu(ni->type), err); + err = -EIO; + goto undo_alloc; + } + /* Extend the attribute record to fit the bigger mapping pairs array. */ + attr_len = le32_to_cpu(a->length); + err = ntfs_attr_record_resize(m, a, mp_size + + le16_to_cpu(a->data.non_resident.mapping_pairs_offset)); + if (unlikely(err)) { + BUG_ON(err != -ENOSPC); + // TODO: Deal with this by moving this extent to a new mft + // record or by starting a new extent in a new mft record, + // possibly by extending this extent partially and filling it + // and creating a new extent for the remainder, or by making + // other attributes non-resident and/or by moving other + // attributes out of this mft record. + if (start < 0 || start >= allocated_size) + ntfs_error(vol->sb, "Not enough space in the mft " + "record for the extended attribute " + "record. This case is not " + "implemented yet."); + err = -EOPNOTSUPP; + goto undo_alloc; + } + mp_rebuilt = TRUE; + /* Generate the mapping pairs array directly into the attr record. */ + err = ntfs_mapping_pairs_build(vol, (u8*)a + + le16_to_cpu(a->data.non_resident.mapping_pairs_offset), + mp_size, rl2, ll, -1, NULL); + if (unlikely(err)) { + if (start < 0 || start >= allocated_size) + ntfs_error(vol->sb, "Cannot extend allocation of " + "inode 0x%lx, attribute type 0x%x, " + "because building the mapping pairs " + "failed with error code %i.", vi->i_ino, + (unsigned)le32_to_cpu(ni->type), err); + err = -EIO; + goto undo_alloc; + } + /* Update the highest_vcn. */ + a->data.non_resident.highest_vcn = cpu_to_sle64((new_alloc_size >> + vol->cluster_size_bits) - 1); + /* + * We now have extended the allocated size of the attribute. Reflect + * this in the ntfs_inode structure and the attribute record. + */ + if (a->data.non_resident.lowest_vcn) { + /* + * We are not in the first attribute extent, switch to it, but + * first ensure the changes will make it to disk later. + */ + flush_dcache_mft_record_page(ctx->ntfs_ino); + mark_mft_record_dirty(ctx->ntfs_ino); + ntfs_attr_reinit_search_ctx(ctx); + err = ntfs_attr_lookup(ni->type, ni->name, ni->name_len, + CASE_SENSITIVE, 0, NULL, 0, ctx); + if (unlikely(err)) + goto restore_undo_alloc; + /* @m is not used any more so no need to set it. */ + a = ctx->attr; + } + write_lock_irqsave(&ni->size_lock, flags); + ni->allocated_size = new_alloc_size; + a->data.non_resident.allocated_size = cpu_to_sle64(new_alloc_size); + /* + * FIXME: This would fail if @ni is a directory, $MFT, or an index, + * since those can have sparse/compressed set. For example can be + * set compressed even though it is not compressed itself and in that + * case the bit means that files are to be created compressed in the + * directory... At present this is ok as this code is only called for + * regular files, and only for their $DATA attribute(s). + * FIXME: The calculation is wrong if we created a hole above. For now + * it does not matter as we never create holes. + */ + if (NInoSparse(ni) || NInoCompressed(ni)) { + ni->itype.compressed.size += new_alloc_size - allocated_size; + a->data.non_resident.compressed_size = + cpu_to_sle64(ni->itype.compressed.size); + vi->i_blocks = ni->itype.compressed.size >> 9; + } else + vi->i_blocks = new_alloc_size >> 9; + write_unlock_irqrestore(&ni->size_lock, flags); +alloc_done: + if (new_data_size >= 0) { + BUG_ON(new_data_size < + sle64_to_cpu(a->data.non_resident.data_size)); + a->data.non_resident.data_size = cpu_to_sle64(new_data_size); + } +flush_done: + /* Ensure the changes make it to disk. */ + flush_dcache_mft_record_page(ctx->ntfs_ino); + mark_mft_record_dirty(ctx->ntfs_ino); +done: + ntfs_attr_put_search_ctx(ctx); + unmap_mft_record(base_ni); + up_write(&ni->runlist.lock); + ntfs_debug("Done, new_allocated_size 0x%llx.", + (unsigned long long)new_alloc_size); + return new_alloc_size; +restore_undo_alloc: + if (start < 0 || start >= allocated_size) + ntfs_error(vol->sb, "Cannot complete extension of allocation " + "of inode 0x%lx, attribute type 0x%x, because " + "lookup of first attribute extent failed with " + "error code %i.", vi->i_ino, + (unsigned)le32_to_cpu(ni->type), err); + if (err == -ENOENT) + err = -EIO; + ntfs_attr_reinit_search_ctx(ctx); + if (ntfs_attr_lookup(ni->type, ni->name, ni->name_len, CASE_SENSITIVE, + allocated_size >> vol->cluster_size_bits, NULL, 0, + ctx)) { + ntfs_error(vol->sb, "Failed to find last attribute extent of " + "attribute in error code path. Run chkdsk to " + "recover."); + write_lock_irqsave(&ni->size_lock, flags); + ni->allocated_size = new_alloc_size; + /* + * FIXME: This would fail if @ni is a directory... See above. + * FIXME: The calculation is wrong if we created a hole above. + * For now it does not matter as we never create holes. + */ + if (NInoSparse(ni) || NInoCompressed(ni)) { + ni->itype.compressed.size += new_alloc_size - + allocated_size; + vi->i_blocks = ni->itype.compressed.size >> 9; + } else + vi->i_blocks = new_alloc_size >> 9; + write_unlock_irqrestore(&ni->size_lock, flags); + ntfs_attr_put_search_ctx(ctx); + unmap_mft_record(base_ni); + up_write(&ni->runlist.lock); + /* + * The only thing that is now wrong is the allocated size of the + * base attribute extent which chkdsk should be able to fix. + */ + NVolSetErrors(vol); + return err; + } + ctx->attr->data.non_resident.highest_vcn = cpu_to_sle64( + (allocated_size >> vol->cluster_size_bits) - 1); +undo_alloc: + ll = allocated_size >> vol->cluster_size_bits; + if (ntfs_cluster_free(ni, ll, -1, ctx) < 0) { + ntfs_error(vol->sb, "Failed to release allocated cluster(s) " + "in error code path. Run chkdsk to recover " + "the lost cluster(s)."); + NVolSetErrors(vol); + } + m = ctx->mrec; + a = ctx->attr; + /* + * If the runlist truncation fails and/or the search context is no + * longer valid, we cannot resize the attribute record or build the + * mapping pairs array thus we mark the inode bad so that no access to + * the freed clusters can happen. + */ + if (ntfs_rl_truncate_nolock(vol, &ni->runlist, ll) || IS_ERR(m)) { + ntfs_error(vol->sb, "Failed to %s in error code path. Run " + "chkdsk to recover.", IS_ERR(m) ? + "restore attribute search context" : + "truncate attribute runlist"); + make_bad_inode(vi); + make_bad_inode(VFS_I(base_ni)); + NVolSetErrors(vol); + } else if (mp_rebuilt) { + if (ntfs_attr_record_resize(m, a, attr_len)) { + ntfs_error(vol->sb, "Failed to restore attribute " + "record in error code path. Run " + "chkdsk to recover."); + make_bad_inode(vi); + make_bad_inode(VFS_I(base_ni)); + NVolSetErrors(vol); + } else /* if (success) */ { + if (ntfs_mapping_pairs_build(vol, (u8*)a + le16_to_cpu( + a->data.non_resident. + mapping_pairs_offset), attr_len - + le16_to_cpu(a->data.non_resident. + mapping_pairs_offset), rl2, ll, -1, + NULL)) { + ntfs_error(vol->sb, "Failed to restore " + "mapping pairs array in error " + "code path. Run chkdsk to " + "recover."); + make_bad_inode(vi); + make_bad_inode(VFS_I(base_ni)); + NVolSetErrors(vol); + } + flush_dcache_mft_record_page(ctx->ntfs_ino); + mark_mft_record_dirty(ctx->ntfs_ino); + } + } +err_out: + if (ctx) + ntfs_attr_put_search_ctx(ctx); + if (m) + unmap_mft_record(base_ni); + up_write(&ni->runlist.lock); +conv_err_out: + ntfs_debug("Failed. Returning error code %i.", err); + return err; +} + +/** * ntfs_attr_set - fill (a part of) an attribute with a byte * @ni: ntfs inode describing the attribute to fill * @ofs: offset inside the attribute at which to start to fill @@ -1773,6 +2592,8 @@ int ntfs_attr_set(ntfs_inode *ni, const s64 ofs, const s64 cnt, const u8 val) /* Finally unlock and release the page. */ unlock_page(page); page_cache_release(page); + balance_dirty_pages_ratelimited(mapping); + cond_resched(); } /* If there is a last partial page, need to do it the slow way. */ if (end_ofs) { diff --git a/fs/ntfs/attrib.h b/fs/ntfs/attrib.h index 0618ed6fd7b3..9074886b44ba 100644 --- a/fs/ntfs/attrib.h +++ b/fs/ntfs/attrib.h @@ -60,14 +60,15 @@ typedef struct { ATTR_RECORD *base_attr; } ntfs_attr_search_ctx; -extern int ntfs_map_runlist_nolock(ntfs_inode *ni, VCN vcn); +extern int ntfs_map_runlist_nolock(ntfs_inode *ni, VCN vcn, + ntfs_attr_search_ctx *ctx); extern int ntfs_map_runlist(ntfs_inode *ni, VCN vcn); extern LCN ntfs_attr_vcn_to_lcn_nolock(ntfs_inode *ni, const VCN vcn, const BOOL write_locked); extern runlist_element *ntfs_attr_find_vcn_nolock(ntfs_inode *ni, - const VCN vcn, const BOOL write_locked); + const VCN vcn, ntfs_attr_search_ctx *ctx); int ntfs_attr_lookup(const ATTR_TYPE type, const ntfschar *name, const u32 name_len, const IGNORE_CASE_BOOL ic, @@ -102,7 +103,10 @@ extern int ntfs_attr_record_resize(MFT_RECORD *m, ATTR_RECORD *a, u32 new_size); extern int ntfs_resident_attr_value_resize(MFT_RECORD *m, ATTR_RECORD *a, const u32 new_size); -extern int ntfs_attr_make_non_resident(ntfs_inode *ni); +extern int ntfs_attr_make_non_resident(ntfs_inode *ni, const u32 data_size); + +extern s64 ntfs_attr_extend_allocation(ntfs_inode *ni, s64 new_alloc_size, + const s64 new_data_size, const s64 data_start); extern int ntfs_attr_set(ntfs_inode *ni, const s64 ofs, const s64 cnt, const u8 val); diff --git a/fs/ntfs/bitmap.c b/fs/ntfs/bitmap.c index 12cf2e30c7dd..7a190cdc60e2 100644 --- a/fs/ntfs/bitmap.c +++ b/fs/ntfs/bitmap.c @@ -1,7 +1,7 @@ /* * bitmap.c - NTFS kernel bitmap handling. Part of the Linux-NTFS project. * - * Copyright (c) 2004 Anton Altaparmakov + * Copyright (c) 2004-2005 Anton Altaparmakov * * This program/include file is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License as published @@ -90,7 +90,8 @@ int __ntfs_bitmap_set_bits_in_run(struct inode *vi, const s64 start_bit, /* If the first byte is partial, modify the appropriate bits in it. */ if (bit) { u8 *byte = kaddr + pos; - while ((bit & 7) && cnt--) { + while ((bit & 7) && cnt) { + cnt--; if (value) *byte |= 1 << bit++; else diff --git a/fs/ntfs/file.c b/fs/ntfs/file.c index be9fd1dd423d..727533891813 100644 --- a/fs/ntfs/file.c +++ b/fs/ntfs/file.c @@ -19,11 +19,24 @@ * Foundation,Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA */ -#include <linux/pagemap.h> #include <linux/buffer_head.h> +#include <linux/pagemap.h> +#include <linux/pagevec.h> +#include <linux/sched.h> +#include <linux/swap.h> +#include <linux/uio.h> +#include <linux/writeback.h> +#include <asm/page.h> +#include <asm/uaccess.h> + +#include "attrib.h" +#include "bitmap.h" #include "inode.h" #include "debug.h" +#include "lcnalloc.h" +#include "malloc.h" +#include "mft.h" #include "ntfs.h" /** @@ -56,6 +69,2185 @@ static int ntfs_file_open(struct inode *vi, struct file *filp) #ifdef NTFS_RW /** + * ntfs_attr_extend_initialized - extend the initialized size of an attribute + * @ni: ntfs inode of the attribute to extend + * @new_init_size: requested new initialized size in bytes + * @cached_page: store any allocated but unused page here + * @lru_pvec: lru-buffering pagevec of the caller + * + * Extend the initialized size of an attribute described by the ntfs inode @ni + * to @new_init_size bytes. This involves zeroing any non-sparse space between + * the old initialized size and @new_init_size both in the page cache and on + * disk (if relevant complete pages are already uptodate in the page cache then + * these are simply marked dirty). + * + * As a side-effect, the file size (vfs inode->i_size) may be incremented as, + * in the resident attribute case, it is tied to the initialized size and, in + * the non-resident attribute case, it may not fall below the initialized size. + * + * Note that if the attribute is resident, we do not need to touch the page + * cache at all. This is because if the page cache page is not uptodate we + * bring it uptodate later, when doing the write to the mft record since we + * then already have the page mapped. And if the page is uptodate, the + * non-initialized region will already have been zeroed when the page was + * brought uptodate and the region may in fact already have been overwritten + * with new data via mmap() based writes, so we cannot just zero it. And since + * POSIX specifies that the behaviour of resizing a file whilst it is mmap()ped + * is unspecified, we choose not to do zeroing and thus we do not need to touch + * the page at all. For a more detailed explanation see ntfs_truncate() in + * fs/ntfs/inode.c. + * + * @cached_page and @lru_pvec are just optimizations for dealing with multiple + * pages. + * + * Return 0 on success and -errno on error. In the case that an error is + * encountered it is possible that the initialized size will already have been + * incremented some way towards @new_init_size but it is guaranteed that if + * this is the case, the necessary zeroing will also have happened and that all + * metadata is self-consistent. + * + * Locking: i_sem on the vfs inode corrseponsind to the ntfs inode @ni must be + * held by the caller. + */ +static int ntfs_attr_extend_initialized(ntfs_inode *ni, const s64 new_init_size, + struct page **cached_page, struct pagevec *lru_pvec) +{ + s64 old_init_size; + loff_t old_i_size; + pgoff_t index, end_index; + unsigned long flags; + struct inode *vi = VFS_I(ni); + ntfs_inode *base_ni; + MFT_RECORD *m = NULL; + ATTR_RECORD *a; + ntfs_attr_search_ctx *ctx = NULL; + struct address_space *mapping; + struct page *page = NULL; + u8 *kattr; + int err; + u32 attr_len; + + read_lock_irqsave(&ni->size_lock, flags); + old_init_size = ni->initialized_size; + old_i_size = i_size_read(vi); + BUG_ON(new_init_size > ni->allocated_size); + read_unlock_irqrestore(&ni->size_lock, flags); + ntfs_debug("Entering for i_ino 0x%lx, attribute type 0x%x, " + "old_initialized_size 0x%llx, " + "new_initialized_size 0x%llx, i_size 0x%llx.", + vi->i_ino, (unsigned)le32_to_cpu(ni->type), + (unsigned long long)old_init_size, + (unsigned long long)new_init_size, old_i_size); + if (!NInoAttr(ni)) + base_ni = ni; + else + base_ni = ni->ext.base_ntfs_ino; + /* Use goto to reduce indentation and we need the label below anyway. */ + if (NInoNonResident(ni)) + goto do_non_resident_extend; + BUG_ON(old_init_size != old_i_size); + m = map_mft_record(base_ni); + if (IS_ERR(m)) { + err = PTR_ERR(m); + m = NULL; + goto err_out; + } + ctx = ntfs_attr_get_search_ctx(base_ni, m); + if (unlikely(!ctx)) { + err = -ENOMEM; + goto err_out; + } + err = ntfs_attr_lookup(ni->type, ni->name, ni->name_len, + CASE_SENSITIVE, 0, NULL, 0, ctx); + if (unlikely(err)) { + if (err == -ENOENT) + err = -EIO; + goto err_out; + } + m = ctx->mrec; + a = ctx->attr; + BUG_ON(a->non_resident); + /* The total length of the attribute value. */ + attr_len = le32_to_cpu(a->data.resident.value_length); + BUG_ON(old_i_size != (loff_t)attr_len); + /* + * Do the zeroing in the mft record and update the attribute size in + * the mft record. + */ + kattr = (u8*)a + le16_to_cpu(a->data.resident.value_offset); + memset(kattr + attr_len, 0, new_init_size - attr_len); + a->data.resident.value_length = cpu_to_le32((u32)new_init_size); + /* Finally, update the sizes in the vfs and ntfs inodes. */ + write_lock_irqsave(&ni->size_lock, flags); + i_size_write(vi, new_init_size); + ni->initialized_size = new_init_size; + write_unlock_irqrestore(&ni->size_lock, flags); + goto done; +do_non_resident_extend: + /* + * If the new initialized size @new_init_size exceeds the current file + * size (vfs inode->i_size), we need to extend the file size to the + * new initialized size. + */ + if (new_init_size > old_i_size) { + m = map_mft_record(base_ni); + if (IS_ERR(m)) { + err = PTR_ERR(m); + m = NULL; + goto err_out; + } + ctx = ntfs_attr_get_search_ctx(base_ni, m); + if (unlikely(!ctx)) { + err = -ENOMEM; + goto err_out; + } + err = ntfs_attr_lookup(ni->type, ni->name, ni->name_len, + CASE_SENSITIVE, 0, NULL, 0, ctx); + if (unlikely(err)) { + if (err == -ENOENT) + err = -EIO; + goto err_out; + } + m = ctx->mrec; + a = ctx->attr; + BUG_ON(!a->non_resident); + BUG_ON(old_i_size != (loff_t) + sle64_to_cpu(a->data.non_resident.data_size)); + a->data.non_resident.data_size = cpu_to_sle64(new_init_size); + flush_dcache_mft_record_page(ctx->ntfs_ino); + mark_mft_record_dirty(ctx->ntfs_ino); + /* Update the file size in the vfs inode. */ + i_size_write(vi, new_init_size); + ntfs_attr_put_search_ctx(ctx); + ctx = NULL; + unmap_mft_record(base_ni); + m = NULL; + } + mapping = vi->i_mapping; + index = old_init_size >> PAGE_CACHE_SHIFT; + end_index = (new_init_size + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; + do { + /* + * Read the page. If the page is not present, this will zero + * the uninitialized regions for us. + */ + page = read_cache_page(mapping, index, + (filler_t*)mapping->a_ops->readpage, NULL); + if (IS_ERR(page)) { + err = PTR_ERR(page); + goto init_err_out; + } + wait_on_page_locked(page); + if (unlikely(!PageUptodate(page) || PageError(page))) { + page_cache_release(page); + err = -EIO; + goto init_err_out; + } + /* + * Update the initialized size in the ntfs inode. This is + * enough to make ntfs_writepage() work. + */ + write_lock_irqsave(&ni->size_lock, flags); + ni->initialized_size = (index + 1) << PAGE_CACHE_SHIFT; + if (ni->initialized_size > new_init_size) + ni->initialized_size = new_init_size; + write_unlock_irqrestore(&ni->size_lock, flags); + /* Set the page dirty so it gets written out. */ + set_page_dirty(page); + page_cache_release(page); + /* + * Play nice with the vm and the rest of the system. This is + * very much needed as we can potentially be modifying the + * initialised size from a very small value to a really huge + * value, e.g. + * f = open(somefile, O_TRUNC); + * truncate(f, 10GiB); + * seek(f, 10GiB); + * write(f, 1); + * And this would mean we would be marking dirty hundreds of + * thousands of pages or as in the above example more than + * two and a half million pages! + * + * TODO: For sparse pages could optimize this workload by using + * the FsMisc / MiscFs page bit as a "PageIsSparse" bit. This + * would be set in readpage for sparse pages and here we would + * not need to mark dirty any pages which have this bit set. + * The only caveat is that we have to clear the bit everywhere + * where we allocate any clusters that lie in the page or that + * contain the page. + * + * TODO: An even greater optimization would be for us to only + * call readpage() on pages which are not in sparse regions as + * determined from the runlist. This would greatly reduce the + * number of pages we read and make dirty in the case of sparse + * files. + */ + balance_dirty_pages_ratelimited(mapping); + cond_resched(); + } while (++index < end_index); + read_lock_irqsave(&ni->size_lock, flags); + BUG_ON(ni->initialized_size != new_init_size); + read_unlock_irqrestore(&ni->size_lock, flags); + /* Now bring in sync the initialized_size in the mft record. */ + m = map_mft_record(base_ni); + if (IS_ERR(m)) { + err = PTR_ERR(m); + m = NULL; + goto init_err_out; + } + ctx = ntfs_attr_get_search_ctx(base_ni, m); + if (unlikely(!ctx)) { + err = -ENOMEM; + goto init_err_out; + } + err = ntfs_attr_lookup(ni->type, ni->name, ni->name_len, + CASE_SENSITIVE, 0, NULL, 0, ctx); + if (unlikely(err)) { + if (err == -ENOENT) + err = -EIO; + goto init_err_out; + } + m = ctx->mrec; + a = ctx->attr; + BUG_ON(!a->non_resident); + a->data.non_resident.initialized_size = cpu_to_sle64(new_init_size); +done: + flush_dcache_mft_record_page(ctx->ntfs_ino); + mark_mft_record_dirty(ctx->ntfs_ino); + if (ctx) + ntfs_attr_put_search_ctx(ctx); + if (m) + unmap_mft_record(base_ni); + ntfs_debug("Done, initialized_size 0x%llx, i_size 0x%llx.", + (unsigned long long)new_init_size, i_size_read(vi)); + return 0; +init_err_out: + write_lock_irqsave(&ni->size_lock, flags); + ni->initialized_size = old_init_size; + write_unlock_irqrestore(&ni->size_lock, flags); +err_out: + if (ctx) + ntfs_attr_put_search_ctx(ctx); + if (m) + unmap_mft_record(base_ni); + ntfs_debug("Failed. Returning error code %i.", err); + return err; +} + +/** + * ntfs_fault_in_pages_readable - + * + * Fault a number of userspace pages into pagetables. + * + * Unlike include/linux/pagemap.h::fault_in_pages_readable(), this one copes + * with more than two userspace pages as well as handling the single page case + * elegantly. + * + * If you find this difficult to understand, then think of the while loop being + * the following code, except that we do without the integer variable ret: + * + * do { + * ret = __get_user(c, uaddr); + * uaddr += PAGE_SIZE; + * } while (!ret && uaddr < end); + * + * Note, the final __get_user() may well run out-of-bounds of the user buffer, + * but _not_ out-of-bounds of the page the user buffer belongs to, and since + * this is only a read and not a write, and since it is still in the same page, + * it should not matter and this makes the code much simpler. + */ +static inline void ntfs_fault_in_pages_readable(const char __user *uaddr, + int bytes) +{ + const char __user *end; + volatile char c; + + /* Set @end to the first byte outside the last page we care about. */ + end = (const char __user*)PAGE_ALIGN((ptrdiff_t __user)uaddr + bytes); + + while (!__get_user(c, uaddr) && (uaddr += PAGE_SIZE, uaddr < end)) + ; +} + +/** + * ntfs_fault_in_pages_readable_iovec - + * + * Same as ntfs_fault_in_pages_readable() but operates on an array of iovecs. + */ +static inline void ntfs_fault_in_pages_readable_iovec(const struct iovec *iov, + size_t iov_ofs, int bytes) +{ + do { + const char __user *buf; + unsigned len; + + buf = iov->iov_base + iov_ofs; + len = iov->iov_len - iov_ofs; + if (len > bytes) + len = bytes; + ntfs_fault_in_pages_readable(buf, len); + bytes -= len; + iov++; + iov_ofs = 0; + } while (bytes); +} + +/** + * __ntfs_grab_cache_pages - obtain a number of locked pages + * @mapping: address space mapping from which to obtain page cache pages + * @index: starting index in @mapping at which to begin obtaining pages + * @nr_pages: number of page cache pages to obtain + * @pages: array of pages in which to return the obtained page cache pages + * @cached_page: allocated but as yet unused page + * @lru_pvec: lru-buffering pagevec of caller + * + * Obtain @nr_pages locked page cache pages from the mapping @maping and + * starting at index @index. + * + * If a page is newly created, increment its refcount and add it to the + * caller's lru-buffering pagevec @lru_pvec. + * + * This is the same as mm/filemap.c::__grab_cache_page(), except that @nr_pages + * are obtained at once instead of just one page and that 0 is returned on + * success and -errno on error. + * + * Note, the page locks are obtained in ascending page index order. + */ +static inline int __ntfs_grab_cache_pages(struct address_space *mapping, + pgoff_t index, const unsigned nr_pages, struct page **pages, + struct page **cached_page, struct pagevec *lru_pvec) +{ + int err, nr; + + BUG_ON(!nr_pages); + err = nr = 0; + do { + pages[nr] = find_lock_page(mapping, index); + if (!pages[nr]) { + if (!*cached_page) { + *cached_page = page_cache_alloc(mapping); + if (unlikely(!*cached_page)) { + err = -ENOMEM; + goto err_out; + } + } + err = add_to_page_cache(*cached_page, mapping, index, + GFP_KERNEL); + if (unlikely(err)) { + if (err == -EEXIST) + continue; + goto err_out; + } + pages[nr] = *cached_page; + page_cache_get(*cached_page); + if (unlikely(!pagevec_add(lru_pvec, *cached_page))) + __pagevec_lru_add(lru_pvec); + *cached_page = NULL; + } + index++; + nr++; + } while (nr < nr_pages); +out: + return err; +err_out: + while (nr > 0) { + unlock_page(pages[--nr]); + page_cache_release(pages[nr]); + } + goto out; +} + +static inline int ntfs_submit_bh_for_read(struct buffer_head *bh) +{ + lock_buffer(bh); + get_bh(bh); + bh->b_end_io = end_buffer_read_sync; + return submit_bh(READ, bh); +} + +/** + * ntfs_prepare_pages_for_non_resident_write - prepare pages for receiving data + * @pages: array of destination pages + * @nr_pages: number of pages in @pages + * @pos: byte position in file at which the write begins + * @bytes: number of bytes to be written + * + * This is called for non-resident attributes from ntfs_file_buffered_write() + * with i_sem held on the inode (@pages[0]->mapping->host). There are + * @nr_pages pages in @pages which are locked but not kmap()ped. The source + * data has not yet been copied into the @pages. + * + * Need to fill any holes with actual clusters, allocate buffers if necessary, + * ensure all the buffers are mapped, and bring uptodate any buffers that are + * only partially being written to. + * + * If @nr_pages is greater than one, we are guaranteed that the cluster size is + * greater than PAGE_CACHE_SIZE, that all pages in @pages are entirely inside + * the same cluster and that they are the entirety of that cluster, and that + * the cluster is sparse, i.e. we need to allocate a cluster to fill the hole. + * + * i_size is not to be modified yet. + * + * Return 0 on success or -errno on error. + */ +static int ntfs_prepare_pages_for_non_resident_write(struct page **pages, + unsigned nr_pages, s64 pos, size_t bytes) +{ + VCN vcn, highest_vcn = 0, cpos, cend, bh_cpos, bh_cend; + LCN lcn; + s64 bh_pos, vcn_len, end, initialized_size; + sector_t lcn_block; + struct page *page; + struct inode *vi; + ntfs_inode *ni, *base_ni = NULL; + ntfs_volume *vol; + runlist_element *rl, *rl2; + struct buffer_head *bh, *head, *wait[2], **wait_bh = wait; + ntfs_attr_search_ctx *ctx = NULL; + MFT_RECORD *m = NULL; + ATTR_RECORD *a = NULL; + unsigned long flags; + u32 attr_rec_len = 0; + unsigned blocksize, u; + int err, mp_size; + BOOL rl_write_locked, was_hole, is_retry; + unsigned char blocksize_bits; + struct { + u8 runlist_merged:1; + u8 mft_attr_mapped:1; + u8 mp_rebuilt:1; + u8 attr_switched:1; + } status = { 0, 0, 0, 0 }; + + BUG_ON(!nr_pages); + BUG_ON(!pages); + BUG_ON(!*pages); + vi = pages[0]->mapping->host; + ni = NTFS_I(vi); + vol = ni->vol; + ntfs_debug("Entering for inode 0x%lx, attribute type 0x%x, start page " + "index 0x%lx, nr_pages 0x%x, pos 0x%llx, bytes 0x%zx.", + vi->i_ino, ni->type, pages[0]->index, nr_pages, + (long long)pos, bytes); + blocksize_bits = vi->i_blkbits; + blocksize = 1 << blocksize_bits; + u = 0; + do { + struct page *page = pages[u]; + /* + * create_empty_buffers() will create uptodate/dirty buffers if + * the page is uptodate/dirty. + */ + if (!page_has_buffers(page)) { + create_empty_buffers(page, blocksize, 0); + if (unlikely(!page_has_buffers(page))) + return -ENOMEM; + } + } while (++u < nr_pages); + rl_write_locked = FALSE; + rl = NULL; + err = 0; + vcn = lcn = -1; + vcn_len = 0; + lcn_block = -1; + was_hole = FALSE; + cpos = pos >> vol->cluster_size_bits; + end = pos + bytes; + cend = (end + vol->cluster_size - 1) >> vol->cluster_size_bits; + /* + * Loop over each page and for each page over each buffer. Use goto to + * reduce indentation. + */ + u = 0; +do_next_page: + page = pages[u]; + bh_pos = (s64)page->index << PAGE_CACHE_SHIFT; + bh = head = page_buffers(page); + do { + VCN cdelta; + s64 bh_end; + unsigned bh_cofs; + + /* Clear buffer_new on all buffers to reinitialise state. */ + if (buffer_new(bh)) + clear_buffer_new(bh); + bh_end = bh_pos + blocksize; + bh_cpos = bh_pos >> vol->cluster_size_bits; + bh_cofs = bh_pos & vol->cluster_size_mask; + if (buffer_mapped(bh)) { + /* + * The buffer is already mapped. If it is uptodate, + * ignore it. + */ + if (buffer_uptodate(bh)) + continue; + /* + * The buffer is not uptodate. If the page is uptodate + * set the buffer uptodate and otherwise ignore it. + */ + if (PageUptodate(page)) { + set_buffer_uptodate(bh); + continue; + } + /* + * Neither the page nor the buffer are uptodate. If + * the buffer is only partially being written to, we + * need to read it in before the write, i.e. now. + */ + if ((bh_pos < pos && bh_end > pos) || + (bh_pos < end && bh_end > end)) { + /* + * If the buffer is fully or partially within + * the initialized size, do an actual read. + * Otherwise, simply zero the buffer. + */ + read_lock_irqsave(&ni->size_lock, flags); + initialized_size = ni->initialized_size; + read_unlock_irqrestore(&ni->size_lock, flags); + if (bh_pos < initialized_size) { + ntfs_submit_bh_for_read(bh); + *wait_bh++ = bh; + } else { + u8 *kaddr = kmap_atomic(page, KM_USER0); + memset(kaddr + bh_offset(bh), 0, + blocksize); + kunmap_atomic(kaddr, KM_USER0); + flush_dcache_page(page); + set_buffer_uptodate(bh); + } + } + continue; + } + /* Unmapped buffer. Need to map it. */ + bh->b_bdev = vol->sb->s_bdev; + /* + * If the current buffer is in the same clusters as the map + * cache, there is no need to check the runlist again. The + * map cache is made up of @vcn, which is the first cached file + * cluster, @vcn_len which is the number of cached file + * clusters, @lcn is the device cluster corresponding to @vcn, + * and @lcn_block is the block number corresponding to @lcn. + */ + cdelta = bh_cpos - vcn; + if (likely(!cdelta || (cdelta > 0 && cdelta < vcn_len))) { +map_buffer_cached: + BUG_ON(lcn < 0); + bh->b_blocknr = lcn_block + + (cdelta << (vol->cluster_size_bits - + blocksize_bits)) + + (bh_cofs >> blocksize_bits); + set_buffer_mapped(bh); + /* + * If the page is uptodate so is the buffer. If the + * buffer is fully outside the write, we ignore it if + * it was already allocated and we mark it dirty so it + * gets written out if we allocated it. On the other + * hand, if we allocated the buffer but we are not + * marking it dirty we set buffer_new so we can do + * error recovery. + */ + if (PageUptodate(page)) { + if (!buffer_uptodate(bh)) + set_buffer_uptodate(bh); + if (unlikely(was_hole)) { + /* We allocated the buffer. */ + unmap_underlying_metadata(bh->b_bdev, + bh->b_blocknr); + if (bh_end <= pos || bh_pos >= end) + mark_buffer_dirty(bh); + else + set_buffer_new(bh); + } + continue; + } + /* Page is _not_ uptodate. */ + if (likely(!was_hole)) { + /* + * Buffer was already allocated. If it is not + * uptodate and is only partially being written + * to, we need to read it in before the write, + * i.e. now. + */ + if (!buffer_uptodate(bh) && bh_pos < end && + bh_end > pos && + (bh_pos < pos || + bh_end > end)) { + /* + * If the buffer is fully or partially + * within the initialized size, do an + * actual read. Otherwise, simply zero + * the buffer. + */ + read_lock_irqsave(&ni->size_lock, + flags); + initialized_size = ni->initialized_size; + read_unlock_irqrestore(&ni->size_lock, + flags); + if (bh_pos < initialized_size) { + ntfs_submit_bh_for_read(bh); + *wait_bh++ = bh; + } else { + u8 *kaddr = kmap_atomic(page, + KM_USER0); + memset(kaddr + bh_offset(bh), + 0, blocksize); + kunmap_atomic(kaddr, KM_USER0); + flush_dcache_page(page); + set_buffer_uptodate(bh); + } + } + continue; + } + /* We allocated the buffer. */ + unmap_underlying_metadata(bh->b_bdev, bh->b_blocknr); + /* + * If the buffer is fully outside the write, zero it, + * set it uptodate, and mark it dirty so it gets + * written out. If it is partially being written to, + * zero region surrounding the write but leave it to + * commit write to do anything else. Finally, if the + * buffer is fully being overwritten, do nothing. + */ + if (bh_end <= pos || bh_pos >= end) { + if (!buffer_uptodate(bh)) { + u8 *kaddr = kmap_atomic(page, KM_USER0); + memset(kaddr + bh_offset(bh), 0, + blocksize); + kunmap_atomic(kaddr, KM_USER0); + flush_dcache_page(page); + set_buffer_uptodate(bh); + } + mark_buffer_dirty(bh); + continue; + } + set_buffer_new(bh); + if (!buffer_uptodate(bh) && + (bh_pos < pos || bh_end > end)) { + u8 *kaddr; + unsigned pofs; + + kaddr = kmap_atomic(page, KM_USER0); + if (bh_pos < pos) { + pofs = bh_pos & ~PAGE_CACHE_MASK; + memset(kaddr + pofs, 0, pos - bh_pos); + } + if (bh_end > end) { + pofs = end & ~PAGE_CACHE_MASK; + memset(kaddr + pofs, 0, bh_end - end); + } + kunmap_atomic(kaddr, KM_USER0); + flush_dcache_page(page); + } + continue; + } + /* + * Slow path: this is the first buffer in the cluster. If it + * is outside allocated size and is not uptodate, zero it and + * set it uptodate. + */ + read_lock_irqsave(&ni->size_lock, flags); + initialized_size = ni->allocated_size; + read_unlock_irqrestore(&ni->size_lock, flags); + if (bh_pos > initialized_size) { + if (PageUptodate(page)) { + if (!buffer_uptodate(bh)) + set_buffer_uptodate(bh); + } else if (!buffer_uptodate(bh)) { + u8 *kaddr = kmap_atomic(page, KM_USER0); + memset(kaddr + bh_offset(bh), 0, blocksize); + kunmap_atomic(kaddr, KM_USER0); + flush_dcache_page(page); + set_buffer_uptodate(bh); + } + continue; + } + is_retry = FALSE; + if (!rl) { + down_read(&ni->runlist.lock); +retry_remap: + rl = ni->runlist.rl; + } + if (likely(rl != NULL)) { + /* Seek to element containing target cluster. */ + while (rl->length && rl[1].vcn <= bh_cpos) + rl++; + lcn = ntfs_rl_vcn_to_lcn(rl, bh_cpos); + if (likely(lcn >= 0)) { + /* + * Successful remap, setup the map cache and + * use that to deal with the buffer. + */ + was_hole = FALSE; + vcn = bh_cpos; + vcn_len = rl[1].vcn - vcn; + lcn_block = lcn << (vol->cluster_size_bits - + blocksize_bits); + cdelta = 0; + /* + * If the number of remaining clusters touched + * by the write is smaller or equal to the + * number of cached clusters, unlock the + * runlist as the map cache will be used from + * now on. + */ + if (likely(vcn + vcn_len >= cend)) { + if (rl_write_locked) { + up_write(&ni->runlist.lock); + rl_write_locked = FALSE; + } else + up_read(&ni->runlist.lock); + rl = NULL; + } + goto map_buffer_cached; + } + } else + lcn = LCN_RL_NOT_MAPPED; + /* + * If it is not a hole and not out of bounds, the runlist is + * probably unmapped so try to map it now. + */ + if (unlikely(lcn != LCN_HOLE && lcn != LCN_ENOENT)) { + if (likely(!is_retry && lcn == LCN_RL_NOT_MAPPED)) { + /* Attempt to map runlist. */ + if (!rl_write_locked) { + /* + * We need the runlist locked for + * writing, so if it is locked for + * reading relock it now and retry in + * case it changed whilst we dropped + * the lock. + */ + up_read(&ni->runlist.lock); + down_write(&ni->runlist.lock); + rl_write_locked = TRUE; + goto retry_remap; + } + err = ntfs_map_runlist_nolock(ni, bh_cpos, + NULL); + if (likely(!err)) { + is_retry = TRUE; + goto retry_remap; + } + /* + * If @vcn is out of bounds, pretend @lcn is + * LCN_ENOENT. As long as the buffer is out + * of bounds this will work fine. + */ + if (err == -ENOENT) { + lcn = LCN_ENOENT; + err = 0; + goto rl_not_mapped_enoent; + } + } else + err = -EIO; + /* Failed to map the buffer, even after retrying. */ + bh->b_blocknr = -1; + ntfs_error(vol->sb, "Failed to write to inode 0x%lx, " + "attribute type 0x%x, vcn 0x%llx, " + "vcn offset 0x%x, because its " + "location on disk could not be " + "determined%s (error code %i).", + ni->mft_no, ni->type, + (unsigned long long)bh_cpos, + (unsigned)bh_pos & + vol->cluster_size_mask, + is_retry ? " even after retrying" : "", + err); + break; + } +rl_not_mapped_enoent: + /* + * The buffer is in a hole or out of bounds. We need to fill + * the hole, unless the buffer is in a cluster which is not + * touched by the write, in which case we just leave the buffer + * unmapped. This can only happen when the cluster size is + * less than the page cache size. + */ + if (unlikely(vol->cluster_size < PAGE_CACHE_SIZE)) { + bh_cend = (bh_end + vol->cluster_size - 1) >> + vol->cluster_size_bits; + if ((bh_cend <= cpos || bh_cpos >= cend)) { + bh->b_blocknr = -1; + /* + * If the buffer is uptodate we skip it. If it + * is not but the page is uptodate, we can set + * the buffer uptodate. If the page is not + * uptodate, we can clear the buffer and set it + * uptodate. Whether this is worthwhile is + * debatable and this could be removed. + */ + if (PageUptodate(page)) { + if (!buffer_uptodate(bh)) + set_buffer_uptodate(bh); + } else if (!buffer_uptodate(bh)) { + u8 *kaddr = kmap_atomic(page, KM_USER0); + memset(kaddr + bh_offset(bh), 0, + blocksize); + kunmap_atomic(kaddr, KM_USER0); + flush_dcache_page(page); + set_buffer_uptodate(bh); + } + continue; + } + } + /* + * Out of bounds buffer is invalid if it was not really out of + * bounds. + */ + BUG_ON(lcn != LCN_HOLE); + /* + * We need the runlist locked for writing, so if it is locked + * for reading relock it now and retry in case it changed + * whilst we dropped the lock. + */ + BUG_ON(!rl); + if (!rl_write_locked) { + up_read(&ni->runlist.lock); + down_write(&ni->runlist.lock); + rl_write_locked = TRUE; + goto retry_remap; + } + /* Find the previous last allocated cluster. */ + BUG_ON(rl->lcn != LCN_HOLE); + lcn = -1; + rl2 = rl; + while (--rl2 >= ni->runlist.rl) { + if (rl2->lcn >= 0) { + lcn = rl2->lcn + rl2->length; + break; + } + } + rl2 = ntfs_cluster_alloc(vol, bh_cpos, 1, lcn, DATA_ZONE, + FALSE); + if (IS_ERR(rl2)) { + err = PTR_ERR(rl2); + ntfs_debug("Failed to allocate cluster, error code %i.", + err); + break; + } + lcn = rl2->lcn; + rl = ntfs_runlists_merge(ni->runlist.rl, rl2); + if (IS_ERR(rl)) { + err = PTR_ERR(rl); + if (err != -ENOMEM) + err = -EIO; + if (ntfs_cluster_free_from_rl(vol, rl2)) { + ntfs_error(vol->sb, "Failed to release " + "allocated cluster in error " + "code path. Run chkdsk to " + "recover the lost cluster."); + NVolSetErrors(vol); + } + ntfs_free(rl2); + break; + } + ni->runlist.rl = rl; + status.runlist_merged = 1; + ntfs_debug("Allocated cluster, lcn 0x%llx.", lcn); + /* Map and lock the mft record and get the attribute record. */ + if (!NInoAttr(ni)) + base_ni = ni; + else + base_ni = ni->ext.base_ntfs_ino; + m = map_mft_record(base_ni); + if (IS_ERR(m)) { + err = PTR_ERR(m); + break; + } + ctx = ntfs_attr_get_search_ctx(base_ni, m); + if (unlikely(!ctx)) { + err = -ENOMEM; + unmap_mft_record(base_ni); + break; + } + status.mft_attr_mapped = 1; + err = ntfs_attr_lookup(ni->type, ni->name, ni->name_len, + CASE_SENSITIVE, bh_cpos, NULL, 0, ctx); + if (unlikely(err)) { + if (err == -ENOENT) + err = -EIO; + break; + } + m = ctx->mrec; + a = ctx->attr; + /* + * Find the runlist element with which the attribute extent + * starts. Note, we cannot use the _attr_ version because we + * have mapped the mft record. That is ok because we know the + * runlist fragment must be mapped already to have ever gotten + * here, so we can just use the _rl_ version. + */ + vcn = sle64_to_cpu(a->data.non_resident.lowest_vcn); + rl2 = ntfs_rl_find_vcn_nolock(rl, vcn); + BUG_ON(!rl2); + BUG_ON(!rl2->length); + BUG_ON(rl2->lcn < LCN_HOLE); + highest_vcn = sle64_to_cpu(a->data.non_resident.highest_vcn); + /* + * If @highest_vcn is zero, calculate the real highest_vcn + * (which can really be zero). + */ + if (!highest_vcn) + highest_vcn = (sle64_to_cpu( + a->data.non_resident.allocated_size) >> + vol->cluster_size_bits) - 1; + /* + * Determine the size of the mapping pairs array for the new + * extent, i.e. the old extent with the hole filled. + */ + mp_size = ntfs_get_size_for_mapping_pairs(vol, rl2, vcn, + highest_vcn); + if (unlikely(mp_size <= 0)) { + if (!(err = mp_size)) + err = -EIO; + ntfs_debug("Failed to get size for mapping pairs " + "array, error code %i.", err); + break; + } + /* + * Resize the attribute record to fit the new mapping pairs + * array. + */ + attr_rec_len = le32_to_cpu(a->length); + err = ntfs_attr_record_resize(m, a, mp_size + le16_to_cpu( + a->data.non_resident.mapping_pairs_offset)); + if (unlikely(err)) { + BUG_ON(err != -ENOSPC); + // TODO: Deal with this by using the current attribute + // and fill it with as much of the mapping pairs + // array as possible. Then loop over each attribute + // extent rewriting the mapping pairs arrays as we go + // along and if when we reach the end we have not + // enough space, try to resize the last attribute + // extent and if even that fails, add a new attribute + // extent. + // We could also try to resize at each step in the hope + // that we will not need to rewrite every single extent. + // Note, we may need to decompress some extents to fill + // the runlist as we are walking the extents... + ntfs_error(vol->sb, "Not enough space in the mft " + "record for the extended attribute " + "record. This case is not " + "implemented yet."); + err = -EOPNOTSUPP; + break ; + } + status.mp_rebuilt = 1; + /* + * Generate the mapping pairs array directly into the attribute + * record. + */ + err = ntfs_mapping_pairs_build(vol, (u8*)a + le16_to_cpu( + a->data.non_resident.mapping_pairs_offset), + mp_size, rl2, vcn, highest_vcn, NULL); + if (unlikely(err)) { + ntfs_error(vol->sb, "Cannot fill hole in inode 0x%lx, " + "attribute type 0x%x, because building " + "the mapping pairs failed with error " + "code %i.", vi->i_ino, + (unsigned)le32_to_cpu(ni->type), err); + err = -EIO; + break; + } + /* Update the highest_vcn but only if it was not set. */ + if (unlikely(!a->data.non_resident.highest_vcn)) + a->data.non_resident.highest_vcn = + cpu_to_sle64(highest_vcn); + /* + * If the attribute is sparse/compressed, update the compressed + * size in the ntfs_inode structure and the attribute record. + */ + if (likely(NInoSparse(ni) || NInoCompressed(ni))) { + /* + * If we are not in the first attribute extent, switch + * to it, but first ensure the changes will make it to + * disk later. + */ + if (a->data.non_resident.lowest_vcn) { + flush_dcache_mft_record_page(ctx->ntfs_ino); + mark_mft_record_dirty(ctx->ntfs_ino); + ntfs_attr_reinit_search_ctx(ctx); + err = ntfs_attr_lookup(ni->type, ni->name, + ni->name_len, CASE_SENSITIVE, + 0, NULL, 0, ctx); + if (unlikely(err)) { + status.attr_switched = 1; + break; + } + /* @m is not used any more so do not set it. */ + a = ctx->attr; + } + write_lock_irqsave(&ni->size_lock, flags); + ni->itype.compressed.size += vol->cluster_size; + a->data.non_resident.compressed_size = + cpu_to_sle64(ni->itype.compressed.size); + write_unlock_irqrestore(&ni->size_lock, flags); + } + /* Ensure the changes make it to disk. */ + flush_dcache_mft_record_page(ctx->ntfs_ino); + mark_mft_record_dirty(ctx->ntfs_ino); + ntfs_attr_put_search_ctx(ctx); + unmap_mft_record(base_ni); + /* Successfully filled the hole. */ + status.runlist_merged = 0; + status.mft_attr_mapped = 0; + status.mp_rebuilt = 0; + /* Setup the map cache and use that to deal with the buffer. */ + was_hole = TRUE; + vcn = bh_cpos; + vcn_len = 1; + lcn_block = lcn << (vol->cluster_size_bits - blocksize_bits); + cdelta = 0; + /* + * If the number of remaining clusters in the @pages is smaller + * or equal to the number of cached clusters, unlock the + * runlist as the map cache will be used from now on. + */ + if (likely(vcn + vcn_len >= cend)) { + up_write(&ni->runlist.lock); + rl_write_locked = FALSE; + rl = NULL; + } + goto map_buffer_cached; + } while (bh_pos += blocksize, (bh = bh->b_this_page) != head); + /* If there are no errors, do the next page. */ + if (likely(!err && ++u < nr_pages)) + goto do_next_page; + /* If there are no errors, release the runlist lock if we took it. */ + if (likely(!err)) { + if (unlikely(rl_write_locked)) { + up_write(&ni->runlist.lock); + rl_write_locked = FALSE; + } else if (unlikely(rl)) + up_read(&ni->runlist.lock); + rl = NULL; + } + /* If we issued read requests, let them complete. */ + read_lock_irqsave(&ni->size_lock, flags); + initialized_size = ni->initialized_size; + read_unlock_irqrestore(&ni->size_lock, flags); + while (wait_bh > wait) { + bh = *--wait_bh; + wait_on_buffer(bh); + if (likely(buffer_uptodate(bh))) { + page = bh->b_page; + bh_pos = ((s64)page->index << PAGE_CACHE_SHIFT) + + bh_offset(bh); + /* + * If the buffer overflows the initialized size, need + * to zero the overflowing region. + */ + if (unlikely(bh_pos + blocksize > initialized_size)) { + u8 *kaddr; + int ofs = 0; + + if (likely(bh_pos < initialized_size)) + ofs = initialized_size - bh_pos; + kaddr = kmap_atomic(page, KM_USER0); + memset(kaddr + bh_offset(bh) + ofs, 0, + blocksize - ofs); + kunmap_atomic(kaddr, KM_USER0); + flush_dcache_page(page); + } + } else /* if (unlikely(!buffer_uptodate(bh))) */ + err = -EIO; + } + if (likely(!err)) { + /* Clear buffer_new on all buffers. */ + u = 0; + do { + bh = head = page_buffers(pages[u]); + do { + if (buffer_new(bh)) + clear_buffer_new(bh); + } while ((bh = bh->b_this_page) != head); + } while (++u < nr_pages); + ntfs_debug("Done."); + return err; + } + if (status.attr_switched) { + /* Get back to the attribute extent we modified. */ + ntfs_attr_reinit_search_ctx(ctx); + if (ntfs_attr_lookup(ni->type, ni->name, ni->name_len, + CASE_SENSITIVE, bh_cpos, NULL, 0, ctx)) { + ntfs_error(vol->sb, "Failed to find required " + "attribute extent of attribute in " + "error code path. Run chkdsk to " + "recover."); + write_lock_irqsave(&ni->size_lock, flags); + ni->itype.compressed.size += vol->cluster_size; + write_unlock_irqrestore(&ni->size_lock, flags); + flush_dcache_mft_record_page(ctx->ntfs_ino); + mark_mft_record_dirty(ctx->ntfs_ino); + /* + * The only thing that is now wrong is the compressed + * size of the base attribute extent which chkdsk + * should be able to fix. + */ + NVolSetErrors(vol); + } else { + m = ctx->mrec; + a = ctx->attr; + status.attr_switched = 0; + } + } + /* + * If the runlist has been modified, need to restore it by punching a + * hole into it and we then need to deallocate the on-disk cluster as + * well. Note, we only modify the runlist if we are able to generate a + * new mapping pairs array, i.e. only when the mapped attribute extent + * is not switched. + */ + if (status.runlist_merged && !status.attr_switched) { + BUG_ON(!rl_write_locked); + /* Make the file cluster we allocated sparse in the runlist. */ + if (ntfs_rl_punch_nolock(vol, &ni->runlist, bh_cpos, 1)) { + ntfs_error(vol->sb, "Failed to punch hole into " + "attribute runlist in error code " + "path. Run chkdsk to recover the " + "lost cluster."); + make_bad_inode(vi); + make_bad_inode(VFS_I(base_ni)); + NVolSetErrors(vol); + } else /* if (success) */ { + status.runlist_merged = 0; + /* + * Deallocate the on-disk cluster we allocated but only + * if we succeeded in punching its vcn out of the + * runlist. + */ + down_write(&vol->lcnbmp_lock); + if (ntfs_bitmap_clear_bit(vol->lcnbmp_ino, lcn)) { + ntfs_error(vol->sb, "Failed to release " + "allocated cluster in error " + "code path. Run chkdsk to " + "recover the lost cluster."); + NVolSetErrors(vol); + } + up_write(&vol->lcnbmp_lock); + } + } + /* + * Resize the attribute record to its old size and rebuild the mapping + * pairs array. Note, we only can do this if the runlist has been + * restored to its old state which also implies that the mapped + * attribute extent is not switched. + */ + if (status.mp_rebuilt && !status.runlist_merged) { + if (ntfs_attr_record_resize(m, a, attr_rec_len)) { + ntfs_error(vol->sb, "Failed to restore attribute " + "record in error code path. Run " + "chkdsk to recover."); + make_bad_inode(vi); + make_bad_inode(VFS_I(base_ni)); + NVolSetErrors(vol); + } else /* if (success) */ { + if (ntfs_mapping_pairs_build(vol, (u8*)a + + le16_to_cpu(a->data.non_resident. + mapping_pairs_offset), attr_rec_len - + le16_to_cpu(a->data.non_resident. + mapping_pairs_offset), ni->runlist.rl, + vcn, highest_vcn, NULL)) { + ntfs_error(vol->sb, "Failed to restore " + "mapping pairs array in error " + "code path. Run chkdsk to " + "recover."); + make_bad_inode(vi); + make_bad_inode(VFS_I(base_ni)); + NVolSetErrors(vol); + } + flush_dcache_mft_record_page(ctx->ntfs_ino); + mark_mft_record_dirty(ctx->ntfs_ino); + } + } + /* Release the mft record and the attribute. */ + if (status.mft_attr_mapped) { + ntfs_attr_put_search_ctx(ctx); + unmap_mft_record(base_ni); + } + /* Release the runlist lock. */ + if (rl_write_locked) + up_write(&ni->runlist.lock); + else if (rl) + up_read(&ni->runlist.lock); + /* + * Zero out any newly allocated blocks to avoid exposing stale data. + * If BH_New is set, we know that the block was newly allocated above + * and that it has not been fully zeroed and marked dirty yet. + */ + nr_pages = u; + u = 0; + end = bh_cpos << vol->cluster_size_bits; + do { + page = pages[u]; + bh = head = page_buffers(page); + do { + if (u == nr_pages && + ((s64)page->index << PAGE_CACHE_SHIFT) + + bh_offset(bh) >= end) + break; + if (!buffer_new(bh)) + continue; + clear_buffer_new(bh); + if (!buffer_uptodate(bh)) { + if (PageUptodate(page)) + set_buffer_uptodate(bh); + else { + u8 *kaddr = kmap_atomic(page, KM_USER0); + memset(kaddr + bh_offset(bh), 0, + blocksize); + kunmap_atomic(kaddr, KM_USER0); + flush_dcache_page(page); + set_buffer_uptodate(bh); + } + } + mark_buffer_dirty(bh); + } while ((bh = bh->b_this_page) != head); + } while (++u <= nr_pages); + ntfs_error(vol->sb, "Failed. Returning error code %i.", err); + return err; +} + +/* + * Copy as much as we can into the pages and return the number of bytes which + * were sucessfully copied. If a fault is encountered then clear the pages + * out to (ofs + bytes) and return the number of bytes which were copied. + */ +static inline size_t ntfs_copy_from_user(struct page **pages, + unsigned nr_pages, unsigned ofs, const char __user *buf, + size_t bytes) +{ + struct page **last_page = pages + nr_pages; + char *kaddr; + size_t total = 0; + unsigned len; + int left; + + do { + len = PAGE_CACHE_SIZE - ofs; + if (len > bytes) + len = bytes; + kaddr = kmap_atomic(*pages, KM_USER0); + left = __copy_from_user_inatomic(kaddr + ofs, buf, len); + kunmap_atomic(kaddr, KM_USER0); + if (unlikely(left)) { + /* Do it the slow way. */ + kaddr = kmap(*pages); + left = __copy_from_user(kaddr + ofs, buf, len); + kunmap(*pages); + if (unlikely(left)) + goto err_out; + } + total += len; + bytes -= len; + if (!bytes) + break; + buf += len; + ofs = 0; + } while (++pages < last_page); +out: + return total; +err_out: + total += len - left; + /* Zero the rest of the target like __copy_from_user(). */ + while (++pages < last_page) { + bytes -= len; + if (!bytes) + break; + len = PAGE_CACHE_SIZE; + if (len > bytes) + len = bytes; + kaddr = kmap_atomic(*pages, KM_USER0); + memset(kaddr, 0, len); + kunmap_atomic(kaddr, KM_USER0); + } + goto out; +} + +static size_t __ntfs_copy_from_user_iovec(char *vaddr, + const struct iovec *iov, size_t iov_ofs, size_t bytes) +{ + size_t total = 0; + + while (1) { + const char __user *buf = iov->iov_base + iov_ofs; + unsigned len; + size_t left; + + len = iov->iov_len - iov_ofs; + if (len > bytes) + len = bytes; + left = __copy_from_user_inatomic(vaddr, buf, len); + total += len; + bytes -= len; + vaddr += len; + if (unlikely(left)) { + /* + * Zero the rest of the target like __copy_from_user(). + */ + memset(vaddr, 0, bytes); + total -= left; + break; + } + if (!bytes) + break; + iov++; + iov_ofs = 0; + } + return total; +} + +static inline void ntfs_set_next_iovec(const struct iovec **iovp, + size_t *iov_ofsp, size_t bytes) +{ + const struct iovec *iov = *iovp; + size_t iov_ofs = *iov_ofsp; + + while (bytes) { + unsigned len; + + len = iov->iov_len - iov_ofs; + if (len > bytes) + len = bytes; + bytes -= len; + iov_ofs += len; + if (iov->iov_len == iov_ofs) { + iov++; + iov_ofs = 0; + } + } + *iovp = iov; + *iov_ofsp = iov_ofs; +} + +/* + * This has the same side-effects and return value as ntfs_copy_from_user(). + * The difference is that on a fault we need to memset the remainder of the + * pages (out to offset + bytes), to emulate ntfs_copy_from_user()'s + * single-segment behaviour. + * + * We call the same helper (__ntfs_copy_from_user_iovec()) both when atomic and + * when not atomic. This is ok because __ntfs_copy_from_user_iovec() calls + * __copy_from_user_inatomic() and it is ok to call this when non-atomic. In + * fact, the only difference between __copy_from_user_inatomic() and + * __copy_from_user() is that the latter calls might_sleep(). And on many + * architectures __copy_from_user_inatomic() is just defined to + * __copy_from_user() so it makes no difference at all on those architectures. + */ +static inline size_t ntfs_copy_from_user_iovec(struct page **pages, + unsigned nr_pages, unsigned ofs, const struct iovec **iov, + size_t *iov_ofs, size_t bytes) +{ + struct page **last_page = pages + nr_pages; + char *kaddr; + size_t copied, len, total = 0; + + do { + len = PAGE_CACHE_SIZE - ofs; + if (len > bytes) + len = bytes; + kaddr = kmap_atomic(*pages, KM_USER0); + copied = __ntfs_copy_from_user_iovec(kaddr + ofs, + *iov, *iov_ofs, len); + kunmap_atomic(kaddr, KM_USER0); + if (unlikely(copied != len)) { + /* Do it the slow way. */ + kaddr = kmap(*pages); + copied = __ntfs_copy_from_user_iovec(kaddr + ofs, + *iov, *iov_ofs, len); + kunmap(*pages); + if (unlikely(copied != len)) + goto err_out; + } + total += len; + bytes -= len; + if (!bytes) + break; + ntfs_set_next_iovec(iov, iov_ofs, len); + ofs = 0; + } while (++pages < last_page); +out: + return total; +err_out: + total += copied; + /* Zero the rest of the target like __copy_from_user(). */ + while (++pages < last_page) { + bytes -= len; + if (!bytes) + break; + len = PAGE_CACHE_SIZE; + if (len > bytes) + len = bytes; + kaddr = kmap_atomic(*pages, KM_USER0); + memset(kaddr, 0, len); + kunmap_atomic(kaddr, KM_USER0); + } + goto out; +} + +static inline void ntfs_flush_dcache_pages(struct page **pages, + unsigned nr_pages) +{ + BUG_ON(!nr_pages); + do { + /* + * Warning: Do not do the decrement at the same time as the + * call because flush_dcache_page() is a NULL macro on i386 + * and hence the decrement never happens. + */ + flush_dcache_page(pages[nr_pages]); + } while (--nr_pages > 0); +} + +/** + * ntfs_commit_pages_after_non_resident_write - commit the received data + * @pages: array of destination pages + * @nr_pages: number of pages in @pages + * @pos: byte position in file at which the write begins + * @bytes: number of bytes to be written + * + * See description of ntfs_commit_pages_after_write(), below. + */ +static inline int ntfs_commit_pages_after_non_resident_write( + struct page **pages, const unsigned nr_pages, + s64 pos, size_t bytes) +{ + s64 end, initialized_size; + struct inode *vi; + ntfs_inode *ni, *base_ni; + struct buffer_head *bh, *head; + ntfs_attr_search_ctx *ctx; + MFT_RECORD *m; + ATTR_RECORD *a; + unsigned long flags; + unsigned blocksize, u; + int err; + + vi = pages[0]->mapping->host; + ni = NTFS_I(vi); + blocksize = 1 << vi->i_blkbits; + end = pos + bytes; + u = 0; + do { + s64 bh_pos; + struct page *page; + BOOL partial; + + page = pages[u]; + bh_pos = (s64)page->index << PAGE_CACHE_SHIFT; + bh = head = page_buffers(page); + partial = FALSE; + do { + s64 bh_end; + + bh_end = bh_pos + blocksize; + if (bh_end <= pos || bh_pos >= end) { + if (!buffer_uptodate(bh)) + partial = TRUE; + } else { + set_buffer_uptodate(bh); + mark_buffer_dirty(bh); + } + } while (bh_pos += blocksize, (bh = bh->b_this_page) != head); + /* + * If all buffers are now uptodate but the page is not, set the + * page uptodate. + */ + if (!partial && !PageUptodate(page)) + SetPageUptodate(page); + } while (++u < nr_pages); + /* + * Finally, if we do not need to update initialized_size or i_size we + * are finished. + */ + read_lock_irqsave(&ni->size_lock, flags); + initialized_size = ni->initialized_size; + read_unlock_irqrestore(&ni->size_lock, flags); + if (end <= initialized_size) { + ntfs_debug("Done."); + return 0; + } + /* + * Update initialized_size/i_size as appropriate, both in the inode and + * the mft record. + */ + if (!NInoAttr(ni)) + base_ni = ni; + else + base_ni = ni->ext.base_ntfs_ino; + /* Map, pin, and lock the mft record. */ + m = map_mft_record(base_ni); + if (IS_ERR(m)) { + err = PTR_ERR(m); + m = NULL; + ctx = NULL; + goto err_out; + } + BUG_ON(!NInoNonResident(ni)); + ctx = ntfs_attr_get_search_ctx(base_ni, m); + if (unlikely(!ctx)) { + err = -ENOMEM; + goto err_out; + } + err = ntfs_attr_lookup(ni->type, ni->name, ni->name_len, + CASE_SENSITIVE, 0, NULL, 0, ctx); + if (unlikely(err)) { + if (err == -ENOENT) + err = -EIO; + goto err_out; + } + a = ctx->attr; + BUG_ON(!a->non_resident); + write_lock_irqsave(&ni->size_lock, flags); + BUG_ON(end > ni->allocated_size); + ni->initialized_size = end; + a->data.non_resident.initialized_size = cpu_to_sle64(end); + if (end > i_size_read(vi)) { + i_size_write(vi, end); + a->data.non_resident.data_size = + a->data.non_resident.initialized_size; + } + write_unlock_irqrestore(&ni->size_lock, flags); + /* Mark the mft record dirty, so it gets written back. */ + flush_dcache_mft_record_page(ctx->ntfs_ino); + mark_mft_record_dirty(ctx->ntfs_ino); + ntfs_attr_put_search_ctx(ctx); + unmap_mft_record(base_ni); + ntfs_debug("Done."); + return 0; +err_out: + if (ctx) + ntfs_attr_put_search_ctx(ctx); + if (m) + unmap_mft_record(base_ni); + ntfs_error(vi->i_sb, "Failed to update initialized_size/i_size (error " + "code %i).", err); + if (err != -ENOMEM) { + NVolSetErrors(ni->vol); + make_bad_inode(VFS_I(base_ni)); + make_bad_inode(vi); + } + return err; +} + +/** + * ntfs_commit_pages_after_write - commit the received data + * @pages: array of destination pages + * @nr_pages: number of pages in @pages + * @pos: byte position in file at which the write begins + * @bytes: number of bytes to be written + * + * This is called from ntfs_file_buffered_write() with i_sem held on the inode + * (@pages[0]->mapping->host). There are @nr_pages pages in @pages which are + * locked but not kmap()ped. The source data has already been copied into the + * @page. ntfs_prepare_pages_for_non_resident_write() has been called before + * the data was copied (for non-resident attributes only) and it returned + * success. + * + * Need to set uptodate and mark dirty all buffers within the boundary of the + * write. If all buffers in a page are uptodate we set the page uptodate, too. + * + * Setting the buffers dirty ensures that they get written out later when + * ntfs_writepage() is invoked by the VM. + * + * Finally, we need to update i_size and initialized_size as appropriate both + * in the inode and the mft record. + * + * This is modelled after fs/buffer.c::generic_commit_write(), which marks + * buffers uptodate and dirty, sets the page uptodate if all buffers in the + * page are uptodate, and updates i_size if the end of io is beyond i_size. In + * that case, it also marks the inode dirty. + * + * If things have gone as outlined in + * ntfs_prepare_pages_for_non_resident_write(), we do not need to do any page + * content modifications here for non-resident attributes. For resident + * attributes we need to do the uptodate bringing here which we combine with + * the copying into the mft record which means we save one atomic kmap. + * + * Return 0 on success or -errno on error. + */ +static int ntfs_commit_pages_after_write(struct page **pages, + const unsigned nr_pages, s64 pos, size_t bytes) +{ + s64 end, initialized_size; + loff_t i_size; + struct inode *vi; + ntfs_inode *ni, *base_ni; + struct page *page; + ntfs_attr_search_ctx *ctx; + MFT_RECORD *m; + ATTR_RECORD *a; + char *kattr, *kaddr; + unsigned long flags; + u32 attr_len; + int err; + + BUG_ON(!nr_pages); + BUG_ON(!pages); + page = pages[0]; + BUG_ON(!page); + vi = page->mapping->host; + ni = NTFS_I(vi); + ntfs_debug("Entering for inode 0x%lx, attribute type 0x%x, start page " + "index 0x%lx, nr_pages 0x%x, pos 0x%llx, bytes 0x%zx.", + vi->i_ino, ni->type, page->index, nr_pages, + (long long)pos, bytes); + if (NInoNonResident(ni)) + return ntfs_commit_pages_after_non_resident_write(pages, + nr_pages, pos, bytes); + BUG_ON(nr_pages > 1); + /* + * Attribute is resident, implying it is not compressed, encrypted, or + * sparse. + */ + if (!NInoAttr(ni)) + base_ni = ni; + else + base_ni = ni->ext.base_ntfs_ino; + BUG_ON(NInoNonResident(ni)); + /* Map, pin, and lock the mft record. */ + m = map_mft_record(base_ni); + if (IS_ERR(m)) { + err = PTR_ERR(m); + m = NULL; + ctx = NULL; + goto err_out; + } + ctx = ntfs_attr_get_search_ctx(base_ni, m); + if (unlikely(!ctx)) { + err = -ENOMEM; + goto err_out; + } + err = ntfs_attr_lookup(ni->type, ni->name, ni->name_len, + CASE_SENSITIVE, 0, NULL, 0, ctx); + if (unlikely(err)) { + if (err == -ENOENT) + err = -EIO; + goto err_out; + } + a = ctx->attr; + BUG_ON(a->non_resident); + /* The total length of the attribute value. */ + attr_len = le32_to_cpu(a->data.resident.value_length); + i_size = i_size_read(vi); + BUG_ON(attr_len != i_size); + BUG_ON(pos > attr_len); + end = pos + bytes; + BUG_ON(end > le32_to_cpu(a->length) - + le16_to_cpu(a->data.resident.value_offset)); + kattr = (u8*)a + le16_to_cpu(a->data.resident.value_offset); + kaddr = kmap_atomic(page, KM_USER0); + /* Copy the received data from the page to the mft record. */ + memcpy(kattr + pos, kaddr + pos, bytes); + /* Update the attribute length if necessary. */ + if (end > attr_len) { + attr_len = end; + a->data.resident.value_length = cpu_to_le32(attr_len); + } + /* + * If the page is not uptodate, bring the out of bounds area(s) + * uptodate by copying data from the mft record to the page. + */ + if (!PageUptodate(page)) { + if (pos > 0) + memcpy(kaddr, kattr, pos); + if (end < attr_len) + memcpy(kaddr + end, kattr + end, attr_len - end); + /* Zero the region outside the end of the attribute value. */ + memset(kaddr + attr_len, 0, PAGE_CACHE_SIZE - attr_len); + flush_dcache_page(page); + SetPageUptodate(page); + } + kunmap_atomic(kaddr, KM_USER0); + /* Update initialized_size/i_size if necessary. */ + read_lock_irqsave(&ni->size_lock, flags); + initialized_size = ni->initialized_size; + BUG_ON(end > ni->allocated_size); + read_unlock_irqrestore(&ni->size_lock, flags); + BUG_ON(initialized_size != i_size); + if (end > initialized_size) { + unsigned long flags; + + write_lock_irqsave(&ni->size_lock, flags); + ni->initialized_size = end; + i_size_write(vi, end); + write_unlock_irqrestore(&ni->size_lock, flags); + } + /* Mark the mft record dirty, so it gets written back. */ + flush_dcache_mft_record_page(ctx->ntfs_ino); + mark_mft_record_dirty(ctx->ntfs_ino); + ntfs_attr_put_search_ctx(ctx); + unmap_mft_record(base_ni); + ntfs_debug("Done."); + return 0; +err_out: + if (err == -ENOMEM) { + ntfs_warning(vi->i_sb, "Error allocating memory required to " + "commit the write."); + if (PageUptodate(page)) { + ntfs_warning(vi->i_sb, "Page is uptodate, setting " + "dirty so the write will be retried " + "later on by the VM."); + /* + * Put the page on mapping->dirty_pages, but leave its + * buffers' dirty state as-is. + */ + __set_page_dirty_nobuffers(page); + err = 0; + } else + ntfs_error(vi->i_sb, "Page is not uptodate. Written " + "data has been lost."); + } else { + ntfs_error(vi->i_sb, "Resident attribute commit write failed " + "with error %i.", err); + NVolSetErrors(ni->vol); + make_bad_inode(VFS_I(base_ni)); + make_bad_inode(vi); + } + if (ctx) + ntfs_attr_put_search_ctx(ctx); + if (m) + unmap_mft_record(base_ni); + return err; +} + +/** + * ntfs_file_buffered_write - + * + * Locking: The vfs is holding ->i_sem on the inode. + */ +static ssize_t ntfs_file_buffered_write(struct kiocb *iocb, + const struct iovec *iov, unsigned long nr_segs, + loff_t pos, loff_t *ppos, size_t count) +{ + struct file *file = iocb->ki_filp; + struct address_space *mapping = file->f_mapping; + struct inode *vi = mapping->host; + ntfs_inode *ni = NTFS_I(vi); + ntfs_volume *vol = ni->vol; + struct page *pages[NTFS_MAX_PAGES_PER_CLUSTER]; + struct page *cached_page = NULL; + char __user *buf = NULL; + s64 end, ll; + VCN last_vcn; + LCN lcn; + unsigned long flags; + size_t bytes, iov_ofs = 0; /* Offset in the current iovec. */ + ssize_t status, written; + unsigned nr_pages; + int err; + struct pagevec lru_pvec; + + ntfs_debug("Entering for i_ino 0x%lx, attribute type 0x%x, " + "pos 0x%llx, count 0x%lx.", + vi->i_ino, (unsigned)le32_to_cpu(ni->type), + (unsigned long long)pos, (unsigned long)count); + if (unlikely(!count)) + return 0; + BUG_ON(NInoMstProtected(ni)); + /* + * If the attribute is not an index root and it is encrypted or + * compressed, we cannot write to it yet. Note we need to check for + * AT_INDEX_ALLOCATION since this is the type of both directory and + * index inodes. + */ + if (ni->type != AT_INDEX_ALLOCATION) { + /* If file is encrypted, deny access, just like NT4. */ + if (NInoEncrypted(ni)) { + /* + * Reminder for later: Encrypted files are _always_ + * non-resident so that the content can always be + * encrypted. + */ + ntfs_debug("Denying write access to encrypted file."); + return -EACCES; + } + if (NInoCompressed(ni)) { + /* Only unnamed $DATA attribute can be compressed. */ + BUG_ON(ni->type != AT_DATA); + BUG_ON(ni->name_len); + /* + * Reminder for later: If resident, the data is not + * actually compressed. Only on the switch to non- + * resident does compression kick in. This is in + * contrast to encrypted files (see above). + */ + ntfs_error(vi->i_sb, "Writing to compressed files is " + "not implemented yet. Sorry."); + return -EOPNOTSUPP; + } + } + /* + * If a previous ntfs_truncate() failed, repeat it and abort if it + * fails again. + */ + if (unlikely(NInoTruncateFailed(ni))) { + down_write(&vi->i_alloc_sem); + err = ntfs_truncate(vi); + up_write(&vi->i_alloc_sem); + if (err || NInoTruncateFailed(ni)) { + if (!err) + err = -EIO; + ntfs_error(vol->sb, "Cannot perform write to inode " + "0x%lx, attribute type 0x%x, because " + "ntfs_truncate() failed (error code " + "%i).", vi->i_ino, + (unsigned)le32_to_cpu(ni->type), err); + return err; + } + } + /* The first byte after the write. */ + end = pos + count; + /* + * If the write goes beyond the allocated size, extend the allocation + * to cover the whole of the write, rounded up to the nearest cluster. + */ + read_lock_irqsave(&ni->size_lock, flags); + ll = ni->allocated_size; + read_unlock_irqrestore(&ni->size_lock, flags); + if (end > ll) { + /* Extend the allocation without changing the data size. */ + ll = ntfs_attr_extend_allocation(ni, end, -1, pos); + if (likely(ll >= 0)) { + BUG_ON(pos >= ll); + /* If the extension was partial truncate the write. */ + if (end > ll) { + ntfs_debug("Truncating write to inode 0x%lx, " + "attribute type 0x%x, because " + "the allocation was only " + "partially extended.", + vi->i_ino, (unsigned) + le32_to_cpu(ni->type)); + end = ll; + count = ll - pos; + } + } else { + err = ll; + read_lock_irqsave(&ni->size_lock, flags); + ll = ni->allocated_size; + read_unlock_irqrestore(&ni->size_lock, flags); + /* Perform a partial write if possible or fail. */ + if (pos < ll) { + ntfs_debug("Truncating write to inode 0x%lx, " + "attribute type 0x%x, because " + "extending the allocation " + "failed (error code %i).", + vi->i_ino, (unsigned) + le32_to_cpu(ni->type), err); + end = ll; + count = ll - pos; + } else { + ntfs_error(vol->sb, "Cannot perform write to " + "inode 0x%lx, attribute type " + "0x%x, because extending the " + "allocation failed (error " + "code %i).", vi->i_ino, + (unsigned) + le32_to_cpu(ni->type), err); + return err; + } + } + } + pagevec_init(&lru_pvec, 0); + written = 0; + /* + * If the write starts beyond the initialized size, extend it up to the + * beginning of the write and initialize all non-sparse space between + * the old initialized size and the new one. This automatically also + * increments the vfs inode->i_size to keep it above or equal to the + * initialized_size. + */ + read_lock_irqsave(&ni->size_lock, flags); + ll = ni->initialized_size; + read_unlock_irqrestore(&ni->size_lock, flags); + if (pos > ll) { + err = ntfs_attr_extend_initialized(ni, pos, &cached_page, + &lru_pvec); + if (err < 0) { + ntfs_error(vol->sb, "Cannot perform write to inode " + "0x%lx, attribute type 0x%x, because " + "extending the initialized size " + "failed (error code %i).", vi->i_ino, + (unsigned)le32_to_cpu(ni->type), err); + status = err; + goto err_out; + } + } + /* + * Determine the number of pages per cluster for non-resident + * attributes. + */ + nr_pages = 1; + if (vol->cluster_size > PAGE_CACHE_SIZE && NInoNonResident(ni)) + nr_pages = vol->cluster_size >> PAGE_CACHE_SHIFT; + /* Finally, perform the actual write. */ + last_vcn = -1; + if (likely(nr_segs == 1)) + buf = iov->iov_base; + do { + VCN vcn; + pgoff_t idx, start_idx; + unsigned ofs, do_pages, u; + size_t copied; + + start_idx = idx = pos >> PAGE_CACHE_SHIFT; + ofs = pos & ~PAGE_CACHE_MASK; + bytes = PAGE_CACHE_SIZE - ofs; + do_pages = 1; + if (nr_pages > 1) { + vcn = pos >> vol->cluster_size_bits; + if (vcn != last_vcn) { + last_vcn = vcn; + /* + * Get the lcn of the vcn the write is in. If + * it is a hole, need to lock down all pages in + * the cluster. + */ + down_read(&ni->runlist.lock); + lcn = ntfs_attr_vcn_to_lcn_nolock(ni, pos >> + vol->cluster_size_bits, FALSE); + up_read(&ni->runlist.lock); + if (unlikely(lcn < LCN_HOLE)) { + status = -EIO; + if (lcn == LCN_ENOMEM) + status = -ENOMEM; + else + ntfs_error(vol->sb, "Cannot " + "perform write to " + "inode 0x%lx, " + "attribute type 0x%x, " + "because the attribute " + "is corrupt.", + vi->i_ino, (unsigned) + le32_to_cpu(ni->type)); + break; + } + if (lcn == LCN_HOLE) { + start_idx = (pos & ~(s64) + vol->cluster_size_mask) + >> PAGE_CACHE_SHIFT; + bytes = vol->cluster_size - (pos & + vol->cluster_size_mask); + do_pages = nr_pages; + } + } + } + if (bytes > count) + bytes = count; + /* + * Bring in the user page(s) that we will copy from _first_. + * Otherwise there is a nasty deadlock on copying from the same + * page(s) as we are writing to, without it/them being marked + * up-to-date. Note, at present there is nothing to stop the + * pages being swapped out between us bringing them into memory + * and doing the actual copying. + */ + if (likely(nr_segs == 1)) + ntfs_fault_in_pages_readable(buf, bytes); + else + ntfs_fault_in_pages_readable_iovec(iov, iov_ofs, bytes); + /* Get and lock @do_pages starting at index @start_idx. */ + status = __ntfs_grab_cache_pages(mapping, start_idx, do_pages, + pages, &cached_page, &lru_pvec); + if (unlikely(status)) + break; + /* + * For non-resident attributes, we need to fill any holes with + * actual clusters and ensure all bufferes are mapped. We also + * need to bring uptodate any buffers that are only partially + * being written to. + */ + if (NInoNonResident(ni)) { + status = ntfs_prepare_pages_for_non_resident_write( + pages, do_pages, pos, bytes); + if (unlikely(status)) { + loff_t i_size; + + do { + unlock_page(pages[--do_pages]); + page_cache_release(pages[do_pages]); + } while (do_pages); + /* + * The write preparation may have instantiated + * allocated space outside i_size. Trim this + * off again. We can ignore any errors in this + * case as we will just be waisting a bit of + * allocated space, which is not a disaster. + */ + i_size = i_size_read(vi); + if (pos + bytes > i_size) + vmtruncate(vi, i_size); + break; + } + } + u = (pos >> PAGE_CACHE_SHIFT) - pages[0]->index; + if (likely(nr_segs == 1)) { + copied = ntfs_copy_from_user(pages + u, do_pages - u, + ofs, buf, bytes); + buf += copied; + } else + copied = ntfs_copy_from_user_iovec(pages + u, + do_pages - u, ofs, &iov, &iov_ofs, + bytes); + ntfs_flush_dcache_pages(pages + u, do_pages - u); + status = ntfs_commit_pages_after_write(pages, do_pages, pos, + bytes); + if (likely(!status)) { + written += copied; + count -= copied; + pos += copied; + if (unlikely(copied != bytes)) + status = -EFAULT; + } + do { + unlock_page(pages[--do_pages]); + mark_page_accessed(pages[do_pages]); + page_cache_release(pages[do_pages]); + } while (do_pages); + if (unlikely(status)) + break; + balance_dirty_pages_ratelimited(mapping); + cond_resched(); + } while (count); +err_out: + *ppos = pos; + if (cached_page) + page_cache_release(cached_page); + /* For now, when the user asks for O_SYNC, we actually give O_DSYNC. */ + if (likely(!status)) { + if (unlikely((file->f_flags & O_SYNC) || IS_SYNC(vi))) { + if (!mapping->a_ops->writepage || !is_sync_kiocb(iocb)) + status = generic_osync_inode(vi, mapping, + OSYNC_METADATA|OSYNC_DATA); + } + } + pagevec_lru_add(&lru_pvec); + ntfs_debug("Done. Returning %s (written 0x%lx, status %li).", + written ? "written" : "status", (unsigned long)written, + (long)status); + return written ? written : status; +} + +/** + * ntfs_file_aio_write_nolock - + */ +static ssize_t ntfs_file_aio_write_nolock(struct kiocb *iocb, + const struct iovec *iov, unsigned long nr_segs, loff_t *ppos) +{ + struct file *file = iocb->ki_filp; + struct address_space *mapping = file->f_mapping; + struct inode *inode = mapping->host; + loff_t pos; + unsigned long seg; + size_t count; /* after file limit checks */ + ssize_t written, err; + + count = 0; + for (seg = 0; seg < nr_segs; seg++) { + const struct iovec *iv = &iov[seg]; + /* + * If any segment has a negative length, or the cumulative + * length ever wraps negative then return -EINVAL. + */ + count += iv->iov_len; + if (unlikely((ssize_t)(count|iv->iov_len) < 0)) + return -EINVAL; + if (access_ok(VERIFY_READ, iv->iov_base, iv->iov_len)) + continue; + if (!seg) + return -EFAULT; + nr_segs = seg; + count -= iv->iov_len; /* This segment is no good */ + break; + } + pos = *ppos; + vfs_check_frozen(inode->i_sb, SB_FREEZE_WRITE); + /* We can write back this queue in page reclaim. */ + current->backing_dev_info = mapping->backing_dev_info; + written = 0; + err = generic_write_checks(file, &pos, &count, S_ISBLK(inode->i_mode)); + if (err) + goto out; + if (!count) + goto out; + err = remove_suid(file->f_dentry); + if (err) + goto out; + inode_update_time(inode, 1); + written = ntfs_file_buffered_write(iocb, iov, nr_segs, pos, ppos, + count); +out: + current->backing_dev_info = NULL; + return written ? written : err; +} + +/** + * ntfs_file_aio_write - + */ +static ssize_t ntfs_file_aio_write(struct kiocb *iocb, const char __user *buf, + size_t count, loff_t pos) +{ + struct file *file = iocb->ki_filp; + struct address_space *mapping = file->f_mapping; + struct inode *inode = mapping->host; + ssize_t ret; + struct iovec local_iov = { .iov_base = (void __user *)buf, + .iov_len = count }; + + BUG_ON(iocb->ki_pos != pos); + + down(&inode->i_sem); + ret = ntfs_file_aio_write_nolock(iocb, &local_iov, 1, &iocb->ki_pos); + up(&inode->i_sem); + if (ret > 0 && ((file->f_flags & O_SYNC) || IS_SYNC(inode))) { + int err = sync_page_range(inode, mapping, pos, ret); + if (err < 0) + ret = err; + } + return ret; +} + +/** + * ntfs_file_writev - + * + * Basically the same as generic_file_writev() except that it ends up calling + * ntfs_file_aio_write_nolock() instead of __generic_file_aio_write_nolock(). + */ +static ssize_t ntfs_file_writev(struct file *file, const struct iovec *iov, + unsigned long nr_segs, loff_t *ppos) +{ + struct address_space *mapping = file->f_mapping; + struct inode *inode = mapping->host; + struct kiocb kiocb; + ssize_t ret; + + down(&inode->i_sem); + init_sync_kiocb(&kiocb, file); + ret = ntfs_file_aio_write_nolock(&kiocb, iov, nr_segs, ppos); + if (ret == -EIOCBQUEUED) + ret = wait_on_sync_kiocb(&kiocb); + up(&inode->i_sem); + if (ret > 0 && ((file->f_flags & O_SYNC) || IS_SYNC(inode))) { + int err = sync_page_range(inode, mapping, *ppos - ret, ret); + if (err < 0) + ret = err; + } + return ret; +} + +/** + * ntfs_file_write - simple wrapper for ntfs_file_writev() + */ +static ssize_t ntfs_file_write(struct file *file, const char __user *buf, + size_t count, loff_t *ppos) +{ + struct iovec local_iov = { .iov_base = (void __user *)buf, + .iov_len = count }; + + return ntfs_file_writev(file, &local_iov, 1, ppos); +} + +/** * ntfs_file_fsync - sync a file to disk * @filp: file to be synced * @dentry: dentry describing the file to sync @@ -113,39 +2305,39 @@ static int ntfs_file_fsync(struct file *filp, struct dentry *dentry, #endif /* NTFS_RW */ struct file_operations ntfs_file_ops = { - .llseek = generic_file_llseek, /* Seek inside file. */ - .read = generic_file_read, /* Read from file. */ - .aio_read = generic_file_aio_read, /* Async read from file. */ - .readv = generic_file_readv, /* Read from file. */ + .llseek = generic_file_llseek, /* Seek inside file. */ + .read = generic_file_read, /* Read from file. */ + .aio_read = generic_file_aio_read, /* Async read from file. */ + .readv = generic_file_readv, /* Read from file. */ #ifdef NTFS_RW - .write = generic_file_write, /* Write to file. */ - .aio_write = generic_file_aio_write, /* Async write to file. */ - .writev = generic_file_writev, /* Write to file. */ - /*.release = ,*/ /* Last file is closed. See - fs/ext2/file.c:: - ext2_release_file() for - how to use this to discard - preallocated space for - write opened files. */ - .fsync = ntfs_file_fsync, /* Sync a file to disk. */ - /*.aio_fsync = ,*/ /* Sync all outstanding async - i/o operations on a - kiocb. */ + .write = ntfs_file_write, /* Write to file. */ + .aio_write = ntfs_file_aio_write, /* Async write to file. */ + .writev = ntfs_file_writev, /* Write to file. */ + /*.release = ,*/ /* Last file is closed. See + fs/ext2/file.c:: + ext2_release_file() for + how to use this to discard + preallocated space for + write opened files. */ + .fsync = ntfs_file_fsync, /* Sync a file to disk. */ + /*.aio_fsync = ,*/ /* Sync all outstanding async + i/o operations on a + kiocb. */ #endif /* NTFS_RW */ - /*.ioctl = ,*/ /* Perform function on the - mounted filesystem. */ - .mmap = generic_file_mmap, /* Mmap file. */ - .open = ntfs_file_open, /* Open file. */ - .sendfile = generic_file_sendfile, /* Zero-copy data send with - the data source being on - the ntfs partition. We - do not need to care about - the data destination. */ - /*.sendpage = ,*/ /* Zero-copy data send with - the data destination being - on the ntfs partition. We - do not need to care about - the data source. */ + /*.ioctl = ,*/ /* Perform function on the + mounted filesystem. */ + .mmap = generic_file_mmap, /* Mmap file. */ + .open = ntfs_file_open, /* Open file. */ + .sendfile = generic_file_sendfile, /* Zero-copy data send with + the data source being on + the ntfs partition. We do + not need to care about the + data destination. */ + /*.sendpage = ,*/ /* Zero-copy data send with + the data destination being + on the ntfs partition. We + do not need to care about + the data source. */ }; struct inode_operations ntfs_file_inode_ops = { diff --git a/fs/ntfs/inode.c b/fs/ntfs/inode.c index 7ec045131808..b24f4c4b2c5c 100644 --- a/fs/ntfs/inode.c +++ b/fs/ntfs/inode.c @@ -30,6 +30,7 @@ #include "debug.h" #include "inode.h" #include "attrib.h" +#include "lcnalloc.h" #include "malloc.h" #include "mft.h" #include "time.h" @@ -2291,11 +2292,16 @@ int ntfs_show_options(struct seq_file *sf, struct vfsmount *mnt) #ifdef NTFS_RW +static const char *es = " Leaving inconsistent metadata. Unmount and run " + "chkdsk."; + /** * ntfs_truncate - called when the i_size of an ntfs inode is changed * @vi: inode for which the i_size was changed * - * We do not support i_size changes yet. + * We only support i_size changes for normal files at present, i.e. not + * compressed and not encrypted. This is enforced in ntfs_setattr(), see + * below. * * The kernel guarantees that @vi is a regular file (S_ISREG() is true) and * that the change is allowed. @@ -2306,80 +2312,499 @@ int ntfs_show_options(struct seq_file *sf, struct vfsmount *mnt) * Returns 0 on success or -errno on error. * * Called with ->i_sem held. In all but one case ->i_alloc_sem is held for - * writing. The only case where ->i_alloc_sem is not held is + * writing. The only case in the kernel where ->i_alloc_sem is not held is * mm/filemap.c::generic_file_buffered_write() where vmtruncate() is called - * with the current i_size as the offset which means that it is a noop as far - * as ntfs_truncate() is concerned. + * with the current i_size as the offset. The analogous place in NTFS is in + * fs/ntfs/file.c::ntfs_file_buffered_write() where we call vmtruncate() again + * without holding ->i_alloc_sem. */ int ntfs_truncate(struct inode *vi) { - ntfs_inode *ni = NTFS_I(vi); + s64 new_size, old_size, nr_freed, new_alloc_size, old_alloc_size; + VCN highest_vcn; + unsigned long flags; + ntfs_inode *base_ni, *ni = NTFS_I(vi); ntfs_volume *vol = ni->vol; ntfs_attr_search_ctx *ctx; MFT_RECORD *m; ATTR_RECORD *a; const char *te = " Leaving file length out of sync with i_size."; - int err; + int err, mp_size, size_change, alloc_change; + u32 attr_len; ntfs_debug("Entering for inode 0x%lx.", vi->i_ino); BUG_ON(NInoAttr(ni)); + BUG_ON(S_ISDIR(vi->i_mode)); + BUG_ON(NInoMstProtected(ni)); BUG_ON(ni->nr_extents < 0); - m = map_mft_record(ni); +retry_truncate: + /* + * Lock the runlist for writing and map the mft record to ensure it is + * safe to mess with the attribute runlist and sizes. + */ + down_write(&ni->runlist.lock); + if (!NInoAttr(ni)) + base_ni = ni; + else + base_ni = ni->ext.base_ntfs_ino; + m = map_mft_record(base_ni); if (IS_ERR(m)) { err = PTR_ERR(m); ntfs_error(vi->i_sb, "Failed to map mft record for inode 0x%lx " "(error code %d).%s", vi->i_ino, err, te); ctx = NULL; m = NULL; - goto err_out; + goto old_bad_out; } - ctx = ntfs_attr_get_search_ctx(ni, m); + ctx = ntfs_attr_get_search_ctx(base_ni, m); if (unlikely(!ctx)) { ntfs_error(vi->i_sb, "Failed to allocate a search context for " "inode 0x%lx (not enough memory).%s", vi->i_ino, te); err = -ENOMEM; - goto err_out; + goto old_bad_out; } err = ntfs_attr_lookup(ni->type, ni->name, ni->name_len, CASE_SENSITIVE, 0, NULL, 0, ctx); if (unlikely(err)) { - if (err == -ENOENT) + if (err == -ENOENT) { ntfs_error(vi->i_sb, "Open attribute is missing from " "mft record. Inode 0x%lx is corrupt. " - "Run chkdsk.", vi->i_ino); - else + "Run chkdsk.%s", vi->i_ino, te); + err = -EIO; + } else ntfs_error(vi->i_sb, "Failed to lookup attribute in " - "inode 0x%lx (error code %d).", - vi->i_ino, err); - goto err_out; + "inode 0x%lx (error code %d).%s", + vi->i_ino, err, te); + goto old_bad_out; } + m = ctx->mrec; a = ctx->attr; - /* If the size has not changed there is nothing to do. */ - if (ntfs_attr_size(a) == i_size_read(vi)) - goto done; - // TODO: Implement the truncate... - ntfs_error(vi->i_sb, "Inode size has changed but this is not " - "implemented yet. Resetting inode size to old value. " - " This is most likely a bug in the ntfs driver!"); - i_size_write(vi, ntfs_attr_size(a)); -done: + /* + * The i_size of the vfs inode is the new size for the attribute value. + */ + new_size = i_size_read(vi); + /* The current size of the attribute value is the old size. */ + old_size = ntfs_attr_size(a); + /* Calculate the new allocated size. */ + if (NInoNonResident(ni)) + new_alloc_size = (new_size + vol->cluster_size - 1) & + ~(s64)vol->cluster_size_mask; + else + new_alloc_size = (new_size + 7) & ~7; + /* The current allocated size is the old allocated size. */ + read_lock_irqsave(&ni->size_lock, flags); + old_alloc_size = ni->allocated_size; + read_unlock_irqrestore(&ni->size_lock, flags); + /* + * The change in the file size. This will be 0 if no change, >0 if the + * size is growing, and <0 if the size is shrinking. + */ + size_change = -1; + if (new_size - old_size >= 0) { + size_change = 1; + if (new_size == old_size) + size_change = 0; + } + /* As above for the allocated size. */ + alloc_change = -1; + if (new_alloc_size - old_alloc_size >= 0) { + alloc_change = 1; + if (new_alloc_size == old_alloc_size) + alloc_change = 0; + } + /* + * If neither the size nor the allocation are being changed there is + * nothing to do. + */ + if (!size_change && !alloc_change) + goto unm_done; + /* If the size is changing, check if new size is allowed in $AttrDef. */ + if (size_change) { + err = ntfs_attr_size_bounds_check(vol, ni->type, new_size); + if (unlikely(err)) { + if (err == -ERANGE) { + ntfs_error(vol->sb, "Truncate would cause the " + "inode 0x%lx to %simum size " + "for its attribute type " + "(0x%x). Aborting truncate.", + vi->i_ino, + new_size > old_size ? "exceed " + "the max" : "go under the min", + le32_to_cpu(ni->type)); + err = -EFBIG; + } else { + ntfs_error(vol->sb, "Inode 0x%lx has unknown " + "attribute type 0x%x. " + "Aborting truncate.", + vi->i_ino, + le32_to_cpu(ni->type)); + err = -EIO; + } + /* Reset the vfs inode size to the old size. */ + i_size_write(vi, old_size); + goto err_out; + } + } + if (NInoCompressed(ni) || NInoEncrypted(ni)) { + ntfs_warning(vi->i_sb, "Changes in inode size are not " + "supported yet for %s files, ignoring.", + NInoCompressed(ni) ? "compressed" : + "encrypted"); + err = -EOPNOTSUPP; + goto bad_out; + } + if (a->non_resident) + goto do_non_resident_truncate; + BUG_ON(NInoNonResident(ni)); + /* Resize the attribute record to best fit the new attribute size. */ + if (new_size < vol->mft_record_size && + !ntfs_resident_attr_value_resize(m, a, new_size)) { + unsigned long flags; + + /* The resize succeeded! */ + flush_dcache_mft_record_page(ctx->ntfs_ino); + mark_mft_record_dirty(ctx->ntfs_ino); + write_lock_irqsave(&ni->size_lock, flags); + /* Update the sizes in the ntfs inode and all is done. */ + ni->allocated_size = le32_to_cpu(a->length) - + le16_to_cpu(a->data.resident.value_offset); + /* + * Note ntfs_resident_attr_value_resize() has already done any + * necessary data clearing in the attribute record. When the + * file is being shrunk vmtruncate() will already have cleared + * the top part of the last partial page, i.e. since this is + * the resident case this is the page with index 0. However, + * when the file is being expanded, the page cache page data + * between the old data_size, i.e. old_size, and the new_size + * has not been zeroed. Fortunately, we do not need to zero it + * either since on one hand it will either already be zero due + * to both readpage and writepage clearing partial page data + * beyond i_size in which case there is nothing to do or in the + * case of the file being mmap()ped at the same time, POSIX + * specifies that the behaviour is unspecified thus we do not + * have to do anything. This means that in our implementation + * in the rare case that the file is mmap()ped and a write + * occured into the mmap()ped region just beyond the file size + * and writepage has not yet been called to write out the page + * (which would clear the area beyond the file size) and we now + * extend the file size to incorporate this dirty region + * outside the file size, a write of the page would result in + * this data being written to disk instead of being cleared. + * Given both POSIX and the Linux mmap(2) man page specify that + * this corner case is undefined, we choose to leave it like + * that as this is much simpler for us as we cannot lock the + * relevant page now since we are holding too many ntfs locks + * which would result in a lock reversal deadlock. + */ + ni->initialized_size = new_size; + write_unlock_irqrestore(&ni->size_lock, flags); + goto unm_done; + } + /* If the above resize failed, this must be an attribute extension. */ + BUG_ON(size_change < 0); + /* + * We have to drop all the locks so we can call + * ntfs_attr_make_non_resident(). This could be optimised by try- + * locking the first page cache page and only if that fails dropping + * the locks, locking the page, and redoing all the locking and + * lookups. While this would be a huge optimisation, it is not worth + * it as this is definitely a slow code path as it only ever can happen + * once for any given file. + */ ntfs_attr_put_search_ctx(ctx); - unmap_mft_record(ni); - NInoClearTruncateFailed(ni); - ntfs_debug("Done."); - return 0; -err_out: - if (err != -ENOMEM) { + unmap_mft_record(base_ni); + up_write(&ni->runlist.lock); + /* + * Not enough space in the mft record, try to make the attribute + * non-resident and if successful restart the truncation process. + */ + err = ntfs_attr_make_non_resident(ni, old_size); + if (likely(!err)) + goto retry_truncate; + /* + * Could not make non-resident. If this is due to this not being + * permitted for this attribute type or there not being enough space, + * try to make other attributes non-resident. Otherwise fail. + */ + if (unlikely(err != -EPERM && err != -ENOSPC)) { + ntfs_error(vol->sb, "Cannot truncate inode 0x%lx, attribute " + "type 0x%x, because the conversion from " + "resident to non-resident attribute failed " + "with error code %i.", vi->i_ino, + (unsigned)le32_to_cpu(ni->type), err); + if (err != -ENOMEM) + err = -EIO; + goto conv_err_out; + } + /* TODO: Not implemented from here, abort. */ + if (err == -ENOSPC) + ntfs_error(vol->sb, "Not enough space in the mft record/on " + "disk for the non-resident attribute value. " + "This case is not implemented yet."); + else /* if (err == -EPERM) */ + ntfs_error(vol->sb, "This attribute type may not be " + "non-resident. This case is not implemented " + "yet."); + err = -EOPNOTSUPP; + goto conv_err_out; +#if 0 + // TODO: Attempt to make other attributes non-resident. + if (!err) + goto do_resident_extend; + /* + * Both the attribute list attribute and the standard information + * attribute must remain in the base inode. Thus, if this is one of + * these attributes, we have to try to move other attributes out into + * extent mft records instead. + */ + if (ni->type == AT_ATTRIBUTE_LIST || + ni->type == AT_STANDARD_INFORMATION) { + // TODO: Attempt to move other attributes into extent mft + // records. + err = -EOPNOTSUPP; + if (!err) + goto do_resident_extend; + goto err_out; + } + // TODO: Attempt to move this attribute to an extent mft record, but + // only if it is not already the only attribute in an mft record in + // which case there would be nothing to gain. + err = -EOPNOTSUPP; + if (!err) + goto do_resident_extend; + /* There is nothing we can do to make enough space. )-: */ + goto err_out; +#endif +do_non_resident_truncate: + BUG_ON(!NInoNonResident(ni)); + if (alloc_change < 0) { + highest_vcn = sle64_to_cpu(a->data.non_resident.highest_vcn); + if (highest_vcn > 0 && + old_alloc_size >> vol->cluster_size_bits > + highest_vcn + 1) { + /* + * This attribute has multiple extents. Not yet + * supported. + */ + ntfs_error(vol->sb, "Cannot truncate inode 0x%lx, " + "attribute type 0x%x, because the " + "attribute is highly fragmented (it " + "consists of multiple extents) and " + "this case is not implemented yet.", + vi->i_ino, + (unsigned)le32_to_cpu(ni->type)); + err = -EOPNOTSUPP; + goto bad_out; + } + } + /* + * If the size is shrinking, need to reduce the initialized_size and + * the data_size before reducing the allocation. + */ + if (size_change < 0) { + /* + * Make the valid size smaller (i_size is already up-to-date). + */ + write_lock_irqsave(&ni->size_lock, flags); + if (new_size < ni->initialized_size) { + ni->initialized_size = new_size; + a->data.non_resident.initialized_size = + cpu_to_sle64(new_size); + } + a->data.non_resident.data_size = cpu_to_sle64(new_size); + write_unlock_irqrestore(&ni->size_lock, flags); + flush_dcache_mft_record_page(ctx->ntfs_ino); + mark_mft_record_dirty(ctx->ntfs_ino); + /* If the allocated size is not changing, we are done. */ + if (!alloc_change) + goto unm_done; + /* + * If the size is shrinking it makes no sense for the + * allocation to be growing. + */ + BUG_ON(alloc_change > 0); + } else /* if (size_change >= 0) */ { + /* + * The file size is growing or staying the same but the + * allocation can be shrinking, growing or staying the same. + */ + if (alloc_change > 0) { + /* + * We need to extend the allocation and possibly update + * the data size. If we are updating the data size, + * since we are not touching the initialized_size we do + * not need to worry about the actual data on disk. + * And as far as the page cache is concerned, there + * will be no pages beyond the old data size and any + * partial region in the last page between the old and + * new data size (or the end of the page if the new + * data size is outside the page) does not need to be + * modified as explained above for the resident + * attribute truncate case. To do this, we simply drop + * the locks we hold and leave all the work to our + * friendly helper ntfs_attr_extend_allocation(). + */ + ntfs_attr_put_search_ctx(ctx); + unmap_mft_record(base_ni); + up_write(&ni->runlist.lock); + err = ntfs_attr_extend_allocation(ni, new_size, + size_change > 0 ? new_size : -1, -1); + /* + * ntfs_attr_extend_allocation() will have done error + * output already. + */ + goto done; + } + if (!alloc_change) + goto alloc_done; + } + /* alloc_change < 0 */ + /* Free the clusters. */ + nr_freed = ntfs_cluster_free(ni, new_alloc_size >> + vol->cluster_size_bits, -1, ctx); + m = ctx->mrec; + a = ctx->attr; + if (unlikely(nr_freed < 0)) { + ntfs_error(vol->sb, "Failed to release cluster(s) (error code " + "%lli). Unmount and run chkdsk to recover " + "the lost cluster(s).", (long long)nr_freed); NVolSetErrors(vol); + nr_freed = 0; + } + /* Truncate the runlist. */ + err = ntfs_rl_truncate_nolock(vol, &ni->runlist, + new_alloc_size >> vol->cluster_size_bits); + /* + * If the runlist truncation failed and/or the search context is no + * longer valid, we cannot resize the attribute record or build the + * mapping pairs array thus we mark the inode bad so that no access to + * the freed clusters can happen. + */ + if (unlikely(err || IS_ERR(m))) { + ntfs_error(vol->sb, "Failed to %s (error code %li).%s", + IS_ERR(m) ? + "restore attribute search context" : + "truncate attribute runlist", + IS_ERR(m) ? PTR_ERR(m) : err, es); + err = -EIO; + goto bad_out; + } + /* Get the size for the shrunk mapping pairs array for the runlist. */ + mp_size = ntfs_get_size_for_mapping_pairs(vol, ni->runlist.rl, 0, -1); + if (unlikely(mp_size <= 0)) { + ntfs_error(vol->sb, "Cannot shrink allocation of inode 0x%lx, " + "attribute type 0x%x, because determining the " + "size for the mapping pairs failed with error " + "code %i.%s", vi->i_ino, + (unsigned)le32_to_cpu(ni->type), mp_size, es); + err = -EIO; + goto bad_out; + } + /* + * Shrink the attribute record for the new mapping pairs array. Note, + * this cannot fail since we are making the attribute smaller thus by + * definition there is enough space to do so. + */ + attr_len = le32_to_cpu(a->length); + err = ntfs_attr_record_resize(m, a, mp_size + + le16_to_cpu(a->data.non_resident.mapping_pairs_offset)); + BUG_ON(err); + /* + * Generate the mapping pairs array directly into the attribute record. + */ + err = ntfs_mapping_pairs_build(vol, (u8*)a + + le16_to_cpu(a->data.non_resident.mapping_pairs_offset), + mp_size, ni->runlist.rl, 0, -1, NULL); + if (unlikely(err)) { + ntfs_error(vol->sb, "Cannot shrink allocation of inode 0x%lx, " + "attribute type 0x%x, because building the " + "mapping pairs failed with error code %i.%s", + vi->i_ino, (unsigned)le32_to_cpu(ni->type), + err, es); + err = -EIO; + goto bad_out; + } + /* Update the allocated/compressed size as well as the highest vcn. */ + a->data.non_resident.highest_vcn = cpu_to_sle64((new_alloc_size >> + vol->cluster_size_bits) - 1); + write_lock_irqsave(&ni->size_lock, flags); + ni->allocated_size = new_alloc_size; + a->data.non_resident.allocated_size = cpu_to_sle64(new_alloc_size); + if (NInoSparse(ni) || NInoCompressed(ni)) { + if (nr_freed) { + ni->itype.compressed.size -= nr_freed << + vol->cluster_size_bits; + BUG_ON(ni->itype.compressed.size < 0); + a->data.non_resident.compressed_size = cpu_to_sle64( + ni->itype.compressed.size); + vi->i_blocks = ni->itype.compressed.size >> 9; + } + } else + vi->i_blocks = new_alloc_size >> 9; + write_unlock_irqrestore(&ni->size_lock, flags); + /* + * We have shrunk the allocation. If this is a shrinking truncate we + * have already dealt with the initialized_size and the data_size above + * and we are done. If the truncate is only changing the allocation + * and not the data_size, we are also done. If this is an extending + * truncate, need to extend the data_size now which is ensured by the + * fact that @size_change is positive. + */ +alloc_done: + /* + * If the size is growing, need to update it now. If it is shrinking, + * we have already updated it above (before the allocation change). + */ + if (size_change > 0) + a->data.non_resident.data_size = cpu_to_sle64(new_size); + /* Ensure the modified mft record is written out. */ + flush_dcache_mft_record_page(ctx->ntfs_ino); + mark_mft_record_dirty(ctx->ntfs_ino); +unm_done: + ntfs_attr_put_search_ctx(ctx); + unmap_mft_record(base_ni); + up_write(&ni->runlist.lock); +done: + /* Update the mtime and ctime on the base inode. */ + inode_update_time(VFS_I(base_ni), 1); + if (likely(!err)) { + NInoClearTruncateFailed(ni); + ntfs_debug("Done."); + } + return err; +old_bad_out: + old_size = -1; +bad_out: + if (err != -ENOMEM && err != -EOPNOTSUPP) { make_bad_inode(vi); + make_bad_inode(VFS_I(base_ni)); + NVolSetErrors(vol); } + if (err != -EOPNOTSUPP) + NInoSetTruncateFailed(ni); + else if (old_size >= 0) + i_size_write(vi, old_size); +err_out: if (ctx) ntfs_attr_put_search_ctx(ctx); if (m) - unmap_mft_record(ni); - NInoSetTruncateFailed(ni); + unmap_mft_record(base_ni); + up_write(&ni->runlist.lock); +out: + ntfs_debug("Failed. Returning error code %i.", err); return err; +conv_err_out: + if (err != -ENOMEM && err != -EOPNOTSUPP) { + make_bad_inode(vi); + make_bad_inode(VFS_I(base_ni)); + NVolSetErrors(vol); + } + if (err != -EOPNOTSUPP) + NInoSetTruncateFailed(ni); + else + i_size_write(vi, old_size); + goto out; } /** @@ -2420,8 +2845,7 @@ int ntfs_setattr(struct dentry *dentry, struct iattr *attr) err = inode_change_ok(vi, attr); if (err) - return err; - + goto out; /* We do not support NTFS ACLs yet. */ if (ia_valid & (ATTR_UID | ATTR_GID | ATTR_MODE)) { ntfs_warning(vi->i_sb, "Changes in user/group/mode are not " @@ -2429,14 +2853,22 @@ int ntfs_setattr(struct dentry *dentry, struct iattr *attr) err = -EOPNOTSUPP; goto out; } - if (ia_valid & ATTR_SIZE) { if (attr->ia_size != i_size_read(vi)) { - ntfs_warning(vi->i_sb, "Changes in inode size are not " - "supported yet, ignoring."); - err = -EOPNOTSUPP; - // TODO: Implement... - // err = vmtruncate(vi, attr->ia_size); + ntfs_inode *ni = NTFS_I(vi); + /* + * FIXME: For now we do not support resizing of + * compressed or encrypted files yet. + */ + if (NInoCompressed(ni) || NInoEncrypted(ni)) { + ntfs_warning(vi->i_sb, "Changes in inode size " + "are not supported yet for " + "%s files, ignoring.", + NInoCompressed(ni) ? + "compressed" : "encrypted"); + err = -EOPNOTSUPP; + } else + err = vmtruncate(vi, attr->ia_size); if (err || ia_valid == ATTR_SIZE) goto out; } else { diff --git a/fs/ntfs/layout.h b/fs/ntfs/layout.h index 609ad1728ce4..f5678d5d7919 100644 --- a/fs/ntfs/layout.h +++ b/fs/ntfs/layout.h @@ -123,7 +123,7 @@ enum { magic_RCRD = const_cpu_to_le32(0x44524352), /* Log record page. */ /* Found in $LogFile/$DATA. (May be found in $MFT/$DATA, also?) */ - magic_CHKD = const_cpu_to_le32(0x424b4843), /* Modified by chkdsk. */ + magic_CHKD = const_cpu_to_le32(0x444b4843), /* Modified by chkdsk. */ /* Found in all ntfs record containing records. */ magic_BAAD = const_cpu_to_le32(0x44414142), /* Failed multi sector @@ -308,10 +308,8 @@ typedef le16 MFT_RECORD_FLAGS; * The _LE versions are to be applied on little endian MFT_REFs. * Note: The _LE versions will return a CPU endian formatted value! */ -typedef enum { - MFT_REF_MASK_CPU = 0x0000ffffffffffffULL, - MFT_REF_MASK_LE = const_cpu_to_le64(0x0000ffffffffffffULL), -} MFT_REF_CONSTS; +#define MFT_REF_MASK_CPU 0x0000ffffffffffffULL +#define MFT_REF_MASK_LE const_cpu_to_le64(MFT_REF_MASK_CPU) typedef u64 MFT_REF; typedef le64 leMFT_REF; @@ -1023,10 +1021,17 @@ enum { FILE_NAME_POSIX = 0x00, /* This is the largest namespace. It is case sensitive and allows all Unicode characters except for: '\0' and '/'. Beware that in - WinNT/2k files which eg have the same name except for their case - will not be distinguished by the standard utilities and thus a "del - filename" will delete both "filename" and "fileName" without - warning. */ + WinNT/2k/2003 by default files which eg have the same name except + for their case will not be distinguished by the standard utilities + and thus a "del filename" will delete both "filename" and "fileName" + without warning. However if for example Services For Unix (SFU) are + installed and the case sensitive option was enabled at installation + time, then you can create/access/delete such files. + Note that even SFU places restrictions on the filenames beyond the + '\0' and '/' and in particular the following set of characters is + not allowed: '"', '/', '<', '>', '\'. All other characters, + including the ones no allowed in WIN32 namespace are allowed. + Tested with SFU 3.5 (this is now free) running on Windows XP. */ FILE_NAME_WIN32 = 0x01, /* The standard WinNT/2k NTFS long filenames. Case insensitive. All Unicode chars except: '\0', '"', '*', '/', ':', '<', '>', '?', '\', @@ -2369,7 +2374,9 @@ typedef struct { * Extended attribute flags (8-bit). */ enum { - NEED_EA = 0x80 + NEED_EA = 0x80 /* If set the file to which the EA belongs + cannot be interpreted without understanding + the associates extended attributes. */ } __attribute__ ((__packed__)); typedef u8 EA_FLAGS; @@ -2377,20 +2384,20 @@ typedef u8 EA_FLAGS; /* * Attribute: Extended attribute (EA) (0xe0). * - * NOTE: Always non-resident. (Is this true?) + * NOTE: Can be resident or non-resident. * * Like the attribute list and the index buffer list, the EA attribute value is * a sequence of EA_ATTR variable length records. - * - * FIXME: It appears weird that the EA name is not unicode. Is it true? */ typedef struct { le32 next_entry_offset; /* Offset to the next EA_ATTR. */ EA_FLAGS flags; /* Flags describing the EA. */ - u8 ea_name_length; /* Length of the name of the EA in bytes. */ + u8 ea_name_length; /* Length of the name of the EA in bytes + excluding the '\0' byte terminator. */ le16 ea_value_length; /* Byte size of the EA's value. */ - u8 ea_name[0]; /* Name of the EA. */ - u8 ea_value[0]; /* The value of the EA. Immediately follows + u8 ea_name[0]; /* Name of the EA. Note this is ASCII, not + Unicode and it is zero terminated. */ + u8 ea_value[0]; /* The value of the EA. Immediately follows the name. */ } __attribute__ ((__packed__)) EA_ATTR; diff --git a/fs/ntfs/lcnalloc.c b/fs/ntfs/lcnalloc.c index 7b5934290685..29cabf93d2d2 100644 --- a/fs/ntfs/lcnalloc.c +++ b/fs/ntfs/lcnalloc.c @@ -76,6 +76,7 @@ int ntfs_cluster_free_from_rl_nolock(ntfs_volume *vol, * @count: number of clusters to allocate * @start_lcn: starting lcn at which to allocate the clusters (or -1 if none) * @zone: zone from which to allocate the clusters + * @is_extension: if TRUE, this is an attribute extension * * Allocate @count clusters preferably starting at cluster @start_lcn or at the * current allocator position if @start_lcn is -1, on the mounted ntfs volume @@ -86,6 +87,13 @@ int ntfs_cluster_free_from_rl_nolock(ntfs_volume *vol, * @start_vcn specifies the vcn of the first allocated cluster. This makes * merging the resulting runlist with the old runlist easier. * + * If @is_extension is TRUE, the caller is allocating clusters to extend an + * attribute and if it is FALSE, the caller is allocating clusters to fill a + * hole in an attribute. Practically the difference is that if @is_extension + * is TRUE the returned runlist will be terminated with LCN_ENOENT and if + * @is_extension is FALSE the runlist will be terminated with + * LCN_RL_NOT_MAPPED. + * * You need to check the return value with IS_ERR(). If this is false, the * function was successful and the return value is a runlist describing the * allocated cluster(s). If IS_ERR() is true, the function failed and @@ -137,7 +145,8 @@ int ntfs_cluster_free_from_rl_nolock(ntfs_volume *vol, */ runlist_element *ntfs_cluster_alloc(ntfs_volume *vol, const VCN start_vcn, const s64 count, const LCN start_lcn, - const NTFS_CLUSTER_ALLOCATION_ZONES zone) + const NTFS_CLUSTER_ALLOCATION_ZONES zone, + const BOOL is_extension) { LCN zone_start, zone_end, bmp_pos, bmp_initial_pos, last_read_pos, lcn; LCN prev_lcn = 0, prev_run_len = 0, mft_zone_size; @@ -310,7 +319,7 @@ runlist_element *ntfs_cluster_alloc(ntfs_volume *vol, const VCN start_vcn, continue; } bit = 1 << (lcn & 7); - ntfs_debug("bit %i.", bit); + ntfs_debug("bit 0x%x.", bit); /* If the bit is already set, go onto the next one. */ if (*byte & bit) { lcn++; @@ -729,7 +738,7 @@ out: /* Add runlist terminator element. */ if (likely(rl)) { rl[rlpos].vcn = rl[rlpos - 1].vcn + rl[rlpos - 1].length; - rl[rlpos].lcn = LCN_RL_NOT_MAPPED; + rl[rlpos].lcn = is_extension ? LCN_ENOENT : LCN_RL_NOT_MAPPED; rl[rlpos].length = 0; } if (likely(page && !IS_ERR(page))) { @@ -779,53 +788,78 @@ out: /** * __ntfs_cluster_free - free clusters on an ntfs volume - * @vi: vfs inode whose runlist describes the clusters to free - * @start_vcn: vcn in the runlist of @vi at which to start freeing clusters + * @ni: ntfs inode whose runlist describes the clusters to free + * @start_vcn: vcn in the runlist of @ni at which to start freeing clusters * @count: number of clusters to free or -1 for all clusters - * @write_locked: true if the runlist is locked for writing + * @ctx: active attribute search context if present or NULL if not * @is_rollback: true if this is a rollback operation * * Free @count clusters starting at the cluster @start_vcn in the runlist - * described by the vfs inode @vi. + * described by the vfs inode @ni. * * If @count is -1, all clusters from @start_vcn to the end of the runlist are * deallocated. Thus, to completely free all clusters in a runlist, use * @start_vcn = 0 and @count = -1. * + * If @ctx is specified, it is an active search context of @ni and its base mft + * record. This is needed when __ntfs_cluster_free() encounters unmapped + * runlist fragments and allows their mapping. If you do not have the mft + * record mapped, you can specify @ctx as NULL and __ntfs_cluster_free() will + * perform the necessary mapping and unmapping. + * + * Note, __ntfs_cluster_free() saves the state of @ctx on entry and restores it + * before returning. Thus, @ctx will be left pointing to the same attribute on + * return as on entry. However, the actual pointers in @ctx may point to + * different memory locations on return, so you must remember to reset any + * cached pointers from the @ctx, i.e. after the call to __ntfs_cluster_free(), + * you will probably want to do: + * m = ctx->mrec; + * a = ctx->attr; + * Assuming you cache ctx->attr in a variable @a of type ATTR_RECORD * and that + * you cache ctx->mrec in a variable @m of type MFT_RECORD *. + * * @is_rollback should always be FALSE, it is for internal use to rollback * errors. You probably want to use ntfs_cluster_free() instead. * - * Note, ntfs_cluster_free() does not modify the runlist at all, so the caller - * has to deal with it later. + * Note, __ntfs_cluster_free() does not modify the runlist, so you have to + * remove from the runlist or mark sparse the freed runs later. * * Return the number of deallocated clusters (not counting sparse ones) on * success and -errno on error. * - * Locking: - The runlist described by @vi must be locked on entry and is - * locked on return. Note if the runlist is locked for reading the - * lock may be dropped and reacquired. Note the runlist may be - * modified when needed runlist fragments need to be mapped. + * WARNING: If @ctx is supplied, regardless of whether success or failure is + * returned, you need to check IS_ERR(@ctx->mrec) and if TRUE the @ctx + * is no longer valid, i.e. you need to either call + * ntfs_attr_reinit_search_ctx() or ntfs_attr_put_search_ctx() on it. + * In that case PTR_ERR(@ctx->mrec) will give you the error code for + * why the mapping of the old inode failed. + * + * Locking: - The runlist described by @ni must be locked for writing on entry + * and is locked on return. Note the runlist may be modified when + * needed runlist fragments need to be mapped. * - The volume lcn bitmap must be unlocked on entry and is unlocked * on return. * - This function takes the volume lcn bitmap lock for writing and * modifies the bitmap contents. + * - If @ctx is NULL, the base mft record of @ni must not be mapped on + * entry and it will be left unmapped on return. + * - If @ctx is not NULL, the base mft record must be mapped on entry + * and it will be left mapped on return. */ -s64 __ntfs_cluster_free(struct inode *vi, const VCN start_vcn, s64 count, - const BOOL write_locked, const BOOL is_rollback) +s64 __ntfs_cluster_free(ntfs_inode *ni, const VCN start_vcn, s64 count, + ntfs_attr_search_ctx *ctx, const BOOL is_rollback) { s64 delta, to_free, total_freed, real_freed; - ntfs_inode *ni; ntfs_volume *vol; struct inode *lcnbmp_vi; runlist_element *rl; int err; - BUG_ON(!vi); + BUG_ON(!ni); ntfs_debug("Entering for i_ino 0x%lx, start_vcn 0x%llx, count " - "0x%llx.%s", vi->i_ino, (unsigned long long)start_vcn, + "0x%llx.%s", ni->mft_no, (unsigned long long)start_vcn, (unsigned long long)count, is_rollback ? " (rollback)" : ""); - ni = NTFS_I(vi); vol = ni->vol; lcnbmp_vi = vol->lcnbmp_ino; BUG_ON(!lcnbmp_vi); @@ -843,7 +877,7 @@ s64 __ntfs_cluster_free(struct inode *vi, const VCN start_vcn, s64 count, total_freed = real_freed = 0; - rl = ntfs_attr_find_vcn_nolock(ni, start_vcn, write_locked); + rl = ntfs_attr_find_vcn_nolock(ni, start_vcn, ctx); if (IS_ERR(rl)) { if (!is_rollback) ntfs_error(vol->sb, "Failed to find first runlist " @@ -897,7 +931,7 @@ s64 __ntfs_cluster_free(struct inode *vi, const VCN start_vcn, s64 count, /* Attempt to map runlist. */ vcn = rl->vcn; - rl = ntfs_attr_find_vcn_nolock(ni, vcn, write_locked); + rl = ntfs_attr_find_vcn_nolock(ni, vcn, ctx); if (IS_ERR(rl)) { err = PTR_ERR(rl); if (!is_rollback) @@ -965,8 +999,7 @@ err_out: * If rollback fails, set the volume errors flag, emit an error * message, and return the error code. */ - delta = __ntfs_cluster_free(vi, start_vcn, total_freed, write_locked, - TRUE); + delta = __ntfs_cluster_free(ni, start_vcn, total_freed, ctx, TRUE); if (delta < 0) { ntfs_error(vol->sb, "Failed to rollback (error %i). Leaving " "inconsistent metadata! Unmount and run " diff --git a/fs/ntfs/lcnalloc.h b/fs/ntfs/lcnalloc.h index e4d7fb98d685..72cbca7003b2 100644 --- a/fs/ntfs/lcnalloc.h +++ b/fs/ntfs/lcnalloc.h @@ -2,7 +2,7 @@ * lcnalloc.h - Exports for NTFS kernel cluster (de)allocation. Part of the * Linux-NTFS project. * - * Copyright (c) 2004 Anton Altaparmakov + * Copyright (c) 2004-2005 Anton Altaparmakov * * This program/include file is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License as published @@ -27,7 +27,9 @@ #include <linux/fs.h> +#include "attrib.h" #include "types.h" +#include "inode.h" #include "runlist.h" #include "volume.h" @@ -40,44 +42,72 @@ typedef enum { extern runlist_element *ntfs_cluster_alloc(ntfs_volume *vol, const VCN start_vcn, const s64 count, const LCN start_lcn, - const NTFS_CLUSTER_ALLOCATION_ZONES zone); + const NTFS_CLUSTER_ALLOCATION_ZONES zone, + const BOOL is_extension); -extern s64 __ntfs_cluster_free(struct inode *vi, const VCN start_vcn, - s64 count, const BOOL write_locked, const BOOL is_rollback); +extern s64 __ntfs_cluster_free(ntfs_inode *ni, const VCN start_vcn, + s64 count, ntfs_attr_search_ctx *ctx, const BOOL is_rollback); /** * ntfs_cluster_free - free clusters on an ntfs volume - * @vi: vfs inode whose runlist describes the clusters to free - * @start_vcn: vcn in the runlist of @vi at which to start freeing clusters + * @ni: ntfs inode whose runlist describes the clusters to free + * @start_vcn: vcn in the runlist of @ni at which to start freeing clusters * @count: number of clusters to free or -1 for all clusters - * @write_locked: true if the runlist is locked for writing + * @ctx: active attribute search context if present or NULL if not * * Free @count clusters starting at the cluster @start_vcn in the runlist - * described by the vfs inode @vi. + * described by the ntfs inode @ni. * * If @count is -1, all clusters from @start_vcn to the end of the runlist are * deallocated. Thus, to completely free all clusters in a runlist, use * @start_vcn = 0 and @count = -1. * - * Note, ntfs_cluster_free() does not modify the runlist at all, so the caller - * has to deal with it later. + * If @ctx is specified, it is an active search context of @ni and its base mft + * record. This is needed when ntfs_cluster_free() encounters unmapped runlist + * fragments and allows their mapping. If you do not have the mft record + * mapped, you can specify @ctx as NULL and ntfs_cluster_free() will perform + * the necessary mapping and unmapping. + * + * Note, ntfs_cluster_free() saves the state of @ctx on entry and restores it + * before returning. Thus, @ctx will be left pointing to the same attribute on + * return as on entry. However, the actual pointers in @ctx may point to + * different memory locations on return, so you must remember to reset any + * cached pointers from the @ctx, i.e. after the call to ntfs_cluster_free(), + * you will probably want to do: + * m = ctx->mrec; + * a = ctx->attr; + * Assuming you cache ctx->attr in a variable @a of type ATTR_RECORD * and that + * you cache ctx->mrec in a variable @m of type MFT_RECORD *. + * + * Note, ntfs_cluster_free() does not modify the runlist, so you have to remove + * from the runlist or mark sparse the freed runs later. * * Return the number of deallocated clusters (not counting sparse ones) on * success and -errno on error. * - * Locking: - The runlist described by @vi must be locked on entry and is - * locked on return. Note if the runlist is locked for reading the - * lock may be dropped and reacquired. Note the runlist may be - * modified when needed runlist fragments need to be mapped. + * WARNING: If @ctx is supplied, regardless of whether success or failure is + * returned, you need to check IS_ERR(@ctx->mrec) and if TRUE the @ctx + * is no longer valid, i.e. you need to either call + * ntfs_attr_reinit_search_ctx() or ntfs_attr_put_search_ctx() on it. + * In that case PTR_ERR(@ctx->mrec) will give you the error code for + * why the mapping of the old inode failed. + * + * Locking: - The runlist described by @ni must be locked for writing on entry + * and is locked on return. Note the runlist may be modified when + * needed runlist fragments need to be mapped. * - The volume lcn bitmap must be unlocked on entry and is unlocked * on return. * - This function takes the volume lcn bitmap lock for writing and * modifies the bitmap contents. + * - If @ctx is NULL, the base mft record of @ni must not be mapped on + * entry and it will be left unmapped on return. + * - If @ctx is not NULL, the base mft record must be mapped on entry + * and it will be left mapped on return. */ -static inline s64 ntfs_cluster_free(struct inode *vi, const VCN start_vcn, - s64 count, const BOOL write_locked) +static inline s64 ntfs_cluster_free(ntfs_inode *ni, const VCN start_vcn, + s64 count, ntfs_attr_search_ctx *ctx) { - return __ntfs_cluster_free(vi, start_vcn, count, write_locked, FALSE); + return __ntfs_cluster_free(ni, start_vcn, count, ctx, FALSE); } extern int ntfs_cluster_free_from_rl_nolock(ntfs_volume *vol, diff --git a/fs/ntfs/logfile.c b/fs/ntfs/logfile.c index 0173e95500d9..0fd70295cca6 100644 --- a/fs/ntfs/logfile.c +++ b/fs/ntfs/logfile.c @@ -51,7 +51,8 @@ static BOOL ntfs_check_restart_page_header(struct inode *vi, RESTART_PAGE_HEADER *rp, s64 pos) { u32 logfile_system_page_size, logfile_log_page_size; - u16 usa_count, usa_ofs, usa_end, ra_ofs; + u16 ra_ofs, usa_count, usa_ofs, usa_end = 0; + BOOL have_usa = TRUE; ntfs_debug("Entering."); /* @@ -86,6 +87,14 @@ static BOOL ntfs_check_restart_page_header(struct inode *vi, (int)sle16_to_cpu(rp->minor_ver)); return FALSE; } + /* + * If chkdsk has been run the restart page may not be protected by an + * update sequence array. + */ + if (ntfs_is_chkd_record(rp->magic) && !le16_to_cpu(rp->usa_count)) { + have_usa = FALSE; + goto skip_usa_checks; + } /* Verify the size of the update sequence array. */ usa_count = 1 + (logfile_system_page_size >> NTFS_BLOCK_SIZE_BITS); if (usa_count != le16_to_cpu(rp->usa_count)) { @@ -102,6 +111,7 @@ static BOOL ntfs_check_restart_page_header(struct inode *vi, "inconsistent update sequence array offset."); return FALSE; } +skip_usa_checks: /* * Verify the position of the restart area. It must be: * - aligned to 8-byte boundary, @@ -109,7 +119,8 @@ static BOOL ntfs_check_restart_page_header(struct inode *vi, * - within the system page size. */ ra_ofs = le16_to_cpu(rp->restart_area_offset); - if (ra_ofs & 7 || ra_ofs < usa_end || + if (ra_ofs & 7 || (have_usa ? ra_ofs < usa_end : + ra_ofs < sizeof(RESTART_PAGE_HEADER)) || ra_ofs > logfile_system_page_size) { ntfs_error(vi->i_sb, "$LogFile restart page specifies " "inconsistent restart area offset."); @@ -402,8 +413,12 @@ static int ntfs_check_and_load_restart_page(struct inode *vi, idx++; } while (to_read > 0); } - /* Perform the multi sector transfer deprotection on the buffer. */ - if (post_read_mst_fixup((NTFS_RECORD*)trp, + /* + * Perform the multi sector transfer deprotection on the buffer if the + * restart page is protected. + */ + if ((!ntfs_is_chkd_record(trp->magic) || le16_to_cpu(trp->usa_count)) + && post_read_mst_fixup((NTFS_RECORD*)trp, le32_to_cpu(rp->system_page_size))) { /* * A multi sector tranfer error was detected. We only need to @@ -615,11 +630,16 @@ is_empty: * Otherwise just throw it away. */ if (rstr2_lsn > rstr1_lsn) { + ntfs_debug("Using second restart page as it is more " + "recent."); ntfs_free(rstr1_ph); rstr1_ph = rstr2_ph; /* rstr1_lsn = rstr2_lsn; */ - } else + } else { + ntfs_debug("Using first restart page as it is more " + "recent."); ntfs_free(rstr2_ph); + } rstr2_ph = NULL; } /* All consistency checks passed. */ diff --git a/fs/ntfs/logfile.h b/fs/ntfs/logfile.h index 42388f95ea6d..a51f3dd0e9eb 100644 --- a/fs/ntfs/logfile.h +++ b/fs/ntfs/logfile.h @@ -113,7 +113,7 @@ typedef struct { */ enum { RESTART_VOLUME_IS_CLEAN = const_cpu_to_le16(0x0002), - RESTART_SPACE_FILLER = 0xffff, /* gcc: Force enum bit width to 16. */ + RESTART_SPACE_FILLER = const_cpu_to_le16(0xffff), /* gcc: Force enum bit width to 16. */ } __attribute__ ((__packed__)); typedef le16 RESTART_AREA_FLAGS; diff --git a/fs/ntfs/malloc.h b/fs/ntfs/malloc.h index 006946efca8c..e38e402e4103 100644 --- a/fs/ntfs/malloc.h +++ b/fs/ntfs/malloc.h @@ -39,8 +39,7 @@ * If there was insufficient memory to complete the request, return NULL. * Depending on @gfp_mask the allocation may be guaranteed to succeed. */ -static inline void *__ntfs_malloc(unsigned long size, - unsigned int __nocast gfp_mask) +static inline void *__ntfs_malloc(unsigned long size, gfp_t gfp_mask) { if (likely(size <= PAGE_SIZE)) { BUG_ON(!size); diff --git a/fs/ntfs/mft.c b/fs/ntfs/mft.c index 2c32b84385a8..0c65cbb8c5cf 100644 --- a/fs/ntfs/mft.c +++ b/fs/ntfs/mft.c @@ -49,7 +49,8 @@ static inline MFT_RECORD *map_mft_record_page(ntfs_inode *ni) ntfs_volume *vol = ni->vol; struct inode *mft_vi = vol->mft_ino; struct page *page; - unsigned long index, ofs, end_index; + unsigned long index, end_index; + unsigned ofs; BUG_ON(ni->page); /* @@ -58,7 +59,8 @@ static inline MFT_RECORD *map_mft_record_page(ntfs_inode *ni) * overflowing the unsigned long, but I don't think we would ever get * here if the volume was that big... */ - index = ni->mft_no << vol->mft_record_size_bits >> PAGE_CACHE_SHIFT; + index = (u64)ni->mft_no << vol->mft_record_size_bits >> + PAGE_CACHE_SHIFT; ofs = (ni->mft_no << vol->mft_record_size_bits) & ~PAGE_CACHE_MASK; i_size = i_size_read(mft_vi); @@ -1307,7 +1309,7 @@ static int ntfs_mft_bitmap_extend_allocation_nolock(ntfs_volume *vol) ll = mftbmp_ni->allocated_size; read_unlock_irqrestore(&mftbmp_ni->size_lock, flags); rl = ntfs_attr_find_vcn_nolock(mftbmp_ni, - (ll - 1) >> vol->cluster_size_bits, TRUE); + (ll - 1) >> vol->cluster_size_bits, NULL); if (unlikely(IS_ERR(rl) || !rl->length || rl->lcn < 0)) { up_write(&mftbmp_ni->runlist.lock); ntfs_error(vol->sb, "Failed to determine last allocated " @@ -1353,7 +1355,8 @@ static int ntfs_mft_bitmap_extend_allocation_nolock(ntfs_volume *vol) up_write(&vol->lcnbmp_lock); ntfs_unmap_page(page); /* Allocate a cluster from the DATA_ZONE. */ - rl2 = ntfs_cluster_alloc(vol, rl[1].vcn, 1, lcn, DATA_ZONE); + rl2 = ntfs_cluster_alloc(vol, rl[1].vcn, 1, lcn, DATA_ZONE, + TRUE); if (IS_ERR(rl2)) { up_write(&mftbmp_ni->runlist.lock); ntfs_error(vol->sb, "Failed to allocate a cluster for " @@ -1737,7 +1740,7 @@ static int ntfs_mft_data_extend_allocation_nolock(ntfs_volume *vol) ll = mft_ni->allocated_size; read_unlock_irqrestore(&mft_ni->size_lock, flags); rl = ntfs_attr_find_vcn_nolock(mft_ni, - (ll - 1) >> vol->cluster_size_bits, TRUE); + (ll - 1) >> vol->cluster_size_bits, NULL); if (unlikely(IS_ERR(rl) || !rl->length || rl->lcn < 0)) { up_write(&mft_ni->runlist.lock); ntfs_error(vol->sb, "Failed to determine last allocated " @@ -1778,7 +1781,8 @@ static int ntfs_mft_data_extend_allocation_nolock(ntfs_volume *vol) nr > min_nr ? "default" : "minimal", (long long)nr); old_last_vcn = rl[1].vcn; do { - rl2 = ntfs_cluster_alloc(vol, old_last_vcn, nr, lcn, MFT_ZONE); + rl2 = ntfs_cluster_alloc(vol, old_last_vcn, nr, lcn, MFT_ZONE, + TRUE); if (likely(!IS_ERR(rl2))) break; if (PTR_ERR(rl2) != -ENOSPC || nr == min_nr) { @@ -1950,20 +1954,21 @@ restore_undo_alloc: NVolSetErrors(vol); return ret; } - a = ctx->attr; - a->data.non_resident.highest_vcn = cpu_to_sle64(old_last_vcn - 1); + ctx->attr->data.non_resident.highest_vcn = + cpu_to_sle64(old_last_vcn - 1); undo_alloc: - if (ntfs_cluster_free(vol->mft_ino, old_last_vcn, -1, TRUE) < 0) { + if (ntfs_cluster_free(mft_ni, old_last_vcn, -1, ctx) < 0) { ntfs_error(vol->sb, "Failed to free clusters from mft data " "attribute.%s", es); NVolSetErrors(vol); } + a = ctx->attr; if (ntfs_rl_truncate_nolock(vol, &mft_ni->runlist, old_last_vcn)) { ntfs_error(vol->sb, "Failed to truncate mft data attribute " "runlist.%s", es); NVolSetErrors(vol); } - if (mp_rebuilt) { + if (mp_rebuilt && !IS_ERR(ctx->mrec)) { if (ntfs_mapping_pairs_build(vol, (u8*)a + le16_to_cpu( a->data.non_resident.mapping_pairs_offset), old_alen - le16_to_cpu( @@ -1980,6 +1985,10 @@ undo_alloc: } flush_dcache_mft_record_page(ctx->ntfs_ino); mark_mft_record_dirty(ctx->ntfs_ino); + } else if (IS_ERR(ctx->mrec)) { + ntfs_error(vol->sb, "Failed to restore attribute search " + "context.%s", es); + NVolSetErrors(vol); } if (ctx) ntfs_attr_put_search_ctx(ctx); diff --git a/fs/ntfs/super.c b/fs/ntfs/super.c index 453d0d51ea4b..6c16db9e1a8a 100644 --- a/fs/ntfs/super.c +++ b/fs/ntfs/super.c @@ -1447,7 +1447,7 @@ not_enabled: if (unlikely(i_size_read(tmp_ino) < sizeof(USN_HEADER))) { ntfs_error(vol->sb, "Found corrupt $UsnJrnl/$DATA/$Max " "attribute (size is 0x%llx but should be at " - "least 0x%x bytes).", i_size_read(tmp_ino), + "least 0x%zx bytes).", i_size_read(tmp_ino), sizeof(USN_HEADER)); return FALSE; } diff --git a/fs/ntfs/unistr.c b/fs/ntfs/unistr.c index a389a5a16c84..0ea887fc859c 100644 --- a/fs/ntfs/unistr.c +++ b/fs/ntfs/unistr.c @@ -1,7 +1,7 @@ /* * unistr.c - NTFS Unicode string handling. Part of the Linux-NTFS project. * - * Copyright (c) 2001-2004 Anton Altaparmakov + * Copyright (c) 2001-2005 Anton Altaparmakov * * This program/include file is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License as published diff --git a/fs/open.c b/fs/open.c index f0d90cf0495c..f53a5b9ffb7d 100644 --- a/fs/open.c +++ b/fs/open.c @@ -194,7 +194,7 @@ out: return error; } -int do_truncate(struct dentry *dentry, loff_t length) +int do_truncate(struct dentry *dentry, loff_t length, struct file *filp) { int err; struct iattr newattrs; @@ -205,6 +205,10 @@ int do_truncate(struct dentry *dentry, loff_t length) newattrs.ia_size = length; newattrs.ia_valid = ATTR_SIZE | ATTR_CTIME; + if (filp) { + newattrs.ia_file = filp; + newattrs.ia_valid |= ATTR_FILE; + } down(&dentry->d_inode->i_sem); err = notify_change(dentry, &newattrs); @@ -236,7 +240,7 @@ static inline long do_sys_truncate(const char __user * path, loff_t length) if (!S_ISREG(inode->i_mode)) goto dput_and_out; - error = permission(inode,MAY_WRITE,&nd); + error = vfs_permission(&nd, MAY_WRITE); if (error) goto dput_and_out; @@ -262,7 +266,7 @@ static inline long do_sys_truncate(const char __user * path, loff_t length) error = locks_verify_truncate(inode, NULL, length); if (!error) { DQUOT_INIT(inode); - error = do_truncate(nd.dentry, length); + error = do_truncate(nd.dentry, length, NULL); } put_write_access(inode); @@ -314,7 +318,7 @@ static inline long do_sys_ftruncate(unsigned int fd, loff_t length, int small) error = locks_verify_truncate(inode, file, length); if (!error) - error = do_truncate(dentry, length); + error = do_truncate(dentry, length, file); out_putf: fput(file); out: @@ -390,7 +394,7 @@ asmlinkage long sys_utime(char __user * filename, struct utimbuf __user * times) goto dput_and_out; if (current->fsuid != inode->i_uid && - (error = permission(inode,MAY_WRITE,&nd)) != 0) + (error = vfs_permission(&nd, MAY_WRITE)) != 0) goto dput_and_out; } down(&inode->i_sem); @@ -443,7 +447,7 @@ long do_utimes(char __user * filename, struct timeval * times) goto dput_and_out; if (current->fsuid != inode->i_uid && - (error = permission(inode,MAY_WRITE,&nd)) != 0) + (error = vfs_permission(&nd, MAY_WRITE)) != 0) goto dput_and_out; } down(&inode->i_sem); @@ -502,7 +506,7 @@ asmlinkage long sys_access(const char __user * filename, int mode) res = __user_walk(filename, LOOKUP_FOLLOW|LOOKUP_ACCESS, &nd); if (!res) { - res = permission(nd.dentry->d_inode, mode, &nd); + res = vfs_permission(&nd, mode); /* SuS v2 requires we report a read only fs too */ if(!res && (mode & S_IWOTH) && IS_RDONLY(nd.dentry->d_inode) && !special_file(nd.dentry->d_inode->i_mode)) @@ -526,7 +530,7 @@ asmlinkage long sys_chdir(const char __user * filename) if (error) goto out; - error = permission(nd.dentry->d_inode,MAY_EXEC,&nd); + error = vfs_permission(&nd, MAY_EXEC); if (error) goto dput_and_out; @@ -559,7 +563,7 @@ asmlinkage long sys_fchdir(unsigned int fd) if (!S_ISDIR(inode->i_mode)) goto out_putf; - error = permission(inode, MAY_EXEC, NULL); + error = file_permission(file, MAY_EXEC); if (!error) set_fs_pwd(current->fs, mnt, dentry); out_putf: @@ -577,7 +581,7 @@ asmlinkage long sys_chroot(const char __user * filename) if (error) goto out; - error = permission(nd.dentry->d_inode,MAY_EXEC,&nd); + error = vfs_permission(&nd, MAY_EXEC); if (error) goto dput_and_out; @@ -739,7 +743,8 @@ asmlinkage long sys_fchown(unsigned int fd, uid_t user, gid_t group) } static struct file *__dentry_open(struct dentry *dentry, struct vfsmount *mnt, - int flags, struct file *f) + int flags, struct file *f, + int (*open)(struct inode *, struct file *)) { struct inode *inode; int error; @@ -761,11 +766,14 @@ static struct file *__dentry_open(struct dentry *dentry, struct vfsmount *mnt, f->f_op = fops_get(inode->i_fop); file_move(f, &inode->i_sb->s_files); - if (f->f_op && f->f_op->open) { - error = f->f_op->open(inode,f); + if (!open && f->f_op) + open = f->f_op->open; + if (open) { + error = open(inode, f); if (error) goto cleanup_all; } + f->f_flags &= ~(O_CREAT | O_EXCL | O_NOCTTY | O_TRUNC); file_ra_state_init(&f->f_ra, f->f_mapping->host->i_mapping); @@ -814,28 +822,79 @@ struct file *filp_open(const char * filename, int flags, int mode) { int namei_flags, error; struct nameidata nd; - struct file *f; namei_flags = flags; if ((namei_flags+1) & O_ACCMODE) namei_flags++; - if (namei_flags & O_TRUNC) - namei_flags |= 2; - - error = -ENFILE; - f = get_empty_filp(); - if (f == NULL) - return ERR_PTR(error); error = open_namei(filename, namei_flags, mode, &nd); if (!error) - return __dentry_open(nd.dentry, nd.mnt, flags, f); + return nameidata_to_filp(&nd, flags); - put_filp(f); return ERR_PTR(error); } EXPORT_SYMBOL(filp_open); +/** + * lookup_instantiate_filp - instantiates the open intent filp + * @nd: pointer to nameidata + * @dentry: pointer to dentry + * @open: open callback + * + * Helper for filesystems that want to use lookup open intents and pass back + * a fully instantiated struct file to the caller. + * This function is meant to be called from within a filesystem's + * lookup method. + * Note that in case of error, nd->intent.open.file is destroyed, but the + * path information remains valid. + * If the open callback is set to NULL, then the standard f_op->open() + * filesystem callback is substituted. + */ +struct file *lookup_instantiate_filp(struct nameidata *nd, struct dentry *dentry, + int (*open)(struct inode *, struct file *)) +{ + if (IS_ERR(nd->intent.open.file)) + goto out; + if (IS_ERR(dentry)) + goto out_err; + nd->intent.open.file = __dentry_open(dget(dentry), mntget(nd->mnt), + nd->intent.open.flags - 1, + nd->intent.open.file, + open); +out: + return nd->intent.open.file; +out_err: + release_open_intent(nd); + nd->intent.open.file = (struct file *)dentry; + goto out; +} +EXPORT_SYMBOL_GPL(lookup_instantiate_filp); + +/** + * nameidata_to_filp - convert a nameidata to an open filp. + * @nd: pointer to nameidata + * @flags: open flags + * + * Note that this function destroys the original nameidata + */ +struct file *nameidata_to_filp(struct nameidata *nd, int flags) +{ + struct file *filp; + + /* Pick up the filp from the open intent */ + filp = nd->intent.open.file; + /* Has the filesystem initialised the file for us? */ + if (filp->f_dentry == NULL) + filp = __dentry_open(nd->dentry, nd->mnt, flags, filp, NULL); + else + path_release(nd); + return filp; +} + +/* + * dentry_open() will have done dput(dentry) and mntput(mnt) if it returns an + * error. + */ struct file *dentry_open(struct dentry *dentry, struct vfsmount *mnt, int flags) { int error; @@ -843,10 +902,13 @@ struct file *dentry_open(struct dentry *dentry, struct vfsmount *mnt, int flags) error = -ENFILE; f = get_empty_filp(); - if (f == NULL) + if (f == NULL) { + dput(dentry); + mntput(mnt); return ERR_PTR(error); + } - return __dentry_open(dentry, mnt, flags, f); + return __dentry_open(dentry, mnt, flags, f, NULL); } EXPORT_SYMBOL(dentry_open); diff --git a/fs/openpromfs/inode.c b/fs/openpromfs/inode.c index 1be11ce96b0f..aeb0106890e4 100644 --- a/fs/openpromfs/inode.c +++ b/fs/openpromfs/inode.c @@ -1088,8 +1088,7 @@ static void __exit exit_openprom_fs(void) unregister_filesystem(&openprom_fs_type); free_pages ((unsigned long)nodes, alloced); for (i = 0; i < aliases_nodes; i++) - if (alias_names [i]) - kfree (alias_names [i]); + kfree (alias_names [i]); nodes = NULL; } diff --git a/fs/partitions/check.c b/fs/partitions/check.c index 77e178f13162..8dc1822a7022 100644 --- a/fs/partitions/check.c +++ b/fs/partitions/check.c @@ -192,6 +192,7 @@ check_partition(struct gendisk *hd, struct block_device *bdev) struct part_attribute { struct attribute attr; ssize_t (*show)(struct hd_struct *,char *); + ssize_t (*store)(struct hd_struct *,const char *, size_t); }; static ssize_t @@ -201,14 +202,33 @@ part_attr_show(struct kobject * kobj, struct attribute * attr, char * page) struct part_attribute * part_attr = container_of(attr,struct part_attribute,attr); ssize_t ret = 0; if (part_attr->show) - ret = part_attr->show(p,page); + ret = part_attr->show(p, page); + return ret; +} +static ssize_t +part_attr_store(struct kobject * kobj, struct attribute * attr, + const char *page, size_t count) +{ + struct hd_struct * p = container_of(kobj,struct hd_struct,kobj); + struct part_attribute * part_attr = container_of(attr,struct part_attribute,attr); + ssize_t ret = 0; + + if (part_attr->store) + ret = part_attr->store(p, page, count); return ret; } static struct sysfs_ops part_sysfs_ops = { .show = part_attr_show, + .store = part_attr_store, }; +static ssize_t part_uevent_store(struct hd_struct * p, + const char *page, size_t count) +{ + kobject_hotplug(&p->kobj, KOBJ_ADD); + return count; +} static ssize_t part_dev_read(struct hd_struct * p, char *page) { struct gendisk *disk = container_of(p->kobj.parent,struct gendisk,kobj); @@ -226,9 +246,13 @@ static ssize_t part_size_read(struct hd_struct * p, char *page) static ssize_t part_stat_read(struct hd_struct * p, char *page) { return sprintf(page, "%8u %8llu %8u %8llu\n", - p->reads, (unsigned long long)p->read_sectors, - p->writes, (unsigned long long)p->write_sectors); + p->ios[0], (unsigned long long)p->sectors[0], + p->ios[1], (unsigned long long)p->sectors[1]); } +static struct part_attribute part_attr_uevent = { + .attr = {.name = "uevent", .mode = S_IWUSR }, + .store = part_uevent_store +}; static struct part_attribute part_attr_dev = { .attr = {.name = "dev", .mode = S_IRUGO }, .show = part_dev_read @@ -247,6 +271,7 @@ static struct part_attribute part_attr_stat = { }; static struct attribute * default_attrs[] = { + &part_attr_uevent.attr, &part_attr_dev.attr, &part_attr_start.attr, &part_attr_size.attr, @@ -278,7 +303,8 @@ void delete_partition(struct gendisk *disk, int part) disk->part[part-1] = NULL; p->start_sect = 0; p->nr_sects = 0; - p->reads = p->writes = p->read_sectors = p->write_sectors = 0; + p->ios[0] = p->ios[1] = 0; + p->sectors[0] = p->sectors[1] = 0; devfs_remove("%s/part%d", disk->devfs_name, part); kobject_unregister(&p->kobj); } @@ -430,7 +456,7 @@ void del_gendisk(struct gendisk *disk) disk->flags &= ~GENHD_FL_UP; unlink_gendisk(disk); disk_stat_set_all(disk, 0); - disk->stamp = disk->stamp_idle = 0; + disk->stamp = 0; devfs_remove_disk(disk); diff --git a/fs/partitions/ibm.c b/fs/partitions/ibm.c index d59dcbf2bd4a..6327bcb2d73d 100644 --- a/fs/partitions/ibm.c +++ b/fs/partitions/ibm.c @@ -29,7 +29,7 @@ * cyl-cyl-head-head structure */ static inline int -cchh2blk (cchh_t *ptr, struct hd_geometry *geo) { +cchh2blk (struct vtoc_cchh *ptr, struct hd_geometry *geo) { return ptr->cc * geo->heads * geo->sectors + ptr->hh * geo->sectors; } @@ -40,7 +40,7 @@ cchh2blk (cchh_t *ptr, struct hd_geometry *geo) { * cyl-cyl-head-head-block structure */ static inline int -cchhb2blk (cchhb_t *ptr, struct hd_geometry *geo) { +cchhb2blk (struct vtoc_cchhb *ptr, struct hd_geometry *geo) { return ptr->cc * geo->heads * geo->sectors + ptr->hh * geo->sectors + ptr->b; @@ -56,7 +56,7 @@ ibm_partition(struct parsed_partitions *state, struct block_device *bdev) struct hd_geometry *geo; char type[5] = {0,}; char name[7] = {0,}; - volume_label_t *vlabel; + struct vtoc_volume_label *vlabel; unsigned char *data; Sector sect; @@ -64,7 +64,8 @@ ibm_partition(struct parsed_partitions *state, struct block_device *bdev) goto out_noinfo; if ((geo = kmalloc(sizeof(struct hd_geometry), GFP_KERNEL)) == NULL) goto out_nogeo; - if ((vlabel = kmalloc(sizeof(volume_label_t), GFP_KERNEL)) == NULL) + if ((vlabel = kmalloc(sizeof(struct vtoc_volume_label), + GFP_KERNEL)) == NULL) goto out_novlab; if (ioctl_by_bdev(bdev, BIODASDINFO, (unsigned long)info) != 0 || @@ -86,7 +87,7 @@ ibm_partition(struct parsed_partitions *state, struct block_device *bdev) strncpy(name, data + 8, 6); else strncpy(name, data + 4, 6); - memcpy (vlabel, data, sizeof(volume_label_t)); + memcpy (vlabel, data, sizeof(struct vtoc_volume_label)); put_dev_sector(sect); EBCASC(type, 4); @@ -129,9 +130,9 @@ ibm_partition(struct parsed_partitions *state, struct block_device *bdev) counter = 0; while ((data = read_dev_sector(bdev, blk*(blocksize/512), §)) != NULL) { - format1_label_t f1; + struct vtoc_format1_label f1; - memcpy(&f1, data, sizeof(format1_label_t)); + memcpy(&f1, data, sizeof(struct vtoc_format1_label)); put_dev_sector(sect); /* skip FMT4 / FMT5 / FMT7 labels */ diff --git a/fs/partitions/ultrix.c b/fs/partitions/ultrix.c index 8a8d4d9db314..ec852c11dce4 100644 --- a/fs/partitions/ultrix.c +++ b/fs/partitions/ultrix.c @@ -7,6 +7,7 @@ */ #include "check.h" +#include "ultrix.h" int ultrix_partition(struct parsed_partitions *state, struct block_device *bdev) { diff --git a/fs/pnode.c b/fs/pnode.c new file mode 100644 index 000000000000..aeeec8ba8dd2 --- /dev/null +++ b/fs/pnode.c @@ -0,0 +1,305 @@ +/* + * linux/fs/pnode.c + * + * (C) Copyright IBM Corporation 2005. + * Released under GPL v2. + * Author : Ram Pai (linuxram@us.ibm.com) + * + */ +#include <linux/namespace.h> +#include <linux/mount.h> +#include <linux/fs.h> +#include "pnode.h" + +/* return the next shared peer mount of @p */ +static inline struct vfsmount *next_peer(struct vfsmount *p) +{ + return list_entry(p->mnt_share.next, struct vfsmount, mnt_share); +} + +static inline struct vfsmount *first_slave(struct vfsmount *p) +{ + return list_entry(p->mnt_slave_list.next, struct vfsmount, mnt_slave); +} + +static inline struct vfsmount *next_slave(struct vfsmount *p) +{ + return list_entry(p->mnt_slave.next, struct vfsmount, mnt_slave); +} + +static int do_make_slave(struct vfsmount *mnt) +{ + struct vfsmount *peer_mnt = mnt, *master = mnt->mnt_master; + struct vfsmount *slave_mnt; + + /* + * slave 'mnt' to a peer mount that has the + * same root dentry. If none is available than + * slave it to anything that is available. + */ + while ((peer_mnt = next_peer(peer_mnt)) != mnt && + peer_mnt->mnt_root != mnt->mnt_root) ; + + if (peer_mnt == mnt) { + peer_mnt = next_peer(mnt); + if (peer_mnt == mnt) + peer_mnt = NULL; + } + list_del_init(&mnt->mnt_share); + + if (peer_mnt) + master = peer_mnt; + + if (master) { + list_for_each_entry(slave_mnt, &mnt->mnt_slave_list, mnt_slave) + slave_mnt->mnt_master = master; + list_del(&mnt->mnt_slave); + list_add(&mnt->mnt_slave, &master->mnt_slave_list); + list_splice(&mnt->mnt_slave_list, master->mnt_slave_list.prev); + INIT_LIST_HEAD(&mnt->mnt_slave_list); + } else { + struct list_head *p = &mnt->mnt_slave_list; + while (!list_empty(p)) { + slave_mnt = list_entry(p->next, + struct vfsmount, mnt_slave); + list_del_init(&slave_mnt->mnt_slave); + slave_mnt->mnt_master = NULL; + } + } + mnt->mnt_master = master; + CLEAR_MNT_SHARED(mnt); + INIT_LIST_HEAD(&mnt->mnt_slave_list); + return 0; +} + +void change_mnt_propagation(struct vfsmount *mnt, int type) +{ + if (type == MS_SHARED) { + set_mnt_shared(mnt); + return; + } + do_make_slave(mnt); + if (type != MS_SLAVE) { + list_del_init(&mnt->mnt_slave); + mnt->mnt_master = NULL; + if (type == MS_UNBINDABLE) + mnt->mnt_flags |= MNT_UNBINDABLE; + } +} + +/* + * get the next mount in the propagation tree. + * @m: the mount seen last + * @origin: the original mount from where the tree walk initiated + */ +static struct vfsmount *propagation_next(struct vfsmount *m, + struct vfsmount *origin) +{ + /* are there any slaves of this mount? */ + if (!IS_MNT_NEW(m) && !list_empty(&m->mnt_slave_list)) + return first_slave(m); + + while (1) { + struct vfsmount *next; + struct vfsmount *master = m->mnt_master; + + if ( master == origin->mnt_master ) { + next = next_peer(m); + return ((next == origin) ? NULL : next); + } else if (m->mnt_slave.next != &master->mnt_slave_list) + return next_slave(m); + + /* back at master */ + m = master; + } +} + +/* + * return the source mount to be used for cloning + * + * @dest the current destination mount + * @last_dest the last seen destination mount + * @last_src the last seen source mount + * @type return CL_SLAVE if the new mount has to be + * cloned as a slave. + */ +static struct vfsmount *get_source(struct vfsmount *dest, + struct vfsmount *last_dest, + struct vfsmount *last_src, + int *type) +{ + struct vfsmount *p_last_src = NULL; + struct vfsmount *p_last_dest = NULL; + *type = CL_PROPAGATION;; + + if (IS_MNT_SHARED(dest)) + *type |= CL_MAKE_SHARED; + + while (last_dest != dest->mnt_master) { + p_last_dest = last_dest; + p_last_src = last_src; + last_dest = last_dest->mnt_master; + last_src = last_src->mnt_master; + } + + if (p_last_dest) { + do { + p_last_dest = next_peer(p_last_dest); + } while (IS_MNT_NEW(p_last_dest)); + } + + if (dest != p_last_dest) { + *type |= CL_SLAVE; + return last_src; + } else + return p_last_src; +} + +/* + * mount 'source_mnt' under the destination 'dest_mnt' at + * dentry 'dest_dentry'. And propagate that mount to + * all the peer and slave mounts of 'dest_mnt'. + * Link all the new mounts into a propagation tree headed at + * source_mnt. Also link all the new mounts using ->mnt_list + * headed at source_mnt's ->mnt_list + * + * @dest_mnt: destination mount. + * @dest_dentry: destination dentry. + * @source_mnt: source mount. + * @tree_list : list of heads of trees to be attached. + */ +int propagate_mnt(struct vfsmount *dest_mnt, struct dentry *dest_dentry, + struct vfsmount *source_mnt, struct list_head *tree_list) +{ + struct vfsmount *m, *child; + int ret = 0; + struct vfsmount *prev_dest_mnt = dest_mnt; + struct vfsmount *prev_src_mnt = source_mnt; + LIST_HEAD(tmp_list); + LIST_HEAD(umount_list); + + for (m = propagation_next(dest_mnt, dest_mnt); m; + m = propagation_next(m, dest_mnt)) { + int type; + struct vfsmount *source; + + if (IS_MNT_NEW(m)) + continue; + + source = get_source(m, prev_dest_mnt, prev_src_mnt, &type); + + if (!(child = copy_tree(source, source->mnt_root, type))) { + ret = -ENOMEM; + list_splice(tree_list, tmp_list.prev); + goto out; + } + + if (is_subdir(dest_dentry, m->mnt_root)) { + mnt_set_mountpoint(m, dest_dentry, child); + list_add_tail(&child->mnt_hash, tree_list); + } else { + /* + * This can happen if the parent mount was bind mounted + * on some subdirectory of a shared/slave mount. + */ + list_add_tail(&child->mnt_hash, &tmp_list); + } + prev_dest_mnt = m; + prev_src_mnt = child; + } +out: + spin_lock(&vfsmount_lock); + while (!list_empty(&tmp_list)) { + child = list_entry(tmp_list.next, struct vfsmount, mnt_hash); + list_del_init(&child->mnt_hash); + umount_tree(child, 0, &umount_list); + } + spin_unlock(&vfsmount_lock); + release_mounts(&umount_list); + return ret; +} + +/* + * return true if the refcount is greater than count + */ +static inline int do_refcount_check(struct vfsmount *mnt, int count) +{ + int mycount = atomic_read(&mnt->mnt_count); + return (mycount > count); +} + +/* + * check if the mount 'mnt' can be unmounted successfully. + * @mnt: the mount to be checked for unmount + * NOTE: unmounting 'mnt' would naturally propagate to all + * other mounts its parent propagates to. + * Check if any of these mounts that **do not have submounts** + * have more references than 'refcnt'. If so return busy. + */ +int propagate_mount_busy(struct vfsmount *mnt, int refcnt) +{ + struct vfsmount *m, *child; + struct vfsmount *parent = mnt->mnt_parent; + int ret = 0; + + if (mnt == parent) + return do_refcount_check(mnt, refcnt); + + /* + * quickly check if the current mount can be unmounted. + * If not, we don't have to go checking for all other + * mounts + */ + if (!list_empty(&mnt->mnt_mounts) || do_refcount_check(mnt, refcnt)) + return 1; + + for (m = propagation_next(parent, parent); m; + m = propagation_next(m, parent)) { + child = __lookup_mnt(m, mnt->mnt_mountpoint, 0); + if (child && list_empty(&child->mnt_mounts) && + (ret = do_refcount_check(child, 1))) + break; + } + return ret; +} + +/* + * NOTE: unmounting 'mnt' naturally propagates to all other mounts its + * parent propagates to. + */ +static void __propagate_umount(struct vfsmount *mnt) +{ + struct vfsmount *parent = mnt->mnt_parent; + struct vfsmount *m; + + BUG_ON(parent == mnt); + + for (m = propagation_next(parent, parent); m; + m = propagation_next(m, parent)) { + + struct vfsmount *child = __lookup_mnt(m, + mnt->mnt_mountpoint, 0); + /* + * umount the child only if the child has no + * other children + */ + if (child && list_empty(&child->mnt_mounts)) { + list_del(&child->mnt_hash); + list_add_tail(&child->mnt_hash, &mnt->mnt_hash); + } + } +} + +/* + * collect all mounts that receive propagation from the mount in @list, + * and return these additional mounts in the same list. + * @list: the list of mounts to be unmounted. + */ +int propagate_umount(struct list_head *list) +{ + struct vfsmount *mnt; + + list_for_each_entry(mnt, list, mnt_hash) + __propagate_umount(mnt); + return 0; +} diff --git a/fs/pnode.h b/fs/pnode.h new file mode 100644 index 000000000000..020e1bb60fdb --- /dev/null +++ b/fs/pnode.h @@ -0,0 +1,37 @@ +/* + * linux/fs/pnode.h + * + * (C) Copyright IBM Corporation 2005. + * Released under GPL v2. + * + */ +#ifndef _LINUX_PNODE_H +#define _LINUX_PNODE_H + +#include <linux/list.h> +#include <linux/mount.h> + +#define IS_MNT_SHARED(mnt) (mnt->mnt_flags & MNT_SHARED) +#define IS_MNT_SLAVE(mnt) (mnt->mnt_master) +#define IS_MNT_NEW(mnt) (!mnt->mnt_namespace) +#define CLEAR_MNT_SHARED(mnt) (mnt->mnt_flags &= ~MNT_SHARED) +#define IS_MNT_UNBINDABLE(mnt) (mnt->mnt_flags & MNT_UNBINDABLE) + +#define CL_EXPIRE 0x01 +#define CL_SLAVE 0x02 +#define CL_COPY_ALL 0x04 +#define CL_MAKE_SHARED 0x08 +#define CL_PROPAGATION 0x10 + +static inline void set_mnt_shared(struct vfsmount *mnt) +{ + mnt->mnt_flags &= ~MNT_PNODE_MASK; + mnt->mnt_flags |= MNT_SHARED; +} + +void change_mnt_propagation(struct vfsmount *, int); +int propagate_mnt(struct vfsmount *, struct dentry *, struct vfsmount *, + struct list_head *); +int propagate_umount(struct list_head *); +int propagate_mount_busy(struct vfsmount *, int); +#endif /* _LINUX_PNODE_H */ diff --git a/fs/posix_acl.c b/fs/posix_acl.c index 296480e96dd5..6c8dcf7613fd 100644 --- a/fs/posix_acl.c +++ b/fs/posix_acl.c @@ -35,7 +35,7 @@ EXPORT_SYMBOL(posix_acl_permission); * Allocate a new ACL with the specified number of entries. */ struct posix_acl * -posix_acl_alloc(int count, unsigned int __nocast flags) +posix_acl_alloc(int count, gfp_t flags) { const size_t size = sizeof(struct posix_acl) + count * sizeof(struct posix_acl_entry); @@ -51,7 +51,7 @@ posix_acl_alloc(int count, unsigned int __nocast flags) * Clone an ACL. */ struct posix_acl * -posix_acl_clone(const struct posix_acl *acl, unsigned int __nocast flags) +posix_acl_clone(const struct posix_acl *acl, gfp_t flags) { struct posix_acl *clone = NULL; @@ -185,7 +185,7 @@ posix_acl_equiv_mode(const struct posix_acl *acl, mode_t *mode_p) * Create an ACL representing the file mode permission bits of an inode. */ struct posix_acl * -posix_acl_from_mode(mode_t mode, unsigned int __nocast flags) +posix_acl_from_mode(mode_t mode, gfp_t flags) { struct posix_acl *acl = posix_acl_alloc(3, flags); if (!acl) diff --git a/fs/proc/array.c b/fs/proc/array.c index d84eecacbeaf..3e1239e4b303 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -438,7 +438,7 @@ static int do_task_stat(struct task_struct *task, char * buffer, int whole) jiffies_to_clock_t(it_real_value), start_time, vsize, - mm ? get_mm_counter(mm, rss) : 0, /* you might want to shift this left 3 */ + mm ? get_mm_rss(mm) : 0, rsslim, mm ? mm->start_code : 0, mm ? mm->end_code : 0, diff --git a/fs/proc/base.c b/fs/proc/base.c index fb34f88a4a74..634355e16986 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -70,6 +70,7 @@ #include <linux/seccomp.h> #include <linux/cpuset.h> #include <linux/audit.h> +#include <linux/poll.h> #include "internal.h" /* @@ -103,7 +104,9 @@ enum pid_directory_inos { PROC_TGID_NUMA_MAPS, PROC_TGID_MOUNTS, PROC_TGID_WCHAN, +#ifdef CONFIG_MMU PROC_TGID_SMAPS, +#endif #ifdef CONFIG_SCHEDSTATS PROC_TGID_SCHEDSTAT, #endif @@ -141,7 +144,9 @@ enum pid_directory_inos { PROC_TID_NUMA_MAPS, PROC_TID_MOUNTS, PROC_TID_WCHAN, +#ifdef CONFIG_MMU PROC_TID_SMAPS, +#endif #ifdef CONFIG_SCHEDSTATS PROC_TID_SCHEDSTAT, #endif @@ -195,7 +200,9 @@ static struct pid_entry tgid_base_stuff[] = { E(PROC_TGID_ROOT, "root", S_IFLNK|S_IRWXUGO), E(PROC_TGID_EXE, "exe", S_IFLNK|S_IRWXUGO), E(PROC_TGID_MOUNTS, "mounts", S_IFREG|S_IRUGO), +#ifdef CONFIG_MMU E(PROC_TGID_SMAPS, "smaps", S_IFREG|S_IRUGO), +#endif #ifdef CONFIG_SECURITY E(PROC_TGID_ATTR, "attr", S_IFDIR|S_IRUGO|S_IXUGO), #endif @@ -235,7 +242,9 @@ static struct pid_entry tid_base_stuff[] = { E(PROC_TID_ROOT, "root", S_IFLNK|S_IRWXUGO), E(PROC_TID_EXE, "exe", S_IFLNK|S_IRWXUGO), E(PROC_TID_MOUNTS, "mounts", S_IFREG|S_IRUGO), +#ifdef CONFIG_MMU E(PROC_TID_SMAPS, "smaps", S_IFREG|S_IRUGO), +#endif #ifdef CONFIG_SECURITY E(PROC_TID_ATTR, "attr", S_IFDIR|S_IRUGO|S_IXUGO), #endif @@ -343,7 +352,8 @@ static int proc_root_link(struct inode *inode, struct dentry **dentry, struct vf /* Same as proc_root_link, but this addionally tries to get fs from other * threads in the group */ -static int proc_task_root_link(struct inode *inode, struct dentry **dentry, struct vfsmount **mnt) +static int proc_task_root_link(struct inode *inode, struct dentry **dentry, + struct vfsmount **mnt) { struct fs_struct *fs; int result = -ENOENT; @@ -357,9 +367,10 @@ static int proc_task_root_link(struct inode *inode, struct dentry **dentry, stru } else { /* Try to get fs from other threads */ task_unlock(leader); - struct task_struct *task = leader; read_lock(&tasklist_lock); - if (pid_alive(task)) { + if (pid_alive(leader)) { + struct task_struct *task = leader; + while ((task = next_thread(task)) != leader) { task_lock(task); fs = task->fs; @@ -628,6 +639,7 @@ static struct file_operations proc_numa_maps_operations = { }; #endif +#ifdef CONFIG_MMU extern struct seq_operations proc_pid_smaps_op; static int smaps_open(struct inode *inode, struct file *file) { @@ -646,28 +658,41 @@ static struct file_operations proc_smaps_operations = { .llseek = seq_lseek, .release = seq_release, }; +#endif extern struct seq_operations mounts_op; +struct proc_mounts { + struct seq_file m; + int event; +}; + static int mounts_open(struct inode *inode, struct file *file) { struct task_struct *task = proc_task(inode); - int ret = seq_open(file, &mounts_op); + struct namespace *namespace; + struct proc_mounts *p; + int ret = -EINVAL; - if (!ret) { - struct seq_file *m = file->private_data; - struct namespace *namespace; - task_lock(task); - namespace = task->namespace; - if (namespace) - get_namespace(namespace); - task_unlock(task); - - if (namespace) - m->private = namespace; - else { - seq_release(inode, file); - ret = -EINVAL; + task_lock(task); + namespace = task->namespace; + if (namespace) + get_namespace(namespace); + task_unlock(task); + + if (namespace) { + ret = -ENOMEM; + p = kmalloc(sizeof(struct proc_mounts), GFP_KERNEL); + if (p) { + file->private_data = &p->m; + ret = seq_open(file, &mounts_op); + if (!ret) { + p->m.private = namespace; + p->event = namespace->event; + return 0; + } + kfree(p); } + put_namespace(namespace); } return ret; } @@ -680,11 +705,30 @@ static int mounts_release(struct inode *inode, struct file *file) return seq_release(inode, file); } +static unsigned mounts_poll(struct file *file, poll_table *wait) +{ + struct proc_mounts *p = file->private_data; + struct namespace *ns = p->m.private; + unsigned res = 0; + + poll_wait(file, &ns->poll, wait); + + spin_lock(&vfsmount_lock); + if (p->event != ns->event) { + p->event = ns->event; + res = POLLERR; + } + spin_unlock(&vfsmount_lock); + + return res; +} + static struct file_operations proc_mounts_operations = { .open = mounts_open, .read = seq_read, .llseek = seq_lseek, .release = mounts_release, + .poll = mounts_poll, }; #define PROC_BLOCK_SIZE (3*1024) /* 4K page size but our output routines use some slack for overruns */ @@ -1679,10 +1723,12 @@ static struct dentry *proc_pident_lookup(struct inode *dir, case PROC_TGID_MOUNTS: inode->i_fop = &proc_mounts_operations; break; +#ifdef CONFIG_MMU case PROC_TID_SMAPS: case PROC_TGID_SMAPS: inode->i_fop = &proc_smaps_operations; break; +#endif #ifdef CONFIG_SECURITY case PROC_TID_ATTR: inode->i_nlink = 2; diff --git a/fs/proc/generic.c b/fs/proc/generic.c index 8a8c34461d48..b638fb500743 100644 --- a/fs/proc/generic.c +++ b/fs/proc/generic.c @@ -533,7 +533,7 @@ static void proc_kill_inodes(struct proc_dir_entry *de) */ file_list_lock(); list_for_each(p, &sb->s_files) { - struct file * filp = list_entry(p, struct file, f_list); + struct file * filp = list_entry(p, struct file, f_u.fu_list); struct dentry * dentry = filp->f_dentry; struct inode * inode; struct file_operations *fops; diff --git a/fs/proc/inode.c b/fs/proc/inode.c index effa6c0c467a..e6a818a93f3d 100644 --- a/fs/proc/inode.c +++ b/fs/proc/inode.c @@ -156,10 +156,13 @@ struct inode *proc_get_inode(struct super_block *sb, unsigned int ino, WARN_ON(de && de->deleted); + if (de != NULL && !try_module_get(de->owner)) + goto out_mod; + inode = iget(sb, ino); if (!inode) - goto out_fail; - + goto out_ino; + PROC_I(inode)->pde = de; if (de) { if (de->mode) { @@ -171,20 +174,20 @@ struct inode *proc_get_inode(struct super_block *sb, unsigned int ino, inode->i_size = de->size; if (de->nlink) inode->i_nlink = de->nlink; - if (!try_module_get(de->owner)) - goto out_fail; if (de->proc_iops) inode->i_op = de->proc_iops; if (de->proc_fops) inode->i_fop = de->proc_fops; } -out: return inode; -out_fail: +out_ino: + if (de != NULL) + module_put(de->owner); +out_mod: de_put(de); - goto out; + return NULL; } int proc_fill_super(struct super_block *s, void *data, int silent) diff --git a/fs/proc/nommu.c b/fs/proc/nommu.c index f3bf016d5ee3..cff10ab1af63 100644 --- a/fs/proc/nommu.c +++ b/fs/proc/nommu.c @@ -91,6 +91,7 @@ static void *nommu_vma_list_start(struct seq_file *m, loff_t *_pos) next = _rb; break; } + pos--; } return next; diff --git a/fs/proc/proc_devtree.c b/fs/proc/proc_devtree.c index 6fd57f154197..fb117b74809e 100644 --- a/fs/proc/proc_devtree.c +++ b/fs/proc/proc_devtree.c @@ -49,6 +49,39 @@ static int property_read_proc(char *page, char **start, off_t off, */ /* + * Add a property to a node + */ +static struct proc_dir_entry * +__proc_device_tree_add_prop(struct proc_dir_entry *de, struct property *pp) +{ + struct proc_dir_entry *ent; + + /* + * Unfortunately proc_register puts each new entry + * at the beginning of the list. So we rearrange them. + */ + ent = create_proc_read_entry(pp->name, + strncmp(pp->name, "security-", 9) + ? S_IRUGO : S_IRUSR, de, + property_read_proc, pp); + if (ent == NULL) + return NULL; + + if (!strncmp(pp->name, "security-", 9)) + ent->size = 0; /* don't leak number of password chars */ + else + ent->size = pp->length; + + return ent; +} + + +void proc_device_tree_add_prop(struct proc_dir_entry *pde, struct property *prop) +{ + __proc_device_tree_add_prop(pde, prop); +} + +/* * Process a node, adding entries for its children and its properties. */ void proc_device_tree_add_node(struct device_node *np, @@ -57,11 +90,9 @@ void proc_device_tree_add_node(struct device_node *np, struct property *pp; struct proc_dir_entry *ent; struct device_node *child; - struct proc_dir_entry *list = NULL, **lastp; const char *p; set_node_proc_entry(np, de); - lastp = &list; for (child = NULL; (child = of_get_next_child(np, child));) { p = strrchr(child->full_name, '/'); if (!p) @@ -71,9 +102,6 @@ void proc_device_tree_add_node(struct device_node *np, ent = proc_mkdir(p, de); if (ent == 0) break; - *lastp = ent; - ent->next = NULL; - lastp = &ent->next; proc_device_tree_add_node(child, ent); } of_node_put(child); @@ -84,7 +112,7 @@ void proc_device_tree_add_node(struct device_node *np, * properties are quite unimportant for us though, thus we * simply "skip" them here, but we do have to check. */ - for (ent = list; ent != NULL; ent = ent->next) + for (ent = de->subdir; ent != NULL; ent = ent->next) if (!strcmp(ent->name, pp->name)) break; if (ent != NULL) { @@ -94,25 +122,10 @@ void proc_device_tree_add_node(struct device_node *np, continue; } - /* - * Unfortunately proc_register puts each new entry - * at the beginning of the list. So we rearrange them. - */ - ent = create_proc_read_entry(pp->name, - strncmp(pp->name, "security-", 9) - ? S_IRUGO : S_IRUSR, de, - property_read_proc, pp); + ent = __proc_device_tree_add_prop(de, pp); if (ent == 0) break; - if (!strncmp(pp->name, "security-", 9)) - ent->size = 0; /* don't leak number of password chars */ - else - ent->size = pp->length; - ent->next = NULL; - *lastp = ent; - lastp = &ent->next; } - de->subdir = list; } /* diff --git a/fs/proc/proc_misc.c b/fs/proc/proc_misc.c index a3453555a94e..5b6b0b6038a7 100644 --- a/fs/proc/proc_misc.c +++ b/fs/proc/proc_misc.c @@ -629,12 +629,4 @@ void __init proc_misc_init(void) if (entry) entry->proc_fops = &proc_sysrq_trigger_operations; #endif -#ifdef CONFIG_PPC32 - { - extern struct file_operations ppc_htab_operations; - entry = create_proc_entry("ppc_htab", S_IRUGO|S_IWUSR, NULL); - if (entry) - entry->proc_fops = &ppc_htab_operations; - } -#endif } diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index c7ef3e48e35b..50bd5a8f0446 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -14,22 +14,41 @@ char *task_mem(struct mm_struct *mm, char *buffer) { unsigned long data, text, lib; + unsigned long hiwater_vm, total_vm, hiwater_rss, total_rss; + + /* + * Note: to minimize their overhead, mm maintains hiwater_vm and + * hiwater_rss only when about to *lower* total_vm or rss. Any + * collector of these hiwater stats must therefore get total_vm + * and rss too, which will usually be the higher. Barriers? not + * worth the effort, such snapshots can always be inconsistent. + */ + hiwater_vm = total_vm = mm->total_vm; + if (hiwater_vm < mm->hiwater_vm) + hiwater_vm = mm->hiwater_vm; + hiwater_rss = total_rss = get_mm_rss(mm); + if (hiwater_rss < mm->hiwater_rss) + hiwater_rss = mm->hiwater_rss; data = mm->total_vm - mm->shared_vm - mm->stack_vm; text = (PAGE_ALIGN(mm->end_code) - (mm->start_code & PAGE_MASK)) >> 10; lib = (mm->exec_vm << (PAGE_SHIFT-10)) - text; buffer += sprintf(buffer, + "VmPeak:\t%8lu kB\n" "VmSize:\t%8lu kB\n" "VmLck:\t%8lu kB\n" + "VmHWM:\t%8lu kB\n" "VmRSS:\t%8lu kB\n" "VmData:\t%8lu kB\n" "VmStk:\t%8lu kB\n" "VmExe:\t%8lu kB\n" "VmLib:\t%8lu kB\n" "VmPTE:\t%8lu kB\n", - (mm->total_vm - mm->reserved_vm) << (PAGE_SHIFT-10), + hiwater_vm << (PAGE_SHIFT-10), + (total_vm - mm->reserved_vm) << (PAGE_SHIFT-10), mm->locked_vm << (PAGE_SHIFT-10), - get_mm_counter(mm, rss) << (PAGE_SHIFT-10), + hiwater_rss << (PAGE_SHIFT-10), + total_rss << (PAGE_SHIFT-10), data << (PAGE_SHIFT-10), mm->stack_vm << (PAGE_SHIFT-10), text, lib, (PTRS_PER_PTE*sizeof(pte_t)*mm->nr_ptes) >> 10); @@ -44,13 +63,11 @@ unsigned long task_vsize(struct mm_struct *mm) int task_statm(struct mm_struct *mm, int *shared, int *text, int *data, int *resident) { - int rss = get_mm_counter(mm, rss); - - *shared = rss - get_mm_counter(mm, anon_rss); + *shared = get_mm_counter(mm, file_rss); *text = (PAGE_ALIGN(mm->end_code) - (mm->start_code & PAGE_MASK)) >> PAGE_SHIFT; *data = mm->total_vm - mm->shared_vm; - *resident = rss; + *resident = *shared + get_mm_counter(mm, anon_rss); return mm->total_vm; } @@ -178,7 +195,7 @@ static int show_map_internal(struct seq_file *m, void *v, struct mem_size_stats static int show_map(struct seq_file *m, void *v) { - return show_map_internal(m, v, 0); + return show_map_internal(m, v, NULL); } static void smaps_pte_range(struct vm_area_struct *vma, pmd_t *pmd, @@ -186,13 +203,14 @@ static void smaps_pte_range(struct vm_area_struct *vma, pmd_t *pmd, struct mem_size_stats *mss) { pte_t *pte, ptent; + spinlock_t *ptl; unsigned long pfn; struct page *page; - pte = pte_offset_map(pmd, addr); + pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl); do { ptent = *pte; - if (pte_none(ptent) || !pte_present(ptent)) + if (!pte_present(ptent)) continue; mss->resident += PAGE_SIZE; @@ -213,8 +231,8 @@ static void smaps_pte_range(struct vm_area_struct *vma, pmd_t *pmd, mss->private_clean += PAGE_SIZE; } } while (pte++, addr += PAGE_SIZE, addr != end); - pte_unmap(pte - 1); - cond_resched_lock(&vma->vm_mm->page_table_lock); + pte_unmap_unlock(pte - 1, ptl); + cond_resched(); } static inline void smaps_pmd_range(struct vm_area_struct *vma, pud_t *pud, @@ -268,17 +286,11 @@ static inline void smaps_pgd_range(struct vm_area_struct *vma, static int show_smap(struct seq_file *m, void *v) { struct vm_area_struct *vma = v; - struct mm_struct *mm = vma->vm_mm; struct mem_size_stats mss; memset(&mss, 0, sizeof mss); - - if (mm) { - spin_lock(&mm->page_table_lock); + if (vma->vm_mm) smaps_pgd_range(vma, vma->vm_start, vma->vm_end, &mss); - spin_unlock(&mm->page_table_lock); - } - return show_map_internal(m, v, &mss); } @@ -390,12 +402,11 @@ struct numa_maps { /* * Calculate numa node maps for a vma */ -static struct numa_maps *get_numa_maps(const struct vm_area_struct *vma) +static struct numa_maps *get_numa_maps(struct vm_area_struct *vma) { + int i; struct page *page; unsigned long vaddr; - struct mm_struct *mm = vma->vm_mm; - int i; struct numa_maps *md = kmalloc(sizeof(struct numa_maps), GFP_KERNEL); if (!md) @@ -407,9 +418,8 @@ static struct numa_maps *get_numa_maps(const struct vm_area_struct *vma) for_each_node(i) md->node[i] =0; - spin_lock(&mm->page_table_lock); for (vaddr = vma->vm_start; vaddr < vma->vm_end; vaddr += PAGE_SIZE) { - page = follow_page(mm, vaddr, 0); + page = follow_page(vma, vaddr, 0); if (page) { int count = page_mapcount(page); @@ -422,8 +432,8 @@ static struct numa_maps *get_numa_maps(const struct vm_area_struct *vma) md->anon++; md->node[page_to_nid(page)]++; } + cond_resched(); } - spin_unlock(&mm->page_table_lock); return md; } @@ -469,7 +479,7 @@ static int show_numa_map(struct seq_file *m, void *v) seq_printf(m, " interleave={"); first = 1; for_each_node(n) { - if (test_bit(n, pol->v.nodes)) { + if (node_isset(n, pol->v.nodes)) { if (!first) seq_putc(m,','); else diff --git a/fs/quota.c b/fs/quota.c index f5d1cff55196..612e04db4b93 100644 --- a/fs/quota.c +++ b/fs/quota.c @@ -15,6 +15,7 @@ #include <linux/security.h> #include <linux/syscalls.h> #include <linux/buffer_head.h> +#include <linux/quotaops.h> /* Check validity of generic quotactl commands */ static int generic_quotactl_valid(struct super_block *sb, int type, int cmd, qid_t id) @@ -118,6 +119,10 @@ static int xqm_quotactl_valid(struct super_block *sb, int type, int cmd, qid_t i if (!sb->s_qcop->get_xquota) return -ENOSYS; break; + case Q_XQUOTASYNC: + if (!sb->s_qcop->quota_sync) + return -ENOSYS; + break; default: return -EINVAL; } @@ -128,7 +133,7 @@ static int xqm_quotactl_valid(struct super_block *sb, int type, int cmd, qid_t i (type == XQM_GRPQUOTA && !in_egroup_p(id))) && !capable(CAP_SYS_ADMIN)) return -EPERM; - } else if (cmd != Q_XGETQSTAT) { + } else if (cmd != Q_XGETQSTAT && cmd != Q_XQUOTASYNC) { if (!capable(CAP_SYS_ADMIN)) return -EPERM; } @@ -322,6 +327,8 @@ static int do_quotactl(struct super_block *sb, int type, int cmd, qid_t id, void return -EFAULT; return 0; } + case Q_XQUOTASYNC: + return sb->s_qcop->quota_sync(sb, type); /* We never reach here unless validity check is broken */ default: BUG(); diff --git a/fs/read_write.c b/fs/read_write.c index b60324aaa2b6..a091ee4f430d 100644 --- a/fs/read_write.c +++ b/fs/read_write.c @@ -499,6 +499,9 @@ static ssize_t do_readv_writev(int type, struct file *file, ret = rw_verify_area(type, file, pos, tot_len); if (ret) goto out; + ret = security_file_permission(file, type == READ ? MAY_READ : MAY_WRITE); + if (ret) + goto out; fnv = NULL; if (type == READ) { diff --git a/fs/reiserfs/file.c b/fs/reiserfs/file.c index c20babd6216d..7892a865b58a 100644 --- a/fs/reiserfs/file.c +++ b/fs/reiserfs/file.c @@ -251,12 +251,12 @@ static int reiserfs_allocate_blocks_for_region(struct reiserfs_transaction_handl blocks_to_allocate, blocks_to_allocate); if (res != CARRY_ON) { - res = -ENOSPC; + res = res == QUOTA_EXCEEDED ? -EDQUOT : -ENOSPC; pathrelse(&path); goto error_exit; } } else { - res = -ENOSPC; + res = res == QUOTA_EXCEEDED ? -EDQUOT : -ENOSPC; pathrelse(&path); goto error_exit; } diff --git a/fs/reiserfs/fix_node.c b/fs/reiserfs/fix_node.c index 2706e2adffab..45829889dcdc 100644 --- a/fs/reiserfs/fix_node.c +++ b/fs/reiserfs/fix_node.c @@ -2022,7 +2022,7 @@ static int get_neighbors(struct tree_balance *p_s_tb, int n_h) } #ifdef CONFIG_REISERFS_CHECK -void *reiserfs_kmalloc(size_t size, int flags, struct super_block *s) +void *reiserfs_kmalloc(size_t size, gfp_t flags, struct super_block *s) { void *vp; static size_t malloced; diff --git a/fs/reiserfs/hashes.c b/fs/reiserfs/hashes.c index 37c1306eb9b7..a3ec238fd9e0 100644 --- a/fs/reiserfs/hashes.c +++ b/fs/reiserfs/hashes.c @@ -19,6 +19,7 @@ // #include <linux/kernel.h> +#include <linux/reiserfs_fs.h> #include <asm/types.h> #include <asm/bug.h> diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c index d76ee6c4f9b8..0a044ad98885 100644 --- a/fs/reiserfs/inode.c +++ b/fs/reiserfs/inode.c @@ -2194,7 +2194,7 @@ static int map_block_for_writepage(struct inode *inode, INITIALIZE_PATH(path); int pos_in_item; int jbegin_count = JOURNAL_PER_BALANCE_CNT; - loff_t byte_offset = (block << inode->i_sb->s_blocksize_bits) + 1; + loff_t byte_offset = ((loff_t)block << inode->i_sb->s_blocksize_bits)+1; int retval; int use_get_block = 0; int bytes_copied = 0; @@ -2842,7 +2842,7 @@ static int reiserfs_set_page_dirty(struct page *page) * even in -o notail mode, we can't be sure an old mount without -o notail * didn't create files with tails. */ -static int reiserfs_releasepage(struct page *page, int unused_gfp_flags) +static int reiserfs_releasepage(struct page *page, gfp_t unused_gfp_flags) { struct inode *inode = page->mapping->host; struct reiserfs_journal *j = SB_JOURNAL(inode->i_sb); diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index 4b15761434bc..68b7b78638ff 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -2757,6 +2757,15 @@ int journal_init(struct super_block *p_s_sb, const char *j_dev_name, journal->j_cnode_used = 0; journal->j_must_wait = 0; + if (journal->j_cnode_free == 0) { + reiserfs_warning(p_s_sb, "journal-2004: Journal cnode memory " + "allocation failed (%ld bytes). Journal is " + "too large for available memory. Usually " + "this is due to a journal that is too large.", + sizeof (struct reiserfs_journal_cnode) * num_cnodes); + goto free_and_return; + } + init_journal_hash(p_s_sb); jl = journal->j_current_jl; jl->j_list_bitmap = get_list_bitmap(p_s_sb, jl); diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index 44b02fc02ebe..42afb5bef111 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -1024,12 +1024,8 @@ static int reiserfs_parse_options(struct super_block *s, char *options, /* strin strcpy(REISERFS_SB(s)->s_qf_names[qtype], arg); *mount_options |= 1 << REISERFS_QUOTA; } else { - if (REISERFS_SB(s)->s_qf_names[qtype]) { - kfree(REISERFS_SB(s)-> - s_qf_names[qtype]); - REISERFS_SB(s)->s_qf_names[qtype] = - NULL; - } + kfree(REISERFS_SB(s)->s_qf_names[qtype]); + REISERFS_SB(s)->s_qf_names[qtype] = NULL; } } if (c == 'f') { @@ -1158,11 +1154,10 @@ static int reiserfs_remount(struct super_block *s, int *mount_flags, char *arg) if (!reiserfs_parse_options (s, arg, &mount_options, &blocks, NULL, &commit_max_age)) { #ifdef CONFIG_QUOTA - for (i = 0; i < MAXQUOTAS; i++) - if (REISERFS_SB(s)->s_qf_names[i]) { - kfree(REISERFS_SB(s)->s_qf_names[i]); - REISERFS_SB(s)->s_qf_names[i] = NULL; - } + for (i = 0; i < MAXQUOTAS; i++) { + kfree(REISERFS_SB(s)->s_qf_names[i]); + REISERFS_SB(s)->s_qf_names[i] = NULL; + } #endif return -EINVAL; } @@ -1940,13 +1935,11 @@ static int reiserfs_fill_super(struct super_block *s, void *data, int silent) brelse(SB_BUFFER_WITH_SB(s)); #ifdef CONFIG_QUOTA for (j = 0; j < MAXQUOTAS; j++) { - if (sbi->s_qf_names[j]) - kfree(sbi->s_qf_names[j]); + kfree(sbi->s_qf_names[j]); + sbi->s_qf_names[j] = NULL; } #endif - if (sbi != NULL) { - kfree(sbi); - } + kfree(sbi); s->s_fs_info = NULL; return errval; diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c index 87ac9dc8b381..72e120798677 100644 --- a/fs/reiserfs/xattr.c +++ b/fs/reiserfs/xattr.c @@ -453,7 +453,7 @@ static struct page *reiserfs_get_page(struct inode *dir, unsigned long n) struct page *page; /* We can deadlock if we try to free dentries, and an unlink/rmdir has just occured - GFP_NOFS avoids this */ - mapping->flags = (mapping->flags & ~__GFP_BITS_MASK) | GFP_NOFS; + mapping_set_gfp_mask(mapping, GFP_NOFS); page = read_cache_page(mapping, n, (filler_t *) mapping->a_ops->readpage, NULL); if (!IS_ERR(page)) { diff --git a/fs/reiserfs/xattr_acl.c b/fs/reiserfs/xattr_acl.c index 6703efa3c430..a47ac9aac8b2 100644 --- a/fs/reiserfs/xattr_acl.c +++ b/fs/reiserfs/xattr_acl.c @@ -296,8 +296,7 @@ reiserfs_set_acl(struct inode *inode, int type, struct posix_acl *acl) } } - if (value) - kfree(value); + kfree(value); if (!error) { /* Release the old one */ diff --git a/fs/relayfs/buffers.c b/fs/relayfs/buffers.c index 2aa8e2719999..84e21ffa5ca8 100644 --- a/fs/relayfs/buffers.c +++ b/fs/relayfs/buffers.c @@ -109,7 +109,7 @@ static void *relay_alloc_buf(struct rchan_buf *buf, unsigned long size) if (unlikely(!buf->page_array[i])) goto depopulate; } - mem = vmap(buf->page_array, n_pages, GFP_KERNEL, PAGE_KERNEL); + mem = vmap(buf->page_array, n_pages, VM_MAP, PAGE_KERNEL); if (!mem) goto depopulate; diff --git a/fs/seq_file.c b/fs/seq_file.c index 38ef913767ff..7c40570b71dc 100644 --- a/fs/seq_file.c +++ b/fs/seq_file.c @@ -28,13 +28,17 @@ */ int seq_open(struct file *file, struct seq_operations *op) { - struct seq_file *p = kmalloc(sizeof(*p), GFP_KERNEL); - if (!p) - return -ENOMEM; + struct seq_file *p = file->private_data; + + if (!p) { + p = kmalloc(sizeof(*p), GFP_KERNEL); + if (!p) + return -ENOMEM; + file->private_data = p; + } memset(p, 0, sizeof(*p)); sema_init(&p->sem, 1); p->op = op; - file->private_data = p; /* * Wrappers around seq_open(e.g. swaps_open) need to be diff --git a/fs/smbfs/request.c b/fs/smbfs/request.c index 2d85dd7415bb..a0f296d9928a 100644 --- a/fs/smbfs/request.c +++ b/fs/smbfs/request.c @@ -786,8 +786,7 @@ int smb_request_recv(struct smb_sb_info *server) /* We should never be called with any of these states */ case SMB_RECV_END: case SMB_RECV_REQUEST: - server->rstate = SMB_RECV_END; - break; + BUG(); } if (result < 0) { diff --git a/fs/smbfs/symlink.c b/fs/smbfs/symlink.c index 0c64bc3a0127..cdc53c4fb381 100644 --- a/fs/smbfs/symlink.c +++ b/fs/smbfs/symlink.c @@ -45,7 +45,7 @@ static void *smb_follow_link(struct dentry *dentry, struct nameidata *nd) int len = smb_proc_read_link(server_from_dentry(dentry), dentry, link, PATH_MAX - 1); if (len < 0) { - putname(link); + __putname(link); link = ERR_PTR(len); } else { link[len] = 0; @@ -59,7 +59,7 @@ static void smb_put_link(struct dentry *dentry, struct nameidata *nd, void *p) { char *s = nd_get_link(nd); if (!IS_ERR(s)) - putname(s); + __putname(s); } struct inode_operations smb_link_inode_operations = diff --git a/fs/super.c b/fs/super.c index 6e57ee252e14..6689dded3c84 100644 --- a/fs/super.c +++ b/fs/super.c @@ -171,6 +171,7 @@ void deactivate_super(struct super_block *s) if (atomic_dec_and_lock(&s->s_active, &sb_lock)) { s->s_count -= S_BIAS-1; spin_unlock(&sb_lock); + DQUOT_OFF(s); down_write(&s->s_umount); fs->kill_sb(s); put_filesystem(fs); @@ -474,8 +475,6 @@ rescan: return NULL; } -EXPORT_SYMBOL(user_get_super); - asmlinkage long sys_ustat(unsigned dev, struct ustat __user * ubuf) { struct super_block *s; @@ -513,7 +512,7 @@ static void mark_files_ro(struct super_block *sb) struct file *f; file_list_lock(); - list_for_each_entry(f, &sb->s_files, f_list) { + list_for_each_entry(f, &sb->s_files, f_u.fu_list) { if (S_ISREG(f->f_dentry->d_inode->i_mode) && file_count(f)) f->f_mode &= ~FMODE_WRITE; } diff --git a/fs/udf/file.c b/fs/udf/file.c index bb40d63f328f..01f520c71dc1 100644 --- a/fs/udf/file.c +++ b/fs/udf/file.c @@ -186,7 +186,7 @@ int udf_ioctl(struct inode *inode, struct file *filp, unsigned int cmd, { int result = -EINVAL; - if ( permission(inode, MAY_READ, NULL) != 0 ) + if ( file_permission(filp, MAY_READ) != 0 ) { udf_debug("no permission to access inode %lu\n", inode->i_ino); diff --git a/fs/udf/udf_sb.h b/fs/udf/udf_sb.h index 0e54922daa09..663669810be6 100644 --- a/fs/udf/udf_sb.h +++ b/fs/udf/udf_sb.h @@ -39,8 +39,7 @@ static inline struct udf_sb_info *UDF_SB(struct super_block *sb) {\ if (UDF_SB(X))\ {\ - if (UDF_SB_PARTMAPS(X))\ - kfree(UDF_SB_PARTMAPS(X));\ + kfree(UDF_SB_PARTMAPS(X));\ UDF_SB_PARTMAPS(X) = NULL;\ }\ } diff --git a/fs/ufs/super.c b/fs/ufs/super.c index f036d694ba5a..54828ebcf1ba 100644 --- a/fs/ufs/super.c +++ b/fs/ufs/super.c @@ -472,13 +472,14 @@ static int ufs_read_cylinder_structures (struct super_block *sb) { return 1; failed: - if (base) kfree (base); + kfree (base); if (sbi->s_ucg) { for (i = 0; i < uspi->s_ncg; i++) - if (sbi->s_ucg[i]) brelse (sbi->s_ucg[i]); + if (sbi->s_ucg[i]) + brelse (sbi->s_ucg[i]); kfree (sbi->s_ucg); for (i = 0; i < UFS_MAX_GROUP_LOADED; i++) - if (sbi->s_ucpi[i]) kfree (sbi->s_ucpi[i]); + kfree (sbi->s_ucpi[i]); } UFSD(("EXIT (FAILED)\n")) return 0; @@ -981,9 +982,10 @@ magic_found: dalloc_failed: iput(inode); failed: - if (ubh) ubh_brelse_uspi (uspi); - if (uspi) kfree (uspi); - if (sbi) kfree(sbi); + if (ubh) + ubh_brelse_uspi (uspi); + kfree (uspi); + kfree(sbi); sb->s_fs_info = NULL; UFSD(("EXIT (FAILED)\n")) return -EINVAL; diff --git a/fs/vfat/namei.c b/fs/vfat/namei.c index 1c6f6b57ef1c..ef46939c0c1a 100644 --- a/fs/vfat/namei.c +++ b/fs/vfat/namei.c @@ -621,8 +621,7 @@ static int vfat_build_slots(struct inode *dir, const unsigned char *name, } /* build the entry of long file name */ - for (cksum = i = 0; i < 11; i++) - cksum = (((cksum&1)<<7)|((cksum&0xfe)>>1)) + msdos_name[i]; + cksum = fat_checksum(msdos_name); *nr_slots = usize / 13; for (ps = slots, i = *nr_slots; i > 0; i--, ps++) { @@ -888,10 +887,10 @@ static int vfat_rename(struct inode *old_dir, struct dentry *old_dentry, { struct buffer_head *dotdot_bh; struct msdos_dir_entry *dotdot_de; - loff_t dotdot_i_pos; struct inode *old_inode, *new_inode; struct fat_slot_info old_sinfo, sinfo; struct timespec ts; + loff_t dotdot_i_pos, new_i_pos; int err, is_dir, update_dotdot, corrupt = 0; old_sinfo.bh = sinfo.bh = dotdot_bh = NULL; @@ -914,31 +913,24 @@ static int vfat_rename(struct inode *old_dir, struct dentry *old_dentry, ts = CURRENT_TIME_SEC; if (new_inode) { - err = vfat_find(new_dir, &new_dentry->d_name, &sinfo); - if (err) - goto out; - if (MSDOS_I(new_inode)->i_pos != sinfo.i_pos) { - /* WTF??? Cry and fail. */ - printk(KERN_WARNING "vfat_rename: fs corrupted\n"); - goto out; - } - if (is_dir) { err = fat_dir_empty(new_inode); if (err) goto out; } + new_i_pos = MSDOS_I(new_inode)->i_pos; fat_detach(new_inode); } else { err = vfat_add_entry(new_dir, &new_dentry->d_name, is_dir, 0, &ts, &sinfo); if (err) goto out; + new_i_pos = sinfo.i_pos; } new_dir->i_version++; fat_detach(old_inode); - fat_attach(old_inode, sinfo.i_pos); + fat_attach(old_inode, new_i_pos); if (IS_DIRSYNC(new_dir)) { err = fat_sync_inode(old_inode); if (err) @@ -1002,7 +994,7 @@ error_inode: fat_detach(old_inode); fat_attach(old_inode, old_sinfo.i_pos); if (new_inode) { - fat_attach(new_inode, sinfo.i_pos); + fat_attach(new_inode, new_i_pos); if (corrupt) corrupt |= fat_sync_inode(new_inode); } else { diff --git a/fs/xattr.c b/fs/xattr.c index 3f9c64bea151..a9db22557998 100644 --- a/fs/xattr.c +++ b/fs/xattr.c @@ -74,8 +74,7 @@ setxattr(struct dentry *d, char __user *name, void __user *value, } out: up(&d->d_inode->i_sem); - if (kvalue) - kfree(kvalue); + kfree(kvalue); return error; } @@ -143,7 +142,7 @@ getxattr(struct dentry *d, char __user *name, void __user *value, size_t size) if (size) { if (size > XATTR_SIZE_MAX) size = XATTR_SIZE_MAX; - kvalue = kmalloc(size, GFP_KERNEL); + kvalue = kzalloc(size, GFP_KERNEL); if (!kvalue) return -ENOMEM; } @@ -154,11 +153,15 @@ getxattr(struct dentry *d, char __user *name, void __user *value, size_t size) error = -EOPNOTSUPP; if (d->d_inode->i_op && d->d_inode->i_op->getxattr) error = d->d_inode->i_op->getxattr(d, kname, kvalue, size); - else if (!strncmp(kname, XATTR_SECURITY_PREFIX, - sizeof XATTR_SECURITY_PREFIX - 1)) { + + if (!strncmp(kname, XATTR_SECURITY_PREFIX, + sizeof XATTR_SECURITY_PREFIX - 1)) { const char *suffix = kname + sizeof XATTR_SECURITY_PREFIX - 1; - error = security_inode_getsecurity(d->d_inode, suffix, kvalue, - size); + int rv = security_inode_getsecurity(d->d_inode, suffix, kvalue, + size, error); + /* Security module active: overwrite error value */ + if (rv != -EOPNOTSUPP) + error = rv; } if (error > 0) { if (size && copy_to_user(value, kvalue, error)) @@ -169,8 +172,7 @@ getxattr(struct dentry *d, char __user *name, void __user *value, size_t size) error = -E2BIG; } out: - if (kvalue) - kfree(kvalue); + kfree(kvalue); return error; } @@ -255,8 +257,7 @@ listxattr(struct dentry *d, char __user *list, size_t size) error = -E2BIG; } out: - if (klist) - kfree(klist); + kfree(klist); return error; } diff --git a/fs/xfs/Kconfig b/fs/xfs/Kconfig index 8e8f32dabe53..bac27d66151d 100644 --- a/fs/xfs/Kconfig +++ b/fs/xfs/Kconfig @@ -24,7 +24,7 @@ config XFS_EXPORT default y config XFS_QUOTA - tristate "XFS Quota support" + bool "XFS Quota support" depends on XFS_FS help If you say Y here, you will be able to set limits for disk usage on diff --git a/fs/xfs/Makefile-linux-2.6 b/fs/xfs/Makefile-linux-2.6 index d8c87fa21ad1..97bd4743b461 100644 --- a/fs/xfs/Makefile-linux-2.6 +++ b/fs/xfs/Makefile-linux-2.6 @@ -109,7 +109,6 @@ xfs-y += xfs_alloc.o \ xfs_dfrag.o \ xfs_log.o \ xfs_log_recover.o \ - xfs_macros.o \ xfs_mount.o \ xfs_rename.o \ xfs_trans.o \ diff --git a/fs/xfs/linux-2.6/kmem.c b/fs/xfs/linux-2.6/kmem.c index 4b184559f231..aba7fcf881a2 100644 --- a/fs/xfs/linux-2.6/kmem.c +++ b/fs/xfs/linux-2.6/kmem.c @@ -1,55 +1,38 @@ /* - * Copyright (c) 2000-2004 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include <linux/sched.h> #include <linux/mm.h> #include <linux/vmalloc.h> #include <linux/highmem.h> #include <linux/swap.h> #include <linux/blkdev.h> - #include "time.h" #include "kmem.h" #define MAX_VMALLOCS 6 #define MAX_SLAB_SIZE 0x20000 - void * kmem_alloc(size_t size, unsigned int __nocast flags) { - int retries = 0; - unsigned int lflags = kmem_flags_convert(flags); - void *ptr; + int retries = 0; + gfp_t lflags = kmem_flags_convert(flags); + void *ptr; do { if (size < MAX_SLAB_SIZE || retries > MAX_VMALLOCS) @@ -107,9 +90,9 @@ kmem_realloc(void *ptr, size_t newsize, size_t oldsize, void * kmem_zone_alloc(kmem_zone_t *zone, unsigned int __nocast flags) { - int retries = 0; - unsigned int lflags = kmem_flags_convert(flags); - void *ptr; + int retries = 0; + gfp_t lflags = kmem_flags_convert(flags); + void *ptr; do { ptr = kmem_cache_alloc(zone, lflags); diff --git a/fs/xfs/linux-2.6/kmem.h b/fs/xfs/linux-2.6/kmem.h index 109fcf27e256..c64a29cdfff3 100644 --- a/fs/xfs/linux-2.6/kmem.h +++ b/fs/xfs/linux-2.6/kmem.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2004 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_SUPPORT_KMEM_H__ #define __XFS_SUPPORT_KMEM_H__ @@ -44,8 +30,8 @@ #define KM_NOFS 0x0004u #define KM_MAYFAIL 0x0008u -#define kmem_zone kmem_cache_s -#define kmem_zone_t kmem_cache_t +#define kmem_zone kmem_cache +#define kmem_zone_t struct kmem_cache typedef unsigned long xfs_pflags_t; @@ -81,9 +67,9 @@ typedef unsigned long xfs_pflags_t; *(NSTATEP) = *(OSTATEP); \ } while (0) -static __inline unsigned int kmem_flags_convert(unsigned int __nocast flags) +static __inline gfp_t kmem_flags_convert(unsigned int __nocast flags) { - unsigned int lflags = __GFP_NOWARN; /* we'll report problems, if need be */ + gfp_t lflags = __GFP_NOWARN; /* we'll report problems, if need be */ #ifdef DEBUG if (unlikely(flags & ~(KM_SLEEP|KM_NOSLEEP|KM_NOFS|KM_MAYFAIL))) { @@ -102,7 +88,7 @@ static __inline unsigned int kmem_flags_convert(unsigned int __nocast flags) if (PFLAGS_TEST_FSTRANS() || (flags & KM_NOFS)) lflags &= ~__GFP_FS; } - + return lflags; } @@ -125,17 +111,16 @@ kmem_zone_destroy(kmem_zone_t *zone) BUG(); } -extern void *kmem_zone_zalloc(kmem_zone_t *, unsigned int __nocast); -extern void *kmem_zone_alloc(kmem_zone_t *, unsigned int __nocast); +extern void *kmem_zone_zalloc(kmem_zone_t *, unsigned int __nocast); +extern void *kmem_zone_alloc(kmem_zone_t *, unsigned int __nocast); -extern void *kmem_alloc(size_t, unsigned int __nocast); -extern void *kmem_realloc(void *, size_t, size_t, - unsigned int __nocast); -extern void *kmem_zalloc(size_t, unsigned int __nocast); -extern void kmem_free(void *, size_t); +extern void *kmem_alloc(size_t, unsigned int __nocast); +extern void *kmem_realloc(void *, size_t, size_t, unsigned int __nocast); +extern void *kmem_zalloc(size_t, unsigned int __nocast); +extern void kmem_free(void *, size_t); typedef struct shrinker *kmem_shaker_t; -typedef int (*kmem_shake_func_t)(int, unsigned int); +typedef int (*kmem_shake_func_t)(int, gfp_t); static __inline kmem_shaker_t kmem_shake_register(kmem_shake_func_t sfunc) @@ -150,7 +135,7 @@ kmem_shake_deregister(kmem_shaker_t shrinker) } static __inline int -kmem_shake_allow(unsigned int gfp_mask) +kmem_shake_allow(gfp_t gfp_mask) { return (gfp_mask & __GFP_WAIT); } diff --git a/fs/xfs/linux-2.6/mrlock.h b/fs/xfs/linux-2.6/mrlock.h index d2c11a098ff2..16b44c3c2362 100644 --- a/fs/xfs/linux-2.6/mrlock.h +++ b/fs/xfs/linux-2.6/mrlock.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2004 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_SUPPORT_MRLOCK_H__ #define __XFS_SUPPORT_MRLOCK_H__ diff --git a/fs/xfs/linux-2.6/mutex.h b/fs/xfs/linux-2.6/mutex.h index 0b296bb944cb..ce773d89a923 100644 --- a/fs/xfs/linux-2.6/mutex.h +++ b/fs/xfs/linux-2.6/mutex.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_SUPPORT_MUTEX_H__ #define __XFS_SUPPORT_MUTEX_H__ diff --git a/fs/xfs/linux-2.6/sema.h b/fs/xfs/linux-2.6/sema.h index 30b67b4e1cbf..194a84490bd1 100644 --- a/fs/xfs/linux-2.6/sema.h +++ b/fs/xfs/linux-2.6/sema.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_SUPPORT_SEMA_H__ #define __XFS_SUPPORT_SEMA_H__ diff --git a/fs/xfs/linux-2.6/spin.h b/fs/xfs/linux-2.6/spin.h index 0039504069a5..50a6191178f4 100644 --- a/fs/xfs/linux-2.6/spin.h +++ b/fs/xfs/linux-2.6/spin.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_SUPPORT_SPIN_H__ #define __XFS_SUPPORT_SPIN_H__ diff --git a/fs/xfs/linux-2.6/sv.h b/fs/xfs/linux-2.6/sv.h index 821d3167e05b..9a8ad481b008 100644 --- a/fs/xfs/linux-2.6/sv.h +++ b/fs/xfs/linux-2.6/sv.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_SUPPORT_SV_H__ #define __XFS_SUPPORT_SV_H__ diff --git a/fs/xfs/linux-2.6/time.h b/fs/xfs/linux-2.6/time.h index b0d2873ab274..387e695a184c 100644 --- a/fs/xfs/linux-2.6/time.h +++ b/fs/xfs/linux-2.6/time.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_SUPPORT_TIME_H__ #define __XFS_SUPPORT_TIME_H__ diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c index c6c077978fe3..94d3cdfbf9b8 100644 --- a/fs/xfs/linux-2.6/xfs_aops.c +++ b/fs/xfs/linux-2.6/xfs_aops.c @@ -1,39 +1,26 @@ /* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_sb.h" +#include "xfs_ag.h" #include "xfs_dir.h" #include "xfs_dir2.h" #include "xfs_trans.h" @@ -42,13 +29,13 @@ #include "xfs_bmap_btree.h" #include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_alloc.h" -#include "xfs_btree.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" #include "xfs_inode.h" +#include "xfs_alloc.h" +#include "xfs_btree.h" #include "xfs_error.h" #include "xfs_rw.h" #include "xfs_iomap.h" @@ -761,8 +748,9 @@ xfs_page_state_convert( if (page->index >= end_index) { if ((page->index >= end_index + 1) || !(i_size_read(inode) & (PAGE_CACHE_SIZE - 1))) { - err = -EIO; - goto error; + if (startio) + unlock_page(page); + return 0; } } @@ -948,13 +936,15 @@ __linvfs_get_block( { vnode_t *vp = LINVFS_GET_VP(inode); xfs_iomap_t iomap; + xfs_off_t offset; + ssize_t size; int retpbbm = 1; int error; - ssize_t size; - loff_t offset = (loff_t)iblock << inode->i_blkbits; + offset = (xfs_off_t)iblock << inode->i_blkbits; if (blocks) - size = blocks << inode->i_blkbits; + size = (ssize_t) min_t(xfs_off_t, LONG_MAX, + (xfs_off_t)blocks << inode->i_blkbits); else size = 1 << inode->i_blkbits; @@ -967,8 +957,8 @@ __linvfs_get_block( return 0; if (iomap.iomap_bn != IOMAP_DADDR_NULL) { - xfs_daddr_t bn; - loff_t delta; + xfs_daddr_t bn; + xfs_off_t delta; /* For unwritten extents do not report a disk address on * the read case (treat as if we're reading into a hole). @@ -1000,9 +990,8 @@ __linvfs_get_block( */ if (create && ((!buffer_mapped(bh_result) && !buffer_uptodate(bh_result)) || - (offset >= i_size_read(inode)) || (iomap.iomap_flags & IOMAP_NEW))) { + (offset >= i_size_read(inode)) || (iomap.iomap_flags & IOMAP_NEW))) set_buffer_new(bh_result); - } if (iomap.iomap_flags & IOMAP_DELAY) { BUG_ON(direct); @@ -1014,9 +1003,11 @@ __linvfs_get_block( } if (blocks) { - bh_result->b_size = (ssize_t)min( - (loff_t)(iomap.iomap_bsize - iomap.iomap_delta), - (loff_t)(blocks << inode->i_blkbits)); + ASSERT(iomap.iomap_bsize - iomap.iomap_delta > 0); + offset = min_t(xfs_off_t, + iomap.iomap_bsize - iomap.iomap_delta, + (xfs_off_t)blocks << inode->i_blkbits); + bh_result->b_size = (u32) min_t(xfs_off_t, UINT_MAX, offset); } return 0; @@ -1296,7 +1287,7 @@ linvfs_invalidate_page( STATIC int linvfs_release_page( struct page *page, - int gfp_mask) + gfp_t gfp_mask) { struct inode *inode = page->mapping->host; int dirty, delalloc, unmapped, unwritten; diff --git a/fs/xfs/linux-2.6/xfs_aops.h b/fs/xfs/linux-2.6/xfs_aops.h index 2fa62974a04d..4720758a9ade 100644 --- a/fs/xfs/linux-2.6/xfs_aops.h +++ b/fs/xfs/linux-2.6/xfs_aops.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2005 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_AOPS_H__ #define __XFS_AOPS_H__ diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c index e82cf72ac599..6fe21d2b8847 100644 --- a/fs/xfs/linux-2.6/xfs_buf.c +++ b/fs/xfs/linux-2.6/xfs_buf.c @@ -1,46 +1,20 @@ /* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. - * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ - */ - -/* - * The xfs_buf.c code provides an abstract buffer cache model on top - * of the Linux page cache. Cached metadata blocks for a file system - * are hashed to the inode for the block device. xfs_buf.c assembles - * buffers (xfs_buf_t) on demand to aggregate such cached pages for I/O. - * - * Written by Steve Lord, Jim Mostek, Russell Cattelan - * and Rajagopal Ananthanarayanan ("ananth") at SGI. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include <linux/stddef.h> #include <linux/errno.h> #include <linux/slab.h> @@ -55,25 +29,16 @@ #include <linux/blkdev.h> #include <linux/hash.h> #include <linux/kthread.h> - #include "xfs_linux.h" -/* - * File wide globals - */ - STATIC kmem_cache_t *pagebuf_zone; STATIC kmem_shaker_t pagebuf_shake; -STATIC int xfsbufd_wakeup(int, unsigned int); +STATIC int xfsbufd_wakeup(int, gfp_t); STATIC void pagebuf_delwri_queue(xfs_buf_t *, int); STATIC struct workqueue_struct *xfslogd_workqueue; struct workqueue_struct *xfsdatad_workqueue; -/* - * Pagebuf debugging - */ - #ifdef PAGEBUF_TRACE void pagebuf_trace( @@ -112,10 +77,6 @@ ktrace_t *pagebuf_trace_buf; # define PB_GET_OWNER(pb) do { } while (0) #endif -/* - * Pagebuf allocation / freeing. - */ - #define pb_to_gfp(flags) \ ((((flags) & PBF_READ_AHEAD) ? __GFP_NORETRY : \ ((flags) & PBF_DONT_BLOCK) ? GFP_NOFS : GFP_KERNEL) | __GFP_NOWARN) @@ -123,7 +84,6 @@ ktrace_t *pagebuf_trace_buf; #define pb_to_km(flags) \ (((flags) & PBF_DONT_BLOCK) ? KM_NOFS : KM_SLEEP) - #define pagebuf_allocate(flags) \ kmem_zone_alloc(pagebuf_zone, pb_to_km(flags)) #define pagebuf_deallocate(pb) \ @@ -181,8 +141,9 @@ set_page_region( size_t offset, size_t length) { - page->private |= page_region_mask(offset, length); - if (page->private == ~0UL) + set_page_private(page, + page_private(page) | page_region_mask(offset, length)); + if (page_private(page) == ~0UL) SetPageUptodate(page); } @@ -194,7 +155,7 @@ test_page_region( { unsigned long mask = page_region_mask(offset, length); - return (mask && (page->private & mask) == mask); + return (mask && (page_private(page) & mask) == mask); } /* @@ -285,7 +246,7 @@ _pagebuf_initialize( * most cases but may be reset (e.g. XFS recovery). */ pb->pb_buffer_length = pb->pb_count_desired = range_length; - pb->pb_flags = flags | PBF_NONE; + pb->pb_flags = flags; pb->pb_bn = XFS_BUF_DADDR_NULL; atomic_set(&pb->pb_pin_count, 0); init_waitqueue_head(&pb->pb_waiters); @@ -383,7 +344,7 @@ _pagebuf_lookup_pages( size_t blocksize = bp->pb_target->pbr_bsize; size_t size = bp->pb_count_desired; size_t nbytes, offset; - int gfp_mask = pb_to_gfp(flags); + gfp_t gfp_mask = pb_to_gfp(flags); unsigned short page_count, i; pgoff_t first; loff_t end; @@ -457,14 +418,8 @@ _pagebuf_lookup_pages( unlock_page(bp->pb_pages[i]); } - if (page_count) { - /* if we have any uptodate pages, mark that in the buffer */ - bp->pb_flags &= ~PBF_NONE; - - /* if some pages aren't uptodate, mark that in the buffer */ - if (page_count != bp->pb_page_count) - bp->pb_flags |= PBF_PARTIAL; - } + if (page_count == bp->pb_page_count) + bp->pb_flags |= PBF_DONE; PB_TRACE(bp, "lookup_pages", (long)page_count); return error; @@ -675,7 +630,7 @@ xfs_buf_read_flags( pb = xfs_buf_get_flags(target, ioff, isize, flags); if (pb) { - if (PBF_NOT_DONE(pb)) { + if (!XFS_BUF_ISDONE(pb)) { PB_TRACE(pb, "read", (unsigned long)flags); XFS_STATS_INC(pb_get_read); pagebuf_iostart(pb, flags); @@ -812,7 +767,7 @@ pagebuf_get_no_daddr( bp = pagebuf_allocate(0); if (unlikely(bp == NULL)) goto fail; - _pagebuf_initialize(bp, target, 0, len, PBF_FORCEIO); + _pagebuf_initialize(bp, target, 0, len, 0); try_again: data = kmem_alloc(malloc_len, KM_SLEEP | KM_MAYFAIL); @@ -875,39 +830,18 @@ pagebuf_rele( PB_TRACE(pb, "rele", pb->pb_relse); - /* - * pagebuf_lookup buffers are not hashed, not delayed write, - * and don't have their own release routines. Special case. - */ - if (unlikely(!hash)) { - ASSERT(!pb->pb_relse); - if (atomic_dec_and_test(&pb->pb_hold)) - xfs_buf_free(pb); - return; - } - if (atomic_dec_and_lock(&pb->pb_hold, &hash->bh_lock)) { - int do_free = 1; - if (pb->pb_relse) { atomic_inc(&pb->pb_hold); spin_unlock(&hash->bh_lock); (*(pb->pb_relse)) (pb); - spin_lock(&hash->bh_lock); - do_free = 0; - } - - if (pb->pb_flags & PBF_FS_MANAGED) { - do_free = 0; - } - - if (do_free) { - ASSERT((pb->pb_flags & (PBF_DELWRI|_PBF_DELWRI_Q)) == 0); - list_del_init(&pb->pb_hash_list); + } else if (pb->pb_flags & PBF_FS_MANAGED) { spin_unlock(&hash->bh_lock); - pagebuf_free(pb); } else { + ASSERT(!(pb->pb_flags & (PBF_DELWRI|_PBF_DELWRI_Q))); + list_del_init(&pb->pb_hash_list); spin_unlock(&hash->bh_lock); + pagebuf_free(pb); } } else { /* @@ -1120,21 +1054,18 @@ pagebuf_iodone_work( void pagebuf_iodone( xfs_buf_t *pb, - int dataio, int schedule) { pb->pb_flags &= ~(PBF_READ | PBF_WRITE); - if (pb->pb_error == 0) { - pb->pb_flags &= ~(PBF_PARTIAL | PBF_NONE); - } + if (pb->pb_error == 0) + pb->pb_flags |= PBF_DONE; PB_TRACE(pb, "iodone", pb->pb_iodone); if ((pb->pb_iodone) || (pb->pb_flags & PBF_ASYNC)) { if (schedule) { INIT_WORK(&pb->pb_iodone_work, pagebuf_iodone_work, pb); - queue_work(dataio ? xfsdatad_workqueue : - xfslogd_workqueue, &pb->pb_iodone_work); + queue_work(xfslogd_workqueue, &pb->pb_iodone_work); } else { pagebuf_iodone_work(pb); } @@ -1234,7 +1165,7 @@ _pagebuf_iodone( { if (atomic_dec_and_test(&pb->pb_io_remaining) == 1) { pb->pb_locked = 0; - pagebuf_iodone(pb, (pb->pb_flags & PBF_FS_DATAIOD), schedule); + pagebuf_iodone(pb, schedule); } } @@ -1303,6 +1234,11 @@ _pagebuf_ioapply( rw = (pb->pb_flags & PBF_READ) ? READ : WRITE; } + if (pb->pb_flags & PBF_ORDERED) { + ASSERT(!(pb->pb_flags & PBF_READ)); + rw = WRITE_BARRIER; + } + /* Special code path for reading a sub page size pagebuf in -- * we populate up the whole page, and hence the other metadata * in the same page. This optimization is only valid when the @@ -1750,7 +1686,7 @@ STATIC int xfsbufd_force_sleep; STATIC int xfsbufd_wakeup( int priority, - unsigned int mask) + gfp_t mask) { if (xfsbufd_force_sleep) return 0; @@ -1780,8 +1716,8 @@ xfsbufd( xfsbufd_force_sleep = 0; } - schedule_timeout_interruptible - (xfs_buf_timer_centisecs * msecs_to_jiffies(10)); + schedule_timeout_interruptible( + xfs_buf_timer_centisecs * msecs_to_jiffies(10)); age = xfs_buf_age_centisecs * msecs_to_jiffies(10); spin_lock(&pbd_delwrite_lock); @@ -1890,14 +1826,22 @@ xfs_flush_buftarg( return pincount; } -STATIC int -xfs_buf_daemons_start(void) +int __init +pagebuf_init(void) { int error = -ENOMEM; +#ifdef PAGEBUF_TRACE + pagebuf_trace_buf = ktrace_alloc(PAGEBUF_TRACE_SIZE, KM_SLEEP); +#endif + + pagebuf_zone = kmem_zone_init(sizeof(xfs_buf_t), "xfs_buf"); + if (!pagebuf_zone) + goto out_free_trace_buf; + xfslogd_workqueue = create_workqueue("xfslogd"); if (!xfslogd_workqueue) - goto out; + goto out_free_buf_zone; xfsdatad_workqueue = create_workqueue("xfsdatad"); if (!xfsdatad_workqueue) @@ -1908,82 +1852,37 @@ xfs_buf_daemons_start(void) error = PTR_ERR(xfsbufd_task); goto out_destroy_xfsdatad_workqueue; } + + pagebuf_shake = kmem_shake_register(xfsbufd_wakeup); + if (!pagebuf_shake) + goto out_stop_xfsbufd; + return 0; + out_stop_xfsbufd: + kthread_stop(xfsbufd_task); out_destroy_xfsdatad_workqueue: destroy_workqueue(xfsdatad_workqueue); out_destroy_xfslogd_workqueue: destroy_workqueue(xfslogd_workqueue); - out: - return error; -} - -/* - * Note: do not mark as __exit, it is called from pagebuf_terminate. - */ -STATIC void -xfs_buf_daemons_stop(void) -{ - kthread_stop(xfsbufd_task); - destroy_workqueue(xfslogd_workqueue); - destroy_workqueue(xfsdatad_workqueue); -} - -/* - * Initialization and Termination - */ - -int __init -pagebuf_init(void) -{ - int error = -ENOMEM; - - pagebuf_zone = kmem_zone_init(sizeof(xfs_buf_t), "xfs_buf"); - if (!pagebuf_zone) - goto out; - -#ifdef PAGEBUF_TRACE - pagebuf_trace_buf = ktrace_alloc(PAGEBUF_TRACE_SIZE, KM_SLEEP); -#endif - - error = xfs_buf_daemons_start(); - if (error) - goto out_free_buf_zone; - - pagebuf_shake = kmem_shake_register(xfsbufd_wakeup); - if (!pagebuf_shake) { - error = -ENOMEM; - goto out_stop_daemons; - } - - return 0; - - out_stop_daemons: - xfs_buf_daemons_stop(); out_free_buf_zone: + kmem_zone_destroy(pagebuf_zone); + out_free_trace_buf: #ifdef PAGEBUF_TRACE ktrace_free(pagebuf_trace_buf); #endif - kmem_zone_destroy(pagebuf_zone); - out: return error; } - -/* - * pagebuf_terminate. - * - * Note: do not mark as __exit, this is also called from the __init code. - */ void pagebuf_terminate(void) { - xfs_buf_daemons_stop(); - + kmem_shake_deregister(pagebuf_shake); + kthread_stop(xfsbufd_task); + destroy_workqueue(xfsdatad_workqueue); + destroy_workqueue(xfslogd_workqueue); + kmem_zone_destroy(pagebuf_zone); #ifdef PAGEBUF_TRACE ktrace_free(pagebuf_trace_buf); #endif - - kmem_zone_destroy(pagebuf_zone); - kmem_shake_deregister(pagebuf_shake); } diff --git a/fs/xfs/linux-2.6/xfs_buf.h b/fs/xfs/linux-2.6/xfs_buf.h index 67c19f799232..237a35b915d1 100644 --- a/fs/xfs/linux-2.6/xfs_buf.h +++ b/fs/xfs/linux-2.6/xfs_buf.h @@ -1,39 +1,20 @@ /* - * Copyright (c) 2000-2004 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - -/* - * Written by Steve Lord, Jim Mostek, Russell Cattelan at SGI - */ - #ifndef __XFS_BUF_H__ #define __XFS_BUF_H__ @@ -69,15 +50,12 @@ typedef enum page_buf_flags_e { /* pb_flags values */ PBF_READ = (1 << 0), /* buffer intended for reading from device */ PBF_WRITE = (1 << 1), /* buffer intended for writing to device */ PBF_MAPPED = (1 << 2), /* buffer mapped (pb_addr valid) */ - PBF_PARTIAL = (1 << 3), /* buffer partially read */ PBF_ASYNC = (1 << 4), /* initiator will not wait for completion */ - PBF_NONE = (1 << 5), /* buffer not read at all */ + PBF_DONE = (1 << 5), /* all pages in the buffer uptodate */ PBF_DELWRI = (1 << 6), /* buffer has dirty pages */ PBF_STALE = (1 << 7), /* buffer has been staled, do not find it */ PBF_FS_MANAGED = (1 << 8), /* filesystem controls freeing memory */ - PBF_FS_DATAIOD = (1 << 9), /* schedule IO completion on fs datad */ - PBF_FORCEIO = (1 << 10), /* ignore any cache state */ - PBF_FLUSH = (1 << 11), /* flush disk write cache */ + PBF_ORDERED = (1 << 11), /* use ordered writes */ PBF_READ_AHEAD = (1 << 12), /* asynchronous read-ahead */ /* flags used only as arguments to access routines */ @@ -92,9 +70,6 @@ typedef enum page_buf_flags_e { /* pb_flags values */ _PBF_DELWRI_Q = (1 << 21), /* buffer on delwri queue */ } page_buf_flags_t; -#define PBF_UPDATE (PBF_READ | PBF_WRITE) -#define PBF_NOT_DONE(pb) (((pb)->pb_flags & (PBF_PARTIAL|PBF_NONE)) != 0) -#define PBF_DONE(pb) (((pb)->pb_flags & (PBF_PARTIAL|PBF_NONE)) == 0) typedef struct xfs_bufhash { struct list_head bh_list; @@ -258,7 +233,6 @@ extern void pagebuf_unlock( /* unlock buffer */ extern void pagebuf_iodone( /* mark buffer I/O complete */ xfs_buf_t *, /* buffer to mark */ - int, /* use data/log helper thread. */ int); /* run completion locally, or in * a helper thread. */ @@ -378,21 +352,21 @@ extern void pagebuf_trace( #define XFS_BUF_GETERROR(x) pagebuf_geterror(x) #define XFS_BUF_ISERROR(x) (pagebuf_geterror(x)?1:0) -#define XFS_BUF_DONE(x) ((x)->pb_flags &= ~(PBF_PARTIAL|PBF_NONE)) -#define XFS_BUF_UNDONE(x) ((x)->pb_flags |= PBF_PARTIAL|PBF_NONE) -#define XFS_BUF_ISDONE(x) (!(PBF_NOT_DONE(x))) +#define XFS_BUF_DONE(x) ((x)->pb_flags |= PBF_DONE) +#define XFS_BUF_UNDONE(x) ((x)->pb_flags &= ~PBF_DONE) +#define XFS_BUF_ISDONE(x) ((x)->pb_flags & PBF_DONE) -#define XFS_BUF_BUSY(x) ((x)->pb_flags |= PBF_FORCEIO) -#define XFS_BUF_UNBUSY(x) ((x)->pb_flags &= ~PBF_FORCEIO) +#define XFS_BUF_BUSY(x) do { } while (0) +#define XFS_BUF_UNBUSY(x) do { } while (0) #define XFS_BUF_ISBUSY(x) (1) #define XFS_BUF_ASYNC(x) ((x)->pb_flags |= PBF_ASYNC) #define XFS_BUF_UNASYNC(x) ((x)->pb_flags &= ~PBF_ASYNC) #define XFS_BUF_ISASYNC(x) ((x)->pb_flags & PBF_ASYNC) -#define XFS_BUF_FLUSH(x) ((x)->pb_flags |= PBF_FLUSH) -#define XFS_BUF_UNFLUSH(x) ((x)->pb_flags &= ~PBF_FLUSH) -#define XFS_BUF_ISFLUSH(x) ((x)->pb_flags & PBF_FLUSH) +#define XFS_BUF_ORDERED(x) ((x)->pb_flags |= PBF_ORDERED) +#define XFS_BUF_UNORDERED(x) ((x)->pb_flags &= ~PBF_ORDERED) +#define XFS_BUF_ISORDERED(x) ((x)->pb_flags & PBF_ORDERED) #define XFS_BUF_SHUT(x) printk("XFS_BUF_SHUT not implemented yet\n") #define XFS_BUF_UNSHUT(x) printk("XFS_BUF_UNSHUT not implemented yet\n") @@ -412,9 +386,6 @@ extern void pagebuf_trace( #define XFS_BUF_BP_ISMAPPED(bp) 1 -#define XFS_BUF_DATAIO(x) ((x)->pb_flags |= PBF_FS_DATAIOD) -#define XFS_BUF_UNDATAIO(x) ((x)->pb_flags &= ~PBF_FS_DATAIOD) - #define XFS_BUF_IODONE_FUNC(buf) (buf)->pb_iodone #define XFS_BUF_SET_IODONE_FUNC(buf, func) \ (buf)->pb_iodone = (func) @@ -510,7 +481,7 @@ static inline void xfs_buf_relse(xfs_buf_t *bp) pagebuf_trace(bp, id, NULL, (void *)__builtin_return_address(0)) #define xfs_biodone(pb) \ - pagebuf_iodone(pb, (pb->pb_flags & PBF_FS_DATAIOD), 0) + pagebuf_iodone(pb, 0) #define xfs_biomove(pb, off, len, data, rw) \ pagebuf_iomove((pb), (off), (len), (data), \ diff --git a/fs/xfs/linux-2.6/xfs_cred.h b/fs/xfs/linux-2.6/xfs_cred.h index 00c45849d41a..4af491024727 100644 --- a/fs/xfs/linux-2.6/xfs_cred.h +++ b/fs/xfs/linux-2.6/xfs_cred.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_CRED_H__ #define __XFS_CRED_H__ diff --git a/fs/xfs/linux-2.6/xfs_export.c b/fs/xfs/linux-2.6/xfs_export.c index f372a1a5e168..80eb249f2fa0 100644 --- a/fs/xfs/linux-2.6/xfs_export.c +++ b/fs/xfs/linux-2.6/xfs_export.c @@ -1,35 +1,20 @@ /* - * Copyright (c) 2004-2005 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2004-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" #include "xfs_types.h" #include "xfs_dmapi.h" @@ -41,7 +26,7 @@ #include "xfs_export.h" /* - * XFS encode and decodes the fileid portion of NFS filehandles + * XFS encodes and decodes the fileid portion of NFS filehandles * itself instead of letting the generic NFS code do it. This * allows filesystems with 64 bit inode numbers to be exported. * @@ -51,7 +36,6 @@ * remains in that code. */ - STATIC struct dentry * linvfs_decode_fh( struct super_block *sb, @@ -92,7 +76,7 @@ linvfs_decode_fh( p = xfs_fileid_decode_fid2(p, &pfid, is64); parent = &pfid; } - + fh = (__u32 *)&ifid; return find_exported_dentry(sb, fh, parent, acceptable, context); } @@ -112,9 +96,8 @@ linvfs_encode_fh( int is64 = 0; #if XFS_BIG_INUMS vfs_t *vfs = LINVFS_GET_VFS(inode->i_sb); - xfs_mount_t *mp = XFS_VFSTOM(vfs); - - if (!(mp->m_flags & XFS_MOUNT_32BITINOOPT)) { + + if (!(vfs->vfs_flag & VFS_32BITINODES)) { /* filesystem may contain 64bit inode numbers */ is64 = XFS_FILEID_TYPE_64FLAG; } diff --git a/fs/xfs/linux-2.6/xfs_export.h b/fs/xfs/linux-2.6/xfs_export.h index 60b2abac1c18..e5b0559700a4 100644 --- a/fs/xfs/linux-2.6/xfs_export.h +++ b/fs/xfs/linux-2.6/xfs_export.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2005 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_EXPORT_H__ #define __XFS_EXPORT_H__ diff --git a/fs/xfs/linux-2.6/xfs_file.c b/fs/xfs/linux-2.6/xfs_file.c index 3881622bcf08..06111d0bbae4 100644 --- a/fs/xfs/linux-2.6/xfs_file.c +++ b/fs/xfs/linux-2.6/xfs_file.c @@ -1,39 +1,26 @@ /* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_sb.h" +#include "xfs_ag.h" #include "xfs_dir.h" #include "xfs_dir2.h" #include "xfs_trans.h" diff --git a/fs/xfs/linux-2.6/xfs_fs_subr.c b/fs/xfs/linux-2.6/xfs_fs_subr.c index 05ebd30ec96f..f89340c61bf2 100644 --- a/fs/xfs/linux-2.6/xfs_fs_subr.c +++ b/fs/xfs/linux-2.6/xfs_fs_subr.c @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #include "xfs.h" @@ -117,6 +103,8 @@ fs_flush_pages( if (VN_CACHED(vp)) { filemap_fdatawrite(ip->i_mapping); + if (flags & XFS_B_ASYNC) + return 0; filemap_fdatawait(ip->i_mapping); } diff --git a/fs/xfs/linux-2.6/xfs_fs_subr.h b/fs/xfs/linux-2.6/xfs_fs_subr.h index 2db9ddbd4567..aee9ccdd18f7 100644 --- a/fs/xfs/linux-2.6/xfs_fs_subr.h +++ b/fs/xfs/linux-2.6/xfs_fs_subr.h @@ -1,48 +1,29 @@ /* - * Copyright (c) 2000, 2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000,2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ - */ -#ifndef __XFS_SUBR_H__ -#define __XFS_SUBR_H__ - -/* - * Utilities shared among file system implementations. + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ +#ifndef __XFS_FS_SUBR_H__ +#define __XFS_FS_SUBR_H__ struct cred; - -extern int fs_noerr(void); -extern int fs_nosys(void); -extern void fs_noval(void); -extern void fs_tosspages(bhv_desc_t *, xfs_off_t, xfs_off_t, int); -extern void fs_flushinval_pages(bhv_desc_t *, xfs_off_t, xfs_off_t, int); -extern int fs_flush_pages(bhv_desc_t *, xfs_off_t, xfs_off_t, uint64_t, int); +extern int fs_noerr(void); +extern int fs_nosys(void); +extern void fs_noval(void); +extern void fs_tosspages(bhv_desc_t *, xfs_off_t, xfs_off_t, int); +extern void fs_flushinval_pages(bhv_desc_t *, xfs_off_t, xfs_off_t, int); +extern int fs_flush_pages(bhv_desc_t *, xfs_off_t, xfs_off_t, uint64_t, int); #endif /* __XFS_FS_SUBR_H__ */ diff --git a/fs/xfs/linux-2.6/xfs_globals.c b/fs/xfs/linux-2.6/xfs_globals.c index a6da5b4fd240..6e8085f34635 100644 --- a/fs/xfs/linux-2.6/xfs_globals.c +++ b/fs/xfs/linux-2.6/xfs_globals.c @@ -1,40 +1,20 @@ /* - * Copyright (c) 2000-2004 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - -/* - * This file contains globals needed by XFS that were normally defined - * somewhere else in IRIX. - */ - #include "xfs.h" #include "xfs_cred.h" #include "xfs_sysctl.h" diff --git a/fs/xfs/linux-2.6/xfs_globals.h b/fs/xfs/linux-2.6/xfs_globals.h index e81e2f38a853..e1a22bfcf865 100644 --- a/fs/xfs/linux-2.6/xfs_globals.h +++ b/fs/xfs/linux-2.6/xfs_globals.h @@ -1,42 +1,23 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_GLOBALS_H__ #define __XFS_GLOBALS_H__ -/* - * This file declares globals needed by XFS that were normally defined - * somewhere else in IRIX. - */ - extern uint64_t xfs_panic_mask; /* set to cause more panics */ extern unsigned long xfs_physmem; extern struct cred *sys_cred; diff --git a/fs/xfs/linux-2.6/xfs_ioctl.c b/fs/xfs/linux-2.6/xfs_ioctl.c index 6a3326bcd8d0..b78b5eb9e96c 100644 --- a/fs/xfs/linux-2.6/xfs_ioctl.c +++ b/fs/xfs/linux-2.6/xfs_ioctl.c @@ -1,67 +1,52 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" - #include "xfs_fs.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" +#include "xfs_ag.h" #include "xfs_dir.h" #include "xfs_dir2.h" #include "xfs_alloc.h" #include "xfs_dmapi.h" #include "xfs_mount.h" -#include "xfs_alloc_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_btree.h" -#include "xfs_ialloc.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dir2_sf.h" #include "xfs_dinode.h" #include "xfs_inode.h" -#include "xfs_bmap.h" -#include "xfs_bit.h" +#include "xfs_btree.h" +#include "xfs_ialloc.h" #include "xfs_rtalloc.h" -#include "xfs_error.h" #include "xfs_itable.h" +#include "xfs_error.h" #include "xfs_rw.h" #include "xfs_acl.h" #include "xfs_cap.h" #include "xfs_mac.h" #include "xfs_attr.h" +#include "xfs_bmap.h" #include "xfs_buf_item.h" #include "xfs_utils.h" #include "xfs_dfrag.h" diff --git a/fs/xfs/linux-2.6/xfs_ioctl32.c b/fs/xfs/linux-2.6/xfs_ioctl32.c index 4636b7f86f1f..c83ae15bb0e6 100644 --- a/fs/xfs/linux-2.6/xfs_ioctl32.c +++ b/fs/xfs/linux-2.6/xfs_ioctl32.c @@ -1,35 +1,20 @@ /* - * Copyright (c) 2004-2005 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2004-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include <linux/config.h> #include <linux/compat.h> #include <linux/init.h> @@ -39,7 +24,6 @@ #include <linux/types.h> #include <linux/fs.h> #include <asm/uaccess.h> - #include "xfs.h" #include "xfs_types.h" #include "xfs_fs.h" diff --git a/fs/xfs/linux-2.6/xfs_ioctl32.h b/fs/xfs/linux-2.6/xfs_ioctl32.h index c874793a1dc9..011c273bec50 100644 --- a/fs/xfs/linux-2.6/xfs_ioctl32.h +++ b/fs/xfs/linux-2.6/xfs_ioctl32.h @@ -1,34 +1,24 @@ /* - * Copyright (c) 2004-2005 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2004-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ +#ifndef __XFS_IOCTL32_H__ +#define __XFS_IOCTL32_H__ + +extern long linvfs_compat_ioctl(struct file *, unsigned, unsigned long); +extern long linvfs_compat_invis_ioctl(struct file *f, unsigned, unsigned long); -long linvfs_compat_ioctl(struct file *f, unsigned cmd, unsigned long arg); -long linvfs_compat_invis_ioctl(struct file *f, unsigned cmd, unsigned long arg); +#endif /* __XFS_IOCTL32_H__ */ diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c index 77708a8c9f87..14215a7db59f 100644 --- a/fs/xfs/linux-2.6/xfs_iops.c +++ b/fs/xfs/linux-2.6/xfs_iops.c @@ -1,39 +1,25 @@ /* - * Copyright (c) 2000-2004 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" #include "xfs_fs.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" @@ -43,18 +29,17 @@ #include "xfs_dmapi.h" #include "xfs_quota.h" #include "xfs_mount.h" -#include "xfs_alloc_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_btree.h" -#include "xfs_ialloc.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" #include "xfs_inode.h" #include "xfs_bmap.h" -#include "xfs_bit.h" +#include "xfs_btree.h" +#include "xfs_ialloc.h" #include "xfs_rtalloc.h" #include "xfs_error.h" #include "xfs_itable.h" @@ -69,6 +54,137 @@ #include <linux/xattr.h> #include <linux/namei.h> +/* + * Change the requested timestamp in the given inode. + * We don't lock across timestamp updates, and we don't log them but + * we do record the fact that there is dirty information in core. + * + * NOTE -- callers MUST combine XFS_ICHGTIME_MOD or XFS_ICHGTIME_CHG + * with XFS_ICHGTIME_ACC to be sure that access time + * update will take. Calling first with XFS_ICHGTIME_ACC + * and then XFS_ICHGTIME_MOD may fail to modify the access + * timestamp if the filesystem is mounted noacctm. + */ +void +xfs_ichgtime( + xfs_inode_t *ip, + int flags) +{ + struct inode *inode = LINVFS_GET_IP(XFS_ITOV(ip)); + timespec_t tv; + + /* + * We're not supposed to change timestamps in readonly-mounted + * filesystems. Throw it away if anyone asks us. + */ + if (unlikely(IS_RDONLY(inode))) + return; + + /* + * Don't update access timestamps on reads if mounted "noatime". + * Throw it away if anyone asks us. + */ + if (unlikely( + (ip->i_mount->m_flags & XFS_MOUNT_NOATIME || IS_NOATIME(inode)) && + (flags & (XFS_ICHGTIME_ACC|XFS_ICHGTIME_MOD|XFS_ICHGTIME_CHG)) == + XFS_ICHGTIME_ACC)) + return; + + nanotime(&tv); + if (flags & XFS_ICHGTIME_MOD) { + inode->i_mtime = tv; + ip->i_d.di_mtime.t_sec = (__int32_t)tv.tv_sec; + ip->i_d.di_mtime.t_nsec = (__int32_t)tv.tv_nsec; + } + if (flags & XFS_ICHGTIME_ACC) { + inode->i_atime = tv; + ip->i_d.di_atime.t_sec = (__int32_t)tv.tv_sec; + ip->i_d.di_atime.t_nsec = (__int32_t)tv.tv_nsec; + } + if (flags & XFS_ICHGTIME_CHG) { + inode->i_ctime = tv; + ip->i_d.di_ctime.t_sec = (__int32_t)tv.tv_sec; + ip->i_d.di_ctime.t_nsec = (__int32_t)tv.tv_nsec; + } + + /* + * We update the i_update_core field _after_ changing + * the timestamps in order to coordinate properly with + * xfs_iflush() so that we don't lose timestamp updates. + * This keeps us from having to hold the inode lock + * while doing this. We use the SYNCHRONIZE macro to + * ensure that the compiler does not reorder the update + * of i_update_core above the timestamp updates above. + */ + SYNCHRONIZE(); + ip->i_update_core = 1; + if (!(inode->i_state & I_LOCK)) + mark_inode_dirty_sync(inode); +} + +/* + * Variant on the above which avoids querying the system clock + * in situations where we know the Linux inode timestamps have + * just been updated (and so we can update our inode cheaply). + * We also skip the readonly and noatime checks here, they are + * also catered for already. + */ +void +xfs_ichgtime_fast( + xfs_inode_t *ip, + struct inode *inode, + int flags) +{ + timespec_t *tvp; + + /* + * We're not supposed to change timestamps in readonly-mounted + * filesystems. Throw it away if anyone asks us. + */ + if (unlikely(IS_RDONLY(inode))) + return; + + /* + * Don't update access timestamps on reads if mounted "noatime". + * Throw it away if anyone asks us. + */ + if (unlikely( + (ip->i_mount->m_flags & XFS_MOUNT_NOATIME || IS_NOATIME(inode)) && + ((flags & (XFS_ICHGTIME_ACC|XFS_ICHGTIME_MOD|XFS_ICHGTIME_CHG)) == + XFS_ICHGTIME_ACC))) + return; + + if (flags & XFS_ICHGTIME_MOD) { + tvp = &inode->i_mtime; + ip->i_d.di_mtime.t_sec = (__int32_t)tvp->tv_sec; + ip->i_d.di_mtime.t_nsec = (__int32_t)tvp->tv_nsec; + } + if (flags & XFS_ICHGTIME_ACC) { + tvp = &inode->i_atime; + ip->i_d.di_atime.t_sec = (__int32_t)tvp->tv_sec; + ip->i_d.di_atime.t_nsec = (__int32_t)tvp->tv_nsec; + } + if (flags & XFS_ICHGTIME_CHG) { + tvp = &inode->i_ctime; + ip->i_d.di_ctime.t_sec = (__int32_t)tvp->tv_sec; + ip->i_d.di_ctime.t_nsec = (__int32_t)tvp->tv_nsec; + } + + /* + * We update the i_update_core field _after_ changing + * the timestamps in order to coordinate properly with + * xfs_iflush() so that we don't lose timestamp updates. + * This keeps us from having to hold the inode lock + * while doing this. We use the SYNCHRONIZE macro to + * ensure that the compiler does not reorder the update + * of i_update_core above the timestamp updates above. + */ + SYNCHRONIZE(); + ip->i_update_core = 1; + if (!(inode->i_state & I_LOCK)) + mark_inode_dirty_sync(inode); +} + /* * Pull the link count and size up from the xfs inode to the linux inode diff --git a/fs/xfs/linux-2.6/xfs_iops.h b/fs/xfs/linux-2.6/xfs_iops.h index 6a69a62c36b0..ee784b63acbf 100644 --- a/fs/xfs/linux-2.6/xfs_iops.h +++ b/fs/xfs/linux-2.6/xfs_iops.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_IOPS_H__ #define __XFS_IOPS_H__ @@ -48,4 +34,8 @@ extern void linvfs_unwritten_done(struct buffer_head *, int); extern int xfs_ioctl(struct bhv_desc *, struct inode *, struct file *, int, unsigned int, void __user *); +struct xfs_inode; +extern void xfs_ichgtime(struct xfs_inode *, int); +extern void xfs_ichgtime_fast(struct xfs_inode *, struct inode *, int); + #endif /* __XFS_IOPS_H__ */ diff --git a/fs/xfs/linux-2.6/xfs_linux.h b/fs/xfs/linux-2.6/xfs_linux.h index 68c5d885ed9c..d8e21ba0cccc 100644 --- a/fs/xfs/linux-2.6/xfs_linux.h +++ b/fs/xfs/linux-2.6/xfs_linux.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2004 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_LINUX__ #define __XFS_LINUX__ @@ -86,7 +72,6 @@ #include <linux/init.h> #include <linux/list.h> #include <linux/proc_fs.h> -#include <linux/version.h> #include <linux/sort.h> #include <asm/page.h> @@ -197,10 +182,6 @@ static inline void set_buffer_unwritten_io(struct buffer_head *bh) /* bytes to clicks */ #define btoc(x) (((__psunsigned_t)(x)+(NBPC-1))>>BPCSHIFT) -#ifndef CELL_CAPABLE -#define FSC_NOTIFY_NAME_CHANGED(vp) -#endif - #ifndef ENOATTR #define ENOATTR ENODATA /* Attribute not found */ #endif @@ -235,30 +216,18 @@ static inline void set_buffer_unwritten_io(struct buffer_head *bh) #define Q_XSETPQLIM XQM_CMD(10) /* set projects disk limits */ #define Q_XGETPQUOTA XQM_CMD(11) /* get projects disk limits */ -/* IRIX uses a dynamic sizing algorithm (ndquot = 200 + numprocs*2) */ -/* we may well need to fine-tune this if it ever becomes an issue. */ -#define DQUOT_MAX_HEURISTIC 1024 /* NR_DQUOTS */ -#define ndquot DQUOT_MAX_HEURISTIC - -/* IRIX uses the current size of the name cache to guess a good value */ -/* - this isn't the same but is a good enough starting point for now. */ -#define DQUOT_HASH_HEURISTIC files_stat.nr_files - -/* IRIX inodes maintain the project ID also, zero this field on Linux */ -#define DEFAULT_PROJID 0 -#define dfltprid DEFAULT_PROJID - +#define dfltprid 0 #define MAXPATHLEN 1024 #define MIN(a,b) (min(a,b)) #define MAX(a,b) (max(a,b)) #define howmany(x, y) (((x)+((y)-1))/(y)) #define roundup(x, y) ((((x)+((y)-1))/(y))*(y)) -#define qsort(a,n,s,fn) sort(a,n,s,fn,NULL) /* * Various platform dependent calls that don't fit anywhere else */ +#define xfs_sort(a,n,s,fn) sort(a,n,s,fn,NULL) #define xfs_stack_trace() dump_stack() #define xfs_itruncate_data(ip, off) \ (-vmtruncate(LINVFS_GET_IP(XFS_ITOV(ip)), (off))) diff --git a/fs/xfs/linux-2.6/xfs_lrw.c b/fs/xfs/linux-2.6/xfs_lrw.c index 3b5fabe8dae9..279e9bc92aba 100644 --- a/fs/xfs/linux-2.6/xfs_lrw.c +++ b/fs/xfs/linux-2.6/xfs_lrw.c @@ -1,44 +1,25 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ -/* - * fs/xfs/linux/xfs_lrw.c (Linux Read Write stuff) - * - */ - #include "xfs.h" - #include "xfs_fs.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" @@ -48,18 +29,17 @@ #include "xfs_dmapi.h" #include "xfs_quota.h" #include "xfs_mount.h" -#include "xfs_alloc_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_btree.h" -#include "xfs_ialloc.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" #include "xfs_inode.h" #include "xfs_bmap.h" -#include "xfs_bit.h" +#include "xfs_btree.h" +#include "xfs_ialloc.h" #include "xfs_rtalloc.h" #include "xfs_error.h" #include "xfs_itable.h" @@ -302,7 +282,7 @@ xfs_read( xfs_iunlock(ip, XFS_IOLOCK_SHARED); if (likely(!(ioflags & IO_INVIS))) - xfs_ichgtime(ip, XFS_ICHGTIME_ACC); + xfs_ichgtime_fast(ip, inode, XFS_ICHGTIME_ACC); unlock_isem: if (unlikely(ioflags & IO_ISDIRECT)) @@ -367,7 +347,7 @@ xfs_sendfile( XFS_STATS_ADD(xs_read_bytes, ret); if (likely(!(ioflags & IO_INVIS))) - xfs_ichgtime(ip, XFS_ICHGTIME_ACC); + xfs_ichgtime_fast(ip, LINVFS_GET_IP(vp), XFS_ICHGTIME_ACC); return ret; } @@ -732,15 +712,10 @@ start: } } - /* - * On Linux, generic_file_write updates the times even if - * no data is copied in so long as the write had a size. - * - * We must update xfs' times since revalidate will overcopy xfs. - */ - if (!(ioflags & IO_INVIS)) { - xfs_ichgtime(xip, XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG); + if (likely(!(ioflags & IO_INVIS))) { inode_update_time(inode, 1); + xfs_ichgtime_fast(xip, inode, + XFS_ICHGTIME_MOD | XFS_ICHGTIME_CHG); } /* diff --git a/fs/xfs/linux-2.6/xfs_lrw.h b/fs/xfs/linux-2.6/xfs_lrw.h index 6294dcdb797c..38864a88d42d 100644 --- a/fs/xfs/linux-2.6/xfs_lrw.h +++ b/fs/xfs/linux-2.6/xfs_lrw.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_LRW_H__ #define __XFS_LRW_H__ @@ -107,9 +93,4 @@ extern ssize_t xfs_sendfile(struct bhv_desc *, struct file *, extern int xfs_dev_is_read_only(struct xfs_mount *, char *); -#define XFS_FSB_TO_DB_IO(io,fsb) \ - (((io)->io_flags & XFS_IOCORE_RT) ? \ - XFS_FSB_TO_BB((io)->io_mount, (fsb)) : \ - XFS_FSB_TO_DADDR((io)->io_mount, (fsb))) - #endif /* __XFS_LRW_H__ */ diff --git a/fs/xfs/linux-2.6/xfs_stats.c b/fs/xfs/linux-2.6/xfs_stats.c index aaf5ddba47f3..6c40a74be7c8 100644 --- a/fs/xfs/linux-2.6/xfs_stats.c +++ b/fs/xfs/linux-2.6/xfs_stats.c @@ -1,35 +1,20 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" #include <linux/proc_fs.h> diff --git a/fs/xfs/linux-2.6/xfs_stats.h b/fs/xfs/linux-2.6/xfs_stats.h index 3f756a6c3eb0..50027c4a5618 100644 --- a/fs/xfs/linux-2.6/xfs_stats.h +++ b/fs/xfs/linux-2.6/xfs_stats.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_STATS_H__ #define __XFS_STATS_H__ diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c index 2302454d8d47..6116b5bf433e 100644 --- a/fs/xfs/linux-2.6/xfs_super.c +++ b/fs/xfs/linux-2.6/xfs_super.c @@ -1,60 +1,45 @@ /* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" - -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" #include "xfs_clnt.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" +#include "xfs_ag.h" #include "xfs_dir.h" #include "xfs_dir2.h" #include "xfs_alloc.h" #include "xfs_dmapi.h" #include "xfs_quota.h" #include "xfs_mount.h" -#include "xfs_alloc_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_btree.h" -#include "xfs_ialloc.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" #include "xfs_inode.h" +#include "xfs_btree.h" +#include "xfs_ialloc.h" #include "xfs_bmap.h" -#include "xfs_bit.h" #include "xfs_rtalloc.h" #include "xfs_error.h" #include "xfs_itable.h" @@ -189,7 +174,7 @@ xfs_revalidate_inode( break; } - inode->i_blksize = PAGE_CACHE_SIZE; + inode->i_blksize = xfs_preferred_iosize(mp); inode->i_generation = ip->i_d.di_gen; i_size_write(inode, ip->i_d.di_size); inode->i_blocks = @@ -278,6 +263,72 @@ xfs_blkdev_put( close_bdev_excl(bdev); } +/* + * Try to write out the superblock using barriers. + */ +STATIC int +xfs_barrier_test( + xfs_mount_t *mp) +{ + xfs_buf_t *sbp = xfs_getsb(mp, 0); + int error; + + XFS_BUF_UNDONE(sbp); + XFS_BUF_UNREAD(sbp); + XFS_BUF_UNDELAYWRITE(sbp); + XFS_BUF_WRITE(sbp); + XFS_BUF_UNASYNC(sbp); + XFS_BUF_ORDERED(sbp); + + xfsbdstrat(mp, sbp); + error = xfs_iowait(sbp); + + /* + * Clear all the flags we set and possible error state in the + * buffer. We only did the write to try out whether barriers + * worked and shouldn't leave any traces in the superblock + * buffer. + */ + XFS_BUF_DONE(sbp); + XFS_BUF_ERROR(sbp, 0); + XFS_BUF_UNORDERED(sbp); + + xfs_buf_relse(sbp); + return error; +} + +void +xfs_mountfs_check_barriers(xfs_mount_t *mp) +{ + int error; + + if (mp->m_logdev_targp != mp->m_ddev_targp) { + xfs_fs_cmn_err(CE_NOTE, mp, + "Disabling barriers, not supported with external log device"); + mp->m_flags &= ~XFS_MOUNT_BARRIER; + } + + if (mp->m_ddev_targp->pbr_bdev->bd_disk->queue->ordered == + QUEUE_ORDERED_NONE) { + xfs_fs_cmn_err(CE_NOTE, mp, + "Disabling barriers, not supported by the underlying device"); + mp->m_flags &= ~XFS_MOUNT_BARRIER; + } + + error = xfs_barrier_test(mp); + if (error) { + xfs_fs_cmn_err(CE_NOTE, mp, + "Disabling barriers, trial barrier write failed"); + mp->m_flags &= ~XFS_MOUNT_BARRIER; + } +} + +void +xfs_blkdev_issue_flush( + xfs_buftarg_t *buftarg) +{ + blkdev_issue_flush(buftarg->pbr_bdev, NULL); +} STATIC struct inode * linvfs_alloc_inode( @@ -701,6 +752,18 @@ linvfs_show_options( } STATIC int +linvfs_quotasync( + struct super_block *sb, + int type) +{ + struct vfs *vfsp = LINVFS_GET_VFS(sb); + int error; + + VFS_QUOTACTL(vfsp, Q_XQUOTASYNC, 0, (caddr_t)NULL, error); + return -error; +} + +STATIC int linvfs_getxstate( struct super_block *sb, struct fs_quota_stat *fqs) @@ -868,6 +931,7 @@ STATIC struct super_operations linvfs_sops = { }; STATIC struct quotactl_ops linvfs_qops = { + .quota_sync = linvfs_quotasync, .get_xstate = linvfs_getxstate, .set_xstate = linvfs_setxstate, .get_xquota = linvfs_getxquota, diff --git a/fs/xfs/linux-2.6/xfs_super.h b/fs/xfs/linux-2.6/xfs_super.h index ec7e0035c731..df59408dca06 100644 --- a/fs/xfs/linux-2.6/xfs_super.h +++ b/fs/xfs/linux-2.6/xfs_super.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2004 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_SUPER_H__ #define __XFS_SUPER_H__ @@ -132,6 +118,7 @@ extern void xfs_flush_device(struct xfs_inode *); extern int xfs_blkdev_get(struct xfs_mount *, const char *, struct block_device **); extern void xfs_blkdev_put(struct block_device *); +extern void xfs_blkdev_issue_flush(struct xfs_buftarg *); extern struct export_operations linvfs_export_ops; diff --git a/fs/xfs/linux-2.6/xfs_sysctl.c b/fs/xfs/linux-2.6/xfs_sysctl.c index 0dc010356f4d..a02564972420 100644 --- a/fs/xfs/linux-2.6/xfs_sysctl.c +++ b/fs/xfs/linux-2.6/xfs_sysctl.c @@ -1,44 +1,26 @@ /* - * Copyright (c) 2001-2004 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2001-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" -#include "xfs_rw.h" #include <linux/sysctl.h> #include <linux/proc_fs.h> - static struct ctl_table_header *xfs_table_header; - #ifdef CONFIG_PROC_FS STATIC int xfs_stats_clear_proc_handler( @@ -76,7 +58,7 @@ xfs_stats_clear_proc_handler( STATIC ctl_table xfs_table[] = { {XFS_RESTRICT_CHOWN, "restrict_chown", &xfs_params.restrict_chown.val, sizeof(int), 0644, NULL, &proc_dointvec_minmax, - &sysctl_intvec, NULL, + &sysctl_intvec, NULL, &xfs_params.restrict_chown.min, &xfs_params.restrict_chown.max}, {XFS_SGID_INHERIT, "irix_sgid_inherit", &xfs_params.sgid_inherit.val, @@ -86,22 +68,22 @@ STATIC ctl_table xfs_table[] = { {XFS_SYMLINK_MODE, "irix_symlink_mode", &xfs_params.symlink_mode.val, sizeof(int), 0644, NULL, &proc_dointvec_minmax, - &sysctl_intvec, NULL, + &sysctl_intvec, NULL, &xfs_params.symlink_mode.min, &xfs_params.symlink_mode.max}, {XFS_PANIC_MASK, "panic_mask", &xfs_params.panic_mask.val, sizeof(int), 0644, NULL, &proc_dointvec_minmax, - &sysctl_intvec, NULL, + &sysctl_intvec, NULL, &xfs_params.panic_mask.min, &xfs_params.panic_mask.max}, {XFS_ERRLEVEL, "error_level", &xfs_params.error_level.val, sizeof(int), 0644, NULL, &proc_dointvec_minmax, - &sysctl_intvec, NULL, + &sysctl_intvec, NULL, &xfs_params.error_level.min, &xfs_params.error_level.max}, {XFS_SYNCD_TIMER, "xfssyncd_centisecs", &xfs_params.syncd_timer.val, sizeof(int), 0644, NULL, &proc_dointvec_minmax, - &sysctl_intvec, NULL, + &sysctl_intvec, NULL, &xfs_params.syncd_timer.min, &xfs_params.syncd_timer.max}, {XFS_INHERIT_SYNC, "inherit_sync", &xfs_params.inherit_sync.val, @@ -118,7 +100,7 @@ STATIC ctl_table xfs_table[] = { sizeof(int), 0644, NULL, &proc_dointvec_minmax, &sysctl_intvec, NULL, &xfs_params.inherit_noatim.min, &xfs_params.inherit_noatim.max}, - + {XFS_BUF_TIMER, "xfsbufd_centisecs", &xfs_params.xfs_buf_timer.val, sizeof(int), 0644, NULL, &proc_dointvec_minmax, &sysctl_intvec, NULL, @@ -136,14 +118,14 @@ STATIC ctl_table xfs_table[] = { {XFS_ROTORSTEP, "rotorstep", &xfs_params.rotorstep.val, sizeof(int), 0644, NULL, &proc_dointvec_minmax, - &sysctl_intvec, NULL, + &sysctl_intvec, NULL, &xfs_params.rotorstep.min, &xfs_params.rotorstep.max}, /* please keep this the last entry */ #ifdef CONFIG_PROC_FS {XFS_STATS_CLEAR, "stats_clear", &xfs_params.stats_clear.val, sizeof(int), 0644, NULL, &xfs_stats_clear_proc_handler, - &sysctl_intvec, NULL, + &sysctl_intvec, NULL, &xfs_params.stats_clear.min, &xfs_params.stats_clear.max}, #endif /* CONFIG_PROC_FS */ diff --git a/fs/xfs/linux-2.6/xfs_sysctl.h b/fs/xfs/linux-2.6/xfs_sysctl.h index a39a95020a58..bc8c11f13722 100644 --- a/fs/xfs/linux-2.6/xfs_sysctl.h +++ b/fs/xfs/linux-2.6/xfs_sysctl.h @@ -1,35 +1,20 @@ /* - * Copyright (c) 2001-2004 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2001-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #ifndef __XFS_SYSCTL_H__ #define __XFS_SYSCTL_H__ diff --git a/fs/xfs/linux-2.6/xfs_version.h b/fs/xfs/linux-2.6/xfs_version.h index 96f96394417e..f8d279d7563a 100644 --- a/fs/xfs/linux-2.6/xfs_version.h +++ b/fs/xfs/linux-2.6/xfs_version.h @@ -1,34 +1,22 @@ /* - * Copyright (c) 2001-2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2001-2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ +#ifndef __XFS_VERSION_H__ +#define __XFS_VERSION_H__ /* * Dummy file that can contain a timestamp to put into the @@ -36,9 +24,6 @@ * running */ -#ifndef __XFS_VERSION_H__ -#define __XFS_VERSION_H__ - #define XFS_VERSION_STRING "SGI XFS" #endif /* __XFS_VERSION_H__ */ diff --git a/fs/xfs/linux-2.6/xfs_vfs.c b/fs/xfs/linux-2.6/xfs_vfs.c index 34cc902ec119..c855d62e5344 100644 --- a/fs/xfs/linux-2.6/xfs_vfs.c +++ b/fs/xfs/linux-2.6/xfs_vfs.c @@ -1,38 +1,22 @@ /* - * Copyright (c) 2000-2004 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" #include "xfs_fs.h" -#include "xfs_macros.h" #include "xfs_inum.h" #include "xfs_log.h" #include "xfs_clnt.h" diff --git a/fs/xfs/linux-2.6/xfs_vfs.h b/fs/xfs/linux-2.6/xfs_vfs.h index f0ab574fb47a..57caf9eddee0 100644 --- a/fs/xfs/linux-2.6/xfs_vfs.h +++ b/fs/xfs/linux-2.6/xfs_vfs.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2004 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_VFS_H__ #define __XFS_VFS_H__ @@ -95,6 +81,7 @@ typedef enum { #define VFS_RDONLY 0x0001 /* read-only vfs */ #define VFS_GRPID 0x0002 /* group-ID assigned from directory */ #define VFS_DMI 0x0004 /* filesystem has the DMI enabled */ +#define VFS_32BITINODES 0x0008 /* do not use inums above 32 bits */ #define VFS_END 0x0008 /* max flag */ #define SYNC_ATTR 0x0001 /* sync attributes */ diff --git a/fs/xfs/linux-2.6/xfs_vnode.c b/fs/xfs/linux-2.6/xfs_vnode.c index 268f45bf6a9a..e9bbcb4d6243 100644 --- a/fs/xfs/linux-2.6/xfs_vnode.c +++ b/fs/xfs/linux-2.6/xfs_vnode.c @@ -1,38 +1,22 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" - uint64_t vn_generation; /* vnode generation number */ DEFINE_SPINLOCK(vnumber_lock); @@ -44,7 +28,6 @@ DEFINE_SPINLOCK(vnumber_lock); #define vptosync(v) (&vsync[((unsigned long)v) % NVSYNC]) STATIC wait_queue_head_t vsync[NVSYNC]; - void vn_init(void) { @@ -124,6 +107,7 @@ vn_revalidate_core( inode->i_mtime = vap->va_mtime; inode->i_ctime = vap->va_ctime; inode->i_atime = vap->va_atime; + inode->i_blksize = vap->va_blocksize; if (vap->va_xflags & XFS_XFLAG_IMMUTABLE) inode->i_flags |= S_IMMUTABLE; else diff --git a/fs/xfs/linux-2.6/xfs_vnode.h b/fs/xfs/linux-2.6/xfs_vnode.h index 35f306cebb87..f2bbb327c081 100644 --- a/fs/xfs/linux-2.6/xfs_vnode.h +++ b/fs/xfs/linux-2.6/xfs_vnode.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA * * Portions Copyright (c) 1989, 1993 * The Regents of the University of California. All rights reserved. @@ -216,11 +202,12 @@ typedef void (*vop_rwunlock_t)(bhv_desc_t *, vrwlock_t); typedef int (*vop_bmap_t)(bhv_desc_t *, xfs_off_t, ssize_t, int, struct xfs_iomap *, int *); typedef int (*vop_reclaim_t)(bhv_desc_t *); -typedef int (*vop_attr_get_t)(bhv_desc_t *, char *, char *, int *, int, - struct cred *); -typedef int (*vop_attr_set_t)(bhv_desc_t *, char *, char *, int, int, - struct cred *); -typedef int (*vop_attr_remove_t)(bhv_desc_t *, char *, int, struct cred *); +typedef int (*vop_attr_get_t)(bhv_desc_t *, const char *, char *, int *, + int, struct cred *); +typedef int (*vop_attr_set_t)(bhv_desc_t *, const char *, char *, int, + int, struct cred *); +typedef int (*vop_attr_remove_t)(bhv_desc_t *, const char *, + int, struct cred *); typedef int (*vop_attr_list_t)(bhv_desc_t *, char *, int, int, struct attrlist_cursor_kern *, struct cred *); typedef void (*vop_link_removed_t)(bhv_desc_t *, vnode_t *, int); @@ -566,13 +553,6 @@ static __inline__ void vn_flagclr(struct vnode *vp, uint flag) } /* - * Update modify/access/change times on the vnode - */ -#define VN_MTIMESET(vp, tvp) (LINVFS_GET_IP(vp)->i_mtime = *(tvp)) -#define VN_ATIMESET(vp, tvp) (LINVFS_GET_IP(vp)->i_atime = *(tvp)) -#define VN_CTIMESET(vp, tvp) (LINVFS_GET_IP(vp)->i_ctime = *(tvp)) - -/* * Dealing with bad inodes */ static inline void vn_mark_bad(struct vnode *vp) @@ -603,6 +583,7 @@ static inline int VN_BAD(struct vnode *vp) #define ATTR_LAZY 0x80 /* set/get attributes lazily */ #define ATTR_NONBLOCK 0x100 /* return EAGAIN if operation would block */ #define ATTR_NOLOCK 0x200 /* Don't grab any conflicting locks */ +#define ATTR_NOSIZETOK 0x400 /* Don't get the SIZE token */ /* * Flags to VOP_FSYNC and VOP_RECLAIM. diff --git a/fs/xfs/quota/Makefile b/fs/xfs/quota/Makefile deleted file mode 100644 index 7a4f725b2824..000000000000 --- a/fs/xfs/quota/Makefile +++ /dev/null @@ -1 +0,0 @@ -include $(TOPDIR)/fs/xfs/quota/Makefile-linux-$(VERSION).$(PATCHLEVEL) diff --git a/fs/xfs/quota/Makefile-linux-2.6 b/fs/xfs/quota/Makefile-linux-2.6 deleted file mode 100644 index 93e60e839355..000000000000 --- a/fs/xfs/quota/Makefile-linux-2.6 +++ /dev/null @@ -1,53 +0,0 @@ -# -# Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. -# -# This program is free software; you can redistribute it and/or modify it -# under the terms of version 2 of the GNU General Public License as -# published by the Free Software Foundation. -# -# This program is distributed in the hope that it would be useful, but -# WITHOUT ANY WARRANTY; without even the implied warranty of -# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. -# -# Further, this software is distributed without any warranty that it is -# free of the rightful claim of any third person regarding infringement -# or the like. Any license provided herein, whether implied or -# otherwise, applies only to this software file. Patent licenses, if -# any, provided herein do not apply to combinations of this program with -# other software, or any other product whatsoever. -# -# You should have received a copy of the GNU General Public License along -# with this program; if not, write the Free Software Foundation, Inc., 59 -# Temple Place - Suite 330, Boston MA 02111-1307, USA. -# -# Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, -# Mountain View, CA 94043, or: -# -# http://www.sgi.com -# -# For further information regarding this notice, see: -# -# http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ -# - -EXTRA_CFLAGS += -I $(TOPDIR)/fs/xfs -I $(TOPDIR)/fs/xfs/linux-2.6 - -ifeq ($(CONFIG_XFS_DEBUG),y) - EXTRA_CFLAGS += -g -DDEBUG - #EXTRA_CFLAGS += -DQUOTADEBUG -endif -ifeq ($(CONFIG_XFS_TRACE),y) - EXTRA_CFLAGS += -DXFS_DQUOT_TRACE - EXTRA_CFLAGS += -DXFS_VNODE_TRACE -endif - -xfs-$(CONFIG_XFS_QUOTA) += xfs_dquot.o \ - xfs_dquot_item.o \ - xfs_trans_dquot.o \ - xfs_qm_syscalls.o \ - xfs_qm_bhv.o \ - xfs_qm.o - -ifeq ($(CONFIG_XFS_QUOTA),y) -xfs-$(CONFIG_PROC_FS) += xfs_qm_stats.o -endif diff --git a/fs/xfs/quota/xfs_dquot.c b/fs/xfs/quota/xfs_dquot.c index e2e8d35fa4d0..00b5043dfa5a 100644 --- a/fs/xfs/quota/xfs_dquot.c +++ b/fs/xfs/quota/xfs_dquot.c @@ -1,39 +1,25 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" #include "xfs_fs.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" @@ -43,18 +29,17 @@ #include "xfs_dmapi.h" #include "xfs_quota.h" #include "xfs_mount.h" -#include "xfs_alloc_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_btree.h" -#include "xfs_ialloc.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" #include "xfs_inode.h" +#include "xfs_btree.h" +#include "xfs_ialloc.h" #include "xfs_bmap.h" -#include "xfs_bit.h" #include "xfs_rtalloc.h" #include "xfs_error.h" #include "xfs_itable.h" @@ -66,7 +51,6 @@ #include "xfs_buf_item.h" #include "xfs_trans_space.h" #include "xfs_trans_priv.h" - #include "xfs_qm.h" @@ -112,7 +96,7 @@ xfs_qm_dqinit( brandnewdquot = xfs_qm_dqalloc_incore(&dqp); dqp->dq_flags = type; - INT_SET(dqp->q_core.d_id, ARCH_CONVERT, id); + dqp->q_core.d_id = cpu_to_be32(id); dqp->q_mount = mp; /* @@ -194,10 +178,10 @@ xfs_qm_dqinit_core( /* * Caller has zero'd the entire dquot 'chunk' already. */ - INT_SET(d->dd_diskdq.d_magic, ARCH_CONVERT, XFS_DQUOT_MAGIC); - INT_SET(d->dd_diskdq.d_version, ARCH_CONVERT, XFS_DQUOT_VERSION); - INT_SET(d->dd_diskdq.d_id, ARCH_CONVERT, id); - INT_SET(d->dd_diskdq.d_flags, ARCH_CONVERT, type); + d->dd_diskdq.d_magic = cpu_to_be16(XFS_DQUOT_MAGIC); + d->dd_diskdq.d_version = XFS_DQUOT_VERSION; + d->dd_diskdq.d_id = cpu_to_be32(id); + d->dd_diskdq.d_flags = type; } @@ -227,19 +211,13 @@ __xfs_dqtrace_entry( (void *)(__psint_t)dqp->q_nrefs, (void *)(__psint_t)dqp->dq_flags, (void *)(__psint_t)dqp->q_res_bcount, - (void *)(__psint_t)INT_GET(dqp->q_core.d_bcount, - ARCH_CONVERT), - (void *)(__psint_t)INT_GET(dqp->q_core.d_icount, - ARCH_CONVERT), - (void *)(__psint_t)INT_GET(dqp->q_core.d_blk_hardlimit, - ARCH_CONVERT), - (void *)(__psint_t)INT_GET(dqp->q_core.d_blk_softlimit, - ARCH_CONVERT), - (void *)(__psint_t)INT_GET(dqp->q_core.d_ino_hardlimit, - ARCH_CONVERT), - (void *)(__psint_t)INT_GET(dqp->q_core.d_ino_softlimit, - ARCH_CONVERT), - (void *)(__psint_t)INT_GET(dqp->q_core.d_id, ARCH_CONVERT), + (void *)(__psint_t)be64_to_cpu(dqp->q_core.d_bcount), + (void *)(__psint_t)be64_to_cpu(dqp->q_core.d_icount), + (void *)(__psint_t)be64_to_cpu(dqp->q_core.d_blk_hardlimit), + (void *)(__psint_t)be64_to_cpu(dqp->q_core.d_blk_softlimit), + (void *)(__psint_t)be64_to_cpu(dqp->q_core.d_ino_hardlimit), + (void *)(__psint_t)be64_to_cpu(dqp->q_core.d_ino_softlimit), + (void *)(__psint_t)be32_to_cpu(dqp->q_core.d_id), (void *)(__psint_t)current_pid(), (void *)(__psint_t)ino, (void *)(__psint_t)retaddr, @@ -264,17 +242,17 @@ xfs_qm_adjust_dqlimits( ASSERT(d->d_id); if (q->qi_bsoftlimit && !d->d_blk_softlimit) - INT_SET(d->d_blk_softlimit, ARCH_CONVERT, q->qi_bsoftlimit); + d->d_blk_softlimit = cpu_to_be64(q->qi_bsoftlimit); if (q->qi_bhardlimit && !d->d_blk_hardlimit) - INT_SET(d->d_blk_hardlimit, ARCH_CONVERT, q->qi_bhardlimit); + d->d_blk_hardlimit = cpu_to_be64(q->qi_bhardlimit); if (q->qi_isoftlimit && !d->d_ino_softlimit) - INT_SET(d->d_ino_softlimit, ARCH_CONVERT, q->qi_isoftlimit); + d->d_ino_softlimit = cpu_to_be64(q->qi_isoftlimit); if (q->qi_ihardlimit && !d->d_ino_hardlimit) - INT_SET(d->d_ino_hardlimit, ARCH_CONVERT, q->qi_ihardlimit); + d->d_ino_hardlimit = cpu_to_be64(q->qi_ihardlimit); if (q->qi_rtbsoftlimit && !d->d_rtb_softlimit) - INT_SET(d->d_rtb_softlimit, ARCH_CONVERT, q->qi_rtbsoftlimit); + d->d_rtb_softlimit = cpu_to_be64(q->qi_rtbsoftlimit); if (q->qi_rtbhardlimit && !d->d_rtb_hardlimit) - INT_SET(d->d_rtb_hardlimit, ARCH_CONVERT, q->qi_rtbhardlimit); + d->d_rtb_hardlimit = cpu_to_be64(q->qi_rtbhardlimit); } /* @@ -298,81 +276,81 @@ xfs_qm_adjust_dqtimers( ASSERT(d->d_id); #ifdef QUOTADEBUG - if (INT_GET(d->d_blk_hardlimit, ARCH_CONVERT)) - ASSERT(INT_GET(d->d_blk_softlimit, ARCH_CONVERT) <= - INT_GET(d->d_blk_hardlimit, ARCH_CONVERT)); - if (INT_GET(d->d_ino_hardlimit, ARCH_CONVERT)) - ASSERT(INT_GET(d->d_ino_softlimit, ARCH_CONVERT) <= - INT_GET(d->d_ino_hardlimit, ARCH_CONVERT)); - if (INT_GET(d->d_rtb_hardlimit, ARCH_CONVERT)) - ASSERT(INT_GET(d->d_rtb_softlimit, ARCH_CONVERT) <= - INT_GET(d->d_rtb_hardlimit, ARCH_CONVERT)); + if (d->d_blk_hardlimit) + ASSERT(be64_to_cpu(d->d_blk_softlimit) <= + be64_to_cpu(d->d_blk_hardlimit)); + if (d->d_ino_hardlimit) + ASSERT(be64_to_cpu(d->d_ino_softlimit) <= + be64_to_cpu(d->d_ino_hardlimit)); + if (d->d_rtb_hardlimit) + ASSERT(be64_to_cpu(d->d_rtb_softlimit) <= + be64_to_cpu(d->d_rtb_hardlimit)); #endif if (!d->d_btimer) { - if ((INT_GET(d->d_blk_softlimit, ARCH_CONVERT) && - (INT_GET(d->d_bcount, ARCH_CONVERT) >= - INT_GET(d->d_blk_softlimit, ARCH_CONVERT))) || - (INT_GET(d->d_blk_hardlimit, ARCH_CONVERT) && - (INT_GET(d->d_bcount, ARCH_CONVERT) >= - INT_GET(d->d_blk_hardlimit, ARCH_CONVERT)))) { - INT_SET(d->d_btimer, ARCH_CONVERT, - get_seconds() + XFS_QI_BTIMELIMIT(mp)); + if ((d->d_blk_softlimit && + (be64_to_cpu(d->d_bcount) >= + be64_to_cpu(d->d_blk_softlimit))) || + (d->d_blk_hardlimit && + (be64_to_cpu(d->d_bcount) >= + be64_to_cpu(d->d_blk_hardlimit)))) { + d->d_btimer = cpu_to_be32(get_seconds() + + XFS_QI_BTIMELIMIT(mp)); } else { d->d_bwarns = 0; } } else { if ((!d->d_blk_softlimit || - (INT_GET(d->d_bcount, ARCH_CONVERT) < - INT_GET(d->d_blk_softlimit, ARCH_CONVERT))) && + (be64_to_cpu(d->d_bcount) < + be64_to_cpu(d->d_blk_softlimit))) && (!d->d_blk_hardlimit || - (INT_GET(d->d_bcount, ARCH_CONVERT) < - INT_GET(d->d_blk_hardlimit, ARCH_CONVERT)))) { + (be64_to_cpu(d->d_bcount) < + be64_to_cpu(d->d_blk_hardlimit)))) { d->d_btimer = 0; } } if (!d->d_itimer) { - if ((INT_GET(d->d_ino_softlimit, ARCH_CONVERT) && - (INT_GET(d->d_icount, ARCH_CONVERT) >= - INT_GET(d->d_ino_softlimit, ARCH_CONVERT))) || - (INT_GET(d->d_ino_hardlimit, ARCH_CONVERT) && - (INT_GET(d->d_icount, ARCH_CONVERT) >= - INT_GET(d->d_ino_hardlimit, ARCH_CONVERT)))) { - INT_SET(d->d_itimer, ARCH_CONVERT, - get_seconds() + XFS_QI_ITIMELIMIT(mp)); + if ((d->d_ino_softlimit && + (be64_to_cpu(d->d_icount) >= + be64_to_cpu(d->d_ino_softlimit))) || + (d->d_ino_hardlimit && + (be64_to_cpu(d->d_icount) >= + be64_to_cpu(d->d_ino_hardlimit)))) { + d->d_itimer = cpu_to_be32(get_seconds() + + XFS_QI_ITIMELIMIT(mp)); } else { d->d_iwarns = 0; } } else { if ((!d->d_ino_softlimit || - (INT_GET(d->d_icount, ARCH_CONVERT) < - INT_GET(d->d_ino_softlimit, ARCH_CONVERT))) && + (be64_to_cpu(d->d_icount) < + be64_to_cpu(d->d_ino_softlimit))) && (!d->d_ino_hardlimit || - (INT_GET(d->d_icount, ARCH_CONVERT) < - INT_GET(d->d_ino_hardlimit, ARCH_CONVERT)))) { + (be64_to_cpu(d->d_icount) < + be64_to_cpu(d->d_ino_hardlimit)))) { d->d_itimer = 0; } } if (!d->d_rtbtimer) { - if ((INT_GET(d->d_rtb_softlimit, ARCH_CONVERT) && - (INT_GET(d->d_rtbcount, ARCH_CONVERT) >= - INT_GET(d->d_rtb_softlimit, ARCH_CONVERT))) || - (INT_GET(d->d_rtb_hardlimit, ARCH_CONVERT) && - (INT_GET(d->d_rtbcount, ARCH_CONVERT) >= - INT_GET(d->d_rtb_hardlimit, ARCH_CONVERT)))) { - INT_SET(d->d_rtbtimer, ARCH_CONVERT, - get_seconds() + XFS_QI_RTBTIMELIMIT(mp)); + if ((d->d_rtb_softlimit && + (be64_to_cpu(d->d_rtbcount) >= + be64_to_cpu(d->d_rtb_softlimit))) || + (d->d_rtb_hardlimit && + (be64_to_cpu(d->d_rtbcount) >= + be64_to_cpu(d->d_rtb_hardlimit)))) { + d->d_rtbtimer = cpu_to_be32(get_seconds() + + XFS_QI_RTBTIMELIMIT(mp)); } else { d->d_rtbwarns = 0; } } else { if ((!d->d_rtb_softlimit || - (INT_GET(d->d_rtbcount, ARCH_CONVERT) < - INT_GET(d->d_rtb_softlimit, ARCH_CONVERT))) && + (be64_to_cpu(d->d_rtbcount) < + be64_to_cpu(d->d_rtb_softlimit))) && (!d->d_rtb_hardlimit || - (INT_GET(d->d_rtbcount, ARCH_CONVERT) < - INT_GET(d->d_rtb_hardlimit, ARCH_CONVERT)))) { + (be64_to_cpu(d->d_rtbcount) < + be64_to_cpu(d->d_rtb_hardlimit)))) { d->d_rtbtimer = 0; } } @@ -490,7 +468,7 @@ xfs_qm_dqalloc( * Make a chunk of dquots out of this buffer and log * the entire thing. */ - xfs_qm_init_dquot_blk(tp, mp, INT_GET(dqp->q_core.d_id, ARCH_CONVERT), + xfs_qm_init_dquot_blk(tp, mp, be32_to_cpu(dqp->q_core.d_id), dqp->dq_flags & XFS_DQ_ALLTYPES, bp); /* @@ -554,7 +532,7 @@ xfs_qm_dqtobp( xfs_trans_t *tp = (tpp ? *tpp : NULL); mp = dqp->q_mount; - id = INT_GET(dqp->q_core.d_id, ARCH_CONVERT); + id = be32_to_cpu(dqp->q_core.d_id); nmaps = 1; newdquot = B_FALSE; @@ -563,8 +541,7 @@ xfs_qm_dqtobp( */ if (dqp->q_blkno == (xfs_daddr_t) 0) { /* We use the id as an index */ - dqp->q_fileoffset = (xfs_fileoff_t) ((uint)id / - XFS_QM_DQPERBLK(mp)); + dqp->q_fileoffset = (xfs_fileoff_t)id / XFS_QM_DQPERBLK(mp); nmaps = 1; quotip = XFS_DQ_TO_QIP(dqp); xfs_ilock(quotip, XFS_ILOCK_SHARED); @@ -694,16 +671,16 @@ xfs_qm_dqread( /* copy everything from disk dquot to the incore dquot */ memcpy(&dqp->q_core, ddqp, sizeof(xfs_disk_dquot_t)); - ASSERT(INT_GET(dqp->q_core.d_id, ARCH_CONVERT) == id); + ASSERT(be32_to_cpu(dqp->q_core.d_id) == id); xfs_qm_dquot_logitem_init(dqp); /* * Reservation counters are defined as reservation plus current usage * to avoid having to add everytime. */ - dqp->q_res_bcount = INT_GET(ddqp->d_bcount, ARCH_CONVERT); - dqp->q_res_icount = INT_GET(ddqp->d_icount, ARCH_CONVERT); - dqp->q_res_rtbcount = INT_GET(ddqp->d_rtbcount, ARCH_CONVERT); + dqp->q_res_bcount = be64_to_cpu(ddqp->d_bcount); + dqp->q_res_icount = be64_to_cpu(ddqp->d_icount); + dqp->q_res_rtbcount = be64_to_cpu(ddqp->d_rtbcount); /* Mark the buf so that this will stay incore a little longer */ XFS_BUF_SET_VTYPE_REF(bp, B_FS_DQUOT, XFS_DQUOT_REF); @@ -829,7 +806,7 @@ xfs_qm_dqlookup( * dqlock to look at the id field of the dquot, since the * id can't be modified without the hashlock anyway. */ - if (INT_GET(dqp->q_core.d_id, ARCH_CONVERT) == id && dqp->q_mount == mp) { + if (be32_to_cpu(dqp->q_core.d_id) == id && dqp->q_mount == mp) { xfs_dqtrace_entry(dqp, "DQFOUND BY LOOKUP"); /* * All in core dquots must be on the dqlist of mp @@ -860,7 +837,7 @@ xfs_qm_dqlookup( * id couldn't have changed; we had the hashlock all * along */ - ASSERT(INT_GET(dqp->q_core.d_id, ARCH_CONVERT) == id); + ASSERT(be32_to_cpu(dqp->q_core.d_id) == id); if (flist_locked) { if (dqp->q_nrefs != 0) { @@ -1282,7 +1259,7 @@ xfs_qm_dqflush( return (error); } - if (xfs_qm_dqcheck(&dqp->q_core, INT_GET(ddqp->d_id, ARCH_CONVERT), + if (xfs_qm_dqcheck(&dqp->q_core, be32_to_cpu(ddqp->d_id), 0, XFS_QMOPT_DOWARN, "dqflush (incore copy)")) { xfs_force_shutdown(dqp->q_mount, XFS_CORRUPT_INCORE); return XFS_ERROR(EIO); @@ -1435,8 +1412,8 @@ xfs_dqlock2( { if (d1 && d2) { ASSERT(d1 != d2); - if (INT_GET(d1->q_core.d_id, ARCH_CONVERT) > - INT_GET(d2->q_core.d_id, ARCH_CONVERT)) { + if (be32_to_cpu(d1->q_core.d_id) > + be32_to_cpu(d2->q_core.d_id)) { xfs_dqlock(d2); xfs_dqlock(d1); } else { @@ -1558,33 +1535,33 @@ xfs_qm_dqprint(xfs_dquot_t *dqp) { cmn_err(CE_DEBUG, "-----------KERNEL DQUOT----------------"); cmn_err(CE_DEBUG, "---- dquotID = %d", - (int)INT_GET(dqp->q_core.d_id, ARCH_CONVERT)); + (int)be32_to_cpu(dqp->q_core.d_id)); cmn_err(CE_DEBUG, "---- type = %s", DQFLAGTO_TYPESTR(dqp)); cmn_err(CE_DEBUG, "---- fs = 0x%p", dqp->q_mount); cmn_err(CE_DEBUG, "---- blkno = 0x%x", (int) dqp->q_blkno); cmn_err(CE_DEBUG, "---- boffset = 0x%x", (int) dqp->q_bufoffset); cmn_err(CE_DEBUG, "---- blkhlimit = %Lu (0x%x)", - INT_GET(dqp->q_core.d_blk_hardlimit, ARCH_CONVERT), - (int) INT_GET(dqp->q_core.d_blk_hardlimit, ARCH_CONVERT)); + be64_to_cpu(dqp->q_core.d_blk_hardlimit), + (int)be64_to_cpu(dqp->q_core.d_blk_hardlimit)); cmn_err(CE_DEBUG, "---- blkslimit = %Lu (0x%x)", - INT_GET(dqp->q_core.d_blk_softlimit, ARCH_CONVERT), - (int)INT_GET(dqp->q_core.d_blk_softlimit, ARCH_CONVERT)); + be64_to_cpu(dqp->q_core.d_blk_softlimit), + (int)be64_to_cpu(dqp->q_core.d_blk_softlimit)); cmn_err(CE_DEBUG, "---- inohlimit = %Lu (0x%x)", - INT_GET(dqp->q_core.d_ino_hardlimit, ARCH_CONVERT), - (int)INT_GET(dqp->q_core.d_ino_hardlimit, ARCH_CONVERT)); + be64_to_cpu(dqp->q_core.d_ino_hardlimit), + (int)be64_to_cpu(dqp->q_core.d_ino_hardlimit)); cmn_err(CE_DEBUG, "---- inoslimit = %Lu (0x%x)", - INT_GET(dqp->q_core.d_ino_softlimit, ARCH_CONVERT), - (int)INT_GET(dqp->q_core.d_ino_softlimit, ARCH_CONVERT)); + be64_to_cpu(dqp->q_core.d_ino_softlimit), + (int)be64_to_cpu(dqp->q_core.d_ino_softlimit)); cmn_err(CE_DEBUG, "---- bcount = %Lu (0x%x)", - INT_GET(dqp->q_core.d_bcount, ARCH_CONVERT), - (int)INT_GET(dqp->q_core.d_bcount, ARCH_CONVERT)); + be64_to_cpu(dqp->q_core.d_bcount), + (int)be64_to_cpu(dqp->q_core.d_bcount)); cmn_err(CE_DEBUG, "---- icount = %Lu (0x%x)", - INT_GET(dqp->q_core.d_icount, ARCH_CONVERT), - (int)INT_GET(dqp->q_core.d_icount, ARCH_CONVERT)); + be64_to_cpu(dqp->q_core.d_icount), + (int)be64_to_cpu(dqp->q_core.d_icount)); cmn_err(CE_DEBUG, "---- btimer = %d", - (int)INT_GET(dqp->q_core.d_btimer, ARCH_CONVERT)); + (int)be32_to_cpu(dqp->q_core.d_btimer)); cmn_err(CE_DEBUG, "---- itimer = %d", - (int)INT_GET(dqp->q_core.d_itimer, ARCH_CONVERT)); + (int)be32_to_cpu(dqp->q_core.d_itimer)); cmn_err(CE_DEBUG, "---------------------------"); } #endif diff --git a/fs/xfs/quota/xfs_dquot.h b/fs/xfs/quota/xfs_dquot.h index 8ebc87176c78..c0c629663a5c 100644 --- a/fs/xfs/quota/xfs_dquot.h +++ b/fs/xfs/quota/xfs_dquot.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_DQUOT_H__ #define __XFS_DQUOT_H__ diff --git a/fs/xfs/quota/xfs_dquot_item.c b/fs/xfs/quota/xfs_dquot_item.c index e74eaa7dd1bc..2f69822344e5 100644 --- a/fs/xfs/quota/xfs_dquot_item.c +++ b/fs/xfs/quota/xfs_dquot_item.c @@ -1,39 +1,25 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" #include "xfs_fs.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" @@ -43,18 +29,17 @@ #include "xfs_dmapi.h" #include "xfs_quota.h" #include "xfs_mount.h" -#include "xfs_alloc_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_btree.h" -#include "xfs_ialloc.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" #include "xfs_inode.h" #include "xfs_bmap.h" -#include "xfs_bit.h" +#include "xfs_btree.h" +#include "xfs_ialloc.h" #include "xfs_rtalloc.h" #include "xfs_error.h" #include "xfs_itable.h" @@ -65,10 +50,8 @@ #include "xfs_attr.h" #include "xfs_buf_item.h" #include "xfs_trans_priv.h" - #include "xfs_qm.h" - /* * returns the number of iovecs needed to log the given dquot item. */ @@ -467,7 +450,7 @@ xfs_qm_dquot_logitem_init( lp->qli_item.li_mountp = dqp->q_mount; lp->qli_dquot = dqp; lp->qli_format.qlf_type = XFS_LI_DQUOT; - lp->qli_format.qlf_id = INT_GET(dqp->q_core.d_id, ARCH_CONVERT); + lp->qli_format.qlf_id = be32_to_cpu(dqp->q_core.d_id); lp->qli_format.qlf_blkno = dqp->q_blkno; lp->qli_format.qlf_len = 1; /* diff --git a/fs/xfs/quota/xfs_dquot_item.h b/fs/xfs/quota/xfs_dquot_item.h index 9c6500dabcaa..5a632531f843 100644 --- a/fs/xfs/quota/xfs_dquot_item.h +++ b/fs/xfs/quota/xfs_dquot_item.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_DQUOT_ITEM_H__ #define __XFS_DQUOT_ITEM_H__ diff --git a/fs/xfs/quota/xfs_qm.c b/fs/xfs/quota/xfs_qm.c index efde16e0a913..1aea42d71a64 100644 --- a/fs/xfs/quota/xfs_qm.c +++ b/fs/xfs/quota/xfs_qm.c @@ -1,39 +1,25 @@ /* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" #include "xfs_fs.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_clnt.h" #include "xfs_trans.h" #include "xfs_sb.h" @@ -44,21 +30,20 @@ #include "xfs_dmapi.h" #include "xfs_quota.h" #include "xfs_mount.h" -#include "xfs_alloc_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_btree.h" -#include "xfs_ialloc.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" #include "xfs_inode.h" -#include "xfs_bmap.h" -#include "xfs_bit.h" +#include "xfs_btree.h" +#include "xfs_ialloc.h" +#include "xfs_itable.h" #include "xfs_rtalloc.h" #include "xfs_error.h" -#include "xfs_itable.h" +#include "xfs_bmap.h" #include "xfs_rw.h" #include "xfs_acl.h" #include "xfs_cap.h" @@ -67,7 +52,6 @@ #include "xfs_buf_item.h" #include "xfs_trans_space.h" #include "xfs_utils.h" - #include "xfs_qm.h" /* @@ -76,8 +60,9 @@ * quota functionality, including maintaining the freelist and hash * tables of dquots. */ -mutex_t xfs_Gqm_lock; +mutex_t xfs_Gqm_lock; struct xfs_qm *xfs_Gqm; +uint ndquot; kmem_zone_t *qm_dqzone; kmem_zone_t *qm_dqtrxzone; @@ -107,10 +92,10 @@ extern mutex_t qcheck_lock; for (dqp = (l)->qh_next; dqp != NULL; dqp = dqp->NXT) { \ cmn_err(CE_DEBUG, " %d. \"%d (%s)\" " \ "bcnt = %d, icnt = %d, refs = %d", \ - ++i, (int) INT_GET(dqp->q_core.d_id, ARCH_CONVERT), \ + ++i, (int) be32_to_cpu(dqp->q_core.d_id), \ DQFLAGTO_TYPESTR(dqp), \ - (int) INT_GET(dqp->q_core.d_bcount, ARCH_CONVERT), \ - (int) INT_GET(dqp->q_core.d_icount, ARCH_CONVERT), \ + (int) be64_to_cpu(dqp->q_core.d_bcount), \ + (int) be64_to_cpu(dqp->q_core.d_icount), \ (int) dqp->q_nrefs); } \ } #else @@ -124,25 +109,25 @@ extern mutex_t qcheck_lock; STATIC struct xfs_qm * xfs_Gqm_init(void) { - xfs_qm_t *xqm; - int hsize, i; - - xqm = kmem_zalloc(sizeof(xfs_qm_t), KM_SLEEP); - ASSERT(xqm); + xfs_dqhash_t *udqhash, *gdqhash; + xfs_qm_t *xqm; + uint i, hsize, flags = KM_SLEEP | KM_MAYFAIL; /* * Initialize the dquot hash tables. */ - hsize = (DQUOT_HASH_HEURISTIC < XFS_QM_NCSIZE_THRESHOLD) ? - XFS_QM_HASHSIZE_LOW : XFS_QM_HASHSIZE_HIGH; - xqm->qm_dqhashmask = hsize - 1; + hsize = XFS_QM_HASHSIZE_HIGH; + while (!(udqhash = kmem_zalloc(hsize * sizeof(xfs_dqhash_t), flags))) { + if ((hsize >>= 1) <= XFS_QM_HASHSIZE_LOW) + flags = KM_SLEEP; + } + gdqhash = kmem_zalloc(hsize * sizeof(xfs_dqhash_t), KM_SLEEP); + ndquot = hsize << 8; - xqm->qm_usr_dqhtable = (xfs_dqhash_t *)kmem_zalloc(hsize * - sizeof(xfs_dqhash_t), - KM_SLEEP); - xqm->qm_grp_dqhtable = (xfs_dqhash_t *)kmem_zalloc(hsize * - sizeof(xfs_dqhash_t), - KM_SLEEP); + xqm = kmem_zalloc(sizeof(xfs_qm_t), KM_SLEEP); + xqm->qm_dqhashmask = hsize - 1; + xqm->qm_usr_dqhtable = udqhash; + xqm->qm_grp_dqhtable = gdqhash; ASSERT(xqm->qm_usr_dqhtable != NULL); ASSERT(xqm->qm_grp_dqhtable != NULL); @@ -743,7 +728,7 @@ xfs_qm_dqattach_one( */ if (udqhint && (dqp = udqhint->q_gdquot) && - (INT_GET(dqp->q_core.d_id, ARCH_CONVERT) == id)) { + (be32_to_cpu(dqp->q_core.d_id) == id)) { ASSERT(XFS_DQ_IS_LOCKED(udqhint)); xfs_dqlock(dqp); XFS_DQHOLD(dqp); @@ -1213,42 +1198,24 @@ xfs_qm_init_quotainfo( * a user or group before he or she can not perform any * more writing. If it is zero, a default is used. */ - qinf->qi_btimelimit = - INT_GET(ddqp->d_btimer, ARCH_CONVERT) ? - INT_GET(ddqp->d_btimer, ARCH_CONVERT) : - XFS_QM_BTIMELIMIT; - qinf->qi_itimelimit = - INT_GET(ddqp->d_itimer, ARCH_CONVERT) ? - INT_GET(ddqp->d_itimer, ARCH_CONVERT) : - XFS_QM_ITIMELIMIT; - qinf->qi_rtbtimelimit = - INT_GET(ddqp->d_rtbtimer, ARCH_CONVERT) ? - INT_GET(ddqp->d_rtbtimer, ARCH_CONVERT) : - XFS_QM_RTBTIMELIMIT; - qinf->qi_bwarnlimit = - INT_GET(ddqp->d_bwarns, ARCH_CONVERT) ? - INT_GET(ddqp->d_bwarns, ARCH_CONVERT) : - XFS_QM_BWARNLIMIT; - qinf->qi_iwarnlimit = - INT_GET(ddqp->d_iwarns, ARCH_CONVERT) ? - INT_GET(ddqp->d_iwarns, ARCH_CONVERT) : - XFS_QM_IWARNLIMIT; - qinf->qi_rtbwarnlimit = - INT_GET(ddqp->d_rtbwarns, ARCH_CONVERT) ? - INT_GET(ddqp->d_rtbwarns, ARCH_CONVERT) : - XFS_QM_RTBWARNLIMIT; - qinf->qi_bhardlimit = - INT_GET(ddqp->d_blk_hardlimit, ARCH_CONVERT); - qinf->qi_bsoftlimit = - INT_GET(ddqp->d_blk_softlimit, ARCH_CONVERT); - qinf->qi_ihardlimit = - INT_GET(ddqp->d_ino_hardlimit, ARCH_CONVERT); - qinf->qi_isoftlimit = - INT_GET(ddqp->d_ino_softlimit, ARCH_CONVERT); - qinf->qi_rtbhardlimit = - INT_GET(ddqp->d_rtb_hardlimit, ARCH_CONVERT); - qinf->qi_rtbsoftlimit = - INT_GET(ddqp->d_rtb_softlimit, ARCH_CONVERT); + qinf->qi_btimelimit = ddqp->d_btimer ? + be32_to_cpu(ddqp->d_btimer) : XFS_QM_BTIMELIMIT; + qinf->qi_itimelimit = ddqp->d_itimer ? + be32_to_cpu(ddqp->d_itimer) : XFS_QM_ITIMELIMIT; + qinf->qi_rtbtimelimit = ddqp->d_rtbtimer ? + be32_to_cpu(ddqp->d_rtbtimer) : XFS_QM_RTBTIMELIMIT; + qinf->qi_bwarnlimit = ddqp->d_bwarns ? + be16_to_cpu(ddqp->d_bwarns) : XFS_QM_BWARNLIMIT; + qinf->qi_iwarnlimit = ddqp->d_iwarns ? + be16_to_cpu(ddqp->d_iwarns) : XFS_QM_IWARNLIMIT; + qinf->qi_rtbwarnlimit = ddqp->d_rtbwarns ? + be16_to_cpu(ddqp->d_rtbwarns) : XFS_QM_RTBWARNLIMIT; + qinf->qi_bhardlimit = be64_to_cpu(ddqp->d_blk_hardlimit); + qinf->qi_bsoftlimit = be64_to_cpu(ddqp->d_blk_softlimit); + qinf->qi_ihardlimit = be64_to_cpu(ddqp->d_ino_hardlimit); + qinf->qi_isoftlimit = be64_to_cpu(ddqp->d_ino_softlimit); + qinf->qi_rtbhardlimit = be64_to_cpu(ddqp->d_rtb_hardlimit); + qinf->qi_rtbsoftlimit = be64_to_cpu(ddqp->d_rtb_softlimit); /* * We sent the XFS_QMOPT_DQSUSER flag to dqget because @@ -1527,15 +1494,15 @@ xfs_qm_reset_dqcounts( */ (void) xfs_qm_dqcheck(ddq, id+j, type, XFS_QMOPT_DQREPAIR, "xfs_quotacheck"); - INT_SET(ddq->d_bcount, ARCH_CONVERT, 0ULL); - INT_SET(ddq->d_icount, ARCH_CONVERT, 0ULL); - INT_SET(ddq->d_rtbcount, ARCH_CONVERT, 0ULL); - INT_SET(ddq->d_btimer, ARCH_CONVERT, (time_t)0); - INT_SET(ddq->d_itimer, ARCH_CONVERT, (time_t)0); - INT_SET(ddq->d_rtbtimer, ARCH_CONVERT, (time_t)0); - INT_SET(ddq->d_bwarns, ARCH_CONVERT, 0UL); - INT_SET(ddq->d_iwarns, ARCH_CONVERT, 0UL); - INT_SET(ddq->d_rtbwarns, ARCH_CONVERT, 0UL); + ddq->d_bcount = 0; + ddq->d_icount = 0; + ddq->d_rtbcount = 0; + ddq->d_btimer = 0; + ddq->d_itimer = 0; + ddq->d_rtbtimer = 0; + ddq->d_bwarns = 0; + ddq->d_iwarns = 0; + ddq->d_rtbwarns = 0; ddq = (xfs_disk_dquot_t *) ((xfs_dqblk_t *)ddq + 1); } @@ -1708,14 +1675,14 @@ xfs_qm_quotacheck_dqadjust( * Adjust the inode count and the block count to reflect this inode's * resource usage. */ - INT_MOD(dqp->q_core.d_icount, ARCH_CONVERT, +1); + be64_add(&dqp->q_core.d_icount, 1); dqp->q_res_icount++; if (nblks) { - INT_MOD(dqp->q_core.d_bcount, ARCH_CONVERT, nblks); + be64_add(&dqp->q_core.d_bcount, nblks); dqp->q_res_bcount += nblks; } if (rtblks) { - INT_MOD(dqp->q_core.d_rtbcount, ARCH_CONVERT, rtblks); + be64_add(&dqp->q_core.d_rtbcount, rtblks); dqp->q_res_rtbcount += rtblks; } @@ -2202,7 +2169,7 @@ xfs_qm_shake_freelist( xfs_dqtrace_entry(dqp, "DQSHAKE: UNLINKING"); #ifdef QUOTADEBUG cmn_err(CE_DEBUG, "Shake 0x%p, ID 0x%x\n", - dqp, INT_GET(dqp->q_core.d_id, ARCH_CONVERT)); + dqp, be32_to_cpu(dqp->q_core.d_id)); #endif ASSERT(dqp->q_nrefs == 0); nextdqp = dqp->dq_flnext; @@ -2670,7 +2637,7 @@ xfs_qm_vop_chown_reserve( XFS_QMOPT_RES_RTBLKS : XFS_QMOPT_RES_REGBLKS; if (XFS_IS_UQUOTA_ON(mp) && udqp && - ip->i_d.di_uid != (uid_t)INT_GET(udqp->q_core.d_id, ARCH_CONVERT)) { + ip->i_d.di_uid != (uid_t)be32_to_cpu(udqp->q_core.d_id)) { delblksudq = udqp; /* * If there are delayed allocation blocks, then we have to @@ -2683,10 +2650,10 @@ xfs_qm_vop_chown_reserve( } } if (XFS_IS_OQUOTA_ON(ip->i_mount) && gdqp) { - if ((XFS_IS_GQUOTA_ON(ip->i_mount) && ip->i_d.di_gid != - INT_GET(gdqp->q_core.d_id, ARCH_CONVERT)) || - (XFS_IS_PQUOTA_ON(ip->i_mount) && ip->i_d.di_projid != - INT_GET(gdqp->q_core.d_id, ARCH_CONVERT))) { + if ((XFS_IS_GQUOTA_ON(ip->i_mount) && + ip->i_d.di_gid != be32_to_cpu(gdqp->q_core.d_id)) || + (XFS_IS_PQUOTA_ON(ip->i_mount) && + ip->i_d.di_projid != be32_to_cpu(gdqp->q_core.d_id))) { delblksgdq = gdqp; if (delblks) { ASSERT(ip->i_gdquot); @@ -2776,7 +2743,7 @@ xfs_qm_vop_dqattach_and_dqmod_newinode( xfs_dqunlock(udqp); ASSERT(ip->i_udquot == NULL); ip->i_udquot = udqp; - ASSERT(ip->i_d.di_uid == INT_GET(udqp->q_core.d_id, ARCH_CONVERT)); + ASSERT(ip->i_d.di_uid == be32_to_cpu(udqp->q_core.d_id)); xfs_trans_mod_dquot(tp, udqp, XFS_TRANS_DQ_ICOUNT, 1); } if (gdqp) { @@ -2785,7 +2752,7 @@ xfs_qm_vop_dqattach_and_dqmod_newinode( xfs_dqunlock(gdqp); ASSERT(ip->i_gdquot == NULL); ip->i_gdquot = gdqp; - ASSERT(ip->i_d.di_gid == INT_GET(gdqp->q_core.d_id, ARCH_CONVERT)); + ASSERT(ip->i_d.di_gid == be32_to_cpu(gdqp->q_core.d_id)); xfs_trans_mod_dquot(tp, gdqp, XFS_TRANS_DQ_ICOUNT, 1); } } diff --git a/fs/xfs/quota/xfs_qm.h b/fs/xfs/quota/xfs_qm.h index 0b00b3c67015..12da259f2fcb 100644 --- a/fs/xfs/quota/xfs_qm.h +++ b/fs/xfs/quota/xfs_qm.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_QM_H__ #define __XFS_QM_H__ @@ -40,6 +26,7 @@ struct xfs_qm; struct xfs_inode; +extern uint ndquot; extern mutex_t xfs_Gqm_lock; extern struct xfs_qm *xfs_Gqm; extern kmem_zone_t *qm_dqzone; @@ -65,9 +52,8 @@ extern kmem_zone_t *qm_dqtrxzone; /* * Dquot hashtable constants/threshold values. */ -#define XFS_QM_NCSIZE_THRESHOLD 5000 -#define XFS_QM_HASHSIZE_LOW 32 -#define XFS_QM_HASHSIZE_HIGH 64 +#define XFS_QM_HASHSIZE_LOW (NBPP / sizeof(xfs_dqhash_t)) +#define XFS_QM_HASHSIZE_HIGH ((NBPP * 4) / sizeof(xfs_dqhash_t)) /* * We output a cmn_err when quotachecking a quota file with more than diff --git a/fs/xfs/quota/xfs_qm_bhv.c b/fs/xfs/quota/xfs_qm_bhv.c index 8890a18a99d8..d9d2993de435 100644 --- a/fs/xfs/quota/xfs_qm_bhv.c +++ b/fs/xfs/quota/xfs_qm_bhv.c @@ -1,70 +1,55 @@ /* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" #include "xfs_fs.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_clnt.h" #include "xfs_trans.h" #include "xfs_sb.h" +#include "xfs_ag.h" #include "xfs_dir.h" #include "xfs_dir2.h" #include "xfs_alloc.h" #include "xfs_dmapi.h" #include "xfs_quota.h" #include "xfs_mount.h" -#include "xfs_alloc_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_btree.h" -#include "xfs_ialloc.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" #include "xfs_inode.h" +#include "xfs_ialloc.h" +#include "xfs_itable.h" +#include "xfs_btree.h" #include "xfs_bmap.h" -#include "xfs_bit.h" #include "xfs_rtalloc.h" #include "xfs_error.h" -#include "xfs_itable.h" #include "xfs_rw.h" #include "xfs_acl.h" #include "xfs_cap.h" #include "xfs_mac.h" #include "xfs_attr.h" #include "xfs_buf_item.h" - #include "xfs_qm.h" #define MNTOPT_QUOTA "quota" /* disk quotas (user) */ diff --git a/fs/xfs/quota/xfs_qm_stats.c b/fs/xfs/quota/xfs_qm_stats.c index 29978e037fee..0570f7733550 100644 --- a/fs/xfs/quota/xfs_qm_stats.c +++ b/fs/xfs/quota/xfs_qm_stats.c @@ -1,69 +1,54 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" #include "xfs_fs.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" +#include "xfs_ag.h" #include "xfs_dir.h" #include "xfs_dir2.h" #include "xfs_alloc.h" #include "xfs_dmapi.h" #include "xfs_quota.h" #include "xfs_mount.h" -#include "xfs_alloc_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_btree.h" -#include "xfs_ialloc.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" #include "xfs_inode.h" +#include "xfs_ialloc.h" +#include "xfs_itable.h" #include "xfs_bmap.h" -#include "xfs_bit.h" +#include "xfs_btree.h" #include "xfs_rtalloc.h" #include "xfs_error.h" -#include "xfs_itable.h" #include "xfs_rw.h" #include "xfs_acl.h" #include "xfs_cap.h" #include "xfs_mac.h" #include "xfs_attr.h" #include "xfs_buf_item.h" - #include "xfs_qm.h" struct xqmstats xqmstats; diff --git a/fs/xfs/quota/xfs_qm_stats.h b/fs/xfs/quota/xfs_qm_stats.h index 8093c5c284ec..a50ffabcf554 100644 --- a/fs/xfs/quota/xfs_qm_stats.h +++ b/fs/xfs/quota/xfs_qm_stats.h @@ -1,38 +1,23 @@ /* - * Copyright (c) 2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2002 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_QM_STATS_H__ #define __XFS_QM_STATS_H__ - #if defined(CONFIG_PROC_FS) && !defined(XFS_STATS_OFF) /* diff --git a/fs/xfs/quota/xfs_qm_syscalls.c b/fs/xfs/quota/xfs_qm_syscalls.c index 15e02e8a9d4f..24690e1af659 100644 --- a/fs/xfs/quota/xfs_qm_syscalls.c +++ b/fs/xfs/quota/xfs_qm_syscalls.c @@ -1,62 +1,48 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" #include "xfs_fs.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" +#include "xfs_ag.h" #include "xfs_dir.h" #include "xfs_dir2.h" #include "xfs_alloc.h" #include "xfs_dmapi.h" #include "xfs_quota.h" #include "xfs_mount.h" -#include "xfs_alloc_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_btree.h" -#include "xfs_ialloc.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" #include "xfs_inode.h" +#include "xfs_ialloc.h" +#include "xfs_itable.h" #include "xfs_bmap.h" -#include "xfs_bit.h" +#include "xfs_btree.h" #include "xfs_rtalloc.h" #include "xfs_error.h" -#include "xfs_itable.h" #include "xfs_rw.h" #include "xfs_acl.h" #include "xfs_cap.h" @@ -64,7 +50,6 @@ #include "xfs_attr.h" #include "xfs_buf_item.h" #include "xfs_utils.h" - #include "xfs_qm.h" #ifdef DEBUG @@ -109,10 +94,7 @@ xfs_qm_quotactl( vfsp = bhvtovfs(bdp); mp = XFS_VFSTOM(vfsp); - if (addr == NULL && cmd != Q_SYNC) - return XFS_ERROR(EINVAL); - if (id < 0 && cmd != Q_SYNC) - return XFS_ERROR(EINVAL); + ASSERT(addr != NULL || cmd == Q_XQUOTASYNC); /* * The following commands are valid even when quotaoff. @@ -122,7 +104,7 @@ xfs_qm_quotactl( /* * Truncate quota files. quota must be off. */ - if (XFS_IS_QUOTA_ON(mp) || addr == NULL) + if (XFS_IS_QUOTA_ON(mp)) return XFS_ERROR(EINVAL); if (vfsp->vfs_flag & VFS_RDONLY) return XFS_ERROR(EROFS); @@ -140,8 +122,6 @@ xfs_qm_quotactl( * QUOTAON - enabling quota enforcement. * Quota accounting must be turned on at mount time. */ - if (addr == NULL) - return XFS_ERROR(EINVAL); if (vfsp->vfs_flag & VFS_RDONLY) return XFS_ERROR(EROFS); return (xfs_qm_scall_quotaon(mp, @@ -152,6 +132,9 @@ xfs_qm_quotactl( return XFS_ERROR(EROFS); break; + case Q_XQUOTASYNC: + return (xfs_sync_inodes(mp, SYNC_DELWRI, 0, NULL)); + default: break; } @@ -655,13 +638,13 @@ xfs_qm_scall_setqlim( */ hard = (newlim->d_fieldmask & FS_DQ_BHARD) ? (xfs_qcnt_t) XFS_BB_TO_FSB(mp, newlim->d_blk_hardlimit) : - INT_GET(ddq->d_blk_hardlimit, ARCH_CONVERT); + be64_to_cpu(ddq->d_blk_hardlimit); soft = (newlim->d_fieldmask & FS_DQ_BSOFT) ? (xfs_qcnt_t) XFS_BB_TO_FSB(mp, newlim->d_blk_softlimit) : - INT_GET(ddq->d_blk_softlimit, ARCH_CONVERT); + be64_to_cpu(ddq->d_blk_softlimit); if (hard == 0 || hard >= soft) { - INT_SET(ddq->d_blk_hardlimit, ARCH_CONVERT, hard); - INT_SET(ddq->d_blk_softlimit, ARCH_CONVERT, soft); + ddq->d_blk_hardlimit = cpu_to_be64(hard); + ddq->d_blk_softlimit = cpu_to_be64(soft); if (id == 0) { mp->m_quotainfo->qi_bhardlimit = hard; mp->m_quotainfo->qi_bsoftlimit = soft; @@ -671,13 +654,13 @@ xfs_qm_scall_setqlim( } hard = (newlim->d_fieldmask & FS_DQ_RTBHARD) ? (xfs_qcnt_t) XFS_BB_TO_FSB(mp, newlim->d_rtb_hardlimit) : - INT_GET(ddq->d_rtb_hardlimit, ARCH_CONVERT); + be64_to_cpu(ddq->d_rtb_hardlimit); soft = (newlim->d_fieldmask & FS_DQ_RTBSOFT) ? (xfs_qcnt_t) XFS_BB_TO_FSB(mp, newlim->d_rtb_softlimit) : - INT_GET(ddq->d_rtb_softlimit, ARCH_CONVERT); + be64_to_cpu(ddq->d_rtb_softlimit); if (hard == 0 || hard >= soft) { - INT_SET(ddq->d_rtb_hardlimit, ARCH_CONVERT, hard); - INT_SET(ddq->d_rtb_softlimit, ARCH_CONVERT, soft); + ddq->d_rtb_hardlimit = cpu_to_be64(hard); + ddq->d_rtb_softlimit = cpu_to_be64(soft); if (id == 0) { mp->m_quotainfo->qi_rtbhardlimit = hard; mp->m_quotainfo->qi_rtbsoftlimit = soft; @@ -688,13 +671,13 @@ xfs_qm_scall_setqlim( hard = (newlim->d_fieldmask & FS_DQ_IHARD) ? (xfs_qcnt_t) newlim->d_ino_hardlimit : - INT_GET(ddq->d_ino_hardlimit, ARCH_CONVERT); + be64_to_cpu(ddq->d_ino_hardlimit); soft = (newlim->d_fieldmask & FS_DQ_ISOFT) ? (xfs_qcnt_t) newlim->d_ino_softlimit : - INT_GET(ddq->d_ino_softlimit, ARCH_CONVERT); + be64_to_cpu(ddq->d_ino_softlimit); if (hard == 0 || hard >= soft) { - INT_SET(ddq->d_ino_hardlimit, ARCH_CONVERT, hard); - INT_SET(ddq->d_ino_softlimit, ARCH_CONVERT, soft); + ddq->d_ino_hardlimit = cpu_to_be64(hard); + ddq->d_ino_softlimit = cpu_to_be64(soft); if (id == 0) { mp->m_quotainfo->qi_ihardlimit = hard; mp->m_quotainfo->qi_isoftlimit = soft; @@ -707,11 +690,11 @@ xfs_qm_scall_setqlim( * Update warnings counter(s) if requested */ if (newlim->d_fieldmask & FS_DQ_BWARNS) - INT_SET(ddq->d_bwarns, ARCH_CONVERT, newlim->d_bwarns); + ddq->d_bwarns = cpu_to_be16(newlim->d_bwarns); if (newlim->d_fieldmask & FS_DQ_IWARNS) - INT_SET(ddq->d_iwarns, ARCH_CONVERT, newlim->d_iwarns); + ddq->d_iwarns = cpu_to_be16(newlim->d_iwarns); if (newlim->d_fieldmask & FS_DQ_RTBWARNS) - INT_SET(ddq->d_rtbwarns, ARCH_CONVERT, newlim->d_rtbwarns); + ddq->d_rtbwarns = cpu_to_be16(newlim->d_rtbwarns); if (id == 0) { /* @@ -723,15 +706,15 @@ xfs_qm_scall_setqlim( */ if (newlim->d_fieldmask & FS_DQ_BTIMER) { mp->m_quotainfo->qi_btimelimit = newlim->d_btimer; - INT_SET(ddq->d_btimer, ARCH_CONVERT, newlim->d_btimer); + ddq->d_btimer = cpu_to_be32(newlim->d_btimer); } if (newlim->d_fieldmask & FS_DQ_ITIMER) { mp->m_quotainfo->qi_itimelimit = newlim->d_itimer; - INT_SET(ddq->d_itimer, ARCH_CONVERT, newlim->d_itimer); + ddq->d_itimer = cpu_to_be32(newlim->d_itimer); } if (newlim->d_fieldmask & FS_DQ_RTBTIMER) { mp->m_quotainfo->qi_rtbtimelimit = newlim->d_rtbtimer; - INT_SET(ddq->d_rtbtimer, ARCH_CONVERT, newlim->d_rtbtimer); + ddq->d_rtbtimer = cpu_to_be32(newlim->d_rtbtimer); } if (newlim->d_fieldmask & FS_DQ_BWARNS) mp->m_quotainfo->qi_bwarnlimit = newlim->d_bwarns; @@ -902,33 +885,27 @@ xfs_qm_export_dquot( { memset(dst, 0, sizeof(*dst)); dst->d_version = FS_DQUOT_VERSION; /* different from src->d_version */ - dst->d_flags = - xfs_qm_export_qtype_flags(INT_GET(src->d_flags, ARCH_CONVERT)); - dst->d_id = INT_GET(src->d_id, ARCH_CONVERT); - dst->d_blk_hardlimit = (__uint64_t) - XFS_FSB_TO_BB(mp, INT_GET(src->d_blk_hardlimit, ARCH_CONVERT)); - dst->d_blk_softlimit = (__uint64_t) - XFS_FSB_TO_BB(mp, INT_GET(src->d_blk_softlimit, ARCH_CONVERT)); - dst->d_ino_hardlimit = (__uint64_t) - INT_GET(src->d_ino_hardlimit, ARCH_CONVERT); - dst->d_ino_softlimit = (__uint64_t) - INT_GET(src->d_ino_softlimit, ARCH_CONVERT); - dst->d_bcount = (__uint64_t) - XFS_FSB_TO_BB(mp, INT_GET(src->d_bcount, ARCH_CONVERT)); - dst->d_icount = (__uint64_t) INT_GET(src->d_icount, ARCH_CONVERT); - dst->d_btimer = (__uint32_t) INT_GET(src->d_btimer, ARCH_CONVERT); - dst->d_itimer = (__uint32_t) INT_GET(src->d_itimer, ARCH_CONVERT); - dst->d_iwarns = INT_GET(src->d_iwarns, ARCH_CONVERT); - dst->d_bwarns = INT_GET(src->d_bwarns, ARCH_CONVERT); - - dst->d_rtb_hardlimit = (__uint64_t) - XFS_FSB_TO_BB(mp, INT_GET(src->d_rtb_hardlimit, ARCH_CONVERT)); - dst->d_rtb_softlimit = (__uint64_t) - XFS_FSB_TO_BB(mp, INT_GET(src->d_rtb_softlimit, ARCH_CONVERT)); - dst->d_rtbcount = (__uint64_t) - XFS_FSB_TO_BB(mp, INT_GET(src->d_rtbcount, ARCH_CONVERT)); - dst->d_rtbtimer = (__uint32_t) INT_GET(src->d_rtbtimer, ARCH_CONVERT); - dst->d_rtbwarns = INT_GET(src->d_rtbwarns, ARCH_CONVERT); + dst->d_flags = xfs_qm_export_qtype_flags(src->d_flags); + dst->d_id = be32_to_cpu(src->d_id); + dst->d_blk_hardlimit = + XFS_FSB_TO_BB(mp, be64_to_cpu(src->d_blk_hardlimit)); + dst->d_blk_softlimit = + XFS_FSB_TO_BB(mp, be64_to_cpu(src->d_blk_softlimit)); + dst->d_ino_hardlimit = be64_to_cpu(src->d_ino_hardlimit); + dst->d_ino_softlimit = be64_to_cpu(src->d_ino_softlimit); + dst->d_bcount = XFS_FSB_TO_BB(mp, be64_to_cpu(src->d_bcount)); + dst->d_icount = be64_to_cpu(src->d_icount); + dst->d_btimer = be32_to_cpu(src->d_btimer); + dst->d_itimer = be32_to_cpu(src->d_itimer); + dst->d_iwarns = be16_to_cpu(src->d_iwarns); + dst->d_bwarns = be16_to_cpu(src->d_bwarns); + dst->d_rtb_hardlimit = + XFS_FSB_TO_BB(mp, be64_to_cpu(src->d_rtb_hardlimit)); + dst->d_rtb_softlimit = + XFS_FSB_TO_BB(mp, be64_to_cpu(src->d_rtb_softlimit)); + dst->d_rtbcount = XFS_FSB_TO_BB(mp, be64_to_cpu(src->d_rtbcount)); + dst->d_rtbtimer = be32_to_cpu(src->d_rtbtimer); + dst->d_rtbwarns = be16_to_cpu(src->d_rtbwarns); /* * Internally, we don't reset all the timers when quota enforcement @@ -1222,10 +1199,10 @@ xfs_qm_dqtest_failed( qmtest_nfails++; if (error) cmn_err(CE_DEBUG, "quotacheck failed id=%d, err=%d\nreason: %s", - INT_GET(d->d_id, ARCH_CONVERT), error, reason); + d->d_id, error, reason); else cmn_err(CE_DEBUG, "quotacheck failed id=%d (%s) [%d != %d]", - INT_GET(d->d_id, ARCH_CONVERT), reason, (int)a, (int)b); + d->d_id, reason, (int)a, (int)b); xfs_qm_dqtest_print(d); if (dqp) xfs_qm_dqprint(dqp); @@ -1237,21 +1214,21 @@ xfs_dqtest_cmp2( xfs_dquot_t *dqp) { int err = 0; - if (INT_GET(dqp->q_core.d_icount, ARCH_CONVERT) != d->d_icount) { + if (be64_to_cpu(dqp->q_core.d_icount) != d->d_icount) { xfs_qm_dqtest_failed(d, dqp, "icount mismatch", - INT_GET(dqp->q_core.d_icount, ARCH_CONVERT), + be64_to_cpu(dqp->q_core.d_icount), d->d_icount, 0); err++; } - if (INT_GET(dqp->q_core.d_bcount, ARCH_CONVERT) != d->d_bcount) { + if (be64_to_cpu(dqp->q_core.d_bcount) != d->d_bcount) { xfs_qm_dqtest_failed(d, dqp, "bcount mismatch", - INT_GET(dqp->q_core.d_bcount, ARCH_CONVERT), + be64_to_cpu(dqp->q_core.d_bcount), d->d_bcount, 0); err++; } - if (INT_GET(dqp->q_core.d_blk_softlimit, ARCH_CONVERT) && - INT_GET(dqp->q_core.d_bcount, ARCH_CONVERT) >= - INT_GET(dqp->q_core.d_blk_softlimit, ARCH_CONVERT)) { + if (dqp->q_core.d_blk_softlimit && + be64_to_cpu(dqp->q_core.d_bcount) >= + be64_to_cpu(dqp->q_core.d_blk_softlimit)) { if (!dqp->q_core.d_btimer && dqp->q_core.d_id) { cmn_err(CE_DEBUG, "%d [%s] [0x%p] BLK TIMER NOT STARTED", @@ -1259,9 +1236,9 @@ xfs_dqtest_cmp2( err++; } } - if (INT_GET(dqp->q_core.d_ino_softlimit, ARCH_CONVERT) && - INT_GET(dqp->q_core.d_icount, ARCH_CONVERT) >= - INT_GET(dqp->q_core.d_ino_softlimit, ARCH_CONVERT)) { + if (dqp->q_core.d_ino_softlimit && + be64_to_cpu(dqp->q_core.d_icount) >= + be64_to_cpu(dqp->q_core.d_ino_softlimit)) { if (!dqp->q_core.d_itimer && dqp->q_core.d_id) { cmn_err(CE_DEBUG, "%d [%s] [0x%p] INO TIMER NOT STARTED", diff --git a/fs/xfs/quota/xfs_quota_priv.h b/fs/xfs/quota/xfs_quota_priv.h index bf413e70ec07..7a9f3beb818c 100644 --- a/fs/xfs/quota/xfs_quota_priv.h +++ b/fs/xfs/quota/xfs_quota_priv.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_QUOTA_PRIV_H__ #define __XFS_QUOTA_PRIV_H__ diff --git a/fs/xfs/quota/xfs_trans_dquot.c b/fs/xfs/quota/xfs_trans_dquot.c index 3b99daf8a640..3290975d31f7 100644 --- a/fs/xfs/quota/xfs_trans_dquot.c +++ b/fs/xfs/quota/xfs_trans_dquot.c @@ -1,39 +1,25 @@ /* - * Copyright (c) 2000-2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2002 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" #include "xfs_fs.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" @@ -43,21 +29,20 @@ #include "xfs_dmapi.h" #include "xfs_quota.h" #include "xfs_mount.h" -#include "xfs_alloc_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_btree.h" -#include "xfs_ialloc.h" #include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" #include "xfs_dinode.h" #include "xfs_inode.h" +#include "xfs_ialloc.h" +#include "xfs_itable.h" +#include "xfs_btree.h" #include "xfs_bmap.h" -#include "xfs_bit.h" #include "xfs_rtalloc.h" #include "xfs_error.h" -#include "xfs_itable.h" #include "xfs_rw.h" #include "xfs_acl.h" #include "xfs_cap.h" @@ -65,7 +50,6 @@ #include "xfs_attr.h" #include "xfs_buf_item.h" #include "xfs_trans_priv.h" - #include "xfs_qm.h" STATIC void xfs_trans_alloc_dqinfo(xfs_trans_t *); @@ -429,25 +413,25 @@ xfs_trans_apply_dquot_deltas( qtrx->qt_delrtb_delta; #ifdef QUOTADEBUG if (totalbdelta < 0) - ASSERT(INT_GET(d->d_bcount, ARCH_CONVERT) >= + ASSERT(be64_to_cpu(d->d_bcount) >= (xfs_qcnt_t) -totalbdelta); if (totalrtbdelta < 0) - ASSERT(INT_GET(d->d_rtbcount, ARCH_CONVERT) >= + ASSERT(be64_to_cpu(d->d_rtbcount) >= (xfs_qcnt_t) -totalrtbdelta); if (qtrx->qt_icount_delta < 0) - ASSERT(INT_GET(d->d_icount, ARCH_CONVERT) >= + ASSERT(be64_to_cpu(d->d_icount) >= (xfs_qcnt_t) -qtrx->qt_icount_delta); #endif if (totalbdelta) - INT_MOD(d->d_bcount, ARCH_CONVERT, (xfs_qcnt_t)totalbdelta); + be64_add(&d->d_bcount, (xfs_qcnt_t)totalbdelta); if (qtrx->qt_icount_delta) - INT_MOD(d->d_icount, ARCH_CONVERT, (xfs_qcnt_t)qtrx->qt_icount_delta); + be64_add(&d->d_icount, (xfs_qcnt_t)qtrx->qt_icount_delta); if (totalrtbdelta) - INT_MOD(d->d_rtbcount, ARCH_CONVERT, (xfs_qcnt_t)totalrtbdelta); + be64_add(&d->d_rtbcount, (xfs_qcnt_t)totalrtbdelta); /* * Get any default limits in use. @@ -531,11 +515,11 @@ xfs_trans_apply_dquot_deltas( } ASSERT(dqp->q_res_bcount >= - INT_GET(dqp->q_core.d_bcount, ARCH_CONVERT)); + be64_to_cpu(dqp->q_core.d_bcount)); ASSERT(dqp->q_res_icount >= - INT_GET(dqp->q_core.d_icount, ARCH_CONVERT)); + be64_to_cpu(dqp->q_core.d_icount)); ASSERT(dqp->q_res_rtbcount >= - INT_GET(dqp->q_core.d_rtbcount, ARCH_CONVERT)); + be64_to_cpu(dqp->q_core.d_rtbcount)); } /* * Do the group quotas next @@ -642,26 +626,26 @@ xfs_trans_dqresv( } ASSERT(XFS_DQ_IS_LOCKED(dqp)); if (flags & XFS_TRANS_DQ_RES_BLKS) { - hardlimit = INT_GET(dqp->q_core.d_blk_hardlimit, ARCH_CONVERT); + hardlimit = be64_to_cpu(dqp->q_core.d_blk_hardlimit); if (!hardlimit) hardlimit = q->qi_bhardlimit; - softlimit = INT_GET(dqp->q_core.d_blk_softlimit, ARCH_CONVERT); + softlimit = be64_to_cpu(dqp->q_core.d_blk_softlimit); if (!softlimit) softlimit = q->qi_bsoftlimit; - timer = INT_GET(dqp->q_core.d_btimer, ARCH_CONVERT); - warns = INT_GET(dqp->q_core.d_bwarns, ARCH_CONVERT); + timer = be32_to_cpu(dqp->q_core.d_btimer); + warns = be16_to_cpu(dqp->q_core.d_bwarns); warnlimit = XFS_QI_BWARNLIMIT(dqp->q_mount); resbcountp = &dqp->q_res_bcount; } else { ASSERT(flags & XFS_TRANS_DQ_RES_RTBLKS); - hardlimit = INT_GET(dqp->q_core.d_rtb_hardlimit, ARCH_CONVERT); + hardlimit = be64_to_cpu(dqp->q_core.d_rtb_hardlimit); if (!hardlimit) hardlimit = q->qi_rtbhardlimit; - softlimit = INT_GET(dqp->q_core.d_rtb_softlimit, ARCH_CONVERT); + softlimit = be64_to_cpu(dqp->q_core.d_rtb_softlimit); if (!softlimit) softlimit = q->qi_rtbsoftlimit; - timer = INT_GET(dqp->q_core.d_rtbtimer, ARCH_CONVERT); - warns = INT_GET(dqp->q_core.d_rtbwarns, ARCH_CONVERT); + timer = be32_to_cpu(dqp->q_core.d_rtbtimer); + warns = be16_to_cpu(dqp->q_core.d_rtbwarns); warnlimit = XFS_QI_RTBWARNLIMIT(dqp->q_mount); resbcountp = &dqp->q_res_rtbcount; } @@ -700,16 +684,14 @@ xfs_trans_dqresv( } } if (ninos > 0) { - count = INT_GET(dqp->q_core.d_icount, ARCH_CONVERT); - timer = INT_GET(dqp->q_core.d_itimer, ARCH_CONVERT); - warns = INT_GET(dqp->q_core.d_iwarns, ARCH_CONVERT); + count = be64_to_cpu(dqp->q_core.d_icount); + timer = be32_to_cpu(dqp->q_core.d_itimer); + warns = be16_to_cpu(dqp->q_core.d_iwarns); warnlimit = XFS_QI_IWARNLIMIT(dqp->q_mount); - hardlimit = INT_GET(dqp->q_core.d_ino_hardlimit, - ARCH_CONVERT); + hardlimit = be64_to_cpu(dqp->q_core.d_ino_hardlimit); if (!hardlimit) hardlimit = q->qi_ihardlimit; - softlimit = INT_GET(dqp->q_core.d_ino_softlimit, - ARCH_CONVERT); + softlimit = be64_to_cpu(dqp->q_core.d_ino_softlimit); if (!softlimit) softlimit = q->qi_isoftlimit; if (hardlimit > 0ULL && count >= hardlimit) { @@ -756,9 +738,9 @@ xfs_trans_dqresv( XFS_TRANS_DQ_RES_INOS, ninos); } - ASSERT(dqp->q_res_bcount >= INT_GET(dqp->q_core.d_bcount, ARCH_CONVERT)); - ASSERT(dqp->q_res_rtbcount >= INT_GET(dqp->q_core.d_rtbcount, ARCH_CONVERT)); - ASSERT(dqp->q_res_icount >= INT_GET(dqp->q_core.d_icount, ARCH_CONVERT)); + ASSERT(dqp->q_res_bcount >= be64_to_cpu(dqp->q_core.d_bcount)); + ASSERT(dqp->q_res_rtbcount >= be64_to_cpu(dqp->q_core.d_rtbcount)); + ASSERT(dqp->q_res_icount >= be64_to_cpu(dqp->q_core.d_icount)); error_return: if (! (flags & XFS_QMOPT_DQLOCK)) { diff --git a/fs/xfs/support/debug.c b/fs/xfs/support/debug.c index 4e1a5ec22fa3..bb6dc91ea261 100644 --- a/fs/xfs/support/debug.c +++ b/fs/xfs/support/debug.c @@ -1,38 +1,22 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "debug.h" #include "spin.h" - #include <asm/page.h> #include <linux/sched.h> #include <linux/kernel.h> diff --git a/fs/xfs/support/debug.h b/fs/xfs/support/debug.h index c5b9365a7e2a..aff558664c32 100644 --- a/fs/xfs/support/debug.h +++ b/fs/xfs/support/debug.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2004 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_SUPPORT_DEBUG_H__ #define __XFS_SUPPORT_DEBUG_H__ @@ -41,9 +27,10 @@ #define CE_ALERT 1 /* alert */ #define CE_PANIC 0 /* panic */ -extern void icmn_err(int, char *, va_list); -/* PRINTFLIKE2 */ -extern void cmn_err(int, char *, ...); +extern void icmn_err(int, char *, va_list) + __attribute__ ((format (printf, 2, 0))); +extern void cmn_err(int, char *, ...) + __attribute__ ((format (printf, 2, 3))); #ifndef STATIC # define STATIC static diff --git a/fs/xfs/support/ktrace.c b/fs/xfs/support/ktrace.c index fa8394f9437d..841aa4c15b8a 100644 --- a/fs/xfs/support/ktrace.c +++ b/fs/xfs/support/ktrace.c @@ -1,35 +1,20 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include <xfs.h> static kmem_zone_t *ktrace_hdr_zone; @@ -65,7 +50,7 @@ ktrace_uninit(void) * number of entries. */ ktrace_t * -ktrace_alloc(int nentries, int sleep) +ktrace_alloc(int nentries, unsigned int __nocast sleep) { ktrace_t *ktp; ktrace_entry_t *ktep; diff --git a/fs/xfs/support/ktrace.h b/fs/xfs/support/ktrace.h index 92d1a1a5d04b..0d73216287c0 100644 --- a/fs/xfs/support/ktrace.h +++ b/fs/xfs/support/ktrace.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_SUPPORT_KTRACE_H__ #define __XFS_SUPPORT_KTRACE_H__ @@ -66,7 +52,7 @@ typedef struct ktrace_snap { extern void ktrace_init(int zentries); extern void ktrace_uninit(void); -extern ktrace_t *ktrace_alloc(int, int); +extern ktrace_t *ktrace_alloc(int, unsigned int __nocast); extern void ktrace_free(ktrace_t *); extern void ktrace_enter( diff --git a/fs/xfs/support/move.c b/fs/xfs/support/move.c index 15b5194f16b2..caefa17b80fe 100644 --- a/fs/xfs/support/move.c +++ b/fs/xfs/support/move.c @@ -1,35 +1,20 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include <xfs.h> /* Read from kernel buffer at src to user/kernel buffer defined diff --git a/fs/xfs/support/move.h b/fs/xfs/support/move.h index 3d406dc1c89e..97a2498d2da3 100644 --- a/fs/xfs/support/move.h +++ b/fs/xfs/support/move.h @@ -1,34 +1,20 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ - * * Portions Copyright (c) 1982, 1986, 1993, 1994 * The Regents of the University of California. All rights reserved. * diff --git a/fs/xfs/support/qsort.c b/fs/xfs/support/qsort.c deleted file mode 100644 index 1ec824140cf7..000000000000 --- a/fs/xfs/support/qsort.c +++ /dev/null @@ -1,155 +0,0 @@ -/* - * Copyright (c) 1992, 1993 - * The Regents of the University of California. All rights reserved. - * - * Redistribution and use in source and binary forms, with or without - * modification, are permitted provided that the following conditions - * are met: - * 1. Redistributions of source code must retain the above copyright - * notice, this list of conditions and the following disclaimer. - * 2. Redistributions in binary form must reproduce the above copyright - * notice, this list of conditions and the following disclaimer in the - * documentation and/or other materials provided with the distribution. - * 3. Neither the name of the University nor the names of its contributors - * may be used to endorse or promote products derived from this software - * without specific prior written permission. - * - * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND - * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE - * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE - * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE - * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL - * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS - * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) - * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT - * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY - * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF - * SUCH DAMAGE. - */ - -#include <linux/kernel.h> -#include <linux/string.h> - -/* - * Qsort routine from Bentley & McIlroy's "Engineering a Sort Function". - */ -#define swapcode(TYPE, parmi, parmj, n) { \ - long i = (n) / sizeof (TYPE); \ - register TYPE *pi = (TYPE *) (parmi); \ - register TYPE *pj = (TYPE *) (parmj); \ - do { \ - register TYPE t = *pi; \ - *pi++ = *pj; \ - *pj++ = t; \ - } while (--i > 0); \ -} - -#define SWAPINIT(a, es) swaptype = ((char *)a - (char *)0) % sizeof(long) || \ - es % sizeof(long) ? 2 : es == sizeof(long)? 0 : 1; - -static __inline void -swapfunc(char *a, char *b, int n, int swaptype) -{ - if (swaptype <= 1) - swapcode(long, a, b, n) - else - swapcode(char, a, b, n) -} - -#define swap(a, b) \ - if (swaptype == 0) { \ - long t = *(long *)(a); \ - *(long *)(a) = *(long *)(b); \ - *(long *)(b) = t; \ - } else \ - swapfunc(a, b, es, swaptype) - -#define vecswap(a, b, n) if ((n) > 0) swapfunc(a, b, n, swaptype) - -static __inline char * -med3(char *a, char *b, char *c, int (*cmp)(const void *, const void *)) -{ - return cmp(a, b) < 0 ? - (cmp(b, c) < 0 ? b : (cmp(a, c) < 0 ? c : a )) - :(cmp(b, c) > 0 ? b : (cmp(a, c) < 0 ? a : c )); -} - -void -qsort(void *aa, size_t n, size_t es, int (*cmp)(const void *, const void *)) -{ - char *pa, *pb, *pc, *pd, *pl, *pm, *pn; - int d, r, swaptype, swap_cnt; - register char *a = aa; - -loop: SWAPINIT(a, es); - swap_cnt = 0; - if (n < 7) { - for (pm = (char *)a + es; pm < (char *) a + n * es; pm += es) - for (pl = pm; pl > (char *) a && cmp(pl - es, pl) > 0; - pl -= es) - swap(pl, pl - es); - return; - } - pm = (char *)a + (n / 2) * es; - if (n > 7) { - pl = (char *)a; - pn = (char *)a + (n - 1) * es; - if (n > 40) { - d = (n / 8) * es; - pl = med3(pl, pl + d, pl + 2 * d, cmp); - pm = med3(pm - d, pm, pm + d, cmp); - pn = med3(pn - 2 * d, pn - d, pn, cmp); - } - pm = med3(pl, pm, pn, cmp); - } - swap(a, pm); - pa = pb = (char *)a + es; - - pc = pd = (char *)a + (n - 1) * es; - for (;;) { - while (pb <= pc && (r = cmp(pb, a)) <= 0) { - if (r == 0) { - swap_cnt = 1; - swap(pa, pb); - pa += es; - } - pb += es; - } - while (pb <= pc && (r = cmp(pc, a)) >= 0) { - if (r == 0) { - swap_cnt = 1; - swap(pc, pd); - pd -= es; - } - pc -= es; - } - if (pb > pc) - break; - swap(pb, pc); - swap_cnt = 1; - pb += es; - pc -= es; - } - if (swap_cnt == 0) { /* Switch to insertion sort */ - for (pm = (char *) a + es; pm < (char *) a + n * es; pm += es) - for (pl = pm; pl > (char *) a && cmp(pl - es, pl) > 0; - pl -= es) - swap(pl, pl - es); - return; - } - - pn = (char *)a + n * es; - r = min(pa - (char *)a, pb - pa); - vecswap(a, pb - r, r); - r = min((long)(pd - pc), (long)(pn - pd - es)); - vecswap(pb, pn - r, r); - if ((r = pb - pa) > es) - qsort(a, r / es, es, cmp); - if ((r = pd - pc) > es) { - /* Iterate rather than recurse to save stack space */ - a = pn - r; - n = r / es; - goto loop; - } -/* qsort(pn - r, r / es, es, cmp);*/ -} diff --git a/fs/xfs/support/qsort.h b/fs/xfs/support/qsort.h deleted file mode 100644 index 94263106d716..000000000000 --- a/fs/xfs/support/qsort.h +++ /dev/null @@ -1,41 +0,0 @@ -/* - * Copyright (c) 2000-2002 Silicon Graphics, Inc. All Rights Reserved. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. - * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ - */ - -#ifndef QSORT_H -#define QSORT_H - -extern void qsort (void *const pbase, - size_t total_elems, - size_t size, - int (*cmp)(const void *, const void *)); - -#endif diff --git a/fs/xfs/support/uuid.c b/fs/xfs/support/uuid.c index 81f40cfcb267..70ce40914c8a 100644 --- a/fs/xfs/support/uuid.c +++ b/fs/xfs/support/uuid.c @@ -1,35 +1,20 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include <xfs.h> static mutex_t uuid_monitor; diff --git a/fs/xfs/support/uuid.h b/fs/xfs/support/uuid.h index 5220ea58ba2b..b6f5922199ba 100644 --- a/fs/xfs/support/uuid.h +++ b/fs/xfs/support/uuid.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_SUPPORT_UUID_H__ #define __XFS_SUPPORT_UUID_H__ @@ -36,13 +22,13 @@ typedef struct { unsigned char __u_bits[16]; } uuid_t; -void uuid_init(void); -void uuid_create_nil(uuid_t *uuid); -int uuid_is_nil(uuid_t *uuid); -int uuid_equal(uuid_t *uuid1, uuid_t *uuid2); -void uuid_getnodeuniq(uuid_t *uuid, int fsid [2]); -__uint64_t uuid_hash64(uuid_t *uuid); -int uuid_table_insert(uuid_t *uuid); -void uuid_table_remove(uuid_t *uuid); +extern void uuid_init(void); +extern void uuid_create_nil(uuid_t *uuid); +extern int uuid_is_nil(uuid_t *uuid); +extern int uuid_equal(uuid_t *uuid1, uuid_t *uuid2); +extern void uuid_getnodeuniq(uuid_t *uuid, int fsid [2]); +extern __uint64_t uuid_hash64(uuid_t *uuid); +extern int uuid_table_insert(uuid_t *uuid); +extern void uuid_table_remove(uuid_t *uuid); #endif /* __XFS_SUPPORT_UUID_H__ */ diff --git a/fs/xfs/xfs.h b/fs/xfs/xfs.h index 7e276dcaf4dc..1a48dbb902a7 100644 --- a/fs/xfs/xfs.h +++ b/fs/xfs/xfs.h @@ -1,40 +1,21 @@ /* - * Copyright (c) 2000-2004 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_H__ #define __XFS_H__ - #include <linux-2.6/xfs_linux.h> - -#include <xfs_fs.h> -#include <xfs_macros.h> - #endif /* __XFS_H__ */ diff --git a/fs/xfs/xfs_acl.c b/fs/xfs/xfs_acl.c index 92fd1d67f878..cc9c91b9e771 100644 --- a/fs/xfs/xfs_acl.c +++ b/fs/xfs/xfs_acl.c @@ -1,49 +1,37 @@ /* - * Copyright (c) 2001-2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2001-2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" - +#include "xfs_fs.h" +#include "xfs_types.h" +#include "xfs_bit.h" #include "xfs_inum.h" +#include "xfs_ag.h" #include "xfs_dir.h" #include "xfs_dir2.h" -#include "xfs_alloc_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_btree.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" #include "xfs_inode.h" +#include "xfs_btree.h" #include "xfs_acl.h" #include "xfs_mac.h" #include "xfs_attr.h" @@ -155,7 +143,7 @@ posix_acl_xattr_to_xfs( } /* - * Comparison function called from qsort(). + * Comparison function called from xfs_sort(). * Primary key is ae_tag, secondary key is ae_id. */ STATIC int @@ -189,8 +177,8 @@ posix_acl_xfs_to_xattr( return -ERANGE; /* Need to sort src XFS ACL by <ae_tag,ae_id> */ - qsort(src->acl_entry, src->acl_cnt, sizeof(src->acl_entry[0]), - xfs_acl_entry_compare); + xfs_sort(src->acl_entry, src->acl_cnt, sizeof(src->acl_entry[0]), + xfs_acl_entry_compare); dest->a_version = cpu_to_le32(POSIX_ACL_XATTR_VERSION); dest_entry = &dest->a_entries[0]; @@ -448,6 +436,7 @@ xfs_acl_access( int seen_userobj = 0; matched.ae_tag = 0; /* Invalid type */ + matched.ae_perm = 0; md >>= 6; /* Normalize the bits for comparison */ for (i = 0; i < fap->acl_cnt; i++) { diff --git a/fs/xfs/xfs_acl.h b/fs/xfs/xfs_acl.h index 0363eb46d357..f9315bc960cb 100644 --- a/fs/xfs/xfs_acl.h +++ b/fs/xfs/xfs_acl.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2001-2004 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2001-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_ACL_H__ #define __XFS_ACL_H__ diff --git a/fs/xfs/xfs_ag.h b/fs/xfs/xfs_ag.h index 96b70f7fba39..a96e2ffce0cc 100644 --- a/fs/xfs/xfs_ag.h +++ b/fs/xfs/xfs_ag.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_AG_H__ #define __XFS_AG_H__ @@ -46,18 +32,9 @@ struct xfs_trans; #define XFS_AGI_MAGIC 0x58414749 /* 'XAGI' */ #define XFS_AGF_VERSION 1 #define XFS_AGI_VERSION 1 -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_AGF_GOOD_VERSION) -int xfs_agf_good_version(unsigned v); -#define XFS_AGF_GOOD_VERSION(v) xfs_agf_good_version(v) -#else -#define XFS_AGF_GOOD_VERSION(v) ((v) == XFS_AGF_VERSION) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_AGI_GOOD_VERSION) -int xfs_agi_good_version(unsigned v); -#define XFS_AGI_GOOD_VERSION(v) xfs_agi_good_version(v) -#else -#define XFS_AGI_GOOD_VERSION(v) ((v) == XFS_AGI_VERSION) -#endif + +#define XFS_AGF_GOOD_VERSION(v) ((v) == XFS_AGF_VERSION) +#define XFS_AGI_GOOD_VERSION(v) ((v) == XFS_AGI_VERSION) /* * Btree number 0 is bno, 1 is cnt. This value gives the size of the @@ -71,27 +48,26 @@ int xfs_agi_good_version(unsigned v); * are > 64k, our value cannot be confused for an EFS superblock's. */ -typedef struct xfs_agf -{ +typedef struct xfs_agf { /* * Common allocation group header information */ - __uint32_t agf_magicnum; /* magic number == XFS_AGF_MAGIC */ - __uint32_t agf_versionnum; /* header version == XFS_AGF_VERSION */ - xfs_agnumber_t agf_seqno; /* sequence # starting from 0 */ - xfs_agblock_t agf_length; /* size in blocks of a.g. */ + __be32 agf_magicnum; /* magic number == XFS_AGF_MAGIC */ + __be32 agf_versionnum; /* header version == XFS_AGF_VERSION */ + __be32 agf_seqno; /* sequence # starting from 0 */ + __be32 agf_length; /* size in blocks of a.g. */ /* * Freespace information */ - xfs_agblock_t agf_roots[XFS_BTNUM_AGF]; /* root blocks */ - __uint32_t agf_spare0; /* spare field */ - __uint32_t agf_levels[XFS_BTNUM_AGF]; /* btree levels */ - __uint32_t agf_spare1; /* spare field */ - __uint32_t agf_flfirst; /* first freelist block's index */ - __uint32_t agf_fllast; /* last freelist block's index */ - __uint32_t agf_flcount; /* count of blocks in freelist */ - xfs_extlen_t agf_freeblks; /* total free blocks */ - xfs_extlen_t agf_longest; /* longest free space */ + __be32 agf_roots[XFS_BTNUM_AGF]; /* root blocks */ + __be32 agf_spare0; /* spare field */ + __be32 agf_levels[XFS_BTNUM_AGF]; /* btree levels */ + __be32 agf_spare1; /* spare field */ + __be32 agf_flfirst; /* first freelist block's index */ + __be32 agf_fllast; /* last freelist block's index */ + __be32 agf_flcount; /* count of blocks in freelist */ + __be32 agf_freeblks; /* total free blocks */ + __be32 agf_longest; /* longest free space */ } xfs_agf_t; #define XFS_AGF_MAGICNUM 0x00000001 @@ -110,43 +86,39 @@ typedef struct xfs_agf /* disk block (xfs_daddr_t) in the AG */ #define XFS_AGF_DADDR(mp) ((xfs_daddr_t)(1 << (mp)->m_sectbb_log)) -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_AGF_BLOCK) -xfs_agblock_t xfs_agf_block(struct xfs_mount *mp); -#define XFS_AGF_BLOCK(mp) xfs_agf_block(mp) -#else -#define XFS_AGF_BLOCK(mp) XFS_HDR_BLOCK(mp, XFS_AGF_DADDR(mp)) -#endif +#define XFS_AGF_BLOCK(mp) XFS_HDR_BLOCK(mp, XFS_AGF_DADDR(mp)) +#define XFS_BUF_TO_AGF(bp) ((xfs_agf_t *)XFS_BUF_PTR(bp)) + /* * Size of the unlinked inode hash table in the agi. */ #define XFS_AGI_UNLINKED_BUCKETS 64 -typedef struct xfs_agi -{ +typedef struct xfs_agi { /* * Common allocation group header information */ - __uint32_t agi_magicnum; /* magic number == XFS_AGI_MAGIC */ - __uint32_t agi_versionnum; /* header version == XFS_AGI_VERSION */ - xfs_agnumber_t agi_seqno; /* sequence # starting from 0 */ - xfs_agblock_t agi_length; /* size in blocks of a.g. */ + __be32 agi_magicnum; /* magic number == XFS_AGI_MAGIC */ + __be32 agi_versionnum; /* header version == XFS_AGI_VERSION */ + __be32 agi_seqno; /* sequence # starting from 0 */ + __be32 agi_length; /* size in blocks of a.g. */ /* * Inode information * Inodes are mapped by interpreting the inode number, so no * mapping data is needed here. */ - xfs_agino_t agi_count; /* count of allocated inodes */ - xfs_agblock_t agi_root; /* root of inode btree */ - __uint32_t agi_level; /* levels in inode btree */ - xfs_agino_t agi_freecount; /* number of free inodes */ - xfs_agino_t agi_newino; /* new inode just allocated */ - xfs_agino_t agi_dirino; /* last directory inode chunk */ + __be32 agi_count; /* count of allocated inodes */ + __be32 agi_root; /* root of inode btree */ + __be32 agi_level; /* levels in inode btree */ + __be32 agi_freecount; /* number of free inodes */ + __be32 agi_newino; /* new inode just allocated */ + __be32 agi_dirino; /* last directory inode chunk */ /* * Hash table of inodes which have been unlinked but are * still being referenced. */ - xfs_agino_t agi_unlinked[XFS_AGI_UNLINKED_BUCKETS]; + __be32 agi_unlinked[XFS_AGI_UNLINKED_BUCKETS]; } xfs_agi_t; #define XFS_AGI_MAGICNUM 0x00000001 @@ -165,25 +137,17 @@ typedef struct xfs_agi /* disk block (xfs_daddr_t) in the AG */ #define XFS_AGI_DADDR(mp) ((xfs_daddr_t)(2 << (mp)->m_sectbb_log)) -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_AGI_BLOCK) -xfs_agblock_t xfs_agi_block(struct xfs_mount *mp); -#define XFS_AGI_BLOCK(mp) xfs_agi_block(mp) -#else -#define XFS_AGI_BLOCK(mp) XFS_HDR_BLOCK(mp, XFS_AGI_DADDR(mp)) -#endif +#define XFS_AGI_BLOCK(mp) XFS_HDR_BLOCK(mp, XFS_AGI_DADDR(mp)) +#define XFS_BUF_TO_AGI(bp) ((xfs_agi_t *)XFS_BUF_PTR(bp)) /* * The third a.g. block contains the a.g. freelist, an array * of block pointers to blocks owned by the allocation btree code. */ #define XFS_AGFL_DADDR(mp) ((xfs_daddr_t)(3 << (mp)->m_sectbb_log)) -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_AGFL_BLOCK) -xfs_agblock_t xfs_agfl_block(struct xfs_mount *mp); -#define XFS_AGFL_BLOCK(mp) xfs_agfl_block(mp) -#else -#define XFS_AGFL_BLOCK(mp) XFS_HDR_BLOCK(mp, XFS_AGFL_DADDR(mp)) -#endif +#define XFS_AGFL_BLOCK(mp) XFS_HDR_BLOCK(mp, XFS_AGFL_DADDR(mp)) #define XFS_AGFL_SIZE(mp) ((mp)->m_sb.sb_sectsize / sizeof(xfs_agblock_t)) +#define XFS_BUF_TO_AGFL(bp) ((xfs_agfl_t *)XFS_BUF_PTR(bp)) typedef struct xfs_agfl { xfs_agblock_t agfl_bno[1]; /* actually XFS_AGFL_SIZE(mp) */ @@ -230,116 +194,38 @@ typedef struct xfs_perag xfs_perag_busy_t *pagb_list; /* unstable blocks */ } xfs_perag_t; -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_AG_MAXLEVELS) -int xfs_ag_maxlevels(struct xfs_mount *mp); -#define XFS_AG_MAXLEVELS(mp) xfs_ag_maxlevels(mp) -#else -#define XFS_AG_MAXLEVELS(mp) ((mp)->m_ag_maxlevels) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_MIN_FREELIST) -int xfs_min_freelist(xfs_agf_t *a, struct xfs_mount *mp); -#define XFS_MIN_FREELIST(a,mp) xfs_min_freelist(a,mp) -#else -#define XFS_MIN_FREELIST(a,mp) \ - XFS_MIN_FREELIST_RAW( \ - INT_GET((a)->agf_levels[XFS_BTNUM_BNOi], ARCH_CONVERT), \ - INT_GET((a)->agf_levels[XFS_BTNUM_CNTi], ARCH_CONVERT), mp) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_MIN_FREELIST_PAG) -int xfs_min_freelist_pag(xfs_perag_t *pag, struct xfs_mount *mp); -#define XFS_MIN_FREELIST_PAG(pag,mp) xfs_min_freelist_pag(pag,mp) -#else -#define XFS_MIN_FREELIST_PAG(pag,mp) \ - XFS_MIN_FREELIST_RAW((uint_t)(pag)->pagf_levels[XFS_BTNUM_BNOi], \ - (uint_t)(pag)->pagf_levels[XFS_BTNUM_CNTi], mp) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_MIN_FREELIST_RAW) -int xfs_min_freelist_raw(int bl, int cl, struct xfs_mount *mp); -#define XFS_MIN_FREELIST_RAW(bl,cl,mp) xfs_min_freelist_raw(bl,cl,mp) -#else +#define XFS_AG_MAXLEVELS(mp) ((mp)->m_ag_maxlevels) #define XFS_MIN_FREELIST_RAW(bl,cl,mp) \ - (MIN(bl + 1, XFS_AG_MAXLEVELS(mp)) + \ - MIN(cl + 1, XFS_AG_MAXLEVELS(mp))) -#endif + (MIN(bl + 1, XFS_AG_MAXLEVELS(mp)) + MIN(cl + 1, XFS_AG_MAXLEVELS(mp))) +#define XFS_MIN_FREELIST(a,mp) \ + (XFS_MIN_FREELIST_RAW( \ + be32_to_cpu((a)->agf_levels[XFS_BTNUM_BNOi]), \ + be32_to_cpu((a)->agf_levels[XFS_BTNUM_CNTi]), mp)) +#define XFS_MIN_FREELIST_PAG(pag,mp) \ + (XFS_MIN_FREELIST_RAW( \ + (uint_t)(pag)->pagf_levels[XFS_BTNUM_BNOi], \ + (uint_t)(pag)->pagf_levels[XFS_BTNUM_CNTi], mp)) -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_AGB_TO_FSB) -xfs_fsblock_t xfs_agb_to_fsb(struct xfs_mount *mp, xfs_agnumber_t agno, - xfs_agblock_t agbno); -#define XFS_AGB_TO_FSB(mp,agno,agbno) xfs_agb_to_fsb(mp,agno,agbno) -#else -#define XFS_AGB_TO_FSB(mp,agno,agbno) \ +#define XFS_AGB_TO_FSB(mp,agno,agbno) \ (((xfs_fsblock_t)(agno) << (mp)->m_sb.sb_agblklog) | (agbno)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_FSB_TO_AGNO) -xfs_agnumber_t xfs_fsb_to_agno(struct xfs_mount *mp, xfs_fsblock_t fsbno); -#define XFS_FSB_TO_AGNO(mp,fsbno) xfs_fsb_to_agno(mp,fsbno) -#else -#define XFS_FSB_TO_AGNO(mp,fsbno) \ +#define XFS_FSB_TO_AGNO(mp,fsbno) \ ((xfs_agnumber_t)((fsbno) >> (mp)->m_sb.sb_agblklog)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_FSB_TO_AGBNO) -xfs_agblock_t xfs_fsb_to_agbno(struct xfs_mount *mp, xfs_fsblock_t fsbno); -#define XFS_FSB_TO_AGBNO(mp,fsbno) xfs_fsb_to_agbno(mp,fsbno) -#else -#define XFS_FSB_TO_AGBNO(mp,fsbno) \ +#define XFS_FSB_TO_AGBNO(mp,fsbno) \ ((xfs_agblock_t)((fsbno) & XFS_MASK32LO((mp)->m_sb.sb_agblklog))) -#endif - -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_AGB_TO_DADDR) -xfs_daddr_t xfs_agb_to_daddr(struct xfs_mount *mp, xfs_agnumber_t agno, - xfs_agblock_t agbno); -#define XFS_AGB_TO_DADDR(mp,agno,agbno) xfs_agb_to_daddr(mp,agno,agbno) -#else -#define XFS_AGB_TO_DADDR(mp,agno,agbno) \ - ((xfs_daddr_t)(XFS_FSB_TO_BB(mp, \ - (xfs_fsblock_t)(agno) * (mp)->m_sb.sb_agblocks + (agbno)))) -#endif -/* - * XFS_DADDR_TO_AGNO and XFS_DADDR_TO_AGBNO moved to xfs_mount.h - * to avoid header file ordering change - */ - -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_AG_DADDR) -xfs_daddr_t xfs_ag_daddr(struct xfs_mount *mp, xfs_agnumber_t agno, - xfs_daddr_t d); -#define XFS_AG_DADDR(mp,agno,d) xfs_ag_daddr(mp,agno,d) -#else -#define XFS_AG_DADDR(mp,agno,d) (XFS_AGB_TO_DADDR(mp, agno, 0) + (d)) -#endif - -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BUF_TO_AGF) -xfs_agf_t *xfs_buf_to_agf(struct xfs_buf *bp); -#define XFS_BUF_TO_AGF(bp) xfs_buf_to_agf(bp) -#else -#define XFS_BUF_TO_AGF(bp) ((xfs_agf_t *)XFS_BUF_PTR(bp)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BUF_TO_AGI) -xfs_agi_t *xfs_buf_to_agi(struct xfs_buf *bp); -#define XFS_BUF_TO_AGI(bp) xfs_buf_to_agi(bp) -#else -#define XFS_BUF_TO_AGI(bp) ((xfs_agi_t *)XFS_BUF_PTR(bp)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BUF_TO_AGFL) -xfs_agfl_t *xfs_buf_to_agfl(struct xfs_buf *bp); -#define XFS_BUF_TO_AGFL(bp) xfs_buf_to_agfl(bp) -#else -#define XFS_BUF_TO_AGFL(bp) ((xfs_agfl_t *)XFS_BUF_PTR(bp)) -#endif +#define XFS_AGB_TO_DADDR(mp,agno,agbno) \ + ((xfs_daddr_t)XFS_FSB_TO_BB(mp, \ + (xfs_fsblock_t)(agno) * (mp)->m_sb.sb_agblocks + (agbno))) +#define XFS_AG_DADDR(mp,agno,d) (XFS_AGB_TO_DADDR(mp, agno, 0) + (d)) /* * For checking for bad ranges of xfs_daddr_t's, covering multiple * allocation groups or a single xfs_daddr_t that's a superblock copy. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_AG_CHECK_DADDR) -void xfs_ag_check_daddr(struct xfs_mount *mp, xfs_daddr_t d, xfs_extlen_t len); -#define XFS_AG_CHECK_DADDR(mp,d,len) xfs_ag_check_daddr(mp,d,len) -#else #define XFS_AG_CHECK_DADDR(mp,d,len) \ ((len) == 1 ? \ ASSERT((d) == XFS_SB_DADDR || \ XFS_DADDR_TO_AGBNO(mp, d) != XFS_SB_DADDR) : \ ASSERT(XFS_DADDR_TO_AGNO(mp, d) == \ XFS_DADDR_TO_AGNO(mp, (d) + (len) - 1))) -#endif #endif /* __XFS_AG_H__ */ diff --git a/fs/xfs/xfs_alloc.c b/fs/xfs/xfs_alloc.c index dcfe19703620..f4328e1e2a74 100644 --- a/fs/xfs/xfs_alloc.c +++ b/fs/xfs/xfs_alloc.c @@ -1,56 +1,44 @@ /* - * Copyright (c) 2000-2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ - */ - -/* - * Free space allocation for XFS. + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #include "xfs.h" -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" #include "xfs_dir.h" +#include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" -#include "xfs_alloc_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" +#include "xfs_dir_sf.h" +#include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" +#include "xfs_dinode.h" +#include "xfs_inode.h" #include "xfs_btree.h" #include "xfs_ialloc.h" #include "xfs_alloc.h" -#include "xfs_bit.h" #include "xfs_error.h" @@ -243,8 +231,8 @@ xfs_alloc_fix_minleft( if (args->minleft == 0) return 1; agf = XFS_BUF_TO_AGF(args->agbp); - diff = INT_GET(agf->agf_freeblks, ARCH_CONVERT) - + INT_GET(agf->agf_flcount, ARCH_CONVERT) + diff = be32_to_cpu(agf->agf_freeblks) + + be32_to_cpu(agf->agf_flcount) - args->len - args->minleft; if (diff >= 0) return 1; @@ -319,7 +307,8 @@ xfs_alloc_fixup_trees( bnoblock = XFS_BUF_TO_ALLOC_BLOCK(bno_cur->bc_bufs[0]); cntblock = XFS_BUF_TO_ALLOC_BLOCK(cnt_cur->bc_bufs[0]); XFS_WANT_CORRUPTED_RETURN( - INT_GET(bnoblock->bb_numrecs, ARCH_CONVERT) == INT_GET(cntblock->bb_numrecs, ARCH_CONVERT)); + be16_to_cpu(bnoblock->bb_numrecs) == + be16_to_cpu(cntblock->bb_numrecs)); } } #endif @@ -505,21 +494,17 @@ xfs_alloc_trace_modagf( (void *)str, (void *)mp, (void *)(__psint_t)flags, - (void *)(__psunsigned_t)INT_GET(agf->agf_seqno, ARCH_CONVERT), - (void *)(__psunsigned_t)INT_GET(agf->agf_length, ARCH_CONVERT), - (void *)(__psunsigned_t)INT_GET(agf->agf_roots[XFS_BTNUM_BNO], - ARCH_CONVERT), - (void *)(__psunsigned_t)INT_GET(agf->agf_roots[XFS_BTNUM_CNT], - ARCH_CONVERT), - (void *)(__psunsigned_t)INT_GET(agf->agf_levels[XFS_BTNUM_BNO], - ARCH_CONVERT), - (void *)(__psunsigned_t)INT_GET(agf->agf_levels[XFS_BTNUM_CNT], - ARCH_CONVERT), - (void *)(__psunsigned_t)INT_GET(agf->agf_flfirst, ARCH_CONVERT), - (void *)(__psunsigned_t)INT_GET(agf->agf_fllast, ARCH_CONVERT), - (void *)(__psunsigned_t)INT_GET(agf->agf_flcount, ARCH_CONVERT), - (void *)(__psunsigned_t)INT_GET(agf->agf_freeblks, ARCH_CONVERT), - (void *)(__psunsigned_t)INT_GET(agf->agf_longest, ARCH_CONVERT)); + (void *)(__psunsigned_t)be32_to_cpu(agf->agf_seqno), + (void *)(__psunsigned_t)be32_to_cpu(agf->agf_length), + (void *)(__psunsigned_t)be32_to_cpu(agf->agf_roots[XFS_BTNUM_BNO]), + (void *)(__psunsigned_t)be32_to_cpu(agf->agf_roots[XFS_BTNUM_CNT]), + (void *)(__psunsigned_t)be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNO]), + (void *)(__psunsigned_t)be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNT]), + (void *)(__psunsigned_t)be32_to_cpu(agf->agf_flfirst), + (void *)(__psunsigned_t)be32_to_cpu(agf->agf_fllast), + (void *)(__psunsigned_t)be32_to_cpu(agf->agf_flcount), + (void *)(__psunsigned_t)be32_to_cpu(agf->agf_freeblks), + (void *)(__psunsigned_t)be32_to_cpu(agf->agf_longest)); } STATIC void @@ -612,12 +597,12 @@ xfs_alloc_ag_vextent( if (!(args->wasfromfl)) { agf = XFS_BUF_TO_AGF(args->agbp); - INT_MOD(agf->agf_freeblks, ARCH_CONVERT, -(args->len)); + be32_add(&agf->agf_freeblks, -(args->len)); xfs_trans_agblocks_delta(args->tp, -((long)(args->len))); args->pag->pagf_freeblks -= args->len; - ASSERT(INT_GET(agf->agf_freeblks, ARCH_CONVERT) - <= INT_GET(agf->agf_length, ARCH_CONVERT)); + ASSERT(be32_to_cpu(agf->agf_freeblks) <= + be32_to_cpu(agf->agf_length)); TRACE_MODAGF(NULL, agf, XFS_AGF_FREEBLKS); xfs_alloc_log_agf(args->tp, args->agbp, XFS_AGF_FREEBLKS); @@ -723,8 +708,7 @@ xfs_alloc_ag_vextent_exact( cnt_cur = xfs_btree_init_cursor(args->mp, args->tp, args->agbp, args->agno, XFS_BTNUM_CNT, NULL, 0); ASSERT(args->agbno + args->len <= - INT_GET(XFS_BUF_TO_AGF(args->agbp)->agf_length, - ARCH_CONVERT)); + be32_to_cpu(XFS_BUF_TO_AGF(args->agbp)->agf_length)); if ((error = xfs_alloc_fixup_trees(cnt_cur, bno_cur, fbno, flen, args->agbno, args->len, XFSA_FIXUP_BNO_OK))) { xfs_btree_del_cursor(cnt_cur, XFS_BTREE_ERROR); @@ -897,8 +881,7 @@ xfs_alloc_ag_vextent_near( goto error0; XFS_WANT_CORRUPTED_GOTO(i == 1, error0); ltend = ltbno + ltlen; - ASSERT(ltend <= INT_GET(XFS_BUF_TO_AGF(args->agbp)->agf_length, - ARCH_CONVERT)); + ASSERT(ltend <= be32_to_cpu(XFS_BUF_TO_AGF(args->agbp)->agf_length)); args->len = blen; if (!xfs_alloc_fix_minleft(args)) { xfs_btree_del_cursor(cnt_cur, XFS_BTREE_NOERROR); @@ -1253,8 +1236,7 @@ xfs_alloc_ag_vextent_near( ltlen, <new); ASSERT(ltnew >= ltbno); ASSERT(ltnew + rlen <= ltend); - ASSERT(ltnew + rlen <= INT_GET(XFS_BUF_TO_AGF(args->agbp)->agf_length, - ARCH_CONVERT)); + ASSERT(ltnew + rlen <= be32_to_cpu(XFS_BUF_TO_AGF(args->agbp)->agf_length)); args->agbno = ltnew; if ((error = xfs_alloc_fixup_trees(cnt_cur, bno_cur_lt, ltbno, ltlen, ltnew, rlen, XFSA_FIXUP_BNO_OK))) @@ -1417,8 +1399,7 @@ xfs_alloc_ag_vextent_size( args->agbno = rbno; XFS_WANT_CORRUPTED_GOTO( args->agbno + args->len <= - INT_GET(XFS_BUF_TO_AGF(args->agbp)->agf_length, - ARCH_CONVERT), + be32_to_cpu(XFS_BUF_TO_AGF(args->agbp)->agf_length), error0); TRACE_ALLOC("normal", args); return 0; @@ -1466,8 +1447,8 @@ xfs_alloc_ag_vextent_small( * freelist. */ else if (args->minlen == 1 && args->alignment == 1 && !args->isfl && - (INT_GET(XFS_BUF_TO_AGF(args->agbp)->agf_flcount, - ARCH_CONVERT) > args->minleft)) { + (be32_to_cpu(XFS_BUF_TO_AGF(args->agbp)->agf_flcount) + > args->minleft)) { if ((error = xfs_alloc_get_freelist(args->tp, args->agbp, &fbno))) goto error0; if (fbno != NULLAGBLOCK) { @@ -1482,8 +1463,7 @@ xfs_alloc_ag_vextent_small( args->agbno = fbno; XFS_WANT_CORRUPTED_GOTO( args->agbno + args->len <= - INT_GET(XFS_BUF_TO_AGF(args->agbp)->agf_length, - ARCH_CONVERT), + be32_to_cpu(XFS_BUF_TO_AGF(args->agbp)->agf_length), error0); args->wasfromfl = 1; TRACE_ALLOC("freelist", args); @@ -1757,12 +1737,12 @@ xfs_free_ag_extent( agf = XFS_BUF_TO_AGF(agbp); pag = &mp->m_perag[agno]; - INT_MOD(agf->agf_freeblks, ARCH_CONVERT, len); + be32_add(&agf->agf_freeblks, len); xfs_trans_agblocks_delta(tp, len); pag->pagf_freeblks += len; XFS_WANT_CORRUPTED_GOTO( - INT_GET(agf->agf_freeblks, ARCH_CONVERT) - <= INT_GET(agf->agf_length, ARCH_CONVERT), + be32_to_cpu(agf->agf_freeblks) <= + be32_to_cpu(agf->agf_length), error0); TRACE_MODAGF(NULL, agf, XFS_AGF_FREEBLKS); xfs_alloc_log_agf(tp, agbp, XFS_AGF_FREEBLKS); @@ -1909,18 +1889,18 @@ xfs_alloc_fix_freelist( */ agf = XFS_BUF_TO_AGF(agbp); need = XFS_MIN_FREELIST(agf, mp); - delta = need > INT_GET(agf->agf_flcount, ARCH_CONVERT) ? - (need - INT_GET(agf->agf_flcount, ARCH_CONVERT)) : 0; + delta = need > be32_to_cpu(agf->agf_flcount) ? + (need - be32_to_cpu(agf->agf_flcount)) : 0; /* * If there isn't enough total or single-extent, reject it. */ - longest = INT_GET(agf->agf_longest, ARCH_CONVERT); + longest = be32_to_cpu(agf->agf_longest); longest = (longest > delta) ? (longest - delta) : - (INT_GET(agf->agf_flcount, ARCH_CONVERT) > 0 || longest > 0); + (be32_to_cpu(agf->agf_flcount) > 0 || longest > 0); if (args->minlen + args->alignment + args->minalignslop - 1 > longest || (args->minleft && - (int)(INT_GET(agf->agf_freeblks, ARCH_CONVERT) + - INT_GET(agf->agf_flcount, ARCH_CONVERT) - need - args->total) < + (int)(be32_to_cpu(agf->agf_freeblks) + + be32_to_cpu(agf->agf_flcount) - need - args->total) < (int)args->minleft)) { xfs_trans_brelse(tp, agbp); args->agbp = NULL; @@ -1929,7 +1909,7 @@ xfs_alloc_fix_freelist( /* * Make the freelist shorter if it's too long. */ - while (INT_GET(agf->agf_flcount, ARCH_CONVERT) > need) { + while (be32_to_cpu(agf->agf_flcount) > need) { xfs_buf_t *bp; if ((error = xfs_alloc_get_freelist(tp, agbp, &bno))) @@ -1956,9 +1936,9 @@ xfs_alloc_fix_freelist( /* * Make the freelist longer if it's too short. */ - while (INT_GET(agf->agf_flcount, ARCH_CONVERT) < need) { + while (be32_to_cpu(agf->agf_flcount) < need) { targs.agbno = 0; - targs.maxlen = need - INT_GET(agf->agf_flcount, ARCH_CONVERT); + targs.maxlen = need - be32_to_cpu(agf->agf_flcount); /* * Allocate as many blocks as possible at once. */ @@ -2018,19 +1998,19 @@ xfs_alloc_get_freelist( */ mp = tp->t_mountp; if ((error = xfs_alloc_read_agfl(mp, tp, - INT_GET(agf->agf_seqno, ARCH_CONVERT), &agflbp))) + be32_to_cpu(agf->agf_seqno), &agflbp))) return error; agfl = XFS_BUF_TO_AGFL(agflbp); /* * Get the block number and update the data structures. */ - bno = INT_GET(agfl->agfl_bno[INT_GET(agf->agf_flfirst, ARCH_CONVERT)], ARCH_CONVERT); - INT_MOD(agf->agf_flfirst, ARCH_CONVERT, 1); + bno = INT_GET(agfl->agfl_bno[be32_to_cpu(agf->agf_flfirst)], ARCH_CONVERT); + be32_add(&agf->agf_flfirst, 1); xfs_trans_brelse(tp, agflbp); - if (INT_GET(agf->agf_flfirst, ARCH_CONVERT) == XFS_AGFL_SIZE(mp)) + if (be32_to_cpu(agf->agf_flfirst) == XFS_AGFL_SIZE(mp)) agf->agf_flfirst = 0; - pag = &mp->m_perag[INT_GET(agf->agf_seqno, ARCH_CONVERT)]; - INT_MOD(agf->agf_flcount, ARCH_CONVERT, -1); + pag = &mp->m_perag[be32_to_cpu(agf->agf_seqno)]; + be32_add(&agf->agf_flcount, -1); xfs_trans_agflist_delta(tp, -1); pag->pagf_flcount--; TRACE_MODAGF(NULL, agf, XFS_AGF_FLFIRST | XFS_AGF_FLCOUNT); @@ -2045,7 +2025,7 @@ xfs_alloc_get_freelist( * the freeing transaction must be pushed to disk NOW by forcing * to disk all iclogs up that transaction's LSN. */ - xfs_alloc_search_busy(tp, INT_GET(agf->agf_seqno, ARCH_CONVERT), bno, 1); + xfs_alloc_search_busy(tp, be32_to_cpu(agf->agf_seqno), bno, 1); return 0; } @@ -2123,18 +2103,18 @@ xfs_alloc_put_freelist( mp = tp->t_mountp; if (!agflbp && (error = xfs_alloc_read_agfl(mp, tp, - INT_GET(agf->agf_seqno, ARCH_CONVERT), &agflbp))) + be32_to_cpu(agf->agf_seqno), &agflbp))) return error; agfl = XFS_BUF_TO_AGFL(agflbp); - INT_MOD(agf->agf_fllast, ARCH_CONVERT, 1); - if (INT_GET(agf->agf_fllast, ARCH_CONVERT) == XFS_AGFL_SIZE(mp)) + be32_add(&agf->agf_fllast, 1); + if (be32_to_cpu(agf->agf_fllast) == XFS_AGFL_SIZE(mp)) agf->agf_fllast = 0; - pag = &mp->m_perag[INT_GET(agf->agf_seqno, ARCH_CONVERT)]; - INT_MOD(agf->agf_flcount, ARCH_CONVERT, 1); + pag = &mp->m_perag[be32_to_cpu(agf->agf_seqno)]; + be32_add(&agf->agf_flcount, 1); xfs_trans_agflist_delta(tp, 1); pag->pagf_flcount++; - ASSERT(INT_GET(agf->agf_flcount, ARCH_CONVERT) <= XFS_AGFL_SIZE(mp)); - blockp = &agfl->agfl_bno[INT_GET(agf->agf_fllast, ARCH_CONVERT)]; + ASSERT(be32_to_cpu(agf->agf_flcount) <= XFS_AGFL_SIZE(mp)); + blockp = &agfl->agfl_bno[be32_to_cpu(agf->agf_fllast)]; INT_SET(*blockp, ARCH_CONVERT, bno); TRACE_MODAGF(NULL, agf, XFS_AGF_FLLAST | XFS_AGF_FLCOUNT); xfs_alloc_log_agf(tp, agbp, XFS_AGF_FLLAST | XFS_AGF_FLCOUNT); @@ -2181,14 +2161,12 @@ xfs_alloc_read_agf( */ agf = XFS_BUF_TO_AGF(bp); agf_ok = - INT_GET(agf->agf_magicnum, ARCH_CONVERT) == XFS_AGF_MAGIC && - XFS_AGF_GOOD_VERSION( - INT_GET(agf->agf_versionnum, ARCH_CONVERT)) && - INT_GET(agf->agf_freeblks, ARCH_CONVERT) <= - INT_GET(agf->agf_length, ARCH_CONVERT) && - INT_GET(agf->agf_flfirst, ARCH_CONVERT) < XFS_AGFL_SIZE(mp) && - INT_GET(agf->agf_fllast, ARCH_CONVERT) < XFS_AGFL_SIZE(mp) && - INT_GET(agf->agf_flcount, ARCH_CONVERT) <= XFS_AGFL_SIZE(mp); + be32_to_cpu(agf->agf_magicnum) == XFS_AGF_MAGIC && + XFS_AGF_GOOD_VERSION(be32_to_cpu(agf->agf_versionnum)) && + be32_to_cpu(agf->agf_freeblks) <= be32_to_cpu(agf->agf_length) && + be32_to_cpu(agf->agf_flfirst) < XFS_AGFL_SIZE(mp) && + be32_to_cpu(agf->agf_fllast) < XFS_AGFL_SIZE(mp) && + be32_to_cpu(agf->agf_flcount) <= XFS_AGFL_SIZE(mp); if (unlikely(XFS_TEST_ERROR(!agf_ok, mp, XFS_ERRTAG_ALLOC_READ_AGF, XFS_RANDOM_ALLOC_READ_AGF))) { XFS_CORRUPTION_ERROR("xfs_alloc_read_agf", @@ -2198,13 +2176,13 @@ xfs_alloc_read_agf( } pag = &mp->m_perag[agno]; if (!pag->pagf_init) { - pag->pagf_freeblks = INT_GET(agf->agf_freeblks, ARCH_CONVERT); - pag->pagf_flcount = INT_GET(agf->agf_flcount, ARCH_CONVERT); - pag->pagf_longest = INT_GET(agf->agf_longest, ARCH_CONVERT); + pag->pagf_freeblks = be32_to_cpu(agf->agf_freeblks); + pag->pagf_flcount = be32_to_cpu(agf->agf_flcount); + pag->pagf_longest = be32_to_cpu(agf->agf_longest); pag->pagf_levels[XFS_BTNUM_BNOi] = - INT_GET(agf->agf_levels[XFS_BTNUM_BNOi], ARCH_CONVERT); + be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNOi]); pag->pagf_levels[XFS_BTNUM_CNTi] = - INT_GET(agf->agf_levels[XFS_BTNUM_CNTi], ARCH_CONVERT); + be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNTi]); spinlock_init(&pag->pagb_lock, "xfspagb"); pag->pagb_list = kmem_zalloc(XFS_PAGB_NUM_SLOTS * sizeof(xfs_perag_busy_t), KM_SLEEP); @@ -2212,13 +2190,13 @@ xfs_alloc_read_agf( } #ifdef DEBUG else if (!XFS_FORCED_SHUTDOWN(mp)) { - ASSERT(pag->pagf_freeblks == INT_GET(agf->agf_freeblks, ARCH_CONVERT)); - ASSERT(pag->pagf_flcount == INT_GET(agf->agf_flcount, ARCH_CONVERT)); - ASSERT(pag->pagf_longest == INT_GET(agf->agf_longest, ARCH_CONVERT)); + ASSERT(pag->pagf_freeblks == be32_to_cpu(agf->agf_freeblks)); + ASSERT(pag->pagf_flcount == be32_to_cpu(agf->agf_flcount)); + ASSERT(pag->pagf_longest == be32_to_cpu(agf->agf_longest)); ASSERT(pag->pagf_levels[XFS_BTNUM_BNOi] == - INT_GET(agf->agf_levels[XFS_BTNUM_BNOi], ARCH_CONVERT)); + be32_to_cpu(agf->agf_levels[XFS_BTNUM_BNOi])); ASSERT(pag->pagf_levels[XFS_BTNUM_CNTi] == - INT_GET(agf->agf_levels[XFS_BTNUM_CNTi], ARCH_CONVERT)); + be32_to_cpu(agf->agf_levels[XFS_BTNUM_CNTi])); } #endif XFS_BUF_SET_VTYPE_REF(bp, B_FS_AGF, XFS_AGF_REF); @@ -2467,7 +2445,7 @@ xfs_free_extent( #ifdef DEBUG ASSERT(args.agbp != NULL); agf = XFS_BUF_TO_AGF(args.agbp); - ASSERT(args.agbno + len <= INT_GET(agf->agf_length, ARCH_CONVERT)); + ASSERT(args.agbno + len <= be32_to_cpu(agf->agf_length)); #endif error = xfs_free_ag_extent(tp, args.agbp, args.agno, args.agbno, len, 0); diff --git a/fs/xfs/xfs_alloc.h b/fs/xfs/xfs_alloc.h index 72329c86351c..3546dea27b7d 100644 --- a/fs/xfs/xfs_alloc.h +++ b/fs/xfs/xfs_alloc.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_ALLOC_H__ #define __XFS_ALLOC_H__ diff --git a/fs/xfs/xfs_alloc_btree.c b/fs/xfs/xfs_alloc_btree.c index e0355a12d946..a1d92da86ccd 100644 --- a/fs/xfs/xfs_alloc_btree.c +++ b/fs/xfs/xfs_alloc_btree.c @@ -1,53 +1,41 @@ /* - * Copyright (c) 2000-2001 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2001,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - -/* - * Free space allocation for XFS. - */ - #include "xfs.h" -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" #include "xfs_dir.h" +#include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" +#include "xfs_bmap_btree.h" #include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_bmap_btree.h" +#include "xfs_dir_sf.h" +#include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" +#include "xfs_dinode.h" +#include "xfs_inode.h" #include "xfs_btree.h" #include "xfs_ialloc.h" #include "xfs_alloc.h" @@ -129,7 +117,7 @@ xfs_alloc_delrec( /* * Fail if we're off the end of the block. */ - if (ptr > INT_GET(block->bb_numrecs, ARCH_CONVERT)) { + if (ptr > be16_to_cpu(block->bb_numrecs)) { *stat = 0; return 0; } @@ -143,18 +131,18 @@ xfs_alloc_delrec( lkp = XFS_ALLOC_KEY_ADDR(block, 1, cur); lpp = XFS_ALLOC_PTR_ADDR(block, 1, cur); #ifdef DEBUG - for (i = ptr; i < INT_GET(block->bb_numrecs, ARCH_CONVERT); i++) { - if ((error = xfs_btree_check_sptr(cur, INT_GET(lpp[i], ARCH_CONVERT), level))) + for (i = ptr; i < be16_to_cpu(block->bb_numrecs); i++) { + if ((error = xfs_btree_check_sptr(cur, be32_to_cpu(lpp[i]), level))) return error; } #endif - if (ptr < INT_GET(block->bb_numrecs, ARCH_CONVERT)) { + if (ptr < be16_to_cpu(block->bb_numrecs)) { memmove(&lkp[ptr - 1], &lkp[ptr], - (INT_GET(block->bb_numrecs, ARCH_CONVERT) - ptr) * sizeof(*lkp)); /* INT_: mem copy */ + (be16_to_cpu(block->bb_numrecs) - ptr) * sizeof(*lkp)); memmove(&lpp[ptr - 1], &lpp[ptr], - (INT_GET(block->bb_numrecs, ARCH_CONVERT) - ptr) * sizeof(*lpp)); /* INT_: mem copy */ - xfs_alloc_log_ptrs(cur, bp, ptr, INT_GET(block->bb_numrecs, ARCH_CONVERT) - 1); - xfs_alloc_log_keys(cur, bp, ptr, INT_GET(block->bb_numrecs, ARCH_CONVERT) - 1); + (be16_to_cpu(block->bb_numrecs) - ptr) * sizeof(*lpp)); + xfs_alloc_log_ptrs(cur, bp, ptr, be16_to_cpu(block->bb_numrecs) - 1); + xfs_alloc_log_keys(cur, bp, ptr, be16_to_cpu(block->bb_numrecs) - 1); } } /* @@ -163,25 +151,25 @@ xfs_alloc_delrec( */ else { lrp = XFS_ALLOC_REC_ADDR(block, 1, cur); - if (ptr < INT_GET(block->bb_numrecs, ARCH_CONVERT)) { + if (ptr < be16_to_cpu(block->bb_numrecs)) { memmove(&lrp[ptr - 1], &lrp[ptr], - (INT_GET(block->bb_numrecs, ARCH_CONVERT) - ptr) * sizeof(*lrp)); - xfs_alloc_log_recs(cur, bp, ptr, INT_GET(block->bb_numrecs, ARCH_CONVERT) - 1); + (be16_to_cpu(block->bb_numrecs) - ptr) * sizeof(*lrp)); + xfs_alloc_log_recs(cur, bp, ptr, be16_to_cpu(block->bb_numrecs) - 1); } /* * If it's the first record in the block, we'll need a key * structure to pass up to the next level (updkey). */ if (ptr == 1) { - key.ar_startblock = lrp->ar_startblock; /* INT_: direct copy */ - key.ar_blockcount = lrp->ar_blockcount; /* INT_: direct copy */ + key.ar_startblock = lrp->ar_startblock; + key.ar_blockcount = lrp->ar_blockcount; lkp = &key; } } /* * Decrement and log the number of entries in the block. */ - INT_MOD(block->bb_numrecs, ARCH_CONVERT, -1); + be16_add(&block->bb_numrecs, -1); xfs_alloc_log_block(cur->bc_tp, bp, XFS_BB_NUMRECS); /* * See if the longest free extent in the allocation group was @@ -194,24 +182,24 @@ xfs_alloc_delrec( if (level == 0 && cur->bc_btnum == XFS_BTNUM_CNT && - INT_GET(block->bb_rightsib, ARCH_CONVERT) == NULLAGBLOCK && - ptr > INT_GET(block->bb_numrecs, ARCH_CONVERT)) { - ASSERT(ptr == INT_GET(block->bb_numrecs, ARCH_CONVERT) + 1); + be32_to_cpu(block->bb_rightsib) == NULLAGBLOCK && + ptr > be16_to_cpu(block->bb_numrecs)) { + ASSERT(ptr == be16_to_cpu(block->bb_numrecs) + 1); /* * There are still records in the block. Grab the size * from the last one. */ - if (INT_GET(block->bb_numrecs, ARCH_CONVERT)) { - rrp = XFS_ALLOC_REC_ADDR(block, INT_GET(block->bb_numrecs, ARCH_CONVERT), cur); - INT_COPY(agf->agf_longest, rrp->ar_blockcount, ARCH_CONVERT); + if (be16_to_cpu(block->bb_numrecs)) { + rrp = XFS_ALLOC_REC_ADDR(block, be16_to_cpu(block->bb_numrecs), cur); + agf->agf_longest = rrp->ar_blockcount; } /* * No free extents left. */ else agf->agf_longest = 0; - mp->m_perag[INT_GET(agf->agf_seqno, ARCH_CONVERT)].pagf_longest = - INT_GET(agf->agf_longest, ARCH_CONVERT); + mp->m_perag[be32_to_cpu(agf->agf_seqno)].pagf_longest = + be32_to_cpu(agf->agf_longest); xfs_alloc_log_agf(cur->bc_tp, cur->bc_private.a.agbp, XFS_AGF_LONGEST); } @@ -225,15 +213,15 @@ xfs_alloc_delrec( * and it's NOT the leaf level, * then we can get rid of this level. */ - if (INT_GET(block->bb_numrecs, ARCH_CONVERT) == 1 && level > 0) { + if (be16_to_cpu(block->bb_numrecs) == 1 && level > 0) { /* * lpp is still set to the first pointer in the block. * Make it the new root of the btree. */ - bno = INT_GET(agf->agf_roots[cur->bc_btnum], ARCH_CONVERT); - INT_COPY(agf->agf_roots[cur->bc_btnum], *lpp, ARCH_CONVERT); - INT_MOD(agf->agf_levels[cur->bc_btnum], ARCH_CONVERT, -1); - mp->m_perag[INT_GET(agf->agf_seqno, ARCH_CONVERT)].pagf_levels[cur->bc_btnum]--; + bno = be32_to_cpu(agf->agf_roots[cur->bc_btnum]); + agf->agf_roots[cur->bc_btnum] = *lpp; + be32_add(&agf->agf_levels[cur->bc_btnum], -1); + mp->m_perag[be32_to_cpu(agf->agf_seqno)].pagf_levels[cur->bc_btnum]--; /* * Put this buffer/block on the ag's freelist. */ @@ -255,7 +243,7 @@ xfs_alloc_delrec( * that freed the block. */ xfs_alloc_mark_busy(cur->bc_tp, - INT_GET(agf->agf_seqno, ARCH_CONVERT), bno, 1); + be32_to_cpu(agf->agf_seqno), bno, 1); xfs_trans_agbtree_delta(cur->bc_tp, -1); xfs_alloc_log_agf(cur->bc_tp, cur->bc_private.a.agbp, @@ -281,7 +269,7 @@ xfs_alloc_delrec( * If the number of records remaining in the block is at least * the minimum, we're done. */ - if (INT_GET(block->bb_numrecs, ARCH_CONVERT) >= XFS_ALLOC_BLOCK_MINRECS(level, cur)) { + if (be16_to_cpu(block->bb_numrecs) >= XFS_ALLOC_BLOCK_MINRECS(level, cur)) { if (level > 0 && (error = xfs_alloc_decrement(cur, level, &i))) return error; *stat = 1; @@ -292,8 +280,8 @@ xfs_alloc_delrec( * tree balanced. Look at the left and right sibling blocks to * see if we can re-balance by moving only one record. */ - rbno = INT_GET(block->bb_rightsib, ARCH_CONVERT); - lbno = INT_GET(block->bb_leftsib, ARCH_CONVERT); + rbno = be32_to_cpu(block->bb_rightsib); + lbno = be32_to_cpu(block->bb_leftsib); bno = NULLAGBLOCK; ASSERT(rbno != NULLAGBLOCK || lbno != NULLAGBLOCK); /* @@ -330,18 +318,18 @@ xfs_alloc_delrec( /* * Grab the current block number, for future use. */ - bno = INT_GET(right->bb_leftsib, ARCH_CONVERT); + bno = be32_to_cpu(right->bb_leftsib); /* * If right block is full enough so that removing one entry * won't make it too empty, and left-shifting an entry out * of right to us works, we're done. */ - if (INT_GET(right->bb_numrecs, ARCH_CONVERT) - 1 >= + if (be16_to_cpu(right->bb_numrecs) - 1 >= XFS_ALLOC_BLOCK_MINRECS(level, cur)) { if ((error = xfs_alloc_lshift(tcur, level, &i))) goto error0; if (i) { - ASSERT(INT_GET(block->bb_numrecs, ARCH_CONVERT) >= + ASSERT(be16_to_cpu(block->bb_numrecs) >= XFS_ALLOC_BLOCK_MINRECS(level, cur)); xfs_btree_del_cursor(tcur, XFS_BTREE_NOERROR); @@ -358,7 +346,7 @@ xfs_alloc_delrec( * future reference, and fix up the temp cursor to point * to our block again (last record). */ - rrecs = INT_GET(right->bb_numrecs, ARCH_CONVERT); + rrecs = be16_to_cpu(right->bb_numrecs); if (lbno != NULLAGBLOCK) { i = xfs_btree_firstrec(tcur, level); XFS_WANT_CORRUPTED_GOTO(i == 1, error0); @@ -394,18 +382,18 @@ xfs_alloc_delrec( /* * Grab the current block number, for future use. */ - bno = INT_GET(left->bb_rightsib, ARCH_CONVERT); + bno = be32_to_cpu(left->bb_rightsib); /* * If left block is full enough so that removing one entry * won't make it too empty, and right-shifting an entry out * of left to us works, we're done. */ - if (INT_GET(left->bb_numrecs, ARCH_CONVERT) - 1 >= + if (be16_to_cpu(left->bb_numrecs) - 1 >= XFS_ALLOC_BLOCK_MINRECS(level, cur)) { if ((error = xfs_alloc_rshift(tcur, level, &i))) goto error0; if (i) { - ASSERT(INT_GET(block->bb_numrecs, ARCH_CONVERT) >= + ASSERT(be16_to_cpu(block->bb_numrecs) >= XFS_ALLOC_BLOCK_MINRECS(level, cur)); xfs_btree_del_cursor(tcur, XFS_BTREE_NOERROR); @@ -419,7 +407,7 @@ xfs_alloc_delrec( * Otherwise, grab the number of records in right for * future reference. */ - lrecs = INT_GET(left->bb_numrecs, ARCH_CONVERT); + lrecs = be16_to_cpu(left->bb_numrecs); } /* * Delete the temp cursor, we're done with it. @@ -433,7 +421,7 @@ xfs_alloc_delrec( * See if we can join with the left neighbor block. */ if (lbno != NULLAGBLOCK && - lrecs + INT_GET(block->bb_numrecs, ARCH_CONVERT) <= XFS_ALLOC_BLOCK_MAXRECS(level, cur)) { + lrecs + be16_to_cpu(block->bb_numrecs) <= XFS_ALLOC_BLOCK_MAXRECS(level, cur)) { /* * Set "right" to be the starting block, * "left" to be the left neighbor. @@ -453,7 +441,7 @@ xfs_alloc_delrec( * If that won't work, see if we can join with the right neighbor block. */ else if (rbno != NULLAGBLOCK && - rrecs + INT_GET(block->bb_numrecs, ARCH_CONVERT) <= + rrecs + be16_to_cpu(block->bb_numrecs) <= XFS_ALLOC_BLOCK_MAXRECS(level, cur)) { /* * Set "left" to be the starting block, @@ -488,31 +476,34 @@ xfs_alloc_delrec( /* * It's a non-leaf. Move keys and pointers. */ - lkp = XFS_ALLOC_KEY_ADDR(left, INT_GET(left->bb_numrecs, ARCH_CONVERT) + 1, cur); - lpp = XFS_ALLOC_PTR_ADDR(left, INT_GET(left->bb_numrecs, ARCH_CONVERT) + 1, cur); + lkp = XFS_ALLOC_KEY_ADDR(left, be16_to_cpu(left->bb_numrecs) + 1, cur); + lpp = XFS_ALLOC_PTR_ADDR(left, be16_to_cpu(left->bb_numrecs) + 1, cur); rkp = XFS_ALLOC_KEY_ADDR(right, 1, cur); rpp = XFS_ALLOC_PTR_ADDR(right, 1, cur); #ifdef DEBUG - for (i = 0; i < INT_GET(right->bb_numrecs, ARCH_CONVERT); i++) { - if ((error = xfs_btree_check_sptr(cur, INT_GET(rpp[i], ARCH_CONVERT), level))) + for (i = 0; i < be16_to_cpu(right->bb_numrecs); i++) { + if ((error = xfs_btree_check_sptr(cur, be32_to_cpu(rpp[i]), level))) return error; } #endif - memcpy(lkp, rkp, INT_GET(right->bb_numrecs, ARCH_CONVERT) * sizeof(*lkp)); /* INT_: structure copy */ - memcpy(lpp, rpp, INT_GET(right->bb_numrecs, ARCH_CONVERT) * sizeof(*lpp)); /* INT_: structure copy */ - xfs_alloc_log_keys(cur, lbp, INT_GET(left->bb_numrecs, ARCH_CONVERT) + 1, - INT_GET(left->bb_numrecs, ARCH_CONVERT) + INT_GET(right->bb_numrecs, ARCH_CONVERT)); - xfs_alloc_log_ptrs(cur, lbp, INT_GET(left->bb_numrecs, ARCH_CONVERT) + 1, - INT_GET(left->bb_numrecs, ARCH_CONVERT) + INT_GET(right->bb_numrecs, ARCH_CONVERT)); + memcpy(lkp, rkp, be16_to_cpu(right->bb_numrecs) * sizeof(*lkp)); + memcpy(lpp, rpp, be16_to_cpu(right->bb_numrecs) * sizeof(*lpp)); + xfs_alloc_log_keys(cur, lbp, be16_to_cpu(left->bb_numrecs) + 1, + be16_to_cpu(left->bb_numrecs) + + be16_to_cpu(right->bb_numrecs)); + xfs_alloc_log_ptrs(cur, lbp, be16_to_cpu(left->bb_numrecs) + 1, + be16_to_cpu(left->bb_numrecs) + + be16_to_cpu(right->bb_numrecs)); } else { /* * It's a leaf. Move records. */ - lrp = XFS_ALLOC_REC_ADDR(left, INT_GET(left->bb_numrecs, ARCH_CONVERT) + 1, cur); + lrp = XFS_ALLOC_REC_ADDR(left, be16_to_cpu(left->bb_numrecs) + 1, cur); rrp = XFS_ALLOC_REC_ADDR(right, 1, cur); - memcpy(lrp, rrp, INT_GET(right->bb_numrecs, ARCH_CONVERT) * sizeof(*lrp)); - xfs_alloc_log_recs(cur, lbp, INT_GET(left->bb_numrecs, ARCH_CONVERT) + 1, - INT_GET(left->bb_numrecs, ARCH_CONVERT) + INT_GET(right->bb_numrecs, ARCH_CONVERT)); + memcpy(lrp, rrp, be16_to_cpu(right->bb_numrecs) * sizeof(*lrp)); + xfs_alloc_log_recs(cur, lbp, be16_to_cpu(left->bb_numrecs) + 1, + be16_to_cpu(left->bb_numrecs) + + be16_to_cpu(right->bb_numrecs)); } /* * If we joined with the left neighbor, set the buffer in the @@ -520,7 +511,7 @@ xfs_alloc_delrec( */ if (bp != lbp) { xfs_btree_setbuf(cur, level, lbp); - cur->bc_ptrs[level] += INT_GET(left->bb_numrecs, ARCH_CONVERT); + cur->bc_ptrs[level] += be16_to_cpu(left->bb_numrecs); } /* * If we joined with the right neighbor and there's a level above @@ -532,28 +523,28 @@ xfs_alloc_delrec( /* * Fix up the number of records in the surviving block. */ - INT_MOD(left->bb_numrecs, ARCH_CONVERT, INT_GET(right->bb_numrecs, ARCH_CONVERT)); + be16_add(&left->bb_numrecs, be16_to_cpu(right->bb_numrecs)); /* * Fix up the right block pointer in the surviving block, and log it. */ - left->bb_rightsib = right->bb_rightsib; /* INT_: direct copy */ + left->bb_rightsib = right->bb_rightsib; xfs_alloc_log_block(cur->bc_tp, lbp, XFS_BB_NUMRECS | XFS_BB_RIGHTSIB); /* * If there is a right sibling now, make it point to the * remaining block. */ - if (INT_GET(left->bb_rightsib, ARCH_CONVERT) != NULLAGBLOCK) { + if (be32_to_cpu(left->bb_rightsib) != NULLAGBLOCK) { xfs_alloc_block_t *rrblock; xfs_buf_t *rrbp; if ((error = xfs_btree_read_bufs(mp, cur->bc_tp, - cur->bc_private.a.agno, INT_GET(left->bb_rightsib, ARCH_CONVERT), 0, + cur->bc_private.a.agno, be32_to_cpu(left->bb_rightsib), 0, &rrbp, XFS_ALLOC_BTREE_REF))) return error; rrblock = XFS_BUF_TO_ALLOC_BLOCK(rrbp); if ((error = xfs_btree_check_sblock(cur, rrblock, level, rrbp))) return error; - INT_SET(rrblock->bb_leftsib, ARCH_CONVERT, lbno); + rrblock->bb_leftsib = cpu_to_be32(lbno); xfs_alloc_log_block(cur->bc_tp, rrbp, XFS_BB_LEFTSIB); } /* @@ -574,10 +565,9 @@ xfs_alloc_delrec( * busy block is allocated, the iclog is pushed up to the * LSN that freed the block. */ - xfs_alloc_mark_busy(cur->bc_tp, - INT_GET(agf->agf_seqno, ARCH_CONVERT), bno, 1); - + xfs_alloc_mark_busy(cur->bc_tp, be32_to_cpu(agf->agf_seqno), bno, 1); xfs_trans_agbtree_delta(cur->bc_tp, -1); + /* * Adjust the current level's cursor so that we're left referring * to the right node, after we're done. @@ -625,7 +615,15 @@ xfs_alloc_insrec( int ptr; /* index in btree block for this rec */ xfs_alloc_rec_t *rp; /* pointer to btree records */ - ASSERT(INT_GET(recp->ar_blockcount, ARCH_CONVERT) > 0); + ASSERT(be32_to_cpu(recp->ar_blockcount) > 0); + + /* + * GCC doesn't understand the (arguably complex) control flow in + * this function and complains about uninitialized structure fields + * without this. + */ + memset(&nrec, 0, sizeof(nrec)); + /* * If we made it to the root level, allocate a new root block * and we're done. @@ -641,8 +639,8 @@ xfs_alloc_insrec( /* * Make a key out of the record data to be inserted, and save it. */ - key.ar_startblock = recp->ar_startblock; /* INT_: direct copy */ - key.ar_blockcount = recp->ar_blockcount; /* INT_: direct copy */ + key.ar_startblock = recp->ar_startblock; + key.ar_blockcount = recp->ar_blockcount; optr = ptr = cur->bc_ptrs[level]; /* * If we're off the left edge, return failure. @@ -663,7 +661,7 @@ xfs_alloc_insrec( /* * Check that the new entry is being inserted in the right place. */ - if (ptr <= INT_GET(block->bb_numrecs, ARCH_CONVERT)) { + if (ptr <= be16_to_cpu(block->bb_numrecs)) { if (level == 0) { rp = XFS_ALLOC_REC_ADDR(block, ptr, cur); xfs_btree_check_rec(cur->bc_btnum, recp, rp); @@ -679,7 +677,7 @@ xfs_alloc_insrec( * If the block is full, we can't insert the new entry until we * make the block un-full. */ - if (INT_GET(block->bb_numrecs, ARCH_CONVERT) == XFS_ALLOC_BLOCK_MAXRECS(level, cur)) { + if (be16_to_cpu(block->bb_numrecs) == XFS_ALLOC_BLOCK_MAXRECS(level, cur)) { /* * First, try shifting an entry to the right neighbor. */ @@ -716,8 +714,8 @@ xfs_alloc_insrec( return error; #endif ptr = cur->bc_ptrs[level]; - nrec.ar_startblock = nkey.ar_startblock; /* INT_: direct copy */ - nrec.ar_blockcount = nkey.ar_blockcount; /* INT_: direct copy */ + nrec.ar_startblock = nkey.ar_startblock; + nrec.ar_blockcount = nkey.ar_blockcount; } /* * Otherwise the insert fails. @@ -741,15 +739,15 @@ xfs_alloc_insrec( kp = XFS_ALLOC_KEY_ADDR(block, 1, cur); pp = XFS_ALLOC_PTR_ADDR(block, 1, cur); #ifdef DEBUG - for (i = INT_GET(block->bb_numrecs, ARCH_CONVERT); i >= ptr; i--) { - if ((error = xfs_btree_check_sptr(cur, INT_GET(pp[i - 1], ARCH_CONVERT), level))) + for (i = be16_to_cpu(block->bb_numrecs); i >= ptr; i--) { + if ((error = xfs_btree_check_sptr(cur, be32_to_cpu(pp[i - 1]), level))) return error; } #endif memmove(&kp[ptr], &kp[ptr - 1], - (INT_GET(block->bb_numrecs, ARCH_CONVERT) - ptr + 1) * sizeof(*kp)); /* INT_: copy */ + (be16_to_cpu(block->bb_numrecs) - ptr + 1) * sizeof(*kp)); memmove(&pp[ptr], &pp[ptr - 1], - (INT_GET(block->bb_numrecs, ARCH_CONVERT) - ptr + 1) * sizeof(*pp)); /* INT_: copy */ + (be16_to_cpu(block->bb_numrecs) - ptr + 1) * sizeof(*pp)); #ifdef DEBUG if ((error = xfs_btree_check_sptr(cur, *bnop, level))) return error; @@ -758,12 +756,12 @@ xfs_alloc_insrec( * Now stuff the new data in, bump numrecs and log the new data. */ kp[ptr - 1] = key; - INT_SET(pp[ptr - 1], ARCH_CONVERT, *bnop); - INT_MOD(block->bb_numrecs, ARCH_CONVERT, +1); - xfs_alloc_log_keys(cur, bp, ptr, INT_GET(block->bb_numrecs, ARCH_CONVERT)); - xfs_alloc_log_ptrs(cur, bp, ptr, INT_GET(block->bb_numrecs, ARCH_CONVERT)); + pp[ptr - 1] = cpu_to_be32(*bnop); + be16_add(&block->bb_numrecs, 1); + xfs_alloc_log_keys(cur, bp, ptr, be16_to_cpu(block->bb_numrecs)); + xfs_alloc_log_ptrs(cur, bp, ptr, be16_to_cpu(block->bb_numrecs)); #ifdef DEBUG - if (ptr < INT_GET(block->bb_numrecs, ARCH_CONVERT)) + if (ptr < be16_to_cpu(block->bb_numrecs)) xfs_btree_check_key(cur->bc_btnum, kp + ptr - 1, kp + ptr); #endif @@ -773,16 +771,16 @@ xfs_alloc_insrec( */ rp = XFS_ALLOC_REC_ADDR(block, 1, cur); memmove(&rp[ptr], &rp[ptr - 1], - (INT_GET(block->bb_numrecs, ARCH_CONVERT) - ptr + 1) * sizeof(*rp)); + (be16_to_cpu(block->bb_numrecs) - ptr + 1) * sizeof(*rp)); /* * Now stuff the new record in, bump numrecs * and log the new data. */ rp[ptr - 1] = *recp; /* INT_: struct copy */ - INT_MOD(block->bb_numrecs, ARCH_CONVERT, +1); - xfs_alloc_log_recs(cur, bp, ptr, INT_GET(block->bb_numrecs, ARCH_CONVERT)); + be16_add(&block->bb_numrecs, 1); + xfs_alloc_log_recs(cur, bp, ptr, be16_to_cpu(block->bb_numrecs)); #ifdef DEBUG - if (ptr < INT_GET(block->bb_numrecs, ARCH_CONVERT)) + if (ptr < be16_to_cpu(block->bb_numrecs)) xfs_btree_check_rec(cur->bc_btnum, rp + ptr - 1, rp + ptr); #endif @@ -804,16 +802,16 @@ xfs_alloc_insrec( agf = XFS_BUF_TO_AGF(cur->bc_private.a.agbp); if (level == 0 && cur->bc_btnum == XFS_BTNUM_CNT && - INT_GET(block->bb_rightsib, ARCH_CONVERT) == NULLAGBLOCK && - INT_GET(recp->ar_blockcount, ARCH_CONVERT) > INT_GET(agf->agf_longest, ARCH_CONVERT)) { + be32_to_cpu(block->bb_rightsib) == NULLAGBLOCK && + be32_to_cpu(recp->ar_blockcount) > be32_to_cpu(agf->agf_longest)) { /* * If this is a leaf in the by-size btree and there * is no right sibling block and this block is bigger * than the previous longest block, update it. */ - INT_COPY(agf->agf_longest, recp->ar_blockcount, ARCH_CONVERT); - cur->bc_mp->m_perag[INT_GET(agf->agf_seqno, ARCH_CONVERT)].pagf_longest - = INT_GET(recp->ar_blockcount, ARCH_CONVERT); + agf->agf_longest = recp->ar_blockcount; + cur->bc_mp->m_perag[be32_to_cpu(agf->agf_seqno)].pagf_longest + = be32_to_cpu(recp->ar_blockcount); xfs_alloc_log_agf(cur->bc_tp, cur->bc_private.a.agbp, XFS_AGF_LONGEST); } @@ -923,8 +921,9 @@ xfs_alloc_log_recs( agf = XFS_BUF_TO_AGF(cur->bc_private.a.agbp); for (p = &rp[rfirst - 1]; p <= &rp[rlast - 1]; p++) - ASSERT(INT_GET(p->ar_startblock, ARCH_CONVERT) + INT_GET(p->ar_blockcount, ARCH_CONVERT) <= - INT_GET(agf->agf_length, ARCH_CONVERT)); + ASSERT(be32_to_cpu(p->ar_startblock) + + be32_to_cpu(p->ar_blockcount) <= + be32_to_cpu(agf->agf_length)); } #endif first = (int)((xfs_caddr_t)&rp[rfirst - 1] - (xfs_caddr_t)block); @@ -961,8 +960,8 @@ xfs_alloc_lookup( xfs_agf_t *agf; /* a.g. freespace header */ agf = XFS_BUF_TO_AGF(cur->bc_private.a.agbp); - agno = INT_GET(agf->agf_seqno, ARCH_CONVERT); - agbno = INT_GET(agf->agf_roots[cur->bc_btnum], ARCH_CONVERT); + agno = be32_to_cpu(agf->agf_seqno); + agbno = be32_to_cpu(agf->agf_roots[cur->bc_btnum]); } /* * Iterate over each level in the btree, starting at the root. @@ -1029,7 +1028,7 @@ xfs_alloc_lookup( * Set low and high entry numbers, 1-based. */ low = 1; - if (!(high = INT_GET(block->bb_numrecs, ARCH_CONVERT))) { + if (!(high = be16_to_cpu(block->bb_numrecs))) { /* * If the block is empty, the tree must * be an empty leaf. @@ -1058,14 +1057,14 @@ xfs_alloc_lookup( xfs_alloc_key_t *kkp; kkp = kkbase + keyno - 1; - startblock = INT_GET(kkp->ar_startblock, ARCH_CONVERT); - blockcount = INT_GET(kkp->ar_blockcount, ARCH_CONVERT); + startblock = be32_to_cpu(kkp->ar_startblock); + blockcount = be32_to_cpu(kkp->ar_blockcount); } else { xfs_alloc_rec_t *krp; krp = krbase + keyno - 1; - startblock = INT_GET(krp->ar_startblock, ARCH_CONVERT); - blockcount = INT_GET(krp->ar_blockcount, ARCH_CONVERT); + startblock = be32_to_cpu(krp->ar_startblock); + blockcount = be32_to_cpu(krp->ar_blockcount); } /* * Compute difference to get next direction. @@ -1105,7 +1104,7 @@ xfs_alloc_lookup( */ if (diff > 0 && --keyno < 1) keyno = 1; - agbno = INT_GET(*XFS_ALLOC_PTR_ADDR(block, keyno, cur), ARCH_CONVERT); + agbno = be32_to_cpu(*XFS_ALLOC_PTR_ADDR(block, keyno, cur)); #ifdef DEBUG if ((error = xfs_btree_check_sptr(cur, agbno, level))) return error; @@ -1124,8 +1123,8 @@ xfs_alloc_lookup( * not the last block, we're in the wrong block. */ if (dir == XFS_LOOKUP_GE && - keyno > INT_GET(block->bb_numrecs, ARCH_CONVERT) && - INT_GET(block->bb_rightsib, ARCH_CONVERT) != NULLAGBLOCK) { + keyno > be16_to_cpu(block->bb_numrecs) && + be32_to_cpu(block->bb_rightsib) != NULLAGBLOCK) { int i; cur->bc_ptrs[0] = keyno; @@ -1142,7 +1141,7 @@ xfs_alloc_lookup( /* * Return if we succeeded or not. */ - if (keyno == 0 || keyno > INT_GET(block->bb_numrecs, ARCH_CONVERT)) + if (keyno == 0 || keyno > be16_to_cpu(block->bb_numrecs)) *stat = 0; else *stat = ((dir != XFS_LOOKUP_EQ) || (diff == 0)); @@ -1185,7 +1184,7 @@ xfs_alloc_lshift( /* * If we've got no left sibling then we can't shift an entry left. */ - if (INT_GET(right->bb_leftsib, ARCH_CONVERT) == NULLAGBLOCK) { + if (be32_to_cpu(right->bb_leftsib) == NULLAGBLOCK) { *stat = 0; return 0; } @@ -1201,8 +1200,8 @@ xfs_alloc_lshift( * Set up the left neighbor as "left". */ if ((error = xfs_btree_read_bufs(cur->bc_mp, cur->bc_tp, - cur->bc_private.a.agno, INT_GET(right->bb_leftsib, ARCH_CONVERT), 0, &lbp, - XFS_ALLOC_BTREE_REF))) + cur->bc_private.a.agno, be32_to_cpu(right->bb_leftsib), + 0, &lbp, XFS_ALLOC_BTREE_REF))) return error; left = XFS_BUF_TO_ALLOC_BLOCK(lbp); if ((error = xfs_btree_check_sblock(cur, left, level, lbp))) @@ -1210,11 +1209,11 @@ xfs_alloc_lshift( /* * If it's full, it can't take another entry. */ - if (INT_GET(left->bb_numrecs, ARCH_CONVERT) == XFS_ALLOC_BLOCK_MAXRECS(level, cur)) { + if (be16_to_cpu(left->bb_numrecs) == XFS_ALLOC_BLOCK_MAXRECS(level, cur)) { *stat = 0; return 0; } - nrec = INT_GET(left->bb_numrecs, ARCH_CONVERT) + 1; + nrec = be16_to_cpu(left->bb_numrecs) + 1; /* * If non-leaf, copy a key and a ptr to the left block. */ @@ -1229,7 +1228,7 @@ xfs_alloc_lshift( lpp = XFS_ALLOC_PTR_ADDR(left, nrec, cur); rpp = XFS_ALLOC_PTR_ADDR(right, 1, cur); #ifdef DEBUG - if ((error = xfs_btree_check_sptr(cur, INT_GET(*rpp, ARCH_CONVERT), level))) + if ((error = xfs_btree_check_sptr(cur, be32_to_cpu(*rpp), level))) return error; #endif *lpp = *rpp; /* INT_: copy */ @@ -1251,30 +1250,30 @@ xfs_alloc_lshift( /* * Bump and log left's numrecs, decrement and log right's numrecs. */ - INT_MOD(left->bb_numrecs, ARCH_CONVERT, +1); + be16_add(&left->bb_numrecs, 1); xfs_alloc_log_block(cur->bc_tp, lbp, XFS_BB_NUMRECS); - INT_MOD(right->bb_numrecs, ARCH_CONVERT, -1); + be16_add(&right->bb_numrecs, -1); xfs_alloc_log_block(cur->bc_tp, rbp, XFS_BB_NUMRECS); /* * Slide the contents of right down one entry. */ if (level > 0) { #ifdef DEBUG - for (i = 0; i < INT_GET(right->bb_numrecs, ARCH_CONVERT); i++) { - if ((error = xfs_btree_check_sptr(cur, INT_GET(rpp[i + 1], ARCH_CONVERT), + for (i = 0; i < be16_to_cpu(right->bb_numrecs); i++) { + if ((error = xfs_btree_check_sptr(cur, be32_to_cpu(rpp[i + 1]), level))) return error; } #endif - memmove(rkp, rkp + 1, INT_GET(right->bb_numrecs, ARCH_CONVERT) * sizeof(*rkp)); - memmove(rpp, rpp + 1, INT_GET(right->bb_numrecs, ARCH_CONVERT) * sizeof(*rpp)); - xfs_alloc_log_keys(cur, rbp, 1, INT_GET(right->bb_numrecs, ARCH_CONVERT)); - xfs_alloc_log_ptrs(cur, rbp, 1, INT_GET(right->bb_numrecs, ARCH_CONVERT)); + memmove(rkp, rkp + 1, be16_to_cpu(right->bb_numrecs) * sizeof(*rkp)); + memmove(rpp, rpp + 1, be16_to_cpu(right->bb_numrecs) * sizeof(*rpp)); + xfs_alloc_log_keys(cur, rbp, 1, be16_to_cpu(right->bb_numrecs)); + xfs_alloc_log_ptrs(cur, rbp, 1, be16_to_cpu(right->bb_numrecs)); } else { - memmove(rrp, rrp + 1, INT_GET(right->bb_numrecs, ARCH_CONVERT) * sizeof(*rrp)); - xfs_alloc_log_recs(cur, rbp, 1, INT_GET(right->bb_numrecs, ARCH_CONVERT)); - key.ar_startblock = rrp->ar_startblock; /* INT_: direct copy */ - key.ar_blockcount = rrp->ar_blockcount; /* INT_: direct copy */ + memmove(rrp, rrp + 1, be16_to_cpu(right->bb_numrecs) * sizeof(*rrp)); + xfs_alloc_log_recs(cur, rbp, 1, be16_to_cpu(right->bb_numrecs)); + key.ar_startblock = rrp->ar_startblock; + key.ar_blockcount = rrp->ar_blockcount; rkp = &key; } /* @@ -1339,9 +1338,9 @@ xfs_alloc_newroot( xfs_agnumber_t seqno; agf = XFS_BUF_TO_AGF(cur->bc_private.a.agbp); - INT_SET(agf->agf_roots[cur->bc_btnum], ARCH_CONVERT, nbno); - INT_MOD(agf->agf_levels[cur->bc_btnum], ARCH_CONVERT, 1); - seqno = INT_GET(agf->agf_seqno, ARCH_CONVERT); + agf->agf_roots[cur->bc_btnum] = cpu_to_be32(nbno); + be32_add(&agf->agf_levels[cur->bc_btnum], 1); + seqno = be32_to_cpu(agf->agf_seqno); mp->m_perag[seqno].pagf_levels[cur->bc_btnum]++; xfs_alloc_log_agf(cur->bc_tp, cur->bc_private.a.agbp, XFS_AGF_ROOTS | XFS_AGF_LEVELS); @@ -1358,12 +1357,12 @@ xfs_alloc_newroot( if ((error = xfs_btree_check_sblock(cur, left, cur->bc_nlevels - 1, lbp))) return error; #endif - if (INT_GET(left->bb_rightsib, ARCH_CONVERT) != NULLAGBLOCK) { + if (be32_to_cpu(left->bb_rightsib) != NULLAGBLOCK) { /* * Our block is left, pick up the right block. */ lbno = XFS_DADDR_TO_AGBNO(mp, XFS_BUF_ADDR(lbp)); - rbno = INT_GET(left->bb_rightsib, ARCH_CONVERT); + rbno = be32_to_cpu(left->bb_rightsib); if ((error = xfs_btree_read_bufs(mp, cur->bc_tp, cur->bc_private.a.agno, rbno, 0, &rbp, XFS_ALLOC_BTREE_REF))) @@ -1380,7 +1379,7 @@ xfs_alloc_newroot( rbp = lbp; right = left; rbno = XFS_DADDR_TO_AGBNO(mp, XFS_BUF_ADDR(rbp)); - lbno = INT_GET(right->bb_leftsib, ARCH_CONVERT); + lbno = be32_to_cpu(right->bb_leftsib); if ((error = xfs_btree_read_bufs(mp, cur->bc_tp, cur->bc_private.a.agno, lbno, 0, &lbp, XFS_ALLOC_BTREE_REF))) @@ -1394,11 +1393,11 @@ xfs_alloc_newroot( /* * Fill in the new block's btree header and log it. */ - INT_SET(new->bb_magic, ARCH_CONVERT, xfs_magics[cur->bc_btnum]); - INT_SET(new->bb_level, ARCH_CONVERT, (__uint16_t)cur->bc_nlevels); - INT_SET(new->bb_numrecs, ARCH_CONVERT, 2); - INT_SET(new->bb_leftsib, ARCH_CONVERT, NULLAGBLOCK); - INT_SET(new->bb_rightsib, ARCH_CONVERT, NULLAGBLOCK); + new->bb_magic = cpu_to_be32(xfs_magics[cur->bc_btnum]); + new->bb_level = cpu_to_be16(cur->bc_nlevels); + new->bb_numrecs = cpu_to_be16(2); + new->bb_leftsib = cpu_to_be32(NULLAGBLOCK); + new->bb_rightsib = cpu_to_be32(NULLAGBLOCK); xfs_alloc_log_block(cur->bc_tp, nbp, XFS_BB_ALL_BITS); ASSERT(lbno != NULLAGBLOCK && rbno != NULLAGBLOCK); /* @@ -1408,18 +1407,18 @@ xfs_alloc_newroot( xfs_alloc_key_t *kp; /* btree key pointer */ kp = XFS_ALLOC_KEY_ADDR(new, 1, cur); - if (INT_GET(left->bb_level, ARCH_CONVERT) > 0) { + if (be16_to_cpu(left->bb_level) > 0) { kp[0] = *XFS_ALLOC_KEY_ADDR(left, 1, cur); /* INT_: structure copy */ kp[1] = *XFS_ALLOC_KEY_ADDR(right, 1, cur);/* INT_: structure copy */ } else { xfs_alloc_rec_t *rp; /* btree record pointer */ rp = XFS_ALLOC_REC_ADDR(left, 1, cur); - kp[0].ar_startblock = rp->ar_startblock; /* INT_: direct copy */ - kp[0].ar_blockcount = rp->ar_blockcount; /* INT_: direct copy */ + kp[0].ar_startblock = rp->ar_startblock; + kp[0].ar_blockcount = rp->ar_blockcount; rp = XFS_ALLOC_REC_ADDR(right, 1, cur); - kp[1].ar_startblock = rp->ar_startblock; /* INT_: direct copy */ - kp[1].ar_blockcount = rp->ar_blockcount; /* INT_: direct copy */ + kp[1].ar_startblock = rp->ar_startblock; + kp[1].ar_blockcount = rp->ar_blockcount; } } xfs_alloc_log_keys(cur, nbp, 1, 2); @@ -1430,8 +1429,8 @@ xfs_alloc_newroot( xfs_alloc_ptr_t *pp; /* btree address pointer */ pp = XFS_ALLOC_PTR_ADDR(new, 1, cur); - INT_SET(pp[0], ARCH_CONVERT, lbno); - INT_SET(pp[1], ARCH_CONVERT, rbno); + pp[0] = cpu_to_be32(lbno); + pp[1] = cpu_to_be32(rbno); } xfs_alloc_log_ptrs(cur, nbp, 1, 2); /* @@ -1476,7 +1475,7 @@ xfs_alloc_rshift( /* * If we've got no right sibling then we can't shift an entry right. */ - if (INT_GET(left->bb_rightsib, ARCH_CONVERT) == NULLAGBLOCK) { + if (be32_to_cpu(left->bb_rightsib) == NULLAGBLOCK) { *stat = 0; return 0; } @@ -1484,7 +1483,7 @@ xfs_alloc_rshift( * If the cursor entry is the one that would be moved, don't * do it... it's too complicated. */ - if (cur->bc_ptrs[level] >= INT_GET(left->bb_numrecs, ARCH_CONVERT)) { + if (cur->bc_ptrs[level] >= be16_to_cpu(left->bb_numrecs)) { *stat = 0; return 0; } @@ -1492,8 +1491,8 @@ xfs_alloc_rshift( * Set up the right neighbor as "right". */ if ((error = xfs_btree_read_bufs(cur->bc_mp, cur->bc_tp, - cur->bc_private.a.agno, INT_GET(left->bb_rightsib, ARCH_CONVERT), 0, &rbp, - XFS_ALLOC_BTREE_REF))) + cur->bc_private.a.agno, be32_to_cpu(left->bb_rightsib), + 0, &rbp, XFS_ALLOC_BTREE_REF))) return error; right = XFS_BUF_TO_ALLOC_BLOCK(rbp); if ((error = xfs_btree_check_sblock(cur, right, level, rbp))) @@ -1501,7 +1500,7 @@ xfs_alloc_rshift( /* * If it's full, it can't take another entry. */ - if (INT_GET(right->bb_numrecs, ARCH_CONVERT) == XFS_ALLOC_BLOCK_MAXRECS(level, cur)) { + if (be16_to_cpu(right->bb_numrecs) == XFS_ALLOC_BLOCK_MAXRECS(level, cur)) { *stat = 0; return 0; } @@ -1514,47 +1513,47 @@ xfs_alloc_rshift( xfs_alloc_ptr_t *lpp; /* address pointer for left block */ xfs_alloc_ptr_t *rpp; /* address pointer for right block */ - lkp = XFS_ALLOC_KEY_ADDR(left, INT_GET(left->bb_numrecs, ARCH_CONVERT), cur); - lpp = XFS_ALLOC_PTR_ADDR(left, INT_GET(left->bb_numrecs, ARCH_CONVERT), cur); + lkp = XFS_ALLOC_KEY_ADDR(left, be16_to_cpu(left->bb_numrecs), cur); + lpp = XFS_ALLOC_PTR_ADDR(left, be16_to_cpu(left->bb_numrecs), cur); rkp = XFS_ALLOC_KEY_ADDR(right, 1, cur); rpp = XFS_ALLOC_PTR_ADDR(right, 1, cur); #ifdef DEBUG - for (i = INT_GET(right->bb_numrecs, ARCH_CONVERT) - 1; i >= 0; i--) { - if ((error = xfs_btree_check_sptr(cur, INT_GET(rpp[i], ARCH_CONVERT), level))) + for (i = be16_to_cpu(right->bb_numrecs) - 1; i >= 0; i--) { + if ((error = xfs_btree_check_sptr(cur, be32_to_cpu(rpp[i]), level))) return error; } #endif - memmove(rkp + 1, rkp, INT_GET(right->bb_numrecs, ARCH_CONVERT) * sizeof(*rkp)); - memmove(rpp + 1, rpp, INT_GET(right->bb_numrecs, ARCH_CONVERT) * sizeof(*rpp)); + memmove(rkp + 1, rkp, be16_to_cpu(right->bb_numrecs) * sizeof(*rkp)); + memmove(rpp + 1, rpp, be16_to_cpu(right->bb_numrecs) * sizeof(*rpp)); #ifdef DEBUG - if ((error = xfs_btree_check_sptr(cur, INT_GET(*lpp, ARCH_CONVERT), level))) + if ((error = xfs_btree_check_sptr(cur, be32_to_cpu(*lpp), level))) return error; #endif *rkp = *lkp; /* INT_: copy */ *rpp = *lpp; /* INT_: copy */ - xfs_alloc_log_keys(cur, rbp, 1, INT_GET(right->bb_numrecs, ARCH_CONVERT) + 1); - xfs_alloc_log_ptrs(cur, rbp, 1, INT_GET(right->bb_numrecs, ARCH_CONVERT) + 1); + xfs_alloc_log_keys(cur, rbp, 1, be16_to_cpu(right->bb_numrecs) + 1); + xfs_alloc_log_ptrs(cur, rbp, 1, be16_to_cpu(right->bb_numrecs) + 1); xfs_btree_check_key(cur->bc_btnum, rkp, rkp + 1); } else { xfs_alloc_rec_t *lrp; /* record pointer for left block */ xfs_alloc_rec_t *rrp; /* record pointer for right block */ - lrp = XFS_ALLOC_REC_ADDR(left, INT_GET(left->bb_numrecs, ARCH_CONVERT), cur); + lrp = XFS_ALLOC_REC_ADDR(left, be16_to_cpu(left->bb_numrecs), cur); rrp = XFS_ALLOC_REC_ADDR(right, 1, cur); - memmove(rrp + 1, rrp, INT_GET(right->bb_numrecs, ARCH_CONVERT) * sizeof(*rrp)); + memmove(rrp + 1, rrp, be16_to_cpu(right->bb_numrecs) * sizeof(*rrp)); *rrp = *lrp; - xfs_alloc_log_recs(cur, rbp, 1, INT_GET(right->bb_numrecs, ARCH_CONVERT) + 1); - key.ar_startblock = rrp->ar_startblock; /* INT_: direct copy */ - key.ar_blockcount = rrp->ar_blockcount; /* INT_: direct copy */ + xfs_alloc_log_recs(cur, rbp, 1, be16_to_cpu(right->bb_numrecs) + 1); + key.ar_startblock = rrp->ar_startblock; + key.ar_blockcount = rrp->ar_blockcount; rkp = &key; xfs_btree_check_rec(cur->bc_btnum, rrp, rrp + 1); } /* * Decrement and log left's numrecs, bump and log right's numrecs. */ - INT_MOD(left->bb_numrecs, ARCH_CONVERT, -1); + be16_add(&left->bb_numrecs, -1); xfs_alloc_log_block(cur->bc_tp, lbp, XFS_BB_NUMRECS); - INT_MOD(right->bb_numrecs, ARCH_CONVERT, +1); + be16_add(&right->bb_numrecs, 1); xfs_alloc_log_block(cur->bc_tp, rbp, XFS_BB_NUMRECS); /* * Using a temporary cursor, update the parent key values of the @@ -1627,17 +1626,17 @@ xfs_alloc_split( /* * Fill in the btree header for the new block. */ - INT_SET(right->bb_magic, ARCH_CONVERT, xfs_magics[cur->bc_btnum]); - right->bb_level = left->bb_level; /* INT_: direct copy */ - INT_SET(right->bb_numrecs, ARCH_CONVERT, (__uint16_t)(INT_GET(left->bb_numrecs, ARCH_CONVERT) / 2)); + right->bb_magic = cpu_to_be32(xfs_magics[cur->bc_btnum]); + right->bb_level = left->bb_level; + right->bb_numrecs = cpu_to_be16(be16_to_cpu(left->bb_numrecs) / 2); /* * Make sure that if there's an odd number of entries now, that * each new block will have the same number of entries. */ - if ((INT_GET(left->bb_numrecs, ARCH_CONVERT) & 1) && - cur->bc_ptrs[level] <= INT_GET(right->bb_numrecs, ARCH_CONVERT) + 1) - INT_MOD(right->bb_numrecs, ARCH_CONVERT, +1); - i = INT_GET(left->bb_numrecs, ARCH_CONVERT) - INT_GET(right->bb_numrecs, ARCH_CONVERT) + 1; + if ((be16_to_cpu(left->bb_numrecs) & 1) && + cur->bc_ptrs[level] <= be16_to_cpu(right->bb_numrecs) + 1) + be16_add(&right->bb_numrecs, 1); + i = be16_to_cpu(left->bb_numrecs) - be16_to_cpu(right->bb_numrecs) + 1; /* * For non-leaf blocks, copy keys and addresses over to the new block. */ @@ -1652,15 +1651,15 @@ xfs_alloc_split( rkp = XFS_ALLOC_KEY_ADDR(right, 1, cur); rpp = XFS_ALLOC_PTR_ADDR(right, 1, cur); #ifdef DEBUG - for (i = 0; i < INT_GET(right->bb_numrecs, ARCH_CONVERT); i++) { - if ((error = xfs_btree_check_sptr(cur, INT_GET(lpp[i], ARCH_CONVERT), level))) + for (i = 0; i < be16_to_cpu(right->bb_numrecs); i++) { + if ((error = xfs_btree_check_sptr(cur, be32_to_cpu(lpp[i]), level))) return error; } #endif - memcpy(rkp, lkp, INT_GET(right->bb_numrecs, ARCH_CONVERT) * sizeof(*rkp)); /* INT_: copy */ - memcpy(rpp, lpp, INT_GET(right->bb_numrecs, ARCH_CONVERT) * sizeof(*rpp)); /* INT_: copy */ - xfs_alloc_log_keys(cur, rbp, 1, INT_GET(right->bb_numrecs, ARCH_CONVERT)); - xfs_alloc_log_ptrs(cur, rbp, 1, INT_GET(right->bb_numrecs, ARCH_CONVERT)); + memcpy(rkp, lkp, be16_to_cpu(right->bb_numrecs) * sizeof(*rkp)); + memcpy(rpp, lpp, be16_to_cpu(right->bb_numrecs) * sizeof(*rpp)); + xfs_alloc_log_keys(cur, rbp, 1, be16_to_cpu(right->bb_numrecs)); + xfs_alloc_log_ptrs(cur, rbp, 1, be16_to_cpu(right->bb_numrecs)); *keyp = *rkp; } /* @@ -1672,38 +1671,38 @@ xfs_alloc_split( lrp = XFS_ALLOC_REC_ADDR(left, i, cur); rrp = XFS_ALLOC_REC_ADDR(right, 1, cur); - memcpy(rrp, lrp, INT_GET(right->bb_numrecs, ARCH_CONVERT) * sizeof(*rrp)); - xfs_alloc_log_recs(cur, rbp, 1, INT_GET(right->bb_numrecs, ARCH_CONVERT)); - keyp->ar_startblock = rrp->ar_startblock; /* INT_: direct copy */ - keyp->ar_blockcount = rrp->ar_blockcount; /* INT_: direct copy */ + memcpy(rrp, lrp, be16_to_cpu(right->bb_numrecs) * sizeof(*rrp)); + xfs_alloc_log_recs(cur, rbp, 1, be16_to_cpu(right->bb_numrecs)); + keyp->ar_startblock = rrp->ar_startblock; + keyp->ar_blockcount = rrp->ar_blockcount; } /* * Find the left block number by looking in the buffer. * Adjust numrecs, sibling pointers. */ lbno = XFS_DADDR_TO_AGBNO(cur->bc_mp, XFS_BUF_ADDR(lbp)); - INT_MOD(left->bb_numrecs, ARCH_CONVERT, -(INT_GET(right->bb_numrecs, ARCH_CONVERT))); - right->bb_rightsib = left->bb_rightsib; /* INT_: direct copy */ - INT_SET(left->bb_rightsib, ARCH_CONVERT, rbno); - INT_SET(right->bb_leftsib, ARCH_CONVERT, lbno); + be16_add(&left->bb_numrecs, -(be16_to_cpu(right->bb_numrecs))); + right->bb_rightsib = left->bb_rightsib; + left->bb_rightsib = cpu_to_be32(rbno); + right->bb_leftsib = cpu_to_be32(lbno); xfs_alloc_log_block(cur->bc_tp, rbp, XFS_BB_ALL_BITS); xfs_alloc_log_block(cur->bc_tp, lbp, XFS_BB_NUMRECS | XFS_BB_RIGHTSIB); /* * If there's a block to the new block's right, make that block * point back to right instead of to left. */ - if (INT_GET(right->bb_rightsib, ARCH_CONVERT) != NULLAGBLOCK) { + if (be32_to_cpu(right->bb_rightsib) != NULLAGBLOCK) { xfs_alloc_block_t *rrblock; /* rr btree block */ xfs_buf_t *rrbp; /* buffer for rrblock */ if ((error = xfs_btree_read_bufs(cur->bc_mp, cur->bc_tp, - cur->bc_private.a.agno, INT_GET(right->bb_rightsib, ARCH_CONVERT), 0, + cur->bc_private.a.agno, be32_to_cpu(right->bb_rightsib), 0, &rrbp, XFS_ALLOC_BTREE_REF))) return error; rrblock = XFS_BUF_TO_ALLOC_BLOCK(rrbp); if ((error = xfs_btree_check_sblock(cur, rrblock, level, rrbp))) return error; - INT_SET(rrblock->bb_leftsib, ARCH_CONVERT, rbno); + rrblock->bb_leftsib = cpu_to_be32(rbno); xfs_alloc_log_block(cur->bc_tp, rrbp, XFS_BB_LEFTSIB); } /* @@ -1711,9 +1710,9 @@ xfs_alloc_split( * If it's just pointing past the last entry in left, then we'll * insert there, so don't change anything in that case. */ - if (cur->bc_ptrs[level] > INT_GET(left->bb_numrecs, ARCH_CONVERT) + 1) { + if (cur->bc_ptrs[level] > be16_to_cpu(left->bb_numrecs) + 1) { xfs_btree_setbuf(cur, level, rbp); - cur->bc_ptrs[level] -= INT_GET(left->bb_numrecs, ARCH_CONVERT); + cur->bc_ptrs[level] -= be16_to_cpu(left->bb_numrecs); } /* * If there are more levels, we'll need another cursor which refers to @@ -1811,7 +1810,7 @@ xfs_alloc_decrement( /* * If we just went off the left edge of the tree, return failure. */ - if (INT_GET(block->bb_leftsib, ARCH_CONVERT) == NULLAGBLOCK) { + if (be32_to_cpu(block->bb_leftsib) == NULLAGBLOCK) { *stat = 0; return 0; } @@ -1840,7 +1839,7 @@ xfs_alloc_decrement( xfs_agblock_t agbno; /* block number of btree block */ xfs_buf_t *bp; /* buffer pointer for block */ - agbno = INT_GET(*XFS_ALLOC_PTR_ADDR(block, cur->bc_ptrs[lev], cur), ARCH_CONVERT); + agbno = be32_to_cpu(*XFS_ALLOC_PTR_ADDR(block, cur->bc_ptrs[lev], cur)); if ((error = xfs_btree_read_bufs(cur->bc_mp, cur->bc_tp, cur->bc_private.a.agno, agbno, 0, &bp, XFS_ALLOC_BTREE_REF))) @@ -1850,7 +1849,7 @@ xfs_alloc_decrement( block = XFS_BUF_TO_ALLOC_BLOCK(bp); if ((error = xfs_btree_check_sblock(cur, block, lev, bp))) return error; - cur->bc_ptrs[lev] = INT_GET(block->bb_numrecs, ARCH_CONVERT); + cur->bc_ptrs[lev] = be16_to_cpu(block->bb_numrecs); } *stat = 1; return 0; @@ -1917,7 +1916,7 @@ xfs_alloc_get_rec( /* * Off the right end or left end, return failure. */ - if (ptr > INT_GET(block->bb_numrecs, ARCH_CONVERT) || ptr <= 0) { + if (ptr > be16_to_cpu(block->bb_numrecs) || ptr <= 0) { *stat = 0; return 0; } @@ -1928,8 +1927,8 @@ xfs_alloc_get_rec( xfs_alloc_rec_t *rec; /* record data */ rec = XFS_ALLOC_REC_ADDR(block, ptr, cur); - *bno = INT_GET(rec->ar_startblock, ARCH_CONVERT); - *len = INT_GET(rec->ar_blockcount, ARCH_CONVERT); + *bno = be32_to_cpu(rec->ar_startblock); + *len = be32_to_cpu(rec->ar_blockcount); } *stat = 1; return 0; @@ -1968,14 +1967,14 @@ xfs_alloc_increment( * Increment the ptr at this level. If we're still in the block * then we're done. */ - if (++cur->bc_ptrs[level] <= INT_GET(block->bb_numrecs, ARCH_CONVERT)) { + if (++cur->bc_ptrs[level] <= be16_to_cpu(block->bb_numrecs)) { *stat = 1; return 0; } /* * If we just went off the right edge of the tree, return failure. */ - if (INT_GET(block->bb_rightsib, ARCH_CONVERT) == NULLAGBLOCK) { + if (be32_to_cpu(block->bb_rightsib) == NULLAGBLOCK) { *stat = 0; return 0; } @@ -1990,7 +1989,7 @@ xfs_alloc_increment( if ((error = xfs_btree_check_sblock(cur, block, lev, bp))) return error; #endif - if (++cur->bc_ptrs[lev] <= INT_GET(block->bb_numrecs, ARCH_CONVERT)) + if (++cur->bc_ptrs[lev] <= be16_to_cpu(block->bb_numrecs)) break; /* * Read-ahead the right block, we're going to read it @@ -2010,7 +2009,7 @@ xfs_alloc_increment( lev > level; ) { xfs_agblock_t agbno; /* block number of btree block */ - agbno = INT_GET(*XFS_ALLOC_PTR_ADDR(block, cur->bc_ptrs[lev], cur), ARCH_CONVERT); + agbno = be32_to_cpu(*XFS_ALLOC_PTR_ADDR(block, cur->bc_ptrs[lev], cur)); if ((error = xfs_btree_read_bufs(cur->bc_mp, cur->bc_tp, cur->bc_private.a.agno, agbno, 0, &bp, XFS_ALLOC_BTREE_REF))) @@ -2045,8 +2044,8 @@ xfs_alloc_insert( level = 0; nbno = NULLAGBLOCK; - INT_SET(nrec.ar_startblock, ARCH_CONVERT, cur->bc_rec.a.ar_startblock); - INT_SET(nrec.ar_blockcount, ARCH_CONVERT, cur->bc_rec.a.ar_blockcount); + nrec.ar_startblock = cpu_to_be32(cur->bc_rec.a.ar_startblock); + nrec.ar_blockcount = cpu_to_be32(cur->bc_rec.a.ar_blockcount); ncur = (xfs_btree_cur_t *)0; pcur = cur; /* @@ -2167,8 +2166,8 @@ xfs_alloc_update( /* * Fill in the new contents and log them. */ - INT_SET(rp->ar_startblock, ARCH_CONVERT, bno); - INT_SET(rp->ar_blockcount, ARCH_CONVERT, len); + rp->ar_startblock = cpu_to_be32(bno); + rp->ar_blockcount = cpu_to_be32(len); xfs_alloc_log_recs(cur, cur->bc_bufs[0], ptr, ptr); } /* @@ -2177,15 +2176,15 @@ xfs_alloc_update( * extent in the a.g., which we cache in the a.g. freelist header. */ if (cur->bc_btnum == XFS_BTNUM_CNT && - INT_GET(block->bb_rightsib, ARCH_CONVERT) == NULLAGBLOCK && - ptr == INT_GET(block->bb_numrecs, ARCH_CONVERT)) { + be32_to_cpu(block->bb_rightsib) == NULLAGBLOCK && + ptr == be16_to_cpu(block->bb_numrecs)) { xfs_agf_t *agf; /* a.g. freespace header */ xfs_agnumber_t seqno; agf = XFS_BUF_TO_AGF(cur->bc_private.a.agbp); - seqno = INT_GET(agf->agf_seqno, ARCH_CONVERT); + seqno = be32_to_cpu(agf->agf_seqno); cur->bc_mp->m_perag[seqno].pagf_longest = len; - INT_SET(agf->agf_longest, ARCH_CONVERT, len); + agf->agf_longest = cpu_to_be32(len); xfs_alloc_log_agf(cur->bc_tp, cur->bc_private.a.agbp, XFS_AGF_LONGEST); } @@ -2195,8 +2194,8 @@ xfs_alloc_update( if (ptr == 1) { xfs_alloc_key_t key; /* key containing [bno, len] */ - INT_SET(key.ar_startblock, ARCH_CONVERT, bno); - INT_SET(key.ar_blockcount, ARCH_CONVERT, len); + key.ar_startblock = cpu_to_be32(bno); + key.ar_blockcount = cpu_to_be32(len); if ((error = xfs_alloc_updkey(cur, &key, 1))) return error; } diff --git a/fs/xfs/xfs_alloc_btree.h b/fs/xfs/xfs_alloc_btree.h index ed5161a572ef..bce81c7a4fdc 100644 --- a/fs/xfs/xfs_alloc_btree.h +++ b/fs/xfs/xfs_alloc_btree.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_ALLOC_BTREE_H__ #define __XFS_ALLOC_BTREE_H__ @@ -52,48 +38,29 @@ struct xfs_mount; /* * Data record/key structure */ -typedef struct xfs_alloc_rec -{ +typedef struct xfs_alloc_rec { + __be32 ar_startblock; /* starting block number */ + __be32 ar_blockcount; /* count of free blocks */ +} xfs_alloc_rec_t, xfs_alloc_key_t; + +typedef struct xfs_alloc_rec_incore { xfs_agblock_t ar_startblock; /* starting block number */ xfs_extlen_t ar_blockcount; /* count of free blocks */ -} xfs_alloc_rec_t, xfs_alloc_key_t; +} xfs_alloc_rec_incore_t; -typedef xfs_agblock_t xfs_alloc_ptr_t; /* btree pointer type */ - /* btree block header type */ +/* btree pointer type */ +typedef __be32 xfs_alloc_ptr_t; +/* btree block header type */ typedef struct xfs_btree_sblock xfs_alloc_block_t; -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BUF_TO_ALLOC_BLOCK) -xfs_alloc_block_t *xfs_buf_to_alloc_block(struct xfs_buf *bp); -#define XFS_BUF_TO_ALLOC_BLOCK(bp) xfs_buf_to_alloc_block(bp) -#else -#define XFS_BUF_TO_ALLOC_BLOCK(bp) ((xfs_alloc_block_t *)(XFS_BUF_PTR(bp))) -#endif +#define XFS_BUF_TO_ALLOC_BLOCK(bp) ((xfs_alloc_block_t *)XFS_BUF_PTR(bp)) /* * Real block structures have a size equal to the disk block size. */ - -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_ALLOC_BLOCK_SIZE) -int xfs_alloc_block_size(int lev, struct xfs_btree_cur *cur); -#define XFS_ALLOC_BLOCK_SIZE(lev,cur) xfs_alloc_block_size(lev,cur) -#else #define XFS_ALLOC_BLOCK_SIZE(lev,cur) (1 << (cur)->bc_blocklog) -#endif - -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_ALLOC_BLOCK_MAXRECS) -int xfs_alloc_block_maxrecs(int lev, struct xfs_btree_cur *cur); -#define XFS_ALLOC_BLOCK_MAXRECS(lev,cur) xfs_alloc_block_maxrecs(lev,cur) -#else -#define XFS_ALLOC_BLOCK_MAXRECS(lev,cur) \ - ((cur)->bc_mp->m_alloc_mxr[lev != 0]) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_ALLOC_BLOCK_MINRECS) -int xfs_alloc_block_minrecs(int lev, struct xfs_btree_cur *cur); -#define XFS_ALLOC_BLOCK_MINRECS(lev,cur) xfs_alloc_block_minrecs(lev,cur) -#else -#define XFS_ALLOC_BLOCK_MINRECS(lev,cur) \ - ((cur)->bc_mp->m_alloc_mnr[lev != 0]) -#endif +#define XFS_ALLOC_BLOCK_MAXRECS(lev,cur) ((cur)->bc_mp->m_alloc_mxr[lev != 0]) +#define XFS_ALLOC_BLOCK_MINRECS(lev,cur) ((cur)->bc_mp->m_alloc_mnr[lev != 0]) /* * Minimum and maximum blocksize and sectorsize. @@ -113,145 +80,80 @@ int xfs_alloc_block_minrecs(int lev, struct xfs_btree_cur *cur); * Block numbers in the AG: * SB is sector 0, AGF is sector 1, AGI is sector 2, AGFL is sector 3. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BNO_BLOCK) -xfs_agblock_t xfs_bno_block(struct xfs_mount *mp); -#define XFS_BNO_BLOCK(mp) xfs_bno_block(mp) -#else #define XFS_BNO_BLOCK(mp) ((xfs_agblock_t)(XFS_AGFL_BLOCK(mp) + 1)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_CNT_BLOCK) -xfs_agblock_t xfs_cnt_block(struct xfs_mount *mp); -#define XFS_CNT_BLOCK(mp) xfs_cnt_block(mp) -#else #define XFS_CNT_BLOCK(mp) ((xfs_agblock_t)(XFS_BNO_BLOCK(mp) + 1)) -#endif /* * Record, key, and pointer address macros for btree blocks. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_ALLOC_REC_ADDR) -xfs_alloc_rec_t *xfs_alloc_rec_addr(xfs_alloc_block_t *bb, int i, - struct xfs_btree_cur *cur); -#define XFS_ALLOC_REC_ADDR(bb,i,cur) xfs_alloc_rec_addr(bb,i,cur) -#else #define XFS_ALLOC_REC_ADDR(bb,i,cur) \ - XFS_BTREE_REC_ADDR(XFS_ALLOC_BLOCK_SIZE(0,cur), xfs_alloc, bb, i, \ - XFS_ALLOC_BLOCK_MAXRECS(0, cur)) -#endif + XFS_BTREE_REC_ADDR(XFS_ALLOC_BLOCK_SIZE(0,cur), xfs_alloc, \ + bb, i, XFS_ALLOC_BLOCK_MAXRECS(0, cur)) -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_ALLOC_KEY_ADDR) -xfs_alloc_key_t *xfs_alloc_key_addr(xfs_alloc_block_t *bb, int i, - struct xfs_btree_cur *cur); -#define XFS_ALLOC_KEY_ADDR(bb,i,cur) xfs_alloc_key_addr(bb,i,cur) -#else #define XFS_ALLOC_KEY_ADDR(bb,i,cur) \ - XFS_BTREE_KEY_ADDR(XFS_ALLOC_BLOCK_SIZE(1,cur), xfs_alloc, bb, i, \ - XFS_ALLOC_BLOCK_MAXRECS(1, cur)) -#endif + XFS_BTREE_KEY_ADDR(XFS_ALLOC_BLOCK_SIZE(1,cur), xfs_alloc, \ + bb, i, XFS_ALLOC_BLOCK_MAXRECS(1, cur)) -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_ALLOC_PTR_ADDR) -xfs_alloc_ptr_t *xfs_alloc_ptr_addr(xfs_alloc_block_t *bb, int i, - struct xfs_btree_cur *cur); -#define XFS_ALLOC_PTR_ADDR(bb,i,cur) xfs_alloc_ptr_addr(bb,i,cur) -#else #define XFS_ALLOC_PTR_ADDR(bb,i,cur) \ - XFS_BTREE_PTR_ADDR(XFS_ALLOC_BLOCK_SIZE(1,cur), xfs_alloc, bb, i, \ - XFS_ALLOC_BLOCK_MAXRECS(1, cur)) -#endif - -/* - * Prototypes for externally visible routines. - */ + XFS_BTREE_PTR_ADDR(XFS_ALLOC_BLOCK_SIZE(1,cur), xfs_alloc, \ + bb, i, XFS_ALLOC_BLOCK_MAXRECS(1, cur)) /* * Decrement cursor by one record at the level. * For nonzero levels the leaf-ward information is untouched. */ -int /* error */ -xfs_alloc_decrement( - struct xfs_btree_cur *cur, /* btree cursor */ - int level, /* level in btree, 0 is leaf */ - int *stat); /* success/failure */ +extern int xfs_alloc_decrement(struct xfs_btree_cur *cur, int level, int *stat); /* * Delete the record pointed to by cur. * The cursor refers to the place where the record was (could be inserted) * when the operation returns. */ -int /* error */ -xfs_alloc_delete( - struct xfs_btree_cur *cur, /* btree cursor */ - int *stat); /* success/failure */ +extern int xfs_alloc_delete(struct xfs_btree_cur *cur, int *stat); /* * Get the data from the pointed-to record. */ -int /* error */ -xfs_alloc_get_rec( - struct xfs_btree_cur *cur, /* btree cursor */ - xfs_agblock_t *bno, /* output: starting block of extent */ - xfs_extlen_t *len, /* output: length of extent */ - int *stat); /* output: success/failure */ +extern int xfs_alloc_get_rec(struct xfs_btree_cur *cur, xfs_agblock_t *bno, + xfs_extlen_t *len, int *stat); /* * Increment cursor by one record at the level. * For nonzero levels the leaf-ward information is untouched. */ -int /* error */ -xfs_alloc_increment( - struct xfs_btree_cur *cur, /* btree cursor */ - int level, /* level in btree, 0 is leaf */ - int *stat); /* success/failure */ +extern int xfs_alloc_increment(struct xfs_btree_cur *cur, int level, int *stat); /* * Insert the current record at the point referenced by cur. * The cursor may be inconsistent on return if splits have been done. */ -int /* error */ -xfs_alloc_insert( - struct xfs_btree_cur *cur, /* btree cursor */ - int *stat); /* success/failure */ +extern int xfs_alloc_insert(struct xfs_btree_cur *cur, int *stat); /* * Lookup the record equal to [bno, len] in the btree given by cur. */ -int /* error */ -xfs_alloc_lookup_eq( - struct xfs_btree_cur *cur, /* btree cursor */ - xfs_agblock_t bno, /* starting block of extent */ - xfs_extlen_t len, /* length of extent */ - int *stat); /* success/failure */ +extern int xfs_alloc_lookup_eq(struct xfs_btree_cur *cur, xfs_agblock_t bno, + xfs_extlen_t len, int *stat); /* * Lookup the first record greater than or equal to [bno, len] * in the btree given by cur. */ -int /* error */ -xfs_alloc_lookup_ge( - struct xfs_btree_cur *cur, /* btree cursor */ - xfs_agblock_t bno, /* starting block of extent */ - xfs_extlen_t len, /* length of extent */ - int *stat); /* success/failure */ +extern int xfs_alloc_lookup_ge(struct xfs_btree_cur *cur, xfs_agblock_t bno, + xfs_extlen_t len, int *stat); /* * Lookup the first record less than or equal to [bno, len] * in the btree given by cur. */ -int /* error */ -xfs_alloc_lookup_le( - struct xfs_btree_cur *cur, /* btree cursor */ - xfs_agblock_t bno, /* starting block of extent */ - xfs_extlen_t len, /* length of extent */ - int *stat); /* success/failure */ +extern int xfs_alloc_lookup_le(struct xfs_btree_cur *cur, xfs_agblock_t bno, + xfs_extlen_t len, int *stat); /* * Update the record referred to by cur, to the value given by [bno, len]. * This either works (return 0) or gets an EFSCORRUPTED error. */ -int /* error */ -xfs_alloc_update( - struct xfs_btree_cur *cur, /* btree cursor */ - xfs_agblock_t bno, /* starting block of extent */ - xfs_extlen_t len); /* length of extent */ +extern int xfs_alloc_update(struct xfs_btree_cur *cur, xfs_agblock_t bno, + xfs_extlen_t len); #endif /* __XFS_ALLOC_BTREE_H__ */ diff --git a/fs/xfs/xfs_arch.h b/fs/xfs/xfs_arch.h index 5ab0dd885b1b..68e5051d8e24 100644 --- a/fs/xfs/xfs_arch.h +++ b/fs/xfs/xfs_arch.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_ARCH_H__ #define __XFS_ARCH_H__ @@ -168,6 +154,21 @@ } \ } +static inline void be16_add(__be16 *a, __s16 b) +{ + *a = cpu_to_be16(be16_to_cpu(*a) + b); +} + +static inline void be32_add(__be32 *a, __s32 b) +{ + *a = cpu_to_be32(be32_to_cpu(*a) + b); +} + +static inline void be64_add(__be64 *a, __s64 b) +{ + *a = cpu_to_be64(be64_to_cpu(*a) + b); +} + /* * In directories inode numbers are stored as unaligned arrays of unsigned * 8bit integers on disk. diff --git a/fs/xfs/xfs_attr.c b/fs/xfs/xfs_attr.c index a41ad3a5e554..5484eeb460c8 100644 --- a/fs/xfs/xfs_attr.c +++ b/fs/xfs/xfs_attr.c @@ -1,41 +1,26 @@ /* - * Copyright (c) 2000-2004 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" - -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" @@ -43,27 +28,26 @@ #include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" -#include "xfs_alloc_btree.h" +#include "xfs_da_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_alloc.h" -#include "xfs_btree.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" -#include "xfs_inode_item.h" #include "xfs_inode.h" +#include "xfs_alloc.h" +#include "xfs_btree.h" +#include "xfs_inode_item.h" #include "xfs_bmap.h" -#include "xfs_da_btree.h" #include "xfs_attr.h" #include "xfs_attr_leaf.h" #include "xfs_error.h" -#include "xfs_bit.h" #include "xfs_quota.h" -#include "xfs_rw.h" #include "xfs_trans_space.h" #include "xfs_acl.h" +#include "xfs_rw.h" /* * xfs_attr.c @@ -122,7 +106,7 @@ ktrace_t *xfs_attr_trace_buf; *========================================================================*/ int -xfs_attr_fetch(xfs_inode_t *ip, char *name, int namelen, +xfs_attr_fetch(xfs_inode_t *ip, const char *name, int namelen, char *value, int *valuelenp, int flags, struct cred *cred) { xfs_da_args_t args; @@ -177,7 +161,7 @@ xfs_attr_fetch(xfs_inode_t *ip, char *name, int namelen, } int -xfs_attr_get(bhv_desc_t *bdp, char *name, char *value, int *valuelenp, +xfs_attr_get(bhv_desc_t *bdp, const char *name, char *value, int *valuelenp, int flags, struct cred *cred) { xfs_inode_t *ip = XFS_BHVTOI(bdp); @@ -200,40 +184,18 @@ xfs_attr_get(bhv_desc_t *bdp, char *name, char *value, int *valuelenp, return(error); } -/*ARGSUSED*/ -int /* error */ -xfs_attr_set(bhv_desc_t *bdp, char *name, char *value, int valuelen, int flags, - struct cred *cred) +STATIC int +xfs_attr_set_int(xfs_inode_t *dp, const char *name, int namelen, + char *value, int valuelen, int flags) { xfs_da_args_t args; - xfs_inode_t *dp; xfs_fsblock_t firstblock; xfs_bmap_free_t flist; int error, err2, committed; int local, size; uint nblks; - xfs_mount_t *mp; + xfs_mount_t *mp = dp->i_mount; int rsvd = (flags & ATTR_ROOT) != 0; - int namelen; - - namelen = strlen(name); - if (namelen >= MAXNAMELEN) - return EFAULT; /* match IRIX behaviour */ - - XFS_STATS_INC(xs_attr_set); - - dp = XFS_BHVTOI(bdp); - mp = dp->i_mount; - if (XFS_FORCED_SHUTDOWN(mp)) - return (EIO); - - xfs_ilock(dp, XFS_ILOCK_SHARED); - if (!(flags & ATTR_SECURE) && - (error = xfs_iaccess(dp, S_IWUSR, cred))) { - xfs_iunlock(dp, XFS_ILOCK_SHARED); - return(XFS_ERROR(error)); - } - xfs_iunlock(dp, XFS_ILOCK_SHARED); /* * Attach the dquots to the inode. @@ -242,12 +204,18 @@ xfs_attr_set(bhv_desc_t *bdp, char *name, char *value, int valuelen, int flags, return (error); /* + * Determine space new attribute will use, and if it would be + * "local" or "remote" (note: local != inline). + */ + size = xfs_attr_leaf_newentsize(namelen, valuelen, + mp->m_sb.sb_blocksize, &local); + + /* * If the inode doesn't have an attribute fork, add one. * (inode must not be locked when we call this routine) */ if (XFS_IFORK_Q(dp) == 0) { - error = xfs_bmap_add_attrfork(dp, rsvd); - if (error) + if ((error = xfs_bmap_add_attrfork(dp, size, rsvd))) return(error); } @@ -265,13 +233,9 @@ xfs_attr_set(bhv_desc_t *bdp, char *name, char *value, int valuelen, int flags, args.firstblock = &firstblock; args.flist = &flist; args.whichfork = XFS_ATTR_FORK; + args.addname = 1; args.oknoent = 1; - /* Determine space new attribute will use, and if it will be inline - * or out of line. - */ - size = xfs_attr_leaf_newentsize(&args, mp->m_sb.sb_blocksize, &local); - nblks = XFS_DAENTER_SPACE_RES(mp, XFS_ATTR_FORK); if (local) { if (size > (mp->m_sb.sb_blocksize >> 1)) { @@ -343,7 +307,7 @@ xfs_attr_set(bhv_desc_t *bdp, char *name, char *value, int valuelen, int flags, * Build initial attribute list (if required). */ if (dp->i_d.di_aformat == XFS_DINODE_FMT_EXTENTS) - (void)xfs_attr_shortform_create(&args); + xfs_attr_shortform_create(&args); /* * Try to add the attr to the attribute list in @@ -456,32 +420,21 @@ out: return(error); } -/* - * Generic handler routine to remove a name from an attribute list. - * Transitions attribute list from Btree to shortform as necessary. - */ -/*ARGSUSED*/ -int /* error */ -xfs_attr_remove(bhv_desc_t *bdp, char *name, int flags, struct cred *cred) +int +xfs_attr_set(bhv_desc_t *bdp, const char *name, char *value, int valuelen, int flags, + struct cred *cred) { - xfs_da_args_t args; - xfs_inode_t *dp; - xfs_fsblock_t firstblock; - xfs_bmap_free_t flist; - int error; - xfs_mount_t *mp; - int namelen; + xfs_inode_t *dp; + int namelen, error; - ASSERT(MAXNAMELEN-1<=0xff); /* length is stored in uint8 */ namelen = strlen(name); - if (namelen>=MAXNAMELEN) - return EFAULT; /* match irix behaviour */ + if (namelen >= MAXNAMELEN) + return EFAULT; /* match IRIX behaviour */ - XFS_STATS_INC(xs_attr_remove); + XFS_STATS_INC(xs_attr_set); dp = XFS_BHVTOI(bdp); - mp = dp->i_mount; - if (XFS_FORCED_SHUTDOWN(mp)) + if (XFS_FORCED_SHUTDOWN(dp->i_mount)) return (EIO); xfs_ilock(dp, XFS_ILOCK_SHARED); @@ -489,14 +442,25 @@ xfs_attr_remove(bhv_desc_t *bdp, char *name, int flags, struct cred *cred) (error = xfs_iaccess(dp, S_IWUSR, cred))) { xfs_iunlock(dp, XFS_ILOCK_SHARED); return(XFS_ERROR(error)); - } else if (XFS_IFORK_Q(dp) == 0 || - (dp->i_d.di_aformat == XFS_DINODE_FMT_EXTENTS && - dp->i_d.di_anextents == 0)) { - xfs_iunlock(dp, XFS_ILOCK_SHARED); - return(XFS_ERROR(ENOATTR)); } xfs_iunlock(dp, XFS_ILOCK_SHARED); + return xfs_attr_set_int(dp, name, namelen, value, valuelen, flags); +} + +/* + * Generic handler routine to remove a name from an attribute list. + * Transitions attribute list from Btree to shortform as necessary. + */ +STATIC int +xfs_attr_remove_int(xfs_inode_t *dp, const char *name, int namelen, int flags) +{ + xfs_da_args_t args; + xfs_fsblock_t firstblock; + xfs_bmap_free_t flist; + int error; + xfs_mount_t *mp = dp->i_mount; + /* * Fill in the arg structure for this request. */ @@ -544,7 +508,6 @@ xfs_attr_remove(bhv_desc_t *bdp, char *name, int flags, struct cred *cred) XFS_ATTRRM_LOG_COUNT))) { xfs_trans_cancel(args.trans, 0); return(error); - } xfs_ilock(dp, XFS_ILOCK_EXCL); @@ -612,6 +575,38 @@ out: return(error); } +int +xfs_attr_remove(bhv_desc_t *bdp, const char *name, int flags, struct cred *cred) +{ + xfs_inode_t *dp; + int namelen, error; + + namelen = strlen(name); + if (namelen >= MAXNAMELEN) + return EFAULT; /* match IRIX behaviour */ + + XFS_STATS_INC(xs_attr_remove); + + dp = XFS_BHVTOI(bdp); + if (XFS_FORCED_SHUTDOWN(dp->i_mount)) + return (EIO); + + xfs_ilock(dp, XFS_ILOCK_SHARED); + if (!(flags & ATTR_SECURE) && + (error = xfs_iaccess(dp, S_IWUSR, cred))) { + xfs_iunlock(dp, XFS_ILOCK_SHARED); + return(XFS_ERROR(error)); + } else if (XFS_IFORK_Q(dp) == 0 || + (dp->i_d.di_aformat == XFS_DINODE_FMT_EXTENTS && + dp->i_d.di_anextents == 0)) { + xfs_iunlock(dp, XFS_ILOCK_SHARED); + return(XFS_ERROR(ENOATTR)); + } + xfs_iunlock(dp, XFS_ILOCK_SHARED); + + return xfs_attr_remove_int(dp, name, namelen, flags); +} + /* * Generate a list of extended attribute names and optionally * also value lengths. Positive return value follows the XFS @@ -811,7 +806,7 @@ out: STATIC int xfs_attr_shortform_addname(xfs_da_args_t *args) { - int newsize, retval; + int newsize, forkoff, retval; retval = xfs_attr_shortform_lookup(args); if ((args->flags & ATTR_REPLACE) && (retval == ENOATTR)) { @@ -823,16 +818,18 @@ xfs_attr_shortform_addname(xfs_da_args_t *args) ASSERT(retval == 0); } + if (args->namelen >= XFS_ATTR_SF_ENTSIZE_MAX || + args->valuelen >= XFS_ATTR_SF_ENTSIZE_MAX) + return(XFS_ERROR(ENOSPC)); + newsize = XFS_ATTR_SF_TOTSIZE(args->dp); newsize += XFS_ATTR_SF_ENTSIZE_BYNAME(args->namelen, args->valuelen); - if ((newsize <= XFS_IFORK_ASIZE(args->dp)) && - (args->namelen < XFS_ATTR_SF_ENTSIZE_MAX) && - (args->valuelen < XFS_ATTR_SF_ENTSIZE_MAX)) { - retval = xfs_attr_shortform_add(args); - ASSERT(retval == 0); - } else { + + forkoff = xfs_attr_shortform_bytesfit(args->dp, newsize); + if (!forkoff) return(XFS_ERROR(ENOSPC)); - } + + xfs_attr_shortform_add(args, forkoff); return(0); } @@ -852,7 +849,7 @@ xfs_attr_leaf_addname(xfs_da_args_t *args) { xfs_inode_t *dp; xfs_dabuf_t *bp; - int retval, error, committed; + int retval, error, committed, forkoff; /* * Read the (only) block in the attribute list in. @@ -995,9 +992,9 @@ xfs_attr_leaf_addname(xfs_da_args_t *args) /* * If the result is small enough, shrink it all into the inode. */ - if (xfs_attr_shortform_allfit(bp, dp)) { + if ((forkoff = xfs_attr_shortform_allfit(bp, dp))) { XFS_BMAP_INIT(args->flist, args->firstblock); - error = xfs_attr_leaf_to_shortform(bp, args); + error = xfs_attr_leaf_to_shortform(bp, args, forkoff); /* bp is gone due to xfs_da_shrink_inode */ if (!error) { error = xfs_bmap_finish(&args->trans, @@ -1049,8 +1046,7 @@ xfs_attr_leaf_removename(xfs_da_args_t *args) { xfs_inode_t *dp; xfs_dabuf_t *bp; - int committed; - int error; + int error, committed, forkoff; /* * Remove the attribute. @@ -1075,9 +1071,9 @@ xfs_attr_leaf_removename(xfs_da_args_t *args) /* * If the result is small enough, shrink it all into the inode. */ - if (xfs_attr_shortform_allfit(bp, dp)) { + if ((forkoff = xfs_attr_shortform_allfit(bp, dp))) { XFS_BMAP_INIT(args->flist, args->firstblock); - error = xfs_attr_leaf_to_shortform(bp, args); + error = xfs_attr_leaf_to_shortform(bp, args, forkoff); /* bp is gone due to xfs_da_shrink_inode */ if (!error) { error = xfs_bmap_finish(&args->trans, args->flist, @@ -1448,7 +1444,7 @@ xfs_attr_node_removename(xfs_da_args_t *args) xfs_da_state_blk_t *blk; xfs_inode_t *dp; xfs_dabuf_t *bp; - int retval, error, committed; + int retval, error, committed, forkoff; /* * Tie a string around our finger to remind us where we are. @@ -1569,9 +1565,9 @@ xfs_attr_node_removename(xfs_da_args_t *args) bp->data)->hdr.info.magic, ARCH_CONVERT) == XFS_ATTR_LEAF_MAGIC); - if (xfs_attr_shortform_allfit(bp, dp)) { + if ((forkoff = xfs_attr_shortform_allfit(bp, dp))) { XFS_BMAP_INIT(args->flist, args->firstblock); - error = xfs_attr_leaf_to_shortform(bp, args); + error = xfs_attr_leaf_to_shortform(bp, args, forkoff); /* bp is gone due to xfs_da_shrink_inode */ if (!error) { error = xfs_bmap_finish(&args->trans, diff --git a/fs/xfs/xfs_attr.h b/fs/xfs/xfs_attr.h index 45ab1c542baf..b2c7b9fcded3 100644 --- a/fs/xfs/xfs_attr.h +++ b/fs/xfs/xfs_attr.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000, 2002-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000,2002-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_ATTR_H__ #define __XFS_ATTR_H__ @@ -172,15 +158,15 @@ struct xfs_da_args; /* * Overall external interface routines. */ -int xfs_attr_get(bhv_desc_t *, char *, char *, int *, int, struct cred *); -int xfs_attr_set(bhv_desc_t *, char *, char *, int, int, struct cred *); -int xfs_attr_remove(bhv_desc_t *, char *, int, struct cred *); +int xfs_attr_get(bhv_desc_t *, const char *, char *, int *, int, struct cred *); +int xfs_attr_set(bhv_desc_t *, const char *, char *, int, int, struct cred *); +int xfs_attr_remove(bhv_desc_t *, const char *, int, struct cred *); int xfs_attr_list(bhv_desc_t *, char *, int, int, struct attrlist_cursor_kern *, struct cred *); int xfs_attr_inactive(struct xfs_inode *dp); int xfs_attr_shortform_getvalue(struct xfs_da_args *); -int xfs_attr_fetch(struct xfs_inode *, char *, int, +int xfs_attr_fetch(struct xfs_inode *, const char *, int, char *, int *, int, struct cred *); #endif /* __XFS_ATTR_H__ */ diff --git a/fs/xfs/xfs_attr_leaf.c b/fs/xfs/xfs_attr_leaf.c index 1cdd574c63a9..1c7421840c18 100644 --- a/fs/xfs/xfs_attr_leaf.c +++ b/fs/xfs/xfs_attr_leaf.c @@ -1,46 +1,26 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ -/* - * xfs_attr_leaf.c - * - * GROT: figure out how to recover gracefully when bmap returns ENOSPC. - */ - #include "xfs.h" - -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" @@ -48,23 +28,22 @@ #include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" -#include "xfs_alloc_btree.h" +#include "xfs_da_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" #include "xfs_alloc.h" #include "xfs_btree.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" -#include "xfs_inode_item.h" #include "xfs_inode.h" +#include "xfs_inode_item.h" #include "xfs_bmap.h" -#include "xfs_da_btree.h" #include "xfs_attr.h" #include "xfs_attr_leaf.h" #include "xfs_error.h" -#include "xfs_bit.h" /* * xfs_attr_leaf.c @@ -118,13 +97,82 @@ STATIC int xfs_attr_put_listent(xfs_attr_list_context_t *context, /*======================================================================== - * External routines when dirsize < XFS_LITINO(mp). + * External routines when attribute fork size < XFS_LITINO(mp). *========================================================================*/ /* - * Create the initial contents of a shortform attribute list. + * Query whether the requested number of additional bytes of extended + * attribute space will be able to fit inline. + * Returns zero if not, else the di_forkoff fork offset to be used in the + * literal area for attribute data once the new bytes have been added. + * + * di_forkoff must be 8 byte aligned, hence is stored as a >>3 value; + * special case for dev/uuid inodes, they have fixed size data forks. */ int +xfs_attr_shortform_bytesfit(xfs_inode_t *dp, int bytes) +{ + int offset; + int minforkoff; /* lower limit on valid forkoff locations */ + int maxforkoff; /* upper limit on valid forkoff locations */ + xfs_mount_t *mp = dp->i_mount; + + offset = (XFS_LITINO(mp) - bytes) >> 3; /* rounded down */ + + switch (dp->i_d.di_format) { + case XFS_DINODE_FMT_DEV: + minforkoff = roundup(sizeof(xfs_dev_t), 8) >> 3; + return (offset >= minforkoff) ? minforkoff : 0; + case XFS_DINODE_FMT_UUID: + minforkoff = roundup(sizeof(uuid_t), 8) >> 3; + return (offset >= minforkoff) ? minforkoff : 0; + } + + if (unlikely(mp->m_flags & XFS_MOUNT_COMPAT_ATTR)) { + if (bytes <= XFS_IFORK_ASIZE(dp)) + return mp->m_attroffset >> 3; + return 0; + } + + /* data fork btree root can have at least this many key/ptr pairs */ + minforkoff = MAX(dp->i_df.if_bytes, XFS_BMDR_SPACE_CALC(MINDBTPTRS)); + minforkoff = roundup(minforkoff, 8) >> 3; + + /* attr fork btree root can have at least this many key/ptr pairs */ + maxforkoff = XFS_LITINO(mp) - XFS_BMDR_SPACE_CALC(MINABTPTRS); + maxforkoff = maxforkoff >> 3; /* rounded down */ + + if (offset >= minforkoff && offset < maxforkoff) + return offset; + if (offset >= maxforkoff) + return maxforkoff; + return 0; +} + +/* + * Switch on the ATTR2 superblock bit (implies also FEATURES2) + */ +STATIC void +xfs_sbversion_add_attr2(xfs_mount_t *mp, xfs_trans_t *tp) +{ + unsigned long s; + + if (!(mp->m_flags & XFS_MOUNT_COMPAT_ATTR) && + !(XFS_SB_VERSION_HASATTR2(&mp->m_sb))) { + s = XFS_SB_LOCK(mp); + if (!XFS_SB_VERSION_HASATTR2(&mp->m_sb)) { + XFS_SB_VERSION_ADDATTR2(&mp->m_sb); + XFS_SB_UNLOCK(mp, s); + xfs_mod_sb(tp, XFS_SB_VERSIONNUM | XFS_SB_FEATURES2); + } else + XFS_SB_UNLOCK(mp, s); + } +} + +/* + * Create the initial contents of a shortform attribute list. + */ +void xfs_attr_shortform_create(xfs_da_args_t *args) { xfs_attr_sf_hdr_t *hdr; @@ -148,29 +196,37 @@ xfs_attr_shortform_create(xfs_da_args_t *args) hdr->count = 0; INT_SET(hdr->totsize, ARCH_CONVERT, sizeof(*hdr)); xfs_trans_log_inode(args->trans, dp, XFS_ILOG_CORE | XFS_ILOG_ADATA); - return(0); } /* * Add a name/value pair to the shortform attribute list. * Overflow from the inode has already been checked for. */ -int -xfs_attr_shortform_add(xfs_da_args_t *args) +void +xfs_attr_shortform_add(xfs_da_args_t *args, int forkoff) { xfs_attr_shortform_t *sf; xfs_attr_sf_entry_t *sfe; int i, offset, size; + xfs_mount_t *mp; xfs_inode_t *dp; xfs_ifork_t *ifp; dp = args->dp; + mp = dp->i_mount; + dp->i_d.di_forkoff = forkoff; + dp->i_df.if_ext_max = + XFS_IFORK_DSIZE(dp) / (uint)sizeof(xfs_bmbt_rec_t); + dp->i_afp->if_ext_max = + XFS_IFORK_ASIZE(dp) / (uint)sizeof(xfs_bmbt_rec_t); + ifp = dp->i_afp; ASSERT(ifp->if_flags & XFS_IFINLINE); sf = (xfs_attr_shortform_t *)ifp->if_u1.if_data; sfe = &sf->list[0]; for (i = 0; i < INT_GET(sf->hdr.count, ARCH_CONVERT); sfe = XFS_ATTR_SF_NEXTENTRY(sfe), i++) { +#ifdef DEBUG if (sfe->namelen != args->namelen) continue; if (memcmp(args->name, sfe->nameval, args->namelen) != 0) @@ -181,7 +237,8 @@ xfs_attr_shortform_add(xfs_da_args_t *args) if (((args->flags & ATTR_ROOT) != 0) != ((sfe->flags & XFS_ATTR_ROOT) != 0)) continue; - return(XFS_ERROR(EEXIST)); + ASSERT(0); +#endif } offset = (char *)sfe - (char *)sf; @@ -200,11 +257,11 @@ xfs_attr_shortform_add(xfs_da_args_t *args) INT_MOD(sf->hdr.totsize, ARCH_CONVERT, size); xfs_trans_log_inode(args->trans, dp, XFS_ILOG_CORE | XFS_ILOG_ADATA); - return(0); + xfs_sbversion_add_attr2(mp, args->trans); } /* - * Remove a name from the shortform attribute list structure. + * Remove an attribute from the shortform attribute list structure. */ int xfs_attr_shortform_remove(xfs_da_args_t *args) @@ -212,17 +269,16 @@ xfs_attr_shortform_remove(xfs_da_args_t *args) xfs_attr_shortform_t *sf; xfs_attr_sf_entry_t *sfe; int base, size=0, end, totsize, i; + xfs_mount_t *mp; xfs_inode_t *dp; - /* - * Remove the attribute. - */ dp = args->dp; + mp = dp->i_mount; base = sizeof(xfs_attr_sf_hdr_t); sf = (xfs_attr_shortform_t *)dp->i_afp->if_u1.if_data; sfe = &sf->list[0]; - for (i = 0; i < INT_GET(sf->hdr.count, ARCH_CONVERT); - sfe = XFS_ATTR_SF_NEXTENTRY(sfe), + end = INT_GET(sf->hdr.count, ARCH_CONVERT); + for (i = 0; i < end; sfe = XFS_ATTR_SF_NEXTENTRY(sfe), base += size, i++) { size = XFS_ATTR_SF_ENTSIZE(sfe); if (sfe->namelen != args->namelen) @@ -237,19 +293,53 @@ xfs_attr_shortform_remove(xfs_da_args_t *args) continue; break; } - if (i == INT_GET(sf->hdr.count, ARCH_CONVERT)) + if (i == end) return(XFS_ERROR(ENOATTR)); + /* + * Fix up the attribute fork data, covering the hole + */ end = base + size; totsize = INT_GET(sf->hdr.totsize, ARCH_CONVERT); - if (end != totsize) { - memmove(&((char *)sf)[base], &((char *)sf)[end], - totsize - end); - } + if (end != totsize) + memmove(&((char *)sf)[base], &((char *)sf)[end], totsize - end); INT_MOD(sf->hdr.count, ARCH_CONVERT, -1); INT_MOD(sf->hdr.totsize, ARCH_CONVERT, -size); - xfs_idata_realloc(dp, -size, XFS_ATTR_FORK); - xfs_trans_log_inode(args->trans, dp, XFS_ILOG_CORE | XFS_ILOG_ADATA); + + /* + * Fix up the start offset of the attribute fork + */ + totsize -= size; + if (totsize == sizeof(xfs_attr_sf_hdr_t) && !args->addname && + !(mp->m_flags & XFS_MOUNT_COMPAT_ATTR)) { + /* + * Last attribute now removed, revert to original + * inode format making all literal area available + * to the data fork once more. + */ + xfs_idestroy_fork(dp, XFS_ATTR_FORK); + dp->i_d.di_forkoff = 0; + dp->i_d.di_aformat = XFS_DINODE_FMT_EXTENTS; + ASSERT(dp->i_d.di_anextents == 0); + ASSERT(dp->i_afp == NULL); + dp->i_df.if_ext_max = + XFS_IFORK_DSIZE(dp) / (uint)sizeof(xfs_bmbt_rec_t); + xfs_trans_log_inode(args->trans, dp, XFS_ILOG_CORE); + } else { + xfs_idata_realloc(dp, -size, XFS_ATTR_FORK); + dp->i_d.di_forkoff = xfs_attr_shortform_bytesfit(dp, totsize); + ASSERT(dp->i_d.di_forkoff); + ASSERT(totsize > sizeof(xfs_attr_sf_hdr_t) || args->addname || + (mp->m_flags & XFS_MOUNT_COMPAT_ATTR)); + dp->i_afp->if_ext_max = + XFS_IFORK_ASIZE(dp) / (uint)sizeof(xfs_bmbt_rec_t); + dp->i_df.if_ext_max = + XFS_IFORK_DSIZE(dp) / (uint)sizeof(xfs_bmbt_rec_t); + xfs_trans_log_inode(args->trans, dp, + XFS_ILOG_CORE | XFS_ILOG_ADATA); + } + + xfs_sbversion_add_attr2(mp, args->trans); return(0); } @@ -561,7 +651,7 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context) /* * Sort the entries on hash then entno. */ - qsort(sbuf, nsbuf, sizeof(*sbuf), xfs_attr_shortform_compare); + xfs_sort(sbuf, nsbuf, sizeof(*sbuf), xfs_attr_shortform_compare); /* * Re-find our place IN THE SORTED LIST. @@ -649,14 +739,17 @@ xfs_attr_shortform_allfit(xfs_dabuf_t *bp, xfs_inode_t *dp) + name_loc->namelen + INT_GET(name_loc->valuelen, ARCH_CONVERT); } - return( bytes < XFS_IFORK_ASIZE(dp) ); + if (!(dp->i_mount->m_flags & XFS_MOUNT_COMPAT_ATTR) && + (bytes == sizeof(struct xfs_attr_sf_hdr))) + return(-1); + return(xfs_attr_shortform_bytesfit(dp, bytes)); } /* * Convert a leaf attribute list to shortform attribute list */ int -xfs_attr_leaf_to_shortform(xfs_dabuf_t *bp, xfs_da_args_t *args) +xfs_attr_leaf_to_shortform(xfs_dabuf_t *bp, xfs_da_args_t *args, int forkoff) { xfs_attr_leafblock_t *leaf; xfs_attr_leaf_entry_t *entry; @@ -683,9 +776,27 @@ xfs_attr_leaf_to_shortform(xfs_dabuf_t *bp, xfs_da_args_t *args) error = xfs_da_shrink_inode(args, 0, bp); if (error) goto out; - error = xfs_attr_shortform_create(args); - if (error) + + if (forkoff == -1) { + ASSERT(!(dp->i_mount->m_flags & XFS_MOUNT_COMPAT_ATTR)); + + /* + * Last attribute was removed, revert to original + * inode format making all literal area available + * to the data fork once more. + */ + xfs_idestroy_fork(dp, XFS_ATTR_FORK); + dp->i_d.di_forkoff = 0; + dp->i_d.di_aformat = XFS_DINODE_FMT_EXTENTS; + ASSERT(dp->i_d.di_anextents == 0); + ASSERT(dp->i_afp == NULL); + dp->i_df.if_ext_max = + XFS_IFORK_DSIZE(dp) / (uint)sizeof(xfs_bmbt_rec_t); + xfs_trans_log_inode(args->trans, dp, XFS_ILOG_CORE); goto out; + } + + xfs_attr_shortform_create(args); /* * Copy the attributes @@ -713,7 +824,7 @@ xfs_attr_leaf_to_shortform(xfs_dabuf_t *bp, xfs_da_args_t *args) nargs.hashval = INT_GET(entry->hashval, ARCH_CONVERT); nargs.flags = (entry->flags & XFS_ATTR_SECURE) ? ATTR_SECURE : ((entry->flags & XFS_ATTR_ROOT) ? ATTR_ROOT : 0); - xfs_attr_shortform_add(&nargs); + xfs_attr_shortform_add(&nargs, forkoff); } error = 0; @@ -898,7 +1009,7 @@ xfs_attr_leaf_add(xfs_dabuf_t *bp, xfs_da_args_t *args) ASSERT((args->index >= 0) && (args->index <= INT_GET(leaf->hdr.count, ARCH_CONVERT))); hdr = &leaf->hdr; - entsize = xfs_attr_leaf_newentsize(args, + entsize = xfs_attr_leaf_newentsize(args->namelen, args->valuelen, args->trans->t_mountp->m_sb.sb_blocksize, NULL); /* @@ -995,13 +1106,14 @@ xfs_attr_leaf_add_work(xfs_dabuf_t *bp, xfs_da_args_t *args, int mapindex) mp = args->trans->t_mountp; ASSERT(INT_GET(map->base, ARCH_CONVERT) < XFS_LBSIZE(mp)); ASSERT((INT_GET(map->base, ARCH_CONVERT) & 0x3) == 0); - ASSERT(INT_GET(map->size, ARCH_CONVERT) - >= xfs_attr_leaf_newentsize(args, - mp->m_sb.sb_blocksize, NULL)); + ASSERT(INT_GET(map->size, ARCH_CONVERT) >= + xfs_attr_leaf_newentsize(args->namelen, args->valuelen, + mp->m_sb.sb_blocksize, NULL)); ASSERT(INT_GET(map->size, ARCH_CONVERT) < XFS_LBSIZE(mp)); ASSERT((INT_GET(map->size, ARCH_CONVERT) & 0x3) == 0); INT_MOD(map->size, ARCH_CONVERT, - -xfs_attr_leaf_newentsize(args, mp->m_sb.sb_blocksize, &tmp)); + -xfs_attr_leaf_newentsize(args->namelen, args->valuelen, + mp->m_sb.sb_blocksize, &tmp)); INT_SET(entry->nameidx, ARCH_CONVERT, INT_GET(map->base, ARCH_CONVERT) + INT_GET(map->size, ARCH_CONVERT)); @@ -1357,8 +1469,10 @@ xfs_attr_leaf_figure_balance(xfs_da_state_t *state, half = (max+1) * sizeof(*entry); half += INT_GET(hdr1->usedbytes, ARCH_CONVERT) + INT_GET(hdr2->usedbytes, ARCH_CONVERT) - + xfs_attr_leaf_newentsize(state->args, - state->blocksize, NULL); + + xfs_attr_leaf_newentsize( + state->args->namelen, + state->args->valuelen, + state->blocksize, NULL); half /= 2; lastdelta = state->blocksize; entry = &leaf1->entries[0]; @@ -1370,9 +1484,10 @@ xfs_attr_leaf_figure_balance(xfs_da_state_t *state, */ if (count == blk1->index) { tmp = totallen + sizeof(*entry) + - xfs_attr_leaf_newentsize(state->args, - state->blocksize, - NULL); + xfs_attr_leaf_newentsize( + state->args->namelen, + state->args->valuelen, + state->blocksize, NULL); if (XFS_ATTR_ABS(half - tmp) > lastdelta) break; lastdelta = XFS_ATTR_ABS(half - tmp); @@ -1408,9 +1523,10 @@ xfs_attr_leaf_figure_balance(xfs_da_state_t *state, totallen -= count * sizeof(*entry); if (foundit) { totallen -= sizeof(*entry) + - xfs_attr_leaf_newentsize(state->args, - state->blocksize, - NULL); + xfs_attr_leaf_newentsize( + state->args->namelen, + state->args->valuelen, + state->blocksize, NULL); } *countarg = count; @@ -2253,17 +2369,17 @@ xfs_attr_leaf_entsize(xfs_attr_leafblock_t *leaf, int index) * a "local" or a "remote" attribute. */ int -xfs_attr_leaf_newentsize(xfs_da_args_t *args, int blocksize, int *local) +xfs_attr_leaf_newentsize(int namelen, int valuelen, int blocksize, int *local) { int size; - size = XFS_ATTR_LEAF_ENTSIZE_LOCAL(args->namelen, args->valuelen); + size = XFS_ATTR_LEAF_ENTSIZE_LOCAL(namelen, valuelen); if (size < XFS_ATTR_LEAF_ENTSIZE_LOCAL_MAX(blocksize)) { if (local) { *local = 1; } } else { - size = XFS_ATTR_LEAF_ENTSIZE_REMOTE(args->namelen); + size = XFS_ATTR_LEAF_ENTSIZE_REMOTE(namelen); if (local) { *local = 0; } diff --git a/fs/xfs/xfs_attr_leaf.h b/fs/xfs/xfs_attr_leaf.h index 0a4cfad6df91..f6143ff251a0 100644 --- a/fs/xfs/xfs_attr_leaf.h +++ b/fs/xfs/xfs_attr_leaf.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000, 2002-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000,2002-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_ATTR_LEAF_H__ #define __XFS_ATTR_LEAF_H__ @@ -146,65 +132,58 @@ typedef struct xfs_attr_leaf_name_remote xfs_attr_leaf_name_remote_t; /* * Cast typed pointers for "local" and "remote" name/value structs. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_ATTR_LEAF_NAME_REMOTE) -xfs_attr_leaf_name_remote_t * -xfs_attr_leaf_name_remote(xfs_attr_leafblock_t *leafp, int idx); #define XFS_ATTR_LEAF_NAME_REMOTE(leafp,idx) \ xfs_attr_leaf_name_remote(leafp,idx) -#else -#define XFS_ATTR_LEAF_NAME_REMOTE(leafp,idx) /* remote name struct ptr */ \ - ((xfs_attr_leaf_name_remote_t *) \ - &((char *)(leafp))[ INT_GET((leafp)->entries[idx].nameidx, ARCH_CONVERT) ]) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_ATTR_LEAF_NAME_LOCAL) -xfs_attr_leaf_name_local_t * -xfs_attr_leaf_name_local(xfs_attr_leafblock_t *leafp, int idx); +static inline xfs_attr_leaf_name_remote_t * +xfs_attr_leaf_name_remote(xfs_attr_leafblock_t *leafp, int idx) +{ + return (xfs_attr_leaf_name_remote_t *) &((char *) + (leafp))[INT_GET((leafp)->entries[idx].nameidx, ARCH_CONVERT)]; +} + #define XFS_ATTR_LEAF_NAME_LOCAL(leafp,idx) \ xfs_attr_leaf_name_local(leafp,idx) -#else -#define XFS_ATTR_LEAF_NAME_LOCAL(leafp,idx) /* local name struct ptr */ \ - ((xfs_attr_leaf_name_local_t *) \ - &((char *)(leafp))[ INT_GET((leafp)->entries[idx].nameidx, ARCH_CONVERT) ]) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_ATTR_LEAF_NAME) -char *xfs_attr_leaf_name(xfs_attr_leafblock_t *leafp, int idx); +static inline xfs_attr_leaf_name_local_t * +xfs_attr_leaf_name_local(xfs_attr_leafblock_t *leafp, int idx) +{ + return (xfs_attr_leaf_name_local_t *) &((char *) + (leafp))[INT_GET((leafp)->entries[idx].nameidx, ARCH_CONVERT)]; +} + #define XFS_ATTR_LEAF_NAME(leafp,idx) xfs_attr_leaf_name(leafp,idx) -#else -#define XFS_ATTR_LEAF_NAME(leafp,idx) /* generic name struct ptr */ \ - (&((char *)(leafp))[ INT_GET((leafp)->entries[idx].nameidx, ARCH_CONVERT) ]) -#endif +static inline char *xfs_attr_leaf_name(xfs_attr_leafblock_t *leafp, int idx) +{ + return (&((char *) + (leafp))[INT_GET((leafp)->entries[idx].nameidx, ARCH_CONVERT)]); +} /* * Calculate total bytes used (including trailing pad for alignment) for * a "local" name/value structure, a "remote" name/value structure, and * a pointer which might be either. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_ATTR_LEAF_ENTSIZE_REMOTE) -int xfs_attr_leaf_entsize_remote(int nlen); #define XFS_ATTR_LEAF_ENTSIZE_REMOTE(nlen) \ xfs_attr_leaf_entsize_remote(nlen) -#else -#define XFS_ATTR_LEAF_ENTSIZE_REMOTE(nlen) /* space for remote struct */ \ - (((uint)sizeof(xfs_attr_leaf_name_remote_t) - 1 + (nlen) + \ - XFS_ATTR_LEAF_NAME_ALIGN - 1) & ~(XFS_ATTR_LEAF_NAME_ALIGN - 1)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_ATTR_LEAF_ENTSIZE_LOCAL) -int xfs_attr_leaf_entsize_local(int nlen, int vlen); +static inline int xfs_attr_leaf_entsize_remote(int nlen) +{ + return ((uint)sizeof(xfs_attr_leaf_name_remote_t) - 1 + (nlen) + \ + XFS_ATTR_LEAF_NAME_ALIGN - 1) & ~(XFS_ATTR_LEAF_NAME_ALIGN - 1); +} + #define XFS_ATTR_LEAF_ENTSIZE_LOCAL(nlen,vlen) \ xfs_attr_leaf_entsize_local(nlen,vlen) -#else -#define XFS_ATTR_LEAF_ENTSIZE_LOCAL(nlen,vlen) /* space for local struct */ \ - (((uint)sizeof(xfs_attr_leaf_name_local_t) - 1 + (nlen) + (vlen) + \ - XFS_ATTR_LEAF_NAME_ALIGN - 1) & ~(XFS_ATTR_LEAF_NAME_ALIGN - 1)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_ATTR_LEAF_ENTSIZE_LOCAL_MAX) -int xfs_attr_leaf_entsize_local_max(int bsize); +static inline int xfs_attr_leaf_entsize_local(int nlen, int vlen) +{ + return ((uint)sizeof(xfs_attr_leaf_name_local_t) - 1 + (nlen) + (vlen) + + XFS_ATTR_LEAF_NAME_ALIGN - 1) & ~(XFS_ATTR_LEAF_NAME_ALIGN - 1); +} + #define XFS_ATTR_LEAF_ENTSIZE_LOCAL_MAX(bsize) \ xfs_attr_leaf_entsize_local_max(bsize) -#else -#define XFS_ATTR_LEAF_ENTSIZE_LOCAL_MAX(bsize) /* max local struct size */ \ - (((bsize) >> 1) + ((bsize) >> 2)) -#endif +static inline int xfs_attr_leaf_entsize_local_max(int bsize) +{ + return (((bsize) >> 1) + ((bsize) >> 2)); +} /*======================================================================== @@ -237,23 +216,25 @@ typedef struct xfs_attr_inactive_list { *========================================================================*/ /* - * Internal routines when dirsize < XFS_LITINO(mp). + * Internal routines when attribute fork size < XFS_LITINO(mp). */ -int xfs_attr_shortform_create(struct xfs_da_args *args); -int xfs_attr_shortform_add(struct xfs_da_args *add); +void xfs_attr_shortform_create(struct xfs_da_args *args); +void xfs_attr_shortform_add(struct xfs_da_args *args, int forkoff); int xfs_attr_shortform_lookup(struct xfs_da_args *args); int xfs_attr_shortform_getvalue(struct xfs_da_args *args); int xfs_attr_shortform_to_leaf(struct xfs_da_args *args); -int xfs_attr_shortform_remove(struct xfs_da_args *remove); +int xfs_attr_shortform_remove(struct xfs_da_args *args); int xfs_attr_shortform_list(struct xfs_attr_list_context *context); int xfs_attr_shortform_allfit(struct xfs_dabuf *bp, struct xfs_inode *dp); +int xfs_attr_shortform_bytesfit(xfs_inode_t *dp, int bytes); + /* - * Internal routines when dirsize == XFS_LBSIZE(mp). + * Internal routines when attribute fork size == XFS_LBSIZE(mp). */ int xfs_attr_leaf_to_node(struct xfs_da_args *args); int xfs_attr_leaf_to_shortform(struct xfs_dabuf *bp, - struct xfs_da_args *args); + struct xfs_da_args *args, int forkoff); int xfs_attr_leaf_clearflag(struct xfs_da_args *args); int xfs_attr_leaf_setflag(struct xfs_da_args *args); int xfs_attr_leaf_flipflags(xfs_da_args_t *args); @@ -289,7 +270,7 @@ int xfs_attr_root_inactive(struct xfs_trans **trans, struct xfs_inode *dp); xfs_dahash_t xfs_attr_leaf_lasthash(struct xfs_dabuf *bp, int *count); int xfs_attr_leaf_order(struct xfs_dabuf *leaf1_bp, struct xfs_dabuf *leaf2_bp); -int xfs_attr_leaf_newentsize(struct xfs_da_args *args, int blocksize, +int xfs_attr_leaf_newentsize(int namelen, int valuelen, int blocksize, int *local); int xfs_attr_rolltrans(struct xfs_trans **transp, struct xfs_inode *dp); diff --git a/fs/xfs/xfs_attr_sf.h b/fs/xfs/xfs_attr_sf.h index ef7d2942d306..ffed6ca81a52 100644 --- a/fs/xfs/xfs_attr_sf.h +++ b/fs/xfs/xfs_attr_sf.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000, 2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000,2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_ATTR_SF_H__ #define __XFS_ATTR_SF_H__ @@ -71,38 +57,17 @@ typedef struct xfs_attr_sf_sort { char *name; /* name value, pointer into buffer */ } xfs_attr_sf_sort_t; -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_ATTR_SF_ENTSIZE_BYNAME) -int xfs_attr_sf_entsize_byname(int nlen, int vlen); -#define XFS_ATTR_SF_ENTSIZE_BYNAME(nlen,vlen) \ - xfs_attr_sf_entsize_byname(nlen,vlen) -#else #define XFS_ATTR_SF_ENTSIZE_BYNAME(nlen,vlen) /* space name/value uses */ \ - ((int)sizeof(xfs_attr_sf_entry_t)-1 + (nlen)+(vlen)) -#endif + (((int)sizeof(xfs_attr_sf_entry_t)-1 + (nlen)+(vlen))) #define XFS_ATTR_SF_ENTSIZE_MAX /* max space for name&value */ \ ((1 << (NBBY*(int)sizeof(__uint8_t))) - 1) -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_ATTR_SF_ENTSIZE) -int xfs_attr_sf_entsize(xfs_attr_sf_entry_t *sfep); -#define XFS_ATTR_SF_ENTSIZE(sfep) xfs_attr_sf_entsize(sfep) -#else #define XFS_ATTR_SF_ENTSIZE(sfep) /* space an entry uses */ \ ((int)sizeof(xfs_attr_sf_entry_t)-1 + (sfep)->namelen+(sfep)->valuelen) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_ATTR_SF_NEXTENTRY) -xfs_attr_sf_entry_t *xfs_attr_sf_nextentry(xfs_attr_sf_entry_t *sfep); -#define XFS_ATTR_SF_NEXTENTRY(sfep) xfs_attr_sf_nextentry(sfep) -#else #define XFS_ATTR_SF_NEXTENTRY(sfep) /* next entry in struct */ \ - ((xfs_attr_sf_entry_t *) \ - ((char *)(sfep) + XFS_ATTR_SF_ENTSIZE(sfep))) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_ATTR_SF_TOTSIZE) -int xfs_attr_sf_totsize(struct xfs_inode *dp); -#define XFS_ATTR_SF_TOTSIZE(dp) xfs_attr_sf_totsize(dp) -#else + ((xfs_attr_sf_entry_t *)((char *)(sfep) + XFS_ATTR_SF_ENTSIZE(sfep))) #define XFS_ATTR_SF_TOTSIZE(dp) /* total space in use */ \ - (INT_GET(((xfs_attr_shortform_t *)((dp)->i_afp->if_u1.if_data))->hdr.totsize, ARCH_CONVERT)) -#endif + (INT_GET(((xfs_attr_shortform_t *) \ + ((dp)->i_afp->if_u1.if_data))->hdr.totsize, ARCH_CONVERT)) #if defined(XFS_ATTR_TRACE) /* diff --git a/fs/xfs/xfs_behavior.c b/fs/xfs/xfs_behavior.c index 16088e175ecc..9880adae3938 100644 --- a/fs/xfs/xfs_behavior.c +++ b/fs/xfs/xfs_behavior.c @@ -1,34 +1,19 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. - * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #include "xfs.h" diff --git a/fs/xfs/xfs_behavior.h b/fs/xfs/xfs_behavior.h index d5ed5a843921..2cd89bb5ab10 100644 --- a/fs/xfs/xfs_behavior.h +++ b/fs/xfs/xfs_behavior.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_BEHAVIOR_H__ #define __XFS_BEHAVIOR_H__ diff --git a/fs/xfs/xfs_bit.c b/fs/xfs/xfs_bit.c index 76c9ad3875ef..43be6a7e47c6 100644 --- a/fs/xfs/xfs_bit.c +++ b/fs/xfs/xfs_bit.c @@ -1,45 +1,29 @@ /* - * Copyright (c) 2000-2004 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - -/* - * XFS bit manipulation routines, used in non-realtime code. - */ - #include "xfs.h" #include "xfs_bit.h" #include "xfs_log.h" #include "xfs_trans.h" #include "xfs_buf_item.h" +/* + * XFS bit manipulation routines, used in non-realtime code. + */ #ifndef HAVE_ARCH_HIGHBIT /* diff --git a/fs/xfs/xfs_bit.h b/fs/xfs/xfs_bit.h index 1e7f57ddf7a8..0bbe56817542 100644 --- a/fs/xfs/xfs_bit.h +++ b/fs/xfs/xfs_bit.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000, 2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000,2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_BIT_H__ #define __XFS_BIT_H__ @@ -39,30 +25,26 @@ /* * masks with n high/low bits set, 32-bit values & 64-bit values */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_MASK32HI) -__uint32_t xfs_mask32hi(int n); #define XFS_MASK32HI(n) xfs_mask32hi(n) -#else -#define XFS_MASK32HI(n) ((__uint32_t)-1 << (32 - (n))) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_MASK64HI) -__uint64_t xfs_mask64hi(int n); +static inline __uint32_t xfs_mask32hi(int n) +{ + return (__uint32_t)-1 << (32 - (n)); +} #define XFS_MASK64HI(n) xfs_mask64hi(n) -#else -#define XFS_MASK64HI(n) ((__uint64_t)-1 << (64 - (n))) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_MASK32LO) -__uint32_t xfs_mask32lo(int n); +static inline __uint64_t xfs_mask64hi(int n) +{ + return (__uint64_t)-1 << (64 - (n)); +} #define XFS_MASK32LO(n) xfs_mask32lo(n) -#else -#define XFS_MASK32LO(n) (((__uint32_t)1 << (n)) - 1) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_MASK64LO) -__uint64_t xfs_mask64lo(int n); +static inline __uint32_t xfs_mask32lo(int n) +{ + return ((__uint32_t)1 << (n)) - 1; +} #define XFS_MASK64LO(n) xfs_mask64lo(n) -#else -#define XFS_MASK64LO(n) (((__uint64_t)1 << (n)) - 1) -#endif +static inline __uint64_t xfs_mask64lo(int n) +{ + return ((__uint64_t)1 << (n)) - 1; +} /* Get high bit set out of 32-bit argument, -1 if none set */ extern int xfs_highbit32(__uint32_t v); diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c index 3e76def1283d..e415a4698e9c 100644 --- a/fs/xfs/xfs_bmap.c +++ b/fs/xfs/xfs_bmap.c @@ -1,68 +1,53 @@ /* - * Copyright (c) 2000-2004 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" - -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" #include "xfs_dir.h" #include "xfs_dir2.h" -#include "xfs_dmapi.h" -#include "xfs_mount.h" -#include "xfs_alloc_btree.h" +#include "xfs_da_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_btree.h" -#include "xfs_ialloc.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" -#include "xfs_inode_item.h" #include "xfs_inode.h" +#include "xfs_btree.h" +#include "xfs_dmapi.h" +#include "xfs_mount.h" +#include "xfs_ialloc.h" #include "xfs_itable.h" +#include "xfs_inode_item.h" #include "xfs_extfree_item.h" #include "xfs_alloc.h" #include "xfs_bmap.h" #include "xfs_rtalloc.h" #include "xfs_error.h" -#include "xfs_da_btree.h" #include "xfs_dir_leaf.h" -#include "xfs_bit.h" +#include "xfs_attr_leaf.h" #include "xfs_rw.h" #include "xfs_quota.h" #include "xfs_trans_space.h" @@ -438,6 +423,12 @@ xfs_bmap_count_leaves( int numrecs, int *count); +STATIC int +xfs_bmap_disk_count_leaves( + xfs_bmbt_rec_t *frp, + int numrecs, + int *count); + /* * Bmap internal routines. */ @@ -2772,8 +2763,8 @@ xfs_bmap_btree_to_extents( ASSERT(ifp->if_flags & XFS_IFEXTENTS); ASSERT(XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_BTREE); rblock = ifp->if_broot; - ASSERT(INT_GET(rblock->bb_level, ARCH_CONVERT) == 1); - ASSERT(INT_GET(rblock->bb_numrecs, ARCH_CONVERT) == 1); + ASSERT(be16_to_cpu(rblock->bb_level) == 1); + ASSERT(be16_to_cpu(rblock->bb_numrecs) == 1); ASSERT(XFS_BMAP_BROOT_MAXRECS(ifp->if_broot_bytes) == 1); mp = ip->i_mount; pp = XFS_BMAP_BROOT_PTR_ADDR(rblock, 1, ifp->if_broot_bytes); @@ -3216,11 +3207,11 @@ xfs_bmap_extents_to_btree( * Fill in the root. */ block = ifp->if_broot; - INT_SET(block->bb_magic, ARCH_CONVERT, XFS_BMAP_MAGIC); - INT_SET(block->bb_level, ARCH_CONVERT, 1); - INT_SET(block->bb_numrecs, ARCH_CONVERT, 1); - INT_SET(block->bb_leftsib, ARCH_CONVERT, NULLDFSBNO); - INT_SET(block->bb_rightsib, ARCH_CONVERT, NULLDFSBNO); + block->bb_magic = cpu_to_be32(XFS_BMAP_MAGIC); + block->bb_level = cpu_to_be16(1); + block->bb_numrecs = cpu_to_be16(1); + block->bb_leftsib = cpu_to_be64(NULLDFSBNO); + block->bb_rightsib = cpu_to_be64(NULLDFSBNO); /* * Need a cursor. Can't allocate until bb_level is filled in. */ @@ -3273,10 +3264,10 @@ xfs_bmap_extents_to_btree( * Fill in the child block. */ ablock = XFS_BUF_TO_BMBT_BLOCK(abp); - INT_SET(ablock->bb_magic, ARCH_CONVERT, XFS_BMAP_MAGIC); + ablock->bb_magic = cpu_to_be32(XFS_BMAP_MAGIC); ablock->bb_level = 0; - INT_SET(ablock->bb_leftsib, ARCH_CONVERT, NULLDFSBNO); - INT_SET(ablock->bb_rightsib, ARCH_CONVERT, NULLDFSBNO); + ablock->bb_leftsib = cpu_to_be64(NULLDFSBNO); + ablock->bb_rightsib = cpu_to_be64(NULLDFSBNO); arp = XFS_BMAP_REC_IADDR(ablock, 1, cur); nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t); for (ep = ifp->if_u1.if_extents, cnt = i = 0; i < nextents; i++, ep++) { @@ -3286,8 +3277,8 @@ xfs_bmap_extents_to_btree( arp++; cnt++; } } - INT_SET(ablock->bb_numrecs, ARCH_CONVERT, cnt); - ASSERT(INT_GET(ablock->bb_numrecs, ARCH_CONVERT) == XFS_IFORK_NEXTENTS(ip, whichfork)); + ASSERT(cnt == XFS_IFORK_NEXTENTS(ip, whichfork)); + ablock->bb_numrecs = cpu_to_be16(cnt); /* * Fill in the root key and pointer. */ @@ -3301,7 +3292,7 @@ xfs_bmap_extents_to_btree( * the root is at the right level. */ xfs_bmbt_log_block(cur, abp, XFS_BB_ALL_BITS); - xfs_bmbt_log_recs(cur, abp, 1, INT_GET(ablock->bb_numrecs, ARCH_CONVERT)); + xfs_bmbt_log_recs(cur, abp, 1, be16_to_cpu(ablock->bb_numrecs)); ASSERT(*curp == NULL); *curp = cur; *logflagsp = XFS_ILOG_CORE | XFS_ILOG_FBROOT(whichfork); @@ -3337,6 +3328,29 @@ xfs_bmap_insert_exlist( } /* + * Helper routine to reset inode di_forkoff field when switching + * attribute fork from local to extent format - we reset it where + * possible to make space available for inline data fork extents. + */ +STATIC void +xfs_bmap_forkoff_reset( + xfs_mount_t *mp, + xfs_inode_t *ip, + int whichfork) +{ + if (whichfork == XFS_ATTR_FORK && + (ip->i_d.di_format != XFS_DINODE_FMT_DEV) && + (ip->i_d.di_format != XFS_DINODE_FMT_UUID) && + ((mp->m_attroffset >> 3) > ip->i_d.di_forkoff)) { + ip->i_d.di_forkoff = mp->m_attroffset >> 3; + ip->i_df.if_ext_max = XFS_IFORK_DSIZE(ip) / + (uint)sizeof(xfs_bmbt_rec_t); + ip->i_afp->if_ext_max = XFS_IFORK_ASIZE(ip) / + (uint)sizeof(xfs_bmbt_rec_t); + } +} + +/* * Convert a local file to an extents file. * This code is out of bounds for data forks of regular files, * since the file data needs to get logged so things will stay consistent. @@ -3403,6 +3417,7 @@ xfs_bmap_local_to_extents( memcpy((char *)XFS_BUF_PTR(bp), ifp->if_u1.if_data, ifp->if_bytes); xfs_trans_log_buf(tp, bp, 0, ifp->if_bytes - 1); + xfs_bmap_forkoff_reset(args.mp, ip, whichfork); xfs_idata_realloc(ip, -ifp->if_bytes, whichfork); xfs_iext_realloc(ip, 1, whichfork); ep = ifp->if_u1.if_extents; @@ -3413,8 +3428,10 @@ xfs_bmap_local_to_extents( XFS_TRANS_MOD_DQUOT_BYINO(args.mp, tp, ip, XFS_TRANS_DQ_BCOUNT, 1L); flags |= XFS_ILOG_FEXT(whichfork); - } else + } else { ASSERT(XFS_IFORK_NEXTENTS(ip, whichfork) == 0); + xfs_bmap_forkoff_reset(ip->i_mount, ip, whichfork); + } ifp->if_flags &= ~XFS_IFINLINE; ifp->if_flags |= XFS_IFEXTENTS; XFS_IFORK_FMT_SET(ip, whichfork, XFS_DINODE_FMT_EXTENTS); @@ -3796,22 +3813,24 @@ xfs_bunmap_trace( int /* error code */ xfs_bmap_add_attrfork( xfs_inode_t *ip, /* incore inode pointer */ - int rsvd) /* OK to allocated reserved blocks in trans */ + int size, /* space new attribute needs */ + int rsvd) /* xact may use reserved blks */ { - int blks; /* space reservation */ - int committed; /* xaction was committed */ - int error; /* error return value */ xfs_fsblock_t firstblock; /* 1st block/ag allocated */ xfs_bmap_free_t flist; /* freed extent list */ - int logflags; /* logging flags */ xfs_mount_t *mp; /* mount structure */ - unsigned long s; /* spinlock spl value */ xfs_trans_t *tp; /* transaction pointer */ + unsigned long s; /* spinlock spl value */ + int blks; /* space reservation */ + int version = 1; /* superblock attr version */ + int committed; /* xaction was committed */ + int logflags; /* logging flags */ + int error; /* error return value */ + ASSERT(XFS_IFORK_Q(ip) == 0); ASSERT(ip->i_df.if_ext_max == XFS_IFORK_DSIZE(ip) / (uint)sizeof(xfs_bmbt_rec_t)); - if (XFS_IFORK_Q(ip)) - return 0; + mp = ip->i_mount; ASSERT(!XFS_NOT_DQATTACHED(mp, ip)); tp = xfs_trans_alloc(mp, XFS_TRANS_ADDAFORK); @@ -3853,7 +3872,11 @@ xfs_bmap_add_attrfork( case XFS_DINODE_FMT_LOCAL: case XFS_DINODE_FMT_EXTENTS: case XFS_DINODE_FMT_BTREE: - ip->i_d.di_forkoff = mp->m_attroffset >> 3; + ip->i_d.di_forkoff = xfs_attr_shortform_bytesfit(ip, size); + if (!ip->i_d.di_forkoff) + ip->i_d.di_forkoff = mp->m_attroffset >> 3; + else if (!(mp->m_flags & XFS_MOUNT_COMPAT_ATTR)) + version = 2; break; default: ASSERT(0); @@ -3890,12 +3913,22 @@ xfs_bmap_add_attrfork( xfs_trans_log_inode(tp, ip, logflags); if (error) goto error2; - if (!XFS_SB_VERSION_HASATTR(&mp->m_sb)) { + if (!XFS_SB_VERSION_HASATTR(&mp->m_sb) || + (!XFS_SB_VERSION_HASATTR2(&mp->m_sb) && version == 2)) { + __int64_t sbfields = 0; + s = XFS_SB_LOCK(mp); if (!XFS_SB_VERSION_HASATTR(&mp->m_sb)) { XFS_SB_VERSION_ADDATTR(&mp->m_sb); + sbfields |= XFS_SB_VERSIONNUM; + } + if (!XFS_SB_VERSION_HASATTR2(&mp->m_sb) && version == 2) { + XFS_SB_VERSION_ADDATTR2(&mp->m_sb); + sbfields |= (XFS_SB_VERSIONNUM | XFS_SB_FEATURES2); + } + if (sbfields) { XFS_SB_UNLOCK(mp, s); - xfs_mod_sb(tp, XFS_SB_VERSIONNUM); + xfs_mod_sb(tp, sbfields); } else XFS_SB_UNLOCK(mp, s); } @@ -3988,13 +4021,19 @@ xfs_bmap_compute_maxlevels( * (a signed 32-bit number, xfs_extnum_t), or by di_anextents * (a signed 16-bit number, xfs_aextnum_t). */ - maxleafents = (whichfork == XFS_DATA_FORK) ? MAXEXTNUM : MAXAEXTNUM; + if (whichfork == XFS_DATA_FORK) { + maxleafents = MAXEXTNUM; + sz = (mp->m_flags & XFS_MOUNT_COMPAT_ATTR) ? + mp->m_attroffset : XFS_BMDR_SPACE_CALC(MINDBTPTRS); + } else { + maxleafents = MAXAEXTNUM; + sz = (mp->m_flags & XFS_MOUNT_COMPAT_ATTR) ? + mp->m_sb.sb_inodesize - mp->m_attroffset : + XFS_BMDR_SPACE_CALC(MINABTPTRS); + } + maxrootrecs = (int)XFS_BTREE_BLOCK_MAXRECS(sz, xfs_bmdr, 0); minleafrecs = mp->m_bmap_dmnr[0]; minnoderecs = mp->m_bmap_dmnr[1]; - sz = (whichfork == XFS_DATA_FORK) ? - mp->m_attroffset : - mp->m_sb.sb_inodesize - mp->m_attroffset; - maxrootrecs = (int)XFS_BTREE_BLOCK_MAXRECS(sz, xfs_bmdr, 0); maxblocks = (maxleafents + minleafrecs - 1) / minleafrecs; for (level = 1; maxblocks > 1; level++) { if (maxblocks <= maxrootrecs) @@ -4332,8 +4371,8 @@ xfs_bmap_read_extents( /* * Root level must use BMAP_BROOT_PTR_ADDR macro to get ptr out. */ - ASSERT(INT_GET(block->bb_level, ARCH_CONVERT) > 0); - level = INT_GET(block->bb_level, ARCH_CONVERT); + level = be16_to_cpu(block->bb_level); + ASSERT(level > 0); pp = XFS_BMAP_BROOT_PTR_ADDR(block, 1, ifp->if_broot_bytes); ASSERT(INT_GET(*pp, ARCH_CONVERT) != NULLDFSBNO); ASSERT(XFS_FSB_TO_AGNO(mp, INT_GET(*pp, ARCH_CONVERT)) < mp->m_sb.sb_agcount); @@ -4376,7 +4415,7 @@ xfs_bmap_read_extents( xfs_extnum_t num_recs; - num_recs = INT_GET(block->bb_numrecs, ARCH_CONVERT); + num_recs = be16_to_cpu(block->bb_numrecs); if (unlikely(i + num_recs > room)) { ASSERT(i + num_recs <= room); xfs_fs_cmn_err(CE_WARN, ip->i_mount, @@ -4393,7 +4432,7 @@ xfs_bmap_read_extents( /* * Read-ahead the next leaf block, if any. */ - nextbno = INT_GET(block->bb_rightsib, ARCH_CONVERT); + nextbno = be64_to_cpu(block->bb_rightsib); if (nextbno != NULLFSBLOCK) xfs_btree_reada_bufl(mp, nextbno, 1); /* @@ -4650,7 +4689,7 @@ xfs_bmapi( } if (wr && *firstblock == NULLFSBLOCK) { if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_BTREE) - minleft = INT_GET(ifp->if_broot->bb_level, ARCH_CONVERT) + 1; + minleft = be16_to_cpu(ifp->if_broot->bb_level) + 1; else minleft = 1; } else @@ -5692,12 +5731,13 @@ xfs_getbmap( out.bmv_offset = XFS_FSB_TO_BB(mp, map[i].br_startoff); out.bmv_length = XFS_FSB_TO_BB(mp, map[i].br_blockcount); ASSERT(map[i].br_startblock != DELAYSTARTBLOCK); - if (prealloced && - map[i].br_startblock == HOLESTARTBLOCK && - out.bmv_offset + out.bmv_length == bmvend) { - /* - * came to hole at end of file - */ + if (map[i].br_startblock == HOLESTARTBLOCK && + ((prealloced && out.bmv_offset + out.bmv_length == bmvend) || + whichfork == XFS_ATTR_FORK )) { + /* + * came to hole at end of file or the end of + attribute fork + */ goto unlock_and_return; } else { out.bmv_block = @@ -5927,10 +5967,10 @@ xfs_check_block( xfs_bmbt_ptr_t *pp, *thispa; /* pointer to block address */ xfs_bmbt_key_t *prevp, *keyp; - ASSERT(INT_GET(block->bb_level, ARCH_CONVERT) > 0); + ASSERT(be16_to_cpu(block->bb_level) > 0); prevp = NULL; - for( i = 1; i <= INT_GET(block->bb_numrecs, ARCH_CONVERT);i++) { + for( i = 1; i <= be16_to_cpu(block->bb_numrecs); i++) { dmxr = mp->m_bmap_dmxr[0]; if (root) { @@ -5955,7 +5995,7 @@ xfs_check_block( pp = XFS_BTREE_PTR_ADDR(mp->m_sb.sb_blocksize, xfs_bmbt, block, i, dmxr); } - for (j = i+1; j <= INT_GET(block->bb_numrecs, ARCH_CONVERT); j++) { + for (j = i+1; j <= be16_to_cpu(block->bb_numrecs); j++) { if (root) { thispa = XFS_BMAP_BROOT_PTR_ADDR(block, j, sz); } else { @@ -6008,8 +6048,8 @@ xfs_bmap_check_leaf_extents( /* * Root level must use BMAP_BROOT_PTR_ADDR macro to get ptr out. */ - ASSERT(INT_GET(block->bb_level, ARCH_CONVERT) > 0); - level = INT_GET(block->bb_level, ARCH_CONVERT); + level = be16_to_cpu(block->bb_level); + ASSERT(level > 0); xfs_check_block(block, mp, 1, ifp->if_broot_bytes); pp = XFS_BMAP_BROOT_PTR_ADDR(block, 1, ifp->if_broot_bytes); ASSERT(INT_GET(*pp, ARCH_CONVERT) != NULLDFSBNO); @@ -6069,13 +6109,13 @@ xfs_bmap_check_leaf_extents( xfs_extnum_t num_recs; - num_recs = INT_GET(block->bb_numrecs, ARCH_CONVERT); + num_recs = be16_to_cpu(block->bb_numrecs); /* * Read-ahead the next leaf block, if any. */ - nextbno = INT_GET(block->bb_rightsib, ARCH_CONVERT); + nextbno = be64_to_cpu(block->bb_rightsib); /* * Check all the extents to make sure they are OK. @@ -6131,7 +6171,7 @@ error0: xfs_trans_brelse(NULL, bp); error_norelse: cmn_err(CE_WARN, "%s: BAD after btree leaves for %d extents", - i, __FUNCTION__); + __FUNCTION__, i); panic("%s: CORRUPTED BTREE OR SOMETHING", __FUNCTION__); return; } @@ -6172,8 +6212,8 @@ xfs_bmap_count_blocks( * Root level must use BMAP_BROOT_PTR_ADDR macro to get ptr out. */ block = ifp->if_broot; - ASSERT(INT_GET(block->bb_level, ARCH_CONVERT) > 0); - level = INT_GET(block->bb_level, ARCH_CONVERT); + level = be16_to_cpu(block->bb_level); + ASSERT(level > 0); pp = XFS_BMAP_BROOT_PTR_ADDR(block, 1, ifp->if_broot_bytes); ASSERT(INT_GET(*pp, ARCH_CONVERT) != NULLDFSBNO); ASSERT(XFS_FSB_TO_AGNO(mp, INT_GET(*pp, ARCH_CONVERT)) < mp->m_sb.sb_agcount); @@ -6218,14 +6258,14 @@ xfs_bmap_count_tree( if (--level) { /* Not at node above leafs, count this level of nodes */ - nextbno = INT_GET(block->bb_rightsib, ARCH_CONVERT); + nextbno = be64_to_cpu(block->bb_rightsib); while (nextbno != NULLFSBLOCK) { if ((error = xfs_btree_read_bufl(mp, tp, nextbno, 0, &nbp, XFS_BMAP_BTREE_REF))) return error; *count += 1; nextblock = XFS_BUF_TO_BMBT_BLOCK(nbp); - nextbno = INT_GET(nextblock->bb_rightsib, ARCH_CONVERT); + nextbno = be64_to_cpu(nextblock->bb_rightsib); xfs_trans_brelse(tp, nbp); } @@ -6244,11 +6284,11 @@ xfs_bmap_count_tree( } else { /* count all level 1 nodes and their leaves */ for (;;) { - nextbno = INT_GET(block->bb_rightsib, ARCH_CONVERT); - numrecs = INT_GET(block->bb_numrecs, ARCH_CONVERT); + nextbno = be64_to_cpu(block->bb_rightsib); + numrecs = be16_to_cpu(block->bb_numrecs); frp = XFS_BTREE_REC_ADDR(mp->m_sb.sb_blocksize, xfs_bmbt, block, 1, mp->m_bmap_dmxr[0]); - if (unlikely(xfs_bmap_count_leaves(frp, numrecs, count) < 0)) { + if (unlikely(xfs_bmap_disk_count_leaves(frp, numrecs, count) < 0)) { xfs_trans_brelse(tp, bp); XFS_ERROR_REPORT("xfs_bmap_count_tree(2)", XFS_ERRLEVEL_LOW, mp); @@ -6280,6 +6320,22 @@ xfs_bmap_count_leaves( int b; for ( b = 1; b <= numrecs; b++, frp++) + *count += xfs_bmbt_get_blockcount(frp); + return 0; +} + +/* + * Count leaf blocks given a pointer to an extent list originally in btree format. + */ +int +xfs_bmap_disk_count_leaves( + xfs_bmbt_rec_t *frp, + int numrecs, + int *count) +{ + int b; + + for ( b = 1; b <= numrecs; b++, frp++) *count += xfs_bmbt_disk_get_blockcount(frp); return 0; } diff --git a/fs/xfs/xfs_bmap.h b/fs/xfs/xfs_bmap.h index e6d22ec9b2e4..2e0717a01309 100644 --- a/fs/xfs/xfs_bmap.h +++ b/fs/xfs/xfs_bmap.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_BMAP_H__ #define __XFS_BMAP_H__ @@ -77,12 +63,11 @@ typedef struct xfs_bmap_free /* combine contig. space */ #define XFS_BMAPI_CONTIG 0x400 /* must allocate only one extent */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BMAPI_AFLAG) -int xfs_bmapi_aflag(int w); #define XFS_BMAPI_AFLAG(w) xfs_bmapi_aflag(w) -#else -#define XFS_BMAPI_AFLAG(w) ((w) == XFS_ATTR_FORK ? XFS_BMAPI_ATTRFORK : 0) -#endif +static inline int xfs_bmapi_aflag(int w) +{ + return (w == XFS_ATTR_FORK ? XFS_BMAPI_ATTRFORK : 0); +} /* * Special values for xfs_bmbt_irec_t br_startblock field. @@ -90,14 +75,12 @@ int xfs_bmapi_aflag(int w); #define DELAYSTARTBLOCK ((xfs_fsblock_t)-1LL) #define HOLESTARTBLOCK ((xfs_fsblock_t)-2LL) -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BMAP_INIT) -void xfs_bmap_init(xfs_bmap_free_t *flp, xfs_fsblock_t *fbp); #define XFS_BMAP_INIT(flp,fbp) xfs_bmap_init(flp,fbp) -#else -#define XFS_BMAP_INIT(flp,fbp) \ +static inline void xfs_bmap_init(xfs_bmap_free_t *flp, xfs_fsblock_t *fbp) +{ ((flp)->xbf_first = NULL, (flp)->xbf_count = 0, \ - (flp)->xbf_low = 0, *(fbp) = NULLFSBLOCK) -#endif + (flp)->xbf_low = 0, *(fbp) = NULLFSBLOCK); +} /* * Argument structure for xfs_bmap_alloc. @@ -156,7 +139,8 @@ xfs_bmap_trace_exlist( int /* error code */ xfs_bmap_add_attrfork( struct xfs_inode *ip, /* incore inode pointer */ - int rsvd); /* flag for reserved block allocation */ + int size, /* space needed for new attribute */ + int rsvd); /* flag for reserved block allocation */ /* * Add the extent to the list of extents to be free at transaction end. diff --git a/fs/xfs/xfs_bmap_btree.c b/fs/xfs/xfs_bmap_btree.c index 09a77b17565b..3f1383d160e8 100644 --- a/fs/xfs/xfs_bmap_btree.c +++ b/fs/xfs/xfs_bmap_btree.c @@ -1,41 +1,26 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" - -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" @@ -43,20 +28,19 @@ #include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" -#include "xfs_alloc_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_btree.h" -#include "xfs_ialloc.h" -#include "xfs_itable.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" -#include "xfs_inode_item.h" #include "xfs_inode.h" +#include "xfs_inode_item.h" #include "xfs_alloc.h" -#include "xfs_bit.h" +#include "xfs_btree.h" +#include "xfs_ialloc.h" +#include "xfs_itable.h" #include "xfs_bmap.h" #include "xfs_error.h" #include "xfs_quota.h" @@ -382,7 +366,7 @@ xfs_bmbt_delrec( return 0; } block = xfs_bmbt_get_block(cur, level, &bp); - numrecs = INT_GET(block->bb_numrecs, ARCH_CONVERT); + numrecs = be16_to_cpu(block->bb_numrecs); #ifdef DEBUG if ((error = xfs_btree_check_lblock(cur, block, level, bp))) { XFS_BMBT_TRACE_CURSOR(cur, ERROR); @@ -427,7 +411,7 @@ xfs_bmbt_delrec( } } numrecs--; - INT_SET(block->bb_numrecs, ARCH_CONVERT, numrecs); + block->bb_numrecs = cpu_to_be16(numrecs); xfs_bmbt_log_block(cur, bp, XFS_BB_NUMRECS); /* * We're at the root level. @@ -463,8 +447,8 @@ xfs_bmbt_delrec( *stat = 1; return 0; } - rbno = INT_GET(block->bb_rightsib, ARCH_CONVERT); - lbno = INT_GET(block->bb_leftsib, ARCH_CONVERT); + rbno = be64_to_cpu(block->bb_rightsib); + lbno = be64_to_cpu(block->bb_leftsib); /* * One child of root, need to get a chance to copy its contents * into the root and delete it. Can't go up to next level, @@ -508,15 +492,15 @@ xfs_bmbt_delrec( goto error0; } #endif - bno = INT_GET(right->bb_leftsib, ARCH_CONVERT); - if (INT_GET(right->bb_numrecs, ARCH_CONVERT) - 1 >= + bno = be64_to_cpu(right->bb_leftsib); + if (be16_to_cpu(right->bb_numrecs) - 1 >= XFS_BMAP_BLOCK_IMINRECS(level, cur)) { if ((error = xfs_bmbt_lshift(tcur, level, &i))) { XFS_BMBT_TRACE_CURSOR(cur, ERROR); goto error0; } if (i) { - ASSERT(INT_GET(block->bb_numrecs, ARCH_CONVERT) >= + ASSERT(be16_to_cpu(block->bb_numrecs) >= XFS_BMAP_BLOCK_IMINRECS(level, tcur)); xfs_btree_del_cursor(tcur, XFS_BTREE_NOERROR); tcur = NULL; @@ -533,7 +517,7 @@ xfs_bmbt_delrec( return 0; } } - rrecs = INT_GET(right->bb_numrecs, ARCH_CONVERT); + rrecs = be16_to_cpu(right->bb_numrecs); if (lbno != NULLFSBLOCK) { i = xfs_btree_firstrec(tcur, level); XFS_WANT_CORRUPTED_GOTO(i == 1, error0); @@ -564,15 +548,15 @@ xfs_bmbt_delrec( goto error0; } #endif - bno = INT_GET(left->bb_rightsib, ARCH_CONVERT); - if (INT_GET(left->bb_numrecs, ARCH_CONVERT) - 1 >= + bno = be64_to_cpu(left->bb_rightsib); + if (be16_to_cpu(left->bb_numrecs) - 1 >= XFS_BMAP_BLOCK_IMINRECS(level, cur)) { if ((error = xfs_bmbt_rshift(tcur, level, &i))) { XFS_BMBT_TRACE_CURSOR(cur, ERROR); goto error0; } if (i) { - ASSERT(INT_GET(block->bb_numrecs, ARCH_CONVERT) >= + ASSERT(be16_to_cpu(block->bb_numrecs) >= XFS_BMAP_BLOCK_IMINRECS(level, tcur)); xfs_btree_del_cursor(tcur, XFS_BTREE_NOERROR); tcur = NULL; @@ -583,14 +567,14 @@ xfs_bmbt_delrec( return 0; } } - lrecs = INT_GET(left->bb_numrecs, ARCH_CONVERT); + lrecs = be16_to_cpu(left->bb_numrecs); } xfs_btree_del_cursor(tcur, XFS_BTREE_NOERROR); tcur = NULL; mp = cur->bc_mp; ASSERT(bno != NULLFSBLOCK); if (lbno != NULLFSBLOCK && - lrecs + INT_GET(block->bb_numrecs, ARCH_CONVERT) <= XFS_BMAP_BLOCK_IMAXRECS(level, cur)) { + lrecs + be16_to_cpu(block->bb_numrecs) <= XFS_BMAP_BLOCK_IMAXRECS(level, cur)) { rbno = bno; right = block; rbp = bp; @@ -605,7 +589,7 @@ xfs_bmbt_delrec( goto error0; } } else if (rbno != NULLFSBLOCK && - rrecs + INT_GET(block->bb_numrecs, ARCH_CONVERT) <= + rrecs + be16_to_cpu(block->bb_numrecs) <= XFS_BMAP_BLOCK_IMAXRECS(level, cur)) { lbno = bno; left = block; @@ -620,7 +604,7 @@ xfs_bmbt_delrec( XFS_BMBT_TRACE_CURSOR(cur, ERROR); goto error0; } - lrecs = INT_GET(left->bb_numrecs, ARCH_CONVERT); + lrecs = be16_to_cpu(left->bb_numrecs); } else { if (level > 0 && (error = xfs_bmbt_decrement(cur, level, &i))) { XFS_BMBT_TRACE_CURSOR(cur, ERROR); @@ -630,8 +614,8 @@ xfs_bmbt_delrec( *stat = 1; return 0; } - numlrecs = INT_GET(left->bb_numrecs, ARCH_CONVERT); - numrrecs = INT_GET(right->bb_numrecs, ARCH_CONVERT); + numlrecs = be16_to_cpu(left->bb_numrecs); + numrrecs = be16_to_cpu(right->bb_numrecs); if (level > 0) { lkp = XFS_BMAP_KEY_IADDR(left, numlrecs + 1, cur); lpp = XFS_BMAP_PTR_IADDR(left, numlrecs + 1, cur); @@ -655,12 +639,12 @@ xfs_bmbt_delrec( memcpy(lrp, rrp, numrrecs * sizeof(*lrp)); xfs_bmbt_log_recs(cur, lbp, numlrecs + 1, numlrecs + numrrecs); } - INT_MOD(left->bb_numrecs, ARCH_CONVERT, numrrecs); - left->bb_rightsib = right->bb_rightsib; /* INT_: direct copy */ + be16_add(&left->bb_numrecs, numrrecs); + left->bb_rightsib = right->bb_rightsib; xfs_bmbt_log_block(cur, lbp, XFS_BB_RIGHTSIB | XFS_BB_NUMRECS); - if (INT_GET(left->bb_rightsib, ARCH_CONVERT) != NULLDFSBNO) { + if (be64_to_cpu(left->bb_rightsib) != NULLDFSBNO) { if ((error = xfs_btree_read_bufl(mp, cur->bc_tp, - INT_GET(left->bb_rightsib, ARCH_CONVERT), + be64_to_cpu(left->bb_rightsib), 0, &rrbp, XFS_BMAP_BTREE_REF))) { XFS_BMBT_TRACE_CURSOR(cur, ERROR); goto error0; @@ -670,7 +654,7 @@ xfs_bmbt_delrec( XFS_BMBT_TRACE_CURSOR(cur, ERROR); goto error0; } - INT_SET(rrblock->bb_leftsib, ARCH_CONVERT, lbno); + rrblock->bb_leftsib = cpu_to_be64(lbno); xfs_bmbt_log_block(cur, rrbp, XFS_BB_LEFTSIB); } xfs_bmap_add_free(XFS_DADDR_TO_FSB(mp, XFS_BUF_ADDR(rbp)), 1, @@ -727,7 +711,7 @@ xfs_bmbt_get_rec( if ((error = xfs_btree_check_lblock(cur, block, 0, bp))) return error; #endif - if (ptr > INT_GET(block->bb_numrecs, ARCH_CONVERT) || ptr <= 0) { + if (ptr > be16_to_cpu(block->bb_numrecs) || ptr <= 0) { *stat = 0; return 0; } @@ -788,7 +772,7 @@ xfs_bmbt_insrec( } XFS_STATS_INC(xs_bmbt_insrec); block = xfs_bmbt_get_block(cur, level, &bp); - numrecs = INT_GET(block->bb_numrecs, ARCH_CONVERT); + numrecs = be16_to_cpu(block->bb_numrecs); #ifdef DEBUG if ((error = xfs_btree_check_lblock(cur, block, level, bp))) { XFS_BMBT_TRACE_CURSOR(cur, ERROR); @@ -870,7 +854,7 @@ xfs_bmbt_insrec( } } } - numrecs = INT_GET(block->bb_numrecs, ARCH_CONVERT); + numrecs = be16_to_cpu(block->bb_numrecs); if (level > 0) { kp = XFS_BMAP_KEY_IADDR(block, 1, cur); pp = XFS_BMAP_PTR_IADDR(block, 1, cur); @@ -897,7 +881,7 @@ xfs_bmbt_insrec( kp[ptr - 1] = key; INT_SET(pp[ptr - 1], ARCH_CONVERT, *bnop); numrecs++; - INT_SET(block->bb_numrecs, ARCH_CONVERT, numrecs); + block->bb_numrecs = cpu_to_be16(numrecs); xfs_bmbt_log_keys(cur, bp, ptr, numrecs); xfs_bmbt_log_ptrs(cur, bp, ptr, numrecs); } else { @@ -906,7 +890,7 @@ xfs_bmbt_insrec( (numrecs - ptr + 1) * sizeof(*rp)); rp[ptr - 1] = *recp; numrecs++; - INT_SET(block->bb_numrecs, ARCH_CONVERT, numrecs); + block->bb_numrecs = cpu_to_be16(numrecs); xfs_bmbt_log_recs(cur, bp, ptr, numrecs); } xfs_bmbt_log_block(cur, bp, XFS_BB_NUMRECS); @@ -971,7 +955,7 @@ xfs_bmbt_killroot( /* * Give up if the root has multiple children. */ - if (INT_GET(block->bb_numrecs, ARCH_CONVERT) != 1) { + if (be16_to_cpu(block->bb_numrecs) != 1) { XFS_BMBT_TRACE_CURSOR(cur, EXIT); return 0; } @@ -982,37 +966,37 @@ xfs_bmbt_killroot( */ cbp = cur->bc_bufs[level - 1]; cblock = XFS_BUF_TO_BMBT_BLOCK(cbp); - if (INT_GET(cblock->bb_numrecs, ARCH_CONVERT) > XFS_BMAP_BLOCK_DMAXRECS(level, cur)) { + if (be16_to_cpu(cblock->bb_numrecs) > XFS_BMAP_BLOCK_DMAXRECS(level, cur)) { XFS_BMBT_TRACE_CURSOR(cur, EXIT); return 0; } - ASSERT(INT_GET(cblock->bb_leftsib, ARCH_CONVERT) == NULLDFSBNO); - ASSERT(INT_GET(cblock->bb_rightsib, ARCH_CONVERT) == NULLDFSBNO); + ASSERT(be64_to_cpu(cblock->bb_leftsib) == NULLDFSBNO); + ASSERT(be64_to_cpu(cblock->bb_rightsib) == NULLDFSBNO); ip = cur->bc_private.b.ip; ifp = XFS_IFORK_PTR(ip, cur->bc_private.b.whichfork); ASSERT(XFS_BMAP_BLOCK_IMAXRECS(level, cur) == XFS_BMAP_BROOT_MAXRECS(ifp->if_broot_bytes)); - i = (int)(INT_GET(cblock->bb_numrecs, ARCH_CONVERT) - XFS_BMAP_BLOCK_IMAXRECS(level, cur)); + i = (int)(be16_to_cpu(cblock->bb_numrecs) - XFS_BMAP_BLOCK_IMAXRECS(level, cur)); if (i) { xfs_iroot_realloc(ip, i, cur->bc_private.b.whichfork); block = ifp->if_broot; } - INT_MOD(block->bb_numrecs, ARCH_CONVERT, i); - ASSERT(INT_GET(block->bb_numrecs, ARCH_CONVERT) == INT_GET(cblock->bb_numrecs, ARCH_CONVERT)); + be16_add(&block->bb_numrecs, i); + ASSERT(block->bb_numrecs == cblock->bb_numrecs); kp = XFS_BMAP_KEY_IADDR(block, 1, cur); ckp = XFS_BMAP_KEY_IADDR(cblock, 1, cur); - memcpy(kp, ckp, INT_GET(block->bb_numrecs, ARCH_CONVERT) * sizeof(*kp)); + memcpy(kp, ckp, be16_to_cpu(block->bb_numrecs) * sizeof(*kp)); pp = XFS_BMAP_PTR_IADDR(block, 1, cur); cpp = XFS_BMAP_PTR_IADDR(cblock, 1, cur); #ifdef DEBUG - for (i = 0; i < INT_GET(cblock->bb_numrecs, ARCH_CONVERT); i++) { + for (i = 0; i < be16_to_cpu(cblock->bb_numrecs); i++) { if ((error = xfs_btree_check_lptr(cur, INT_GET(cpp[i], ARCH_CONVERT), level - 1))) { XFS_BMBT_TRACE_CURSOR(cur, ERROR); return error; } } #endif - memcpy(pp, cpp, INT_GET(block->bb_numrecs, ARCH_CONVERT) * sizeof(*pp)); + memcpy(pp, cpp, be16_to_cpu(block->bb_numrecs) * sizeof(*pp)); xfs_bmap_add_free(XFS_DADDR_TO_FSB(cur->bc_mp, XFS_BUF_ADDR(cbp)), 1, cur->bc_private.b.flist, cur->bc_mp); ip->i_d.di_nblocks--; @@ -1020,7 +1004,7 @@ xfs_bmbt_killroot( XFS_TRANS_DQ_BCOUNT, -1L); xfs_trans_binval(cur->bc_tp, cbp); cur->bc_bufs[level - 1] = NULL; - INT_MOD(block->bb_level, ARCH_CONVERT, -1); + be16_add(&block->bb_level, -1); xfs_trans_log_inode(cur->bc_tp, ip, XFS_ILOG_CORE | XFS_ILOG_FBROOT(cur->bc_private.b.whichfork)); cur->bc_nlevels--; @@ -1176,7 +1160,7 @@ xfs_bmbt_lookup( else krbase = XFS_BMAP_REC_IADDR(block, 1, cur); low = 1; - if (!(high = INT_GET(block->bb_numrecs, ARCH_CONVERT))) { + if (!(high = be16_to_cpu(block->bb_numrecs))) { ASSERT(level == 0); cur->bc_ptrs[0] = dir != XFS_LOOKUP_LE; XFS_BMBT_TRACE_CURSOR(cur, EXIT); @@ -1223,8 +1207,8 @@ xfs_bmbt_lookup( * If ge search and we went off the end of the block, but it's * not the last block, we're in the wrong block. */ - if (dir == XFS_LOOKUP_GE && keyno > INT_GET(block->bb_numrecs, ARCH_CONVERT) && - INT_GET(block->bb_rightsib, ARCH_CONVERT) != NULLDFSBNO) { + if (dir == XFS_LOOKUP_GE && keyno > be16_to_cpu(block->bb_numrecs) && + be64_to_cpu(block->bb_rightsib) != NULLDFSBNO) { cur->bc_ptrs[0] = keyno; if ((error = xfs_bmbt_increment(cur, 0, &i))) { XFS_BMBT_TRACE_CURSOR(cur, ERROR); @@ -1239,7 +1223,7 @@ xfs_bmbt_lookup( else if (dir == XFS_LOOKUP_LE && diff > 0) keyno--; cur->bc_ptrs[0] = keyno; - if (keyno == 0 || keyno > INT_GET(block->bb_numrecs, ARCH_CONVERT)) { + if (keyno == 0 || keyno > be16_to_cpu(block->bb_numrecs)) { XFS_BMBT_TRACE_CURSOR(cur, EXIT); *stat = 0; } else { @@ -1296,7 +1280,7 @@ xfs_bmbt_lshift( return error; } #endif - if (INT_GET(right->bb_leftsib, ARCH_CONVERT) == NULLDFSBNO) { + if (be64_to_cpu(right->bb_leftsib) == NULLDFSBNO) { XFS_BMBT_TRACE_CURSOR(cur, EXIT); *stat = 0; return 0; @@ -1307,7 +1291,7 @@ xfs_bmbt_lshift( return 0; } mp = cur->bc_mp; - if ((error = xfs_btree_read_bufl(mp, cur->bc_tp, INT_GET(right->bb_leftsib, ARCH_CONVERT), 0, + if ((error = xfs_btree_read_bufl(mp, cur->bc_tp, be64_to_cpu(right->bb_leftsib), 0, &lbp, XFS_BMAP_BTREE_REF))) { XFS_BMBT_TRACE_CURSOR(cur, ERROR); return error; @@ -1317,12 +1301,12 @@ xfs_bmbt_lshift( XFS_BMBT_TRACE_CURSOR(cur, ERROR); return error; } - if (INT_GET(left->bb_numrecs, ARCH_CONVERT) == XFS_BMAP_BLOCK_IMAXRECS(level, cur)) { + if (be16_to_cpu(left->bb_numrecs) == XFS_BMAP_BLOCK_IMAXRECS(level, cur)) { XFS_BMBT_TRACE_CURSOR(cur, EXIT); *stat = 0; return 0; } - lrecs = INT_GET(left->bb_numrecs, ARCH_CONVERT) + 1; + lrecs = be16_to_cpu(left->bb_numrecs) + 1; if (level > 0) { lkp = XFS_BMAP_KEY_IADDR(left, lrecs, cur); rkp = XFS_BMAP_KEY_IADDR(right, 1, cur); @@ -1344,7 +1328,7 @@ xfs_bmbt_lshift( *lrp = *rrp; xfs_bmbt_log_recs(cur, lbp, lrecs, lrecs); } - INT_SET(left->bb_numrecs, ARCH_CONVERT, lrecs); + left->bb_numrecs = cpu_to_be16(lrecs); xfs_bmbt_log_block(cur, lbp, XFS_BB_NUMRECS); #ifdef DEBUG if (level > 0) @@ -1352,8 +1336,8 @@ xfs_bmbt_lshift( else xfs_btree_check_rec(XFS_BTNUM_BMAP, lrp - 1, lrp); #endif - rrecs = INT_GET(right->bb_numrecs, ARCH_CONVERT) - 1; - INT_SET(right->bb_numrecs, ARCH_CONVERT, rrecs); + rrecs = be16_to_cpu(right->bb_numrecs) - 1; + right->bb_numrecs = cpu_to_be16(rrecs); xfs_bmbt_log_block(cur, rbp, XFS_BB_NUMRECS); if (level > 0) { #ifdef DEBUG @@ -1430,18 +1414,18 @@ xfs_bmbt_rshift( return error; } #endif - if (INT_GET(left->bb_rightsib, ARCH_CONVERT) == NULLDFSBNO) { + if (be64_to_cpu(left->bb_rightsib) == NULLDFSBNO) { XFS_BMBT_TRACE_CURSOR(cur, EXIT); *stat = 0; return 0; } - if (cur->bc_ptrs[level] >= INT_GET(left->bb_numrecs, ARCH_CONVERT)) { + if (cur->bc_ptrs[level] >= be16_to_cpu(left->bb_numrecs)) { XFS_BMBT_TRACE_CURSOR(cur, EXIT); *stat = 0; return 0; } mp = cur->bc_mp; - if ((error = xfs_btree_read_bufl(mp, cur->bc_tp, INT_GET(left->bb_rightsib, ARCH_CONVERT), 0, + if ((error = xfs_btree_read_bufl(mp, cur->bc_tp, be64_to_cpu(left->bb_rightsib), 0, &rbp, XFS_BMAP_BTREE_REF))) { XFS_BMBT_TRACE_CURSOR(cur, ERROR); return error; @@ -1451,26 +1435,26 @@ xfs_bmbt_rshift( XFS_BMBT_TRACE_CURSOR(cur, ERROR); return error; } - if (INT_GET(right->bb_numrecs, ARCH_CONVERT) == XFS_BMAP_BLOCK_IMAXRECS(level, cur)) { + if (be16_to_cpu(right->bb_numrecs) == XFS_BMAP_BLOCK_IMAXRECS(level, cur)) { XFS_BMBT_TRACE_CURSOR(cur, EXIT); *stat = 0; return 0; } if (level > 0) { - lkp = XFS_BMAP_KEY_IADDR(left, INT_GET(left->bb_numrecs, ARCH_CONVERT), cur); - lpp = XFS_BMAP_PTR_IADDR(left, INT_GET(left->bb_numrecs, ARCH_CONVERT), cur); + lkp = XFS_BMAP_KEY_IADDR(left, be16_to_cpu(left->bb_numrecs), cur); + lpp = XFS_BMAP_PTR_IADDR(left, be16_to_cpu(left->bb_numrecs), cur); rkp = XFS_BMAP_KEY_IADDR(right, 1, cur); rpp = XFS_BMAP_PTR_IADDR(right, 1, cur); #ifdef DEBUG - for (i = INT_GET(right->bb_numrecs, ARCH_CONVERT) - 1; i >= 0; i--) { + for (i = be16_to_cpu(right->bb_numrecs) - 1; i >= 0; i--) { if ((error = xfs_btree_check_lptr(cur, INT_GET(rpp[i], ARCH_CONVERT), level))) { XFS_BMBT_TRACE_CURSOR(cur, ERROR); return error; } } #endif - memmove(rkp + 1, rkp, INT_GET(right->bb_numrecs, ARCH_CONVERT) * sizeof(*rkp)); - memmove(rpp + 1, rpp, INT_GET(right->bb_numrecs, ARCH_CONVERT) * sizeof(*rpp)); + memmove(rkp + 1, rkp, be16_to_cpu(right->bb_numrecs) * sizeof(*rkp)); + memmove(rpp + 1, rpp, be16_to_cpu(right->bb_numrecs) * sizeof(*rpp)); #ifdef DEBUG if ((error = xfs_btree_check_lptr(cur, INT_GET(*lpp, ARCH_CONVERT), level))) { XFS_BMBT_TRACE_CURSOR(cur, ERROR); @@ -1479,21 +1463,21 @@ xfs_bmbt_rshift( #endif *rkp = *lkp; *rpp = *lpp; /* INT_: direct copy */ - xfs_bmbt_log_keys(cur, rbp, 1, INT_GET(right->bb_numrecs, ARCH_CONVERT) + 1); - xfs_bmbt_log_ptrs(cur, rbp, 1, INT_GET(right->bb_numrecs, ARCH_CONVERT) + 1); + xfs_bmbt_log_keys(cur, rbp, 1, be16_to_cpu(right->bb_numrecs) + 1); + xfs_bmbt_log_ptrs(cur, rbp, 1, be16_to_cpu(right->bb_numrecs) + 1); } else { - lrp = XFS_BMAP_REC_IADDR(left, INT_GET(left->bb_numrecs, ARCH_CONVERT), cur); + lrp = XFS_BMAP_REC_IADDR(left, be16_to_cpu(left->bb_numrecs), cur); rrp = XFS_BMAP_REC_IADDR(right, 1, cur); - memmove(rrp + 1, rrp, INT_GET(right->bb_numrecs, ARCH_CONVERT) * sizeof(*rrp)); + memmove(rrp + 1, rrp, be16_to_cpu(right->bb_numrecs) * sizeof(*rrp)); *rrp = *lrp; - xfs_bmbt_log_recs(cur, rbp, 1, INT_GET(right->bb_numrecs, ARCH_CONVERT) + 1); + xfs_bmbt_log_recs(cur, rbp, 1, be16_to_cpu(right->bb_numrecs) + 1); INT_SET(key.br_startoff, ARCH_CONVERT, xfs_bmbt_disk_get_startoff(rrp)); rkp = &key; } - INT_MOD(left->bb_numrecs, ARCH_CONVERT, -1); + be16_add(&left->bb_numrecs, -1); xfs_bmbt_log_block(cur, lbp, XFS_BB_NUMRECS); - INT_MOD(right->bb_numrecs, ARCH_CONVERT, +1); + be16_add(&right->bb_numrecs, 1); #ifdef DEBUG if (level > 0) xfs_btree_check_key(XFS_BTNUM_BMAP, rkp, rkp + 1); @@ -1624,47 +1608,47 @@ xfs_bmbt_split( return error; } #endif - INT_SET(right->bb_magic, ARCH_CONVERT, XFS_BMAP_MAGIC); - right->bb_level = left->bb_level; /* INT_: direct copy */ - INT_SET(right->bb_numrecs, ARCH_CONVERT, (__uint16_t)(INT_GET(left->bb_numrecs, ARCH_CONVERT) / 2)); - if ((INT_GET(left->bb_numrecs, ARCH_CONVERT) & 1) && - cur->bc_ptrs[level] <= INT_GET(right->bb_numrecs, ARCH_CONVERT) + 1) - INT_MOD(right->bb_numrecs, ARCH_CONVERT, +1); - i = INT_GET(left->bb_numrecs, ARCH_CONVERT) - INT_GET(right->bb_numrecs, ARCH_CONVERT) + 1; + right->bb_magic = cpu_to_be32(XFS_BMAP_MAGIC); + right->bb_level = left->bb_level; + right->bb_numrecs = cpu_to_be16(be16_to_cpu(left->bb_numrecs) / 2); + if ((be16_to_cpu(left->bb_numrecs) & 1) && + cur->bc_ptrs[level] <= be16_to_cpu(right->bb_numrecs) + 1) + be16_add(&right->bb_numrecs, 1); + i = be16_to_cpu(left->bb_numrecs) - be16_to_cpu(right->bb_numrecs) + 1; if (level > 0) { lkp = XFS_BMAP_KEY_IADDR(left, i, cur); lpp = XFS_BMAP_PTR_IADDR(left, i, cur); rkp = XFS_BMAP_KEY_IADDR(right, 1, cur); rpp = XFS_BMAP_PTR_IADDR(right, 1, cur); #ifdef DEBUG - for (i = 0; i < INT_GET(right->bb_numrecs, ARCH_CONVERT); i++) { + for (i = 0; i < be16_to_cpu(right->bb_numrecs); i++) { if ((error = xfs_btree_check_lptr(cur, INT_GET(lpp[i], ARCH_CONVERT), level))) { XFS_BMBT_TRACE_CURSOR(cur, ERROR); return error; } } #endif - memcpy(rkp, lkp, INT_GET(right->bb_numrecs, ARCH_CONVERT) * sizeof(*rkp)); - memcpy(rpp, lpp, INT_GET(right->bb_numrecs, ARCH_CONVERT) * sizeof(*rpp)); - xfs_bmbt_log_keys(cur, rbp, 1, INT_GET(right->bb_numrecs, ARCH_CONVERT)); - xfs_bmbt_log_ptrs(cur, rbp, 1, INT_GET(right->bb_numrecs, ARCH_CONVERT)); + memcpy(rkp, lkp, be16_to_cpu(right->bb_numrecs) * sizeof(*rkp)); + memcpy(rpp, lpp, be16_to_cpu(right->bb_numrecs) * sizeof(*rpp)); + xfs_bmbt_log_keys(cur, rbp, 1, be16_to_cpu(right->bb_numrecs)); + xfs_bmbt_log_ptrs(cur, rbp, 1, be16_to_cpu(right->bb_numrecs)); keyp->br_startoff = INT_GET(rkp->br_startoff, ARCH_CONVERT); } else { lrp = XFS_BMAP_REC_IADDR(left, i, cur); rrp = XFS_BMAP_REC_IADDR(right, 1, cur); - memcpy(rrp, lrp, INT_GET(right->bb_numrecs, ARCH_CONVERT) * sizeof(*rrp)); - xfs_bmbt_log_recs(cur, rbp, 1, INT_GET(right->bb_numrecs, ARCH_CONVERT)); + memcpy(rrp, lrp, be16_to_cpu(right->bb_numrecs) * sizeof(*rrp)); + xfs_bmbt_log_recs(cur, rbp, 1, be16_to_cpu(right->bb_numrecs)); keyp->br_startoff = xfs_bmbt_disk_get_startoff(rrp); } - INT_MOD(left->bb_numrecs, ARCH_CONVERT, -(INT_GET(right->bb_numrecs, ARCH_CONVERT))); - right->bb_rightsib = left->bb_rightsib; /* INT_: direct copy */ - INT_SET(left->bb_rightsib, ARCH_CONVERT, args.fsbno); - INT_SET(right->bb_leftsib, ARCH_CONVERT, lbno); + be16_add(&left->bb_numrecs, -(be16_to_cpu(right->bb_numrecs))); + right->bb_rightsib = left->bb_rightsib; + left->bb_rightsib = cpu_to_be64(args.fsbno); + right->bb_leftsib = cpu_to_be64(lbno); xfs_bmbt_log_block(cur, rbp, XFS_BB_ALL_BITS); xfs_bmbt_log_block(cur, lbp, XFS_BB_NUMRECS | XFS_BB_RIGHTSIB); - if (INT_GET(right->bb_rightsib, ARCH_CONVERT) != NULLDFSBNO) { + if (be64_to_cpu(right->bb_rightsib) != NULLDFSBNO) { if ((error = xfs_btree_read_bufl(args.mp, args.tp, - INT_GET(right->bb_rightsib, ARCH_CONVERT), 0, &rrbp, + be64_to_cpu(right->bb_rightsib), 0, &rrbp, XFS_BMAP_BTREE_REF))) { XFS_BMBT_TRACE_CURSOR(cur, ERROR); return error; @@ -1674,12 +1658,12 @@ xfs_bmbt_split( XFS_BMBT_TRACE_CURSOR(cur, ERROR); return error; } - INT_SET(rrblock->bb_leftsib, ARCH_CONVERT, args.fsbno); + rrblock->bb_leftsib = cpu_to_be64(args.fsbno); xfs_bmbt_log_block(cur, rrbp, XFS_BB_LEFTSIB); } - if (cur->bc_ptrs[level] > INT_GET(left->bb_numrecs, ARCH_CONVERT) + 1) { + if (cur->bc_ptrs[level] > be16_to_cpu(left->bb_numrecs) + 1) { xfs_btree_setbuf(cur, level, rbp); - cur->bc_ptrs[level] -= INT_GET(left->bb_numrecs, ARCH_CONVERT); + cur->bc_ptrs[level] -= be16_to_cpu(left->bb_numrecs); } if (level + 1 < cur->bc_nlevels) { if ((error = xfs_btree_dup_cursor(cur, curp))) { @@ -1751,18 +1735,18 @@ xfs_bmdr_to_bmbt( xfs_bmbt_key_t *tkp; xfs_bmbt_ptr_t *tpp; - INT_SET(rblock->bb_magic, ARCH_CONVERT, XFS_BMAP_MAGIC); - rblock->bb_level = dblock->bb_level; /* both in on-disk format */ - ASSERT(INT_GET(rblock->bb_level, ARCH_CONVERT) > 0); - rblock->bb_numrecs = dblock->bb_numrecs;/* both in on-disk format */ - INT_SET(rblock->bb_leftsib, ARCH_CONVERT, NULLDFSBNO); - INT_SET(rblock->bb_rightsib, ARCH_CONVERT, NULLDFSBNO); + rblock->bb_magic = cpu_to_be32(XFS_BMAP_MAGIC); + rblock->bb_level = dblock->bb_level; + ASSERT(be16_to_cpu(rblock->bb_level) > 0); + rblock->bb_numrecs = dblock->bb_numrecs; + rblock->bb_leftsib = cpu_to_be64(NULLDFSBNO); + rblock->bb_rightsib = cpu_to_be64(NULLDFSBNO); dmxr = (int)XFS_BTREE_BLOCK_MAXRECS(dblocklen, xfs_bmdr, 0); fkp = XFS_BTREE_KEY_ADDR(dblocklen, xfs_bmdr, dblock, 1, dmxr); tkp = XFS_BMAP_BROOT_KEY_ADDR(rblock, 1, rblocklen); fpp = XFS_BTREE_PTR_ADDR(dblocklen, xfs_bmdr, dblock, 1, dmxr); tpp = XFS_BMAP_BROOT_PTR_ADDR(rblock, 1, rblocklen); - dmxr = INT_GET(dblock->bb_numrecs, ARCH_CONVERT); + dmxr = be16_to_cpu(dblock->bb_numrecs); memcpy(tkp, fkp, sizeof(*fkp) * dmxr); memcpy(tpp, fpp, sizeof(*fpp) * dmxr); /* INT_: direct copy */ } @@ -1805,7 +1789,7 @@ xfs_bmbt_decrement( return error; } #endif - if (INT_GET(block->bb_leftsib, ARCH_CONVERT) == NULLDFSBNO) { + if (be64_to_cpu(block->bb_leftsib) == NULLDFSBNO) { XFS_BMBT_TRACE_CURSOR(cur, EXIT); *stat = 0; return 0; @@ -1837,7 +1821,7 @@ xfs_bmbt_decrement( XFS_BMBT_TRACE_CURSOR(cur, ERROR); return error; } - cur->bc_ptrs[lev] = INT_GET(block->bb_numrecs, ARCH_CONVERT); + cur->bc_ptrs[lev] = be16_to_cpu(block->bb_numrecs); } XFS_BMBT_TRACE_CURSOR(cur, EXIT); *stat = 1; @@ -2123,12 +2107,12 @@ xfs_bmbt_increment( return error; } #endif - if (++cur->bc_ptrs[level] <= INT_GET(block->bb_numrecs, ARCH_CONVERT)) { + if (++cur->bc_ptrs[level] <= be16_to_cpu(block->bb_numrecs)) { XFS_BMBT_TRACE_CURSOR(cur, EXIT); *stat = 1; return 0; } - if (INT_GET(block->bb_rightsib, ARCH_CONVERT) == NULLDFSBNO) { + if (be64_to_cpu(block->bb_rightsib) == NULLDFSBNO) { XFS_BMBT_TRACE_CURSOR(cur, EXIT); *stat = 0; return 0; @@ -2141,7 +2125,7 @@ xfs_bmbt_increment( return error; } #endif - if (++cur->bc_ptrs[lev] <= INT_GET(block->bb_numrecs, ARCH_CONVERT)) + if (++cur->bc_ptrs[lev] <= be16_to_cpu(block->bb_numrecs)) break; if (lev < cur->bc_nlevels - 1) xfs_btree_readahead(cur, lev, XFS_BTCUR_RIGHTRA); @@ -2403,23 +2387,23 @@ xfs_bmbt_newroot( bp = xfs_btree_get_bufl(args.mp, cur->bc_tp, args.fsbno, 0); cblock = XFS_BUF_TO_BMBT_BLOCK(bp); *cblock = *block; - INT_MOD(block->bb_level, ARCH_CONVERT, +1); - INT_SET(block->bb_numrecs, ARCH_CONVERT, 1); + be16_add(&block->bb_level, 1); + block->bb_numrecs = cpu_to_be16(1); cur->bc_nlevels++; cur->bc_ptrs[level + 1] = 1; kp = XFS_BMAP_KEY_IADDR(block, 1, cur); ckp = XFS_BMAP_KEY_IADDR(cblock, 1, cur); - memcpy(ckp, kp, INT_GET(cblock->bb_numrecs, ARCH_CONVERT) * sizeof(*kp)); + memcpy(ckp, kp, be16_to_cpu(cblock->bb_numrecs) * sizeof(*kp)); cpp = XFS_BMAP_PTR_IADDR(cblock, 1, cur); #ifdef DEBUG - for (i = 0; i < INT_GET(cblock->bb_numrecs, ARCH_CONVERT); i++) { + for (i = 0; i < be16_to_cpu(cblock->bb_numrecs); i++) { if ((error = xfs_btree_check_lptr(cur, INT_GET(pp[i], ARCH_CONVERT), level))) { XFS_BMBT_TRACE_CURSOR(cur, ERROR); return error; } } #endif - memcpy(cpp, pp, INT_GET(cblock->bb_numrecs, ARCH_CONVERT) * sizeof(*pp)); + memcpy(cpp, pp, be16_to_cpu(cblock->bb_numrecs) * sizeof(*pp)); #ifdef DEBUG if ((error = xfs_btree_check_lptr(cur, (xfs_bmbt_ptr_t)args.fsbno, level))) { @@ -2428,7 +2412,7 @@ xfs_bmbt_newroot( } #endif INT_SET(*pp, ARCH_CONVERT, args.fsbno); - xfs_iroot_realloc(cur->bc_private.b.ip, 1 - INT_GET(cblock->bb_numrecs, ARCH_CONVERT), + xfs_iroot_realloc(cur->bc_private.b.ip, 1 - be16_to_cpu(cblock->bb_numrecs), cur->bc_private.b.whichfork); xfs_btree_setbuf(cur, level, bp); /* @@ -2436,8 +2420,8 @@ xfs_bmbt_newroot( * the root is at the right level. */ xfs_bmbt_log_block(cur, bp, XFS_BB_ALL_BITS); - xfs_bmbt_log_keys(cur, bp, 1, INT_GET(cblock->bb_numrecs, ARCH_CONVERT)); - xfs_bmbt_log_ptrs(cur, bp, 1, INT_GET(cblock->bb_numrecs, ARCH_CONVERT)); + xfs_bmbt_log_keys(cur, bp, 1, be16_to_cpu(cblock->bb_numrecs)); + xfs_bmbt_log_ptrs(cur, bp, 1, be16_to_cpu(cblock->bb_numrecs)); XFS_BMBT_TRACE_CURSOR(cur, EXIT); *logflags |= XFS_ILOG_CORE | XFS_ILOG_FBROOT(cur->bc_private.b.whichfork); @@ -2705,18 +2689,18 @@ xfs_bmbt_to_bmdr( xfs_bmbt_key_t *tkp; xfs_bmbt_ptr_t *tpp; - ASSERT(INT_GET(rblock->bb_magic, ARCH_CONVERT) == XFS_BMAP_MAGIC); - ASSERT(INT_GET(rblock->bb_leftsib, ARCH_CONVERT) == NULLDFSBNO); - ASSERT(INT_GET(rblock->bb_rightsib, ARCH_CONVERT) == NULLDFSBNO); - ASSERT(INT_GET(rblock->bb_level, ARCH_CONVERT) > 0); - dblock->bb_level = rblock->bb_level; /* both in on-disk format */ - dblock->bb_numrecs = rblock->bb_numrecs;/* both in on-disk format */ + ASSERT(be32_to_cpu(rblock->bb_magic) == XFS_BMAP_MAGIC); + ASSERT(be64_to_cpu(rblock->bb_leftsib) == NULLDFSBNO); + ASSERT(be64_to_cpu(rblock->bb_rightsib) == NULLDFSBNO); + ASSERT(be16_to_cpu(rblock->bb_level) > 0); + dblock->bb_level = rblock->bb_level; + dblock->bb_numrecs = rblock->bb_numrecs; dmxr = (int)XFS_BTREE_BLOCK_MAXRECS(dblocklen, xfs_bmdr, 0); fkp = XFS_BMAP_BROOT_KEY_ADDR(rblock, 1, rblocklen); tkp = XFS_BTREE_KEY_ADDR(dblocklen, xfs_bmdr, dblock, 1, dmxr); fpp = XFS_BMAP_BROOT_PTR_ADDR(rblock, 1, rblocklen); tpp = XFS_BTREE_PTR_ADDR(dblocklen, xfs_bmdr, dblock, 1, dmxr); - dmxr = INT_GET(dblock->bb_numrecs, ARCH_CONVERT); + dmxr = be16_to_cpu(dblock->bb_numrecs); memcpy(tkp, fkp, sizeof(*fkp) * dmxr); memcpy(tpp, fpp, sizeof(*fpp) * dmxr); /* INT_: direct copy */ } diff --git a/fs/xfs/xfs_bmap_btree.h b/fs/xfs/xfs_bmap_btree.h index 2cf4fe45cbcb..e095a2d344ae 100644 --- a/fs/xfs/xfs_bmap_btree.h +++ b/fs/xfs/xfs_bmap_btree.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000,2002-2004 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000,2002-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_BMAP_BTREE_H__ #define __XFS_BMAP_BTREE_H__ @@ -42,10 +28,9 @@ struct xfs_inode; /* * Bmap root header, on-disk form only. */ -typedef struct xfs_bmdr_block -{ - __uint16_t bb_level; /* 0 is a leaf */ - __uint16_t bb_numrecs; /* current # of data records */ +typedef struct xfs_bmdr_block { + __be16 bb_level; /* 0 is a leaf */ + __be16 bb_numrecs; /* current # of data records */ } xfs_bmdr_block_t; /* @@ -114,31 +99,31 @@ typedef xfs_bmbt_rec_64_t xfs_bmbt_rec_t, xfs_bmdr_rec_t; (((((xfs_fsblock_t)1) << STARTBLOCKMASKBITS) - 1) << STARTBLOCKVALBITS) #define DSTARTBLOCKMASK \ (((((xfs_dfsbno_t)1) << DSTARTBLOCKMASKBITS) - 1) << STARTBLOCKVALBITS) -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_ISNULLSTARTBLOCK) -int isnullstartblock(xfs_fsblock_t x); + #define ISNULLSTARTBLOCK(x) isnullstartblock(x) -#else -#define ISNULLSTARTBLOCK(x) (((x) & STARTBLOCKMASK) == STARTBLOCKMASK) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_ISNULLDSTARTBLOCK) -int isnulldstartblock(xfs_dfsbno_t x); +static inline int isnullstartblock(xfs_fsblock_t x) +{ + return ((x) & STARTBLOCKMASK) == STARTBLOCKMASK; +} + #define ISNULLDSTARTBLOCK(x) isnulldstartblock(x) -#else -#define ISNULLDSTARTBLOCK(x) (((x) & DSTARTBLOCKMASK) == DSTARTBLOCKMASK) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_NULLSTARTBLOCK) -xfs_fsblock_t nullstartblock(int k); +static inline int isnulldstartblock(xfs_dfsbno_t x) +{ + return ((x) & DSTARTBLOCKMASK) == DSTARTBLOCKMASK; +} + #define NULLSTARTBLOCK(k) nullstartblock(k) -#else -#define NULLSTARTBLOCK(k) \ - ((ASSERT(k < (1 << STARTBLOCKVALBITS))), (STARTBLOCKMASK | (k))) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_STARTBLOCKVAL) -xfs_filblks_t startblockval(xfs_fsblock_t x); +static inline xfs_fsblock_t nullstartblock(int k) +{ + ASSERT(k < (1 << STARTBLOCKVALBITS)); + return STARTBLOCKMASK | (k); +} + #define STARTBLOCKVAL(x) startblockval(x) -#else -#define STARTBLOCKVAL(x) ((xfs_filblks_t)((x) & ~STARTBLOCKMASK)) -#endif +static inline xfs_filblks_t startblockval(xfs_fsblock_t x) +{ + return (xfs_filblks_t)((x) & ~STARTBLOCKMASK); +} /* * Possible extent formats. @@ -159,14 +144,9 @@ typedef enum { /* * Extent state and extent format macros. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_EXTFMT_INODE ) -xfs_exntfmt_t xfs_extfmt_inode(struct xfs_inode *ip); -#define XFS_EXTFMT_INODE(x) xfs_extfmt_inode(x) -#else -#define XFS_EXTFMT_INODE(x) \ - (XFS_SB_VERSION_HASEXTFLGBIT(&((x)->i_mount->m_sb)) ? \ - XFS_EXTFMT_HASSTATE : XFS_EXTFMT_NOSTATE) -#endif +#define XFS_EXTFMT_INODE(x) \ + (XFS_SB_VERSION_HASEXTFLGBIT(&((x)->i_mount->m_sb)) ? \ + XFS_EXTFMT_HASSTATE : XFS_EXTFMT_NOSTATE) #define ISUNWRITTEN(x) ((x)->br_state == XFS_EXT_UNWRITTEN) /* @@ -192,248 +172,110 @@ typedef xfs_dfsbno_t xfs_bmbt_ptr_t, xfs_bmdr_ptr_t; /* btree pointer type */ /* btree block header type */ typedef struct xfs_btree_lblock xfs_bmbt_block_t; -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BUF_TO_BMBT_BLOCK) -xfs_bmbt_block_t *xfs_buf_to_bmbt_block(struct xfs_buf *bp); -#define XFS_BUF_TO_BMBT_BLOCK(bp) xfs_buf_to_bmbt_block(bp) -#else -#define XFS_BUF_TO_BMBT_BLOCK(bp) ((xfs_bmbt_block_t *)(XFS_BUF_PTR(bp))) -#endif +#define XFS_BUF_TO_BMBT_BLOCK(bp) ((xfs_bmbt_block_t *)XFS_BUF_PTR(bp)) -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BMAP_RBLOCK_DSIZE) -int xfs_bmap_rblock_dsize(int lev, struct xfs_btree_cur *cur); -#define XFS_BMAP_RBLOCK_DSIZE(lev,cur) xfs_bmap_rblock_dsize(lev,cur) -#else -#define XFS_BMAP_RBLOCK_DSIZE(lev,cur) ((cur)->bc_private.b.forksize) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BMAP_RBLOCK_ISIZE) -int xfs_bmap_rblock_isize(int lev, struct xfs_btree_cur *cur); -#define XFS_BMAP_RBLOCK_ISIZE(lev,cur) xfs_bmap_rblock_isize(lev,cur) -#else -#define XFS_BMAP_RBLOCK_ISIZE(lev,cur) \ +#define XFS_BMAP_IBLOCK_SIZE(lev,cur) (1 << (cur)->bc_blocklog) +#define XFS_BMAP_RBLOCK_DSIZE(lev,cur) ((cur)->bc_private.b.forksize) +#define XFS_BMAP_RBLOCK_ISIZE(lev,cur) \ ((int)XFS_IFORK_PTR((cur)->bc_private.b.ip, \ - (cur)->bc_private.b.whichfork)->if_broot_bytes) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BMAP_IBLOCK_SIZE) -int xfs_bmap_iblock_size(int lev, struct xfs_btree_cur *cur); -#define XFS_BMAP_IBLOCK_SIZE(lev,cur) xfs_bmap_iblock_size(lev,cur) -#else -#define XFS_BMAP_IBLOCK_SIZE(lev,cur) (1 << (cur)->bc_blocklog) -#endif + (cur)->bc_private.b.whichfork)->if_broot_bytes) -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BMAP_BLOCK_DSIZE) -int xfs_bmap_block_dsize(int lev, struct xfs_btree_cur *cur); -#define XFS_BMAP_BLOCK_DSIZE(lev,cur) xfs_bmap_block_dsize(lev,cur) -#else -#define XFS_BMAP_BLOCK_DSIZE(lev,cur) \ - ((lev) == (cur)->bc_nlevels - 1 ? \ - XFS_BMAP_RBLOCK_DSIZE(lev,cur) : \ - XFS_BMAP_IBLOCK_SIZE(lev,cur)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BMAP_BLOCK_ISIZE) -int xfs_bmap_block_isize(int lev, struct xfs_btree_cur *cur); -#define XFS_BMAP_BLOCK_ISIZE(lev,cur) xfs_bmap_block_isize(lev,cur) -#else -#define XFS_BMAP_BLOCK_ISIZE(lev,cur) \ - ((lev) == (cur)->bc_nlevels - 1 ? \ - XFS_BMAP_RBLOCK_ISIZE(lev,cur) : \ - XFS_BMAP_IBLOCK_SIZE(lev,cur)) -#endif +#define XFS_BMAP_BLOCK_DSIZE(lev,cur) \ + (((lev) == (cur)->bc_nlevels - 1 ? \ + XFS_BMAP_RBLOCK_DSIZE(lev,cur) : XFS_BMAP_IBLOCK_SIZE(lev,cur))) +#define XFS_BMAP_BLOCK_ISIZE(lev,cur) \ + (((lev) == (cur)->bc_nlevels - 1 ? \ + XFS_BMAP_RBLOCK_ISIZE(lev,cur) : XFS_BMAP_IBLOCK_SIZE(lev,cur))) -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BMAP_BLOCK_DMAXRECS) -int xfs_bmap_block_dmaxrecs(int lev, struct xfs_btree_cur *cur); -#define XFS_BMAP_BLOCK_DMAXRECS(lev,cur) xfs_bmap_block_dmaxrecs(lev,cur) -#else #define XFS_BMAP_BLOCK_DMAXRECS(lev,cur) \ - ((lev) == (cur)->bc_nlevels - 1 ? \ + (((lev) == (cur)->bc_nlevels - 1 ? \ XFS_BTREE_BLOCK_MAXRECS(XFS_BMAP_RBLOCK_DSIZE(lev,cur), \ xfs_bmdr, (lev) == 0) : \ - ((cur)->bc_mp->m_bmap_dmxr[(lev) != 0])) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BMAP_BLOCK_IMAXRECS) -int xfs_bmap_block_imaxrecs(int lev, struct xfs_btree_cur *cur); -#define XFS_BMAP_BLOCK_IMAXRECS(lev,cur) xfs_bmap_block_imaxrecs(lev,cur) -#else + ((cur)->bc_mp->m_bmap_dmxr[(lev) != 0]))) #define XFS_BMAP_BLOCK_IMAXRECS(lev,cur) \ - ((lev) == (cur)->bc_nlevels - 1 ? \ - XFS_BTREE_BLOCK_MAXRECS(XFS_BMAP_RBLOCK_ISIZE(lev,cur), \ - xfs_bmbt, (lev) == 0) : \ - ((cur)->bc_mp->m_bmap_dmxr[(lev) != 0])) -#endif + (((lev) == (cur)->bc_nlevels - 1 ? \ + XFS_BTREE_BLOCK_MAXRECS(XFS_BMAP_RBLOCK_ISIZE(lev,cur),\ + xfs_bmbt, (lev) == 0) : \ + ((cur)->bc_mp->m_bmap_dmxr[(lev) != 0]))) -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BMAP_BLOCK_DMINRECS) -int xfs_bmap_block_dminrecs(int lev, struct xfs_btree_cur *cur); -#define XFS_BMAP_BLOCK_DMINRECS(lev,cur) xfs_bmap_block_dminrecs(lev,cur) -#else #define XFS_BMAP_BLOCK_DMINRECS(lev,cur) \ - ((lev) == (cur)->bc_nlevels - 1 ? \ - XFS_BTREE_BLOCK_MINRECS(XFS_BMAP_RBLOCK_DSIZE(lev,cur), \ - xfs_bmdr, (lev) == 0) : \ - ((cur)->bc_mp->m_bmap_dmnr[(lev) != 0])) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BMAP_BLOCK_IMINRECS) -int xfs_bmap_block_iminrecs(int lev, struct xfs_btree_cur *cur); -#define XFS_BMAP_BLOCK_IMINRECS(lev,cur) xfs_bmap_block_iminrecs(lev,cur) -#else + (((lev) == (cur)->bc_nlevels - 1 ? \ + XFS_BTREE_BLOCK_MINRECS(XFS_BMAP_RBLOCK_DSIZE(lev,cur),\ + xfs_bmdr, (lev) == 0) : \ + ((cur)->bc_mp->m_bmap_dmnr[(lev) != 0]))) #define XFS_BMAP_BLOCK_IMINRECS(lev,cur) \ - ((lev) == (cur)->bc_nlevels - 1 ? \ - XFS_BTREE_BLOCK_MINRECS(XFS_BMAP_RBLOCK_ISIZE(lev,cur), \ - xfs_bmbt, (lev) == 0) : \ - ((cur)->bc_mp->m_bmap_dmnr[(lev) != 0])) -#endif - -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BMAP_REC_DADDR) -xfs_bmbt_rec_t * -xfs_bmap_rec_daddr(xfs_bmbt_block_t *bb, int i, struct xfs_btree_cur *cur); -#define XFS_BMAP_REC_DADDR(bb,i,cur) xfs_bmap_rec_daddr(bb,i,cur) -#else -#define XFS_BMAP_REC_DADDR(bb,i,cur) \ - XFS_BTREE_REC_ADDR(XFS_BMAP_BLOCK_DSIZE( \ - INT_GET((bb)->bb_level, ARCH_CONVERT), cur), \ - xfs_bmbt, bb, i, XFS_BMAP_BLOCK_DMAXRECS( \ - INT_GET((bb)->bb_level, ARCH_CONVERT), cur)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BMAP_REC_IADDR) -xfs_bmbt_rec_t * -xfs_bmap_rec_iaddr(xfs_bmbt_block_t *bb, int i, struct xfs_btree_cur *cur); -#define XFS_BMAP_REC_IADDR(bb,i,cur) xfs_bmap_rec_iaddr(bb,i,cur) -#else -#define XFS_BMAP_REC_IADDR(bb,i,cur) \ - XFS_BTREE_REC_ADDR(XFS_BMAP_BLOCK_ISIZE( \ - INT_GET((bb)->bb_level, ARCH_CONVERT), cur), \ - xfs_bmbt, bb, i, XFS_BMAP_BLOCK_IMAXRECS( \ - INT_GET((bb)->bb_level, ARCH_CONVERT), cur)) -#endif - -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BMAP_KEY_DADDR) -xfs_bmbt_key_t * -xfs_bmap_key_daddr(xfs_bmbt_block_t *bb, int i, struct xfs_btree_cur *cur); -#define XFS_BMAP_KEY_DADDR(bb,i,cur) xfs_bmap_key_daddr(bb,i,cur) -#else -#define XFS_BMAP_KEY_DADDR(bb,i,cur) \ - XFS_BTREE_KEY_ADDR(XFS_BMAP_BLOCK_DSIZE( \ - INT_GET((bb)->bb_level, ARCH_CONVERT), cur), \ - xfs_bmbt, bb, i, XFS_BMAP_BLOCK_DMAXRECS( \ - INT_GET((bb)->bb_level, ARCH_CONVERT), cur)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BMAP_KEY_IADDR) -xfs_bmbt_key_t * -xfs_bmap_key_iaddr(xfs_bmbt_block_t *bb, int i, struct xfs_btree_cur *cur); -#define XFS_BMAP_KEY_IADDR(bb,i,cur) xfs_bmap_key_iaddr(bb,i,cur) -#else -#define XFS_BMAP_KEY_IADDR(bb,i,cur) \ - XFS_BTREE_KEY_ADDR(XFS_BMAP_BLOCK_ISIZE( \ - INT_GET((bb)->bb_level, ARCH_CONVERT), cur), \ - xfs_bmbt, bb, i, XFS_BMAP_BLOCK_IMAXRECS( \ - INT_GET((bb)->bb_level, ARCH_CONVERT), cur)) -#endif - -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BMAP_PTR_DADDR) -xfs_bmbt_ptr_t * -xfs_bmap_ptr_daddr(xfs_bmbt_block_t *bb, int i, struct xfs_btree_cur *cur); -#define XFS_BMAP_PTR_DADDR(bb,i,cur) xfs_bmap_ptr_daddr(bb,i,cur) -#else -#define XFS_BMAP_PTR_DADDR(bb,i,cur) \ - XFS_BTREE_PTR_ADDR(XFS_BMAP_BLOCK_DSIZE( \ - INT_GET((bb)->bb_level, ARCH_CONVERT), cur), \ - xfs_bmbt, bb, i, XFS_BMAP_BLOCK_DMAXRECS( \ - INT_GET((bb)->bb_level, ARCH_CONVERT), cur)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BMAP_PTR_IADDR) -xfs_bmbt_ptr_t * -xfs_bmap_ptr_iaddr(xfs_bmbt_block_t *bb, int i, struct xfs_btree_cur *cur); -#define XFS_BMAP_PTR_IADDR(bb,i,cur) xfs_bmap_ptr_iaddr(bb,i,cur) -#else -#define XFS_BMAP_PTR_IADDR(bb,i,cur) \ - XFS_BTREE_PTR_ADDR(XFS_BMAP_BLOCK_ISIZE( \ - INT_GET((bb)->bb_level, ARCH_CONVERT), cur), \ - xfs_bmbt, bb, i, XFS_BMAP_BLOCK_IMAXRECS( \ - INT_GET((bb)->bb_level, ARCH_CONVERT), cur)) -#endif + (((lev) == (cur)->bc_nlevels - 1 ? \ + XFS_BTREE_BLOCK_MINRECS(XFS_BMAP_RBLOCK_ISIZE(lev,cur),\ + xfs_bmbt, (lev) == 0) : \ + ((cur)->bc_mp->m_bmap_dmnr[(lev) != 0]))) + +#define XFS_BMAP_REC_DADDR(bb,i,cur) \ + (XFS_BTREE_REC_ADDR(XFS_BMAP_BLOCK_DSIZE( \ + be16_to_cpu((bb)->bb_level), cur), \ + xfs_bmbt, bb, i, XFS_BMAP_BLOCK_DMAXRECS( \ + be16_to_cpu((bb)->bb_level), cur))) +#define XFS_BMAP_REC_IADDR(bb,i,cur) \ + (XFS_BTREE_REC_ADDR(XFS_BMAP_BLOCK_ISIZE( \ + be16_to_cpu((bb)->bb_level), cur), \ + xfs_bmbt, bb, i, XFS_BMAP_BLOCK_IMAXRECS( \ + be16_to_cpu((bb)->bb_level), cur))) + +#define XFS_BMAP_KEY_DADDR(bb,i,cur) \ + (XFS_BTREE_KEY_ADDR(XFS_BMAP_BLOCK_DSIZE( \ + be16_to_cpu((bb)->bb_level), cur), \ + xfs_bmbt, bb, i, XFS_BMAP_BLOCK_DMAXRECS( \ + be16_to_cpu((bb)->bb_level), cur))) +#define XFS_BMAP_KEY_IADDR(bb,i,cur) \ + (XFS_BTREE_KEY_ADDR(XFS_BMAP_BLOCK_ISIZE( \ + be16_to_cpu((bb)->bb_level), cur), \ + xfs_bmbt, bb, i, XFS_BMAP_BLOCK_IMAXRECS( \ + be16_to_cpu((bb)->bb_level), cur))) + +#define XFS_BMAP_PTR_DADDR(bb,i,cur) \ + (XFS_BTREE_PTR_ADDR(XFS_BMAP_BLOCK_DSIZE( \ + be16_to_cpu((bb)->bb_level), cur), \ + xfs_bmbt, bb, i, XFS_BMAP_BLOCK_DMAXRECS( \ + be16_to_cpu((bb)->bb_level), cur))) +#define XFS_BMAP_PTR_IADDR(bb,i,cur) \ + (XFS_BTREE_PTR_ADDR(XFS_BMAP_BLOCK_ISIZE( \ + be16_to_cpu((bb)->bb_level), cur), \ + xfs_bmbt, bb, i, XFS_BMAP_BLOCK_IMAXRECS( \ + be16_to_cpu((bb)->bb_level), cur))) /* * These are to be used when we know the size of the block and * we don't have a cursor. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BMAP_BROOT_REC_ADDR) -xfs_bmbt_rec_t *xfs_bmap_broot_rec_addr(xfs_bmbt_block_t *bb, int i, int sz); -#define XFS_BMAP_BROOT_REC_ADDR(bb,i,sz) xfs_bmap_broot_rec_addr(bb,i,sz) -#else #define XFS_BMAP_BROOT_REC_ADDR(bb,i,sz) \ - XFS_BTREE_REC_ADDR(sz,xfs_bmbt,bb,i,XFS_BMAP_BROOT_MAXRECS(sz)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BMAP_BROOT_KEY_ADDR) -xfs_bmbt_key_t *xfs_bmap_broot_key_addr(xfs_bmbt_block_t *bb, int i, int sz); -#define XFS_BMAP_BROOT_KEY_ADDR(bb,i,sz) xfs_bmap_broot_key_addr(bb,i,sz) -#else + (XFS_BTREE_REC_ADDR(sz,xfs_bmbt,bb,i,XFS_BMAP_BROOT_MAXRECS(sz))) #define XFS_BMAP_BROOT_KEY_ADDR(bb,i,sz) \ - XFS_BTREE_KEY_ADDR(sz,xfs_bmbt,bb,i,XFS_BMAP_BROOT_MAXRECS(sz)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BMAP_BROOT_PTR_ADDR) -xfs_bmbt_ptr_t *xfs_bmap_broot_ptr_addr(xfs_bmbt_block_t *bb, int i, int sz); -#define XFS_BMAP_BROOT_PTR_ADDR(bb,i,sz) xfs_bmap_broot_ptr_addr(bb,i,sz) -#else + (XFS_BTREE_KEY_ADDR(sz,xfs_bmbt,bb,i,XFS_BMAP_BROOT_MAXRECS(sz))) #define XFS_BMAP_BROOT_PTR_ADDR(bb,i,sz) \ - XFS_BTREE_PTR_ADDR(sz,xfs_bmbt,bb,i,XFS_BMAP_BROOT_MAXRECS(sz)) -#endif + (XFS_BTREE_PTR_ADDR(sz,xfs_bmbt,bb,i,XFS_BMAP_BROOT_MAXRECS(sz))) + +#define XFS_BMAP_BROOT_NUMRECS(bb) be16_to_cpu((bb)->bb_numrecs) +#define XFS_BMAP_BROOT_MAXRECS(sz) XFS_BTREE_BLOCK_MAXRECS(sz,xfs_bmbt,0) -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BMAP_BROOT_NUMRECS) -int xfs_bmap_broot_numrecs(xfs_bmdr_block_t *bb); -#define XFS_BMAP_BROOT_NUMRECS(bb) xfs_bmap_broot_numrecs(bb) -#else -#define XFS_BMAP_BROOT_NUMRECS(bb) (INT_GET((bb)->bb_numrecs, ARCH_CONVERT)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BMAP_BROOT_MAXRECS) -int xfs_bmap_broot_maxrecs(int sz); -#define XFS_BMAP_BROOT_MAXRECS(sz) xfs_bmap_broot_maxrecs(sz) -#else -#define XFS_BMAP_BROOT_MAXRECS(sz) XFS_BTREE_BLOCK_MAXRECS(sz,xfs_bmbt,0) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BMAP_BROOT_SPACE_CALC) -int xfs_bmap_broot_space_calc(int nrecs); -#define XFS_BMAP_BROOT_SPACE_CALC(nrecs) xfs_bmap_broot_space_calc(nrecs) -#else #define XFS_BMAP_BROOT_SPACE_CALC(nrecs) \ - ((int)(sizeof(xfs_bmbt_block_t) + \ - ((nrecs) * (sizeof(xfs_bmbt_key_t) + sizeof(xfs_bmbt_ptr_t))))) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BMAP_BROOT_SPACE) -int xfs_bmap_broot_space(xfs_bmdr_block_t *bb); -#define XFS_BMAP_BROOT_SPACE(bb) xfs_bmap_broot_space(bb) -#else + (int)(sizeof(xfs_bmbt_block_t) + \ + ((nrecs) * (sizeof(xfs_bmbt_key_t) + sizeof(xfs_bmbt_ptr_t)))) + #define XFS_BMAP_BROOT_SPACE(bb) \ - XFS_BMAP_BROOT_SPACE_CALC(INT_GET((bb)->bb_numrecs, ARCH_CONVERT)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BMDR_SPACE_CALC) -int xfs_bmdr_space_calc(int nrecs); -#define XFS_BMDR_SPACE_CALC(nrecs) xfs_bmdr_space_calc(nrecs) -#else -#define XFS_BMDR_SPACE_CALC(nrecs) \ - ((int)(sizeof(xfs_bmdr_block_t) + \ - ((nrecs) * (sizeof(xfs_bmbt_key_t) + sizeof(xfs_bmbt_ptr_t))))) -#endif + (XFS_BMAP_BROOT_SPACE_CALC(be16_to_cpu((bb)->bb_numrecs))) +#define XFS_BMDR_SPACE_CALC(nrecs) \ + (int)(sizeof(xfs_bmdr_block_t) + \ + ((nrecs) * (sizeof(xfs_bmbt_key_t) + sizeof(xfs_bmbt_ptr_t)))) /* * Maximum number of bmap btree levels. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BM_MAXLEVELS) -int xfs_bm_maxlevels(struct xfs_mount *mp, int w); -#define XFS_BM_MAXLEVELS(mp,w) xfs_bm_maxlevels(mp,w) -#else -#define XFS_BM_MAXLEVELS(mp,w) ((mp)->m_bm_maxlevels[w]) -#endif +#define XFS_BM_MAXLEVELS(mp,w) ((mp)->m_bm_maxlevels[(w)]) -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BMAP_SANITY_CHECK) -int xfs_bmap_sanity_check(struct xfs_mount *mp, xfs_bmbt_block_t *bb, - int level); -#define XFS_BMAP_SANITY_CHECK(mp,bb,level) \ - xfs_bmap_sanity_check(mp,bb,level) -#else -#define XFS_BMAP_SANITY_CHECK(mp,bb,level) \ - (INT_GET((bb)->bb_magic, ARCH_CONVERT) == XFS_BMAP_MAGIC && \ - INT_GET((bb)->bb_level, ARCH_CONVERT) == level && \ - INT_GET((bb)->bb_numrecs, ARCH_CONVERT) > 0 && \ - INT_GET((bb)->bb_numrecs, ARCH_CONVERT) <= (mp)->m_bmap_dmxr[(level) != 0]) -#endif +#define XFS_BMAP_SANITY_CHECK(mp,bb,level) \ + (be32_to_cpu((bb)->bb_magic) == XFS_BMAP_MAGIC && \ + be16_to_cpu((bb)->bb_level) == level && \ + be16_to_cpu((bb)->bb_numrecs) > 0 && \ + be16_to_cpu((bb)->bb_numrecs) <= (mp)->m_bmap_dmxr[(level) != 0]) #ifdef __KERNEL__ @@ -459,234 +301,84 @@ extern ktrace_t *xfs_bmbt_trace_buf; /* * Prototypes for xfs_bmap.c to call. */ - -void -xfs_bmdr_to_bmbt( - xfs_bmdr_block_t *, - int, - xfs_bmbt_block_t *, - int); - -int -xfs_bmbt_decrement( - struct xfs_btree_cur *, - int, - int *); - -int -xfs_bmbt_delete( - struct xfs_btree_cur *, - int *); - -void -xfs_bmbt_get_all( - xfs_bmbt_rec_t *r, - xfs_bmbt_irec_t *s); - -xfs_bmbt_block_t * -xfs_bmbt_get_block( - struct xfs_btree_cur *cur, - int level, - struct xfs_buf **bpp); - -xfs_filblks_t -xfs_bmbt_get_blockcount( - xfs_bmbt_rec_t *r); - -xfs_fsblock_t -xfs_bmbt_get_startblock( - xfs_bmbt_rec_t *r); - -xfs_fileoff_t -xfs_bmbt_get_startoff( - xfs_bmbt_rec_t *r); - -xfs_exntst_t -xfs_bmbt_get_state( - xfs_bmbt_rec_t *r); +extern void xfs_bmdr_to_bmbt(xfs_bmdr_block_t *, int, xfs_bmbt_block_t *, int); +extern int xfs_bmbt_decrement(struct xfs_btree_cur *, int, int *); +extern int xfs_bmbt_delete(struct xfs_btree_cur *, int *); +extern void xfs_bmbt_get_all(xfs_bmbt_rec_t *r, xfs_bmbt_irec_t *s); +extern xfs_bmbt_block_t *xfs_bmbt_get_block(struct xfs_btree_cur *cur, + int, struct xfs_buf **bpp); +extern xfs_filblks_t xfs_bmbt_get_blockcount(xfs_bmbt_rec_t *r); +extern xfs_fsblock_t xfs_bmbt_get_startblock(xfs_bmbt_rec_t *r); +extern xfs_fileoff_t xfs_bmbt_get_startoff(xfs_bmbt_rec_t *r); +extern xfs_exntst_t xfs_bmbt_get_state(xfs_bmbt_rec_t *r); #ifndef XFS_NATIVE_HOST -void -xfs_bmbt_disk_get_all( - xfs_bmbt_rec_t *r, - xfs_bmbt_irec_t *s); - -xfs_exntst_t -xfs_bmbt_disk_get_state( - xfs_bmbt_rec_t *r); - -xfs_filblks_t -xfs_bmbt_disk_get_blockcount( - xfs_bmbt_rec_t *r); - -xfs_fsblock_t -xfs_bmbt_disk_get_startblock( - xfs_bmbt_rec_t *r); - -xfs_fileoff_t -xfs_bmbt_disk_get_startoff( - xfs_bmbt_rec_t *r); - -#else -#define xfs_bmbt_disk_get_all(r, s) \ - xfs_bmbt_get_all(r, s) -#define xfs_bmbt_disk_get_state(r) \ - xfs_bmbt_get_state(r) -#define xfs_bmbt_disk_get_blockcount(r) \ - xfs_bmbt_get_blockcount(r) -#define xfs_bmbt_disk_get_startblock(r) \ - xfs_bmbt_get_blockcount(r) -#define xfs_bmbt_disk_get_startoff(r) \ - xfs_bmbt_get_startoff(r) +extern void xfs_bmbt_disk_get_all(xfs_bmbt_rec_t *r, xfs_bmbt_irec_t *s); +extern xfs_exntst_t xfs_bmbt_disk_get_state(xfs_bmbt_rec_t *r); +extern xfs_filblks_t xfs_bmbt_disk_get_blockcount(xfs_bmbt_rec_t *r); +extern xfs_fsblock_t xfs_bmbt_disk_get_startblock(xfs_bmbt_rec_t *r); +extern xfs_fileoff_t xfs_bmbt_disk_get_startoff(xfs_bmbt_rec_t *r); +#else +#define xfs_bmbt_disk_get_all(r, s) xfs_bmbt_get_all(r, s) +#define xfs_bmbt_disk_get_state(r) xfs_bmbt_get_state(r) +#define xfs_bmbt_disk_get_blockcount(r) xfs_bmbt_get_blockcount(r) +#define xfs_bmbt_disk_get_startblock(r) xfs_bmbt_get_blockcount(r) +#define xfs_bmbt_disk_get_startoff(r) xfs_bmbt_get_startoff(r) #endif /* XFS_NATIVE_HOST */ -int -xfs_bmbt_increment( - struct xfs_btree_cur *, - int, - int *); - -int -xfs_bmbt_insert( - struct xfs_btree_cur *, - int *); - -void -xfs_bmbt_log_block( - struct xfs_btree_cur *, - struct xfs_buf *, - int); - -void -xfs_bmbt_log_recs( - struct xfs_btree_cur *, - struct xfs_buf *, - int, - int); - -int -xfs_bmbt_lookup_eq( - struct xfs_btree_cur *, - xfs_fileoff_t, - xfs_fsblock_t, - xfs_filblks_t, - int *); - -int -xfs_bmbt_lookup_ge( - struct xfs_btree_cur *, - xfs_fileoff_t, - xfs_fsblock_t, - xfs_filblks_t, - int *); +extern int xfs_bmbt_increment(struct xfs_btree_cur *, int, int *); +extern int xfs_bmbt_insert(struct xfs_btree_cur *, int *); +extern void xfs_bmbt_log_block(struct xfs_btree_cur *, struct xfs_buf *, int); +extern void xfs_bmbt_log_recs(struct xfs_btree_cur *, struct xfs_buf *, int, + int); +extern int xfs_bmbt_lookup_eq(struct xfs_btree_cur *, xfs_fileoff_t, + xfs_fsblock_t, xfs_filblks_t, int *); +extern int xfs_bmbt_lookup_ge(struct xfs_btree_cur *, xfs_fileoff_t, + xfs_fsblock_t, xfs_filblks_t, int *); /* * Give the bmap btree a new root block. Copy the old broot contents * down into a real block and make the broot point to it. */ -int /* error */ -xfs_bmbt_newroot( - struct xfs_btree_cur *cur, /* btree cursor */ - int *logflags, /* logging flags for inode */ - int *stat); /* return status - 0 fail */ - -void -xfs_bmbt_set_all( - xfs_bmbt_rec_t *r, - xfs_bmbt_irec_t *s); - -void -xfs_bmbt_set_allf( - xfs_bmbt_rec_t *r, - xfs_fileoff_t o, - xfs_fsblock_t b, - xfs_filblks_t c, - xfs_exntst_t v); - -void -xfs_bmbt_set_blockcount( - xfs_bmbt_rec_t *r, - xfs_filblks_t v); - -void -xfs_bmbt_set_startblock( - xfs_bmbt_rec_t *r, - xfs_fsblock_t v); - -void -xfs_bmbt_set_startoff( - xfs_bmbt_rec_t *r, - xfs_fileoff_t v); - -void -xfs_bmbt_set_state( - xfs_bmbt_rec_t *r, - xfs_exntst_t v); +extern int xfs_bmbt_newroot(struct xfs_btree_cur *cur, int *lflags, int *stat); + +extern void xfs_bmbt_set_all(xfs_bmbt_rec_t *r, xfs_bmbt_irec_t *s); +extern void xfs_bmbt_set_allf(xfs_bmbt_rec_t *r, xfs_fileoff_t o, + xfs_fsblock_t b, xfs_filblks_t c, xfs_exntst_t v); +extern void xfs_bmbt_set_blockcount(xfs_bmbt_rec_t *r, xfs_filblks_t v); +extern void xfs_bmbt_set_startblock(xfs_bmbt_rec_t *r, xfs_fsblock_t v); +extern void xfs_bmbt_set_startoff(xfs_bmbt_rec_t *r, xfs_fileoff_t v); +extern void xfs_bmbt_set_state(xfs_bmbt_rec_t *r, xfs_exntst_t v); #ifndef XFS_NATIVE_HOST -void -xfs_bmbt_disk_set_all( - xfs_bmbt_rec_t *r, - xfs_bmbt_irec_t *s); - -void -xfs_bmbt_disk_set_allf( - xfs_bmbt_rec_t *r, - xfs_fileoff_t o, - xfs_fsblock_t b, - xfs_filblks_t c, - xfs_exntst_t v); +extern void xfs_bmbt_disk_set_all(xfs_bmbt_rec_t *r, xfs_bmbt_irec_t *s); +extern void xfs_bmbt_disk_set_allf(xfs_bmbt_rec_t *r, xfs_fileoff_t o, + xfs_fsblock_t b, xfs_filblks_t c, xfs_exntst_t v); #else -#define xfs_bmbt_disk_set_all(r, s) \ - xfs_bmbt_set_all(r, s) -#define xfs_bmbt_disk_set_allf(r, o, b, c, v) \ - xfs_bmbt_set_allf(r, o, b, c, v) +#define xfs_bmbt_disk_set_all(r, s) xfs_bmbt_set_all(r, s) +#define xfs_bmbt_disk_set_allf(r, o, b, c, v) xfs_bmbt_set_allf(r, o, b, c, v) #endif /* XFS_NATIVE_HOST */ -void -xfs_bmbt_to_bmdr( - xfs_bmbt_block_t *, - int, - xfs_bmdr_block_t *, - int); - -int -xfs_bmbt_update( - struct xfs_btree_cur *, - xfs_fileoff_t, - xfs_fsblock_t, - xfs_filblks_t, - xfs_exntst_t); +extern void xfs_bmbt_to_bmdr(xfs_bmbt_block_t *, int, xfs_bmdr_block_t *, int); +extern int xfs_bmbt_update(struct xfs_btree_cur *, xfs_fileoff_t, + xfs_fsblock_t, xfs_filblks_t, xfs_exntst_t); #ifdef DEBUG /* * Get the data from the pointed-to record. */ -int -xfs_bmbt_get_rec( - struct xfs_btree_cur *, - xfs_fileoff_t *, - xfs_fsblock_t *, - xfs_filblks_t *, - xfs_exntst_t *, - int *); +extern int xfs_bmbt_get_rec(struct xfs_btree_cur *, xfs_fileoff_t *, + xfs_fsblock_t *, xfs_filblks_t *, + xfs_exntst_t *, int *); #endif - /* * Search an extent list for the extent which includes block * bno. */ -xfs_bmbt_rec_t * -xfs_bmap_do_search_extents( - xfs_bmbt_rec_t *, - xfs_extnum_t, - xfs_extnum_t, - xfs_fileoff_t, - int *, - xfs_extnum_t *, - xfs_bmbt_irec_t *, - xfs_bmbt_irec_t *); +xfs_bmbt_rec_t *xfs_bmap_do_search_extents(xfs_bmbt_rec_t *, + xfs_extnum_t, xfs_extnum_t, xfs_fileoff_t, int *, + xfs_extnum_t *, xfs_bmbt_irec_t *, xfs_bmbt_irec_t *); #endif /* __KERNEL__ */ diff --git a/fs/xfs/xfs_btree.c b/fs/xfs/xfs_btree.c index 0cc63d657a14..52d5d095fc35 100644 --- a/fs/xfs/xfs_btree.c +++ b/fs/xfs/xfs_btree.c @@ -1,45 +1,26 @@ /* - * Copyright (c) 2000-2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - -/* - * This file contains common code for the space manager's btree implementations. - */ - #include "xfs.h" - -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" @@ -47,17 +28,16 @@ #include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" -#include "xfs_alloc_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_btree.h" -#include "xfs_ialloc.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" #include "xfs_inode.h" -#include "xfs_bit.h" +#include "xfs_btree.h" +#include "xfs_ialloc.h" #include "xfs_error.h" /* @@ -110,11 +90,14 @@ xfs_btree_maxrecs( switch (cur->bc_btnum) { case XFS_BTNUM_BNO: case XFS_BTNUM_CNT: - return (int)XFS_ALLOC_BLOCK_MAXRECS(INT_GET(block->bb_h.bb_level, ARCH_CONVERT), cur); + return (int)XFS_ALLOC_BLOCK_MAXRECS( + be16_to_cpu(block->bb_h.bb_level), cur); case XFS_BTNUM_BMAP: - return (int)XFS_BMAP_BLOCK_IMAXRECS(INT_GET(block->bb_h.bb_level, ARCH_CONVERT), cur); + return (int)XFS_BMAP_BLOCK_IMAXRECS( + be16_to_cpu(block->bb_h.bb_level), cur); case XFS_BTNUM_INO: - return (int)XFS_INOBT_BLOCK_MAXRECS(INT_GET(block->bb_h.bb_level, ARCH_CONVERT), cur); + return (int)XFS_INOBT_BLOCK_MAXRECS( + be16_to_cpu(block->bb_h.bb_level), cur); default: ASSERT(0); return 0; @@ -160,7 +143,7 @@ xfs_btree_check_key( k1 = ak1; k2 = ak2; - ASSERT(INT_GET(k1->ar_startblock, ARCH_CONVERT) < INT_GET(k2->ar_startblock, ARCH_CONVERT)); + ASSERT(be32_to_cpu(k1->ar_startblock) < be32_to_cpu(k2->ar_startblock)); break; } case XFS_BTNUM_CNT: { @@ -169,9 +152,9 @@ xfs_btree_check_key( k1 = ak1; k2 = ak2; - ASSERT(INT_GET(k1->ar_blockcount, ARCH_CONVERT) < INT_GET(k2->ar_blockcount, ARCH_CONVERT) || - (INT_GET(k1->ar_blockcount, ARCH_CONVERT) == INT_GET(k2->ar_blockcount, ARCH_CONVERT) && - INT_GET(k1->ar_startblock, ARCH_CONVERT) < INT_GET(k2->ar_startblock, ARCH_CONVERT))); + ASSERT(be32_to_cpu(k1->ar_blockcount) < be32_to_cpu(k2->ar_blockcount) || + (k1->ar_blockcount == k2->ar_blockcount && + be32_to_cpu(k1->ar_startblock) < be32_to_cpu(k2->ar_startblock))); break; } case XFS_BTNUM_BMAP: { @@ -214,16 +197,16 @@ xfs_btree_check_lblock( mp = cur->bc_mp; lblock_ok = - INT_GET(block->bb_magic, ARCH_CONVERT) == xfs_magics[cur->bc_btnum] && - INT_GET(block->bb_level, ARCH_CONVERT) == level && - INT_GET(block->bb_numrecs, ARCH_CONVERT) <= + be32_to_cpu(block->bb_magic) == xfs_magics[cur->bc_btnum] && + be16_to_cpu(block->bb_level) == level && + be16_to_cpu(block->bb_numrecs) <= xfs_btree_maxrecs(cur, (xfs_btree_block_t *)block) && block->bb_leftsib && - (INT_GET(block->bb_leftsib, ARCH_CONVERT) == NULLDFSBNO || - XFS_FSB_SANITY_CHECK(mp, INT_GET(block->bb_leftsib, ARCH_CONVERT))) && + (be64_to_cpu(block->bb_leftsib) == NULLDFSBNO || + XFS_FSB_SANITY_CHECK(mp, be64_to_cpu(block->bb_leftsib))) && block->bb_rightsib && - (INT_GET(block->bb_rightsib, ARCH_CONVERT) == NULLDFSBNO || - XFS_FSB_SANITY_CHECK(mp, INT_GET(block->bb_rightsib, ARCH_CONVERT))); + (be64_to_cpu(block->bb_rightsib) == NULLDFSBNO || + XFS_FSB_SANITY_CHECK(mp, be64_to_cpu(block->bb_rightsib))); if (unlikely(XFS_TEST_ERROR(!lblock_ok, mp, XFS_ERRTAG_BTREE_CHECK_LBLOCK, XFS_RANDOM_BTREE_CHECK_LBLOCK))) { if (bp) @@ -271,8 +254,9 @@ xfs_btree_check_rec( r1 = ar1; r2 = ar2; - ASSERT(INT_GET(r1->ar_startblock, ARCH_CONVERT) + INT_GET(r1->ar_blockcount, ARCH_CONVERT) <= - INT_GET(r2->ar_startblock, ARCH_CONVERT)); + ASSERT(be32_to_cpu(r1->ar_startblock) + + be32_to_cpu(r1->ar_blockcount) <= + be32_to_cpu(r2->ar_startblock)); break; } case XFS_BTNUM_CNT: { @@ -281,9 +265,9 @@ xfs_btree_check_rec( r1 = ar1; r2 = ar2; - ASSERT(INT_GET(r1->ar_blockcount, ARCH_CONVERT) < INT_GET(r2->ar_blockcount, ARCH_CONVERT) || - (INT_GET(r1->ar_blockcount, ARCH_CONVERT) == INT_GET(r2->ar_blockcount, ARCH_CONVERT) && - INT_GET(r1->ar_startblock, ARCH_CONVERT) < INT_GET(r2->ar_startblock, ARCH_CONVERT))); + ASSERT(be32_to_cpu(r1->ar_blockcount) < be32_to_cpu(r2->ar_blockcount) || + (r1->ar_blockcount == r2->ar_blockcount && + be32_to_cpu(r1->ar_startblock) < be32_to_cpu(r2->ar_startblock))); break; } case XFS_BTNUM_BMAP: { @@ -331,17 +315,17 @@ xfs_btree_check_sblock( agbp = cur->bc_private.a.agbp; agf = XFS_BUF_TO_AGF(agbp); - agflen = INT_GET(agf->agf_length, ARCH_CONVERT); + agflen = be32_to_cpu(agf->agf_length); sblock_ok = - INT_GET(block->bb_magic, ARCH_CONVERT) == xfs_magics[cur->bc_btnum] && - INT_GET(block->bb_level, ARCH_CONVERT) == level && - INT_GET(block->bb_numrecs, ARCH_CONVERT) <= + be32_to_cpu(block->bb_magic) == xfs_magics[cur->bc_btnum] && + be16_to_cpu(block->bb_level) == level && + be16_to_cpu(block->bb_numrecs) <= xfs_btree_maxrecs(cur, (xfs_btree_block_t *)block) && - (INT_GET(block->bb_leftsib, ARCH_CONVERT) == NULLAGBLOCK || - INT_GET(block->bb_leftsib, ARCH_CONVERT) < agflen) && + (be32_to_cpu(block->bb_leftsib) == NULLAGBLOCK || + be32_to_cpu(block->bb_leftsib) < agflen) && block->bb_leftsib && - (INT_GET(block->bb_rightsib, ARCH_CONVERT) == NULLAGBLOCK || - INT_GET(block->bb_rightsib, ARCH_CONVERT) < agflen) && + (be32_to_cpu(block->bb_rightsib) == NULLAGBLOCK || + be32_to_cpu(block->bb_rightsib) < agflen) && block->bb_rightsib; if (unlikely(XFS_TEST_ERROR(!sblock_ok, cur->bc_mp, XFS_ERRTAG_BTREE_CHECK_SBLOCK, @@ -372,7 +356,7 @@ xfs_btree_check_sptr( XFS_WANT_CORRUPTED_RETURN( level > 0 && ptr != NULLAGBLOCK && ptr != 0 && - ptr < INT_GET(agf->agf_length, ARCH_CONVERT)); + ptr < be32_to_cpu(agf->agf_length)); return 0; } @@ -611,15 +595,15 @@ xfs_btree_init_cursor( case XFS_BTNUM_BNO: case XFS_BTNUM_CNT: agf = XFS_BUF_TO_AGF(agbp); - nlevels = INT_GET(agf->agf_levels[btnum], ARCH_CONVERT); + nlevels = be32_to_cpu(agf->agf_levels[btnum]); break; case XFS_BTNUM_BMAP: ifp = XFS_IFORK_PTR(ip, whichfork); - nlevels = INT_GET(ifp->if_broot->bb_level, ARCH_CONVERT) + 1; + nlevels = be16_to_cpu(ifp->if_broot->bb_level) + 1; break; case XFS_BTNUM_INO: agi = XFS_BUF_TO_AGI(agbp); - nlevels = INT_GET(agi->agi_level, ARCH_CONVERT); + nlevels = be32_to_cpu(agi->agi_level); break; default: ASSERT(0); @@ -683,9 +667,9 @@ xfs_btree_islastblock( block = xfs_btree_get_block(cur, level, &bp); xfs_btree_check_block(cur, block, level, bp); if (XFS_BTREE_LONG_PTRS(cur->bc_btnum)) - return INT_GET(block->bb_u.l.bb_rightsib, ARCH_CONVERT) == NULLDFSBNO; + return be64_to_cpu(block->bb_u.l.bb_rightsib) == NULLDFSBNO; else - return INT_GET(block->bb_u.s.bb_rightsib, ARCH_CONVERT) == NULLAGBLOCK; + return be32_to_cpu(block->bb_u.s.bb_rightsib) == NULLAGBLOCK; } /* @@ -713,7 +697,7 @@ xfs_btree_lastrec( /* * Set the ptr value to numrecs, that's the last record/key. */ - cur->bc_ptrs[level] = INT_GET(block->bb_h.bb_numrecs, ARCH_CONVERT); + cur->bc_ptrs[level] = be16_to_cpu(block->bb_h.bb_numrecs); return 1; } @@ -883,38 +867,38 @@ xfs_btree_readahead_core( case XFS_BTNUM_BNO: case XFS_BTNUM_CNT: a = XFS_BUF_TO_ALLOC_BLOCK(cur->bc_bufs[lev]); - if ((lr & XFS_BTCUR_LEFTRA) && INT_GET(a->bb_leftsib, ARCH_CONVERT) != NULLAGBLOCK) { + if ((lr & XFS_BTCUR_LEFTRA) && be32_to_cpu(a->bb_leftsib) != NULLAGBLOCK) { xfs_btree_reada_bufs(cur->bc_mp, cur->bc_private.a.agno, - INT_GET(a->bb_leftsib, ARCH_CONVERT), 1); + be32_to_cpu(a->bb_leftsib), 1); rval++; } - if ((lr & XFS_BTCUR_RIGHTRA) && INT_GET(a->bb_rightsib, ARCH_CONVERT) != NULLAGBLOCK) { + if ((lr & XFS_BTCUR_RIGHTRA) && be32_to_cpu(a->bb_rightsib) != NULLAGBLOCK) { xfs_btree_reada_bufs(cur->bc_mp, cur->bc_private.a.agno, - INT_GET(a->bb_rightsib, ARCH_CONVERT), 1); + be32_to_cpu(a->bb_rightsib), 1); rval++; } break; case XFS_BTNUM_BMAP: b = XFS_BUF_TO_BMBT_BLOCK(cur->bc_bufs[lev]); - if ((lr & XFS_BTCUR_LEFTRA) && INT_GET(b->bb_leftsib, ARCH_CONVERT) != NULLDFSBNO) { - xfs_btree_reada_bufl(cur->bc_mp, INT_GET(b->bb_leftsib, ARCH_CONVERT), 1); + if ((lr & XFS_BTCUR_LEFTRA) && be64_to_cpu(b->bb_leftsib) != NULLDFSBNO) { + xfs_btree_reada_bufl(cur->bc_mp, be64_to_cpu(b->bb_leftsib), 1); rval++; } - if ((lr & XFS_BTCUR_RIGHTRA) && INT_GET(b->bb_rightsib, ARCH_CONVERT) != NULLDFSBNO) { - xfs_btree_reada_bufl(cur->bc_mp, INT_GET(b->bb_rightsib, ARCH_CONVERT), 1); + if ((lr & XFS_BTCUR_RIGHTRA) && be64_to_cpu(b->bb_rightsib) != NULLDFSBNO) { + xfs_btree_reada_bufl(cur->bc_mp, be64_to_cpu(b->bb_rightsib), 1); rval++; } break; case XFS_BTNUM_INO: i = XFS_BUF_TO_INOBT_BLOCK(cur->bc_bufs[lev]); - if ((lr & XFS_BTCUR_LEFTRA) && INT_GET(i->bb_leftsib, ARCH_CONVERT) != NULLAGBLOCK) { + if ((lr & XFS_BTCUR_LEFTRA) && be32_to_cpu(i->bb_leftsib) != NULLAGBLOCK) { xfs_btree_reada_bufs(cur->bc_mp, cur->bc_private.i.agno, - INT_GET(i->bb_leftsib, ARCH_CONVERT), 1); + be32_to_cpu(i->bb_leftsib), 1); rval++; } - if ((lr & XFS_BTCUR_RIGHTRA) && INT_GET(i->bb_rightsib, ARCH_CONVERT) != NULLAGBLOCK) { + if ((lr & XFS_BTCUR_RIGHTRA) && be32_to_cpu(i->bb_rightsib) != NULLAGBLOCK) { xfs_btree_reada_bufs(cur->bc_mp, cur->bc_private.i.agno, - INT_GET(i->bb_rightsib, ARCH_CONVERT), 1); + be32_to_cpu(i->bb_rightsib), 1); rval++; } break; @@ -946,14 +930,14 @@ xfs_btree_setbuf( return; b = XFS_BUF_TO_BLOCK(bp); if (XFS_BTREE_LONG_PTRS(cur->bc_btnum)) { - if (INT_GET(b->bb_u.l.bb_leftsib, ARCH_CONVERT) == NULLDFSBNO) + if (be64_to_cpu(b->bb_u.l.bb_leftsib) == NULLDFSBNO) cur->bc_ra[lev] |= XFS_BTCUR_LEFTRA; - if (INT_GET(b->bb_u.l.bb_rightsib, ARCH_CONVERT) == NULLDFSBNO) + if (be64_to_cpu(b->bb_u.l.bb_rightsib) == NULLDFSBNO) cur->bc_ra[lev] |= XFS_BTCUR_RIGHTRA; } else { - if (INT_GET(b->bb_u.s.bb_leftsib, ARCH_CONVERT) == NULLAGBLOCK) + if (be32_to_cpu(b->bb_u.s.bb_leftsib) == NULLAGBLOCK) cur->bc_ra[lev] |= XFS_BTCUR_LEFTRA; - if (INT_GET(b->bb_u.s.bb_rightsib, ARCH_CONVERT) == NULLAGBLOCK) + if (be32_to_cpu(b->bb_u.s.bb_rightsib) == NULLAGBLOCK) cur->bc_ra[lev] |= XFS_BTCUR_RIGHTRA; } } diff --git a/fs/xfs/xfs_btree.h b/fs/xfs/xfs_btree.h index 09b4e1532a35..44f1bd98064a 100644 --- a/fs/xfs/xfs_btree.h +++ b/fs/xfs/xfs_btree.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2001 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2001,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_BTREE_H__ #define __XFS_BTREE_H__ @@ -53,25 +39,23 @@ struct xfs_trans; /* * Short form header: space allocation btrees. */ -typedef struct xfs_btree_sblock -{ - __uint32_t bb_magic; /* magic number for block type */ - __uint16_t bb_level; /* 0 is a leaf */ - __uint16_t bb_numrecs; /* current # of data records */ - xfs_agblock_t bb_leftsib; /* left sibling block or NULLAGBLOCK */ - xfs_agblock_t bb_rightsib; /* right sibling block or NULLAGBLOCK */ +typedef struct xfs_btree_sblock { + __be32 bb_magic; /* magic number for block type */ + __be16 bb_level; /* 0 is a leaf */ + __be16 bb_numrecs; /* current # of data records */ + __be32 bb_leftsib; /* left sibling block or NULLAGBLOCK */ + __be32 bb_rightsib; /* right sibling block or NULLAGBLOCK */ } xfs_btree_sblock_t; /* * Long form header: bmap btrees. */ -typedef struct xfs_btree_lblock -{ - __uint32_t bb_magic; /* magic number for block type */ - __uint16_t bb_level; /* 0 is a leaf */ - __uint16_t bb_numrecs; /* current # of data records */ - xfs_dfsbno_t bb_leftsib; /* left sibling block or NULLDFSBNO */ - xfs_dfsbno_t bb_rightsib; /* right sibling block or NULLDFSBNO */ +typedef struct xfs_btree_lblock { + __be32 bb_magic; /* magic number for block type */ + __be16 bb_level; /* 0 is a leaf */ + __be16 bb_numrecs; /* current # of data records */ + __be64 bb_leftsib; /* left sibling block or NULLDFSBNO */ + __be64 bb_rightsib; /* right sibling block or NULLDFSBNO */ } xfs_btree_lblock_t; /* @@ -79,24 +63,23 @@ typedef struct xfs_btree_lblock */ typedef struct xfs_btree_hdr { - __uint32_t bb_magic; /* magic number for block type */ - __uint16_t bb_level; /* 0 is a leaf */ - __uint16_t bb_numrecs; /* current # of data records */ + __be32 bb_magic; /* magic number for block type */ + __be16 bb_level; /* 0 is a leaf */ + __be16 bb_numrecs; /* current # of data records */ } xfs_btree_hdr_t; -typedef struct xfs_btree_block -{ +typedef struct xfs_btree_block { xfs_btree_hdr_t bb_h; /* header */ - union { - struct { - xfs_agblock_t bb_leftsib; - xfs_agblock_t bb_rightsib; - } s; /* short form pointers */ + union { + struct { + __be32 bb_leftsib; + __be32 bb_rightsib; + } s; /* short form pointers */ struct { - xfs_dfsbno_t bb_leftsib; - xfs_dfsbno_t bb_rightsib; - } l; /* long form pointers */ - } bb_u; /* rest */ + __be64 bb_leftsib; + __be64 bb_rightsib; + } l; /* long form pointers */ + } bb_u; /* rest */ } xfs_btree_block_t; /* @@ -113,12 +96,7 @@ typedef struct xfs_btree_block /* * Boolean to select which form of xfs_btree_block_t.bb_u to use. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BTREE_LONG_PTRS) -int xfs_btree_long_ptrs(xfs_btnum_t btnum); -#define XFS_BTREE_LONG_PTRS(btnum) ((btnum) == XFS_BTNUM_BMAP) -#else #define XFS_BTREE_LONG_PTRS(btnum) ((btnum) == XFS_BTNUM_BMAP) -#endif /* * Magic numbers for btree blocks. @@ -165,7 +143,7 @@ typedef struct xfs_btree_cur struct xfs_trans *bc_tp; /* transaction we're in, if any */ struct xfs_mount *bc_mp; /* file system mount struct */ union { - xfs_alloc_rec_t a; + xfs_alloc_rec_incore_t a; xfs_bmbt_irec_t b; xfs_inobt_rec_t i; } bc_rec; /* current insert/search record value */ @@ -205,24 +183,10 @@ typedef struct xfs_btree_cur /* * Convert from buffer to btree block header. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BUF_TO_BLOCK) -xfs_btree_block_t *xfs_buf_to_block(struct xfs_buf *bp); -#define XFS_BUF_TO_BLOCK(bp) xfs_buf_to_block(bp) -#else -#define XFS_BUF_TO_BLOCK(bp) ((xfs_btree_block_t *)(XFS_BUF_PTR(bp))) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BUF_TO_LBLOCK) -xfs_btree_lblock_t *xfs_buf_to_lblock(struct xfs_buf *bp); -#define XFS_BUF_TO_LBLOCK(bp) xfs_buf_to_lblock(bp) -#else -#define XFS_BUF_TO_LBLOCK(bp) ((xfs_btree_lblock_t *)(XFS_BUF_PTR(bp))) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BUF_TO_SBLOCK) -xfs_btree_sblock_t *xfs_buf_to_sblock(struct xfs_buf *bp); -#define XFS_BUF_TO_SBLOCK(bp) xfs_buf_to_sblock(bp) -#else -#define XFS_BUF_TO_SBLOCK(bp) ((xfs_btree_sblock_t *)(XFS_BUF_PTR(bp))) -#endif +#define XFS_BUF_TO_BLOCK(bp) ((xfs_btree_block_t *)XFS_BUF_PTR(bp)) +#define XFS_BUF_TO_LBLOCK(bp) ((xfs_btree_lblock_t *)XFS_BUF_PTR(bp)) +#define XFS_BUF_TO_SBLOCK(bp) ((xfs_btree_sblock_t *)XFS_BUF_PTR(bp)) + #ifdef __KERNEL__ @@ -477,106 +441,33 @@ xfs_btree_setbuf( /* * Min and max functions for extlen, agblock, fileoff, and filblks types. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_EXTLEN_MIN) -xfs_extlen_t xfs_extlen_min(xfs_extlen_t a, xfs_extlen_t b); -#define XFS_EXTLEN_MIN(a,b) xfs_extlen_min(a,b) -#else #define XFS_EXTLEN_MIN(a,b) \ ((xfs_extlen_t)(a) < (xfs_extlen_t)(b) ? \ - (xfs_extlen_t)(a) : (xfs_extlen_t)(b)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_EXTLEN_MAX) -xfs_extlen_t xfs_extlen_max(xfs_extlen_t a, xfs_extlen_t b); -#define XFS_EXTLEN_MAX(a,b) xfs_extlen_max(a,b) -#else + (xfs_extlen_t)(a) : (xfs_extlen_t)(b)) #define XFS_EXTLEN_MAX(a,b) \ ((xfs_extlen_t)(a) > (xfs_extlen_t)(b) ? \ - (xfs_extlen_t)(a) : (xfs_extlen_t)(b)) -#endif - -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_AGBLOCK_MIN) -xfs_agblock_t xfs_agblock_min(xfs_agblock_t a, xfs_agblock_t b); -#define XFS_AGBLOCK_MIN(a,b) xfs_agblock_min(a,b) -#else + (xfs_extlen_t)(a) : (xfs_extlen_t)(b)) #define XFS_AGBLOCK_MIN(a,b) \ ((xfs_agblock_t)(a) < (xfs_agblock_t)(b) ? \ - (xfs_agblock_t)(a) : (xfs_agblock_t)(b)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_AGBLOCK_MAX) -xfs_agblock_t xfs_agblock_max(xfs_agblock_t a, xfs_agblock_t b); -#define XFS_AGBLOCK_MAX(a,b) xfs_agblock_max(a,b) -#else + (xfs_agblock_t)(a) : (xfs_agblock_t)(b)) #define XFS_AGBLOCK_MAX(a,b) \ ((xfs_agblock_t)(a) > (xfs_agblock_t)(b) ? \ - (xfs_agblock_t)(a) : (xfs_agblock_t)(b)) -#endif - -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_FILEOFF_MIN) -xfs_fileoff_t xfs_fileoff_min(xfs_fileoff_t a, xfs_fileoff_t b); -#define XFS_FILEOFF_MIN(a,b) xfs_fileoff_min(a,b) -#else + (xfs_agblock_t)(a) : (xfs_agblock_t)(b)) #define XFS_FILEOFF_MIN(a,b) \ ((xfs_fileoff_t)(a) < (xfs_fileoff_t)(b) ? \ - (xfs_fileoff_t)(a) : (xfs_fileoff_t)(b)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_FILEOFF_MAX) -xfs_fileoff_t xfs_fileoff_max(xfs_fileoff_t a, xfs_fileoff_t b); -#define XFS_FILEOFF_MAX(a,b) xfs_fileoff_max(a,b) -#else + (xfs_fileoff_t)(a) : (xfs_fileoff_t)(b)) #define XFS_FILEOFF_MAX(a,b) \ ((xfs_fileoff_t)(a) > (xfs_fileoff_t)(b) ? \ - (xfs_fileoff_t)(a) : (xfs_fileoff_t)(b)) -#endif - -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_FILBLKS_MIN) -xfs_filblks_t xfs_filblks_min(xfs_filblks_t a, xfs_filblks_t b); -#define XFS_FILBLKS_MIN(a,b) xfs_filblks_min(a,b) -#else + (xfs_fileoff_t)(a) : (xfs_fileoff_t)(b)) #define XFS_FILBLKS_MIN(a,b) \ ((xfs_filblks_t)(a) < (xfs_filblks_t)(b) ? \ - (xfs_filblks_t)(a) : (xfs_filblks_t)(b)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_FILBLKS_MAX) -xfs_filblks_t xfs_filblks_max(xfs_filblks_t a, xfs_filblks_t b); -#define XFS_FILBLKS_MAX(a,b) xfs_filblks_max(a,b) -#else + (xfs_filblks_t)(a) : (xfs_filblks_t)(b)) #define XFS_FILBLKS_MAX(a,b) \ ((xfs_filblks_t)(a) > (xfs_filblks_t)(b) ? \ - (xfs_filblks_t)(a) : (xfs_filblks_t)(b)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_FSB_SANITY_CHECK) -int xfs_fsb_sanity_check(struct xfs_mount *mp, xfs_fsblock_t fsb); -#define XFS_FSB_SANITY_CHECK(mp,fsb) xfs_fsb_sanity_check(mp,fsb) -#else + (xfs_filblks_t)(a) : (xfs_filblks_t)(b)) + #define XFS_FSB_SANITY_CHECK(mp,fsb) \ (XFS_FSB_TO_AGNO(mp, fsb) < mp->m_sb.sb_agcount && \ - XFS_FSB_TO_AGBNO(mp, fsb) < mp->m_sb.sb_agblocks) -#endif - -/* - * Macros to set EFSCORRUPTED & return/branch. - */ -#define XFS_WANT_CORRUPTED_GOTO(x,l) \ - { \ - int fs_is_ok = (x); \ - ASSERT(fs_is_ok); \ - if (unlikely(!fs_is_ok)) { \ - XFS_ERROR_REPORT("XFS_WANT_CORRUPTED_GOTO", \ - XFS_ERRLEVEL_LOW, NULL); \ - error = XFS_ERROR(EFSCORRUPTED); \ - goto l; \ - } \ - } - -#define XFS_WANT_CORRUPTED_RETURN(x) \ - { \ - int fs_is_ok = (x); \ - ASSERT(fs_is_ok); \ - if (unlikely(!fs_is_ok)) { \ - XFS_ERROR_REPORT("XFS_WANT_CORRUPTED_RETURN", \ - XFS_ERRLEVEL_LOW, NULL); \ - return XFS_ERROR(EFSCORRUPTED); \ - } \ - } + XFS_FSB_TO_AGBNO(mp, fsb) < mp->m_sb.sb_agblocks) #endif /* __XFS_BTREE_H__ */ diff --git a/fs/xfs/xfs_buf_item.c b/fs/xfs/xfs_buf_item.c index a264657acfd9..07e2324152b1 100644 --- a/fs/xfs/xfs_buf_item.c +++ b/fs/xfs/xfs_buf_item.c @@ -1,57 +1,33 @@ /* - * Copyright (c) 2000-2004 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ - */ - -/* - * This file contains the implementation of the xfs_buf_log_item. - * It contains the item operations used to manipulate the buf log - * items as well as utility routines used by the buffer specific - * transaction routines. + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" - -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" -#include "xfs_buf_item.h" #include "xfs_sb.h" #include "xfs_dir.h" #include "xfs_dmapi.h" #include "xfs_mount.h" +#include "xfs_buf_item.h" #include "xfs_trans_priv.h" -#include "xfs_rw.h" -#include "xfs_bit.h" #include "xfs_error.h" diff --git a/fs/xfs/xfs_buf_item.h b/fs/xfs/xfs_buf_item.h index 01aed5f2d579..07c708c2b529 100644 --- a/fs/xfs/xfs_buf_item.h +++ b/fs/xfs/xfs_buf_item.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2001 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2001,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_BUF_ITEM_H__ #define __XFS_BUF_ITEM_H__ diff --git a/fs/xfs/xfs_cap.h b/fs/xfs/xfs_cap.h index 2deac7303758..433ec537f9bd 100644 --- a/fs/xfs/xfs_cap.h +++ b/fs/xfs/xfs_cap.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_CAP_H__ #define __XFS_CAP_H__ diff --git a/fs/xfs/xfs_clnt.h b/fs/xfs/xfs_clnt.h index b3215ffe0be8..328a528b926d 100644 --- a/fs/xfs/xfs_clnt.h +++ b/fs/xfs/xfs_clnt.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_CLNT_H__ #define __XFS_CLNT_H__ @@ -55,6 +41,7 @@ */ struct xfs_mount_args { int flags; /* flags -> see XFSMNT_... macros below */ + int flags2; /* flags -> see XFSMNT2_... macros below */ int logbufs; /* Number of log buffers, -1 to default */ int logbufsize; /* Size of log buffers, -1 to default */ char fsname[MAXNAMELEN+1]; /* data device name */ @@ -68,9 +55,9 @@ struct xfs_mount_args { }; /* - * XFS mount option flags + * XFS mount option flags -- args->flags1 */ -#define XFSMNT_CHKLOG 0x00000001 /* check log */ +#define XFSMNT_COMPAT_ATTR 0x00000001 /* do not use ATTR2 format */ #define XFSMNT_WSYNC 0x00000002 /* safe mode nfs mount * compatible */ #define XFSMNT_INO64 0x00000004 /* move inode numbers up @@ -91,7 +78,7 @@ struct xfs_mount_args { #define XFSMNT_SHARED 0x00001000 /* shared XFS mount */ #define XFSMNT_IOSIZE 0x00002000 /* optimize for I/O size */ #define XFSMNT_OSYNCISOSYNC 0x00004000 /* o_sync is REALLY o_sync */ - /* (osyncisdsync is now default) */ + /* (osyncisdsync is default) */ #define XFSMNT_32BITINODES 0x00200000 /* restrict inodes to 32 * bits of address space */ #define XFSMNT_GQUOTA 0x00400000 /* group quota accounting */ @@ -99,12 +86,19 @@ struct xfs_mount_args { * enforcement */ #define XFSMNT_NOUUID 0x01000000 /* Ignore fs uuid */ #define XFSMNT_DMAPI 0x02000000 /* enable dmapi/xdsm */ -#define XFSMNT_NOLOGFLUSH 0x04000000 /* Don't flush for log blocks */ +#define XFSMNT_BARRIER 0x04000000 /* use write barriers */ #define XFSMNT_IDELETE 0x08000000 /* inode cluster delete */ #define XFSMNT_SWALLOC 0x10000000 /* turn on stripe width * allocation */ #define XFSMNT_IHASHSIZE 0x20000000 /* inode hash table size */ #define XFSMNT_DIRSYNC 0x40000000 /* sync creat,link,unlink,rename * symlink,mkdir,rmdir,mknod */ +#define XFSMNT_FLAGS2 0x80000000 /* more flags set in flags2 */ + +/* + * XFS mount option flags -- args->flags2 + */ +#define XFSMNT2_COMPAT_IOSIZE 0x00000001 /* don't report large preferred + * I/O size in stat(2) */ #endif /* __XFS_CLNT_H__ */ diff --git a/fs/xfs/xfs_da_btree.c b/fs/xfs/xfs_da_btree.c index 8b792ddf2164..473671fa5c13 100644 --- a/fs/xfs/xfs_da_btree.c +++ b/fs/xfs/xfs_da_btree.c @@ -1,41 +1,26 @@ /* - * Copyright (c) 2000-2004 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" - -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" @@ -43,19 +28,19 @@ #include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" -#include "xfs_alloc_btree.h" +#include "xfs_da_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_alloc.h" -#include "xfs_btree.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" -#include "xfs_inode_item.h" #include "xfs_inode.h" +#include "xfs_inode_item.h" +#include "xfs_alloc.h" +#include "xfs_btree.h" #include "xfs_bmap.h" -#include "xfs_da_btree.h" #include "xfs_attr.h" #include "xfs_attr_leaf.h" #include "xfs_dir_leaf.h" @@ -64,7 +49,6 @@ #include "xfs_dir2_block.h" #include "xfs_dir2_node.h" #include "xfs_error.h" -#include "xfs_bit.h" /* * xfs_da_btree.c @@ -190,9 +174,6 @@ xfs_da_split(xfs_da_state_t *state) */ switch (oldblk->magic) { case XFS_ATTR_LEAF_MAGIC: -#ifndef __KERNEL__ - return(ENOTTY); -#else error = xfs_attr_leaf_split(state, oldblk, newblk); if ((error != 0) && (error != ENOSPC)) { return(error); /* GROT: attr is inconsistent */ @@ -218,7 +199,6 @@ xfs_da_split(xfs_da_state_t *state) return(error); /* GROT: attr inconsistent */ addblk = newblk; break; -#endif case XFS_DIR_LEAF_MAGIC: ASSERT(XFS_DIR_IS_V1(state->mp)); error = xfs_dir_leaf_split(state, oldblk, newblk); @@ -449,7 +429,8 @@ xfs_da_node_split(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk, /* * With V2 the extra block is data or freespace. */ - useextra = state->extravalid && XFS_DIR_IS_V1(state->mp); + useextra = state->extravalid && (XFS_DIR_IS_V1(state->mp) || + state->args->whichfork == XFS_ATTR_FORK); newcount = 1 + useextra; /* * Do we have to split the node? @@ -706,18 +687,12 @@ xfs_da_join(xfs_da_state_t *state) */ switch (drop_blk->magic) { case XFS_ATTR_LEAF_MAGIC: -#ifndef __KERNEL__ - error = ENOTTY; -#else error = xfs_attr_leaf_toosmall(state, &action); -#endif if (error) return(error); if (action == 0) return(0); -#ifdef __KERNEL__ xfs_attr_leaf_unbalance(state, drop_blk, save_blk); -#endif break; case XFS_DIR_LEAF_MAGIC: ASSERT(XFS_DIR_IS_V1(state->mp)); @@ -973,13 +948,11 @@ xfs_da_fixhashpath(xfs_da_state_t *state, xfs_da_state_path_t *path) level = path->active-1; blk = &path->blk[ level ]; switch (blk->magic) { -#ifdef __KERNEL__ case XFS_ATTR_LEAF_MAGIC: lasthash = xfs_attr_leaf_lasthash(blk->bp, &count); if (count == 0) return; break; -#endif case XFS_DIR_LEAF_MAGIC: ASSERT(XFS_DIR_IS_V1(state->mp)); lasthash = xfs_dir_leaf_lasthash(blk->bp, &count); @@ -1220,12 +1193,10 @@ xfs_da_node_lookup_int(xfs_da_state_t *state, int *result) blkno = INT_GET(btree->before, ARCH_CONVERT); } } -#ifdef __KERNEL__ else if (INT_GET(curr->magic, ARCH_CONVERT) == XFS_ATTR_LEAF_MAGIC) { blk->hashval = xfs_attr_leaf_lasthash(blk->bp, NULL); break; } -#endif else if (INT_GET(curr->magic, ARCH_CONVERT) == XFS_DIR_LEAF_MAGIC) { blk->hashval = xfs_dir_leaf_lasthash(blk->bp, NULL); break; @@ -1252,13 +1223,11 @@ xfs_da_node_lookup_int(xfs_da_state_t *state, int *result) retval = xfs_dir2_leafn_lookup_int(blk->bp, args, &blk->index, state); } -#ifdef __KERNEL__ else if (blk->magic == XFS_ATTR_LEAF_MAGIC) { retval = xfs_attr_leaf_lookup_int(blk->bp, args); blk->index = args->index; args->blkno = blk->blkno; } -#endif if (((retval == ENOENT) || (retval == ENOATTR)) && (blk->hashval == args->hashval)) { error = xfs_da_path_shift(state, &state->path, 1, 1, @@ -1268,12 +1237,10 @@ xfs_da_node_lookup_int(xfs_da_state_t *state, int *result) if (retval == 0) { continue; } -#ifdef __KERNEL__ else if (blk->magic == XFS_ATTR_LEAF_MAGIC) { /* path_shift() gives ENOENT */ retval = XFS_ERROR(ENOATTR); } -#endif } break; } @@ -1312,11 +1279,9 @@ xfs_da_blk_link(xfs_da_state_t *state, xfs_da_state_blk_t *old_blk, ASSERT(old_blk->magic == new_blk->magic); switch (old_blk->magic) { -#ifdef __KERNEL__ case XFS_ATTR_LEAF_MAGIC: before = xfs_attr_leaf_order(old_blk->bp, new_blk->bp); break; -#endif case XFS_DIR_LEAF_MAGIC: ASSERT(XFS_DIR_IS_V1(state->mp)); before = xfs_dir_leaf_order(old_blk->bp, new_blk->bp); @@ -1587,12 +1552,10 @@ xfs_da_path_shift(xfs_da_state_t *state, xfs_da_state_path_t *path, ASSERT(level == path->active-1); blk->index = 0; switch(blk->magic) { -#ifdef __KERNEL__ case XFS_ATTR_LEAF_MAGIC: blk->hashval = xfs_attr_leaf_lasthash(blk->bp, NULL); break; -#endif case XFS_DIR_LEAF_MAGIC: ASSERT(XFS_DIR_IS_V1(state->mp)); blk->hashval = xfs_dir_leaf_lasthash(blk->bp, @@ -1626,19 +1589,10 @@ xfs_da_path_shift(xfs_da_state_t *state, xfs_da_state_path_t *path, * This is implemented with some source-level loop unrolling. */ xfs_dahash_t -xfs_da_hashname(uchar_t *name, int namelen) +xfs_da_hashname(const uchar_t *name, int namelen) { xfs_dahash_t hash; -#ifdef SLOWVERSION - /* - * This is the old one-byte-at-a-time version. - */ - for (hash = 0; namelen > 0; namelen--) - hash = *name++ ^ rol32(hash, 7); - - return(hash); -#else /* * Do four characters at a time as long as we can. */ @@ -1657,12 +1611,9 @@ xfs_da_hashname(uchar_t *name, int namelen) return (name[0] << 7) ^ (name[1] << 0) ^ rol32(hash, 7 * 2); case 1: return (name[0] << 0) ^ rol32(hash, 7 * 1); - case 0: + default: /* case 0: */ return hash; } - /* NOTREACHED */ -#endif - return 0; /* keep gcc happy */ } /* @@ -2200,20 +2151,16 @@ xfs_da_do_buf( error = bp ? XFS_BUF_GETERROR(bp) : XFS_ERROR(EIO); break; case 1: -#ifndef __KERNEL__ case 2: -#endif bp = NULL; error = xfs_trans_read_buf(mp, trans, mp->m_ddev_targp, mappedbno, nmapped, 0, &bp); break; -#ifdef __KERNEL__ case 3: xfs_baread(mp->m_ddev_targp, mappedbno, nmapped); error = 0; bp = NULL; break; -#endif } if (error) { if (bp) diff --git a/fs/xfs/xfs_da_btree.h b/fs/xfs/xfs_da_btree.h index 3a9b9e809c60..41352113721a 100644 --- a/fs/xfs/xfs_da_btree.h +++ b/fs/xfs/xfs_da_btree.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000, 2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000,2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_DA_BTREE_H__ #define __XFS_DA_BTREE_H__ @@ -92,72 +78,24 @@ typedef struct xfs_da_node_entry xfs_da_node_entry_t; #define XFS_DA_MAXHASH ((xfs_dahash_t)-1) /* largest valid hash value */ -/* - * Macros used by directory code to interface to the filesystem. - */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_LBSIZE) -int xfs_lbsize(struct xfs_mount *mp); -#define XFS_LBSIZE(mp) xfs_lbsize(mp) -#else -#define XFS_LBSIZE(mp) ((mp)->m_sb.sb_blocksize) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_LBLOG) -int xfs_lblog(struct xfs_mount *mp); -#define XFS_LBLOG(mp) xfs_lblog(mp) -#else -#define XFS_LBLOG(mp) ((mp)->m_sb.sb_blocklog) -#endif +#define XFS_LBSIZE(mp) (mp)->m_sb.sb_blocksize +#define XFS_LBLOG(mp) (mp)->m_sb.sb_blocklog -/* - * Macros used by directory code to interface to the kernel - */ - -/* - * Macros used to manipulate directory off_t's - */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DA_MAKE_BNOENTRY) -__uint32_t xfs_da_make_bnoentry(struct xfs_mount *mp, xfs_dablk_t bno, - int entry); #define XFS_DA_MAKE_BNOENTRY(mp,bno,entry) \ - xfs_da_make_bnoentry(mp,bno,entry) -#else -#define XFS_DA_MAKE_BNOENTRY(mp,bno,entry) \ (((bno) << (mp)->m_dircook_elog) | (entry)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DA_MAKE_COOKIE) -xfs_off_t xfs_da_make_cookie(struct xfs_mount *mp, xfs_dablk_t bno, int entry, - xfs_dahash_t hash); #define XFS_DA_MAKE_COOKIE(mp,bno,entry,hash) \ - xfs_da_make_cookie(mp,bno,entry,hash) -#else -#define XFS_DA_MAKE_COOKIE(mp,bno,entry,hash) \ (((xfs_off_t)XFS_DA_MAKE_BNOENTRY(mp, bno, entry) << 32) | (hash)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DA_COOKIE_HASH) -xfs_dahash_t xfs_da_cookie_hash(struct xfs_mount *mp, xfs_off_t cookie); -#define XFS_DA_COOKIE_HASH(mp,cookie) xfs_da_cookie_hash(mp,cookie) -#else -#define XFS_DA_COOKIE_HASH(mp,cookie) ((xfs_dahash_t)(cookie)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DA_COOKIE_BNO) -xfs_dablk_t xfs_da_cookie_bno(struct xfs_mount *mp, xfs_off_t cookie); -#define XFS_DA_COOKIE_BNO(mp,cookie) xfs_da_cookie_bno(mp,cookie) -#else -#define XFS_DA_COOKIE_BNO(mp,cookie) \ - (((xfs_off_t)(cookie) >> 31) == -1LL ? \ +#define XFS_DA_COOKIE_HASH(mp,cookie) ((xfs_dahash_t)cookie) +#define XFS_DA_COOKIE_BNO(mp,cookie) \ + ((((xfs_off_t)(cookie) >> 31) == -1LL ? \ (xfs_dablk_t)0 : \ - (xfs_dablk_t)((xfs_off_t)(cookie) >> ((mp)->m_dircook_elog + 32))) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DA_COOKIE_ENTRY) -int xfs_da_cookie_entry(struct xfs_mount *mp, xfs_off_t cookie); -#define XFS_DA_COOKIE_ENTRY(mp,cookie) xfs_da_cookie_entry(mp,cookie) -#else -#define XFS_DA_COOKIE_ENTRY(mp,cookie) \ - (((xfs_off_t)(cookie) >> 31) == -1LL ? \ + (xfs_dablk_t)((xfs_off_t)(cookie) >> \ + ((mp)->m_dircook_elog + 32)))) +#define XFS_DA_COOKIE_ENTRY(mp,cookie) \ + ((((xfs_off_t)(cookie) >> 31) == -1LL ? \ (xfs_dablk_t)0 : \ (xfs_dablk_t)(((xfs_off_t)(cookie) >> 32) & \ - ((1 << (mp)->m_dircook_elog) - 1))) -#endif + ((1 << (mp)->m_dircook_elog) - 1)))) /*======================================================================== @@ -168,7 +106,7 @@ int xfs_da_cookie_entry(struct xfs_mount *mp, xfs_off_t cookie); * Structure to ease passing around component names. */ typedef struct xfs_da_args { - uchar_t *name; /* string (maybe not NULL terminated) */ + const uchar_t *name; /* string (maybe not NULL terminated) */ int namelen; /* length of string (maybe no NULL) */ uchar_t *value; /* set of bytes (maybe contain NULLs) */ int valuelen; /* length of value */ @@ -314,7 +252,7 @@ xfs_daddr_t xfs_da_reada_buf(struct xfs_trans *trans, struct xfs_inode *dp, int xfs_da_shrink_inode(xfs_da_args_t *args, xfs_dablk_t dead_blkno, xfs_dabuf_t *dead_buf); -uint xfs_da_hashname(uchar_t *name_string, int name_length); +uint xfs_da_hashname(const uchar_t *name_string, int name_length); uint xfs_da_log2_roundup(uint i); xfs_da_state_t *xfs_da_state_alloc(void); void xfs_da_state_free(xfs_da_state_t *state); diff --git a/fs/xfs/xfs_dfrag.c b/fs/xfs/xfs_dfrag.c index 681be5c93af5..070259a4254c 100644 --- a/fs/xfs/xfs_dfrag.c +++ b/fs/xfs/xfs_dfrag.c @@ -1,58 +1,44 @@ /* - * Copyright (c) 2000-2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" +#include "xfs_ag.h" #include "xfs_dir.h" #include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" -#include "xfs_ag.h" -#include "xfs_alloc_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_btree.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" -#include "xfs_inode_item.h" #include "xfs_inode.h" +#include "xfs_inode_item.h" #include "xfs_bmap.h" +#include "xfs_btree.h" #include "xfs_ialloc.h" #include "xfs_itable.h" #include "xfs_dfrag.h" @@ -65,9 +51,9 @@ */ int xfs_swapext( - xfs_swapext_t __user *sxp) + xfs_swapext_t __user *sxu) { - xfs_swapext_t sx; + xfs_swapext_t *sxp; xfs_inode_t *ip=NULL, *tip=NULL, *ips[2]; xfs_trans_t *tp; xfs_mount_t *mp; @@ -76,20 +62,29 @@ xfs_swapext( vnode_t *vp, *tvp; bhv_desc_t *bdp, *tbdp; vn_bhv_head_t *bhp, *tbhp; - uint lock_flags=0; + static uint lock_flags = XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL; int ilf_fields, tilf_fields; int error = 0; - xfs_ifork_t tempif, *ifp, *tifp; + xfs_ifork_t *tempifp, *ifp, *tifp; __uint64_t tmp; int aforkblks = 0; int taforkblks = 0; - int locked = 0; + char locked = 0; - if (copy_from_user(&sx, sxp, sizeof(sx))) - return XFS_ERROR(EFAULT); + sxp = kmem_alloc(sizeof(xfs_swapext_t), KM_MAYFAIL); + tempifp = kmem_alloc(sizeof(xfs_ifork_t), KM_MAYFAIL); + if (!sxp || !tempifp) { + error = XFS_ERROR(ENOMEM); + goto error0; + } + + if (copy_from_user(sxp, sxu, sizeof(xfs_swapext_t))) { + error = XFS_ERROR(EFAULT); + goto error0; + } /* Pull information for the target fd */ - if (((fp = fget((int)sx.sx_fdtarget)) == NULL) || + if (((fp = fget((int)sxp->sx_fdtarget)) == NULL) || ((vp = LINVFS_GET_VP(fp->f_dentry->d_inode)) == NULL)) { error = XFS_ERROR(EINVAL); goto error0; @@ -104,7 +99,7 @@ xfs_swapext( ip = XFS_BHVTOI(bdp); } - if (((tfp = fget((int)sx.sx_fdtmp)) == NULL) || + if (((tfp = fget((int)sxp->sx_fdtmp)) == NULL) || ((tvp = LINVFS_GET_VP(tfp->f_dentry->d_inode)) == NULL)) { error = XFS_ERROR(EINVAL); goto error0; @@ -131,7 +126,7 @@ xfs_swapext( mp = ip->i_mount; - sbp = &sx.sx_stat; + sbp = &sxp->sx_stat; if (XFS_FORCED_SHUTDOWN(mp)) { error = XFS_ERROR(EIO); @@ -148,7 +143,7 @@ xfs_swapext( ips[0] = tip; ips[1] = ip; } - lock_flags = XFS_ILOCK_EXCL | XFS_IOLOCK_EXCL; + xfs_lock_inodes(ips, 2, 0, lock_flags); /* Check permissions */ @@ -192,9 +187,9 @@ xfs_swapext( } /* Verify all data are being swapped */ - if (sx.sx_offset != 0 || - sx.sx_length != ip->i_d.di_size || - sx.sx_length != tip->i_d.di_size) { + if (sxp->sx_offset != 0 || + sxp->sx_length != ip->i_d.di_size || + sxp->sx_length != tip->i_d.di_size) { error = XFS_ERROR(EFAULT); goto error0; } @@ -255,7 +250,8 @@ xfs_swapext( xfs_iunlock(ip, XFS_IOLOCK_EXCL); xfs_iunlock(tip, XFS_IOLOCK_EXCL); xfs_trans_cancel(tp, 0); - return error; + locked = 0; + goto error0; } xfs_lock_inodes(ips, 2, 0, XFS_ILOCK_EXCL); @@ -266,10 +262,8 @@ xfs_swapext( (ip->i_d.di_aformat != XFS_DINODE_FMT_LOCAL)) { error = xfs_bmap_count_blocks(tp, ip, XFS_ATTR_FORK, &aforkblks); if (error) { - xfs_iunlock(ip, lock_flags); - xfs_iunlock(tip, lock_flags); xfs_trans_cancel(tp, 0); - return error; + goto error0; } } if ( ((XFS_IFORK_Q(tip) != 0) && (tip->i_d.di_anextents > 0)) && @@ -277,10 +271,8 @@ xfs_swapext( error = xfs_bmap_count_blocks(tp, tip, XFS_ATTR_FORK, &taforkblks); if (error) { - xfs_iunlock(ip, lock_flags); - xfs_iunlock(tip, lock_flags); xfs_trans_cancel(tp, 0); - return error; + goto error0; } } @@ -289,9 +281,9 @@ xfs_swapext( */ ifp = &ip->i_df; tifp = &tip->i_df; - tempif = *ifp; /* struct copy */ - *ifp = *tifp; /* struct copy */ - *tifp = tempif; /* struct copy */ + *tempifp = *ifp; /* struct copy */ + *ifp = *tifp; /* struct copy */ + *tifp = *tempifp; /* struct copy */ /* * Fix the on-disk inode values @@ -369,11 +361,7 @@ xfs_swapext( } error = xfs_trans_commit(tp, XFS_TRANS_SWAPEXT, NULL); - - fput(fp); - fput(tfp); - - return error; + locked = 0; error0: if (locked) { @@ -381,8 +369,15 @@ xfs_swapext( xfs_iunlock(tip, lock_flags); } - if (fp != NULL) fput(fp); - if (tfp != NULL) fput(tfp); + if (fp != NULL) + fput(fp); + if (tfp != NULL) + fput(tfp); + + if (sxp != NULL) + kmem_free(sxp, sizeof(xfs_swapext_t)); + if (tempifp != NULL) + kmem_free(tempifp, sizeof(xfs_ifork_t)); return error; } diff --git a/fs/xfs/xfs_dfrag.h b/fs/xfs/xfs_dfrag.h index 904860594b8f..f678559abc45 100644 --- a/fs/xfs/xfs_dfrag.h +++ b/fs/xfs/xfs_dfrag.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_DFRAG_H__ #define __XFS_DFRAG_H__ diff --git a/fs/xfs/xfs_dinode.h b/fs/xfs/xfs_dinode.h index f5c932b064e6..c5a0e537ff1a 100644 --- a/fs/xfs/xfs_dinode.h +++ b/fs/xfs/xfs_dinode.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000, 2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000,2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_DINODE_H__ #define __XFS_DINODE_H__ @@ -37,13 +23,8 @@ struct xfs_mount; #define XFS_DINODE_VERSION_1 1 #define XFS_DINODE_VERSION_2 2 -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DINODE_GOOD_VERSION) -int xfs_dinode_good_version(int v); -#define XFS_DINODE_GOOD_VERSION(v) xfs_dinode_good_version(v) -#else -#define XFS_DINODE_GOOD_VERSION(v) (((v) == XFS_DINODE_VERSION_1) || \ - ((v) == XFS_DINODE_VERSION_2)) -#endif +#define XFS_DINODE_GOOD_VERSION(v) \ + (((v) == XFS_DINODE_VERSION_1 || (v) == XFS_DINODE_VERSION_2)) #define XFS_DINODE_MAGIC 0x494e /* 'IN' */ /* @@ -184,75 +165,30 @@ typedef enum xfs_dinode_fmt /* * Inode size for given fs. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_LITINO) -int xfs_litino(struct xfs_mount *mp); -#define XFS_LITINO(mp) xfs_litino(mp) -#else #define XFS_LITINO(mp) ((mp)->m_litino) -#endif #define XFS_BROOT_SIZE_ADJ \ (sizeof(xfs_bmbt_block_t) - sizeof(xfs_bmdr_block_t)) /* - * Fork identifiers. Here so utilities can use them without including - * xfs_inode.h. - */ -#define XFS_DATA_FORK 0 -#define XFS_ATTR_FORK 1 - -/* * Inode data & attribute fork sizes, per inode. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_CFORK_Q) -int xfs_cfork_q_disk(xfs_dinode_core_t *dcp); -int xfs_cfork_q(xfs_dinode_core_t *dcp); -#define XFS_CFORK_Q_DISK(dcp) xfs_cfork_q_disk(dcp) -#define XFS_CFORK_Q(dcp) xfs_cfork_q(dcp) -#else -#define XFS_CFORK_Q_DISK(dcp) ((dcp)->di_forkoff != 0) #define XFS_CFORK_Q(dcp) ((dcp)->di_forkoff != 0) +#define XFS_CFORK_Q_DISK(dcp) ((dcp)->di_forkoff != 0) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_CFORK_BOFF) -int xfs_cfork_boff_disk(xfs_dinode_core_t *dcp); -int xfs_cfork_boff(xfs_dinode_core_t *dcp); -#define XFS_CFORK_BOFF_DISK(dcp) xfs_cfork_boff_disk(dcp) -#define XFS_CFORK_BOFF(dcp) xfs_cfork_boff(dcp) -#else -#define XFS_CFORK_BOFF_DISK(dcp) ((int)(INT_GET((dcp)->di_forkoff, ARCH_CONVERT) << 3)) #define XFS_CFORK_BOFF(dcp) ((int)((dcp)->di_forkoff << 3)) +#define XFS_CFORK_BOFF_DISK(dcp) \ + ((int)(INT_GET((dcp)->di_forkoff, ARCH_CONVERT) << 3)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_CFORK_DSIZE) -int xfs_cfork_dsize_disk(xfs_dinode_core_t *dcp, struct xfs_mount *mp); -int xfs_cfork_dsize(xfs_dinode_core_t *dcp, struct xfs_mount *mp); -#define XFS_CFORK_DSIZE_DISK(dcp,mp) xfs_cfork_dsize_disk(dcp,mp) -#define XFS_CFORK_DSIZE(dcp,mp) xfs_cfork_dsize(dcp,mp) -#else #define XFS_CFORK_DSIZE_DISK(dcp,mp) \ (XFS_CFORK_Q_DISK(dcp) ? XFS_CFORK_BOFF_DISK(dcp) : XFS_LITINO(mp)) #define XFS_CFORK_DSIZE(dcp,mp) \ (XFS_CFORK_Q(dcp) ? XFS_CFORK_BOFF(dcp) : XFS_LITINO(mp)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_CFORK_ASIZE) -int xfs_cfork_asize_disk(xfs_dinode_core_t *dcp, struct xfs_mount *mp); -int xfs_cfork_asize(xfs_dinode_core_t *dcp, struct xfs_mount *mp); -#define XFS_CFORK_ASIZE_DISK(dcp,mp) xfs_cfork_asize_disk(dcp,mp) -#define XFS_CFORK_ASIZE(dcp,mp) xfs_cfork_asize(dcp,mp) -#else #define XFS_CFORK_ASIZE_DISK(dcp,mp) \ (XFS_CFORK_Q_DISK(dcp) ? XFS_LITINO(mp) - XFS_CFORK_BOFF_DISK(dcp) : 0) #define XFS_CFORK_ASIZE(dcp,mp) \ (XFS_CFORK_Q(dcp) ? XFS_LITINO(mp) - XFS_CFORK_BOFF(dcp) : 0) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_CFORK_SIZE) -int xfs_cfork_size_disk(xfs_dinode_core_t *dcp, struct xfs_mount *mp, int w); -int xfs_cfork_size(xfs_dinode_core_t *dcp, struct xfs_mount *mp, int w); -#define XFS_CFORK_SIZE_DISK(dcp,mp,w) xfs_cfork_size_disk(dcp,mp,w) -#define XFS_CFORK_SIZE(dcp,mp,w) xfs_cfork_size(dcp,mp,w) -#else #define XFS_CFORK_SIZE_DISK(dcp,mp,w) \ ((w) == XFS_DATA_FORK ? \ XFS_CFORK_DSIZE_DISK(dcp, mp) : \ @@ -261,93 +197,26 @@ int xfs_cfork_size(xfs_dinode_core_t *dcp, struct xfs_mount *mp, int w); ((w) == XFS_DATA_FORK ? \ XFS_CFORK_DSIZE(dcp, mp) : XFS_CFORK_ASIZE(dcp, mp)) -#endif - -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DFORK_DSIZE) -int xfs_dfork_dsize(xfs_dinode_t *dip, struct xfs_mount *mp); -#define XFS_DFORK_DSIZE(dip,mp) xfs_dfork_dsize(dip,mp) -#else -#define XFS_DFORK_DSIZE(dip,mp) XFS_CFORK_DSIZE_DISK(&(dip)->di_core, mp) - -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DFORK_ASIZE) -int xfs_dfork_asize(xfs_dinode_t *dip, struct xfs_mount *mp); -#define XFS_DFORK_ASIZE(dip,mp) xfs_dfork_asize(dip,mp) -#else -#define XFS_DFORK_ASIZE(dip,mp) XFS_CFORK_ASIZE_DISK(&(dip)->di_core, mp) +#define XFS_DFORK_DSIZE(dip,mp) \ + XFS_CFORK_DSIZE_DISK(&(dip)->di_core, mp) +#define XFS_DFORK_ASIZE(dip,mp) \ + XFS_CFORK_ASIZE_DISK(&(dip)->di_core, mp) +#define XFS_DFORK_SIZE(dip,mp,w) \ + XFS_CFORK_SIZE_DISK(&(dip)->di_core, mp, w) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DFORK_SIZE) -int xfs_dfork_size(xfs_dinode_t *dip, struct xfs_mount *mp, int w); -#define XFS_DFORK_SIZE(dip,mp,w) xfs_dfork_size(dip,mp,w) -#else -#define XFS_DFORK_SIZE(dip,mp,w) XFS_CFORK_SIZE_DISK(&(dip)->di_core, mp, w) - -#endif - -/* - * Macros for accessing per-fork disk inode information. - */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DFORK_Q) -int xfs_dfork_q(xfs_dinode_t *dip); -#define XFS_DFORK_Q(dip) xfs_dfork_q(dip) -#else #define XFS_DFORK_Q(dip) XFS_CFORK_Q_DISK(&(dip)->di_core) - -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DFORK_BOFF) -int xfs_dfork_boff(xfs_dinode_t *dip); -#define XFS_DFORK_BOFF(dip) xfs_dfork_boff(dip) -#else #define XFS_DFORK_BOFF(dip) XFS_CFORK_BOFF_DISK(&(dip)->di_core) - -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DFORK_DPTR) -char *xfs_dfork_dptr(xfs_dinode_t *dip); -#define XFS_DFORK_DPTR(dip) xfs_dfork_dptr(dip) -#else #define XFS_DFORK_DPTR(dip) ((dip)->di_u.di_c) - -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DFORK_APTR) -char *xfs_dfork_aptr(xfs_dinode_t *dip); -#define XFS_DFORK_APTR(dip) xfs_dfork_aptr(dip) -#else -#define XFS_DFORK_APTR(dip) ((dip)->di_u.di_c + XFS_DFORK_BOFF(dip)) - -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DFORK_PTR) -char *xfs_dfork_ptr(xfs_dinode_t *dip, int w); -#define XFS_DFORK_PTR(dip,w) xfs_dfork_ptr(dip,w) -#else +#define XFS_DFORK_APTR(dip) \ + ((dip)->di_u.di_c + XFS_DFORK_BOFF(dip)) #define XFS_DFORK_PTR(dip,w) \ ((w) == XFS_DATA_FORK ? XFS_DFORK_DPTR(dip) : XFS_DFORK_APTR(dip)) - -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_CFORK_FORMAT) -int xfs_cfork_format(xfs_dinode_core_t *dcp, int w); -#define XFS_CFORK_FORMAT(dcp,w) xfs_cfork_format(dcp,w) -#else -#define XFS_CFORK_FORMAT(dcp,w) \ +#define XFS_CFORK_FORMAT(dcp,w) \ ((w) == XFS_DATA_FORK ? (dcp)->di_format : (dcp)->di_aformat) - -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_CFORK_FMT_SET) -void xfs_cfork_fmt_set(xfs_dinode_core_t *dcp, int w, int n); -#define XFS_CFORK_FMT_SET(dcp,w,n) xfs_cfork_fmt_set(dcp,w,n) -#else #define XFS_CFORK_FMT_SET(dcp,w,n) \ ((w) == XFS_DATA_FORK ? \ - ((dcp)->di_format = (n)) : \ - ((dcp)->di_aformat = (n))) + ((dcp)->di_format = (n)) : ((dcp)->di_aformat = (n))) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_CFORK_NEXTENTS) -int xfs_cfork_nextents_disk(xfs_dinode_core_t *dcp, int w); -int xfs_cfork_nextents(xfs_dinode_core_t *dcp, int w); -#define XFS_CFORK_NEXTENTS_DISK(dcp,w) xfs_cfork_nextents_disk(dcp,w) -#define XFS_CFORK_NEXTENTS(dcp,w) xfs_cfork_nextents(dcp,w) -#else #define XFS_CFORK_NEXTENTS_DISK(dcp,w) \ ((w) == XFS_DATA_FORK ? \ INT_GET((dcp)->di_nextents, ARCH_CONVERT) : \ @@ -355,31 +224,13 @@ int xfs_cfork_nextents(xfs_dinode_core_t *dcp, int w); #define XFS_CFORK_NEXTENTS(dcp,w) \ ((w) == XFS_DATA_FORK ? (dcp)->di_nextents : (dcp)->di_anextents) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_CFORK_NEXT_SET) -void xfs_cfork_next_set(xfs_dinode_core_t *dcp, int w, int n); -#define XFS_CFORK_NEXT_SET(dcp,w,n) xfs_cfork_next_set(dcp,w,n) -#else #define XFS_CFORK_NEXT_SET(dcp,w,n) \ ((w) == XFS_DATA_FORK ? \ - ((dcp)->di_nextents = (n)) : \ - ((dcp)->di_anextents = (n))) - -#endif + ((dcp)->di_nextents = (n)) : ((dcp)->di_anextents = (n))) -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DFORK_NEXTENTS) -int xfs_dfork_nextents(xfs_dinode_t *dip, int w); -#define XFS_DFORK_NEXTENTS(dip,w) xfs_dfork_nextents(dip,w) -#else #define XFS_DFORK_NEXTENTS(dip,w) XFS_CFORK_NEXTENTS_DISK(&(dip)->di_core, w) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BUF_TO_DINODE) -xfs_dinode_t *xfs_buf_to_dinode(struct xfs_buf *bp); -#define XFS_BUF_TO_DINODE(bp) xfs_buf_to_dinode(bp) -#else -#define XFS_BUF_TO_DINODE(bp) ((xfs_dinode_t *)(XFS_BUF_PTR(bp))) -#endif +#define XFS_BUF_TO_DINODE(bp) ((xfs_dinode_t *)XFS_BUF_PTR(bp)) /* * Values for di_flags diff --git a/fs/xfs/xfs_dir.c b/fs/xfs/xfs_dir.c index ba30bc7682f2..3dd30391f551 100644 --- a/fs/xfs/xfs_dir.c +++ b/fs/xfs/xfs_dir.c @@ -1,59 +1,43 @@ /* - * Copyright (c) 2000-2001 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2001,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" - -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_dir.h" #include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" -#include "xfs_alloc_btree.h" +#include "xfs_da_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" #include "xfs_alloc.h" #include "xfs_btree.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" #include "xfs_inode.h" #include "xfs_bmap.h" -#include "xfs_da_btree.h" #include "xfs_dir_leaf.h" #include "xfs_error.h" @@ -192,11 +176,23 @@ xfs_dir_mount(xfs_mount_t *mp) uint shortcount, leafcount, count; mp->m_dirversion = 1; - shortcount = (mp->m_attroffset - (uint)sizeof(xfs_dir_sf_hdr_t)) / - (uint)sizeof(xfs_dir_sf_entry_t); - leafcount = (XFS_LBSIZE(mp) - (uint)sizeof(xfs_dir_leaf_hdr_t)) / - ((uint)sizeof(xfs_dir_leaf_entry_t) + - (uint)sizeof(xfs_dir_leaf_name_t)); + if (mp->m_flags & XFS_MOUNT_COMPAT_ATTR) { + shortcount = (mp->m_attroffset - + (uint)sizeof(xfs_dir_sf_hdr_t)) / + (uint)sizeof(xfs_dir_sf_entry_t); + leafcount = (XFS_LBSIZE(mp) - + (uint)sizeof(xfs_dir_leaf_hdr_t)) / + ((uint)sizeof(xfs_dir_leaf_entry_t) + + (uint)sizeof(xfs_dir_leaf_name_t)); + } else { + shortcount = (XFS_BMDR_SPACE_CALC(MINABTPTRS) - + (uint)sizeof(xfs_dir_sf_hdr_t)) / + (uint)sizeof(xfs_dir_sf_entry_t); + leafcount = (XFS_LBSIZE(mp) - + (uint)sizeof(xfs_dir_leaf_hdr_t)) / + ((uint)sizeof(xfs_dir_leaf_entry_t) + + (uint)sizeof(xfs_dir_leaf_name_t)); + } count = shortcount > leafcount ? shortcount : leafcount; mp->m_dircook_elog = xfs_da_log2_roundup(count + 1); ASSERT(mp->m_dircook_elog <= mp->m_sb.sb_blocklog); diff --git a/fs/xfs/xfs_dir.h b/fs/xfs/xfs_dir.h index 4dbc9f54cca5..488defe86ba6 100644 --- a/fs/xfs/xfs_dir.h +++ b/fs/xfs/xfs_dir.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_DIR_H__ #define __XFS_DIR_H__ diff --git a/fs/xfs/xfs_dir2.c b/fs/xfs/xfs_dir2.c index 49fc0a3695ae..022c8398ab62 100644 --- a/fs/xfs/xfs_dir2.c +++ b/fs/xfs/xfs_dir2.c @@ -1,46 +1,26 @@ /* - * Copyright (c) 2000-2001 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2001,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - -/* - * XFS v2 directory implmentation. - * Top-level and utility routines. - */ - #include "xfs.h" - -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" @@ -48,16 +28,16 @@ #include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" -#include "xfs_alloc_btree.h" +#include "xfs_da_btree.h" #include "xfs_bmap_btree.h" -#include "xfs_attr_sf.h" +#include "xfs_alloc_btree.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" -#include "xfs_inode_item.h" #include "xfs_inode.h" +#include "xfs_inode_item.h" #include "xfs_bmap.h" -#include "xfs_da_btree.h" #include "xfs_dir_leaf.h" #include "xfs_dir2_data.h" #include "xfs_dir2_leaf.h" @@ -65,7 +45,6 @@ #include "xfs_dir2_node.h" #include "xfs_dir2_trace.h" #include "xfs_error.h" -#include "xfs_bit.h" /* * Declarations for interface routines. diff --git a/fs/xfs/xfs_dir2.h b/fs/xfs/xfs_dir2.h index 8f4fc7f23bcd..7e24ffeda9e1 100644 --- a/fs/xfs/xfs_dir2.h +++ b/fs/xfs/xfs_dir2.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2001 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2001,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_DIR2_H__ #define __XFS_DIR2_H__ diff --git a/fs/xfs/xfs_dir2_block.c b/fs/xfs/xfs_dir2_block.c index bc4c40fcd479..31bc99faa704 100644 --- a/fs/xfs/xfs_dir2_block.c +++ b/fs/xfs/xfs_dir2_block.c @@ -1,61 +1,39 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - -/* - * xfs_dir2_block.c - * XFS V2 directory implementation, single-block form. - * See xfs_dir2_block.h for the format. - */ - #include "xfs.h" - -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_dir.h" #include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" +#include "xfs_da_btree.h" #include "xfs_bmap_btree.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" -#include "xfs_inode_item.h" #include "xfs_inode.h" -#include "xfs_da_btree.h" +#include "xfs_inode_item.h" #include "xfs_dir_leaf.h" #include "xfs_dir2_data.h" #include "xfs_dir2_leaf.h" @@ -1234,7 +1212,7 @@ xfs_dir2_sf_to_block( /* * Sort the leaf entries by hash value. */ - qsort(blp, INT_GET(btp->count, ARCH_CONVERT), sizeof(*blp), xfs_dir2_block_sort); + xfs_sort(blp, INT_GET(btp->count, ARCH_CONVERT), sizeof(*blp), xfs_dir2_block_sort); /* * Log the leaf entry area and tail. * Already logged the header in data_init, ignore needlog. diff --git a/fs/xfs/xfs_dir2_block.h b/fs/xfs/xfs_dir2_block.h index 5a578b84e246..a2e5cb98a838 100644 --- a/fs/xfs/xfs_dir2_block.h +++ b/fs/xfs/xfs_dir2_block.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2001 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2001,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_DIR2_BLOCK_H__ #define __XFS_DIR2_BLOCK_H__ @@ -74,53 +60,37 @@ typedef struct xfs_dir2_block { /* * Pointer to the leaf header embedded in a data block (1-block format) */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_BLOCK_TAIL_P) -xfs_dir2_block_tail_t * -xfs_dir2_block_tail_p(struct xfs_mount *mp, xfs_dir2_block_t *block); #define XFS_DIR2_BLOCK_TAIL_P(mp,block) xfs_dir2_block_tail_p(mp,block) -#else -#define XFS_DIR2_BLOCK_TAIL_P(mp,block) \ - (((xfs_dir2_block_tail_t *)((char *)(block) + (mp)->m_dirblksize)) - 1) -#endif +static inline xfs_dir2_block_tail_t * +xfs_dir2_block_tail_p(struct xfs_mount *mp, xfs_dir2_block_t *block) +{ + return (((xfs_dir2_block_tail_t *) + ((char *)(block) + (mp)->m_dirblksize)) - 1); +} /* * Pointer to the leaf entries embedded in a data block (1-block format) */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_BLOCK_LEAF_P) -struct xfs_dir2_leaf_entry *xfs_dir2_block_leaf_p(xfs_dir2_block_tail_t *btp); -#define XFS_DIR2_BLOCK_LEAF_P(btp) \ - xfs_dir2_block_leaf_p(btp) -#else -#define XFS_DIR2_BLOCK_LEAF_P(btp) \ - (((struct xfs_dir2_leaf_entry *)(btp)) - INT_GET((btp)->count, ARCH_CONVERT)) -#endif +#define XFS_DIR2_BLOCK_LEAF_P(btp) xfs_dir2_block_leaf_p(btp) +static inline struct xfs_dir2_leaf_entry * +xfs_dir2_block_leaf_p(xfs_dir2_block_tail_t *btp) +{ + return (((struct xfs_dir2_leaf_entry *) + (btp)) - INT_GET((btp)->count, ARCH_CONVERT)); +} /* * Function declarations. */ - -extern int - xfs_dir2_block_addname(struct xfs_da_args *args); - -extern int - xfs_dir2_block_getdents(struct xfs_trans *tp, struct xfs_inode *dp, - struct uio *uio, int *eofp, struct xfs_dirent *dbp, - xfs_dir2_put_t put); - -extern int - xfs_dir2_block_lookup(struct xfs_da_args *args); - -extern int - xfs_dir2_block_removename(struct xfs_da_args *args); - -extern int - xfs_dir2_block_replace(struct xfs_da_args *args); - -extern int - xfs_dir2_leaf_to_block(struct xfs_da_args *args, struct xfs_dabuf *lbp, - struct xfs_dabuf *dbp); - -extern int - xfs_dir2_sf_to_block(struct xfs_da_args *args); +extern int xfs_dir2_block_addname(struct xfs_da_args *args); +extern int xfs_dir2_block_getdents(struct xfs_trans *tp, struct xfs_inode *dp, + struct uio *uio, int *eofp, + struct xfs_dirent *dbp, xfs_dir2_put_t put); +extern int xfs_dir2_block_lookup(struct xfs_da_args *args); +extern int xfs_dir2_block_removename(struct xfs_da_args *args); +extern int xfs_dir2_block_replace(struct xfs_da_args *args); +extern int xfs_dir2_leaf_to_block(struct xfs_da_args *args, + struct xfs_dabuf *lbp, struct xfs_dabuf *dbp); +extern int xfs_dir2_sf_to_block(struct xfs_da_args *args); #endif /* __XFS_DIR2_BLOCK_H__ */ diff --git a/fs/xfs/xfs_dir2_data.c b/fs/xfs/xfs_dir2_data.c index a0aa0e44ff9d..5b7c47e2f14a 100644 --- a/fs/xfs/xfs_dir2_data.c +++ b/fs/xfs/xfs_dir2_data.c @@ -1,60 +1,38 @@ /* - * Copyright (c) 2000-2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ - */ - -/* - * xfs_dir2_data.c - * Core data block handling routines for XFS V2 directories. - * See xfs_dir2_data.h for data structures. + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" - -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_dir.h" #include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" +#include "xfs_da_btree.h" #include "xfs_bmap_btree.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" #include "xfs_inode.h" -#include "xfs_da_btree.h" #include "xfs_dir_leaf.h" #include "xfs_dir2_data.h" #include "xfs_dir2_leaf.h" diff --git a/fs/xfs/xfs_dir2_data.h b/fs/xfs/xfs_dir2_data.h index 476cac920bf5..5e3a7f9ec735 100644 --- a/fs/xfs/xfs_dir2_data.h +++ b/fs/xfs/xfs_dir2_data.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_DIR2_DATA_H__ #define __XFS_DIR2_DATA_H__ @@ -137,88 +123,65 @@ typedef struct xfs_dir2_data { /* * Size of a data entry. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_DATA_ENTSIZE) -int xfs_dir2_data_entsize(int n); #define XFS_DIR2_DATA_ENTSIZE(n) xfs_dir2_data_entsize(n) -#else -#define XFS_DIR2_DATA_ENTSIZE(n) \ - ((int)(roundup(offsetof(xfs_dir2_data_entry_t, name[0]) + (n) + \ - (uint)sizeof(xfs_dir2_data_off_t), XFS_DIR2_DATA_ALIGN))) -#endif +static inline int xfs_dir2_data_entsize(int n) +{ + return (int)roundup(offsetof(xfs_dir2_data_entry_t, name[0]) + (n) + \ + (uint)sizeof(xfs_dir2_data_off_t), XFS_DIR2_DATA_ALIGN); +} /* * Pointer to an entry's tag word. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_DATA_ENTRY_TAG_P) -xfs_dir2_data_off_t *xfs_dir2_data_entry_tag_p(xfs_dir2_data_entry_t *dep); #define XFS_DIR2_DATA_ENTRY_TAG_P(dep) xfs_dir2_data_entry_tag_p(dep) -#else -#define XFS_DIR2_DATA_ENTRY_TAG_P(dep) \ - ((xfs_dir2_data_off_t *)\ - ((char *)(dep) + XFS_DIR2_DATA_ENTSIZE((dep)->namelen) - \ - (uint)sizeof(xfs_dir2_data_off_t))) -#endif +static inline xfs_dir2_data_off_t * +xfs_dir2_data_entry_tag_p(xfs_dir2_data_entry_t *dep) +{ + return (xfs_dir2_data_off_t *) \ + ((char *)(dep) + XFS_DIR2_DATA_ENTSIZE((dep)->namelen) - \ + (uint)sizeof(xfs_dir2_data_off_t)); +} /* * Pointer to a freespace's tag word. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_DATA_UNUSED_TAG_P) -xfs_dir2_data_off_t *xfs_dir2_data_unused_tag_p(xfs_dir2_data_unused_t *dup); #define XFS_DIR2_DATA_UNUSED_TAG_P(dup) \ xfs_dir2_data_unused_tag_p(dup) -#else -#define XFS_DIR2_DATA_UNUSED_TAG_P(dup) \ - ((xfs_dir2_data_off_t *)\ - ((char *)(dup) + INT_GET((dup)->length, ARCH_CONVERT) \ - - (uint)sizeof(xfs_dir2_data_off_t))) -#endif +static inline xfs_dir2_data_off_t * +xfs_dir2_data_unused_tag_p(xfs_dir2_data_unused_t *dup) +{ + return (xfs_dir2_data_off_t *) \ + ((char *)(dup) + INT_GET((dup)->length, ARCH_CONVERT) \ + - (uint)sizeof(xfs_dir2_data_off_t)); +} /* * Function declarations. */ - #ifdef DEBUG -extern void - xfs_dir2_data_check(struct xfs_inode *dp, struct xfs_dabuf *bp); +extern void xfs_dir2_data_check(struct xfs_inode *dp, struct xfs_dabuf *bp); #else #define xfs_dir2_data_check(dp,bp) #endif - -extern xfs_dir2_data_free_t * - xfs_dir2_data_freefind(xfs_dir2_data_t *d, - xfs_dir2_data_unused_t *dup); - -extern xfs_dir2_data_free_t * - xfs_dir2_data_freeinsert(xfs_dir2_data_t *d, - xfs_dir2_data_unused_t *dup, int *loghead); - -extern void - xfs_dir2_data_freescan(struct xfs_mount *mp, xfs_dir2_data_t *d, - int *loghead, char *aendp); - -extern int - xfs_dir2_data_init(struct xfs_da_args *args, xfs_dir2_db_t blkno, - struct xfs_dabuf **bpp); - -extern void - xfs_dir2_data_log_entry(struct xfs_trans *tp, struct xfs_dabuf *bp, +extern xfs_dir2_data_free_t *xfs_dir2_data_freefind(xfs_dir2_data_t *d, + xfs_dir2_data_unused_t *dup); +extern xfs_dir2_data_free_t *xfs_dir2_data_freeinsert(xfs_dir2_data_t *d, + xfs_dir2_data_unused_t *dup, int *loghead); +extern void xfs_dir2_data_freescan(struct xfs_mount *mp, xfs_dir2_data_t *d, + int *loghead, char *aendp); +extern int xfs_dir2_data_init(struct xfs_da_args *args, xfs_dir2_db_t blkno, + struct xfs_dabuf **bpp); +extern void xfs_dir2_data_log_entry(struct xfs_trans *tp, struct xfs_dabuf *bp, xfs_dir2_data_entry_t *dep); - -extern void - xfs_dir2_data_log_header(struct xfs_trans *tp, struct xfs_dabuf *bp); - -extern void - xfs_dir2_data_log_unused(struct xfs_trans *tp, struct xfs_dabuf *bp, - xfs_dir2_data_unused_t *dup); - -extern void - xfs_dir2_data_make_free(struct xfs_trans *tp, struct xfs_dabuf *bp, +extern void xfs_dir2_data_log_header(struct xfs_trans *tp, + struct xfs_dabuf *bp); +extern void xfs_dir2_data_log_unused(struct xfs_trans *tp, struct xfs_dabuf *bp, + xfs_dir2_data_unused_t *dup); +extern void xfs_dir2_data_make_free(struct xfs_trans *tp, struct xfs_dabuf *bp, xfs_dir2_data_aoff_t offset, xfs_dir2_data_aoff_t len, int *needlogp, int *needscanp); - -extern void - xfs_dir2_data_use_free(struct xfs_trans *tp, struct xfs_dabuf *bp, +extern void xfs_dir2_data_use_free(struct xfs_trans *tp, struct xfs_dabuf *bp, xfs_dir2_data_unused_t *dup, xfs_dir2_data_aoff_t offset, xfs_dir2_data_aoff_t len, int *needlogp, diff --git a/fs/xfs/xfs_dir2_leaf.c b/fs/xfs/xfs_dir2_leaf.c index 056f5283904b..d342b6b55239 100644 --- a/fs/xfs/xfs_dir2_leaf.c +++ b/fs/xfs/xfs_dir2_leaf.c @@ -1,49 +1,26 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ - */ - -/* - * xfs_dir2_leaf.c - * XFS directory version 2 implementation - single leaf form - * see xfs_dir2_leaf.h for data structures. - * These directories have multiple XFS_DIR2_DATA blocks and one - * XFS_DIR2_LEAF1 block containing the hash table and freespace map. + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" - -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" @@ -51,6 +28,7 @@ #include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" +#include "xfs_da_btree.h" #include "xfs_bmap_btree.h" #include "xfs_attr_sf.h" #include "xfs_dir_sf.h" @@ -58,14 +36,12 @@ #include "xfs_dinode.h" #include "xfs_inode.h" #include "xfs_bmap.h" -#include "xfs_da_btree.h" #include "xfs_dir2_data.h" #include "xfs_dir2_leaf.h" #include "xfs_dir2_block.h" #include "xfs_dir2_node.h" #include "xfs_dir2_trace.h" #include "xfs_error.h" -#include "xfs_bit.h" /* * Local function declarations. diff --git a/fs/xfs/xfs_dir2_leaf.h b/fs/xfs/xfs_dir2_leaf.h index 3303cd6f4c00..1393993d61e9 100644 --- a/fs/xfs/xfs_dir2_leaf.h +++ b/fs/xfs/xfs_dir2_leaf.h @@ -1,41 +1,23 @@ /* - * Copyright (c) 2000-2001 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2001,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_DIR2_LEAF_H__ #define __XFS_DIR2_LEAF_H__ -/* - * Directory version 2, leaf block structures. - */ - struct uio; struct xfs_dabuf; struct xfs_da_args; @@ -44,10 +26,6 @@ struct xfs_mount; struct xfs_trans; /* - * Constants. - */ - -/* * Offset of the leaf/node space. First block in this space * is the btree root. */ @@ -57,10 +35,6 @@ struct xfs_trans; XFS_DIR2_BYTE_TO_DB(mp, XFS_DIR2_LEAF_OFFSET) /* - * Types. - */ - -/* * Offset in data space of a data entry. */ typedef __uint32_t xfs_dir2_dataptr_t; @@ -68,10 +42,6 @@ typedef __uint32_t xfs_dir2_dataptr_t; #define XFS_DIR2_NULL_DATAPTR ((xfs_dir2_dataptr_t)0) /* - * Structures. - */ - -/* * Leaf block header. */ typedef struct xfs_dir2_leaf_hdr { @@ -109,245 +79,193 @@ typedef struct xfs_dir2_leaf { } xfs_dir2_leaf_t; /* - * Macros. - * The DB blocks are logical directory block numbers, not filesystem blocks. + * DB blocks here are logical directory block numbers, not filesystem blocks. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_MAX_LEAF_ENTS) -int -xfs_dir2_max_leaf_ents(struct xfs_mount *mp); -#define XFS_DIR2_MAX_LEAF_ENTS(mp) \ - xfs_dir2_max_leaf_ents(mp) -#else -#define XFS_DIR2_MAX_LEAF_ENTS(mp) \ - ((int)(((mp)->m_dirblksize - (uint)sizeof(xfs_dir2_leaf_hdr_t)) / \ - (uint)sizeof(xfs_dir2_leaf_entry_t))) -#endif +#define XFS_DIR2_MAX_LEAF_ENTS(mp) xfs_dir2_max_leaf_ents(mp) +static inline int xfs_dir2_max_leaf_ents(struct xfs_mount *mp) +{ + return (int)(((mp)->m_dirblksize - (uint)sizeof(xfs_dir2_leaf_hdr_t)) / + (uint)sizeof(xfs_dir2_leaf_entry_t)); +} /* * Get address of the bestcount field in the single-leaf block. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_LEAF_TAIL_P) -xfs_dir2_leaf_tail_t * -xfs_dir2_leaf_tail_p(struct xfs_mount *mp, xfs_dir2_leaf_t *lp); -#define XFS_DIR2_LEAF_TAIL_P(mp,lp) \ - xfs_dir2_leaf_tail_p(mp, lp) -#else -#define XFS_DIR2_LEAF_TAIL_P(mp,lp) \ - ((xfs_dir2_leaf_tail_t *)\ - ((char *)(lp) + (mp)->m_dirblksize - \ - (uint)sizeof(xfs_dir2_leaf_tail_t))) -#endif +#define XFS_DIR2_LEAF_TAIL_P(mp,lp) xfs_dir2_leaf_tail_p(mp, lp) +static inline xfs_dir2_leaf_tail_t * +xfs_dir2_leaf_tail_p(struct xfs_mount *mp, xfs_dir2_leaf_t *lp) +{ + return (xfs_dir2_leaf_tail_t *) + ((char *)(lp) + (mp)->m_dirblksize - + (uint)sizeof(xfs_dir2_leaf_tail_t)); +} /* * Get address of the bests array in the single-leaf block. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_LEAF_BESTS_P) -xfs_dir2_data_off_t * -xfs_dir2_leaf_bests_p(xfs_dir2_leaf_tail_t *ltp); #define XFS_DIR2_LEAF_BESTS_P(ltp) xfs_dir2_leaf_bests_p(ltp) -#else -#define XFS_DIR2_LEAF_BESTS_P(ltp) \ - ((xfs_dir2_data_off_t *)(ltp) - INT_GET((ltp)->bestcount, ARCH_CONVERT)) -#endif +static inline xfs_dir2_data_off_t * +xfs_dir2_leaf_bests_p(xfs_dir2_leaf_tail_t *ltp) +{ + return (xfs_dir2_data_off_t *) + (ltp) - INT_GET((ltp)->bestcount, ARCH_CONVERT); +} /* * Convert dataptr to byte in file space */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_DATAPTR_TO_BYTE) -xfs_dir2_off_t -xfs_dir2_dataptr_to_byte(struct xfs_mount *mp, xfs_dir2_dataptr_t dp); #define XFS_DIR2_DATAPTR_TO_BYTE(mp,dp) xfs_dir2_dataptr_to_byte(mp, dp) -#else -#define XFS_DIR2_DATAPTR_TO_BYTE(mp,dp) \ - ((xfs_dir2_off_t)(dp) << XFS_DIR2_DATA_ALIGN_LOG) -#endif +static inline xfs_dir2_off_t +xfs_dir2_dataptr_to_byte(struct xfs_mount *mp, xfs_dir2_dataptr_t dp) +{ + return (xfs_dir2_off_t)(dp) << XFS_DIR2_DATA_ALIGN_LOG; +} /* * Convert byte in file space to dataptr. It had better be aligned. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_BYTE_TO_DATAPTR) -xfs_dir2_dataptr_t -xfs_dir2_byte_to_dataptr(struct xfs_mount *mp, xfs_dir2_off_t by); #define XFS_DIR2_BYTE_TO_DATAPTR(mp,by) xfs_dir2_byte_to_dataptr(mp,by) -#else -#define XFS_DIR2_BYTE_TO_DATAPTR(mp,by) \ - ((xfs_dir2_dataptr_t)((by) >> XFS_DIR2_DATA_ALIGN_LOG)) -#endif +static inline xfs_dir2_dataptr_t +xfs_dir2_byte_to_dataptr(struct xfs_mount *mp, xfs_dir2_off_t by) +{ + return (xfs_dir2_dataptr_t)((by) >> XFS_DIR2_DATA_ALIGN_LOG); +} + +/* + * Convert byte in space to (DB) block + */ +#define XFS_DIR2_BYTE_TO_DB(mp,by) xfs_dir2_byte_to_db(mp, by) +static inline xfs_dir2_db_t +xfs_dir2_byte_to_db(struct xfs_mount *mp, xfs_dir2_off_t by) +{ + return (xfs_dir2_db_t)((by) >> \ + ((mp)->m_sb.sb_blocklog + (mp)->m_sb.sb_dirblklog)); +} /* * Convert dataptr to a block number */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_DATAPTR_TO_DB) -xfs_dir2_db_t -xfs_dir2_dataptr_to_db(struct xfs_mount *mp, xfs_dir2_dataptr_t dp); #define XFS_DIR2_DATAPTR_TO_DB(mp,dp) xfs_dir2_dataptr_to_db(mp, dp) -#else -#define XFS_DIR2_DATAPTR_TO_DB(mp,dp) \ - XFS_DIR2_BYTE_TO_DB(mp, XFS_DIR2_DATAPTR_TO_BYTE(mp, dp)) -#endif +static inline xfs_dir2_db_t +xfs_dir2_dataptr_to_db(struct xfs_mount *mp, xfs_dir2_dataptr_t dp) +{ + return XFS_DIR2_BYTE_TO_DB(mp, XFS_DIR2_DATAPTR_TO_BYTE(mp, dp)); +} + +/* + * Convert byte in space to offset in a block + */ +#define XFS_DIR2_BYTE_TO_OFF(mp,by) xfs_dir2_byte_to_off(mp, by) +static inline xfs_dir2_data_aoff_t +xfs_dir2_byte_to_off(struct xfs_mount *mp, xfs_dir2_off_t by) +{ + return (xfs_dir2_data_aoff_t)((by) & \ + ((1 << ((mp)->m_sb.sb_blocklog + (mp)->m_sb.sb_dirblklog)) - 1)); +} /* * Convert dataptr to a byte offset in a block */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_DATAPTR_TO_OFF) -xfs_dir2_data_aoff_t -xfs_dir2_dataptr_to_off(struct xfs_mount *mp, xfs_dir2_dataptr_t dp); #define XFS_DIR2_DATAPTR_TO_OFF(mp,dp) xfs_dir2_dataptr_to_off(mp, dp) -#else -#define XFS_DIR2_DATAPTR_TO_OFF(mp,dp) \ - XFS_DIR2_BYTE_TO_OFF(mp, XFS_DIR2_DATAPTR_TO_BYTE(mp, dp)) -#endif +static inline xfs_dir2_data_aoff_t +xfs_dir2_dataptr_to_off(struct xfs_mount *mp, xfs_dir2_dataptr_t dp) +{ + return XFS_DIR2_BYTE_TO_OFF(mp, XFS_DIR2_DATAPTR_TO_BYTE(mp, dp)); +} /* * Convert block and offset to byte in space */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_DB_OFF_TO_BYTE) -xfs_dir2_off_t -xfs_dir2_db_off_to_byte(struct xfs_mount *mp, xfs_dir2_db_t db, - xfs_dir2_data_aoff_t o); #define XFS_DIR2_DB_OFF_TO_BYTE(mp,db,o) \ xfs_dir2_db_off_to_byte(mp, db, o) -#else -#define XFS_DIR2_DB_OFF_TO_BYTE(mp,db,o) \ - (((xfs_dir2_off_t)(db) << \ - ((mp)->m_sb.sb_blocklog + (mp)->m_sb.sb_dirblklog)) + (o)) -#endif +static inline xfs_dir2_off_t +xfs_dir2_db_off_to_byte(struct xfs_mount *mp, xfs_dir2_db_t db, + xfs_dir2_data_aoff_t o) +{ + return ((xfs_dir2_off_t)(db) << \ + ((mp)->m_sb.sb_blocklog + (mp)->m_sb.sb_dirblklog)) + (o); +} /* - * Convert byte in space to (DB) block + * Convert block (DB) to block (dablk) */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_BYTE_TO_DB) -xfs_dir2_db_t xfs_dir2_byte_to_db(struct xfs_mount *mp, xfs_dir2_off_t by); -#define XFS_DIR2_BYTE_TO_DB(mp,by) xfs_dir2_byte_to_db(mp, by) -#else -#define XFS_DIR2_BYTE_TO_DB(mp,by) \ - ((xfs_dir2_db_t)((by) >> \ - ((mp)->m_sb.sb_blocklog + (mp)->m_sb.sb_dirblklog))) -#endif +#define XFS_DIR2_DB_TO_DA(mp,db) xfs_dir2_db_to_da(mp, db) +static inline xfs_dablk_t +xfs_dir2_db_to_da(struct xfs_mount *mp, xfs_dir2_db_t db) +{ + return (xfs_dablk_t)((db) << (mp)->m_sb.sb_dirblklog); +} /* * Convert byte in space to (DA) block */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_BYTE_TO_DA) -xfs_dablk_t xfs_dir2_byte_to_da(struct xfs_mount *mp, xfs_dir2_off_t by); #define XFS_DIR2_BYTE_TO_DA(mp,by) xfs_dir2_byte_to_da(mp, by) -#else -#define XFS_DIR2_BYTE_TO_DA(mp,by) \ - XFS_DIR2_DB_TO_DA(mp, XFS_DIR2_BYTE_TO_DB(mp, by)) -#endif - -/* - * Convert byte in space to offset in a block - */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_BYTE_TO_OFF) -xfs_dir2_data_aoff_t -xfs_dir2_byte_to_off(struct xfs_mount *mp, xfs_dir2_off_t by); -#define XFS_DIR2_BYTE_TO_OFF(mp,by) xfs_dir2_byte_to_off(mp, by) -#else -#define XFS_DIR2_BYTE_TO_OFF(mp,by) \ - ((xfs_dir2_data_aoff_t)((by) & \ - ((1 << ((mp)->m_sb.sb_blocklog + \ - (mp)->m_sb.sb_dirblklog)) - 1))) -#endif +static inline xfs_dablk_t +xfs_dir2_byte_to_da(struct xfs_mount *mp, xfs_dir2_off_t by) +{ + return XFS_DIR2_DB_TO_DA(mp, XFS_DIR2_BYTE_TO_DB(mp, by)); +} /* * Convert block and offset to dataptr */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_DB_OFF_TO_DATAPTR) -xfs_dir2_dataptr_t -xfs_dir2_db_off_to_dataptr(struct xfs_mount *mp, xfs_dir2_db_t db, - xfs_dir2_data_aoff_t o); #define XFS_DIR2_DB_OFF_TO_DATAPTR(mp,db,o) \ xfs_dir2_db_off_to_dataptr(mp, db, o) -#else -#define XFS_DIR2_DB_OFF_TO_DATAPTR(mp,db,o) \ - XFS_DIR2_BYTE_TO_DATAPTR(mp, XFS_DIR2_DB_OFF_TO_BYTE(mp, db, o)) -#endif - -/* - * Convert block (DB) to block (dablk) - */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_DB_TO_DA) -xfs_dablk_t xfs_dir2_db_to_da(struct xfs_mount *mp, xfs_dir2_db_t db); -#define XFS_DIR2_DB_TO_DA(mp,db) xfs_dir2_db_to_da(mp, db) -#else -#define XFS_DIR2_DB_TO_DA(mp,db) \ - ((xfs_dablk_t)((db) << (mp)->m_sb.sb_dirblklog)) -#endif +static inline xfs_dir2_dataptr_t +xfs_dir2_db_off_to_dataptr(struct xfs_mount *mp, xfs_dir2_db_t db, + xfs_dir2_data_aoff_t o) +{ + return XFS_DIR2_BYTE_TO_DATAPTR(mp, XFS_DIR2_DB_OFF_TO_BYTE(mp, db, o)); +} /* * Convert block (dablk) to block (DB) */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_DA_TO_DB) -xfs_dir2_db_t xfs_dir2_da_to_db(struct xfs_mount *mp, xfs_dablk_t da); #define XFS_DIR2_DA_TO_DB(mp,da) xfs_dir2_da_to_db(mp, da) -#else -#define XFS_DIR2_DA_TO_DB(mp,da) \ - ((xfs_dir2_db_t)((da) >> (mp)->m_sb.sb_dirblklog)) -#endif +static inline xfs_dir2_db_t +xfs_dir2_da_to_db(struct xfs_mount *mp, xfs_dablk_t da) +{ + return (xfs_dir2_db_t)((da) >> (mp)->m_sb.sb_dirblklog); +} /* * Convert block (dablk) to byte offset in space */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_DA_TO_BYTE) -xfs_dir2_off_t xfs_dir2_da_to_byte(struct xfs_mount *mp, xfs_dablk_t da); #define XFS_DIR2_DA_TO_BYTE(mp,da) xfs_dir2_da_to_byte(mp, da) -#else -#define XFS_DIR2_DA_TO_BYTE(mp,da) \ - XFS_DIR2_DB_OFF_TO_BYTE(mp, XFS_DIR2_DA_TO_DB(mp, da), 0) -#endif +static inline xfs_dir2_off_t +xfs_dir2_da_to_byte(struct xfs_mount *mp, xfs_dablk_t da) +{ + return XFS_DIR2_DB_OFF_TO_BYTE(mp, XFS_DIR2_DA_TO_DB(mp, da), 0); +} /* * Function declarations. */ - -extern int - xfs_dir2_block_to_leaf(struct xfs_da_args *args, struct xfs_dabuf *dbp); - -extern int - xfs_dir2_leaf_addname(struct xfs_da_args *args); - -extern void - xfs_dir2_leaf_compact(struct xfs_da_args *args, struct xfs_dabuf *bp); - -extern void - xfs_dir2_leaf_compact_x1(struct xfs_dabuf *bp, int *indexp, - int *lowstalep, int *highstalep, int *lowlogp, - int *highlogp); - -extern int - xfs_dir2_leaf_getdents(struct xfs_trans *tp, struct xfs_inode *dp, - struct uio *uio, int *eofp, struct xfs_dirent *dbp, - xfs_dir2_put_t put); - -extern int - xfs_dir2_leaf_init(struct xfs_da_args *args, xfs_dir2_db_t bno, - struct xfs_dabuf **bpp, int magic); - -extern void - xfs_dir2_leaf_log_ents(struct xfs_trans *tp, struct xfs_dabuf *bp, - int first, int last); - -extern void - xfs_dir2_leaf_log_header(struct xfs_trans *tp, struct xfs_dabuf *bp); - -extern int - xfs_dir2_leaf_lookup(struct xfs_da_args *args); - -extern int - xfs_dir2_leaf_removename(struct xfs_da_args *args); - -extern int - xfs_dir2_leaf_replace(struct xfs_da_args *args); - -extern int - xfs_dir2_leaf_search_hash(struct xfs_da_args *args, - struct xfs_dabuf *lbp); -extern int - xfs_dir2_leaf_trim_data(struct xfs_da_args *args, struct xfs_dabuf *lbp, xfs_dir2_db_t db); - -extern int - xfs_dir2_node_to_leaf(struct xfs_da_state *state); +extern int xfs_dir2_block_to_leaf(struct xfs_da_args *args, + struct xfs_dabuf *dbp); +extern int xfs_dir2_leaf_addname(struct xfs_da_args *args); +extern void xfs_dir2_leaf_compact(struct xfs_da_args *args, + struct xfs_dabuf *bp); +extern void xfs_dir2_leaf_compact_x1(struct xfs_dabuf *bp, int *indexp, + int *lowstalep, int *highstalep, + int *lowlogp, int *highlogp); +extern int xfs_dir2_leaf_getdents(struct xfs_trans *tp, struct xfs_inode *dp, + struct uio *uio, int *eofp, + struct xfs_dirent *dbp, xfs_dir2_put_t put); +extern int xfs_dir2_leaf_init(struct xfs_da_args *args, xfs_dir2_db_t bno, + struct xfs_dabuf **bpp, int magic); +extern void xfs_dir2_leaf_log_ents(struct xfs_trans *tp, struct xfs_dabuf *bp, + int first, int last); +extern void xfs_dir2_leaf_log_header(struct xfs_trans *tp, + struct xfs_dabuf *bp); +extern int xfs_dir2_leaf_lookup(struct xfs_da_args *args); +extern int xfs_dir2_leaf_removename(struct xfs_da_args *args); +extern int xfs_dir2_leaf_replace(struct xfs_da_args *args); +extern int xfs_dir2_leaf_search_hash(struct xfs_da_args *args, + struct xfs_dabuf *lbp); +extern int xfs_dir2_leaf_trim_data(struct xfs_da_args *args, + struct xfs_dabuf *lbp, xfs_dir2_db_t db); +extern int xfs_dir2_node_to_leaf(struct xfs_da_state *state); #endif /* __XFS_DIR2_LEAF_H__ */ diff --git a/fs/xfs/xfs_dir2_node.c b/fs/xfs/xfs_dir2_node.c index a7615d86bfb7..641f8633d254 100644 --- a/fs/xfs/xfs_dir2_node.c +++ b/fs/xfs/xfs_dir2_node.c @@ -1,61 +1,39 @@ /* - * Copyright (c) 2000-2004 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ - */ - -/* - * xfs_dir2_node.c - * XFS directory implementation, version 2, node form files - * See data structures in xfs_dir2_node.h and xfs_da_btree.h. + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" - -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_dir.h" #include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" +#include "xfs_da_btree.h" #include "xfs_bmap_btree.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" #include "xfs_inode.h" #include "xfs_bmap.h" -#include "xfs_da_btree.h" #include "xfs_dir2_data.h" #include "xfs_dir2_leaf.h" #include "xfs_dir2_block.h" diff --git a/fs/xfs/xfs_dir2_node.h b/fs/xfs/xfs_dir2_node.h index 96db420c7c5c..0ab8fbd59512 100644 --- a/fs/xfs/xfs_dir2_node.h +++ b/fs/xfs/xfs_dir2_node.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_DIR2_NODE_H__ #define __XFS_DIR2_NODE_H__ @@ -45,10 +31,6 @@ struct xfs_inode; struct xfs_trans; /* - * Constants. - */ - -/* * Offset of the freespace index. */ #define XFS_DIR2_FREE_SPACE 2 @@ -58,9 +40,6 @@ struct xfs_trans; #define XFS_DIR2_FREE_MAGIC 0x58443246 /* XD2F */ -/* - * Structures. - */ typedef struct xfs_dir2_free_hdr { __uint32_t magic; /* XFS_DIR2_FREE_MAGIC */ __int32_t firstdb; /* db of first entry */ @@ -73,87 +52,53 @@ typedef struct xfs_dir2_free { xfs_dir2_data_off_t bests[1]; /* best free counts */ /* unused entries are -1 */ } xfs_dir2_free_t; + #define XFS_DIR2_MAX_FREE_BESTS(mp) \ (((mp)->m_dirblksize - (uint)sizeof(xfs_dir2_free_hdr_t)) / \ (uint)sizeof(xfs_dir2_data_off_t)) /* - * Macros. - */ - -/* * Convert data space db to the corresponding free db. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_DB_TO_FDB) -xfs_dir2_db_t -xfs_dir2_db_to_fdb(struct xfs_mount *mp, xfs_dir2_db_t db); #define XFS_DIR2_DB_TO_FDB(mp,db) xfs_dir2_db_to_fdb(mp, db) -#else -#define XFS_DIR2_DB_TO_FDB(mp,db) \ - (XFS_DIR2_FREE_FIRSTDB(mp) + (db) / XFS_DIR2_MAX_FREE_BESTS(mp)) -#endif +static inline xfs_dir2_db_t +xfs_dir2_db_to_fdb(struct xfs_mount *mp, xfs_dir2_db_t db) +{ + return (XFS_DIR2_FREE_FIRSTDB(mp) + (db) / XFS_DIR2_MAX_FREE_BESTS(mp)); +} /* * Convert data space db to the corresponding index in a free db. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_DB_TO_FDINDEX) -int -xfs_dir2_db_to_fdindex(struct xfs_mount *mp, xfs_dir2_db_t db); #define XFS_DIR2_DB_TO_FDINDEX(mp,db) xfs_dir2_db_to_fdindex(mp, db) -#else -#define XFS_DIR2_DB_TO_FDINDEX(mp,db) ((db) % XFS_DIR2_MAX_FREE_BESTS(mp)) -#endif - -/* - * Functions. - */ - -extern void - xfs_dir2_free_log_bests(struct xfs_trans *tp, struct xfs_dabuf *bp, - int first, int last); - -extern int - xfs_dir2_leaf_to_node(struct xfs_da_args *args, struct xfs_dabuf *lbp); - -extern xfs_dahash_t - xfs_dir2_leafn_lasthash(struct xfs_dabuf *bp, int *count); - -extern int - xfs_dir2_leafn_lookup_int(struct xfs_dabuf *bp, - struct xfs_da_args *args, int *indexp, - struct xfs_da_state *state); - -extern int - xfs_dir2_leafn_order(struct xfs_dabuf *leaf1_bp, - struct xfs_dabuf *leaf2_bp); - -extern int - xfs_dir2_leafn_split(struct xfs_da_state *state, - struct xfs_da_state_blk *oldblk, - struct xfs_da_state_blk *newblk); - -extern int - xfs_dir2_leafn_toosmall(struct xfs_da_state *state, int *action); - -extern void - xfs_dir2_leafn_unbalance(struct xfs_da_state *state, - struct xfs_da_state_blk *drop_blk, - struct xfs_da_state_blk *save_blk); - -extern int - xfs_dir2_node_addname(struct xfs_da_args *args); - -extern int - xfs_dir2_node_lookup(struct xfs_da_args *args); - -extern int - xfs_dir2_node_removename(struct xfs_da_args *args); - -extern int - xfs_dir2_node_replace(struct xfs_da_args *args); - -extern int - xfs_dir2_node_trim_free(struct xfs_da_args *args, xfs_fileoff_t fo, - int *rvalp); +static inline int +xfs_dir2_db_to_fdindex(struct xfs_mount *mp, xfs_dir2_db_t db) +{ + return ((db) % XFS_DIR2_MAX_FREE_BESTS(mp)); +} + +extern void xfs_dir2_free_log_bests(struct xfs_trans *tp, struct xfs_dabuf *bp, + int first, int last); +extern int xfs_dir2_leaf_to_node(struct xfs_da_args *args, + struct xfs_dabuf *lbp); +extern xfs_dahash_t xfs_dir2_leafn_lasthash(struct xfs_dabuf *bp, int *count); +extern int xfs_dir2_leafn_lookup_int(struct xfs_dabuf *bp, + struct xfs_da_args *args, int *indexp, + struct xfs_da_state *state); +extern int xfs_dir2_leafn_order(struct xfs_dabuf *leaf1_bp, + struct xfs_dabuf *leaf2_bp); +extern int xfs_dir2_leafn_split(struct xfs_da_state *state, + struct xfs_da_state_blk *oldblk, + struct xfs_da_state_blk *newblk); +extern int xfs_dir2_leafn_toosmall(struct xfs_da_state *state, int *action); +extern void xfs_dir2_leafn_unbalance(struct xfs_da_state *state, + struct xfs_da_state_blk *drop_blk, + struct xfs_da_state_blk *save_blk); +extern int xfs_dir2_node_addname(struct xfs_da_args *args); +extern int xfs_dir2_node_lookup(struct xfs_da_args *args); +extern int xfs_dir2_node_removename(struct xfs_da_args *args); +extern int xfs_dir2_node_replace(struct xfs_da_args *args); +extern int xfs_dir2_node_trim_free(struct xfs_da_args *args, xfs_fileoff_t fo, + int *rvalp); #endif /* __XFS_DIR2_NODE_H__ */ diff --git a/fs/xfs/xfs_dir2_sf.c b/fs/xfs/xfs_dir2_sf.c index 6bbc61674411..ec8e7476c8b7 100644 --- a/fs/xfs/xfs_dir2_sf.c +++ b/fs/xfs/xfs_dir2_sf.c @@ -1,60 +1,39 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - -/* - * xfs_dir2_sf.c - * Shortform directory implementation for v2 directories. - */ - #include "xfs.h" - -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_dir.h" #include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" +#include "xfs_da_btree.h" #include "xfs_bmap_btree.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" -#include "xfs_inode_item.h" #include "xfs_inode.h" -#include "xfs_da_btree.h" +#include "xfs_inode_item.h" #include "xfs_dir_leaf.h" #include "xfs_error.h" #include "xfs_dir2_data.h" @@ -107,7 +86,7 @@ xfs_dir2_block_sfsize( int isdotdot; /* entry is ".." */ xfs_mount_t *mp; /* mount structure pointer */ int namelen; /* total name bytes */ - xfs_ino_t parent; /* parent inode number */ + xfs_ino_t parent = 0; /* parent inode number */ int size=0; /* total computed size */ mp = dp->i_mount; @@ -298,11 +277,11 @@ xfs_dir2_sf_addname( int incr_isize; /* total change in size */ int new_isize; /* di_size after adding name */ int objchange; /* changing to 8-byte inodes */ - xfs_dir2_data_aoff_t offset; /* offset for new entry */ + xfs_dir2_data_aoff_t offset = 0; /* offset for new entry */ int old_isize; /* di_size before adding name */ int pick; /* which algorithm to use */ xfs_dir2_sf_t *sfp; /* shortform structure */ - xfs_dir2_sf_entry_t *sfep; /* shortform entry */ + xfs_dir2_sf_entry_t *sfep = NULL; /* shortform entry */ xfs_dir2_trace_args("sf_addname", args); ASSERT(xfs_dir2_sf_lookup(args) == ENOENT); diff --git a/fs/xfs/xfs_dir2_sf.h b/fs/xfs/xfs_dir2_sf.h index bac6f5a2a312..42f015b70018 100644 --- a/fs/xfs/xfs_dir2_sf.h +++ b/fs/xfs/xfs_dir2_sf.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2001 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2001,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_DIR2_SF_H__ #define __XFS_DIR2_SF_H__ @@ -104,140 +90,106 @@ typedef struct xfs_dir2_sf { xfs_dir2_sf_entry_t list[1]; /* shortform entries */ } xfs_dir2_sf_t; -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_SF_HDR_SIZE) -int xfs_dir2_sf_hdr_size(int i8count); #define XFS_DIR2_SF_HDR_SIZE(i8count) xfs_dir2_sf_hdr_size(i8count) -#else -#define XFS_DIR2_SF_HDR_SIZE(i8count) \ - ((uint)sizeof(xfs_dir2_sf_hdr_t) - \ - ((i8count) == 0) * \ - ((uint)sizeof(xfs_dir2_ino8_t) - (uint)sizeof(xfs_dir2_ino4_t))) -#endif - -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_SF_INUMBERP) -xfs_dir2_inou_t *xfs_dir2_sf_inumberp(xfs_dir2_sf_entry_t *sfep); +static inline int xfs_dir2_sf_hdr_size(int i8count) +{ + return ((uint)sizeof(xfs_dir2_sf_hdr_t) - \ + ((i8count) == 0) * \ + ((uint)sizeof(xfs_dir2_ino8_t) - (uint)sizeof(xfs_dir2_ino4_t))); +} + #define XFS_DIR2_SF_INUMBERP(sfep) xfs_dir2_sf_inumberp(sfep) -#else -#define XFS_DIR2_SF_INUMBERP(sfep) \ - ((xfs_dir2_inou_t *)&(sfep)->name[(sfep)->namelen]) -#endif - -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_SF_GET_INUMBER) -xfs_intino_t xfs_dir2_sf_get_inumber(xfs_dir2_sf_t *sfp, xfs_dir2_inou_t *from); -#define XFS_DIR2_SF_GET_INUMBER(sfp, from) \ - xfs_dir2_sf_get_inumber(sfp, from) +static inline xfs_dir2_inou_t *xfs_dir2_sf_inumberp(xfs_dir2_sf_entry_t *sfep) +{ + return (xfs_dir2_inou_t *)&(sfep)->name[(sfep)->namelen]; +} -#else -#define XFS_DIR2_SF_GET_INUMBER(sfp, from) \ - ((sfp)->hdr.i8count == 0 ? \ +#define XFS_DIR2_SF_GET_INUMBER(sfp, from) \ + xfs_dir2_sf_get_inumber(sfp, from) +static inline xfs_intino_t +xfs_dir2_sf_get_inumber(xfs_dir2_sf_t *sfp, xfs_dir2_inou_t *from) +{ + return ((sfp)->hdr.i8count == 0 ? \ (xfs_intino_t)XFS_GET_DIR_INO4((from)->i4) : \ - (xfs_intino_t)XFS_GET_DIR_INO8((from)->i8)) -#endif + (xfs_intino_t)XFS_GET_DIR_INO8((from)->i8)); +} -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_SF_PUT_INUMBER) -void xfs_dir2_sf_put_inumber(xfs_dir2_sf_t *sfp, xfs_ino_t *from, - xfs_dir2_inou_t *to); -#define XFS_DIR2_SF_PUT_INUMBER(sfp,from,to) \ +#define XFS_DIR2_SF_PUT_INUMBER(sfp,from,to) \ xfs_dir2_sf_put_inumber(sfp,from,to) -#else -#define XFS_DIR2_SF_PUT_INUMBER(sfp,from,to) \ - if ((sfp)->hdr.i8count == 0) { \ - XFS_PUT_DIR_INO4(*(from), (to)->i4); \ - } else { \ - XFS_PUT_DIR_INO8(*(from), (to)->i8); \ - } -#endif - -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_SF_GET_OFFSET) -xfs_dir2_data_aoff_t xfs_dir2_sf_get_offset(xfs_dir2_sf_entry_t *sfep); +static inline void xfs_dir2_sf_put_inumber(xfs_dir2_sf_t *sfp, xfs_ino_t *from, + xfs_dir2_inou_t *to) +{ + if ((sfp)->hdr.i8count == 0) + XFS_PUT_DIR_INO4(*(from), (to)->i4); + else + XFS_PUT_DIR_INO8(*(from), (to)->i8); +} + #define XFS_DIR2_SF_GET_OFFSET(sfep) \ xfs_dir2_sf_get_offset(sfep) -#else -#define XFS_DIR2_SF_GET_OFFSET(sfep) \ - INT_GET_UNALIGNED_16_BE(&(sfep)->offset.i) -#endif +static inline xfs_dir2_data_aoff_t +xfs_dir2_sf_get_offset(xfs_dir2_sf_entry_t *sfep) +{ + return INT_GET_UNALIGNED_16_BE(&(sfep)->offset.i); +} -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_SF_PUT_OFFSET) -void xfs_dir2_sf_put_offset(xfs_dir2_sf_entry_t *sfep, - xfs_dir2_data_aoff_t off); #define XFS_DIR2_SF_PUT_OFFSET(sfep,off) \ xfs_dir2_sf_put_offset(sfep,off) -#else -#define XFS_DIR2_SF_PUT_OFFSET(sfep,off) \ - INT_SET_UNALIGNED_16_BE(&(sfep)->offset.i,off) -#endif +static inline void +xfs_dir2_sf_put_offset(xfs_dir2_sf_entry_t *sfep, xfs_dir2_data_aoff_t off) +{ + INT_SET_UNALIGNED_16_BE(&(sfep)->offset.i, off); +} -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_SF_ENTSIZE_BYNAME) -int xfs_dir2_sf_entsize_byname(xfs_dir2_sf_t *sfp, int len); #define XFS_DIR2_SF_ENTSIZE_BYNAME(sfp,len) \ xfs_dir2_sf_entsize_byname(sfp,len) -#else -#define XFS_DIR2_SF_ENTSIZE_BYNAME(sfp,len) /* space a name uses */ \ - ((uint)sizeof(xfs_dir2_sf_entry_t) - 1 + (len) - \ - ((sfp)->hdr.i8count == 0) * \ - ((uint)sizeof(xfs_dir2_ino8_t) - (uint)sizeof(xfs_dir2_ino4_t))) -#endif - -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_SF_ENTSIZE_BYENTRY) -int xfs_dir2_sf_entsize_byentry(xfs_dir2_sf_t *sfp, xfs_dir2_sf_entry_t *sfep); +static inline int xfs_dir2_sf_entsize_byname(xfs_dir2_sf_t *sfp, int len) +{ + return ((uint)sizeof(xfs_dir2_sf_entry_t) - 1 + (len) - \ + ((sfp)->hdr.i8count == 0) * \ + ((uint)sizeof(xfs_dir2_ino8_t) - (uint)sizeof(xfs_dir2_ino4_t))); +} + #define XFS_DIR2_SF_ENTSIZE_BYENTRY(sfp,sfep) \ xfs_dir2_sf_entsize_byentry(sfp,sfep) -#else -#define XFS_DIR2_SF_ENTSIZE_BYENTRY(sfp,sfep) /* space an entry uses */ \ - ((uint)sizeof(xfs_dir2_sf_entry_t) - 1 + (sfep)->namelen - \ - ((sfp)->hdr.i8count == 0) * \ - ((uint)sizeof(xfs_dir2_ino8_t) - (uint)sizeof(xfs_dir2_ino4_t))) -#endif - -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_SF_FIRSTENTRY) -xfs_dir2_sf_entry_t *xfs_dir2_sf_firstentry(xfs_dir2_sf_t *sfp); +static inline int +xfs_dir2_sf_entsize_byentry(xfs_dir2_sf_t *sfp, xfs_dir2_sf_entry_t *sfep) +{ + return ((uint)sizeof(xfs_dir2_sf_entry_t) - 1 + (sfep)->namelen - \ + ((sfp)->hdr.i8count == 0) * \ + ((uint)sizeof(xfs_dir2_ino8_t) - (uint)sizeof(xfs_dir2_ino4_t))); +} + #define XFS_DIR2_SF_FIRSTENTRY(sfp) xfs_dir2_sf_firstentry(sfp) -#else -#define XFS_DIR2_SF_FIRSTENTRY(sfp) /* first entry in struct */ \ - ((xfs_dir2_sf_entry_t *) \ - ((char *)(sfp) + XFS_DIR2_SF_HDR_SIZE(sfp->hdr.i8count))) -#endif - -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR2_SF_NEXTENTRY) -xfs_dir2_sf_entry_t *xfs_dir2_sf_nextentry(xfs_dir2_sf_t *sfp, - xfs_dir2_sf_entry_t *sfep); -#define XFS_DIR2_SF_NEXTENTRY(sfp,sfep) xfs_dir2_sf_nextentry(sfp,sfep) -#else -#define XFS_DIR2_SF_NEXTENTRY(sfp,sfep) /* next entry in struct */ \ - ((xfs_dir2_sf_entry_t *) \ - ((char *)(sfep) + XFS_DIR2_SF_ENTSIZE_BYENTRY(sfp,sfep))) -#endif +static inline xfs_dir2_sf_entry_t *xfs_dir2_sf_firstentry(xfs_dir2_sf_t *sfp) +{ + return ((xfs_dir2_sf_entry_t *) \ + ((char *)(sfp) + XFS_DIR2_SF_HDR_SIZE(sfp->hdr.i8count))); +} + +#define XFS_DIR2_SF_NEXTENTRY(sfp,sfep) xfs_dir2_sf_nextentry(sfp,sfep) +static inline xfs_dir2_sf_entry_t * +xfs_dir2_sf_nextentry(xfs_dir2_sf_t *sfp, xfs_dir2_sf_entry_t *sfep) +{ + return ((xfs_dir2_sf_entry_t *) \ + ((char *)(sfep) + XFS_DIR2_SF_ENTSIZE_BYENTRY(sfp,sfep))); +} /* * Functions. */ - -extern int - xfs_dir2_block_sfsize(struct xfs_inode *dp, - struct xfs_dir2_block *block, - xfs_dir2_sf_hdr_t *sfhp); - -extern int - xfs_dir2_block_to_sf(struct xfs_da_args *args, struct xfs_dabuf *bp, - int size, xfs_dir2_sf_hdr_t *sfhp); - -extern int - xfs_dir2_sf_addname(struct xfs_da_args *args); - -extern int - xfs_dir2_sf_create(struct xfs_da_args *args, xfs_ino_t pino); - -extern int - xfs_dir2_sf_getdents(struct xfs_inode *dp, struct uio *uio, int *eofp, - struct xfs_dirent *dbp, xfs_dir2_put_t put); - -extern int - xfs_dir2_sf_lookup(struct xfs_da_args *args); - -extern int - xfs_dir2_sf_removename(struct xfs_da_args *args); - -extern int - xfs_dir2_sf_replace(struct xfs_da_args *args); +extern int xfs_dir2_block_sfsize(struct xfs_inode *dp, + struct xfs_dir2_block *block, + xfs_dir2_sf_hdr_t *sfhp); +extern int xfs_dir2_block_to_sf(struct xfs_da_args *args, struct xfs_dabuf *bp, + int size, xfs_dir2_sf_hdr_t *sfhp); +extern int xfs_dir2_sf_addname(struct xfs_da_args *args); +extern int xfs_dir2_sf_create(struct xfs_da_args *args, xfs_ino_t pino); +extern int xfs_dir2_sf_getdents(struct xfs_inode *dp, struct uio *uio, + int *eofp, struct xfs_dirent *dbp, + xfs_dir2_put_t put); +extern int xfs_dir2_sf_lookup(struct xfs_da_args *args); +extern int xfs_dir2_sf_removename(struct xfs_da_args *args); +extern int xfs_dir2_sf_replace(struct xfs_da_args *args); #endif /* __XFS_DIR2_SF_H__ */ diff --git a/fs/xfs/xfs_dir2_trace.c b/fs/xfs/xfs_dir2_trace.c index 9d6417393140..c626943b4112 100644 --- a/fs/xfs/xfs_dir2_trace.c +++ b/fs/xfs/xfs_dir2_trace.c @@ -1,52 +1,33 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ - */ - -/* - * xfs_dir2_trace.c - * Tracing for xfs v2 directories. + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #include "xfs.h" - +#include "xfs_fs.h" #include "xfs_types.h" #include "xfs_inum.h" #include "xfs_dir.h" #include "xfs_dir2.h" +#include "xfs_da_btree.h" #include "xfs_bmap_btree.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" #include "xfs_inode.h" -#include "xfs_da_btree.h" #include "xfs_dir2_trace.h" #ifdef XFS_DIR2_TRACE diff --git a/fs/xfs/xfs_dir2_trace.h b/fs/xfs/xfs_dir2_trace.h index 0a178bffa806..ca3c754f4822 100644 --- a/fs/xfs/xfs_dir2_trace.h +++ b/fs/xfs/xfs_dir2_trace.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_DIR2_TRACE_H__ #define __XFS_DIR2_TRACE_H__ diff --git a/fs/xfs/xfs_dir_leaf.c b/fs/xfs/xfs_dir_leaf.c index c2ea6171fb0e..950df31efc46 100644 --- a/fs/xfs/xfs_dir_leaf.c +++ b/fs/xfs/xfs_dir_leaf.c @@ -1,66 +1,44 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - -/* - * xfs_dir_leaf.c - * - * GROT: figure out how to recover gracefully when bmap returns ENOSPC. - */ - #include "xfs.h" - -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_dir.h" #include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" -#include "xfs_alloc_btree.h" +#include "xfs_da_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_alloc.h" -#include "xfs_btree.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" -#include "xfs_inode_item.h" #include "xfs_inode.h" +#include "xfs_inode_item.h" +#include "xfs_alloc.h" +#include "xfs_btree.h" #include "xfs_bmap.h" -#include "xfs_da_btree.h" #include "xfs_dir_leaf.h" #include "xfs_error.h" @@ -508,7 +486,7 @@ xfs_dir_shortform_getdents(xfs_inode_t *dp, uio_t *uio, int *eofp, /* * Sort the entries on hash then entno. */ - qsort(sbuf, nsbuf, sizeof(*sbuf), xfs_dir_shortform_compare); + xfs_sort(sbuf, nsbuf, sizeof(*sbuf), xfs_dir_shortform_compare); /* * Stuff in last entry. */ @@ -650,7 +628,7 @@ xfs_dir_leaf_to_shortform(xfs_da_args_t *iargs) xfs_dir_leaf_name_t *namest; xfs_da_args_t args; xfs_inode_t *dp; - xfs_ino_t parent; + xfs_ino_t parent = 0; char *tmpbuffer; int retval, i; xfs_dabuf_t *bp; diff --git a/fs/xfs/xfs_dir_leaf.h b/fs/xfs/xfs_dir_leaf.h index 480bffc1f29f..ab6b09eef9ab 100644 --- a/fs/xfs/xfs_dir_leaf.h +++ b/fs/xfs/xfs_dir_leaf.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2001 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2001,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_DIR_LEAF_H__ #define __XFS_DIR_LEAF_H__ @@ -152,30 +138,26 @@ typedef struct xfs_dir_put_args struct uio *uio; /* uio control structure */ } xfs_dir_put_args_t; -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR_LEAF_ENTSIZE_BYNAME) -int xfs_dir_leaf_entsize_byname(int len); #define XFS_DIR_LEAF_ENTSIZE_BYNAME(len) xfs_dir_leaf_entsize_byname(len) -#else -#define XFS_DIR_LEAF_ENTSIZE_BYNAME(len) /* space a name will use */ \ - ((uint)sizeof(xfs_dir_leaf_name_t)-1 + len) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR_LEAF_ENTSIZE_BYENTRY) -int xfs_dir_leaf_entsize_byentry(xfs_dir_leaf_entry_t *entry); +static inline int xfs_dir_leaf_entsize_byname(int len) +{ + return (uint)sizeof(xfs_dir_leaf_name_t)-1 + len; +} + #define XFS_DIR_LEAF_ENTSIZE_BYENTRY(entry) \ xfs_dir_leaf_entsize_byentry(entry) -#else -#define XFS_DIR_LEAF_ENTSIZE_BYENTRY(entry) /* space an entry will use */ \ - ((uint)sizeof(xfs_dir_leaf_name_t)-1 + (entry)->namelen) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR_LEAF_NAMESTRUCT) -xfs_dir_leaf_name_t * -xfs_dir_leaf_namestruct(xfs_dir_leafblock_t *leafp, int offset); +static inline int xfs_dir_leaf_entsize_byentry(xfs_dir_leaf_entry_t *entry) +{ + return (uint)sizeof(xfs_dir_leaf_name_t)-1 + (entry)->namelen; +} + #define XFS_DIR_LEAF_NAMESTRUCT(leafp,offset) \ xfs_dir_leaf_namestruct(leafp,offset) -#else -#define XFS_DIR_LEAF_NAMESTRUCT(leafp,offset) /* point to name struct */ \ - ((xfs_dir_leaf_name_t *)&((char *)(leafp))[offset]) -#endif +static inline xfs_dir_leaf_name_t * +xfs_dir_leaf_namestruct(xfs_dir_leafblock_t *leafp, int offset) +{ + return (xfs_dir_leaf_name_t *)&((char *)(leafp))[offset]; +} /*======================================================================== * Function prototypes for the kernel. @@ -190,7 +172,7 @@ int xfs_dir_shortform_lookup(struct xfs_da_args *args); int xfs_dir_shortform_to_leaf(struct xfs_da_args *args); int xfs_dir_shortform_removename(struct xfs_da_args *args); int xfs_dir_shortform_getdents(struct xfs_inode *dp, struct uio *uio, int *eofp, - struct xfs_dirent *dbp, xfs_dir_put_t put); + struct xfs_dirent *dbp, xfs_dir_put_t put); int xfs_dir_shortform_replace(struct xfs_da_args *args); /* @@ -237,7 +219,6 @@ int xfs_dir_put_dirent64_direct(xfs_dir_put_args_t *pa); int xfs_dir_put_dirent64_uio(xfs_dir_put_args_t *pa); int xfs_dir_ino_validate(struct xfs_mount *mp, xfs_ino_t ino); - /* * Global data. */ diff --git a/fs/xfs/xfs_dir_sf.h b/fs/xfs/xfs_dir_sf.h index a61bcfc2a87d..fe44c6f4d560 100644 --- a/fs/xfs/xfs_dir_sf.h +++ b/fs/xfs/xfs_dir_sf.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_DIR_SF_H__ #define __XFS_DIR_SF_H__ @@ -76,49 +62,44 @@ typedef struct xfs_dir_sf_sort { char *name; /* name value, pointer into buffer */ } xfs_dir_sf_sort_t; -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR_SF_GET_DIRINO) -void xfs_dir_sf_get_dirino(xfs_dir_ino_t *from, xfs_ino_t *to); -#define XFS_DIR_SF_GET_DIRINO(from,to) xfs_dir_sf_get_dirino(from, to) -#else -#define XFS_DIR_SF_GET_DIRINO(from,to) (*(to) = XFS_GET_DIR_INO8(*from)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR_SF_PUT_DIRINO) -void xfs_dir_sf_put_dirino(xfs_ino_t *from, xfs_dir_ino_t *to); -#define XFS_DIR_SF_PUT_DIRINO(from,to) xfs_dir_sf_put_dirino(from, to) -#else -#define XFS_DIR_SF_PUT_DIRINO(from,to) XFS_PUT_DIR_INO8(*(from), *(to)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR_SF_ENTSIZE_BYNAME) -int xfs_dir_sf_entsize_byname(int len); -#define XFS_DIR_SF_ENTSIZE_BYNAME(len) xfs_dir_sf_entsize_byname(len) -#else -#define XFS_DIR_SF_ENTSIZE_BYNAME(len) /* space a name uses */ \ - ((uint)sizeof(xfs_dir_sf_entry_t)-1 + (len)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR_SF_ENTSIZE_BYENTRY) -int xfs_dir_sf_entsize_byentry(xfs_dir_sf_entry_t *sfep); +#define XFS_DIR_SF_GET_DIRINO(from,to) xfs_dir_sf_get_dirino(from, to) +static inline void xfs_dir_sf_get_dirino(xfs_dir_ino_t *from, xfs_ino_t *to) +{ + *(to) = XFS_GET_DIR_INO8(*from); +} + +#define XFS_DIR_SF_PUT_DIRINO(from,to) xfs_dir_sf_put_dirino(from, to) +static inline void xfs_dir_sf_put_dirino(xfs_ino_t *from, xfs_dir_ino_t *to) +{ + XFS_PUT_DIR_INO8(*(from), *(to)); +} + +#define XFS_DIR_SF_ENTSIZE_BYNAME(len) xfs_dir_sf_entsize_byname(len) +static inline int xfs_dir_sf_entsize_byname(int len) +{ + return (uint)sizeof(xfs_dir_sf_entry_t)-1 + (len); +} + #define XFS_DIR_SF_ENTSIZE_BYENTRY(sfep) xfs_dir_sf_entsize_byentry(sfep) -#else -#define XFS_DIR_SF_ENTSIZE_BYENTRY(sfep) /* space an entry uses */ \ - ((uint)sizeof(xfs_dir_sf_entry_t)-1 + (sfep)->namelen) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR_SF_NEXTENTRY) -xfs_dir_sf_entry_t *xfs_dir_sf_nextentry(xfs_dir_sf_entry_t *sfep); +static inline int xfs_dir_sf_entsize_byentry(xfs_dir_sf_entry_t *sfep) +{ + return (uint)sizeof(xfs_dir_sf_entry_t)-1 + (sfep)->namelen; +} + #define XFS_DIR_SF_NEXTENTRY(sfep) xfs_dir_sf_nextentry(sfep) -#else -#define XFS_DIR_SF_NEXTENTRY(sfep) /* next entry in struct */ \ - ((xfs_dir_sf_entry_t *) \ - ((char *)(sfep) + XFS_DIR_SF_ENTSIZE_BYENTRY(sfep))) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DIR_SF_ALLFIT) -int xfs_dir_sf_allfit(int count, int totallen); +static inline xfs_dir_sf_entry_t *xfs_dir_sf_nextentry(xfs_dir_sf_entry_t *sfep) +{ + return (xfs_dir_sf_entry_t *) \ + ((char *)(sfep) + XFS_DIR_SF_ENTSIZE_BYENTRY(sfep)); +} + #define XFS_DIR_SF_ALLFIT(count,totallen) \ xfs_dir_sf_allfit(count,totallen) -#else -#define XFS_DIR_SF_ALLFIT(count,totallen) /* will all entries fit? */ \ - ((uint)sizeof(xfs_dir_sf_hdr_t) + \ - ((uint)sizeof(xfs_dir_sf_entry_t)-1)*(count) + (totallen)) -#endif +static inline int xfs_dir_sf_allfit(int count, int totallen) +{ + return ((uint)sizeof(xfs_dir_sf_hdr_t) + \ + ((uint)sizeof(xfs_dir_sf_entry_t)-1)*(count) + (totallen)); +} #if defined(XFS_DIR_TRACE) diff --git a/fs/xfs/xfs_dmapi.h b/fs/xfs/xfs_dmapi.h index 19e872856f6b..864bf6955689 100644 --- a/fs/xfs/xfs_dmapi.h +++ b/fs/xfs/xfs_dmapi.h @@ -1,37 +1,24 @@ /* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_DMAPI_H__ #define __XFS_DMAPI_H__ +#include <linux/version.h> /* Values used to define the on-disk version of dm_attrname_t. All * on-disk attribute names start with the 8-byte string "SGI_DMI_". * diff --git a/fs/xfs/xfs_dmops.c b/fs/xfs/xfs_dmops.c index cec54ba800eb..629795b3b3d5 100644 --- a/fs/xfs/xfs_dmops.c +++ b/fs/xfs/xfs_dmops.c @@ -1,40 +1,25 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #include "xfs.h" - -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" diff --git a/fs/xfs/xfs_error.c b/fs/xfs/xfs_error.c index dcd3fdd5c1f7..d7b6b5d16704 100644 --- a/fs/xfs/xfs_error.c +++ b/fs/xfs/xfs_error.c @@ -1,51 +1,35 @@ /* - * Copyright (c) 2000-2001 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2001,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" - -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" #include "xfs_log.h" -#include "xfs_sb.h" +#include "xfs_inum.h" #include "xfs_trans.h" +#include "xfs_sb.h" #include "xfs_dir.h" #include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" #include "xfs_bmap_btree.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" #include "xfs_inode.h" #include "xfs_utils.h" diff --git a/fs/xfs/xfs_error.h b/fs/xfs/xfs_error.h index 52ee2b90b5ed..06d8a8426c16 100644 --- a/fs/xfs/xfs_error.h +++ b/fs/xfs/xfs_error.h @@ -1,39 +1,25 @@ /* - * Copyright (c) 2000-2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_ERROR_H__ #define __XFS_ERROR_H__ #define prdev(fmt,targ,args...) \ - printk("XFS: device %s- " fmt "\n", XFS_BUFTARG_NAME(targ), ## args) + printk("XFS: device %s - " fmt "\n", XFS_BUFTARG_NAME(targ), ## args) #define XFS_ERECOVER 1 /* Failure to recover log */ #define XFS_ELOGSTAT 2 /* Failure to stat log in user space */ @@ -54,24 +40,10 @@ extern int xfs_error_trap(int); struct xfs_mount; -extern void -xfs_error_report( - char *tag, - int level, - struct xfs_mount *mp, - char *fname, - int linenum, - inst_t *ra); - -extern void -xfs_corruption_error( - char *tag, - int level, - struct xfs_mount *mp, - void *p, - char *fname, - int linenum, - inst_t *ra); +extern void xfs_error_report(char *tag, int level, struct xfs_mount *mp, + char *fname, int linenum, inst_t *ra); +extern void xfs_corruption_error(char *tag, int level, struct xfs_mount *mp, + void *p, char *fname, int linenum, inst_t *ra); #define XFS_ERROR_REPORT(e, lvl, mp) \ xfs_error_report(e, lvl, mp, __FILE__, __LINE__, __return_address) @@ -84,6 +56,32 @@ xfs_corruption_error( #define XFS_ERRLEVEL_HIGH 5 /* + * Macros to set EFSCORRUPTED & return/branch. + */ +#define XFS_WANT_CORRUPTED_GOTO(x,l) \ + { \ + int fs_is_ok = (x); \ + ASSERT(fs_is_ok); \ + if (unlikely(!fs_is_ok)) { \ + XFS_ERROR_REPORT("XFS_WANT_CORRUPTED_GOTO", \ + XFS_ERRLEVEL_LOW, NULL); \ + error = XFS_ERROR(EFSCORRUPTED); \ + goto l; \ + } \ + } + +#define XFS_WANT_CORRUPTED_RETURN(x) \ + { \ + int fs_is_ok = (x); \ + ASSERT(fs_is_ok); \ + if (unlikely(!fs_is_ok)) { \ + XFS_ERROR_REPORT("XFS_WANT_CORRUPTED_RETURN", \ + XFS_ERRLEVEL_LOW, NULL); \ + return XFS_ERROR(EFSCORRUPTED); \ + } \ + } + +/* * error injection tags - the labels can be anything you want * but each tag should have its own unique number */ @@ -139,8 +137,8 @@ xfs_corruption_error( #define XFS_RANDOM_BMAPIFORMAT XFS_RANDOM_DEFAULT #if (defined(DEBUG) || defined(INDUCE_IO_ERROR)) -extern int xfs_error_test(int, int *, char *, int, char *, unsigned long); -void xfs_error_test_init(void); +extern int xfs_error_test(int, int *, char *, int, char *, unsigned long); +extern void xfs_error_test_init(void); #define XFS_NUM_INJECT_ERROR 10 @@ -156,12 +154,10 @@ void xfs_error_test_init(void); (rf))) #endif /* __ANSI_CPP__ */ -int xfs_errortag_add(int error_tag, xfs_mount_t *mp); -int xfs_errortag_clear(int error_tag, xfs_mount_t *mp); - -int xfs_errortag_clearall(xfs_mount_t *mp); -int xfs_errortag_clearall_umount(int64_t fsid, char *fsname, - int loud); +extern int xfs_errortag_add(int error_tag, xfs_mount_t *mp); +extern int xfs_errortag_clear(int error_tag, xfs_mount_t *mp); +extern int xfs_errortag_clearall(xfs_mount_t *mp); +extern int xfs_errortag_clearall_umount(int64_t fsid, char *fsname, int loud); #else #define XFS_TEST_ERROR(expr, mp, tag, rf) (expr) #define xfs_errortag_add(tag, mp) (ENOSYS) @@ -185,9 +181,9 @@ int xfs_errortag_clearall_umount(int64_t fsid, char *fsname, struct xfs_mount; /* PRINTFLIKE4 */ -void xfs_cmn_err(int panic_tag, int level, struct xfs_mount *mp, +extern void xfs_cmn_err(int panic_tag, int level, struct xfs_mount *mp, char *fmt, ...); /* PRINTFLIKE3 */ -void xfs_fs_cmn_err(int level, struct xfs_mount *mp, char *fmt, ...); +extern void xfs_fs_cmn_err(int level, struct xfs_mount *mp, char *fmt, ...); #endif /* __XFS_ERROR_H__ */ diff --git a/fs/xfs/xfs_extfree_item.c b/fs/xfs/xfs_extfree_item.c index cc7d1494a45d..f19282ec8549 100644 --- a/fs/xfs/xfs_extfree_item.c +++ b/fs/xfs/xfs_extfree_item.c @@ -1,46 +1,25 @@ /* - * Copyright (c) 2000-2001 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2001,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ - */ - -/* - * This file contains the implementation of the xfs_efi_log_item - * and xfs_efd_log_item items. + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" - -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_buf_item.h" #include "xfs_sb.h" diff --git a/fs/xfs/xfs_extfree_item.h b/fs/xfs/xfs_extfree_item.h index d433bac9f59d..5bf681708fec 100644 --- a/fs/xfs/xfs_extfree_item.h +++ b/fs/xfs/xfs_extfree_item.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_EXTFREE_ITEM_H__ #define __XFS_EXTFREE_ITEM_H__ diff --git a/fs/xfs/xfs_fs.h b/fs/xfs/xfs_fs.h index 095af0a5cff3..ba096f80f48d 100644 --- a/fs/xfs/xfs_fs.h +++ b/fs/xfs/xfs_fs.h @@ -1,34 +1,19 @@ /* - * Copyright (c) 1995-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 1995-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2.1 of the GNU Lesser General Public License - * as published by the Free Software Foundation. + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU Lesser General Public - * License along with this program; if not, write the Free Software - * Foundation, Inc., 59 Temple Place - Suite 330, Boston MA 02111-1307, - * USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_FS_H__ #define __XFS_FS_H__ @@ -251,6 +236,7 @@ typedef struct xfs_fsop_resblks { #define XFS_FSOP_GEOM_FLAGS_DIRV2 0x0080 /* directory version 2 */ #define XFS_FSOP_GEOM_FLAGS_LOGV2 0x0100 /* log format version 2 */ #define XFS_FSOP_GEOM_FLAGS_SECTOR 0x0200 /* sector sizes >1BB */ +#define XFS_FSOP_GEOM_FLAGS_ATTR2 0x0400 /* inline attributes rework */ /* diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c index ca535d613190..d1236d6f4045 100644 --- a/fs/xfs/xfs_fsops.c +++ b/fs/xfs/xfs_fsops.c @@ -1,66 +1,51 @@ /* - * Copyright (c) 2000-2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" +#include "xfs_bit.h" #include "xfs_inum.h" #include "xfs_log.h" #include "xfs_trans.h" #include "xfs_sb.h" +#include "xfs_ag.h" #include "xfs_dir.h" +#include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" -#include "xfs_ag.h" -#include "xfs_alloc_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" +#include "xfs_dir_sf.h" +#include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" +#include "xfs_dinode.h" +#include "xfs_inode.h" +#include "xfs_inode_item.h" #include "xfs_btree.h" #include "xfs_error.h" #include "xfs_alloc.h" #include "xfs_ialloc.h" #include "xfs_fsops.h" #include "xfs_itable.h" -#include "xfs_rw.h" -#include "xfs_refcache.h" #include "xfs_trans_space.h" #include "xfs_rtalloc.h" -#include "xfs_dir2.h" -#include "xfs_attr_sf.h" -#include "xfs_dir_sf.h" -#include "xfs_dir2_sf.h" -#include "xfs_dinode.h" -#include "xfs_inode.h" -#include "xfs_inode_item.h" +#include "xfs_rw.h" /* * File system operations @@ -110,7 +95,9 @@ xfs_fs_geometry( (XFS_SB_VERSION_HASDIRV2(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_DIRV2 : 0) | (XFS_SB_VERSION_HASSECTOR(&mp->m_sb) ? - XFS_FSOP_GEOM_FLAGS_SECTOR : 0); + XFS_FSOP_GEOM_FLAGS_SECTOR : 0) | + (XFS_SB_VERSION_HASATTR2(&mp->m_sb) ? + XFS_FSOP_GEOM_FLAGS_ATTR2 : 0); geo->logsectsize = XFS_SB_VERSION_HASSECTOR(&mp->m_sb) ? mp->m_sb.sb_logsectsize : BBSIZE; geo->rtsectsize = mp->m_sb.sb_blocksize; @@ -184,7 +171,7 @@ xfs_growfs_data_private( memset(&mp->m_perag[oagcount], 0, (nagcount - oagcount) * sizeof(xfs_perag_t)); mp->m_flags |= XFS_MOUNT_32BITINODES; - nagimax = xfs_initialize_perag(mp, nagcount); + nagimax = xfs_initialize_perag(XFS_MTOVFS(mp), mp, nagcount); up_write(&mp->m_peraglock); } tp = xfs_trans_alloc(mp, XFS_TRANS_GROWFS); @@ -204,28 +191,26 @@ xfs_growfs_data_private( XFS_FSS_TO_BB(mp, 1), 0); agf = XFS_BUF_TO_AGF(bp); memset(agf, 0, mp->m_sb.sb_sectsize); - INT_SET(agf->agf_magicnum, ARCH_CONVERT, XFS_AGF_MAGIC); - INT_SET(agf->agf_versionnum, ARCH_CONVERT, XFS_AGF_VERSION); - INT_SET(agf->agf_seqno, ARCH_CONVERT, agno); + agf->agf_magicnum = cpu_to_be32(XFS_AGF_MAGIC); + agf->agf_versionnum = cpu_to_be32(XFS_AGF_VERSION); + agf->agf_seqno = cpu_to_be32(agno); if (agno == nagcount - 1) agsize = nb - (agno * (xfs_rfsblock_t)mp->m_sb.sb_agblocks); else agsize = mp->m_sb.sb_agblocks; - INT_SET(agf->agf_length, ARCH_CONVERT, agsize); - INT_SET(agf->agf_roots[XFS_BTNUM_BNOi], ARCH_CONVERT, - XFS_BNO_BLOCK(mp)); - INT_SET(agf->agf_roots[XFS_BTNUM_CNTi], ARCH_CONVERT, - XFS_CNT_BLOCK(mp)); - INT_SET(agf->agf_levels[XFS_BTNUM_BNOi], ARCH_CONVERT, 1); - INT_SET(agf->agf_levels[XFS_BTNUM_CNTi], ARCH_CONVERT, 1); + agf->agf_length = cpu_to_be32(agsize); + agf->agf_roots[XFS_BTNUM_BNOi] = cpu_to_be32(XFS_BNO_BLOCK(mp)); + agf->agf_roots[XFS_BTNUM_CNTi] = cpu_to_be32(XFS_CNT_BLOCK(mp)); + agf->agf_levels[XFS_BTNUM_BNOi] = cpu_to_be32(1); + agf->agf_levels[XFS_BTNUM_CNTi] = cpu_to_be32(1); agf->agf_flfirst = 0; - INT_SET(agf->agf_fllast, ARCH_CONVERT, XFS_AGFL_SIZE(mp) - 1); + agf->agf_fllast = cpu_to_be32(XFS_AGFL_SIZE(mp) - 1); agf->agf_flcount = 0; tmpsize = agsize - XFS_PREALLOC_BLOCKS(mp); - INT_SET(agf->agf_freeblks, ARCH_CONVERT, tmpsize); - INT_SET(agf->agf_longest, ARCH_CONVERT, tmpsize); + agf->agf_freeblks = cpu_to_be32(tmpsize); + agf->agf_longest = cpu_to_be32(tmpsize); error = xfs_bwrite(mp, bp); if (error) { goto error0; @@ -238,19 +223,18 @@ xfs_growfs_data_private( XFS_FSS_TO_BB(mp, 1), 0); agi = XFS_BUF_TO_AGI(bp); memset(agi, 0, mp->m_sb.sb_sectsize); - INT_SET(agi->agi_magicnum, ARCH_CONVERT, XFS_AGI_MAGIC); - INT_SET(agi->agi_versionnum, ARCH_CONVERT, XFS_AGI_VERSION); - INT_SET(agi->agi_seqno, ARCH_CONVERT, agno); - INT_SET(agi->agi_length, ARCH_CONVERT, agsize); + agi->agi_magicnum = cpu_to_be32(XFS_AGI_MAGIC); + agi->agi_versionnum = cpu_to_be32(XFS_AGI_VERSION); + agi->agi_seqno = cpu_to_be32(agno); + agi->agi_length = cpu_to_be32(agsize); agi->agi_count = 0; - INT_SET(agi->agi_root, ARCH_CONVERT, XFS_IBT_BLOCK(mp)); - INT_SET(agi->agi_level, ARCH_CONVERT, 1); + agi->agi_root = cpu_to_be32(XFS_IBT_BLOCK(mp)); + agi->agi_level = cpu_to_be32(1); agi->agi_freecount = 0; - INT_SET(agi->agi_newino, ARCH_CONVERT, NULLAGINO); - INT_SET(agi->agi_dirino, ARCH_CONVERT, NULLAGINO); + agi->agi_newino = cpu_to_be32(NULLAGINO); + agi->agi_dirino = cpu_to_be32(NULLAGINO); for (bucket = 0; bucket < XFS_AGI_UNLINKED_BUCKETS; bucket++) - INT_SET(agi->agi_unlinked[bucket], ARCH_CONVERT, - NULLAGINO); + agi->agi_unlinked[bucket] = cpu_to_be32(NULLAGINO); error = xfs_bwrite(mp, bp); if (error) { goto error0; @@ -263,17 +247,16 @@ xfs_growfs_data_private( BTOBB(mp->m_sb.sb_blocksize), 0); block = XFS_BUF_TO_SBLOCK(bp); memset(block, 0, mp->m_sb.sb_blocksize); - INT_SET(block->bb_magic, ARCH_CONVERT, XFS_ABTB_MAGIC); + block->bb_magic = cpu_to_be32(XFS_ABTB_MAGIC); block->bb_level = 0; - INT_SET(block->bb_numrecs, ARCH_CONVERT, 1); - INT_SET(block->bb_leftsib, ARCH_CONVERT, NULLAGBLOCK); - INT_SET(block->bb_rightsib, ARCH_CONVERT, NULLAGBLOCK); + block->bb_numrecs = cpu_to_be16(1); + block->bb_leftsib = cpu_to_be32(NULLAGBLOCK); + block->bb_rightsib = cpu_to_be32(NULLAGBLOCK); arec = XFS_BTREE_REC_ADDR(mp->m_sb.sb_blocksize, xfs_alloc, block, 1, mp->m_alloc_mxr[0]); - INT_SET(arec->ar_startblock, ARCH_CONVERT, - XFS_PREALLOC_BLOCKS(mp)); - INT_SET(arec->ar_blockcount, ARCH_CONVERT, - agsize - INT_GET(arec->ar_startblock, ARCH_CONVERT)); + arec->ar_startblock = cpu_to_be32(XFS_PREALLOC_BLOCKS(mp)); + arec->ar_blockcount = cpu_to_be32( + agsize - be32_to_cpu(arec->ar_startblock)); error = xfs_bwrite(mp, bp); if (error) { goto error0; @@ -286,18 +269,17 @@ xfs_growfs_data_private( BTOBB(mp->m_sb.sb_blocksize), 0); block = XFS_BUF_TO_SBLOCK(bp); memset(block, 0, mp->m_sb.sb_blocksize); - INT_SET(block->bb_magic, ARCH_CONVERT, XFS_ABTC_MAGIC); + block->bb_magic = cpu_to_be32(XFS_ABTC_MAGIC); block->bb_level = 0; - INT_SET(block->bb_numrecs, ARCH_CONVERT, 1); - INT_SET(block->bb_leftsib, ARCH_CONVERT, NULLAGBLOCK); - INT_SET(block->bb_rightsib, ARCH_CONVERT, NULLAGBLOCK); + block->bb_numrecs = cpu_to_be16(1); + block->bb_leftsib = cpu_to_be32(NULLAGBLOCK); + block->bb_rightsib = cpu_to_be32(NULLAGBLOCK); arec = XFS_BTREE_REC_ADDR(mp->m_sb.sb_blocksize, xfs_alloc, block, 1, mp->m_alloc_mxr[0]); - INT_SET(arec->ar_startblock, ARCH_CONVERT, - XFS_PREALLOC_BLOCKS(mp)); - INT_SET(arec->ar_blockcount, ARCH_CONVERT, - agsize - INT_GET(arec->ar_startblock, ARCH_CONVERT)); - nfree += INT_GET(arec->ar_blockcount, ARCH_CONVERT); + arec->ar_startblock = cpu_to_be32(XFS_PREALLOC_BLOCKS(mp)); + arec->ar_blockcount = cpu_to_be32( + agsize - be32_to_cpu(arec->ar_startblock)); + nfree += be32_to_cpu(arec->ar_blockcount); error = xfs_bwrite(mp, bp); if (error) { goto error0; @@ -310,11 +292,11 @@ xfs_growfs_data_private( BTOBB(mp->m_sb.sb_blocksize), 0); block = XFS_BUF_TO_SBLOCK(bp); memset(block, 0, mp->m_sb.sb_blocksize); - INT_SET(block->bb_magic, ARCH_CONVERT, XFS_IBT_MAGIC); + block->bb_magic = cpu_to_be32(XFS_IBT_MAGIC); block->bb_level = 0; block->bb_numrecs = 0; - INT_SET(block->bb_leftsib, ARCH_CONVERT, NULLAGBLOCK); - INT_SET(block->bb_rightsib, ARCH_CONVERT, NULLAGBLOCK); + block->bb_leftsib = cpu_to_be32(NULLAGBLOCK); + block->bb_rightsib = cpu_to_be32(NULLAGBLOCK); error = xfs_bwrite(mp, bp); if (error) { goto error0; @@ -334,10 +316,9 @@ xfs_growfs_data_private( } ASSERT(bp); agi = XFS_BUF_TO_AGI(bp); - INT_MOD(agi->agi_length, ARCH_CONVERT, new); + be32_add(&agi->agi_length, new); ASSERT(nagcount == oagcount || - INT_GET(agi->agi_length, ARCH_CONVERT) == - mp->m_sb.sb_agblocks); + be32_to_cpu(agi->agi_length) == mp->m_sb.sb_agblocks); xfs_ialloc_log_agi(tp, bp, XFS_AGI_LENGTH); /* * Change agf length. @@ -348,14 +329,14 @@ xfs_growfs_data_private( } ASSERT(bp); agf = XFS_BUF_TO_AGF(bp); - INT_MOD(agf->agf_length, ARCH_CONVERT, new); - ASSERT(INT_GET(agf->agf_length, ARCH_CONVERT) == - INT_GET(agi->agi_length, ARCH_CONVERT)); + be32_add(&agf->agf_length, new); + ASSERT(be32_to_cpu(agf->agf_length) == + be32_to_cpu(agi->agi_length)); /* * Free the new space. */ error = xfs_free_extent(tp, XFS_AGB_TO_FSB(mp, agno, - INT_GET(agf->agf_length, ARCH_CONVERT) - new), new); + be32_to_cpu(agf->agf_length) - new), new); if (error) { goto error0; } @@ -569,7 +550,7 @@ xfs_fs_goingdown( struct vfs *vfsp = XFS_MTOVFS(mp); struct super_block *sb = freeze_bdev(vfsp->vfs_super->s_bdev); - if (sb) { + if (sb && !IS_ERR(sb)) { xfs_force_shutdown(mp, XFS_FORCE_UMOUNT); thaw_bdev(sb->s_bdev, sb); } diff --git a/fs/xfs/xfs_fsops.h b/fs/xfs/xfs_fsops.h index b61486173a61..f32713f14f9a 100644 --- a/fs/xfs/xfs_fsops.h +++ b/fs/xfs/xfs_fsops.h @@ -1,67 +1,29 @@ /* - * Copyright (c) 2000-2001 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2001,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_FSOPS_H__ #define __XFS_FSOPS_H__ -int -xfs_fs_geometry( - xfs_mount_t *mp, - xfs_fsop_geom_t *geo, - int new_version); - -int -xfs_growfs_data( - xfs_mount_t *mp, - xfs_growfs_data_t *in); - -int -xfs_growfs_log( - xfs_mount_t *mp, - xfs_growfs_log_t *in); - -int -xfs_fs_counts( - xfs_mount_t *mp, - xfs_fsop_counts_t *cnt); - -int -xfs_reserve_blocks( - xfs_mount_t *mp, - __uint64_t *inval, - xfs_fsop_resblks_t *outval); - -int -xfs_fs_goingdown( - xfs_mount_t *mp, - __uint32_t inflags); +extern int xfs_fs_geometry(xfs_mount_t *mp, xfs_fsop_geom_t *geo, int nversion); +extern int xfs_growfs_data(xfs_mount_t *mp, xfs_growfs_data_t *in); +extern int xfs_growfs_log(xfs_mount_t *mp, xfs_growfs_log_t *in); +extern int xfs_fs_counts(xfs_mount_t *mp, xfs_fsop_counts_t *cnt); +extern int xfs_reserve_blocks(xfs_mount_t *mp, __uint64_t *inval, + xfs_fsop_resblks_t *outval); +extern int xfs_fs_goingdown(xfs_mount_t *mp, __uint32_t inflags); #endif /* __XFS_FSOPS_H__ */ diff --git a/fs/xfs/xfs_ialloc.c b/fs/xfs/xfs_ialloc.c index ce5fee9eaec5..8f3fae1aa98a 100644 --- a/fs/xfs/xfs_ialloc.c +++ b/fs/xfs/xfs_ialloc.c @@ -1,41 +1,26 @@ /* - * Copyright (c) 2000-2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" - -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" @@ -43,18 +28,17 @@ #include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" -#include "xfs_alloc_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_btree.h" -#include "xfs_ialloc.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" #include "xfs_inode.h" +#include "xfs_btree.h" +#include "xfs_ialloc.h" #include "xfs_alloc.h" -#include "xfs_bit.h" #include "xfs_rtalloc.h" #include "xfs_error.h" #include "xfs_bmap.h" @@ -194,8 +178,8 @@ xfs_ialloc_ag_alloc( * Ideally they should be spaced out through the a.g. * For now, just allocate blocks up front. */ - args.agbno = INT_GET(agi->agi_root, ARCH_CONVERT); - args.fsbno = XFS_AGB_TO_FSB(args.mp, INT_GET(agi->agi_seqno, ARCH_CONVERT), + args.agbno = be32_to_cpu(agi->agi_root); + args.fsbno = XFS_AGB_TO_FSB(args.mp, be32_to_cpu(agi->agi_seqno), args.agbno); /* * Allocate a fixed-size extent of inodes. @@ -217,9 +201,9 @@ xfs_ialloc_ag_alloc( */ if (isaligned && args.fsbno == NULLFSBLOCK) { args.type = XFS_ALLOCTYPE_NEAR_BNO; - args.agbno = INT_GET(agi->agi_root, ARCH_CONVERT); + args.agbno = be32_to_cpu(agi->agi_root); args.fsbno = XFS_AGB_TO_FSB(args.mp, - INT_GET(agi->agi_seqno, ARCH_CONVERT), args.agbno); + be32_to_cpu(agi->agi_seqno), args.agbno); if (XFS_SB_VERSION_HASALIGN(&args.mp->m_sb) && args.mp->m_sb.sb_inoalignmt >= XFS_B_TO_FSBT(args.mp, XFS_INODE_CLUSTER_SIZE(args.mp))) @@ -274,7 +258,7 @@ xfs_ialloc_ag_alloc( /* * Get the block. */ - d = XFS_AGB_TO_DADDR(args.mp, INT_GET(agi->agi_seqno, ARCH_CONVERT), + d = XFS_AGB_TO_DADDR(args.mp, be32_to_cpu(agi->agi_seqno), args.agbno + (j * blks_per_cluster)); fbuf = xfs_trans_get_buf(tp, args.mp->m_ddev_targp, d, args.mp->m_bsize * blks_per_cluster, @@ -294,17 +278,17 @@ xfs_ialloc_ag_alloc( } xfs_trans_inode_alloc_buf(tp, fbuf); } - INT_MOD(agi->agi_count, ARCH_CONVERT, newlen); - INT_MOD(agi->agi_freecount, ARCH_CONVERT, newlen); + be32_add(&agi->agi_count, newlen); + be32_add(&agi->agi_freecount, newlen); down_read(&args.mp->m_peraglock); - args.mp->m_perag[INT_GET(agi->agi_seqno, ARCH_CONVERT)].pagi_freecount += newlen; + args.mp->m_perag[be32_to_cpu(agi->agi_seqno)].pagi_freecount += newlen; up_read(&args.mp->m_peraglock); - INT_SET(agi->agi_newino, ARCH_CONVERT, newino); + agi->agi_newino = cpu_to_be32(newino); /* * Insert records describing the new inode chunk into the btree. */ cur = xfs_btree_init_cursor(args.mp, tp, agbp, - INT_GET(agi->agi_seqno, ARCH_CONVERT), + be32_to_cpu(agi->agi_seqno), XFS_BTNUM_INO, (xfs_inode_t *)0, 0); for (thisino = newino; thisino < newino + newlen; @@ -544,7 +528,7 @@ xfs_dialloc( return 0; } agi = XFS_BUF_TO_AGI(agbp); - ASSERT(INT_GET(agi->agi_magicnum, ARCH_CONVERT) == XFS_AGI_MAGIC); + ASSERT(be32_to_cpu(agi->agi_magicnum) == XFS_AGI_MAGIC); } else { /* * Continue where we left off before. In this case, we @@ -552,12 +536,12 @@ xfs_dialloc( */ agbp = *IO_agbp; agi = XFS_BUF_TO_AGI(agbp); - ASSERT(INT_GET(agi->agi_magicnum, ARCH_CONVERT) == XFS_AGI_MAGIC); - ASSERT(INT_GET(agi->agi_freecount, ARCH_CONVERT) > 0); + ASSERT(be32_to_cpu(agi->agi_magicnum) == XFS_AGI_MAGIC); + ASSERT(be32_to_cpu(agi->agi_freecount) > 0); } mp = tp->t_mountp; agcount = mp->m_sb.sb_agcount; - agno = INT_GET(agi->agi_seqno, ARCH_CONVERT); + agno = be32_to_cpu(agi->agi_seqno); tagno = agno; pagno = XFS_INO_TO_AGNO(mp, parent); pagino = XFS_INO_TO_AGINO(mp, parent); @@ -605,7 +589,7 @@ xfs_dialloc( * can commit the current transaction and call * us again where we left off. */ - ASSERT(INT_GET(agi->agi_freecount, ARCH_CONVERT) > 0); + ASSERT(be32_to_cpu(agi->agi_freecount) > 0); *alloc_done = B_TRUE; *IO_agbp = agbp; *inop = NULLFSINO; @@ -636,7 +620,7 @@ nextag: if (error) goto nextag; agi = XFS_BUF_TO_AGI(agbp); - ASSERT(INT_GET(agi->agi_magicnum, ARCH_CONVERT) == XFS_AGI_MAGIC); + ASSERT(be32_to_cpu(agi->agi_magicnum) == XFS_AGI_MAGIC); } /* * Here with an allocation group that has a free inode. @@ -645,14 +629,14 @@ nextag: */ agno = tagno; *IO_agbp = NULL; - cur = xfs_btree_init_cursor(mp, tp, agbp, INT_GET(agi->agi_seqno, ARCH_CONVERT), + cur = xfs_btree_init_cursor(mp, tp, agbp, be32_to_cpu(agi->agi_seqno), XFS_BTNUM_INO, (xfs_inode_t *)0, 0); /* * If pagino is 0 (this is the root inode allocation) use newino. * This must work because we've just allocated some. */ if (!pagino) - pagino = INT_GET(agi->agi_newino, ARCH_CONVERT); + pagino = be32_to_cpu(agi->agi_newino); #ifdef DEBUG if (cur->bc_nlevels == 1) { int freecount = 0; @@ -670,7 +654,7 @@ nextag: goto error0; } while (i == 1); - ASSERT(freecount == INT_GET(agi->agi_freecount, ARCH_CONVERT) || + ASSERT(freecount == be32_to_cpu(agi->agi_freecount) || XFS_FORCED_SHUTDOWN(mp)); } #endif @@ -829,9 +813,9 @@ nextag: * In a different a.g. from the parent. * See if the most recently allocated block has any free. */ - else if (INT_GET(agi->agi_newino, ARCH_CONVERT) != NULLAGINO) { + else if (be32_to_cpu(agi->agi_newino) != NULLAGINO) { if ((error = xfs_inobt_lookup_eq(cur, - INT_GET(agi->agi_newino, ARCH_CONVERT), 0, 0, &i))) + be32_to_cpu(agi->agi_newino), 0, 0, &i))) goto error0; if (i == 1 && (error = xfs_inobt_get_rec(cur, &rec.ir_startino, @@ -878,7 +862,7 @@ nextag: if ((error = xfs_inobt_update(cur, rec.ir_startino, rec.ir_freecount, rec.ir_free))) goto error0; - INT_MOD(agi->agi_freecount, ARCH_CONVERT, -1); + be32_add(&agi->agi_freecount, -1); xfs_ialloc_log_agi(tp, agbp, XFS_AGI_FREECOUNT); down_read(&mp->m_peraglock); mp->m_perag[tagno].pagi_freecount--; @@ -898,7 +882,7 @@ nextag: if ((error = xfs_inobt_increment(cur, 0, &i))) goto error0; } while (i == 1); - ASSERT(freecount == INT_GET(agi->agi_freecount, ARCH_CONVERT) || + ASSERT(freecount == be32_to_cpu(agi->agi_freecount) || XFS_FORCED_SHUTDOWN(mp)); } #endif @@ -957,8 +941,11 @@ xfs_difree( agino = XFS_INO_TO_AGINO(mp, inode); if (inode != XFS_AGINO_TO_INO(mp, agno, agino)) { cmn_err(CE_WARN, - "xfs_difree: inode != XFS_AGINO_TO_INO() (%d != %d) on %s. Returning EINVAL.", - inode, XFS_AGINO_TO_INO(mp, agno, agino), mp->m_fsname); + "xfs_difree: inode != XFS_AGINO_TO_INO() " + "(%llu != %llu) on %s. Returning EINVAL.", + (unsigned long long)inode, + (unsigned long long)XFS_AGINO_TO_INO(mp, agno, agino), + mp->m_fsname); ASSERT(0); return XFS_ERROR(EINVAL); } @@ -983,8 +970,8 @@ xfs_difree( return error; } agi = XFS_BUF_TO_AGI(agbp); - ASSERT(INT_GET(agi->agi_magicnum, ARCH_CONVERT) == XFS_AGI_MAGIC); - ASSERT(agbno < INT_GET(agi->agi_length, ARCH_CONVERT)); + ASSERT(be32_to_cpu(agi->agi_magicnum) == XFS_AGI_MAGIC); + ASSERT(agbno < be32_to_cpu(agi->agi_length)); /* * Initialize the cursor. */ @@ -1006,7 +993,7 @@ xfs_difree( goto error0; } } while (i == 1); - ASSERT(freecount == INT_GET(agi->agi_freecount, ARCH_CONVERT) || + ASSERT(freecount == be32_to_cpu(agi->agi_freecount) || XFS_FORCED_SHUTDOWN(mp)); } #endif @@ -1055,8 +1042,8 @@ xfs_difree( * to be freed when the transaction is committed. */ ilen = XFS_IALLOC_INODES(mp); - INT_MOD(agi->agi_count, ARCH_CONVERT, -ilen); - INT_MOD(agi->agi_freecount, ARCH_CONVERT, -(ilen - 1)); + be32_add(&agi->agi_count, -ilen); + be32_add(&agi->agi_freecount, -(ilen - 1)); xfs_ialloc_log_agi(tp, agbp, XFS_AGI_COUNT | XFS_AGI_FREECOUNT); down_read(&mp->m_peraglock); mp->m_perag[agno].pagi_freecount -= ilen - 1; @@ -1085,7 +1072,7 @@ xfs_difree( /* * Change the inode free counts and log the ag/sb changes. */ - INT_MOD(agi->agi_freecount, ARCH_CONVERT, 1); + be32_add(&agi->agi_freecount, 1); xfs_ialloc_log_agi(tp, agbp, XFS_AGI_FREECOUNT); down_read(&mp->m_peraglock); mp->m_perag[agno].pagi_freecount++; @@ -1111,7 +1098,7 @@ xfs_difree( goto error0; } } while (i == 1); - ASSERT(freecount == INT_GET(agi->agi_freecount, ARCH_CONVERT) || + ASSERT(freecount == be32_to_cpu(agi->agi_freecount) || XFS_FORCED_SHUTDOWN(mp)); } #endif @@ -1320,7 +1307,7 @@ xfs_ialloc_log_agi( xfs_agi_t *agi; /* allocation group header */ agi = XFS_BUF_TO_AGI(bp); - ASSERT(INT_GET(agi->agi_magicnum, ARCH_CONVERT) == XFS_AGI_MAGIC); + ASSERT(be32_to_cpu(agi->agi_magicnum) == XFS_AGI_MAGIC); #endif /* * Compute byte offsets for the first and last fields. @@ -1362,9 +1349,8 @@ xfs_ialloc_read_agi( */ agi = XFS_BUF_TO_AGI(bp); agi_ok = - INT_GET(agi->agi_magicnum, ARCH_CONVERT) == XFS_AGI_MAGIC && - XFS_AGI_GOOD_VERSION( - INT_GET(agi->agi_versionnum, ARCH_CONVERT)); + be32_to_cpu(agi->agi_magicnum) == XFS_AGI_MAGIC && + XFS_AGI_GOOD_VERSION(be32_to_cpu(agi->agi_versionnum)); if (unlikely(XFS_TEST_ERROR(!agi_ok, mp, XFS_ERRTAG_IALLOC_READ_AGI, XFS_RANDOM_IALLOC_READ_AGI))) { XFS_CORRUPTION_ERROR("xfs_ialloc_read_agi", XFS_ERRLEVEL_LOW, @@ -1374,16 +1360,15 @@ xfs_ialloc_read_agi( } pag = &mp->m_perag[agno]; if (!pag->pagi_init) { - pag->pagi_freecount = INT_GET(agi->agi_freecount, ARCH_CONVERT); + pag->pagi_freecount = be32_to_cpu(agi->agi_freecount); pag->pagi_init = 1; } else { /* * It's possible for these to be out of sync if * we are in the middle of a forced shutdown. */ - ASSERT(pag->pagi_freecount == - INT_GET(agi->agi_freecount, ARCH_CONVERT) - || XFS_FORCED_SHUTDOWN(mp)); + ASSERT(pag->pagi_freecount == be32_to_cpu(agi->agi_freecount) || + XFS_FORCED_SHUTDOWN(mp)); } #ifdef DEBUG diff --git a/fs/xfs/xfs_ialloc.h b/fs/xfs/xfs_ialloc.h index db6d0015cecf..7f5debe1acb6 100644 --- a/fs/xfs/xfs_ialloc.h +++ b/fs/xfs/xfs_ialloc.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_IALLOC_H__ #define __XFS_IALLOC_H__ @@ -40,18 +26,8 @@ struct xfs_trans; /* * Allocation parameters for inode allocation. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_IALLOC_INODES) -int xfs_ialloc_inodes(struct xfs_mount *mp); -#define XFS_IALLOC_INODES(mp) xfs_ialloc_inodes(mp) -#else -#define XFS_IALLOC_INODES(mp) ((mp)->m_ialloc_inos) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_IALLOC_BLOCKS) -xfs_extlen_t xfs_ialloc_blocks(struct xfs_mount *mp); -#define XFS_IALLOC_BLOCKS(mp) xfs_ialloc_blocks(mp) -#else -#define XFS_IALLOC_BLOCKS(mp) ((mp)->m_ialloc_blks) -#endif +#define XFS_IALLOC_INODES(mp) (mp)->m_ialloc_inos +#define XFS_IALLOC_BLOCKS(mp) (mp)->m_ialloc_blks /* * For small block file systems, move inodes in clusters of this size. @@ -67,31 +43,25 @@ xfs_extlen_t xfs_ialloc_blocks(struct xfs_mount *mp); /* * Make an inode pointer out of the buffer/offset. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_MAKE_IPTR) -struct xfs_dinode *xfs_make_iptr(struct xfs_mount *mp, struct xfs_buf *b, int o); #define XFS_MAKE_IPTR(mp,b,o) xfs_make_iptr(mp,b,o) -#else -#define XFS_MAKE_IPTR(mp,b,o) \ - ((xfs_dinode_t *)(xfs_buf_offset(b, (o) << (mp)->m_sb.sb_inodelog))) -#endif +static inline struct xfs_dinode * +xfs_make_iptr(struct xfs_mount *mp, struct xfs_buf *b, int o) +{ + return (xfs_dinode_t *) + (xfs_buf_offset(b, o << (mp)->m_sb.sb_inodelog)); +} /* * Find a free (set) bit in the inode bitmask. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_IALLOC_FIND_FREE) -int xfs_ialloc_find_free(xfs_inofree_t *fp); #define XFS_IALLOC_FIND_FREE(fp) xfs_ialloc_find_free(fp) -#else -#define XFS_IALLOC_FIND_FREE(fp) xfs_lowbit64(*(fp)) -#endif +static inline int xfs_ialloc_find_free(xfs_inofree_t *fp) +{ + return xfs_lowbit64(*fp); +} #ifdef __KERNEL__ - -/* - * Prototypes for visible xfs_ialloc.c routines. - */ - /* * Allocate an inode on disk. * Mode is used to tell whether the new inode will need space, and whether diff --git a/fs/xfs/xfs_ialloc_btree.c b/fs/xfs/xfs_ialloc_btree.c index 2d4daecec990..60c65683462d 100644 --- a/fs/xfs/xfs_ialloc_btree.c +++ b/fs/xfs/xfs_ialloc_btree.c @@ -1,63 +1,46 @@ /* - * Copyright (c) 2000-2001 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2001,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" - -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" #include "xfs_dir.h" +#include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" -#include "xfs_alloc_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" +#include "xfs_dir_sf.h" +#include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" +#include "xfs_dinode.h" +#include "xfs_inode.h" #include "xfs_btree.h" #include "xfs_ialloc.h" #include "xfs_alloc.h" #include "xfs_error.h" -/* - * Inode allocation management for XFS. - */ - -/* - * Prototypes for internal functions. - */ - STATIC void xfs_inobt_log_block(xfs_trans_t *, xfs_buf_t *, int); STATIC void xfs_inobt_log_keys(xfs_btree_cur_t *, xfs_buf_t *, int, int); STATIC void xfs_inobt_log_ptrs(xfs_btree_cur_t *, xfs_buf_t *, int, int); @@ -70,10 +53,6 @@ STATIC int xfs_inobt_split(xfs_btree_cur_t *, int, xfs_agblock_t *, STATIC int xfs_inobt_updkey(xfs_btree_cur_t *, xfs_inobt_key_t *, int); /* - * Internal functions. - */ - -/* * Single level of the xfs_inobt_delete record deletion routine. * Delete record pointed to by cur/level. * Remove the record from its block then rebalance the tree. @@ -139,7 +118,7 @@ xfs_inobt_delrec( * Fail if we're off the end of the block. */ - numrecs = INT_GET(block->bb_numrecs, ARCH_CONVERT); + numrecs = be16_to_cpu(block->bb_numrecs); if (ptr > numrecs) { *stat = 0; return 0; @@ -154,7 +133,7 @@ xfs_inobt_delrec( pp = XFS_INOBT_PTR_ADDR(block, 1, cur); #ifdef DEBUG for (i = ptr; i < numrecs; i++) { - if ((error = xfs_btree_check_sptr(cur, INT_GET(pp[i], ARCH_CONVERT), level))) + if ((error = xfs_btree_check_sptr(cur, be32_to_cpu(pp[i]), level))) return error; } #endif @@ -191,7 +170,7 @@ xfs_inobt_delrec( * Decrement and log the number of entries in the block. */ numrecs--; - INT_SET(block->bb_numrecs, ARCH_CONVERT, numrecs); + block->bb_numrecs = cpu_to_be16(numrecs); xfs_inobt_log_block(cur->bc_tp, bp, XFS_BB_NUMRECS); /* * Is this the root level? If so, we're almost done. @@ -210,9 +189,9 @@ xfs_inobt_delrec( * pp is still set to the first pointer in the block. * Make it the new root of the btree. */ - bno = INT_GET(agi->agi_root, ARCH_CONVERT); + bno = be32_to_cpu(agi->agi_root); agi->agi_root = *pp; - INT_MOD(agi->agi_level, ARCH_CONVERT, -1); + be32_add(&agi->agi_level, -1); /* * Free the block. */ @@ -255,8 +234,8 @@ xfs_inobt_delrec( * tree balanced. Look at the left and right sibling blocks to * see if we can re-balance by moving only one record. */ - rbno = INT_GET(block->bb_rightsib, ARCH_CONVERT); - lbno = INT_GET(block->bb_leftsib, ARCH_CONVERT); + rbno = be32_to_cpu(block->bb_rightsib); + lbno = be32_to_cpu(block->bb_leftsib); bno = NULLAGBLOCK; ASSERT(rbno != NULLAGBLOCK || lbno != NULLAGBLOCK); /* @@ -293,18 +272,18 @@ xfs_inobt_delrec( /* * Grab the current block number, for future use. */ - bno = INT_GET(right->bb_leftsib, ARCH_CONVERT); + bno = be32_to_cpu(right->bb_leftsib); /* * If right block is full enough so that removing one entry * won't make it too empty, and left-shifting an entry out * of right to us works, we're done. */ - if (INT_GET(right->bb_numrecs, ARCH_CONVERT) - 1 >= + if (be16_to_cpu(right->bb_numrecs) - 1 >= XFS_INOBT_BLOCK_MINRECS(level, cur)) { if ((error = xfs_inobt_lshift(tcur, level, &i))) goto error0; if (i) { - ASSERT(INT_GET(block->bb_numrecs, ARCH_CONVERT) >= + ASSERT(be16_to_cpu(block->bb_numrecs) >= XFS_INOBT_BLOCK_MINRECS(level, cur)); xfs_btree_del_cursor(tcur, XFS_BTREE_NOERROR); @@ -321,7 +300,7 @@ xfs_inobt_delrec( * future reference, and fix up the temp cursor to point * to our block again (last record). */ - rrecs = INT_GET(right->bb_numrecs, ARCH_CONVERT); + rrecs = be16_to_cpu(right->bb_numrecs); if (lbno != NULLAGBLOCK) { xfs_btree_firstrec(tcur, level); if ((error = xfs_inobt_decrement(tcur, level, &i))) @@ -353,18 +332,18 @@ xfs_inobt_delrec( /* * Grab the current block number, for future use. */ - bno = INT_GET(left->bb_rightsib, ARCH_CONVERT); + bno = be32_to_cpu(left->bb_rightsib); /* * If left block is full enough so that removing one entry * won't make it too empty, and right-shifting an entry out * of left to us works, we're done. */ - if (INT_GET(left->bb_numrecs, ARCH_CONVERT) - 1 >= + if (be16_to_cpu(left->bb_numrecs) - 1 >= XFS_INOBT_BLOCK_MINRECS(level, cur)) { if ((error = xfs_inobt_rshift(tcur, level, &i))) goto error0; if (i) { - ASSERT(INT_GET(block->bb_numrecs, ARCH_CONVERT) >= + ASSERT(be16_to_cpu(block->bb_numrecs) >= XFS_INOBT_BLOCK_MINRECS(level, cur)); xfs_btree_del_cursor(tcur, XFS_BTREE_NOERROR); @@ -378,7 +357,7 @@ xfs_inobt_delrec( * Otherwise, grab the number of records in right for * future reference. */ - lrecs = INT_GET(left->bb_numrecs, ARCH_CONVERT); + lrecs = be16_to_cpu(left->bb_numrecs); } /* * Delete the temp cursor, we're done with it. @@ -399,14 +378,14 @@ xfs_inobt_delrec( */ rbno = bno; right = block; - rrecs = INT_GET(right->bb_numrecs, ARCH_CONVERT); + rrecs = be16_to_cpu(right->bb_numrecs); rbp = bp; if ((error = xfs_btree_read_bufs(mp, cur->bc_tp, cur->bc_private.i.agno, lbno, 0, &lbp, XFS_INO_BTREE_REF))) return error; left = XFS_BUF_TO_INOBT_BLOCK(lbp); - lrecs = INT_GET(left->bb_numrecs, ARCH_CONVERT); + lrecs = be16_to_cpu(left->bb_numrecs); if ((error = xfs_btree_check_sblock(cur, left, level, lbp))) return error; } @@ -421,14 +400,14 @@ xfs_inobt_delrec( */ lbno = bno; left = block; - lrecs = INT_GET(left->bb_numrecs, ARCH_CONVERT); + lrecs = be16_to_cpu(left->bb_numrecs); lbp = bp; if ((error = xfs_btree_read_bufs(mp, cur->bc_tp, cur->bc_private.i.agno, rbno, 0, &rbp, XFS_INO_BTREE_REF))) return error; right = XFS_BUF_TO_INOBT_BLOCK(rbp); - rrecs = INT_GET(right->bb_numrecs, ARCH_CONVERT); + rrecs = be16_to_cpu(right->bb_numrecs); if ((error = xfs_btree_check_sblock(cur, right, level, rbp))) return error; } @@ -456,7 +435,7 @@ xfs_inobt_delrec( rpp = XFS_INOBT_PTR_ADDR(right, 1, cur); #ifdef DEBUG for (i = 0; i < rrecs; i++) { - if ((error = xfs_btree_check_sptr(cur, INT_GET(rpp[i], ARCH_CONVERT), level))) + if ((error = xfs_btree_check_sptr(cur, be32_to_cpu(rpp[i]), level))) return error; } #endif @@ -492,7 +471,7 @@ xfs_inobt_delrec( * Fix up the number of records in the surviving block. */ lrecs += rrecs; - INT_SET(left->bb_numrecs, ARCH_CONVERT, lrecs); + left->bb_numrecs = cpu_to_be16(lrecs); /* * Fix up the right block pointer in the surviving block, and log it. */ @@ -502,18 +481,18 @@ xfs_inobt_delrec( * If there is a right sibling now, make it point to the * remaining block. */ - if (INT_GET(left->bb_rightsib, ARCH_CONVERT) != NULLAGBLOCK) { + if (be32_to_cpu(left->bb_rightsib) != NULLAGBLOCK) { xfs_inobt_block_t *rrblock; xfs_buf_t *rrbp; if ((error = xfs_btree_read_bufs(mp, cur->bc_tp, - cur->bc_private.i.agno, INT_GET(left->bb_rightsib, ARCH_CONVERT), 0, + cur->bc_private.i.agno, be32_to_cpu(left->bb_rightsib), 0, &rrbp, XFS_INO_BTREE_REF))) return error; rrblock = XFS_BUF_TO_INOBT_BLOCK(rrbp); if ((error = xfs_btree_check_sblock(cur, rrblock, level, rrbp))) return error; - INT_SET(rrblock->bb_leftsib, ARCH_CONVERT, lbno); + rrblock->bb_leftsib = cpu_to_be32(lbno); xfs_inobt_log_block(cur->bc_tp, rrbp, XFS_BB_LEFTSIB); } /* @@ -572,6 +551,13 @@ xfs_inobt_insrec( xfs_inobt_rec_t *rp=NULL; /* pointer to btree records */ /* + * GCC doesn't understand the (arguably complex) control flow in + * this function and complains about uninitialized structure fields + * without this. + */ + memset(&nrec, 0, sizeof(nrec)); + + /* * If we made it to the root level, allocate a new root block * and we're done. */ @@ -598,7 +584,7 @@ xfs_inobt_insrec( */ bp = cur->bc_bufs[level]; block = XFS_BUF_TO_INOBT_BLOCK(bp); - numrecs = INT_GET(block->bb_numrecs, ARCH_CONVERT); + numrecs = be16_to_cpu(block->bb_numrecs); #ifdef DEBUG if ((error = xfs_btree_check_sblock(cur, block, level, bp))) return error; @@ -672,7 +658,7 @@ xfs_inobt_insrec( * At this point we know there's room for our new entry in the block * we're pointing at. */ - numrecs = INT_GET(block->bb_numrecs, ARCH_CONVERT); + numrecs = be16_to_cpu(block->bb_numrecs); if (level > 0) { /* * It's a non-leaf entry. Make a hole for the new data @@ -682,7 +668,7 @@ xfs_inobt_insrec( pp = XFS_INOBT_PTR_ADDR(block, 1, cur); #ifdef DEBUG for (i = numrecs; i >= ptr; i--) { - if ((error = xfs_btree_check_sptr(cur, INT_GET(pp[i - 1], ARCH_CONVERT), level))) + if ((error = xfs_btree_check_sptr(cur, be32_to_cpu(pp[i - 1]), level))) return error; } #endif @@ -698,9 +684,9 @@ xfs_inobt_insrec( return error; #endif kp[ptr - 1] = key; /* INT_: struct copy */ - INT_SET(pp[ptr - 1], ARCH_CONVERT, *bnop); + pp[ptr - 1] = cpu_to_be32(*bnop); numrecs++; - INT_SET(block->bb_numrecs, ARCH_CONVERT, numrecs); + block->bb_numrecs = cpu_to_be16(numrecs); xfs_inobt_log_keys(cur, bp, ptr, numrecs); xfs_inobt_log_ptrs(cur, bp, ptr, numrecs); } else { @@ -716,7 +702,7 @@ xfs_inobt_insrec( */ rp[ptr - 1] = *recp; /* INT_: struct copy */ numrecs++; - INT_SET(block->bb_numrecs, ARCH_CONVERT, numrecs); + block->bb_numrecs = cpu_to_be16(numrecs); xfs_inobt_log_recs(cur, bp, ptr, numrecs); } /* @@ -871,8 +857,8 @@ xfs_inobt_lookup( xfs_agi_t *agi; /* a.g. inode header */ agi = XFS_BUF_TO_AGI(cur->bc_private.i.agbp); - agno = INT_GET(agi->agi_seqno, ARCH_CONVERT); - agbno = INT_GET(agi->agi_root, ARCH_CONVERT); + agno = be32_to_cpu(agi->agi_seqno); + agbno = be32_to_cpu(agi->agi_root); } /* * Iterate over each level in the btree, starting at the root. @@ -939,7 +925,7 @@ xfs_inobt_lookup( * Set low and high entry numbers, 1-based. */ low = 1; - if (!(high = INT_GET(block->bb_numrecs, ARCH_CONVERT))) { + if (!(high = be16_to_cpu(block->bb_numrecs))) { /* * If the block is empty, the tree must * be an empty leaf. @@ -1006,7 +992,7 @@ xfs_inobt_lookup( */ if (diff > 0 && --keyno < 1) keyno = 1; - agbno = INT_GET(*XFS_INOBT_PTR_ADDR(block, keyno, cur), ARCH_CONVERT); + agbno = be32_to_cpu(*XFS_INOBT_PTR_ADDR(block, keyno, cur)); #ifdef DEBUG if ((error = xfs_btree_check_sptr(cur, agbno, level))) return error; @@ -1025,8 +1011,8 @@ xfs_inobt_lookup( * not the last block, we're in the wrong block. */ if (dir == XFS_LOOKUP_GE && - keyno > INT_GET(block->bb_numrecs, ARCH_CONVERT) && - INT_GET(block->bb_rightsib, ARCH_CONVERT) != NULLAGBLOCK) { + keyno > be16_to_cpu(block->bb_numrecs) && + be32_to_cpu(block->bb_rightsib) != NULLAGBLOCK) { int i; cur->bc_ptrs[0] = keyno; @@ -1043,7 +1029,7 @@ xfs_inobt_lookup( /* * Return if we succeeded or not. */ - if (keyno == 0 || keyno > INT_GET(block->bb_numrecs, ARCH_CONVERT)) + if (keyno == 0 || keyno > be16_to_cpu(block->bb_numrecs)) *stat = 0; else *stat = ((dir != XFS_LOOKUP_EQ) || (diff == 0)); @@ -1089,7 +1075,7 @@ xfs_inobt_lshift( /* * If we've got no left sibling then we can't shift an entry left. */ - if (INT_GET(right->bb_leftsib, ARCH_CONVERT) == NULLAGBLOCK) { + if (be32_to_cpu(right->bb_leftsib) == NULLAGBLOCK) { *stat = 0; return 0; } @@ -1105,8 +1091,8 @@ xfs_inobt_lshift( * Set up the left neighbor as "left". */ if ((error = xfs_btree_read_bufs(cur->bc_mp, cur->bc_tp, - cur->bc_private.i.agno, INT_GET(right->bb_leftsib, ARCH_CONVERT), 0, &lbp, - XFS_INO_BTREE_REF))) + cur->bc_private.i.agno, be32_to_cpu(right->bb_leftsib), + 0, &lbp, XFS_INO_BTREE_REF))) return error; left = XFS_BUF_TO_INOBT_BLOCK(lbp); if ((error = xfs_btree_check_sblock(cur, left, level, lbp))) @@ -1114,11 +1100,11 @@ xfs_inobt_lshift( /* * If it's full, it can't take another entry. */ - if (INT_GET(left->bb_numrecs, ARCH_CONVERT) == XFS_INOBT_BLOCK_MAXRECS(level, cur)) { + if (be16_to_cpu(left->bb_numrecs) == XFS_INOBT_BLOCK_MAXRECS(level, cur)) { *stat = 0; return 0; } - nrec = INT_GET(left->bb_numrecs, ARCH_CONVERT) + 1; + nrec = be16_to_cpu(left->bb_numrecs) + 1; /* * If non-leaf, copy a key and a ptr to the left block. */ @@ -1130,7 +1116,7 @@ xfs_inobt_lshift( lpp = XFS_INOBT_PTR_ADDR(left, nrec, cur); rpp = XFS_INOBT_PTR_ADDR(right, 1, cur); #ifdef DEBUG - if ((error = xfs_btree_check_sptr(cur, INT_GET(*rpp, ARCH_CONVERT), level))) + if ((error = xfs_btree_check_sptr(cur, be32_to_cpu(*rpp), level))) return error; #endif *lpp = *rpp; /* INT_: no-change copy */ @@ -1148,7 +1134,7 @@ xfs_inobt_lshift( /* * Bump and log left's numrecs, decrement and log right's numrecs. */ - INT_MOD(left->bb_numrecs, ARCH_CONVERT, +1); + be16_add(&left->bb_numrecs, 1); xfs_inobt_log_block(cur->bc_tp, lbp, XFS_BB_NUMRECS); #ifdef DEBUG if (level > 0) @@ -1156,26 +1142,26 @@ xfs_inobt_lshift( else xfs_btree_check_rec(cur->bc_btnum, lrp - 1, lrp); #endif - INT_MOD(right->bb_numrecs, ARCH_CONVERT, -1); + be16_add(&right->bb_numrecs, -1); xfs_inobt_log_block(cur->bc_tp, rbp, XFS_BB_NUMRECS); /* * Slide the contents of right down one entry. */ if (level > 0) { #ifdef DEBUG - for (i = 0; i < INT_GET(right->bb_numrecs, ARCH_CONVERT); i++) { - if ((error = xfs_btree_check_sptr(cur, INT_GET(rpp[i + 1], ARCH_CONVERT), + for (i = 0; i < be16_to_cpu(right->bb_numrecs); i++) { + if ((error = xfs_btree_check_sptr(cur, be32_to_cpu(rpp[i + 1]), level))) return error; } #endif - memmove(rkp, rkp + 1, INT_GET(right->bb_numrecs, ARCH_CONVERT) * sizeof(*rkp)); - memmove(rpp, rpp + 1, INT_GET(right->bb_numrecs, ARCH_CONVERT) * sizeof(*rpp)); - xfs_inobt_log_keys(cur, rbp, 1, INT_GET(right->bb_numrecs, ARCH_CONVERT)); - xfs_inobt_log_ptrs(cur, rbp, 1, INT_GET(right->bb_numrecs, ARCH_CONVERT)); + memmove(rkp, rkp + 1, be16_to_cpu(right->bb_numrecs) * sizeof(*rkp)); + memmove(rpp, rpp + 1, be16_to_cpu(right->bb_numrecs) * sizeof(*rpp)); + xfs_inobt_log_keys(cur, rbp, 1, be16_to_cpu(right->bb_numrecs)); + xfs_inobt_log_ptrs(cur, rbp, 1, be16_to_cpu(right->bb_numrecs)); } else { - memmove(rrp, rrp + 1, INT_GET(right->bb_numrecs, ARCH_CONVERT) * sizeof(*rrp)); - xfs_inobt_log_recs(cur, rbp, 1, INT_GET(right->bb_numrecs, ARCH_CONVERT)); + memmove(rrp, rrp + 1, be16_to_cpu(right->bb_numrecs) * sizeof(*rrp)); + xfs_inobt_log_recs(cur, rbp, 1, be16_to_cpu(right->bb_numrecs)); key.ir_startino = rrp->ir_startino; /* INT_: direct copy */ rkp = &key; } @@ -1227,7 +1213,7 @@ xfs_inobt_newroot( args.tp = cur->bc_tp; args.mp = cur->bc_mp; args.fsbno = XFS_AGB_TO_FSB(args.mp, cur->bc_private.i.agno, - INT_GET(agi->agi_root, ARCH_CONVERT)); + be32_to_cpu(agi->agi_root)); args.mod = args.minleft = args.alignment = args.total = args.wasdel = args.isfl = args.userdata = args.minalignslop = 0; args.minlen = args.maxlen = args.prod = 1; @@ -1247,8 +1233,8 @@ xfs_inobt_newroot( /* * Set the root data in the a.g. inode structure. */ - INT_SET(agi->agi_root, ARCH_CONVERT, args.agbno); - INT_MOD(agi->agi_level, ARCH_CONVERT, 1); + agi->agi_root = cpu_to_be32(args.agbno); + be32_add(&agi->agi_level, 1); xfs_ialloc_log_agi(args.tp, cur->bc_private.i.agbp, XFS_AGI_ROOT | XFS_AGI_LEVEL); /* @@ -1263,14 +1249,14 @@ xfs_inobt_newroot( if ((error = xfs_btree_check_sblock(cur, block, cur->bc_nlevels - 1, bp))) return error; #endif - if (INT_GET(block->bb_rightsib, ARCH_CONVERT) != NULLAGBLOCK) { + if (be32_to_cpu(block->bb_rightsib) != NULLAGBLOCK) { /* * Our block is left, pick up the right block. */ lbp = bp; lbno = XFS_DADDR_TO_AGBNO(args.mp, XFS_BUF_ADDR(lbp)); left = block; - rbno = INT_GET(left->bb_rightsib, ARCH_CONVERT); + rbno = be32_to_cpu(left->bb_rightsib); if ((error = xfs_btree_read_bufs(args.mp, args.tp, args.agno, rbno, 0, &rbp, XFS_INO_BTREE_REF))) return error; @@ -1287,7 +1273,7 @@ xfs_inobt_newroot( rbp = bp; rbno = XFS_DADDR_TO_AGBNO(args.mp, XFS_BUF_ADDR(rbp)); right = block; - lbno = INT_GET(right->bb_leftsib, ARCH_CONVERT); + lbno = be32_to_cpu(right->bb_leftsib); if ((error = xfs_btree_read_bufs(args.mp, args.tp, args.agno, lbno, 0, &lbp, XFS_INO_BTREE_REF))) return error; @@ -1301,18 +1287,18 @@ xfs_inobt_newroot( /* * Fill in the new block's btree header and log it. */ - INT_SET(new->bb_magic, ARCH_CONVERT, xfs_magics[cur->bc_btnum]); - INT_SET(new->bb_level, ARCH_CONVERT, (__uint16_t)cur->bc_nlevels); - INT_SET(new->bb_numrecs, ARCH_CONVERT, 2); - INT_SET(new->bb_leftsib, ARCH_CONVERT, NULLAGBLOCK); - INT_SET(new->bb_rightsib, ARCH_CONVERT, NULLAGBLOCK); + new->bb_magic = cpu_to_be32(xfs_magics[cur->bc_btnum]); + new->bb_level = cpu_to_be16(cur->bc_nlevels); + new->bb_numrecs = cpu_to_be16(2); + new->bb_leftsib = cpu_to_be32(NULLAGBLOCK); + new->bb_rightsib = cpu_to_be32(NULLAGBLOCK); xfs_inobt_log_block(args.tp, nbp, XFS_BB_ALL_BITS); ASSERT(lbno != NULLAGBLOCK && rbno != NULLAGBLOCK); /* * Fill in the key data in the new root. */ kp = XFS_INOBT_KEY_ADDR(new, 1, cur); - if (INT_GET(left->bb_level, ARCH_CONVERT) > 0) { + if (be16_to_cpu(left->bb_level) > 0) { kp[0] = *XFS_INOBT_KEY_ADDR(left, 1, cur); /* INT_: struct copy */ kp[1] = *XFS_INOBT_KEY_ADDR(right, 1, cur); /* INT_: struct copy */ } else { @@ -1326,8 +1312,8 @@ xfs_inobt_newroot( * Fill in the pointer data in the new root. */ pp = XFS_INOBT_PTR_ADDR(new, 1, cur); - INT_SET(pp[0], ARCH_CONVERT, lbno); - INT_SET(pp[1], ARCH_CONVERT, rbno); + pp[0] = cpu_to_be32(lbno); + pp[1] = cpu_to_be32(rbno); xfs_inobt_log_ptrs(cur, nbp, 1, 2); /* * Fix up the cursor. @@ -1376,7 +1362,7 @@ xfs_inobt_rshift( /* * If we've got no right sibling then we can't shift an entry right. */ - if (INT_GET(left->bb_rightsib, ARCH_CONVERT) == NULLAGBLOCK) { + if (be32_to_cpu(left->bb_rightsib) == NULLAGBLOCK) { *stat = 0; return 0; } @@ -1384,7 +1370,7 @@ xfs_inobt_rshift( * If the cursor entry is the one that would be moved, don't * do it... it's too complicated. */ - if (cur->bc_ptrs[level] >= INT_GET(left->bb_numrecs, ARCH_CONVERT)) { + if (cur->bc_ptrs[level] >= be16_to_cpu(left->bb_numrecs)) { *stat = 0; return 0; } @@ -1392,8 +1378,8 @@ xfs_inobt_rshift( * Set up the right neighbor as "right". */ if ((error = xfs_btree_read_bufs(cur->bc_mp, cur->bc_tp, - cur->bc_private.i.agno, INT_GET(left->bb_rightsib, ARCH_CONVERT), 0, &rbp, - XFS_INO_BTREE_REF))) + cur->bc_private.i.agno, be32_to_cpu(left->bb_rightsib), + 0, &rbp, XFS_INO_BTREE_REF))) return error; right = XFS_BUF_TO_INOBT_BLOCK(rbp); if ((error = xfs_btree_check_sblock(cur, right, level, rbp))) @@ -1401,7 +1387,7 @@ xfs_inobt_rshift( /* * If it's full, it can't take another entry. */ - if (INT_GET(right->bb_numrecs, ARCH_CONVERT) == XFS_INOBT_BLOCK_MAXRECS(level, cur)) { + if (be16_to_cpu(right->bb_numrecs) == XFS_INOBT_BLOCK_MAXRECS(level, cur)) { *stat = 0; return 0; } @@ -1410,41 +1396,41 @@ xfs_inobt_rshift( * copy the last left block entry to the hole. */ if (level > 0) { - lkp = XFS_INOBT_KEY_ADDR(left, INT_GET(left->bb_numrecs, ARCH_CONVERT), cur); - lpp = XFS_INOBT_PTR_ADDR(left, INT_GET(left->bb_numrecs, ARCH_CONVERT), cur); + lkp = XFS_INOBT_KEY_ADDR(left, be16_to_cpu(left->bb_numrecs), cur); + lpp = XFS_INOBT_PTR_ADDR(left, be16_to_cpu(left->bb_numrecs), cur); rkp = XFS_INOBT_KEY_ADDR(right, 1, cur); rpp = XFS_INOBT_PTR_ADDR(right, 1, cur); #ifdef DEBUG - for (i = INT_GET(right->bb_numrecs, ARCH_CONVERT) - 1; i >= 0; i--) { - if ((error = xfs_btree_check_sptr(cur, INT_GET(rpp[i], ARCH_CONVERT), level))) + for (i = be16_to_cpu(right->bb_numrecs) - 1; i >= 0; i--) { + if ((error = xfs_btree_check_sptr(cur, be32_to_cpu(rpp[i]), level))) return error; } #endif - memmove(rkp + 1, rkp, INT_GET(right->bb_numrecs, ARCH_CONVERT) * sizeof(*rkp)); - memmove(rpp + 1, rpp, INT_GET(right->bb_numrecs, ARCH_CONVERT) * sizeof(*rpp)); + memmove(rkp + 1, rkp, be16_to_cpu(right->bb_numrecs) * sizeof(*rkp)); + memmove(rpp + 1, rpp, be16_to_cpu(right->bb_numrecs) * sizeof(*rpp)); #ifdef DEBUG - if ((error = xfs_btree_check_sptr(cur, INT_GET(*lpp, ARCH_CONVERT), level))) + if ((error = xfs_btree_check_sptr(cur, be32_to_cpu(*lpp), level))) return error; #endif *rkp = *lkp; /* INT_: no change copy */ *rpp = *lpp; /* INT_: no change copy */ - xfs_inobt_log_keys(cur, rbp, 1, INT_GET(right->bb_numrecs, ARCH_CONVERT) + 1); - xfs_inobt_log_ptrs(cur, rbp, 1, INT_GET(right->bb_numrecs, ARCH_CONVERT) + 1); + xfs_inobt_log_keys(cur, rbp, 1, be16_to_cpu(right->bb_numrecs) + 1); + xfs_inobt_log_ptrs(cur, rbp, 1, be16_to_cpu(right->bb_numrecs) + 1); } else { - lrp = XFS_INOBT_REC_ADDR(left, INT_GET(left->bb_numrecs, ARCH_CONVERT), cur); + lrp = XFS_INOBT_REC_ADDR(left, be16_to_cpu(left->bb_numrecs), cur); rrp = XFS_INOBT_REC_ADDR(right, 1, cur); - memmove(rrp + 1, rrp, INT_GET(right->bb_numrecs, ARCH_CONVERT) * sizeof(*rrp)); + memmove(rrp + 1, rrp, be16_to_cpu(right->bb_numrecs) * sizeof(*rrp)); *rrp = *lrp; - xfs_inobt_log_recs(cur, rbp, 1, INT_GET(right->bb_numrecs, ARCH_CONVERT) + 1); + xfs_inobt_log_recs(cur, rbp, 1, be16_to_cpu(right->bb_numrecs) + 1); key.ir_startino = rrp->ir_startino; /* INT_: direct copy */ rkp = &key; } /* * Decrement and log left's numrecs, bump and log right's numrecs. */ - INT_MOD(left->bb_numrecs, ARCH_CONVERT, -1); + be16_add(&left->bb_numrecs, -1); xfs_inobt_log_block(cur->bc_tp, lbp, XFS_BB_NUMRECS); - INT_MOD(right->bb_numrecs, ARCH_CONVERT, +1); + be16_add(&right->bb_numrecs, 1); #ifdef DEBUG if (level > 0) xfs_btree_check_key(cur->bc_btnum, rkp, rkp + 1); @@ -1536,17 +1522,17 @@ xfs_inobt_split( /* * Fill in the btree header for the new block. */ - INT_SET(right->bb_magic, ARCH_CONVERT, xfs_magics[cur->bc_btnum]); - right->bb_level = left->bb_level; /* INT_: direct copy */ - INT_SET(right->bb_numrecs, ARCH_CONVERT, (__uint16_t)(INT_GET(left->bb_numrecs, ARCH_CONVERT) / 2)); + right->bb_magic = cpu_to_be32(xfs_magics[cur->bc_btnum]); + right->bb_level = left->bb_level; + right->bb_numrecs = cpu_to_be16(be16_to_cpu(left->bb_numrecs) / 2); /* * Make sure that if there's an odd number of entries now, that * each new block will have the same number of entries. */ - if ((INT_GET(left->bb_numrecs, ARCH_CONVERT) & 1) && - cur->bc_ptrs[level] <= INT_GET(right->bb_numrecs, ARCH_CONVERT) + 1) - INT_MOD(right->bb_numrecs, ARCH_CONVERT, +1); - i = INT_GET(left->bb_numrecs, ARCH_CONVERT) - INT_GET(right->bb_numrecs, ARCH_CONVERT) + 1; + if ((be16_to_cpu(left->bb_numrecs) & 1) && + cur->bc_ptrs[level] <= be16_to_cpu(right->bb_numrecs) + 1) + be16_add(&right->bb_numrecs, 1); + i = be16_to_cpu(left->bb_numrecs) - be16_to_cpu(right->bb_numrecs) + 1; /* * For non-leaf blocks, copy keys and addresses over to the new block. */ @@ -1556,15 +1542,15 @@ xfs_inobt_split( rkp = XFS_INOBT_KEY_ADDR(right, 1, cur); rpp = XFS_INOBT_PTR_ADDR(right, 1, cur); #ifdef DEBUG - for (i = 0; i < INT_GET(right->bb_numrecs, ARCH_CONVERT); i++) { - if ((error = xfs_btree_check_sptr(cur, INT_GET(lpp[i], ARCH_CONVERT), level))) + for (i = 0; i < be16_to_cpu(right->bb_numrecs); i++) { + if ((error = xfs_btree_check_sptr(cur, be32_to_cpu(lpp[i]), level))) return error; } #endif - memcpy(rkp, lkp, INT_GET(right->bb_numrecs, ARCH_CONVERT) * sizeof(*rkp)); - memcpy(rpp, lpp, INT_GET(right->bb_numrecs, ARCH_CONVERT) * sizeof(*rpp)); - xfs_inobt_log_keys(cur, rbp, 1, INT_GET(right->bb_numrecs, ARCH_CONVERT)); - xfs_inobt_log_ptrs(cur, rbp, 1, INT_GET(right->bb_numrecs, ARCH_CONVERT)); + memcpy(rkp, lkp, be16_to_cpu(right->bb_numrecs) * sizeof(*rkp)); + memcpy(rpp, lpp, be16_to_cpu(right->bb_numrecs) * sizeof(*rpp)); + xfs_inobt_log_keys(cur, rbp, 1, be16_to_cpu(right->bb_numrecs)); + xfs_inobt_log_ptrs(cur, rbp, 1, be16_to_cpu(right->bb_numrecs)); *keyp = *rkp; } /* @@ -1573,36 +1559,36 @@ xfs_inobt_split( else { lrp = XFS_INOBT_REC_ADDR(left, i, cur); rrp = XFS_INOBT_REC_ADDR(right, 1, cur); - memcpy(rrp, lrp, INT_GET(right->bb_numrecs, ARCH_CONVERT) * sizeof(*rrp)); - xfs_inobt_log_recs(cur, rbp, 1, INT_GET(right->bb_numrecs, ARCH_CONVERT)); + memcpy(rrp, lrp, be16_to_cpu(right->bb_numrecs) * sizeof(*rrp)); + xfs_inobt_log_recs(cur, rbp, 1, be16_to_cpu(right->bb_numrecs)); keyp->ir_startino = rrp->ir_startino; /* INT_: direct copy */ } /* * Find the left block number by looking in the buffer. * Adjust numrecs, sibling pointers. */ - INT_MOD(left->bb_numrecs, ARCH_CONVERT, -(INT_GET(right->bb_numrecs, ARCH_CONVERT))); - right->bb_rightsib = left->bb_rightsib; /* INT_: direct copy */ - INT_SET(left->bb_rightsib, ARCH_CONVERT, args.agbno); - INT_SET(right->bb_leftsib, ARCH_CONVERT, lbno); + be16_add(&left->bb_numrecs, -(be16_to_cpu(right->bb_numrecs))); + right->bb_rightsib = left->bb_rightsib; + left->bb_rightsib = cpu_to_be32(args.agbno); + right->bb_leftsib = cpu_to_be32(lbno); xfs_inobt_log_block(args.tp, rbp, XFS_BB_ALL_BITS); xfs_inobt_log_block(args.tp, lbp, XFS_BB_NUMRECS | XFS_BB_RIGHTSIB); /* * If there's a block to the new block's right, make that block * point back to right instead of to left. */ - if (INT_GET(right->bb_rightsib, ARCH_CONVERT) != NULLAGBLOCK) { + if (be32_to_cpu(right->bb_rightsib) != NULLAGBLOCK) { xfs_inobt_block_t *rrblock; /* rr btree block */ xfs_buf_t *rrbp; /* buffer for rrblock */ if ((error = xfs_btree_read_bufs(args.mp, args.tp, args.agno, - INT_GET(right->bb_rightsib, ARCH_CONVERT), 0, &rrbp, + be32_to_cpu(right->bb_rightsib), 0, &rrbp, XFS_INO_BTREE_REF))) return error; rrblock = XFS_BUF_TO_INOBT_BLOCK(rrbp); if ((error = xfs_btree_check_sblock(cur, rrblock, level, rrbp))) return error; - INT_SET(rrblock->bb_leftsib, ARCH_CONVERT, args.agbno); + rrblock->bb_leftsib = cpu_to_be32(args.agbno); xfs_inobt_log_block(args.tp, rrbp, XFS_BB_LEFTSIB); } /* @@ -1610,9 +1596,9 @@ xfs_inobt_split( * If it's just pointing past the last entry in left, then we'll * insert there, so don't change anything in that case. */ - if (cur->bc_ptrs[level] > INT_GET(left->bb_numrecs, ARCH_CONVERT) + 1) { + if (cur->bc_ptrs[level] > be16_to_cpu(left->bb_numrecs) + 1) { xfs_btree_setbuf(cur, level, rbp); - cur->bc_ptrs[level] -= INT_GET(left->bb_numrecs, ARCH_CONVERT); + cur->bc_ptrs[level] -= be16_to_cpu(left->bb_numrecs); } /* * If there are more levels, we'll need another cursor which refers @@ -1710,7 +1696,7 @@ xfs_inobt_decrement( /* * If we just went off the left edge of the tree, return failure. */ - if (INT_GET(block->bb_leftsib, ARCH_CONVERT) == NULLAGBLOCK) { + if (be32_to_cpu(block->bb_leftsib) == NULLAGBLOCK) { *stat = 0; return 0; } @@ -1739,7 +1725,7 @@ xfs_inobt_decrement( xfs_agblock_t agbno; /* block number of btree block */ xfs_buf_t *bp; /* buffer containing btree block */ - agbno = INT_GET(*XFS_INOBT_PTR_ADDR(block, cur->bc_ptrs[lev], cur), ARCH_CONVERT); + agbno = be32_to_cpu(*XFS_INOBT_PTR_ADDR(block, cur->bc_ptrs[lev], cur)); if ((error = xfs_btree_read_bufs(cur->bc_mp, cur->bc_tp, cur->bc_private.i.agno, agbno, 0, &bp, XFS_INO_BTREE_REF))) @@ -1749,7 +1735,7 @@ xfs_inobt_decrement( block = XFS_BUF_TO_INOBT_BLOCK(bp); if ((error = xfs_btree_check_sblock(cur, block, lev, bp))) return error; - cur->bc_ptrs[lev] = INT_GET(block->bb_numrecs, ARCH_CONVERT); + cur->bc_ptrs[lev] = be16_to_cpu(block->bb_numrecs); } *stat = 1; return 0; @@ -1821,7 +1807,7 @@ xfs_inobt_get_rec( /* * Off the right end or left end, return failure. */ - if (ptr > INT_GET(block->bb_numrecs, ARCH_CONVERT) || ptr <= 0) { + if (ptr > be16_to_cpu(block->bb_numrecs) || ptr <= 0) { *stat = 0; return 0; } @@ -1869,14 +1855,14 @@ xfs_inobt_increment( * Increment the ptr at this level. If we're still in the block * then we're done. */ - if (++cur->bc_ptrs[level] <= INT_GET(block->bb_numrecs, ARCH_CONVERT)) { + if (++cur->bc_ptrs[level] <= be16_to_cpu(block->bb_numrecs)) { *stat = 1; return 0; } /* * If we just went off the right edge of the tree, return failure. */ - if (INT_GET(block->bb_rightsib, ARCH_CONVERT) == NULLAGBLOCK) { + if (be32_to_cpu(block->bb_rightsib) == NULLAGBLOCK) { *stat = 0; return 0; } @@ -1891,7 +1877,7 @@ xfs_inobt_increment( if ((error = xfs_btree_check_sblock(cur, block, lev, bp))) return error; #endif - if (++cur->bc_ptrs[lev] <= INT_GET(block->bb_numrecs, ARCH_CONVERT)) + if (++cur->bc_ptrs[lev] <= be16_to_cpu(block->bb_numrecs)) break; /* * Read-ahead the right block, we're going to read it @@ -1911,7 +1897,7 @@ xfs_inobt_increment( lev > level; ) { xfs_agblock_t agbno; /* block number of btree block */ - agbno = INT_GET(*XFS_INOBT_PTR_ADDR(block, cur->bc_ptrs[lev], cur), ARCH_CONVERT); + agbno = be32_to_cpu(*XFS_INOBT_PTR_ADDR(block, cur->bc_ptrs[lev], cur)); if ((error = xfs_btree_read_bufs(cur->bc_mp, cur->bc_tp, cur->bc_private.i.agno, agbno, 0, &bp, XFS_INO_BTREE_REF))) diff --git a/fs/xfs/xfs_ialloc_btree.h b/fs/xfs/xfs_ialloc_btree.h index 44be188674a6..ae3904cb1ee8 100644 --- a/fs/xfs/xfs_ialloc_btree.h +++ b/fs/xfs/xfs_ialloc_btree.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_IALLOC_BTREE_H__ #define __XFS_IALLOC_BTREE_H__ @@ -51,14 +37,12 @@ typedef __uint64_t xfs_inofree_t; #define XFS_INODES_PER_CHUNK_LOG (XFS_NBBYLOG + 3) #define XFS_INOBT_ALL_FREE ((xfs_inofree_t)-1) -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_INOBT_MASKN) -xfs_inofree_t xfs_inobt_maskn(int i, int n); #define XFS_INOBT_MASKN(i,n) xfs_inobt_maskn(i,n) -#else -#define XFS_INOBT_MASKN(i,n) \ - ((((n) >= XFS_INODES_PER_CHUNK ? \ - (xfs_inofree_t)0 : ((xfs_inofree_t)1 << (n))) - 1) << (i)) -#endif +static inline xfs_inofree_t xfs_inobt_maskn(int i, int n) +{ + return (((n) >= XFS_INODES_PER_CHUNK ? \ + (xfs_inofree_t)0 : ((xfs_inofree_t)1 << (n))) - 1) << (i); +} /* * Data record structure @@ -78,241 +62,116 @@ typedef struct xfs_inobt_key xfs_agino_t ir_startino; /* starting inode number */ } xfs_inobt_key_t; -typedef xfs_agblock_t xfs_inobt_ptr_t; /* btree pointer type */ - /* btree block header type */ +/* btree pointer type */ +typedef __be32 xfs_inobt_ptr_t; + +/* btree block header type */ typedef struct xfs_btree_sblock xfs_inobt_block_t; -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BUF_TO_INOBT_BLOCK) -xfs_inobt_block_t *xfs_buf_to_inobt_block(struct xfs_buf *bp); -#define XFS_BUF_TO_INOBT_BLOCK(bp) xfs_buf_to_inobt_block(bp) -#else -#define XFS_BUF_TO_INOBT_BLOCK(bp) ((xfs_inobt_block_t *)(XFS_BUF_PTR(bp))) -#endif +#define XFS_BUF_TO_INOBT_BLOCK(bp) ((xfs_inobt_block_t *)XFS_BUF_PTR(bp)) /* * Bit manipulations for ir_free. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_INOBT_MASK) -xfs_inofree_t xfs_inobt_mask(int i); -#define XFS_INOBT_MASK(i) xfs_inobt_mask(i) -#else #define XFS_INOBT_MASK(i) ((xfs_inofree_t)1 << (i)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_INOBT_IS_FREE) -int xfs_inobt_is_free(xfs_inobt_rec_t *rp, int i); -#define XFS_INOBT_IS_FREE(rp,i) xfs_inobt_is_free(rp,i) -#define XFS_INOBT_IS_FREE_DISK(rp,i) xfs_inobt_is_free_disk(rp,i) -#else -#define XFS_INOBT_IS_FREE(rp,i) \ - (((rp)->ir_free & XFS_INOBT_MASK(i)) != 0) -#define XFS_INOBT_IS_FREE_DISK(rp,i) \ - ((INT_GET((rp)->ir_free, ARCH_CONVERT) & XFS_INOBT_MASK(i)) != 0) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_INOBT_SET_FREE) -void xfs_inobt_set_free(xfs_inobt_rec_t *rp, int i); -#define XFS_INOBT_SET_FREE(rp,i) xfs_inobt_set_free(rp,i) -#else +#define XFS_INOBT_IS_FREE(rp,i) \ + (((rp)->ir_free & XFS_INOBT_MASK(i)) != 0) +#define XFS_INOBT_IS_FREE_DISK(rp,i) \ + ((INT_GET((rp)->ir_free,ARCH_CONVERT) & XFS_INOBT_MASK(i)) != 0) #define XFS_INOBT_SET_FREE(rp,i) ((rp)->ir_free |= XFS_INOBT_MASK(i)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_INOBT_CLR_FREE) -void xfs_inobt_clr_free(xfs_inobt_rec_t *rp, int i); -#define XFS_INOBT_CLR_FREE(rp,i) xfs_inobt_clr_free(rp,i) -#else #define XFS_INOBT_CLR_FREE(rp,i) ((rp)->ir_free &= ~XFS_INOBT_MASK(i)) -#endif /* * Real block structures have a size equal to the disk block size. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_INOBT_BLOCK_SIZE) -int xfs_inobt_block_size(int lev, struct xfs_btree_cur *cur); -#define XFS_INOBT_BLOCK_SIZE(lev,cur) xfs_inobt_block_size(lev,cur) -#else #define XFS_INOBT_BLOCK_SIZE(lev,cur) (1 << (cur)->bc_blocklog) -#endif - -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_INOBT_BLOCK_MAXRECS) -int xfs_inobt_block_maxrecs(int lev, struct xfs_btree_cur *cur); -#define XFS_INOBT_BLOCK_MAXRECS(lev,cur) xfs_inobt_block_maxrecs(lev,cur) -#else -#define XFS_INOBT_BLOCK_MAXRECS(lev,cur) \ - ((cur)->bc_mp->m_inobt_mxr[lev != 0]) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_INOBT_BLOCK_MINRECS) -int xfs_inobt_block_minrecs(int lev, struct xfs_btree_cur *cur); -#define XFS_INOBT_BLOCK_MINRECS(lev,cur) xfs_inobt_block_minrecs(lev,cur) -#else -#define XFS_INOBT_BLOCK_MINRECS(lev,cur) \ - ((cur)->bc_mp->m_inobt_mnr[lev != 0]) -#endif - -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_INOBT_IS_LAST_REC) -int xfs_inobt_is_last_rec(struct xfs_btree_cur *cur); -#define XFS_INOBT_IS_LAST_REC(cur) xfs_inobt_is_last_rec(cur) -#else +#define XFS_INOBT_BLOCK_MAXRECS(lev,cur) ((cur)->bc_mp->m_inobt_mxr[lev != 0]) +#define XFS_INOBT_BLOCK_MINRECS(lev,cur) ((cur)->bc_mp->m_inobt_mnr[lev != 0]) #define XFS_INOBT_IS_LAST_REC(cur) \ - ((cur)->bc_ptrs[0] == \ - INT_GET(XFS_BUF_TO_INOBT_BLOCK((cur)->bc_bufs[0])->bb_numrecs, ARCH_CONVERT)) -#endif + ((cur)->bc_ptrs[0] == be16_to_cpu(XFS_BUF_TO_INOBT_BLOCK((cur)->bc_bufs[0])->bb_numrecs)) /* * Maximum number of inode btree levels. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_IN_MAXLEVELS) -int xfs_in_maxlevels(struct xfs_mount *mp); -#define XFS_IN_MAXLEVELS(mp) xfs_in_maxlevels(mp) -#else #define XFS_IN_MAXLEVELS(mp) ((mp)->m_in_maxlevels) -#endif /* * block numbers in the AG. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_IBT_BLOCK) -xfs_agblock_t xfs_ibt_block(struct xfs_mount *mp); -#define XFS_IBT_BLOCK(mp) xfs_ibt_block(mp) -#else -#define XFS_IBT_BLOCK(mp) ((xfs_agblock_t)(XFS_CNT_BLOCK(mp) + 1)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_PREALLOC_BLOCKS) -xfs_agblock_t xfs_prealloc_blocks(struct xfs_mount *mp); -#define XFS_PREALLOC_BLOCKS(mp) xfs_prealloc_blocks(mp) -#else -#define XFS_PREALLOC_BLOCKS(mp) ((xfs_agblock_t)(XFS_IBT_BLOCK(mp) + 1)) -#endif +#define XFS_IBT_BLOCK(mp) ((xfs_agblock_t)(XFS_CNT_BLOCK(mp) + 1)) +#define XFS_PREALLOC_BLOCKS(mp) ((xfs_agblock_t)(XFS_IBT_BLOCK(mp) + 1)) /* * Record, key, and pointer address macros for btree blocks. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_INOBT_REC_ADDR) -xfs_inobt_rec_t * -xfs_inobt_rec_addr(xfs_inobt_block_t *bb, int i, struct xfs_btree_cur *cur); -#define XFS_INOBT_REC_ADDR(bb,i,cur) xfs_inobt_rec_addr(bb,i,cur) -#else -#define XFS_INOBT_REC_ADDR(bb,i,cur) \ - XFS_BTREE_REC_ADDR(XFS_INOBT_BLOCK_SIZE(0,cur), xfs_inobt, bb, i, \ - XFS_INOBT_BLOCK_MAXRECS(0, cur)) -#endif - -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_INOBT_KEY_ADDR) -xfs_inobt_key_t * -xfs_inobt_key_addr(xfs_inobt_block_t *bb, int i, struct xfs_btree_cur *cur); -#define XFS_INOBT_KEY_ADDR(bb,i,cur) xfs_inobt_key_addr(bb,i,cur) -#else -#define XFS_INOBT_KEY_ADDR(bb,i,cur) \ - XFS_BTREE_KEY_ADDR(XFS_INOBT_BLOCK_SIZE(1,cur), xfs_inobt, bb, i, \ - XFS_INOBT_BLOCK_MAXRECS(1, cur)) -#endif +#define XFS_INOBT_REC_ADDR(bb,i,cur) \ + (XFS_BTREE_REC_ADDR(XFS_INOBT_BLOCK_SIZE(0,cur), xfs_inobt, bb, \ + i, XFS_INOBT_BLOCK_MAXRECS(0, cur))) +#define XFS_INOBT_KEY_ADDR(bb,i,cur) \ + (XFS_BTREE_KEY_ADDR(XFS_INOBT_BLOCK_SIZE(1,cur), xfs_inobt, bb, \ + i, XFS_INOBT_BLOCK_MAXRECS(1, cur))) -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_INOBT_PTR_ADDR) -xfs_inobt_ptr_t * -xfs_inobt_ptr_addr(xfs_inobt_block_t *bb, int i, struct xfs_btree_cur *cur); -#define XFS_INOBT_PTR_ADDR(bb,i,cur) xfs_inobt_ptr_addr(bb,i,cur) -#else -#define XFS_INOBT_PTR_ADDR(bb,i,cur) \ - XFS_BTREE_PTR_ADDR(XFS_INOBT_BLOCK_SIZE(1,cur), xfs_inobt, bb, i, \ - XFS_INOBT_BLOCK_MAXRECS(1, cur)) -#endif - -/* - * Prototypes for externally visible routines. - */ +#define XFS_INOBT_PTR_ADDR(bb,i,cur) \ + (XFS_BTREE_PTR_ADDR(XFS_INOBT_BLOCK_SIZE(1,cur), xfs_inobt, bb, \ + i, XFS_INOBT_BLOCK_MAXRECS(1, cur))) /* * Decrement cursor by one record at the level. * For nonzero levels the leaf-ward information is untouched. */ -int /* error */ -xfs_inobt_decrement( - struct xfs_btree_cur *cur, /* btree cursor */ - int level, /* level in btree, 0 is leaf */ - int *stat); /* success/failure */ +extern int xfs_inobt_decrement(struct xfs_btree_cur *cur, int level, int *stat); /* * Delete the record pointed to by cur. * The cursor refers to the place where the record was (could be inserted) * when the operation returns. */ -int /* error */ -xfs_inobt_delete( - struct xfs_btree_cur *cur, /* btree cursor */ - int *stat); /* success/failure */ +extern int xfs_inobt_delete(struct xfs_btree_cur *cur, int *stat); /* * Get the data from the pointed-to record. */ -int /* error */ -xfs_inobt_get_rec( - struct xfs_btree_cur *cur, /* btree cursor */ - xfs_agino_t *ino, /* output: starting inode of chunk */ - __int32_t *fcnt, /* output: number of free inodes */ - xfs_inofree_t *free, /* output: free inode mask */ - int *stat); /* output: success/failure */ +extern int xfs_inobt_get_rec(struct xfs_btree_cur *cur, xfs_agino_t *ino, + __int32_t *fcnt, xfs_inofree_t *free, int *stat); /* * Increment cursor by one record at the level. * For nonzero levels the leaf-ward information is untouched. */ -int /* error */ -xfs_inobt_increment( - struct xfs_btree_cur *cur, /* btree cursor */ - int level, /* level in btree, 0 is leaf */ - int *stat); /* success/failure */ +extern int xfs_inobt_increment(struct xfs_btree_cur *cur, int level, int *stat); /* * Insert the current record at the point referenced by cur. * The cursor may be inconsistent on return if splits have been done. */ -int /* error */ -xfs_inobt_insert( - struct xfs_btree_cur *cur, /* btree cursor */ - int *stat); /* success/failure */ +extern int xfs_inobt_insert(struct xfs_btree_cur *cur, int *stat); /* * Lookup the record equal to ino in the btree given by cur. */ -int /* error */ -xfs_inobt_lookup_eq( - struct xfs_btree_cur *cur, /* btree cursor */ - xfs_agino_t ino, /* starting inode of chunk */ - __int32_t fcnt, /* free inode count */ - xfs_inofree_t free, /* free inode mask */ - int *stat); /* success/failure */ +extern int xfs_inobt_lookup_eq(struct xfs_btree_cur *cur, xfs_agino_t ino, + __int32_t fcnt, xfs_inofree_t free, int *stat); /* * Lookup the first record greater than or equal to ino * in the btree given by cur. */ -int /* error */ -xfs_inobt_lookup_ge( - struct xfs_btree_cur *cur, /* btree cursor */ - xfs_agino_t ino, /* starting inode of chunk */ - __int32_t fcnt, /* free inode count */ - xfs_inofree_t free, /* free inode mask */ - int *stat); /* success/failure */ +extern int xfs_inobt_lookup_ge(struct xfs_btree_cur *cur, xfs_agino_t ino, + __int32_t fcnt, xfs_inofree_t free, int *stat); /* * Lookup the first record less than or equal to ino * in the btree given by cur. */ -int /* error */ -xfs_inobt_lookup_le( - struct xfs_btree_cur *cur, /* btree cursor */ - xfs_agino_t ino, /* starting inode of chunk */ - __int32_t fcnt, /* free inode count */ - xfs_inofree_t free, /* free inode mask */ - int *stat); /* success/failure */ +extern int xfs_inobt_lookup_le(struct xfs_btree_cur *cur, xfs_agino_t ino, + __int32_t fcnt, xfs_inofree_t free, int *stat); /* * Update the record referred to by cur, to the value given * by [ino, fcnt, free]. * This either works (return 0) or gets an EFSCORRUPTED error. */ -int /* error */ -xfs_inobt_update( - struct xfs_btree_cur *cur, /* btree cursor */ - xfs_agino_t ino, /* starting inode of chunk */ - __int32_t fcnt, /* free inode count */ - xfs_inofree_t free); /* free inode mask */ +extern int xfs_inobt_update(struct xfs_btree_cur *cur, xfs_agino_t ino, + __int32_t fcnt, xfs_inofree_t free); #endif /* __XFS_IALLOC_BTREE_H__ */ diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c index 0d9ae8fb4138..fc19eedbd11b 100644 --- a/fs/xfs/xfs_iget.c +++ b/fs/xfs/xfs_iget.c @@ -1,43 +1,26 @@ /* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - -#include <linux/delay.h> - #include "xfs.h" - -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" @@ -45,19 +28,18 @@ #include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" -#include "xfs_alloc_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_btree.h" -#include "xfs_ialloc.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" #include "xfs_inode.h" +#include "xfs_btree.h" +#include "xfs_ialloc.h" #include "xfs_quota.h" #include "xfs_utils.h" -#include "xfs_bit.h" /* * Initialize the inode hash table for the newly mounted file system. diff --git a/fs/xfs/xfs_imap.h b/fs/xfs/xfs_imap.h index e385064a066a..d36450003983 100644 --- a/fs/xfs/xfs_imap.h +++ b/fs/xfs/xfs_imap.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_IMAP_H__ #define __XFS_IMAP_H__ diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index db43308aae93..df0d4572d70a 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -1,40 +1,27 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" +#include "xfs_imap.h" #include "xfs_trans.h" #include "xfs_trans_priv.h" #include "xfs_sb.h" @@ -43,24 +30,22 @@ #include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" -#include "xfs_alloc_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_btree.h" -#include "xfs_imap.h" -#include "xfs_alloc.h" -#include "xfs_ialloc.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" -#include "xfs_inode_item.h" #include "xfs_inode.h" -#include "xfs_bmap.h" #include "xfs_buf_item.h" +#include "xfs_inode_item.h" +#include "xfs_btree.h" +#include "xfs_alloc.h" +#include "xfs_ialloc.h" +#include "xfs_bmap.h" #include "xfs_rw.h" #include "xfs_error.h" -#include "xfs_bit.h" #include "xfs_utils.h" #include "xfs_dir2_trace.h" #include "xfs_quota.h" @@ -194,9 +179,10 @@ xfs_inotobp( if ((imap.im_blkno + imap.im_len) > XFS_FSB_TO_BB(mp, mp->m_sb.sb_dblocks)) { cmn_err(CE_WARN, - "xfs_inotobp: inode number (%d + %d) maps to a block outside the bounds " + "xfs_inotobp: inode number (%llu + %d) maps to a block outside the bounds " "of the file system %s. Returning EINVAL.", - imap.im_blkno, imap.im_len,mp->m_fsname); + (unsigned long long)imap.im_blkno, + imap.im_len, mp->m_fsname); return XFS_ERROR(EINVAL); } @@ -1878,8 +1864,8 @@ xfs_iunlink( */ agi = XFS_BUF_TO_AGI(agibp); agi_ok = - INT_GET(agi->agi_magicnum, ARCH_CONVERT) == XFS_AGI_MAGIC && - XFS_AGI_GOOD_VERSION(INT_GET(agi->agi_versionnum, ARCH_CONVERT)); + be32_to_cpu(agi->agi_magicnum) == XFS_AGI_MAGIC && + XFS_AGI_GOOD_VERSION(be32_to_cpu(agi->agi_versionnum)); if (unlikely(XFS_TEST_ERROR(!agi_ok, mp, XFS_ERRTAG_IUNLINK, XFS_RANDOM_IUNLINK))) { XFS_CORRUPTION_ERROR("xfs_iunlink", XFS_ERRLEVEL_LOW, mp, agi); @@ -1894,9 +1880,9 @@ xfs_iunlink( ASSERT(agino != 0); bucket_index = agino % XFS_AGI_UNLINKED_BUCKETS; ASSERT(agi->agi_unlinked[bucket_index]); - ASSERT(INT_GET(agi->agi_unlinked[bucket_index], ARCH_CONVERT) != agino); + ASSERT(be32_to_cpu(agi->agi_unlinked[bucket_index]) != agino); - if (INT_GET(agi->agi_unlinked[bucket_index], ARCH_CONVERT) != NULLAGINO) { + if (be32_to_cpu(agi->agi_unlinked[bucket_index]) != NULLAGINO) { /* * There is already another inode in the bucket we need * to add ourselves to. Add us at the front of the list. @@ -1923,7 +1909,7 @@ xfs_iunlink( * Point the bucket head pointer at the inode being inserted. */ ASSERT(agino != 0); - INT_SET(agi->agi_unlinked[bucket_index], ARCH_CONVERT, agino); + agi->agi_unlinked[bucket_index] = cpu_to_be32(agino); offset = offsetof(xfs_agi_t, agi_unlinked) + (sizeof(xfs_agino_t) * bucket_index); xfs_trans_log_buf(tp, agibp, offset, @@ -1981,8 +1967,8 @@ xfs_iunlink_remove( */ agi = XFS_BUF_TO_AGI(agibp); agi_ok = - INT_GET(agi->agi_magicnum, ARCH_CONVERT) == XFS_AGI_MAGIC && - XFS_AGI_GOOD_VERSION(INT_GET(agi->agi_versionnum, ARCH_CONVERT)); + be32_to_cpu(agi->agi_magicnum) == XFS_AGI_MAGIC && + XFS_AGI_GOOD_VERSION(be32_to_cpu(agi->agi_versionnum)); if (unlikely(XFS_TEST_ERROR(!agi_ok, mp, XFS_ERRTAG_IUNLINK_REMOVE, XFS_RANDOM_IUNLINK_REMOVE))) { XFS_CORRUPTION_ERROR("xfs_iunlink_remove", XFS_ERRLEVEL_LOW, @@ -2000,10 +1986,10 @@ xfs_iunlink_remove( agino = XFS_INO_TO_AGINO(mp, ip->i_ino); ASSERT(agino != 0); bucket_index = agino % XFS_AGI_UNLINKED_BUCKETS; - ASSERT(INT_GET(agi->agi_unlinked[bucket_index], ARCH_CONVERT) != NULLAGINO); + ASSERT(be32_to_cpu(agi->agi_unlinked[bucket_index]) != NULLAGINO); ASSERT(agi->agi_unlinked[bucket_index]); - if (INT_GET(agi->agi_unlinked[bucket_index], ARCH_CONVERT) == agino) { + if (be32_to_cpu(agi->agi_unlinked[bucket_index]) == agino) { /* * We're at the head of the list. Get the inode's * on-disk buffer to see if there is anyone after us @@ -2037,7 +2023,7 @@ xfs_iunlink_remove( */ ASSERT(next_agino != 0); ASSERT(next_agino != agino); - INT_SET(agi->agi_unlinked[bucket_index], ARCH_CONVERT, next_agino); + agi->agi_unlinked[bucket_index] = cpu_to_be32(next_agino); offset = offsetof(xfs_agi_t, agi_unlinked) + (sizeof(xfs_agino_t) * bucket_index); xfs_trans_log_buf(tp, agibp, offset, @@ -2046,7 +2032,7 @@ xfs_iunlink_remove( /* * We need to search the list for the inode being freed. */ - next_agino = INT_GET(agi->agi_unlinked[bucket_index], ARCH_CONVERT); + next_agino = be32_to_cpu(agi->agi_unlinked[bucket_index]); last_ibp = NULL; while (next_agino != agino) { /* @@ -3687,73 +3673,6 @@ xfs_iroundup( return( 0 ); } -/* - * Change the requested timestamp in the given inode. - * We don't lock across timestamp updates, and we don't log them but - * we do record the fact that there is dirty information in core. - * - * NOTE -- callers MUST combine XFS_ICHGTIME_MOD or XFS_ICHGTIME_CHG - * with XFS_ICHGTIME_ACC to be sure that access time - * update will take. Calling first with XFS_ICHGTIME_ACC - * and then XFS_ICHGTIME_MOD may fail to modify the access - * timestamp if the filesystem is mounted noacctm. - */ -void -xfs_ichgtime(xfs_inode_t *ip, - int flags) -{ - timespec_t tv; - vnode_t *vp = XFS_ITOV(ip); - struct inode *inode = LINVFS_GET_IP(vp); - - /* - * We're not supposed to change timestamps in readonly-mounted - * filesystems. Throw it away if anyone asks us. - */ - if (unlikely(vp->v_vfsp->vfs_flag & VFS_RDONLY)) - return; - - /* - * Don't update access timestamps on reads if mounted "noatime" - * Throw it away if anyone asks us. - */ - if ((ip->i_mount->m_flags & XFS_MOUNT_NOATIME || IS_NOATIME(inode)) && - ((flags & (XFS_ICHGTIME_ACC|XFS_ICHGTIME_MOD|XFS_ICHGTIME_CHG)) - == XFS_ICHGTIME_ACC)) - return; - - nanotime(&tv); - if (flags & XFS_ICHGTIME_MOD) { - VN_MTIMESET(vp, &tv); - ip->i_d.di_mtime.t_sec = (__int32_t)tv.tv_sec; - ip->i_d.di_mtime.t_nsec = (__int32_t)tv.tv_nsec; - } - if (flags & XFS_ICHGTIME_ACC) { - VN_ATIMESET(vp, &tv); - ip->i_d.di_atime.t_sec = (__int32_t)tv.tv_sec; - ip->i_d.di_atime.t_nsec = (__int32_t)tv.tv_nsec; - } - if (flags & XFS_ICHGTIME_CHG) { - VN_CTIMESET(vp, &tv); - ip->i_d.di_ctime.t_sec = (__int32_t)tv.tv_sec; - ip->i_d.di_ctime.t_nsec = (__int32_t)tv.tv_nsec; - } - - /* - * We update the i_update_core field _after_ changing - * the timestamps in order to coordinate properly with - * xfs_iflush() so that we don't lose timestamp updates. - * This keeps us from having to hold the inode lock - * while doing this. We use the SYNCHRONIZE macro to - * ensure that the compiler does not reorder the update - * of i_update_core above the timestamp updates above. - */ - SYNCHRONIZE(); - ip->i_update_core = 1; - if (!(inode->i_state & I_LOCK)) - mark_inode_dirty_sync(inode); -} - #ifdef XFS_ILOCK_TRACE ktrace_t *xfs_ilock_trace_buf; diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index 54d9e54c7c95..124d30e6143b 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -1,38 +1,30 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_INODE_H__ #define __XFS_INODE_H__ /* + * Fork identifiers. + */ +#define XFS_DATA_FORK 0 +#define XFS_ATTR_FORK 1 + +/* * File incore extent information, present for each of data & attr forks. */ #define XFS_INLINE_EXTS 2 @@ -107,24 +99,6 @@ extern void xfs_ilock_trace(struct xfs_inode *, int, unsigned int, inst_t *); #define xfs_ilock_trace(i,n,f,ra) #endif -/* - * This structure is used to communicate which extents of a file - * were holes when a write started from xfs_write_file() to - * xfs_strat_read(). This is necessary so that we can know which - * blocks need to be zeroed when they are read in in xfs_strat_read() - * if they weren\'t allocated when the buffer given to xfs_strat_read() - * was mapped. - * - * We keep a list of these attached to the inode. The list is - * protected by the inode lock and the fact that the io lock is - * held exclusively by writers. - */ -typedef struct xfs_gap { - struct xfs_gap *xg_next; - xfs_fileoff_t xg_offset_fsb; - xfs_extlen_t xg_count_fsb; -} xfs_gap_t; - typedef struct dm_attrs_s { __uint32_t da_dmevmask; /* DMIG event mask */ __uint16_t da_dmstate; /* DMIG state info */ @@ -311,60 +285,16 @@ typedef struct xfs_inode { /* * Fork handling. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_IFORK_PTR) -xfs_ifork_t *xfs_ifork_ptr(xfs_inode_t *ip, int w); -#define XFS_IFORK_PTR(ip,w) xfs_ifork_ptr(ip,w) -#else -#define XFS_IFORK_PTR(ip,w) ((w) == XFS_DATA_FORK ? &(ip)->i_df : (ip)->i_afp) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_IFORK_Q) -int xfs_ifork_q(xfs_inode_t *ip); -#define XFS_IFORK_Q(ip) xfs_ifork_q(ip) -#else +#define XFS_IFORK_PTR(ip,w) \ + ((w) == XFS_DATA_FORK ? &(ip)->i_df : (ip)->i_afp) #define XFS_IFORK_Q(ip) XFS_CFORK_Q(&(ip)->i_d) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_IFORK_DSIZE) -int xfs_ifork_dsize(xfs_inode_t *ip); -#define XFS_IFORK_DSIZE(ip) xfs_ifork_dsize(ip) -#else #define XFS_IFORK_DSIZE(ip) XFS_CFORK_DSIZE(&ip->i_d, ip->i_mount) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_IFORK_ASIZE) -int xfs_ifork_asize(xfs_inode_t *ip); -#define XFS_IFORK_ASIZE(ip) xfs_ifork_asize(ip) -#else #define XFS_IFORK_ASIZE(ip) XFS_CFORK_ASIZE(&ip->i_d, ip->i_mount) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_IFORK_SIZE) -int xfs_ifork_size(xfs_inode_t *ip, int w); -#define XFS_IFORK_SIZE(ip,w) xfs_ifork_size(ip,w) -#else #define XFS_IFORK_SIZE(ip,w) XFS_CFORK_SIZE(&ip->i_d, ip->i_mount, w) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_IFORK_FORMAT) -int xfs_ifork_format(xfs_inode_t *ip, int w); -#define XFS_IFORK_FORMAT(ip,w) xfs_ifork_format(ip,w) -#else #define XFS_IFORK_FORMAT(ip,w) XFS_CFORK_FORMAT(&ip->i_d, w) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_IFORK_FMT_SET) -void xfs_ifork_fmt_set(xfs_inode_t *ip, int w, int n); -#define XFS_IFORK_FMT_SET(ip,w,n) xfs_ifork_fmt_set(ip,w,n) -#else #define XFS_IFORK_FMT_SET(ip,w,n) XFS_CFORK_FMT_SET(&ip->i_d, w, n) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_IFORK_NEXTENTS) -int xfs_ifork_nextents(xfs_inode_t *ip, int w); -#define XFS_IFORK_NEXTENTS(ip,w) xfs_ifork_nextents(ip,w) -#else #define XFS_IFORK_NEXTENTS(ip,w) XFS_CFORK_NEXTENTS(&ip->i_d, w) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_IFORK_NEXT_SET) -void xfs_ifork_next_set(xfs_inode_t *ip, int w, int n); -#define XFS_IFORK_NEXT_SET(ip,w,n) xfs_ifork_next_set(ip,w,n) -#else #define XFS_IFORK_NEXT_SET(ip,w,n) XFS_CFORK_NEXT_SET(&ip->i_d, w, n) -#endif #ifdef __KERNEL__ @@ -388,6 +318,7 @@ void xfs_ifork_next_set(xfs_inode_t *ip, int w, int n); #define XFS_ILOCK_EXCL 0x004 #define XFS_ILOCK_SHARED 0x008 #define XFS_IUNLOCK_NONOTIFY 0x010 +/* XFS_IOLOCK_NESTED 0x020 */ #define XFS_EXTENT_TOKEN_RD 0x040 #define XFS_SIZE_TOKEN_RD 0x080 #define XFS_EXTSIZE_RD (XFS_EXTENT_TOKEN_RD|XFS_SIZE_TOKEN_RD) @@ -395,7 +326,7 @@ void xfs_ifork_next_set(xfs_inode_t *ip, int w, int n); #define XFS_EXTENT_TOKEN_WR (XFS_EXTENT_TOKEN_RD | XFS_WILLLEND) #define XFS_SIZE_TOKEN_WR (XFS_SIZE_TOKEN_RD | XFS_WILLLEND) #define XFS_EXTSIZE_WR (XFS_EXTSIZE_RD | XFS_WILLLEND) - +/* XFS_SIZE_TOKEN_WANT 0x200 */ #define XFS_LOCK_MASK \ (XFS_IOLOCK_EXCL | XFS_IOLOCK_SHARED | XFS_ILOCK_EXCL | \ @@ -417,28 +348,11 @@ void xfs_ifork_next_set(xfs_inode_t *ip, int w, int n); #define XFS_ITRUNC_DEFINITE 0x1 #define XFS_ITRUNC_MAYBE 0x2 -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_ITOV) -struct vnode *xfs_itov(xfs_inode_t *ip); -#define XFS_ITOV(ip) xfs_itov(ip) -#else #define XFS_ITOV(ip) BHV_TO_VNODE(XFS_ITOBHV(ip)) -#endif #define XFS_ITOV_NULL(ip) BHV_TO_VNODE_NULL(XFS_ITOBHV(ip)) -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_ITOBHV) -struct bhv_desc *xfs_itobhv(xfs_inode_t *ip); -#define XFS_ITOBHV(ip) xfs_itobhv(ip) -#else #define XFS_ITOBHV(ip) ((struct bhv_desc *)(&((ip)->i_bhv_desc))) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BHVTOI) -xfs_inode_t *xfs_bhvtoi(struct bhv_desc *bhvp); -#define XFS_BHVTOI(bhvp) xfs_bhvtoi(bhvp) -#else -#define XFS_BHVTOI(bhvp) \ - ((xfs_inode_t *)((char *)(bhvp) - \ - (char *)&(((xfs_inode_t *)0)->i_bhv_desc))) -#endif - +#define XFS_BHVTOI(bhvp) ((xfs_inode_t *)((char *)(bhvp) - \ + (char *)&(((xfs_inode_t *)0)->i_bhv_desc))) #define BHV_IS_XFS(bdp) (BHV_OPS(bdp) == &xfs_vnodeops) /* diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c index 50e2cadf9091..7f3363c621e1 100644 --- a/fs/xfs/xfs_inode_item.c +++ b/fs/xfs/xfs_inode_item.c @@ -1,66 +1,46 @@ /* - * Copyright (c) 2000-2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ - */ - -/* - * This file contains the implementation of the xfs_inode_log_item. - * It contains the item operations used to manipulate the inode log - * items as well as utility routines used by the inode specific - * transaction routines. + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #include "xfs.h" -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_buf_item.h" #include "xfs_sb.h" +#include "xfs_ag.h" #include "xfs_dir.h" #include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" #include "xfs_trans_priv.h" -#include "xfs_ag.h" -#include "xfs_alloc_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_btree.h" -#include "xfs_ialloc.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" -#include "xfs_inode_item.h" #include "xfs_inode.h" +#include "xfs_inode_item.h" +#include "xfs_btree.h" +#include "xfs_ialloc.h" #include "xfs_rw.h" diff --git a/fs/xfs/xfs_inode_item.h b/fs/xfs/xfs_inode_item.h index d8775e0d6291..c5dbf93b6661 100644 --- a/fs/xfs/xfs_inode_item.h +++ b/fs/xfs/xfs_inode_item.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_INODE_ITEM_H__ #define __XFS_INODE_ITEM_H__ @@ -159,38 +145,33 @@ typedef struct xfs_inode_log_item { } xfs_inode_log_item_t; -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_ILOG_FDATA) -int xfs_ilog_fdata(int w); #define XFS_ILOG_FDATA(w) xfs_ilog_fdata(w) -#else -#define XFS_ILOG_FDATA(w) \ - ((w) == XFS_DATA_FORK ? XFS_ILOG_DDATA : XFS_ILOG_ADATA) -#endif +static inline int xfs_ilog_fdata(int w) +{ + return (w == XFS_DATA_FORK ? XFS_ILOG_DDATA : XFS_ILOG_ADATA); +} #endif /* __KERNEL__ */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_ILOG_FBROOT) -int xfs_ilog_fbroot(int w); #define XFS_ILOG_FBROOT(w) xfs_ilog_fbroot(w) -#else -#define XFS_ILOG_FBROOT(w) \ - ((w) == XFS_DATA_FORK ? XFS_ILOG_DBROOT : XFS_ILOG_ABROOT) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_ILOG_FEXT) -int xfs_ilog_fext(int w); +static inline int xfs_ilog_fbroot(int w) +{ + return (w == XFS_DATA_FORK ? XFS_ILOG_DBROOT : XFS_ILOG_ABROOT); +} + #define XFS_ILOG_FEXT(w) xfs_ilog_fext(w) -#else -#define XFS_ILOG_FEXT(w) \ - ((w) == XFS_DATA_FORK ? XFS_ILOG_DEXT : XFS_ILOG_AEXT) -#endif +static inline int xfs_ilog_fext(int w) +{ + return (w == XFS_DATA_FORK ? XFS_ILOG_DEXT : XFS_ILOG_AEXT); +} #ifdef __KERNEL__ -void xfs_inode_item_init(struct xfs_inode *, struct xfs_mount *); -void xfs_inode_item_destroy(struct xfs_inode *); -void xfs_iflush_done(struct xfs_buf *, xfs_inode_log_item_t *); -void xfs_istale_done(struct xfs_buf *, xfs_inode_log_item_t *); -void xfs_iflush_abort(struct xfs_inode *); +extern void xfs_inode_item_init(struct xfs_inode *, struct xfs_mount *); +extern void xfs_inode_item_destroy(struct xfs_inode *); +extern void xfs_iflush_done(struct xfs_buf *, xfs_inode_log_item_t *); +extern void xfs_istale_done(struct xfs_buf *, xfs_inode_log_item_t *); +extern void xfs_iflush_abort(struct xfs_inode *); #endif /* __KERNEL__ */ diff --git a/fs/xfs/xfs_inum.h b/fs/xfs/xfs_inum.h index a3af2d5a6eb7..7a28191cb0de 100644 --- a/fs/xfs/xfs_inum.h +++ b/fs/xfs/xfs_inum.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_INUM_H__ #define __XFS_INUM_H__ @@ -58,109 +44,31 @@ typedef __uint32_t xfs_intino_t; struct xfs_mount; -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_INO_MASK) -__uint32_t xfs_ino_mask(int k); -#define XFS_INO_MASK(k) xfs_ino_mask(k) -#else -#define XFS_INO_MASK(k) ((__uint32_t)((1ULL << (k)) - 1)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_INO_OFFSET_BITS) -int xfs_ino_offset_bits(struct xfs_mount *mp); -#define XFS_INO_OFFSET_BITS(mp) xfs_ino_offset_bits(mp) -#else -#define XFS_INO_OFFSET_BITS(mp) ((mp)->m_sb.sb_inopblog) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_INO_AGBNO_BITS) -int xfs_ino_agbno_bits(struct xfs_mount *mp); -#define XFS_INO_AGBNO_BITS(mp) xfs_ino_agbno_bits(mp) -#else -#define XFS_INO_AGBNO_BITS(mp) ((mp)->m_sb.sb_agblklog) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_INO_AGINO_BITS) -int xfs_ino_agino_bits(struct xfs_mount *mp); -#define XFS_INO_AGINO_BITS(mp) xfs_ino_agino_bits(mp) -#else -#define XFS_INO_AGINO_BITS(mp) ((mp)->m_agino_log) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_INO_AGNO_BITS) -int xfs_ino_agno_bits(struct xfs_mount *mp); -#define XFS_INO_AGNO_BITS(mp) xfs_ino_agno_bits(mp) -#else -#define XFS_INO_AGNO_BITS(mp) ((mp)->m_agno_log) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_INO_BITS) -int xfs_ino_bits(struct xfs_mount *mp); -#define XFS_INO_BITS(mp) xfs_ino_bits(mp) -#else -#define XFS_INO_BITS(mp) (XFS_INO_AGNO_BITS(mp) + XFS_INO_AGINO_BITS(mp)) -#endif - -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_INO_TO_AGNO) -xfs_agnumber_t xfs_ino_to_agno(struct xfs_mount *mp, xfs_ino_t i); -#define XFS_INO_TO_AGNO(mp,i) xfs_ino_to_agno(mp,i) -#else -#define XFS_INO_TO_AGNO(mp,i) \ +#define XFS_INO_MASK(k) (__uint32_t)((1ULL << (k)) - 1) +#define XFS_INO_OFFSET_BITS(mp) (mp)->m_sb.sb_inopblog +#define XFS_INO_AGBNO_BITS(mp) (mp)->m_sb.sb_agblklog +#define XFS_INO_AGINO_BITS(mp) (mp)->m_agino_log +#define XFS_INO_AGNO_BITS(mp) (mp)->m_agno_log +#define XFS_INO_BITS(mp) \ + XFS_INO_AGNO_BITS(mp) + XFS_INO_AGINO_BITS(mp) +#define XFS_INO_TO_AGNO(mp,i) \ ((xfs_agnumber_t)((i) >> XFS_INO_AGINO_BITS(mp))) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_INO_TO_AGINO) -xfs_agino_t xfs_ino_to_agino(struct xfs_mount *mp, xfs_ino_t i); -#define XFS_INO_TO_AGINO(mp,i) xfs_ino_to_agino(mp,i) -#else -#define XFS_INO_TO_AGINO(mp,i) \ +#define XFS_INO_TO_AGINO(mp,i) \ ((xfs_agino_t)(i) & XFS_INO_MASK(XFS_INO_AGINO_BITS(mp))) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_INO_TO_AGBNO) -xfs_agblock_t xfs_ino_to_agbno(struct xfs_mount *mp, xfs_ino_t i); -#define XFS_INO_TO_AGBNO(mp,i) xfs_ino_to_agbno(mp,i) -#else -#define XFS_INO_TO_AGBNO(mp,i) \ +#define XFS_INO_TO_AGBNO(mp,i) \ (((xfs_agblock_t)(i) >> XFS_INO_OFFSET_BITS(mp)) & \ - XFS_INO_MASK(XFS_INO_AGBNO_BITS(mp))) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_INO_TO_OFFSET) -int xfs_ino_to_offset(struct xfs_mount *mp, xfs_ino_t i); -#define XFS_INO_TO_OFFSET(mp,i) xfs_ino_to_offset(mp,i) -#else -#define XFS_INO_TO_OFFSET(mp,i) \ + XFS_INO_MASK(XFS_INO_AGBNO_BITS(mp))) +#define XFS_INO_TO_OFFSET(mp,i) \ ((int)(i) & XFS_INO_MASK(XFS_INO_OFFSET_BITS(mp))) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_INO_TO_FSB) -xfs_fsblock_t xfs_ino_to_fsb(struct xfs_mount *mp, xfs_ino_t i); -#define XFS_INO_TO_FSB(mp,i) xfs_ino_to_fsb(mp,i) -#else -#define XFS_INO_TO_FSB(mp,i) \ +#define XFS_INO_TO_FSB(mp,i) \ XFS_AGB_TO_FSB(mp, XFS_INO_TO_AGNO(mp,i), XFS_INO_TO_AGBNO(mp,i)) -#endif - -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_AGINO_TO_INO) -xfs_ino_t -xfs_agino_to_ino(struct xfs_mount *mp, xfs_agnumber_t a, xfs_agino_t i); -#define XFS_AGINO_TO_INO(mp,a,i) xfs_agino_to_ino(mp,a,i) -#else #define XFS_AGINO_TO_INO(mp,a,i) \ (((xfs_ino_t)(a) << XFS_INO_AGINO_BITS(mp)) | (i)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_AGINO_TO_AGBNO) -xfs_agblock_t xfs_agino_to_agbno(struct xfs_mount *mp, xfs_agino_t i); -#define XFS_AGINO_TO_AGBNO(mp,i) xfs_agino_to_agbno(mp,i) -#else #define XFS_AGINO_TO_AGBNO(mp,i) ((i) >> XFS_INO_OFFSET_BITS(mp)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_AGINO_TO_OFFSET) -int xfs_agino_to_offset(struct xfs_mount *mp, xfs_agino_t i); -#define XFS_AGINO_TO_OFFSET(mp,i) xfs_agino_to_offset(mp,i) -#else #define XFS_AGINO_TO_OFFSET(mp,i) \ ((i) & XFS_INO_MASK(XFS_INO_OFFSET_BITS(mp))) -#endif - -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_OFFBNO_TO_AGINO) -xfs_agino_t xfs_offbno_to_agino(struct xfs_mount *mp, xfs_agblock_t b, int o); -#define XFS_OFFBNO_TO_AGINO(mp,b,o) xfs_offbno_to_agino(mp,b,o) -#else #define XFS_OFFBNO_TO_AGINO(mp,b,o) \ ((xfs_agino_t)(((b) << XFS_INO_OFFSET_BITS(mp)) | (o))) -#endif #if XFS_BIG_INUMS #define XFS_MAXINUMBER ((xfs_ino_t)((1ULL << 56) - 1ULL)) diff --git a/fs/xfs/xfs_iocore.c b/fs/xfs/xfs_iocore.c index 414ec496845f..a07815661a8c 100644 --- a/fs/xfs/xfs_iocore.c +++ b/fs/xfs/xfs_iocore.c @@ -1,40 +1,26 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" @@ -42,22 +28,21 @@ #include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" -#include "xfs_alloc_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_itable.h" -#include "xfs_btree.h" -#include "xfs_alloc.h" -#include "xfs_ialloc.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" -#include "xfs_inode_item.h" #include "xfs_inode.h" +#include "xfs_inode_item.h" +#include "xfs_itable.h" +#include "xfs_btree.h" +#include "xfs_alloc.h" +#include "xfs_ialloc.h" #include "xfs_bmap.h" #include "xfs_error.h" -#include "xfs_bit.h" #include "xfs_rw.h" #include "xfs_quota.h" #include "xfs_trans_space.h" diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c index d0f5be63cddb..45a77a3a6c07 100644 --- a/fs/xfs/xfs_iomap.c +++ b/fs/xfs/xfs_iomap.c @@ -1,40 +1,25 @@ /* - * Copyright (c) 2000-2004 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" - #include "xfs_fs.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" @@ -44,16 +29,16 @@ #include "xfs_dmapi.h" #include "xfs_quota.h" #include "xfs_mount.h" -#include "xfs_alloc_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_btree.h" -#include "xfs_ialloc.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" #include "xfs_inode.h" +#include "xfs_ialloc.h" +#include "xfs_btree.h" #include "xfs_bmap.h" #include "xfs_bit.h" #include "xfs_rtalloc.h" @@ -379,17 +364,15 @@ xfs_iomap_write_direct( xfs_fileoff_t offset_fsb; xfs_fileoff_t last_fsb; xfs_filblks_t count_fsb; - xfs_fsize_t isize; xfs_fsblock_t firstfsb; - int nimaps, maps; + int nimaps; int error; int bmapi_flag; int quota_flag; int rt; xfs_trans_t *tp; - xfs_bmbt_irec_t imap[XFS_WRITE_IMAPS], *imapp; + xfs_bmbt_irec_t imap; xfs_bmap_free_t free_list; - int aeof; xfs_filblks_t qblocks, resblks; int committed; int resrtextents; @@ -402,15 +385,6 @@ xfs_iomap_write_direct( if (error) return XFS_ERROR(error); - maps = min(XFS_WRITE_IMAPS, *nmaps); - nimaps = maps; - - isize = ip->i_d.di_size; - aeof = (offset + count) > isize; - - if (io->io_new_size > isize) - isize = io->io_new_size; - offset_fsb = XFS_B_TO_FSBT(mp, offset); last_fsb = XFS_B_TO_FSB(mp, ((xfs_ufsize_t)(offset + count))); count_fsb = last_fsb - offset_fsb; @@ -479,9 +453,8 @@ xfs_iomap_write_direct( */ XFS_BMAP_INIT(&free_list, &firstfsb); nimaps = 1; - imapp = &imap[0]; error = xfs_bmapi(tp, ip, offset_fsb, count_fsb, - bmapi_flag, &firstfsb, 0, imapp, &nimaps, &free_list); + bmapi_flag, &firstfsb, 0, &imap, &nimaps, &free_list); if (error) goto error0; @@ -503,7 +476,7 @@ xfs_iomap_write_direct( goto error_out; } - *ret_imap = imap[0]; + *ret_imap = imap; *nmaps = 1; if ( !(io->io_flags & XFS_IOCORE_RT) && !ret_imap->br_startblock) { cmn_err(CE_PANIC,"Access to block zero: fs <%s> inode: %lld " diff --git a/fs/xfs/xfs_iomap.h b/fs/xfs/xfs_iomap.h index 4daaa5212102..3ce204a524b0 100644 --- a/fs/xfs/xfs_iomap.h +++ b/fs/xfs/xfs_iomap.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2003-2005 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2003-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_IOMAP_H__ #define __XFS_IOMAP_H__ @@ -83,7 +69,7 @@ typedef struct xfs_iomap { xfs_buftarg_t *iomap_target; xfs_off_t iomap_offset; /* offset of mapping, bytes */ xfs_off_t iomap_bsize; /* size of mapping, bytes */ - size_t iomap_delta; /* offset into mapping, bytes */ + xfs_off_t iomap_delta; /* offset into mapping, bytes */ iomap_flags_t iomap_flags; } xfs_iomap_t; diff --git a/fs/xfs/xfs_itable.c b/fs/xfs/xfs_itable.c index 8fbc8d378188..f63646ead816 100644 --- a/fs/xfs/xfs_itable.c +++ b/fs/xfs/xfs_itable.c @@ -1,59 +1,45 @@ /* - * Copyright (c) 2000-2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" +#include "xfs_ag.h" #include "xfs_dir.h" #include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" -#include "xfs_ag.h" -#include "xfs_alloc_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_btree.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" #include "xfs_inode.h" #include "xfs_ialloc.h" #include "xfs_itable.h" #include "xfs_error.h" +#include "xfs_btree.h" #ifndef HAVE_USERACC #define useracc(ubuffer, size, flags, foo) (0) @@ -462,7 +448,7 @@ xfs_bulkstat( while (error) { agino += XFS_INODES_PER_CHUNK; if (XFS_AGINO_TO_AGBNO(mp, agino) >= - INT_GET(agi->agi_length, ARCH_CONVERT)) + be32_to_cpu(agi->agi_length)) break; error = xfs_inobt_lookup_ge(cur, agino, 0, 0, &tmp); diff --git a/fs/xfs/xfs_itable.h b/fs/xfs/xfs_itable.h index 2be9d1805ab2..047d834ed210 100644 --- a/fs/xfs/xfs_itable.h +++ b/fs/xfs/xfs_itable.h @@ -1,33 +1,18 @@ /* * Copyright (c) 2000-2001 Silicon Graphics, Inc. All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_ITABLE_H__ #define __XFS_ITABLE_H__ diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index 54a6f1142403..29af51275ca9 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -1,58 +1,47 @@ /* - * Copyright (c) 2000-2004 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - -/* - * High level interface routines for log manager - */ - #include "xfs.h" -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" -#include "xfs_ag.h" -#include "xfs_sb.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" +#include "xfs_sb.h" +#include "xfs_ag.h" #include "xfs_dir.h" +#include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" #include "xfs_error.h" #include "xfs_log_priv.h" #include "xfs_buf_item.h" +#include "xfs_bmap_btree.h" #include "xfs_alloc_btree.h" +#include "xfs_ialloc_btree.h" #include "xfs_log_recover.h" -#include "xfs_bit.h" -#include "xfs_rw.h" #include "xfs_trans_priv.h" +#include "xfs_dir_sf.h" +#include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" +#include "xfs_dinode.h" +#include "xfs_inode.h" +#include "xfs_rw.h" #define xlog_write_adv_cnt(ptr, len, off, bytes) \ @@ -93,8 +82,11 @@ STATIC int xlog_state_release_iclog(xlog_t *log, STATIC void xlog_state_switch_iclogs(xlog_t *log, xlog_in_core_t *iclog, int eventual_size); -STATIC int xlog_state_sync(xlog_t *log, xfs_lsn_t lsn, uint flags); -STATIC int xlog_state_sync_all(xlog_t *log, uint flags); +STATIC int xlog_state_sync(xlog_t *log, + xfs_lsn_t lsn, + uint flags, + int *log_flushed); +STATIC int xlog_state_sync_all(xlog_t *log, uint flags, int *log_flushed); STATIC void xlog_state_want_sync(xlog_t *log, xlog_in_core_t *iclog); /* local functions to manipulate grant head */ @@ -119,8 +111,7 @@ STATIC xlog_ticket_t *xlog_ticket_get(xlog_t *log, uint flags); STATIC void xlog_ticket_put(xlog_t *log, xlog_ticket_t *ticket); -/* local debug functions */ -#if defined(DEBUG) && !defined(XLOG_NOLOG) +#if defined(DEBUG) STATIC void xlog_verify_dest_ptr(xlog_t *log, __psint_t ptr); STATIC void xlog_verify_grant_head(xlog_t *log, int equals); STATIC void xlog_verify_iclog(xlog_t *log, xlog_in_core_t *iclog, @@ -136,26 +127,7 @@ STATIC void xlog_verify_tail_lsn(xlog_t *log, xlog_in_core_t *iclog, STATIC int xlog_iclogs_empty(xlog_t *log); -#ifdef DEBUG -int xlog_do_error = 0; -int xlog_req_num = 0; -int xlog_error_mod = 33; -#endif - -#define XLOG_FORCED_SHUTDOWN(log) (log->l_flags & XLOG_IO_ERROR) - -/* - * 0 => disable log manager - * 1 => enable log manager - * 2 => enable log manager and log debugging - */ -#if defined(XLOG_NOLOG) || defined(DEBUG) -int xlog_debug = 1; -xfs_buftarg_t *xlog_target; -#endif - #if defined(XFS_LOG_TRACE) - void xlog_trace_loggrant(xlog_t *log, xlog_ticket_t *tic, xfs_caddr_t string) { @@ -191,31 +163,16 @@ xlog_trace_loggrant(xlog_t *log, xlog_ticket_t *tic, xfs_caddr_t string) void xlog_trace_iclog(xlog_in_core_t *iclog, uint state) { - pid_t pid; - - pid = current_pid(); - if (!iclog->ic_trace) iclog->ic_trace = ktrace_alloc(256, KM_SLEEP); ktrace_enter(iclog->ic_trace, (void *)((unsigned long)state), - (void *)((unsigned long)pid), - (void *)0, - (void *)0, - (void *)0, - (void *)0, - (void *)0, - (void *)0, - (void *)0, - (void *)0, - (void *)0, - (void *)0, - (void *)0, - (void *)0, - (void *)0, - (void *)0); + (void *)((unsigned long)current_pid()), + (void *)NULL, (void *)NULL, (void *)NULL, (void *)NULL, + (void *)NULL, (void *)NULL, (void *)NULL, (void *)NULL, + (void *)NULL, (void *)NULL, (void *)NULL, (void *)NULL, + (void *)NULL, (void *)NULL); } - #else #define xlog_trace_loggrant(log,tic,string) #define xlog_trace_iclog(iclog,state) @@ -252,11 +209,6 @@ xfs_log_done(xfs_mount_t *mp, xlog_ticket_t *ticket = (xfs_log_ticket_t) xtic; xfs_lsn_t lsn = 0; -#if defined(DEBUG) || defined(XLOG_NOLOG) - if (!xlog_debug && xlog_target == log->l_targ) - return 0; -#endif - if (XLOG_FORCED_SHUTDOWN(log) || /* * If nothing was ever written, don't write out commit record. @@ -312,33 +264,28 @@ xfs_log_done(xfs_mount_t *mp, * semaphore. */ int -xfs_log_force(xfs_mount_t *mp, - xfs_lsn_t lsn, - uint flags) +_xfs_log_force( + xfs_mount_t *mp, + xfs_lsn_t lsn, + uint flags, + int *log_flushed) { - int rval; - xlog_t *log = mp->m_log; + xlog_t *log = mp->m_log; + int dummy; -#if defined(DEBUG) || defined(XLOG_NOLOG) - if (!xlog_debug && xlog_target == log->l_targ) - return 0; -#endif + if (!log_flushed) + log_flushed = &dummy; ASSERT(flags & XFS_LOG_FORCE); XFS_STATS_INC(xs_log_force); - if ((log->l_flags & XLOG_IO_ERROR) == 0) { - if (lsn == 0) - rval = xlog_state_sync_all(log, flags); - else - rval = xlog_state_sync(log, lsn, flags); - } else { - rval = XFS_ERROR(EIO); - } - - return rval; - + if (log->l_flags & XLOG_IO_ERROR) + return XFS_ERROR(EIO); + if (lsn == 0) + return xlog_state_sync_all(log, flags, log_flushed); + else + return xlog_state_sync(log, lsn, flags, log_flushed); } /* xfs_log_force */ /* @@ -356,10 +303,6 @@ xfs_log_notify(xfs_mount_t *mp, /* mount of partition */ xlog_in_core_t *iclog = (xlog_in_core_t *)iclog_hndl; int abortflg, spl; -#if defined(DEBUG) || defined(XLOG_NOLOG) - if (!xlog_debug && xlog_target == log->l_targ) - return 0; -#endif cb->cb_next = NULL; spl = LOG_LOCK(log); abortflg = (iclog->ic_state & XLOG_STATE_IOERROR); @@ -410,13 +353,8 @@ xfs_log_reserve(xfs_mount_t *mp, { xlog_t *log = mp->m_log; xlog_ticket_t *internal_ticket; - int retval; + int retval = 0; -#if defined(DEBUG) || defined(XLOG_NOLOG) - if (!xlog_debug && xlog_target == log->l_targ) - return 0; -#endif - retval = 0; ASSERT(client == XFS_TRANSACTION || client == XFS_LOG); ASSERT((flags & XFS_LOG_NOSLEEP) == 0); @@ -478,12 +416,6 @@ xfs_log_mount(xfs_mount_t *mp, mp->m_log = xlog_alloc_log(mp, log_target, blk_offset, num_bblks); -#if defined(DEBUG) || defined(XLOG_NOLOG) - if (!xlog_debug) { - cmn_err(CE_NOTE, "log dev: %s", XFS_BUFTARG_NAME(log_target)); - return 0; - } -#endif /* * skip log recovery on a norecovery mount. pretend it all * just worked. @@ -587,11 +519,6 @@ xfs_log_unmount_write(xfs_mount_t *mp) __uint32_t pad2; /* may as well make it 64 bits */ } magic = { XLOG_UNMOUNT_TYPE, 0, 0 }; -#if defined(DEBUG) || defined(XLOG_NOLOG) - if (!xlog_debug && xlog_target == log->l_targ) - return 0; -#endif - /* * Don't write out unmount record on read-only mounts. * Or, if we are doing a forced umount (typically because of IO errors). @@ -718,12 +645,6 @@ xfs_log_write(xfs_mount_t * mp, int error; xlog_t *log = mp->m_log; -#if defined(DEBUG) || defined(XLOG_NOLOG) - if (!xlog_debug && xlog_target == log->l_targ) { - *start_lsn = 0; - return 0; - } -#endif if (XLOG_FORCED_SHUTDOWN(log)) return XFS_ERROR(EIO); @@ -743,11 +664,6 @@ xfs_log_move_tail(xfs_mount_t *mp, int need_bytes, free_bytes, cycle, bytes; SPLDECL(s); -#if defined(DEBUG) || defined(XLOG_NOLOG) - if (!xlog_debug && xlog_target == log->l_targ) - return; -#endif - /* XXXsup tmp */ if (XLOG_FORCED_SHUTDOWN(log)) return; ASSERT(!XFS_FORCED_SHUTDOWN(mp)); @@ -1034,51 +950,22 @@ xlog_get_iclog_buffer_size(xfs_mount_t *mp, int size; int xhdrs; -#if defined(DEBUG) || defined(XLOG_NOLOG) - /* - * When logbufs == 0, someone has disabled the log from the FSTAB - * file. This is not a documented feature. We need to set xlog_debug - * to zero (this deactivates the log) and set xlog_target to the - * appropriate device. Only one filesystem may be affected as such - * since this is just a performance hack to test what we might be able - * to get if the log were not present. - */ - if (mp->m_logbufs == 0) { - xlog_debug = 0; - xlog_target = log->l_targ; - log->l_iclog_bufs = XLOG_MIN_ICLOGS; - } else -#endif - { - /* - * This is the normal path. If m_logbufs == -1, then the - * admin has chosen to use the system defaults for logbuffers. - */ - if (mp->m_logbufs == -1) { - if (xfs_physmem <= btoc(128*1024*1024)) { - log->l_iclog_bufs = XLOG_MIN_ICLOGS; - } else if (xfs_physmem <= btoc(400*1024*1024)) { - log->l_iclog_bufs = XLOG_MED_ICLOGS; - } else { - /* 256K with 32K bufs */ - log->l_iclog_bufs = XLOG_MAX_ICLOGS; - } - } else - log->l_iclog_bufs = mp->m_logbufs; - -#if defined(DEBUG) || defined(XLOG_NOLOG) - /* We are reactivating a filesystem after it was inactive */ - if (log->l_targ == xlog_target) { - xlog_target = NULL; - xlog_debug = 1; + if (mp->m_logbufs <= 0) { + if (xfs_physmem <= btoc(128*1024*1024)) { + log->l_iclog_bufs = XLOG_MIN_ICLOGS; + } else if (xfs_physmem <= btoc(400*1024*1024)) { + log->l_iclog_bufs = XLOG_MED_ICLOGS; + } else { /* 256K with 32K bufs */ + log->l_iclog_bufs = XLOG_MAX_ICLOGS; } -#endif + } else { + log->l_iclog_bufs = mp->m_logbufs; } /* * Buffer size passed in from mount system call. */ - if (mp->m_logbsize != -1) { + if (mp->m_logbsize > 0) { size = log->l_iclog_size = mp->m_logbsize; log->l_iclog_size_log = 0; while (size != 1) { @@ -1101,7 +988,7 @@ xlog_get_iclog_buffer_size(xfs_mount_t *mp, log->l_iclog_hsize = BBSIZE; log->l_iclog_heads = 1; } - return; + goto done; } /* @@ -1128,7 +1015,7 @@ xlog_get_iclog_buffer_size(xfs_mount_t *mp, if (mp->m_sb.sb_blocksize >= 16*1024) { log->l_iclog_size = XLOG_BIG_RECORD_BSIZE; log->l_iclog_size_log = XLOG_BIG_RECORD_BSHIFT; - if (mp->m_logbufs == -1) { + if (mp->m_logbufs <= 0) { switch (mp->m_sb.sb_blocksize) { case 16*1024: /* 16 KB */ log->l_iclog_bufs = 3; @@ -1145,6 +1032,12 @@ xlog_get_iclog_buffer_size(xfs_mount_t *mp, } } } + +done: /* are we being asked to make the sizes selected above visible? */ + if (mp->m_logbufs == 0) + mp->m_logbufs = log->l_iclog_bufs; + if (mp->m_logbsize == 0) + mp->m_logbsize = log->l_iclog_size; } /* xlog_get_iclog_buffer_size */ @@ -1467,14 +1360,13 @@ xlog_sync(xlog_t *log, XFS_BUF_BUSY(bp); XFS_BUF_ASYNC(bp); /* - * Do a disk write cache flush for the log block. - * This is a bit of a sledgehammer, it would be better - * to use a tag barrier here that just prevents reordering. + * Do an ordered write for the log block. + * * It may not be needed to flush the first split block in the log wrap * case, but do it anyways to be safe -AK */ - if (!(log->l_mp->m_flags & XFS_MOUNT_NOLOGFLUSH)) - XFS_BUF_FLUSH(bp); + if (log->l_mp->m_flags & XFS_MOUNT_BARRIER) + XFS_BUF_ORDERED(bp); ASSERT(XFS_BUF_ADDR(bp) <= log->l_logBBsize-1); ASSERT(XFS_BUF_ADDR(bp) + BTOBB(count) <= log->l_logBBsize); @@ -1505,8 +1397,8 @@ xlog_sync(xlog_t *log, XFS_BUF_SET_FSPRIVATE(bp, iclog); XFS_BUF_BUSY(bp); XFS_BUF_ASYNC(bp); - if (!(log->l_mp->m_flags & XFS_MOUNT_NOLOGFLUSH)) - XFS_BUF_FLUSH(bp); + if (log->l_mp->m_flags & XFS_MOUNT_BARRIER) + XFS_BUF_ORDERED(bp); dptr = XFS_BUF_PTR(bp); /* * Bump the cycle numbers at the start of each block @@ -2951,7 +2843,7 @@ xlog_state_switch_iclogs(xlog_t *log, * not in the active nor dirty state. */ STATIC int -xlog_state_sync_all(xlog_t *log, uint flags) +xlog_state_sync_all(xlog_t *log, uint flags, int *log_flushed) { xlog_in_core_t *iclog; xfs_lsn_t lsn; @@ -3000,6 +2892,7 @@ xlog_state_sync_all(xlog_t *log, uint flags) if (xlog_state_release_iclog(log, iclog)) return XFS_ERROR(EIO); + *log_flushed = 1; s = LOG_LOCK(log); if (INT_GET(iclog->ic_header.h_lsn, ARCH_CONVERT) == lsn && iclog->ic_state != XLOG_STATE_DIRTY) @@ -3043,6 +2936,7 @@ maybe_sleep: */ if (iclog->ic_state & XLOG_STATE_IOERROR) return XFS_ERROR(EIO); + *log_flushed = 1; } else { @@ -3068,7 +2962,8 @@ no_sleep: int xlog_state_sync(xlog_t *log, xfs_lsn_t lsn, - uint flags) + uint flags, + int *log_flushed) { xlog_in_core_t *iclog; int already_slept = 0; @@ -3120,6 +3015,7 @@ try_again: XFS_STATS_INC(xs_log_force_sleep); sv_wait(&iclog->ic_prev->ic_writesema, PSWP, &log->l_icloglock, s); + *log_flushed = 1; already_slept = 1; goto try_again; } else { @@ -3128,6 +3024,7 @@ try_again: LOG_UNLOCK(log, s); if (xlog_state_release_iclog(log, iclog)) return XFS_ERROR(EIO); + *log_flushed = 1; s = LOG_LOCK(log); } } @@ -3152,6 +3049,7 @@ try_again: */ if (iclog->ic_state & XLOG_STATE_IOERROR) return XFS_ERROR(EIO); + *log_flushed = 1; } else { /* just return */ LOG_UNLOCK(log, s); } @@ -3392,7 +3290,7 @@ xlog_ticket_get(xlog_t *log, * ****************************************************************************** */ -#if defined(DEBUG) && !defined(XLOG_NOLOG) +#if defined(DEBUG) /* * Make sure that the destination ptr is within the valid data region of * one of the iclogs. This uses backup pointers stored in a different @@ -3533,7 +3431,9 @@ xlog_verify_iclog(xlog_t *log, } } if (clientid != XFS_TRANSACTION && clientid != XFS_LOG) - cmn_err(CE_WARN, "xlog_verify_iclog: invalid clientid %d op 0x%p offset 0x%x", clientid, ophead, field_offset); + cmn_err(CE_WARN, "xlog_verify_iclog: " + "invalid clientid %d op 0x%p offset 0x%lx", + clientid, ophead, (unsigned long)field_offset); /* check length */ field_offset = (__psint_t) @@ -3554,7 +3454,7 @@ xlog_verify_iclog(xlog_t *log, ptr += sizeof(xlog_op_header_t) + op_len; } } /* xlog_verify_iclog */ -#endif /* DEBUG && !XLOG_NOLOG */ +#endif /* * Mark all iclogs IOERROR. LOG_LOCK is held by the caller. @@ -3604,6 +3504,7 @@ xfs_log_force_umount( xlog_ticket_t *tic; xlog_t *log; int retval; + int dummy; SPLDECL(s); SPLDECL(s2); @@ -3682,7 +3583,7 @@ xfs_log_force_umount( * Force the incore logs to disk before shutting the * log down completely. */ - xlog_state_sync_all(log, XFS_LOG_FORCE|XFS_LOG_SYNC); + xlog_state_sync_all(log, XFS_LOG_FORCE|XFS_LOG_SYNC, &dummy); s2 = LOG_LOCK(log); retval = xlog_state_ioerror(log); LOG_UNLOCK(log, s2); diff --git a/fs/xfs/xfs_log.h b/fs/xfs/xfs_log.h index 18961119fc65..158829ca56f6 100644 --- a/fs/xfs/xfs_log.h +++ b/fs/xfs/xfs_log.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_LOG_H__ #define __XFS_LOG_H__ @@ -174,9 +160,12 @@ xfs_lsn_t xfs_log_done(struct xfs_mount *mp, xfs_log_ticket_t ticket, void **iclog, uint flags); -int xfs_log_force(struct xfs_mount *mp, - xfs_lsn_t lsn, - uint flags); +int _xfs_log_force(struct xfs_mount *mp, + xfs_lsn_t lsn, + uint flags, + int *log_forced); +#define xfs_log_force(mp, lsn, flags) \ + _xfs_log_force(mp, lsn, flags, NULL); int xfs_log_mount(struct xfs_mount *mp, struct xfs_buftarg *log_target, xfs_daddr_t start_block, diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h index a884cea82fca..4518b188ade6 100644 --- a/fs/xfs/xfs_log_priv.h +++ b/fs/xfs/xfs_log_priv.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_LOG_PRIV_H__ #define __XFS_LOG_PRIV_H__ @@ -35,6 +21,7 @@ struct xfs_buf; struct ktrace; struct log; +struct xlog_ticket; struct xfs_buf_cancel; struct xfs_mount; @@ -120,77 +107,6 @@ struct xfs_mount; ((i) >> 24) #endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XLOG_GRANT_SUB_SPACE) -void xlog_grant_sub_space(struct log *log, int bytes, int type); -#define XLOG_GRANT_SUB_SPACE(log,bytes,type) \ - xlog_grant_sub_space(log,bytes,type) -#else -#define XLOG_GRANT_SUB_SPACE(log,bytes,type) \ - { \ - if (type == 'w') { \ - (log)->l_grant_write_bytes -= (bytes); \ - if ((log)->l_grant_write_bytes < 0) { \ - (log)->l_grant_write_bytes += (log)->l_logsize; \ - (log)->l_grant_write_cycle--; \ - } \ - } else { \ - (log)->l_grant_reserve_bytes -= (bytes); \ - if ((log)->l_grant_reserve_bytes < 0) { \ - (log)->l_grant_reserve_bytes += (log)->l_logsize;\ - (log)->l_grant_reserve_cycle--; \ - } \ - } \ - } -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XLOG_GRANT_ADD_SPACE) -void xlog_grant_add_space(struct log *log, int bytes, int type); -#define XLOG_GRANT_ADD_SPACE(log,bytes,type) \ - xlog_grant_add_space(log,bytes,type) -#else -#define XLOG_GRANT_ADD_SPACE(log,bytes,type) \ - { \ - if (type == 'w') { \ - (log)->l_grant_write_bytes += (bytes); \ - if ((log)->l_grant_write_bytes > (log)->l_logsize) { \ - (log)->l_grant_write_bytes -= (log)->l_logsize; \ - (log)->l_grant_write_cycle++; \ - } \ - } else { \ - (log)->l_grant_reserve_bytes += (bytes); \ - if ((log)->l_grant_reserve_bytes > (log)->l_logsize) { \ - (log)->l_grant_reserve_bytes -= (log)->l_logsize;\ - (log)->l_grant_reserve_cycle++; \ - } \ - } \ - } -#endif -#define XLOG_INS_TICKETQ(q,tic) \ - { \ - if (q) { \ - (tic)->t_next = (q); \ - (tic)->t_prev = (q)->t_prev; \ - (q)->t_prev->t_next = (tic); \ - (q)->t_prev = (tic); \ - } else { \ - (tic)->t_prev = (tic)->t_next = (tic); \ - (q) = (tic); \ - } \ - (tic)->t_flags |= XLOG_TIC_IN_Q; \ - } -#define XLOG_DEL_TICKETQ(q,tic) \ - { \ - if ((tic) == (tic)->t_next) { \ - (q) = NULL; \ - } else { \ - (q) = (tic)->t_next; \ - (tic)->t_next->t_prev = (tic)->t_prev; \ - (tic)->t_prev->t_next = (tic)->t_next; \ - } \ - (tic)->t_next = (tic)->t_prev = NULL; \ - (tic)->t_flags &= ~XLOG_TIC_IN_Q; \ - } - - #define GRANT_LOCK(log) mutex_spinlock(&(log)->l_grant_lock) #define GRANT_UNLOCK(log, s) mutex_spinunlock(&(log)->l_grant_lock, s) #define LOG_LOCK(log) mutex_spinlock(&(log)->l_icloglock) @@ -576,6 +492,65 @@ typedef struct log { * alignment mask */ } xlog_t; +#define XLOG_FORCED_SHUTDOWN(log) ((log)->l_flags & XLOG_IO_ERROR) + +#define XLOG_GRANT_SUB_SPACE(log,bytes,type) \ + { \ + if (type == 'w') { \ + (log)->l_grant_write_bytes -= (bytes); \ + if ((log)->l_grant_write_bytes < 0) { \ + (log)->l_grant_write_bytes += (log)->l_logsize; \ + (log)->l_grant_write_cycle--; \ + } \ + } else { \ + (log)->l_grant_reserve_bytes -= (bytes); \ + if ((log)->l_grant_reserve_bytes < 0) { \ + (log)->l_grant_reserve_bytes += (log)->l_logsize;\ + (log)->l_grant_reserve_cycle--; \ + } \ + } \ + } +#define XLOG_GRANT_ADD_SPACE(log,bytes,type) \ + { \ + if (type == 'w') { \ + (log)->l_grant_write_bytes += (bytes); \ + if ((log)->l_grant_write_bytes > (log)->l_logsize) { \ + (log)->l_grant_write_bytes -= (log)->l_logsize; \ + (log)->l_grant_write_cycle++; \ + } \ + } else { \ + (log)->l_grant_reserve_bytes += (bytes); \ + if ((log)->l_grant_reserve_bytes > (log)->l_logsize) { \ + (log)->l_grant_reserve_bytes -= (log)->l_logsize;\ + (log)->l_grant_reserve_cycle++; \ + } \ + } \ + } +#define XLOG_INS_TICKETQ(q, tic) \ + { \ + if (q) { \ + (tic)->t_next = (q); \ + (tic)->t_prev = (q)->t_prev; \ + (q)->t_prev->t_next = (tic); \ + (q)->t_prev = (tic); \ + } else { \ + (tic)->t_prev = (tic)->t_next = (tic); \ + (q) = (tic); \ + } \ + (tic)->t_flags |= XLOG_TIC_IN_Q; \ + } +#define XLOG_DEL_TICKETQ(q, tic) \ + { \ + if ((tic) == (tic)->t_next) { \ + (q) = NULL; \ + } else { \ + (q) = (tic)->t_next; \ + (tic)->t_next->t_prev = (tic)->t_prev; \ + (tic)->t_prev->t_next = (tic)->t_next; \ + } \ + (tic)->t_next = (tic)->t_prev = NULL; \ + (tic)->t_flags &= ~XLOG_TIC_IN_Q; \ + } /* common routines */ extern xfs_lsn_t xlog_assign_tail_lsn(struct xfs_mount *mp); diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c index 14faabaabf29..8ab7df768063 100644 --- a/fs/xfs/xfs_log_recover.c +++ b/fs/xfs/xfs_log_recover.c @@ -1,66 +1,51 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" -#include "xfs_ag.h" -#include "xfs_sb.h" +#include "xfs_inum.h" #include "xfs_trans.h" +#include "xfs_sb.h" +#include "xfs_ag.h" #include "xfs_dir.h" #include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" #include "xfs_error.h" #include "xfs_bmap_btree.h" -#include "xfs_alloc.h" -#include "xfs_attr_sf.h" +#include "xfs_alloc_btree.h" +#include "xfs_ialloc_btree.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" -#include "xfs_imap.h" -#include "xfs_inode_item.h" #include "xfs_inode.h" -#include "xfs_ialloc_btree.h" +#include "xfs_inode_item.h" +#include "xfs_imap.h" +#include "xfs_alloc.h" #include "xfs_ialloc.h" #include "xfs_log_priv.h" #include "xfs_buf_item.h" -#include "xfs_alloc_btree.h" #include "xfs_log_recover.h" #include "xfs_extfree_item.h" #include "xfs_trans_priv.h" -#include "xfs_bit.h" #include "xfs_quota.h" #include "xfs_rw.h" @@ -2013,79 +1998,74 @@ xfs_qm_dqcheck( * This is all fine; things are still consistent, and we haven't lost * any quota information. Just don't complain about bad dquot blks. */ - if (INT_GET(ddq->d_magic, ARCH_CONVERT) != XFS_DQUOT_MAGIC) { + if (be16_to_cpu(ddq->d_magic) != XFS_DQUOT_MAGIC) { if (flags & XFS_QMOPT_DOWARN) cmn_err(CE_ALERT, "%s : XFS dquot ID 0x%x, magic 0x%x != 0x%x", - str, id, - INT_GET(ddq->d_magic, ARCH_CONVERT), XFS_DQUOT_MAGIC); + str, id, be16_to_cpu(ddq->d_magic), XFS_DQUOT_MAGIC); errs++; } - if (INT_GET(ddq->d_version, ARCH_CONVERT) != XFS_DQUOT_VERSION) { + if (ddq->d_version != XFS_DQUOT_VERSION) { if (flags & XFS_QMOPT_DOWARN) cmn_err(CE_ALERT, "%s : XFS dquot ID 0x%x, version 0x%x != 0x%x", - str, id, - INT_GET(ddq->d_magic, ARCH_CONVERT), XFS_DQUOT_VERSION); + str, id, ddq->d_version, XFS_DQUOT_VERSION); errs++; } - if (INT_GET(ddq->d_flags, ARCH_CONVERT) != XFS_DQ_USER && - INT_GET(ddq->d_flags, ARCH_CONVERT) != XFS_DQ_PROJ && - INT_GET(ddq->d_flags, ARCH_CONVERT) != XFS_DQ_GROUP) { + if (ddq->d_flags != XFS_DQ_USER && + ddq->d_flags != XFS_DQ_PROJ && + ddq->d_flags != XFS_DQ_GROUP) { if (flags & XFS_QMOPT_DOWARN) cmn_err(CE_ALERT, "%s : XFS dquot ID 0x%x, unknown flags 0x%x", - str, id, INT_GET(ddq->d_flags, ARCH_CONVERT)); + str, id, ddq->d_flags); errs++; } - if (id != -1 && id != INT_GET(ddq->d_id, ARCH_CONVERT)) { + if (id != -1 && id != be32_to_cpu(ddq->d_id)) { if (flags & XFS_QMOPT_DOWARN) cmn_err(CE_ALERT, "%s : ondisk-dquot 0x%p, ID mismatch: " "0x%x expected, found id 0x%x", - str, ddq, id, INT_GET(ddq->d_id, ARCH_CONVERT)); + str, ddq, id, be32_to_cpu(ddq->d_id)); errs++; } if (!errs && ddq->d_id) { - if (INT_GET(ddq->d_blk_softlimit, ARCH_CONVERT) && - INT_GET(ddq->d_bcount, ARCH_CONVERT) >= - INT_GET(ddq->d_blk_softlimit, ARCH_CONVERT)) { + if (ddq->d_blk_softlimit && + be64_to_cpu(ddq->d_bcount) >= + be64_to_cpu(ddq->d_blk_softlimit)) { if (!ddq->d_btimer) { if (flags & XFS_QMOPT_DOWARN) cmn_err(CE_ALERT, "%s : Dquot ID 0x%x (0x%p) " "BLK TIMER NOT STARTED", - str, (int) - INT_GET(ddq->d_id, ARCH_CONVERT), ddq); + str, (int)be32_to_cpu(ddq->d_id), ddq); errs++; } } - if (INT_GET(ddq->d_ino_softlimit, ARCH_CONVERT) && - INT_GET(ddq->d_icount, ARCH_CONVERT) >= - INT_GET(ddq->d_ino_softlimit, ARCH_CONVERT)) { + if (ddq->d_ino_softlimit && + be64_to_cpu(ddq->d_icount) >= + be64_to_cpu(ddq->d_ino_softlimit)) { if (!ddq->d_itimer) { if (flags & XFS_QMOPT_DOWARN) cmn_err(CE_ALERT, "%s : Dquot ID 0x%x (0x%p) " "INODE TIMER NOT STARTED", - str, (int) - INT_GET(ddq->d_id, ARCH_CONVERT), ddq); + str, (int)be32_to_cpu(ddq->d_id), ddq); errs++; } } - if (INT_GET(ddq->d_rtb_softlimit, ARCH_CONVERT) && - INT_GET(ddq->d_rtbcount, ARCH_CONVERT) >= - INT_GET(ddq->d_rtb_softlimit, ARCH_CONVERT)) { + if (ddq->d_rtb_softlimit && + be64_to_cpu(ddq->d_rtbcount) >= + be64_to_cpu(ddq->d_rtb_softlimit)) { if (!ddq->d_rtbtimer) { if (flags & XFS_QMOPT_DOWARN) cmn_err(CE_ALERT, "%s : Dquot ID 0x%x (0x%p) " "RTBLK TIMER NOT STARTED", - str, (int) - INT_GET(ddq->d_id, ARCH_CONVERT), ddq); + str, (int)be32_to_cpu(ddq->d_id), ddq); errs++; } } @@ -2103,10 +2083,11 @@ xfs_qm_dqcheck( ASSERT(id != -1); ASSERT(flags & XFS_QMOPT_DQREPAIR); memset(d, 0, sizeof(xfs_dqblk_t)); - INT_SET(d->dd_diskdq.d_magic, ARCH_CONVERT, XFS_DQUOT_MAGIC); - INT_SET(d->dd_diskdq.d_version, ARCH_CONVERT, XFS_DQUOT_VERSION); - INT_SET(d->dd_diskdq.d_id, ARCH_CONVERT, id); - INT_SET(d->dd_diskdq.d_flags, ARCH_CONVERT, type); + + d->dd_diskdq.d_magic = cpu_to_be16(XFS_DQUOT_MAGIC); + d->dd_diskdq.d_version = XFS_DQUOT_VERSION; + d->dd_diskdq.d_flags = type; + d->dd_diskdq.d_id = cpu_to_be32(id); return errs; } @@ -2226,8 +2207,9 @@ xlog_recover_do_buffer_trans( break; default: xfs_fs_cmn_err(CE_ALERT, log->l_mp, - "xfs_log_recover: unknown buffer type 0x%x, dev %s", - buf_f->blf_type, XFS_BUFTARG_NAME(log->l_targ)); + "xfs_log_recover: unknown buffer type 0x%x, logdev %s", + buf_f->blf_type, log->l_mp->m_logname ? + log->l_mp->m_logname : "internal"); XFS_ERROR_REPORT("xlog_recover_do_buffer_trans", XFS_ERRLEVEL_LOW, log->l_mp); return XFS_ERROR(EFSCORRUPTED); @@ -3178,13 +3160,12 @@ xlog_recover_clear_agi_bucket( } agi = XFS_BUF_TO_AGI(agibp); - if (INT_GET(agi->agi_magicnum, ARCH_CONVERT) != XFS_AGI_MAGIC) { + if (be32_to_cpu(agi->agi_magicnum) != XFS_AGI_MAGIC) { xfs_trans_cancel(tp, XFS_TRANS_ABORT); return; } - ASSERT(INT_GET(agi->agi_magicnum, ARCH_CONVERT) == XFS_AGI_MAGIC); - INT_SET(agi->agi_unlinked[bucket], ARCH_CONVERT, NULLAGINO); + agi->agi_unlinked[bucket] = cpu_to_be32(NULLAGINO); offset = offsetof(xfs_agi_t, agi_unlinked) + (sizeof(xfs_agino_t) * bucket); xfs_trans_log_buf(tp, agibp, offset, @@ -3243,12 +3224,11 @@ xlog_recover_process_iunlinks( XFS_AG_DADDR(mp, agno, XFS_AGI_DADDR(mp))); } agi = XFS_BUF_TO_AGI(agibp); - ASSERT(XFS_AGI_MAGIC == - INT_GET(agi->agi_magicnum, ARCH_CONVERT)); + ASSERT(XFS_AGI_MAGIC == be32_to_cpu(agi->agi_magicnum)); for (bucket = 0; bucket < XFS_AGI_UNLINKED_BUCKETS; bucket++) { - agino = INT_GET(agi->agi_unlinked[bucket], ARCH_CONVERT); + agino = be32_to_cpu(agi->agi_unlinked[bucket]); while (agino != NULLAGINO) { /* @@ -3336,8 +3316,8 @@ xlog_recover_process_iunlinks( XFS_AGI_DADDR(mp))); } agi = XFS_BUF_TO_AGI(agibp); - ASSERT(XFS_AGI_MAGIC == INT_GET( - agi->agi_magicnum, ARCH_CONVERT)); + ASSERT(XFS_AGI_MAGIC == be32_to_cpu( + agi->agi_magicnum)); } } @@ -3938,8 +3918,9 @@ xlog_recover( } cmn_err(CE_NOTE, - "Starting XFS recovery on filesystem: %s (dev: %s)", - log->l_mp->m_fsname, XFS_BUFTARG_NAME(log->l_targ)); + "Starting XFS recovery on filesystem: %s (logdev: %s)", + log->l_mp->m_fsname, log->l_mp->m_logname ? + log->l_mp->m_logname : "internal"); error = xlog_do_recover(log, head_blk, tail_blk); log->l_flags |= XLOG_RECOVERY_NEEDED; @@ -3987,8 +3968,9 @@ xlog_recover_finish( xlog_recover_check_summary(log); cmn_err(CE_NOTE, - "Ending XFS recovery on filesystem: %s (dev: %s)", - log->l_mp->m_fsname, XFS_BUFTARG_NAME(log->l_targ)); + "Ending XFS recovery on filesystem: %s (logdev: %s)", + log->l_mp->m_fsname, log->l_mp->m_logname ? + log->l_mp->m_logname : "internal"); log->l_flags &= ~XLOG_RECOVERY_NEEDED; } else { cmn_err(CE_DEBUG, @@ -4038,14 +4020,12 @@ xlog_recover_check_summary( mp, agfbp, agfdaddr); } agfp = XFS_BUF_TO_AGF(agfbp); - ASSERT(XFS_AGF_MAGIC == - INT_GET(agfp->agf_magicnum, ARCH_CONVERT)); - ASSERT(XFS_AGF_GOOD_VERSION( - INT_GET(agfp->agf_versionnum, ARCH_CONVERT))); - ASSERT(INT_GET(agfp->agf_seqno, ARCH_CONVERT) == agno); - - freeblks += INT_GET(agfp->agf_freeblks, ARCH_CONVERT) + - INT_GET(agfp->agf_flcount, ARCH_CONVERT); + ASSERT(XFS_AGF_MAGIC == be32_to_cpu(agfp->agf_magicnum)); + ASSERT(XFS_AGF_GOOD_VERSION(be32_to_cpu(agfp->agf_versionnum))); + ASSERT(be32_to_cpu(agfp->agf_seqno) == agno); + + freeblks += be32_to_cpu(agfp->agf_freeblks) + + be32_to_cpu(agfp->agf_flcount); xfs_buf_relse(agfbp); agidaddr = XFS_AG_DADDR(mp, agno, XFS_AGI_DADDR(mp)); @@ -4056,14 +4036,12 @@ xlog_recover_check_summary( mp, agibp, agidaddr); } agip = XFS_BUF_TO_AGI(agibp); - ASSERT(XFS_AGI_MAGIC == - INT_GET(agip->agi_magicnum, ARCH_CONVERT)); - ASSERT(XFS_AGI_GOOD_VERSION( - INT_GET(agip->agi_versionnum, ARCH_CONVERT))); - ASSERT(INT_GET(agip->agi_seqno, ARCH_CONVERT) == agno); - - itotal += INT_GET(agip->agi_count, ARCH_CONVERT); - ifree += INT_GET(agip->agi_freecount, ARCH_CONVERT); + ASSERT(XFS_AGI_MAGIC == be32_to_cpu(agip->agi_magicnum)); + ASSERT(XFS_AGI_GOOD_VERSION(be32_to_cpu(agip->agi_versionnum))); + ASSERT(be32_to_cpu(agip->agi_seqno) == agno); + + itotal += be32_to_cpu(agip->agi_count); + ifree += be32_to_cpu(agip->agi_freecount); xfs_buf_relse(agibp); } diff --git a/fs/xfs/xfs_log_recover.h b/fs/xfs/xfs_log_recover.h index 42158b442b55..b22545555301 100644 --- a/fs/xfs/xfs_log_recover.h +++ b/fs/xfs/xfs_log_recover.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_LOG_RECOVER_H__ #define __XFS_LOG_RECOVER_H__ diff --git a/fs/xfs/xfs_mac.h b/fs/xfs/xfs_mac.h index 8d59aaffeb8e..18e0e98e03d0 100644 --- a/fs/xfs/xfs_mac.h +++ b/fs/xfs/xfs_mac.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2001-2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2001-2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_MAC_H__ #define __XFS_MAC_H__ diff --git a/fs/xfs/xfs_macros.c b/fs/xfs/xfs_macros.c deleted file mode 100644 index 698c2cd62858..000000000000 --- a/fs/xfs/xfs_macros.c +++ /dev/null @@ -1,2141 +0,0 @@ -/* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. - * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ - */ - -#define XFS_MACRO_C - -#include "xfs.h" -#include "xfs_macros.h" -#include "xfs_types.h" -#include "xfs_inum.h" -#include "xfs_log.h" -#include "xfs_trans.h" -#include "xfs_sb.h" -#include "xfs_ag.h" -#include "xfs_dir.h" -#include "xfs_dir2.h" -#include "xfs_dmapi.h" -#include "xfs_mount.h" -#include "xfs_alloc_btree.h" -#include "xfs_ialloc_btree.h" -#include "xfs_bmap_btree.h" -#include "xfs_btree.h" -#include "xfs_attr_sf.h" -#include "xfs_dir_sf.h" -#include "xfs_dir2_sf.h" -#include "xfs_dinode.h" -#include "xfs_ialloc.h" -#include "xfs_inode_item.h" -#include "xfs_inode.h" -#include "xfs_bmap.h" -#include "xfs_rw.h" -#include "xfs_log_priv.h" -#include "xfs_da_btree.h" -#include "xfs_attr_leaf.h" -#include "xfs_dir_leaf.h" -#include "xfs_dir2_data.h" -#include "xfs_dir2_leaf.h" -#include "xfs_dir2_block.h" -#include "xfs_dir2_node.h" -#include "xfs_bit.h" - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_ISNULLDSTARTBLOCK) -int -isnulldstartblock(xfs_dfsbno_t x) -{ - return ISNULLDSTARTBLOCK(x); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_ISNULLSTARTBLOCK) -int -isnullstartblock(xfs_fsblock_t x) -{ - return ISNULLSTARTBLOCK(x); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_NULLSTARTBLOCK) -xfs_fsblock_t -nullstartblock(int k) -{ - return NULLSTARTBLOCK(k); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_STARTBLOCKVAL) -xfs_filblks_t -startblockval(xfs_fsblock_t x) -{ - return STARTBLOCKVAL(x); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_AG_CHECK_DADDR) -void -xfs_ag_check_daddr(xfs_mount_t *mp, xfs_daddr_t d, xfs_extlen_t len) -{ - XFS_AG_CHECK_DADDR(mp, d, len); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_AG_DADDR) -xfs_daddr_t -xfs_ag_daddr(xfs_mount_t *mp, xfs_agnumber_t agno, xfs_daddr_t d) -{ - return XFS_AG_DADDR(mp, agno, d); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_AG_MAXLEVELS) -int -xfs_ag_maxlevels(xfs_mount_t *mp) -{ - return XFS_AG_MAXLEVELS(mp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_AGB_TO_DADDR) -xfs_daddr_t -xfs_agb_to_daddr(xfs_mount_t *mp, xfs_agnumber_t agno, xfs_agblock_t agbno) -{ - return XFS_AGB_TO_DADDR(mp, agno, agbno); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_AGB_TO_FSB) -xfs_fsblock_t -xfs_agb_to_fsb(xfs_mount_t *mp, xfs_agnumber_t agno, xfs_agblock_t agbno) -{ - return XFS_AGB_TO_FSB(mp, agno, agbno); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_AGBLOCK_MAX) -xfs_agblock_t -xfs_agblock_max(xfs_agblock_t a, xfs_agblock_t b) -{ - return XFS_AGBLOCK_MAX(a, b); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_AGBLOCK_MIN) -xfs_agblock_t -xfs_agblock_min(xfs_agblock_t a, xfs_agblock_t b) -{ - return XFS_AGBLOCK_MIN(a, b); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_AGF_BLOCK) -xfs_agblock_t -xfs_agf_block(xfs_mount_t *mp) -{ - return XFS_AGF_BLOCK(mp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_AGF_GOOD_VERSION) -int -xfs_agf_good_version(unsigned v) -{ - return XFS_AGF_GOOD_VERSION(v); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_AGFL_BLOCK) -xfs_agblock_t -xfs_agfl_block(xfs_mount_t *mp) -{ - return XFS_AGFL_BLOCK(mp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_AGI_BLOCK) -xfs_agblock_t -xfs_agi_block(xfs_mount_t *mp) -{ - return XFS_AGI_BLOCK(mp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_AGI_GOOD_VERSION) -int -xfs_agi_good_version(unsigned v) -{ - return XFS_AGI_GOOD_VERSION(v); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_AGINO_TO_AGBNO) -xfs_agblock_t -xfs_agino_to_agbno(xfs_mount_t *mp, xfs_agino_t i) -{ - return XFS_AGINO_TO_AGBNO(mp, i); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_AGINO_TO_INO) -xfs_ino_t -xfs_agino_to_ino(xfs_mount_t *mp, xfs_agnumber_t a, xfs_agino_t i) -{ - return XFS_AGINO_TO_INO(mp, a, i); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_AGINO_TO_OFFSET) -int -xfs_agino_to_offset(xfs_mount_t *mp, xfs_agino_t i) -{ - return XFS_AGINO_TO_OFFSET(mp, i); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_ALLOC_BLOCK_MAXRECS) -int -xfs_alloc_block_maxrecs(int lev, xfs_btree_cur_t *cur) -{ - return XFS_ALLOC_BLOCK_MAXRECS(lev, cur); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_ALLOC_BLOCK_MINRECS) -int -xfs_alloc_block_minrecs(int lev, xfs_btree_cur_t *cur) -{ - return XFS_ALLOC_BLOCK_MINRECS(lev, cur); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_ALLOC_BLOCK_SIZE) -/*ARGSUSED1*/ -int -xfs_alloc_block_size(int lev, xfs_btree_cur_t *cur) -{ - return XFS_ALLOC_BLOCK_SIZE(lev, cur); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_ALLOC_KEY_ADDR) -/*ARGSUSED3*/ -xfs_alloc_key_t * -xfs_alloc_key_addr(xfs_alloc_block_t *bb, int i, xfs_btree_cur_t *cur) -{ - return XFS_ALLOC_KEY_ADDR(bb, i, cur); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_ALLOC_PTR_ADDR) -xfs_alloc_ptr_t * -xfs_alloc_ptr_addr(xfs_alloc_block_t *bb, int i, xfs_btree_cur_t *cur) -{ - return XFS_ALLOC_PTR_ADDR(bb, i, cur); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_ALLOC_REC_ADDR) -/*ARGSUSED3*/ -xfs_alloc_rec_t * -xfs_alloc_rec_addr(xfs_alloc_block_t *bb, int i, xfs_btree_cur_t *cur) -{ - return XFS_ALLOC_REC_ADDR(bb, i, cur); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_ATTR_LEAF_ENTSIZE_LOCAL) -int -xfs_attr_leaf_entsize_local(int nlen, int vlen) -{ - return XFS_ATTR_LEAF_ENTSIZE_LOCAL(nlen, vlen); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_ATTR_LEAF_ENTSIZE_LOCAL_MAX) -int -xfs_attr_leaf_entsize_local_max(int bsize) -{ - return XFS_ATTR_LEAF_ENTSIZE_LOCAL_MAX(bsize); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_ATTR_LEAF_ENTSIZE_REMOTE) -int -xfs_attr_leaf_entsize_remote(int nlen) -{ - return XFS_ATTR_LEAF_ENTSIZE_REMOTE(nlen); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_ATTR_LEAF_NAME) -char * -xfs_attr_leaf_name(xfs_attr_leafblock_t *leafp, int idx) -{ - return XFS_ATTR_LEAF_NAME(leafp, idx); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_ATTR_LEAF_NAME_LOCAL) -xfs_attr_leaf_name_local_t * -xfs_attr_leaf_name_local(xfs_attr_leafblock_t *leafp, int idx) -{ - return XFS_ATTR_LEAF_NAME_LOCAL(leafp, idx); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_ATTR_LEAF_NAME_REMOTE) -xfs_attr_leaf_name_remote_t * -xfs_attr_leaf_name_remote(xfs_attr_leafblock_t *leafp, int idx) -{ - return XFS_ATTR_LEAF_NAME_REMOTE(leafp, idx); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_ATTR_SF_ENTSIZE) -int -xfs_attr_sf_entsize(xfs_attr_sf_entry_t *sfep) -{ - return XFS_ATTR_SF_ENTSIZE(sfep); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_ATTR_SF_ENTSIZE_BYNAME) -int -xfs_attr_sf_entsize_byname(int nlen, int vlen) -{ - return XFS_ATTR_SF_ENTSIZE_BYNAME(nlen, vlen); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_ATTR_SF_NEXTENTRY) -xfs_attr_sf_entry_t * -xfs_attr_sf_nextentry(xfs_attr_sf_entry_t *sfep) -{ - return XFS_ATTR_SF_NEXTENTRY(sfep); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_ATTR_SF_TOTSIZE) -int -xfs_attr_sf_totsize(xfs_inode_t *dp) -{ - return XFS_ATTR_SF_TOTSIZE(dp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BHVTOI) -xfs_inode_t * -xfs_bhvtoi(bhv_desc_t *bhvp) -{ - return XFS_BHVTOI(bhvp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BHVTOM) -xfs_mount_t * -xfs_bhvtom(bhv_desc_t *bdp) -{ - return XFS_BHVTOM(bdp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_VFSTOM) -xfs_mount_t * -xfs_vfstom(vfs_t *vfs) -{ - return XFS_VFSTOM(vfs); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BM_MAXLEVELS) -int -xfs_bm_maxlevels(xfs_mount_t *mp, int w) -{ - return XFS_BM_MAXLEVELS(mp, w); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BMAP_BLOCK_DMAXRECS) -int -xfs_bmap_block_dmaxrecs(int lev, xfs_btree_cur_t *cur) -{ - return XFS_BMAP_BLOCK_DMAXRECS(lev, cur); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BMAP_BLOCK_DMINRECS) -int -xfs_bmap_block_dminrecs(int lev, xfs_btree_cur_t *cur) -{ - return XFS_BMAP_BLOCK_DMINRECS(lev, cur); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BMAP_BLOCK_DSIZE) -int -xfs_bmap_block_dsize(int lev, xfs_btree_cur_t *cur) -{ - return XFS_BMAP_BLOCK_DSIZE(lev, cur); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BMAP_BLOCK_IMAXRECS) -int -xfs_bmap_block_imaxrecs(int lev, xfs_btree_cur_t *cur) -{ - return XFS_BMAP_BLOCK_IMAXRECS(lev, cur); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BMAP_BLOCK_IMINRECS) -int -xfs_bmap_block_iminrecs(int lev, xfs_btree_cur_t *cur) -{ - return XFS_BMAP_BLOCK_IMINRECS(lev, cur); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BMAP_BLOCK_ISIZE) -int -xfs_bmap_block_isize(int lev, xfs_btree_cur_t *cur) -{ - return XFS_BMAP_BLOCK_ISIZE(lev, cur); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BMAP_BROOT_KEY_ADDR) -/*ARGSUSED3*/ -xfs_bmbt_key_t * -xfs_bmap_broot_key_addr(xfs_bmbt_block_t *bb, int i, int sz) -{ - return XFS_BMAP_BROOT_KEY_ADDR(bb, i, sz); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BMAP_BROOT_MAXRECS) -int -xfs_bmap_broot_maxrecs(int sz) -{ - return XFS_BMAP_BROOT_MAXRECS(sz); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BMAP_BROOT_NUMRECS) -int -xfs_bmap_broot_numrecs(xfs_bmdr_block_t *bb) -{ - return XFS_BMAP_BROOT_NUMRECS(bb); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BMAP_BROOT_PTR_ADDR) -xfs_bmbt_ptr_t * -xfs_bmap_broot_ptr_addr(xfs_bmbt_block_t *bb, int i, int sz) -{ - return XFS_BMAP_BROOT_PTR_ADDR(bb, i, sz); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BMAP_BROOT_REC_ADDR) -/*ARGSUSED3*/ -xfs_bmbt_rec_t * -xfs_bmap_broot_rec_addr(xfs_bmbt_block_t *bb, int i, int sz) -{ - return XFS_BMAP_BROOT_REC_ADDR(bb, i, sz); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BMAP_BROOT_SPACE) -int -xfs_bmap_broot_space(xfs_bmdr_block_t *bb) -{ - return XFS_BMAP_BROOT_SPACE(bb); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BMAP_BROOT_SPACE_CALC) -int -xfs_bmap_broot_space_calc(int nrecs) -{ - return XFS_BMAP_BROOT_SPACE_CALC(nrecs); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BMAP_IBLOCK_SIZE) -/*ARGSUSED1*/ -int -xfs_bmap_iblock_size(int lev, xfs_btree_cur_t *cur) -{ - return XFS_BMAP_IBLOCK_SIZE(lev, cur); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BMAP_INIT) -void -xfs_bmap_init(xfs_bmap_free_t *flp, xfs_fsblock_t *fbp) -{ - XFS_BMAP_INIT(flp, fbp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BMAP_KEY_DADDR) -/*ARGSUSED3*/ -xfs_bmbt_key_t * -xfs_bmap_key_daddr(xfs_bmbt_block_t *bb, int i, xfs_btree_cur_t *cur) -{ - return XFS_BMAP_KEY_DADDR(bb, i, cur); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BMAP_KEY_IADDR) -/*ARGSUSED3*/ -xfs_bmbt_key_t * -xfs_bmap_key_iaddr(xfs_bmbt_block_t *bb, int i, xfs_btree_cur_t *cur) -{ - return XFS_BMAP_KEY_IADDR(bb, i, cur); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BMAP_PTR_DADDR) -xfs_bmbt_ptr_t * -xfs_bmap_ptr_daddr(xfs_bmbt_block_t *bb, int i, xfs_btree_cur_t *cur) -{ - return XFS_BMAP_PTR_DADDR(bb, i, cur); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BMAP_PTR_IADDR) -xfs_bmbt_ptr_t * -xfs_bmap_ptr_iaddr(xfs_bmbt_block_t *bb, int i, xfs_btree_cur_t *cur) -{ - return XFS_BMAP_PTR_IADDR(bb, i, cur); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BMAP_RBLOCK_DSIZE) -/*ARGSUSED1*/ -int -xfs_bmap_rblock_dsize(int lev, xfs_btree_cur_t *cur) -{ - return XFS_BMAP_RBLOCK_DSIZE(lev, cur); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BMAP_RBLOCK_ISIZE) -/*ARGSUSED1*/ -int -xfs_bmap_rblock_isize(int lev, xfs_btree_cur_t *cur) -{ - return XFS_BMAP_RBLOCK_ISIZE(lev, cur); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BMAP_REC_DADDR) -/*ARGSUSED3*/ -xfs_bmbt_rec_t * -xfs_bmap_rec_daddr(xfs_bmbt_block_t *bb, int i, xfs_btree_cur_t *cur) -{ - return XFS_BMAP_REC_DADDR(bb, i, cur); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BMAP_REC_IADDR) -/*ARGSUSED3*/ -xfs_bmbt_rec_t * -xfs_bmap_rec_iaddr(xfs_bmbt_block_t *bb, int i, xfs_btree_cur_t *cur) -{ - return XFS_BMAP_REC_IADDR(bb, i, cur); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BMAP_SANITY_CHECK) -int -xfs_bmap_sanity_check(xfs_mount_t *mp, xfs_bmbt_block_t *bb, int level) -{ - return XFS_BMAP_SANITY_CHECK(mp, bb, level); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BMAPI_AFLAG) -int -xfs_bmapi_aflag(int w) -{ - return XFS_BMAPI_AFLAG(w); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BMDR_SPACE_CALC) -int -xfs_bmdr_space_calc(int nrecs) -{ - return XFS_BMDR_SPACE_CALC(nrecs); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BNO_BLOCK) -xfs_agblock_t -xfs_bno_block(xfs_mount_t *mp) -{ - return XFS_BNO_BLOCK(mp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BTREE_LONG_PTRS) -int -xfs_btree_long_ptrs(xfs_btnum_t btnum) -{ - return XFS_BTREE_LONG_PTRS(btnum); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BUF_TO_AGF) -xfs_agf_t * -xfs_buf_to_agf(xfs_buf_t *bp) -{ - return XFS_BUF_TO_AGF(bp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BUF_TO_AGFL) -xfs_agfl_t * -xfs_buf_to_agfl(xfs_buf_t *bp) -{ - return XFS_BUF_TO_AGFL(bp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BUF_TO_AGI) -xfs_agi_t * -xfs_buf_to_agi(xfs_buf_t *bp) -{ - return XFS_BUF_TO_AGI(bp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BUF_TO_ALLOC_BLOCK) -xfs_alloc_block_t * -xfs_buf_to_alloc_block(xfs_buf_t *bp) -{ - return XFS_BUF_TO_ALLOC_BLOCK(bp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BUF_TO_BLOCK) -xfs_btree_block_t * -xfs_buf_to_block(xfs_buf_t *bp) -{ - return XFS_BUF_TO_BLOCK(bp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BUF_TO_BMBT_BLOCK) -xfs_bmbt_block_t * -xfs_buf_to_bmbt_block(xfs_buf_t *bp) -{ - return XFS_BUF_TO_BMBT_BLOCK(bp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BUF_TO_DINODE) -xfs_dinode_t * -xfs_buf_to_dinode(xfs_buf_t *bp) -{ - return XFS_BUF_TO_DINODE(bp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BUF_TO_INOBT_BLOCK) -xfs_inobt_block_t * -xfs_buf_to_inobt_block(xfs_buf_t *bp) -{ - return XFS_BUF_TO_INOBT_BLOCK(bp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BUF_TO_LBLOCK) -xfs_btree_lblock_t * -xfs_buf_to_lblock(xfs_buf_t *bp) -{ - return XFS_BUF_TO_LBLOCK(bp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BUF_TO_SBLOCK) -xfs_btree_sblock_t * -xfs_buf_to_sblock(xfs_buf_t *bp) -{ - return XFS_BUF_TO_SBLOCK(bp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_BUF_TO_SBP) -xfs_sb_t * -xfs_buf_to_sbp(xfs_buf_t *bp) -{ - return XFS_BUF_TO_SBP(bp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_CFORK_ASIZE) -int -xfs_cfork_asize_disk(xfs_dinode_core_t *dcp, xfs_mount_t *mp) -{ - return XFS_CFORK_ASIZE_DISK(dcp, mp); -} -int -xfs_cfork_asize(xfs_dinode_core_t *dcp, xfs_mount_t *mp) -{ - return XFS_CFORK_ASIZE(dcp, mp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_CFORK_BOFF) -int -xfs_cfork_boff_disk(xfs_dinode_core_t *dcp) -{ - return XFS_CFORK_BOFF_DISK(dcp); -} -int -xfs_cfork_boff(xfs_dinode_core_t *dcp) -{ - return XFS_CFORK_BOFF(dcp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_CFORK_DSIZE) -int -xfs_cfork_dsize_disk(xfs_dinode_core_t *dcp, xfs_mount_t *mp) -{ - return XFS_CFORK_DSIZE_DISK(dcp, mp); -} -int -xfs_cfork_dsize(xfs_dinode_core_t *dcp, xfs_mount_t *mp) -{ - return XFS_CFORK_DSIZE(dcp, mp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_CFORK_FMT_SET) -void -xfs_cfork_fmt_set(xfs_dinode_core_t *dcp, int w, int n) -{ - XFS_CFORK_FMT_SET(dcp, w, n); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_CFORK_FORMAT) -int -xfs_cfork_format(xfs_dinode_core_t *dcp, int w) -{ - return XFS_CFORK_FORMAT(dcp, w); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_CFORK_NEXT_SET) -void -xfs_cfork_next_set(xfs_dinode_core_t *dcp, int w, int n) -{ - XFS_CFORK_NEXT_SET(dcp, w, n); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_CFORK_NEXTENTS) -int -xfs_cfork_nextents_disk(xfs_dinode_core_t *dcp, int w) -{ - return XFS_CFORK_NEXTENTS_DISK(dcp, w); -} -int -xfs_cfork_nextents(xfs_dinode_core_t *dcp, int w) -{ - return XFS_CFORK_NEXTENTS(dcp, w); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_CFORK_Q) -int -xfs_cfork_q_disk(xfs_dinode_core_t *dcp) -{ - return XFS_CFORK_Q_DISK(dcp); -} -int -xfs_cfork_q(xfs_dinode_core_t *dcp) -{ - return XFS_CFORK_Q(dcp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_CFORK_SIZE) -int -xfs_cfork_size_disk(xfs_dinode_core_t *dcp, xfs_mount_t *mp, int w) -{ - return XFS_CFORK_SIZE_DISK(dcp, mp, w); -} -int -xfs_cfork_size(xfs_dinode_core_t *dcp, xfs_mount_t *mp, int w) -{ - return XFS_CFORK_SIZE(dcp, mp, w); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_CNT_BLOCK) -xfs_agblock_t -xfs_cnt_block(xfs_mount_t *mp) -{ - return XFS_CNT_BLOCK(mp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DA_COOKIE_BNO) -xfs_dablk_t -xfs_da_cookie_bno(xfs_mount_t *mp, xfs_off_t cookie) -{ - return XFS_DA_COOKIE_BNO(mp, cookie); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DA_COOKIE_ENTRY) -int -xfs_da_cookie_entry(xfs_mount_t *mp, xfs_off_t cookie) -{ - return XFS_DA_COOKIE_ENTRY(mp, cookie); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DA_COOKIE_HASH) -/*ARGSUSED1*/ -xfs_dahash_t -xfs_da_cookie_hash(xfs_mount_t *mp, xfs_off_t cookie) -{ - return XFS_DA_COOKIE_HASH(mp, cookie); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DA_MAKE_BNOENTRY) -__uint32_t -xfs_da_make_bnoentry(xfs_mount_t *mp, xfs_dablk_t bno, int entry) -{ - return XFS_DA_MAKE_BNOENTRY(mp, bno, entry); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DA_MAKE_COOKIE) -xfs_off_t -xfs_da_make_cookie(xfs_mount_t *mp, xfs_dablk_t bno, int entry, - xfs_dahash_t hash) -{ - return XFS_DA_MAKE_COOKIE(mp, bno, entry, hash); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DADDR_TO_AGBNO) -xfs_agblock_t -xfs_daddr_to_agbno(xfs_mount_t *mp, xfs_daddr_t d) -{ - return XFS_DADDR_TO_AGBNO(mp, d); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DADDR_TO_AGNO) -xfs_agnumber_t -xfs_daddr_to_agno(xfs_mount_t *mp, xfs_daddr_t d) -{ - return XFS_DADDR_TO_AGNO(mp, d); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DADDR_TO_FSB) -xfs_fsblock_t -xfs_daddr_to_fsb(xfs_mount_t *mp, xfs_daddr_t d) -{ - return XFS_DADDR_TO_FSB(mp, d); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DFORK_APTR) -char * -xfs_dfork_aptr(xfs_dinode_t *dip) -{ - return XFS_DFORK_APTR(dip); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DFORK_ASIZE) -int -xfs_dfork_asize(xfs_dinode_t *dip, xfs_mount_t *mp) -{ - return XFS_DFORK_ASIZE(dip, mp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DFORK_BOFF) -int -xfs_dfork_boff(xfs_dinode_t *dip) -{ - return XFS_DFORK_BOFF(dip); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DFORK_DPTR) -char * -xfs_dfork_dptr(xfs_dinode_t *dip) -{ - return XFS_DFORK_DPTR(dip); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DFORK_DSIZE) -int -xfs_dfork_dsize(xfs_dinode_t *dip, xfs_mount_t *mp) -{ - return XFS_DFORK_DSIZE(dip, mp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DFORK_NEXTENTS) -int -xfs_dfork_nextents(xfs_dinode_t *dip, int w) -{ - return XFS_DFORK_NEXTENTS(dip, w); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DFORK_PTR) -char * -xfs_dfork_ptr(xfs_dinode_t *dip, int w) -{ - return XFS_DFORK_PTR(dip, w); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DFORK_Q) -int -xfs_dfork_q(xfs_dinode_t *dip) -{ - return XFS_DFORK_Q(dip); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DFORK_SIZE) -int -xfs_dfork_size(xfs_dinode_t *dip, xfs_mount_t *mp, int w) -{ - return XFS_DFORK_SIZE(dip, mp, w); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DINODE_GOOD_VERSION) -int -xfs_dinode_good_version(int v) -{ - return XFS_DINODE_GOOD_VERSION(v); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR_LEAF_ENTSIZE_BYENTRY) -int -xfs_dir_leaf_entsize_byentry(xfs_dir_leaf_entry_t *entry) -{ - return XFS_DIR_LEAF_ENTSIZE_BYENTRY(entry); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR_LEAF_ENTSIZE_BYNAME) -int -xfs_dir_leaf_entsize_byname(int len) -{ - return XFS_DIR_LEAF_ENTSIZE_BYNAME(len); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR_LEAF_NAMESTRUCT) -xfs_dir_leaf_name_t * -xfs_dir_leaf_namestruct(xfs_dir_leafblock_t *leafp, int offset) -{ - return XFS_DIR_LEAF_NAMESTRUCT(leafp, offset); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR_SF_ALLFIT) -int -xfs_dir_sf_allfit(int count, int totallen) -{ - return XFS_DIR_SF_ALLFIT(count, totallen); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR_SF_ENTSIZE_BYENTRY) -int -xfs_dir_sf_entsize_byentry(xfs_dir_sf_entry_t *sfep) -{ - return XFS_DIR_SF_ENTSIZE_BYENTRY(sfep); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR_SF_ENTSIZE_BYNAME) -int -xfs_dir_sf_entsize_byname(int len) -{ - return XFS_DIR_SF_ENTSIZE_BYNAME(len); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR_SF_GET_DIRINO) -void -xfs_dir_sf_get_dirino(xfs_dir_ino_t *from, xfs_ino_t *to) -{ - XFS_DIR_SF_GET_DIRINO(from, to); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR_SF_NEXTENTRY) -xfs_dir_sf_entry_t * -xfs_dir_sf_nextentry(xfs_dir_sf_entry_t *sfep) -{ - return XFS_DIR_SF_NEXTENTRY(sfep); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR_SF_PUT_DIRINO) -void -xfs_dir_sf_put_dirino(xfs_ino_t *from, xfs_dir_ino_t *to) -{ - XFS_DIR_SF_PUT_DIRINO(from, to); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_BLOCK_LEAF_P) -xfs_dir2_leaf_entry_t * -xfs_dir2_block_leaf_p(xfs_dir2_block_tail_t *btp) -{ - return XFS_DIR2_BLOCK_LEAF_P(btp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_BLOCK_TAIL_P) -xfs_dir2_block_tail_t * -xfs_dir2_block_tail_p(xfs_mount_t *mp, xfs_dir2_block_t *block) -{ - return XFS_DIR2_BLOCK_TAIL_P(mp, block); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_BYTE_TO_DA) -xfs_dablk_t -xfs_dir2_byte_to_da(xfs_mount_t *mp, xfs_dir2_off_t by) -{ - return XFS_DIR2_BYTE_TO_DA(mp, by); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_BYTE_TO_DATAPTR) -/* ARGSUSED */ -xfs_dir2_dataptr_t -xfs_dir2_byte_to_dataptr(xfs_mount_t *mp, xfs_dir2_off_t by) -{ - return XFS_DIR2_BYTE_TO_DATAPTR(mp, by); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_BYTE_TO_DB) -xfs_dir2_db_t -xfs_dir2_byte_to_db(xfs_mount_t *mp, xfs_dir2_off_t by) -{ - return XFS_DIR2_BYTE_TO_DB(mp, by); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_BYTE_TO_OFF) -xfs_dir2_data_aoff_t -xfs_dir2_byte_to_off(xfs_mount_t *mp, xfs_dir2_off_t by) -{ - return XFS_DIR2_BYTE_TO_OFF(mp, by); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_DA_TO_BYTE) -xfs_dir2_off_t -xfs_dir2_da_to_byte(xfs_mount_t *mp, xfs_dablk_t da) -{ - return XFS_DIR2_DA_TO_BYTE(mp, da); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_DA_TO_DB) -xfs_dir2_db_t -xfs_dir2_da_to_db(xfs_mount_t *mp, xfs_dablk_t da) -{ - return XFS_DIR2_DA_TO_DB(mp, da); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_DATA_ENTRY_TAG_P) -xfs_dir2_data_off_t * -xfs_dir2_data_entry_tag_p(xfs_dir2_data_entry_t *dep) -{ - return XFS_DIR2_DATA_ENTRY_TAG_P(dep); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_DATA_ENTSIZE) -int -xfs_dir2_data_entsize(int n) -{ - return XFS_DIR2_DATA_ENTSIZE(n); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_DATA_UNUSED_TAG_P) -xfs_dir2_data_off_t * -xfs_dir2_data_unused_tag_p(xfs_dir2_data_unused_t *dup) -{ - return XFS_DIR2_DATA_UNUSED_TAG_P(dup); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_DATAPTR_TO_BYTE) -/* ARGSUSED */ -xfs_dir2_off_t -xfs_dir2_dataptr_to_byte(xfs_mount_t *mp, xfs_dir2_dataptr_t dp) -{ - return XFS_DIR2_DATAPTR_TO_BYTE(mp, dp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_DATAPTR_TO_DB) -xfs_dir2_db_t -xfs_dir2_dataptr_to_db(xfs_mount_t *mp, xfs_dir2_dataptr_t dp) -{ - return XFS_DIR2_DATAPTR_TO_DB(mp, dp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_DATAPTR_TO_OFF) -xfs_dir2_data_aoff_t -xfs_dir2_dataptr_to_off(xfs_mount_t *mp, xfs_dir2_dataptr_t dp) -{ - return XFS_DIR2_DATAPTR_TO_OFF(mp, dp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_DB_OFF_TO_BYTE) -xfs_dir2_off_t -xfs_dir2_db_off_to_byte(xfs_mount_t *mp, xfs_dir2_db_t db, - xfs_dir2_data_aoff_t o) -{ - return XFS_DIR2_DB_OFF_TO_BYTE(mp, db, o); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_DB_OFF_TO_DATAPTR) -xfs_dir2_dataptr_t -xfs_dir2_db_off_to_dataptr(xfs_mount_t *mp, xfs_dir2_db_t db, - xfs_dir2_data_aoff_t o) -{ - return XFS_DIR2_DB_OFF_TO_DATAPTR(mp, db, o); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_DB_TO_DA) -xfs_dablk_t -xfs_dir2_db_to_da(xfs_mount_t *mp, xfs_dir2_db_t db) -{ - return XFS_DIR2_DB_TO_DA(mp, db); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_DB_TO_FDB) -xfs_dir2_db_t -xfs_dir2_db_to_fdb(xfs_mount_t *mp, xfs_dir2_db_t db) -{ - return XFS_DIR2_DB_TO_FDB(mp, db); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_DB_TO_FDINDEX) -int -xfs_dir2_db_to_fdindex(xfs_mount_t *mp, xfs_dir2_db_t db) -{ - return XFS_DIR2_DB_TO_FDINDEX(mp, db); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_LEAF_BESTS_P) -xfs_dir2_data_off_t * -xfs_dir2_leaf_bests_p(xfs_dir2_leaf_tail_t *ltp) -{ - return XFS_DIR2_LEAF_BESTS_P(ltp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_LEAF_TAIL_P) -xfs_dir2_leaf_tail_t * -xfs_dir2_leaf_tail_p(xfs_mount_t *mp, xfs_dir2_leaf_t *lp) -{ - return XFS_DIR2_LEAF_TAIL_P(mp, lp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_MAX_LEAF_ENTS) -int -xfs_dir2_max_leaf_ents(xfs_mount_t *mp) -{ - return XFS_DIR2_MAX_LEAF_ENTS(mp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_SF_ENTSIZE_BYENTRY) -int -xfs_dir2_sf_entsize_byentry(xfs_dir2_sf_t *sfp, xfs_dir2_sf_entry_t *sfep) -{ - return XFS_DIR2_SF_ENTSIZE_BYENTRY(sfp, sfep); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_SF_FIRSTENTRY) -xfs_dir2_sf_entry_t * -xfs_dir2_sf_firstentry(xfs_dir2_sf_t *sfp) -{ - return XFS_DIR2_SF_FIRSTENTRY(sfp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_SF_ENTSIZE_BYNAME) -int -xfs_dir2_sf_entsize_byname(xfs_dir2_sf_t *sfp, int len) -{ - return XFS_DIR2_SF_ENTSIZE_BYNAME(sfp, len); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_SF_GET_INUMBER) -xfs_intino_t -xfs_dir2_sf_get_inumber(xfs_dir2_sf_t *sfp, xfs_dir2_inou_t *from) -{ - return XFS_DIR2_SF_GET_INUMBER(sfp, from); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_SF_GET_OFFSET) -xfs_dir2_data_aoff_t -xfs_dir2_sf_get_offset(xfs_dir2_sf_entry_t *sfep) -{ - return XFS_DIR2_SF_GET_OFFSET(sfep); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_SF_HDR_SIZE) -int -xfs_dir2_sf_hdr_size(int i8count) -{ - return XFS_DIR2_SF_HDR_SIZE(i8count); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_SF_INUMBERP) -xfs_dir2_inou_t * -xfs_dir2_sf_inumberp(xfs_dir2_sf_entry_t *sfep) -{ - return XFS_DIR2_SF_INUMBERP(sfep); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_SF_NEXTENTRY) -xfs_dir2_sf_entry_t * -xfs_dir2_sf_nextentry(xfs_dir2_sf_t *sfp, xfs_dir2_sf_entry_t *sfep) -{ - return XFS_DIR2_SF_NEXTENTRY(sfp, sfep); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_SF_PUT_INUMBER) -void -xfs_dir2_sf_put_inumber(xfs_dir2_sf_t *sfp, xfs_ino_t *from, xfs_dir2_inou_t *to) -{ - XFS_DIR2_SF_PUT_INUMBER(sfp, from, to); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_DIR2_SF_PUT_OFFSET) -void -xfs_dir2_sf_put_offset(xfs_dir2_sf_entry_t *sfep, xfs_dir2_data_aoff_t off) -{ - XFS_DIR2_SF_PUT_OFFSET(sfep, off); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_EXTFMT_INODE ) -xfs_exntfmt_t -xfs_extfmt_inode(struct xfs_inode *ip) -{ - return XFS_EXTFMT_INODE(ip); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_EXTLEN_MAX) -xfs_extlen_t -xfs_extlen_max(xfs_extlen_t a, xfs_extlen_t b) -{ - return XFS_EXTLEN_MAX(a, b); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_EXTLEN_MIN) -xfs_extlen_t -xfs_extlen_min(xfs_extlen_t a, xfs_extlen_t b) -{ - return XFS_EXTLEN_MIN(a, b); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_FILBLKS_MAX) -xfs_filblks_t -xfs_filblks_max(xfs_filblks_t a, xfs_filblks_t b) -{ - return XFS_FILBLKS_MAX(a, b); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_FILBLKS_MIN) -xfs_filblks_t -xfs_filblks_min(xfs_filblks_t a, xfs_filblks_t b) -{ - return XFS_FILBLKS_MIN(a, b); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_FILEOFF_MAX) -xfs_fileoff_t -xfs_fileoff_max(xfs_fileoff_t a, xfs_fileoff_t b) -{ - return XFS_FILEOFF_MAX(a, b); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_FILEOFF_MIN) -xfs_fileoff_t -xfs_fileoff_min(xfs_fileoff_t a, xfs_fileoff_t b) -{ - return XFS_FILEOFF_MIN(a, b); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_FSB_SANITY_CHECK) -int -xfs_fsb_sanity_check(xfs_mount_t *mp, xfs_fsblock_t fsbno) -{ - return XFS_FSB_SANITY_CHECK(mp, fsbno); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_FSB_TO_AGBNO) -xfs_agblock_t -xfs_fsb_to_agbno(xfs_mount_t *mp, xfs_fsblock_t fsbno) -{ - return XFS_FSB_TO_AGBNO(mp, fsbno); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_FSB_TO_AGNO) -xfs_agnumber_t -xfs_fsb_to_agno(xfs_mount_t *mp, xfs_fsblock_t fsbno) -{ - return XFS_FSB_TO_AGNO(mp, fsbno); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_FSB_TO_DADDR) -xfs_daddr_t -xfs_fsb_to_daddr(xfs_mount_t *mp, xfs_fsblock_t fsbno) -{ - return XFS_FSB_TO_DADDR(mp, fsbno); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_FSB_TO_DB) -xfs_daddr_t -xfs_fsb_to_db(xfs_inode_t *ip, xfs_fsblock_t fsb) -{ - return XFS_FSB_TO_DB(ip, fsb); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_HDR_BLOCK) -xfs_agblock_t -xfs_hdr_block(xfs_mount_t *mp, xfs_daddr_t d) -{ - return XFS_HDR_BLOCK(mp, d); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_IALLOC_BLOCKS) -xfs_extlen_t -xfs_ialloc_blocks(xfs_mount_t *mp) -{ - return XFS_IALLOC_BLOCKS(mp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_IALLOC_FIND_FREE) -int -xfs_ialloc_find_free(xfs_inofree_t *fp) -{ - return XFS_IALLOC_FIND_FREE(fp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_IALLOC_INODES) -int -xfs_ialloc_inodes(xfs_mount_t *mp) -{ - return XFS_IALLOC_INODES(mp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_IBT_BLOCK) -xfs_agblock_t -xfs_ibt_block(xfs_mount_t *mp) -{ - return XFS_IBT_BLOCK(mp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_IFORK_ASIZE) -int -xfs_ifork_asize(xfs_inode_t *ip) -{ - return XFS_IFORK_ASIZE(ip); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_IFORK_DSIZE) -int -xfs_ifork_dsize(xfs_inode_t *ip) -{ - return XFS_IFORK_DSIZE(ip); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_IFORK_FMT_SET) -void -xfs_ifork_fmt_set(xfs_inode_t *ip, int w, int n) -{ - XFS_IFORK_FMT_SET(ip, w, n); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_IFORK_FORMAT) -int -xfs_ifork_format(xfs_inode_t *ip, int w) -{ - return XFS_IFORK_FORMAT(ip, w); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_IFORK_NEXT_SET) -void -xfs_ifork_next_set(xfs_inode_t *ip, int w, int n) -{ - XFS_IFORK_NEXT_SET(ip, w, n); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_IFORK_NEXTENTS) -int -xfs_ifork_nextents(xfs_inode_t *ip, int w) -{ - return XFS_IFORK_NEXTENTS(ip, w); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_IFORK_PTR) -xfs_ifork_t * -xfs_ifork_ptr(xfs_inode_t *ip, int w) -{ - return XFS_IFORK_PTR(ip, w); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_IFORK_Q) -int -xfs_ifork_q(xfs_inode_t *ip) -{ - return XFS_IFORK_Q(ip); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_IFORK_SIZE) -int -xfs_ifork_size(xfs_inode_t *ip, int w) -{ - return XFS_IFORK_SIZE(ip, w); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_ILOG_FBROOT) -int -xfs_ilog_fbroot(int w) -{ - return XFS_ILOG_FBROOT(w); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_ILOG_FDATA) -int -xfs_ilog_fdata(int w) -{ - return XFS_ILOG_FDATA(w); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_ILOG_FEXT) -int -xfs_ilog_fext(int w) -{ - return XFS_ILOG_FEXT(w); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_IN_MAXLEVELS) -int -xfs_in_maxlevels(xfs_mount_t *mp) -{ - return XFS_IN_MAXLEVELS(mp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_INO_AGBNO_BITS) -int -xfs_ino_agbno_bits(xfs_mount_t *mp) -{ - return XFS_INO_AGBNO_BITS(mp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_INO_AGINO_BITS) -int -xfs_ino_agino_bits(xfs_mount_t *mp) -{ - return XFS_INO_AGINO_BITS(mp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_INO_AGNO_BITS) -int -xfs_ino_agno_bits(xfs_mount_t *mp) -{ - return XFS_INO_AGNO_BITS(mp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_INO_BITS) -int -xfs_ino_bits(xfs_mount_t *mp) -{ - return XFS_INO_BITS(mp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_INO_MASK) -__uint32_t -xfs_ino_mask(int k) -{ - return XFS_INO_MASK(k); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_INO_OFFSET_BITS) -int -xfs_ino_offset_bits(xfs_mount_t *mp) -{ - return XFS_INO_OFFSET_BITS(mp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_INO_TO_AGBNO) -xfs_agblock_t -xfs_ino_to_agbno(xfs_mount_t *mp, xfs_ino_t i) -{ - return XFS_INO_TO_AGBNO(mp, i); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_INO_TO_AGINO) -xfs_agino_t -xfs_ino_to_agino(xfs_mount_t *mp, xfs_ino_t i) -{ - return XFS_INO_TO_AGINO(mp, i); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_INO_TO_AGNO) -xfs_agnumber_t -xfs_ino_to_agno(xfs_mount_t *mp, xfs_ino_t i) -{ - return XFS_INO_TO_AGNO(mp, i); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_INO_TO_FSB) -xfs_fsblock_t -xfs_ino_to_fsb(xfs_mount_t *mp, xfs_ino_t i) -{ - return XFS_INO_TO_FSB(mp, i); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_INO_TO_OFFSET) -int -xfs_ino_to_offset(xfs_mount_t *mp, xfs_ino_t i) -{ - return XFS_INO_TO_OFFSET(mp, i); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_INOBT_BLOCK_MAXRECS) -int -xfs_inobt_block_maxrecs(int lev, xfs_btree_cur_t *cur) -{ - return XFS_INOBT_BLOCK_MAXRECS(lev, cur); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_INOBT_BLOCK_MINRECS) -int -xfs_inobt_block_minrecs(int lev, xfs_btree_cur_t *cur) -{ - return XFS_INOBT_BLOCK_MINRECS(lev, cur); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_INOBT_BLOCK_SIZE) -/*ARGSUSED1*/ -int -xfs_inobt_block_size(int lev, xfs_btree_cur_t *cur) -{ - return XFS_INOBT_BLOCK_SIZE(lev, cur); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_INOBT_CLR_FREE) -void -xfs_inobt_clr_free(xfs_inobt_rec_t *rp, int i) -{ - XFS_INOBT_CLR_FREE(rp, i); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_INOBT_IS_FREE) -int -xfs_inobt_is_free(xfs_inobt_rec_t *rp, int i) -{ - return XFS_INOBT_IS_FREE(rp, i); -} -int -xfs_inobt_is_free_disk(xfs_inobt_rec_t *rp, int i) -{ - return XFS_INOBT_IS_FREE_DISK(rp, i); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_INOBT_IS_LAST_REC) -int -xfs_inobt_is_last_rec(xfs_btree_cur_t *cur) -{ - return XFS_INOBT_IS_LAST_REC(cur); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_INOBT_KEY_ADDR) -/*ARGSUSED3*/ -xfs_inobt_key_t * -xfs_inobt_key_addr(xfs_inobt_block_t *bb, int i, xfs_btree_cur_t *cur) -{ - return XFS_INOBT_KEY_ADDR(bb, i, cur); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_INOBT_MASK) -xfs_inofree_t -xfs_inobt_mask(int i) -{ - return XFS_INOBT_MASK(i); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_INOBT_MASKN) -xfs_inofree_t -xfs_inobt_maskn(int i, int n) -{ - return XFS_INOBT_MASKN(i, n); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_INOBT_PTR_ADDR) -xfs_inobt_ptr_t * -xfs_inobt_ptr_addr(xfs_inobt_block_t *bb, int i, xfs_btree_cur_t *cur) -{ - return XFS_INOBT_PTR_ADDR(bb, i, cur); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_INOBT_REC_ADDR) -/*ARGSUSED3*/ -xfs_inobt_rec_t * -xfs_inobt_rec_addr(xfs_inobt_block_t *bb, int i, xfs_btree_cur_t *cur) -{ - return XFS_INOBT_REC_ADDR(bb, i, cur); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_INOBT_SET_FREE) -void -xfs_inobt_set_free(xfs_inobt_rec_t *rp, int i) -{ - XFS_INOBT_SET_FREE(rp, i); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_ITOBHV) -bhv_desc_t * -xfs_itobhv(xfs_inode_t *ip) -{ - return XFS_ITOBHV(ip); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_ITOV) -vnode_t * -xfs_itov(xfs_inode_t *ip) -{ - return XFS_ITOV(ip); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_LBLOG) -int -xfs_lblog(xfs_mount_t *mp) -{ - return XFS_LBLOG(mp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_LBSIZE) -int -xfs_lbsize(xfs_mount_t *mp) -{ - return XFS_LBSIZE(mp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_LIC_ALL_FREE) -void -xfs_lic_all_free(xfs_log_item_chunk_t *cp) -{ - XFS_LIC_ALL_FREE(cp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_LIC_ARE_ALL_FREE) -int -xfs_lic_are_all_free(xfs_log_item_chunk_t *cp) -{ - return XFS_LIC_ARE_ALL_FREE(cp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_LIC_CLAIM) -void -xfs_lic_claim(xfs_log_item_chunk_t *cp, int slot) -{ - XFS_LIC_CLAIM(cp, slot); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_LIC_DESC_TO_CHUNK) -xfs_log_item_chunk_t * -xfs_lic_desc_to_chunk(xfs_log_item_desc_t *dp) -{ - return XFS_LIC_DESC_TO_CHUNK(dp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_LIC_DESC_TO_SLOT) -int -xfs_lic_desc_to_slot(xfs_log_item_desc_t *dp) -{ - return XFS_LIC_DESC_TO_SLOT(dp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_LIC_INIT) -void -xfs_lic_init(xfs_log_item_chunk_t *cp) -{ - XFS_LIC_INIT(cp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_LIC_INIT_SLOT) -void -xfs_lic_init_slot(xfs_log_item_chunk_t *cp, int slot) -{ - XFS_LIC_INIT_SLOT(cp, slot); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_LIC_ISFREE) -int -xfs_lic_isfree(xfs_log_item_chunk_t *cp, int slot) -{ - return XFS_LIC_ISFREE(cp, slot); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_LIC_RELSE) -void -xfs_lic_relse(xfs_log_item_chunk_t *cp, int slot) -{ - XFS_LIC_RELSE(cp, slot); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_LIC_SLOT) -xfs_log_item_desc_t * -xfs_lic_slot(xfs_log_item_chunk_t *cp, int slot) -{ - return XFS_LIC_SLOT(cp, slot); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_LIC_VACANCY) -int -xfs_lic_vacancy(xfs_log_item_chunk_t *cp) -{ - return XFS_LIC_VACANCY(cp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_LITINO) -int -xfs_litino(xfs_mount_t *mp) -{ - return XFS_LITINO(mp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_MAKE_IPTR) -xfs_dinode_t * -xfs_make_iptr(xfs_mount_t *mp, xfs_buf_t *b, int o) -{ - return XFS_MAKE_IPTR(mp, b, o); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_MASK32HI) -__uint32_t -xfs_mask32hi(int n) -{ - return XFS_MASK32HI(n); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_MASK32LO) -__uint32_t -xfs_mask32lo(int n) -{ - return XFS_MASK32LO(n); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_MASK64HI) -__uint64_t -xfs_mask64hi(int n) -{ - return XFS_MASK64HI(n); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_MASK64LO) -__uint64_t -xfs_mask64lo(int n) -{ - return XFS_MASK64LO(n); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_MIN_FREELIST) -int -xfs_min_freelist(xfs_agf_t *a, xfs_mount_t *mp) -{ - return XFS_MIN_FREELIST(a, mp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_MIN_FREELIST_PAG) -int -xfs_min_freelist_pag(xfs_perag_t *pag, xfs_mount_t *mp) -{ - return XFS_MIN_FREELIST_PAG(pag, mp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_MIN_FREELIST_RAW) -int -xfs_min_freelist_raw(uint bl, uint cl, xfs_mount_t *mp) -{ - return XFS_MIN_FREELIST_RAW(bl, cl, mp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_MTOVFS) -vfs_t * -xfs_mtovfs(xfs_mount_t *mp) -{ - return XFS_MTOVFS(mp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_OFFBNO_TO_AGINO) -xfs_agino_t -xfs_offbno_to_agino(xfs_mount_t *mp, xfs_agblock_t b, int o) -{ - return XFS_OFFBNO_TO_AGINO(mp, b, o); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_PREALLOC_BLOCKS) -xfs_agblock_t -xfs_prealloc_blocks(xfs_mount_t *mp) -{ - return XFS_PREALLOC_BLOCKS(mp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_SB_BLOCK) -xfs_agblock_t -xfs_sb_block(xfs_mount_t *mp) -{ - return XFS_SB_BLOCK(mp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_SB_GOOD_VERSION) -int -xfs_sb_good_version(xfs_sb_t *sbp) -{ - return XFS_SB_GOOD_VERSION(sbp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_SB_VERSION_ADDATTR) -void -xfs_sb_version_addattr(xfs_sb_t *sbp) -{ - XFS_SB_VERSION_ADDATTR(sbp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_SB_VERSION_ADDDALIGN) -void -xfs_sb_version_adddalign(xfs_sb_t *sbp) -{ - XFS_SB_VERSION_ADDDALIGN(sbp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_SB_VERSION_ADDNLINK) -void -xfs_sb_version_addnlink(xfs_sb_t *sbp) -{ - XFS_SB_VERSION_ADDNLINK(sbp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_SB_VERSION_ADDQUOTA) -void -xfs_sb_version_addquota(xfs_sb_t *sbp) -{ - XFS_SB_VERSION_ADDQUOTA(sbp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_SB_VERSION_ADDSHARED) -void -xfs_sb_version_addshared(xfs_sb_t *sbp) -{ - XFS_SB_VERSION_ADDSHARED(sbp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_SB_VERSION_HASALIGN) -int -xfs_sb_version_hasalign(xfs_sb_t *sbp) -{ - return XFS_SB_VERSION_HASALIGN(sbp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_SB_VERSION_HASATTR) -int -xfs_sb_version_hasattr(xfs_sb_t *sbp) -{ - return XFS_SB_VERSION_HASATTR(sbp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_SB_VERSION_HASDALIGN) -int -xfs_sb_version_hasdalign(xfs_sb_t *sbp) -{ - return XFS_SB_VERSION_HASDALIGN(sbp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_SB_VERSION_HASDIRV2) -int -xfs_sb_version_hasdirv2(xfs_sb_t *sbp) -{ - return XFS_SB_VERSION_HASDIRV2(sbp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_SB_VERSION_HASEXTFLGBIT) -int -xfs_sb_version_hasextflgbit(xfs_sb_t *sbp) -{ - return XFS_SB_VERSION_HASEXTFLGBIT(sbp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_SB_VERSION_HASNLINK) -int -xfs_sb_version_hasnlink(xfs_sb_t *sbp) -{ - return XFS_SB_VERSION_HASNLINK(sbp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_SB_VERSION_HASQUOTA) -int -xfs_sb_version_hasquota(xfs_sb_t *sbp) -{ - return XFS_SB_VERSION_HASQUOTA(sbp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_SB_VERSION_HASSHARED) -int -xfs_sb_version_hasshared(xfs_sb_t *sbp) -{ - return XFS_SB_VERSION_HASSHARED(sbp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_SB_VERSION_NUM) -int -xfs_sb_version_num(xfs_sb_t *sbp) -{ - return XFS_SB_VERSION_NUM(sbp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_SB_VERSION_SUBALIGN) -void -xfs_sb_version_subalign(xfs_sb_t *sbp) -{ - XFS_SB_VERSION_SUBALIGN(sbp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_SB_VERSION_SUBSHARED) -void -xfs_sb_version_subshared(xfs_sb_t *sbp) -{ - XFS_SB_VERSION_SUBSHARED(sbp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_SB_VERSION_HASLOGV2) -int -xfs_sb_version_haslogv2(xfs_sb_t *sbp) -{ - return XFS_SB_VERSION_HASLOGV2(sbp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_SB_VERSION_HASSECTOR) -int -xfs_sb_version_hassector(xfs_sb_t *sbp) -{ - return XFS_SB_VERSION_HASSECTOR(sbp); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_SB_VERSION_TONEW) -unsigned -xfs_sb_version_tonew(unsigned v) -{ - return XFS_SB_VERSION_TONEW(v); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_SB_VERSION_TOOLD) -unsigned -xfs_sb_version_toold(unsigned v) -{ - return XFS_SB_VERSION_TOOLD(v); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XLOG_GRANT_ADD_SPACE) -void -xlog_grant_add_space(xlog_t *log, int bytes, int type) -{ - XLOG_GRANT_ADD_SPACE(log, bytes, type); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XLOG_GRANT_SUB_SPACE) -void -xlog_grant_sub_space(xlog_t *log, int bytes, int type) -{ - XLOG_GRANT_SUB_SPACE(log, bytes, type); -} -#endif - -#if XFS_WANT_FUNCS_C || (XFS_WANT_SPACE_C && XFSSO_XFS_SB_VERSION_HASMOREBITS) -int -xfs_sb_version_hasmorebits(xfs_sb_t *sbp) -{ - return XFS_SB_VERSION_HASMOREBITS(sbp); -} -#endif - diff --git a/fs/xfs/xfs_macros.h b/fs/xfs/xfs_macros.h deleted file mode 100644 index 0a9307514a48..000000000000 --- a/fs/xfs/xfs_macros.h +++ /dev/null @@ -1,104 +0,0 @@ -/* - * Copyright (c) 2000 Silicon Graphics, Inc. All Rights Reserved. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as - * published by the Free Software Foundation. - * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. - * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ - */ -#ifndef __XFS_MACROS_H__ -#define __XFS_MACROS_H__ - -/* - * Set for debug kernels and simulation - * These replacements save space. - * Used in xfs_macros.c. - */ -#define XFS_WANT_SPACE_C \ - (!defined(_STANDALONE) && defined(DEBUG)) - -/* - * Set for debug simulation and kernel builds, but not for standalone. - * These replacements do not save space. - * Used in xfs_macros.c. - */ -#define XFS_WANT_FUNCS_C \ - (!defined(_STANDALONE) && defined(DEBUG)) - -/* - * Corresponding names used in .h files. - */ -#define XFS_WANT_SPACE (XFS_WANT_SPACE_C && !defined(XFS_MACRO_C)) -#define XFS_WANT_FUNCS (XFS_WANT_FUNCS_C && !defined(XFS_MACRO_C)) - -/* - * These are the macros that get turned into functions to save space. - */ -#define XFSSO_NULLSTARTBLOCK 1 -#define XFSSO_XFS_AGB_TO_DADDR 1 -#define XFSSO_XFS_AGB_TO_FSB 1 -#define XFSSO_XFS_AGINO_TO_INO 1 -#define XFSSO_XFS_ALLOC_BLOCK_MINRECS 1 -#define XFSSO_XFS_ATTR_SF_NEXTENTRY 1 -#define XFSSO_XFS_BMAP_BLOCK_DMAXRECS 1 -#define XFSSO_XFS_BMAP_BLOCK_IMAXRECS 1 -#define XFSSO_XFS_BMAP_BLOCK_IMINRECS 1 -#define XFSSO_XFS_BMAP_INIT 1 -#define XFSSO_XFS_BMAP_PTR_IADDR 1 -#define XFSSO_XFS_BMAP_SANITY_CHECK 1 -#define XFSSO_XFS_BMAPI_AFLAG 1 -#define XFSSO_XFS_CFORK_SIZE 1 -#define XFSSO_XFS_DA_COOKIE_BNO 1 -#define XFSSO_XFS_DA_COOKIE_ENTRY 1 -#define XFSSO_XFS_DADDR_TO_AGBNO 1 -#define XFSSO_XFS_DADDR_TO_FSB 1 -#define XFSSO_XFS_DFORK_PTR 1 -#define XFSSO_XFS_DIR_SF_GET_DIRINO 1 -#define XFSSO_XFS_DIR_SF_NEXTENTRY 1 -#define XFSSO_XFS_DIR_SF_PUT_DIRINO 1 -#define XFSSO_XFS_FILBLKS_MIN 1 -#define XFSSO_XFS_FSB_SANITY_CHECK 1 -#define XFSSO_XFS_FSB_TO_DADDR 1 -#define XFSSO_XFS_FSB_TO_DB 1 -#define XFSSO_XFS_IALLOC_INODES 1 -#define XFSSO_XFS_IFORK_ASIZE 1 -#define XFSSO_XFS_IFORK_DSIZE 1 -#define XFSSO_XFS_IFORK_FORMAT 1 -#define XFSSO_XFS_IFORK_NEXT_SET 1 -#define XFSSO_XFS_IFORK_NEXTENTS 1 -#define XFSSO_XFS_IFORK_PTR 1 -#define XFSSO_XFS_ILOG_FBROOT 1 -#define XFSSO_XFS_ILOG_FEXT 1 -#define XFSSO_XFS_INO_MASK 1 -#define XFSSO_XFS_INO_TO_FSB 1 -#define XFSSO_XFS_INODE_CLEAR_READ_AHEAD 1 -#define XFSSO_XFS_MIN_FREELIST 1 -#define XFSSO_XFS_SB_GOOD_VERSION 1 -#define XFSSO_XFS_SB_VERSION_HASNLINK 1 -#define XFSSO_XLOG_GRANT_ADD_SPACE 1 -#define XFSSO_XLOG_GRANT_SUB_SPACE 1 - -#endif /* __XFS_MACROS_H__ */ diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index 82e1646e6243..541d5dd474be 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -1,40 +1,26 @@ /* - * Copyright (c) 2000-2004 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" @@ -42,21 +28,20 @@ #include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" -#include "xfs_alloc_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_btree.h" -#include "xfs_ialloc.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" #include "xfs_inode.h" +#include "xfs_btree.h" +#include "xfs_ialloc.h" #include "xfs_alloc.h" #include "xfs_rtalloc.h" #include "xfs_bmap.h" #include "xfs_error.h" -#include "xfs_bit.h" #include "xfs_rw.h" #include "xfs_quota.h" #include "xfs_fsops.h" @@ -180,6 +165,10 @@ xfs_mount_free( if (mp->m_fsname != NULL) kmem_free(mp->m_fsname, mp->m_fsname_len); + if (mp->m_rtname != NULL) + kmem_free(mp->m_rtname, strlen(mp->m_rtname) + 1); + if (mp->m_logname != NULL) + kmem_free(mp->m_logname, strlen(mp->m_logname) + 1); if (remove_bhv) { struct vfs *vfsp = XFS_MTOVFS(mp); @@ -318,7 +307,7 @@ xfs_mount_validate_sb( "XFS: Attempted to mount file system with blocksize %d bytes", sbp->sb_blocksize); cmn_err(CE_WARN, - "XFS: Only page-sized (%d) or less blocksizes currently work.", + "XFS: Only page-sized (%ld) or less blocksizes currently work.", PAGE_SIZE); return XFS_ERROR(ENOSYS); } @@ -327,7 +316,10 @@ xfs_mount_validate_sb( } xfs_agnumber_t -xfs_initialize_perag(xfs_mount_t *mp, xfs_agnumber_t agcount) +xfs_initialize_perag( + struct vfs *vfs, + xfs_mount_t *mp, + xfs_agnumber_t agcount) { xfs_agnumber_t index, max_metadata; xfs_perag_t *pag; @@ -343,7 +335,7 @@ xfs_initialize_perag(xfs_mount_t *mp, xfs_agnumber_t agcount) /* Clear the mount flag if no inode can overflow 32 bits * on this filesystem, or if specifically requested.. */ - if ((mp->m_flags & XFS_MOUNT_32BITINOOPT) && ino > max_inum) { + if ((vfs->vfs_flag & VFS_32BITINODES) && ino > max_inum) { mp->m_flags |= XFS_MOUNT_32BITINODES; } else { mp->m_flags &= ~XFS_MOUNT_32BITINODES; @@ -360,7 +352,7 @@ xfs_initialize_perag(xfs_mount_t *mp, xfs_agnumber_t agcount) icount = sbp->sb_dblocks * sbp->sb_imax_pct; do_div(icount, 100); icount += sbp->sb_agblocks - 1; - do_div(icount, mp->m_ialloc_blks); + do_div(icount, sbp->sb_agblocks); max_metadata = icount; } else { max_metadata = agcount; @@ -584,12 +576,13 @@ xfs_mount_common(xfs_mount_t *mp, xfs_sb_t *sbp) ASSERT(sbp->sb_inodesize >= 256 && sbp->sb_inodesize <= 2048); switch (sbp->sb_inodesize) { case 256: - mp->m_attroffset = XFS_LITINO(mp) - XFS_BMDR_SPACE_CALC(2); + mp->m_attroffset = XFS_LITINO(mp) - + XFS_BMDR_SPACE_CALC(MINABTPTRS); break; case 512: case 1024: case 2048: - mp->m_attroffset = XFS_BMDR_SPACE_CALC(12); + mp->m_attroffset = XFS_BMDR_SPACE_CALC(6 * MINABTPTRS); break; default: ASSERT(0); @@ -954,7 +947,7 @@ xfs_mountfs( mp->m_perag = kmem_zalloc(sbp->sb_agcount * sizeof(xfs_perag_t), KM_SLEEP); - mp->m_maxagi = xfs_initialize_perag(mp, sbp->sb_agcount); + mp->m_maxagi = xfs_initialize_perag(vfsp, mp, sbp->sb_agcount); /* * log's mount-time initialization. Perform 1st part recovery if needed diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index 5affba38a577..08b2e0a5d807 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -1,38 +1,23 @@ /* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_MOUNT_H__ #define __XFS_MOUNT_H__ - typedef struct xfs_trans_reservations { uint tr_write; /* extent alloc trans */ uint tr_itruncate; /* truncate trans */ @@ -57,7 +42,6 @@ typedef struct xfs_trans_reservations { uint tr_growrtfree; /* grow realtime freeing */ } xfs_trans_reservations_t; - #ifndef __KERNEL__ /* * Moved here from xfs_ag.h to avoid reordering header files @@ -80,6 +64,9 @@ struct xfs_iocore; struct xfs_bmbt_irec; struct xfs_bmap_free; +extern struct vfsops xfs_vfsops; +extern struct vnodeops xfs_vnodeops; + #define AIL_LOCK_T lock_t #define AIL_LOCKINIT(x,y) spinlock_init(x,y) #define AIL_LOCK_DESTROY(x) spinlock_destroy(x) @@ -292,6 +279,8 @@ typedef struct xfs_mount { struct xfs_buf *m_sb_bp; /* buffer for superblock */ char *m_fsname; /* filesystem name */ int m_fsname_len; /* strlen of fs name */ + char *m_rtname; /* realtime device name */ + char *m_logname; /* external log device name */ int m_bsize; /* fs logical block size */ xfs_agnumber_t m_agfrotor; /* last ag where space found */ xfs_agnumber_t m_agirotor; /* last ag dir inode alloced */ @@ -344,7 +333,7 @@ typedef struct xfs_mount { sema_t m_growlock; /* growfs mutex */ int m_fixedfsid[2]; /* unchanged for life of FS */ uint m_dmevmask; /* DMI events for this FS */ - uint m_flags; /* global mount flags */ + __uint64_t m_flags; /* global mount flags */ uint m_attroffset; /* inode attribute offset */ uint m_dir_node_ents; /* #entries in a dir danode */ uint m_attr_node_ents; /* #entries in attr danode */ @@ -389,38 +378,41 @@ typedef struct xfs_mount { /* * Flags for m_flags. */ -#define XFS_MOUNT_WSYNC 0x00000001 /* for nfs - all metadata ops +#define XFS_MOUNT_WSYNC (1ULL << 0) /* for nfs - all metadata ops must be synchronous except for space allocations */ -#define XFS_MOUNT_INO64 0x00000002 - /* 0x00000004 -- currently unused */ - /* 0x00000008 -- currently unused */ -#define XFS_MOUNT_FS_SHUTDOWN 0x00000010 /* atomic stop of all filesystem +#define XFS_MOUNT_INO64 (1ULL << 1) + /* (1ULL << 2) -- currently unused */ + /* (1ULL << 3) -- currently unused */ +#define XFS_MOUNT_FS_SHUTDOWN (1ULL << 4) /* atomic stop of all filesystem operations, typically for disk errors in metadata */ -#define XFS_MOUNT_NOATIME 0x00000020 /* don't modify inode access +#define XFS_MOUNT_NOATIME (1ULL << 5) /* don't modify inode access times on reads */ -#define XFS_MOUNT_RETERR 0x00000040 /* return alignment errors to +#define XFS_MOUNT_RETERR (1ULL << 6) /* return alignment errors to user */ -#define XFS_MOUNT_NOALIGN 0x00000080 /* turn off stripe alignment +#define XFS_MOUNT_NOALIGN (1ULL << 7) /* turn off stripe alignment allocations */ - /* 0x00000100 -- currently unused */ - /* 0x00000200 -- currently unused */ -#define XFS_MOUNT_NORECOVERY 0x00000400 /* no recovery - dirty fs */ -#define XFS_MOUNT_SHARED 0x00000800 /* shared mount */ -#define XFS_MOUNT_DFLT_IOSIZE 0x00001000 /* set default i/o size */ -#define XFS_MOUNT_OSYNCISOSYNC 0x00002000 /* o_sync is REALLY o_sync */ +#define XFS_MOUNT_COMPAT_ATTR (1ULL << 8) /* do not use attr2 format */ + /* (1ULL << 9) -- currently unused */ +#define XFS_MOUNT_NORECOVERY (1ULL << 10) /* no recovery - dirty fs */ +#define XFS_MOUNT_SHARED (1ULL << 11) /* shared mount */ +#define XFS_MOUNT_DFLT_IOSIZE (1ULL << 12) /* set default i/o size */ +#define XFS_MOUNT_OSYNCISOSYNC (1ULL << 13) /* o_sync is REALLY o_sync */ /* osyncisdsync is now default*/ -#define XFS_MOUNT_32BITINODES 0x00004000 /* do not create inodes above +#define XFS_MOUNT_32BITINODES (1ULL << 14) /* do not create inodes above * 32 bits in size */ -#define XFS_MOUNT_32BITINOOPT 0x00008000 /* saved mount option state */ -#define XFS_MOUNT_NOUUID 0x00010000 /* ignore uuid during mount */ -#define XFS_MOUNT_NOLOGFLUSH 0x00020000 -#define XFS_MOUNT_IDELETE 0x00040000 /* delete empty inode clusters*/ -#define XFS_MOUNT_SWALLOC 0x00080000 /* turn on stripe width + /* (1ULL << 15) -- currently unused */ +#define XFS_MOUNT_NOUUID (1ULL << 16) /* ignore uuid during mount */ +#define XFS_MOUNT_BARRIER (1ULL << 17) +#define XFS_MOUNT_IDELETE (1ULL << 18) /* delete empty inode clusters*/ +#define XFS_MOUNT_SWALLOC (1ULL << 19) /* turn on stripe width * allocation */ -#define XFS_MOUNT_IHASHSIZE 0x00100000 /* inode hash table size */ -#define XFS_MOUNT_DIRSYNC 0x00200000 /* synchronous directory ops */ +#define XFS_MOUNT_IHASHSIZE (1ULL << 20) /* inode hash table size */ +#define XFS_MOUNT_DIRSYNC (1ULL << 21) /* synchronous directory ops */ +#define XFS_MOUNT_COMPAT_IOSIZE (1ULL << 22) /* don't report large preferred + * I/O size in stat() */ + /* * Default minimum read and write sizes. @@ -442,6 +434,30 @@ typedef struct xfs_mount { #define XFS_WSYNC_READIO_LOG 15 /* 32K */ #define XFS_WSYNC_WRITEIO_LOG 14 /* 16K */ +/* + * Allow large block sizes to be reported to userspace programs if the + * "largeio" mount option is used. + * + * If compatibility mode is specified, simply return the basic unit of caching + * so that we don't get inefficient read/modify/write I/O from user apps. + * Otherwise.... + * + * If the underlying volume is a stripe, then return the stripe width in bytes + * as the recommended I/O size. It is not a stripe and we've set a default + * buffered I/O size, return that, otherwise return the compat default. + */ +static inline unsigned long +xfs_preferred_iosize(xfs_mount_t *mp) +{ + if (mp->m_flags & XFS_MOUNT_COMPAT_IOSIZE) + return PAGE_CACHE_SIZE; + return (mp->m_swidth ? + (mp->m_swidth << mp->m_sb.sb_blocklog) : + ((mp->m_flags & XFS_MOUNT_DFLT_IOSIZE) ? + (1 << (int)MAX(mp->m_readio_log, mp->m_writeio_log)) : + PAGE_CACHE_SIZE)); +} + #define XFS_MAXIOFFSET(mp) ((mp)->m_maxioffset) #define XFS_FORCED_SHUTDOWN(mp) ((mp)->m_flags & XFS_MOUNT_FS_SHUTDOWN) @@ -474,57 +490,41 @@ typedef struct xfs_mount { /* * Macros for getting from mount to vfs and back. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_MTOVFS) -struct vfs *xfs_mtovfs(xfs_mount_t *mp); #define XFS_MTOVFS(mp) xfs_mtovfs(mp) -#else -#define XFS_MTOVFS(mp) (bhvtovfs(&(mp)->m_bhv)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BHVTOM) -xfs_mount_t *xfs_bhvtom(bhv_desc_t *bdp); -#define XFS_BHVTOM(bdp) xfs_bhvtom(bdp) -#else -#define XFS_BHVTOM(bdp) ((xfs_mount_t *)BHV_PDATA(bdp)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_VFSTOM) -xfs_mount_t *xfs_vfstom(vfs_t *vfs); -#define XFS_VFSTOM(vfs) xfs_vfstom(vfs) -#else -#define XFS_VFSTOM(vfs) \ - (XFS_BHVTOM(bhv_lookup(VFS_BHVHEAD(vfs), &xfs_vfsops))) -#endif +static inline struct vfs *xfs_mtovfs(xfs_mount_t *mp) +{ + return bhvtovfs(&mp->m_bhv); +} +#define XFS_BHVTOM(bdp) xfs_bhvtom(bdp) +static inline xfs_mount_t *xfs_bhvtom(bhv_desc_t *bdp) +{ + return (xfs_mount_t *)BHV_PDATA(bdp); +} -/* - * Moved here from xfs_ag.h to avoid reordering header files - */ +#define XFS_VFSTOM(vfs) xfs_vfstom(vfs) +static inline xfs_mount_t *xfs_vfstom(vfs_t *vfs) +{ + return XFS_BHVTOM(bhv_lookup(VFS_BHVHEAD(vfs), &xfs_vfsops)); +} -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DADDR_TO_AGNO) -xfs_agnumber_t xfs_daddr_to_agno(struct xfs_mount *mp, xfs_daddr_t d); #define XFS_DADDR_TO_AGNO(mp,d) xfs_daddr_to_agno(mp,d) -#else - -static inline xfs_agnumber_t XFS_DADDR_TO_AGNO(xfs_mount_t *mp, xfs_daddr_t d) +static inline xfs_agnumber_t +xfs_daddr_to_agno(struct xfs_mount *mp, xfs_daddr_t d) { - d = XFS_BB_TO_FSBT(mp, d); - do_div(d, mp->m_sb.sb_agblocks); - return (xfs_agnumber_t) d; + xfs_daddr_t ld = XFS_BB_TO_FSBT(mp, d); + do_div(ld, mp->m_sb.sb_agblocks); + return (xfs_agnumber_t) ld; } -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DADDR_TO_AGBNO) -xfs_agblock_t xfs_daddr_to_agbno(struct xfs_mount *mp, xfs_daddr_t d); #define XFS_DADDR_TO_AGBNO(mp,d) xfs_daddr_to_agbno(mp,d) -#else - -static inline xfs_agblock_t XFS_DADDR_TO_AGBNO(xfs_mount_t *mp, xfs_daddr_t d) +static inline xfs_agblock_t +xfs_daddr_to_agbno(struct xfs_mount *mp, xfs_daddr_t d) { - d = XFS_BB_TO_FSBT(mp, d); - return (xfs_agblock_t) do_div(d, mp->m_sb.sb_agblocks); + xfs_daddr_t ld = XFS_BB_TO_FSBT(mp, d); + return (xfs_agblock_t) do_div(ld, mp->m_sb.sb_agblocks); } -#endif - /* * This structure is for use by the xfs_mod_incore_sb_batch() routine. */ @@ -542,6 +542,7 @@ extern xfs_mount_t *xfs_mount_init(void); extern void xfs_mod_sb(xfs_trans_t *, __int64_t); extern void xfs_mount_free(xfs_mount_t *mp, int remove_bhv); extern int xfs_mountfs(struct vfs *, xfs_mount_t *mp, int); +extern void xfs_mountfs_check_barriers(xfs_mount_t *mp); extern int xfs_unmountfs(xfs_mount_t *, struct cred *); extern void xfs_unmountfs_close(xfs_mount_t *, struct cred *); @@ -555,12 +556,11 @@ extern int xfs_readsb(xfs_mount_t *mp); extern void xfs_freesb(xfs_mount_t *); extern void xfs_do_force_shutdown(bhv_desc_t *, int, char *, int); extern int xfs_syncsub(xfs_mount_t *, int, int, int *); -extern xfs_agnumber_t xfs_initialize_perag(xfs_mount_t *, xfs_agnumber_t); +extern int xfs_sync_inodes(xfs_mount_t *, int, int, int *); +extern xfs_agnumber_t xfs_initialize_perag(struct vfs *, xfs_mount_t *, + xfs_agnumber_t); extern void xfs_xlatesb(void *, struct xfs_sb *, int, __int64_t); -extern struct vfsops xfs_vfsops; -extern struct vnodeops xfs_vnodeops; - extern struct xfs_dmops xfs_dmcore_stub; extern struct xfs_qmops xfs_qmcore_stub; extern struct xfs_ioops xfs_iocore_xfs; diff --git a/fs/xfs/xfs_qmops.c b/fs/xfs/xfs_qmops.c index a6cd6324e946..1408a32eef88 100644 --- a/fs/xfs/xfs_qmops.c +++ b/fs/xfs/xfs_qmops.c @@ -1,40 +1,25 @@ /* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #include "xfs.h" - -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" diff --git a/fs/xfs/xfs_quota.h b/fs/xfs/xfs_quota.h index 32cb79752d5d..82a08baf437b 100644 --- a/fs/xfs/xfs_quota.h +++ b/fs/xfs/xfs_quota.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_QUOTA_H__ #define __XFS_QUOTA_H__ @@ -42,7 +28,7 @@ * uid_t and gid_t are hard-coded to 32 bits in the inode. * Hence, an 'id' in a dquot is 32 bits.. */ -typedef __int32_t xfs_dqid_t; +typedef __uint32_t xfs_dqid_t; /* * Eventhough users may not have quota limits occupying all 64-bits, @@ -59,28 +45,28 @@ typedef __uint16_t xfs_qwarncnt_t; * to construct the on disk structure. */ typedef struct xfs_disk_dquot { -/*16*/ u_int16_t d_magic; /* dquot magic = XFS_DQUOT_MAGIC */ -/*8 */ u_int8_t d_version; /* dquot version */ -/*8 */ u_int8_t d_flags; /* XFS_DQ_USER/PROJ/GROUP */ -/*32*/ xfs_dqid_t d_id; /* user,project,group id */ -/*64*/ xfs_qcnt_t d_blk_hardlimit;/* absolute limit on disk blks */ -/*64*/ xfs_qcnt_t d_blk_softlimit;/* preferred limit on disk blks */ -/*64*/ xfs_qcnt_t d_ino_hardlimit;/* maximum # allocated inodes */ -/*64*/ xfs_qcnt_t d_ino_softlimit;/* preferred inode limit */ -/*64*/ xfs_qcnt_t d_bcount; /* disk blocks owned by the user */ -/*64*/ xfs_qcnt_t d_icount; /* inodes owned by the user */ -/*32*/ __int32_t d_itimer; /* zero if within inode limits if not, + __be16 d_magic; /* dquot magic = XFS_DQUOT_MAGIC */ + __u8 d_version; /* dquot version */ + __u8 d_flags; /* XFS_DQ_USER/PROJ/GROUP */ + __be32 d_id; /* user,project,group id */ + __be64 d_blk_hardlimit;/* absolute limit on disk blks */ + __be64 d_blk_softlimit;/* preferred limit on disk blks */ + __be64 d_ino_hardlimit;/* maximum # allocated inodes */ + __be64 d_ino_softlimit;/* preferred inode limit */ + __be64 d_bcount; /* disk blocks owned by the user */ + __be64 d_icount; /* inodes owned by the user */ + __be32 d_itimer; /* zero if within inode limits if not, this is when we refuse service */ -/*32*/ __int32_t d_btimer; /* similar to above; for disk blocks */ -/*16*/ xfs_qwarncnt_t d_iwarns; /* warnings issued wrt num inodes */ -/*16*/ xfs_qwarncnt_t d_bwarns; /* warnings issued wrt disk blocks */ -/*32*/ __int32_t d_pad0; /* 64 bit align */ -/*64*/ xfs_qcnt_t d_rtb_hardlimit;/* absolute limit on realtime blks */ -/*64*/ xfs_qcnt_t d_rtb_softlimit;/* preferred limit on RT disk blks */ -/*64*/ xfs_qcnt_t d_rtbcount; /* realtime blocks owned */ -/*32*/ __int32_t d_rtbtimer; /* similar to above; for RT disk blocks */ -/*16*/ xfs_qwarncnt_t d_rtbwarns; /* warnings issued wrt RT disk blocks */ -/*16*/ __uint16_t d_pad; + __be32 d_btimer; /* similar to above; for disk blocks */ + __be16 d_iwarns; /* warnings issued wrt num inodes */ + __be16 d_bwarns; /* warnings issued wrt disk blocks */ + __be32 d_pad0; /* 64 bit align */ + __be64 d_rtb_hardlimit;/* absolute limit on realtime blks */ + __be64 d_rtb_softlimit;/* preferred limit on RT disk blks */ + __be64 d_rtbcount; /* realtime blocks owned */ + __be32 d_rtbtimer; /* similar to above; for RT disk blocks */ + __be16 d_rtbwarns; /* warnings issued wrt RT disk blocks */ + __be16 d_pad; } xfs_disk_dquot_t; /* diff --git a/fs/xfs/xfs_refcache.h b/fs/xfs/xfs_refcache.h index cd8ddfd35d69..2dec79edb510 100644 --- a/fs/xfs/xfs_refcache.h +++ b/fs/xfs/xfs_refcache.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_REFCACHE_H__ #define __XFS_REFCACHE_H__ diff --git a/fs/xfs/xfs_rename.c b/fs/xfs/xfs_rename.c index 23b48ac1cb7e..4d4e8f4e768e 100644 --- a/fs/xfs/xfs_rename.c +++ b/fs/xfs/xfs_rename.c @@ -1,60 +1,45 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_dir.h" #include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" +#include "xfs_da_btree.h" #include "xfs_bmap_btree.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" -#include "xfs_inode_item.h" #include "xfs_inode.h" +#include "xfs_inode_item.h" #include "xfs_bmap.h" #include "xfs_error.h" #include "xfs_quota.h" #include "xfs_refcache.h" #include "xfs_utils.h" #include "xfs_trans_space.h" -#include "xfs_da_btree.h" #include "xfs_dir_leaf.h" @@ -620,8 +605,6 @@ xfs_rename( IRELE(target_ip); } - FSC_NOTIFY_NAME_CHANGED(XFS_ITOV(src_ip)); - IRELE(src_ip); /* Fall through to std_return with error = 0 or errno from diff --git a/fs/xfs/xfs_rtalloc.c b/fs/xfs/xfs_rtalloc.c index 2c37822d1012..06fc061c50fc 100644 --- a/fs/xfs/xfs_rtalloc.c +++ b/fs/xfs/xfs_rtalloc.c @@ -1,44 +1,26 @@ /* - * Copyright (c) 2000-2004 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - -/* - * Free realtime space allocation for XFS. - */ - #include "xfs.h" -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" @@ -46,19 +28,18 @@ #include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" -#include "xfs_alloc_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_btree.h" -#include "xfs_ialloc.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" #include "xfs_inode.h" +#include "xfs_btree.h" +#include "xfs_ialloc.h" #include "xfs_alloc.h" #include "xfs_bmap.h" -#include "xfs_bit.h" #include "xfs_rtalloc.h" #include "xfs_fsops.h" #include "xfs_error.h" diff --git a/fs/xfs/xfs_rtalloc.h b/fs/xfs/xfs_rtalloc.h index e2710264c054..0e0b4d2ec202 100644 --- a/fs/xfs/xfs_rtalloc.h +++ b/fs/xfs/xfs_rtalloc.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_RTALLOC_H__ #define __XFS_RTALLOC_H__ diff --git a/fs/xfs/xfs_rw.c b/fs/xfs/xfs_rw.c index d3ff7aef33ba..c4b20872f07d 100644 --- a/fs/xfs/xfs_rw.c +++ b/fs/xfs/xfs_rw.c @@ -1,40 +1,26 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" @@ -42,20 +28,20 @@ #include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" -#include "xfs_alloc_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" +#include "xfs_dir_sf.h" +#include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" +#include "xfs_dinode.h" +#include "xfs_inode.h" +#include "xfs_inode_item.h" #include "xfs_itable.h" #include "xfs_btree.h" #include "xfs_alloc.h" #include "xfs_ialloc.h" #include "xfs_attr.h" -#include "xfs_attr_sf.h" -#include "xfs_dir_sf.h" -#include "xfs_dir2_sf.h" -#include "xfs_dinode.h" -#include "xfs_inode_item.h" -#include "xfs_inode.h" #include "xfs_bmap.h" #include "xfs_acl.h" #include "xfs_mac.h" @@ -264,7 +250,7 @@ xfs_ioerror_alert( { cmn_err(CE_ALERT, "I/O error in filesystem (\"%s\") meta-data dev %s block 0x%llx" - " (\"%s\") error %d buf count %u", + " (\"%s\") error %d buf count %zd", (!mp || !mp->m_fsname) ? "(fs name not set)" : mp->m_fsname, XFS_BUFTARG_NAME(bp->pb_target), (__uint64_t)blkno, diff --git a/fs/xfs/xfs_rw.h b/fs/xfs/xfs_rw.h index c8b10bf8f530..de85eefb7966 100644 --- a/fs/xfs/xfs_rw.h +++ b/fs/xfs/xfs_rw.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_RW_H__ #define __XFS_RW_H__ @@ -68,87 +54,44 @@ struct xfs_mount; * file is a real time file or not, because the bmap code * does. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_FSB_TO_DB) -xfs_daddr_t xfs_fsb_to_db(struct xfs_inode *ip, xfs_fsblock_t fsb); #define XFS_FSB_TO_DB(ip,fsb) xfs_fsb_to_db(ip,fsb) -#else -#define XFS_FSB_TO_DB(ip,fsb) \ - (((ip)->i_d.di_flags & XFS_DIFLAG_REALTIME) ? \ +static inline xfs_daddr_t +xfs_fsb_to_db(struct xfs_inode *ip, xfs_fsblock_t fsb) +{ + return (((ip)->i_d.di_flags & XFS_DIFLAG_REALTIME) ? \ (xfs_daddr_t)XFS_FSB_TO_BB((ip)->i_mount, (fsb)) : \ - XFS_FSB_TO_DADDR((ip)->i_mount, (fsb))) -#endif - -#define XFS_FSB_TO_DB_IO(io,fsb) \ - (((io)->io_flags & XFS_IOCORE_RT) ? \ + XFS_FSB_TO_DADDR((ip)->i_mount, (fsb))); +} +#define XFS_FSB_TO_DB_IO(io,fsb) xfs_fsb_to_db_io(io,fsb) +static inline xfs_daddr_t +xfs_fsb_to_db_io(struct xfs_iocore *io, xfs_fsblock_t fsb) +{ + return (((io)->io_flags & XFS_IOCORE_RT) ? \ XFS_FSB_TO_BB((io)->io_mount, (fsb)) : \ - XFS_FSB_TO_DADDR((io)->io_mount, (fsb))) + XFS_FSB_TO_DADDR((io)->io_mount, (fsb))); +} /* * Prototypes for functions in xfs_rw.c. */ - -int -xfs_write_clear_setuid( - struct xfs_inode *ip); - -int -xfs_bwrite( - struct xfs_mount *mp, - struct xfs_buf *bp); - -int -xfs_bioerror( - struct xfs_buf *b); - -int -xfs_bioerror_relse( - struct xfs_buf *b); - -int -xfs_read_buf( - struct xfs_mount *mp, - xfs_buftarg_t *target, - xfs_daddr_t blkno, - int len, - uint flags, - struct xfs_buf **bpp); - -void -xfs_ioerror_alert( - char *func, - struct xfs_mount *mp, - xfs_buf_t *bp, - xfs_daddr_t blkno); - +extern int xfs_write_clear_setuid(struct xfs_inode *ip); +extern int xfs_bwrite(struct xfs_mount *mp, struct xfs_buf *bp); +extern int xfs_bioerror(struct xfs_buf *bp); +extern int xfs_bioerror_relse(struct xfs_buf *bp); +extern int xfs_read_buf(struct xfs_mount *mp, xfs_buftarg_t *btp, + xfs_daddr_t blkno, int len, uint flags, + struct xfs_buf **bpp); +extern void xfs_ioerror_alert(char *func, struct xfs_mount *mp, + xfs_buf_t *bp, xfs_daddr_t blkno); /* * Prototypes for functions in xfs_vnodeops.c. */ - -int -xfs_rwlock( - bhv_desc_t *bdp, - vrwlock_t write_lock); - -void -xfs_rwunlock( - bhv_desc_t *bdp, - vrwlock_t write_lock); - -int -xfs_change_file_space( - bhv_desc_t *bdp, - int cmd, - xfs_flock64_t *bf, - xfs_off_t offset, - cred_t *credp, - int flags); - -int -xfs_set_dmattrs( - bhv_desc_t *bdp, - u_int evmask, - u_int16_t state, - cred_t *credp); +extern int xfs_rwlock(bhv_desc_t *bdp, vrwlock_t write_lock); +extern void xfs_rwunlock(bhv_desc_t *bdp, vrwlock_t write_lock); +extern int xfs_change_file_space(bhv_desc_t *bdp, int cmd, xfs_flock64_t *bf, + xfs_off_t offset, cred_t *credp, int flags); +extern int xfs_set_dmattrs(bhv_desc_t *bdp, u_int evmask, u_int16_t state, + cred_t *credp); #endif /* __XFS_RW_H__ */ diff --git a/fs/xfs/xfs_sb.h b/fs/xfs/xfs_sb.h index ad090a834ced..4a17d335f897 100644 --- a/fs/xfs/xfs_sb.h +++ b/fs/xfs/xfs_sb.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2001 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_SB_H__ #define __XFS_SB_H__ @@ -72,7 +58,8 @@ struct xfs_mount; XFS_SB_VERSION_DALIGNBIT | \ XFS_SB_VERSION_SHAREDBIT | \ XFS_SB_VERSION_LOGV2BIT | \ - XFS_SB_VERSION_SECTORBIT) + XFS_SB_VERSION_SECTORBIT | \ + XFS_SB_VERSION_MOREBITSBIT) #define XFS_SB_VERSION_OKSASHBITS \ (XFS_SB_VERSION_NUMBITS | \ XFS_SB_VERSION_REALFBITS | \ @@ -103,12 +90,15 @@ struct xfs_mount; */ #define XFS_SB_VERSION2_REALFBITS 0x00ffffff /* Mask: features */ #define XFS_SB_VERSION2_RESERVED1BIT 0x00000001 +#define XFS_SB_VERSION2_RESERVED2BIT 0x00000002 +#define XFS_SB_VERSION2_RESERVED4BIT 0x00000004 +#define XFS_SB_VERSION2_ATTR2BIT 0x00000008 /* Inline attr rework */ #define XFS_SB_VERSION2_SASHFBITS 0xff000000 /* Mask: features that require changing PROM and SASH */ #define XFS_SB_VERSION2_OKREALFBITS \ - (0) + (XFS_SB_VERSION2_ATTR2BIT) #define XFS_SB_VERSION2_OKSASHFBITS \ (0) #define XFS_SB_VERSION2_OKREALBITS \ @@ -118,8 +108,7 @@ struct xfs_mount; /* * mkfs macro to set up sb_features2 word */ -#define XFS_SB_VERSION2_MKFS(xyz) \ - ((xyz) ? 0 : 0) +#define XFS_SB_VERSION2_MKFS(resvd1, sbcntr) 0 typedef struct xfs_sb { @@ -176,7 +165,7 @@ typedef struct xfs_sb __uint8_t sb_logsectlog; /* log2 of the log sector size */ __uint16_t sb_logsectsize; /* sector size for the log, bytes */ __uint32_t sb_logsunit; /* stripe unit size for the log */ - __uint32_t sb_features2; /* additonal feature bits */ + __uint32_t sb_features2; /* additional feature bits */ } xfs_sb_t; /* @@ -216,12 +205,15 @@ typedef enum { #define XFS_SB_SHARED_VN XFS_SB_MVAL(SHARED_VN) #define XFS_SB_UNIT XFS_SB_MVAL(UNIT) #define XFS_SB_WIDTH XFS_SB_MVAL(WIDTH) +#define XFS_SB_FEATURES2 XFS_SB_MVAL(FEATURES2) #define XFS_SB_NUM_BITS ((int)XFS_SBS_FIELDCOUNT) #define XFS_SB_ALL_BITS ((1LL << XFS_SB_NUM_BITS) - 1) #define XFS_SB_MOD_BITS \ (XFS_SB_UUID | XFS_SB_ROOTINO | XFS_SB_RBMINO | XFS_SB_RSUMINO | \ XFS_SB_VERSIONNUM | XFS_SB_UQUOTINO | XFS_SB_GQUOTINO | \ - XFS_SB_QFLAGS | XFS_SB_SHARED_VN | XFS_SB_UNIT | XFS_SB_WIDTH) + XFS_SB_QFLAGS | XFS_SB_SHARED_VN | XFS_SB_UNIT | XFS_SB_WIDTH | \ + XFS_SB_FEATURES2) + /* * Misc. Flags - warning - these will be cleared by xfs_repair unless @@ -235,42 +227,33 @@ typedef enum { */ #define XFS_SB_MAX_SHARED_VN 0 -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_SB_VERSION_NUM) -int xfs_sb_version_num(xfs_sb_t *sbp); -#define XFS_SB_VERSION_NUM(sbp) xfs_sb_version_num(sbp) -#else #define XFS_SB_VERSION_NUM(sbp) ((sbp)->sb_versionnum & XFS_SB_VERSION_NUMBITS) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_SB_GOOD_VERSION) -int xfs_sb_good_version(xfs_sb_t *sbp); #define XFS_SB_GOOD_VERSION(sbp) xfs_sb_good_version(sbp) -#else -#define XFS_SB_GOOD_VERSION_INT(sbp) \ - ((((sbp)->sb_versionnum >= XFS_SB_VERSION_1) && \ - ((sbp)->sb_versionnum <= XFS_SB_VERSION_3)) || \ - ((XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ - !(((sbp)->sb_versionnum & ~XFS_SB_VERSION_OKREALBITS) || \ - (((sbp)->sb_versionnum & XFS_SB_VERSION_MOREBITSBIT) && \ - ((sbp)->sb_features2 & ~XFS_SB_VERSION2_OKREALBITS))) - #ifdef __KERNEL__ -#define XFS_SB_GOOD_VERSION(sbp) \ - (XFS_SB_GOOD_VERSION_INT(sbp) && \ - (sbp)->sb_shared_vn <= XFS_SB_MAX_SHARED_VN) )) +static inline int xfs_sb_good_version(xfs_sb_t *sbp) +{ + return (((sbp->sb_versionnum >= XFS_SB_VERSION_1) && \ + (sbp->sb_versionnum <= XFS_SB_VERSION_3)) || \ + ((XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ + !((sbp->sb_versionnum & ~XFS_SB_VERSION_OKREALBITS) || \ + ((sbp->sb_versionnum & XFS_SB_VERSION_MOREBITSBIT) && \ + (sbp->sb_features2 & ~XFS_SB_VERSION2_OKREALBITS))) && \ + (sbp->sb_shared_vn <= XFS_SB_MAX_SHARED_VN))); +} #else -/* - * extra 2 paren's here (( to unconfuse paren-matching editors - * like vi because XFS_SB_GOOD_VERSION_INT is a partial expression - * and the two XFS_SB_GOOD_VERSION's each 2 more close paren's to - * complete the expression. - */ -#define XFS_SB_GOOD_VERSION(sbp) \ - (XFS_SB_GOOD_VERSION_INT(sbp) && \ - (!((sbp)->sb_versionnum & XFS_SB_VERSION_SHAREDBIT) || \ - (sbp)->sb_shared_vn <= XFS_SB_MAX_SHARED_VN)) )) +static inline int xfs_sb_good_version(xfs_sb_t *sbp) +{ + return (((sbp->sb_versionnum >= XFS_SB_VERSION_1) && \ + (sbp->sb_versionnum <= XFS_SB_VERSION_3)) || \ + ((XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ + !((sbp->sb_versionnum & ~XFS_SB_VERSION_OKREALBITS) || \ + ((sbp->sb_versionnum & XFS_SB_VERSION_MOREBITSBIT) && \ + (sbp->sb_features2 & ~XFS_SB_VERSION2_OKREALBITS))) && \ + (!(sbp->sb_versionnum & XFS_SB_VERSION_SHAREDBIT) || \ + (sbp->sb_shared_vn <= XFS_SB_MAX_SHARED_VN)))); +} #endif /* __KERNEL__ */ -#endif #define XFS_SB_GOOD_SASH_VERSION(sbp) \ ((((sbp)->sb_versionnum >= XFS_SB_VERSION_1) && \ @@ -278,275 +261,218 @@ int xfs_sb_good_version(xfs_sb_t *sbp); ((XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ !((sbp)->sb_versionnum & ~XFS_SB_VERSION_OKSASHBITS))) -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_SB_VERSION_TONEW) -unsigned xfs_sb_version_tonew(unsigned v); #define XFS_SB_VERSION_TONEW(v) xfs_sb_version_tonew(v) -#else -#define XFS_SB_VERSION_TONEW(v) \ - ((((v) == XFS_SB_VERSION_1) ? \ +static inline unsigned xfs_sb_version_tonew(unsigned v) +{ + return ((((v) == XFS_SB_VERSION_1) ? \ 0 : \ (((v) == XFS_SB_VERSION_2) ? \ XFS_SB_VERSION_ATTRBIT : \ (XFS_SB_VERSION_ATTRBIT | XFS_SB_VERSION_NLINKBIT))) | \ - XFS_SB_VERSION_4) -#endif + XFS_SB_VERSION_4); +} -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_SB_VERSION_TOOLD) -unsigned xfs_sb_version_toold(unsigned v); #define XFS_SB_VERSION_TOOLD(v) xfs_sb_version_toold(v) -#else -#define XFS_SB_VERSION_TOOLD(v) \ - (((v) & (XFS_SB_VERSION_QUOTABIT | XFS_SB_VERSION_ALIGNBIT)) ? \ +static inline unsigned xfs_sb_version_toold(unsigned v) +{ + return (((v) & (XFS_SB_VERSION_QUOTABIT | XFS_SB_VERSION_ALIGNBIT)) ? \ 0 : \ (((v) & XFS_SB_VERSION_NLINKBIT) ? \ XFS_SB_VERSION_3 : \ (((v) & XFS_SB_VERSION_ATTRBIT) ? \ XFS_SB_VERSION_2 : \ - XFS_SB_VERSION_1))) -#endif + XFS_SB_VERSION_1))); +} -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_SB_VERSION_HASATTR) -int xfs_sb_version_hasattr(xfs_sb_t *sbp); #define XFS_SB_VERSION_HASATTR(sbp) xfs_sb_version_hasattr(sbp) -#else -#define XFS_SB_VERSION_HASATTR(sbp) \ - (((sbp)->sb_versionnum == XFS_SB_VERSION_2) || \ - ((sbp)->sb_versionnum == XFS_SB_VERSION_3) || \ - ((XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ - ((sbp)->sb_versionnum & XFS_SB_VERSION_ATTRBIT))) -#endif +static inline int xfs_sb_version_hasattr(xfs_sb_t *sbp) +{ + return ((sbp)->sb_versionnum == XFS_SB_VERSION_2) || \ + ((sbp)->sb_versionnum == XFS_SB_VERSION_3) || \ + ((XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ + ((sbp)->sb_versionnum & XFS_SB_VERSION_ATTRBIT)); +} -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_SB_VERSION_ADDATTR) -void xfs_sb_version_addattr(xfs_sb_t *sbp); #define XFS_SB_VERSION_ADDATTR(sbp) xfs_sb_version_addattr(sbp) -#else -#define XFS_SB_VERSION_ADDATTR(sbp) \ - ((sbp)->sb_versionnum = \ - (((sbp)->sb_versionnum == XFS_SB_VERSION_1) ? \ +static inline void xfs_sb_version_addattr(xfs_sb_t *sbp) +{ + (sbp)->sb_versionnum = (((sbp)->sb_versionnum == XFS_SB_VERSION_1) ? \ XFS_SB_VERSION_2 : \ ((XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) ? \ ((sbp)->sb_versionnum | XFS_SB_VERSION_ATTRBIT) : \ - (XFS_SB_VERSION_4 | XFS_SB_VERSION_ATTRBIT)))) -#endif + (XFS_SB_VERSION_4 | XFS_SB_VERSION_ATTRBIT))); +} -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_SB_VERSION_HASNLINK) -int xfs_sb_version_hasnlink(xfs_sb_t *sbp); #define XFS_SB_VERSION_HASNLINK(sbp) xfs_sb_version_hasnlink(sbp) -#else -#define XFS_SB_VERSION_HASNLINK(sbp) \ - (((sbp)->sb_versionnum == XFS_SB_VERSION_3) || \ - ((XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ - ((sbp)->sb_versionnum & XFS_SB_VERSION_NLINKBIT))) -#endif +static inline int xfs_sb_version_hasnlink(xfs_sb_t *sbp) +{ + return ((sbp)->sb_versionnum == XFS_SB_VERSION_3) || \ + ((XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ + ((sbp)->sb_versionnum & XFS_SB_VERSION_NLINKBIT)); +} -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_SB_VERSION_ADDNLINK) -void xfs_sb_version_addnlink(xfs_sb_t *sbp); #define XFS_SB_VERSION_ADDNLINK(sbp) xfs_sb_version_addnlink(sbp) -#else -#define XFS_SB_VERSION_ADDNLINK(sbp) \ - ((sbp)->sb_versionnum = \ - ((sbp)->sb_versionnum <= XFS_SB_VERSION_2 ? \ +static inline void xfs_sb_version_addnlink(xfs_sb_t *sbp) +{ + (sbp)->sb_versionnum = ((sbp)->sb_versionnum <= XFS_SB_VERSION_2 ? \ XFS_SB_VERSION_3 : \ - ((sbp)->sb_versionnum | XFS_SB_VERSION_NLINKBIT))) -#endif + ((sbp)->sb_versionnum | XFS_SB_VERSION_NLINKBIT)); +} -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_SB_VERSION_HASQUOTA) -int xfs_sb_version_hasquota(xfs_sb_t *sbp); #define XFS_SB_VERSION_HASQUOTA(sbp) xfs_sb_version_hasquota(sbp) -#else -#define XFS_SB_VERSION_HASQUOTA(sbp) \ - ((XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ - ((sbp)->sb_versionnum & XFS_SB_VERSION_QUOTABIT)) -#endif +static inline int xfs_sb_version_hasquota(xfs_sb_t *sbp) +{ + return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ + ((sbp)->sb_versionnum & XFS_SB_VERSION_QUOTABIT); +} -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_SB_VERSION_ADDQUOTA) -void xfs_sb_version_addquota(xfs_sb_t *sbp); #define XFS_SB_VERSION_ADDQUOTA(sbp) xfs_sb_version_addquota(sbp) -#else -#define XFS_SB_VERSION_ADDQUOTA(sbp) \ - ((sbp)->sb_versionnum = \ - (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4 ? \ - ((sbp)->sb_versionnum | XFS_SB_VERSION_QUOTABIT) : \ - (XFS_SB_VERSION_TONEW((sbp)->sb_versionnum) | \ - XFS_SB_VERSION_QUOTABIT))) -#endif - -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_SB_VERSION_HASALIGN) -int xfs_sb_version_hasalign(xfs_sb_t *sbp); +static inline void xfs_sb_version_addquota(xfs_sb_t *sbp) +{ + (sbp)->sb_versionnum = \ + (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4 ? \ + ((sbp)->sb_versionnum | XFS_SB_VERSION_QUOTABIT) : \ + (XFS_SB_VERSION_TONEW((sbp)->sb_versionnum) | \ + XFS_SB_VERSION_QUOTABIT)); +} + #define XFS_SB_VERSION_HASALIGN(sbp) xfs_sb_version_hasalign(sbp) -#else -#define XFS_SB_VERSION_HASALIGN(sbp) \ - ((XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ - ((sbp)->sb_versionnum & XFS_SB_VERSION_ALIGNBIT)) -#endif +static inline int xfs_sb_version_hasalign(xfs_sb_t *sbp) +{ + return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ + ((sbp)->sb_versionnum & XFS_SB_VERSION_ALIGNBIT); +} -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_SB_VERSION_SUBALIGN) -void xfs_sb_version_subalign(xfs_sb_t *sbp); #define XFS_SB_VERSION_SUBALIGN(sbp) xfs_sb_version_subalign(sbp) -#else -#define XFS_SB_VERSION_SUBALIGN(sbp) \ - ((sbp)->sb_versionnum = \ - XFS_SB_VERSION_TOOLD((sbp)->sb_versionnum & ~XFS_SB_VERSION_ALIGNBIT)) -#endif +static inline void xfs_sb_version_subalign(xfs_sb_t *sbp) +{ + (sbp)->sb_versionnum = \ + XFS_SB_VERSION_TOOLD((sbp)->sb_versionnum & ~XFS_SB_VERSION_ALIGNBIT); +} -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_SB_VERSION_HASDALIGN) -int xfs_sb_version_hasdalign(xfs_sb_t *sbp); #define XFS_SB_VERSION_HASDALIGN(sbp) xfs_sb_version_hasdalign(sbp) -#else -#define XFS_SB_VERSION_HASDALIGN(sbp) \ - ((XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ - ((sbp)->sb_versionnum & XFS_SB_VERSION_DALIGNBIT)) -#endif +static inline int xfs_sb_version_hasdalign(xfs_sb_t *sbp) +{ + return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ + ((sbp)->sb_versionnum & XFS_SB_VERSION_DALIGNBIT); +} -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_SB_VERSION_ADDDALIGN) -int xfs_sb_version_adddalign(xfs_sb_t *sbp); #define XFS_SB_VERSION_ADDDALIGN(sbp) xfs_sb_version_adddalign(sbp) -#else -#define XFS_SB_VERSION_ADDDALIGN(sbp) \ - ((sbp)->sb_versionnum = \ - ((sbp)->sb_versionnum | XFS_SB_VERSION_DALIGNBIT)) -#endif +static inline int xfs_sb_version_adddalign(xfs_sb_t *sbp) +{ + return (sbp)->sb_versionnum = \ + ((sbp)->sb_versionnum | XFS_SB_VERSION_DALIGNBIT); +} -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_SB_VERSION_HASSHARED) -int xfs_sb_version_hasshared(xfs_sb_t *sbp); #define XFS_SB_VERSION_HASSHARED(sbp) xfs_sb_version_hasshared(sbp) -#else -#define XFS_SB_VERSION_HASSHARED(sbp) \ - ((XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ - ((sbp)->sb_versionnum & XFS_SB_VERSION_SHAREDBIT)) -#endif +static inline int xfs_sb_version_hasshared(xfs_sb_t *sbp) +{ + return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ + ((sbp)->sb_versionnum & XFS_SB_VERSION_SHAREDBIT); +} -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_SB_VERSION_ADDSHARED) -int xfs_sb_version_addshared(xfs_sb_t *sbp); #define XFS_SB_VERSION_ADDSHARED(sbp) xfs_sb_version_addshared(sbp) -#else -#define XFS_SB_VERSION_ADDSHARED(sbp) \ - ((sbp)->sb_versionnum = \ - ((sbp)->sb_versionnum | XFS_SB_VERSION_SHAREDBIT)) -#endif +static inline int xfs_sb_version_addshared(xfs_sb_t *sbp) +{ + return (sbp)->sb_versionnum = \ + ((sbp)->sb_versionnum | XFS_SB_VERSION_SHAREDBIT); +} -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_SB_VERSION_SUBSHARED) -int xfs_sb_version_subshared(xfs_sb_t *sbp); #define XFS_SB_VERSION_SUBSHARED(sbp) xfs_sb_version_subshared(sbp) -#else -#define XFS_SB_VERSION_SUBSHARED(sbp) \ - ((sbp)->sb_versionnum = \ - ((sbp)->sb_versionnum & ~XFS_SB_VERSION_SHAREDBIT)) -#endif +static inline int xfs_sb_version_subshared(xfs_sb_t *sbp) +{ + return (sbp)->sb_versionnum = \ + ((sbp)->sb_versionnum & ~XFS_SB_VERSION_SHAREDBIT); +} -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_SB_VERSION_HASDIRV2) -int xfs_sb_version_hasdirv2(xfs_sb_t *sbp); #define XFS_SB_VERSION_HASDIRV2(sbp) xfs_sb_version_hasdirv2(sbp) -#else -#define XFS_SB_VERSION_HASDIRV2(sbp) \ - ((XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ - ((sbp)->sb_versionnum & XFS_SB_VERSION_DIRV2BIT)) -#endif +static inline int xfs_sb_version_hasdirv2(xfs_sb_t *sbp) +{ + return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ + ((sbp)->sb_versionnum & XFS_SB_VERSION_DIRV2BIT); +} -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_SB_VERSION_HASLOGV2) -int xfs_sb_version_haslogv2(xfs_sb_t *sbp); #define XFS_SB_VERSION_HASLOGV2(sbp) xfs_sb_version_haslogv2(sbp) -#else -#define XFS_SB_VERSION_HASLOGV2(sbp) \ - ((XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ - ((sbp)->sb_versionnum & XFS_SB_VERSION_LOGV2BIT)) -#endif +static inline int xfs_sb_version_haslogv2(xfs_sb_t *sbp) +{ + return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ + ((sbp)->sb_versionnum & XFS_SB_VERSION_LOGV2BIT); +} -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_SB_VERSION_HASEXTFLGBIT) -int xfs_sb_version_hasextflgbit(xfs_sb_t *sbp); #define XFS_SB_VERSION_HASEXTFLGBIT(sbp) xfs_sb_version_hasextflgbit(sbp) -#else -#define XFS_SB_VERSION_HASEXTFLGBIT(sbp) \ - ((XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ - ((sbp)->sb_versionnum & XFS_SB_VERSION_EXTFLGBIT)) -#endif +static inline int xfs_sb_version_hasextflgbit(xfs_sb_t *sbp) +{ + return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ + ((sbp)->sb_versionnum & XFS_SB_VERSION_EXTFLGBIT); +} -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_SB_VERSION_ADDEXTFLGBIT) -int xfs_sb_version_addextflgbit(xfs_sb_t *sbp); #define XFS_SB_VERSION_ADDEXTFLGBIT(sbp) xfs_sb_version_addextflgbit(sbp) -#else -#define XFS_SB_VERSION_ADDEXTFLGBIT(sbp) \ - ((sbp)->sb_versionnum = \ - ((sbp)->sb_versionnum | XFS_SB_VERSION_EXTFLGBIT)) -#endif +static inline int xfs_sb_version_addextflgbit(xfs_sb_t *sbp) +{ + return (sbp)->sb_versionnum = \ + ((sbp)->sb_versionnum | XFS_SB_VERSION_EXTFLGBIT); +} -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_SB_VERSION_SUBEXTFLGBIT) -int xfs_sb_version_subextflgbit(xfs_sb_t *sbp); #define XFS_SB_VERSION_SUBEXTFLGBIT(sbp) xfs_sb_version_subextflgbit(sbp) -#else -#define XFS_SB_VERSION_SUBEXTFLGBIT(sbp) \ - ((sbp)->sb_versionnum = \ - ((sbp)->sb_versionnum & ~XFS_SB_VERSION_EXTFLGBIT)) -#endif +static inline int xfs_sb_version_subextflgbit(xfs_sb_t *sbp) +{ + return (sbp)->sb_versionnum = \ + ((sbp)->sb_versionnum & ~XFS_SB_VERSION_EXTFLGBIT); +} -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_SB_VERSION_HASSECTOR) -int xfs_sb_version_hassector(xfs_sb_t *sbp); #define XFS_SB_VERSION_HASSECTOR(sbp) xfs_sb_version_hassector(sbp) -#else -#define XFS_SB_VERSION_HASSECTOR(sbp) \ - ((XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ - ((sbp)->sb_versionnum & XFS_SB_VERSION_SECTORBIT)) -#endif +static inline int xfs_sb_version_hassector(xfs_sb_t *sbp) +{ + return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ + ((sbp)->sb_versionnum & XFS_SB_VERSION_SECTORBIT); +} -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_SB_VERSION_HASMOREBITSBIT) -int xfs_sb_version_hasmorebits(xfs_sb_t *sbp); #define XFS_SB_VERSION_HASMOREBITS(sbp) xfs_sb_version_hasmorebits(sbp) -#else -#define XFS_SB_VERSION_HASMOREBITS(sbp) \ - ((XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ - ((sbp)->sb_versionnum & XFS_SB_VERSION_MOREBITSBIT)) -#endif +static inline int xfs_sb_version_hasmorebits(xfs_sb_t *sbp) +{ + return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ + ((sbp)->sb_versionnum & XFS_SB_VERSION_MOREBITSBIT); +} /* * sb_features2 bit version macros. * - * For example, for a bit defined as XFS_SB_VERSION2_YBIT, has a macro: + * For example, for a bit defined as XFS_SB_VERSION2_FUNBIT, has a macro: * - * SB_VERSION_HASYBIT(xfs_sb_t *sbp) + * SB_VERSION_HASFUNBIT(xfs_sb_t *sbp) * ((XFS_SB_VERSION_HASMOREBITS(sbp) && - * ((sbp)->sb_versionnum & XFS_SB_VERSION2_YBIT) + * ((sbp)->sb_features2 & XFS_SB_VERSION2_FUNBIT) */ +#define XFS_SB_VERSION_HASATTR2(sbp) xfs_sb_version_hasattr2(sbp) +static inline int xfs_sb_version_hasattr2(xfs_sb_t *sbp) +{ + return (XFS_SB_VERSION_HASMOREBITS(sbp)) && \ + ((sbp)->sb_features2 & XFS_SB_VERSION2_ATTR2BIT); +} + +#define XFS_SB_VERSION_ADDATTR2(sbp) xfs_sb_version_addattr2(sbp) +static inline void xfs_sb_version_addattr2(xfs_sb_t *sbp) +{ + ((sbp)->sb_versionnum = \ + ((sbp)->sb_versionnum | XFS_SB_VERSION_MOREBITSBIT), \ + ((sbp)->sb_features2 = \ + ((sbp)->sb_features2 | XFS_SB_VERSION2_ATTR2BIT))); +} + /* * end of superblock version macros */ -#define XFS_SB_DADDR ((xfs_daddr_t)0) /* daddr in filesystem/ag */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_SB_BLOCK) -xfs_agblock_t xfs_sb_block(struct xfs_mount *mp); -#define XFS_SB_BLOCK(mp) xfs_sb_block(mp) -#else +#define XFS_SB_DADDR ((xfs_daddr_t)0) /* daddr in filesystem/ag */ #define XFS_SB_BLOCK(mp) XFS_HDR_BLOCK(mp, XFS_SB_DADDR) -#endif - -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_HDR_BLOCK) -xfs_agblock_t xfs_hdr_block(struct xfs_mount *mp, xfs_daddr_t d); -#define XFS_HDR_BLOCK(mp,d) xfs_hdr_block(mp,d) -#else -#define XFS_HDR_BLOCK(mp,d) ((xfs_agblock_t)(XFS_BB_TO_FSBT(mp,d))) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_DADDR_TO_FSB) -xfs_fsblock_t xfs_daddr_to_fsb(struct xfs_mount *mp, xfs_daddr_t d); -#define XFS_DADDR_TO_FSB(mp,d) xfs_daddr_to_fsb(mp,d) -#else -#define XFS_DADDR_TO_FSB(mp,d) \ - XFS_AGB_TO_FSB(mp, XFS_DADDR_TO_AGNO(mp,d), XFS_DADDR_TO_AGBNO(mp,d)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_FSB_TO_DADDR) -xfs_daddr_t xfs_fsb_to_daddr(struct xfs_mount *mp, xfs_fsblock_t fsbno); -#define XFS_FSB_TO_DADDR(mp,fsbno) xfs_fsb_to_daddr(mp,fsbno) -#else -#define XFS_FSB_TO_DADDR(mp,fsbno) \ - XFS_AGB_TO_DADDR(mp, XFS_FSB_TO_AGNO(mp,fsbno), \ - XFS_FSB_TO_AGBNO(mp,fsbno)) -#endif - -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_BUF_TO_SBP) -xfs_sb_t *xfs_buf_to_sbp(struct xfs_buf *bp); -#define XFS_BUF_TO_SBP(bp) xfs_buf_to_sbp(bp) -#else #define XFS_BUF_TO_SBP(bp) ((xfs_sb_t *)XFS_BUF_PTR(bp)) -#endif + +#define XFS_HDR_BLOCK(mp,d) ((xfs_agblock_t)XFS_BB_TO_FSBT(mp,d)) +#define XFS_DADDR_TO_FSB(mp,d) XFS_AGB_TO_FSB(mp, \ + XFS_DADDR_TO_AGNO(mp,d), XFS_DADDR_TO_AGBNO(mp,d)) +#define XFS_FSB_TO_DADDR(mp,fsbno) XFS_AGB_TO_DADDR(mp, \ + XFS_FSB_TO_AGNO(mp,fsbno), XFS_FSB_TO_AGBNO(mp,fsbno)) /* * File system sector to basic block conversions. diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c index 92efe272b83d..279e043d7323 100644 --- a/fs/xfs/xfs_trans.c +++ b/fs/xfs/xfs_trans.c @@ -1,40 +1,26 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" @@ -43,21 +29,21 @@ #include "xfs_dmapi.h" #include "xfs_mount.h" #include "xfs_error.h" -#include "xfs_trans_priv.h" -#include "xfs_alloc_btree.h" +#include "xfs_da_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_btree.h" -#include "xfs_ialloc.h" -#include "xfs_alloc.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" #include "xfs_inode.h" +#include "xfs_btree.h" +#include "xfs_ialloc.h" +#include "xfs_alloc.h" #include "xfs_bmap.h" -#include "xfs_da_btree.h" #include "xfs_quota.h" +#include "xfs_trans_priv.h" #include "xfs_trans_space.h" @@ -190,12 +176,8 @@ xfs_trans_dup( XFS_LBC_INIT(&(ntp->t_busy)); ASSERT(tp->t_flags & XFS_TRANS_PERM_LOG_RES); - -#if defined(XLOG_NOLOG) || defined(DEBUG) - ASSERT(!xlog_debug || tp->t_ticket != NULL); -#else ASSERT(tp->t_ticket != NULL); -#endif + ntp->t_flags = XFS_TRANS_PERM_LOG_RES | (tp->t_flags & XFS_TRANS_RESERVE); ntp->t_ticket = tp->t_ticket; ntp->t_blk_res = tp->t_blk_res - tp->t_blk_res_used; @@ -661,10 +643,11 @@ xfs_trans_unreserve_and_mod_sb( */ /*ARGSUSED*/ int -xfs_trans_commit( +_xfs_trans_commit( xfs_trans_t *tp, uint flags, - xfs_lsn_t *commit_lsn_p) + xfs_lsn_t *commit_lsn_p, + int *log_flushed) { xfs_log_iovec_t *log_vector; int nvec; @@ -676,9 +659,6 @@ xfs_trans_commit( int sync; #define XFS_TRANS_LOGVEC_COUNT 16 xfs_log_iovec_t log_vector_fast[XFS_TRANS_LOGVEC_COUNT]; -#if defined(XLOG_NOLOG) || defined(DEBUG) - static xfs_lsn_t trans_lsn = 1; -#endif void *commit_iclog; int shutdown; @@ -729,11 +709,7 @@ shut_us_down: *commit_lsn_p = commit_lsn; return (shutdown); } -#if defined(XLOG_NOLOG) || defined(DEBUG) - ASSERT(!xlog_debug || tp->t_ticket != NULL); -#else ASSERT(tp->t_ticket != NULL); -#endif /* * If we need to update the superblock, then do it now. @@ -750,14 +726,10 @@ shut_us_down: * by using a vector from the stack when it fits. */ nvec = xfs_trans_count_vecs(tp); - if (nvec == 0) { xfs_force_shutdown(mp, XFS_LOG_IO_ERROR); goto shut_us_down; - } - - - if (nvec <= XFS_TRANS_LOGVEC_COUNT) { + } else if (nvec <= XFS_TRANS_LOGVEC_COUNT) { log_vector = log_vector_fast; } else { log_vector = (xfs_log_iovec_t *)kmem_alloc(nvec * @@ -771,30 +743,14 @@ shut_us_down: */ xfs_trans_fill_vecs(tp, log_vector); - /* - * Ignore errors here. xfs_log_done would do the right thing. - * We need to put the ticket, etc. away. - */ - error = xfs_log_write(mp, log_vector, nvec, tp->t_ticket, - &(tp->t_lsn)); + error = xfs_log_write(mp, log_vector, nvec, tp->t_ticket, &(tp->t_lsn)); -#if defined(XLOG_NOLOG) || defined(DEBUG) - if (xlog_debug) { - commit_lsn = xfs_log_done(mp, tp->t_ticket, - &commit_iclog, log_flags); - } else { - commit_lsn = 0; - tp->t_lsn = trans_lsn++; - } -#else /* - * This is the regular case. At this point (after the call finishes), - * the transaction is committed incore and could go out to disk at - * any time. However, all the items associated with the transaction - * are still locked and pinned in memory. + * The transaction is committed incore here, and can go out to disk + * at any time after this call. However, all the items associated + * with the transaction are still locked and pinned in memory. */ commit_lsn = xfs_log_done(mp, tp->t_ticket, &commit_iclog, log_flags); -#endif tp->t_commit_lsn = commit_lsn; if (nvec > XFS_TRANS_LOGVEC_COUNT) { @@ -893,9 +849,11 @@ shut_us_down: * log out now and wait for it. */ if (sync) { - if (!error) - error = xfs_log_force(mp, commit_lsn, - XFS_LOG_FORCE | XFS_LOG_SYNC); + if (!error) { + error = _xfs_log_force(mp, commit_lsn, + XFS_LOG_FORCE | XFS_LOG_SYNC, + log_flushed); + } XFS_STATS_INC(xs_trans_sync); } else { XFS_STATS_INC(xs_trans_async); diff --git a/fs/xfs/xfs_trans.h b/fs/xfs/xfs_trans.h index a263aec8b3a6..a889963fdd14 100644 --- a/fs/xfs/xfs_trans.h +++ b/fs/xfs/xfs_trans.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_TRANS_H__ #define __XFS_TRANS_H__ @@ -135,19 +121,6 @@ typedef struct xfs_ail_entry { struct xfs_log_item *ail_back; /* AIL back pointer */ } xfs_ail_entry_t; -/* - * This structure is passed as a parameter to xfs_trans_push_ail() - * and is used to track the what LSN the waiting processes are - * waiting to become unused. - */ -typedef struct xfs_ail_ticket { - xfs_lsn_t at_lsn; /* lsn waitin for */ - struct xfs_ail_ticket *at_forw; /* wait list ptr */ - struct xfs_ail_ticket *at_back; /* wait list ptr */ - sv_t at_sema; /* wait sema */ -} xfs_ail_ticket_t; - - typedef struct xfs_log_item { xfs_ail_entry_t li_ail; /* AIL pointers */ xfs_lsn_t li_lsn; /* last on-disk lsn */ @@ -247,68 +220,67 @@ typedef struct xfs_log_item_chunk { * lic_unused to the right value (0 matches all free). The * lic_descs.lid_index values are set up as each desc is allocated. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_LIC_INIT) -void xfs_lic_init(xfs_log_item_chunk_t *cp); #define XFS_LIC_INIT(cp) xfs_lic_init(cp) -#else -#define XFS_LIC_INIT(cp) ((cp)->lic_free = XFS_LIC_FREEMASK) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_LIC_INIT_SLOT) -void xfs_lic_init_slot(xfs_log_item_chunk_t *cp, int slot); +static inline void xfs_lic_init(xfs_log_item_chunk_t *cp) +{ + cp->lic_free = XFS_LIC_FREEMASK; +} + #define XFS_LIC_INIT_SLOT(cp,slot) xfs_lic_init_slot(cp, slot) -#else -#define XFS_LIC_INIT_SLOT(cp,slot) \ - ((cp)->lic_descs[slot].lid_index = (unsigned char)(slot)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_LIC_VACANCY) -int xfs_lic_vacancy(xfs_log_item_chunk_t *cp); +static inline void xfs_lic_init_slot(xfs_log_item_chunk_t *cp, int slot) +{ + cp->lic_descs[slot].lid_index = (unsigned char)(slot); +} + #define XFS_LIC_VACANCY(cp) xfs_lic_vacancy(cp) -#else -#define XFS_LIC_VACANCY(cp) (((cp)->lic_free) & XFS_LIC_FREEMASK) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_LIC_ALL_FREE) -void xfs_lic_all_free(xfs_log_item_chunk_t *cp); +static inline int xfs_lic_vacancy(xfs_log_item_chunk_t *cp) +{ + return cp->lic_free & XFS_LIC_FREEMASK; +} + #define XFS_LIC_ALL_FREE(cp) xfs_lic_all_free(cp) -#else -#define XFS_LIC_ALL_FREE(cp) ((cp)->lic_free = XFS_LIC_FREEMASK) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_LIC_ARE_ALL_FREE) -int xfs_lic_are_all_free(xfs_log_item_chunk_t *cp); +static inline void xfs_lic_all_free(xfs_log_item_chunk_t *cp) +{ + cp->lic_free = XFS_LIC_FREEMASK; +} + #define XFS_LIC_ARE_ALL_FREE(cp) xfs_lic_are_all_free(cp) -#else -#define XFS_LIC_ARE_ALL_FREE(cp) (((cp)->lic_free & XFS_LIC_FREEMASK) ==\ - XFS_LIC_FREEMASK) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_LIC_ISFREE) -int xfs_lic_isfree(xfs_log_item_chunk_t *cp, int slot); +static inline int xfs_lic_are_all_free(xfs_log_item_chunk_t *cp) +{ + return ((cp->lic_free & XFS_LIC_FREEMASK) == XFS_LIC_FREEMASK); +} + #define XFS_LIC_ISFREE(cp,slot) xfs_lic_isfree(cp,slot) -#else -#define XFS_LIC_ISFREE(cp,slot) ((cp)->lic_free & (1 << (slot))) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_LIC_CLAIM) -void xfs_lic_claim(xfs_log_item_chunk_t *cp, int slot); +static inline int xfs_lic_isfree(xfs_log_item_chunk_t *cp, int slot) +{ + return (cp->lic_free & (1 << slot)); +} + #define XFS_LIC_CLAIM(cp,slot) xfs_lic_claim(cp,slot) -#else -#define XFS_LIC_CLAIM(cp,slot) ((cp)->lic_free &= ~(1 << (slot))) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_LIC_RELSE) -void xfs_lic_relse(xfs_log_item_chunk_t *cp, int slot); +static inline void xfs_lic_claim(xfs_log_item_chunk_t *cp, int slot) +{ + cp->lic_free &= ~(1 << slot); +} + #define XFS_LIC_RELSE(cp,slot) xfs_lic_relse(cp,slot) -#else -#define XFS_LIC_RELSE(cp,slot) ((cp)->lic_free |= 1 << (slot)) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_LIC_SLOT) -xfs_log_item_desc_t *xfs_lic_slot(xfs_log_item_chunk_t *cp, int slot); +static inline void xfs_lic_relse(xfs_log_item_chunk_t *cp, int slot) +{ + cp->lic_free |= 1 << slot; +} + #define XFS_LIC_SLOT(cp,slot) xfs_lic_slot(cp,slot) -#else -#define XFS_LIC_SLOT(cp,slot) (&((cp)->lic_descs[slot])) -#endif -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_LIC_DESC_TO_SLOT) -int xfs_lic_desc_to_slot(xfs_log_item_desc_t *dp); +static inline xfs_log_item_desc_t * +xfs_lic_slot(xfs_log_item_chunk_t *cp, int slot) +{ + return &(cp->lic_descs[slot]); +} + #define XFS_LIC_DESC_TO_SLOT(dp) xfs_lic_desc_to_slot(dp) -#else -#define XFS_LIC_DESC_TO_SLOT(dp) ((uint)((dp)->lid_index)) -#endif +static inline int xfs_lic_desc_to_slot(xfs_log_item_desc_t *dp) +{ + return (uint)dp->lid_index; +} + /* * Calculate the address of a chunk given a descriptor pointer: * dp - dp->lid_index give the address of the start of the lic_descs array. @@ -316,15 +288,14 @@ int xfs_lic_desc_to_slot(xfs_log_item_desc_t *dp); * All of this yields the address of the chunk, which is * cast to a chunk pointer. */ -#if XFS_WANT_FUNCS || (XFS_WANT_SPACE && XFSSO_XFS_LIC_DESC_TO_CHUNK) -xfs_log_item_chunk_t *xfs_lic_desc_to_chunk(xfs_log_item_desc_t *dp); #define XFS_LIC_DESC_TO_CHUNK(dp) xfs_lic_desc_to_chunk(dp) -#else -#define XFS_LIC_DESC_TO_CHUNK(dp) ((xfs_log_item_chunk_t*) \ - (((xfs_caddr_t)((dp) - (dp)->lid_index)) -\ - (xfs_caddr_t)(((xfs_log_item_chunk_t*) \ - 0)->lic_descs))) -#endif +static inline xfs_log_item_chunk_t * +xfs_lic_desc_to_chunk(xfs_log_item_desc_t *dp) +{ + return (xfs_log_item_chunk_t*) \ + (((xfs_caddr_t)((dp) - (dp)->lid_index)) - \ + (xfs_caddr_t)(((xfs_log_item_chunk_t*)0)->lic_descs)); +} #ifdef __KERNEL__ /* @@ -341,7 +312,7 @@ typedef struct xfs_log_busy_slot { #define XFS_LBC_NUM_SLOTS 31 typedef struct xfs_log_busy_chunk { struct xfs_log_busy_chunk *lbc_next; - uint lbc_free; /* bitmask of free slots */ + uint lbc_free; /* free slots bitmask */ ushort lbc_unused; /* first unused */ xfs_log_busy_slot_t lbc_busy[XFS_LBC_NUM_SLOTS]; } xfs_log_busy_chunk_t; @@ -1025,7 +996,12 @@ void xfs_trans_log_efd_extent(xfs_trans_t *, struct xfs_efd_log_item *, xfs_fsblock_t, xfs_extlen_t); -int xfs_trans_commit(xfs_trans_t *, uint flags, xfs_lsn_t *); +int _xfs_trans_commit(xfs_trans_t *, + uint flags, + xfs_lsn_t *, + int *); +#define xfs_trans_commit(tp, flags, lsn) \ + _xfs_trans_commit(tp, flags, lsn, NULL) void xfs_trans_cancel(xfs_trans_t *, int); void xfs_trans_ail_init(struct xfs_mount *); xfs_lsn_t xfs_trans_push_ail(struct xfs_mount *, xfs_lsn_t); diff --git a/fs/xfs/xfs_trans_ail.c b/fs/xfs/xfs_trans_ail.c index 2a71b4f91bfa..19ab24af1c1c 100644 --- a/fs/xfs/xfs_trans_ail.c +++ b/fs/xfs/xfs_trans_ail.c @@ -1,40 +1,25 @@ /* - * Copyright (c) 2000-2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_dir.h" diff --git a/fs/xfs/xfs_trans_buf.c b/fs/xfs/xfs_trans_buf.c index e733293dd7f4..c74c31ebc81c 100644 --- a/fs/xfs/xfs_trans_buf.c +++ b/fs/xfs/xfs_trans_buf.c @@ -1,47 +1,42 @@ /* - * Copyright (c) 2000-2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" -#include "xfs_buf_item.h" #include "xfs_sb.h" #include "xfs_ag.h" #include "xfs_dir.h" +#include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" +#include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" +#include "xfs_ialloc_btree.h" +#include "xfs_dir_sf.h" +#include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" +#include "xfs_dinode.h" +#include "xfs_inode.h" +#include "xfs_buf_item.h" #include "xfs_trans_priv.h" #include "xfs_error.h" #include "xfs_rw.h" diff --git a/fs/xfs/xfs_trans_extfree.c b/fs/xfs/xfs_trans_extfree.c index 93259a15f983..7d7d627f25df 100644 --- a/fs/xfs/xfs_trans_extfree.c +++ b/fs/xfs/xfs_trans_extfree.c @@ -1,40 +1,25 @@ /* - * Copyright (c) 2000 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_dir.h" diff --git a/fs/xfs/xfs_trans_inode.c b/fs/xfs/xfs_trans_inode.c index 7e7631ca4979..e341409172d2 100644 --- a/fs/xfs/xfs_trans_inode.c +++ b/fs/xfs/xfs_trans_inode.c @@ -1,40 +1,26 @@ /* - * Copyright (c) 2000 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" @@ -42,18 +28,18 @@ #include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" -#include "xfs_trans_priv.h" -#include "xfs_alloc_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_btree.h" -#include "xfs_ialloc.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" -#include "xfs_inode_item.h" #include "xfs_inode.h" +#include "xfs_btree.h" +#include "xfs_ialloc.h" +#include "xfs_trans_priv.h" +#include "xfs_inode_item.h" #ifdef XFS_TRANS_DEBUG STATIC void diff --git a/fs/xfs/xfs_trans_item.c b/fs/xfs/xfs_trans_item.c index 1b8a756d80ed..486147ef0e3d 100644 --- a/fs/xfs/xfs_trans_item.c +++ b/fs/xfs/xfs_trans_item.c @@ -1,40 +1,25 @@ /* - * Copyright (c) 2000-2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" STATIC int xfs_trans_unlock_chunk(xfs_log_item_chunk_t *, diff --git a/fs/xfs/xfs_trans_priv.h b/fs/xfs/xfs_trans_priv.h index d4dae7d06afc..13edab8a9e94 100644 --- a/fs/xfs/xfs_trans_priv.h +++ b/fs/xfs/xfs_trans_priv.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000, 2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000,2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_TRANS_PRIV_H__ #define __XFS_TRANS_PRIV_H__ diff --git a/fs/xfs/xfs_trans_space.h b/fs/xfs/xfs_trans_space.h index e91d173f4ed3..7fe3792b18df 100644 --- a/fs/xfs/xfs_trans_space.h +++ b/fs/xfs/xfs_trans_space.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_TRANS_SPACE_H__ #define __XFS_TRANS_SPACE_H__ diff --git a/fs/xfs/xfs_types.h b/fs/xfs/xfs_types.h index 16f5371ce102..104f64a98790 100644 --- a/fs/xfs/xfs_types.h +++ b/fs/xfs/xfs_types.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2003 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_TYPES_H__ #define __XFS_TYPES_H__ @@ -154,6 +140,12 @@ typedef __uint8_t xfs_arch_t; /* architecture of an xfs fs */ #define MAXAEXTNUM ((xfs_aextnum_t)0x7fff) /* signed short */ /* + * Min numbers of data/attr fork btree root pointers. + */ +#define MINDBTPTRS 3 +#define MINABTPTRS 2 + +/* * MAXNAMELEN is the length (including the terminating null) of * the longest permissible file (component) name. */ diff --git a/fs/xfs/xfs_utils.c b/fs/xfs/xfs_utils.c index 11351f08d438..fefe1d60377f 100644 --- a/fs/xfs/xfs_utils.c +++ b/fs/xfs/xfs_utils.c @@ -1,53 +1,40 @@ /* - * Copyright (c) 2000-2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" +#include "xfs_ag.h" #include "xfs_dir.h" #include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" #include "xfs_bmap_btree.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" -#include "xfs_inode_item.h" #include "xfs_inode.h" +#include "xfs_inode_item.h" #include "xfs_bmap.h" #include "xfs_error.h" #include "xfs_quota.h" diff --git a/fs/xfs/xfs_utils.h b/fs/xfs/xfs_utils.h index 01d98b4b7af7..472661a3b6d8 100644 --- a/fs/xfs/xfs_utils.h +++ b/fs/xfs/xfs_utils.h @@ -1,33 +1,19 @@ /* - * Copyright (c) 2000-2002 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2002,2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ #ifndef __XFS_UTILS_H__ #define __XFS_UTILS_H__ diff --git a/fs/xfs/xfs_vfsops.c b/fs/xfs/xfs_vfsops.c index f1a904e23ade..7bdbd991ab1c 100644 --- a/fs/xfs/xfs_vfsops.c +++ b/fs/xfs/xfs_vfsops.c @@ -1,74 +1,58 @@ /* - * XFS filesystem operations. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * Copyright (c) 2000-2005 Silicon Graphics, Inc. All Rights Reserved. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. - * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" +#include "xfs_ag.h" #include "xfs_dir.h" #include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" +#include "xfs_da_btree.h" #include "xfs_bmap_btree.h" #include "xfs_ialloc_btree.h" #include "xfs_alloc_btree.h" -#include "xfs_btree.h" -#include "xfs_alloc.h" -#include "xfs_ialloc.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" -#include "xfs_inode_item.h" #include "xfs_inode.h" -#include "xfs_ag.h" +#include "xfs_inode_item.h" +#include "xfs_btree.h" +#include "xfs_alloc.h" +#include "xfs_ialloc.h" +#include "xfs_quota.h" #include "xfs_error.h" #include "xfs_bmap.h" -#include "xfs_da_btree.h" #include "xfs_rw.h" #include "xfs_refcache.h" #include "xfs_buf_item.h" -#include "xfs_extfree_item.h" -#include "xfs_quota.h" +#include "xfs_log_priv.h" #include "xfs_dir2_trace.h" +#include "xfs_extfree_item.h" #include "xfs_acl.h" #include "xfs_attr.h" #include "xfs_clnt.h" -#include "xfs_log_priv.h" STATIC int xfs_sync(bhv_desc_t *, int, cred_t *); @@ -230,9 +214,7 @@ xfs_start_flags( } if (ap->logbufs != -1 && -#if defined(DEBUG) || defined(XLOG_NOLOG) ap->logbufs != 0 && -#endif (ap->logbufs < XLOG_MIN_ICLOGS || ap->logbufs > XLOG_MAX_ICLOGS)) { cmn_err(CE_WARN, @@ -242,6 +224,7 @@ xfs_start_flags( } mp->m_logbufs = ap->logbufs; if (ap->logbufsize != -1 && + ap->logbufsize != 0 && ap->logbufsize != 16 * 1024 && ap->logbufsize != 32 * 1024 && ap->logbufsize != 64 * 1024 && @@ -257,6 +240,14 @@ xfs_start_flags( mp->m_fsname_len = strlen(ap->fsname) + 1; mp->m_fsname = kmem_alloc(mp->m_fsname_len, KM_SLEEP); strcpy(mp->m_fsname, ap->fsname); + if (ap->rtname[0]) { + mp->m_rtname = kmem_alloc(strlen(ap->rtname) + 1, KM_SLEEP); + strcpy(mp->m_rtname, ap->rtname); + } + if (ap->logname[0]) { + mp->m_logname = kmem_alloc(strlen(ap->logname) + 1, KM_SLEEP); + strcpy(mp->m_logname, ap->logname); + } if (ap->flags & XFSMNT_WSYNC) mp->m_flags |= XFS_MOUNT_WSYNC; @@ -268,21 +259,16 @@ xfs_start_flags( #endif if (ap->flags & XFSMNT_NOATIME) mp->m_flags |= XFS_MOUNT_NOATIME; - if (ap->flags & XFSMNT_RETERR) mp->m_flags |= XFS_MOUNT_RETERR; - if (ap->flags & XFSMNT_NOALIGN) mp->m_flags |= XFS_MOUNT_NOALIGN; - if (ap->flags & XFSMNT_SWALLOC) mp->m_flags |= XFS_MOUNT_SWALLOC; - if (ap->flags & XFSMNT_OSYNCISOSYNC) mp->m_flags |= XFS_MOUNT_OSYNCISOSYNC; - if (ap->flags & XFSMNT_32BITINODES) - mp->m_flags |= (XFS_MOUNT_32BITINODES | XFS_MOUNT_32BITINOOPT); + mp->m_flags |= XFS_MOUNT_32BITINODES; if (ap->flags & XFSMNT_IOSIZE) { if (ap->iosizelog > XFS_MAX_IO_LOG || @@ -300,12 +286,15 @@ xfs_start_flags( if (ap->flags & XFSMNT_IHASHSIZE) mp->m_flags |= XFS_MOUNT_IHASHSIZE; - if (ap->flags & XFSMNT_IDELETE) mp->m_flags |= XFS_MOUNT_IDELETE; - if (ap->flags & XFSMNT_DIRSYNC) mp->m_flags |= XFS_MOUNT_DIRSYNC; + if (ap->flags & XFSMNT_COMPAT_ATTR) + mp->m_flags |= XFS_MOUNT_COMPAT_ATTR; + + if (ap->flags2 & XFSMNT2_COMPAT_IOSIZE) + mp->m_flags |= XFS_MOUNT_COMPAT_IOSIZE; /* * no recovery flag requires a read-only mount @@ -321,8 +310,8 @@ xfs_start_flags( if (ap->flags & XFSMNT_NOUUID) mp->m_flags |= XFS_MOUNT_NOUUID; - if (ap->flags & XFSMNT_NOLOGFLUSH) - mp->m_flags |= XFS_MOUNT_NOLOGFLUSH; + if (ap->flags & XFSMNT_BARRIER) + mp->m_flags |= XFS_MOUNT_BARRIER; return 0; } @@ -393,6 +382,10 @@ xfs_finish_flags( return XFS_ERROR(EINVAL); } + if (XFS_SB_VERSION_HASATTR2(&mp->m_sb)) { + mp->m_flags &= ~XFS_MOUNT_COMPAT_ATTR; + } + return 0; } @@ -512,8 +505,14 @@ xfs_mount( goto error2; error = XFS_IOINIT(vfsp, args, flags); - if (!error) - return 0; + if (error) + goto error2; + + if ((args->flags & XFSMNT_BARRIER) && + !(XFS_MTOVFS(mp)->vfs_flag & VFS_RDONLY)) + xfs_mountfs_check_barriers(mp); + return 0; + error2: if (mp->m_sb_bp) xfs_freesb(mp); @@ -656,19 +655,24 @@ xfs_mntupdate( else mp->m_flags &= ~XFS_MOUNT_NOATIME; - if (!(vfsp->vfs_flag & VFS_RDONLY)) { - VFS_SYNC(vfsp, SYNC_FSDATA|SYNC_BDFLUSH|SYNC_ATTR, NULL, error); + if ((vfsp->vfs_flag & VFS_RDONLY) && + !(*flags & MS_RDONLY)) { + vfsp->vfs_flag &= ~VFS_RDONLY; + + if (args->flags & XFSMNT_BARRIER) + xfs_mountfs_check_barriers(mp); } - if (*flags & MS_RDONLY) { + if (!(vfsp->vfs_flag & VFS_RDONLY) && + (*flags & MS_RDONLY)) { + VFS_SYNC(vfsp, SYNC_FSDATA|SYNC_BDFLUSH|SYNC_ATTR, NULL, error); + xfs_quiesce_fs(mp); /* Ok now write out an unmount record */ xfs_log_unmount_write(mp); xfs_unmountfs_writesb(mp); vfsp->vfs_flag |= VFS_RDONLY; - } else { - vfsp->vfs_flag &= ~VFS_RDONLY; } return 0; @@ -892,7 +896,7 @@ xfs_sync( * only available by calling this routine. * */ -STATIC int +int xfs_sync_inodes( xfs_mount_t *mp, int flags, @@ -976,7 +980,7 @@ xfs_sync_inodes( ipointer = (xfs_iptr_t *)kmem_zalloc(sizeof(xfs_iptr_t), KM_SLEEP); fflag = XFS_B_ASYNC; /* default is don't wait */ - if (flags & SYNC_BDFLUSH) + if (flags & (SYNC_BDFLUSH | SYNC_DELWRI)) fflag = XFS_B_DELWRI; if (flags & SYNC_WAIT) fflag = 0; /* synchronous overrides all */ @@ -1628,11 +1632,17 @@ xfs_vget( #define MNTOPT_ALLOCSIZE "allocsize" /* preferred allocation size */ #define MNTOPT_IHASHSIZE "ihashsize" /* size of inode hash table */ #define MNTOPT_NORECOVERY "norecovery" /* don't run XFS recovery */ -#define MNTOPT_NOLOGFLUSH "nologflush" /* don't hard flush on log writes */ +#define MNTOPT_BARRIER "barrier" /* use writer barriers for log write and + * unwritten extent conversion */ #define MNTOPT_OSYNCISOSYNC "osyncisosync" /* o_sync is REALLY o_sync */ #define MNTOPT_64BITINODE "inode64" /* inodes can be allocated anywhere */ #define MNTOPT_IKEEP "ikeep" /* do not free empty inode clusters */ #define MNTOPT_NOIKEEP "noikeep" /* free empty inode clusters */ +#define MNTOPT_LARGEIO "largeio" /* report large I/O sizes in stat() */ +#define MNTOPT_NOLARGEIO "nolargeio" /* do not report large I/O sizes + * in stat(). */ +#define MNTOPT_ATTR2 "attr2" /* do use attr2 attribute format */ +#define MNTOPT_NOATTR2 "noattr2" /* do not use attr2 attribute format */ STATIC unsigned long suffix_strtoul(const char *cp, char **endp, unsigned int base) @@ -1669,12 +1679,15 @@ xfs_parseargs( int dsunit, dswidth, vol_dsunit, vol_dswidth; int iosize; + args->flags2 |= XFSMNT2_COMPAT_IOSIZE; + args->flags |= XFSMNT_COMPAT_ATTR; + #if 0 /* XXX: off by default, until some remaining issues ironed out */ args->flags |= XFSMNT_IDELETE; /* default to on */ #endif if (!options) - return 0; + goto done; iosize = dsunit = dswidth = vol_dsunit = vol_dswidth = 0; @@ -1791,12 +1804,20 @@ xfs_parseargs( #endif } else if (!strcmp(this_char, MNTOPT_NOUUID)) { args->flags |= XFSMNT_NOUUID; - } else if (!strcmp(this_char, MNTOPT_NOLOGFLUSH)) { - args->flags |= XFSMNT_NOLOGFLUSH; + } else if (!strcmp(this_char, MNTOPT_BARRIER)) { + args->flags |= XFSMNT_BARRIER; } else if (!strcmp(this_char, MNTOPT_IKEEP)) { args->flags &= ~XFSMNT_IDELETE; } else if (!strcmp(this_char, MNTOPT_NOIKEEP)) { args->flags |= XFSMNT_IDELETE; + } else if (!strcmp(this_char, MNTOPT_LARGEIO)) { + args->flags2 &= ~XFSMNT2_COMPAT_IOSIZE; + } else if (!strcmp(this_char, MNTOPT_NOLARGEIO)) { + args->flags2 |= XFSMNT2_COMPAT_IOSIZE; + } else if (!strcmp(this_char, MNTOPT_ATTR2)) { + args->flags &= ~XFSMNT_COMPAT_ATTR; + } else if (!strcmp(this_char, MNTOPT_NOATTR2)) { + args->flags |= XFSMNT_COMPAT_ATTR; } else if (!strcmp(this_char, "osyncisdsync")) { /* no-op, this is now the default */ printk("XFS: osyncisdsync is now the default, option is deprecated.\n"); @@ -1846,6 +1867,11 @@ printk("XFS: irixsgid is now a sysctl(2) variable, option is deprecated.\n"); args->sunit = args->swidth = 0; } +done: + if (args->flags & XFSMNT_32BITINODES) + vfsp->vfs_flag |= VFS_32BITINODES; + if (args->flags2) + args->flags |= XFSMNT_FLAGS2; return 0; } @@ -1866,7 +1892,7 @@ xfs_showargs( { XFS_MOUNT_NOUUID, "," MNTOPT_NOUUID }, { XFS_MOUNT_NORECOVERY, "," MNTOPT_NORECOVERY }, { XFS_MOUNT_OSYNCISOSYNC, "," MNTOPT_OSYNCISOSYNC }, - { XFS_MOUNT_NOLOGFLUSH, "," MNTOPT_NOLOGFLUSH }, + { XFS_MOUNT_BARRIER, "," MNTOPT_BARRIER }, { XFS_MOUNT_IDELETE, "," MNTOPT_NOIKEEP }, { 0, NULL } }; @@ -1883,21 +1909,20 @@ xfs_showargs( seq_printf(m, "," MNTOPT_IHASHSIZE "=%d", mp->m_ihsize); if (mp->m_flags & XFS_MOUNT_DFLT_IOSIZE) - seq_printf(m, "," MNTOPT_ALLOCSIZE "=%d", 1<<mp->m_writeio_log); + seq_printf(m, "," MNTOPT_ALLOCSIZE "=%dk", + (int)(1 << mp->m_writeio_log) >> 10); if (mp->m_logbufs > 0) seq_printf(m, "," MNTOPT_LOGBUFS "=%d", mp->m_logbufs); if (mp->m_logbsize > 0) - seq_printf(m, "," MNTOPT_LOGBSIZE "=%d", mp->m_logbsize); + seq_printf(m, "," MNTOPT_LOGBSIZE "=%dk", mp->m_logbsize >> 10); - if (mp->m_ddev_targp != mp->m_logdev_targp) - seq_printf(m, "," MNTOPT_LOGDEV "=%s", - XFS_BUFTARG_NAME(mp->m_logdev_targp)); + if (mp->m_logname) + seq_printf(m, "," MNTOPT_LOGDEV "=%s", mp->m_logname); - if (mp->m_rtdev_targp && mp->m_ddev_targp != mp->m_rtdev_targp) - seq_printf(m, "," MNTOPT_RTDEV "=%s", - XFS_BUFTARG_NAME(mp->m_rtdev_targp)); + if (mp->m_rtname) + seq_printf(m, "," MNTOPT_RTDEV "=%s", mp->m_rtname); if (mp->m_dalign > 0) seq_printf(m, "," MNTOPT_SUNIT "=%d", @@ -1907,7 +1932,13 @@ xfs_showargs( seq_printf(m, "," MNTOPT_SWIDTH "=%d", (int)XFS_FSB_TO_BB(mp, mp->m_swidth)); - if (!(mp->m_flags & XFS_MOUNT_32BITINOOPT)) + if (!(mp->m_flags & XFS_MOUNT_COMPAT_ATTR)) + seq_printf(m, "," MNTOPT_ATTR2); + + if (!(mp->m_flags & XFS_MOUNT_COMPAT_IOSIZE)) + seq_printf(m, "," MNTOPT_LARGEIO); + + if (!(vfsp->vfs_flag & VFS_32BITINODES)) seq_printf(m, "," MNTOPT_64BITINODE); if (vfsp->vfs_flag & VFS_GRPID) diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index 58bfe629b933..e03fa2a3d5ed 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -1,40 +1,26 @@ /* - * Copyright (c) 2000-2005 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2000-2005 Silicon Graphics, Inc. + * All Rights Reserved. * - * This program is free software; you can redistribute it and/or modify it - * under the terms of version 2 of the GNU General Public License as + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as * published by the Free Software Foundation. * - * This program is distributed in the hope that it would be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + * This program is distributed in the hope that it would be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. * - * Further, this software is distributed without any warranty that it is - * free of the rightful claim of any third person regarding infringement - * or the like. Any license provided herein, whether implied or - * otherwise, applies only to this software file. Patent licenses, if - * any, provided herein do not apply to combinations of this program with - * other software, or any other product whatsoever. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write the Free Software Foundation, Inc., 59 - * Temple Place - Suite 330, Boston MA 02111-1307, USA. - * - * Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy, - * Mountain View, CA 94043, or: - * - * http://www.sgi.com - * - * For further information regarding this notice, see: - * - * http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/ + * You should have received a copy of the GNU General Public License + * along with this program; if not, write the Free Software Foundation, + * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */ - #include "xfs.h" -#include "xfs_macros.h" +#include "xfs_fs.h" #include "xfs_types.h" -#include "xfs_inum.h" +#include "xfs_bit.h" #include "xfs_log.h" +#include "xfs_inum.h" #include "xfs_trans.h" #include "xfs_sb.h" #include "xfs_ag.h" @@ -42,33 +28,32 @@ #include "xfs_dir2.h" #include "xfs_dmapi.h" #include "xfs_mount.h" -#include "xfs_alloc_btree.h" +#include "xfs_da_btree.h" #include "xfs_bmap_btree.h" +#include "xfs_alloc_btree.h" #include "xfs_ialloc_btree.h" -#include "xfs_itable.h" -#include "xfs_btree.h" -#include "xfs_ialloc.h" -#include "xfs_alloc.h" -#include "xfs_attr_sf.h" #include "xfs_dir_sf.h" #include "xfs_dir2_sf.h" +#include "xfs_attr_sf.h" #include "xfs_dinode.h" -#include "xfs_inode_item.h" #include "xfs_inode.h" +#include "xfs_inode_item.h" +#include "xfs_dir_leaf.h" +#include "xfs_itable.h" +#include "xfs_btree.h" +#include "xfs_ialloc.h" +#include "xfs_alloc.h" #include "xfs_bmap.h" -#include "xfs_da_btree.h" #include "xfs_attr.h" #include "xfs_rw.h" -#include "xfs_refcache.h" #include "xfs_error.h" -#include "xfs_bit.h" -#include "xfs_rtalloc.h" #include "xfs_quota.h" #include "xfs_utils.h" +#include "xfs_rtalloc.h" +#include "xfs_refcache.h" #include "xfs_trans_space.h" -#include "xfs_dir_leaf.h" -#include "xfs_mac.h" #include "xfs_log_priv.h" +#include "xfs_mac.h" /* @@ -181,40 +166,7 @@ xfs_getattr( vap->va_rdev = 0; if (!(ip->i_d.di_flags & XFS_DIFLAG_REALTIME)) { - -#if 0 - /* Large block sizes confuse various - * user space programs, so letting the - * stripe size through is not a good - * idea for now. - */ - vap->va_blocksize = mp->m_swidth ? - /* - * If the underlying volume is a stripe, then - * return the stripe width in bytes as the - * recommended I/O size. - */ - (mp->m_swidth << mp->m_sb.sb_blocklog) : - /* - * Return the largest of the preferred buffer - * sizes since doing small I/Os into larger - * buffers causes buffers to be decommissioned. - * The value returned is in bytes. - */ - (1 << (int)MAX(mp->m_readio_log, - mp->m_writeio_log)); - -#else - vap->va_blocksize = - /* - * Return the largest of the preferred buffer - * sizes since doing small I/Os into larger - * buffers causes buffers to be decommissioned. - * The value returned is in bytes. - */ - 1 << (int)MAX(mp->m_readio_log, - mp->m_writeio_log); -#endif + vap->va_blocksize = xfs_preferred_iosize(mp); } else { /* @@ -581,8 +533,7 @@ xfs_setattr( /* * Can't change extent size if any extents are allocated. */ - if ((ip->i_d.di_nextents || ip->i_delayed_blks) && - (mask & XFS_AT_EXTSIZE) && + if (ip->i_d.di_nextents && (mask & XFS_AT_EXTSIZE) && ((ip->i_d.di_extsize << mp->m_sb.sb_blocklog) != vap->va_extsize) ) { code = XFS_ERROR(EINVAL); /* EFBIG? */ @@ -610,7 +561,8 @@ xfs_setattr( /* * Can't change realtime flag if any extents are allocated. */ - if (ip->i_d.di_nextents && (mask & XFS_AT_XFLAGS) && + if ((ip->i_d.di_nextents || ip->i_delayed_blks) && + (mask & XFS_AT_XFLAGS) && (ip->i_d.di_flags & XFS_DIFLAG_REALTIME) != (vap->va_xflags & XFS_XFLAG_REALTIME)) { code = XFS_ERROR(EINVAL); /* EFBIG? */ @@ -674,8 +626,10 @@ xfs_setattr( */ if (mask & XFS_AT_SIZE) { code = 0; - if (vap->va_size > ip->i_d.di_size) + if ((vap->va_size > ip->i_d.di_size) && + (flags & ATTR_NOSIZETOK) == 0) { code = xfs_igrow_start(ip, vap->va_size, credp); + } xfs_iunlock(ip, XFS_ILOCK_EXCL); if (!code) code = xfs_itruncate_data(ip, vap->va_size); @@ -1118,6 +1072,7 @@ xfs_fsync( xfs_inode_t *ip; xfs_trans_t *tp; int error; + int log_flushed = 0, changed = 1; vn_trace_entry(BHV_TO_VNODE(bdp), __FUNCTION__, (inst_t *)__return_address); @@ -1171,10 +1126,18 @@ xfs_fsync( xfs_iunlock(ip, XFS_ILOCK_SHARED); if (xfs_ipincount(ip)) { - xfs_log_force(ip->i_mount, (xfs_lsn_t)0, + _xfs_log_force(ip->i_mount, (xfs_lsn_t)0, XFS_LOG_FORCE | ((flag & FSYNC_WAIT) - ? XFS_LOG_SYNC : 0)); + ? XFS_LOG_SYNC : 0), + &log_flushed); + } else { + /* + * If the inode is not pinned and nothing + * has changed we don't need to flush the + * cache. + */ + changed = 0; } error = 0; } else { @@ -1210,10 +1173,27 @@ xfs_fsync( xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); if (flag & FSYNC_WAIT) xfs_trans_set_sync(tp); - error = xfs_trans_commit(tp, 0, NULL); + error = _xfs_trans_commit(tp, 0, NULL, &log_flushed); xfs_iunlock(ip, XFS_ILOCK_EXCL); } + + if ((ip->i_mount->m_flags & XFS_MOUNT_BARRIER) && changed) { + /* + * If the log write didn't issue an ordered tag we need + * to flush the disk cache for the data device now. + */ + if (!log_flushed) + xfs_blkdev_issue_flush(ip->i_mount->m_ddev_targp); + + /* + * If this inode is on the RT dev we need to flush that + * cache aswell. + */ + if (ip->i_d.di_flags & XFS_DIFLAG_REALTIME) + xfs_blkdev_issue_flush(ip->i_mount->m_rtdev_targp); + } + return error; } @@ -3978,8 +3958,9 @@ xfs_finish_reclaim_all(xfs_mount_t *mp, int noblock) } } XFS_MOUNT_IUNLOCK(mp); - xfs_finish_reclaim(ip, noblock, - XFS_IFLUSH_DELWRI_ELSE_ASYNC); + if (xfs_finish_reclaim(ip, noblock, + XFS_IFLUSH_DELWRI_ELSE_ASYNC)) + delay(1); purged = 1; break; } |
