diff options
Diffstat (limited to 'fs')
169 files changed, 27896 insertions, 1971 deletions
diff --git a/fs/Kconfig b/fs/Kconfig index 599de54451af..133dcc8a4150 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -140,6 +140,73 @@ config EXT3_FS_SECURITY If you are not using a security module that requires using extended attributes for file security labels, say N. +config EXT4DEV_FS + tristate "Ext4dev/ext4 extended fs support development (EXPERIMENTAL)" + depends on EXPERIMENTAL + select JBD2 + help + Ext4dev is a predecessor filesystem of the next generation + extended fs ext4, based on ext3 filesystem code. It will be + renamed ext4 fs later, once ext4dev is mature and stabilized. + + Unlike the change from ext2 filesystem to ext3 filesystem, + the on-disk format of ext4dev is not the same as ext3 any more: + it is based on extent maps and it supports 48-bit physical block + numbers. These combined on-disk format changes will allow + ext4dev/ext4 to handle more than 16 TB filesystem volumes -- + a hard limit that ext3 cannot overcome without changing the + on-disk format. + + Other than extent maps and 48-bit block numbers, ext4dev also is + likely to have other new features such as persistent preallocation, + high resolution time stamps, and larger file support etc. These + features will be added to ext4dev gradually. + + To compile this file system support as a module, choose M here. The + module will be called ext4dev. Be aware, however, that the filesystem + of your root partition (the one containing the directory /) cannot + be compiled as a module, and so this could be dangerous. + + If unsure, say N. + +config EXT4DEV_FS_XATTR + bool "Ext4dev extended attributes" + depends on EXT4DEV_FS + default y + help + Extended attributes are name:value pairs associated with inodes by + the kernel or by users (see the attr(5) manual page, or visit + <http://acl.bestbits.at/> for details). + + If unsure, say N. + + You need this for POSIX ACL support on ext4dev/ext4. + +config EXT4DEV_FS_POSIX_ACL + bool "Ext4dev POSIX Access Control Lists" + depends on EXT4DEV_FS_XATTR + select FS_POSIX_ACL + help + POSIX Access Control Lists (ACLs) support permissions for users and + groups beyond the owner/group/world scheme. + + To learn more about Access Control Lists, visit the POSIX ACLs for + Linux website <http://acl.bestbits.at/>. + + If you don't know what Access Control Lists are, say N + +config EXT4DEV_FS_SECURITY + bool "Ext4dev Security Labels" + depends on EXT4DEV_FS_XATTR + help + Security labels support alternative access control models + implemented by security modules like SELinux. This option + enables an extended attribute handler for file security + labels in the ext4dev/ext4 filesystem. + + If you are not using a security module that requires using + extended attributes for file security labels, say N. + config JBD tristate help @@ -172,12 +239,44 @@ config JBD_DEBUG generated. To turn debugging off again, do "echo 0 > /proc/sys/fs/jbd-debug". +config JBD2 + tristate + help + This is a generic journaling layer for block devices that support + both 32-bit and 64-bit block numbers. It is currently used by + the ext4dev/ext4 filesystem, but it could also be used to add + journal support to other file systems or block devices such + as RAID or LVM. + + If you are using ext4dev/ext4, you need to say Y here. If you are not + using ext4dev/ext4 then you will probably want to say N. + + To compile this device as a module, choose M here. The module will be + called jbd2. If you are compiling ext4dev/ext4 into the kernel, + you cannot compile this code as a module. + +config JBD2_DEBUG + bool "JBD2 (ext4dev/ext4) debugging support" + depends on JBD2 + help + If you are using the ext4dev/ext4 journaled file system (or + potentially any other filesystem/device using JBD2), this option + allows you to enable debugging output while the system is running, + in order to help track down any problems you are having. + By default, the debugging output will be turned off. + + If you select Y here, then you will be able to turn on debugging + with "echo N > /proc/sys/fs/jbd2-debug", where N is a number between + 1 and 5. The higher the number, the more debugging output is + generated. To turn debugging off again, do + "echo 0 > /proc/sys/fs/jbd2-debug". + config FS_MBCACHE -# Meta block cache for Extended Attributes (ext2/ext3) +# Meta block cache for Extended Attributes (ext2/ext3/ext4) tristate - depends on EXT2_FS_XATTR || EXT3_FS_XATTR - default y if EXT2_FS=y || EXT3_FS=y - default m if EXT2_FS=m || EXT3_FS=m + depends on EXT2_FS_XATTR || EXT3_FS_XATTR || EXT4DEV_FS_XATTR + default y if EXT2_FS=y || EXT3_FS=y || EXT4DEV_FS=y + default m if EXT2_FS=m || EXT3_FS=m || EXT4DEV_FS=m config REISERFS_FS tristate "Reiserfs support" @@ -535,6 +634,10 @@ config FUSE_FS If you want to develop a userspace FS, or if you want to use a filesystem based on FUSE, answer Y or M. +config GENERIC_ACL + bool + select FS_POSIX_ACL + if BLOCK menu "CD-ROM/DVD Filesystems" @@ -1674,6 +1777,7 @@ config RPCSEC_GSS_KRB5 select CRYPTO select CRYPTO_MD5 select CRYPTO_DES + select CRYPTO_CBC help Provides for secure RPC calls by means of a gss-api mechanism based on Kerberos V5. This is required for @@ -1692,6 +1796,7 @@ config RPCSEC_GSS_SPKM3 select CRYPTO_MD5 select CRYPTO_DES select CRYPTO_CAST5 + select CRYPTO_CBC help Provides for secure RPC calls by means of a gss-api mechanism based on the SPKM3 public-key mechanism. @@ -1887,7 +1992,7 @@ config CIFS_EXPERIMENTAL config CIFS_UPCALL bool "Kerberos/SPNEGO advanced session setup (EXPERIMENTAL)" depends on CIFS_EXPERIMENTAL - select CONNECTOR + depends on CONNECTOR help Enables an upcall mechanism for CIFS which will be used to contact userspace helper utilities to provide SPNEGO packaged Kerberos @@ -1981,10 +2086,6 @@ config 9P_FS If unsure, say N. -config GENERIC_ACL - bool - select FS_POSIX_ACL - endmenu if BLOCK diff --git a/fs/Makefile b/fs/Makefile index df614eacee86..9a5ce9323bfd 100644 --- a/fs/Makefile +++ b/fs/Makefile @@ -62,7 +62,9 @@ obj-$(CONFIG_DLM) += dlm/ # Do not add any filesystems before this line obj-$(CONFIG_REISERFS_FS) += reiserfs/ obj-$(CONFIG_EXT3_FS) += ext3/ # Before ext2 so root fs can be ext3 +obj-$(CONFIG_EXT4DEV_FS) += ext4/ # Before ext2 so root fs can be ext4dev obj-$(CONFIG_JBD) += jbd/ +obj-$(CONFIG_JBD2) += jbd2/ obj-$(CONFIG_EXT2_FS) += ext2/ obj-$(CONFIG_CRAMFS) += cramfs/ obj-$(CONFIG_RAMFS) += ramfs/ diff --git a/fs/afs/dir.c b/fs/afs/dir.c index cf8a2cb28505..a6ec75c56fcf 100644 --- a/fs/afs/dir.c +++ b/fs/afs/dir.c @@ -211,8 +211,8 @@ static int afs_dir_open(struct inode *inode, struct file *file) { _enter("{%lu}", inode->i_ino); - BUG_ON(sizeof(union afs_dir_block) != 2048); - BUG_ON(sizeof(union afs_dirent) != 32); + BUILD_BUG_ON(sizeof(union afs_dir_block) != 2048); + BUILD_BUG_ON(sizeof(union afs_dirent) != 32); if (AFS_FS_I(inode)->flags & AFS_VNODE_DELETED) return -ENOENT; @@ -446,8 +446,8 @@ static struct dentry *afs_dir_lookup(struct inode *dir, struct dentry *dentry, _enter("{%lu},%p{%s}", dir->i_ino, dentry, dentry->d_name.name); /* insanity checks first */ - BUG_ON(sizeof(union afs_dir_block) != 2048); - BUG_ON(sizeof(union afs_dirent) != 32); + BUILD_BUG_ON(sizeof(union afs_dir_block) != 2048); + BUILD_BUG_ON(sizeof(union afs_dirent) != 32); if (dentry->d_name.len > 255) { _leave(" = -ENAMETOOLONG"); diff --git a/fs/autofs/autofs_i.h b/fs/autofs/autofs_i.h index c7700d9b3f96..906ba5ce2261 100644 --- a/fs/autofs/autofs_i.h +++ b/fs/autofs/autofs_i.h @@ -149,6 +149,7 @@ extern const struct file_operations autofs_root_operations; /* Initializing function */ int autofs_fill_super(struct super_block *, void *, int); +void autofs_kill_sb(struct super_block *sb); /* Queue management functions */ diff --git a/fs/autofs/dirhash.c b/fs/autofs/dirhash.c index 3fded389d06b..bf8c8af98004 100644 --- a/fs/autofs/dirhash.c +++ b/fs/autofs/dirhash.c @@ -246,5 +246,4 @@ void autofs_hash_nuke(struct autofs_sb_info *sbi) kfree(ent); } } - shrink_dcache_sb(sbi->sb); } diff --git a/fs/autofs/init.c b/fs/autofs/init.c index aca123752406..cea5219b4f37 100644 --- a/fs/autofs/init.c +++ b/fs/autofs/init.c @@ -24,7 +24,7 @@ static struct file_system_type autofs_fs_type = { .owner = THIS_MODULE, .name = "autofs", .get_sb = autofs_get_sb, - .kill_sb = kill_anon_super, + .kill_sb = autofs_kill_sb, }; static int __init init_autofs_fs(void) diff --git a/fs/autofs/inode.c b/fs/autofs/inode.c index 2c9759baad61..54c518c89e4c 100644 --- a/fs/autofs/inode.c +++ b/fs/autofs/inode.c @@ -20,7 +20,7 @@ #include "autofs_i.h" #include <linux/module.h> -static void autofs_put_super(struct super_block *sb) +void autofs_kill_sb(struct super_block *sb) { struct autofs_sb_info *sbi = autofs_sbi(sb); unsigned int n; @@ -37,13 +37,13 @@ static void autofs_put_super(struct super_block *sb) kfree(sb->s_fs_info); DPRINTK(("autofs: shutting down\n")); + kill_anon_super(sb); } static void autofs_read_inode(struct inode *inode); static struct super_operations autofs_sops = { .read_inode = autofs_read_inode, - .put_super = autofs_put_super, .statfs = simple_statfs, }; diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h index 480ab178cba5..b13f32c8aeee 100644 --- a/fs/autofs4/autofs_i.h +++ b/fs/autofs4/autofs_i.h @@ -94,7 +94,6 @@ struct autofs_wait_queue { struct autofs_sb_info { u32 magic; - struct dentry *root; int pipefd; struct file *pipe; pid_t oz_pgrp; @@ -229,4 +228,4 @@ out: } void autofs4_dentry_release(struct dentry *); - +extern void autofs4_kill_sb(struct super_block *); diff --git a/fs/autofs4/init.c b/fs/autofs4/init.c index 5d9193332bef..723a1c5e361b 100644 --- a/fs/autofs4/init.c +++ b/fs/autofs4/init.c @@ -24,7 +24,7 @@ static struct file_system_type autofs_fs_type = { .owner = THIS_MODULE, .name = "autofs", .get_sb = autofs_get_sb, - .kill_sb = kill_anon_super, + .kill_sb = autofs4_kill_sb, }; static int __init init_autofs4_fs(void) diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c index 800ce876caec..51fd8595bf85 100644 --- a/fs/autofs4/inode.c +++ b/fs/autofs4/inode.c @@ -96,7 +96,7 @@ void autofs4_free_ino(struct autofs_info *ino) */ static void autofs4_force_release(struct autofs_sb_info *sbi) { - struct dentry *this_parent = sbi->root; + struct dentry *this_parent = sbi->sb->s_root; struct list_head *next; spin_lock(&dcache_lock); @@ -127,7 +127,7 @@ resume: spin_lock(&dcache_lock); } - if (this_parent != sbi->root) { + if (this_parent != sbi->sb->s_root) { struct dentry *dentry = this_parent; next = this_parent->d_u.d_child.next; @@ -140,15 +140,9 @@ resume: goto resume; } spin_unlock(&dcache_lock); - - dput(sbi->root); - sbi->root = NULL; - shrink_dcache_sb(sbi->sb); - - return; } -static void autofs4_put_super(struct super_block *sb) +void autofs4_kill_sb(struct super_block *sb) { struct autofs_sb_info *sbi = autofs4_sbi(sb); @@ -163,6 +157,7 @@ static void autofs4_put_super(struct super_block *sb) kfree(sbi); DPRINTK("shutting down"); + kill_anon_super(sb); } static int autofs4_show_options(struct seq_file *m, struct vfsmount *mnt) @@ -189,7 +184,6 @@ static int autofs4_show_options(struct seq_file *m, struct vfsmount *mnt) } static struct super_operations autofs4_sops = { - .put_super = autofs4_put_super, .statfs = simple_statfs, .show_options = autofs4_show_options, }; @@ -315,7 +309,6 @@ int autofs4_fill_super(struct super_block *s, void *data, int silent) s->s_fs_info = sbi; sbi->magic = AUTOFS_SBI_MAGIC; - sbi->root = NULL; sbi->pipefd = -1; sbi->catatonic = 0; sbi->exp_timeout = 0; @@ -397,13 +390,6 @@ int autofs4_fill_super(struct super_block *s, void *data, int silent) sbi->pipefd = pipefd; /* - * Take a reference to the root dentry so we get a chance to - * clean up the dentry tree on umount. - * See autofs4_force_release. - */ - sbi->root = dget(root); - - /* * Success! Install the root dentry now to indicate completion. */ s->s_root = root; diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c index ce103e7b0bc3..c0a6c8d445c7 100644 --- a/fs/autofs4/waitq.c +++ b/fs/autofs4/waitq.c @@ -45,7 +45,6 @@ void autofs4_catatonic_mode(struct autofs_sb_info *sbi) fput(sbi->pipe); /* Close the pipe */ sbi->pipe = NULL; } - shrink_dcache_sb(sbi->sb); } static int autofs4_write(struct file *file, const void *addr, int bytes) diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 06435f3665f4..79b05a1a4365 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -1152,7 +1152,7 @@ static int dump_write(struct file *file, const void *addr, int nr) static int dump_seek(struct file *file, loff_t off) { if (file->f_op->llseek && file->f_op->llseek != no_llseek) { - if (file->f_op->llseek(file, off, 1) != off) + if (file->f_op->llseek(file, off, SEEK_CUR) < 0) return 0; } else { char *buf = (char *)get_zeroed_page(GFP_KERNEL); @@ -1220,7 +1220,7 @@ static int notesize(struct memelfnote *en) static int alignfile(struct file *file, loff_t *foffset) { - char buf[4] = { 0, }; + static const char buf[4] = { 0, }; DUMP_WRITE(buf, roundup(*foffset, 4) - *foffset, foffset); return 1; } @@ -1569,7 +1569,8 @@ static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file) DUMP_WRITE(elf, sizeof(*elf)); offset += sizeof(*elf); /* Elf header */ - offset += (segs+1) * sizeof(struct elf_phdr); /* Program headers */ + offset += (segs + 1) * sizeof(struct elf_phdr); /* Program headers */ + foffset = offset; /* Write notes phdr entry */ { @@ -1586,8 +1587,6 @@ static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file) DUMP_WRITE(&phdr, sizeof(phdr)); } - foffset = offset; - dataoff = offset = roundup(offset, ELF_EXEC_PAGESIZE); /* Write program headers for segments dump */ @@ -1612,7 +1611,6 @@ static int elf_core_dump(long signr, struct pt_regs *regs, struct file *file) phdr.p_align = ELF_EXEC_PAGESIZE; DUMP_WRITE(&phdr, sizeof(phdr)); - foffset += sizeof(phdr); } #ifdef ELF_CORE_WRITE_EXTRA_PHDRS @@ -79,7 +79,6 @@ static struct bio_set *fs_bio_set; 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; /* * see comment near bvec_array define! @@ -98,10 +97,12 @@ static inline struct bio_vec *bvec_alloc_bs(gfp_t gfp_mask, int nr, unsigned lon * idx now points to the pool we want to allocate from */ - bp = bvec_slabs + *idx; bvl = mempool_alloc(bs->bvec_pools[*idx], gfp_mask); - if (bvl) + if (bvl) { + struct biovec_slab *bp = bvec_slabs + *idx; + memset(bvl, 0, bp->nr_vecs * sizeof(struct bio_vec)); + } return bvl; } @@ -166,7 +167,7 @@ struct bio *bio_alloc_bioset(gfp_t gfp_mask, int nr_iovecs, struct bio_set *bs) bio_init(bio); if (likely(nr_iovecs)) { - unsigned long idx; + unsigned long idx = 0; /* shut up gcc */ bvl = bvec_alloc_bs(gfp_mask, nr_iovecs, &idx, bs); if (unlikely(!bvl)) { diff --git a/fs/block_dev.c b/fs/block_dev.c index bc8f27cc4483..36c0e7af9d0f 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -642,34 +642,47 @@ static void free_bd_holder(struct bd_holder *bo) } /** + * find_bd_holder - find matching struct bd_holder from the block device + * + * @bdev: struct block device to be searched + * @bo: target struct bd_holder + * + * Returns matching entry with @bo in @bdev->bd_holder_list. + * If found, increment the reference count and return the pointer. + * If not found, returns NULL. + */ +static struct bd_holder *find_bd_holder(struct block_device *bdev, + struct bd_holder *bo) +{ + struct bd_holder *tmp; + + list_for_each_entry(tmp, &bdev->bd_holder_list, list) + if (tmp->sdir == bo->sdir) { + tmp->count++; + return tmp; + } + + return NULL; +} + +/** * add_bd_holder - create sysfs symlinks for bd_claim() relationship * * @bdev: block device to be bd_claimed * @bo: preallocated and initialized by alloc_bd_holder() * - * If there is no matching entry with @bo in @bdev->bd_holder_list, - * add @bo to the list, create symlinks. + * Add @bo to @bdev->bd_holder_list, create symlinks. * - * Returns 0 if symlinks are created or already there. - * Returns -ve if something fails and @bo can be freed. + * Returns 0 if symlinks are created. + * Returns -ve if something fails. */ static int add_bd_holder(struct block_device *bdev, struct bd_holder *bo) { - struct bd_holder *tmp; int ret; if (!bo) return -EINVAL; - list_for_each_entry(tmp, &bdev->bd_holder_list, list) { - if (tmp->sdir == bo->sdir) { - tmp->count++; - /* We've already done what we need to do here. */ - free_bd_holder(bo); - return 0; - } - } - if (!bd_holder_grab_dirs(bdev, bo)) return -EBUSY; @@ -740,7 +753,7 @@ static int bd_claim_by_kobject(struct block_device *bdev, void *holder, struct kobject *kobj) { int res; - struct bd_holder *bo; + struct bd_holder *bo, *found; if (!kobj) return -EINVAL; @@ -751,9 +764,16 @@ static int bd_claim_by_kobject(struct block_device *bdev, void *holder, mutex_lock_nested(&bdev->bd_mutex, BD_MUTEX_PARTITION); res = bd_claim(bdev, holder); - if (res == 0) - res = add_bd_holder(bdev, bo); - if (res) + if (res == 0) { + found = find_bd_holder(bdev, bo); + if (found == NULL) { + res = add_bd_holder(bdev, bo); + if (res) + bd_release(bdev); + } + } + + if (res || found) free_bd_holder(bo); mutex_unlock(&bdev->bd_mutex); @@ -1131,6 +1151,8 @@ static int blkdev_open(struct inode * inode, struct file * filp) filp->f_flags |= O_LARGEFILE; bdev = bd_acquire(inode); + if (bdev == NULL) + return -ENOMEM; res = do_open(bdev, filp, BD_MUTEX_NORMAL); if (res) diff --git a/fs/buffer.c b/fs/buffer.c index eeb8ac1aa856..35527dca1dbc 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -452,6 +452,7 @@ static void end_buffer_async_write(struct buffer_head *bh, int uptodate) bdevname(bh->b_bdev, b)); } set_bit(AS_EIO, &page->mapping->flags); + set_buffer_write_io_error(bh); clear_buffer_uptodate(bh); SetPageError(page); } @@ -571,6 +572,10 @@ EXPORT_SYMBOL(mark_buffer_async_write); static inline void __remove_assoc_queue(struct buffer_head *bh) { list_del_init(&bh->b_assoc_buffers); + WARN_ON(!bh->b_assoc_map); + if (buffer_write_io_error(bh)) + set_bit(AS_EIO, &bh->b_assoc_map->flags); + bh->b_assoc_map = NULL; } int inode_has_buffers(struct inode *inode) @@ -669,6 +674,7 @@ void mark_buffer_dirty_inode(struct buffer_head *bh, struct inode *inode) spin_lock(&buffer_mapping->private_lock); list_move_tail(&bh->b_assoc_buffers, &mapping->private_list); + bh->b_assoc_map = mapping; spin_unlock(&buffer_mapping->private_lock); } } @@ -765,7 +771,7 @@ static int fsync_buffers_list(spinlock_t *lock, struct list_head *list) spin_lock(lock); while (!list_empty(list)) { bh = BH_ENTRY(list->next); - list_del_init(&bh->b_assoc_buffers); + __remove_assoc_queue(bh); if (buffer_dirty(bh) || buffer_locked(bh)) { list_add(&bh->b_assoc_buffers, &tmp); if (buffer_dirty(bh)) { @@ -786,7 +792,7 @@ static int fsync_buffers_list(spinlock_t *lock, struct list_head *list) while (!list_empty(&tmp)) { bh = BH_ENTRY(tmp.prev); - __remove_assoc_queue(bh); + list_del_init(&bh->b_assoc_buffers); get_bh(bh); spin_unlock(lock); wait_on_buffer(bh); @@ -1042,8 +1048,21 @@ grow_buffers(struct block_device *bdev, sector_t block, int size) } while ((size << sizebits) < PAGE_SIZE); index = block >> sizebits; - block = index << sizebits; + /* + * Check for a block which wants to lie outside our maximum possible + * pagecache index. (this comparison is done using sector_t types). + */ + if (unlikely(index != block >> sizebits)) { + char b[BDEVNAME_SIZE]; + + printk(KERN_ERR "%s: requested out-of-range block %llu for " + "device %s\n", + __FUNCTION__, (unsigned long long)block, + bdevname(bdev, b)); + return -EIO; + } + block = index << sizebits; /* Create a page with the proper size buffers.. */ page = grow_dev_page(bdev, block, index, size); if (!page) @@ -1070,12 +1089,16 @@ __getblk_slow(struct block_device *bdev, sector_t block, int size) for (;;) { struct buffer_head * bh; + int ret; bh = __find_get_block(bdev, block, size); if (bh) return bh; - if (!grow_buffers(bdev, block, size)) + ret = grow_buffers(bdev, block, size); + if (ret < 0) + return NULL; + if (ret == 0) free_more_memory(); } } @@ -1150,6 +1173,7 @@ void __bforget(struct buffer_head *bh) spin_lock(&buffer_mapping->private_lock); list_del_init(&bh->b_assoc_buffers); + bh->b_assoc_map = NULL; spin_unlock(&buffer_mapping->private_lock); } __brelse(bh); @@ -1837,6 +1861,7 @@ static int __block_prepare_write(struct inode *inode, struct page *page, clear_buffer_new(bh); kaddr = kmap_atomic(page, KM_USER0); memset(kaddr+block_start, 0, bh->b_size); + flush_dcache_page(page); kunmap_atomic(kaddr, KM_USER0); set_buffer_uptodate(bh); mark_buffer_dirty(bh); @@ -2343,6 +2368,7 @@ failed: */ kaddr = kmap_atomic(page, KM_USER0); memset(kaddr, 0, PAGE_CACHE_SIZE); + flush_dcache_page(page); kunmap_atomic(kaddr, KM_USER0); SetPageUptodate(page); set_page_dirty(page); diff --git a/fs/cifs/CHANGES b/fs/cifs/CHANGES index 1eb9a2ec0a3b..0b3c37ef52e0 100644 --- a/fs/cifs/CHANGES +++ b/fs/cifs/CHANGES @@ -1,6 +1,11 @@ Version 1.46 ------------ Support deep tree mounts. Better support OS/2, Win9x (DOS) time stamps. +Allow null user to be specified on mount ("username="). Do not return +EINVAL on readdir when filldir fails due to overwritten blocksize +(fixes FC problem). Return error in rename 2nd attempt retry (ie report +if rename by handle also fails, after rename by path fails, we were +not reporting whether the retry worked or not). Version 1.45 ------------ diff --git a/fs/cifs/cifsacl.h b/fs/cifs/cifsacl.h index d0776ac2b804..5eff35d6e564 100644 --- a/fs/cifs/cifsacl.h +++ b/fs/cifs/cifsacl.h @@ -31,8 +31,8 @@ struct cifs_sid { } __attribute__((packed)); /* everyone */ -extern const struct cifs_sid sid_everyone; +/* extern const struct cifs_sid sid_everyone;*/ /* group users */ -extern const struct cifs_sid sid_user; +/* extern const struct cifs_sid sid_user;*/ #endif /* _CIFSACL_H */ diff --git a/fs/cifs/cifsencrypt.h b/fs/cifs/cifsencrypt.h index 03e359b32861..152fa2dcfc6c 100644 --- a/fs/cifs/cifsencrypt.h +++ b/fs/cifs/cifsencrypt.h @@ -27,8 +27,6 @@ extern void mdfour(unsigned char *out, unsigned char *in, int n); /* smbdes.c */ extern void E_P16(unsigned char *p14, unsigned char *p16); extern void E_P24(unsigned char *p21, unsigned char *c8, unsigned char *p24); -extern void D_P16(unsigned char *p14, unsigned char *in, unsigned char *out); -extern void E_old_pw_hash(unsigned char *, unsigned char *, unsigned char *); diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index c00c654f2e11..84976cdbe713 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -63,6 +63,7 @@ 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; +static struct super_operations cifs_super_ops; 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"); @@ -198,10 +199,12 @@ cifs_statfs(struct dentry *dentry, struct kstatfs *buf) /* Only need to call the old QFSInfo if failed on newer one */ if(rc) - rc = CIFSSMBQFSInfo(xid, pTcon, buf); + if(pTcon->ses->capabilities & CAP_NT_SMBS) + rc = CIFSSMBQFSInfo(xid, pTcon, buf); /* not supported by OS2 */ - /* Old Windows servers do not support level 103, retry with level - one if old server failed the previous call */ + /* Some old Windows servers also do not support level 103, retry with + older level one if old server failed the previous call or we + bypassed it because we detected that this was an older LANMAN sess */ if(rc) rc = SMBOldQFSInfo(xid, pTcon, buf); /* @@ -435,13 +438,21 @@ static void cifs_umount_begin(struct vfsmount * vfsmnt, int flags) return; } +#ifdef CONFIG_CIFS_STATS2 +static int cifs_show_stats(struct seq_file *s, struct vfsmount *mnt) +{ + /* BB FIXME */ + return 0; +} +#endif + static int cifs_remount(struct super_block *sb, int *flags, char *data) { *flags |= MS_NODIRATIME; return 0; } -struct super_operations cifs_super_ops = { +static struct super_operations cifs_super_ops = { .read_inode = cifs_read_inode, .put_super = cifs_put_super, .statfs = cifs_statfs, @@ -454,6 +465,9 @@ struct super_operations cifs_super_ops = { .show_options = cifs_show_options, .umount_begin = cifs_umount_begin, .remount_fs = cifs_remount, +#ifdef CONFIG_CIFS_STATS2 + .show_stats = cifs_show_stats, +#endif }; static int @@ -495,7 +509,7 @@ static ssize_t cifs_file_aio_write(struct kiocb *iocb, const struct iovec *iov, static loff_t cifs_llseek(struct file *file, loff_t offset, int origin) { /* origin == SEEK_END => we must revalidate the cached file length */ - if (origin == 2) { + if (origin == SEEK_END) { int retval = cifs_revalidate(file->f_dentry); if (retval < 0) return (loff_t)retval; @@ -903,7 +917,7 @@ init_cifs(void) #ifdef CONFIG_PROC_FS cifs_proc_init(); #endif - INIT_LIST_HEAD(&GlobalServerList); /* BB not implemented yet */ +/* INIT_LIST_HEAD(&GlobalServerList);*/ /* BB not implemented yet */ INIT_LIST_HEAD(&GlobalSMBSessionList); INIT_LIST_HEAD(&GlobalTreeConnectionList); INIT_LIST_HEAD(&GlobalOplock_Q); @@ -931,6 +945,7 @@ init_cifs(void) GlobalCurrentXid = 0; GlobalTotalActiveXid = 0; GlobalMaxActiveXid = 0; + memset(Local_System_Name, 0, 15); rwlock_init(&GlobalSMBSeslock); spin_lock_init(&GlobalMid_Lock); diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h index bea875d9a46a..a243f779b363 100644 --- a/fs/cifs/cifsfs.h +++ b/fs/cifs/cifsfs.h @@ -36,7 +36,7 @@ extern const struct address_space_operations cifs_addr_ops; extern const struct address_space_operations cifs_addr_ops_smallbuf; /* Functions related to super block operations */ -extern struct super_operations cifs_super_ops; +/* extern struct super_operations cifs_super_ops;*/ extern void cifs_read_inode(struct inode *); extern void cifs_delete_inode(struct inode *); /* extern void cifs_write_inode(struct inode *); *//* BB not needed yet */ diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index b24006c47df1..74d3ccbb103b 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -153,7 +153,7 @@ struct TCP_Server_Info { char sessid[4]; /* unique token id for this session */ /* (returned on Negotiate */ int capabilities; /* allow selective disabling of caps by smb sess */ - __u16 timeZone; + int timeAdj; /* Adjust for difference in server time zone in sec */ __u16 CurrentMid; /* multiplex id - rotating counter */ char cryptKey[CIFS_CRYPTO_KEY_SIZE]; /* 16th byte of RFC1001 workstation name is always null */ @@ -203,9 +203,14 @@ struct cifsSesInfo { char * domainName; char * password; }; -/* session flags */ +/* no more than one of the following three session flags may be set */ #define CIFS_SES_NT4 1 - +#define CIFS_SES_OS2 2 +#define CIFS_SES_W9X 4 +/* following flag is set for old servers such as OS2 (and Win95?) + which do not negotiate NTLM or POSIX dialects, but instead + negotiate one of the older LANMAN dialects */ +#define CIFS_SES_LANMAN 8 /* * there is one of these for each connection to a resource on a particular * session @@ -512,7 +517,8 @@ require use of the stronger protocol */ * This list helps improve performance and eliminate the messages indicating * that we had a communications error talking to the server in this list. */ -GLOBAL_EXTERN struct servers_not_supported *NotSuppList; /*@z4a */ +/* Feature not supported */ +/* GLOBAL_EXTERN struct servers_not_supported *NotSuppList; */ /* * The following is a hash table of all the users we know about. @@ -568,7 +574,6 @@ GLOBAL_EXTERN unsigned int lookupCacheEnabled; GLOBAL_EXTERN unsigned int extended_security; /* if on, session setup sent with more secure ntlmssp2 challenge/resp */ GLOBAL_EXTERN unsigned int sign_CIFS_PDUs; /* enable smb packet signing */ -GLOBAL_EXTERN unsigned int secFlags; GLOBAL_EXTERN unsigned int linuxExtEnabled;/*enable Linux/Unix CIFS extensions*/ GLOBAL_EXTERN unsigned int CIFSMaxBufSize; /* max size not including hdr */ GLOBAL_EXTERN unsigned int cifs_min_rcv; /* min size of big ntwrk buf pool */ diff --git a/fs/cifs/cifspdu.h b/fs/cifs/cifspdu.h index 81df2bf8e75a..6df9dadba647 100644 --- a/fs/cifs/cifspdu.h +++ b/fs/cifs/cifspdu.h @@ -26,7 +26,8 @@ #ifdef CONFIG_CIFS_WEAK_PW_HASH #define LANMAN_PROT 0 -#define CIFS_PROT 1 +#define LANMAN2_PROT 1 +#define CIFS_PROT 2 #else #define CIFS_PROT 0 #endif @@ -408,6 +409,8 @@ typedef struct negotiate_req { /* Dialect index is 13 for LANMAN */ +#define MIN_TZ_ADJ (15 * 60) /* minimum grid for timezones in seconds */ + typedef struct lanman_neg_rsp { struct smb_hdr hdr; /* wct = 13 */ __le16 DialectIndex; @@ -417,7 +420,10 @@ typedef struct lanman_neg_rsp { __le16 MaxNumberVcs; __le16 RawMode; __le32 SessionKey; - __le32 ServerTime; + struct { + __le16 Time; + __le16 Date; + } __attribute__((packed)) SrvTime; __le16 ServerTimeZone; __le16 EncryptionKeyLength; __le16 Reserved; @@ -674,7 +680,7 @@ typedef union smb_com_tree_disconnect { /* as an altetnative can use flag on typedef struct smb_com_close_req { struct smb_hdr hdr; /* wct = 3 */ __u16 FileID; - __u32 LastWriteTime; /* should be zero */ + __u32 LastWriteTime; /* should be zero or -1 */ __u16 ByteCount; /* 0 */ } __attribute__((packed)) CLOSE_REQ; diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index b35c55c3c8bb..f1f8225102f0 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -50,12 +50,12 @@ extern int SendReceive(const unsigned int /* xid */ , struct cifsSesInfo *, extern int SendReceive2(const unsigned int /* xid */ , struct cifsSesInfo *, struct kvec *, int /* nvec to send */, int * /* type of buf returned */ , const int long_op); -extern int SendReceiveBlockingLock(const unsigned int /* xid */ , struct cifsTconInfo *, +extern int SendReceiveBlockingLock(const unsigned int /* xid */ , + struct cifsTconInfo *, struct smb_hdr * /* input */ , struct smb_hdr * /* out */ , int * /* bytes returned */); -extern int checkSMBhdr(struct smb_hdr *smb, __u16 mid); -extern int checkSMB(struct smb_hdr *smb, __u16 mid, int length); +extern int checkSMB(struct smb_hdr *smb, __u16 mid, unsigned int length); extern int is_valid_oplock_break(struct smb_hdr *smb, struct TCP_Server_Info *); extern int is_size_safe_to_change(struct cifsInodeInfo *); extern struct cifsFileInfo *find_writable_file(struct cifsInodeInfo *); @@ -80,6 +80,9 @@ extern struct oplock_q_entry * AllocOplockQEntry(struct inode *, u16, extern void DeleteOplockQEntry(struct oplock_q_entry *); extern struct timespec cifs_NTtimeToUnix(u64 /* utc nanoseconds since 1601 */ ); extern u64 cifs_UnixTimeToNT(struct timespec); +extern __le64 cnvrtDosCifsTm(__u16 date, __u16 time); +extern struct timespec cnvrtDosUnixTm(__u16 date, __u16 time); + extern int cifs_get_inode_info(struct inode **pinode, const unsigned char *search_path, FILE_ALL_INFO * pfile_info, @@ -116,6 +119,7 @@ extern int CIFSFindClose(const int, struct cifsTconInfo *tcon, extern int CIFSSMBQPathInfo(const int xid, struct cifsTconInfo *tcon, const unsigned char *searchName, FILE_ALL_INFO * findData, + int legacy /* whether to use old info level */, const struct nls_table *nls_codepage, int remap); extern int SMBQueryInformation(const int xid, struct cifsTconInfo *tcon, const unsigned char *searchName, @@ -279,8 +283,6 @@ extern void sesInfoFree(struct cifsSesInfo *); extern struct cifsTconInfo *tconInfoAlloc(void); extern void tconInfoFree(struct cifsTconInfo *); -extern int cifs_reconnect(struct TCP_Server_Info *server); - extern int cifs_sign_smb(struct smb_hdr *, struct TCP_Server_Info *,__u32 *); extern int cifs_sign_smb2(struct kvec *iov, int n_vec, struct TCP_Server_Info *, __u32 *); diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 075d8fb3d376..098790eb2aa1 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -46,6 +46,7 @@ static struct { } protocols[] = { #ifdef CONFIG_CIFS_WEAK_PW_HASH {LANMAN_PROT, "\2LM1.2X002"}, + {LANMAN2_PROT, "\2LANMAN2.1"}, #endif /* weak password hashing for legacy clients */ {CIFS_PROT, "\2NT LM 0.12"}, {POSIX_PROT, "\2POSIX 2"}, @@ -58,6 +59,7 @@ static struct { } protocols[] = { #ifdef CONFIG_CIFS_WEAK_PW_HASH {LANMAN_PROT, "\2LM1.2X002"}, + {LANMAN2_PROT, "\2LANMAN2.1"}, #endif /* weak password hashing for legacy clients */ {CIFS_PROT, "\2NT LM 0.12"}, {BAD_PROT, "\2"} @@ -67,13 +69,13 @@ static struct { /* define the number of elements in the cifs dialect array */ #ifdef CONFIG_CIFS_POSIX #ifdef CONFIG_CIFS_WEAK_PW_HASH -#define CIFS_NUM_PROT 3 +#define CIFS_NUM_PROT 4 #else #define CIFS_NUM_PROT 2 #endif /* CIFS_WEAK_PW_HASH */ #else /* not posix */ #ifdef CONFIG_CIFS_WEAK_PW_HASH -#define CIFS_NUM_PROT 2 +#define CIFS_NUM_PROT 3 #else #define CIFS_NUM_PROT 1 #endif /* CONFIG_CIFS_WEAK_PW_HASH */ @@ -397,6 +399,7 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) struct TCP_Server_Info * server; u16 count; unsigned int secFlags; + u16 dialect; if(ses->server) server = ses->server; @@ -436,9 +439,10 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) if (rc != 0) goto neg_err_exit; - cFYI(1,("Dialect: %d", pSMBr->DialectIndex)); + dialect = le16_to_cpu(pSMBr->DialectIndex); + cFYI(1,("Dialect: %d", dialect)); /* Check wct = 1 error case */ - if((pSMBr->hdr.WordCount < 13) || (pSMBr->DialectIndex == BAD_PROT)) { + if((pSMBr->hdr.WordCount < 13) || (dialect == BAD_PROT)) { /* core returns wct = 1, but we do not ask for core - otherwise small wct just comes when dialect index is -1 indicating we could not negotiate a common dialect */ @@ -446,7 +450,9 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) goto neg_err_exit; #ifdef CONFIG_CIFS_WEAK_PW_HASH } else if((pSMBr->hdr.WordCount == 13) - && (pSMBr->DialectIndex == LANMAN_PROT)) { + && ((dialect == LANMAN_PROT) + || (dialect == LANMAN2_PROT))) { + __s16 tmp; struct lanman_neg_rsp * rsp = (struct lanman_neg_rsp *)pSMBr; if((secFlags & CIFSSEC_MAY_LANMAN) || @@ -472,12 +478,44 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) server->maxRw = 0;/* we do not need to use raw anyway */ server->capabilities = CAP_MPX_MODE; } - server->timeZone = le16_to_cpu(rsp->ServerTimeZone); + tmp = (__s16)le16_to_cpu(rsp->ServerTimeZone); + if (tmp == -1) { + /* OS/2 often does not set timezone therefore + * we must use server time to calc time zone. + * Could deviate slightly from the right zone. + * Smallest defined timezone difference is 15 minutes + * (i.e. Nepal). Rounding up/down is done to match + * this requirement. + */ + int val, seconds, remain, result; + struct timespec ts, utc; + utc = CURRENT_TIME; + ts = cnvrtDosUnixTm(le16_to_cpu(rsp->SrvTime.Date), + le16_to_cpu(rsp->SrvTime.Time)); + cFYI(1,("SrvTime: %d sec since 1970 (utc: %d) diff: %d", + (int)ts.tv_sec, (int)utc.tv_sec, + (int)(utc.tv_sec - ts.tv_sec))); + val = (int)(utc.tv_sec - ts.tv_sec); + seconds = val < 0 ? -val : val; + result = (seconds / MIN_TZ_ADJ) * MIN_TZ_ADJ; + remain = seconds % MIN_TZ_ADJ; + if(remain >= (MIN_TZ_ADJ / 2)) + result += MIN_TZ_ADJ; + if(val < 0) + result = - result; + server->timeAdj = result; + } else { + server->timeAdj = (int)tmp; + server->timeAdj *= 60; /* also in seconds */ + } + cFYI(1,("server->timeAdj: %d seconds", server->timeAdj)); + /* BB get server time for time conversions and add code to use it and timezone since this is not UTC */ - if (rsp->EncryptionKeyLength == cpu_to_le16(CIFS_CRYPTO_KEY_SIZE)) { + if (rsp->EncryptionKeyLength == + cpu_to_le16(CIFS_CRYPTO_KEY_SIZE)) { memcpy(server->cryptKey, rsp->EncryptionKey, CIFS_CRYPTO_KEY_SIZE); } else if (server->secMode & SECMODE_PW_ENCRYPT) { @@ -531,7 +569,8 @@ CIFSSMBNegotiate(unsigned int xid, struct cifsSesInfo *ses) cFYI(0, ("Max buf = %d", ses->server->maxBuf)); GETU32(ses->server->sessid) = le32_to_cpu(pSMBr->SessionKey); server->capabilities = le32_to_cpu(pSMBr->Capabilities); - server->timeZone = le16_to_cpu(pSMBr->ServerTimeZone); + server->timeAdj = (int)(__s16)le16_to_cpu(pSMBr->ServerTimeZone); + server->timeAdj *= 60; if (pSMBr->EncryptionKeyLength == CIFS_CRYPTO_KEY_SIZE) { memcpy(server->cryptKey, pSMBr->u.EncryptionKey, CIFS_CRYPTO_KEY_SIZE); @@ -1617,7 +1656,7 @@ CIFSSMBClose(const int xid, struct cifsTconInfo *tcon, int smb_file_id) pSMBr = (CLOSE_RSP *)pSMB; /* BB removeme BB */ pSMB->FileID = (__u16) smb_file_id; - pSMB->LastWriteTime = 0; + pSMB->LastWriteTime = 0xFFFFFFFF; pSMB->ByteCount = 0; rc = SendReceive(xid, tcon->ses, (struct smb_hdr *) pSMB, (struct smb_hdr *) pSMBr, &bytes_returned, 0); @@ -2773,9 +2812,11 @@ GetExtAttrOut: /* security id for everyone */ -const struct cifs_sid sid_everyone = {1, 1, {0, 0, 0, 0, 0, 0}, {0, 0, 0, 0}}; +const static struct cifs_sid sid_everyone = + {1, 1, {0, 0, 0, 0, 0, 0}, {0, 0, 0, 0}}; /* group users */ -const struct cifs_sid sid_user = {1, 2 , {0, 0, 0, 0, 0, 5}, {32, 545, 0, 0}}; +const static struct cifs_sid sid_user = + {1, 2 , {0, 0, 0, 0, 0, 5}, {32, 545, 0, 0}}; /* Convert CIFS ACL to POSIX form */ static int parse_sec_desc(struct cifs_sid * psec_desc, int acl_len) @@ -2856,7 +2897,6 @@ qsec_out: return rc; } - /* Legacy Query Path Information call for lookup to old servers such as Win9x/WinME */ int SMBQueryInformation(const int xid, struct cifsTconInfo *tcon, @@ -2898,7 +2938,16 @@ QInfRetry: if (rc) { cFYI(1, ("Send error in QueryInfo = %d", rc)); } else if (pFinfo) { /* decode response */ + struct timespec ts; + __u32 time = le32_to_cpu(pSMBr->last_write_time); + /* BB FIXME - add time zone adjustment BB */ memset(pFinfo, 0, sizeof(FILE_ALL_INFO)); + ts.tv_nsec = 0; + ts.tv_sec = time; + /* decode time fields */ + pFinfo->ChangeTime = cpu_to_le64(cifs_UnixTimeToNT(ts)); + pFinfo->LastWriteTime = pFinfo->ChangeTime; + pFinfo->LastAccessTime = 0; pFinfo->AllocationSize = cpu_to_le64(le32_to_cpu(pSMBr->size)); pFinfo->EndOfFile = pFinfo->AllocationSize; @@ -2922,6 +2971,7 @@ int CIFSSMBQPathInfo(const int xid, struct cifsTconInfo *tcon, const unsigned char *searchName, FILE_ALL_INFO * pFindData, + int legacy /* old style infolevel */, const struct nls_table *nls_codepage, int remap) { /* level 263 SMB_QUERY_FILE_ALL_INFO */ @@ -2970,7 +3020,10 @@ QPathInfoRetry: byte_count = params + 1 /* pad */ ; pSMB->TotalParameterCount = cpu_to_le16(params); pSMB->ParameterCount = pSMB->TotalParameterCount; - pSMB->InformationLevel = cpu_to_le16(SMB_QUERY_FILE_ALL_INFO); + if(legacy) + pSMB->InformationLevel = cpu_to_le16(SMB_INFO_STANDARD); + else + pSMB->InformationLevel = cpu_to_le16(SMB_QUERY_FILE_ALL_INFO); pSMB->Reserved4 = 0; pSMB->hdr.smb_buf_length += byte_count; pSMB->ByteCount = cpu_to_le16(byte_count); @@ -2982,13 +3035,24 @@ QPathInfoRetry: } else { /* decode response */ rc = validate_t2((struct smb_t2_rsp *)pSMBr); - if (rc || (pSMBr->ByteCount < 40)) + if (rc) /* BB add auto retry on EOPNOTSUPP? */ + rc = -EIO; + else if (!legacy && (pSMBr->ByteCount < 40)) rc = -EIO; /* bad smb */ + else if(legacy && (pSMBr->ByteCount < 24)) + rc = -EIO; /* 24 or 26 expected but we do not read last field */ else if (pFindData){ + int size; __u16 data_offset = le16_to_cpu(pSMBr->t2.DataOffset); + if(legacy) /* we do not read the last field, EAsize, fortunately + since it varies by subdialect and on Set vs. Get, is + two bytes or 4 bytes depending but we don't care here */ + size = sizeof(FILE_INFO_STANDARD); + else + size = sizeof(FILE_ALL_INFO); memcpy((char *) pFindData, (char *) &pSMBr->hdr.Protocol + - data_offset, sizeof (FILE_ALL_INFO)); + data_offset, size); } else rc = -ENOMEM; } @@ -3613,6 +3677,14 @@ getDFSRetry: strncpy(pSMB->RequestFileName, searchName, name_len); } + if(ses->server) { + if(ses->server->secMode & + (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED)) + pSMB->hdr.Flags2 |= SMBFLG2_SECURITY_SIGNATURE; + } + + pSMB->hdr.Uid = ses->Suid; + params = 2 /* level */ + name_len /*includes null */ ; pSMB->TotalDataCount = 0; pSMB->DataCount = 0; diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index c78762051da4..71f77914ce93 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -109,7 +109,7 @@ static int ipv6_connect(struct sockaddr_in6 *psin_server, * wake up waiters on reconnection? - (not needed currently) */ -int +static int cifs_reconnect(struct TCP_Server_Info *server) { int rc = 0; @@ -771,13 +771,18 @@ cifs_parse_mount_options(char *options, const char *devname,struct smb_vol *vol) separator[0] = ','; separator[1] = 0; - memset(vol->source_rfc1001_name,0x20,15); - for(i=0;i < strnlen(utsname()->nodename,15);i++) { - /* does not have to be a perfect mapping since the field is - informational, only used for servers that do not support - port 445 and it can be overridden at mount time */ - vol->source_rfc1001_name[i] = - toupper(utsname()->nodename[i]); + if (Local_System_Name[0] != 0) + memcpy(vol->source_rfc1001_name, Local_System_Name,15); + else { + char *nodename = utsname()->nodename; + int n = strnlen(nodename,15); + memset(vol->source_rfc1001_name,0x20,15); + for(i=0 ; i < n ; i++) { + /* does not have to be perfect mapping since field is + informational, only used for servers that do not support + port 445 and it can be overridden at mount time */ + vol->source_rfc1001_name[i] = toupper(nodename[i]); + } } vol->source_rfc1001_name[15] = 0; /* null target name indicates to use *SMBSERVR default called name @@ -817,10 +822,13 @@ cifs_parse_mount_options(char *options, const char *devname,struct smb_vol *vol) } else if (strnicmp(data, "nouser_xattr",12) == 0) { vol->no_xattr = 1; } else if (strnicmp(data, "user", 4) == 0) { - if (!value || !*value) { + if (!value) { printk(KERN_WARNING "CIFS: invalid or missing username\n"); return 1; /* needs_arg; */ + } else if(!*value) { + /* null user, ie anonymous, authentication */ + vol->nullauth = 1; } if (strnlen(value, 200) < 200) { vol->username = value; @@ -1637,6 +1645,8 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, /* BB fixme parse for domain name here */ cFYI(1, ("Username: %s ", volume_info.username)); + } else if (volume_info.nullauth) { + cFYI(1,("null user")); } else { cifserror("No username specified"); /* In userspace mount helper we can get user name from alternate @@ -3215,7 +3225,9 @@ CIFSTCon(unsigned int xid, struct cifsSesInfo *ses, } /* else do not bother copying these informational fields */ } - if(smb_buffer_response->WordCount == 3) + if((smb_buffer_response->WordCount == 3) || + (smb_buffer_response->WordCount == 7)) + /* field is in same location */ tcon->Flags = le16_to_cpu(pSMBr->OptionalSupport); else tcon->Flags = 0; @@ -3312,19 +3324,21 @@ int cifs_setup_session(unsigned int xid, struct cifsSesInfo *pSesInfo, first_time = 1; } if (!rc) { + pSesInfo->flags = 0; pSesInfo->capabilities = pSesInfo->server->capabilities; if(linuxExtEnabled == 0) pSesInfo->capabilities &= (~CAP_UNIX); /* pSesInfo->sequence_number = 0;*/ - cFYI(1,("Security Mode: 0x%x Capabilities: 0x%x Time Zone: %d", + cFYI(1,("Security Mode: 0x%x Capabilities: 0x%x TimeAdjust: %d", pSesInfo->server->secMode, pSesInfo->server->capabilities, - pSesInfo->server->timeZone)); + pSesInfo->server->timeAdj)); if(experimEnabled < 2) rc = CIFS_SessSetup(xid, pSesInfo, first_time, nls_info); else if (extended_security - && (pSesInfo->capabilities & CAP_EXTENDED_SECURITY) + && (pSesInfo->capabilities + & CAP_EXTENDED_SECURITY) && (pSesInfo->server->secType == NTLMSSP)) { rc = -EOPNOTSUPP; } else if (extended_security @@ -3338,7 +3352,7 @@ int cifs_setup_session(unsigned int xid, struct cifsSesInfo *pSesInfo, if (!rc) { if(ntlmv2_flag) { char * v2_response; - cFYI(1,("Can use more secure NTLM version 2 password hash")); + cFYI(1,("more secure NTLM ver2 hash")); if(CalcNTLMv2_partial_mac_key(pSesInfo, nls_info)) { rc = -ENOMEM; diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 976a691c5a68..7e056b9b49e8 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -1806,13 +1806,6 @@ static int cifs_readpages(struct file *file, struct address_space *mapping, } if ((rc < 0) || (smb_read_data == NULL)) { cFYI(1, ("Read error in readpages: %d", rc)); - /* clean up remaing pages off list */ - while (!list_empty(page_list) && (i < num_pages)) { - page = list_entry(page_list->prev, struct page, - lru); - list_del(&page->lru); - page_cache_release(page); - } break; } else if (bytes_read > 0) { pSMBr = (struct smb_com_read_rsp *)smb_read_data; @@ -1831,13 +1824,7 @@ static int cifs_readpages(struct file *file, struct address_space *mapping, this case is ok - if we are at server EOF we will hit it on next read */ - /* while (!list_empty(page_list) && (i < num_pages)) { - page = list_entry(page_list->prev, - struct page, list); - list_del(&page->list); - page_cache_release(page); - } - break; */ + /* break; */ } } else { cFYI(1, ("No bytes read (%d) at offset %lld . " @@ -1845,14 +1832,6 @@ static int cifs_readpages(struct file *file, struct address_space *mapping, bytes_read, offset)); /* BB turn off caching and do new lookup on file size at server? */ - while (!list_empty(page_list) && (i < num_pages)) { - page = list_entry(page_list->prev, struct page, - lru); - list_del(&page->lru); - - /* BB removeme - replace with zero of page? */ - page_cache_release(page); - } break; } if (smb_read_data) { diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index 6b90ef98e4cf..dffe295825f4 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -337,6 +337,7 @@ 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, + 0 /* not legacy */, 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 @@ -384,8 +385,10 @@ int cifs_get_inode_info(struct inode **pinode, /* get new inode */ if (*pinode == NULL) { *pinode = new_inode(sb); - if (*pinode == NULL) + if (*pinode == NULL) { + kfree(buf); return -ENOMEM; + } /* Is an i_ino of zero legal? Can we use that to check if the server supports returning inode numbers? Are there other sanity checks we can use to ensure that @@ -431,8 +434,11 @@ int cifs_get_inode_info(struct inode **pinode, (pTcon->ses->server->maxBuf - MAX_CIFS_HDR_SIZE) & 0xFFFFFE00;*/ /* Linux can not store file creation time so ignore it */ - inode->i_atime = - cifs_NTtimeToUnix(le64_to_cpu(pfindData->LastAccessTime)); + if(pfindData->LastAccessTime) + inode->i_atime = cifs_NTtimeToUnix + (le64_to_cpu(pfindData->LastAccessTime)); + else /* do not need to use current_fs_time - time not stored */ + inode->i_atime = CURRENT_TIME; inode->i_mtime = cifs_NTtimeToUnix(le64_to_cpu(pfindData->LastWriteTime)); inode->i_ctime = @@ -879,10 +885,14 @@ int cifs_rename(struct inode *source_inode, struct dentry *source_direntry, kmalloc(2 * sizeof(FILE_UNIX_BASIC_INFO), GFP_KERNEL); if (info_buf_source != NULL) { info_buf_target = info_buf_source + 1; - rc = CIFSSMBUnixQPathInfo(xid, pTcon, fromName, - info_buf_source, cifs_sb_source->local_nls, - cifs_sb_source->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); + if (pTcon->ses->capabilities & CAP_UNIX) + rc = CIFSSMBUnixQPathInfo(xid, pTcon, fromName, + info_buf_source, + cifs_sb_source->local_nls, + cifs_sb_source->mnt_cifs_flags & + CIFS_MOUNT_MAP_SPECIAL_CHR); + /* else rc is still EEXIST so will fall through to + unlink the target and retry rename */ if (rc == 0) { rc = CIFSSMBUnixQPathInfo(xid, pTcon, toName, info_buf_target, @@ -931,7 +941,7 @@ int cifs_rename(struct inode *source_inode, struct dentry *source_direntry, cifs_sb_source->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); if (rc==0) { - CIFSSMBRenameOpenFile(xid, pTcon, netfid, toName, + rc = CIFSSMBRenameOpenFile(xid, pTcon, netfid, toName, cifs_sb_source->local_nls, cifs_sb_source->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); diff --git a/fs/cifs/link.c b/fs/cifs/link.c index a57f5d6e6213..0bee8b7e521a 100644 --- a/fs/cifs/link.c +++ b/fs/cifs/link.c @@ -254,7 +254,11 @@ cifs_readlink(struct dentry *direntry, char __user *pBuffer, int buflen) tmpbuffer, len - 1, cifs_sb->local_nls); - else { + else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UNX_EMUL) { + cERROR(1,("SFU style symlinks not implemented yet")); + /* add open and read as in fs/cifs/inode.c */ + + } else { rc = CIFSSMBOpen(xid, pTcon, full_path, FILE_OPEN, GENERIC_READ, OPEN_REPARSE_POINT,&fid, &oplock, NULL, cifs_sb->local_nls, diff --git a/fs/cifs/md5.c b/fs/cifs/md5.c index 7aa23490541f..ccebf9b7eb86 100644 --- a/fs/cifs/md5.c +++ b/fs/cifs/md5.c @@ -252,10 +252,11 @@ MD5Transform(__u32 buf[4], __u32 const in[16]) buf[3] += d; } +#if 0 /* currently unused */ /*********************************************************************** the rfc 2104 version of hmac_md5 initialisation. ***********************************************************************/ -void +static void hmac_md5_init_rfc2104(unsigned char *key, int key_len, struct HMACMD5Context *ctx) { @@ -289,6 +290,7 @@ hmac_md5_init_rfc2104(unsigned char *key, int key_len, MD5Init(&ctx->ctx); MD5Update(&ctx->ctx, ctx->k_ipad, 64); } +#endif /*********************************************************************** the microsoft version of hmac_md5 initialisation. @@ -350,7 +352,8 @@ hmac_md5_final(unsigned char *digest, struct HMACMD5Context *ctx) single function to calculate an HMAC MD5 digest from data. use the microsoft hmacmd5 init method because the key is 16 bytes. ************************************************************/ -void +#if 0 /* currently unused */ +static void hmac_md5(unsigned char key[16], unsigned char *data, int data_len, unsigned char *digest) { @@ -361,3 +364,4 @@ hmac_md5(unsigned char key[16], unsigned char *data, int data_len, } hmac_md5_final(digest, &ctx); } +#endif diff --git a/fs/cifs/md5.h b/fs/cifs/md5.h index 00e1c5394fe1..f7d4f4197bac 100644 --- a/fs/cifs/md5.h +++ b/fs/cifs/md5.h @@ -27,12 +27,12 @@ void MD5Final(unsigned char digest[16], struct MD5Context *context); /* The following definitions come from lib/hmacmd5.c */ -void hmac_md5_init_rfc2104(unsigned char *key, int key_len, - struct HMACMD5Context *ctx); +/* void hmac_md5_init_rfc2104(unsigned char *key, int key_len, + struct HMACMD5Context *ctx);*/ void hmac_md5_init_limK_to_64(const unsigned char *key, int key_len, struct HMACMD5Context *ctx); void hmac_md5_update(const unsigned char *text, int text_len, struct HMACMD5Context *ctx); void hmac_md5_final(unsigned char *digest, struct HMACMD5Context *ctx); -void hmac_md5(unsigned char key[16], unsigned char *data, int data_len, - unsigned char *digest); +/* void hmac_md5(unsigned char key[16], unsigned char *data, int data_len, + unsigned char *digest);*/ diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index 22c937e5884f..bbc9cd34b6ea 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -389,7 +389,7 @@ header_assemble(struct smb_hdr *buffer, char smb_command /* command */ , return; } -int +static int checkSMBhdr(struct smb_hdr *smb, __u16 mid) { /* Make sure that this really is an SMB, that it is a response, @@ -418,26 +418,42 @@ checkSMBhdr(struct smb_hdr *smb, __u16 mid) } int -checkSMB(struct smb_hdr *smb, __u16 mid, int length) +checkSMB(struct smb_hdr *smb, __u16 mid, unsigned int length) { __u32 len = smb->smb_buf_length; __u32 clc_len; /* calculated length */ cFYI(0, ("checkSMB Length: 0x%x, smb_buf_length: 0x%x", length, len)); - if (((unsigned int)length < 2 + sizeof (struct smb_hdr)) || - (len > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE - 4)) { - if ((unsigned int)length < 2 + sizeof (struct smb_hdr)) { - if (((unsigned int)length >= - sizeof (struct smb_hdr) - 1) + + if (length < 2 + sizeof (struct smb_hdr)) { + if ((length >= sizeof (struct smb_hdr) - 1) && (smb->Status.CifsError != 0)) { - smb->WordCount = 0; - /* some error cases do not return wct and bcc */ + smb->WordCount = 0; + /* some error cases do not return wct and bcc */ + return 0; + } else if ((length == sizeof(struct smb_hdr) + 1) && + (smb->WordCount == 0)) { + char * tmp = (char *)smb; + /* Need to work around a bug in two servers here */ + /* First, check if the part of bcc they sent was zero */ + if (tmp[sizeof(struct smb_hdr)] == 0) { + /* some servers return only half of bcc + * on simple responses (wct, bcc both zero) + * in particular have seen this on + * ulogoffX and FindClose. This leaves + * one byte of bcc potentially unitialized + */ + /* zero rest of bcc */ + tmp[sizeof(struct smb_hdr)+1] = 0; return 0; - } else { - cERROR(1, ("Length less than smb header size")); } + cERROR(1,("rcvd invalid byte count (bcc)")); + } else { + cERROR(1, ("Length less than smb header size")); } - if (len > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE - 4) - cERROR(1, ("smb length greater than MaxBufSize, mid=%d", + return 1; + } + if (len > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE - 4) { + cERROR(1, ("smb length greater than MaxBufSize, mid=%d", smb->Mid)); return 1; } @@ -446,7 +462,7 @@ checkSMB(struct smb_hdr *smb, __u16 mid, int length) return 1; clc_len = smbCalcSize_LE(smb); - if(4 + len != (unsigned int)length) { + if(4 + len != length) { cERROR(1, ("Length read does not match RFC1001 length %d",len)); return 1; } diff --git a/fs/cifs/netmisc.c b/fs/cifs/netmisc.c index ce87550e918f..992e80edc720 100644 --- a/fs/cifs/netmisc.c +++ b/fs/cifs/netmisc.c @@ -909,3 +909,61 @@ cifs_UnixTimeToNT(struct timespec t) /* Convert to 100ns intervals and then add the NTFS time offset. */ return (u64) t.tv_sec * 10000000 + t.tv_nsec/100 + NTFS_TIME_OFFSET; } + +static int total_days_of_prev_months[] = +{0, 31, 59, 90, 120, 151, 181, 212, 243, 273, 304, 334}; + + +__le64 cnvrtDosCifsTm(__u16 date, __u16 time) +{ + return cpu_to_le64(cifs_UnixTimeToNT(cnvrtDosUnixTm(date, time))); +} + +struct timespec cnvrtDosUnixTm(__u16 date, __u16 time) +{ + struct timespec ts; + int sec, min, days, month, year; + SMB_TIME * st = (SMB_TIME *)&time; + SMB_DATE * sd = (SMB_DATE *)&date; + + cFYI(1,("date %d time %d",date, time)); + + sec = 2 * st->TwoSeconds; + min = st->Minutes; + if((sec > 59) || (min > 59)) + cERROR(1,("illegal time min %d sec %d", min, sec)); + sec += (min * 60); + sec += 60 * 60 * st->Hours; + if(st->Hours > 24) + cERROR(1,("illegal hours %d",st->Hours)); + days = sd->Day; + month = sd->Month; + if((days > 31) || (month > 12)) + cERROR(1,("illegal date, month %d day: %d", month, days)); + month -= 1; + days += total_days_of_prev_months[month]; + days += 3652; /* account for difference in days between 1980 and 1970 */ + year = sd->Year; + days += year * 365; + days += (year/4); /* leap year */ + /* generalized leap year calculation is more complex, ie no leap year + for years/100 except for years/400, but since the maximum number for DOS + year is 2**7, the last year is 1980+127, which means we need only + consider 2 special case years, ie the years 2000 and 2100, and only + adjust for the lack of leap year for the year 2100, as 2000 was a + leap year (divisable by 400) */ + if(year >= 120) /* the year 2100 */ + days = days - 1; /* do not count leap year for the year 2100 */ + + /* adjust for leap year where we are still before leap day */ + if(year != 120) + days -= ((year & 0x03) == 0) && (month < 2 ? 1 : 0); + sec += 24 * 60 * 60 * days; + + ts.tv_sec = sec; + + /* cFYI(1,("sec after cnvrt dos to unix time %d",sec)); */ + + ts.tv_nsec = 0; + return ts; +} diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c index b27b34537bf2..ed18c3965f7b 100644 --- a/fs/cifs/readdir.c +++ b/fs/cifs/readdir.c @@ -106,6 +106,17 @@ static int construct_dentry(struct qstr *qstring, struct file *file, return rc; } +static void AdjustForTZ(struct cifsTconInfo * tcon, struct inode * inode) +{ + if((tcon) && (tcon->ses) && (tcon->ses->server)) { + inode->i_ctime.tv_sec += tcon->ses->server->timeAdj; + inode->i_mtime.tv_sec += tcon->ses->server->timeAdj; + inode->i_atime.tv_sec += tcon->ses->server->timeAdj; + } + return; +} + + static void fill_in_inode(struct inode *tmp_inode, int new_buf_type, char * buf, int *pobject_type, int isNewInode) { @@ -135,16 +146,23 @@ static void fill_in_inode(struct inode *tmp_inode, int new_buf_type, tmp_inode->i_ctime = cifs_NTtimeToUnix(le64_to_cpu(pfindData->ChangeTime)); } else { /* legacy, OS2 and DOS style */ +/* struct timespec ts;*/ FIND_FILE_STANDARD_INFO * pfindData = (FIND_FILE_STANDARD_INFO *)buf; + tmp_inode->i_mtime = cnvrtDosUnixTm( + le16_to_cpu(pfindData->LastWriteDate), + le16_to_cpu(pfindData->LastWriteTime)); + tmp_inode->i_atime = cnvrtDosUnixTm( + le16_to_cpu(pfindData->LastAccessDate), + le16_to_cpu(pfindData->LastAccessTime)); + tmp_inode->i_ctime = cnvrtDosUnixTm( + le16_to_cpu(pfindData->LastWriteDate), + le16_to_cpu(pfindData->LastWriteTime)); + AdjustForTZ(cifs_sb->tcon, tmp_inode); attr = le16_to_cpu(pfindData->Attributes); allocation_size = le32_to_cpu(pfindData->AllocationSize); end_of_file = le32_to_cpu(pfindData->DataSize); - tmp_inode->i_atime = CURRENT_TIME; - /* tmp_inode->i_mtime = BB FIXME - add dos time handling - tmp_inode->i_ctime = 0; BB FIXME */ - } /* Linux can not store file creation time unfortunately so ignore it */ @@ -878,6 +896,10 @@ static int cifs_filldir(char *pfindEntry, struct file *file, tmp_inode->i_ino,obj_type); if(rc) { cFYI(1,("filldir rc = %d",rc)); + /* we can not return filldir errors to the caller + since they are "normal" when the stat blocksize + is too small - we return remapped error instead */ + rc = -EOVERFLOW; } dput(tmp_dentry); @@ -938,6 +960,7 @@ static int cifs_save_resume_key(const char *current_entry, filename = &pFindData->FileName[0]; /* one byte length, no name conversion */ len = (unsigned int)pFindData->FileNameLength; + cifsFile->srch_inf.resume_key = pFindData->ResumeKey; } else { cFYI(1,("Unknown findfirst level %d",level)); return -EINVAL; @@ -1055,6 +1078,11 @@ int cifs_readdir(struct file *file, void *direntry, filldir_t filldir) we want to check for that here? */ rc = cifs_filldir(current_entry, file, filldir, direntry, tmp_buf, max_len); + if(rc == -EOVERFLOW) { + rc = 0; + break; + } + file->f_pos++; if(file->f_pos == cifsFile->srch_inf.index_of_last_entry) { diff --git a/fs/cifs/sess.c b/fs/cifs/sess.c index 22b4c35dcfe3..a8a083543ba0 100644 --- a/fs/cifs/sess.c +++ b/fs/cifs/sess.c @@ -268,6 +268,10 @@ static int decode_ascii_ssetup(char ** pbcc_area, int bleft, struct cifsSesInfo ses->serverOS = kzalloc(len + 1, GFP_KERNEL); if(ses->serverOS) strncpy(ses->serverOS, bcc_ptr, len); + if(strncmp(ses->serverOS, "OS/2",4) == 0) { + cFYI(1,("OS/2 server")); + ses->flags |= CIFS_SES_OS2; + } bcc_ptr += len + 1; bleft -= len + 1; @@ -290,16 +294,11 @@ static int decode_ascii_ssetup(char ** pbcc_area, int bleft, struct cifsSesInfo if(len > bleft) return rc; - if(ses->serverDomain) - kfree(ses->serverDomain); - - ses->serverDomain = kzalloc(len + 1, GFP_KERNEL); - if(ses->serverOS) - strncpy(ses->serverOS, bcc_ptr, len); - - bcc_ptr += len + 1; - bleft -= len + 1; - + /* No domain field in LANMAN case. Domain is + returned by old servers in the SMB negprot response */ + /* BB For newer servers which do not support Unicode, + but thus do return domain here we could add parsing + for it later, but it is not very important */ cFYI(1,("ascii: bytes left %d",bleft)); return rc; @@ -366,6 +365,8 @@ CIFS_SessSetup(unsigned int xid, struct cifsSesInfo *ses, int first_time, str_area = kmalloc(2000, GFP_KERNEL); bcc_ptr = str_area; + ses->flags &= ~CIFS_SES_LANMAN; + if(type == LANMAN) { #ifdef CONFIG_CIFS_WEAK_PW_HASH char lnm_session_key[CIFS_SESS_KEY_SIZE]; @@ -377,7 +378,7 @@ CIFS_SessSetup(unsigned int xid, struct cifsSesInfo *ses, int first_time, /* and copy into bcc */ calc_lanman_hash(ses, lnm_session_key); - + ses->flags |= CIFS_SES_LANMAN; /* #ifdef CONFIG_CIFS_DEBUG2 cifs_dump_mem("cryptkey: ",ses->server->cryptKey, CIFS_SESS_KEY_SIZE); diff --git a/fs/cifs/smbdes.c b/fs/cifs/smbdes.c index efaa044523a7..7a1b2b961ec8 100644 --- a/fs/cifs/smbdes.c +++ b/fs/cifs/smbdes.c @@ -364,20 +364,20 @@ E_P24(unsigned char *p21, unsigned char *c8, unsigned char *p24) smbhash(p24 + 16, c8, p21 + 14, 1); } -void +#if 0 /* currently unsued */ +static void D_P16(unsigned char *p14, unsigned char *in, unsigned char *out) { smbhash(out, in, p14, 0); smbhash(out + 8, in + 8, p14 + 7, 0); } -void +static void E_old_pw_hash(unsigned char *p14, unsigned char *in, unsigned char *out) { smbhash(out, in, p14, 1); smbhash(out + 8, in + 8, p14 + 7, 1); } -#if 0 /* these routines are currently unneeded, but may be needed later */ void diff --git a/fs/cifs/smbencrypt.c b/fs/cifs/smbencrypt.c index f518c5e45035..4b25ba92180d 100644 --- a/fs/cifs/smbencrypt.c +++ b/fs/cifs/smbencrypt.c @@ -51,11 +51,8 @@ void SMBencrypt(unsigned char *passwd, unsigned char *c8, unsigned char *p24); void E_md4hash(const unsigned char *passwd, unsigned char *p16); -void nt_lm_owf_gen(char *pwd, unsigned char nt_p16[16], unsigned char p16[16]); static void SMBOWFencrypt(unsigned char passwd[16], unsigned char *c8, unsigned char p24[24]); -void NTLMSSPOWFencrypt(unsigned char passwd[8], - unsigned char *ntlmchalresp, unsigned char p24[24]); void SMBNTencrypt(unsigned char *passwd, unsigned char *c8, unsigned char *p24); /* @@ -144,8 +141,9 @@ E_md4hash(const unsigned char *passwd, unsigned char *p16) memset(wpwd,0,129 * 2); } +#if 0 /* currently unused */ /* Does both the NT and LM owfs of a user's password */ -void +static void nt_lm_owf_gen(char *pwd, unsigned char nt_p16[16], unsigned char p16[16]) { char passwd[514]; @@ -171,6 +169,7 @@ nt_lm_owf_gen(char *pwd, unsigned char nt_p16[16], unsigned char p16[16]) /* clear out local copy of user's password (just being paranoid). */ memset(passwd, '\0', sizeof (passwd)); } +#endif /* Does the NTLMv2 owfs of a user's password */ #if 0 /* function not needed yet - but will be soon */ @@ -223,7 +222,8 @@ SMBOWFencrypt(unsigned char passwd[16], unsigned char *c8, } /* Does the des encryption from the FIRST 8 BYTES of the NT or LM MD4 hash. */ -void +#if 0 /* currently unused */ +static void NTLMSSPOWFencrypt(unsigned char passwd[8], unsigned char *ntlmchalresp, unsigned char p24[24]) { @@ -235,6 +235,7 @@ NTLMSSPOWFencrypt(unsigned char passwd[8], E_P24(p21, ntlmchalresp, p24); } +#endif /* Does the NT MD4 hash then des encryption. */ diff --git a/fs/compat.c b/fs/compat.c index 50624d4a70c6..8d0a0018a7d2 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -1835,9 +1835,12 @@ asmlinkage long compat_sys_pselect7(int n, compat_ulong_t __user *inp, } while (!ret && !timeout && tsp && (ts.tv_sec || ts.tv_nsec)); - if (ret == 0 && tsp && !(current->personality & STICKY_TIMEOUTS)) { + if (tsp) { struct compat_timespec rts; + if (current->personality & STICKY_TIMEOUTS) + goto sticky; + rts.tv_sec = timeout / HZ; rts.tv_nsec = (timeout % HZ) * (NSEC_PER_SEC/HZ); if (rts.tv_nsec >= NSEC_PER_SEC) { @@ -1846,8 +1849,19 @@ asmlinkage long compat_sys_pselect7(int n, compat_ulong_t __user *inp, } if (compat_timespec_compare(&rts, &ts) >= 0) rts = ts; - if (copy_to_user(tsp, &rts, sizeof(rts))) - ret = -EFAULT; + if (copy_to_user(tsp, &rts, sizeof(rts))) { +sticky: + /* + * If an application puts its timeval in read-only + * memory, we don't want the Linux-specific update to + * the timeval to cause a fault after the select has + * completed successfully. However, because we're not + * updating the timeval, we can't restart the system + * call. + */ + if (ret == -ERESTARTNOHAND) + ret = -EINTR; + } } if (ret == -ERESTARTNOHAND) { diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c index 27ca1aa30562..a91f2628c981 100644 --- a/fs/compat_ioctl.c +++ b/fs/compat_ioctl.c @@ -2438,13 +2438,17 @@ HANDLE_IOCTL(0x1260, broken_blkgetsize) HANDLE_IOCTL(BLKFRAGET, w_long) HANDLE_IOCTL(BLKSECTGET, w_long) HANDLE_IOCTL(BLKPG, blkpg_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) -HANDLE_IOCTL(HDIO_GET_32BIT, hdio_ioctl_trans) HANDLE_IOCTL(HDIO_GET_MULTCOUNT, hdio_ioctl_trans) +HANDLE_IOCTL(HDIO_GET_KEEPSETTINGS, hdio_ioctl_trans) +HANDLE_IOCTL(HDIO_GET_32BIT, hdio_ioctl_trans) HANDLE_IOCTL(HDIO_GET_NOWERR, hdio_ioctl_trans) +HANDLE_IOCTL(HDIO_GET_DMA, hdio_ioctl_trans) HANDLE_IOCTL(HDIO_GET_NICE, hdio_ioctl_trans) +HANDLE_IOCTL(HDIO_GET_WCACHE, hdio_ioctl_trans) +HANDLE_IOCTL(HDIO_GET_ACOUSTIC, hdio_ioctl_trans) +HANDLE_IOCTL(HDIO_GET_ADDRESS, hdio_ioctl_trans) +HANDLE_IOCTL(HDIO_GET_BUSSTATE, hdio_ioctl_trans) HANDLE_IOCTL(FDSETPRM32, fd_ioctl_trans) HANDLE_IOCTL(FDDEFPRM32, fd_ioctl_trans) HANDLE_IOCTL(FDGETPRM32, fd_ioctl_trans) diff --git a/fs/configfs/file.c b/fs/configfs/file.c index e6d5754a715e..cf33fac68c84 100644 --- a/fs/configfs/file.c +++ b/fs/configfs/file.c @@ -275,13 +275,14 @@ static int check_perm(struct inode * inode, struct file * file) * it in file->private_data for easy access. */ buffer = kzalloc(sizeof(struct configfs_buffer),GFP_KERNEL); - if (buffer) { - init_MUTEX(&buffer->sem); - buffer->needs_read_fill = 1; - buffer->ops = ops; - file->private_data = buffer; - } else + if (!buffer) { error = -ENOMEM; + goto Enomem; + } + init_MUTEX(&buffer->sem); + buffer->needs_read_fill = 1; + buffer->ops = ops; + file->private_data = buffer; goto Done; Einval: @@ -289,6 +290,7 @@ static int check_perm(struct inode * inode, struct file * file) goto Done; Eaccess: error = -EACCES; + Enomem: module_put(attr->ca_owner); Done: if (error && item) diff --git a/fs/dcache.c b/fs/dcache.c index 2355bddad8de..fd4a428998ef 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -478,11 +478,12 @@ static void prune_dcache(int count, struct super_block *sb) up_read(s_umount); } spin_unlock(&dentry->d_lock); - /* Cannot remove the first dentry, and it isn't appropriate - * to move it to the head of the list, so give up, and try - * later + /* + * Insert dentry at the head of the list as inserting at the + * tail leads to a cycle. */ - break; + list_add(&dentry->d_lru, &dentry_unused); + dentry_stat.nr_unused++; } spin_unlock(&dcache_lock); } @@ -549,6 +550,142 @@ repeat: } /* + * destroy a single subtree of dentries for unmount + * - see the comments on shrink_dcache_for_umount() for a description of the + * locking + */ +static void shrink_dcache_for_umount_subtree(struct dentry *dentry) +{ + struct dentry *parent; + unsigned detached = 0; + + BUG_ON(!IS_ROOT(dentry)); + + /* detach this root from the system */ + spin_lock(&dcache_lock); + if (!list_empty(&dentry->d_lru)) { + dentry_stat.nr_unused--; + list_del_init(&dentry->d_lru); + } + __d_drop(dentry); + spin_unlock(&dcache_lock); + + for (;;) { + /* descend to the first leaf in the current subtree */ + while (!list_empty(&dentry->d_subdirs)) { + struct dentry *loop; + + /* this is a branch with children - detach all of them + * from the system in one go */ + spin_lock(&dcache_lock); + list_for_each_entry(loop, &dentry->d_subdirs, + d_u.d_child) { + if (!list_empty(&loop->d_lru)) { + dentry_stat.nr_unused--; + list_del_init(&loop->d_lru); + } + + __d_drop(loop); + cond_resched_lock(&dcache_lock); + } + spin_unlock(&dcache_lock); + + /* move to the first child */ + dentry = list_entry(dentry->d_subdirs.next, + struct dentry, d_u.d_child); + } + + /* consume the dentries from this leaf up through its parents + * until we find one with children or run out altogether */ + do { + struct inode *inode; + + if (atomic_read(&dentry->d_count) != 0) { + printk(KERN_ERR + "BUG: Dentry %p{i=%lx,n=%s}" + " still in use (%d)" + " [unmount of %s %s]\n", + dentry, + dentry->d_inode ? + dentry->d_inode->i_ino : 0UL, + dentry->d_name.name, + atomic_read(&dentry->d_count), + dentry->d_sb->s_type->name, + dentry->d_sb->s_id); + BUG(); + } + + parent = dentry->d_parent; + if (parent == dentry) + parent = NULL; + else + atomic_dec(&parent->d_count); + + list_del(&dentry->d_u.d_child); + detached++; + + inode = dentry->d_inode; + if (inode) { + dentry->d_inode = NULL; + list_del_init(&dentry->d_alias); + if (dentry->d_op && dentry->d_op->d_iput) + dentry->d_op->d_iput(dentry, inode); + else + iput(inode); + } + + d_free(dentry); + + /* finished when we fall off the top of the tree, + * otherwise we ascend to the parent and move to the + * next sibling if there is one */ + if (!parent) + goto out; + + dentry = parent; + + } while (list_empty(&dentry->d_subdirs)); + + dentry = list_entry(dentry->d_subdirs.next, + struct dentry, d_u.d_child); + } +out: + /* several dentries were freed, need to correct nr_dentry */ + spin_lock(&dcache_lock); + dentry_stat.nr_dentry -= detached; + spin_unlock(&dcache_lock); +} + +/* + * destroy the dentries attached to a superblock on unmounting + * - we don't need to use dentry->d_lock, and only need dcache_lock when + * removing the dentry from the system lists and hashes because: + * - the superblock is detached from all mountings and open files, so the + * dentry trees will not be rearranged by the VFS + * - s_umount is write-locked, so the memory pressure shrinker will ignore + * any dentries belonging to this superblock that it comes across + * - the filesystem itself is no longer permitted to rearrange the dentries + * in this superblock + */ +void shrink_dcache_for_umount(struct super_block *sb) +{ + struct dentry *dentry; + + if (down_read_trylock(&sb->s_umount)) + BUG(); + + dentry = sb->s_root; + sb->s_root = NULL; + atomic_dec(&dentry->d_count); + shrink_dcache_for_umount_subtree(dentry); + + while (!hlist_empty(&sb->s_anon)) { + dentry = hlist_entry(sb->s_anon.first, struct dentry, d_hash); + shrink_dcache_for_umount_subtree(dentry); + } +} + +/* * Search for at least 1 mount point in the dentry's subdirs. * We descend to the next level whenever the d_subdirs * list is non-empty and continue searching. @@ -1339,23 +1476,21 @@ static void switch_names(struct dentry *dentry, struct dentry *target) * deleted it. */ -/** - * d_move - move a dentry +/* + * d_move_locked - move a dentry * @dentry: entry to move * @target: new dentry * * Update the dcache to reflect the move of a file name. Negative * dcache entries should not be moved in this way. */ - -void d_move(struct dentry * dentry, struct dentry * target) +static void d_move_locked(struct dentry * dentry, struct dentry * target) { struct hlist_head *list; if (!dentry->d_inode) printk(KERN_WARNING "VFS: moving negative dcache entry\n"); - spin_lock(&dcache_lock); write_seqlock(&rename_lock); /* * XXXX: do we really need to take target->d_lock? @@ -1406,7 +1541,81 @@ already_unhashed: fsnotify_d_move(dentry); spin_unlock(&dentry->d_lock); write_sequnlock(&rename_lock); +} + +/** + * d_move - move a dentry + * @dentry: entry to move + * @target: new dentry + * + * Update the dcache to reflect the move of a file name. Negative + * dcache entries should not be moved in this way. + */ + +void d_move(struct dentry * dentry, struct dentry * target) +{ + spin_lock(&dcache_lock); + d_move_locked(dentry, target); + spin_unlock(&dcache_lock); +} + +/* + * Helper that returns 1 if p1 is a parent of p2, else 0 + */ +static int d_isparent(struct dentry *p1, struct dentry *p2) +{ + struct dentry *p; + + for (p = p2; p->d_parent != p; p = p->d_parent) { + if (p->d_parent == p1) + return 1; + } + return 0; +} + +/* + * This helper attempts to cope with remotely renamed directories + * + * It assumes that the caller is already holding + * dentry->d_parent->d_inode->i_mutex and the dcache_lock + * + * Note: If ever the locking in lock_rename() changes, then please + * remember to update this too... + * + * On return, dcache_lock will have been unlocked. + */ +static struct dentry *__d_unalias(struct dentry *dentry, struct dentry *alias) +{ + struct mutex *m1 = NULL, *m2 = NULL; + struct dentry *ret; + + /* If alias and dentry share a parent, then no extra locks required */ + if (alias->d_parent == dentry->d_parent) + goto out_unalias; + + /* Check for loops */ + ret = ERR_PTR(-ELOOP); + if (d_isparent(alias, dentry)) + goto out_err; + + /* See lock_rename() */ + ret = ERR_PTR(-EBUSY); + if (!mutex_trylock(&dentry->d_sb->s_vfs_rename_mutex)) + goto out_err; + m1 = &dentry->d_sb->s_vfs_rename_mutex; + if (!mutex_trylock(&alias->d_parent->d_inode->i_mutex)) + goto out_err; + m2 = &alias->d_parent->d_inode->i_mutex; +out_unalias: + d_move_locked(alias, dentry); + ret = alias; +out_err: spin_unlock(&dcache_lock); + if (m2) + mutex_unlock(m2); + if (m1) + mutex_unlock(m1); + return ret; } /* @@ -1451,7 +1660,7 @@ static void __d_materialise_dentry(struct dentry *dentry, struct dentry *anon) */ struct dentry *d_materialise_unique(struct dentry *dentry, struct inode *inode) { - struct dentry *alias, *actual; + struct dentry *actual; BUG_ON(!d_unhashed(dentry)); @@ -1463,26 +1672,27 @@ struct dentry *d_materialise_unique(struct dentry *dentry, struct inode *inode) goto found_lock; } - /* See if a disconnected directory already exists as an anonymous root - * that we should splice into the tree instead */ - if (S_ISDIR(inode->i_mode) && (alias = __d_find_alias(inode, 1))) { - spin_lock(&alias->d_lock); - - /* Is this a mountpoint that we could splice into our tree? */ - if (IS_ROOT(alias)) - goto connect_mountpoint; - - if (alias->d_name.len == dentry->d_name.len && - alias->d_parent == dentry->d_parent && - memcmp(alias->d_name.name, - dentry->d_name.name, - dentry->d_name.len) == 0) - goto replace_with_alias; - - spin_unlock(&alias->d_lock); - - /* Doh! Seem to be aliasing directories for some reason... */ - dput(alias); + if (S_ISDIR(inode->i_mode)) { + struct dentry *alias; + + /* Does an aliased dentry already exist? */ + alias = __d_find_alias(inode, 0); + if (alias) { + actual = alias; + /* Is this an anonymous mountpoint that we could splice + * into our tree? */ + if (IS_ROOT(alias)) { + spin_lock(&alias->d_lock); + __d_materialise_dentry(dentry, alias); + __d_drop(alias); + goto found; + } + /* Nope, but we must(!) avoid directory aliasing */ + actual = __d_unalias(dentry, alias); + if (IS_ERR(actual)) + dput(alias); + goto out_nolock; + } } /* Add a unique reference */ @@ -1498,7 +1708,7 @@ found: _d_rehash(actual); spin_unlock(&actual->d_lock); spin_unlock(&dcache_lock); - +out_nolock: if (actual == dentry) { security_d_instantiate(dentry, inode); return NULL; @@ -1507,16 +1717,6 @@ found: iput(inode); return actual; - /* Convert the anonymous/root alias into an ordinary dentry */ -connect_mountpoint: - __d_materialise_dentry(dentry, alias); - - /* Replace the candidate dentry with the alias in the tree */ -replace_with_alias: - __d_drop(alias); - actual = alias; - goto found; - shouldnt_be_hashed: spin_unlock(&dcache_lock); BUG(); diff --git a/fs/dlm/Kconfig b/fs/dlm/Kconfig index 490f85b3fa59..81b2c6465eeb 100644 --- a/fs/dlm/Kconfig +++ b/fs/dlm/Kconfig @@ -1,10 +1,9 @@ menu "Distributed Lock Manager" - depends on INET && EXPERIMENTAL + depends on INET && IP_SCTP && EXPERIMENTAL config DLM tristate "Distributed Lock Manager (DLM)" depends on IPV6 || IPV6=n - depends on IP_SCTP select CONFIGFS_FS help A general purpose distributed lock manager for kernel or userspace diff --git a/fs/dlm/lowcomms.c b/fs/dlm/lowcomms.c index 7bcea7c5addb..6da6b14d5a61 100644 --- a/fs/dlm/lowcomms.c +++ b/fs/dlm/lowcomms.c @@ -519,6 +519,7 @@ static int receive_from_sock(void) msg.msg_flags = 0; msg.msg_control = incmsg; msg.msg_controllen = sizeof(incmsg); + msg.msg_iovlen = 1; /* I don't see why this circular buffer stuff is necessary for SCTP * which is a packet-based protocol, but the whole thing breaks under @@ -548,7 +549,7 @@ static int receive_from_sock(void) } len = iov[0].iov_len + iov[1].iov_len; - r = ret = kernel_recvmsg(sctp_con.sock, &msg, iov, 1, len, + r = ret = kernel_recvmsg(sctp_con.sock, &msg, iov, msg.msg_iovlen, len, MSG_NOSIGNAL | MSG_DONTWAIT); if (ret <= 0) goto out_close; diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c index ed35a9712fa1..136175a69332 100644 --- a/fs/ecryptfs/crypto.c +++ b/fs/ecryptfs/crypto.c @@ -94,25 +94,53 @@ static int ecryptfs_calculate_md5(char *dst, struct ecryptfs_crypt_stat *crypt_stat, char *src, int len) { - int rc = 0; struct scatterlist sg; + struct hash_desc desc = { + .tfm = crypt_stat->hash_tfm, + .flags = CRYPTO_TFM_REQ_MAY_SLEEP + }; + int rc = 0; - mutex_lock(&crypt_stat->cs_md5_tfm_mutex); + mutex_lock(&crypt_stat->cs_hash_tfm_mutex); sg_init_one(&sg, (u8 *)src, len); - if (!crypt_stat->md5_tfm) { - crypt_stat->md5_tfm = - crypto_alloc_tfm("md5", CRYPTO_TFM_REQ_MAY_SLEEP); - if (!crypt_stat->md5_tfm) { - rc = -ENOMEM; + if (!desc.tfm) { + desc.tfm = crypto_alloc_hash(ECRYPTFS_DEFAULT_HASH, 0, + CRYPTO_ALG_ASYNC); + if (IS_ERR(desc.tfm)) { + rc = PTR_ERR(desc.tfm); ecryptfs_printk(KERN_ERR, "Error attempting to " - "allocate crypto context\n"); + "allocate crypto context; rc = [%d]\n", + rc); goto out; } + crypt_stat->hash_tfm = desc.tfm; } - crypto_digest_init(crypt_stat->md5_tfm); - crypto_digest_update(crypt_stat->md5_tfm, &sg, 1); - crypto_digest_final(crypt_stat->md5_tfm, dst); - mutex_unlock(&crypt_stat->cs_md5_tfm_mutex); + crypto_hash_init(&desc); + crypto_hash_update(&desc, &sg, len); + crypto_hash_final(&desc, dst); + mutex_unlock(&crypt_stat->cs_hash_tfm_mutex); +out: + return rc; +} + +int ecryptfs_crypto_api_algify_cipher_name(char **algified_name, + char *cipher_name, + char *chaining_modifier) +{ + int cipher_name_len = strlen(cipher_name); + int chaining_modifier_len = strlen(chaining_modifier); + int algified_name_len; + int rc; + + algified_name_len = (chaining_modifier_len + cipher_name_len + 3); + (*algified_name) = kmalloc(algified_name_len, GFP_KERNEL); + if (!(*algified_name)) { + rc = -ENOMEM; + goto out; + } + snprintf((*algified_name), algified_name_len, "%s(%s)", + chaining_modifier, cipher_name); + rc = 0; out: return rc; } @@ -178,7 +206,7 @@ ecryptfs_init_crypt_stat(struct ecryptfs_crypt_stat *crypt_stat) memset((void *)crypt_stat, 0, sizeof(struct ecryptfs_crypt_stat)); mutex_init(&crypt_stat->cs_mutex); mutex_init(&crypt_stat->cs_tfm_mutex); - mutex_init(&crypt_stat->cs_md5_tfm_mutex); + mutex_init(&crypt_stat->cs_hash_tfm_mutex); ECRYPTFS_SET_FLAG(crypt_stat->flags, ECRYPTFS_STRUCT_INITIALIZED); } @@ -191,9 +219,9 @@ ecryptfs_init_crypt_stat(struct ecryptfs_crypt_stat *crypt_stat) void ecryptfs_destruct_crypt_stat(struct ecryptfs_crypt_stat *crypt_stat) { if (crypt_stat->tfm) - crypto_free_tfm(crypt_stat->tfm); - if (crypt_stat->md5_tfm) - crypto_free_tfm(crypt_stat->md5_tfm); + crypto_free_blkcipher(crypt_stat->tfm); + if (crypt_stat->hash_tfm) + crypto_free_hash(crypt_stat->hash_tfm); memset(crypt_stat, 0, sizeof(struct ecryptfs_crypt_stat)); } @@ -203,7 +231,7 @@ void ecryptfs_destruct_mount_crypt_stat( if (mount_crypt_stat->global_auth_tok_key) key_put(mount_crypt_stat->global_auth_tok_key); if (mount_crypt_stat->global_key_tfm) - crypto_free_tfm(mount_crypt_stat->global_key_tfm); + crypto_free_blkcipher(mount_crypt_stat->global_key_tfm); memset(mount_crypt_stat, 0, sizeof(struct ecryptfs_mount_crypt_stat)); } @@ -269,6 +297,11 @@ static int encrypt_scatterlist(struct ecryptfs_crypt_stat *crypt_stat, struct scatterlist *src_sg, int size, unsigned char *iv) { + struct blkcipher_desc desc = { + .tfm = crypt_stat->tfm, + .info = iv, + .flags = CRYPTO_TFM_REQ_MAY_SLEEP + }; int rc = 0; BUG_ON(!crypt_stat || !crypt_stat->tfm @@ -282,8 +315,8 @@ static int encrypt_scatterlist(struct ecryptfs_crypt_stat *crypt_stat, } /* Consider doing this once, when the file is opened */ mutex_lock(&crypt_stat->cs_tfm_mutex); - rc = crypto_cipher_setkey(crypt_stat->tfm, crypt_stat->key, - crypt_stat->key_size); + rc = crypto_blkcipher_setkey(crypt_stat->tfm, crypt_stat->key, + crypt_stat->key_size); if (rc) { ecryptfs_printk(KERN_ERR, "Error setting key; rc = [%d]\n", rc); @@ -292,7 +325,7 @@ static int encrypt_scatterlist(struct ecryptfs_crypt_stat *crypt_stat, goto out; } ecryptfs_printk(KERN_DEBUG, "Encrypting [%d] bytes.\n", size); - crypto_cipher_encrypt_iv(crypt_stat->tfm, dest_sg, src_sg, size, iv); + crypto_blkcipher_encrypt_iv(&desc, dest_sg, src_sg, size); mutex_unlock(&crypt_stat->cs_tfm_mutex); out: return rc; @@ -675,12 +708,17 @@ static int decrypt_scatterlist(struct ecryptfs_crypt_stat *crypt_stat, struct scatterlist *src_sg, int size, unsigned char *iv) { + struct blkcipher_desc desc = { + .tfm = crypt_stat->tfm, + .info = iv, + .flags = CRYPTO_TFM_REQ_MAY_SLEEP + }; int rc = 0; /* Consider doing this once, when the file is opened */ mutex_lock(&crypt_stat->cs_tfm_mutex); - rc = crypto_cipher_setkey(crypt_stat->tfm, crypt_stat->key, - crypt_stat->key_size); + rc = crypto_blkcipher_setkey(crypt_stat->tfm, crypt_stat->key, + crypt_stat->key_size); if (rc) { ecryptfs_printk(KERN_ERR, "Error setting key; rc = [%d]\n", rc); @@ -689,8 +727,7 @@ static int decrypt_scatterlist(struct ecryptfs_crypt_stat *crypt_stat, goto out; } ecryptfs_printk(KERN_DEBUG, "Decrypting [%d] bytes.\n", size); - rc = crypto_cipher_decrypt_iv(crypt_stat->tfm, dest_sg, src_sg, size, - iv); + rc = crypto_blkcipher_decrypt_iv(&desc, dest_sg, src_sg, size); mutex_unlock(&crypt_stat->cs_tfm_mutex); if (rc) { ecryptfs_printk(KERN_ERR, "Error decrypting; rc = [%d]\n", @@ -759,6 +796,7 @@ ecryptfs_decrypt_page_offset(struct ecryptfs_crypt_stat *crypt_stat, */ int ecryptfs_init_crypt_ctx(struct ecryptfs_crypt_stat *crypt_stat) { + char *full_alg_name; int rc = -EINVAL; if (!crypt_stat->cipher) { @@ -775,16 +813,24 @@ int ecryptfs_init_crypt_ctx(struct ecryptfs_crypt_stat *crypt_stat) goto out; } mutex_lock(&crypt_stat->cs_tfm_mutex); - crypt_stat->tfm = crypto_alloc_tfm(crypt_stat->cipher, - ECRYPTFS_DEFAULT_CHAINING_MODE - | CRYPTO_TFM_REQ_WEAK_KEY); - mutex_unlock(&crypt_stat->cs_tfm_mutex); + rc = ecryptfs_crypto_api_algify_cipher_name(&full_alg_name, + crypt_stat->cipher, "cbc"); + if (rc) + goto out; + crypt_stat->tfm = crypto_alloc_blkcipher(full_alg_name, 0, + CRYPTO_ALG_ASYNC); + kfree(full_alg_name); if (!crypt_stat->tfm) { ecryptfs_printk(KERN_ERR, "cryptfs: init_crypt_ctx(): " "Error initializing cipher [%s]\n", crypt_stat->cipher); + mutex_unlock(&crypt_stat->cs_tfm_mutex); goto out; } + crypto_blkcipher_set_flags(crypt_stat->tfm, + (ECRYPTFS_DEFAULT_CHAINING_MODE + | CRYPTO_TFM_REQ_WEAK_KEY)); + mutex_unlock(&crypt_stat->cs_tfm_mutex); rc = 0; out: return rc; @@ -1145,28 +1191,28 @@ int ecryptfs_cipher_code_to_string(char *str, u16 cipher_code) int ecryptfs_read_header_region(char *data, struct dentry *dentry, struct vfsmount *mnt) { - struct file *file; + struct file *lower_file; mm_segment_t oldfs; int rc; - mnt = mntget(mnt); - file = dentry_open(dentry, mnt, O_RDONLY); - if (IS_ERR(file)) { - ecryptfs_printk(KERN_DEBUG, "Error opening file to " - "read header region\n"); - mntput(mnt); - rc = PTR_ERR(file); + if ((rc = ecryptfs_open_lower_file(&lower_file, dentry, mnt, + O_RDONLY))) { + printk(KERN_ERR + "Error opening lower_file to read header region\n"); goto out; } - file->f_pos = 0; + lower_file->f_pos = 0; oldfs = get_fs(); set_fs(get_ds()); /* For releases 0.1 and 0.2, all of the header information * fits in the first data extent-sized region. */ - rc = file->f_op->read(file, (char __user *)data, - ECRYPTFS_DEFAULT_EXTENT_SIZE, &file->f_pos); + rc = lower_file->f_op->read(lower_file, (char __user *)data, + ECRYPTFS_DEFAULT_EXTENT_SIZE, &lower_file->f_pos); set_fs(oldfs); - fput(file); + if ((rc = ecryptfs_close_lower_file(lower_file))) { + printk(KERN_ERR "Error closing lower_file\n"); + goto out; + } rc = 0; out: return rc; @@ -1573,84 +1619,52 @@ out: /** * ecryptfs_process_cipher - Perform cipher initialization. - * @tfm: Crypto context set by this function * @key_tfm: Crypto context for key material, set by this function - * @cipher_name: Name of the cipher. - * @key_size: Size of the key in bytes. + * @cipher_name: Name of the cipher + * @key_size: Size of the key in bytes * * Returns zero on success. Any crypto_tfm structs allocated here * should be released by other functions, such as on a superblock put * event, regardless of whether this function succeeds for fails. */ int -ecryptfs_process_cipher(struct crypto_tfm **tfm, struct crypto_tfm **key_tfm, - char *cipher_name, size_t key_size) +ecryptfs_process_cipher(struct crypto_blkcipher **key_tfm, char *cipher_name, + size_t *key_size) { char dummy_key[ECRYPTFS_MAX_KEY_BYTES]; + char *full_alg_name; int rc; - *tfm = *key_tfm = NULL; - if (key_size > ECRYPTFS_MAX_KEY_BYTES) { + *key_tfm = NULL; + if (*key_size > ECRYPTFS_MAX_KEY_BYTES) { rc = -EINVAL; printk(KERN_ERR "Requested key size is [%Zd] bytes; maximum " - "allowable is [%d]\n", key_size, ECRYPTFS_MAX_KEY_BYTES); + "allowable is [%d]\n", *key_size, ECRYPTFS_MAX_KEY_BYTES); goto out; } - *tfm = crypto_alloc_tfm(cipher_name, (ECRYPTFS_DEFAULT_CHAINING_MODE - | CRYPTO_TFM_REQ_WEAK_KEY)); - if (!(*tfm)) { - rc = -EINVAL; - printk(KERN_ERR "Unable to allocate crypto cipher with name " - "[%s]\n", cipher_name); + rc = ecryptfs_crypto_api_algify_cipher_name(&full_alg_name, cipher_name, + "ecb"); + if (rc) goto out; - } - *key_tfm = crypto_alloc_tfm(cipher_name, CRYPTO_TFM_REQ_WEAK_KEY); - if (!(*key_tfm)) { - rc = -EINVAL; + *key_tfm = crypto_alloc_blkcipher(full_alg_name, 0, CRYPTO_ALG_ASYNC); + kfree(full_alg_name); + if (IS_ERR(*key_tfm)) { + rc = PTR_ERR(*key_tfm); printk(KERN_ERR "Unable to allocate crypto cipher with name " - "[%s]\n", cipher_name); - goto out; - } - if (key_size < crypto_tfm_alg_min_keysize(*tfm)) { - rc = -EINVAL; - printk(KERN_ERR "Request key size is [%Zd]; minimum key size " - "supported by cipher [%s] is [%d]\n", key_size, - cipher_name, crypto_tfm_alg_min_keysize(*tfm)); - goto out; - } - if (key_size < crypto_tfm_alg_min_keysize(*key_tfm)) { - rc = -EINVAL; - printk(KERN_ERR "Request key size is [%Zd]; minimum key size " - "supported by cipher [%s] is [%d]\n", key_size, - cipher_name, crypto_tfm_alg_min_keysize(*key_tfm)); + "[%s]; rc = [%d]\n", cipher_name, rc); goto out; } - if (key_size > crypto_tfm_alg_max_keysize(*tfm)) { - rc = -EINVAL; - printk(KERN_ERR "Request key size is [%Zd]; maximum key size " - "supported by cipher [%s] is [%d]\n", key_size, - cipher_name, crypto_tfm_alg_min_keysize(*tfm)); - goto out; - } - if (key_size > crypto_tfm_alg_max_keysize(*key_tfm)) { - rc = -EINVAL; - printk(KERN_ERR "Request key size is [%Zd]; maximum key size " - "supported by cipher [%s] is [%d]\n", key_size, - cipher_name, crypto_tfm_alg_min_keysize(*key_tfm)); - goto out; - } - get_random_bytes(dummy_key, key_size); - rc = crypto_cipher_setkey(*tfm, dummy_key, key_size); - if (rc) { - printk(KERN_ERR "Error attempting to set key of size [%Zd] for " - "cipher [%s]; rc = [%d]\n", key_size, cipher_name, rc); - rc = -EINVAL; - goto out; + crypto_blkcipher_set_flags(*key_tfm, CRYPTO_TFM_REQ_WEAK_KEY); + if (*key_size == 0) { + struct blkcipher_alg *alg = crypto_blkcipher_alg(*key_tfm); + + *key_size = alg->max_keysize; } - rc = crypto_cipher_setkey(*key_tfm, dummy_key, key_size); + get_random_bytes(dummy_key, *key_size); + rc = crypto_blkcipher_setkey(*key_tfm, dummy_key, *key_size); if (rc) { printk(KERN_ERR "Error attempting to set key of size [%Zd] for " - "cipher [%s]; rc = [%d]\n", key_size, cipher_name, rc); + "cipher [%s]; rc = [%d]\n", *key_size, cipher_name, rc); rc = -EINVAL; goto out; } diff --git a/fs/ecryptfs/dentry.c b/fs/ecryptfs/dentry.c index f0d2a433242b..0b9992ab990f 100644 --- a/fs/ecryptfs/dentry.c +++ b/fs/ecryptfs/dentry.c @@ -24,6 +24,7 @@ #include <linux/dcache.h> #include <linux/namei.h> +#include <linux/mount.h> #include "ecryptfs_kernel.h" /** @@ -76,8 +77,13 @@ static void ecryptfs_d_release(struct dentry *dentry) if (ecryptfs_dentry_to_private(dentry)) kmem_cache_free(ecryptfs_dentry_info_cache, ecryptfs_dentry_to_private(dentry)); - if (lower_dentry) + if (lower_dentry) { + struct vfsmount *lower_mnt = + ecryptfs_dentry_to_lower_mnt(dentry); + + mntput(lower_mnt); dput(lower_dentry); + } return; } diff --git a/fs/ecryptfs/ecryptfs_kernel.h b/fs/ecryptfs/ecryptfs_kernel.h index 872c9958531a..f992533d1692 100644 --- a/fs/ecryptfs/ecryptfs_kernel.h +++ b/fs/ecryptfs/ecryptfs_kernel.h @@ -175,6 +175,7 @@ ecryptfs_get_key_payload_data(struct key *key) #define ECRYPTFS_DEFAULT_CIPHER "aes" #define ECRYPTFS_DEFAULT_KEY_BYTES 16 #define ECRYPTFS_DEFAULT_CHAINING_MODE CRYPTO_TFM_MODE_CBC +#define ECRYPTFS_DEFAULT_HASH "md5" #define ECRYPTFS_TAG_3_PACKET_TYPE 0x8C #define ECRYPTFS_TAG_11_PACKET_TYPE 0xED #define MD5_DIGEST_SIZE 16 @@ -204,15 +205,15 @@ struct ecryptfs_crypt_stat { size_t extent_shift; unsigned int extent_mask; struct ecryptfs_mount_crypt_stat *mount_crypt_stat; - struct crypto_tfm *tfm; - struct crypto_tfm *md5_tfm; /* Crypto context for generating - * the initialization vectors */ + struct crypto_blkcipher *tfm; + struct crypto_hash *hash_tfm; /* Crypto context for generating + * the initialization vectors */ unsigned char cipher[ECRYPTFS_MAX_CIPHER_NAME_SIZE]; unsigned char key[ECRYPTFS_MAX_KEY_BYTES]; unsigned char root_iv[ECRYPTFS_MAX_IV_BYTES]; unsigned char keysigs[ECRYPTFS_MAX_NUM_KEYSIGS][ECRYPTFS_SIG_SIZE_HEX]; struct mutex cs_tfm_mutex; - struct mutex cs_md5_tfm_mutex; + struct mutex cs_hash_tfm_mutex; struct mutex cs_mutex; }; @@ -244,7 +245,7 @@ struct ecryptfs_mount_crypt_stat { struct ecryptfs_auth_tok *global_auth_tok; struct key *global_auth_tok_key; size_t global_default_cipher_key_size; - struct crypto_tfm *global_key_tfm; + struct crypto_blkcipher *global_key_tfm; struct mutex global_key_tfm_mutex; unsigned char global_default_cipher_name[ECRYPTFS_MAX_CIPHER_NAME_SIZE + 1]; @@ -425,6 +426,9 @@ void ecryptfs_destruct_crypt_stat(struct ecryptfs_crypt_stat *crypt_stat); void ecryptfs_destruct_mount_crypt_stat( struct ecryptfs_mount_crypt_stat *mount_crypt_stat); int ecryptfs_init_crypt_ctx(struct ecryptfs_crypt_stat *crypt_stat); +int ecryptfs_crypto_api_algify_cipher_name(char **algified_name, + char *cipher_name, + char *chaining_modifier); int ecryptfs_write_inode_size_to_header(struct file *lower_file, struct inode *lower_inode, struct inode *inode); @@ -473,10 +477,14 @@ ecryptfs_parse_packet_set(struct ecryptfs_crypt_stat *crypt_stat, unsigned char *src, struct dentry *ecryptfs_dentry); int ecryptfs_truncate(struct dentry *dentry, loff_t new_length); int -ecryptfs_process_cipher(struct crypto_tfm **tfm, struct crypto_tfm **key_tfm, - char *cipher_name, size_t key_size); +ecryptfs_process_cipher(struct crypto_blkcipher **key_tfm, char *cipher_name, + size_t *key_size); int ecryptfs_inode_test(struct inode *inode, void *candidate_lower_inode); int ecryptfs_inode_set(struct inode *inode, void *lower_inode); void ecryptfs_init_inode(struct inode *inode, struct inode *lower_inode); +int ecryptfs_open_lower_file(struct file **lower_file, + struct dentry *lower_dentry, + struct vfsmount *lower_mnt, int flags); +int ecryptfs_close_lower_file(struct file *lower_file); #endif /* #ifndef ECRYPTFS_KERNEL_H */ diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c index c8550c9f9cd2..a92ef05eff8f 100644 --- a/fs/ecryptfs/file.c +++ b/fs/ecryptfs/file.c @@ -198,6 +198,33 @@ retry: struct kmem_cache *ecryptfs_file_info_cache; +int ecryptfs_open_lower_file(struct file **lower_file, + struct dentry *lower_dentry, + struct vfsmount *lower_mnt, int flags) +{ + int rc = 0; + + dget(lower_dentry); + mntget(lower_mnt); + *lower_file = dentry_open(lower_dentry, lower_mnt, flags); + if (IS_ERR(*lower_file)) { + printk(KERN_ERR "Error opening lower file for lower_dentry " + "[0x%p], lower_mnt [0x%p], and flags [0x%x]\n", + lower_dentry, lower_mnt, flags); + rc = PTR_ERR(*lower_file); + *lower_file = NULL; + goto out; + } +out: + return rc; +} + +int ecryptfs_close_lower_file(struct file *lower_file) +{ + fput(lower_file); + return 0; +} + /** * ecryptfs_open * @inode: inode speciying file to open @@ -244,19 +271,15 @@ static int ecryptfs_open(struct inode *inode, struct file *file) ECRYPTFS_SET_FLAG(crypt_stat->flags, ECRYPTFS_ENCRYPTED); } mutex_unlock(&crypt_stat->cs_mutex); - /* This mntget & dget is undone via fput when the file is released */ - dget(lower_dentry); lower_flags = file->f_flags; if ((lower_flags & O_ACCMODE) == O_WRONLY) lower_flags = (lower_flags & O_ACCMODE) | O_RDWR; if (file->f_flags & O_APPEND) lower_flags &= ~O_APPEND; lower_mnt = ecryptfs_dentry_to_lower_mnt(ecryptfs_dentry); - mntget(lower_mnt); /* Corresponding fput() in ecryptfs_release() */ - lower_file = dentry_open(lower_dentry, lower_mnt, lower_flags); - if (IS_ERR(lower_file)) { - rc = PTR_ERR(lower_file); + if ((rc = ecryptfs_open_lower_file(&lower_file, lower_dentry, lower_mnt, + lower_flags))) { ecryptfs_printk(KERN_ERR, "Error opening lower file\n"); goto out_puts; } @@ -341,11 +364,16 @@ static int ecryptfs_release(struct inode *inode, struct file *file) struct file *lower_file = ecryptfs_file_to_lower(file); struct ecryptfs_file_info *file_info = ecryptfs_file_to_private(file); struct inode *lower_inode = ecryptfs_inode_to_lower(inode); + int rc; - fput(lower_file); + if ((rc = ecryptfs_close_lower_file(lower_file))) { + printk(KERN_ERR "Error closing lower_file\n"); + goto out; + } inode->i_blocks = lower_inode->i_blocks; kmem_cache_free(ecryptfs_file_info_cache, file_info); - return 0; +out: + return rc; } static int diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index efdd2b7b62d7..ff4865d24f0f 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -231,7 +231,6 @@ static int ecryptfs_initialize_file(struct dentry *ecryptfs_dentry) int lower_flags; struct ecryptfs_crypt_stat *crypt_stat; struct dentry *lower_dentry; - struct dentry *tlower_dentry = NULL; struct file *lower_file; struct inode *inode, *lower_inode; struct vfsmount *lower_mnt; @@ -241,30 +240,19 @@ static int ecryptfs_initialize_file(struct dentry *ecryptfs_dentry) lower_dentry->d_name.name); inode = ecryptfs_dentry->d_inode; crypt_stat = &ecryptfs_inode_to_private(inode)->crypt_stat; - tlower_dentry = dget(lower_dentry); - if (!tlower_dentry) { - rc = -ENOMEM; - ecryptfs_printk(KERN_ERR, "Error dget'ing lower_dentry\n"); - goto out; - } lower_flags = ((O_CREAT | O_WRONLY | O_TRUNC) & O_ACCMODE) | O_RDWR; #if BITS_PER_LONG != 32 lower_flags |= O_LARGEFILE; #endif lower_mnt = ecryptfs_dentry_to_lower_mnt(ecryptfs_dentry); - mntget(lower_mnt); /* Corresponding fput() at end of this function */ - lower_file = dentry_open(tlower_dentry, lower_mnt, lower_flags); - if (IS_ERR(lower_file)) { - rc = PTR_ERR(lower_file); + if ((rc = ecryptfs_open_lower_file(&lower_file, lower_dentry, lower_mnt, + lower_flags))) { ecryptfs_printk(KERN_ERR, "Error opening dentry; rc = [%i]\n", rc); goto out; } - /* fput(lower_file) should handle the puts if we do this */ - lower_file->f_dentry = tlower_dentry; - lower_file->f_vfsmnt = lower_mnt; - lower_inode = tlower_dentry->d_inode; + lower_inode = lower_dentry->d_inode; if (S_ISDIR(ecryptfs_dentry->d_inode->i_mode)) { ecryptfs_printk(KERN_DEBUG, "This is a directory\n"); ECRYPTFS_CLEAR_FLAG(crypt_stat->flags, ECRYPTFS_ENCRYPTED); @@ -285,7 +273,8 @@ static int ecryptfs_initialize_file(struct dentry *ecryptfs_dentry) } rc = grow_file(ecryptfs_dentry, lower_file, inode, lower_inode); out_fput: - fput(lower_file); + if ((rc = ecryptfs_close_lower_file(lower_file))) + printk(KERN_ERR "Error closing lower_file\n"); out: return rc; } @@ -336,7 +325,6 @@ static struct dentry *ecryptfs_lookup(struct inode *dir, struct dentry *dentry, struct dentry *lower_dir_dentry; struct dentry *lower_dentry; struct vfsmount *lower_mnt; - struct dentry *tlower_dentry = NULL; char *encoded_name; unsigned int encoded_namelen; struct ecryptfs_crypt_stat *crypt_stat = NULL; @@ -347,27 +335,32 @@ static struct dentry *ecryptfs_lookup(struct inode *dir, struct dentry *dentry, lower_dir_dentry = ecryptfs_dentry_to_lower(dentry->d_parent); dentry->d_op = &ecryptfs_dops; if ((dentry->d_name.len == 1 && !strcmp(dentry->d_name.name, ".")) - || (dentry->d_name.len == 2 && !strcmp(dentry->d_name.name, ".."))) - goto out_drop; + || (dentry->d_name.len == 2 + && !strcmp(dentry->d_name.name, ".."))) { + d_drop(dentry); + goto out; + } encoded_namelen = ecryptfs_encode_filename(crypt_stat, dentry->d_name.name, dentry->d_name.len, &encoded_name); if (encoded_namelen < 0) { rc = encoded_namelen; - goto out_drop; + d_drop(dentry); + goto out; } ecryptfs_printk(KERN_DEBUG, "encoded_name = [%s]; encoded_namelen " "= [%d]\n", encoded_name, encoded_namelen); lower_dentry = lookup_one_len(encoded_name, lower_dir_dentry, encoded_namelen - 1); kfree(encoded_name); - lower_mnt = mntget(ecryptfs_dentry_to_lower_mnt(dentry->d_parent)); if (IS_ERR(lower_dentry)) { ecryptfs_printk(KERN_ERR, "ERR from lower_dentry\n"); rc = PTR_ERR(lower_dentry); - goto out_drop; + d_drop(dentry); + goto out; } + lower_mnt = mntget(ecryptfs_dentry_to_lower_mnt(dentry->d_parent)); ecryptfs_printk(KERN_DEBUG, "lower_dentry = [%p]; lower_dentry->" "d_name.name = [%s]\n", lower_dentry, lower_dentry->d_name.name); @@ -408,12 +401,6 @@ static struct dentry *ecryptfs_lookup(struct inode *dir, struct dentry *dentry, "as we *think* we are about to unlink\n"); goto out; } - tlower_dentry = dget(lower_dentry); - if (!tlower_dentry || IS_ERR(tlower_dentry)) { - rc = -ENOMEM; - ecryptfs_printk(KERN_ERR, "Cannot dget lower_dentry\n"); - goto out_dput; - } /* Released in this function */ page_virt = (char *)kmem_cache_alloc(ecryptfs_header_cache_2, @@ -425,7 +412,7 @@ static struct dentry *ecryptfs_lookup(struct inode *dir, struct dentry *dentry, goto out_dput; } memset(page_virt, 0, PAGE_CACHE_SIZE); - rc = ecryptfs_read_header_region(page_virt, tlower_dentry, nd->mnt); + rc = ecryptfs_read_header_region(page_virt, lower_dentry, nd->mnt); crypt_stat = &ecryptfs_inode_to_private(dentry->d_inode)->crypt_stat; if (!ECRYPTFS_CHECK_FLAG(crypt_stat->flags, ECRYPTFS_POLICY_APPLIED)) ecryptfs_set_default_sizes(crypt_stat); @@ -448,9 +435,6 @@ static struct dentry *ecryptfs_lookup(struct inode *dir, struct dentry *dentry, out_dput: dput(lower_dentry); - if (tlower_dentry) - dput(tlower_dentry); -out_drop: d_drop(dentry); out: return ERR_PTR(rc); @@ -486,8 +470,8 @@ out_lock: unlock_dir(lower_dir_dentry); dput(lower_new_dentry); dput(lower_old_dentry); - if (!new_dentry->d_inode) - d_drop(new_dentry); + d_drop(new_dentry); + d_drop(old_dentry); return rc; } @@ -576,41 +560,24 @@ out: static int ecryptfs_rmdir(struct inode *dir, struct dentry *dentry) { - int rc = 0; - struct dentry *tdentry = NULL; struct dentry *lower_dentry; - struct dentry *tlower_dentry = NULL; struct dentry *lower_dir_dentry; + int rc; lower_dentry = ecryptfs_dentry_to_lower(dentry); - if (!(tdentry = dget(dentry))) { - rc = -EINVAL; - ecryptfs_printk(KERN_ERR, "Error dget'ing dentry [%p]\n", - dentry); - goto out; - } + dget(dentry); lower_dir_dentry = lock_parent(lower_dentry); - if (!(tlower_dentry = dget(lower_dentry))) { - rc = -EINVAL; - ecryptfs_printk(KERN_ERR, "Error dget'ing lower_dentry " - "[%p]\n", lower_dentry); - goto out; - } + dget(lower_dentry); rc = vfs_rmdir(lower_dir_dentry->d_inode, lower_dentry); - if (!rc) { - d_delete(tlower_dentry); - tlower_dentry = NULL; - } + dput(lower_dentry); + if (!rc) + d_delete(lower_dentry); ecryptfs_copy_attr_times(dir, lower_dir_dentry->d_inode); dir->i_nlink = lower_dir_dentry->d_inode->i_nlink; unlock_dir(lower_dir_dentry); if (!rc) d_drop(dentry); -out: - if (tdentry) - dput(tdentry); - if (tlower_dentry) - dput(tlower_dentry); + dput(dentry); return rc; } @@ -832,12 +799,11 @@ int ecryptfs_truncate(struct dentry *dentry, loff_t new_length) } lower_dentry = ecryptfs_dentry_to_lower(dentry); /* This dget & mntget is released through fput at out_fput: */ - dget(lower_dentry); lower_mnt = ecryptfs_dentry_to_lower_mnt(dentry); - mntget(lower_mnt); - lower_file = dentry_open(lower_dentry, lower_mnt, O_RDWR); - if (unlikely(IS_ERR(lower_file))) { - rc = PTR_ERR(lower_file); + if ((rc = ecryptfs_open_lower_file(&lower_file, lower_dentry, lower_mnt, + O_RDWR))) { + ecryptfs_printk(KERN_ERR, + "Error opening dentry; rc = [%i]\n", rc); goto out_free; } ecryptfs_set_file_lower(&fake_ecryptfs_file, lower_file); @@ -879,7 +845,8 @@ int ecryptfs_truncate(struct dentry *dentry, loff_t new_length) = CURRENT_TIME; mark_inode_dirty_sync(inode); out_fput: - fput(lower_file); + if ((rc = ecryptfs_close_lower_file(lower_file))) + printk(KERN_ERR "Error closing lower_file\n"); out_free: if (ecryptfs_file_to_private(&fake_ecryptfs_file)) kmem_cache_free(ecryptfs_file_info_cache, diff --git a/fs/ecryptfs/keystore.c b/fs/ecryptfs/keystore.c index ba454785a0c5..c3746f56d162 100644 --- a/fs/ecryptfs/keystore.c +++ b/fs/ecryptfs/keystore.c @@ -458,14 +458,16 @@ out: static int decrypt_session_key(struct ecryptfs_auth_tok *auth_tok, struct ecryptfs_crypt_stat *crypt_stat) { - int rc = 0; struct ecryptfs_password *password_s_ptr; - struct crypto_tfm *tfm = NULL; struct scatterlist src_sg[2], dst_sg[2]; struct mutex *tfm_mutex = NULL; /* TODO: Use virt_to_scatterlist for these */ char *encrypted_session_key; char *session_key; + struct blkcipher_desc desc = { + .flags = CRYPTO_TFM_REQ_MAY_SLEEP + }; + int rc = 0; password_s_ptr = &auth_tok->token.password; if (ECRYPTFS_CHECK_FLAG(password_s_ptr->flags, @@ -482,30 +484,37 @@ static int decrypt_session_key(struct ecryptfs_auth_tok *auth_tok, if (!strcmp(crypt_stat->cipher, crypt_stat->mount_crypt_stat->global_default_cipher_name) && crypt_stat->mount_crypt_stat->global_key_tfm) { - tfm = crypt_stat->mount_crypt_stat->global_key_tfm; + desc.tfm = crypt_stat->mount_crypt_stat->global_key_tfm; tfm_mutex = &crypt_stat->mount_crypt_stat->global_key_tfm_mutex; } else { - tfm = crypto_alloc_tfm(crypt_stat->cipher, - CRYPTO_TFM_REQ_WEAK_KEY); - if (!tfm) { - printk(KERN_ERR "Error allocating crypto context\n"); - rc = -ENOMEM; + char *full_alg_name; + + rc = ecryptfs_crypto_api_algify_cipher_name(&full_alg_name, + crypt_stat->cipher, + "ecb"); + if (rc) + goto out; + desc.tfm = crypto_alloc_blkcipher(full_alg_name, 0, + CRYPTO_ALG_ASYNC); + kfree(full_alg_name); + if (IS_ERR(desc.tfm)) { + rc = PTR_ERR(desc.tfm); + printk(KERN_ERR "Error allocating crypto context; " + "rc = [%d]\n", rc); goto out; } - } - if (password_s_ptr->session_key_encryption_key_bytes - < crypto_tfm_alg_min_keysize(tfm)) { - printk(KERN_WARNING "Session key encryption key is [%d] bytes; " - "minimum keysize for selected cipher is [%d] bytes.\n", - password_s_ptr->session_key_encryption_key_bytes, - crypto_tfm_alg_min_keysize(tfm)); - rc = -EINVAL; - goto out; + crypto_blkcipher_set_flags(desc.tfm, CRYPTO_TFM_REQ_WEAK_KEY); } if (tfm_mutex) mutex_lock(tfm_mutex); - crypto_cipher_setkey(tfm, password_s_ptr->session_key_encryption_key, - crypt_stat->key_size); + rc = crypto_blkcipher_setkey(desc.tfm, + password_s_ptr->session_key_encryption_key, + crypt_stat->key_size); + if (rc < 0) { + printk(KERN_ERR "Error setting key for crypto context\n"); + rc = -EINVAL; + goto out_free_tfm; + } /* TODO: virt_to_scatterlist */ encrypted_session_key = (char *)__get_free_page(GFP_KERNEL); if (!encrypted_session_key) { @@ -531,9 +540,12 @@ static int decrypt_session_key(struct ecryptfs_auth_tok *auth_tok, auth_tok->session_key.decrypted_key_size = auth_tok->session_key.encrypted_key_size; dst_sg[0].length = auth_tok->session_key.encrypted_key_size; - /* TODO: Handle error condition */ - crypto_cipher_decrypt(tfm, dst_sg, src_sg, - auth_tok->session_key.encrypted_key_size); + rc = crypto_blkcipher_decrypt(&desc, dst_sg, src_sg, + auth_tok->session_key.encrypted_key_size); + if (rc) { + printk(KERN_ERR "Error decrypting; rc = [%d]\n", rc); + goto out_free_memory; + } auth_tok->session_key.decrypted_key_size = auth_tok->session_key.encrypted_key_size; memcpy(auth_tok->session_key.decrypted_key, session_key, @@ -546,6 +558,7 @@ static int decrypt_session_key(struct ecryptfs_auth_tok *auth_tok, if (ecryptfs_verbosity > 0) ecryptfs_dump_hex(crypt_stat->key, crypt_stat->key_size); +out_free_memory: memset(encrypted_session_key, 0, PAGE_CACHE_SIZE); free_page((unsigned long)encrypted_session_key); memset(session_key, 0, PAGE_CACHE_SIZE); @@ -554,7 +567,7 @@ out_free_tfm: if (tfm_mutex) mutex_unlock(tfm_mutex); else - crypto_free_tfm(tfm); + crypto_free_blkcipher(desc.tfm); out: return rc; } @@ -803,19 +816,21 @@ write_tag_3_packet(char *dest, size_t max, struct ecryptfs_auth_tok *auth_tok, struct ecryptfs_crypt_stat *crypt_stat, struct ecryptfs_key_record *key_rec, size_t *packet_size) { - int rc = 0; - size_t i; size_t signature_is_valid = 0; size_t encrypted_session_key_valid = 0; char session_key_encryption_key[ECRYPTFS_MAX_KEY_BYTES]; struct scatterlist dest_sg[2]; struct scatterlist src_sg[2]; - struct crypto_tfm *tfm = NULL; struct mutex *tfm_mutex = NULL; size_t key_rec_size; size_t packet_size_length; size_t cipher_code; + struct blkcipher_desc desc = { + .tfm = NULL, + .flags = CRYPTO_TFM_REQ_MAY_SLEEP + }; + int rc = 0; (*packet_size) = 0; /* Check for a valid signature on the auth_tok */ @@ -882,33 +897,48 @@ write_tag_3_packet(char *dest, size_t max, struct ecryptfs_auth_tok *auth_tok, if (!strcmp(crypt_stat->cipher, crypt_stat->mount_crypt_stat->global_default_cipher_name) && crypt_stat->mount_crypt_stat->global_key_tfm) { - tfm = crypt_stat->mount_crypt_stat->global_key_tfm; + desc.tfm = crypt_stat->mount_crypt_stat->global_key_tfm; tfm_mutex = &crypt_stat->mount_crypt_stat->global_key_tfm_mutex; - } else - tfm = crypto_alloc_tfm(crypt_stat->cipher, 0); - if (!tfm) { - ecryptfs_printk(KERN_ERR, "Could not initialize crypto " - "context for cipher [%s]\n", - crypt_stat->cipher); - rc = -EINVAL; - goto out; + } else { + char *full_alg_name; + + rc = ecryptfs_crypto_api_algify_cipher_name(&full_alg_name, + crypt_stat->cipher, + "ecb"); + if (rc) + goto out; + desc.tfm = crypto_alloc_blkcipher(full_alg_name, 0, + CRYPTO_ALG_ASYNC); + kfree(full_alg_name); + if (IS_ERR(desc.tfm)) { + rc = PTR_ERR(desc.tfm); + ecryptfs_printk(KERN_ERR, "Could not initialize crypto " + "context for cipher [%s]; rc = [%d]\n", + crypt_stat->cipher, rc); + goto out; + } + crypto_blkcipher_set_flags(desc.tfm, CRYPTO_TFM_REQ_WEAK_KEY); } if (tfm_mutex) mutex_lock(tfm_mutex); - rc = crypto_cipher_setkey(tfm, session_key_encryption_key, - crypt_stat->key_size); + rc = crypto_blkcipher_setkey(desc.tfm, session_key_encryption_key, + crypt_stat->key_size); if (rc < 0) { if (tfm_mutex) mutex_unlock(tfm_mutex); ecryptfs_printk(KERN_ERR, "Error setting key for crypto " - "context\n"); + "context; rc = [%d]\n", rc); goto out; } rc = 0; ecryptfs_printk(KERN_DEBUG, "Encrypting [%d] bytes of the key\n", crypt_stat->key_size); - crypto_cipher_encrypt(tfm, dest_sg, src_sg, - (*key_rec).enc_key_size); + rc = crypto_blkcipher_encrypt(&desc, dest_sg, src_sg, + (*key_rec).enc_key_size); + if (rc) { + printk(KERN_ERR "Error encrypting; rc = [%d]\n", rc); + goto out; + } if (tfm_mutex) mutex_unlock(tfm_mutex); ecryptfs_printk(KERN_DEBUG, "This should be the encrypted key:\n"); @@ -971,8 +1001,8 @@ encrypted_session_key_set: (*key_rec).enc_key_size); (*packet_size) += (*key_rec).enc_key_size; out: - if (tfm && !tfm_mutex) - crypto_free_tfm(tfm); + if (desc.tfm && !tfm_mutex) + crypto_free_blkcipher(desc.tfm); if (rc) (*packet_size) = 0; return rc; diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c index 7a11b8ae6644..a78d87d14baf 100644 --- a/fs/ecryptfs/main.c +++ b/fs/ecryptfs/main.c @@ -104,10 +104,7 @@ int ecryptfs_interpose(struct dentry *lower_dentry, struct dentry *dentry, inode->i_op = &ecryptfs_dir_iops; if (S_ISDIR(lower_inode->i_mode)) inode->i_fop = &ecryptfs_dir_fops; - /* TODO: Is there a better way to identify if the inode is - * special? */ - if (S_ISBLK(lower_inode->i_mode) || S_ISCHR(lower_inode->i_mode) || - S_ISFIFO(lower_inode->i_mode) || S_ISSOCK(lower_inode->i_mode)) + if (special_file(lower_inode->i_mode)) init_special_inode(inode, lower_inode->i_mode, lower_inode->i_rdev); dentry->d_op = &ecryptfs_dops; @@ -211,7 +208,6 @@ static int ecryptfs_parse_options(struct super_block *sb, char *options) char *cipher_name_dst; char *cipher_name_src; char *cipher_key_bytes_src; - struct crypto_tfm *tmp_tfm; int cipher_name_len; if (!options) { @@ -308,25 +304,19 @@ static int ecryptfs_parse_options(struct super_block *sb, char *options) = '\0'; } if (!cipher_key_bytes_set) { - mount_crypt_stat->global_default_cipher_key_size = - ECRYPTFS_DEFAULT_KEY_BYTES; - ecryptfs_printk(KERN_DEBUG, "Cipher key size was not " - "specified. Defaulting to [%d]\n", - mount_crypt_stat-> - global_default_cipher_key_size); + mount_crypt_stat->global_default_cipher_key_size = 0; } rc = ecryptfs_process_cipher( - &tmp_tfm, &mount_crypt_stat->global_key_tfm, mount_crypt_stat->global_default_cipher_name, - mount_crypt_stat->global_default_cipher_key_size); - if (tmp_tfm) - crypto_free_tfm(tmp_tfm); + &mount_crypt_stat->global_default_cipher_key_size); if (rc) { printk(KERN_ERR "Error attempting to initialize cipher [%s] " "with key size [%Zd] bytes; rc = [%d]\n", mount_crypt_stat->global_default_cipher_name, mount_crypt_stat->global_default_cipher_key_size, rc); + mount_crypt_stat->global_key_tfm = NULL; + mount_crypt_stat->global_auth_tok_key = NULL; rc = -EINVAL; goto out; } diff --git a/fs/ecryptfs/super.c b/fs/ecryptfs/super.c index c337c0410fb1..825757ae4867 100644 --- a/fs/ecryptfs/super.c +++ b/fs/ecryptfs/super.c @@ -138,23 +138,6 @@ static void ecryptfs_clear_inode(struct inode *inode) } /** - * ecryptfs_umount_begin - * - * Called in do_umount(). - */ -static void ecryptfs_umount_begin(struct vfsmount *vfsmnt, int flags) -{ - struct vfsmount *lower_mnt = - ecryptfs_dentry_to_lower_mnt(vfsmnt->mnt_sb->s_root); - struct super_block *lower_sb; - - mntput(lower_mnt); - lower_sb = lower_mnt->mnt_sb; - if (lower_sb->s_op->umount_begin) - lower_sb->s_op->umount_begin(lower_mnt, flags); -} - -/** * ecryptfs_show_options * * Prints the directory we are currently mounted over. @@ -193,6 +176,5 @@ struct super_operations ecryptfs_sops = { .statfs = ecryptfs_statfs, .remount_fs = NULL, .clear_inode = ecryptfs_clear_inode, - .umount_begin = ecryptfs_umount_begin, .show_options = ecryptfs_show_options }; diff --git a/fs/eventpoll.c b/fs/eventpoll.c index 557d5b614fae..ae228ec54e94 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -105,6 +105,8 @@ /* Maximum msec timeout value storeable in a long int */ #define EP_MAX_MSTIMEO min(1000ULL * MAX_SCHEDULE_TIMEOUT / HZ, (LONG_MAX - 999ULL) / HZ) +#define EP_MAX_EVENTS (INT_MAX / sizeof(struct epoll_event)) + struct epoll_filefd { struct file *file; @@ -497,7 +499,7 @@ void eventpoll_release_file(struct file *file) */ asmlinkage long sys_epoll_create(int size) { - int error, fd; + int error, fd = -1; struct eventpoll *ep; struct inode *inode; struct file *file; @@ -640,7 +642,6 @@ eexit_1: return error; } -#define MAX_EVENTS (INT_MAX / sizeof(struct epoll_event)) /* * Implement the event wait interface for the eventpoll file. It is the kernel @@ -657,7 +658,7 @@ asmlinkage long sys_epoll_wait(int epfd, struct epoll_event __user *events, current, epfd, events, maxevents, timeout)); /* The maximum number of event must be greater than zero */ - if (maxevents <= 0 || maxevents > MAX_EVENTS) + if (maxevents <= 0 || maxevents > EP_MAX_EVENTS) return -EINVAL; /* Verify that the area passed by the user is writeable */ @@ -699,6 +700,55 @@ eexit_1: } +#ifdef TIF_RESTORE_SIGMASK + +/* + * Implement the event wait interface for the eventpoll file. It is the kernel + * part of the user space epoll_pwait(2). + */ +asmlinkage long sys_epoll_pwait(int epfd, struct epoll_event __user *events, + int maxevents, int timeout, const sigset_t __user *sigmask, + size_t sigsetsize) +{ + int error; + sigset_t ksigmask, sigsaved; + + /* + * If the caller wants a certain signal mask to be set during the wait, + * we apply it here. + */ + if (sigmask) { + if (sigsetsize != sizeof(sigset_t)) + return -EINVAL; + if (copy_from_user(&ksigmask, sigmask, sizeof(ksigmask))) + return -EFAULT; + sigdelsetmask(&ksigmask, sigmask(SIGKILL) | sigmask(SIGSTOP)); + sigprocmask(SIG_SETMASK, &ksigmask, &sigsaved); + } + + error = sys_epoll_wait(epfd, events, maxevents, timeout); + + /* + * If we changed the signal mask, we need to restore the original one. + * In case we've got a signal while waiting, we do not restore the + * signal mask yet, and we allow do_signal() to deliver the signal on + * the way back to userspace, before the signal mask is restored. + */ + if (sigmask) { + if (error == -EINTR) { + memcpy(¤t->saved_sigmask, &sigsaved, + sizeof(sigsaved)); + set_thread_flag(TIF_RESTORE_SIGMASK); + } else + sigprocmask(SIG_SETMASK, &sigsaved, NULL); + } + + return error; +} + +#endif /* #ifdef TIF_RESTORE_SIGMASK */ + + /* * Creates the file descriptor to be used by the epoll interface. */ diff --git a/fs/ext2/super.c b/fs/ext2/super.c index 513cd421ac0b..d8b9abd95d07 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -364,7 +364,6 @@ static int parse_options (char * options, { char * p; substring_t args[MAX_OPT_ARGS]; - unsigned long kind = EXT2_MOUNT_ERRORS_CONT; int option; if (!options) @@ -404,13 +403,19 @@ static int parse_options (char * options, /* *sb_block = match_int(&args[0]); */ break; case Opt_err_panic: - kind = EXT2_MOUNT_ERRORS_PANIC; + clear_opt (sbi->s_mount_opt, ERRORS_CONT); + clear_opt (sbi->s_mount_opt, ERRORS_RO); + set_opt (sbi->s_mount_opt, ERRORS_PANIC); break; case Opt_err_ro: - kind = EXT2_MOUNT_ERRORS_RO; + clear_opt (sbi->s_mount_opt, ERRORS_CONT); + clear_opt (sbi->s_mount_opt, ERRORS_PANIC); + set_opt (sbi->s_mount_opt, ERRORS_RO); break; case Opt_err_cont: - kind = EXT2_MOUNT_ERRORS_CONT; + clear_opt (sbi->s_mount_opt, ERRORS_RO); + clear_opt (sbi->s_mount_opt, ERRORS_PANIC); + set_opt (sbi->s_mount_opt, ERRORS_CONT); break; case Opt_nouid32: set_opt (sbi->s_mount_opt, NO_UID32); @@ -489,7 +494,6 @@ static int parse_options (char * options, return 0; } } - sbi->s_mount_opt |= kind; return 1; } @@ -715,6 +719,8 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent) set_opt(sbi->s_mount_opt, ERRORS_PANIC); else if (le16_to_cpu(sbi->s_es->s_errors) == EXT2_ERRORS_RO) set_opt(sbi->s_mount_opt, ERRORS_RO); + else + set_opt(sbi->s_mount_opt, ERRORS_CONT); sbi->s_resuid = le16_to_cpu(es->s_def_resuid); sbi->s_resgid = le16_to_cpu(es->s_def_resgid); diff --git a/fs/ext3/super.c b/fs/ext3/super.c index 8bfd56ef18ca..afc2d4f42d77 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -1470,6 +1470,8 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent) set_opt(sbi->s_mount_opt, ERRORS_PANIC); else if (le16_to_cpu(sbi->s_es->s_errors) == EXT3_ERRORS_RO) set_opt(sbi->s_mount_opt, ERRORS_RO); + else + set_opt(sbi->s_mount_opt, ERRORS_CONT); sbi->s_resuid = le16_to_cpu(es->s_def_resuid); sbi->s_resgid = le16_to_cpu(es->s_def_resgid); diff --git a/fs/ext4/Makefile b/fs/ext4/Makefile new file mode 100644 index 000000000000..a6acb96ebeb9 --- /dev/null +++ b/fs/ext4/Makefile @@ -0,0 +1,12 @@ +# +# Makefile for the linux ext4-filesystem routines. +# + +obj-$(CONFIG_EXT4DEV_FS) += ext4dev.o + +ext4dev-y := balloc.o bitmap.o dir.o file.o fsync.o ialloc.o inode.o \ + ioctl.o namei.o super.o symlink.o hash.o resize.o extents.o + +ext4dev-$(CONFIG_EXT4DEV_FS_XATTR) += xattr.o xattr_user.o xattr_trusted.o +ext4dev-$(CONFIG_EXT4DEV_FS_POSIX_ACL) += acl.o +ext4dev-$(CONFIG_EXT4DEV_FS_SECURITY) += xattr_security.o diff --git a/fs/ext4/acl.c b/fs/ext4/acl.c new file mode 100644 index 000000000000..9e882546d91a --- /dev/null +++ b/fs/ext4/acl.c @@ -0,0 +1,551 @@ +/* + * linux/fs/ext4/acl.c + * + * Copyright (C) 2001-2003 Andreas Gruenbacher, <agruen@suse.de> + */ + +#include <linux/init.h> +#include <linux/sched.h> +#include <linux/slab.h> +#include <linux/capability.h> +#include <linux/fs.h> +#include <linux/ext4_jbd2.h> +#include <linux/ext4_fs.h> +#include "xattr.h" +#include "acl.h" + +/* + * Convert from filesystem to in-memory representation. + */ +static struct posix_acl * +ext4_acl_from_disk(const void *value, size_t size) +{ + const char *end = (char *)value + size; + int n, count; + struct posix_acl *acl; + + if (!value) + return NULL; + if (size < sizeof(ext4_acl_header)) + return ERR_PTR(-EINVAL); + if (((ext4_acl_header *)value)->a_version != + cpu_to_le32(EXT4_ACL_VERSION)) + return ERR_PTR(-EINVAL); + value = (char *)value + sizeof(ext4_acl_header); + count = ext4_acl_count(size); + if (count < 0) + return ERR_PTR(-EINVAL); + if (count == 0) + return NULL; + acl = posix_acl_alloc(count, GFP_KERNEL); + if (!acl) + return ERR_PTR(-ENOMEM); + for (n=0; n < count; n++) { + ext4_acl_entry *entry = + (ext4_acl_entry *)value; + if ((char *)value + sizeof(ext4_acl_entry_short) > end) + goto fail; + acl->a_entries[n].e_tag = le16_to_cpu(entry->e_tag); + acl->a_entries[n].e_perm = le16_to_cpu(entry->e_perm); + switch(acl->a_entries[n].e_tag) { + case ACL_USER_OBJ: + case ACL_GROUP_OBJ: + case ACL_MASK: + case ACL_OTHER: + value = (char *)value + + sizeof(ext4_acl_entry_short); + acl->a_entries[n].e_id = ACL_UNDEFINED_ID; + break; + + case ACL_USER: + case ACL_GROUP: + value = (char *)value + sizeof(ext4_acl_entry); + if ((char *)value > end) + goto fail; + acl->a_entries[n].e_id = + le32_to_cpu(entry->e_id); + break; + + default: + goto fail; + } + } + if (value != end) + goto fail; + return acl; + +fail: + posix_acl_release(acl); + return ERR_PTR(-EINVAL); +} + +/* + * Convert from in-memory to filesystem representation. + */ +static void * +ext4_acl_to_disk(const struct posix_acl *acl, size_t *size) +{ + ext4_acl_header *ext_acl; + char *e; + size_t n; + + *size = ext4_acl_size(acl->a_count); + ext_acl = kmalloc(sizeof(ext4_acl_header) + acl->a_count * + sizeof(ext4_acl_entry), GFP_KERNEL); + if (!ext_acl) + return ERR_PTR(-ENOMEM); + ext_acl->a_version = cpu_to_le32(EXT4_ACL_VERSION); + e = (char *)ext_acl + sizeof(ext4_acl_header); + for (n=0; n < acl->a_count; n++) { + ext4_acl_entry *entry = (ext4_acl_entry *)e; + entry->e_tag = cpu_to_le16(acl->a_entries[n].e_tag); + entry->e_perm = cpu_to_le16(acl->a_entries[n].e_perm); + switch(acl->a_entries[n].e_tag) { + case ACL_USER: + case ACL_GROUP: + entry->e_id = + cpu_to_le32(acl->a_entries[n].e_id); + e += sizeof(ext4_acl_entry); + break; + + case ACL_USER_OBJ: + case ACL_GROUP_OBJ: + case ACL_MASK: + case ACL_OTHER: + e += sizeof(ext4_acl_entry_short); + break; + + default: + goto fail; + } + } + return (char *)ext_acl; + +fail: + kfree(ext_acl); + return ERR_PTR(-EINVAL); +} + +static inline struct posix_acl * +ext4_iget_acl(struct inode *inode, struct posix_acl **i_acl) +{ + struct posix_acl *acl = EXT4_ACL_NOT_CACHED; + + spin_lock(&inode->i_lock); + if (*i_acl != EXT4_ACL_NOT_CACHED) + acl = posix_acl_dup(*i_acl); + spin_unlock(&inode->i_lock); + + return acl; +} + +static inline void +ext4_iset_acl(struct inode *inode, struct posix_acl **i_acl, + struct posix_acl *acl) +{ + spin_lock(&inode->i_lock); + if (*i_acl != EXT4_ACL_NOT_CACHED) + posix_acl_release(*i_acl); + *i_acl = posix_acl_dup(acl); + spin_unlock(&inode->i_lock); +} + +/* + * Inode operation get_posix_acl(). + * + * inode->i_mutex: don't care + */ +static struct posix_acl * +ext4_get_acl(struct inode *inode, int type) +{ + struct ext4_inode_info *ei = EXT4_I(inode); + int name_index; + char *value = NULL; + struct posix_acl *acl; + int retval; + + if (!test_opt(inode->i_sb, POSIX_ACL)) + return NULL; + + switch(type) { + case ACL_TYPE_ACCESS: + acl = ext4_iget_acl(inode, &ei->i_acl); + if (acl != EXT4_ACL_NOT_CACHED) + return acl; + name_index = EXT4_XATTR_INDEX_POSIX_ACL_ACCESS; + break; + + case ACL_TYPE_DEFAULT: + acl = ext4_iget_acl(inode, &ei->i_default_acl); + if (acl != EXT4_ACL_NOT_CACHED) + return acl; + name_index = EXT4_XATTR_INDEX_POSIX_ACL_DEFAULT; + break; + + default: + return ERR_PTR(-EINVAL); + } + retval = ext4_xattr_get(inode, name_index, "", NULL, 0); + if (retval > 0) { + value = kmalloc(retval, GFP_KERNEL); + if (!value) + return ERR_PTR(-ENOMEM); + retval = ext4_xattr_get(inode, name_index, "", value, retval); + } + if (retval > 0) + acl = ext4_acl_from_disk(value, retval); + else if (retval == -ENODATA || retval == -ENOSYS) + acl = NULL; + else + acl = ERR_PTR(retval); + kfree(value); + + if (!IS_ERR(acl)) { + switch(type) { + case ACL_TYPE_ACCESS: + ext4_iset_acl(inode, &ei->i_acl, acl); + break; + + case ACL_TYPE_DEFAULT: + ext4_iset_acl(inode, &ei->i_default_acl, acl); + break; + } + } + return acl; +} + +/* + * Set the access or default ACL of an inode. + * + * inode->i_mutex: down unless called from ext4_new_inode + */ +static int +ext4_set_acl(handle_t *handle, struct inode *inode, int type, + struct posix_acl *acl) +{ + struct ext4_inode_info *ei = EXT4_I(inode); + int name_index; + void *value = NULL; + size_t size = 0; + int error; + + if (S_ISLNK(inode->i_mode)) + return -EOPNOTSUPP; + + switch(type) { + case ACL_TYPE_ACCESS: + name_index = EXT4_XATTR_INDEX_POSIX_ACL_ACCESS; + if (acl) { + mode_t mode = inode->i_mode; + error = posix_acl_equiv_mode(acl, &mode); + if (error < 0) + return error; + else { + inode->i_mode = mode; + ext4_mark_inode_dirty(handle, inode); + if (error == 0) + acl = NULL; + } + } + break; + + case ACL_TYPE_DEFAULT: + name_index = EXT4_XATTR_INDEX_POSIX_ACL_DEFAULT; + if (!S_ISDIR(inode->i_mode)) + return acl ? -EACCES : 0; + break; + + default: + return -EINVAL; + } + if (acl) { + value = ext4_acl_to_disk(acl, &size); + if (IS_ERR(value)) + return (int)PTR_ERR(value); + } + + error = ext4_xattr_set_handle(handle, inode, name_index, "", + value, size, 0); + + kfree(value); + if (!error) { + switch(type) { + case ACL_TYPE_ACCESS: + ext4_iset_acl(inode, &ei->i_acl, acl); + break; + + case ACL_TYPE_DEFAULT: + ext4_iset_acl(inode, &ei->i_default_acl, acl); + break; + } + } + return error; +} + +static int +ext4_check_acl(struct inode *inode, int mask) +{ + struct posix_acl *acl = ext4_get_acl(inode, ACL_TYPE_ACCESS); + + if (IS_ERR(acl)) + return PTR_ERR(acl); + if (acl) { + int error = posix_acl_permission(inode, acl, mask); + posix_acl_release(acl); + return error; + } + + return -EAGAIN; +} + +int +ext4_permission(struct inode *inode, int mask, struct nameidata *nd) +{ + return generic_permission(inode, mask, ext4_check_acl); +} + +/* + * Initialize the ACLs of a new inode. Called from ext4_new_inode. + * + * dir->i_mutex: down + * inode->i_mutex: up (access to inode is still exclusive) + */ +int +ext4_init_acl(handle_t *handle, struct inode *inode, struct inode *dir) +{ + struct posix_acl *acl = NULL; + int error = 0; + + if (!S_ISLNK(inode->i_mode)) { + if (test_opt(dir->i_sb, POSIX_ACL)) { + acl = ext4_get_acl(dir, ACL_TYPE_DEFAULT); + if (IS_ERR(acl)) + return PTR_ERR(acl); + } + if (!acl) + inode->i_mode &= ~current->fs->umask; + } + if (test_opt(inode->i_sb, POSIX_ACL) && acl) { + struct posix_acl *clone; + mode_t mode; + + if (S_ISDIR(inode->i_mode)) { + error = ext4_set_acl(handle, inode, + ACL_TYPE_DEFAULT, acl); + if (error) + goto cleanup; + } + clone = posix_acl_clone(acl, GFP_KERNEL); + error = -ENOMEM; + if (!clone) + goto cleanup; + + mode = inode->i_mode; + error = posix_acl_create_masq(clone, &mode); + if (error >= 0) { + inode->i_mode = mode; + if (error > 0) { + /* This is an extended ACL */ + error = ext4_set_acl(handle, inode, + ACL_TYPE_ACCESS, clone); + } + } + posix_acl_release(clone); + } +cleanup: + posix_acl_release(acl); + return error; +} + +/* + * Does chmod for an inode that may have an Access Control List. The + * inode->i_mode field must be updated to the desired value by the caller + * before calling this function. + * Returns 0 on success, or a negative error number. + * + * We change the ACL rather than storing some ACL entries in the file + * mode permission bits (which would be more efficient), because that + * would break once additional permissions (like ACL_APPEND, ACL_DELETE + * for directories) are added. There are no more bits available in the + * file mode. + * + * inode->i_mutex: down + */ +int +ext4_acl_chmod(struct inode *inode) +{ + struct posix_acl *acl, *clone; + int error; + + if (S_ISLNK(inode->i_mode)) + return -EOPNOTSUPP; + if (!test_opt(inode->i_sb, POSIX_ACL)) + return 0; + acl = ext4_get_acl(inode, ACL_TYPE_ACCESS); + if (IS_ERR(acl) || !acl) + return PTR_ERR(acl); + clone = posix_acl_clone(acl, GFP_KERNEL); + posix_acl_release(acl); + if (!clone) + return -ENOMEM; + error = posix_acl_chmod_masq(clone, inode->i_mode); + if (!error) { + handle_t *handle; + int retries = 0; + + retry: + handle = ext4_journal_start(inode, + EXT4_DATA_TRANS_BLOCKS(inode->i_sb)); + if (IS_ERR(handle)) { + error = PTR_ERR(handle); + ext4_std_error(inode->i_sb, error); + goto out; + } + error = ext4_set_acl(handle, inode, ACL_TYPE_ACCESS, clone); + ext4_journal_stop(handle); + if (error == -ENOSPC && + ext4_should_retry_alloc(inode->i_sb, &retries)) + goto retry; + } +out: + posix_acl_release(clone); + return error; +} + +/* + * Extended attribute handlers + */ +static size_t +ext4_xattr_list_acl_access(struct inode *inode, char *list, size_t list_len, + const char *name, size_t name_len) +{ + const size_t size = sizeof(POSIX_ACL_XATTR_ACCESS); + + if (!test_opt(inode->i_sb, POSIX_ACL)) + return 0; + if (list && size <= list_len) + memcpy(list, POSIX_ACL_XATTR_ACCESS, size); + return size; +} + +static size_t +ext4_xattr_list_acl_default(struct inode *inode, char *list, size_t list_len, + const char *name, size_t name_len) +{ + const size_t size = sizeof(POSIX_ACL_XATTR_DEFAULT); + + if (!test_opt(inode->i_sb, POSIX_ACL)) + return 0; + if (list && size <= list_len) + memcpy(list, POSIX_ACL_XATTR_DEFAULT, size); + return size; +} + +static int +ext4_xattr_get_acl(struct inode *inode, int type, void *buffer, size_t size) +{ + struct posix_acl *acl; + int error; + + if (!test_opt(inode->i_sb, POSIX_ACL)) + return -EOPNOTSUPP; + + acl = ext4_get_acl(inode, type); + if (IS_ERR(acl)) + return PTR_ERR(acl); + if (acl == NULL) + return -ENODATA; + error = posix_acl_to_xattr(acl, buffer, size); + posix_acl_release(acl); + + return error; +} + +static int +ext4_xattr_get_acl_access(struct inode *inode, const char *name, + void *buffer, size_t size) +{ + if (strcmp(name, "") != 0) + return -EINVAL; + return ext4_xattr_get_acl(inode, ACL_TYPE_ACCESS, buffer, size); +} + +static int +ext4_xattr_get_acl_default(struct inode *inode, const char *name, + void *buffer, size_t size) +{ + if (strcmp(name, "") != 0) + return -EINVAL; + return ext4_xattr_get_acl(inode, ACL_TYPE_DEFAULT, buffer, size); +} + +static int +ext4_xattr_set_acl(struct inode *inode, int type, const void *value, + size_t size) +{ + handle_t *handle; + struct posix_acl *acl; + int error, retries = 0; + + if (!test_opt(inode->i_sb, POSIX_ACL)) + return -EOPNOTSUPP; + if ((current->fsuid != inode->i_uid) && !capable(CAP_FOWNER)) + return -EPERM; + + if (value) { + acl = posix_acl_from_xattr(value, size); + if (IS_ERR(acl)) + return PTR_ERR(acl); + else if (acl) { + error = posix_acl_valid(acl); + if (error) + goto release_and_out; + } + } else + acl = NULL; + +retry: + handle = ext4_journal_start(inode, EXT4_DATA_TRANS_BLOCKS(inode->i_sb)); + if (IS_ERR(handle)) + return PTR_ERR(handle); + error = ext4_set_acl(handle, inode, type, acl); + ext4_journal_stop(handle); + if (error == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries)) + goto retry; + +release_and_out: + posix_acl_release(acl); + return error; +} + +static int +ext4_xattr_set_acl_access(struct inode *inode, const char *name, + const void *value, size_t size, int flags) +{ + if (strcmp(name, "") != 0) + return -EINVAL; + return ext4_xattr_set_acl(inode, ACL_TYPE_ACCESS, value, size); +} + +static int +ext4_xattr_set_acl_default(struct inode *inode, const char *name, + const void *value, size_t size, int flags) +{ + if (strcmp(name, "") != 0) + return -EINVAL; + return ext4_xattr_set_acl(inode, ACL_TYPE_DEFAULT, value, size); +} + +struct xattr_handler ext4_xattr_acl_access_handler = { + .prefix = POSIX_ACL_XATTR_ACCESS, + .list = ext4_xattr_list_acl_access, + .get = ext4_xattr_get_acl_access, + .set = ext4_xattr_set_acl_access, +}; + +struct xattr_handler ext4_xattr_acl_default_handler = { + .prefix = POSIX_ACL_XATTR_DEFAULT, + .list = ext4_xattr_list_acl_default, + .get = ext4_xattr_get_acl_default, + .set = ext4_xattr_set_acl_default, +}; diff --git a/fs/ext4/acl.h b/fs/ext4/acl.h new file mode 100644 index 000000000000..26a5c1abf147 --- /dev/null +++ b/fs/ext4/acl.h @@ -0,0 +1,81 @@ +/* + File: fs/ext4/acl.h + + (C) 2001 Andreas Gruenbacher, <a.gruenbacher@computer.org> +*/ + +#include <linux/posix_acl_xattr.h> + +#define EXT4_ACL_VERSION 0x0001 + +typedef struct { + __le16 e_tag; + __le16 e_perm; + __le32 e_id; +} ext4_acl_entry; + +typedef struct { + __le16 e_tag; + __le16 e_perm; +} ext4_acl_entry_short; + +typedef struct { + __le32 a_version; +} ext4_acl_header; + +static inline size_t ext4_acl_size(int count) +{ + if (count <= 4) { + return sizeof(ext4_acl_header) + + count * sizeof(ext4_acl_entry_short); + } else { + return sizeof(ext4_acl_header) + + 4 * sizeof(ext4_acl_entry_short) + + (count - 4) * sizeof(ext4_acl_entry); + } +} + +static inline int ext4_acl_count(size_t size) +{ + ssize_t s; + size -= sizeof(ext4_acl_header); + s = size - 4 * sizeof(ext4_acl_entry_short); + if (s < 0) { + if (size % sizeof(ext4_acl_entry_short)) + return -1; + return size / sizeof(ext4_acl_entry_short); + } else { + if (s % sizeof(ext4_acl_entry)) + return -1; + return s / sizeof(ext4_acl_entry) + 4; + } +} + +#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL + +/* Value for inode->u.ext4_i.i_acl and inode->u.ext4_i.i_default_acl + if the ACL has not been cached */ +#define EXT4_ACL_NOT_CACHED ((void *)-1) + +/* acl.c */ +extern int ext4_permission (struct inode *, int, struct nameidata *); +extern int ext4_acl_chmod (struct inode *); +extern int ext4_init_acl (handle_t *, struct inode *, struct inode *); + +#else /* CONFIG_EXT4DEV_FS_POSIX_ACL */ +#include <linux/sched.h> +#define ext4_permission NULL + +static inline int +ext4_acl_chmod(struct inode *inode) +{ + return 0; +} + +static inline int +ext4_init_acl(handle_t *handle, struct inode *inode, struct inode *dir) +{ + return 0; +} +#endif /* CONFIG_EXT4DEV_FS_POSIX_ACL */ + diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c new file mode 100644 index 000000000000..5d45582f9517 --- /dev/null +++ b/fs/ext4/balloc.c @@ -0,0 +1,1833 @@ +/* + * linux/fs/ext4/balloc.c + * + * Copyright (C) 1992, 1993, 1994, 1995 + * Remy Card (card@masi.ibp.fr) + * Laboratoire MASI - Institut Blaise Pascal + * Universite Pierre et Marie Curie (Paris VI) + * + * Enhanced block allocation by Stephen Tweedie (sct@redhat.com), 1993 + * Big-endian to little-endian byte-swapping/bitmaps by + * David S. Miller (davem@caip.rutgers.edu), 1995 + */ + +#include <linux/time.h> +#include <linux/capability.h> +#include <linux/fs.h> +#include <linux/jbd2.h> +#include <linux/ext4_fs.h> +#include <linux/ext4_jbd2.h> +#include <linux/quotaops.h> +#include <linux/buffer_head.h> + +/* + * balloc.c contains the blocks allocation and deallocation routines + */ + +/* + * Calculate the block group number and offset, given a block number + */ +void ext4_get_group_no_and_offset(struct super_block *sb, ext4_fsblk_t blocknr, + unsigned long *blockgrpp, ext4_grpblk_t *offsetp) +{ + struct ext4_super_block *es = EXT4_SB(sb)->s_es; + ext4_grpblk_t offset; + + blocknr = blocknr - le32_to_cpu(es->s_first_data_block); + offset = do_div(blocknr, EXT4_BLOCKS_PER_GROUP(sb)); + if (offsetp) + *offsetp = offset; + if (blockgrpp) + *blockgrpp = blocknr; + +} + +/* + * The free blocks are managed by bitmaps. A file system contains several + * blocks groups. Each group contains 1 bitmap block for blocks, 1 bitmap + * block for inodes, N blocks for the inode table and data blocks. + * + * The file system contains group descriptors which are located after the + * super block. Each descriptor contains the number of the bitmap block and + * the free blocks count in the block. The descriptors are loaded in memory + * when a file system is mounted (see ext4_read_super). + */ + + +#define in_range(b, first, len) ((b) >= (first) && (b) <= (first) + (len) - 1) + +/** + * ext4_get_group_desc() -- load group descriptor from disk + * @sb: super block + * @block_group: given block group + * @bh: pointer to the buffer head to store the block + * group descriptor + */ +struct ext4_group_desc * ext4_get_group_desc(struct super_block * sb, + unsigned int block_group, + struct buffer_head ** bh) +{ + unsigned long group_desc; + unsigned long offset; + struct ext4_group_desc * desc; + struct ext4_sb_info *sbi = EXT4_SB(sb); + + if (block_group >= sbi->s_groups_count) { + ext4_error (sb, "ext4_get_group_desc", + "block_group >= groups_count - " + "block_group = %d, groups_count = %lu", + block_group, sbi->s_groups_count); + + return NULL; + } + smp_rmb(); + + group_desc = block_group >> EXT4_DESC_PER_BLOCK_BITS(sb); + offset = block_group & (EXT4_DESC_PER_BLOCK(sb) - 1); + if (!sbi->s_group_desc[group_desc]) { + ext4_error (sb, "ext4_get_group_desc", + "Group descriptor not loaded - " + "block_group = %d, group_desc = %lu, desc = %lu", + block_group, group_desc, offset); + return NULL; + } + + desc = (struct ext4_group_desc *)( + (__u8 *)sbi->s_group_desc[group_desc]->b_data + + offset * EXT4_DESC_SIZE(sb)); + if (bh) + *bh = sbi->s_group_desc[group_desc]; + return desc; +} + +/** + * read_block_bitmap() + * @sb: super block + * @block_group: given block group + * + * Read the bitmap for a given block_group, reading into the specified + * slot in the superblock's bitmap cache. + * + * Return buffer_head on success or NULL in case of failure. + */ +static struct buffer_head * +read_block_bitmap(struct super_block *sb, unsigned int block_group) +{ + struct ext4_group_desc * desc; + struct buffer_head * bh = NULL; + + desc = ext4_get_group_desc (sb, block_group, NULL); + if (!desc) + goto error_out; + bh = sb_bread(sb, ext4_block_bitmap(sb, desc)); + if (!bh) + ext4_error (sb, "read_block_bitmap", + "Cannot read block bitmap - " + "block_group = %d, block_bitmap = %llu", + block_group, + ext4_block_bitmap(sb, desc)); +error_out: + return bh; +} +/* + * The reservation window structure operations + * -------------------------------------------- + * Operations include: + * dump, find, add, remove, is_empty, find_next_reservable_window, etc. + * + * We use a red-black tree to represent per-filesystem reservation + * windows. + * + */ + +/** + * __rsv_window_dump() -- Dump the filesystem block allocation reservation map + * @rb_root: root of per-filesystem reservation rb tree + * @verbose: verbose mode + * @fn: function which wishes to dump the reservation map + * + * If verbose is turned on, it will print the whole block reservation + * windows(start, end). Otherwise, it will only print out the "bad" windows, + * those windows that overlap with their immediate neighbors. + */ +#if 1 +static void __rsv_window_dump(struct rb_root *root, int verbose, + const char *fn) +{ + struct rb_node *n; + struct ext4_reserve_window_node *rsv, *prev; + int bad; + +restart: + n = rb_first(root); + bad = 0; + prev = NULL; + + printk("Block Allocation Reservation Windows Map (%s):\n", fn); + while (n) { + rsv = list_entry(n, struct ext4_reserve_window_node, rsv_node); + if (verbose) + printk("reservation window 0x%p " + "start: %llu, end: %llu\n", + rsv, rsv->rsv_start, rsv->rsv_end); + if (rsv->rsv_start && rsv->rsv_start >= rsv->rsv_end) { + printk("Bad reservation %p (start >= end)\n", + rsv); + bad = 1; + } + if (prev && prev->rsv_end >= rsv->rsv_start) { + printk("Bad reservation %p (prev->end >= start)\n", + rsv); + bad = 1; + } + if (bad) { + if (!verbose) { + printk("Restarting reservation walk in verbose mode\n"); + verbose = 1; + goto restart; + } + } + n = rb_next(n); + prev = rsv; + } + printk("Window map complete.\n"); + if (bad) + BUG(); +} +#define rsv_window_dump(root, verbose) \ + __rsv_window_dump((root), (verbose), __FUNCTION__) +#else +#define rsv_window_dump(root, verbose) do {} while (0) +#endif + +/** + * goal_in_my_reservation() + * @rsv: inode's reservation window + * @grp_goal: given goal block relative to the allocation block group + * @group: the current allocation block group + * @sb: filesystem super block + * + * Test if the given goal block (group relative) is within the file's + * own block reservation window range. + * + * If the reservation window is outside the goal allocation group, return 0; + * grp_goal (given goal block) could be -1, which means no specific + * goal block. In this case, always return 1. + * If the goal block is within the reservation window, return 1; + * otherwise, return 0; + */ +static int +goal_in_my_reservation(struct ext4_reserve_window *rsv, ext4_grpblk_t grp_goal, + unsigned int group, struct super_block * sb) +{ + ext4_fsblk_t group_first_block, group_last_block; + + group_first_block = ext4_group_first_block_no(sb, group); + group_last_block = group_first_block + (EXT4_BLOCKS_PER_GROUP(sb) - 1); + + if ((rsv->_rsv_start > group_last_block) || + (rsv->_rsv_end < group_first_block)) + return 0; + if ((grp_goal >= 0) && ((grp_goal + group_first_block < rsv->_rsv_start) + || (grp_goal + group_first_block > rsv->_rsv_end))) + return 0; + return 1; +} + +/** + * search_reserve_window() + * @rb_root: root of reservation tree + * @goal: target allocation block + * + * Find the reserved window which includes the goal, or the previous one + * if the goal is not in any window. + * Returns NULL if there are no windows or if all windows start after the goal. + */ +static struct ext4_reserve_window_node * +search_reserve_window(struct rb_root *root, ext4_fsblk_t goal) +{ + struct rb_node *n = root->rb_node; + struct ext4_reserve_window_node *rsv; + + if (!n) + return NULL; + + do { + rsv = rb_entry(n, struct ext4_reserve_window_node, rsv_node); + + if (goal < rsv->rsv_start) + n = n->rb_left; + else if (goal > rsv->rsv_end) + n = n->rb_right; + else + return rsv; + } while (n); + /* + * We've fallen off the end of the tree: the goal wasn't inside + * any particular node. OK, the previous node must be to one + * side of the interval containing the goal. If it's the RHS, + * we need to back up one. + */ + if (rsv->rsv_start > goal) { + n = rb_prev(&rsv->rsv_node); + rsv = rb_entry(n, struct ext4_reserve_window_node, rsv_node); + } + return rsv; +} + +/** + * ext4_rsv_window_add() -- Insert a window to the block reservation rb tree. + * @sb: super block + * @rsv: reservation window to add + * + * Must be called with rsv_lock hold. + */ +void ext4_rsv_window_add(struct super_block *sb, + struct ext4_reserve_window_node *rsv) +{ + struct rb_root *root = &EXT4_SB(sb)->s_rsv_window_root; + struct rb_node *node = &rsv->rsv_node; + ext4_fsblk_t start = rsv->rsv_start; + + struct rb_node ** p = &root->rb_node; + struct rb_node * parent = NULL; + struct ext4_reserve_window_node *this; + + while (*p) + { + parent = *p; + this = rb_entry(parent, struct ext4_reserve_window_node, rsv_node); + + if (start < this->rsv_start) + p = &(*p)->rb_left; + else if (start > this->rsv_end) + p = &(*p)->rb_right; + else { + rsv_window_dump(root, 1); + BUG(); + } + } + + rb_link_node(node, parent, p); + rb_insert_color(node, root); +} + +/** + * ext4_rsv_window_remove() -- unlink a window from the reservation rb tree + * @sb: super block + * @rsv: reservation window to remove + * + * Mark the block reservation window as not allocated, and unlink it + * from the filesystem reservation window rb tree. Must be called with + * rsv_lock hold. + */ +static void rsv_window_remove(struct super_block *sb, + struct ext4_reserve_window_node *rsv) +{ + rsv->rsv_start = EXT4_RESERVE_WINDOW_NOT_ALLOCATED; + rsv->rsv_end = EXT4_RESERVE_WINDOW_NOT_ALLOCATED; + rsv->rsv_alloc_hit = 0; + rb_erase(&rsv->rsv_node, &EXT4_SB(sb)->s_rsv_window_root); +} + +/* + * rsv_is_empty() -- Check if the reservation window is allocated. + * @rsv: given reservation window to check + * + * returns 1 if the end block is EXT4_RESERVE_WINDOW_NOT_ALLOCATED. + */ +static inline int rsv_is_empty(struct ext4_reserve_window *rsv) +{ + /* a valid reservation end block could not be 0 */ + return rsv->_rsv_end == EXT4_RESERVE_WINDOW_NOT_ALLOCATED; +} + +/** + * ext4_init_block_alloc_info() + * @inode: file inode structure + * + * Allocate and initialize the reservation window structure, and + * link the window to the ext4 inode structure at last + * + * The reservation window structure is only dynamically allocated + * and linked to ext4 inode the first time the open file + * needs a new block. So, before every ext4_new_block(s) call, for + * regular files, we should check whether the reservation window + * structure exists or not. In the latter case, this function is called. + * Fail to do so will result in block reservation being turned off for that + * open file. + * + * This function is called from ext4_get_blocks_handle(), also called + * when setting the reservation window size through ioctl before the file + * is open for write (needs block allocation). + * + * Needs truncate_mutex protection prior to call this function. + */ +void ext4_init_block_alloc_info(struct inode *inode) +{ + struct ext4_inode_info *ei = EXT4_I(inode); + struct ext4_block_alloc_info *block_i = ei->i_block_alloc_info; + struct super_block *sb = inode->i_sb; + + block_i = kmalloc(sizeof(*block_i), GFP_NOFS); + if (block_i) { + struct ext4_reserve_window_node *rsv = &block_i->rsv_window_node; + + rsv->rsv_start = EXT4_RESERVE_WINDOW_NOT_ALLOCATED; + rsv->rsv_end = EXT4_RESERVE_WINDOW_NOT_ALLOCATED; + + /* + * if filesystem is mounted with NORESERVATION, the goal + * reservation window size is set to zero to indicate + * block reservation is off + */ + if (!test_opt(sb, RESERVATION)) + rsv->rsv_goal_size = 0; + else + rsv->rsv_goal_size = EXT4_DEFAULT_RESERVE_BLOCKS; + rsv->rsv_alloc_hit = 0; + block_i->last_alloc_logical_block = 0; + block_i->last_alloc_physical_block = 0; + } + ei->i_block_alloc_info = block_i; +} + +/** + * ext4_discard_reservation() + * @inode: inode + * + * Discard(free) block reservation window on last file close, or truncate + * or at last iput(). + * + * It is being called in three cases: + * ext4_release_file(): last writer close the file + * ext4_clear_inode(): last iput(), when nobody link to this file. + * ext4_truncate(): when the block indirect map is about to change. + * + */ +void ext4_discard_reservation(struct inode *inode) +{ + struct ext4_inode_info *ei = EXT4_I(inode); + struct ext4_block_alloc_info *block_i = ei->i_block_alloc_info; + struct ext4_reserve_window_node *rsv; + spinlock_t *rsv_lock = &EXT4_SB(inode->i_sb)->s_rsv_window_lock; + + if (!block_i) + return; + + rsv = &block_i->rsv_window_node; + if (!rsv_is_empty(&rsv->rsv_window)) { + spin_lock(rsv_lock); + if (!rsv_is_empty(&rsv->rsv_window)) + rsv_window_remove(inode->i_sb, rsv); + spin_unlock(rsv_lock); + } +} + +/** + * ext4_free_blocks_sb() -- Free given blocks and update quota + * @handle: handle to this transaction + * @sb: super block + * @block: start physcial block to free + * @count: number of blocks to free + * @pdquot_freed_blocks: pointer to quota + */ +void ext4_free_blocks_sb(handle_t *handle, struct super_block *sb, + ext4_fsblk_t block, unsigned long count, + unsigned long *pdquot_freed_blocks) +{ + struct buffer_head *bitmap_bh = NULL; + struct buffer_head *gd_bh; + unsigned long block_group; + ext4_grpblk_t bit; + unsigned long i; + unsigned long overflow; + struct ext4_group_desc * desc; + struct ext4_super_block * es; + struct ext4_sb_info *sbi; + int err = 0, ret; + ext4_grpblk_t group_freed; + + *pdquot_freed_blocks = 0; + sbi = EXT4_SB(sb); + es = sbi->s_es; + if (block < le32_to_cpu(es->s_first_data_block) || + block + count < block || + block + count > ext4_blocks_count(es)) { + ext4_error (sb, "ext4_free_blocks", + "Freeing blocks not in datazone - " + "block = %llu, count = %lu", block, count); + goto error_return; + } + + ext4_debug ("freeing block(s) %llu-%llu\n", block, block + count - 1); + +do_more: + overflow = 0; + ext4_get_group_no_and_offset(sb, block, &block_group, &bit); + /* + * Check to see if we are freeing blocks across a group + * boundary. + */ + if (bit + count > EXT4_BLOCKS_PER_GROUP(sb)) { + overflow = bit + count - EXT4_BLOCKS_PER_GROUP(sb); + count -= overflow; + } + brelse(bitmap_bh); + bitmap_bh = read_block_bitmap(sb, block_group); + if (!bitmap_bh) + goto error_return; + desc = ext4_get_group_desc (sb, block_group, &gd_bh); + if (!desc) + goto error_return; + + if (in_range(ext4_block_bitmap(sb, desc), block, count) || + in_range(ext4_inode_bitmap(sb, desc), block, count) || + in_range(block, ext4_inode_table(sb, desc), sbi->s_itb_per_group) || + in_range(block + count - 1, ext4_inode_table(sb, desc), + sbi->s_itb_per_group)) + ext4_error (sb, "ext4_free_blocks", + "Freeing blocks in system zones - " + "Block = %llu, count = %lu", + block, count); + + /* + * We are about to start releasing blocks in the bitmap, + * so we need undo access. + */ + /* @@@ check errors */ + BUFFER_TRACE(bitmap_bh, "getting undo access"); + err = ext4_journal_get_undo_access(handle, bitmap_bh); + if (err) + goto error_return; + + /* + * We are about to modify some metadata. Call the journal APIs + * to unshare ->b_data if a currently-committing transaction is + * using it + */ + BUFFER_TRACE(gd_bh, "get_write_access"); + err = ext4_journal_get_write_access(handle, gd_bh); + if (err) + goto error_return; + + jbd_lock_bh_state(bitmap_bh); + + for (i = 0, group_freed = 0; i < count; i++) { + /* + * An HJ special. This is expensive... + */ +#ifdef CONFIG_JBD_DEBUG + jbd_unlock_bh_state(bitmap_bh); + { + struct buffer_head *debug_bh; + debug_bh = sb_find_get_block(sb, block + i); + if (debug_bh) { + BUFFER_TRACE(debug_bh, "Deleted!"); + if (!bh2jh(bitmap_bh)->b_committed_data) + BUFFER_TRACE(debug_bh, + "No commited data in bitmap"); + BUFFER_TRACE2(debug_bh, bitmap_bh, "bitmap"); + __brelse(debug_bh); + } + } + jbd_lock_bh_state(bitmap_bh); +#endif + if (need_resched()) { + jbd_unlock_bh_state(bitmap_bh); + cond_resched(); + jbd_lock_bh_state(bitmap_bh); + } + /* @@@ This prevents newly-allocated data from being + * freed and then reallocated within the same + * transaction. + * + * Ideally we would want to allow that to happen, but to + * do so requires making jbd2_journal_forget() capable of + * revoking the queued write of a data block, which + * implies blocking on the journal lock. *forget() + * cannot block due to truncate races. + * + * Eventually we can fix this by making jbd2_journal_forget() + * return a status indicating whether or not it was able + * to revoke the buffer. On successful revoke, it is + * safe not to set the allocation bit in the committed + * bitmap, because we know that there is no outstanding + * activity on the buffer any more and so it is safe to + * reallocate it. + */ + BUFFER_TRACE(bitmap_bh, "set in b_committed_data"); + J_ASSERT_BH(bitmap_bh, + bh2jh(bitmap_bh)->b_committed_data != NULL); + ext4_set_bit_atomic(sb_bgl_lock(sbi, block_group), bit + i, + bh2jh(bitmap_bh)->b_committed_data); + + /* + * We clear the bit in the bitmap after setting the committed + * data bit, because this is the reverse order to that which + * the allocator uses. + */ + BUFFER_TRACE(bitmap_bh, "clear bit"); + if (!ext4_clear_bit_atomic(sb_bgl_lock(sbi, block_group), + bit + i, bitmap_bh->b_data)) { + jbd_unlock_bh_state(bitmap_bh); + ext4_error(sb, __FUNCTION__, + "bit already cleared for block %llu", + (ext4_fsblk_t)(block + i)); + jbd_lock_bh_state(bitmap_bh); + BUFFER_TRACE(bitmap_bh, "bit already cleared"); + } else { + group_freed++; + } + } + jbd_unlock_bh_state(bitmap_bh); + + spin_lock(sb_bgl_lock(sbi, block_group)); + desc->bg_free_blocks_count = + cpu_to_le16(le16_to_cpu(desc->bg_free_blocks_count) + + group_freed); + spin_unlock(sb_bgl_lock(sbi, block_group)); + percpu_counter_mod(&sbi->s_freeblocks_counter, count); + + /* We dirtied the bitmap block */ + BUFFER_TRACE(bitmap_bh, "dirtied bitmap block"); + err = ext4_journal_dirty_metadata(handle, bitmap_bh); + + /* And the group descriptor block */ + BUFFER_TRACE(gd_bh, "dirtied group descriptor block"); + ret = ext4_journal_dirty_metadata(handle, gd_bh); + if (!err) err = ret; + *pdquot_freed_blocks += group_freed; + + if (overflow && !err) { + block += count; + count = overflow; + goto do_more; + } + sb->s_dirt = 1; +error_return: + brelse(bitmap_bh); + ext4_std_error(sb, err); + return; +} + +/** + * ext4_free_blocks() -- Free given blocks and update quota + * @handle: handle for this transaction + * @inode: inode + * @block: start physical block to free + * @count: number of blocks to count + */ +void ext4_free_blocks(handle_t *handle, struct inode *inode, + ext4_fsblk_t block, unsigned long count) +{ + struct super_block * sb; + unsigned long dquot_freed_blocks; + + sb = inode->i_sb; + if (!sb) { + printk ("ext4_free_blocks: nonexistent device"); + return; + } + ext4_free_blocks_sb(handle, sb, block, count, &dquot_freed_blocks); + if (dquot_freed_blocks) + DQUOT_FREE_BLOCK(inode, dquot_freed_blocks); + return; +} + +/** + * ext4_test_allocatable() + * @nr: given allocation block group + * @bh: bufferhead contains the bitmap of the given block group + * + * For ext4 allocations, we must not reuse any blocks which are + * allocated in the bitmap buffer's "last committed data" copy. This + * prevents deletes from freeing up the page for reuse until we have + * committed the delete transaction. + * + * If we didn't do this, then deleting something and reallocating it as + * data would allow the old block to be overwritten before the + * transaction committed (because we force data to disk before commit). + * This would lead to corruption if we crashed between overwriting the + * data and committing the delete. + * + * @@@ We may want to make this allocation behaviour conditional on + * data-writes at some point, and disable it for metadata allocations or + * sync-data inodes. + */ +static int ext4_test_allocatable(ext4_grpblk_t nr, struct buffer_head *bh) +{ + int ret; + struct journal_head *jh = bh2jh(bh); + + if (ext4_test_bit(nr, bh->b_data)) + return 0; + + jbd_lock_bh_state(bh); + if (!jh->b_committed_data) + ret = 1; + else + ret = !ext4_test_bit(nr, jh->b_committed_data); + jbd_unlock_bh_state(bh); + return ret; +} + +/** + * bitmap_search_next_usable_block() + * @start: the starting block (group relative) of the search + * @bh: bufferhead contains the block group bitmap + * @maxblocks: the ending block (group relative) of the reservation + * + * The bitmap search --- search forward alternately through the actual + * bitmap on disk and the last-committed copy in journal, until we find a + * bit free in both bitmaps. + */ +static ext4_grpblk_t +bitmap_search_next_usable_block(ext4_grpblk_t start, struct buffer_head *bh, + ext4_grpblk_t maxblocks) +{ + ext4_grpblk_t next; + struct journal_head *jh = bh2jh(bh); + + while (start < maxblocks) { + next = ext4_find_next_zero_bit(bh->b_data, maxblocks, start); + if (next >= maxblocks) + return -1; + if (ext4_test_allocatable(next, bh)) + return next; + jbd_lock_bh_state(bh); + if (jh->b_committed_data) + start = ext4_find_next_zero_bit(jh->b_committed_data, + maxblocks, next); + jbd_unlock_bh_state(bh); + } + return -1; +} + +/** + * find_next_usable_block() + * @start: the starting block (group relative) to find next + * allocatable block in bitmap. + * @bh: bufferhead contains the block group bitmap + * @maxblocks: the ending block (group relative) for the search + * + * Find an allocatable block in a bitmap. We honor both the bitmap and + * its last-committed copy (if that exists), and perform the "most + * appropriate allocation" algorithm of looking for a free block near + * the initial goal; then for a free byte somewhere in the bitmap; then + * for any free bit in the bitmap. + */ +static ext4_grpblk_t +find_next_usable_block(ext4_grpblk_t start, struct buffer_head *bh, + ext4_grpblk_t maxblocks) +{ + ext4_grpblk_t here, next; + char *p, *r; + + if (start > 0) { + /* + * The goal was occupied; search forward for a free + * block within the next XX blocks. + * + * end_goal is more or less random, but it has to be + * less than EXT4_BLOCKS_PER_GROUP. Aligning up to the + * next 64-bit boundary is simple.. + */ + ext4_grpblk_t end_goal = (start + 63) & ~63; + if (end_goal > maxblocks) + end_goal = maxblocks; + here = ext4_find_next_zero_bit(bh->b_data, end_goal, start); + if (here < end_goal && ext4_test_allocatable(here, bh)) + return here; + ext4_debug("Bit not found near goal\n"); + } + + here = start; + if (here < 0) + here = 0; + + p = ((char *)bh->b_data) + (here >> 3); + r = memscan(p, 0, (maxblocks - here + 7) >> 3); + next = (r - ((char *)bh->b_data)) << 3; + + if (next < maxblocks && next >= start && ext4_test_allocatable(next, bh)) + return next; + + /* + * The bitmap search --- search forward alternately through the actual + * bitmap and the last-committed copy until we find a bit free in + * both + */ + here = bitmap_search_next_usable_block(here, bh, maxblocks); + return here; +} + +/** + * claim_block() + * @block: the free block (group relative) to allocate + * @bh: the bufferhead containts the block group bitmap + * + * We think we can allocate this block in this bitmap. Try to set the bit. + * If that succeeds then check that nobody has allocated and then freed the + * block since we saw that is was not marked in b_committed_data. If it _was_ + * allocated and freed then clear the bit in the bitmap again and return + * zero (failure). + */ +static inline int +claim_block(spinlock_t *lock, ext4_grpblk_t block, struct buffer_head *bh) +{ + struct journal_head *jh = bh2jh(bh); + int ret; + + if (ext4_set_bit_atomic(lock, block, bh->b_data)) + return 0; + jbd_lock_bh_state(bh); + if (jh->b_committed_data && ext4_test_bit(block,jh->b_committed_data)) { + ext4_clear_bit_atomic(lock, block, bh->b_data); + ret = 0; + } else { + ret = 1; + } + jbd_unlock_bh_state(bh); + return ret; +} + +/** + * ext4_try_to_allocate() + * @sb: superblock + * @handle: handle to this transaction + * @group: given allocation block group + * @bitmap_bh: bufferhead holds the block bitmap + * @grp_goal: given target block within the group + * @count: target number of blocks to allocate + * @my_rsv: reservation window + * + * Attempt to allocate blocks within a give range. Set the range of allocation + * first, then find the first free bit(s) from the bitmap (within the range), + * and at last, allocate the blocks by claiming the found free bit as allocated. + * + * To set the range of this allocation: + * if there is a reservation window, only try to allocate block(s) from the + * file's own reservation window; + * Otherwise, the allocation range starts from the give goal block, ends at + * the block group's last block. + * + * If we failed to allocate the desired block then we may end up crossing to a + * new bitmap. In that case we must release write access to the old one via + * ext4_journal_release_buffer(), else we'll run out of credits. + */ +static ext4_grpblk_t +ext4_try_to_allocate(struct super_block *sb, handle_t *handle, int group, + struct buffer_head *bitmap_bh, ext4_grpblk_t grp_goal, + unsigned long *count, struct ext4_reserve_window *my_rsv) +{ + ext4_fsblk_t group_first_block; + ext4_grpblk_t start, end; + unsigned long num = 0; + + /* we do allocation within the reservation window if we have a window */ + if (my_rsv) { + group_first_block = ext4_group_first_block_no(sb, group); + if (my_rsv->_rsv_start >= group_first_block) + start = my_rsv->_rsv_start - group_first_block; + else + /* reservation window cross group boundary */ + start = 0; + end = my_rsv->_rsv_end - group_first_block + 1; + if (end > EXT4_BLOCKS_PER_GROUP(sb)) + /* reservation window crosses group boundary */ + end = EXT4_BLOCKS_PER_GROUP(sb); + if ((start <= grp_goal) && (grp_goal < end)) + start = grp_goal; + else + grp_goal = -1; + } else { + if (grp_goal > 0) + start = grp_goal; + else + start = 0; + end = EXT4_BLOCKS_PER_GROUP(sb); + } + + BUG_ON(start > EXT4_BLOCKS_PER_GROUP(sb)); + +repeat: + if (grp_goal < 0 || !ext4_test_allocatable(grp_goal, bitmap_bh)) { + grp_goal = find_next_usable_block(start, bitmap_bh, end); + if (grp_goal < 0) + goto fail_access; + if (!my_rsv) { + int i; + + for (i = 0; i < 7 && grp_goal > start && + ext4_test_allocatable(grp_goal - 1, + bitmap_bh); + i++, grp_goal--) + ; + } + } + start = grp_goal; + + if (!claim_block(sb_bgl_lock(EXT4_SB(sb), group), + grp_goal, bitmap_bh)) { + /* + * The block was allocated by another thread, or it was + * allocated and then freed by another thread + */ + start++; + grp_goal++; + if (start >= end) + goto fail_access; + goto repeat; + } + num++; + grp_goal++; + while (num < *count && grp_goal < end + && ext4_test_allocatable(grp_goal, bitmap_bh) + && claim_block(sb_bgl_lock(EXT4_SB(sb), group), + grp_goal, bitmap_bh)) { + num++; + grp_goal++; + } + *count = num; + return grp_goal - num; +fail_access: + *count = num; + return -1; +} + +/** + * find_next_reservable_window(): + * find a reservable space within the given range. + * It does not allocate the reservation window for now: + * alloc_new_reservation() will do the work later. + * + * @search_head: the head of the searching list; + * This is not necessarily the list head of the whole filesystem + * + * We have both head and start_block to assist the search + * for the reservable space. The list starts from head, + * but we will shift to the place where start_block is, + * then start from there, when looking for a reservable space. + * + * @size: the target new reservation window size + * + * @group_first_block: the first block we consider to start + * the real search from + * + * @last_block: + * the maximum block number that our goal reservable space + * could start from. This is normally the last block in this + * group. The search will end when we found the start of next + * possible reservable space is out of this boundary. + * This could handle the cross boundary reservation window + * request. + * + * basically we search from the given range, rather than the whole + * reservation double linked list, (start_block, last_block) + * to find a free region that is of my size and has not + * been reserved. + * + */ +static int find_next_reservable_window( + struct ext4_reserve_window_node *search_head, + struct ext4_reserve_window_node *my_rsv, + struct super_block * sb, + ext4_fsblk_t start_block, + ext4_fsblk_t last_block) +{ + struct rb_node *next; + struct ext4_reserve_window_node *rsv, *prev; + ext4_fsblk_t cur; + int size = my_rsv->rsv_goal_size; + + /* TODO: make the start of the reservation window byte-aligned */ + /* cur = *start_block & ~7;*/ + cur = start_block; + rsv = search_head; + if (!rsv) + return -1; + + while (1) { + if (cur <= rsv->rsv_end) + cur = rsv->rsv_end + 1; + + /* TODO? + * in the case we could not find a reservable space + * that is what is expected, during the re-search, we could + * remember what's the largest reservable space we could have + * and return that one. + * + * For now it will fail if we could not find the reservable + * space with expected-size (or more)... + */ + if (cur > last_block) + return -1; /* fail */ + + prev = rsv; + next = rb_next(&rsv->rsv_node); + rsv = list_entry(next,struct ext4_reserve_window_node,rsv_node); + + /* + * Reached the last reservation, we can just append to the + * previous one. + */ + if (!next) + break; + + if (cur + size <= rsv->rsv_start) { + /* + * Found a reserveable space big enough. We could + * have a reservation across the group boundary here + */ + break; + } + } + /* + * we come here either : + * when we reach the end of the whole list, + * and there is empty reservable space after last entry in the list. + * append it to the end of the list. + * + * or we found one reservable space in the middle of the list, + * return the reservation window that we could append to. + * succeed. + */ + + if ((prev != my_rsv) && (!rsv_is_empty(&my_rsv->rsv_window))) + rsv_window_remove(sb, my_rsv); + + /* + * Let's book the whole avaliable window for now. We will check the + * disk bitmap later and then, if there are free blocks then we adjust + * the window size if it's larger than requested. + * Otherwise, we will remove this node from the tree next time + * call find_next_reservable_window. + */ + my_rsv->rsv_start = cur; + my_rsv->rsv_end = cur + size - 1; + my_rsv->rsv_alloc_hit = 0; + + if (prev != my_rsv) + ext4_rsv_window_add(sb, my_rsv); + + return 0; +} + +/** + * alloc_new_reservation()--allocate a new reservation window + * + * To make a new reservation, we search part of the filesystem + * reservation list (the list that inside the group). We try to + * allocate a new reservation window near the allocation goal, + * or the beginning of the group, if there is no goal. + * + * We first find a reservable space after the goal, then from + * there, we check the bitmap for the first free block after + * it. If there is no free block until the end of group, then the + * whole group is full, we failed. Otherwise, check if the free + * block is inside the expected reservable space, if so, we + * succeed. + * If the first free block is outside the reservable space, then + * start from the first free block, we search for next available + * space, and go on. + * + * on succeed, a new reservation will be found and inserted into the list + * It contains at least one free block, and it does not overlap with other + * reservation windows. + * + * failed: we failed to find a reservation window in this group + * + * @rsv: the reservation + * + * @grp_goal: The goal (group-relative). It is where the search for a + * free reservable space should start from. + * if we have a grp_goal(grp_goal >0 ), then start from there, + * no grp_goal(grp_goal = -1), we start from the first block + * of the group. + * + * @sb: the super block + * @group: the group we are trying to allocate in + * @bitmap_bh: the block group block bitmap + * + */ +static int alloc_new_reservation(struct ext4_reserve_window_node *my_rsv, + ext4_grpblk_t grp_goal, struct super_block *sb, + unsigned int group, struct buffer_head *bitmap_bh) +{ + struct ext4_reserve_window_node *search_head; + ext4_fsblk_t group_first_block, group_end_block, start_block; + ext4_grpblk_t first_free_block; + struct rb_root *fs_rsv_root = &EXT4_SB(sb)->s_rsv_window_root; + unsigned long size; + int ret; + spinlock_t *rsv_lock = &EXT4_SB(sb)->s_rsv_window_lock; + + group_first_block = ext4_group_first_block_no(sb, group); + group_end_block = group_first_block + (EXT4_BLOCKS_PER_GROUP(sb) - 1); + + if (grp_goal < 0) + start_block = group_first_block; + else + start_block = grp_goal + group_first_block; + + size = my_rsv->rsv_goal_size; + + if (!rsv_is_empty(&my_rsv->rsv_window)) { + /* + * if the old reservation is cross group boundary + * and if the goal is inside the old reservation window, + * we will come here when we just failed to allocate from + * the first part of the window. We still have another part + * that belongs to the next group. In this case, there is no + * point to discard our window and try to allocate a new one + * in this group(which will fail). we should + * keep the reservation window, just simply move on. + * + * Maybe we could shift the start block of the reservation + * window to the first block of next group. + */ + + if ((my_rsv->rsv_start <= group_end_block) && + (my_rsv->rsv_end > group_end_block) && + (start_block >= my_rsv->rsv_start)) + return -1; + + if ((my_rsv->rsv_alloc_hit > + (my_rsv->rsv_end - my_rsv->rsv_start + 1) / 2)) { + /* + * if the previously allocation hit ratio is + * greater than 1/2, then we double the size of + * the reservation window the next time, + * otherwise we keep the same size window + */ + size = size * 2; + if (size > EXT4_MAX_RESERVE_BLOCKS) + size = EXT4_MAX_RESERVE_BLOCKS; + my_rsv->rsv_goal_size= size; + } + } + + spin_lock(rsv_lock); + /* + * shift the search start to the window near the goal block + */ + search_head = search_reserve_window(fs_rsv_root, start_block); + + /* + * find_next_reservable_window() simply finds a reservable window + * inside the given range(start_block, group_end_block). + * + * To make sure the reservation window has a free bit inside it, we + * need to check the bitmap after we found a reservable window. + */ +retry: + ret = find_next_reservable_window(search_head, my_rsv, sb, + start_block, group_end_block); + + if (ret == -1) { + if (!rsv_is_empty(&my_rsv->rsv_window)) + rsv_window_remove(sb, my_rsv); + spin_unlock(rsv_lock); + return -1; + } + + /* + * On success, find_next_reservable_window() returns the + * reservation window where there is a reservable space after it. + * Before we reserve this reservable space, we need + * to make sure there is at least a free block inside this region. + * + * searching the first free bit on the block bitmap and copy of + * last committed bitmap alternatively, until we found a allocatable + * block. Search start from the start block of the reservable space + * we just found. + */ + spin_unlock(rsv_lock); + first_free_block = bitmap_search_next_usable_block( + my_rsv->rsv_start - group_first_block, + bitmap_bh, group_end_block - group_first_block + 1); + + if (first_free_block < 0) { + /* + * no free block left on the bitmap, no point + * to reserve the space. return failed. + */ + spin_lock(rsv_lock); + if (!rsv_is_empty(&my_rsv->rsv_window)) + rsv_window_remove(sb, my_rsv); + spin_unlock(rsv_lock); + return -1; /* failed */ + } + + start_block = first_free_block + group_first_block; + /* + * check if the first free block is within the + * free space we just reserved + */ + if (start_block >= my_rsv->rsv_start && start_block < my_rsv->rsv_end) + return 0; /* success */ + /* + * if the first free bit we found is out of the reservable space + * continue search for next reservable space, + * start from where the free block is, + * we also shift the list head to where we stopped last time + */ + search_head = my_rsv; + spin_lock(rsv_lock); + goto retry; +} + +/** + * try_to_extend_reservation() + * @my_rsv: given reservation window + * @sb: super block + * @size: the delta to extend + * + * Attempt to expand the reservation window large enough to have + * required number of free blocks + * + * Since ext4_try_to_allocate() will always allocate blocks within + * the reservation window range, if the window size is too small, + * multiple blocks allocation has to stop at the end of the reservation + * window. To make this more efficient, given the total number of + * blocks needed and the current size of the window, we try to + * expand the reservation window size if necessary on a best-effort + * basis before ext4_new_blocks() tries to allocate blocks, + */ +static void try_to_extend_reservation(struct ext4_reserve_window_node *my_rsv, + struct super_block *sb, int size) +{ + struct ext4_reserve_window_node *next_rsv; + struct rb_node *next; + spinlock_t *rsv_lock = &EXT4_SB(sb)->s_rsv_window_lock; + + if (!spin_trylock(rsv_lock)) + return; + + next = rb_next(&my_rsv->rsv_node); + + if (!next) + my_rsv->rsv_end += size; + else { + next_rsv = list_entry(next, struct ext4_reserve_window_node, rsv_node); + + if ((next_rsv->rsv_start - my_rsv->rsv_end - 1) >= size) + my_rsv->rsv_end += size; + else + my_rsv->rsv_end = next_rsv->rsv_start - 1; + } + spin_unlock(rsv_lock); +} + +/** + * ext4_try_to_allocate_with_rsv() + * @sb: superblock + * @handle: handle to this transaction + * @group: given allocation block group + * @bitmap_bh: bufferhead holds the block bitmap + * @grp_goal: given target block within the group + * @count: target number of blocks to allocate + * @my_rsv: reservation window + * @errp: pointer to store the error code + * + * This is the main function used to allocate a new block and its reservation + * window. + * + * Each time when a new block allocation is need, first try to allocate from + * its own reservation. If it does not have a reservation window, instead of + * looking for a free bit on bitmap first, then look up the reservation list to + * see if it is inside somebody else's reservation window, we try to allocate a + * reservation window for it starting from the goal first. Then do the block + * allocation within the reservation window. + * + * This will avoid keeping on searching the reservation list again and + * again when somebody is looking for a free block (without + * reservation), and there are lots of free blocks, but they are all + * being reserved. + * + * We use a red-black tree for the per-filesystem reservation list. + * + */ +static ext4_grpblk_t +ext4_try_to_allocate_with_rsv(struct super_block *sb, handle_t *handle, + unsigned int group, struct buffer_head *bitmap_bh, + ext4_grpblk_t grp_goal, + struct ext4_reserve_window_node * my_rsv, + unsigned long *count, int *errp) +{ + ext4_fsblk_t group_first_block, group_last_block; + ext4_grpblk_t ret = 0; + int fatal; + unsigned long num = *count; + + *errp = 0; + + /* + * Make sure we use undo access for the bitmap, because it is critical + * that we do the frozen_data COW on bitmap buffers in all cases even + * if the buffer is in BJ_Forget state in the committing transaction. + */ + BUFFER_TRACE(bitmap_bh, "get undo access for new block"); + fatal = ext4_journal_get_undo_access(handle, bitmap_bh); + if (fatal) { + *errp = fatal; + return -1; + } + + /* + * we don't deal with reservation when + * filesystem is mounted without reservation + * or the file is not a regular file + * or last attempt to allocate a block with reservation turned on failed + */ + if (my_rsv == NULL ) { + ret = ext4_try_to_allocate(sb, handle, group, bitmap_bh, + grp_goal, count, NULL); + goto out; + } + /* + * grp_goal is a group relative block number (if there is a goal) + * 0 < grp_goal < EXT4_BLOCKS_PER_GROUP(sb) + * first block is a filesystem wide block number + * first block is the block number of the first block in this group + */ + group_first_block = ext4_group_first_block_no(sb, group); + group_last_block = group_first_block + (EXT4_BLOCKS_PER_GROUP(sb) - 1); + + /* + * Basically we will allocate a new block from inode's reservation + * window. + * + * We need to allocate a new reservation window, if: + * a) inode does not have a reservation window; or + * b) last attempt to allocate a block from existing reservation + * failed; or + * c) we come here with a goal and with a reservation window + * + * We do not need to allocate a new reservation window if we come here + * at the beginning with a goal and the goal is inside the window, or + * we don't have a goal but already have a reservation window. + * then we could go to allocate from the reservation window directly. + */ + while (1) { + if (rsv_is_empty(&my_rsv->rsv_window) || (ret < 0) || + !goal_in_my_reservation(&my_rsv->rsv_window, + grp_goal, group, sb)) { + if (my_rsv->rsv_goal_size < *count) + my_rsv->rsv_goal_size = *count; + ret = alloc_new_reservation(my_rsv, grp_goal, sb, + group, bitmap_bh); + if (ret < 0) + break; /* failed */ + + if (!goal_in_my_reservation(&my_rsv->rsv_window, + grp_goal, group, sb)) + grp_goal = -1; + } else if (grp_goal > 0 && + (my_rsv->rsv_end-grp_goal+1) < *count) + try_to_extend_reservation(my_rsv, sb, + *count-my_rsv->rsv_end + grp_goal - 1); + + if ((my_rsv->rsv_start > group_last_block) || + (my_rsv->rsv_end < group_first_block)) { + rsv_window_dump(&EXT4_SB(sb)->s_rsv_window_root, 1); + BUG(); + } + ret = ext4_try_to_allocate(sb, handle, group, bitmap_bh, + grp_goal, &num, &my_rsv->rsv_window); + if (ret >= 0) { + my_rsv->rsv_alloc_hit += num; + *count = num; + break; /* succeed */ + } + num = *count; + } +out: + if (ret >= 0) { + BUFFER_TRACE(bitmap_bh, "journal_dirty_metadata for " + "bitmap block"); + fatal = ext4_journal_dirty_metadata(handle, bitmap_bh); + if (fatal) { + *errp = fatal; + return -1; + } + return ret; + } + + BUFFER_TRACE(bitmap_bh, "journal_release_buffer"); + ext4_journal_release_buffer(handle, bitmap_bh); + return ret; +} + +/** + * ext4_has_free_blocks() + * @sbi: in-core super block structure. + * + * Check if filesystem has at least 1 free block available for allocation. + */ +static int ext4_has_free_blocks(struct ext4_sb_info *sbi) +{ + ext4_fsblk_t free_blocks, root_blocks; + + free_blocks = percpu_counter_read_positive(&sbi->s_freeblocks_counter); + root_blocks = ext4_r_blocks_count(sbi->s_es); + if (free_blocks < root_blocks + 1 && !capable(CAP_SYS_RESOURCE) && + sbi->s_resuid != current->fsuid && + (sbi->s_resgid == 0 || !in_group_p (sbi->s_resgid))) { + return 0; + } + return 1; +} + +/** + * ext4_should_retry_alloc() + * @sb: super block + * @retries number of attemps has been made + * + * ext4_should_retry_alloc() is called when ENOSPC is returned, and if + * it is profitable to retry the operation, this function will wait + * for the current or commiting transaction to complete, and then + * return TRUE. + * + * if the total number of retries exceed three times, return FALSE. + */ +int ext4_should_retry_alloc(struct super_block *sb, int *retries) +{ + if (!ext4_has_free_blocks(EXT4_SB(sb)) || (*retries)++ > 3) + return 0; + + jbd_debug(1, "%s: retrying operation after ENOSPC\n", sb->s_id); + + return jbd2_journal_force_commit_nested(EXT4_SB(sb)->s_journal); +} + +/** + * ext4_new_blocks() -- core block(s) allocation function + * @handle: handle to this transaction + * @inode: file inode + * @goal: given target block(filesystem wide) + * @count: target number of blocks to allocate + * @errp: error code + * + * ext4_new_blocks uses a goal block to assist allocation. It tries to + * allocate block(s) from the block group contains the goal block first. If that + * fails, it will try to allocate block(s) from other block groups without + * any specific goal block. + * + */ +ext4_fsblk_t ext4_new_blocks(handle_t *handle, struct inode *inode, + ext4_fsblk_t goal, unsigned long *count, int *errp) +{ + struct buffer_head *bitmap_bh = NULL; + struct buffer_head *gdp_bh; + unsigned long group_no; + int goal_group; + ext4_grpblk_t grp_target_blk; /* blockgroup relative goal block */ + ext4_grpblk_t grp_alloc_blk; /* blockgroup-relative allocated block*/ + ext4_fsblk_t ret_block; /* filesyetem-wide allocated block */ + int bgi; /* blockgroup iteration index */ + int fatal = 0, err; + int performed_allocation = 0; + ext4_grpblk_t free_blocks; /* number of free blocks in a group */ + struct super_block *sb; + struct ext4_group_desc *gdp; + struct ext4_super_block *es; + struct ext4_sb_info *sbi; + struct ext4_reserve_window_node *my_rsv = NULL; + struct ext4_block_alloc_info *block_i; + unsigned short windowsz = 0; +#ifdef EXT4FS_DEBUG + static int goal_hits, goal_attempts; +#endif + unsigned long ngroups; + unsigned long num = *count; + + *errp = -ENOSPC; + sb = inode->i_sb; + if (!sb) { + printk("ext4_new_block: nonexistent device"); + return 0; + } + + /* + * Check quota for allocation of this block. + */ + if (DQUOT_ALLOC_BLOCK(inode, num)) { + *errp = -EDQUOT; + return 0; + } + + sbi = EXT4_SB(sb); + es = EXT4_SB(sb)->s_es; + ext4_debug("goal=%lu.\n", goal); + /* + * Allocate a block from reservation only when + * filesystem is mounted with reservation(default,-o reservation), and + * it's a regular file, and + * the desired window size is greater than 0 (One could use ioctl + * command EXT4_IOC_SETRSVSZ to set the window size to 0 to turn off + * reservation on that particular file) + */ + block_i = EXT4_I(inode)->i_block_alloc_info; + if (block_i && ((windowsz = block_i->rsv_window_node.rsv_goal_size) > 0)) + my_rsv = &block_i->rsv_window_node; + + if (!ext4_has_free_blocks(sbi)) { + *errp = -ENOSPC; + goto out; + } + + /* + * First, test whether the goal block is free. + */ + if (goal < le32_to_cpu(es->s_first_data_block) || + goal >= ext4_blocks_count(es)) + goal = le32_to_cpu(es->s_first_data_block); + ext4_get_group_no_and_offset(sb, goal, &group_no, &grp_target_blk); + goal_group = group_no; +retry_alloc: + gdp = ext4_get_group_desc(sb, group_no, &gdp_bh); + if (!gdp) + goto io_error; + + free_blocks = le16_to_cpu(gdp->bg_free_blocks_count); + /* + * if there is not enough free blocks to make a new resevation + * turn off reservation for this allocation + */ + if (my_rsv && (free_blocks < windowsz) + && (rsv_is_empty(&my_rsv->rsv_window))) + my_rsv = NULL; + + if (free_blocks > 0) { + bitmap_bh = read_block_bitmap(sb, group_no); + if (!bitmap_bh) + goto io_error; + grp_alloc_blk = ext4_try_to_allocate_with_rsv(sb, handle, + group_no, bitmap_bh, grp_target_blk, + my_rsv, &num, &fatal); + if (fatal) + goto out; + if (grp_alloc_blk >= 0) + goto allocated; + } + + ngroups = EXT4_SB(sb)->s_groups_count; + smp_rmb(); + + /* + * Now search the rest of the groups. We assume that + * i and gdp correctly point to the last group visited. + */ + for (bgi = 0; bgi < ngroups; bgi++) { + group_no++; + if (group_no >= ngroups) + group_no = 0; + gdp = ext4_get_group_desc(sb, group_no, &gdp_bh); + if (!gdp) { + *errp = -EIO; + goto out; + } + free_blocks = le16_to_cpu(gdp->bg_free_blocks_count); + /* + * skip this group if the number of + * free blocks is less than half of the reservation + * window size. + */ + if (free_blocks <= (windowsz/2)) + continue; + + brelse(bitmap_bh); + bitmap_bh = read_block_bitmap(sb, group_no); + if (!bitmap_bh) + goto io_error; + /* + * try to allocate block(s) from this group, without a goal(-1). + */ + grp_alloc_blk = ext4_try_to_allocate_with_rsv(sb, handle, + group_no, bitmap_bh, -1, my_rsv, + &num, &fatal); + if (fatal) + goto out; + if (grp_alloc_blk >= 0) + goto allocated; + } + /* + * We may end up a bogus ealier ENOSPC error due to + * filesystem is "full" of reservations, but + * there maybe indeed free blocks avaliable on disk + * In this case, we just forget about the reservations + * just do block allocation as without reservations. + */ + if (my_rsv) { + my_rsv = NULL; + group_no = goal_group; + goto retry_alloc; + } + /* No space left on the device */ + *errp = -ENOSPC; + goto out; + +allocated: + + ext4_debug("using block group %d(%d)\n", + group_no, gdp->bg_free_blocks_count); + + BUFFER_TRACE(gdp_bh, "get_write_access"); + fatal = ext4_journal_get_write_access(handle, gdp_bh); + if (fatal) + goto out; + + ret_block = grp_alloc_blk + ext4_group_first_block_no(sb, group_no); + + if (in_range(ext4_block_bitmap(sb, gdp), ret_block, num) || + in_range(ext4_block_bitmap(sb, gdp), ret_block, num) || + in_range(ret_block, ext4_inode_table(sb, gdp), + EXT4_SB(sb)->s_itb_per_group) || + in_range(ret_block + num - 1, ext4_inode_table(sb, gdp), + EXT4_SB(sb)->s_itb_per_group)) + ext4_error(sb, "ext4_new_block", + "Allocating block in system zone - " + "blocks from %llu, length %lu", + ret_block, num); + + performed_allocation = 1; + +#ifdef CONFIG_JBD_DEBUG + { + struct buffer_head *debug_bh; + + /* Record bitmap buffer state in the newly allocated block */ + debug_bh = sb_find_get_block(sb, ret_block); + if (debug_bh) { + BUFFER_TRACE(debug_bh, "state when allocated"); + BUFFER_TRACE2(debug_bh, bitmap_bh, "bitmap state"); + brelse(debug_bh); + } + } + jbd_lock_bh_state(bitmap_bh); + spin_lock(sb_bgl_lock(sbi, group_no)); + if (buffer_jbd(bitmap_bh) && bh2jh(bitmap_bh)->b_committed_data) { + int i; + + for (i = 0; i < num; i++) { + if (ext4_test_bit(grp_alloc_blk+i, + bh2jh(bitmap_bh)->b_committed_data)) { + printk("%s: block was unexpectedly set in " + "b_committed_data\n", __FUNCTION__); + } + } + } + ext4_debug("found bit %d\n", grp_alloc_blk); + spin_unlock(sb_bgl_lock(sbi, group_no)); + jbd_unlock_bh_state(bitmap_bh); +#endif + + if (ret_block + num - 1 >= ext4_blocks_count(es)) { + ext4_error(sb, "ext4_new_block", + "block(%llu) >= blocks count(%llu) - " + "block_group = %lu, es == %p ", ret_block, + ext4_blocks_count(es), group_no, es); + goto out; + } + + /* + * It is up to the caller to add the new buffer to a journal + * list of some description. We don't know in advance whether + * the caller wants to use it as metadata or data. + */ + ext4_debug("allocating block %lu. Goal hits %d of %d.\n", + ret_block, goal_hits, goal_attempts); + + spin_lock(sb_bgl_lock(sbi, group_no)); + gdp->bg_free_blocks_count = + cpu_to_le16(le16_to_cpu(gdp->bg_free_blocks_count)-num); + spin_unlock(sb_bgl_lock(sbi, group_no)); + percpu_counter_mod(&sbi->s_freeblocks_counter, -num); + + BUFFER_TRACE(gdp_bh, "journal_dirty_metadata for group descriptor"); + err = ext4_journal_dirty_metadata(handle, gdp_bh); + if (!fatal) + fatal = err; + + sb->s_dirt = 1; + if (fatal) + goto out; + + *errp = 0; + brelse(bitmap_bh); + DQUOT_FREE_BLOCK(inode, *count-num); + *count = num; + return ret_block; + +io_error: + *errp = -EIO; +out: + if (fatal) { + *errp = fatal; + ext4_std_error(sb, fatal); + } + /* + * Undo the block allocation + */ + if (!performed_allocation) + DQUOT_FREE_BLOCK(inode, *count); + brelse(bitmap_bh); + return 0; +} + +ext4_fsblk_t ext4_new_block(handle_t *handle, struct inode *inode, + ext4_fsblk_t goal, int *errp) +{ + unsigned long count = 1; + + return ext4_new_blocks(handle, inode, goal, &count, errp); +} + +/** + * ext4_count_free_blocks() -- count filesystem free blocks + * @sb: superblock + * + * Adds up the number of free blocks from each block group. + */ +ext4_fsblk_t ext4_count_free_blocks(struct super_block *sb) +{ + ext4_fsblk_t desc_count; + struct ext4_group_desc *gdp; + int i; + unsigned long ngroups = EXT4_SB(sb)->s_groups_count; +#ifdef EXT4FS_DEBUG + struct ext4_super_block *es; + ext4_fsblk_t bitmap_count; + unsigned long x; + struct buffer_head *bitmap_bh = NULL; + + es = EXT4_SB(sb)->s_es; + desc_count = 0; + bitmap_count = 0; + gdp = NULL; + + smp_rmb(); + for (i = 0; i < ngroups; i++) { + gdp = ext4_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; + + x = ext4_count_free(bitmap_bh, sb->s_blocksize); + printk("group %d: stored = %d, counted = %lu\n", + i, le16_to_cpu(gdp->bg_free_blocks_count), x); + bitmap_count += x; + } + brelse(bitmap_bh); + printk("ext4_count_free_blocks: stored = %llu" + ", computed = %llu, %llu\n", + EXT4_FREE_BLOCKS_COUNT(es), + desc_count, bitmap_count); + return bitmap_count; +#else + desc_count = 0; + smp_rmb(); + for (i = 0; i < ngroups; i++) { + gdp = ext4_get_group_desc(sb, i, NULL); + if (!gdp) + continue; + desc_count += le16_to_cpu(gdp->bg_free_blocks_count); + } + + return desc_count; +#endif +} + +static inline int +block_in_use(ext4_fsblk_t block, struct super_block *sb, unsigned char *map) +{ + ext4_grpblk_t offset; + + ext4_get_group_no_and_offset(sb, block, NULL, &offset); + return ext4_test_bit (offset, map); +} + +static inline int test_root(int a, int b) +{ + int num = b; + + while (a > num) + num *= b; + return num == a; +} + +static int ext4_group_sparse(int group) +{ + if (group <= 1) + return 1; + if (!(group & 1)) + return 0; + return (test_root(group, 7) || test_root(group, 5) || + test_root(group, 3)); +} + +/** + * ext4_bg_has_super - number of blocks used by the superblock in group + * @sb: superblock for filesystem + * @group: group number to check + * + * Return the number of blocks used by the superblock (primary or backup) + * in this group. Currently this will be only 0 or 1. + */ +int ext4_bg_has_super(struct super_block *sb, int group) +{ + if (EXT4_HAS_RO_COMPAT_FEATURE(sb, + EXT4_FEATURE_RO_COMPAT_SPARSE_SUPER) && + !ext4_group_sparse(group)) + return 0; + return 1; +} + +static unsigned long ext4_bg_num_gdb_meta(struct super_block *sb, int group) +{ + unsigned long metagroup = group / EXT4_DESC_PER_BLOCK(sb); + unsigned long first = metagroup * EXT4_DESC_PER_BLOCK(sb); + unsigned long last = first + EXT4_DESC_PER_BLOCK(sb) - 1; + + if (group == first || group == first + 1 || group == last) + return 1; + return 0; +} + +static unsigned long ext4_bg_num_gdb_nometa(struct super_block *sb, int group) +{ + if (EXT4_HAS_RO_COMPAT_FEATURE(sb, + EXT4_FEATURE_RO_COMPAT_SPARSE_SUPER) && + !ext4_group_sparse(group)) + return 0; + return EXT4_SB(sb)->s_gdb_count; +} + +/** + * ext4_bg_num_gdb - number of blocks used by the group table in group + * @sb: superblock for filesystem + * @group: group number to check + * + * Return the number of blocks used by the group descriptor table + * (primary or backup) in this group. In the future there may be a + * different number of descriptor blocks in each group. + */ +unsigned long ext4_bg_num_gdb(struct super_block *sb, int group) +{ + unsigned long first_meta_bg = + le32_to_cpu(EXT4_SB(sb)->s_es->s_first_meta_bg); + unsigned long metagroup = group / EXT4_DESC_PER_BLOCK(sb); + + if (!EXT4_HAS_INCOMPAT_FEATURE(sb,EXT4_FEATURE_INCOMPAT_META_BG) || + metagroup < first_meta_bg) + return ext4_bg_num_gdb_nometa(sb,group); + + return ext4_bg_num_gdb_meta(sb,group); + +} diff --git a/fs/ext4/bitmap.c b/fs/ext4/bitmap.c new file mode 100644 index 000000000000..11e93c169bcf --- /dev/null +++ b/fs/ext4/bitmap.c @@ -0,0 +1,32 @@ +/* + * linux/fs/ext4/bitmap.c + * + * Copyright (C) 1992, 1993, 1994, 1995 + * Remy Card (card@masi.ibp.fr) + * Laboratoire MASI - Institut Blaise Pascal + * Universite Pierre et Marie Curie (Paris VI) + */ + +#include <linux/buffer_head.h> +#include <linux/jbd2.h> +#include <linux/ext4_fs.h> + +#ifdef EXT4FS_DEBUG + +static int nibblemap[] = {4, 3, 3, 2, 3, 2, 2, 1, 3, 2, 2, 1, 2, 1, 1, 0}; + +unsigned long ext4_count_free (struct buffer_head * map, unsigned int numchars) +{ + unsigned int i; + unsigned long sum = 0; + + if (!map) + return (0); + for (i = 0; i < numchars; i++) + sum += nibblemap[map->b_data[i] & 0xf] + + nibblemap[(map->b_data[i] >> 4) & 0xf]; + return (sum); +} + +#endif /* EXT4FS_DEBUG */ + diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c new file mode 100644 index 000000000000..f8595787a70e --- /dev/null +++ b/fs/ext4/dir.c @@ -0,0 +1,518 @@ +/* + * linux/fs/ext4/dir.c + * + * Copyright (C) 1992, 1993, 1994, 1995 + * Remy Card (card@masi.ibp.fr) + * Laboratoire MASI - Institut Blaise Pascal + * Universite Pierre et Marie Curie (Paris VI) + * + * from + * + * linux/fs/minix/dir.c + * + * Copyright (C) 1991, 1992 Linus Torvalds + * + * ext4 directory handling functions + * + * Big-endian to little-endian byte-swapping/bitmaps by + * David S. Miller (davem@caip.rutgers.edu), 1995 + * + * Hash Tree Directory indexing (c) 2001 Daniel Phillips + * + */ + +#include <linux/fs.h> +#include <linux/jbd2.h> +#include <linux/ext4_fs.h> +#include <linux/buffer_head.h> +#include <linux/smp_lock.h> +#include <linux/slab.h> +#include <linux/rbtree.h> + +static unsigned char ext4_filetype_table[] = { + DT_UNKNOWN, DT_REG, DT_DIR, DT_CHR, DT_BLK, DT_FIFO, DT_SOCK, DT_LNK +}; + +static int ext4_readdir(struct file *, void *, filldir_t); +static int ext4_dx_readdir(struct file * filp, + void * dirent, filldir_t filldir); +static int ext4_release_dir (struct inode * inode, + struct file * filp); + +const struct file_operations ext4_dir_operations = { + .llseek = generic_file_llseek, + .read = generic_read_dir, + .readdir = ext4_readdir, /* we take BKL. needed?*/ + .ioctl = ext4_ioctl, /* BKL held */ +#ifdef CONFIG_COMPAT + .compat_ioctl = ext4_compat_ioctl, +#endif + .fsync = ext4_sync_file, /* BKL held */ +#ifdef CONFIG_EXT4_INDEX + .release = ext4_release_dir, +#endif +}; + + +static unsigned char get_dtype(struct super_block *sb, int filetype) +{ + if (!EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_FILETYPE) || + (filetype >= EXT4_FT_MAX)) + return DT_UNKNOWN; + + return (ext4_filetype_table[filetype]); +} + + +int ext4_check_dir_entry (const char * function, struct inode * dir, + struct ext4_dir_entry_2 * de, + struct buffer_head * bh, + unsigned long offset) +{ + const char * error_msg = NULL; + const int rlen = le16_to_cpu(de->rec_len); + + if (rlen < EXT4_DIR_REC_LEN(1)) + error_msg = "rec_len is smaller than minimal"; + else if (rlen % 4 != 0) + error_msg = "rec_len % 4 != 0"; + else if (rlen < EXT4_DIR_REC_LEN(de->name_len)) + error_msg = "rec_len is too small for name_len"; + else if (((char *) de - bh->b_data) + rlen > dir->i_sb->s_blocksize) + error_msg = "directory entry across blocks"; + else if (le32_to_cpu(de->inode) > + le32_to_cpu(EXT4_SB(dir->i_sb)->s_es->s_inodes_count)) + error_msg = "inode out of bounds"; + + if (error_msg != NULL) + ext4_error (dir->i_sb, function, + "bad entry in directory #%lu: %s - " + "offset=%lu, inode=%lu, rec_len=%d, name_len=%d", + dir->i_ino, error_msg, offset, + (unsigned long) le32_to_cpu(de->inode), + rlen, de->name_len); + return error_msg == NULL ? 1 : 0; +} + +static int ext4_readdir(struct file * filp, + void * dirent, filldir_t filldir) +{ + int error = 0; + unsigned long offset; + int i, stored; + struct ext4_dir_entry_2 *de; + struct super_block *sb; + int err; + struct inode *inode = filp->f_dentry->d_inode; + int ret = 0; + + sb = inode->i_sb; + +#ifdef CONFIG_EXT4_INDEX + if (EXT4_HAS_COMPAT_FEATURE(inode->i_sb, + EXT4_FEATURE_COMPAT_DIR_INDEX) && + ((EXT4_I(inode)->i_flags & EXT4_INDEX_FL) || + ((inode->i_size >> sb->s_blocksize_bits) == 1))) { + err = ext4_dx_readdir(filp, dirent, filldir); + if (err != ERR_BAD_DX_DIR) { + ret = err; + goto out; + } + /* + * We don't set the inode dirty flag since it's not + * critical that it get flushed back to the disk. + */ + EXT4_I(filp->f_dentry->d_inode)->i_flags &= ~EXT4_INDEX_FL; + } +#endif + stored = 0; + offset = filp->f_pos & (sb->s_blocksize - 1); + + while (!error && !stored && filp->f_pos < inode->i_size) { + unsigned long blk = filp->f_pos >> EXT4_BLOCK_SIZE_BITS(sb); + struct buffer_head map_bh; + struct buffer_head *bh = NULL; + + map_bh.b_state = 0; + err = ext4_get_blocks_wrap(NULL, inode, blk, 1, &map_bh, 0, 0); + if (err > 0) { + page_cache_readahead(sb->s_bdev->bd_inode->i_mapping, + &filp->f_ra, + filp, + map_bh.b_blocknr >> + (PAGE_CACHE_SHIFT - inode->i_blkbits), + 1); + bh = ext4_bread(NULL, inode, blk, 0, &err); + } + + /* + * We ignore I/O errors on directories so users have a chance + * of recovering data when there's a bad sector + */ + if (!bh) { + ext4_error (sb, "ext4_readdir", + "directory #%lu contains a hole at offset %lu", + inode->i_ino, (unsigned long)filp->f_pos); + filp->f_pos += sb->s_blocksize - offset; + continue; + } + +revalidate: + /* If the dir block has changed since the last call to + * readdir(2), then we might be pointing to an invalid + * dirent right now. Scan from the start of the block + * to make sure. */ + if (filp->f_version != inode->i_version) { + for (i = 0; i < sb->s_blocksize && i < offset; ) { + de = (struct ext4_dir_entry_2 *) + (bh->b_data + i); + /* It's too expensive to do a full + * dirent test each time round this + * loop, but we do have to test at + * least that it is non-zero. A + * failure will be detected in the + * dirent test below. */ + if (le16_to_cpu(de->rec_len) < + EXT4_DIR_REC_LEN(1)) + break; + i += le16_to_cpu(de->rec_len); + } + offset = i; + filp->f_pos = (filp->f_pos & ~(sb->s_blocksize - 1)) + | offset; + filp->f_version = inode->i_version; + } + + while (!error && filp->f_pos < inode->i_size + && offset < sb->s_blocksize) { + de = (struct ext4_dir_entry_2 *) (bh->b_data + offset); + if (!ext4_check_dir_entry ("ext4_readdir", inode, de, + bh, offset)) { + /* + * On error, skip the f_pos to the next block + */ + filp->f_pos = (filp->f_pos | + (sb->s_blocksize - 1)) + 1; + brelse (bh); + ret = stored; + goto out; + } + offset += le16_to_cpu(de->rec_len); + if (le32_to_cpu(de->inode)) { + /* We might block in the next section + * if the data destination is + * currently swapped out. So, use a + * version stamp to detect whether or + * not the directory has been modified + * during the copy operation. + */ + unsigned long version = filp->f_version; + + error = filldir(dirent, de->name, + de->name_len, + filp->f_pos, + le32_to_cpu(de->inode), + get_dtype(sb, de->file_type)); + if (error) + break; + if (version != filp->f_version) + goto revalidate; + stored ++; + } + filp->f_pos += le16_to_cpu(de->rec_len); + } + offset = 0; + brelse (bh); + } +out: + return ret; +} + +#ifdef CONFIG_EXT4_INDEX +/* + * These functions convert from the major/minor hash to an f_pos + * value. + * + * Currently we only use major hash numer. This is unfortunate, but + * on 32-bit machines, the same VFS interface is used for lseek and + * llseek, so if we use the 64 bit offset, then the 32-bit versions of + * lseek/telldir/seekdir will blow out spectacularly, and from within + * the ext2 low-level routine, we don't know if we're being called by + * a 64-bit version of the system call or the 32-bit version of the + * system call. Worse yet, NFSv2 only allows for a 32-bit readdir + * cookie. Sigh. + */ +#define hash2pos(major, minor) (major >> 1) +#define pos2maj_hash(pos) ((pos << 1) & 0xffffffff) +#define pos2min_hash(pos) (0) + +/* + * This structure holds the nodes of the red-black tree used to store + * the directory entry in hash order. + */ +struct fname { + __u32 hash; + __u32 minor_hash; + struct rb_node rb_hash; + struct fname *next; + __u32 inode; + __u8 name_len; + __u8 file_type; + char name[0]; +}; + +/* + * This functoin implements a non-recursive way of freeing all of the + * nodes in the red-black tree. + */ +static void free_rb_tree_fname(struct rb_root *root) +{ + struct rb_node *n = root->rb_node; + struct rb_node *parent; + struct fname *fname; + + while (n) { + /* Do the node's children first */ + if ((n)->rb_left) { + n = n->rb_left; + continue; + } + if (n->rb_right) { + n = n->rb_right; + continue; + } + /* + * The node has no children; free it, and then zero + * out parent's link to it. Finally go to the + * beginning of the loop and try to free the parent + * node. + */ + parent = rb_parent(n); + fname = rb_entry(n, struct fname, rb_hash); + while (fname) { + struct fname * old = fname; + fname = fname->next; + kfree (old); + } + if (!parent) + root->rb_node = NULL; + else if (parent->rb_left == n) + parent->rb_left = NULL; + else if (parent->rb_right == n) + parent->rb_right = NULL; + n = parent; + } + root->rb_node = NULL; +} + + +static struct dir_private_info *create_dir_info(loff_t pos) +{ + struct dir_private_info *p; + + p = kmalloc(sizeof(struct dir_private_info), GFP_KERNEL); + if (!p) + return NULL; + p->root.rb_node = NULL; + p->curr_node = NULL; + p->extra_fname = NULL; + p->last_pos = 0; + p->curr_hash = pos2maj_hash(pos); + p->curr_minor_hash = pos2min_hash(pos); + p->next_hash = 0; + return p; +} + +void ext4_htree_free_dir_info(struct dir_private_info *p) +{ + free_rb_tree_fname(&p->root); + kfree(p); +} + +/* + * Given a directory entry, enter it into the fname rb tree. + */ +int ext4_htree_store_dirent(struct file *dir_file, __u32 hash, + __u32 minor_hash, + struct ext4_dir_entry_2 *dirent) +{ + struct rb_node **p, *parent = NULL; + struct fname * fname, *new_fn; + struct dir_private_info *info; + int len; + + info = (struct dir_private_info *) dir_file->private_data; + p = &info->root.rb_node; + + /* Create and allocate the fname structure */ + len = sizeof(struct fname) + dirent->name_len + 1; + new_fn = kzalloc(len, GFP_KERNEL); + if (!new_fn) + return -ENOMEM; + new_fn->hash = hash; + new_fn->minor_hash = minor_hash; + new_fn->inode = le32_to_cpu(dirent->inode); + new_fn->name_len = dirent->name_len; + new_fn->file_type = dirent->file_type; + memcpy(new_fn->name, dirent->name, dirent->name_len); + new_fn->name[dirent->name_len] = 0; + + while (*p) { + parent = *p; + fname = rb_entry(parent, struct fname, rb_hash); + + /* + * If the hash and minor hash match up, then we put + * them on a linked list. This rarely happens... + */ + if ((new_fn->hash == fname->hash) && + (new_fn->minor_hash == fname->minor_hash)) { + new_fn->next = fname->next; + fname->next = new_fn; + return 0; + } + + if (new_fn->hash < fname->hash) + p = &(*p)->rb_left; + else if (new_fn->hash > fname->hash) + p = &(*p)->rb_right; + else if (new_fn->minor_hash < fname->minor_hash) + p = &(*p)->rb_left; + else /* if (new_fn->minor_hash > fname->minor_hash) */ + p = &(*p)->rb_right; + } + + rb_link_node(&new_fn->rb_hash, parent, p); + rb_insert_color(&new_fn->rb_hash, &info->root); + return 0; +} + + + +/* + * This is a helper function for ext4_dx_readdir. It calls filldir + * for all entres on the fname linked list. (Normally there is only + * one entry on the linked list, unless there are 62 bit hash collisions.) + */ +static int call_filldir(struct file * filp, void * dirent, + filldir_t filldir, struct fname *fname) +{ + struct dir_private_info *info = filp->private_data; + loff_t curr_pos; + struct inode *inode = filp->f_dentry->d_inode; + struct super_block * sb; + int error; + + sb = inode->i_sb; + + if (!fname) { + printk("call_filldir: called with null fname?!?\n"); + return 0; + } + curr_pos = hash2pos(fname->hash, fname->minor_hash); + while (fname) { + error = filldir(dirent, fname->name, + fname->name_len, curr_pos, + fname->inode, + get_dtype(sb, fname->file_type)); + if (error) { + filp->f_pos = curr_pos; + info->extra_fname = fname->next; + return error; + } + fname = fname->next; + } + return 0; +} + +static int ext4_dx_readdir(struct file * filp, + void * dirent, filldir_t filldir) +{ + struct dir_private_info *info = filp->private_data; + struct inode *inode = filp->f_dentry->d_inode; + struct fname *fname; + int ret; + + if (!info) { + info = create_dir_info(filp->f_pos); + if (!info) + return -ENOMEM; + filp->private_data = info; + } + + if (filp->f_pos == EXT4_HTREE_EOF) + return 0; /* EOF */ + + /* Some one has messed with f_pos; reset the world */ + if (info->last_pos != filp->f_pos) { + free_rb_tree_fname(&info->root); + info->curr_node = NULL; + info->extra_fname = NULL; + info->curr_hash = pos2maj_hash(filp->f_pos); + info->curr_minor_hash = pos2min_hash(filp->f_pos); + } + + /* + * If there are any leftover names on the hash collision + * chain, return them first. + */ + if (info->extra_fname && + call_filldir(filp, dirent, filldir, info->extra_fname)) + goto finished; + + if (!info->curr_node) + info->curr_node = rb_first(&info->root); + + while (1) { + /* + * Fill the rbtree if we have no more entries, + * or the inode has changed since we last read in the + * cached entries. + */ + if ((!info->curr_node) || + (filp->f_version != inode->i_version)) { + info->curr_node = NULL; + free_rb_tree_fname(&info->root); + filp->f_version = inode->i_version; + ret = ext4_htree_fill_tree(filp, info->curr_hash, + info->curr_minor_hash, + &info->next_hash); + if (ret < 0) + return ret; + if (ret == 0) { + filp->f_pos = EXT4_HTREE_EOF; + break; + } + info->curr_node = rb_first(&info->root); + } + + fname = rb_entry(info->curr_node, struct fname, rb_hash); + info->curr_hash = fname->hash; + info->curr_minor_hash = fname->minor_hash; + if (call_filldir(filp, dirent, filldir, fname)) + break; + + info->curr_node = rb_next(info->curr_node); + if (!info->curr_node) { + if (info->next_hash == ~0) { + filp->f_pos = EXT4_HTREE_EOF; + break; + } + info->curr_hash = info->next_hash; + info->curr_minor_hash = 0; + } + } +finished: + info->last_pos = filp->f_pos; + return 0; +} + +static int ext4_release_dir (struct inode * inode, struct file * filp) +{ + if (filp->private_data) + ext4_htree_free_dir_info(filp->private_data); + + return 0; +} + +#endif diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c new file mode 100644 index 000000000000..2608dce18f3e --- /dev/null +++ b/fs/ext4/extents.c @@ -0,0 +1,2152 @@ +/* + * Copyright (c) 2003-2006, Cluster File Systems, Inc, info@clusterfs.com + * Written by Alex Tomas <alex@clusterfs.com> + * + * Architecture independence: + * Copyright (c) 2005, Bull S.A. + * Written by Pierre Peiffer <pierre.peiffer@bull.net> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * This program 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 General Public License for more details. + * + * You should have received a copy of the GNU General Public Licens + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111- + */ + +/* + * Extents support for EXT4 + * + * TODO: + * - ext4*_error() should be used in some situations + * - analyze all BUG()/BUG_ON(), use -EIO where appropriate + * - smart tree reduction + */ + +#include <linux/module.h> +#include <linux/fs.h> +#include <linux/time.h> +#include <linux/ext4_jbd2.h> +#include <linux/jbd.h> +#include <linux/smp_lock.h> +#include <linux/highuid.h> +#include <linux/pagemap.h> +#include <linux/quotaops.h> +#include <linux/string.h> +#include <linux/slab.h> +#include <linux/ext4_fs_extents.h> +#include <asm/uaccess.h> + + +/* + * ext_pblock: + * combine low and high parts of physical block number into ext4_fsblk_t + */ +static inline ext4_fsblk_t ext_pblock(struct ext4_extent *ex) +{ + ext4_fsblk_t block; + + block = le32_to_cpu(ex->ee_start); + block |= ((ext4_fsblk_t) le16_to_cpu(ex->ee_start_hi) << 31) << 1; + return block; +} + +/* + * idx_pblock: + * combine low and high parts of a leaf physical block number into ext4_fsblk_t + */ +static inline ext4_fsblk_t idx_pblock(struct ext4_extent_idx *ix) +{ + ext4_fsblk_t block; + + block = le32_to_cpu(ix->ei_leaf); + block |= ((ext4_fsblk_t) le16_to_cpu(ix->ei_leaf_hi) << 31) << 1; + return block; +} + +/* + * ext4_ext_store_pblock: + * stores a large physical block number into an extent struct, + * breaking it into parts + */ +static inline void ext4_ext_store_pblock(struct ext4_extent *ex, ext4_fsblk_t pb) +{ + ex->ee_start = cpu_to_le32((unsigned long) (pb & 0xffffffff)); + ex->ee_start_hi = cpu_to_le16((unsigned long) ((pb >> 31) >> 1) & 0xffff); +} + +/* + * ext4_idx_store_pblock: + * stores a large physical block number into an index struct, + * breaking it into parts + */ +static inline void ext4_idx_store_pblock(struct ext4_extent_idx *ix, ext4_fsblk_t pb) +{ + ix->ei_leaf = cpu_to_le32((unsigned long) (pb & 0xffffffff)); + ix->ei_leaf_hi = cpu_to_le16((unsigned long) ((pb >> 31) >> 1) & 0xffff); +} + +static int ext4_ext_check_header(const char *function, struct inode *inode, + struct ext4_extent_header *eh) +{ + const char *error_msg = NULL; + + if (unlikely(eh->eh_magic != EXT4_EXT_MAGIC)) { + error_msg = "invalid magic"; + goto corrupted; + } + if (unlikely(eh->eh_max == 0)) { + error_msg = "invalid eh_max"; + goto corrupted; + } + if (unlikely(le16_to_cpu(eh->eh_entries) > le16_to_cpu(eh->eh_max))) { + error_msg = "invalid eh_entries"; + goto corrupted; + } + return 0; + +corrupted: + ext4_error(inode->i_sb, function, + "bad header in inode #%lu: %s - magic %x, " + "entries %u, max %u, depth %u", + inode->i_ino, error_msg, le16_to_cpu(eh->eh_magic), + le16_to_cpu(eh->eh_entries), le16_to_cpu(eh->eh_max), + le16_to_cpu(eh->eh_depth)); + + return -EIO; +} + +static handle_t *ext4_ext_journal_restart(handle_t *handle, int needed) +{ + int err; + + if (handle->h_buffer_credits > needed) + return handle; + if (!ext4_journal_extend(handle, needed)) + return handle; + err = ext4_journal_restart(handle, needed); + + return handle; +} + +/* + * could return: + * - EROFS + * - ENOMEM + */ +static int ext4_ext_get_access(handle_t *handle, struct inode *inode, + struct ext4_ext_path *path) +{ + if (path->p_bh) { + /* path points to block */ + return ext4_journal_get_write_access(handle, path->p_bh); + } + /* path points to leaf/index in inode body */ + /* we use in-core data, no need to protect them */ + return 0; +} + +/* + * could return: + * - EROFS + * - ENOMEM + * - EIO + */ +static int ext4_ext_dirty(handle_t *handle, struct inode *inode, + struct ext4_ext_path *path) +{ + int err; + if (path->p_bh) { + /* path points to block */ + err = ext4_journal_dirty_metadata(handle, path->p_bh); + } else { + /* path points to leaf/index in inode body */ + err = ext4_mark_inode_dirty(handle, inode); + } + return err; +} + +static ext4_fsblk_t ext4_ext_find_goal(struct inode *inode, + struct ext4_ext_path *path, + ext4_fsblk_t block) +{ + struct ext4_inode_info *ei = EXT4_I(inode); + ext4_fsblk_t bg_start; + ext4_grpblk_t colour; + int depth; + + if (path) { + struct ext4_extent *ex; + depth = path->p_depth; + + /* try to predict block placement */ + if ((ex = path[depth].p_ext)) + return ext_pblock(ex)+(block-le32_to_cpu(ex->ee_block)); + + /* it looks like index is empty; + * try to find starting block from index itself */ + if (path[depth].p_bh) + return path[depth].p_bh->b_blocknr; + } + + /* OK. use inode's group */ + bg_start = (ei->i_block_group * EXT4_BLOCKS_PER_GROUP(inode->i_sb)) + + le32_to_cpu(EXT4_SB(inode->i_sb)->s_es->s_first_data_block); + colour = (current->pid % 16) * + (EXT4_BLOCKS_PER_GROUP(inode->i_sb) / 16); + return bg_start + colour + block; +} + +static ext4_fsblk_t +ext4_ext_new_block(handle_t *handle, struct inode *inode, + struct ext4_ext_path *path, + struct ext4_extent *ex, int *err) +{ + ext4_fsblk_t goal, newblock; + + goal = ext4_ext_find_goal(inode, path, le32_to_cpu(ex->ee_block)); + newblock = ext4_new_block(handle, inode, goal, err); + return newblock; +} + +static inline int ext4_ext_space_block(struct inode *inode) +{ + int size; + + size = (inode->i_sb->s_blocksize - sizeof(struct ext4_extent_header)) + / sizeof(struct ext4_extent); +#ifdef AGRESSIVE_TEST + if (size > 6) + size = 6; +#endif + return size; +} + +static inline int ext4_ext_space_block_idx(struct inode *inode) +{ + int size; + + size = (inode->i_sb->s_blocksize - sizeof(struct ext4_extent_header)) + / sizeof(struct ext4_extent_idx); +#ifdef AGRESSIVE_TEST + if (size > 5) + size = 5; +#endif + return size; +} + +static inline int ext4_ext_space_root(struct inode *inode) +{ + int size; + + size = sizeof(EXT4_I(inode)->i_data); + size -= sizeof(struct ext4_extent_header); + size /= sizeof(struct ext4_extent); +#ifdef AGRESSIVE_TEST + if (size > 3) + size = 3; +#endif + return size; +} + +static inline int ext4_ext_space_root_idx(struct inode *inode) +{ + int size; + + size = sizeof(EXT4_I(inode)->i_data); + size -= sizeof(struct ext4_extent_header); + size /= sizeof(struct ext4_extent_idx); +#ifdef AGRESSIVE_TEST + if (size > 4) + size = 4; +#endif + return size; +} + +#ifdef EXT_DEBUG +static void ext4_ext_show_path(struct inode *inode, struct ext4_ext_path *path) +{ + int k, l = path->p_depth; + + ext_debug("path:"); + for (k = 0; k <= l; k++, path++) { + if (path->p_idx) { + ext_debug(" %d->%llu", le32_to_cpu(path->p_idx->ei_block), + idx_pblock(path->p_idx)); + } else if (path->p_ext) { + ext_debug(" %d:%d:%llu ", + le32_to_cpu(path->p_ext->ee_block), + le16_to_cpu(path->p_ext->ee_len), + ext_pblock(path->p_ext)); + } else + ext_debug(" []"); + } + ext_debug("\n"); +} + +static void ext4_ext_show_leaf(struct inode *inode, struct ext4_ext_path *path) +{ + int depth = ext_depth(inode); + struct ext4_extent_header *eh; + struct ext4_extent *ex; + int i; + + if (!path) + return; + + eh = path[depth].p_hdr; + ex = EXT_FIRST_EXTENT(eh); + + for (i = 0; i < le16_to_cpu(eh->eh_entries); i++, ex++) { + ext_debug("%d:%d:%llu ", le32_to_cpu(ex->ee_block), + le16_to_cpu(ex->ee_len), ext_pblock(ex)); + } + ext_debug("\n"); +} +#else +#define ext4_ext_show_path(inode,path) +#define ext4_ext_show_leaf(inode,path) +#endif + +static void ext4_ext_drop_refs(struct ext4_ext_path *path) +{ + int depth = path->p_depth; + int i; + + for (i = 0; i <= depth; i++, path++) + if (path->p_bh) { + brelse(path->p_bh); + path->p_bh = NULL; + } +} + +/* + * ext4_ext_binsearch_idx: + * binary search for the closest index of the given block + */ +static void +ext4_ext_binsearch_idx(struct inode *inode, struct ext4_ext_path *path, int block) +{ + struct ext4_extent_header *eh = path->p_hdr; + struct ext4_extent_idx *r, *l, *m; + + BUG_ON(eh->eh_magic != EXT4_EXT_MAGIC); + BUG_ON(le16_to_cpu(eh->eh_entries) > le16_to_cpu(eh->eh_max)); + BUG_ON(le16_to_cpu(eh->eh_entries) <= 0); + + ext_debug("binsearch for %d(idx): ", block); + + l = EXT_FIRST_INDEX(eh) + 1; + r = EXT_FIRST_INDEX(eh) + le16_to_cpu(eh->eh_entries) - 1; + while (l <= r) { + m = l + (r - l) / 2; + if (block < le32_to_cpu(m->ei_block)) + r = m - 1; + else + l = m + 1; + ext_debug("%p(%u):%p(%u):%p(%u) ", l, l->ei_block, + m, m->ei_block, r, r->ei_block); + } + + path->p_idx = l - 1; + ext_debug(" -> %d->%lld ", le32_to_cpu(path->p_idx->ei_block), + idx_block(path->p_idx)); + +#ifdef CHECK_BINSEARCH + { + struct ext4_extent_idx *chix, *ix; + int k; + + chix = ix = EXT_FIRST_INDEX(eh); + for (k = 0; k < le16_to_cpu(eh->eh_entries); k++, ix++) { + if (k != 0 && + le32_to_cpu(ix->ei_block) <= le32_to_cpu(ix[-1].ei_block)) { + printk("k=%d, ix=0x%p, first=0x%p\n", k, + ix, EXT_FIRST_INDEX(eh)); + printk("%u <= %u\n", + le32_to_cpu(ix->ei_block), + le32_to_cpu(ix[-1].ei_block)); + } + BUG_ON(k && le32_to_cpu(ix->ei_block) + <= le32_to_cpu(ix[-1].ei_block)); + if (block < le32_to_cpu(ix->ei_block)) + break; + chix = ix; + } + BUG_ON(chix != path->p_idx); + } +#endif + +} + +/* + * ext4_ext_binsearch: + * binary search for closest extent of the given block + */ +static void +ext4_ext_binsearch(struct inode *inode, struct ext4_ext_path *path, int block) +{ + struct ext4_extent_header *eh = path->p_hdr; + struct ext4_extent *r, *l, *m; + + BUG_ON(eh->eh_magic != EXT4_EXT_MAGIC); + BUG_ON(le16_to_cpu(eh->eh_entries) > le16_to_cpu(eh->eh_max)); + + if (eh->eh_entries == 0) { + /* + * this leaf is empty: + * we get such a leaf in split/add case + */ + return; + } + + ext_debug("binsearch for %d: ", block); + + l = EXT_FIRST_EXTENT(eh) + 1; + r = EXT_FIRST_EXTENT(eh) + le16_to_cpu(eh->eh_entries) - 1; + + while (l <= r) { + m = l + (r - l) / 2; + if (block < le32_to_cpu(m->ee_block)) + r = m - 1; + else + l = m + 1; + ext_debug("%p(%u):%p(%u):%p(%u) ", l, l->ee_block, + m, m->ee_block, r, r->ee_block); + } + + path->p_ext = l - 1; + ext_debug(" -> %d:%llu:%d ", + le32_to_cpu(path->p_ext->ee_block), + ext_pblock(path->p_ext), + le16_to_cpu(path->p_ext->ee_len)); + +#ifdef CHECK_BINSEARCH + { + struct ext4_extent *chex, *ex; + int k; + + chex = ex = EXT_FIRST_EXTENT(eh); + for (k = 0; k < le16_to_cpu(eh->eh_entries); k++, ex++) { + BUG_ON(k && le32_to_cpu(ex->ee_block) + <= le32_to_cpu(ex[-1].ee_block)); + if (block < le32_to_cpu(ex->ee_block)) + break; + chex = ex; + } + BUG_ON(chex != path->p_ext); + } +#endif + +} + +int ext4_ext_tree_init(handle_t *handle, struct inode *inode) +{ + struct ext4_extent_header *eh; + + eh = ext_inode_hdr(inode); + eh->eh_depth = 0; + eh->eh_entries = 0; + eh->eh_magic = EXT4_EXT_MAGIC; + eh->eh_max = cpu_to_le16(ext4_ext_space_root(inode)); + ext4_mark_inode_dirty(handle, inode); + ext4_ext_invalidate_cache(inode); + return 0; +} + +struct ext4_ext_path * +ext4_ext_find_extent(struct inode *inode, int block, struct ext4_ext_path *path) +{ + struct ext4_extent_header *eh; + struct buffer_head *bh; + short int depth, i, ppos = 0, alloc = 0; + + eh = ext_inode_hdr(inode); + BUG_ON(eh == NULL); + if (ext4_ext_check_header(__FUNCTION__, inode, eh)) + return ERR_PTR(-EIO); + + i = depth = ext_depth(inode); + + /* account possible depth increase */ + if (!path) { + path = kmalloc(sizeof(struct ext4_ext_path) * (depth + 2), + GFP_NOFS); + if (!path) + return ERR_PTR(-ENOMEM); + alloc = 1; + } + memset(path, 0, sizeof(struct ext4_ext_path) * (depth + 1)); + path[0].p_hdr = eh; + + /* walk through the tree */ + while (i) { + ext_debug("depth %d: num %d, max %d\n", + ppos, le16_to_cpu(eh->eh_entries), le16_to_cpu(eh->eh_max)); + ext4_ext_binsearch_idx(inode, path + ppos, block); + path[ppos].p_block = idx_pblock(path[ppos].p_idx); + path[ppos].p_depth = i; + path[ppos].p_ext = NULL; + + bh = sb_bread(inode->i_sb, path[ppos].p_block); + if (!bh) + goto err; + + eh = ext_block_hdr(bh); + ppos++; + BUG_ON(ppos > depth); + path[ppos].p_bh = bh; + path[ppos].p_hdr = eh; + i--; + + if (ext4_ext_check_header(__FUNCTION__, inode, eh)) + goto err; + } + + path[ppos].p_depth = i; + path[ppos].p_hdr = eh; + path[ppos].p_ext = NULL; + path[ppos].p_idx = NULL; + + if (ext4_ext_check_header(__FUNCTION__, inode, eh)) + goto err; + + /* find extent */ + ext4_ext_binsearch(inode, path + ppos, block); + + ext4_ext_show_path(inode, path); + + return path; + +err: + ext4_ext_drop_refs(path); + if (alloc) + kfree(path); + return ERR_PTR(-EIO); +} + +/* + * ext4_ext_insert_index: + * insert new index [@logical;@ptr] into the block at @curp; + * check where to insert: before @curp or after @curp + */ +static int ext4_ext_insert_index(handle_t *handle, struct inode *inode, + struct ext4_ext_path *curp, + int logical, ext4_fsblk_t ptr) +{ + struct ext4_extent_idx *ix; + int len, err; + + if ((err = ext4_ext_get_access(handle, inode, curp))) + return err; + + BUG_ON(logical == le32_to_cpu(curp->p_idx->ei_block)); + len = EXT_MAX_INDEX(curp->p_hdr) - curp->p_idx; + if (logical > le32_to_cpu(curp->p_idx->ei_block)) { + /* insert after */ + if (curp->p_idx != EXT_LAST_INDEX(curp->p_hdr)) { + len = (len - 1) * sizeof(struct ext4_extent_idx); + len = len < 0 ? 0 : len; + ext_debug("insert new index %d after: %d. " + "move %d from 0x%p to 0x%p\n", + logical, ptr, len, + (curp->p_idx + 1), (curp->p_idx + 2)); + memmove(curp->p_idx + 2, curp->p_idx + 1, len); + } + ix = curp->p_idx + 1; + } else { + /* insert before */ + len = len * sizeof(struct ext4_extent_idx); + len = len < 0 ? 0 : len; + ext_debug("insert new index %d before: %d. " + "move %d from 0x%p to 0x%p\n", + logical, ptr, len, + curp->p_idx, (curp->p_idx + 1)); + memmove(curp->p_idx + 1, curp->p_idx, len); + ix = curp->p_idx; + } + + ix->ei_block = cpu_to_le32(logical); + ext4_idx_store_pblock(ix, ptr); + curp->p_hdr->eh_entries = cpu_to_le16(le16_to_cpu(curp->p_hdr->eh_entries)+1); + + BUG_ON(le16_to_cpu(curp->p_hdr->eh_entries) + > le16_to_cpu(curp->p_hdr->eh_max)); + BUG_ON(ix > EXT_LAST_INDEX(curp->p_hdr)); + + err = ext4_ext_dirty(handle, inode, curp); + ext4_std_error(inode->i_sb, err); + + return err; +} + +/* + * ext4_ext_split: + * inserts new subtree into the path, using free index entry + * at depth @at: + * - allocates all needed blocks (new leaf and all intermediate index blocks) + * - makes decision where to split + * - moves remaining extents and index entries (right to the split point) + * into the newly allocated blocks + * - initializes subtree + */ +static int ext4_ext_split(handle_t *handle, struct inode *inode, + struct ext4_ext_path *path, + struct ext4_extent *newext, int at) +{ + struct buffer_head *bh = NULL; + int depth = ext_depth(inode); + struct ext4_extent_header *neh; + struct ext4_extent_idx *fidx; + struct ext4_extent *ex; + int i = at, k, m, a; + ext4_fsblk_t newblock, oldblock; + __le32 border; + ext4_fsblk_t *ablocks = NULL; /* array of allocated blocks */ + int err = 0; + + /* make decision: where to split? */ + /* FIXME: now decision is simplest: at current extent */ + + /* if current leaf will be split, then we should use + * border from split point */ + BUG_ON(path[depth].p_ext > EXT_MAX_EXTENT(path[depth].p_hdr)); + if (path[depth].p_ext != EXT_MAX_EXTENT(path[depth].p_hdr)) { + border = path[depth].p_ext[1].ee_block; + ext_debug("leaf will be split." + " next leaf starts at %d\n", + le32_to_cpu(border)); + } else { + border = newext->ee_block; + ext_debug("leaf will be added." + " next leaf starts at %d\n", + le32_to_cpu(border)); + } + + /* + * If error occurs, then we break processing + * and mark filesystem read-only. index won't + * be inserted and tree will be in consistent + * state. Next mount will repair buffers too. + */ + + /* + * Get array to track all allocated blocks. + * We need this to handle errors and free blocks + * upon them. + */ + ablocks = kmalloc(sizeof(ext4_fsblk_t) * depth, GFP_NOFS); + if (!ablocks) + return -ENOMEM; + memset(ablocks, 0, sizeof(ext4_fsblk_t) * depth); + + /* allocate all needed blocks */ + ext_debug("allocate %d blocks for indexes/leaf\n", depth - at); + for (a = 0; a < depth - at; a++) { + newblock = ext4_ext_new_block(handle, inode, path, newext, &err); + if (newblock == 0) + goto cleanup; + ablocks[a] = newblock; + } + + /* initialize new leaf */ + newblock = ablocks[--a]; + BUG_ON(newblock == 0); + bh = sb_getblk(inode->i_sb, newblock); + if (!bh) { + err = -EIO; + goto cleanup; + } + lock_buffer(bh); + + if ((err = ext4_journal_get_create_access(handle, bh))) + goto cleanup; + + neh = ext_block_hdr(bh); + neh->eh_entries = 0; + neh->eh_max = cpu_to_le16(ext4_ext_space_block(inode)); + neh->eh_magic = EXT4_EXT_MAGIC; + neh->eh_depth = 0; + ex = EXT_FIRST_EXTENT(neh); + + /* move remainder of path[depth] to the new leaf */ + BUG_ON(path[depth].p_hdr->eh_entries != path[depth].p_hdr->eh_max); + /* start copy from next extent */ + /* TODO: we could do it by single memmove */ + m = 0; + path[depth].p_ext++; + while (path[depth].p_ext <= + EXT_MAX_EXTENT(path[depth].p_hdr)) { + ext_debug("move %d:%llu:%d in new leaf %llu\n", + le32_to_cpu(path[depth].p_ext->ee_block), + ext_pblock(path[depth].p_ext), + le16_to_cpu(path[depth].p_ext->ee_len), + newblock); + /*memmove(ex++, path[depth].p_ext++, + sizeof(struct ext4_extent)); + neh->eh_entries++;*/ + path[depth].p_ext++; + m++; + } + if (m) { + memmove(ex, path[depth].p_ext-m, sizeof(struct ext4_extent)*m); + neh->eh_entries = cpu_to_le16(le16_to_cpu(neh->eh_entries)+m); + } + + set_buffer_uptodate(bh); + unlock_buffer(bh); + + if ((err = ext4_journal_dirty_metadata(handle, bh))) + goto cleanup; + brelse(bh); + bh = NULL; + + /* correct old leaf */ + if (m) { + if ((err = ext4_ext_get_access(handle, inode, path + depth))) + goto cleanup; + path[depth].p_hdr->eh_entries = + cpu_to_le16(le16_to_cpu(path[depth].p_hdr->eh_entries)-m); + if ((err = ext4_ext_dirty(handle, inode, path + depth))) + goto cleanup; + + } + + /* create intermediate indexes */ + k = depth - at - 1; + BUG_ON(k < 0); + if (k) + ext_debug("create %d intermediate indices\n", k); + /* insert new index into current index block */ + /* current depth stored in i var */ + i = depth - 1; + while (k--) { + oldblock = newblock; + newblock = ablocks[--a]; + bh = sb_getblk(inode->i_sb, (ext4_fsblk_t)newblock); + if (!bh) { + err = -EIO; + goto cleanup; + } + lock_buffer(bh); + + if ((err = ext4_journal_get_create_access(handle, bh))) + goto cleanup; + + neh = ext_block_hdr(bh); + neh->eh_entries = cpu_to_le16(1); + neh->eh_magic = EXT4_EXT_MAGIC; + neh->eh_max = cpu_to_le16(ext4_ext_space_block_idx(inode)); + neh->eh_depth = cpu_to_le16(depth - i); + fidx = EXT_FIRST_INDEX(neh); + fidx->ei_block = border; + ext4_idx_store_pblock(fidx, oldblock); + + ext_debug("int.index at %d (block %llu): %lu -> %llu\n", i, + newblock, (unsigned long) le32_to_cpu(border), + oldblock); + /* copy indexes */ + m = 0; + path[i].p_idx++; + + ext_debug("cur 0x%p, last 0x%p\n", path[i].p_idx, + EXT_MAX_INDEX(path[i].p_hdr)); + BUG_ON(EXT_MAX_INDEX(path[i].p_hdr) != + EXT_LAST_INDEX(path[i].p_hdr)); + while (path[i].p_idx <= EXT_MAX_INDEX(path[i].p_hdr)) { + ext_debug("%d: move %d:%d in new index %llu\n", i, + le32_to_cpu(path[i].p_idx->ei_block), + idx_pblock(path[i].p_idx), + newblock); + /*memmove(++fidx, path[i].p_idx++, + sizeof(struct ext4_extent_idx)); + neh->eh_entries++; + BUG_ON(neh->eh_entries > neh->eh_max);*/ + path[i].p_idx++; + m++; + } + if (m) { + memmove(++fidx, path[i].p_idx - m, + sizeof(struct ext4_extent_idx) * m); + neh->eh_entries = + cpu_to_le16(le16_to_cpu(neh->eh_entries) + m); + } + set_buffer_uptodate(bh); + unlock_buffer(bh); + + if ((err = ext4_journal_dirty_metadata(handle, bh))) + goto cleanup; + brelse(bh); + bh = NULL; + + /* correct old index */ + if (m) { + err = ext4_ext_get_access(handle, inode, path + i); + if (err) + goto cleanup; + path[i].p_hdr->eh_entries = cpu_to_le16(le16_to_cpu(path[i].p_hdr->eh_entries)-m); + err = ext4_ext_dirty(handle, inode, path + i); + if (err) + goto cleanup; + } + + i--; + } + + /* insert new index */ + if (err) + goto cleanup; + + err = ext4_ext_insert_index(handle, inode, path + at, + le32_to_cpu(border), newblock); + +cleanup: + if (bh) { + if (buffer_locked(bh)) + unlock_buffer(bh); + brelse(bh); + } + + if (err) { + /* free all allocated blocks in error case */ + for (i = 0; i < depth; i++) { + if (!ablocks[i]) + continue; + ext4_free_blocks(handle, inode, ablocks[i], 1); + } + } + kfree(ablocks); + + return err; +} + +/* + * ext4_ext_grow_indepth: + * implements tree growing procedure: + * - allocates new block + * - moves top-level data (index block or leaf) into the new block + * - initializes new top-level, creating index that points to the + * just created block + */ +static int ext4_ext_grow_indepth(handle_t *handle, struct inode *inode, + struct ext4_ext_path *path, + struct ext4_extent *newext) +{ + struct ext4_ext_path *curp = path; + struct ext4_extent_header *neh; + struct ext4_extent_idx *fidx; + struct buffer_head *bh; + ext4_fsblk_t newblock; + int err = 0; + + newblock = ext4_ext_new_block(handle, inode, path, newext, &err); + if (newblock == 0) + return err; + + bh = sb_getblk(inode->i_sb, newblock); + if (!bh) { + err = -EIO; + ext4_std_error(inode->i_sb, err); + return err; + } + lock_buffer(bh); + + if ((err = ext4_journal_get_create_access(handle, bh))) { + unlock_buffer(bh); + goto out; + } + + /* move top-level index/leaf into new block */ + memmove(bh->b_data, curp->p_hdr, sizeof(EXT4_I(inode)->i_data)); + + /* set size of new block */ + neh = ext_block_hdr(bh); + /* old root could have indexes or leaves + * so calculate e_max right way */ + if (ext_depth(inode)) + neh->eh_max = cpu_to_le16(ext4_ext_space_block_idx(inode)); + else + neh->eh_max = cpu_to_le16(ext4_ext_space_block(inode)); + neh->eh_magic = EXT4_EXT_MAGIC; + set_buffer_uptodate(bh); + unlock_buffer(bh); + + if ((err = ext4_journal_dirty_metadata(handle, bh))) + goto out; + + /* create index in new top-level index: num,max,pointer */ + if ((err = ext4_ext_get_access(handle, inode, curp))) + goto out; + + curp->p_hdr->eh_magic = EXT4_EXT_MAGIC; + curp->p_hdr->eh_max = cpu_to_le16(ext4_ext_space_root_idx(inode)); + curp->p_hdr->eh_entries = cpu_to_le16(1); + curp->p_idx = EXT_FIRST_INDEX(curp->p_hdr); + /* FIXME: it works, but actually path[0] can be index */ + curp->p_idx->ei_block = EXT_FIRST_EXTENT(path[0].p_hdr)->ee_block; + ext4_idx_store_pblock(curp->p_idx, newblock); + + neh = ext_inode_hdr(inode); + fidx = EXT_FIRST_INDEX(neh); + ext_debug("new root: num %d(%d), lblock %d, ptr %llu\n", + le16_to_cpu(neh->eh_entries), le16_to_cpu(neh->eh_max), + le32_to_cpu(fidx->ei_block), idx_pblock(fidx)); + + neh->eh_depth = cpu_to_le16(path->p_depth + 1); + err = ext4_ext_dirty(handle, inode, curp); +out: + brelse(bh); + + return err; +} + +/* + * ext4_ext_create_new_leaf: + * finds empty index and adds new leaf. + * if no free index is found, then it requests in-depth growing. + */ +static int ext4_ext_create_new_leaf(handle_t *handle, struct inode *inode, + struct ext4_ext_path *path, + struct ext4_extent *newext) +{ + struct ext4_ext_path *curp; + int depth, i, err = 0; + +repeat: + i = depth = ext_depth(inode); + + /* walk up to the tree and look for free index entry */ + curp = path + depth; + while (i > 0 && !EXT_HAS_FREE_INDEX(curp)) { + i--; + curp--; + } + + /* we use already allocated block for index block, + * so subsequent data blocks should be contiguous */ + if (EXT_HAS_FREE_INDEX(curp)) { + /* if we found index with free entry, then use that + * entry: create all needed subtree and add new leaf */ + err = ext4_ext_split(handle, inode, path, newext, i); + + /* refill path */ + ext4_ext_drop_refs(path); + path = ext4_ext_find_extent(inode, + le32_to_cpu(newext->ee_block), + path); + if (IS_ERR(path)) + err = PTR_ERR(path); + } else { + /* tree is full, time to grow in depth */ + err = ext4_ext_grow_indepth(handle, inode, path, newext); + if (err) + goto out; + + /* refill path */ + ext4_ext_drop_refs(path); + path = ext4_ext_find_extent(inode, + le32_to_cpu(newext->ee_block), + path); + if (IS_ERR(path)) { + err = PTR_ERR(path); + goto out; + } + + /* + * only first (depth 0 -> 1) produces free space; + * in all other cases we have to split the grown tree + */ + depth = ext_depth(inode); + if (path[depth].p_hdr->eh_entries == path[depth].p_hdr->eh_max) { + /* now we need to split */ + goto repeat; + } + } + +out: + return err; +} + +/* + * ext4_ext_next_allocated_block: + * returns allocated block in subsequent extent or EXT_MAX_BLOCK. + * NOTE: it considers block number from index entry as + * allocated block. Thus, index entries have to be consistent + * with leaves. + */ +static unsigned long +ext4_ext_next_allocated_block(struct ext4_ext_path *path) +{ + int depth; + + BUG_ON(path == NULL); + depth = path->p_depth; + + if (depth == 0 && path->p_ext == NULL) + return EXT_MAX_BLOCK; + + while (depth >= 0) { + if (depth == path->p_depth) { + /* leaf */ + if (path[depth].p_ext != + EXT_LAST_EXTENT(path[depth].p_hdr)) + return le32_to_cpu(path[depth].p_ext[1].ee_block); + } else { + /* index */ + if (path[depth].p_idx != + EXT_LAST_INDEX(path[depth].p_hdr)) + return le32_to_cpu(path[depth].p_idx[1].ei_block); + } + depth--; + } + + return EXT_MAX_BLOCK; +} + +/* + * ext4_ext_next_leaf_block: + * returns first allocated block from next leaf or EXT_MAX_BLOCK + */ +static unsigned ext4_ext_next_leaf_block(struct inode *inode, + struct ext4_ext_path *path) +{ + int depth; + + BUG_ON(path == NULL); + depth = path->p_depth; + + /* zero-tree has no leaf blocks at all */ + if (depth == 0) + return EXT_MAX_BLOCK; + + /* go to index block */ + depth--; + + while (depth >= 0) { + if (path[depth].p_idx != + EXT_LAST_INDEX(path[depth].p_hdr)) + return le32_to_cpu(path[depth].p_idx[1].ei_block); + depth--; + } + + return EXT_MAX_BLOCK; +} + +/* + * ext4_ext_correct_indexes: + * if leaf gets modified and modified extent is first in the leaf, + * then we have to correct all indexes above. + * TODO: do we need to correct tree in all cases? + */ +int ext4_ext_correct_indexes(handle_t *handle, struct inode *inode, + struct ext4_ext_path *path) +{ + struct ext4_extent_header *eh; + int depth = ext_depth(inode); + struct ext4_extent *ex; + __le32 border; + int k, err = 0; + + eh = path[depth].p_hdr; + ex = path[depth].p_ext; + BUG_ON(ex == NULL); + BUG_ON(eh == NULL); + + if (depth == 0) { + /* there is no tree at all */ + return 0; + } + + if (ex != EXT_FIRST_EXTENT(eh)) { + /* we correct tree if first leaf got modified only */ + return 0; + } + + /* + * TODO: we need correction if border is smaller than current one + */ + k = depth - 1; + border = path[depth].p_ext->ee_block; + if ((err = ext4_ext_get_access(handle, inode, path + k))) + return err; + path[k].p_idx->ei_block = border; + if ((err = ext4_ext_dirty(handle, inode, path + k))) + return err; + + while (k--) { + /* change all left-side indexes */ + if (path[k+1].p_idx != EXT_FIRST_INDEX(path[k+1].p_hdr)) + break; + if ((err = ext4_ext_get_access(handle, inode, path + k))) + break; + path[k].p_idx->ei_block = border; + if ((err = ext4_ext_dirty(handle, inode, path + k))) + break; + } + + return err; +} + +static int inline +ext4_can_extents_be_merged(struct inode *inode, struct ext4_extent *ex1, + struct ext4_extent *ex2) +{ + if (le32_to_cpu(ex1->ee_block) + le16_to_cpu(ex1->ee_len) != + le32_to_cpu(ex2->ee_block)) + return 0; + + /* + * To allow future support for preallocated extents to be added + * as an RO_COMPAT feature, refuse to merge to extents if + * this can result in the top bit of ee_len being set. + */ + if (le16_to_cpu(ex1->ee_len) + le16_to_cpu(ex2->ee_len) > EXT_MAX_LEN) + return 0; +#ifdef AGRESSIVE_TEST + if (le16_to_cpu(ex1->ee_len) >= 4) + return 0; +#endif + + if (ext_pblock(ex1) + le16_to_cpu(ex1->ee_len) == ext_pblock(ex2)) + return 1; + return 0; +} + +/* + * ext4_ext_insert_extent: + * tries to merge requsted extent into the existing extent or + * inserts requested extent as new one into the tree, + * creating new leaf in the no-space case. + */ +int ext4_ext_insert_extent(handle_t *handle, struct inode *inode, + struct ext4_ext_path *path, + struct ext4_extent *newext) +{ + struct ext4_extent_header * eh; + struct ext4_extent *ex, *fex; + struct ext4_extent *nearex; /* nearest extent */ + struct ext4_ext_path *npath = NULL; + int depth, len, err, next; + + BUG_ON(newext->ee_len == 0); + depth = ext_depth(inode); + ex = path[depth].p_ext; + BUG_ON(path[depth].p_hdr == NULL); + + /* try to insert block into found extent and return */ + if (ex && ext4_can_extents_be_merged(inode, ex, newext)) { + ext_debug("append %d block to %d:%d (from %llu)\n", + le16_to_cpu(newext->ee_len), + le32_to_cpu(ex->ee_block), + le16_to_cpu(ex->ee_len), ext_pblock(ex)); + if ((err = ext4_ext_get_access(handle, inode, path + depth))) + return err; + ex->ee_len = cpu_to_le16(le16_to_cpu(ex->ee_len) + + le16_to_cpu(newext->ee_len)); + eh = path[depth].p_hdr; + nearex = ex; + goto merge; + } + +repeat: + depth = ext_depth(inode); + eh = path[depth].p_hdr; + if (le16_to_cpu(eh->eh_entries) < le16_to_cpu(eh->eh_max)) + goto has_space; + + /* probably next leaf has space for us? */ + fex = EXT_LAST_EXTENT(eh); + next = ext4_ext_next_leaf_block(inode, path); + if (le32_to_cpu(newext->ee_block) > le32_to_cpu(fex->ee_block) + && next != EXT_MAX_BLOCK) { + ext_debug("next leaf block - %d\n", next); + BUG_ON(npath != NULL); + npath = ext4_ext_find_extent(inode, next, NULL); + if (IS_ERR(npath)) + return PTR_ERR(npath); + BUG_ON(npath->p_depth != path->p_depth); + eh = npath[depth].p_hdr; + if (le16_to_cpu(eh->eh_entries) < le16_to_cpu(eh->eh_max)) { + ext_debug("next leaf isnt full(%d)\n", + le16_to_cpu(eh->eh_entries)); + path = npath; + goto repeat; + } + ext_debug("next leaf has no free space(%d,%d)\n", + le16_to_cpu(eh->eh_entries), le16_to_cpu(eh->eh_max)); + } + + /* + * There is no free space in the found leaf. + * We're gonna add a new leaf in the tree. + */ + err = ext4_ext_create_new_leaf(handle, inode, path, newext); + if (err) + goto cleanup; + depth = ext_depth(inode); + eh = path[depth].p_hdr; + +has_space: + nearex = path[depth].p_ext; + + if ((err = ext4_ext_get_access(handle, inode, path + depth))) + goto cleanup; + + if (!nearex) { + /* there is no extent in this leaf, create first one */ + ext_debug("first extent in the leaf: %d:%llu:%d\n", + le32_to_cpu(newext->ee_block), + ext_pblock(newext), + le16_to_cpu(newext->ee_len)); + path[depth].p_ext = EXT_FIRST_EXTENT(eh); + } else if (le32_to_cpu(newext->ee_block) + > le32_to_cpu(nearex->ee_block)) { +/* BUG_ON(newext->ee_block == nearex->ee_block); */ + if (nearex != EXT_LAST_EXTENT(eh)) { + len = EXT_MAX_EXTENT(eh) - nearex; + len = (len - 1) * sizeof(struct ext4_extent); + len = len < 0 ? 0 : len; + ext_debug("insert %d:%llu:%d after: nearest 0x%p, " + "move %d from 0x%p to 0x%p\n", + le32_to_cpu(newext->ee_block), + ext_pblock(newext), + le16_to_cpu(newext->ee_len), + nearex, len, nearex + 1, nearex + 2); + memmove(nearex + 2, nearex + 1, len); + } + path[depth].p_ext = nearex + 1; + } else { + BUG_ON(newext->ee_block == nearex->ee_block); + len = (EXT_MAX_EXTENT(eh) - nearex) * sizeof(struct ext4_extent); + len = len < 0 ? 0 : len; + ext_debug("insert %d:%llu:%d before: nearest 0x%p, " + "move %d from 0x%p to 0x%p\n", + le32_to_cpu(newext->ee_block), + ext_pblock(newext), + le16_to_cpu(newext->ee_len), + nearex, len, nearex + 1, nearex + 2); + memmove(nearex + 1, nearex, len); + path[depth].p_ext = nearex; + } + + eh->eh_entries = cpu_to_le16(le16_to_cpu(eh->eh_entries)+1); + nearex = path[depth].p_ext; + nearex->ee_block = newext->ee_block; + nearex->ee_start = newext->ee_start; + nearex->ee_start_hi = newext->ee_start_hi; + nearex->ee_len = newext->ee_len; + +merge: + /* try to merge extents to the right */ + while (nearex < EXT_LAST_EXTENT(eh)) { + if (!ext4_can_extents_be_merged(inode, nearex, nearex + 1)) + break; + /* merge with next extent! */ + nearex->ee_len = cpu_to_le16(le16_to_cpu(nearex->ee_len) + + le16_to_cpu(nearex[1].ee_len)); + if (nearex + 1 < EXT_LAST_EXTENT(eh)) { + len = (EXT_LAST_EXTENT(eh) - nearex - 1) + * sizeof(struct ext4_extent); + memmove(nearex + 1, nearex + 2, len); + } + eh->eh_entries = cpu_to_le16(le16_to_cpu(eh->eh_entries)-1); + BUG_ON(eh->eh_entries == 0); + } + + /* try to merge extents to the left */ + + /* time to correct all indexes above */ + err = ext4_ext_correct_indexes(handle, inode, path); + if (err) + goto cleanup; + + err = ext4_ext_dirty(handle, inode, path + depth); + +cleanup: + if (npath) { + ext4_ext_drop_refs(npath); + kfree(npath); + } + ext4_ext_tree_changed(inode); + ext4_ext_invalidate_cache(inode); + return err; +} + +int ext4_ext_walk_space(struct inode *inode, unsigned long block, + unsigned long num, ext_prepare_callback func, + void *cbdata) +{ + struct ext4_ext_path *path = NULL; + struct ext4_ext_cache cbex; + struct ext4_extent *ex; + unsigned long next, start = 0, end = 0; + unsigned long last = block + num; + int depth, exists, err = 0; + + BUG_ON(func == NULL); + BUG_ON(inode == NULL); + + while (block < last && block != EXT_MAX_BLOCK) { + num = last - block; + /* find extent for this block */ + path = ext4_ext_find_extent(inode, block, path); + if (IS_ERR(path)) { + err = PTR_ERR(path); + path = NULL; + break; + } + + depth = ext_depth(inode); + BUG_ON(path[depth].p_hdr == NULL); + ex = path[depth].p_ext; + next = ext4_ext_next_allocated_block(path); + + exists = 0; + if (!ex) { + /* there is no extent yet, so try to allocate + * all requested space */ + start = block; + end = block + num; + } else if (le32_to_cpu(ex->ee_block) > block) { + /* need to allocate space before found extent */ + start = block; + end = le32_to_cpu(ex->ee_block); + if (block + num < end) + end = block + num; + } else if (block >= + le32_to_cpu(ex->ee_block) + le16_to_cpu(ex->ee_len)) { + /* need to allocate space after found extent */ + start = block; + end = block + num; + if (end >= next) + end = next; + } else if (block >= le32_to_cpu(ex->ee_block)) { + /* + * some part of requested space is covered + * by found extent + */ + start = block; + end = le32_to_cpu(ex->ee_block) + le16_to_cpu(ex->ee_len); + if (block + num < end) + end = block + num; + exists = 1; + } else { + BUG(); + } + BUG_ON(end <= start); + + if (!exists) { + cbex.ec_block = start; + cbex.ec_len = end - start; + cbex.ec_start = 0; + cbex.ec_type = EXT4_EXT_CACHE_GAP; + } else { + cbex.ec_block = le32_to_cpu(ex->ee_block); + cbex.ec_len = le16_to_cpu(ex->ee_len); + cbex.ec_start = ext_pblock(ex); + cbex.ec_type = EXT4_EXT_CACHE_EXTENT; + } + + BUG_ON(cbex.ec_len == 0); + err = func(inode, path, &cbex, cbdata); + ext4_ext_drop_refs(path); + + if (err < 0) + break; + if (err == EXT_REPEAT) + continue; + else if (err == EXT_BREAK) { + err = 0; + break; + } + + if (ext_depth(inode) != depth) { + /* depth was changed. we have to realloc path */ + kfree(path); + path = NULL; + } + + block = cbex.ec_block + cbex.ec_len; + } + + if (path) { + ext4_ext_drop_refs(path); + kfree(path); + } + + return err; +} + +static inline void +ext4_ext_put_in_cache(struct inode *inode, __u32 block, + __u32 len, __u32 start, int type) +{ + struct ext4_ext_cache *cex; + BUG_ON(len == 0); + cex = &EXT4_I(inode)->i_cached_extent; + cex->ec_type = type; + cex->ec_block = block; + cex->ec_len = len; + cex->ec_start = start; +} + +/* + * ext4_ext_put_gap_in_cache: + * calculate boundaries of the gap that the requested block fits into + * and cache this gap + */ +static inline void +ext4_ext_put_gap_in_cache(struct inode *inode, struct ext4_ext_path *path, + unsigned long block) +{ + int depth = ext_depth(inode); + unsigned long lblock, len; + struct ext4_extent *ex; + + ex = path[depth].p_ext; + if (ex == NULL) { + /* there is no extent yet, so gap is [0;-] */ + lblock = 0; + len = EXT_MAX_BLOCK; + ext_debug("cache gap(whole file):"); + } else if (block < le32_to_cpu(ex->ee_block)) { + lblock = block; + len = le32_to_cpu(ex->ee_block) - block; + ext_debug("cache gap(before): %lu [%lu:%lu]", + (unsigned long) block, + (unsigned long) le32_to_cpu(ex->ee_block), + (unsigned long) le16_to_cpu(ex->ee_len)); + } else if (block >= le32_to_cpu(ex->ee_block) + + le16_to_cpu(ex->ee_len)) { + lblock = le32_to_cpu(ex->ee_block) + + le16_to_cpu(ex->ee_len); + len = ext4_ext_next_allocated_block(path); + ext_debug("cache gap(after): [%lu:%lu] %lu", + (unsigned long) le32_to_cpu(ex->ee_block), + (unsigned long) le16_to_cpu(ex->ee_len), + (unsigned long) block); + BUG_ON(len == lblock); + len = len - lblock; + } else { + lblock = len = 0; + BUG(); + } + + ext_debug(" -> %lu:%lu\n", (unsigned long) lblock, len); + ext4_ext_put_in_cache(inode, lblock, len, 0, EXT4_EXT_CACHE_GAP); +} + +static inline int +ext4_ext_in_cache(struct inode *inode, unsigned long block, + struct ext4_extent *ex) +{ + struct ext4_ext_cache *cex; + + cex = &EXT4_I(inode)->i_cached_extent; + + /* has cache valid data? */ + if (cex->ec_type == EXT4_EXT_CACHE_NO) + return EXT4_EXT_CACHE_NO; + + BUG_ON(cex->ec_type != EXT4_EXT_CACHE_GAP && + cex->ec_type != EXT4_EXT_CACHE_EXTENT); + if (block >= cex->ec_block && block < cex->ec_block + cex->ec_len) { + ex->ee_block = cpu_to_le32(cex->ec_block); + ext4_ext_store_pblock(ex, cex->ec_start); + ex->ee_len = cpu_to_le16(cex->ec_len); + ext_debug("%lu cached by %lu:%lu:%llu\n", + (unsigned long) block, + (unsigned long) cex->ec_block, + (unsigned long) cex->ec_len, + cex->ec_start); + return cex->ec_type; + } + + /* not in cache */ + return EXT4_EXT_CACHE_NO; +} + +/* + * ext4_ext_rm_idx: + * removes index from the index block. + * It's used in truncate case only, thus all requests are for + * last index in the block only. + */ +int ext4_ext_rm_idx(handle_t *handle, struct inode *inode, + struct ext4_ext_path *path) +{ + struct buffer_head *bh; + int err; + ext4_fsblk_t leaf; + + /* free index block */ + path--; + leaf = idx_pblock(path->p_idx); + BUG_ON(path->p_hdr->eh_entries == 0); + if ((err = ext4_ext_get_access(handle, inode, path))) + return err; + path->p_hdr->eh_entries = cpu_to_le16(le16_to_cpu(path->p_hdr->eh_entries)-1); + if ((err = ext4_ext_dirty(handle, inode, path))) + return err; + ext_debug("index is empty, remove it, free block %llu\n", leaf); + bh = sb_find_get_block(inode->i_sb, leaf); + ext4_forget(handle, 1, inode, bh, leaf); + ext4_free_blocks(handle, inode, leaf, 1); + return err; +} + +/* + * ext4_ext_calc_credits_for_insert: + * This routine returns max. credits that the extent tree can consume. + * It should be OK for low-performance paths like ->writepage() + * To allow many writing processes to fit into a single transaction, + * the caller should calculate credits under truncate_mutex and + * pass the actual path. + */ +int inline ext4_ext_calc_credits_for_insert(struct inode *inode, + struct ext4_ext_path *path) +{ + int depth, needed; + + if (path) { + /* probably there is space in leaf? */ + depth = ext_depth(inode); + if (le16_to_cpu(path[depth].p_hdr->eh_entries) + < le16_to_cpu(path[depth].p_hdr->eh_max)) + return 1; + } + + /* + * given 32-bit logical block (4294967296 blocks), max. tree + * can be 4 levels in depth -- 4 * 340^4 == 53453440000. + * Let's also add one more level for imbalance. + */ + depth = 5; + + /* allocation of new data block(s) */ + needed = 2; + + /* + * tree can be full, so it would need to grow in depth: + * allocation + old root + new root + */ + needed += 2 + 1 + 1; + + /* + * Index split can happen, we would need: + * allocate intermediate indexes (bitmap + group) + * + change two blocks at each level, but root (already included) + */ + needed = (depth * 2) + (depth * 2); + + /* any allocation modifies superblock */ + needed += 1; + + return needed; +} + +static int ext4_remove_blocks(handle_t *handle, struct inode *inode, + struct ext4_extent *ex, + unsigned long from, unsigned long to) +{ + struct buffer_head *bh; + int i; + +#ifdef EXTENTS_STATS + { + struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); + unsigned short ee_len = le16_to_cpu(ex->ee_len); + spin_lock(&sbi->s_ext_stats_lock); + sbi->s_ext_blocks += ee_len; + sbi->s_ext_extents++; + if (ee_len < sbi->s_ext_min) + sbi->s_ext_min = ee_len; + if (ee_len > sbi->s_ext_max) + sbi->s_ext_max = ee_len; + if (ext_depth(inode) > sbi->s_depth_max) + sbi->s_depth_max = ext_depth(inode); + spin_unlock(&sbi->s_ext_stats_lock); + } +#endif + if (from >= le32_to_cpu(ex->ee_block) + && to == le32_to_cpu(ex->ee_block) + le16_to_cpu(ex->ee_len) - 1) { + /* tail removal */ + unsigned long num; + ext4_fsblk_t start; + num = le32_to_cpu(ex->ee_block) + le16_to_cpu(ex->ee_len) - from; + start = ext_pblock(ex) + le16_to_cpu(ex->ee_len) - num; + ext_debug("free last %lu blocks starting %llu\n", num, start); + for (i = 0; i < num; i++) { + bh = sb_find_get_block(inode->i_sb, start + i); + ext4_forget(handle, 0, inode, bh, start + i); + } + ext4_free_blocks(handle, inode, start, num); + } else if (from == le32_to_cpu(ex->ee_block) + && to <= le32_to_cpu(ex->ee_block) + le16_to_cpu(ex->ee_len) - 1) { + printk("strange request: removal %lu-%lu from %u:%u\n", + from, to, le32_to_cpu(ex->ee_block), le16_to_cpu(ex->ee_len)); + } else { + printk("strange request: removal(2) %lu-%lu from %u:%u\n", + from, to, le32_to_cpu(ex->ee_block), le16_to_cpu(ex->ee_len)); + } + return 0; +} + +static int +ext4_ext_rm_leaf(handle_t *handle, struct inode *inode, + struct ext4_ext_path *path, unsigned long start) +{ + int err = 0, correct_index = 0; + int depth = ext_depth(inode), credits; + struct ext4_extent_header *eh; + unsigned a, b, block, num; + unsigned long ex_ee_block; + unsigned short ex_ee_len; + struct ext4_extent *ex; + + ext_debug("truncate since %lu in leaf\n", start); + if (!path[depth].p_hdr) + path[depth].p_hdr = ext_block_hdr(path[depth].p_bh); + eh = path[depth].p_hdr; + BUG_ON(eh == NULL); + BUG_ON(le16_to_cpu(eh->eh_entries) > le16_to_cpu(eh->eh_max)); + BUG_ON(eh->eh_magic != EXT4_EXT_MAGIC); + + /* find where to start removing */ + ex = EXT_LAST_EXTENT(eh); + + ex_ee_block = le32_to_cpu(ex->ee_block); + ex_ee_len = le16_to_cpu(ex->ee_len); + + while (ex >= EXT_FIRST_EXTENT(eh) && + ex_ee_block + ex_ee_len > start) { + ext_debug("remove ext %lu:%u\n", ex_ee_block, ex_ee_len); + path[depth].p_ext = ex; + + a = ex_ee_block > start ? ex_ee_block : start; + b = ex_ee_block + ex_ee_len - 1 < EXT_MAX_BLOCK ? + ex_ee_block + ex_ee_len - 1 : EXT_MAX_BLOCK; + + ext_debug(" border %u:%u\n", a, b); + + if (a != ex_ee_block && b != ex_ee_block + ex_ee_len - 1) { + block = 0; + num = 0; + BUG(); + } else if (a != ex_ee_block) { + /* remove tail of the extent */ + block = ex_ee_block; + num = a - block; + } else if (b != ex_ee_block + ex_ee_len - 1) { + /* remove head of the extent */ + block = a; + num = b - a; + /* there is no "make a hole" API yet */ + BUG(); + } else { + /* remove whole extent: excellent! */ + block = ex_ee_block; + num = 0; + BUG_ON(a != ex_ee_block); + BUG_ON(b != ex_ee_block + ex_ee_len - 1); + } + + /* at present, extent can't cross block group: */ + /* leaf + bitmap + group desc + sb + inode */ + credits = 5; + if (ex == EXT_FIRST_EXTENT(eh)) { + correct_index = 1; + credits += (ext_depth(inode)) + 1; + } +#ifdef CONFIG_QUOTA + credits += 2 * EXT4_QUOTA_TRANS_BLOCKS(inode->i_sb); +#endif + + handle = ext4_ext_journal_restart(handle, credits); + if (IS_ERR(handle)) { + err = PTR_ERR(handle); + goto out; + } + + err = ext4_ext_get_access(handle, inode, path + depth); + if (err) + goto out; + + err = ext4_remove_blocks(handle, inode, ex, a, b); + if (err) + goto out; + + if (num == 0) { + /* this extent is removed; mark slot entirely unused */ + ext4_ext_store_pblock(ex, 0); + eh->eh_entries = cpu_to_le16(le16_to_cpu(eh->eh_entries)-1); + } + + ex->ee_block = cpu_to_le32(block); + ex->ee_len = cpu_to_le16(num); + + err = ext4_ext_dirty(handle, inode, path + depth); + if (err) + goto out; + + ext_debug("new extent: %u:%u:%llu\n", block, num, + ext_pblock(ex)); + ex--; + ex_ee_block = le32_to_cpu(ex->ee_block); + ex_ee_len = le16_to_cpu(ex->ee_len); + } + + if (correct_index && eh->eh_entries) + err = ext4_ext_correct_indexes(handle, inode, path); + + /* if this leaf is free, then we should + * remove it from index block above */ + if (err == 0 && eh->eh_entries == 0 && path[depth].p_bh != NULL) + err = ext4_ext_rm_idx(handle, inode, path + depth); + +out: + return err; +} + +/* + * ext4_ext_more_to_rm: + * returns 1 if current index has to be freed (even partial) + */ +static int inline +ext4_ext_more_to_rm(struct ext4_ext_path *path) +{ + BUG_ON(path->p_idx == NULL); + + if (path->p_idx < EXT_FIRST_INDEX(path->p_hdr)) + return 0; + + /* + * if truncate on deeper level happened, it wasn't partial, + * so we have to consider current index for truncation + */ + if (le16_to_cpu(path->p_hdr->eh_entries) == path->p_block) + return 0; + return 1; +} + +int ext4_ext_remove_space(struct inode *inode, unsigned long start) +{ + struct super_block *sb = inode->i_sb; + int depth = ext_depth(inode); + struct ext4_ext_path *path; + handle_t *handle; + int i = 0, err = 0; + + ext_debug("truncate since %lu\n", start); + + /* probably first extent we're gonna free will be last in block */ + handle = ext4_journal_start(inode, depth + 1); + if (IS_ERR(handle)) + return PTR_ERR(handle); + + ext4_ext_invalidate_cache(inode); + + /* + * We start scanning from right side, freeing all the blocks + * after i_size and walking into the tree depth-wise. + */ + path = kmalloc(sizeof(struct ext4_ext_path) * (depth + 1), GFP_KERNEL); + if (path == NULL) { + ext4_journal_stop(handle); + return -ENOMEM; + } + memset(path, 0, sizeof(struct ext4_ext_path) * (depth + 1)); + path[0].p_hdr = ext_inode_hdr(inode); + if (ext4_ext_check_header(__FUNCTION__, inode, path[0].p_hdr)) { + err = -EIO; + goto out; + } + path[0].p_depth = depth; + + while (i >= 0 && err == 0) { + if (i == depth) { + /* this is leaf block */ + err = ext4_ext_rm_leaf(handle, inode, path, start); + /* root level has p_bh == NULL, brelse() eats this */ + brelse(path[i].p_bh); + path[i].p_bh = NULL; + i--; + continue; + } + + /* this is index block */ + if (!path[i].p_hdr) { + ext_debug("initialize header\n"); + path[i].p_hdr = ext_block_hdr(path[i].p_bh); + if (ext4_ext_check_header(__FUNCTION__, inode, + path[i].p_hdr)) { + err = -EIO; + goto out; + } + } + + BUG_ON(le16_to_cpu(path[i].p_hdr->eh_entries) + > le16_to_cpu(path[i].p_hdr->eh_max)); + BUG_ON(path[i].p_hdr->eh_magic != EXT4_EXT_MAGIC); + + if (!path[i].p_idx) { + /* this level hasn't been touched yet */ + path[i].p_idx = EXT_LAST_INDEX(path[i].p_hdr); + path[i].p_block = le16_to_cpu(path[i].p_hdr->eh_entries)+1; + ext_debug("init index ptr: hdr 0x%p, num %d\n", + path[i].p_hdr, + le16_to_cpu(path[i].p_hdr->eh_entries)); + } else { + /* we were already here, see at next index */ + path[i].p_idx--; + } + + ext_debug("level %d - index, first 0x%p, cur 0x%p\n", + i, EXT_FIRST_INDEX(path[i].p_hdr), + path[i].p_idx); + if (ext4_ext_more_to_rm(path + i)) { + /* go to the next level */ + ext_debug("move to level %d (block %llu)\n", + i + 1, idx_pblock(path[i].p_idx)); + memset(path + i + 1, 0, sizeof(*path)); + path[i+1].p_bh = + sb_bread(sb, idx_pblock(path[i].p_idx)); + if (!path[i+1].p_bh) { + /* should we reset i_size? */ + err = -EIO; + break; + } + + /* save actual number of indexes since this + * number is changed at the next iteration */ + path[i].p_block = le16_to_cpu(path[i].p_hdr->eh_entries); + i++; + } else { + /* we finished processing this index, go up */ + if (path[i].p_hdr->eh_entries == 0 && i > 0) { + /* index is empty, remove it; + * handle must be already prepared by the + * truncatei_leaf() */ + err = ext4_ext_rm_idx(handle, inode, path + i); + } + /* root level has p_bh == NULL, brelse() eats this */ + brelse(path[i].p_bh); + path[i].p_bh = NULL; + i--; + ext_debug("return to level %d\n", i); + } + } + + /* TODO: flexible tree reduction should be here */ + if (path->p_hdr->eh_entries == 0) { + /* + * truncate to zero freed all the tree, + * so we need to correct eh_depth + */ + err = ext4_ext_get_access(handle, inode, path); + if (err == 0) { + ext_inode_hdr(inode)->eh_depth = 0; + ext_inode_hdr(inode)->eh_max = + cpu_to_le16(ext4_ext_space_root(inode)); + err = ext4_ext_dirty(handle, inode, path); + } + } +out: + ext4_ext_tree_changed(inode); + ext4_ext_drop_refs(path); + kfree(path); + ext4_journal_stop(handle); + + return err; +} + +/* + * called at mount time + */ +void ext4_ext_init(struct super_block *sb) +{ + /* + * possible initialization would be here + */ + + if (test_opt(sb, EXTENTS)) { + printk("EXT4-fs: file extents enabled"); +#ifdef AGRESSIVE_TEST + printk(", agressive tests"); +#endif +#ifdef CHECK_BINSEARCH + printk(", check binsearch"); +#endif +#ifdef EXTENTS_STATS + printk(", stats"); +#endif + printk("\n"); +#ifdef EXTENTS_STATS + spin_lock_init(&EXT4_SB(sb)->s_ext_stats_lock); + EXT4_SB(sb)->s_ext_min = 1 << 30; + EXT4_SB(sb)->s_ext_max = 0; +#endif + } +} + +/* + * called at umount time + */ +void ext4_ext_release(struct super_block *sb) +{ + if (!test_opt(sb, EXTENTS)) + return; + +#ifdef EXTENTS_STATS + if (EXT4_SB(sb)->s_ext_blocks && EXT4_SB(sb)->s_ext_extents) { + struct ext4_sb_info *sbi = EXT4_SB(sb); + printk(KERN_ERR "EXT4-fs: %lu blocks in %lu extents (%lu ave)\n", + sbi->s_ext_blocks, sbi->s_ext_extents, + sbi->s_ext_blocks / sbi->s_ext_extents); + printk(KERN_ERR "EXT4-fs: extents: %lu min, %lu max, max depth %lu\n", + sbi->s_ext_min, sbi->s_ext_max, sbi->s_depth_max); + } +#endif +} + +int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, + ext4_fsblk_t iblock, + unsigned long max_blocks, struct buffer_head *bh_result, + int create, int extend_disksize) +{ + struct ext4_ext_path *path = NULL; + struct ext4_extent newex, *ex; + ext4_fsblk_t goal, newblock; + int err = 0, depth; + unsigned long allocated = 0; + + __clear_bit(BH_New, &bh_result->b_state); + ext_debug("blocks %d/%lu requested for inode %u\n", (int) iblock, + max_blocks, (unsigned) inode->i_ino); + mutex_lock(&EXT4_I(inode)->truncate_mutex); + + /* check in cache */ + if ((goal = ext4_ext_in_cache(inode, iblock, &newex))) { + if (goal == EXT4_EXT_CACHE_GAP) { + if (!create) { + /* block isn't allocated yet and + * user doesn't want to allocate it */ + goto out2; + } + /* we should allocate requested block */ + } else if (goal == EXT4_EXT_CACHE_EXTENT) { + /* block is already allocated */ + newblock = iblock + - le32_to_cpu(newex.ee_block) + + ext_pblock(&newex); + /* number of remaining blocks in the extent */ + allocated = le16_to_cpu(newex.ee_len) - + (iblock - le32_to_cpu(newex.ee_block)); + goto out; + } else { + BUG(); + } + } + + /* find extent for this block */ + path = ext4_ext_find_extent(inode, iblock, NULL); + if (IS_ERR(path)) { + err = PTR_ERR(path); + path = NULL; + goto out2; + } + + depth = ext_depth(inode); + + /* + * consistent leaf must not be empty; + * this situation is possible, though, _during_ tree modification; + * this is why assert can't be put in ext4_ext_find_extent() + */ + BUG_ON(path[depth].p_ext == NULL && depth != 0); + + if ((ex = path[depth].p_ext)) { + unsigned long ee_block = le32_to_cpu(ex->ee_block); + ext4_fsblk_t ee_start = ext_pblock(ex); + unsigned short ee_len = le16_to_cpu(ex->ee_len); + + /* + * Allow future support for preallocated extents to be added + * as an RO_COMPAT feature: + * Uninitialized extents are treated as holes, except that + * we avoid (fail) allocating new blocks during a write. + */ + if (ee_len > EXT_MAX_LEN) + goto out2; + /* if found extent covers block, simply return it */ + if (iblock >= ee_block && iblock < ee_block + ee_len) { + newblock = iblock - ee_block + ee_start; + /* number of remaining blocks in the extent */ + allocated = ee_len - (iblock - ee_block); + ext_debug("%d fit into %lu:%d -> %llu\n", (int) iblock, + ee_block, ee_len, newblock); + ext4_ext_put_in_cache(inode, ee_block, ee_len, + ee_start, EXT4_EXT_CACHE_EXTENT); + goto out; + } + } + + /* + * requested block isn't allocated yet; + * we couldn't try to create block if create flag is zero + */ + if (!create) { + /* put just found gap into cache to speed up + * subsequent requests */ + ext4_ext_put_gap_in_cache(inode, path, iblock); + goto out2; + } + /* + * Okay, we need to do block allocation. Lazily initialize the block + * allocation info here if necessary. + */ + if (S_ISREG(inode->i_mode) && (!EXT4_I(inode)->i_block_alloc_info)) + ext4_init_block_alloc_info(inode); + + /* allocate new block */ + goal = ext4_ext_find_goal(inode, path, iblock); + allocated = max_blocks; + newblock = ext4_new_blocks(handle, inode, goal, &allocated, &err); + if (!newblock) + goto out2; + ext_debug("allocate new block: goal %llu, found %llu/%lu\n", + goal, newblock, allocated); + + /* try to insert new extent into found leaf and return */ + newex.ee_block = cpu_to_le32(iblock); + ext4_ext_store_pblock(&newex, newblock); + newex.ee_len = cpu_to_le16(allocated); + err = ext4_ext_insert_extent(handle, inode, path, &newex); + if (err) + goto out2; + + if (extend_disksize && inode->i_size > EXT4_I(inode)->i_disksize) + EXT4_I(inode)->i_disksize = inode->i_size; + + /* previous routine could use block we allocated */ + newblock = ext_pblock(&newex); + __set_bit(BH_New, &bh_result->b_state); + + ext4_ext_put_in_cache(inode, iblock, allocated, newblock, + EXT4_EXT_CACHE_EXTENT); +out: + if (allocated > max_blocks) + allocated = max_blocks; + ext4_ext_show_leaf(inode, path); + __set_bit(BH_Mapped, &bh_result->b_state); + bh_result->b_bdev = inode->i_sb->s_bdev; + bh_result->b_blocknr = newblock; +out2: + if (path) { + ext4_ext_drop_refs(path); + kfree(path); + } + mutex_unlock(&EXT4_I(inode)->truncate_mutex); + + return err ? err : allocated; +} + +void ext4_ext_truncate(struct inode * inode, struct page *page) +{ + struct address_space *mapping = inode->i_mapping; + struct super_block *sb = inode->i_sb; + unsigned long last_block; + handle_t *handle; + int err = 0; + + /* + * probably first extent we're gonna free will be last in block + */ + err = ext4_writepage_trans_blocks(inode) + 3; + handle = ext4_journal_start(inode, err); + if (IS_ERR(handle)) { + if (page) { + clear_highpage(page); + flush_dcache_page(page); + unlock_page(page); + page_cache_release(page); + } + return; + } + + if (page) + ext4_block_truncate_page(handle, page, mapping, inode->i_size); + + mutex_lock(&EXT4_I(inode)->truncate_mutex); + ext4_ext_invalidate_cache(inode); + + /* + * TODO: optimization is possible here. + * Probably we need not scan at all, + * because page truncation is enough. + */ + if (ext4_orphan_add(handle, inode)) + goto out_stop; + + /* we have to know where to truncate from in crash case */ + EXT4_I(inode)->i_disksize = inode->i_size; + ext4_mark_inode_dirty(handle, inode); + + last_block = (inode->i_size + sb->s_blocksize - 1) + >> EXT4_BLOCK_SIZE_BITS(sb); + err = ext4_ext_remove_space(inode, last_block); + + /* In a multi-transaction truncate, we only make the final + * transaction synchronous. */ + if (IS_SYNC(inode)) + handle->h_sync = 1; + +out_stop: + /* + * If this was a simple ftruncate() and the file will remain alive, + * then we need to clear up the orphan record which we created above. + * However, if this was a real unlink then we were called by + * ext4_delete_inode(), and we allow that function to clean up the + * orphan info for us. + */ + if (inode->i_nlink) + ext4_orphan_del(handle, inode); + + mutex_unlock(&EXT4_I(inode)->truncate_mutex); + ext4_journal_stop(handle); +} + +/* + * ext4_ext_writepage_trans_blocks: + * calculate max number of blocks we could modify + * in order to allocate new block for an inode + */ +int ext4_ext_writepage_trans_blocks(struct inode *inode, int num) +{ + int needed; + + needed = ext4_ext_calc_credits_for_insert(inode, NULL); + + /* caller wants to allocate num blocks, but note it includes sb */ + needed = needed * num - (num - 1); + +#ifdef CONFIG_QUOTA + needed += 2 * EXT4_QUOTA_TRANS_BLOCKS(inode->i_sb); +#endif + + return needed; +} + +EXPORT_SYMBOL(ext4_mark_inode_dirty); +EXPORT_SYMBOL(ext4_ext_invalidate_cache); +EXPORT_SYMBOL(ext4_ext_insert_extent); +EXPORT_SYMBOL(ext4_ext_walk_space); +EXPORT_SYMBOL(ext4_ext_find_goal); +EXPORT_SYMBOL(ext4_ext_calc_credits_for_insert); + diff --git a/fs/ext4/file.c b/fs/ext4/file.c new file mode 100644 index 000000000000..0b622c0624b7 --- /dev/null +++ b/fs/ext4/file.c @@ -0,0 +1,139 @@ +/* + * linux/fs/ext4/file.c + * + * Copyright (C) 1992, 1993, 1994, 1995 + * Remy Card (card@masi.ibp.fr) + * Laboratoire MASI - Institut Blaise Pascal + * Universite Pierre et Marie Curie (Paris VI) + * + * from + * + * linux/fs/minix/file.c + * + * Copyright (C) 1991, 1992 Linus Torvalds + * + * ext4 fs regular file handling primitives + * + * 64-bit file support on 64-bit platforms by Jakub Jelinek + * (jj@sunsite.ms.mff.cuni.cz) + */ + +#include <linux/time.h> +#include <linux/fs.h> +#include <linux/jbd2.h> +#include <linux/ext4_fs.h> +#include <linux/ext4_jbd2.h> +#include "xattr.h" +#include "acl.h" + +/* + * Called when an inode is released. Note that this is different + * from ext4_file_open: open gets called at every open, but release + * gets called only when /all/ the files are closed. + */ +static int ext4_release_file (struct inode * inode, struct file * filp) +{ + /* if we are the last writer on the inode, drop the block reservation */ + if ((filp->f_mode & FMODE_WRITE) && + (atomic_read(&inode->i_writecount) == 1)) + { + mutex_lock(&EXT4_I(inode)->truncate_mutex); + ext4_discard_reservation(inode); + mutex_unlock(&EXT4_I(inode)->truncate_mutex); + } + if (is_dx(inode) && filp->private_data) + ext4_htree_free_dir_info(filp->private_data); + + return 0; +} + +static ssize_t +ext4_file_write(struct kiocb *iocb, const struct iovec *iov, + unsigned long nr_segs, loff_t pos) +{ + struct file *file = iocb->ki_filp; + struct inode *inode = file->f_dentry->d_inode; + ssize_t ret; + int err; + + ret = generic_file_aio_write(iocb, iov, nr_segs, pos); + + /* + * Skip flushing if there was an error, or if nothing was written. + */ + if (ret <= 0) + return ret; + + /* + * If the inode is IS_SYNC, or is O_SYNC and we are doing data + * journalling then we need to make sure that we force the transaction + * to disk to keep all metadata uptodate synchronously. + */ + if (file->f_flags & O_SYNC) { + /* + * If we are non-data-journaled, then the dirty data has + * already been flushed to backing store by generic_osync_inode, + * and the inode has been flushed too if there have been any + * modifications other than mere timestamp updates. + * + * Open question --- do we care about flushing timestamps too + * if the inode is IS_SYNC? + */ + if (!ext4_should_journal_data(inode)) + return ret; + + goto force_commit; + } + + /* + * So we know that there has been no forced data flush. If the inode + * is marked IS_SYNC, we need to force one ourselves. + */ + if (!IS_SYNC(inode)) + return ret; + + /* + * Open question #2 --- should we force data to disk here too? If we + * don't, the only impact is that data=writeback filesystems won't + * flush data to disk automatically on IS_SYNC, only metadata (but + * historically, that is what ext2 has done.) + */ + +force_commit: + err = ext4_force_commit(inode->i_sb); + if (err) + return err; + return ret; +} + +const struct file_operations ext4_file_operations = { + .llseek = generic_file_llseek, + .read = do_sync_read, + .write = do_sync_write, + .aio_read = generic_file_aio_read, + .aio_write = ext4_file_write, + .ioctl = ext4_ioctl, +#ifdef CONFIG_COMPAT + .compat_ioctl = ext4_compat_ioctl, +#endif + .mmap = generic_file_mmap, + .open = generic_file_open, + .release = ext4_release_file, + .fsync = ext4_sync_file, + .sendfile = generic_file_sendfile, + .splice_read = generic_file_splice_read, + .splice_write = generic_file_splice_write, +}; + +struct inode_operations ext4_file_inode_operations = { + .truncate = ext4_truncate, + .setattr = ext4_setattr, +#ifdef CONFIG_EXT4DEV_FS_XATTR + .setxattr = generic_setxattr, + .getxattr = generic_getxattr, + .listxattr = ext4_listxattr, + .removexattr = generic_removexattr, +#endif + .permission = ext4_permission, +}; + diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c new file mode 100644 index 000000000000..2a167d7131fa --- /dev/null +++ b/fs/ext4/fsync.c @@ -0,0 +1,88 @@ +/* + * linux/fs/ext4/fsync.c + * + * Copyright (C) 1993 Stephen Tweedie (sct@redhat.com) + * from + * Copyright (C) 1992 Remy Card (card@masi.ibp.fr) + * Laboratoire MASI - Institut Blaise Pascal + * Universite Pierre et Marie Curie (Paris VI) + * from + * linux/fs/minix/truncate.c Copyright (C) 1991, 1992 Linus Torvalds + * + * ext4fs fsync primitive + * + * Big-endian to little-endian byte-swapping/bitmaps by + * David S. Miller (davem@caip.rutgers.edu), 1995 + * + * Removed unnecessary code duplication for little endian machines + * and excessive __inline__s. + * Andi Kleen, 1997 + * + * Major simplications and cleanup - we only need to do the metadata, because + * we can depend on generic_block_fdatasync() to sync the data blocks. + */ + +#include <linux/time.h> +#include <linux/fs.h> +#include <linux/sched.h> +#include <linux/writeback.h> +#include <linux/jbd2.h> +#include <linux/ext4_fs.h> +#include <linux/ext4_jbd2.h> + +/* + * akpm: A new design for ext4_sync_file(). + * + * This is only called from sys_fsync(), sys_fdatasync() and sys_msync(). + * There cannot be a transaction open by this task. + * Another task could have dirtied this inode. Its data can be in any + * state in the journalling system. + * + * What we do is just kick off a commit and wait on it. This will snapshot the + * inode to disk. + */ + +int ext4_sync_file(struct file * file, struct dentry *dentry, int datasync) +{ + struct inode *inode = dentry->d_inode; + int ret = 0; + + J_ASSERT(ext4_journal_current_handle() == 0); + + /* + * data=writeback: + * The caller's filemap_fdatawrite()/wait will sync the data. + * sync_inode() will sync the metadata + * + * data=ordered: + * The caller's filemap_fdatawrite() will write the data and + * sync_inode() will write the inode if it is dirty. Then the caller's + * filemap_fdatawait() will wait on the pages. + * + * data=journal: + * filemap_fdatawrite won't do anything (the buffers are clean). + * ext4_force_commit will write the file data into the journal and + * will wait on that. + * filemap_fdatawait() will encounter a ton of newly-dirtied pages + * (they were dirtied by commit). But that's OK - the blocks are + * safe in-journal, which is all fsync() needs to ensure. + */ + if (ext4_should_journal_data(inode)) { + ret = ext4_force_commit(inode->i_sb); + goto out; + } + + /* + * The VFS has written the file data. If the inode is unaltered + * then we need not start a commit. + */ + if (inode->i_state & (I_DIRTY_SYNC|I_DIRTY_DATASYNC)) { + struct writeback_control wbc = { + .sync_mode = WB_SYNC_ALL, + .nr_to_write = 0, /* sys_fsync did this */ + }; + ret = sync_inode(inode, &wbc); + } +out: + return ret; +} diff --git a/fs/ext4/hash.c b/fs/ext4/hash.c new file mode 100644 index 000000000000..a67966385e06 --- /dev/null +++ b/fs/ext4/hash.c @@ -0,0 +1,152 @@ +/* + * linux/fs/ext4/hash.c + * + * Copyright (C) 2002 by Theodore Ts'o + * + * This file is released under the GPL v2. + * + * This file may be redistributed under the terms of the GNU Public + * License. + */ + +#include <linux/fs.h> +#include <linux/jbd2.h> +#include <linux/sched.h> +#include <linux/ext4_fs.h> +#include <linux/cryptohash.h> + +#define DELTA 0x9E3779B9 + +static void TEA_transform(__u32 buf[4], __u32 const in[]) +{ + __u32 sum = 0; + __u32 b0 = buf[0], b1 = buf[1]; + __u32 a = in[0], b = in[1], c = in[2], d = in[3]; + int n = 16; + + do { + sum += DELTA; + b0 += ((b1 << 4)+a) ^ (b1+sum) ^ ((b1 >> 5)+b); + b1 += ((b0 << 4)+c) ^ (b0+sum) ^ ((b0 >> 5)+d); + } while(--n); + + buf[0] += b0; + buf[1] += b1; +} + + +/* The old legacy hash */ +static __u32 dx_hack_hash (const char *name, int len) +{ + __u32 hash0 = 0x12a3fe2d, hash1 = 0x37abe8f9; + while (len--) { + __u32 hash = hash1 + (hash0 ^ (*name++ * 7152373)); + + if (hash & 0x80000000) hash -= 0x7fffffff; + hash1 = hash0; + hash0 = hash; + } + return (hash0 << 1); +} + +static void str2hashbuf(const char *msg, int len, __u32 *buf, int num) +{ + __u32 pad, val; + int i; + + pad = (__u32)len | ((__u32)len << 8); + pad |= pad << 16; + + val = pad; + if (len > num*4) + len = num * 4; + for (i=0; i < len; i++) { + if ((i % 4) == 0) + val = pad; + val = msg[i] + (val << 8); + if ((i % 4) == 3) { + *buf++ = val; + val = pad; + num--; + } + } + if (--num >= 0) + *buf++ = val; + while (--num >= 0) + *buf++ = pad; +} + +/* + * Returns the hash of a filename. If len is 0 and name is NULL, then + * this function can be used to test whether or not a hash version is + * supported. + * + * The seed is an 4 longword (32 bits) "secret" which can be used to + * uniquify a hash. If the seed is all zero's, then some default seed + * may be used. + * + * A particular hash version specifies whether or not the seed is + * represented, and whether or not the returned hash is 32 bits or 64 + * bits. 32 bit hashes will return 0 for the minor hash. + */ +int ext4fs_dirhash(const char *name, int len, struct dx_hash_info *hinfo) +{ + __u32 hash; + __u32 minor_hash = 0; + const char *p; + int i; + __u32 in[8], buf[4]; + + /* Initialize the default seed for the hash checksum functions */ + buf[0] = 0x67452301; + buf[1] = 0xefcdab89; + buf[2] = 0x98badcfe; + buf[3] = 0x10325476; + + /* Check to see if the seed is all zero's */ + if (hinfo->seed) { + for (i=0; i < 4; i++) { + if (hinfo->seed[i]) + break; + } + if (i < 4) + memcpy(buf, hinfo->seed, sizeof(buf)); + } + + switch (hinfo->hash_version) { + case DX_HASH_LEGACY: + hash = dx_hack_hash(name, len); + break; + case DX_HASH_HALF_MD4: + p = name; + while (len > 0) { + str2hashbuf(p, len, in, 8); + half_md4_transform(buf, in); + len -= 32; + p += 32; + } + minor_hash = buf[2]; + hash = buf[1]; + break; + case DX_HASH_TEA: + p = name; + while (len > 0) { + str2hashbuf(p, len, in, 4); + TEA_transform(buf, in); + len -= 16; + p += 16; + } + hash = buf[0]; + minor_hash = buf[1]; + break; + default: + hinfo->hash = 0; + return -1; + } + hash = hash & ~1; + if (hash == (EXT4_HTREE_EOF << 1)) + hash = (EXT4_HTREE_EOF-1) << 1; + hinfo->hash = hash; + hinfo->minor_hash = minor_hash; + return 0; +} diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c new file mode 100644 index 000000000000..c88b439ba5cd --- /dev/null +++ b/fs/ext4/ialloc.c @@ -0,0 +1,772 @@ +/* + * linux/fs/ext4/ialloc.c + * + * Copyright (C) 1992, 1993, 1994, 1995 + * Remy Card (card@masi.ibp.fr) + * Laboratoire MASI - Institut Blaise Pascal + * Universite Pierre et Marie Curie (Paris VI) + * + * BSD ufs-inspired inode and directory allocation by + * Stephen Tweedie (sct@redhat.com), 1993 + * Big-endian to little-endian byte-swapping/bitmaps by + * David S. Miller (davem@caip.rutgers.edu), 1995 + */ + +#include <linux/time.h> +#include <linux/fs.h> +#include <linux/jbd2.h> +#include <linux/ext4_fs.h> +#include <linux/ext4_jbd2.h> +#include <linux/stat.h> +#include <linux/string.h> +#include <linux/quotaops.h> +#include <linux/buffer_head.h> +#include <linux/random.h> +#include <linux/bitops.h> +#include <linux/blkdev.h> +#include <asm/byteorder.h> + +#include "xattr.h" +#include "acl.h" + +/* + * ialloc.c contains the inodes allocation and deallocation routines + */ + +/* + * The free inodes are managed by bitmaps. A file system contains several + * blocks groups. Each group contains 1 bitmap block for blocks, 1 bitmap + * block for inodes, N blocks for the inode table and data blocks. + * + * The file system contains group descriptors which are located after the + * super block. Each descriptor contains the number of the bitmap block and + * the free blocks count in the block. + */ + + +/* + * Read the inode allocation bitmap for a given block_group, reading + * into the specified slot in the superblock's bitmap cache. + * + * Return buffer_head of bitmap on success or NULL. + */ +static struct buffer_head * +read_inode_bitmap(struct super_block * sb, unsigned long block_group) +{ + struct ext4_group_desc *desc; + struct buffer_head *bh = NULL; + + desc = ext4_get_group_desc(sb, block_group, NULL); + if (!desc) + goto error_out; + + bh = sb_bread(sb, ext4_inode_bitmap(sb, desc)); + if (!bh) + ext4_error(sb, "read_inode_bitmap", + "Cannot read inode bitmap - " + "block_group = %lu, inode_bitmap = %llu", + block_group, ext4_inode_bitmap(sb, desc)); +error_out: + return bh; +} + +/* + * NOTE! When we get the inode, we're the only people + * that have access to it, and as such there are no + * race conditions we have to worry about. The inode + * is not on the hash-lists, and it cannot be reached + * through the filesystem because the directory entry + * has been deleted earlier. + * + * HOWEVER: we must make sure that we get no aliases, + * which means that we have to call "clear_inode()" + * _before_ we mark the inode not in use in the inode + * bitmaps. Otherwise a newly created file might use + * the same inode number (not actually the same pointer + * though), and then we'd have two inodes sharing the + * same inode number and space on the harddisk. + */ +void ext4_free_inode (handle_t *handle, struct inode * inode) +{ + struct super_block * sb = inode->i_sb; + int is_directory; + unsigned long ino; + struct buffer_head *bitmap_bh = NULL; + struct buffer_head *bh2; + unsigned long block_group; + unsigned long bit; + struct ext4_group_desc * gdp; + struct ext4_super_block * es; + struct ext4_sb_info *sbi; + int fatal = 0, err; + + if (atomic_read(&inode->i_count) > 1) { + printk ("ext4_free_inode: inode has count=%d\n", + atomic_read(&inode->i_count)); + return; + } + if (inode->i_nlink) { + printk ("ext4_free_inode: inode has nlink=%d\n", + inode->i_nlink); + return; + } + if (!sb) { + printk("ext4_free_inode: inode on nonexistent device\n"); + return; + } + sbi = EXT4_SB(sb); + + ino = inode->i_ino; + ext4_debug ("freeing inode %lu\n", ino); + + /* + * Note: we must free any quota before locking the superblock, + * as writing the quota to disk may need the lock as well. + */ + DQUOT_INIT(inode); + ext4_xattr_delete_inode(handle, inode); + DQUOT_FREE_INODE(inode); + DQUOT_DROP(inode); + + is_directory = S_ISDIR(inode->i_mode); + + /* Do this BEFORE marking the inode not in use or returning an error */ + clear_inode (inode); + + es = EXT4_SB(sb)->s_es; + if (ino < EXT4_FIRST_INO(sb) || ino > le32_to_cpu(es->s_inodes_count)) { + ext4_error (sb, "ext4_free_inode", + "reserved or nonexistent inode %lu", ino); + goto error_return; + } + block_group = (ino - 1) / EXT4_INODES_PER_GROUP(sb); + bit = (ino - 1) % EXT4_INODES_PER_GROUP(sb); + bitmap_bh = read_inode_bitmap(sb, block_group); + if (!bitmap_bh) + goto error_return; + + BUFFER_TRACE(bitmap_bh, "get_write_access"); + fatal = ext4_journal_get_write_access(handle, bitmap_bh); + if (fatal) + goto error_return; + + /* Ok, now we can actually update the inode bitmaps.. */ + if (!ext4_clear_bit_atomic(sb_bgl_lock(sbi, block_group), + bit, bitmap_bh->b_data)) + ext4_error (sb, "ext4_free_inode", + "bit already cleared for inode %lu", ino); + else { + gdp = ext4_get_group_desc (sb, block_group, &bh2); + + BUFFER_TRACE(bh2, "get_write_access"); + fatal = ext4_journal_get_write_access(handle, bh2); + if (fatal) goto error_return; + + if (gdp) { + spin_lock(sb_bgl_lock(sbi, block_group)); + gdp->bg_free_inodes_count = cpu_to_le16( + le16_to_cpu(gdp->bg_free_inodes_count) + 1); + if (is_directory) + gdp->bg_used_dirs_count = cpu_to_le16( + le16_to_cpu(gdp->bg_used_dirs_count) - 1); + spin_unlock(sb_bgl_lock(sbi, block_group)); + percpu_counter_inc(&sbi->s_freeinodes_counter); + if (is_directory) + percpu_counter_dec(&sbi->s_dirs_counter); + + } + BUFFER_TRACE(bh2, "call ext4_journal_dirty_metadata"); + err = ext4_journal_dirty_metadata(handle, bh2); + if (!fatal) fatal = err; + } + BUFFER_TRACE(bitmap_bh, "call ext4_journal_dirty_metadata"); + err = ext4_journal_dirty_metadata(handle, bitmap_bh); + if (!fatal) + fatal = err; + sb->s_dirt = 1; +error_return: + brelse(bitmap_bh); + ext4_std_error(sb, fatal); +} + +/* + * There are two policies for allocating an inode. If the new inode is + * a directory, then a forward search is made for a block group with both + * free space and a low directory-to-inode ratio; if that fails, then of + * the groups with above-average free space, that group with the fewest + * directories already is chosen. + * + * For other inodes, search forward from the parent directory\'s block + * group to find a free inode. + */ +static int find_group_dir(struct super_block *sb, struct inode *parent) +{ + int ngroups = EXT4_SB(sb)->s_groups_count; + unsigned int freei, avefreei; + struct ext4_group_desc *desc, *best_desc = NULL; + struct buffer_head *bh; + int group, best_group = -1; + + freei = percpu_counter_read_positive(&EXT4_SB(sb)->s_freeinodes_counter); + avefreei = freei / ngroups; + + for (group = 0; group < ngroups; group++) { + desc = ext4_get_group_desc (sb, group, &bh); + if (!desc || !desc->bg_free_inodes_count) + continue; + if (le16_to_cpu(desc->bg_free_inodes_count) < avefreei) + continue; + if (!best_desc || + (le16_to_cpu(desc->bg_free_blocks_count) > + le16_to_cpu(best_desc->bg_free_blocks_count))) { + best_group = group; + best_desc = desc; + } + } + return best_group; +} + +/* + * Orlov's allocator for directories. + * + * We always try to spread first-level directories. + * + * If there are blockgroups with both free inodes and free blocks counts + * not worse than average we return one with smallest directory count. + * Otherwise we simply return a random group. + * + * For the rest rules look so: + * + * It's OK to put directory into a group unless + * it has too many directories already (max_dirs) or + * it has too few free inodes left (min_inodes) or + * it has too few free blocks left (min_blocks) or + * it's already running too large debt (max_debt). + * Parent's group is prefered, if it doesn't satisfy these + * conditions we search cyclically through the rest. If none + * of the groups look good we just look for a group with more + * free inodes than average (starting at parent's group). + * + * Debt is incremented each time we allocate a directory and decremented + * when we allocate an inode, within 0--255. + */ + +#define INODE_COST 64 +#define BLOCK_COST 256 + +static int find_group_orlov(struct super_block *sb, struct inode *parent) +{ + int parent_group = EXT4_I(parent)->i_block_group; + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct ext4_super_block *es = sbi->s_es; + int ngroups = sbi->s_groups_count; + int inodes_per_group = EXT4_INODES_PER_GROUP(sb); + unsigned int freei, avefreei; + ext4_fsblk_t freeb, avefreeb; + ext4_fsblk_t blocks_per_dir; + unsigned int ndirs; + int max_debt, max_dirs, min_inodes; + ext4_grpblk_t min_blocks; + int group = -1, i; + struct ext4_group_desc *desc; + struct buffer_head *bh; + + freei = percpu_counter_read_positive(&sbi->s_freeinodes_counter); + avefreei = freei / ngroups; + freeb = percpu_counter_read_positive(&sbi->s_freeblocks_counter); + avefreeb = freeb; + do_div(avefreeb, ngroups); + ndirs = percpu_counter_read_positive(&sbi->s_dirs_counter); + + if ((parent == sb->s_root->d_inode) || + (EXT4_I(parent)->i_flags & EXT4_TOPDIR_FL)) { + int best_ndir = inodes_per_group; + int best_group = -1; + + get_random_bytes(&group, sizeof(group)); + parent_group = (unsigned)group % ngroups; + for (i = 0; i < ngroups; i++) { + group = (parent_group + i) % ngroups; + desc = ext4_get_group_desc (sb, group, &bh); + if (!desc || !desc->bg_free_inodes_count) + continue; + if (le16_to_cpu(desc->bg_used_dirs_count) >= best_ndir) + continue; + if (le16_to_cpu(desc->bg_free_inodes_count) < avefreei) + continue; + if (le16_to_cpu(desc->bg_free_blocks_count) < avefreeb) + continue; + best_group = group; + best_ndir = le16_to_cpu(desc->bg_used_dirs_count); + } + if (best_group >= 0) + return best_group; + goto fallback; + } + + blocks_per_dir = ext4_blocks_count(es) - freeb; + do_div(blocks_per_dir, ndirs); + + max_dirs = ndirs / ngroups + inodes_per_group / 16; + min_inodes = avefreei - inodes_per_group / 4; + min_blocks = avefreeb - EXT4_BLOCKS_PER_GROUP(sb) / 4; + + max_debt = EXT4_BLOCKS_PER_GROUP(sb); + max_debt /= max_t(int, blocks_per_dir, BLOCK_COST); + if (max_debt * INODE_COST > inodes_per_group) + max_debt = inodes_per_group / INODE_COST; + if (max_debt > 255) + max_debt = 255; + if (max_debt == 0) + max_debt = 1; + + for (i = 0; i < ngroups; i++) { + group = (parent_group + i) % ngroups; + desc = ext4_get_group_desc (sb, group, &bh); + if (!desc || !desc->bg_free_inodes_count) + continue; + if (le16_to_cpu(desc->bg_used_dirs_count) >= max_dirs) + continue; + if (le16_to_cpu(desc->bg_free_inodes_count) < min_inodes) + continue; + if (le16_to_cpu(desc->bg_free_blocks_count) < min_blocks) + continue; + return group; + } + +fallback: + for (i = 0; i < ngroups; i++) { + group = (parent_group + i) % ngroups; + desc = ext4_get_group_desc (sb, group, &bh); + if (!desc || !desc->bg_free_inodes_count) + continue; + if (le16_to_cpu(desc->bg_free_inodes_count) >= avefreei) + return group; + } + + if (avefreei) { + /* + * The free-inodes counter is approximate, and for really small + * filesystems the above test can fail to find any blockgroups + */ + avefreei = 0; + goto fallback; + } + + return -1; +} + +static int find_group_other(struct super_block *sb, struct inode *parent) +{ + int parent_group = EXT4_I(parent)->i_block_group; + int ngroups = EXT4_SB(sb)->s_groups_count; + struct ext4_group_desc *desc; + struct buffer_head *bh; + int group, i; + + /* + * Try to place the inode in its parent directory + */ + group = parent_group; + desc = ext4_get_group_desc (sb, group, &bh); + if (desc && le16_to_cpu(desc->bg_free_inodes_count) && + le16_to_cpu(desc->bg_free_blocks_count)) + return group; + + /* + * We're going to place this inode in a different blockgroup from its + * parent. We want to cause files in a common directory to all land in + * the same blockgroup. But we want files which are in a different + * directory which shares a blockgroup with our parent to land in a + * different blockgroup. + * + * So add our directory's i_ino into the starting point for the hash. + */ + group = (group + parent->i_ino) % ngroups; + + /* + * Use a quadratic hash to find a group with a free inode and some free + * blocks. + */ + for (i = 1; i < ngroups; i <<= 1) { + group += i; + if (group >= ngroups) + group -= ngroups; + desc = ext4_get_group_desc (sb, group, &bh); + if (desc && le16_to_cpu(desc->bg_free_inodes_count) && + le16_to_cpu(desc->bg_free_blocks_count)) + return group; + } + + /* + * That failed: try linear search for a free inode, even if that group + * has no free blocks. + */ + group = parent_group; + for (i = 0; i < ngroups; i++) { + if (++group >= ngroups) + group = 0; + desc = ext4_get_group_desc (sb, group, &bh); + if (desc && le16_to_cpu(desc->bg_free_inodes_count)) + return group; + } + + return -1; +} + +/* + * There are two policies for allocating an inode. If the new inode is + * a directory, then a forward search is made for a block group with both + * free space and a low directory-to-inode ratio; if that fails, then of + * the groups with above-average free space, that group with the fewest + * directories already is chosen. + * + * For other inodes, search forward from the parent directory's block + * group to find a free inode. + */ +struct inode *ext4_new_inode(handle_t *handle, struct inode * dir, int mode) +{ + struct super_block *sb; + struct buffer_head *bitmap_bh = NULL; + struct buffer_head *bh2; + int group; + unsigned long ino = 0; + struct inode * inode; + struct ext4_group_desc * gdp = NULL; + struct ext4_super_block * es; + struct ext4_inode_info *ei; + struct ext4_sb_info *sbi; + int err = 0; + struct inode *ret; + int i; + + /* Cannot create files in a deleted directory */ + if (!dir || !dir->i_nlink) + return ERR_PTR(-EPERM); + + sb = dir->i_sb; + inode = new_inode(sb); + if (!inode) + return ERR_PTR(-ENOMEM); + ei = EXT4_I(inode); + + sbi = EXT4_SB(sb); + es = sbi->s_es; + if (S_ISDIR(mode)) { + if (test_opt (sb, OLDALLOC)) + group = find_group_dir(sb, dir); + else + group = find_group_orlov(sb, dir); + } else + group = find_group_other(sb, dir); + + err = -ENOSPC; + if (group == -1) + goto out; + + for (i = 0; i < sbi->s_groups_count; i++) { + err = -EIO; + + gdp = ext4_get_group_desc(sb, group, &bh2); + if (!gdp) + goto fail; + + brelse(bitmap_bh); + bitmap_bh = read_inode_bitmap(sb, group); + if (!bitmap_bh) + goto fail; + + ino = 0; + +repeat_in_this_group: + ino = ext4_find_next_zero_bit((unsigned long *) + bitmap_bh->b_data, EXT4_INODES_PER_GROUP(sb), ino); + if (ino < EXT4_INODES_PER_GROUP(sb)) { + + BUFFER_TRACE(bitmap_bh, "get_write_access"); + err = ext4_journal_get_write_access(handle, bitmap_bh); + if (err) + goto fail; + + if (!ext4_set_bit_atomic(sb_bgl_lock(sbi, group), + ino, bitmap_bh->b_data)) { + /* we won it */ + BUFFER_TRACE(bitmap_bh, + "call ext4_journal_dirty_metadata"); + err = ext4_journal_dirty_metadata(handle, + bitmap_bh); + if (err) + goto fail; + goto got; + } + /* we lost it */ + jbd2_journal_release_buffer(handle, bitmap_bh); + + if (++ino < EXT4_INODES_PER_GROUP(sb)) + goto repeat_in_this_group; + } + + /* + * This case is possible in concurrent environment. It is very + * rare. We cannot repeat the find_group_xxx() call because + * that will simply return the same blockgroup, because the + * group descriptor metadata has not yet been updated. + * So we just go onto the next blockgroup. + */ + if (++group == sbi->s_groups_count) + group = 0; + } + err = -ENOSPC; + goto out; + +got: + ino += group * EXT4_INODES_PER_GROUP(sb) + 1; + if (ino < EXT4_FIRST_INO(sb) || ino > le32_to_cpu(es->s_inodes_count)) { + ext4_error (sb, "ext4_new_inode", + "reserved inode or inode > inodes count - " + "block_group = %d, inode=%lu", group, ino); + err = -EIO; + goto fail; + } + + BUFFER_TRACE(bh2, "get_write_access"); + err = ext4_journal_get_write_access(handle, bh2); + if (err) goto fail; + spin_lock(sb_bgl_lock(sbi, group)); + gdp->bg_free_inodes_count = + cpu_to_le16(le16_to_cpu(gdp->bg_free_inodes_count) - 1); + if (S_ISDIR(mode)) { + gdp->bg_used_dirs_count = + cpu_to_le16(le16_to_cpu(gdp->bg_used_dirs_count) + 1); + } + spin_unlock(sb_bgl_lock(sbi, group)); + BUFFER_TRACE(bh2, "call ext4_journal_dirty_metadata"); + err = ext4_journal_dirty_metadata(handle, bh2); + if (err) goto fail; + + percpu_counter_dec(&sbi->s_freeinodes_counter); + if (S_ISDIR(mode)) + percpu_counter_inc(&sbi->s_dirs_counter); + sb->s_dirt = 1; + + inode->i_uid = current->fsuid; + if (test_opt (sb, GRPID)) + inode->i_gid = dir->i_gid; + else if (dir->i_mode & S_ISGID) { + inode->i_gid = dir->i_gid; + if (S_ISDIR(mode)) + mode |= S_ISGID; + } else + inode->i_gid = current->fsgid; + inode->i_mode = mode; + + inode->i_ino = ino; + /* This is the optimal IO size (for stat), not the fs block size */ + inode->i_blocks = 0; + inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC; + + memset(ei->i_data, 0, sizeof(ei->i_data)); + ei->i_dir_start_lookup = 0; + ei->i_disksize = 0; + + ei->i_flags = EXT4_I(dir)->i_flags & ~EXT4_INDEX_FL; + if (S_ISLNK(mode)) + ei->i_flags &= ~(EXT4_IMMUTABLE_FL|EXT4_APPEND_FL); + /* dirsync only applies to directories */ + if (!S_ISDIR(mode)) + ei->i_flags &= ~EXT4_DIRSYNC_FL; +#ifdef EXT4_FRAGMENTS + ei->i_faddr = 0; + ei->i_frag_no = 0; + ei->i_frag_size = 0; +#endif + ei->i_file_acl = 0; + ei->i_dir_acl = 0; + ei->i_dtime = 0; + ei->i_block_alloc_info = NULL; + ei->i_block_group = group; + + ext4_set_inode_flags(inode); + if (IS_DIRSYNC(inode)) + handle->h_sync = 1; + insert_inode_hash(inode); + spin_lock(&sbi->s_next_gen_lock); + inode->i_generation = sbi->s_next_generation++; + spin_unlock(&sbi->s_next_gen_lock); + + ei->i_state = EXT4_STATE_NEW; + ei->i_extra_isize = + (EXT4_INODE_SIZE(inode->i_sb) > EXT4_GOOD_OLD_INODE_SIZE) ? + sizeof(struct ext4_inode) - EXT4_GOOD_OLD_INODE_SIZE : 0; + + ret = inode; + if(DQUOT_ALLOC_INODE(inode)) { + err = -EDQUOT; + goto fail_drop; + } + + err = ext4_init_acl(handle, inode, dir); + if (err) + goto fail_free_drop; + + err = ext4_init_security(handle,inode, dir); + if (err) + goto fail_free_drop; + + err = ext4_mark_inode_dirty(handle, inode); + if (err) { + ext4_std_error(sb, err); + goto fail_free_drop; + } + if (test_opt(sb, EXTENTS)) { + EXT4_I(inode)->i_flags |= EXT4_EXTENTS_FL; + ext4_ext_tree_init(handle, inode); + if (!EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_EXTENTS)) { + err = ext4_journal_get_write_access(handle, EXT4_SB(sb)->s_sbh); + if (err) goto fail; + EXT4_SET_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_EXTENTS); + BUFFER_TRACE(EXT4_SB(sb)->s_sbh, "call ext4_journal_dirty_metadata"); + err = ext4_journal_dirty_metadata(handle, EXT4_SB(sb)->s_sbh); + } + } + + ext4_debug("allocating inode %lu\n", inode->i_ino); + goto really_out; +fail: + ext4_std_error(sb, err); +out: + iput(inode); + ret = ERR_PTR(err); +really_out: + brelse(bitmap_bh); + return ret; + +fail_free_drop: + DQUOT_FREE_INODE(inode); + +fail_drop: + DQUOT_DROP(inode); + inode->i_flags |= S_NOQUOTA; + inode->i_nlink = 0; + iput(inode); + brelse(bitmap_bh); + return ERR_PTR(err); +} + +/* Verify that we are loading a valid orphan from disk */ +struct inode *ext4_orphan_get(struct super_block *sb, unsigned long ino) +{ + unsigned long max_ino = le32_to_cpu(EXT4_SB(sb)->s_es->s_inodes_count); + unsigned long block_group; + int bit; + struct buffer_head *bitmap_bh = NULL; + struct inode *inode = NULL; + + /* Error cases - e2fsck has already cleaned up for us */ + if (ino > max_ino) { + ext4_warning(sb, __FUNCTION__, + "bad orphan ino %lu! e2fsck was run?", ino); + goto out; + } + + block_group = (ino - 1) / EXT4_INODES_PER_GROUP(sb); + bit = (ino - 1) % EXT4_INODES_PER_GROUP(sb); + bitmap_bh = read_inode_bitmap(sb, block_group); + if (!bitmap_bh) { + ext4_warning(sb, __FUNCTION__, + "inode bitmap error for orphan %lu", ino); + goto out; + } + + /* Having the inode bit set should be a 100% indicator that this + * is a valid orphan (no e2fsck run on fs). Orphans also include + * inodes that were being truncated, so we can't check i_nlink==0. + */ + if (!ext4_test_bit(bit, bitmap_bh->b_data) || + !(inode = iget(sb, ino)) || is_bad_inode(inode) || + NEXT_ORPHAN(inode) > max_ino) { + ext4_warning(sb, __FUNCTION__, + "bad orphan inode %lu! e2fsck was run?", ino); + printk(KERN_NOTICE "ext4_test_bit(bit=%d, block=%llu) = %d\n", + bit, (unsigned long long)bitmap_bh->b_blocknr, + ext4_test_bit(bit, bitmap_bh->b_data)); + printk(KERN_NOTICE "inode=%p\n", inode); + if (inode) { + printk(KERN_NOTICE "is_bad_inode(inode)=%d\n", + is_bad_inode(inode)); + printk(KERN_NOTICE "NEXT_ORPHAN(inode)=%u\n", + NEXT_ORPHAN(inode)); + printk(KERN_NOTICE "max_ino=%lu\n", max_ino); + } + /* Avoid freeing blocks if we got a bad deleted inode */ + if (inode && inode->i_nlink == 0) + inode->i_blocks = 0; + iput(inode); + inode = NULL; + } +out: + brelse(bitmap_bh); + return inode; +} + +unsigned long ext4_count_free_inodes (struct super_block * sb) +{ + unsigned long desc_count; + struct ext4_group_desc *gdp; + int i; +#ifdef EXT4FS_DEBUG + struct ext4_super_block *es; + unsigned long bitmap_count, x; + struct buffer_head *bitmap_bh = NULL; + + es = EXT4_SB(sb)->s_es; + desc_count = 0; + bitmap_count = 0; + gdp = NULL; + for (i = 0; i < EXT4_SB(sb)->s_groups_count; i++) { + gdp = ext4_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 = ext4_count_free(bitmap_bh, EXT4_INODES_PER_GROUP(sb) / 8); + printk("group %d: stored = %d, counted = %lu\n", + i, le16_to_cpu(gdp->bg_free_inodes_count), x); + bitmap_count += x; + } + brelse(bitmap_bh); + printk("ext4_count_free_inodes: stored = %u, computed = %lu, %lu\n", + le32_to_cpu(es->s_free_inodes_count), desc_count, bitmap_count); + return desc_count; +#else + desc_count = 0; + for (i = 0; i < EXT4_SB(sb)->s_groups_count; i++) { + gdp = ext4_get_group_desc (sb, i, NULL); + if (!gdp) + continue; + desc_count += le16_to_cpu(gdp->bg_free_inodes_count); + cond_resched(); + } + return desc_count; +#endif +} + +/* Called at mount-time, super-block is locked */ +unsigned long ext4_count_dirs (struct super_block * sb) +{ + unsigned long count = 0; + int i; + + for (i = 0; i < EXT4_SB(sb)->s_groups_count; i++) { + struct ext4_group_desc *gdp = ext4_get_group_desc (sb, i, NULL); + if (!gdp) + continue; + count += le16_to_cpu(gdp->bg_used_dirs_count); + } + return count; +} + diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c new file mode 100644 index 000000000000..0a60ec5a16db --- /dev/null +++ b/fs/ext4/inode.c @@ -0,0 +1,3233 @@ +/* + * linux/fs/ext4/inode.c + * + * Copyright (C) 1992, 1993, 1994, 1995 + * Remy Card (card@masi.ibp.fr) + * Laboratoire MASI - Institut Blaise Pascal + * Universite Pierre et Marie Curie (Paris VI) + * + * from + * + * linux/fs/minix/inode.c + * + * Copyright (C) 1991, 1992 Linus Torvalds + * + * Goal-directed block allocation by Stephen Tweedie + * (sct@redhat.com), 1993, 1998 + * Big-endian to little-endian byte-swapping/bitmaps by + * David S. Miller (davem@caip.rutgers.edu), 1995 + * 64-bit file support on 64-bit platforms by Jakub Jelinek + * (jj@sunsite.ms.mff.cuni.cz) + * + * Assorted race fixes, rewrite of ext4_get_block() by Al Viro, 2000 + */ + +#include <linux/module.h> +#include <linux/fs.h> +#include <linux/time.h> +#include <linux/ext4_jbd2.h> +#include <linux/jbd2.h> +#include <linux/smp_lock.h> +#include <linux/highuid.h> +#include <linux/pagemap.h> +#include <linux/quotaops.h> +#include <linux/string.h> +#include <linux/buffer_head.h> +#include <linux/writeback.h> +#include <linux/mpage.h> +#include <linux/uio.h> +#include <linux/bio.h> +#include "xattr.h" +#include "acl.h" + +/* + * Test whether an inode is a fast symlink. + */ +static int ext4_inode_is_fast_symlink(struct inode *inode) +{ + int ea_blocks = EXT4_I(inode)->i_file_acl ? + (inode->i_sb->s_blocksize >> 9) : 0; + + return (S_ISLNK(inode->i_mode) && inode->i_blocks - ea_blocks == 0); +} + +/* + * The ext4 forget function must perform a revoke if we are freeing data + * which has been journaled. Metadata (eg. indirect blocks) must be + * revoked in all cases. + * + * "bh" may be NULL: a metadata block may have been freed from memory + * but there may still be a record of it in the journal, and that record + * still needs to be revoked. + */ +int ext4_forget(handle_t *handle, int is_metadata, struct inode *inode, + struct buffer_head *bh, ext4_fsblk_t blocknr) +{ + int err; + + might_sleep(); + + BUFFER_TRACE(bh, "enter"); + + jbd_debug(4, "forgetting bh %p: is_metadata = %d, mode %o, " + "data mode %lx\n", + bh, is_metadata, inode->i_mode, + test_opt(inode->i_sb, DATA_FLAGS)); + + /* Never use the revoke function if we are doing full data + * journaling: there is no need to, and a V1 superblock won't + * support it. Otherwise, only skip the revoke on un-journaled + * data blocks. */ + + if (test_opt(inode->i_sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA || + (!is_metadata && !ext4_should_journal_data(inode))) { + if (bh) { + BUFFER_TRACE(bh, "call jbd2_journal_forget"); + return ext4_journal_forget(handle, bh); + } + return 0; + } + + /* + * data!=journal && (is_metadata || should_journal_data(inode)) + */ + BUFFER_TRACE(bh, "call ext4_journal_revoke"); + err = ext4_journal_revoke(handle, blocknr, bh); + if (err) + ext4_abort(inode->i_sb, __FUNCTION__, + "error %d when attempting revoke", err); + BUFFER_TRACE(bh, "exit"); + return err; +} + +/* + * Work out how many blocks we need to proceed with the next chunk of a + * truncate transaction. + */ +static unsigned long blocks_for_truncate(struct inode *inode) +{ + unsigned long needed; + + needed = inode->i_blocks >> (inode->i_sb->s_blocksize_bits - 9); + + /* Give ourselves just enough room to cope with inodes in which + * i_blocks is corrupt: we've seen disk corruptions in the past + * which resulted in random data in an inode which looked enough + * like a regular file for ext4 to try to delete it. Things + * will go a bit crazy if that happens, but at least we should + * try not to panic the whole kernel. */ + if (needed < 2) + needed = 2; + + /* But we need to bound the transaction so we don't overflow the + * journal. */ + if (needed > EXT4_MAX_TRANS_DATA) + needed = EXT4_MAX_TRANS_DATA; + + return EXT4_DATA_TRANS_BLOCKS(inode->i_sb) + needed; +} + +/* + * Truncate transactions can be complex and absolutely huge. So we need to + * be able to restart the transaction at a conventient checkpoint to make + * sure we don't overflow the journal. + * + * start_transaction gets us a new handle for a truncate transaction, + * and extend_transaction tries to extend the existing one a bit. If + * extend fails, we need to propagate the failure up and restart the + * transaction in the top-level truncate loop. --sct + */ +static handle_t *start_transaction(struct inode *inode) +{ + handle_t *result; + + result = ext4_journal_start(inode, blocks_for_truncate(inode)); + if (!IS_ERR(result)) + return result; + + ext4_std_error(inode->i_sb, PTR_ERR(result)); + return result; +} + +/* + * Try to extend this transaction for the purposes of truncation. + * + * Returns 0 if we managed to create more room. If we can't create more + * room, and the transaction must be restarted we return 1. + */ +static int try_to_extend_transaction(handle_t *handle, struct inode *inode) +{ + if (handle->h_buffer_credits > EXT4_RESERVE_TRANS_BLOCKS) + return 0; + if (!ext4_journal_extend(handle, blocks_for_truncate(inode))) + return 0; + return 1; +} + +/* + * Restart the transaction associated with *handle. This does a commit, + * so before we call here everything must be consistently dirtied against + * this transaction. + */ +static int ext4_journal_test_restart(handle_t *handle, struct inode *inode) +{ + jbd_debug(2, "restarting handle %p\n", handle); + return ext4_journal_restart(handle, blocks_for_truncate(inode)); +} + +/* + * Called at the last iput() if i_nlink is zero. + */ +void ext4_delete_inode (struct inode * inode) +{ + handle_t *handle; + + truncate_inode_pages(&inode->i_data, 0); + + if (is_bad_inode(inode)) + goto no_delete; + + handle = start_transaction(inode); + if (IS_ERR(handle)) { + /* + * If we're going to skip the normal cleanup, we still need to + * make sure that the in-core orphan linked list is properly + * cleaned up. + */ + ext4_orphan_del(NULL, inode); + goto no_delete; + } + + if (IS_SYNC(inode)) + handle->h_sync = 1; + inode->i_size = 0; + if (inode->i_blocks) + ext4_truncate(inode); + /* + * Kill off the orphan record which ext4_truncate created. + * AKPM: I think this can be inside the above `if'. + * Note that ext4_orphan_del() has to be able to cope with the + * deletion of a non-existent orphan - this is because we don't + * know if ext4_truncate() actually created an orphan record. + * (Well, we could do this if we need to, but heck - it works) + */ + ext4_orphan_del(handle, inode); + EXT4_I(inode)->i_dtime = get_seconds(); + + /* + * One subtle ordering requirement: if anything has gone wrong + * (transaction abort, IO errors, whatever), then we can still + * do these next steps (the fs will already have been marked as + * having errors), but we can't free the inode if the mark_dirty + * fails. + */ + if (ext4_mark_inode_dirty(handle, inode)) + /* If that failed, just do the required in-core inode clear. */ + clear_inode(inode); + else + ext4_free_inode(handle, inode); + ext4_journal_stop(handle); + return; +no_delete: + clear_inode(inode); /* We must guarantee clearing of inode... */ +} + +typedef struct { + __le32 *p; + __le32 key; + struct buffer_head *bh; +} Indirect; + +static inline void add_chain(Indirect *p, struct buffer_head *bh, __le32 *v) +{ + p->key = *(p->p = v); + p->bh = bh; +} + +static int verify_chain(Indirect *from, Indirect *to) +{ + while (from <= to && from->key == *from->p) + from++; + return (from > to); +} + +/** + * ext4_block_to_path - parse the block number into array of offsets + * @inode: inode in question (we are only interested in its superblock) + * @i_block: block number to be parsed + * @offsets: array to store the offsets in + * @boundary: set this non-zero if the referred-to block is likely to be + * followed (on disk) by an indirect block. + * + * To store the locations of file's data ext4 uses a data structure common + * for UNIX filesystems - tree of pointers anchored in the inode, with + * data blocks at leaves and indirect blocks in intermediate nodes. + * This function translates the block number into path in that tree - + * return value is the path length and @offsets[n] is the offset of + * pointer to (n+1)th node in the nth one. If @block is out of range + * (negative or too large) warning is printed and zero returned. + * + * Note: function doesn't find node addresses, so no IO is needed. All + * we need to know is the capacity of indirect blocks (taken from the + * inode->i_sb). + */ + +/* + * Portability note: the last comparison (check that we fit into triple + * indirect block) is spelled differently, because otherwise on an + * architecture with 32-bit longs and 8Kb pages we might get into trouble + * if our filesystem had 8Kb blocks. We might use long long, but that would + * kill us on x86. Oh, well, at least the sign propagation does not matter - + * i_block would have to be negative in the very beginning, so we would not + * get there at all. + */ + +static int ext4_block_to_path(struct inode *inode, + long i_block, int offsets[4], int *boundary) +{ + int ptrs = EXT4_ADDR_PER_BLOCK(inode->i_sb); + int ptrs_bits = EXT4_ADDR_PER_BLOCK_BITS(inode->i_sb); + const long direct_blocks = EXT4_NDIR_BLOCKS, + indirect_blocks = ptrs, + double_blocks = (1 << (ptrs_bits * 2)); + int n = 0; + int final = 0; + + if (i_block < 0) { + ext4_warning (inode->i_sb, "ext4_block_to_path", "block < 0"); + } else if (i_block < direct_blocks) { + offsets[n++] = i_block; + final = direct_blocks; + } else if ( (i_block -= direct_blocks) < indirect_blocks) { + offsets[n++] = EXT4_IND_BLOCK; + offsets[n++] = i_block; + final = ptrs; + } else if ((i_block -= indirect_blocks) < double_blocks) { + offsets[n++] = EXT4_DIND_BLOCK; + offsets[n++] = i_block >> ptrs_bits; + offsets[n++] = i_block & (ptrs - 1); + final = ptrs; + } else if (((i_block -= double_blocks) >> (ptrs_bits * 2)) < ptrs) { + offsets[n++] = EXT4_TIND_BLOCK; + offsets[n++] = i_block >> (ptrs_bits * 2); + offsets[n++] = (i_block >> ptrs_bits) & (ptrs - 1); + offsets[n++] = i_block & (ptrs - 1); + final = ptrs; + } else { + ext4_warning(inode->i_sb, "ext4_block_to_path", "block > big"); + } + if (boundary) + *boundary = final - 1 - (i_block & (ptrs - 1)); + return n; +} + +/** + * ext4_get_branch - read the chain of indirect blocks leading to data + * @inode: inode in question + * @depth: depth of the chain (1 - direct pointer, etc.) + * @offsets: offsets of pointers in inode/indirect blocks + * @chain: place to store the result + * @err: here we store the error value + * + * Function fills the array of triples <key, p, bh> and returns %NULL + * if everything went OK or the pointer to the last filled triple + * (incomplete one) otherwise. Upon the return chain[i].key contains + * the number of (i+1)-th block in the chain (as it is stored in memory, + * i.e. little-endian 32-bit), chain[i].p contains the address of that + * number (it points into struct inode for i==0 and into the bh->b_data + * for i>0) and chain[i].bh points to the buffer_head of i-th indirect + * block for i>0 and NULL for i==0. In other words, it holds the block + * numbers of the chain, addresses they were taken from (and where we can + * verify that chain did not change) and buffer_heads hosting these + * numbers. + * + * Function stops when it stumbles upon zero pointer (absent block) + * (pointer to last triple returned, *@err == 0) + * or when it gets an IO error reading an indirect block + * (ditto, *@err == -EIO) + * or when it notices that chain had been changed while it was reading + * (ditto, *@err == -EAGAIN) + * or when it reads all @depth-1 indirect blocks successfully and finds + * the whole chain, all way to the data (returns %NULL, *err == 0). + */ +static Indirect *ext4_get_branch(struct inode *inode, int depth, int *offsets, + Indirect chain[4], int *err) +{ + struct super_block *sb = inode->i_sb; + Indirect *p = chain; + struct buffer_head *bh; + + *err = 0; + /* i_data is not going away, no lock needed */ + add_chain (chain, NULL, EXT4_I(inode)->i_data + *offsets); + if (!p->key) + goto no_block; + while (--depth) { + bh = sb_bread(sb, le32_to_cpu(p->key)); + if (!bh) + goto failure; + /* Reader: pointers */ + if (!verify_chain(chain, p)) + goto changed; + add_chain(++p, bh, (__le32*)bh->b_data + *++offsets); + /* Reader: end */ + if (!p->key) + goto no_block; + } + return NULL; + +changed: + brelse(bh); + *err = -EAGAIN; + goto no_block; +failure: + *err = -EIO; +no_block: + return p; +} + +/** + * ext4_find_near - find a place for allocation with sufficient locality + * @inode: owner + * @ind: descriptor of indirect block. + * + * This function returns the prefered place for block allocation. + * It is used when heuristic for sequential allocation fails. + * Rules are: + * + if there is a block to the left of our position - allocate near it. + * + if pointer will live in indirect block - allocate near that block. + * + if pointer will live in inode - allocate in the same + * cylinder group. + * + * In the latter case we colour the starting block by the callers PID to + * prevent it from clashing with concurrent allocations for a different inode + * in the same block group. The PID is used here so that functionally related + * files will be close-by on-disk. + * + * Caller must make sure that @ind is valid and will stay that way. + */ +static ext4_fsblk_t ext4_find_near(struct inode *inode, Indirect *ind) +{ + struct ext4_inode_info *ei = EXT4_I(inode); + __le32 *start = ind->bh ? (__le32*) ind->bh->b_data : ei->i_data; + __le32 *p; + ext4_fsblk_t bg_start; + ext4_grpblk_t colour; + + /* Try to find previous block */ + for (p = ind->p - 1; p >= start; p--) { + if (*p) + return le32_to_cpu(*p); + } + + /* No such thing, so let's try location of indirect block */ + if (ind->bh) + return ind->bh->b_blocknr; + + /* + * It is going to be referred to from the inode itself? OK, just put it + * into the same cylinder group then. + */ + bg_start = ext4_group_first_block_no(inode->i_sb, ei->i_block_group); + colour = (current->pid % 16) * + (EXT4_BLOCKS_PER_GROUP(inode->i_sb) / 16); + return bg_start + colour; +} + +/** + * ext4_find_goal - find a prefered place for allocation. + * @inode: owner + * @block: block we want + * @chain: chain of indirect blocks + * @partial: pointer to the last triple within a chain + * @goal: place to store the result. + * + * Normally this function find the prefered place for block allocation, + * stores it in *@goal and returns zero. + */ + +static ext4_fsblk_t ext4_find_goal(struct inode *inode, long block, + Indirect chain[4], Indirect *partial) +{ + struct ext4_block_alloc_info *block_i; + + block_i = EXT4_I(inode)->i_block_alloc_info; + + /* + * try the heuristic for sequential allocation, + * failing that at least try to get decent locality. + */ + if (block_i && (block == block_i->last_alloc_logical_block + 1) + && (block_i->last_alloc_physical_block != 0)) { + return block_i->last_alloc_physical_block + 1; + } + + return ext4_find_near(inode, partial); +} + +/** + * ext4_blks_to_allocate: Look up the block map and count the number + * of direct blocks need to be allocated for the given branch. + * + * @branch: chain of indirect blocks + * @k: number of blocks need for indirect blocks + * @blks: number of data blocks to be mapped. + * @blocks_to_boundary: the offset in the indirect block + * + * return the total number of blocks to be allocate, including the + * direct and indirect blocks. + */ +static int ext4_blks_to_allocate(Indirect *branch, int k, unsigned long blks, + int blocks_to_boundary) +{ + unsigned long count = 0; + + /* + * Simple case, [t,d]Indirect block(s) has not allocated yet + * then it's clear blocks on that path have not allocated + */ + if (k > 0) { + /* right now we don't handle cross boundary allocation */ + if (blks < blocks_to_boundary + 1) + count += blks; + else + count += blocks_to_boundary + 1; + return count; + } + + count++; + while (count < blks && count <= blocks_to_boundary && + le32_to_cpu(*(branch[0].p + count)) == 0) { + count++; + } + return count; +} + +/** + * ext4_alloc_blocks: multiple allocate blocks needed for a branch + * @indirect_blks: the number of blocks need to allocate for indirect + * blocks + * + * @new_blocks: on return it will store the new block numbers for + * the indirect blocks(if needed) and the first direct block, + * @blks: on return it will store the total number of allocated + * direct blocks + */ +static int ext4_alloc_blocks(handle_t *handle, struct inode *inode, + ext4_fsblk_t goal, int indirect_blks, int blks, + ext4_fsblk_t new_blocks[4], int *err) +{ + int target, i; + unsigned long count = 0; + int index = 0; + ext4_fsblk_t current_block = 0; + int ret = 0; + + /* + * Here we try to allocate the requested multiple blocks at once, + * on a best-effort basis. + * To build a branch, we should allocate blocks for + * the indirect blocks(if not allocated yet), and at least + * the first direct block of this branch. That's the + * minimum number of blocks need to allocate(required) + */ + target = blks + indirect_blks; + + while (1) { + count = target; + /* allocating blocks for indirect blocks and direct blocks */ + current_block = ext4_new_blocks(handle,inode,goal,&count,err); + if (*err) + goto failed_out; + + target -= count; + /* allocate blocks for indirect blocks */ + while (index < indirect_blks && count) { + new_blocks[index++] = current_block++; + count--; + } + + if (count > 0) + break; + } + + /* save the new block number for the first direct block */ + new_blocks[index] = current_block; + + /* total number of blocks allocated for direct blocks */ + ret = count; + *err = 0; + return ret; +failed_out: + for (i = 0; i <index; i++) + ext4_free_blocks(handle, inode, new_blocks[i], 1); + return ret; +} + +/** + * ext4_alloc_branch - allocate and set up a chain of blocks. + * @inode: owner + * @indirect_blks: number of allocated indirect blocks + * @blks: number of allocated direct blocks + * @offsets: offsets (in the blocks) to store the pointers to next. + * @branch: place to store the chain in. + * + * This function allocates blocks, zeroes out all but the last one, + * links them into chain and (if we are synchronous) writes them to disk. + * In other words, it prepares a branch that can be spliced onto the + * inode. It stores the information about that chain in the branch[], in + * the same format as ext4_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 ext4_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. + * + * If allocation fails we free all blocks we've allocated (and forget + * their buffer_heads) and return the error value the from failed + * ext4_alloc_block() (normally -ENOSPC). Otherwise we set the chain + * as described above and return 0. + */ +static int ext4_alloc_branch(handle_t *handle, struct inode *inode, + int indirect_blks, int *blks, ext4_fsblk_t goal, + int *offsets, Indirect *branch) +{ + int blocksize = inode->i_sb->s_blocksize; + int i, n = 0; + int err = 0; + struct buffer_head *bh; + int num; + ext4_fsblk_t new_blocks[4]; + ext4_fsblk_t current_block; + + num = ext4_alloc_blocks(handle, inode, goal, indirect_blks, + *blks, new_blocks, &err); + if (err) + return err; + + branch[0].key = cpu_to_le32(new_blocks[0]); + /* + * metadata blocks and data blocks are allocated. + */ + for (n = 1; n <= indirect_blks; n++) { + /* + * Get buffer_head for parent block, zero it out + * and set the pointer to new one, then send + * parent to disk. + */ + bh = sb_getblk(inode->i_sb, new_blocks[n-1]); + branch[n].bh = bh; + lock_buffer(bh); + BUFFER_TRACE(bh, "call get_create_access"); + err = ext4_journal_get_create_access(handle, bh); + if (err) { + unlock_buffer(bh); + brelse(bh); + goto failed; + } + + memset(bh->b_data, 0, blocksize); + branch[n].p = (__le32 *) bh->b_data + offsets[n]; + branch[n].key = cpu_to_le32(new_blocks[n]); + *branch[n].p = branch[n].key; + if ( n == indirect_blks) { + current_block = new_blocks[n]; + /* + * End of chain, update the last new metablock of + * the chain to point to the new allocated + * data blocks numbers + */ + for (i=1; i < num; i++) + *(branch[n].p + i) = cpu_to_le32(++current_block); + } + BUFFER_TRACE(bh, "marking uptodate"); + set_buffer_uptodate(bh); + unlock_buffer(bh); + + BUFFER_TRACE(bh, "call ext4_journal_dirty_metadata"); + err = ext4_journal_dirty_metadata(handle, bh); + if (err) + goto failed; + } + *blks = num; + return err; +failed: + /* Allocation failed, free what we already allocated */ + for (i = 1; i <= n ; i++) { + BUFFER_TRACE(branch[i].bh, "call jbd2_journal_forget"); + ext4_journal_forget(handle, branch[i].bh); + } + for (i = 0; i <indirect_blks; i++) + ext4_free_blocks(handle, inode, new_blocks[i], 1); + + ext4_free_blocks(handle, inode, new_blocks[i], num); + + return err; +} + +/** + * ext4_splice_branch - splice the allocated branch onto inode. + * @inode: owner + * @block: (logical) number of block we are adding + * @chain: chain of indirect blocks (with a missing link - see + * ext4_alloc_branch) + * @where: location of missing link + * @num: number of indirect blocks we are adding + * @blks: number of direct blocks we are adding + * + * This function fills the missing link and does all housekeeping needed in + * inode (->i_blocks, etc.). In case of success we end up with the full + * chain to new block and return 0. + */ +static int ext4_splice_branch(handle_t *handle, struct inode *inode, + long block, Indirect *where, int num, int blks) +{ + int i; + int err = 0; + struct ext4_block_alloc_info *block_i; + ext4_fsblk_t current_block; + + block_i = EXT4_I(inode)->i_block_alloc_info; + /* + * If we're splicing into a [td]indirect block (as opposed to the + * inode) then we need to get write access to the [td]indirect block + * before the splice. + */ + if (where->bh) { + BUFFER_TRACE(where->bh, "get_write_access"); + err = ext4_journal_get_write_access(handle, where->bh); + if (err) + goto err_out; + } + /* That's it */ + + *where->p = where->key; + + /* + * Update the host buffer_head or inode to point to more just allocated + * direct blocks blocks + */ + if (num == 0 && blks > 1) { + current_block = le32_to_cpu(where->key) + 1; + for (i = 1; i < blks; i++) + *(where->p + i ) = cpu_to_le32(current_block++); + } + + /* + * update the most recently allocated logical & physical block + * in i_block_alloc_info, to assist find the proper goal block for next + * allocation + */ + if (block_i) { + block_i->last_alloc_logical_block = block + blks - 1; + block_i->last_alloc_physical_block = + le32_to_cpu(where[num].key) + blks - 1; + } + + /* We are done with atomic stuff, now do the rest of housekeeping */ + + inode->i_ctime = CURRENT_TIME_SEC; + ext4_mark_inode_dirty(handle, inode); + + /* had we spliced it onto indirect block? */ + if (where->bh) { + /* + * If we spliced it onto an indirect block, we haven't + * altered the inode. Note however that if it is being spliced + * onto an indirect block at the very end of the file (the + * file is growing) then we *will* alter the inode to reflect + * the new i_size. But that is not done here - it is done in + * generic_commit_write->__mark_inode_dirty->ext4_dirty_inode. + */ + jbd_debug(5, "splicing indirect only\n"); + BUFFER_TRACE(where->bh, "call ext4_journal_dirty_metadata"); + err = ext4_journal_dirty_metadata(handle, where->bh); + if (err) + goto err_out; + } else { + /* + * OK, we spliced it into the inode itself on a direct block. + * Inode was dirtied above. + */ + jbd_debug(5, "splicing direct\n"); + } + return err; + +err_out: + for (i = 1; i <= num; i++) { + BUFFER_TRACE(where[i].bh, "call jbd2_journal_forget"); + ext4_journal_forget(handle, where[i].bh); + ext4_free_blocks(handle,inode,le32_to_cpu(where[i-1].key),1); + } + ext4_free_blocks(handle, inode, le32_to_cpu(where[num].key), blks); + + return err; +} + +/* + * Allocation strategy is simple: if we have to allocate something, we will + * have to go the whole way to leaf. So let's do it before attaching anything + * to tree, set linkage between the newborn blocks, write them if sync is + * required, recheck the path, free and repeat if check fails, otherwise + * set the last missing link (that will protect us from any truncate-generated + * removals - all blocks on the path are immune now) and possibly force the + * write on the parent block. + * That has a nice additional property: no special recovery from the failed + * allocations is needed - we simply release blocks and do not touch anything + * reachable from inode. + * + * `handle' can be NULL if create == 0. + * + * The BKL may not be held on entry here. Be sure to take it early. + * return > 0, # of blocks mapped or allocated. + * return = 0, if plain lookup failed. + * return < 0, error case. + */ +int ext4_get_blocks_handle(handle_t *handle, struct inode *inode, + sector_t iblock, unsigned long maxblocks, + struct buffer_head *bh_result, + int create, int extend_disksize) +{ + int err = -EIO; + int offsets[4]; + Indirect chain[4]; + Indirect *partial; + ext4_fsblk_t goal; + int indirect_blks; + int blocks_to_boundary = 0; + int depth; + struct ext4_inode_info *ei = EXT4_I(inode); + int count = 0; + ext4_fsblk_t first_block = 0; + + + J_ASSERT(!(EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL)); + J_ASSERT(handle != NULL || create == 0); + depth = ext4_block_to_path(inode,iblock,offsets,&blocks_to_boundary); + + if (depth == 0) + goto out; + + partial = ext4_get_branch(inode, depth, offsets, chain, &err); + + /* Simplest case - block found, no allocation needed */ + if (!partial) { + first_block = le32_to_cpu(chain[depth - 1].key); + clear_buffer_new(bh_result); + count++; + /*map more blocks*/ + while (count < maxblocks && count <= blocks_to_boundary) { + ext4_fsblk_t blk; + + if (!verify_chain(chain, partial)) { + /* + * Indirect block might be removed by + * truncate while we were reading it. + * Handling of that case: forget what we've + * got now. Flag the err as EAGAIN, so it + * will reread. + */ + err = -EAGAIN; + count = 0; + break; + } + blk = le32_to_cpu(*(chain[depth-1].p + count)); + + if (blk == first_block + count) + count++; + else + break; + } + if (err != -EAGAIN) + goto got_it; + } + + /* Next simple case - plain lookup or failed read of indirect block */ + if (!create || err == -EIO) + goto cleanup; + + mutex_lock(&ei->truncate_mutex); + + /* + * If the indirect block is missing while we are reading + * the chain(ext4_get_branch() returns -EAGAIN err), or + * if the chain has been changed after we grab the semaphore, + * (either because another process truncated this branch, or + * another get_block allocated this branch) re-grab the chain to see if + * the request block has been allocated or not. + * + * Since we already block the truncate/other get_block + * at this point, we will have the current copy of the chain when we + * splice the branch into the tree. + */ + if (err == -EAGAIN || !verify_chain(chain, partial)) { + while (partial > chain) { + brelse(partial->bh); + partial--; + } + partial = ext4_get_branch(inode, depth, offsets, chain, &err); + if (!partial) { + count++; + mutex_unlock(&ei->truncate_mutex); + if (err) + goto cleanup; + clear_buffer_new(bh_result); + goto got_it; + } + } + + /* + * Okay, we need to do block allocation. Lazily initialize the block + * allocation info here if necessary + */ + if (S_ISREG(inode->i_mode) && (!ei->i_block_alloc_info)) + ext4_init_block_alloc_info(inode); + + goal = ext4_find_goal(inode, iblock, chain, partial); + + /* the number of blocks need to allocate for [d,t]indirect blocks */ + indirect_blks = (chain + depth) - partial - 1; + + /* + * Next look up the indirect map to count the totoal number of + * direct blocks to allocate for this branch. + */ + count = ext4_blks_to_allocate(partial, indirect_blks, + maxblocks, blocks_to_boundary); + /* + * Block out ext4_truncate while we alter the tree + */ + err = ext4_alloc_branch(handle, inode, indirect_blks, &count, goal, + offsets + (partial - chain), partial); + + /* + * The ext4_splice_branch call will free and forget any buffers + * on the new chain if there is a failure, but that risks using + * up transaction credits, especially for bitmaps where the + * credits cannot be returned. Can we handle this somehow? We + * may need to return -EAGAIN upwards in the worst case. --sct + */ + if (!err) + err = ext4_splice_branch(handle, inode, iblock, + partial, indirect_blks, count); + /* + * i_disksize growing is protected by truncate_mutex. Don't forget to + * protect it if you're about to implement concurrent + * ext4_get_block() -bzzz + */ + if (!err && extend_disksize && inode->i_size > ei->i_disksize) + ei->i_disksize = inode->i_size; + mutex_unlock(&ei->truncate_mutex); + if (err) + goto cleanup; + + set_buffer_new(bh_result); +got_it: + map_bh(bh_result, inode->i_sb, le32_to_cpu(chain[depth-1].key)); + if (count > blocks_to_boundary) + set_buffer_boundary(bh_result); + err = count; + /* Clean up and exit */ + partial = chain + depth - 1; /* the whole chain */ +cleanup: + while (partial > chain) { + BUFFER_TRACE(partial->bh, "call brelse"); + brelse(partial->bh); + partial--; + } + BUFFER_TRACE(bh_result, "returned"); +out: + return err; +} + +#define DIO_CREDITS (EXT4_RESERVE_TRANS_BLOCKS + 32) + +static int ext4_get_block(struct inode *inode, sector_t iblock, + struct buffer_head *bh_result, int create) +{ + handle_t *handle = journal_current_handle(); + int ret = 0; + unsigned max_blocks = bh_result->b_size >> inode->i_blkbits; + + if (!create) + goto get_block; /* A read */ + + if (max_blocks == 1) + goto get_block; /* A single block get */ + + if (handle->h_transaction->t_state == T_LOCKED) { + /* + * Huge direct-io writes can hold off commits for long + * periods of time. Let this commit run. + */ + ext4_journal_stop(handle); + handle = ext4_journal_start(inode, DIO_CREDITS); + if (IS_ERR(handle)) + ret = PTR_ERR(handle); + goto get_block; + } + + if (handle->h_buffer_credits <= EXT4_RESERVE_TRANS_BLOCKS) { + /* + * Getting low on buffer credits... + */ + ret = ext4_journal_extend(handle, DIO_CREDITS); + if (ret > 0) { + /* + * Couldn't extend the transaction. Start a new one. + */ + ret = ext4_journal_restart(handle, DIO_CREDITS); + } + } + +get_block: + if (ret == 0) { + ret = ext4_get_blocks_wrap(handle, inode, iblock, + max_blocks, bh_result, create, 0); + if (ret > 0) { + bh_result->b_size = (ret << inode->i_blkbits); + ret = 0; + } + } + return ret; +} + +/* + * `handle' can be NULL if create is zero + */ +struct buffer_head *ext4_getblk(handle_t *handle, struct inode *inode, + long block, int create, int *errp) +{ + struct buffer_head dummy; + int fatal = 0, err; + + J_ASSERT(handle != NULL || create == 0); + + dummy.b_state = 0; + dummy.b_blocknr = -1000; + buffer_trace_init(&dummy.b_history); + err = ext4_get_blocks_wrap(handle, inode, block, 1, + &dummy, create, 1); + /* + * ext4_get_blocks_handle() returns number of blocks + * mapped. 0 in case of a HOLE. + */ + if (err > 0) { + if (err > 1) + WARN_ON(1); + err = 0; + } + *errp = err; + if (!err && 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); + + /* + * Now that we do not always journal data, we should + * keep in mind whether this should always journal the + * new buffer as metadata. For now, regular file + * writes use ext4_get_block instead, so it's not a + * problem. + */ + lock_buffer(bh); + BUFFER_TRACE(bh, "call get_create_access"); + fatal = ext4_journal_get_create_access(handle, bh); + if (!fatal && !buffer_uptodate(bh)) { + memset(bh->b_data,0,inode->i_sb->s_blocksize); + set_buffer_uptodate(bh); + } + unlock_buffer(bh); + BUFFER_TRACE(bh, "call ext4_journal_dirty_metadata"); + err = ext4_journal_dirty_metadata(handle, bh); + if (!fatal) + fatal = err; + } else { + BUFFER_TRACE(bh, "not a new buffer"); + } + if (fatal) { + *errp = fatal; + brelse(bh); + bh = NULL; + } + return bh; + } +err: + return NULL; +} + +struct buffer_head *ext4_bread(handle_t *handle, struct inode *inode, + int block, int create, int *err) +{ + struct buffer_head * bh; + + bh = ext4_getblk(handle, inode, block, create, err); + if (!bh) + return bh; + if (buffer_uptodate(bh)) + return bh; + ll_rw_block(READ_META, 1, &bh); + wait_on_buffer(bh); + if (buffer_uptodate(bh)) + return bh; + put_bh(bh); + *err = -EIO; + return NULL; +} + +static int walk_page_buffers( handle_t *handle, + struct buffer_head *head, + unsigned from, + unsigned to, + int *partial, + int (*fn)( handle_t *handle, + struct buffer_head *bh)) +{ + struct buffer_head *bh; + unsigned block_start, block_end; + unsigned blocksize = head->b_size; + int err, ret = 0; + struct buffer_head *next; + + for ( bh = head, block_start = 0; + ret == 0 && (bh != head || !block_start); + block_start = block_end, bh = next) + { + next = bh->b_this_page; + block_end = block_start + blocksize; + if (block_end <= from || block_start >= to) { + if (partial && !buffer_uptodate(bh)) + *partial = 1; + continue; + } + err = (*fn)(handle, bh); + if (!ret) + ret = err; + } + return ret; +} + +/* + * To preserve ordering, it is essential that the hole instantiation and + * the data write be encapsulated in a single transaction. We cannot + * close off a transaction and start a new one between the ext4_get_block() + * and the commit_write(). So doing the jbd2_journal_start at the start of + * prepare_write() is the right place. + * + * Also, this function can nest inside ext4_writepage() -> + * block_write_full_page(). In that case, we *know* that ext4_writepage() + * has generated enough buffer credits to do the whole page. So we won't + * block on the journal in that case, which is good, because the caller may + * be PF_MEMALLOC. + * + * By accident, ext4 can be reentered when a transaction is open via + * quota file writes. If we were to commit the transaction while thus + * reentered, there can be a deadlock - we would be holding a quota + * lock, and the commit would never complete if another thread had a + * transaction open and was blocking on the quota lock - a ranking + * violation. + * + * So what we do is to rely on the fact that jbd2_journal_stop/journal_start + * will _not_ run commit under these circumstances because handle->h_ref + * is elevated. We'll still have enough credits for the tiny quotafile + * write. + */ +static int do_journal_get_write_access(handle_t *handle, + struct buffer_head *bh) +{ + if (!buffer_mapped(bh) || buffer_freed(bh)) + return 0; + return ext4_journal_get_write_access(handle, bh); +} + +static int ext4_prepare_write(struct file *file, struct page *page, + unsigned from, unsigned to) +{ + struct inode *inode = page->mapping->host; + int ret, needed_blocks = ext4_writepage_trans_blocks(inode); + handle_t *handle; + int retries = 0; + +retry: + handle = ext4_journal_start(inode, needed_blocks); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + goto out; + } + if (test_opt(inode->i_sb, NOBH) && ext4_should_writeback_data(inode)) + ret = nobh_prepare_write(page, from, to, ext4_get_block); + else + ret = block_prepare_write(page, from, to, ext4_get_block); + if (ret) + goto prepare_write_failed; + + if (ext4_should_journal_data(inode)) { + ret = walk_page_buffers(handle, page_buffers(page), + from, to, NULL, do_journal_get_write_access); + } +prepare_write_failed: + if (ret) + ext4_journal_stop(handle); + if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries)) + goto retry; +out: + return ret; +} + +int ext4_journal_dirty_data(handle_t *handle, struct buffer_head *bh) +{ + int err = jbd2_journal_dirty_data(handle, bh); + if (err) + ext4_journal_abort_handle(__FUNCTION__, __FUNCTION__, + bh, handle,err); + return err; +} + +/* For commit_write() in data=journal mode */ +static int commit_write_fn(handle_t *handle, struct buffer_head *bh) +{ + if (!buffer_mapped(bh) || buffer_freed(bh)) + return 0; + set_buffer_uptodate(bh); + return ext4_journal_dirty_metadata(handle, bh); +} + +/* + * We need to pick up the new inode size which generic_commit_write gave us + * `file' can be NULL - eg, when called from page_symlink(). + * + * ext4 never places buffers on inode->i_mapping->private_list. metadata + * buffers are managed internally. + */ +static int ext4_ordered_commit_write(struct file *file, struct page *page, + unsigned from, unsigned to) +{ + handle_t *handle = ext4_journal_current_handle(); + struct inode *inode = page->mapping->host; + int ret = 0, ret2; + + ret = walk_page_buffers(handle, page_buffers(page), + from, to, NULL, ext4_journal_dirty_data); + + if (ret == 0) { + /* + * generic_commit_write() will run mark_inode_dirty() if i_size + * changes. So let's piggyback the i_disksize mark_inode_dirty + * into that. + */ + loff_t new_i_size; + + new_i_size = ((loff_t)page->index << PAGE_CACHE_SHIFT) + to; + if (new_i_size > EXT4_I(inode)->i_disksize) + EXT4_I(inode)->i_disksize = new_i_size; + ret = generic_commit_write(file, page, from, to); + } + ret2 = ext4_journal_stop(handle); + if (!ret) + ret = ret2; + return ret; +} + +static int ext4_writeback_commit_write(struct file *file, struct page *page, + unsigned from, unsigned to) +{ + handle_t *handle = ext4_journal_current_handle(); + struct inode *inode = page->mapping->host; + int ret = 0, ret2; + loff_t new_i_size; + + new_i_size = ((loff_t)page->index << PAGE_CACHE_SHIFT) + to; + if (new_i_size > EXT4_I(inode)->i_disksize) + EXT4_I(inode)->i_disksize = new_i_size; + + if (test_opt(inode->i_sb, NOBH) && ext4_should_writeback_data(inode)) + ret = nobh_commit_write(file, page, from, to); + else + ret = generic_commit_write(file, page, from, to); + + ret2 = ext4_journal_stop(handle); + if (!ret) + ret = ret2; + return ret; +} + +static int ext4_journalled_commit_write(struct file *file, + struct page *page, unsigned from, unsigned to) +{ + handle_t *handle = ext4_journal_current_handle(); + struct inode *inode = page->mapping->host; + int ret = 0, ret2; + int partial = 0; + loff_t pos; + + /* + * Here we duplicate the generic_commit_write() functionality + */ + pos = ((loff_t)page->index << PAGE_CACHE_SHIFT) + to; + + ret = walk_page_buffers(handle, page_buffers(page), from, + to, &partial, commit_write_fn); + if (!partial) + SetPageUptodate(page); + if (pos > inode->i_size) + i_size_write(inode, pos); + EXT4_I(inode)->i_state |= EXT4_STATE_JDATA; + if (inode->i_size > EXT4_I(inode)->i_disksize) { + EXT4_I(inode)->i_disksize = inode->i_size; + ret2 = ext4_mark_inode_dirty(handle, inode); + if (!ret) + ret = ret2; + } + ret2 = ext4_journal_stop(handle); + if (!ret) + ret = ret2; + return ret; +} + +/* + * bmap() is special. It gets used by applications such as lilo and by + * the swapper to find the on-disk block of a specific piece of data. + * + * Naturally, this is dangerous if the block concerned is still in the + * journal. If somebody makes a swapfile on an ext4 data-journaling + * filesystem and enables swap, then they may get a nasty shock when the + * data getting swapped to that swapfile suddenly gets overwritten by + * the original zero's written out previously to the journal and + * awaiting writeback in the kernel's buffer cache. + * + * So, if we see any bmap calls here on a modified, data-journaled file, + * take extra steps to flush any blocks which might be in the cache. + */ +static sector_t ext4_bmap(struct address_space *mapping, sector_t block) +{ + struct inode *inode = mapping->host; + journal_t *journal; + int err; + + if (EXT4_I(inode)->i_state & EXT4_STATE_JDATA) { + /* + * This is a REALLY heavyweight approach, but the use of + * bmap on dirty files is expected to be extremely rare: + * only if we run lilo or swapon on a freshly made file + * do we expect this to happen. + * + * (bmap requires CAP_SYS_RAWIO so this does not + * represent an unprivileged user DOS attack --- we'd be + * in trouble if mortal users could trigger this path at + * will.) + * + * NB. EXT4_STATE_JDATA is not set on files other than + * regular files. If somebody wants to bmap a directory + * or symlink and gets confused because the buffer + * hasn't yet been flushed to disk, they deserve + * everything they get. + */ + + EXT4_I(inode)->i_state &= ~EXT4_STATE_JDATA; + journal = EXT4_JOURNAL(inode); + jbd2_journal_lock_updates(journal); + err = jbd2_journal_flush(journal); + jbd2_journal_unlock_updates(journal); + + if (err) + return 0; + } + + return generic_block_bmap(mapping,block,ext4_get_block); +} + +static int bget_one(handle_t *handle, struct buffer_head *bh) +{ + get_bh(bh); + return 0; +} + +static int bput_one(handle_t *handle, struct buffer_head *bh) +{ + put_bh(bh); + return 0; +} + +static int jbd2_journal_dirty_data_fn(handle_t *handle, struct buffer_head *bh) +{ + if (buffer_mapped(bh)) + return ext4_journal_dirty_data(handle, bh); + return 0; +} + +/* + * Note that we always start a transaction even if we're not journalling + * data. This is to preserve ordering: any hole instantiation within + * __block_write_full_page -> ext4_get_block() should be journalled + * along with the data so we don't crash and then get metadata which + * refers to old data. + * + * In all journalling modes block_write_full_page() will start the I/O. + * + * Problem: + * + * ext4_writepage() -> kmalloc() -> __alloc_pages() -> page_launder() -> + * ext4_writepage() + * + * Similar for: + * + * ext4_file_write() -> generic_file_write() -> __alloc_pages() -> ... + * + * Same applies to ext4_get_block(). We will deadlock on various things like + * lock_journal and i_truncate_mutex. + * + * Setting PF_MEMALLOC here doesn't work - too many internal memory + * allocations fail. + * + * 16May01: If we're reentered then journal_current_handle() will be + * non-zero. We simply *return*. + * + * 1 July 2001: @@@ FIXME: + * In journalled data mode, a data buffer may be metadata against the + * current transaction. But the same file is part of a shared mapping + * and someone does a writepage() on it. + * + * We will move the buffer onto the async_data list, but *after* it has + * been dirtied. So there's a small window where we have dirty data on + * BJ_Metadata. + * + * Note that this only applies to the last partial page in the file. The + * bit which block_write_full_page() uses prepare/commit for. (That's + * broken code anyway: it's wrong for msync()). + * + * It's a rare case: affects the final partial page, for journalled data + * where the file is subject to bith write() and writepage() in the same + * transction. To fix it we'll need a custom block_write_full_page(). + * We'll probably need that anyway for journalling writepage() output. + * + * We don't honour synchronous mounts for writepage(). That would be + * disastrous. Any write() or metadata operation will sync the fs for + * us. + * + * AKPM2: if all the page's buffers are mapped to disk and !data=journal, + * we don't need to open a transaction here. + */ +static int ext4_ordered_writepage(struct page *page, + struct writeback_control *wbc) +{ + struct inode *inode = page->mapping->host; + struct buffer_head *page_bufs; + handle_t *handle = NULL; + int ret = 0; + int err; + + J_ASSERT(PageLocked(page)); + + /* + * We give up here if we're reentered, because it might be for a + * different filesystem. + */ + if (ext4_journal_current_handle()) + goto out_fail; + + handle = ext4_journal_start(inode, ext4_writepage_trans_blocks(inode)); + + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + goto out_fail; + } + + if (!page_has_buffers(page)) { + create_empty_buffers(page, inode->i_sb->s_blocksize, + (1 << BH_Dirty)|(1 << BH_Uptodate)); + } + page_bufs = page_buffers(page); + walk_page_buffers(handle, page_bufs, 0, + PAGE_CACHE_SIZE, NULL, bget_one); + + ret = block_write_full_page(page, ext4_get_block, wbc); + + /* + * The page can become unlocked at any point now, and + * truncate can then come in and change things. So we + * can't touch *page from now on. But *page_bufs is + * safe due to elevated refcount. + */ + + /* + * And attach them to the current transaction. But only if + * block_write_full_page() succeeded. Otherwise they are unmapped, + * and generally junk. + */ + if (ret == 0) { + err = walk_page_buffers(handle, page_bufs, 0, PAGE_CACHE_SIZE, + NULL, jbd2_journal_dirty_data_fn); + if (!ret) + ret = err; + } + walk_page_buffers(handle, page_bufs, 0, + PAGE_CACHE_SIZE, NULL, bput_one); + err = ext4_journal_stop(handle); + if (!ret) + ret = err; + return ret; + +out_fail: + redirty_page_for_writepage(wbc, page); + unlock_page(page); + return ret; +} + +static int ext4_writeback_writepage(struct page *page, + struct writeback_control *wbc) +{ + struct inode *inode = page->mapping->host; + handle_t *handle = NULL; + int ret = 0; + int err; + + if (ext4_journal_current_handle()) + goto out_fail; + + handle = ext4_journal_start(inode, ext4_writepage_trans_blocks(inode)); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + goto out_fail; + } + + if (test_opt(inode->i_sb, NOBH) && ext4_should_writeback_data(inode)) + ret = nobh_writepage(page, ext4_get_block, wbc); + else + ret = block_write_full_page(page, ext4_get_block, wbc); + + err = ext4_journal_stop(handle); + if (!ret) + ret = err; + return ret; + +out_fail: + redirty_page_for_writepage(wbc, page); + unlock_page(page); + return ret; +} + +static int ext4_journalled_writepage(struct page *page, + struct writeback_control *wbc) +{ + struct inode *inode = page->mapping->host; + handle_t *handle = NULL; + int ret = 0; + int err; + + if (ext4_journal_current_handle()) + goto no_write; + + handle = ext4_journal_start(inode, ext4_writepage_trans_blocks(inode)); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + goto no_write; + } + + if (!page_has_buffers(page) || PageChecked(page)) { + /* + * It's mmapped pagecache. Add buffers and journal it. There + * doesn't seem much point in redirtying the page here. + */ + ClearPageChecked(page); + ret = block_prepare_write(page, 0, PAGE_CACHE_SIZE, + ext4_get_block); + if (ret != 0) { + ext4_journal_stop(handle); + goto out_unlock; + } + ret = walk_page_buffers(handle, page_buffers(page), 0, + PAGE_CACHE_SIZE, NULL, do_journal_get_write_access); + + err = walk_page_buffers(handle, page_buffers(page), 0, + PAGE_CACHE_SIZE, NULL, commit_write_fn); + if (ret == 0) + ret = err; + EXT4_I(inode)->i_state |= EXT4_STATE_JDATA; + unlock_page(page); + } else { + /* + * It may be a page full of checkpoint-mode buffers. We don't + * really know unless we go poke around in the buffer_heads. + * But block_write_full_page will do the right thing. + */ + ret = block_write_full_page(page, ext4_get_block, wbc); + } + err = ext4_journal_stop(handle); + if (!ret) + ret = err; +out: + return ret; + +no_write: + redirty_page_for_writepage(wbc, page); +out_unlock: + unlock_page(page); + goto out; +} + +static int ext4_readpage(struct file *file, struct page *page) +{ + return mpage_readpage(page, ext4_get_block); +} + +static int +ext4_readpages(struct file *file, struct address_space *mapping, + struct list_head *pages, unsigned nr_pages) +{ + return mpage_readpages(mapping, pages, nr_pages, ext4_get_block); +} + +static void ext4_invalidatepage(struct page *page, unsigned long offset) +{ + journal_t *journal = EXT4_JOURNAL(page->mapping->host); + + /* + * If it's a full truncate we just forget about the pending dirtying + */ + if (offset == 0) + ClearPageChecked(page); + + jbd2_journal_invalidatepage(journal, page, offset); +} + +static int ext4_releasepage(struct page *page, gfp_t wait) +{ + journal_t *journal = EXT4_JOURNAL(page->mapping->host); + + WARN_ON(PageChecked(page)); + if (!page_has_buffers(page)) + return 0; + return jbd2_journal_try_to_free_buffers(journal, page, wait); +} + +/* + * If the O_DIRECT write will extend the file then add this inode to the + * orphan list. So recovery will truncate it back to the original size + * if the machine crashes during the write. + * + * If the O_DIRECT write is intantiating holes inside i_size and the machine + * crashes then stale disk data _may_ be exposed inside the file. + */ +static ssize_t ext4_direct_IO(int rw, struct kiocb *iocb, + const struct iovec *iov, loff_t offset, + unsigned long nr_segs) +{ + struct file *file = iocb->ki_filp; + struct inode *inode = file->f_mapping->host; + struct ext4_inode_info *ei = EXT4_I(inode); + handle_t *handle = NULL; + ssize_t ret; + int orphan = 0; + size_t count = iov_length(iov, nr_segs); + + if (rw == WRITE) { + loff_t final_size = offset + count; + + handle = ext4_journal_start(inode, DIO_CREDITS); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + goto out; + } + if (final_size > inode->i_size) { + ret = ext4_orphan_add(handle, inode); + if (ret) + goto out_stop; + orphan = 1; + ei->i_disksize = inode->i_size; + } + } + + ret = blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev, iov, + offset, nr_segs, + ext4_get_block, NULL); + + /* + * Reacquire the handle: ext4_get_block() can restart the transaction + */ + handle = journal_current_handle(); + +out_stop: + if (handle) { + int err; + + if (orphan && inode->i_nlink) + ext4_orphan_del(handle, inode); + if (orphan && ret > 0) { + loff_t end = offset + ret; + if (end > inode->i_size) { + ei->i_disksize = end; + i_size_write(inode, end); + /* + * We're going to return a positive `ret' + * here due to non-zero-length I/O, so there's + * no way of reporting error returns from + * ext4_mark_inode_dirty() to userspace. So + * ignore it. + */ + ext4_mark_inode_dirty(handle, inode); + } + } + err = ext4_journal_stop(handle); + if (ret == 0) + ret = err; + } +out: + return ret; +} + +/* + * Pages can be marked dirty completely asynchronously from ext4's journalling + * activity. By filemap_sync_pte(), try_to_unmap_one(), etc. We cannot do + * much here because ->set_page_dirty is called under VFS locks. The page is + * not necessarily locked. + * + * We cannot just dirty the page and leave attached buffers clean, because the + * buffers' dirty state is "definitive". We cannot just set the buffers dirty + * or jbddirty because all the journalling code will explode. + * + * So what we do is to mark the page "pending dirty" and next time writepage + * is called, propagate that into the buffers appropriately. + */ +static int ext4_journalled_set_page_dirty(struct page *page) +{ + SetPageChecked(page); + return __set_page_dirty_nobuffers(page); +} + +static const struct address_space_operations ext4_ordered_aops = { + .readpage = ext4_readpage, + .readpages = ext4_readpages, + .writepage = ext4_ordered_writepage, + .sync_page = block_sync_page, + .prepare_write = ext4_prepare_write, + .commit_write = ext4_ordered_commit_write, + .bmap = ext4_bmap, + .invalidatepage = ext4_invalidatepage, + .releasepage = ext4_releasepage, + .direct_IO = ext4_direct_IO, + .migratepage = buffer_migrate_page, +}; + +static const struct address_space_operations ext4_writeback_aops = { + .readpage = ext4_readpage, + .readpages = ext4_readpages, + .writepage = ext4_writeback_writepage, + .sync_page = block_sync_page, + .prepare_write = ext4_prepare_write, + .commit_write = ext4_writeback_commit_write, + .bmap = ext4_bmap, + .invalidatepage = ext4_invalidatepage, + .releasepage = ext4_releasepage, + .direct_IO = ext4_direct_IO, + .migratepage = buffer_migrate_page, +}; + +static const struct address_space_operations ext4_journalled_aops = { + .readpage = ext4_readpage, + .readpages = ext4_readpages, + .writepage = ext4_journalled_writepage, + .sync_page = block_sync_page, + .prepare_write = ext4_prepare_write, + .commit_write = ext4_journalled_commit_write, + .set_page_dirty = ext4_journalled_set_page_dirty, + .bmap = ext4_bmap, + .invalidatepage = ext4_invalidatepage, + .releasepage = ext4_releasepage, +}; + +void ext4_set_aops(struct inode *inode) +{ + if (ext4_should_order_data(inode)) + inode->i_mapping->a_ops = &ext4_ordered_aops; + else if (ext4_should_writeback_data(inode)) + inode->i_mapping->a_ops = &ext4_writeback_aops; + else + inode->i_mapping->a_ops = &ext4_journalled_aops; +} + +/* + * ext4_block_truncate_page() zeroes out a mapping from file offset `from' + * up to the end of the block which corresponds to `from'. + * This required during truncate. We need to physically zero the tail end + * of that block so it doesn't yield old data if the file is later grown. + */ +int ext4_block_truncate_page(handle_t *handle, struct page *page, + struct address_space *mapping, loff_t from) +{ + ext4_fsblk_t index = from >> PAGE_CACHE_SHIFT; + unsigned offset = from & (PAGE_CACHE_SIZE-1); + unsigned blocksize, iblock, length, pos; + struct inode *inode = mapping->host; + struct buffer_head *bh; + int err = 0; + void *kaddr; + + blocksize = inode->i_sb->s_blocksize; + length = blocksize - (offset & (blocksize - 1)); + iblock = index << (PAGE_CACHE_SHIFT - inode->i_sb->s_blocksize_bits); + + /* + * For "nobh" option, we can only work if we don't need to + * read-in the page - otherwise we create buffers to do the IO. + */ + if (!page_has_buffers(page) && test_opt(inode->i_sb, NOBH) && + ext4_should_writeback_data(inode) && PageUptodate(page)) { + kaddr = kmap_atomic(page, KM_USER0); + memset(kaddr + offset, 0, length); + flush_dcache_page(page); + kunmap_atomic(kaddr, KM_USER0); + set_page_dirty(page); + goto unlock; + } + + if (!page_has_buffers(page)) + create_empty_buffers(page, blocksize, 0); + + /* Find the buffer that contains "offset" */ + bh = page_buffers(page); + pos = blocksize; + while (offset >= pos) { + bh = bh->b_this_page; + iblock++; + pos += blocksize; + } + + err = 0; + if (buffer_freed(bh)) { + BUFFER_TRACE(bh, "freed: skip"); + goto unlock; + } + + if (!buffer_mapped(bh)) { + BUFFER_TRACE(bh, "unmapped"); + ext4_get_block(inode, iblock, bh, 0); + /* unmapped? It's a hole - nothing to do */ + if (!buffer_mapped(bh)) { + BUFFER_TRACE(bh, "still unmapped"); + goto unlock; + } + } + + /* Ok, it's mapped. Make sure it's up-to-date */ + if (PageUptodate(page)) + set_buffer_uptodate(bh); + + if (!buffer_uptodate(bh)) { + err = -EIO; + ll_rw_block(READ, 1, &bh); + wait_on_buffer(bh); + /* Uhhuh. Read error. Complain and punt. */ + if (!buffer_uptodate(bh)) + goto unlock; + } + + if (ext4_should_journal_data(inode)) { + BUFFER_TRACE(bh, "get write access"); + err = ext4_journal_get_write_access(handle, bh); + if (err) + goto unlock; + } + + kaddr = kmap_atomic(page, KM_USER0); + memset(kaddr + offset, 0, length); + flush_dcache_page(page); + kunmap_atomic(kaddr, KM_USER0); + + BUFFER_TRACE(bh, "zeroed end of block"); + + err = 0; + if (ext4_should_journal_data(inode)) { + err = ext4_journal_dirty_metadata(handle, bh); + } else { + if (ext4_should_order_data(inode)) + err = ext4_journal_dirty_data(handle, bh); + mark_buffer_dirty(bh); + } + +unlock: + unlock_page(page); + page_cache_release(page); + return err; +} + +/* + * Probably it should be a library function... search for first non-zero word + * or memcmp with zero_page, whatever is better for particular architecture. + * Linus? + */ +static inline int all_zeroes(__le32 *p, __le32 *q) +{ + while (p < q) + if (*p++) + return 0; + return 1; +} + +/** + * ext4_find_shared - find the indirect blocks for partial truncation. + * @inode: inode in question + * @depth: depth of the affected branch + * @offsets: offsets of pointers in that branch (see ext4_block_to_path) + * @chain: place to store the pointers to partial indirect blocks + * @top: place to the (detached) top of branch + * + * This is a helper function used by ext4_truncate(). + * + * When we do truncate() we may have to clean the ends of several + * indirect blocks but leave the blocks themselves alive. Block is + * partially truncated if some data below the new i_size is refered + * from it (and it is on the path to the first completely truncated + * data block, indeed). We have to free the top of that path along + * with everything to the right of the path. Since no allocation + * past the truncation point is possible until ext4_truncate() + * finishes, we may safely do the latter, but top of branch may + * require special attention - pageout below the truncation point + * might try to populate it. + * + * We atomically detach the top of branch from the tree, store the + * block number of its root in *@top, pointers to buffer_heads of + * partially truncated blocks - in @chain[].bh and pointers to + * their last elements that should not be removed - in + * @chain[].p. Return value is the pointer to last filled element + * of @chain. + * + * The work left to caller to do the actual freeing of subtrees: + * a) free the subtree starting from *@top + * b) free the subtrees whose roots are stored in + * (@chain[i].p+1 .. end of @chain[i].bh->b_data) + * c) free the subtrees growing from the inode past the @chain[0]. + * (no partially truncated stuff there). */ + +static Indirect *ext4_find_shared(struct inode *inode, int depth, + int offsets[4], Indirect chain[4], __le32 *top) +{ + Indirect *partial, *p; + int k, err; + + *top = 0; + /* Make k index the deepest non-null offest + 1 */ + for (k = depth; k > 1 && !offsets[k-1]; k--) + ; + partial = ext4_get_branch(inode, k, offsets, chain, &err); + /* Writer: pointers */ + if (!partial) + partial = chain + k-1; + /* + * If the branch acquired continuation since we've looked at it - + * fine, it should all survive and (new) top doesn't belong to us. + */ + if (!partial->key && *partial->p) + /* Writer: end */ + goto no_top; + for (p=partial; p>chain && all_zeroes((__le32*)p->bh->b_data,p->p); p--) + ; + /* + * OK, we've found the last block that must survive. The rest of our + * branch should be detached before unlocking. However, if that rest + * of branch is all ours and does not grow immediately from the inode + * it's easier to cheat and just decrement partial->p. + */ + if (p == chain + k - 1 && p > chain) { + p->p--; + } else { + *top = *p->p; + /* Nope, don't do this in ext4. Must leave the tree intact */ +#if 0 + *p->p = 0; +#endif + } + /* Writer: end */ + + while(partial > p) { + brelse(partial->bh); + partial--; + } +no_top: + return partial; +} + +/* + * Zero a number of block pointers in either an inode or an indirect block. + * If we restart the transaction we must again get write access to the + * indirect block for further modification. + * + * We release `count' blocks on disk, but (last - first) may be greater + * than `count' because there can be holes in there. + */ +static void ext4_clear_blocks(handle_t *handle, struct inode *inode, + struct buffer_head *bh, ext4_fsblk_t block_to_free, + unsigned long count, __le32 *first, __le32 *last) +{ + __le32 *p; + if (try_to_extend_transaction(handle, inode)) { + if (bh) { + BUFFER_TRACE(bh, "call ext4_journal_dirty_metadata"); + ext4_journal_dirty_metadata(handle, bh); + } + ext4_mark_inode_dirty(handle, inode); + ext4_journal_test_restart(handle, inode); + if (bh) { + BUFFER_TRACE(bh, "retaking write access"); + ext4_journal_get_write_access(handle, bh); + } + } + + /* + * Any buffers which are on the journal will be in memory. We find + * them on the hash table so jbd2_journal_revoke() will run jbd2_journal_forget() + * on them. We've already detached each block from the file, so + * bforget() in jbd2_journal_forget() should be safe. + * + * AKPM: turn on bforget in jbd2_journal_forget()!!! + */ + for (p = first; p < last; p++) { + u32 nr = le32_to_cpu(*p); + if (nr) { + struct buffer_head *bh; + + *p = 0; + bh = sb_find_get_block(inode->i_sb, nr); + ext4_forget(handle, 0, inode, bh, nr); + } + } + + ext4_free_blocks(handle, inode, block_to_free, count); +} + +/** + * ext4_free_data - free a list of data blocks + * @handle: handle for this transaction + * @inode: inode we are dealing with + * @this_bh: indirect buffer_head which contains *@first and *@last + * @first: array of block numbers + * @last: points immediately past the end of array + * + * We are freeing all blocks refered from that array (numbers are stored as + * little-endian 32-bit) and updating @inode->i_blocks appropriately. + * + * We accumulate contiguous runs of blocks to free. Conveniently, if these + * blocks are contiguous then releasing them at one time will only affect one + * or two bitmap blocks (+ group descriptor(s) and superblock) and we won't + * actually use a lot of journal space. + * + * @this_bh will be %NULL if @first and @last point into the inode's direct + * block pointers. + */ +static void ext4_free_data(handle_t *handle, struct inode *inode, + struct buffer_head *this_bh, + __le32 *first, __le32 *last) +{ + ext4_fsblk_t block_to_free = 0; /* Starting block # of a run */ + unsigned long count = 0; /* Number of blocks in the run */ + __le32 *block_to_free_p = NULL; /* Pointer into inode/ind + corresponding to + block_to_free */ + ext4_fsblk_t nr; /* Current block # */ + __le32 *p; /* Pointer into inode/ind + for current block */ + int err; + + if (this_bh) { /* For indirect block */ + BUFFER_TRACE(this_bh, "get_write_access"); + err = ext4_journal_get_write_access(handle, this_bh); + /* Important: if we can't update the indirect pointers + * to the blocks, we can't free them. */ + if (err) + return; + } + + for (p = first; p < last; p++) { + nr = le32_to_cpu(*p); + if (nr) { + /* accumulate blocks to free if they're contiguous */ + if (count == 0) { + block_to_free = nr; + block_to_free_p = p; + count = 1; + } else if (nr == block_to_free + count) { + count++; + } else { + ext4_clear_blocks(handle, inode, this_bh, + block_to_free, + count, block_to_free_p, p); + block_to_free = nr; + block_to_free_p = p; + count = 1; + } + } + } + + if (count > 0) + ext4_clear_blocks(handle, inode, this_bh, block_to_free, + count, block_to_free_p, p); + + if (this_bh) { + BUFFER_TRACE(this_bh, "call ext4_journal_dirty_metadata"); + ext4_journal_dirty_metadata(handle, this_bh); + } +} + +/** + * ext4_free_branches - free an array of branches + * @handle: JBD handle for this transaction + * @inode: inode we are dealing with + * @parent_bh: the buffer_head which contains *@first and *@last + * @first: array of block numbers + * @last: pointer immediately past the end of array + * @depth: depth of the branches to free + * + * We are freeing all blocks refered from these branches (numbers are + * stored as little-endian 32-bit) and updating @inode->i_blocks + * appropriately. + */ +static void ext4_free_branches(handle_t *handle, struct inode *inode, + struct buffer_head *parent_bh, + __le32 *first, __le32 *last, int depth) +{ + ext4_fsblk_t nr; + __le32 *p; + + if (is_handle_aborted(handle)) + return; + + if (depth--) { + struct buffer_head *bh; + int addr_per_block = EXT4_ADDR_PER_BLOCK(inode->i_sb); + p = last; + while (--p >= first) { + nr = le32_to_cpu(*p); + if (!nr) + continue; /* A hole */ + + /* Go read the buffer for the next level down */ + bh = sb_bread(inode->i_sb, nr); + + /* + * A read failure? Report error and clear slot + * (should be rare). + */ + if (!bh) { + ext4_error(inode->i_sb, "ext4_free_branches", + "Read failure, inode=%lu, block=%llu", + inode->i_ino, nr); + continue; + } + + /* This zaps the entire block. Bottom up. */ + BUFFER_TRACE(bh, "free child branches"); + ext4_free_branches(handle, inode, bh, + (__le32*)bh->b_data, + (__le32*)bh->b_data + addr_per_block, + depth); + + /* + * We've probably journalled the indirect block several + * times during the truncate. But it's no longer + * needed and we now drop it from the transaction via + * jbd2_journal_revoke(). + * + * That's easy if it's exclusively part of this + * transaction. But if it's part of the committing + * transaction then jbd2_journal_forget() will simply + * brelse() it. That means that if the underlying + * block is reallocated in ext4_get_block(), + * unmap_underlying_metadata() will find this block + * and will try to get rid of it. damn, damn. + * + * If this block has already been committed to the + * journal, a revoke record will be written. And + * revoke records must be emitted *before* clearing + * this block's bit in the bitmaps. + */ + ext4_forget(handle, 1, inode, bh, bh->b_blocknr); + + /* + * Everything below this this pointer has been + * released. Now let this top-of-subtree go. + * + * We want the freeing of this indirect block to be + * atomic in the journal with the updating of the + * bitmap block which owns it. So make some room in + * the journal. + * + * We zero the parent pointer *after* freeing its + * pointee in the bitmaps, so if extend_transaction() + * for some reason fails to put the bitmap changes and + * the release into the same transaction, recovery + * will merely complain about releasing a free block, + * rather than leaking blocks. + */ + if (is_handle_aborted(handle)) + return; + if (try_to_extend_transaction(handle, inode)) { + ext4_mark_inode_dirty(handle, inode); + ext4_journal_test_restart(handle, inode); + } + + ext4_free_blocks(handle, inode, nr, 1); + + if (parent_bh) { + /* + * The block which we have just freed is + * pointed to by an indirect block: journal it + */ + BUFFER_TRACE(parent_bh, "get_write_access"); + if (!ext4_journal_get_write_access(handle, + parent_bh)){ + *p = 0; + BUFFER_TRACE(parent_bh, + "call ext4_journal_dirty_metadata"); + ext4_journal_dirty_metadata(handle, + parent_bh); + } + } + } + } else { + /* We have reached the bottom of the tree. */ + BUFFER_TRACE(parent_bh, "free data blocks"); + ext4_free_data(handle, inode, parent_bh, first, last); + } +} + +/* + * ext4_truncate() + * + * We block out ext4_get_block() block instantiations across the entire + * transaction, and VFS/VM ensures that ext4_truncate() cannot run + * simultaneously on behalf of the same inode. + * + * As we work through the truncate and commmit bits of it to the journal there + * is one core, guiding principle: the file's tree must always be consistent on + * disk. We must be able to restart the truncate after a crash. + * + * The file's tree may be transiently inconsistent in memory (although it + * probably isn't), but whenever we close off and commit a journal transaction, + * the contents of (the filesystem + the journal) must be consistent and + * restartable. It's pretty simple, really: bottom up, right to left (although + * left-to-right works OK too). + * + * Note that at recovery time, journal replay occurs *before* the restart of + * truncate against the orphan inode list. + * + * The committed inode has the new, desired i_size (which is the same as + * i_disksize in this case). After a crash, ext4_orphan_cleanup() will see + * that this inode's truncate did not complete and it will again call + * ext4_truncate() to have another go. So there will be instantiated blocks + * to the right of the truncation point in a crashed ext4 filesystem. But + * that's fine - as long as they are linked from the inode, the post-crash + * ext4_truncate() run will find them and release them. + */ +void ext4_truncate(struct inode *inode) +{ + handle_t *handle; + struct ext4_inode_info *ei = EXT4_I(inode); + __le32 *i_data = ei->i_data; + int addr_per_block = EXT4_ADDR_PER_BLOCK(inode->i_sb); + struct address_space *mapping = inode->i_mapping; + int offsets[4]; + Indirect chain[4]; + Indirect *partial; + __le32 nr = 0; + int n; + long last_block; + unsigned blocksize = inode->i_sb->s_blocksize; + struct page *page; + + if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) || + S_ISLNK(inode->i_mode))) + return; + if (ext4_inode_is_fast_symlink(inode)) + return; + if (IS_APPEND(inode) || IS_IMMUTABLE(inode)) + return; + + /* + * We have to lock the EOF page here, because lock_page() nests + * outside jbd2_journal_start(). + */ + if ((inode->i_size & (blocksize - 1)) == 0) { + /* Block boundary? Nothing to do */ + page = NULL; + } else { + page = grab_cache_page(mapping, + inode->i_size >> PAGE_CACHE_SHIFT); + if (!page) + return; + } + + if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL) + return ext4_ext_truncate(inode, page); + + handle = start_transaction(inode); + if (IS_ERR(handle)) { + if (page) { + clear_highpage(page); + flush_dcache_page(page); + unlock_page(page); + page_cache_release(page); + } + return; /* AKPM: return what? */ + } + + last_block = (inode->i_size + blocksize-1) + >> EXT4_BLOCK_SIZE_BITS(inode->i_sb); + + if (page) + ext4_block_truncate_page(handle, page, mapping, inode->i_size); + + n = ext4_block_to_path(inode, last_block, offsets, NULL); + if (n == 0) + goto out_stop; /* error */ + + /* + * OK. This truncate is going to happen. We add the inode to the + * orphan list, so that if this truncate spans multiple transactions, + * and we crash, we will resume the truncate when the filesystem + * recovers. It also marks the inode dirty, to catch the new size. + * + * Implication: the file must always be in a sane, consistent + * truncatable state while each transaction commits. + */ + if (ext4_orphan_add(handle, inode)) + goto out_stop; + + /* + * The orphan list entry will now protect us from any crash which + * occurs before the truncate completes, so it is now safe to propagate + * the new, shorter inode size (held for now in i_size) into the + * on-disk inode. We do this via i_disksize, which is the value which + * ext4 *really* writes onto the disk inode. + */ + ei->i_disksize = inode->i_size; + + /* + * From here we block out all ext4_get_block() callers who want to + * modify the block allocation tree. + */ + mutex_lock(&ei->truncate_mutex); + + if (n == 1) { /* direct blocks */ + ext4_free_data(handle, inode, NULL, i_data+offsets[0], + i_data + EXT4_NDIR_BLOCKS); + goto do_indirects; + } + + partial = ext4_find_shared(inode, n, offsets, chain, &nr); + /* Kill the top of shared branch (not detached) */ + if (nr) { + if (partial == chain) { + /* Shared branch grows from the inode */ + ext4_free_branches(handle, inode, NULL, + &nr, &nr+1, (chain+n-1) - partial); + *partial->p = 0; + /* + * We mark the inode dirty prior to restart, + * and prior to stop. No need for it here. + */ + } else { + /* Shared branch grows from an indirect block */ + BUFFER_TRACE(partial->bh, "get_write_access"); + ext4_free_branches(handle, inode, partial->bh, + partial->p, + partial->p+1, (chain+n-1) - partial); + } + } + /* Clear the ends of indirect blocks on the shared branch */ + while (partial > chain) { + ext4_free_branches(handle, inode, partial->bh, partial->p + 1, + (__le32*)partial->bh->b_data+addr_per_block, + (chain+n-1) - partial); + BUFFER_TRACE(partial->bh, "call brelse"); + brelse (partial->bh); + partial--; + } +do_indirects: + /* Kill the remaining (whole) subtrees */ + switch (offsets[0]) { + default: + nr = i_data[EXT4_IND_BLOCK]; + if (nr) { + ext4_free_branches(handle, inode, NULL, &nr, &nr+1, 1); + i_data[EXT4_IND_BLOCK] = 0; + } + case EXT4_IND_BLOCK: + nr = i_data[EXT4_DIND_BLOCK]; + if (nr) { + ext4_free_branches(handle, inode, NULL, &nr, &nr+1, 2); + i_data[EXT4_DIND_BLOCK] = 0; + } + case EXT4_DIND_BLOCK: + nr = i_data[EXT4_TIND_BLOCK]; + if (nr) { + ext4_free_branches(handle, inode, NULL, &nr, &nr+1, 3); + i_data[EXT4_TIND_BLOCK] = 0; + } + case EXT4_TIND_BLOCK: + ; + } + + ext4_discard_reservation(inode); + + mutex_unlock(&ei->truncate_mutex); + inode->i_mtime = inode->i_ctime = CURRENT_TIME_SEC; + ext4_mark_inode_dirty(handle, inode); + + /* + * In a multi-transaction truncate, we only make the final transaction + * synchronous + */ + if (IS_SYNC(inode)) + handle->h_sync = 1; +out_stop: + /* + * If this was a simple ftruncate(), and the file will remain alive + * then we need to clear up the orphan record which we created above. + * However, if this was a real unlink then we were called by + * ext4_delete_inode(), and we allow that function to clean up the + * orphan info for us. + */ + if (inode->i_nlink) + ext4_orphan_del(handle, inode); + + ext4_journal_stop(handle); +} + +static ext4_fsblk_t ext4_get_inode_block(struct super_block *sb, + unsigned long ino, struct ext4_iloc *iloc) +{ + unsigned long desc, group_desc, block_group; + unsigned long offset; + ext4_fsblk_t block; + struct buffer_head *bh; + struct ext4_group_desc * gdp; + + if (!ext4_valid_inum(sb, ino)) { + /* + * This error is already checked for in namei.c unless we are + * looking at an NFS filehandle, in which case no error + * report is needed + */ + return 0; + } + + block_group = (ino - 1) / EXT4_INODES_PER_GROUP(sb); + if (block_group >= EXT4_SB(sb)->s_groups_count) { + ext4_error(sb,"ext4_get_inode_block","group >= groups count"); + return 0; + } + smp_rmb(); + group_desc = block_group >> EXT4_DESC_PER_BLOCK_BITS(sb); + desc = block_group & (EXT4_DESC_PER_BLOCK(sb) - 1); + bh = EXT4_SB(sb)->s_group_desc[group_desc]; + if (!bh) { + ext4_error (sb, "ext4_get_inode_block", + "Descriptor not loaded"); + return 0; + } + + gdp = (struct ext4_group_desc *)((__u8 *)bh->b_data + + desc * EXT4_DESC_SIZE(sb)); + /* + * Figure out the offset within the block group inode table + */ + offset = ((ino - 1) % EXT4_INODES_PER_GROUP(sb)) * + EXT4_INODE_SIZE(sb); + block = ext4_inode_table(sb, gdp) + + (offset >> EXT4_BLOCK_SIZE_BITS(sb)); + + iloc->block_group = block_group; + iloc->offset = offset & (EXT4_BLOCK_SIZE(sb) - 1); + return block; +} + +/* + * ext4_get_inode_loc returns with an extra refcount against the inode's + * underlying buffer_head on success. If 'in_mem' is true, we have all + * data in memory that is needed to recreate the on-disk version of this + * inode. + */ +static int __ext4_get_inode_loc(struct inode *inode, + struct ext4_iloc *iloc, int in_mem) +{ + ext4_fsblk_t block; + struct buffer_head *bh; + + block = ext4_get_inode_block(inode->i_sb, inode->i_ino, iloc); + if (!block) + return -EIO; + + bh = sb_getblk(inode->i_sb, block); + if (!bh) { + ext4_error (inode->i_sb, "ext4_get_inode_loc", + "unable to read inode block - " + "inode=%lu, block=%llu", + inode->i_ino, block); + return -EIO; + } + if (!buffer_uptodate(bh)) { + lock_buffer(bh); + if (buffer_uptodate(bh)) { + /* someone brought it uptodate while we waited */ + unlock_buffer(bh); + goto has_buffer; + } + + /* + * If we have all information of the inode in memory and this + * is the only valid inode in the block, we need not read the + * block. + */ + if (in_mem) { + struct buffer_head *bitmap_bh; + struct ext4_group_desc *desc; + int inodes_per_buffer; + int inode_offset, i; + int block_group; + int start; + + block_group = (inode->i_ino - 1) / + EXT4_INODES_PER_GROUP(inode->i_sb); + inodes_per_buffer = bh->b_size / + EXT4_INODE_SIZE(inode->i_sb); + inode_offset = ((inode->i_ino - 1) % + EXT4_INODES_PER_GROUP(inode->i_sb)); + start = inode_offset & ~(inodes_per_buffer - 1); + + /* Is the inode bitmap in cache? */ + desc = ext4_get_group_desc(inode->i_sb, + block_group, NULL); + if (!desc) + goto make_io; + + bitmap_bh = sb_getblk(inode->i_sb, + ext4_inode_bitmap(inode->i_sb, desc)); + if (!bitmap_bh) + goto make_io; + + /* + * If the inode bitmap isn't in cache then the + * optimisation may end up performing two reads instead + * of one, so skip it. + */ + if (!buffer_uptodate(bitmap_bh)) { + brelse(bitmap_bh); + goto make_io; + } + for (i = start; i < start + inodes_per_buffer; i++) { + if (i == inode_offset) + continue; + if (ext4_test_bit(i, bitmap_bh->b_data)) + break; + } + brelse(bitmap_bh); + if (i == start + inodes_per_buffer) { + /* all other inodes are free, so skip I/O */ + memset(bh->b_data, 0, bh->b_size); + set_buffer_uptodate(bh); + unlock_buffer(bh); + goto has_buffer; + } + } + +make_io: + /* + * There are other valid inodes in the buffer, this inode + * has in-inode xattrs, or we don't have this inode in memory. + * Read the block from disk. + */ + get_bh(bh); + bh->b_end_io = end_buffer_read_sync; + submit_bh(READ_META, bh); + wait_on_buffer(bh); + if (!buffer_uptodate(bh)) { + ext4_error(inode->i_sb, "ext4_get_inode_loc", + "unable to read inode block - " + "inode=%lu, block=%llu", + inode->i_ino, block); + brelse(bh); + return -EIO; + } + } +has_buffer: + iloc->bh = bh; + return 0; +} + +int ext4_get_inode_loc(struct inode *inode, struct ext4_iloc *iloc) +{ + /* We have all inode data except xattrs in memory here. */ + return __ext4_get_inode_loc(inode, iloc, + !(EXT4_I(inode)->i_state & EXT4_STATE_XATTR)); +} + +void ext4_set_inode_flags(struct inode *inode) +{ + unsigned int flags = EXT4_I(inode)->i_flags; + + inode->i_flags &= ~(S_SYNC|S_APPEND|S_IMMUTABLE|S_NOATIME|S_DIRSYNC); + if (flags & EXT4_SYNC_FL) + inode->i_flags |= S_SYNC; + if (flags & EXT4_APPEND_FL) + inode->i_flags |= S_APPEND; + if (flags & EXT4_IMMUTABLE_FL) + inode->i_flags |= S_IMMUTABLE; + if (flags & EXT4_NOATIME_FL) + inode->i_flags |= S_NOATIME; + if (flags & EXT4_DIRSYNC_FL) + inode->i_flags |= S_DIRSYNC; +} + +void ext4_read_inode(struct inode * inode) +{ + struct ext4_iloc iloc; + struct ext4_inode *raw_inode; + struct ext4_inode_info *ei = EXT4_I(inode); + struct buffer_head *bh; + int block; + +#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL + ei->i_acl = EXT4_ACL_NOT_CACHED; + ei->i_default_acl = EXT4_ACL_NOT_CACHED; +#endif + ei->i_block_alloc_info = NULL; + + if (__ext4_get_inode_loc(inode, &iloc, 0)) + goto bad_inode; + bh = iloc.bh; + raw_inode = ext4_raw_inode(&iloc); + inode->i_mode = le16_to_cpu(raw_inode->i_mode); + inode->i_uid = (uid_t)le16_to_cpu(raw_inode->i_uid_low); + inode->i_gid = (gid_t)le16_to_cpu(raw_inode->i_gid_low); + if(!(test_opt (inode->i_sb, NO_UID32))) { + inode->i_uid |= le16_to_cpu(raw_inode->i_uid_high) << 16; + inode->i_gid |= le16_to_cpu(raw_inode->i_gid_high) << 16; + } + inode->i_nlink = le16_to_cpu(raw_inode->i_links_count); + inode->i_size = le32_to_cpu(raw_inode->i_size); + inode->i_atime.tv_sec = le32_to_cpu(raw_inode->i_atime); + inode->i_ctime.tv_sec = le32_to_cpu(raw_inode->i_ctime); + inode->i_mtime.tv_sec = le32_to_cpu(raw_inode->i_mtime); + inode->i_atime.tv_nsec = inode->i_ctime.tv_nsec = inode->i_mtime.tv_nsec = 0; + + ei->i_state = 0; + ei->i_dir_start_lookup = 0; + ei->i_dtime = le32_to_cpu(raw_inode->i_dtime); + /* We now have enough fields to check if the inode was active or not. + * This is needed because nfsd might try to access dead inodes + * the test is that same one that e2fsck uses + * NeilBrown 1999oct15 + */ + if (inode->i_nlink == 0) { + if (inode->i_mode == 0 || + !(EXT4_SB(inode->i_sb)->s_mount_state & EXT4_ORPHAN_FS)) { + /* this inode is deleted */ + brelse (bh); + goto bad_inode; + } + /* The only unlinked inodes we let through here have + * valid i_mode and are being read by the orphan + * recovery code: that's fine, we're about to complete + * the process of deleting those. */ + } + inode->i_blocks = le32_to_cpu(raw_inode->i_blocks); + ei->i_flags = le32_to_cpu(raw_inode->i_flags); +#ifdef EXT4_FRAGMENTS + ei->i_faddr = le32_to_cpu(raw_inode->i_faddr); + ei->i_frag_no = raw_inode->i_frag; + ei->i_frag_size = raw_inode->i_fsize; +#endif + ei->i_file_acl = le32_to_cpu(raw_inode->i_file_acl); + if (EXT4_SB(inode->i_sb)->s_es->s_creator_os != + cpu_to_le32(EXT4_OS_HURD)) + ei->i_file_acl |= + ((__u64)le16_to_cpu(raw_inode->i_file_acl_high)) << 32; + if (!S_ISREG(inode->i_mode)) { + ei->i_dir_acl = le32_to_cpu(raw_inode->i_dir_acl); + } else { + inode->i_size |= + ((__u64)le32_to_cpu(raw_inode->i_size_high)) << 32; + } + ei->i_disksize = inode->i_size; + inode->i_generation = le32_to_cpu(raw_inode->i_generation); + ei->i_block_group = iloc.block_group; + /* + * NOTE! The in-memory inode i_data array is in little-endian order + * even on big-endian machines: we do NOT byteswap the block numbers! + */ + for (block = 0; block < EXT4_N_BLOCKS; block++) + ei->i_data[block] = raw_inode->i_block[block]; + INIT_LIST_HEAD(&ei->i_orphan); + + if (inode->i_ino >= EXT4_FIRST_INO(inode->i_sb) + 1 && + EXT4_INODE_SIZE(inode->i_sb) > EXT4_GOOD_OLD_INODE_SIZE) { + /* + * When mke2fs creates big inodes it does not zero out + * the unused bytes above EXT4_GOOD_OLD_INODE_SIZE, + * so ignore those first few inodes. + */ + ei->i_extra_isize = le16_to_cpu(raw_inode->i_extra_isize); + if (EXT4_GOOD_OLD_INODE_SIZE + ei->i_extra_isize > + EXT4_INODE_SIZE(inode->i_sb)) + goto bad_inode; + if (ei->i_extra_isize == 0) { + /* The extra space is currently unused. Use it. */ + ei->i_extra_isize = sizeof(struct ext4_inode) - + EXT4_GOOD_OLD_INODE_SIZE; + } else { + __le32 *magic = (void *)raw_inode + + EXT4_GOOD_OLD_INODE_SIZE + + ei->i_extra_isize; + if (*magic == cpu_to_le32(EXT4_XATTR_MAGIC)) + ei->i_state |= EXT4_STATE_XATTR; + } + } else + ei->i_extra_isize = 0; + + if (S_ISREG(inode->i_mode)) { + inode->i_op = &ext4_file_inode_operations; + inode->i_fop = &ext4_file_operations; + ext4_set_aops(inode); + } else if (S_ISDIR(inode->i_mode)) { + inode->i_op = &ext4_dir_inode_operations; + inode->i_fop = &ext4_dir_operations; + } else if (S_ISLNK(inode->i_mode)) { + if (ext4_inode_is_fast_symlink(inode)) + inode->i_op = &ext4_fast_symlink_inode_operations; + else { + inode->i_op = &ext4_symlink_inode_operations; + ext4_set_aops(inode); + } + } else { + inode->i_op = &ext4_special_inode_operations; + if (raw_inode->i_block[0]) + init_special_inode(inode, inode->i_mode, + old_decode_dev(le32_to_cpu(raw_inode->i_block[0]))); + else + init_special_inode(inode, inode->i_mode, + new_decode_dev(le32_to_cpu(raw_inode->i_block[1]))); + } + brelse (iloc.bh); + ext4_set_inode_flags(inode); + return; + +bad_inode: + make_bad_inode(inode); + return; +} + +/* + * Post the struct inode info into an on-disk inode location in the + * buffer-cache. This gobbles the caller's reference to the + * buffer_head in the inode location struct. + * + * The caller must have write access to iloc->bh. + */ +static int ext4_do_update_inode(handle_t *handle, + struct inode *inode, + struct ext4_iloc *iloc) +{ + struct ext4_inode *raw_inode = ext4_raw_inode(iloc); + struct ext4_inode_info *ei = EXT4_I(inode); + struct buffer_head *bh = iloc->bh; + int err = 0, rc, block; + + /* For fields not not tracking in the in-memory inode, + * initialise them to zero for new inodes. */ + if (ei->i_state & EXT4_STATE_NEW) + memset(raw_inode, 0, EXT4_SB(inode->i_sb)->s_inode_size); + + raw_inode->i_mode = cpu_to_le16(inode->i_mode); + if(!(test_opt(inode->i_sb, NO_UID32))) { + raw_inode->i_uid_low = cpu_to_le16(low_16_bits(inode->i_uid)); + raw_inode->i_gid_low = cpu_to_le16(low_16_bits(inode->i_gid)); +/* + * Fix up interoperability with old kernels. Otherwise, old inodes get + * re-used with the upper 16 bits of the uid/gid intact + */ + if(!ei->i_dtime) { + raw_inode->i_uid_high = + cpu_to_le16(high_16_bits(inode->i_uid)); + raw_inode->i_gid_high = + cpu_to_le16(high_16_bits(inode->i_gid)); + } else { + raw_inode->i_uid_high = 0; + raw_inode->i_gid_high = 0; + } + } else { + raw_inode->i_uid_low = + cpu_to_le16(fs_high2lowuid(inode->i_uid)); + raw_inode->i_gid_low = + cpu_to_le16(fs_high2lowgid(inode->i_gid)); + raw_inode->i_uid_high = 0; + raw_inode->i_gid_high = 0; + } + raw_inode->i_links_count = cpu_to_le16(inode->i_nlink); + raw_inode->i_size = cpu_to_le32(ei->i_disksize); + raw_inode->i_atime = cpu_to_le32(inode->i_atime.tv_sec); + raw_inode->i_ctime = cpu_to_le32(inode->i_ctime.tv_sec); + raw_inode->i_mtime = cpu_to_le32(inode->i_mtime.tv_sec); + raw_inode->i_blocks = cpu_to_le32(inode->i_blocks); + raw_inode->i_dtime = cpu_to_le32(ei->i_dtime); + raw_inode->i_flags = cpu_to_le32(ei->i_flags); +#ifdef EXT4_FRAGMENTS + raw_inode->i_faddr = cpu_to_le32(ei->i_faddr); + raw_inode->i_frag = ei->i_frag_no; + raw_inode->i_fsize = ei->i_frag_size; +#endif + if (EXT4_SB(inode->i_sb)->s_es->s_creator_os != + cpu_to_le32(EXT4_OS_HURD)) + raw_inode->i_file_acl_high = + cpu_to_le16(ei->i_file_acl >> 32); + raw_inode->i_file_acl = cpu_to_le32(ei->i_file_acl); + if (!S_ISREG(inode->i_mode)) { + raw_inode->i_dir_acl = cpu_to_le32(ei->i_dir_acl); + } else { + raw_inode->i_size_high = + cpu_to_le32(ei->i_disksize >> 32); + if (ei->i_disksize > 0x7fffffffULL) { + struct super_block *sb = inode->i_sb; + if (!EXT4_HAS_RO_COMPAT_FEATURE(sb, + EXT4_FEATURE_RO_COMPAT_LARGE_FILE) || + EXT4_SB(sb)->s_es->s_rev_level == + cpu_to_le32(EXT4_GOOD_OLD_REV)) { + /* If this is the first large file + * created, add a flag to the superblock. + */ + err = ext4_journal_get_write_access(handle, + EXT4_SB(sb)->s_sbh); + if (err) + goto out_brelse; + ext4_update_dynamic_rev(sb); + EXT4_SET_RO_COMPAT_FEATURE(sb, + EXT4_FEATURE_RO_COMPAT_LARGE_FILE); + sb->s_dirt = 1; + handle->h_sync = 1; + err = ext4_journal_dirty_metadata(handle, + EXT4_SB(sb)->s_sbh); + } + } + } + raw_inode->i_generation = cpu_to_le32(inode->i_generation); + if (S_ISCHR(inode->i_mode) || S_ISBLK(inode->i_mode)) { + if (old_valid_dev(inode->i_rdev)) { + raw_inode->i_block[0] = + cpu_to_le32(old_encode_dev(inode->i_rdev)); + raw_inode->i_block[1] = 0; + } else { + raw_inode->i_block[0] = 0; + raw_inode->i_block[1] = + cpu_to_le32(new_encode_dev(inode->i_rdev)); + raw_inode->i_block[2] = 0; + } + } else for (block = 0; block < EXT4_N_BLOCKS; block++) + raw_inode->i_block[block] = ei->i_data[block]; + + if (ei->i_extra_isize) + raw_inode->i_extra_isize = cpu_to_le16(ei->i_extra_isize); + + BUFFER_TRACE(bh, "call ext4_journal_dirty_metadata"); + rc = ext4_journal_dirty_metadata(handle, bh); + if (!err) + err = rc; + ei->i_state &= ~EXT4_STATE_NEW; + +out_brelse: + brelse (bh); + ext4_std_error(inode->i_sb, err); + return err; +} + +/* + * ext4_write_inode() + * + * We are called from a few places: + * + * - Within generic_file_write() for O_SYNC files. + * Here, there will be no transaction running. We wait for any running + * trasnaction to commit. + * + * - Within sys_sync(), kupdate and such. + * We wait on commit, if tol to. + * + * - Within prune_icache() (PF_MEMALLOC == true) + * Here we simply return. We can't afford to block kswapd on the + * journal commit. + * + * In all cases it is actually safe for us to return without doing anything, + * because the inode has been copied into a raw inode buffer in + * ext4_mark_inode_dirty(). This is a correctness thing for O_SYNC and for + * knfsd. + * + * Note that we are absolutely dependent upon all inode dirtiers doing the + * right thing: they *must* call mark_inode_dirty() after dirtying info in + * which we are interested. + * + * It would be a bug for them to not do this. The code: + * + * mark_inode_dirty(inode) + * stuff(); + * inode->i_size = expr; + * + * is in error because a kswapd-driven write_inode() could occur while + * `stuff()' is running, and the new i_size will be lost. Plus the inode + * will no longer be on the superblock's dirty inode list. + */ +int ext4_write_inode(struct inode *inode, int wait) +{ + if (current->flags & PF_MEMALLOC) + return 0; + + if (ext4_journal_current_handle()) { + jbd_debug(0, "called recursively, non-PF_MEMALLOC!\n"); + dump_stack(); + return -EIO; + } + + if (!wait) + return 0; + + return ext4_force_commit(inode->i_sb); +} + +/* + * ext4_setattr() + * + * Called from notify_change. + * + * We want to trap VFS attempts to truncate the file as soon as + * possible. In particular, we want to make sure that when the VFS + * shrinks i_size, we put the inode on the orphan list and modify + * i_disksize immediately, so that during the subsequent flushing of + * dirty pages and freeing of disk blocks, we can guarantee that any + * commit will leave the blocks being flushed in an unused state on + * disk. (On recovery, the inode will get truncated and the blocks will + * be freed, so we have a strong guarantee that no future commit will + * leave these blocks visible to the user.) + * + * Called with inode->sem down. + */ +int ext4_setattr(struct dentry *dentry, struct iattr *attr) +{ + struct inode *inode = dentry->d_inode; + int error, rc = 0; + const unsigned int ia_valid = attr->ia_valid; + + error = inode_change_ok(inode, attr); + if (error) + return error; + + if ((ia_valid & ATTR_UID && attr->ia_uid != inode->i_uid) || + (ia_valid & ATTR_GID && attr->ia_gid != inode->i_gid)) { + handle_t *handle; + + /* (user+group)*(old+new) structure, inode write (sb, + * inode block, ? - but truncate inode update has it) */ + handle = ext4_journal_start(inode, 2*(EXT4_QUOTA_INIT_BLOCKS(inode->i_sb)+ + EXT4_QUOTA_DEL_BLOCKS(inode->i_sb))+3); + if (IS_ERR(handle)) { + error = PTR_ERR(handle); + goto err_out; + } + error = DQUOT_TRANSFER(inode, attr) ? -EDQUOT : 0; + if (error) { + ext4_journal_stop(handle); + return error; + } + /* Update corresponding info in inode so that everything is in + * one transaction */ + if (attr->ia_valid & ATTR_UID) + inode->i_uid = attr->ia_uid; + if (attr->ia_valid & ATTR_GID) + inode->i_gid = attr->ia_gid; + error = ext4_mark_inode_dirty(handle, inode); + ext4_journal_stop(handle); + } + + if (S_ISREG(inode->i_mode) && + attr->ia_valid & ATTR_SIZE && attr->ia_size < inode->i_size) { + handle_t *handle; + + handle = ext4_journal_start(inode, 3); + if (IS_ERR(handle)) { + error = PTR_ERR(handle); + goto err_out; + } + + error = ext4_orphan_add(handle, inode); + EXT4_I(inode)->i_disksize = attr->ia_size; + rc = ext4_mark_inode_dirty(handle, inode); + if (!error) + error = rc; + ext4_journal_stop(handle); + } + + rc = inode_setattr(inode, attr); + + /* If inode_setattr's call to ext4_truncate failed to get a + * transaction handle at all, we need to clean up the in-core + * orphan list manually. */ + if (inode->i_nlink) + ext4_orphan_del(NULL, inode); + + if (!rc && (ia_valid & ATTR_MODE)) + rc = ext4_acl_chmod(inode); + +err_out: + ext4_std_error(inode->i_sb, error); + if (!error) + error = rc; + return error; +} + + +/* + * How many blocks doth make a writepage()? + * + * With N blocks per page, it may be: + * N data blocks + * 2 indirect block + * 2 dindirect + * 1 tindirect + * N+5 bitmap blocks (from the above) + * N+5 group descriptor summary blocks + * 1 inode block + * 1 superblock. + * 2 * EXT4_SINGLEDATA_TRANS_BLOCKS for the quote files + * + * 3 * (N + 5) + 2 + 2 * EXT4_SINGLEDATA_TRANS_BLOCKS + * + * With ordered or writeback data it's the same, less the N data blocks. + * + * If the inode's direct blocks can hold an integral number of pages then a + * page cannot straddle two indirect blocks, and we can only touch one indirect + * and dindirect block, and the "5" above becomes "3". + * + * This still overestimates under most circumstances. If we were to pass the + * start and end offsets in here as well we could do block_to_path() on each + * block and work out the exact number of indirects which are touched. Pah. + */ + +int ext4_writepage_trans_blocks(struct inode *inode) +{ + int bpp = ext4_journal_blocks_per_page(inode); + int indirects = (EXT4_NDIR_BLOCKS % bpp) ? 5 : 3; + int ret; + + if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL) + return ext4_ext_writepage_trans_blocks(inode, bpp); + + if (ext4_should_journal_data(inode)) + ret = 3 * (bpp + indirects) + 2; + else + ret = 2 * (bpp + indirects) + 2; + +#ifdef CONFIG_QUOTA + /* We know that structure was already allocated during DQUOT_INIT so + * we will be updating only the data blocks + inodes */ + ret += 2*EXT4_QUOTA_TRANS_BLOCKS(inode->i_sb); +#endif + + return ret; +} + +/* + * The caller must have previously called ext4_reserve_inode_write(). + * Give this, we know that the caller already has write access to iloc->bh. + */ +int ext4_mark_iloc_dirty(handle_t *handle, + struct inode *inode, struct ext4_iloc *iloc) +{ + int err = 0; + + /* the do_update_inode consumes one bh->b_count */ + get_bh(iloc->bh); + + /* ext4_do_update_inode() does jbd2_journal_dirty_metadata */ + err = ext4_do_update_inode(handle, inode, iloc); + put_bh(iloc->bh); + return err; +} + +/* + * On success, We end up with an outstanding reference count against + * iloc->bh. This _must_ be cleaned up later. + */ + +int +ext4_reserve_inode_write(handle_t *handle, struct inode *inode, + struct ext4_iloc *iloc) +{ + int err = 0; + if (handle) { + err = ext4_get_inode_loc(inode, iloc); + if (!err) { + BUFFER_TRACE(iloc->bh, "get_write_access"); + err = ext4_journal_get_write_access(handle, iloc->bh); + if (err) { + brelse(iloc->bh); + iloc->bh = NULL; + } + } + } + ext4_std_error(inode->i_sb, err); + return err; +} + +/* + * What we do here is to mark the in-core inode as clean with respect to inode + * dirtiness (it may still be data-dirty). + * This means that the in-core inode may be reaped by prune_icache + * without having to perform any I/O. This is a very good thing, + * because *any* task may call prune_icache - even ones which + * have a transaction open against a different journal. + * + * Is this cheating? Not really. Sure, we haven't written the + * inode out, but prune_icache isn't a user-visible syncing function. + * Whenever the user wants stuff synced (sys_sync, sys_msync, sys_fsync) + * we start and wait on commits. + * + * Is this efficient/effective? Well, we're being nice to the system + * by cleaning up our inodes proactively so they can be reaped + * without I/O. But we are potentially leaving up to five seconds' + * worth of inodes floating about which prune_icache wants us to + * write out. One way to fix that would be to get prune_icache() + * to do a write_super() to free up some memory. It has the desired + * effect. + */ +int ext4_mark_inode_dirty(handle_t *handle, struct inode *inode) +{ + struct ext4_iloc iloc; + int err; + + might_sleep(); + err = ext4_reserve_inode_write(handle, inode, &iloc); + if (!err) + err = ext4_mark_iloc_dirty(handle, inode, &iloc); + return err; +} + +/* + * ext4_dirty_inode() is called from __mark_inode_dirty() + * + * We're really interested in the case where a file is being extended. + * i_size has been changed by generic_commit_write() and we thus need + * to include the updated inode in the current transaction. + * + * Also, DQUOT_ALLOC_SPACE() will always dirty the inode when blocks + * are allocated to the file. + * + * If the inode is marked synchronous, we don't honour that here - doing + * so would cause a commit on atime updates, which we don't bother doing. + * We handle synchronous inodes at the highest possible level. + */ +void ext4_dirty_inode(struct inode *inode) +{ + handle_t *current_handle = ext4_journal_current_handle(); + handle_t *handle; + + handle = ext4_journal_start(inode, 2); + if (IS_ERR(handle)) + goto out; + if (current_handle && + current_handle->h_transaction != handle->h_transaction) { + /* This task has a transaction open against a different fs */ + printk(KERN_EMERG "%s: transactions do not match!\n", + __FUNCTION__); + } else { + jbd_debug(5, "marking dirty. outer handle=%p\n", + current_handle); + ext4_mark_inode_dirty(handle, inode); + } + ext4_journal_stop(handle); +out: + return; +} + +#if 0 +/* + * Bind an inode's backing buffer_head into this transaction, to prevent + * it from being flushed to disk early. Unlike + * ext4_reserve_inode_write, this leaves behind no bh reference and + * returns no iloc structure, so the caller needs to repeat the iloc + * lookup to mark the inode dirty later. + */ +static int ext4_pin_inode(handle_t *handle, struct inode *inode) +{ + struct ext4_iloc iloc; + + int err = 0; + if (handle) { + err = ext4_get_inode_loc(inode, &iloc); + if (!err) { + BUFFER_TRACE(iloc.bh, "get_write_access"); + err = jbd2_journal_get_write_access(handle, iloc.bh); + if (!err) + err = ext4_journal_dirty_metadata(handle, + iloc.bh); + brelse(iloc.bh); + } + } + ext4_std_error(inode->i_sb, err); + return err; +} +#endif + +int ext4_change_inode_journal_flag(struct inode *inode, int val) +{ + journal_t *journal; + handle_t *handle; + int err; + + /* + * We have to be very careful here: changing a data block's + * journaling status dynamically is dangerous. If we write a + * data block to the journal, change the status and then delete + * that block, we risk forgetting to revoke the old log record + * from the journal and so a subsequent replay can corrupt data. + * So, first we make sure that the journal is empty and that + * nobody is changing anything. + */ + + journal = EXT4_JOURNAL(inode); + if (is_journal_aborted(journal) || IS_RDONLY(inode)) + return -EROFS; + + jbd2_journal_lock_updates(journal); + jbd2_journal_flush(journal); + + /* + * OK, there are no updates running now, and all cached data is + * synced to disk. We are now in a completely consistent state + * which doesn't have anything in the journal, and we know that + * no filesystem updates are running, so it is safe to modify + * the inode's in-core data-journaling state flag now. + */ + + if (val) + EXT4_I(inode)->i_flags |= EXT4_JOURNAL_DATA_FL; + else + EXT4_I(inode)->i_flags &= ~EXT4_JOURNAL_DATA_FL; + ext4_set_aops(inode); + + jbd2_journal_unlock_updates(journal); + + /* Finally we can mark the inode as dirty. */ + + handle = ext4_journal_start(inode, 1); + if (IS_ERR(handle)) + return PTR_ERR(handle); + + err = ext4_mark_inode_dirty(handle, inode); + handle->h_sync = 1; + ext4_journal_stop(handle); + ext4_std_error(inode->i_sb, err); + + return err; +} diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c new file mode 100644 index 000000000000..22a737c306c7 --- /dev/null +++ b/fs/ext4/ioctl.c @@ -0,0 +1,306 @@ +/* + * linux/fs/ext4/ioctl.c + * + * Copyright (C) 1993, 1994, 1995 + * Remy Card (card@masi.ibp.fr) + * Laboratoire MASI - Institut Blaise Pascal + * Universite Pierre et Marie Curie (Paris VI) + */ + +#include <linux/fs.h> +#include <linux/jbd2.h> +#include <linux/capability.h> +#include <linux/ext4_fs.h> +#include <linux/ext4_jbd2.h> +#include <linux/time.h> +#include <linux/compat.h> +#include <linux/smp_lock.h> +#include <asm/uaccess.h> + +int ext4_ioctl (struct inode * inode, struct file * filp, unsigned int cmd, + unsigned long arg) +{ + struct ext4_inode_info *ei = EXT4_I(inode); + unsigned int flags; + unsigned short rsv_window_size; + + ext4_debug ("cmd = %u, arg = %lu\n", cmd, arg); + + switch (cmd) { + case EXT4_IOC_GETFLAGS: + flags = ei->i_flags & EXT4_FL_USER_VISIBLE; + return put_user(flags, (int __user *) arg); + case EXT4_IOC_SETFLAGS: { + handle_t *handle = NULL; + int err; + struct ext4_iloc iloc; + unsigned int oldflags; + unsigned int jflag; + + if (IS_RDONLY(inode)) + return -EROFS; + + if ((current->fsuid != inode->i_uid) && !capable(CAP_FOWNER)) + return -EACCES; + + if (get_user(flags, (int __user *) arg)) + return -EFAULT; + + if (!S_ISDIR(inode->i_mode)) + flags &= ~EXT4_DIRSYNC_FL; + + mutex_lock(&inode->i_mutex); + oldflags = ei->i_flags; + + /* The JOURNAL_DATA flag is modifiable only by root */ + jflag = flags & EXT4_JOURNAL_DATA_FL; + + /* + * The IMMUTABLE and APPEND_ONLY flags can only be changed by + * the relevant capability. + * + * This test looks nicer. Thanks to Pauline Middelink + */ + if ((flags ^ oldflags) & (EXT4_APPEND_FL | EXT4_IMMUTABLE_FL)) { + if (!capable(CAP_LINUX_IMMUTABLE)) { + mutex_unlock(&inode->i_mutex); + return -EPERM; + } + } + + /* + * The JOURNAL_DATA flag can only be changed by + * the relevant capability. + */ + if ((jflag ^ oldflags) & (EXT4_JOURNAL_DATA_FL)) { + if (!capable(CAP_SYS_RESOURCE)) { + mutex_unlock(&inode->i_mutex); + return -EPERM; + } + } + + + handle = ext4_journal_start(inode, 1); + if (IS_ERR(handle)) { + mutex_unlock(&inode->i_mutex); + return PTR_ERR(handle); + } + if (IS_SYNC(inode)) + handle->h_sync = 1; + err = ext4_reserve_inode_write(handle, inode, &iloc); + if (err) + goto flags_err; + + flags = flags & EXT4_FL_USER_MODIFIABLE; + flags |= oldflags & ~EXT4_FL_USER_MODIFIABLE; + ei->i_flags = flags; + + ext4_set_inode_flags(inode); + inode->i_ctime = CURRENT_TIME_SEC; + + err = ext4_mark_iloc_dirty(handle, inode, &iloc); +flags_err: + ext4_journal_stop(handle); + if (err) { + mutex_unlock(&inode->i_mutex); + return err; + } + + if ((jflag ^ oldflags) & (EXT4_JOURNAL_DATA_FL)) + err = ext4_change_inode_journal_flag(inode, jflag); + mutex_unlock(&inode->i_mutex); + return err; + } + case EXT4_IOC_GETVERSION: + case EXT4_IOC_GETVERSION_OLD: + return put_user(inode->i_generation, (int __user *) arg); + case EXT4_IOC_SETVERSION: + case EXT4_IOC_SETVERSION_OLD: { + handle_t *handle; + struct ext4_iloc iloc; + __u32 generation; + int err; + + if ((current->fsuid != inode->i_uid) && !capable(CAP_FOWNER)) + return -EPERM; + if (IS_RDONLY(inode)) + return -EROFS; + if (get_user(generation, (int __user *) arg)) + return -EFAULT; + + handle = ext4_journal_start(inode, 1); + if (IS_ERR(handle)) + return PTR_ERR(handle); + err = ext4_reserve_inode_write(handle, inode, &iloc); + if (err == 0) { + inode->i_ctime = CURRENT_TIME_SEC; + inode->i_generation = generation; + err = ext4_mark_iloc_dirty(handle, inode, &iloc); + } + ext4_journal_stop(handle); + return err; + } +#ifdef CONFIG_JBD_DEBUG + case EXT4_IOC_WAIT_FOR_READONLY: + /* + * This is racy - by the time we're woken up and running, + * the superblock could be released. And the module could + * have been unloaded. So sue me. + * + * Returns 1 if it slept, else zero. + */ + { + struct super_block *sb = inode->i_sb; + DECLARE_WAITQUEUE(wait, current); + int ret = 0; + + set_current_state(TASK_INTERRUPTIBLE); + add_wait_queue(&EXT4_SB(sb)->ro_wait_queue, &wait); + if (timer_pending(&EXT4_SB(sb)->turn_ro_timer)) { + schedule(); + ret = 1; + } + remove_wait_queue(&EXT4_SB(sb)->ro_wait_queue, &wait); + return ret; + } +#endif + case EXT4_IOC_GETRSVSZ: + if (test_opt(inode->i_sb, RESERVATION) + && S_ISREG(inode->i_mode) + && ei->i_block_alloc_info) { + rsv_window_size = ei->i_block_alloc_info->rsv_window_node.rsv_goal_size; + return put_user(rsv_window_size, (int __user *)arg); + } + return -ENOTTY; + case EXT4_IOC_SETRSVSZ: { + + if (!test_opt(inode->i_sb, RESERVATION) ||!S_ISREG(inode->i_mode)) + return -ENOTTY; + + if (IS_RDONLY(inode)) + return -EROFS; + + if ((current->fsuid != inode->i_uid) && !capable(CAP_FOWNER)) + return -EACCES; + + if (get_user(rsv_window_size, (int __user *)arg)) + return -EFAULT; + + if (rsv_window_size > EXT4_MAX_RESERVE_BLOCKS) + rsv_window_size = EXT4_MAX_RESERVE_BLOCKS; + + /* + * need to allocate reservation structure for this inode + * before set the window size + */ + mutex_lock(&ei->truncate_mutex); + if (!ei->i_block_alloc_info) + ext4_init_block_alloc_info(inode); + + if (ei->i_block_alloc_info){ + struct ext4_reserve_window_node *rsv = &ei->i_block_alloc_info->rsv_window_node; + rsv->rsv_goal_size = rsv_window_size; + } + mutex_unlock(&ei->truncate_mutex); + return 0; + } + case EXT4_IOC_GROUP_EXTEND: { + ext4_fsblk_t n_blocks_count; + struct super_block *sb = inode->i_sb; + int err; + + if (!capable(CAP_SYS_RESOURCE)) + return -EPERM; + + if (IS_RDONLY(inode)) + return -EROFS; + + if (get_user(n_blocks_count, (__u32 __user *)arg)) + return -EFAULT; + + err = ext4_group_extend(sb, EXT4_SB(sb)->s_es, n_blocks_count); + jbd2_journal_lock_updates(EXT4_SB(sb)->s_journal); + jbd2_journal_flush(EXT4_SB(sb)->s_journal); + jbd2_journal_unlock_updates(EXT4_SB(sb)->s_journal); + + return err; + } + case EXT4_IOC_GROUP_ADD: { + struct ext4_new_group_data input; + struct super_block *sb = inode->i_sb; + int err; + + if (!capable(CAP_SYS_RESOURCE)) + return -EPERM; + + if (IS_RDONLY(inode)) + return -EROFS; + + if (copy_from_user(&input, (struct ext4_new_group_input __user *)arg, + sizeof(input))) + return -EFAULT; + + err = ext4_group_add(sb, &input); + jbd2_journal_lock_updates(EXT4_SB(sb)->s_journal); + jbd2_journal_flush(EXT4_SB(sb)->s_journal); + jbd2_journal_unlock_updates(EXT4_SB(sb)->s_journal); + + return err; + } + + default: + return -ENOTTY; + } +} + +#ifdef CONFIG_COMPAT +long ext4_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg) +{ + struct inode *inode = file->f_dentry->d_inode; + int ret; + + /* These are just misnamed, they actually get/put from/to user an int */ + switch (cmd) { + case EXT4_IOC32_GETFLAGS: + cmd = EXT4_IOC_GETFLAGS; + break; + case EXT4_IOC32_SETFLAGS: + cmd = EXT4_IOC_SETFLAGS; + break; + case EXT4_IOC32_GETVERSION: + cmd = EXT4_IOC_GETVERSION; + break; + case EXT4_IOC32_SETVERSION: + cmd = EXT4_IOC_SETVERSION; + break; + case EXT4_IOC32_GROUP_EXTEND: + cmd = EXT4_IOC_GROUP_EXTEND; + break; + case EXT4_IOC32_GETVERSION_OLD: + cmd = EXT4_IOC_GETVERSION_OLD; + break; + case EXT4_IOC32_SETVERSION_OLD: + cmd = EXT4_IOC_SETVERSION_OLD; + break; +#ifdef CONFIG_JBD_DEBUG + case EXT4_IOC32_WAIT_FOR_READONLY: + cmd = EXT4_IOC_WAIT_FOR_READONLY; + break; +#endif + case EXT4_IOC32_GETRSVSZ: + cmd = EXT4_IOC_GETRSVSZ; + break; + case EXT4_IOC32_SETRSVSZ: + cmd = EXT4_IOC_SETRSVSZ; + break; + case EXT4_IOC_GROUP_ADD: + break; + default: + return -ENOIOCTLCMD; + } + lock_kernel(); + ret = ext4_ioctl(inode, file, cmd, (unsigned long) compat_ptr(arg)); + unlock_kernel(); + return ret; +} +#endif diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c new file mode 100644 index 000000000000..8b1bd03d20f5 --- /dev/null +++ b/fs/ext4/namei.c @@ -0,0 +1,2395 @@ +/* + * linux/fs/ext4/namei.c + * + * Copyright (C) 1992, 1993, 1994, 1995 + * Remy Card (card@masi.ibp.fr) + * Laboratoire MASI - Institut Blaise Pascal + * Universite Pierre et Marie Curie (Paris VI) + * + * from + * + * linux/fs/minix/namei.c + * + * Copyright (C) 1991, 1992 Linus Torvalds + * + * Big-endian to little-endian byte-swapping/bitmaps by + * David S. Miller (davem@caip.rutgers.edu), 1995 + * Directory entry file type support and forward compatibility hooks + * for B-tree directories by Theodore Ts'o (tytso@mit.edu), 1998 + * Hash Tree Directory indexing (c) + * Daniel Phillips, 2001 + * Hash Tree Directory indexing porting + * Christopher Li, 2002 + * Hash Tree Directory indexing cleanup + * Theodore Ts'o, 2002 + */ + +#include <linux/fs.h> +#include <linux/pagemap.h> +#include <linux/jbd2.h> +#include <linux/time.h> +#include <linux/ext4_fs.h> +#include <linux/ext4_jbd2.h> +#include <linux/fcntl.h> +#include <linux/stat.h> +#include <linux/string.h> +#include <linux/quotaops.h> +#include <linux/buffer_head.h> +#include <linux/bio.h> +#include <linux/smp_lock.h> + +#include "namei.h" +#include "xattr.h" +#include "acl.h" + +/* + * define how far ahead to read directories while searching them. + */ +#define NAMEI_RA_CHUNKS 2 +#define NAMEI_RA_BLOCKS 4 +#define NAMEI_RA_SIZE (NAMEI_RA_CHUNKS * NAMEI_RA_BLOCKS) +#define NAMEI_RA_INDEX(c,b) (((c) * NAMEI_RA_BLOCKS) + (b)) + +static struct buffer_head *ext4_append(handle_t *handle, + struct inode *inode, + u32 *block, int *err) +{ + struct buffer_head *bh; + + *block = inode->i_size >> inode->i_sb->s_blocksize_bits; + + if ((bh = ext4_bread(handle, inode, *block, 1, err))) { + inode->i_size += inode->i_sb->s_blocksize; + EXT4_I(inode)->i_disksize = inode->i_size; + ext4_journal_get_write_access(handle,bh); + } + return bh; +} + +#ifndef assert +#define assert(test) J_ASSERT(test) +#endif + +#ifndef swap +#define swap(x, y) do { typeof(x) z = x; x = y; y = z; } while (0) +#endif + +#ifdef DX_DEBUG +#define dxtrace(command) command +#else +#define dxtrace(command) +#endif + +struct fake_dirent +{ + __le32 inode; + __le16 rec_len; + u8 name_len; + u8 file_type; +}; + +struct dx_countlimit +{ + __le16 limit; + __le16 count; +}; + +struct dx_entry +{ + __le32 hash; + __le32 block; +}; + +/* + * dx_root_info is laid out so that if it should somehow get overlaid by a + * dirent the two low bits of the hash version will be zero. Therefore, the + * hash version mod 4 should never be 0. Sincerely, the paranoia department. + */ + +struct dx_root +{ + struct fake_dirent dot; + char dot_name[4]; + struct fake_dirent dotdot; + char dotdot_name[4]; + struct dx_root_info + { + __le32 reserved_zero; + u8 hash_version; + u8 info_length; /* 8 */ + u8 indirect_levels; + u8 unused_flags; + } + info; + struct dx_entry entries[0]; +}; + +struct dx_node +{ + struct fake_dirent fake; + struct dx_entry entries[0]; +}; + + +struct dx_frame +{ + struct buffer_head *bh; + struct dx_entry *entries; + struct dx_entry *at; +}; + +struct dx_map_entry +{ + u32 hash; + u32 offs; +}; + +#ifdef CONFIG_EXT4_INDEX +static inline unsigned dx_get_block (struct dx_entry *entry); +static void dx_set_block (struct dx_entry *entry, unsigned value); +static inline unsigned dx_get_hash (struct dx_entry *entry); +static void dx_set_hash (struct dx_entry *entry, unsigned value); +static unsigned dx_get_count (struct dx_entry *entries); +static unsigned dx_get_limit (struct dx_entry *entries); +static void dx_set_count (struct dx_entry *entries, unsigned value); +static void dx_set_limit (struct dx_entry *entries, unsigned value); +static unsigned dx_root_limit (struct inode *dir, unsigned infosize); +static unsigned dx_node_limit (struct inode *dir); +static struct dx_frame *dx_probe(struct dentry *dentry, + struct inode *dir, + struct dx_hash_info *hinfo, + struct dx_frame *frame, + int *err); +static void dx_release (struct dx_frame *frames); +static int dx_make_map (struct ext4_dir_entry_2 *de, int size, + struct dx_hash_info *hinfo, struct dx_map_entry map[]); +static void dx_sort_map(struct dx_map_entry *map, unsigned count); +static struct ext4_dir_entry_2 *dx_move_dirents (char *from, char *to, + struct dx_map_entry *offsets, int count); +static struct ext4_dir_entry_2* dx_pack_dirents (char *base, int size); +static void dx_insert_block (struct dx_frame *frame, u32 hash, u32 block); +static int ext4_htree_next_block(struct inode *dir, __u32 hash, + struct dx_frame *frame, + struct dx_frame *frames, + __u32 *start_hash); +static struct buffer_head * ext4_dx_find_entry(struct dentry *dentry, + struct ext4_dir_entry_2 **res_dir, int *err); +static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry, + struct inode *inode); + +/* + * Future: use high four bits of block for coalesce-on-delete flags + * Mask them off for now. + */ + +static inline unsigned dx_get_block (struct dx_entry *entry) +{ + return le32_to_cpu(entry->block) & 0x00ffffff; +} + +static inline void dx_set_block (struct dx_entry *entry, unsigned value) +{ + entry->block = cpu_to_le32(value); +} + +static inline unsigned dx_get_hash (struct dx_entry *entry) +{ + return le32_to_cpu(entry->hash); +} + +static inline void dx_set_hash (struct dx_entry *entry, unsigned value) +{ + entry->hash = cpu_to_le32(value); +} + +static inline unsigned dx_get_count (struct dx_entry *entries) +{ + return le16_to_cpu(((struct dx_countlimit *) entries)->count); +} + +static inline unsigned dx_get_limit (struct dx_entry *entries) +{ + return le16_to_cpu(((struct dx_countlimit *) entries)->limit); +} + +static inline void dx_set_count (struct dx_entry *entries, unsigned value) +{ + ((struct dx_countlimit *) entries)->count = cpu_to_le16(value); +} + +static inline void dx_set_limit (struct dx_entry *entries, unsigned value) +{ + ((struct dx_countlimit *) entries)->limit = cpu_to_le16(value); +} + +static inline unsigned dx_root_limit (struct inode *dir, unsigned infosize) +{ + unsigned entry_space = dir->i_sb->s_blocksize - EXT4_DIR_REC_LEN(1) - + EXT4_DIR_REC_LEN(2) - infosize; + return 0? 20: entry_space / sizeof(struct dx_entry); +} + +static inline unsigned dx_node_limit (struct inode *dir) +{ + unsigned entry_space = dir->i_sb->s_blocksize - EXT4_DIR_REC_LEN(0); + return 0? 22: entry_space / sizeof(struct dx_entry); +} + +/* + * Debug + */ +#ifdef DX_DEBUG +static void dx_show_index (char * label, struct dx_entry *entries) +{ + int i, n = dx_get_count (entries); + printk("%s index ", label); + for (i = 0; i < n; i++) { + printk("%x->%u ", i? dx_get_hash(entries + i) : + 0, dx_get_block(entries + i)); + } + printk("\n"); +} + +struct stats +{ + unsigned names; + unsigned space; + unsigned bcount; +}; + +static struct stats dx_show_leaf(struct dx_hash_info *hinfo, struct ext4_dir_entry_2 *de, + int size, int show_names) +{ + unsigned names = 0, space = 0; + char *base = (char *) de; + struct dx_hash_info h = *hinfo; + + printk("names: "); + while ((char *) de < base + size) + { + if (de->inode) + { + if (show_names) + { + int len = de->name_len; + char *name = de->name; + while (len--) printk("%c", *name++); + ext4fs_dirhash(de->name, de->name_len, &h); + printk(":%x.%u ", h.hash, + ((char *) de - base)); + } + space += EXT4_DIR_REC_LEN(de->name_len); + names++; + } + de = (struct ext4_dir_entry_2 *) ((char *) de + le16_to_cpu(de->rec_len)); + } + printk("(%i)\n", names); + return (struct stats) { names, space, 1 }; +} + +struct stats dx_show_entries(struct dx_hash_info *hinfo, struct inode *dir, + struct dx_entry *entries, int levels) +{ + unsigned blocksize = dir->i_sb->s_blocksize; + unsigned count = dx_get_count (entries), names = 0, space = 0, i; + unsigned bcount = 0; + struct buffer_head *bh; + int err; + printk("%i indexed blocks...\n", count); + for (i = 0; i < count; i++, entries++) + { + u32 block = dx_get_block(entries), hash = i? dx_get_hash(entries): 0; + u32 range = i < count - 1? (dx_get_hash(entries + 1) - hash): ~hash; + struct stats stats; + printk("%s%3u:%03u hash %8x/%8x ",levels?"":" ", i, block, hash, range); + if (!(bh = ext4_bread (NULL,dir, block, 0,&err))) continue; + stats = levels? + dx_show_entries(hinfo, dir, ((struct dx_node *) bh->b_data)->entries, levels - 1): + dx_show_leaf(hinfo, (struct ext4_dir_entry_2 *) bh->b_data, blocksize, 0); + names += stats.names; + space += stats.space; + bcount += stats.bcount; + brelse (bh); + } + if (bcount) + printk("%snames %u, fullness %u (%u%%)\n", levels?"":" ", + names, space/bcount,(space/bcount)*100/blocksize); + return (struct stats) { names, space, bcount}; +} +#endif /* DX_DEBUG */ + +/* + * Probe for a directory leaf block to search. + * + * dx_probe can return ERR_BAD_DX_DIR, which means there was a format + * error in the directory index, and the caller should fall back to + * searching the directory normally. The callers of dx_probe **MUST** + * check for this error code, and make sure it never gets reflected + * back to userspace. + */ +static struct dx_frame * +dx_probe(struct dentry *dentry, struct inode *dir, + struct dx_hash_info *hinfo, struct dx_frame *frame_in, int *err) +{ + unsigned count, indirect; + struct dx_entry *at, *entries, *p, *q, *m; + struct dx_root *root; + struct buffer_head *bh; + struct dx_frame *frame = frame_in; + u32 hash; + + frame->bh = NULL; + if (dentry) + dir = dentry->d_parent->d_inode; + if (!(bh = ext4_bread (NULL,dir, 0, 0, err))) + goto fail; + root = (struct dx_root *) bh->b_data; + if (root->info.hash_version != DX_HASH_TEA && + root->info.hash_version != DX_HASH_HALF_MD4 && + root->info.hash_version != DX_HASH_LEGACY) { + ext4_warning(dir->i_sb, __FUNCTION__, + "Unrecognised inode hash code %d", + root->info.hash_version); + brelse(bh); + *err = ERR_BAD_DX_DIR; + goto fail; + } + hinfo->hash_version = root->info.hash_version; + hinfo->seed = EXT4_SB(dir->i_sb)->s_hash_seed; + if (dentry) + ext4fs_dirhash(dentry->d_name.name, dentry->d_name.len, hinfo); + hash = hinfo->hash; + + if (root->info.unused_flags & 1) { + ext4_warning(dir->i_sb, __FUNCTION__, + "Unimplemented inode hash flags: %#06x", + root->info.unused_flags); + brelse(bh); + *err = ERR_BAD_DX_DIR; + goto fail; + } + + if ((indirect = root->info.indirect_levels) > 1) { + ext4_warning(dir->i_sb, __FUNCTION__, + "Unimplemented inode hash depth: %#06x", + root->info.indirect_levels); + brelse(bh); + *err = ERR_BAD_DX_DIR; + goto fail; + } + + entries = (struct dx_entry *) (((char *)&root->info) + + root->info.info_length); + assert(dx_get_limit(entries) == dx_root_limit(dir, + root->info.info_length)); + dxtrace (printk("Look up %x", hash)); + while (1) + { + count = dx_get_count(entries); + assert (count && count <= dx_get_limit(entries)); + p = entries + 1; + q = entries + count - 1; + while (p <= q) + { + m = p + (q - p)/2; + dxtrace(printk(".")); + if (dx_get_hash(m) > hash) + q = m - 1; + else + p = m + 1; + } + + if (0) // linear search cross check + { + unsigned n = count - 1; + at = entries; + while (n--) + { + dxtrace(printk(",")); + if (dx_get_hash(++at) > hash) + { + at--; + break; + } + } + assert (at == p - 1); + } + + at = p - 1; + dxtrace(printk(" %x->%u\n", at == entries? 0: dx_get_hash(at), dx_get_block(at))); + frame->bh = bh; + frame->entries = entries; + frame->at = at; + if (!indirect--) return frame; + if (!(bh = ext4_bread (NULL,dir, dx_get_block(at), 0, err))) + goto fail2; + at = entries = ((struct dx_node *) bh->b_data)->entries; + assert (dx_get_limit(entries) == dx_node_limit (dir)); + frame++; + } +fail2: + while (frame >= frame_in) { + brelse(frame->bh); + frame--; + } +fail: + return NULL; +} + +static void dx_release (struct dx_frame *frames) +{ + if (frames[0].bh == NULL) + return; + + if (((struct dx_root *) frames[0].bh->b_data)->info.indirect_levels) + brelse(frames[1].bh); + brelse(frames[0].bh); +} + +/* + * This function increments the frame pointer to search the next leaf + * block, and reads in the necessary intervening nodes if the search + * should be necessary. Whether or not the search is necessary is + * controlled by the hash parameter. If the hash value is even, then + * the search is only continued if the next block starts with that + * hash value. This is used if we are searching for a specific file. + * + * If the hash value is HASH_NB_ALWAYS, then always go to the next block. + * + * This function returns 1 if the caller should continue to search, + * or 0 if it should not. If there is an error reading one of the + * index blocks, it will a negative error code. + * + * If start_hash is non-null, it will be filled in with the starting + * hash of the next page. + */ +static int ext4_htree_next_block(struct inode *dir, __u32 hash, + struct dx_frame *frame, + struct dx_frame *frames, + __u32 *start_hash) +{ + struct dx_frame *p; + struct buffer_head *bh; + int err, num_frames = 0; + __u32 bhash; + + p = frame; + /* + * Find the next leaf page by incrementing the frame pointer. + * If we run out of entries in the interior node, loop around and + * increment pointer in the parent node. When we break out of + * this loop, num_frames indicates the number of interior + * nodes need to be read. + */ + while (1) { + if (++(p->at) < p->entries + dx_get_count(p->entries)) + break; + if (p == frames) + return 0; + num_frames++; + p--; + } + + /* + * If the hash is 1, then continue only if the next page has a + * continuation hash of any value. This is used for readdir + * handling. Otherwise, check to see if the hash matches the + * desired contiuation hash. If it doesn't, return since + * there's no point to read in the successive index pages. + */ + bhash = dx_get_hash(p->at); + if (start_hash) + *start_hash = bhash; + if ((hash & 1) == 0) { + if ((bhash & ~1) != hash) + return 0; + } + /* + * If the hash is HASH_NB_ALWAYS, we always go to the next + * block so no check is necessary + */ + while (num_frames--) { + if (!(bh = ext4_bread(NULL, dir, dx_get_block(p->at), + 0, &err))) + return err; /* Failure */ + p++; + brelse (p->bh); + p->bh = bh; + p->at = p->entries = ((struct dx_node *) bh->b_data)->entries; + } + return 1; +} + + +/* + * p is at least 6 bytes before the end of page + */ +static inline struct ext4_dir_entry_2 *ext4_next_entry(struct ext4_dir_entry_2 *p) +{ + return (struct ext4_dir_entry_2 *)((char*)p + le16_to_cpu(p->rec_len)); +} + +/* + * This function fills a red-black tree with information from a + * directory block. It returns the number directory entries loaded + * into the tree. If there is an error it is returned in err. + */ +static int htree_dirblock_to_tree(struct file *dir_file, + struct inode *dir, int block, + struct dx_hash_info *hinfo, + __u32 start_hash, __u32 start_minor_hash) +{ + struct buffer_head *bh; + struct ext4_dir_entry_2 *de, *top; + int err, count = 0; + + dxtrace(printk("In htree dirblock_to_tree: block %d\n", block)); + if (!(bh = ext4_bread (NULL, dir, block, 0, &err))) + return err; + + de = (struct ext4_dir_entry_2 *) bh->b_data; + top = (struct ext4_dir_entry_2 *) ((char *) de + + dir->i_sb->s_blocksize - + EXT4_DIR_REC_LEN(0)); + for (; de < top; de = ext4_next_entry(de)) { + ext4fs_dirhash(de->name, de->name_len, hinfo); + if ((hinfo->hash < start_hash) || + ((hinfo->hash == start_hash) && + (hinfo->minor_hash < start_minor_hash))) + continue; + if (de->inode == 0) + continue; + if ((err = ext4_htree_store_dirent(dir_file, + hinfo->hash, hinfo->minor_hash, de)) != 0) { + brelse(bh); + return err; + } + count++; + } + brelse(bh); + return count; +} + + +/* + * This function fills a red-black tree with information from a + * directory. We start scanning the directory in hash order, starting + * at start_hash and start_minor_hash. + * + * This function returns the number of entries inserted into the tree, + * or a negative error code. + */ +int ext4_htree_fill_tree(struct file *dir_file, __u32 start_hash, + __u32 start_minor_hash, __u32 *next_hash) +{ + struct dx_hash_info hinfo; + struct ext4_dir_entry_2 *de; + struct dx_frame frames[2], *frame; + struct inode *dir; + int block, err; + int count = 0; + int ret; + __u32 hashval; + + dxtrace(printk("In htree_fill_tree, start hash: %x:%x\n", start_hash, + start_minor_hash)); + dir = dir_file->f_dentry->d_inode; + if (!(EXT4_I(dir)->i_flags & EXT4_INDEX_FL)) { + hinfo.hash_version = EXT4_SB(dir->i_sb)->s_def_hash_version; + hinfo.seed = EXT4_SB(dir->i_sb)->s_hash_seed; + count = htree_dirblock_to_tree(dir_file, dir, 0, &hinfo, + start_hash, start_minor_hash); + *next_hash = ~0; + return count; + } + hinfo.hash = start_hash; + hinfo.minor_hash = 0; + frame = dx_probe(NULL, dir_file->f_dentry->d_inode, &hinfo, frames, &err); + if (!frame) + return err; + + /* Add '.' and '..' from the htree header */ + if (!start_hash && !start_minor_hash) { + de = (struct ext4_dir_entry_2 *) frames[0].bh->b_data; + if ((err = ext4_htree_store_dirent(dir_file, 0, 0, de)) != 0) + goto errout; + count++; + } + if (start_hash < 2 || (start_hash ==2 && start_minor_hash==0)) { + de = (struct ext4_dir_entry_2 *) frames[0].bh->b_data; + de = ext4_next_entry(de); + if ((err = ext4_htree_store_dirent(dir_file, 2, 0, de)) != 0) + goto errout; + count++; + } + + while (1) { + block = dx_get_block(frame->at); + ret = htree_dirblock_to_tree(dir_file, dir, block, &hinfo, + start_hash, start_minor_hash); + if (ret < 0) { + err = ret; + goto errout; + } + count += ret; + hashval = ~0; + ret = ext4_htree_next_block(dir, HASH_NB_ALWAYS, + frame, frames, &hashval); + *next_hash = hashval; + if (ret < 0) { + err = ret; + goto errout; + } + /* + * Stop if: (a) there are no more entries, or + * (b) we have inserted at least one entry and the + * next hash value is not a continuation + */ + if ((ret == 0) || + (count && ((hashval & 1) == 0))) + break; + } + dx_release(frames); + dxtrace(printk("Fill tree: returned %d entries, next hash: %x\n", + count, *next_hash)); + return count; +errout: + dx_release(frames); + return (err); +} + + +/* + * Directory block splitting, compacting + */ + +static int dx_make_map (struct ext4_dir_entry_2 *de, int size, + struct dx_hash_info *hinfo, struct dx_map_entry *map_tail) +{ + int count = 0; + char *base = (char *) de; + struct dx_hash_info h = *hinfo; + + while ((char *) de < base + size) + { + if (de->name_len && de->inode) { + ext4fs_dirhash(de->name, de->name_len, &h); + map_tail--; + map_tail->hash = h.hash; + map_tail->offs = (u32) ((char *) de - base); + count++; + cond_resched(); + } + /* XXX: do we need to check rec_len == 0 case? -Chris */ + de = (struct ext4_dir_entry_2 *) ((char *) de + le16_to_cpu(de->rec_len)); + } + return count; +} + +static void dx_sort_map (struct dx_map_entry *map, unsigned count) +{ + struct dx_map_entry *p, *q, *top = map + count - 1; + int more; + /* Combsort until bubble sort doesn't suck */ + while (count > 2) { + count = count*10/13; + if (count - 9 < 2) /* 9, 10 -> 11 */ + count = 11; + for (p = top, q = p - count; q >= map; p--, q--) + if (p->hash < q->hash) + swap(*p, *q); + } + /* Garden variety bubble sort */ + do { + more = 0; + q = top; + while (q-- > map) { + if (q[1].hash >= q[0].hash) + continue; + swap(*(q+1), *q); + more = 1; + } + } while(more); +} + +static void dx_insert_block(struct dx_frame *frame, u32 hash, u32 block) +{ + struct dx_entry *entries = frame->entries; + struct dx_entry *old = frame->at, *new = old + 1; + int count = dx_get_count(entries); + + assert(count < dx_get_limit(entries)); + assert(old < entries + count); + memmove(new + 1, new, (char *)(entries + count) - (char *)(new)); + dx_set_hash(new, hash); + dx_set_block(new, block); + dx_set_count(entries, count + 1); +} +#endif + + +static void ext4_update_dx_flag(struct inode *inode) +{ + if (!EXT4_HAS_COMPAT_FEATURE(inode->i_sb, + EXT4_FEATURE_COMPAT_DIR_INDEX)) + EXT4_I(inode)->i_flags &= ~EXT4_INDEX_FL; +} + +/* + * NOTE! unlike strncmp, ext4_match returns 1 for success, 0 for failure. + * + * `len <= EXT4_NAME_LEN' is guaranteed by caller. + * `de != NULL' is guaranteed by caller. + */ +static inline int ext4_match (int len, const char * const name, + struct ext4_dir_entry_2 * de) +{ + if (len != de->name_len) + return 0; + if (!de->inode) + return 0; + return !memcmp(name, de->name, len); +} + +/* + * Returns 0 if not found, -1 on failure, and 1 on success + */ +static inline int search_dirblock(struct buffer_head * bh, + struct inode *dir, + struct dentry *dentry, + unsigned long offset, + struct ext4_dir_entry_2 ** res_dir) +{ + struct ext4_dir_entry_2 * de; + char * dlimit; + int de_len; + const char *name = dentry->d_name.name; + int namelen = dentry->d_name.len; + + de = (struct ext4_dir_entry_2 *) bh->b_data; + dlimit = bh->b_data + dir->i_sb->s_blocksize; + while ((char *) de < dlimit) { + /* this code is executed quadratically often */ + /* do minimal checking `by hand' */ + + if ((char *) de + namelen <= dlimit && + ext4_match (namelen, name, de)) { + /* found a match - just to be sure, do a full check */ + if (!ext4_check_dir_entry("ext4_find_entry", + dir, de, bh, offset)) + return -1; + *res_dir = de; + return 1; + } + /* prevent looping on a bad block */ + de_len = le16_to_cpu(de->rec_len); + if (de_len <= 0) + return -1; + offset += de_len; + de = (struct ext4_dir_entry_2 *) ((char *) de + de_len); + } + return 0; +} + + +/* + * ext4_find_entry() + * + * finds an entry in the specified directory with the wanted name. It + * returns the cache buffer in which the entry was found, and the entry + * itself (as a parameter - res_dir). It does NOT read the inode of the + * entry - you'll have to do that yourself if you want to. + * + * The returned buffer_head has ->b_count elevated. The caller is expected + * to brelse() it when appropriate. + */ +static struct buffer_head * ext4_find_entry (struct dentry *dentry, + struct ext4_dir_entry_2 ** res_dir) +{ + struct super_block * sb; + struct buffer_head * bh_use[NAMEI_RA_SIZE]; + struct buffer_head * bh, *ret = NULL; + unsigned long start, block, b; + int ra_max = 0; /* Number of bh's in the readahead + buffer, bh_use[] */ + int ra_ptr = 0; /* Current index into readahead + buffer */ + int num = 0; + int nblocks, i, err; + struct inode *dir = dentry->d_parent->d_inode; + int namelen; + const u8 *name; + unsigned blocksize; + + *res_dir = NULL; + sb = dir->i_sb; + blocksize = sb->s_blocksize; + namelen = dentry->d_name.len; + name = dentry->d_name.name; + if (namelen > EXT4_NAME_LEN) + return NULL; +#ifdef CONFIG_EXT4_INDEX + if (is_dx(dir)) { + bh = ext4_dx_find_entry(dentry, res_dir, &err); + /* + * On success, or if the error was file not found, + * return. Otherwise, fall back to doing a search the + * old fashioned way. + */ + if (bh || (err != ERR_BAD_DX_DIR)) + return bh; + dxtrace(printk("ext4_find_entry: dx failed, falling back\n")); + } +#endif + nblocks = dir->i_size >> EXT4_BLOCK_SIZE_BITS(sb); + start = EXT4_I(dir)->i_dir_start_lookup; + if (start >= nblocks) + start = 0; + block = start; +restart: + do { + /* + * We deal with the read-ahead logic here. + */ + if (ra_ptr >= ra_max) { + /* Refill the readahead buffer */ + ra_ptr = 0; + b = block; + for (ra_max = 0; ra_max < NAMEI_RA_SIZE; ra_max++) { + /* + * Terminate if we reach the end of the + * directory and must wrap, or if our + * search has finished at this block. + */ + if (b >= nblocks || (num && block == start)) { + bh_use[ra_max] = NULL; + break; + } + num++; + bh = ext4_getblk(NULL, dir, b++, 0, &err); + bh_use[ra_max] = bh; + if (bh) + ll_rw_block(READ_META, 1, &bh); + } + } + if ((bh = bh_use[ra_ptr++]) == NULL) + goto next; + wait_on_buffer(bh); + if (!buffer_uptodate(bh)) { + /* read error, skip block & hope for the best */ + ext4_error(sb, __FUNCTION__, "reading directory #%lu " + "offset %lu", dir->i_ino, block); + brelse(bh); + goto next; + } + i = search_dirblock(bh, dir, dentry, + block << EXT4_BLOCK_SIZE_BITS(sb), res_dir); + if (i == 1) { + EXT4_I(dir)->i_dir_start_lookup = block; + ret = bh; + goto cleanup_and_exit; + } else { + brelse(bh); + if (i < 0) + goto cleanup_and_exit; + } + next: + if (++block >= nblocks) + block = 0; + } while (block != start); + + /* + * If the directory has grown while we were searching, then + * search the last part of the directory before giving up. + */ + block = nblocks; + nblocks = dir->i_size >> EXT4_BLOCK_SIZE_BITS(sb); + if (block < nblocks) { + start = 0; + goto restart; + } + +cleanup_and_exit: + /* Clean up the read-ahead blocks */ + for (; ra_ptr < ra_max; ra_ptr++) + brelse (bh_use[ra_ptr]); + return ret; +} + +#ifdef CONFIG_EXT4_INDEX +static struct buffer_head * ext4_dx_find_entry(struct dentry *dentry, + struct ext4_dir_entry_2 **res_dir, int *err) +{ + struct super_block * sb; + struct dx_hash_info hinfo; + u32 hash; + struct dx_frame frames[2], *frame; + struct ext4_dir_entry_2 *de, *top; + struct buffer_head *bh; + unsigned long block; + int retval; + int namelen = dentry->d_name.len; + const u8 *name = dentry->d_name.name; + struct inode *dir = dentry->d_parent->d_inode; + + sb = dir->i_sb; + /* NFS may look up ".." - look at dx_root directory block */ + if (namelen > 2 || name[0] != '.'||(name[1] != '.' && name[1] != '\0')){ + if (!(frame = dx_probe(dentry, NULL, &hinfo, frames, err))) + return NULL; + } else { + frame = frames; + frame->bh = NULL; /* for dx_release() */ + frame->at = (struct dx_entry *)frames; /* hack for zero entry*/ + dx_set_block(frame->at, 0); /* dx_root block is 0 */ + } + hash = hinfo.hash; + do { + block = dx_get_block(frame->at); + if (!(bh = ext4_bread (NULL,dir, block, 0, err))) + goto errout; + de = (struct ext4_dir_entry_2 *) bh->b_data; + top = (struct ext4_dir_entry_2 *) ((char *) de + sb->s_blocksize - + EXT4_DIR_REC_LEN(0)); + for (; de < top; de = ext4_next_entry(de)) + if (ext4_match (namelen, name, de)) { + if (!ext4_check_dir_entry("ext4_find_entry", + dir, de, bh, + (block<<EXT4_BLOCK_SIZE_BITS(sb)) + +((char *)de - bh->b_data))) { + brelse (bh); + goto errout; + } + *res_dir = de; + dx_release (frames); + return bh; + } + brelse (bh); + /* Check to see if we should continue to search */ + retval = ext4_htree_next_block(dir, hash, frame, + frames, NULL); + if (retval < 0) { + ext4_warning(sb, __FUNCTION__, + "error reading index page in directory #%lu", + dir->i_ino); + *err = retval; + goto errout; + } + } while (retval == 1); + + *err = -ENOENT; +errout: + dxtrace(printk("%s not found\n", name)); + dx_release (frames); + return NULL; +} +#endif + +static struct dentry *ext4_lookup(struct inode * dir, struct dentry *dentry, struct nameidata *nd) +{ + struct inode * inode; + struct ext4_dir_entry_2 * de; + struct buffer_head * bh; + + if (dentry->d_name.len > EXT4_NAME_LEN) + return ERR_PTR(-ENAMETOOLONG); + + bh = ext4_find_entry(dentry, &de); + inode = NULL; + if (bh) { + unsigned long ino = le32_to_cpu(de->inode); + brelse (bh); + if (!ext4_valid_inum(dir->i_sb, ino)) { + ext4_error(dir->i_sb, "ext4_lookup", + "bad inode number: %lu", ino); + inode = NULL; + } else + inode = iget(dir->i_sb, ino); + + if (!inode) + return ERR_PTR(-EACCES); + } + return d_splice_alias(inode, dentry); +} + + +struct dentry *ext4_get_parent(struct dentry *child) +{ + unsigned long ino; + struct dentry *parent; + struct inode *inode; + struct dentry dotdot; + struct ext4_dir_entry_2 * de; + struct buffer_head *bh; + + dotdot.d_name.name = ".."; + dotdot.d_name.len = 2; + dotdot.d_parent = child; /* confusing, isn't it! */ + + bh = ext4_find_entry(&dotdot, &de); + inode = NULL; + if (!bh) + return ERR_PTR(-ENOENT); + ino = le32_to_cpu(de->inode); + brelse(bh); + + if (!ext4_valid_inum(child->d_inode->i_sb, ino)) { + ext4_error(child->d_inode->i_sb, "ext4_get_parent", + "bad inode number: %lu", ino); + inode = NULL; + } else + inode = iget(child->d_inode->i_sb, ino); + + if (!inode) + return ERR_PTR(-EACCES); + + parent = d_alloc_anon(inode); + if (!parent) { + iput(inode); + parent = ERR_PTR(-ENOMEM); + } + return parent; +} + +#define S_SHIFT 12 +static unsigned char ext4_type_by_mode[S_IFMT >> S_SHIFT] = { + [S_IFREG >> S_SHIFT] = EXT4_FT_REG_FILE, + [S_IFDIR >> S_SHIFT] = EXT4_FT_DIR, + [S_IFCHR >> S_SHIFT] = EXT4_FT_CHRDEV, + [S_IFBLK >> S_SHIFT] = EXT4_FT_BLKDEV, + [S_IFIFO >> S_SHIFT] = EXT4_FT_FIFO, + [S_IFSOCK >> S_SHIFT] = EXT4_FT_SOCK, + [S_IFLNK >> S_SHIFT] = EXT4_FT_SYMLINK, +}; + +static inline void ext4_set_de_type(struct super_block *sb, + struct ext4_dir_entry_2 *de, + umode_t mode) { + if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_FILETYPE)) + de->file_type = ext4_type_by_mode[(mode & S_IFMT)>>S_SHIFT]; +} + +#ifdef CONFIG_EXT4_INDEX +static struct ext4_dir_entry_2 * +dx_move_dirents(char *from, char *to, struct dx_map_entry *map, int count) +{ + unsigned rec_len = 0; + + while (count--) { + struct ext4_dir_entry_2 *de = (struct ext4_dir_entry_2 *) (from + map->offs); + rec_len = EXT4_DIR_REC_LEN(de->name_len); + memcpy (to, de, rec_len); + ((struct ext4_dir_entry_2 *) to)->rec_len = + cpu_to_le16(rec_len); + de->inode = 0; + map++; + to += rec_len; + } + return (struct ext4_dir_entry_2 *) (to - rec_len); +} + +static struct ext4_dir_entry_2* dx_pack_dirents(char *base, int size) +{ + struct ext4_dir_entry_2 *next, *to, *prev, *de = (struct ext4_dir_entry_2 *) base; + unsigned rec_len = 0; + + prev = to = de; + while ((char*)de < base + size) { + next = (struct ext4_dir_entry_2 *) ((char *) de + + le16_to_cpu(de->rec_len)); + if (de->inode && de->name_len) { + rec_len = EXT4_DIR_REC_LEN(de->name_len); + if (de > to) + memmove(to, de, rec_len); + to->rec_len = cpu_to_le16(rec_len); + prev = to; + to = (struct ext4_dir_entry_2 *) (((char *) to) + rec_len); + } + de = next; + } + return prev; +} + +static struct ext4_dir_entry_2 *do_split(handle_t *handle, struct inode *dir, + struct buffer_head **bh,struct dx_frame *frame, + struct dx_hash_info *hinfo, int *error) +{ + unsigned blocksize = dir->i_sb->s_blocksize; + unsigned count, continued; + struct buffer_head *bh2; + u32 newblock; + u32 hash2; + struct dx_map_entry *map; + char *data1 = (*bh)->b_data, *data2; + unsigned split; + struct ext4_dir_entry_2 *de = NULL, *de2; + int err; + + bh2 = ext4_append (handle, dir, &newblock, error); + if (!(bh2)) { + brelse(*bh); + *bh = NULL; + goto errout; + } + + BUFFER_TRACE(*bh, "get_write_access"); + err = ext4_journal_get_write_access(handle, *bh); + if (err) { + journal_error: + brelse(*bh); + brelse(bh2); + *bh = NULL; + ext4_std_error(dir->i_sb, err); + goto errout; + } + BUFFER_TRACE(frame->bh, "get_write_access"); + err = ext4_journal_get_write_access(handle, frame->bh); + if (err) + goto journal_error; + + data2 = bh2->b_data; + + /* create map in the end of data2 block */ + map = (struct dx_map_entry *) (data2 + blocksize); + count = dx_make_map ((struct ext4_dir_entry_2 *) data1, + blocksize, hinfo, map); + map -= count; + split = count/2; // need to adjust to actual middle + dx_sort_map (map, count); + hash2 = map[split].hash; + continued = hash2 == map[split - 1].hash; + dxtrace(printk("Split block %i at %x, %i/%i\n", + dx_get_block(frame->at), hash2, split, count-split)); + + /* Fancy dance to stay within two buffers */ + de2 = dx_move_dirents(data1, data2, map + split, count - split); + de = dx_pack_dirents(data1,blocksize); + de->rec_len = cpu_to_le16(data1 + blocksize - (char *) de); + de2->rec_len = cpu_to_le16(data2 + blocksize - (char *) de2); + dxtrace(dx_show_leaf (hinfo, (struct ext4_dir_entry_2 *) data1, blocksize, 1)); + dxtrace(dx_show_leaf (hinfo, (struct ext4_dir_entry_2 *) data2, blocksize, 1)); + + /* Which block gets the new entry? */ + if (hinfo->hash >= hash2) + { + swap(*bh, bh2); + de = de2; + } + dx_insert_block (frame, hash2 + continued, newblock); + err = ext4_journal_dirty_metadata (handle, bh2); + if (err) + goto journal_error; + err = ext4_journal_dirty_metadata (handle, frame->bh); + if (err) + goto journal_error; + brelse (bh2); + dxtrace(dx_show_index ("frame", frame->entries)); +errout: + return de; +} +#endif + + +/* + * Add a new entry into a directory (leaf) block. If de is non-NULL, + * it points to a directory entry which is guaranteed to be large + * enough for new directory entry. If de is NULL, then + * add_dirent_to_buf will attempt search the directory block for + * space. It will return -ENOSPC if no space is available, and -EIO + * and -EEXIST if directory entry already exists. + * + * NOTE! bh is NOT released in the case where ENOSPC is returned. In + * all other cases bh is released. + */ +static int add_dirent_to_buf(handle_t *handle, struct dentry *dentry, + struct inode *inode, struct ext4_dir_entry_2 *de, + struct buffer_head * bh) +{ + struct inode *dir = dentry->d_parent->d_inode; + const char *name = dentry->d_name.name; + int namelen = dentry->d_name.len; + unsigned long offset = 0; + unsigned short reclen; + int nlen, rlen, err; + char *top; + + reclen = EXT4_DIR_REC_LEN(namelen); + if (!de) { + de = (struct ext4_dir_entry_2 *)bh->b_data; + top = bh->b_data + dir->i_sb->s_blocksize - reclen; + while ((char *) de <= top) { + if (!ext4_check_dir_entry("ext4_add_entry", dir, de, + bh, offset)) { + brelse (bh); + return -EIO; + } + if (ext4_match (namelen, name, de)) { + brelse (bh); + return -EEXIST; + } + nlen = EXT4_DIR_REC_LEN(de->name_len); + rlen = le16_to_cpu(de->rec_len); + if ((de->inode? rlen - nlen: rlen) >= reclen) + break; + de = (struct ext4_dir_entry_2 *)((char *)de + rlen); + offset += rlen; + } + if ((char *) de > top) + return -ENOSPC; + } + BUFFER_TRACE(bh, "get_write_access"); + err = ext4_journal_get_write_access(handle, bh); + if (err) { + ext4_std_error(dir->i_sb, err); + brelse(bh); + return err; + } + + /* By now the buffer is marked for journaling */ + nlen = EXT4_DIR_REC_LEN(de->name_len); + rlen = le16_to_cpu(de->rec_len); + if (de->inode) { + struct ext4_dir_entry_2 *de1 = (struct ext4_dir_entry_2 *)((char *)de + nlen); + de1->rec_len = cpu_to_le16(rlen - nlen); + de->rec_len = cpu_to_le16(nlen); + de = de1; + } + de->file_type = EXT4_FT_UNKNOWN; + if (inode) { + de->inode = cpu_to_le32(inode->i_ino); + ext4_set_de_type(dir->i_sb, de, inode->i_mode); + } else + de->inode = 0; + de->name_len = namelen; + memcpy (de->name, name, namelen); + /* + * XXX shouldn't update any times until successful + * completion of syscall, but too many callers depend + * on this. + * + * XXX similarly, too many callers depend on + * ext4_new_inode() setting the times, but error + * recovery deletes the inode, so the worst that can + * happen is that the times are slightly out of date + * and/or different from the directory change time. + */ + dir->i_mtime = dir->i_ctime = CURRENT_TIME_SEC; + ext4_update_dx_flag(dir); + dir->i_version++; + ext4_mark_inode_dirty(handle, dir); + BUFFER_TRACE(bh, "call ext4_journal_dirty_metadata"); + err = ext4_journal_dirty_metadata(handle, bh); + if (err) + ext4_std_error(dir->i_sb, err); + brelse(bh); + return 0; +} + +#ifdef CONFIG_EXT4_INDEX +/* + * This converts a one block unindexed directory to a 3 block indexed + * directory, and adds the dentry to the indexed directory. + */ +static int make_indexed_dir(handle_t *handle, struct dentry *dentry, + struct inode *inode, struct buffer_head *bh) +{ + struct inode *dir = dentry->d_parent->d_inode; + const char *name = dentry->d_name.name; + int namelen = dentry->d_name.len; + struct buffer_head *bh2; + struct dx_root *root; + struct dx_frame frames[2], *frame; + struct dx_entry *entries; + struct ext4_dir_entry_2 *de, *de2; + char *data1, *top; + unsigned len; + int retval; + unsigned blocksize; + struct dx_hash_info hinfo; + u32 block; + struct fake_dirent *fde; + + blocksize = dir->i_sb->s_blocksize; + dxtrace(printk("Creating index\n")); + retval = ext4_journal_get_write_access(handle, bh); + if (retval) { + ext4_std_error(dir->i_sb, retval); + brelse(bh); + return retval; + } + root = (struct dx_root *) bh->b_data; + + bh2 = ext4_append (handle, dir, &block, &retval); + if (!(bh2)) { + brelse(bh); + return retval; + } + EXT4_I(dir)->i_flags |= EXT4_INDEX_FL; + data1 = bh2->b_data; + + /* The 0th block becomes the root, move the dirents out */ + fde = &root->dotdot; + de = (struct ext4_dir_entry_2 *)((char *)fde + le16_to_cpu(fde->rec_len)); + len = ((char *) root) + blocksize - (char *) de; + memcpy (data1, de, len); + de = (struct ext4_dir_entry_2 *) data1; + top = data1 + len; + while ((char *)(de2=(void*)de+le16_to_cpu(de->rec_len)) < top) + de = de2; + de->rec_len = cpu_to_le16(data1 + blocksize - (char *) de); + /* Initialize the root; the dot dirents already exist */ + de = (struct ext4_dir_entry_2 *) (&root->dotdot); + de->rec_len = cpu_to_le16(blocksize - EXT4_DIR_REC_LEN(2)); + memset (&root->info, 0, sizeof(root->info)); + root->info.info_length = sizeof(root->info); + root->info.hash_version = EXT4_SB(dir->i_sb)->s_def_hash_version; + entries = root->entries; + dx_set_block (entries, 1); + dx_set_count (entries, 1); + dx_set_limit (entries, dx_root_limit(dir, sizeof(root->info))); + + /* Initialize as for dx_probe */ + hinfo.hash_version = root->info.hash_version; + hinfo.seed = EXT4_SB(dir->i_sb)->s_hash_seed; + ext4fs_dirhash(name, namelen, &hinfo); + frame = frames; + frame->entries = entries; + frame->at = entries; + frame->bh = bh; + bh = bh2; + de = do_split(handle,dir, &bh, frame, &hinfo, &retval); + dx_release (frames); + if (!(de)) + return retval; + + return add_dirent_to_buf(handle, dentry, inode, de, bh); +} +#endif + +/* + * ext4_add_entry() + * + * adds a file entry to the specified directory, using the same + * semantics as ext4_find_entry(). It returns NULL if it failed. + * + * NOTE!! The inode part of 'de' is left at 0 - which means you + * may not sleep between calling this and putting something into + * the entry, as someone else might have used it while you slept. + */ +static int ext4_add_entry (handle_t *handle, struct dentry *dentry, + struct inode *inode) +{ + struct inode *dir = dentry->d_parent->d_inode; + unsigned long offset; + struct buffer_head * bh; + struct ext4_dir_entry_2 *de; + struct super_block * sb; + int retval; +#ifdef CONFIG_EXT4_INDEX + int dx_fallback=0; +#endif + unsigned blocksize; + u32 block, blocks; + + sb = dir->i_sb; + blocksize = sb->s_blocksize; + if (!dentry->d_name.len) + return -EINVAL; +#ifdef CONFIG_EXT4_INDEX + if (is_dx(dir)) { + retval = ext4_dx_add_entry(handle, dentry, inode); + if (!retval || (retval != ERR_BAD_DX_DIR)) + return retval; + EXT4_I(dir)->i_flags &= ~EXT4_INDEX_FL; + dx_fallback++; + ext4_mark_inode_dirty(handle, dir); + } +#endif + blocks = dir->i_size >> sb->s_blocksize_bits; + for (block = 0, offset = 0; block < blocks; block++) { + bh = ext4_bread(handle, dir, block, 0, &retval); + if(!bh) + return retval; + retval = add_dirent_to_buf(handle, dentry, inode, NULL, bh); + if (retval != -ENOSPC) + return retval; + +#ifdef CONFIG_EXT4_INDEX + if (blocks == 1 && !dx_fallback && + EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_DIR_INDEX)) + return make_indexed_dir(handle, dentry, inode, bh); +#endif + brelse(bh); + } + bh = ext4_append(handle, dir, &block, &retval); + if (!bh) + return retval; + de = (struct ext4_dir_entry_2 *) bh->b_data; + de->inode = 0; + de->rec_len = cpu_to_le16(blocksize); + return add_dirent_to_buf(handle, dentry, inode, de, bh); +} + +#ifdef CONFIG_EXT4_INDEX +/* + * Returns 0 for success, or a negative error value + */ +static int ext4_dx_add_entry(handle_t *handle, struct dentry *dentry, + struct inode *inode) +{ + struct dx_frame frames[2], *frame; + struct dx_entry *entries, *at; + struct dx_hash_info hinfo; + struct buffer_head * bh; + struct inode *dir = dentry->d_parent->d_inode; + struct super_block * sb = dir->i_sb; + struct ext4_dir_entry_2 *de; + int err; + + frame = dx_probe(dentry, NULL, &hinfo, frames, &err); + if (!frame) + return err; + entries = frame->entries; + at = frame->at; + + if (!(bh = ext4_bread(handle,dir, dx_get_block(frame->at), 0, &err))) + goto cleanup; + + BUFFER_TRACE(bh, "get_write_access"); + err = ext4_journal_get_write_access(handle, bh); + if (err) + goto journal_error; + + err = add_dirent_to_buf(handle, dentry, inode, NULL, bh); + if (err != -ENOSPC) { + bh = NULL; + goto cleanup; + } + + /* Block full, should compress but for now just split */ + dxtrace(printk("using %u of %u node entries\n", + dx_get_count(entries), dx_get_limit(entries))); + /* Need to split index? */ + if (dx_get_count(entries) == dx_get_limit(entries)) { + u32 newblock; + unsigned icount = dx_get_count(entries); + int levels = frame - frames; + struct dx_entry *entries2; + struct dx_node *node2; + struct buffer_head *bh2; + + if (levels && (dx_get_count(frames->entries) == + dx_get_limit(frames->entries))) { + ext4_warning(sb, __FUNCTION__, + "Directory index full!"); + err = -ENOSPC; + goto cleanup; + } + bh2 = ext4_append (handle, dir, &newblock, &err); + if (!(bh2)) + goto cleanup; + node2 = (struct dx_node *)(bh2->b_data); + entries2 = node2->entries; + node2->fake.rec_len = cpu_to_le16(sb->s_blocksize); + node2->fake.inode = 0; + BUFFER_TRACE(frame->bh, "get_write_access"); + err = ext4_journal_get_write_access(handle, frame->bh); + if (err) + goto journal_error; + if (levels) { + unsigned icount1 = icount/2, icount2 = icount - icount1; + unsigned hash2 = dx_get_hash(entries + icount1); + dxtrace(printk("Split index %i/%i\n", icount1, icount2)); + + BUFFER_TRACE(frame->bh, "get_write_access"); /* index root */ + err = ext4_journal_get_write_access(handle, + frames[0].bh); + if (err) + goto journal_error; + + memcpy ((char *) entries2, (char *) (entries + icount1), + icount2 * sizeof(struct dx_entry)); + dx_set_count (entries, icount1); + dx_set_count (entries2, icount2); + dx_set_limit (entries2, dx_node_limit(dir)); + + /* Which index block gets the new entry? */ + if (at - entries >= icount1) { + frame->at = at = at - entries - icount1 + entries2; + frame->entries = entries = entries2; + swap(frame->bh, bh2); + } + dx_insert_block (frames + 0, hash2, newblock); + dxtrace(dx_show_index ("node", frames[1].entries)); + dxtrace(dx_show_index ("node", + ((struct dx_node *) bh2->b_data)->entries)); + err = ext4_journal_dirty_metadata(handle, bh2); + if (err) + goto journal_error; + brelse (bh2); + } else { + dxtrace(printk("Creating second level index...\n")); + memcpy((char *) entries2, (char *) entries, + icount * sizeof(struct dx_entry)); + dx_set_limit(entries2, dx_node_limit(dir)); + + /* Set up root */ + dx_set_count(entries, 1); + dx_set_block(entries + 0, newblock); + ((struct dx_root *) frames[0].bh->b_data)->info.indirect_levels = 1; + + /* Add new access path frame */ + frame = frames + 1; + frame->at = at = at - entries + entries2; + frame->entries = entries = entries2; + frame->bh = bh2; + err = ext4_journal_get_write_access(handle, + frame->bh); + if (err) + goto journal_error; + } + ext4_journal_dirty_metadata(handle, frames[0].bh); + } + de = do_split(handle, dir, &bh, frame, &hinfo, &err); + if (!de) + goto cleanup; + err = add_dirent_to_buf(handle, dentry, inode, de, bh); + bh = NULL; + goto cleanup; + +journal_error: + ext4_std_error(dir->i_sb, err); +cleanup: + if (bh) + brelse(bh); + dx_release(frames); + return err; +} +#endif + +/* + * ext4_delete_entry deletes a directory entry by merging it with the + * previous entry + */ +static int ext4_delete_entry (handle_t *handle, + struct inode * dir, + struct ext4_dir_entry_2 * de_del, + struct buffer_head * bh) +{ + struct ext4_dir_entry_2 * de, * pde; + int i; + + i = 0; + pde = NULL; + de = (struct ext4_dir_entry_2 *) bh->b_data; + while (i < bh->b_size) { + if (!ext4_check_dir_entry("ext4_delete_entry", dir, de, bh, i)) + return -EIO; + if (de == de_del) { + BUFFER_TRACE(bh, "get_write_access"); + ext4_journal_get_write_access(handle, bh); + if (pde) + pde->rec_len = + cpu_to_le16(le16_to_cpu(pde->rec_len) + + le16_to_cpu(de->rec_len)); + else + de->inode = 0; + dir->i_version++; + BUFFER_TRACE(bh, "call ext4_journal_dirty_metadata"); + ext4_journal_dirty_metadata(handle, bh); + return 0; + } + i += le16_to_cpu(de->rec_len); + pde = de; + de = (struct ext4_dir_entry_2 *) + ((char *) de + le16_to_cpu(de->rec_len)); + } + return -ENOENT; +} + +/* + * ext4_mark_inode_dirty is somewhat expensive, so unlike ext2 we + * do not perform it in these functions. We perform it at the call site, + * if it is needed. + */ +static inline void ext4_inc_count(handle_t *handle, struct inode *inode) +{ + inc_nlink(inode); +} + +static inline void ext4_dec_count(handle_t *handle, struct inode *inode) +{ + drop_nlink(inode); +} + +static int ext4_add_nondir(handle_t *handle, + struct dentry *dentry, struct inode *inode) +{ + int err = ext4_add_entry(handle, dentry, inode); + if (!err) { + ext4_mark_inode_dirty(handle, inode); + d_instantiate(dentry, inode); + return 0; + } + ext4_dec_count(handle, inode); + iput(inode); + return err; +} + +/* + * By the time this is called, we already have created + * the directory cache entry for the new file, but it + * is so far negative - it has no inode. + * + * If the create succeeds, we fill in the inode information + * with d_instantiate(). + */ +static int ext4_create (struct inode * dir, struct dentry * dentry, int mode, + struct nameidata *nd) +{ + handle_t *handle; + struct inode * inode; + int err, retries = 0; + +retry: + handle = ext4_journal_start(dir, EXT4_DATA_TRANS_BLOCKS(dir->i_sb) + + EXT4_INDEX_EXTRA_TRANS_BLOCKS + 3 + + 2*EXT4_QUOTA_INIT_BLOCKS(dir->i_sb)); + if (IS_ERR(handle)) + return PTR_ERR(handle); + + if (IS_DIRSYNC(dir)) + handle->h_sync = 1; + + inode = ext4_new_inode (handle, dir, mode); + err = PTR_ERR(inode); + if (!IS_ERR(inode)) { + inode->i_op = &ext4_file_inode_operations; + inode->i_fop = &ext4_file_operations; + ext4_set_aops(inode); + err = ext4_add_nondir(handle, dentry, inode); + } + ext4_journal_stop(handle); + if (err == -ENOSPC && ext4_should_retry_alloc(dir->i_sb, &retries)) + goto retry; + return err; +} + +static int ext4_mknod (struct inode * dir, struct dentry *dentry, + int mode, dev_t rdev) +{ + handle_t *handle; + struct inode *inode; + int err, retries = 0; + + if (!new_valid_dev(rdev)) + return -EINVAL; + +retry: + handle = ext4_journal_start(dir, EXT4_DATA_TRANS_BLOCKS(dir->i_sb) + + EXT4_INDEX_EXTRA_TRANS_BLOCKS + 3 + + 2*EXT4_QUOTA_INIT_BLOCKS(dir->i_sb)); + if (IS_ERR(handle)) + return PTR_ERR(handle); + + if (IS_DIRSYNC(dir)) + handle->h_sync = 1; + + inode = ext4_new_inode (handle, dir, mode); + err = PTR_ERR(inode); + if (!IS_ERR(inode)) { + init_special_inode(inode, inode->i_mode, rdev); +#ifdef CONFIG_EXT4DEV_FS_XATTR + inode->i_op = &ext4_special_inode_operations; +#endif + err = ext4_add_nondir(handle, dentry, inode); + } + ext4_journal_stop(handle); + if (err == -ENOSPC && ext4_should_retry_alloc(dir->i_sb, &retries)) + goto retry; + return err; +} + +static int ext4_mkdir(struct inode * dir, struct dentry * dentry, int mode) +{ + handle_t *handle; + struct inode * inode; + struct buffer_head * dir_block; + struct ext4_dir_entry_2 * de; + int err, retries = 0; + + if (dir->i_nlink >= EXT4_LINK_MAX) + return -EMLINK; + +retry: + handle = ext4_journal_start(dir, EXT4_DATA_TRANS_BLOCKS(dir->i_sb) + + EXT4_INDEX_EXTRA_TRANS_BLOCKS + 3 + + 2*EXT4_QUOTA_INIT_BLOCKS(dir->i_sb)); + if (IS_ERR(handle)) + return PTR_ERR(handle); + + if (IS_DIRSYNC(dir)) + handle->h_sync = 1; + + inode = ext4_new_inode (handle, dir, S_IFDIR | mode); + err = PTR_ERR(inode); + if (IS_ERR(inode)) + goto out_stop; + + inode->i_op = &ext4_dir_inode_operations; + inode->i_fop = &ext4_dir_operations; + inode->i_size = EXT4_I(inode)->i_disksize = inode->i_sb->s_blocksize; + dir_block = ext4_bread (handle, inode, 0, 1, &err); + if (!dir_block) { + drop_nlink(inode); /* is this nlink == 0? */ + ext4_mark_inode_dirty(handle, inode); + iput (inode); + goto out_stop; + } + BUFFER_TRACE(dir_block, "get_write_access"); + ext4_journal_get_write_access(handle, dir_block); + de = (struct ext4_dir_entry_2 *) dir_block->b_data; + de->inode = cpu_to_le32(inode->i_ino); + de->name_len = 1; + de->rec_len = cpu_to_le16(EXT4_DIR_REC_LEN(de->name_len)); + strcpy (de->name, "."); + ext4_set_de_type(dir->i_sb, de, S_IFDIR); + de = (struct ext4_dir_entry_2 *) + ((char *) de + le16_to_cpu(de->rec_len)); + de->inode = cpu_to_le32(dir->i_ino); + de->rec_len = cpu_to_le16(inode->i_sb->s_blocksize-EXT4_DIR_REC_LEN(1)); + de->name_len = 2; + strcpy (de->name, ".."); + ext4_set_de_type(dir->i_sb, de, S_IFDIR); + inode->i_nlink = 2; + BUFFER_TRACE(dir_block, "call ext4_journal_dirty_metadata"); + ext4_journal_dirty_metadata(handle, dir_block); + brelse (dir_block); + ext4_mark_inode_dirty(handle, inode); + err = ext4_add_entry (handle, dentry, inode); + if (err) { + inode->i_nlink = 0; + ext4_mark_inode_dirty(handle, inode); + iput (inode); + goto out_stop; + } + inc_nlink(dir); + ext4_update_dx_flag(dir); + ext4_mark_inode_dirty(handle, dir); + d_instantiate(dentry, inode); +out_stop: + ext4_journal_stop(handle); + if (err == -ENOSPC && ext4_should_retry_alloc(dir->i_sb, &retries)) + goto retry; + return err; +} + +/* + * routine to check that the specified directory is empty (for rmdir) + */ +static int empty_dir (struct inode * inode) +{ + unsigned long offset; + struct buffer_head * bh; + struct ext4_dir_entry_2 * de, * de1; + struct super_block * sb; + int err = 0; + + sb = inode->i_sb; + if (inode->i_size < EXT4_DIR_REC_LEN(1) + EXT4_DIR_REC_LEN(2) || + !(bh = ext4_bread (NULL, inode, 0, 0, &err))) { + if (err) + ext4_error(inode->i_sb, __FUNCTION__, + "error %d reading directory #%lu offset 0", + err, inode->i_ino); + else + ext4_warning(inode->i_sb, __FUNCTION__, + "bad directory (dir #%lu) - no data block", + inode->i_ino); + return 1; + } + de = (struct ext4_dir_entry_2 *) bh->b_data; + de1 = (struct ext4_dir_entry_2 *) + ((char *) de + le16_to_cpu(de->rec_len)); + if (le32_to_cpu(de->inode) != inode->i_ino || + !le32_to_cpu(de1->inode) || + strcmp (".", de->name) || + strcmp ("..", de1->name)) { + ext4_warning (inode->i_sb, "empty_dir", + "bad directory (dir #%lu) - no `.' or `..'", + inode->i_ino); + brelse (bh); + return 1; + } + offset = le16_to_cpu(de->rec_len) + le16_to_cpu(de1->rec_len); + de = (struct ext4_dir_entry_2 *) + ((char *) de1 + le16_to_cpu(de1->rec_len)); + while (offset < inode->i_size ) { + if (!bh || + (void *) de >= (void *) (bh->b_data+sb->s_blocksize)) { + err = 0; + brelse (bh); + bh = ext4_bread (NULL, inode, + offset >> EXT4_BLOCK_SIZE_BITS(sb), 0, &err); + if (!bh) { + if (err) + ext4_error(sb, __FUNCTION__, + "error %d reading directory" + " #%lu offset %lu", + err, inode->i_ino, offset); + offset += sb->s_blocksize; + continue; + } + de = (struct ext4_dir_entry_2 *) bh->b_data; + } + if (!ext4_check_dir_entry("empty_dir", inode, de, bh, offset)) { + de = (struct ext4_dir_entry_2 *)(bh->b_data + + sb->s_blocksize); + offset = (offset | (sb->s_blocksize - 1)) + 1; + continue; + } + if (le32_to_cpu(de->inode)) { + brelse (bh); + return 0; + } + offset += le16_to_cpu(de->rec_len); + de = (struct ext4_dir_entry_2 *) + ((char *) de + le16_to_cpu(de->rec_len)); + } + brelse (bh); + return 1; +} + +/* ext4_orphan_add() links an unlinked or truncated inode into a list of + * such inodes, starting at the superblock, in case we crash before the + * file is closed/deleted, or in case the inode truncate spans multiple + * transactions and the last transaction is not recovered after a crash. + * + * At filesystem recovery time, we walk this list deleting unlinked + * inodes and truncating linked inodes in ext4_orphan_cleanup(). + */ +int ext4_orphan_add(handle_t *handle, struct inode *inode) +{ + struct super_block *sb = inode->i_sb; + struct ext4_iloc iloc; + int err = 0, rc; + + lock_super(sb); + if (!list_empty(&EXT4_I(inode)->i_orphan)) + goto out_unlock; + + /* Orphan handling is only valid for files with data blocks + * being truncated, or files being unlinked. */ + + /* @@@ FIXME: Observation from aviro: + * I think I can trigger J_ASSERT in ext4_orphan_add(). We block + * here (on lock_super()), so race with ext4_link() which might bump + * ->i_nlink. For, say it, character device. Not a regular file, + * not a directory, not a symlink and ->i_nlink > 0. + */ + J_ASSERT ((S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) || + S_ISLNK(inode->i_mode)) || inode->i_nlink == 0); + + BUFFER_TRACE(EXT4_SB(sb)->s_sbh, "get_write_access"); + err = ext4_journal_get_write_access(handle, EXT4_SB(sb)->s_sbh); + if (err) + goto out_unlock; + + err = ext4_reserve_inode_write(handle, inode, &iloc); + if (err) + goto out_unlock; + + /* Insert this inode at the head of the on-disk orphan list... */ + NEXT_ORPHAN(inode) = le32_to_cpu(EXT4_SB(sb)->s_es->s_last_orphan); + EXT4_SB(sb)->s_es->s_last_orphan = cpu_to_le32(inode->i_ino); + err = ext4_journal_dirty_metadata(handle, EXT4_SB(sb)->s_sbh); + rc = ext4_mark_iloc_dirty(handle, inode, &iloc); + if (!err) + err = rc; + + /* Only add to the head of the in-memory list if all the + * previous operations succeeded. If the orphan_add is going to + * fail (possibly taking the journal offline), we can't risk + * leaving the inode on the orphan list: stray orphan-list + * entries can cause panics at unmount time. + * + * This is safe: on error we're going to ignore the orphan list + * anyway on the next recovery. */ + if (!err) + list_add(&EXT4_I(inode)->i_orphan, &EXT4_SB(sb)->s_orphan); + + jbd_debug(4, "superblock will point to %lu\n", inode->i_ino); + jbd_debug(4, "orphan inode %lu will point to %d\n", + inode->i_ino, NEXT_ORPHAN(inode)); +out_unlock: + unlock_super(sb); + ext4_std_error(inode->i_sb, err); + return err; +} + +/* + * ext4_orphan_del() removes an unlinked or truncated inode from the list + * of such inodes stored on disk, because it is finally being cleaned up. + */ +int ext4_orphan_del(handle_t *handle, struct inode *inode) +{ + struct list_head *prev; + struct ext4_inode_info *ei = EXT4_I(inode); + struct ext4_sb_info *sbi; + unsigned long ino_next; + struct ext4_iloc iloc; + int err = 0; + + lock_super(inode->i_sb); + if (list_empty(&ei->i_orphan)) { + unlock_super(inode->i_sb); + return 0; + } + + ino_next = NEXT_ORPHAN(inode); + prev = ei->i_orphan.prev; + sbi = EXT4_SB(inode->i_sb); + + jbd_debug(4, "remove inode %lu from orphan list\n", inode->i_ino); + + list_del_init(&ei->i_orphan); + + /* If we're on an error path, we may not have a valid + * transaction handle with which to update the orphan list on + * disk, but we still need to remove the inode from the linked + * list in memory. */ + if (!handle) + goto out; + + err = ext4_reserve_inode_write(handle, inode, &iloc); + if (err) + goto out_err; + + if (prev == &sbi->s_orphan) { + jbd_debug(4, "superblock will point to %lu\n", ino_next); + BUFFER_TRACE(sbi->s_sbh, "get_write_access"); + err = ext4_journal_get_write_access(handle, sbi->s_sbh); + if (err) + goto out_brelse; + sbi->s_es->s_last_orphan = cpu_to_le32(ino_next); + err = ext4_journal_dirty_metadata(handle, sbi->s_sbh); + } else { + struct ext4_iloc iloc2; + struct inode *i_prev = + &list_entry(prev, struct ext4_inode_info, i_orphan)->vfs_inode; + + jbd_debug(4, "orphan inode %lu will point to %lu\n", + i_prev->i_ino, ino_next); + err = ext4_reserve_inode_write(handle, i_prev, &iloc2); + if (err) + goto out_brelse; + NEXT_ORPHAN(i_prev) = ino_next; + err = ext4_mark_iloc_dirty(handle, i_prev, &iloc2); + } + if (err) + goto out_brelse; + NEXT_ORPHAN(inode) = 0; + err = ext4_mark_iloc_dirty(handle, inode, &iloc); + +out_err: + ext4_std_error(inode->i_sb, err); +out: + unlock_super(inode->i_sb); + return err; + +out_brelse: + brelse(iloc.bh); + goto out_err; +} + +static int ext4_rmdir (struct inode * dir, struct dentry *dentry) +{ + int retval; + struct inode * inode; + struct buffer_head * bh; + struct ext4_dir_entry_2 * de; + handle_t *handle; + + /* Initialize quotas before so that eventual writes go in + * separate transaction */ + DQUOT_INIT(dentry->d_inode); + handle = ext4_journal_start(dir, EXT4_DELETE_TRANS_BLOCKS(dir->i_sb)); + if (IS_ERR(handle)) + return PTR_ERR(handle); + + retval = -ENOENT; + bh = ext4_find_entry (dentry, &de); + if (!bh) + goto end_rmdir; + + if (IS_DIRSYNC(dir)) + handle->h_sync = 1; + + inode = dentry->d_inode; + + retval = -EIO; + if (le32_to_cpu(de->inode) != inode->i_ino) + goto end_rmdir; + + retval = -ENOTEMPTY; + if (!empty_dir (inode)) + goto end_rmdir; + + retval = ext4_delete_entry(handle, dir, de, bh); + if (retval) + goto end_rmdir; + if (inode->i_nlink != 2) + ext4_warning (inode->i_sb, "ext4_rmdir", + "empty directory has nlink!=2 (%d)", + inode->i_nlink); + inode->i_version++; + clear_nlink(inode); + /* There's no need to set i_disksize: the fact that i_nlink is + * zero will ensure that the right thing happens during any + * recovery. */ + inode->i_size = 0; + ext4_orphan_add(handle, inode); + inode->i_ctime = dir->i_ctime = dir->i_mtime = CURRENT_TIME_SEC; + ext4_mark_inode_dirty(handle, inode); + drop_nlink(dir); + ext4_update_dx_flag(dir); + ext4_mark_inode_dirty(handle, dir); + +end_rmdir: + ext4_journal_stop(handle); + brelse (bh); + return retval; +} + +static int ext4_unlink(struct inode * dir, struct dentry *dentry) +{ + int retval; + struct inode * inode; + struct buffer_head * bh; + struct ext4_dir_entry_2 * de; + handle_t *handle; + + /* Initialize quotas before so that eventual writes go + * in separate transaction */ + DQUOT_INIT(dentry->d_inode); + handle = ext4_journal_start(dir, EXT4_DELETE_TRANS_BLOCKS(dir->i_sb)); + if (IS_ERR(handle)) + return PTR_ERR(handle); + + if (IS_DIRSYNC(dir)) + handle->h_sync = 1; + + retval = -ENOENT; + bh = ext4_find_entry (dentry, &de); + if (!bh) + goto end_unlink; + + inode = dentry->d_inode; + + retval = -EIO; + if (le32_to_cpu(de->inode) != inode->i_ino) + goto end_unlink; + + if (!inode->i_nlink) { + ext4_warning (inode->i_sb, "ext4_unlink", + "Deleting nonexistent file (%lu), %d", + inode->i_ino, inode->i_nlink); + inode->i_nlink = 1; + } + retval = ext4_delete_entry(handle, dir, de, bh); + if (retval) + goto end_unlink; + dir->i_ctime = dir->i_mtime = CURRENT_TIME_SEC; + ext4_update_dx_flag(dir); + ext4_mark_inode_dirty(handle, dir); + drop_nlink(inode); + if (!inode->i_nlink) + ext4_orphan_add(handle, inode); + inode->i_ctime = dir->i_ctime; + ext4_mark_inode_dirty(handle, inode); + retval = 0; + +end_unlink: + ext4_journal_stop(handle); + brelse (bh); + return retval; +} + +static int ext4_symlink (struct inode * dir, + struct dentry *dentry, const char * symname) +{ + handle_t *handle; + struct inode * inode; + int l, err, retries = 0; + + l = strlen(symname)+1; + if (l > dir->i_sb->s_blocksize) + return -ENAMETOOLONG; + +retry: + handle = ext4_journal_start(dir, EXT4_DATA_TRANS_BLOCKS(dir->i_sb) + + EXT4_INDEX_EXTRA_TRANS_BLOCKS + 5 + + 2*EXT4_QUOTA_INIT_BLOCKS(dir->i_sb)); + if (IS_ERR(handle)) + return PTR_ERR(handle); + + if (IS_DIRSYNC(dir)) + handle->h_sync = 1; + + inode = ext4_new_inode (handle, dir, S_IFLNK|S_IRWXUGO); + err = PTR_ERR(inode); + if (IS_ERR(inode)) + goto out_stop; + + if (l > sizeof (EXT4_I(inode)->i_data)) { + inode->i_op = &ext4_symlink_inode_operations; + ext4_set_aops(inode); + /* + * page_symlink() calls into ext4_prepare/commit_write. + * We have a transaction open. All is sweetness. It also sets + * i_size in generic_commit_write(). + */ + err = __page_symlink(inode, symname, l, + mapping_gfp_mask(inode->i_mapping) & ~__GFP_FS); + if (err) { + ext4_dec_count(handle, inode); + ext4_mark_inode_dirty(handle, inode); + iput (inode); + goto out_stop; + } + } else { + inode->i_op = &ext4_fast_symlink_inode_operations; + memcpy((char*)&EXT4_I(inode)->i_data,symname,l); + inode->i_size = l-1; + } + EXT4_I(inode)->i_disksize = inode->i_size; + err = ext4_add_nondir(handle, dentry, inode); +out_stop: + ext4_journal_stop(handle); + if (err == -ENOSPC && ext4_should_retry_alloc(dir->i_sb, &retries)) + goto retry; + return err; +} + +static int ext4_link (struct dentry * old_dentry, + struct inode * dir, struct dentry *dentry) +{ + handle_t *handle; + struct inode *inode = old_dentry->d_inode; + int err, retries = 0; + + if (inode->i_nlink >= EXT4_LINK_MAX) + return -EMLINK; + +retry: + handle = ext4_journal_start(dir, EXT4_DATA_TRANS_BLOCKS(dir->i_sb) + + EXT4_INDEX_EXTRA_TRANS_BLOCKS); + if (IS_ERR(handle)) + return PTR_ERR(handle); + + if (IS_DIRSYNC(dir)) + handle->h_sync = 1; + + inode->i_ctime = CURRENT_TIME_SEC; + ext4_inc_count(handle, inode); + atomic_inc(&inode->i_count); + + err = ext4_add_nondir(handle, dentry, inode); + ext4_journal_stop(handle); + if (err == -ENOSPC && ext4_should_retry_alloc(dir->i_sb, &retries)) + goto retry; + return err; +} + +#define PARENT_INO(buffer) \ + ((struct ext4_dir_entry_2 *) ((char *) buffer + \ + le16_to_cpu(((struct ext4_dir_entry_2 *) buffer)->rec_len)))->inode + +/* + * Anybody can rename anything with this: the permission checks are left to the + * higher-level routines. + */ +static int ext4_rename (struct inode * old_dir, struct dentry *old_dentry, + struct inode * new_dir,struct dentry *new_dentry) +{ + handle_t *handle; + struct inode * old_inode, * new_inode; + struct buffer_head * old_bh, * new_bh, * dir_bh; + struct ext4_dir_entry_2 * old_de, * new_de; + int retval; + + old_bh = new_bh = dir_bh = NULL; + + /* Initialize quotas before so that eventual writes go + * in separate transaction */ + if (new_dentry->d_inode) + DQUOT_INIT(new_dentry->d_inode); + handle = ext4_journal_start(old_dir, 2 * + EXT4_DATA_TRANS_BLOCKS(old_dir->i_sb) + + EXT4_INDEX_EXTRA_TRANS_BLOCKS + 2); + if (IS_ERR(handle)) + return PTR_ERR(handle); + + if (IS_DIRSYNC(old_dir) || IS_DIRSYNC(new_dir)) + handle->h_sync = 1; + + old_bh = ext4_find_entry (old_dentry, &old_de); + /* + * Check for inode number is _not_ due to possible IO errors. + * We might rmdir the source, keep it as pwd of some process + * and merrily kill the link to whatever was created under the + * same name. Goodbye sticky bit ;-< + */ + old_inode = old_dentry->d_inode; + retval = -ENOENT; + if (!old_bh || le32_to_cpu(old_de->inode) != old_inode->i_ino) + goto end_rename; + + new_inode = new_dentry->d_inode; + new_bh = ext4_find_entry (new_dentry, &new_de); + if (new_bh) { + if (!new_inode) { + brelse (new_bh); + new_bh = NULL; + } + } + if (S_ISDIR(old_inode->i_mode)) { + if (new_inode) { + retval = -ENOTEMPTY; + if (!empty_dir (new_inode)) + goto end_rename; + } + retval = -EIO; + dir_bh = ext4_bread (handle, old_inode, 0, 0, &retval); + if (!dir_bh) + goto end_rename; + if (le32_to_cpu(PARENT_INO(dir_bh->b_data)) != old_dir->i_ino) + goto end_rename; + retval = -EMLINK; + if (!new_inode && new_dir!=old_dir && + new_dir->i_nlink >= EXT4_LINK_MAX) + goto end_rename; + } + if (!new_bh) { + retval = ext4_add_entry (handle, new_dentry, old_inode); + if (retval) + goto end_rename; + } else { + BUFFER_TRACE(new_bh, "get write access"); + ext4_journal_get_write_access(handle, new_bh); + new_de->inode = cpu_to_le32(old_inode->i_ino); + if (EXT4_HAS_INCOMPAT_FEATURE(new_dir->i_sb, + EXT4_FEATURE_INCOMPAT_FILETYPE)) + new_de->file_type = old_de->file_type; + new_dir->i_version++; + BUFFER_TRACE(new_bh, "call ext4_journal_dirty_metadata"); + ext4_journal_dirty_metadata(handle, new_bh); + brelse(new_bh); + new_bh = NULL; + } + + /* + * Like most other Unix systems, set the ctime for inodes on a + * rename. + */ + old_inode->i_ctime = CURRENT_TIME_SEC; + ext4_mark_inode_dirty(handle, old_inode); + + /* + * ok, that's it + */ + if (le32_to_cpu(old_de->inode) != old_inode->i_ino || + old_de->name_len != old_dentry->d_name.len || + strncmp(old_de->name, old_dentry->d_name.name, old_de->name_len) || + (retval = ext4_delete_entry(handle, old_dir, + old_de, old_bh)) == -ENOENT) { + /* old_de could have moved from under us during htree split, so + * make sure that we are deleting the right entry. We might + * also be pointing to a stale entry in the unused part of + * old_bh so just checking inum and the name isn't enough. */ + struct buffer_head *old_bh2; + struct ext4_dir_entry_2 *old_de2; + + old_bh2 = ext4_find_entry(old_dentry, &old_de2); + if (old_bh2) { + retval = ext4_delete_entry(handle, old_dir, + old_de2, old_bh2); + brelse(old_bh2); + } + } + if (retval) { + ext4_warning(old_dir->i_sb, "ext4_rename", + "Deleting old file (%lu), %d, error=%d", + old_dir->i_ino, old_dir->i_nlink, retval); + } + + if (new_inode) { + drop_nlink(new_inode); + new_inode->i_ctime = CURRENT_TIME_SEC; + } + old_dir->i_ctime = old_dir->i_mtime = CURRENT_TIME_SEC; + ext4_update_dx_flag(old_dir); + if (dir_bh) { + BUFFER_TRACE(dir_bh, "get_write_access"); + ext4_journal_get_write_access(handle, dir_bh); + PARENT_INO(dir_bh->b_data) = cpu_to_le32(new_dir->i_ino); + BUFFER_TRACE(dir_bh, "call ext4_journal_dirty_metadata"); + ext4_journal_dirty_metadata(handle, dir_bh); + drop_nlink(old_dir); + if (new_inode) { + drop_nlink(new_inode); + } else { + inc_nlink(new_dir); + ext4_update_dx_flag(new_dir); + ext4_mark_inode_dirty(handle, new_dir); + } + } + ext4_mark_inode_dirty(handle, old_dir); + if (new_inode) { + ext4_mark_inode_dirty(handle, new_inode); + if (!new_inode->i_nlink) + ext4_orphan_add(handle, new_inode); + } + retval = 0; + +end_rename: + brelse (dir_bh); + brelse (old_bh); + brelse (new_bh); + ext4_journal_stop(handle); + return retval; +} + +/* + * directories can handle most operations... + */ +struct inode_operations ext4_dir_inode_operations = { + .create = ext4_create, + .lookup = ext4_lookup, + .link = ext4_link, + .unlink = ext4_unlink, + .symlink = ext4_symlink, + .mkdir = ext4_mkdir, + .rmdir = ext4_rmdir, + .mknod = ext4_mknod, + .rename = ext4_rename, + .setattr = ext4_setattr, +#ifdef CONFIG_EXT4DEV_FS_XATTR + .setxattr = generic_setxattr, + .getxattr = generic_getxattr, + .listxattr = ext4_listxattr, + .removexattr = generic_removexattr, +#endif + .permission = ext4_permission, +}; + +struct inode_operations ext4_special_inode_operations = { + .setattr = ext4_setattr, +#ifdef CONFIG_EXT4DEV_FS_XATTR + .setxattr = generic_setxattr, + .getxattr = generic_getxattr, + .listxattr = ext4_listxattr, + .removexattr = generic_removexattr, +#endif + .permission = ext4_permission, +}; diff --git a/fs/ext4/namei.h b/fs/ext4/namei.h new file mode 100644 index 000000000000..5e4dfff36a00 --- /dev/null +++ b/fs/ext4/namei.h @@ -0,0 +1,8 @@ +/* linux/fs/ext4/namei.h + * + * Copyright (C) 2005 Simtec Electronics + * Ben Dooks <ben@simtec.co.uk> + * +*/ + +extern struct dentry *ext4_get_parent(struct dentry *child); diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c new file mode 100644 index 000000000000..4fe49c3661b2 --- /dev/null +++ b/fs/ext4/resize.c @@ -0,0 +1,1050 @@ +/* + * linux/fs/ext4/resize.c + * + * Support for resizing an ext4 filesystem while it is mounted. + * + * Copyright (C) 2001, 2002 Andreas Dilger <adilger@clusterfs.com> + * + * This could probably be made into a module, because it is not often in use. + */ + + +#define EXT4FS_DEBUG + +#include <linux/sched.h> +#include <linux/smp_lock.h> +#include <linux/ext4_jbd2.h> + +#include <linux/errno.h> +#include <linux/slab.h> + + +#define outside(b, first, last) ((b) < (first) || (b) >= (last)) +#define inside(b, first, last) ((b) >= (first) && (b) < (last)) + +static int verify_group_input(struct super_block *sb, + struct ext4_new_group_data *input) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct ext4_super_block *es = sbi->s_es; + ext4_fsblk_t start = ext4_blocks_count(es); + ext4_fsblk_t end = start + input->blocks_count; + unsigned group = input->group; + ext4_fsblk_t itend = input->inode_table + sbi->s_itb_per_group; + unsigned overhead = ext4_bg_has_super(sb, group) ? + (1 + ext4_bg_num_gdb(sb, group) + + le16_to_cpu(es->s_reserved_gdt_blocks)) : 0; + ext4_fsblk_t metaend = start + overhead; + struct buffer_head *bh = NULL; + ext4_grpblk_t free_blocks_count, offset; + int err = -EINVAL; + + input->free_blocks_count = free_blocks_count = + input->blocks_count - 2 - overhead - sbi->s_itb_per_group; + + if (test_opt(sb, DEBUG)) + printk(KERN_DEBUG "EXT4-fs: adding %s group %u: %u blocks " + "(%d free, %u reserved)\n", + ext4_bg_has_super(sb, input->group) ? "normal" : + "no-super", input->group, input->blocks_count, + free_blocks_count, input->reserved_blocks); + + ext4_get_group_no_and_offset(sb, start, NULL, &offset); + if (group != sbi->s_groups_count) + ext4_warning(sb, __FUNCTION__, + "Cannot add at group %u (only %lu groups)", + input->group, sbi->s_groups_count); + else if (offset != 0) + ext4_warning(sb, __FUNCTION__, "Last group not full"); + else if (input->reserved_blocks > input->blocks_count / 5) + ext4_warning(sb, __FUNCTION__, "Reserved blocks too high (%u)", + input->reserved_blocks); + else if (free_blocks_count < 0) + ext4_warning(sb, __FUNCTION__, "Bad blocks count %u", + input->blocks_count); + else if (!(bh = sb_bread(sb, end - 1))) + ext4_warning(sb, __FUNCTION__, + "Cannot read last block (%llu)", + end - 1); + else if (outside(input->block_bitmap, start, end)) + ext4_warning(sb, __FUNCTION__, + "Block bitmap not in group (block %llu)", + (unsigned long long)input->block_bitmap); + else if (outside(input->inode_bitmap, start, end)) + ext4_warning(sb, __FUNCTION__, + "Inode bitmap not in group (block %llu)", + (unsigned long long)input->inode_bitmap); + else if (outside(input->inode_table, start, end) || + outside(itend - 1, start, end)) + ext4_warning(sb, __FUNCTION__, + "Inode table not in group (blocks %llu-%llu)", + (unsigned long long)input->inode_table, itend - 1); + else if (input->inode_bitmap == input->block_bitmap) + ext4_warning(sb, __FUNCTION__, + "Block bitmap same as inode bitmap (%llu)", + (unsigned long long)input->block_bitmap); + else if (inside(input->block_bitmap, input->inode_table, itend)) + ext4_warning(sb, __FUNCTION__, + "Block bitmap (%llu) in inode table (%llu-%llu)", + (unsigned long long)input->block_bitmap, + (unsigned long long)input->inode_table, itend - 1); + else if (inside(input->inode_bitmap, input->inode_table, itend)) + ext4_warning(sb, __FUNCTION__, + "Inode bitmap (%llu) in inode table (%llu-%llu)", + (unsigned long long)input->inode_bitmap, + (unsigned long long)input->inode_table, itend - 1); + else if (inside(input->block_bitmap, start, metaend)) + ext4_warning(sb, __FUNCTION__, + "Block bitmap (%llu) in GDT table" + " (%llu-%llu)", + (unsigned long long)input->block_bitmap, + start, metaend - 1); + else if (inside(input->inode_bitmap, start, metaend)) + ext4_warning(sb, __FUNCTION__, + "Inode bitmap (%llu) in GDT table" + " (%llu-%llu)", + (unsigned long long)input->inode_bitmap, + start, metaend - 1); + else if (inside(input->inode_table, start, metaend) || + inside(itend - 1, start, metaend)) + ext4_warning(sb, __FUNCTION__, + "Inode table (%llu-%llu) overlaps" + "GDT table (%llu-%llu)", + (unsigned long long)input->inode_table, + itend - 1, start, metaend - 1); + else + err = 0; + brelse(bh); + + return err; +} + +static struct buffer_head *bclean(handle_t *handle, struct super_block *sb, + ext4_fsblk_t blk) +{ + struct buffer_head *bh; + int err; + + bh = sb_getblk(sb, blk); + if (!bh) + return ERR_PTR(-EIO); + if ((err = ext4_journal_get_write_access(handle, bh))) { + brelse(bh); + bh = ERR_PTR(err); + } else { + lock_buffer(bh); + memset(bh->b_data, 0, sb->s_blocksize); + set_buffer_uptodate(bh); + unlock_buffer(bh); + } + + return bh; +} + +/* + * To avoid calling the atomic setbit hundreds or thousands of times, we only + * need to use it within a single byte (to ensure we get endianness right). + * We can use memset for the rest of the bitmap as there are no other users. + */ +static void mark_bitmap_end(int start_bit, int end_bit, char *bitmap) +{ + int i; + + if (start_bit >= end_bit) + return; + + ext4_debug("mark end bits +%d through +%d used\n", start_bit, end_bit); + for (i = start_bit; i < ((start_bit + 7) & ~7UL); i++) + ext4_set_bit(i, bitmap); + if (i < end_bit) + memset(bitmap + (i >> 3), 0xff, (end_bit - i) >> 3); +} + +/* + * Set up the block and inode bitmaps, and the inode table for the new group. + * This doesn't need to be part of the main transaction, since we are only + * changing blocks outside the actual filesystem. We still do journaling to + * ensure the recovery is correct in case of a failure just after resize. + * If any part of this fails, we simply abort the resize. + */ +static int setup_new_group_blocks(struct super_block *sb, + struct ext4_new_group_data *input) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + ext4_fsblk_t start = ext4_group_first_block_no(sb, input->group); + int reserved_gdb = ext4_bg_has_super(sb, input->group) ? + le16_to_cpu(sbi->s_es->s_reserved_gdt_blocks) : 0; + unsigned long gdblocks = ext4_bg_num_gdb(sb, input->group); + struct buffer_head *bh; + handle_t *handle; + ext4_fsblk_t block; + ext4_grpblk_t bit; + int i; + int err = 0, err2; + + handle = ext4_journal_start_sb(sb, reserved_gdb + gdblocks + + 2 + sbi->s_itb_per_group); + if (IS_ERR(handle)) + return PTR_ERR(handle); + + lock_super(sb); + if (input->group != sbi->s_groups_count) { + err = -EBUSY; + goto exit_journal; + } + + if (IS_ERR(bh = bclean(handle, sb, input->block_bitmap))) { + err = PTR_ERR(bh); + goto exit_journal; + } + + if (ext4_bg_has_super(sb, input->group)) { + ext4_debug("mark backup superblock %#04lx (+0)\n", start); + ext4_set_bit(0, bh->b_data); + } + + /* Copy all of the GDT blocks into the backup in this group */ + for (i = 0, bit = 1, block = start + 1; + i < gdblocks; i++, block++, bit++) { + struct buffer_head *gdb; + + ext4_debug("update backup group %#04lx (+%d)\n", block, bit); + + gdb = sb_getblk(sb, block); + if (!gdb) { + err = -EIO; + goto exit_bh; + } + if ((err = ext4_journal_get_write_access(handle, gdb))) { + brelse(gdb); + goto exit_bh; + } + lock_buffer(bh); + memcpy(gdb->b_data, sbi->s_group_desc[i]->b_data, bh->b_size); + set_buffer_uptodate(gdb); + unlock_buffer(bh); + ext4_journal_dirty_metadata(handle, gdb); + ext4_set_bit(bit, bh->b_data); + brelse(gdb); + } + + /* Zero out all of the reserved backup group descriptor table blocks */ + for (i = 0, bit = gdblocks + 1, block = start + bit; + i < reserved_gdb; i++, block++, bit++) { + struct buffer_head *gdb; + + ext4_debug("clear reserved block %#04lx (+%d)\n", block, bit); + + if (IS_ERR(gdb = bclean(handle, sb, block))) { + err = PTR_ERR(bh); + goto exit_bh; + } + ext4_journal_dirty_metadata(handle, gdb); + ext4_set_bit(bit, bh->b_data); + brelse(gdb); + } + ext4_debug("mark block bitmap %#04x (+%ld)\n", input->block_bitmap, + input->block_bitmap - start); + ext4_set_bit(input->block_bitmap - start, bh->b_data); + ext4_debug("mark inode bitmap %#04x (+%ld)\n", input->inode_bitmap, + input->inode_bitmap - start); + ext4_set_bit(input->inode_bitmap - start, bh->b_data); + + /* Zero out all of the inode table blocks */ + for (i = 0, block = input->inode_table, bit = block - start; + i < sbi->s_itb_per_group; i++, bit++, block++) { + struct buffer_head *it; + + ext4_debug("clear inode block %#04lx (+%d)\n", block, bit); + if (IS_ERR(it = bclean(handle, sb, block))) { + err = PTR_ERR(it); + goto exit_bh; + } + ext4_journal_dirty_metadata(handle, it); + brelse(it); + ext4_set_bit(bit, bh->b_data); + } + mark_bitmap_end(input->blocks_count, EXT4_BLOCKS_PER_GROUP(sb), + bh->b_data); + ext4_journal_dirty_metadata(handle, bh); + brelse(bh); + + /* Mark unused entries in inode bitmap used */ + ext4_debug("clear inode bitmap %#04x (+%ld)\n", + input->inode_bitmap, input->inode_bitmap - start); + if (IS_ERR(bh = bclean(handle, sb, input->inode_bitmap))) { + err = PTR_ERR(bh); + goto exit_journal; + } + + mark_bitmap_end(EXT4_INODES_PER_GROUP(sb), EXT4_BLOCKS_PER_GROUP(sb), + bh->b_data); + ext4_journal_dirty_metadata(handle, bh); +exit_bh: + brelse(bh); + +exit_journal: + unlock_super(sb); + if ((err2 = ext4_journal_stop(handle)) && !err) + err = err2; + + return err; +} + + +/* + * Iterate through the groups which hold BACKUP superblock/GDT copies in an + * ext4 filesystem. The counters should be initialized to 1, 5, and 7 before + * calling this for the first time. In a sparse filesystem it will be the + * sequence of powers of 3, 5, and 7: 1, 3, 5, 7, 9, 25, 27, 49, 81, ... + * For a non-sparse filesystem it will be every group: 1, 2, 3, 4, ... + */ +static unsigned ext4_list_backups(struct super_block *sb, unsigned *three, + unsigned *five, unsigned *seven) +{ + unsigned *min = three; + int mult = 3; + unsigned ret; + + if (!EXT4_HAS_RO_COMPAT_FEATURE(sb, + EXT4_FEATURE_RO_COMPAT_SPARSE_SUPER)) { + ret = *min; + *min += 1; + return ret; + } + + if (*five < *min) { + min = five; + mult = 5; + } + if (*seven < *min) { + min = seven; + mult = 7; + } + + ret = *min; + *min *= mult; + + return ret; +} + +/* + * Check that all of the backup GDT blocks are held in the primary GDT block. + * It is assumed that they are stored in group order. Returns the number of + * groups in current filesystem that have BACKUPS, or -ve error code. + */ +static int verify_reserved_gdb(struct super_block *sb, + struct buffer_head *primary) +{ + const ext4_fsblk_t blk = primary->b_blocknr; + const unsigned long end = EXT4_SB(sb)->s_groups_count; + unsigned three = 1; + unsigned five = 5; + unsigned seven = 7; + unsigned grp; + __le32 *p = (__le32 *)primary->b_data; + int gdbackups = 0; + + while ((grp = ext4_list_backups(sb, &three, &five, &seven)) < end) { + if (le32_to_cpu(*p++) != + grp * EXT4_BLOCKS_PER_GROUP(sb) + blk){ + ext4_warning(sb, __FUNCTION__, + "reserved GDT %llu" + " missing grp %d (%llu)", + blk, grp, + grp * + (ext4_fsblk_t)EXT4_BLOCKS_PER_GROUP(sb) + + blk); + return -EINVAL; + } + if (++gdbackups > EXT4_ADDR_PER_BLOCK(sb)) + return -EFBIG; + } + + return gdbackups; +} + +/* + * Called when we need to bring a reserved group descriptor table block into + * use from the resize inode. The primary copy of the new GDT block currently + * is an indirect block (under the double indirect block in the resize inode). + * The new backup GDT blocks will be stored as leaf blocks in this indirect + * block, in group order. Even though we know all the block numbers we need, + * we check to ensure that the resize inode has actually reserved these blocks. + * + * Don't need to update the block bitmaps because the blocks are still in use. + * + * We get all of the error cases out of the way, so that we are sure to not + * fail once we start modifying the data on disk, because JBD has no rollback. + */ +static int add_new_gdb(handle_t *handle, struct inode *inode, + struct ext4_new_group_data *input, + struct buffer_head **primary) +{ + struct super_block *sb = inode->i_sb; + struct ext4_super_block *es = EXT4_SB(sb)->s_es; + unsigned long gdb_num = input->group / EXT4_DESC_PER_BLOCK(sb); + ext4_fsblk_t gdblock = EXT4_SB(sb)->s_sbh->b_blocknr + 1 + gdb_num; + struct buffer_head **o_group_desc, **n_group_desc; + struct buffer_head *dind; + int gdbackups; + struct ext4_iloc iloc; + __le32 *data; + int err; + + if (test_opt(sb, DEBUG)) + printk(KERN_DEBUG + "EXT4-fs: ext4_add_new_gdb: adding group block %lu\n", + gdb_num); + + /* + * If we are not using the primary superblock/GDT copy don't resize, + * because the user tools have no way of handling this. Probably a + * bad time to do it anyways. + */ + if (EXT4_SB(sb)->s_sbh->b_blocknr != + le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block)) { + ext4_warning(sb, __FUNCTION__, + "won't resize using backup superblock at %llu", + (unsigned long long)EXT4_SB(sb)->s_sbh->b_blocknr); + return -EPERM; + } + + *primary = sb_bread(sb, gdblock); + if (!*primary) + return -EIO; + + if ((gdbackups = verify_reserved_gdb(sb, *primary)) < 0) { + err = gdbackups; + goto exit_bh; + } + + data = EXT4_I(inode)->i_data + EXT4_DIND_BLOCK; + dind = sb_bread(sb, le32_to_cpu(*data)); + if (!dind) { + err = -EIO; + goto exit_bh; + } + + data = (__le32 *)dind->b_data; + if (le32_to_cpu(data[gdb_num % EXT4_ADDR_PER_BLOCK(sb)]) != gdblock) { + ext4_warning(sb, __FUNCTION__, + "new group %u GDT block %llu not reserved", + input->group, gdblock); + err = -EINVAL; + goto exit_dind; + } + + if ((err = ext4_journal_get_write_access(handle, EXT4_SB(sb)->s_sbh))) + goto exit_dind; + + if ((err = ext4_journal_get_write_access(handle, *primary))) + goto exit_sbh; + + if ((err = ext4_journal_get_write_access(handle, dind))) + goto exit_primary; + + /* ext4_reserve_inode_write() gets a reference on the iloc */ + if ((err = ext4_reserve_inode_write(handle, inode, &iloc))) + goto exit_dindj; + + n_group_desc = kmalloc((gdb_num + 1) * sizeof(struct buffer_head *), + GFP_KERNEL); + if (!n_group_desc) { + err = -ENOMEM; + ext4_warning (sb, __FUNCTION__, + "not enough memory for %lu groups", gdb_num + 1); + goto exit_inode; + } + + /* + * Finally, we have all of the possible failures behind us... + * + * Remove new GDT block from inode double-indirect block and clear out + * the new GDT block for use (which also "frees" the backup GDT blocks + * from the reserved inode). We don't need to change the bitmaps for + * these blocks, because they are marked as in-use from being in the + * reserved inode, and will become GDT blocks (primary and backup). + */ + data[gdb_num % EXT4_ADDR_PER_BLOCK(sb)] = 0; + ext4_journal_dirty_metadata(handle, dind); + brelse(dind); + inode->i_blocks -= (gdbackups + 1) * sb->s_blocksize >> 9; + ext4_mark_iloc_dirty(handle, inode, &iloc); + memset((*primary)->b_data, 0, sb->s_blocksize); + ext4_journal_dirty_metadata(handle, *primary); + + o_group_desc = EXT4_SB(sb)->s_group_desc; + memcpy(n_group_desc, o_group_desc, + EXT4_SB(sb)->s_gdb_count * sizeof(struct buffer_head *)); + n_group_desc[gdb_num] = *primary; + EXT4_SB(sb)->s_group_desc = n_group_desc; + EXT4_SB(sb)->s_gdb_count++; + kfree(o_group_desc); + + es->s_reserved_gdt_blocks = + cpu_to_le16(le16_to_cpu(es->s_reserved_gdt_blocks) - 1); + ext4_journal_dirty_metadata(handle, EXT4_SB(sb)->s_sbh); + + return 0; + +exit_inode: + //ext4_journal_release_buffer(handle, iloc.bh); + brelse(iloc.bh); +exit_dindj: + //ext4_journal_release_buffer(handle, dind); +exit_primary: + //ext4_journal_release_buffer(handle, *primary); +exit_sbh: + //ext4_journal_release_buffer(handle, *primary); +exit_dind: + brelse(dind); +exit_bh: + brelse(*primary); + + ext4_debug("leaving with error %d\n", err); + return err; +} + +/* + * Called when we are adding a new group which has a backup copy of each of + * the GDT blocks (i.e. sparse group) and there are reserved GDT blocks. + * We need to add these reserved backup GDT blocks to the resize inode, so + * that they are kept for future resizing and not allocated to files. + * + * Each reserved backup GDT block will go into a different indirect block. + * The indirect blocks are actually the primary reserved GDT blocks, + * so we know in advance what their block numbers are. We only get the + * double-indirect block to verify it is pointing to the primary reserved + * GDT blocks so we don't overwrite a data block by accident. The reserved + * backup GDT blocks are stored in their reserved primary GDT block. + */ +static int reserve_backup_gdb(handle_t *handle, struct inode *inode, + struct ext4_new_group_data *input) +{ + struct super_block *sb = inode->i_sb; + int reserved_gdb =le16_to_cpu(EXT4_SB(sb)->s_es->s_reserved_gdt_blocks); + struct buffer_head **primary; + struct buffer_head *dind; + struct ext4_iloc iloc; + ext4_fsblk_t blk; + __le32 *data, *end; + int gdbackups = 0; + int res, i; + int err; + + primary = kmalloc(reserved_gdb * sizeof(*primary), GFP_KERNEL); + if (!primary) + return -ENOMEM; + + data = EXT4_I(inode)->i_data + EXT4_DIND_BLOCK; + dind = sb_bread(sb, le32_to_cpu(*data)); + if (!dind) { + err = -EIO; + goto exit_free; + } + + blk = EXT4_SB(sb)->s_sbh->b_blocknr + 1 + EXT4_SB(sb)->s_gdb_count; + data = (__le32 *)dind->b_data + EXT4_SB(sb)->s_gdb_count; + end = (__le32 *)dind->b_data + EXT4_ADDR_PER_BLOCK(sb); + + /* Get each reserved primary GDT block and verify it holds backups */ + for (res = 0; res < reserved_gdb; res++, blk++) { + if (le32_to_cpu(*data) != blk) { + ext4_warning(sb, __FUNCTION__, + "reserved block %llu" + " not at offset %ld", + blk, + (long)(data - (__le32 *)dind->b_data)); + err = -EINVAL; + goto exit_bh; + } + primary[res] = sb_bread(sb, blk); + if (!primary[res]) { + err = -EIO; + goto exit_bh; + } + if ((gdbackups = verify_reserved_gdb(sb, primary[res])) < 0) { + brelse(primary[res]); + err = gdbackups; + goto exit_bh; + } + if (++data >= end) + data = (__le32 *)dind->b_data; + } + + for (i = 0; i < reserved_gdb; i++) { + if ((err = ext4_journal_get_write_access(handle, primary[i]))) { + /* + int j; + for (j = 0; j < i; j++) + ext4_journal_release_buffer(handle, primary[j]); + */ + goto exit_bh; + } + } + + if ((err = ext4_reserve_inode_write(handle, inode, &iloc))) + goto exit_bh; + + /* + * Finally we can add each of the reserved backup GDT blocks from + * the new group to its reserved primary GDT block. + */ + blk = input->group * EXT4_BLOCKS_PER_GROUP(sb); + for (i = 0; i < reserved_gdb; i++) { + int err2; + data = (__le32 *)primary[i]->b_data; + /* printk("reserving backup %lu[%u] = %lu\n", + primary[i]->b_blocknr, gdbackups, + blk + primary[i]->b_blocknr); */ + data[gdbackups] = cpu_to_le32(blk + primary[i]->b_blocknr); + err2 = ext4_journal_dirty_metadata(handle, primary[i]); + if (!err) + err = err2; + } + inode->i_blocks += reserved_gdb * sb->s_blocksize >> 9; + ext4_mark_iloc_dirty(handle, inode, &iloc); + +exit_bh: + while (--res >= 0) + brelse(primary[res]); + brelse(dind); + +exit_free: + kfree(primary); + + return err; +} + +/* + * Update the backup copies of the ext4 metadata. These don't need to be part + * of the main resize transaction, because e2fsck will re-write them if there + * is a problem (basically only OOM will cause a problem). However, we + * _should_ update the backups if possible, in case the primary gets trashed + * for some reason and we need to run e2fsck from a backup superblock. The + * important part is that the new block and inode counts are in the backup + * superblocks, and the location of the new group metadata in the GDT backups. + * + * We do not need lock_super() for this, because these blocks are not + * otherwise touched by the filesystem code when it is mounted. We don't + * need to worry about last changing from sbi->s_groups_count, because the + * worst that can happen is that we do not copy the full number of backups + * at this time. The resize which changed s_groups_count will backup again. + */ +static void update_backups(struct super_block *sb, + int blk_off, char *data, int size) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + const unsigned long last = sbi->s_groups_count; + const int bpg = EXT4_BLOCKS_PER_GROUP(sb); + unsigned three = 1; + unsigned five = 5; + unsigned seven = 7; + unsigned group; + int rest = sb->s_blocksize - size; + handle_t *handle; + int err = 0, err2; + + handle = ext4_journal_start_sb(sb, EXT4_MAX_TRANS_DATA); + if (IS_ERR(handle)) { + group = 1; + err = PTR_ERR(handle); + goto exit_err; + } + + while ((group = ext4_list_backups(sb, &three, &five, &seven)) < last) { + struct buffer_head *bh; + + /* Out of journal space, and can't get more - abort - so sad */ + if (handle->h_buffer_credits == 0 && + ext4_journal_extend(handle, EXT4_MAX_TRANS_DATA) && + (err = ext4_journal_restart(handle, EXT4_MAX_TRANS_DATA))) + break; + + bh = sb_getblk(sb, group * bpg + blk_off); + if (!bh) { + err = -EIO; + break; + } + ext4_debug("update metadata backup %#04lx\n", + (unsigned long)bh->b_blocknr); + if ((err = ext4_journal_get_write_access(handle, bh))) + break; + lock_buffer(bh); + memcpy(bh->b_data, data, size); + if (rest) + memset(bh->b_data + size, 0, rest); + set_buffer_uptodate(bh); + unlock_buffer(bh); + ext4_journal_dirty_metadata(handle, bh); + brelse(bh); + } + if ((err2 = ext4_journal_stop(handle)) && !err) + err = err2; + + /* + * Ugh! Need to have e2fsck write the backup copies. It is too + * late to revert the resize, we shouldn't fail just because of + * the backup copies (they are only needed in case of corruption). + * + * However, if we got here we have a journal problem too, so we + * can't really start a transaction to mark the superblock. + * Chicken out and just set the flag on the hope it will be written + * to disk, and if not - we will simply wait until next fsck. + */ +exit_err: + if (err) { + ext4_warning(sb, __FUNCTION__, + "can't update backup for group %d (err %d), " + "forcing fsck on next reboot", group, err); + sbi->s_mount_state &= ~EXT4_VALID_FS; + sbi->s_es->s_state &= cpu_to_le16(~EXT4_VALID_FS); + mark_buffer_dirty(sbi->s_sbh); + } +} + +/* Add group descriptor data to an existing or new group descriptor block. + * Ensure we handle all possible error conditions _before_ we start modifying + * the filesystem, because we cannot abort the transaction and not have it + * write the data to disk. + * + * If we are on a GDT block boundary, we need to get the reserved GDT block. + * Otherwise, we may need to add backup GDT blocks for a sparse group. + * + * We only need to hold the superblock lock while we are actually adding + * in the new group's counts to the superblock. Prior to that we have + * not really "added" the group at all. We re-check that we are still + * adding in the last group in case things have changed since verifying. + */ +int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct ext4_super_block *es = sbi->s_es; + int reserved_gdb = ext4_bg_has_super(sb, input->group) ? + le16_to_cpu(es->s_reserved_gdt_blocks) : 0; + struct buffer_head *primary = NULL; + struct ext4_group_desc *gdp; + struct inode *inode = NULL; + handle_t *handle; + int gdb_off, gdb_num; + int err, err2; + + gdb_num = input->group / EXT4_DESC_PER_BLOCK(sb); + gdb_off = input->group % EXT4_DESC_PER_BLOCK(sb); + + if (gdb_off == 0 && !EXT4_HAS_RO_COMPAT_FEATURE(sb, + EXT4_FEATURE_RO_COMPAT_SPARSE_SUPER)) { + ext4_warning(sb, __FUNCTION__, + "Can't resize non-sparse filesystem further"); + return -EPERM; + } + + if (ext4_blocks_count(es) + input->blocks_count < + ext4_blocks_count(es)) { + ext4_warning(sb, __FUNCTION__, "blocks_count overflow\n"); + return -EINVAL; + } + + if (le32_to_cpu(es->s_inodes_count) + EXT4_INODES_PER_GROUP(sb) < + le32_to_cpu(es->s_inodes_count)) { + ext4_warning(sb, __FUNCTION__, "inodes_count overflow\n"); + return -EINVAL; + } + + if (reserved_gdb || gdb_off == 0) { + if (!EXT4_HAS_COMPAT_FEATURE(sb, + EXT4_FEATURE_COMPAT_RESIZE_INODE)){ + ext4_warning(sb, __FUNCTION__, + "No reserved GDT blocks, can't resize"); + return -EPERM; + } + inode = iget(sb, EXT4_RESIZE_INO); + if (!inode || is_bad_inode(inode)) { + ext4_warning(sb, __FUNCTION__, + "Error opening resize inode"); + iput(inode); + return -ENOENT; + } + } + + if ((err = verify_group_input(sb, input))) + goto exit_put; + + if ((err = setup_new_group_blocks(sb, input))) + goto exit_put; + + /* + * We will always be modifying at least the superblock and a GDT + * block. If we are adding a group past the last current GDT block, + * we will also modify the inode and the dindirect block. If we + * are adding a group with superblock/GDT backups we will also + * modify each of the reserved GDT dindirect blocks. + */ + handle = ext4_journal_start_sb(sb, + ext4_bg_has_super(sb, input->group) ? + 3 + reserved_gdb : 4); + if (IS_ERR(handle)) { + err = PTR_ERR(handle); + goto exit_put; + } + + lock_super(sb); + if (input->group != sbi->s_groups_count) { + ext4_warning(sb, __FUNCTION__, + "multiple resizers run on filesystem!"); + err = -EBUSY; + goto exit_journal; + } + + if ((err = ext4_journal_get_write_access(handle, sbi->s_sbh))) + goto exit_journal; + + /* + * We will only either add reserved group blocks to a backup group + * or remove reserved blocks for the first group in a new group block. + * Doing both would be mean more complex code, and sane people don't + * use non-sparse filesystems anymore. This is already checked above. + */ + if (gdb_off) { + primary = sbi->s_group_desc[gdb_num]; + if ((err = ext4_journal_get_write_access(handle, primary))) + goto exit_journal; + + if (reserved_gdb && ext4_bg_num_gdb(sb, input->group) && + (err = reserve_backup_gdb(handle, inode, input))) + goto exit_journal; + } else if ((err = add_new_gdb(handle, inode, input, &primary))) + goto exit_journal; + + /* + * OK, now we've set up the new group. Time to make it active. + * + * Current kernels don't lock all allocations via lock_super(), + * so we have to be safe wrt. concurrent accesses the group + * data. So we need to be careful to set all of the relevant + * group descriptor data etc. *before* we enable the group. + * + * The key field here is sbi->s_groups_count: as long as + * that retains its old value, nobody is going to access the new + * group. + * + * So first we update all the descriptor metadata for the new + * group; then we update the total disk blocks count; then we + * update the groups count to enable the group; then finally we + * update the free space counts so that the system can start + * using the new disk blocks. + */ + + /* Update group descriptor block for new group */ + gdp = (struct ext4_group_desc *)primary->b_data + gdb_off; + + ext4_block_bitmap_set(sb, gdp, input->block_bitmap); /* LV FIXME */ + ext4_inode_bitmap_set(sb, gdp, input->inode_bitmap); /* LV FIXME */ + ext4_inode_table_set(sb, gdp, input->inode_table); /* LV FIXME */ + gdp->bg_free_blocks_count = cpu_to_le16(input->free_blocks_count); + gdp->bg_free_inodes_count = cpu_to_le16(EXT4_INODES_PER_GROUP(sb)); + + /* + * Make the new blocks and inodes valid next. We do this before + * increasing the group count so that once the group is enabled, + * all of its blocks and inodes are already valid. + * + * We always allocate group-by-group, then block-by-block or + * inode-by-inode within a group, so enabling these + * blocks/inodes before the group is live won't actually let us + * allocate the new space yet. + */ + ext4_blocks_count_set(es, ext4_blocks_count(es) + + input->blocks_count); + es->s_inodes_count = cpu_to_le32(le32_to_cpu(es->s_inodes_count) + + EXT4_INODES_PER_GROUP(sb)); + + /* + * We need to protect s_groups_count against other CPUs seeing + * inconsistent state in the superblock. + * + * The precise rules we use are: + * + * * Writers of s_groups_count *must* hold lock_super + * AND + * * Writers must perform a smp_wmb() after updating all dependent + * data and before modifying the groups count + * + * * Readers must hold lock_super() over the access + * OR + * * Readers must perform an smp_rmb() after reading the groups count + * and before reading any dependent data. + * + * NB. These rules can be relaxed when checking the group count + * while freeing data, as we can only allocate from a block + * group after serialising against the group count, and we can + * only then free after serialising in turn against that + * allocation. + */ + smp_wmb(); + + /* Update the global fs size fields */ + sbi->s_groups_count++; + + ext4_journal_dirty_metadata(handle, primary); + + /* Update the reserved block counts only once the new group is + * active. */ + ext4_r_blocks_count_set(es, ext4_r_blocks_count(es) + + input->reserved_blocks); + + /* Update the free space counts */ + percpu_counter_mod(&sbi->s_freeblocks_counter, + input->free_blocks_count); + percpu_counter_mod(&sbi->s_freeinodes_counter, + EXT4_INODES_PER_GROUP(sb)); + + ext4_journal_dirty_metadata(handle, sbi->s_sbh); + sb->s_dirt = 1; + +exit_journal: + unlock_super(sb); + if ((err2 = ext4_journal_stop(handle)) && !err) + err = err2; + if (!err) { + update_backups(sb, sbi->s_sbh->b_blocknr, (char *)es, + sizeof(struct ext4_super_block)); + update_backups(sb, primary->b_blocknr, primary->b_data, + primary->b_size); + } +exit_put: + iput(inode); + return err; +} /* ext4_group_add */ + +/* Extend the filesystem to the new number of blocks specified. This entry + * point is only used to extend the current filesystem to the end of the last + * existing group. It can be accessed via ioctl, or by "remount,resize=<size>" + * for emergencies (because it has no dependencies on reserved blocks). + * + * If we _really_ wanted, we could use default values to call ext4_group_add() + * allow the "remount" trick to work for arbitrary resizing, assuming enough + * GDT blocks are reserved to grow to the desired size. + */ +int ext4_group_extend(struct super_block *sb, struct ext4_super_block *es, + ext4_fsblk_t n_blocks_count) +{ + ext4_fsblk_t o_blocks_count; + unsigned long o_groups_count; + ext4_grpblk_t last; + ext4_grpblk_t add; + struct buffer_head * bh; + handle_t *handle; + int err; + unsigned long freed_blocks; + + /* We don't need to worry about locking wrt other resizers just + * yet: we're going to revalidate es->s_blocks_count after + * taking lock_super() below. */ + o_blocks_count = ext4_blocks_count(es); + o_groups_count = EXT4_SB(sb)->s_groups_count; + + if (test_opt(sb, DEBUG)) + printk(KERN_DEBUG "EXT4-fs: extending last group from %llu uto %llu blocks\n", + o_blocks_count, n_blocks_count); + + if (n_blocks_count == 0 || n_blocks_count == o_blocks_count) + return 0; + + if (n_blocks_count > (sector_t)(~0ULL) >> (sb->s_blocksize_bits - 9)) { + printk(KERN_ERR "EXT4-fs: filesystem on %s:" + " too large to resize to %llu blocks safely\n", + sb->s_id, n_blocks_count); + if (sizeof(sector_t) < 8) + ext4_warning(sb, __FUNCTION__, + "CONFIG_LBD not enabled\n"); + return -EINVAL; + } + + if (n_blocks_count < o_blocks_count) { + ext4_warning(sb, __FUNCTION__, + "can't shrink FS - resize aborted"); + return -EBUSY; + } + + /* Handle the remaining blocks in the last group only. */ + ext4_get_group_no_and_offset(sb, o_blocks_count, NULL, &last); + + if (last == 0) { + ext4_warning(sb, __FUNCTION__, + "need to use ext2online to resize further"); + return -EPERM; + } + + add = EXT4_BLOCKS_PER_GROUP(sb) - last; + + if (o_blocks_count + add < o_blocks_count) { + ext4_warning(sb, __FUNCTION__, "blocks_count overflow"); + return -EINVAL; + } + + if (o_blocks_count + add > n_blocks_count) + add = n_blocks_count - o_blocks_count; + + if (o_blocks_count + add < n_blocks_count) + ext4_warning(sb, __FUNCTION__, + "will only finish group (%llu" + " blocks, %u new)", + o_blocks_count + add, add); + + /* See if the device is actually as big as what was requested */ + bh = sb_bread(sb, o_blocks_count + add -1); + if (!bh) { + ext4_warning(sb, __FUNCTION__, + "can't read last block, resize aborted"); + return -ENOSPC; + } + brelse(bh); + + /* We will update the superblock, one block bitmap, and + * one group descriptor via ext4_free_blocks(). + */ + handle = ext4_journal_start_sb(sb, 3); + if (IS_ERR(handle)) { + err = PTR_ERR(handle); + ext4_warning(sb, __FUNCTION__, "error %d on journal start",err); + goto exit_put; + } + + lock_super(sb); + if (o_blocks_count != ext4_blocks_count(es)) { + ext4_warning(sb, __FUNCTION__, + "multiple resizers run on filesystem!"); + unlock_super(sb); + err = -EBUSY; + goto exit_put; + } + + if ((err = ext4_journal_get_write_access(handle, + EXT4_SB(sb)->s_sbh))) { + ext4_warning(sb, __FUNCTION__, + "error %d on journal write access", err); + unlock_super(sb); + ext4_journal_stop(handle); + goto exit_put; + } + ext4_blocks_count_set(es, o_blocks_count + add); + ext4_journal_dirty_metadata(handle, EXT4_SB(sb)->s_sbh); + sb->s_dirt = 1; + unlock_super(sb); + ext4_debug("freeing blocks %lu through %llu\n", o_blocks_count, + o_blocks_count + add); + ext4_free_blocks_sb(handle, sb, o_blocks_count, add, &freed_blocks); + ext4_debug("freed blocks %llu through %llu\n", o_blocks_count, + o_blocks_count + add); + if ((err = ext4_journal_stop(handle))) + goto exit_put; + if (test_opt(sb, DEBUG)) + printk(KERN_DEBUG "EXT4-fs: extended group to %llu blocks\n", + ext4_blocks_count(es)); + update_backups(sb, EXT4_SB(sb)->s_sbh->b_blocknr, (char *)es, + sizeof(struct ext4_super_block)); +exit_put: + return err; +} /* ext4_group_extend */ diff --git a/fs/ext4/super.c b/fs/ext4/super.c new file mode 100644 index 000000000000..b4b022aa2bc2 --- /dev/null +++ b/fs/ext4/super.c @@ -0,0 +1,2829 @@ +/* + * linux/fs/ext4/super.c + * + * Copyright (C) 1992, 1993, 1994, 1995 + * Remy Card (card@masi.ibp.fr) + * Laboratoire MASI - Institut Blaise Pascal + * Universite Pierre et Marie Curie (Paris VI) + * + * from + * + * linux/fs/minix/inode.c + * + * Copyright (C) 1991, 1992 Linus Torvalds + * + * Big-endian to little-endian byte-swapping/bitmaps by + * David S. Miller (davem@caip.rutgers.edu), 1995 + */ + +#include <linux/module.h> +#include <linux/string.h> +#include <linux/fs.h> +#include <linux/time.h> +#include <linux/jbd2.h> +#include <linux/ext4_fs.h> +#include <linux/ext4_jbd2.h> +#include <linux/slab.h> +#include <linux/init.h> +#include <linux/blkdev.h> +#include <linux/parser.h> +#include <linux/smp_lock.h> +#include <linux/buffer_head.h> +#include <linux/vfs.h> +#include <linux/random.h> +#include <linux/mount.h> +#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 ext4_load_journal(struct super_block *, struct ext4_super_block *, + unsigned long journal_devnum); +static int ext4_create_journal(struct super_block *, struct ext4_super_block *, + unsigned int); +static void ext4_commit_super (struct super_block * sb, + struct ext4_super_block * es, + int sync); +static void ext4_mark_recovery_complete(struct super_block * sb, + struct ext4_super_block * es); +static void ext4_clear_journal_err(struct super_block * sb, + struct ext4_super_block * es); +static int ext4_sync_fs(struct super_block *sb, int wait); +static const char *ext4_decode_error(struct super_block * sb, int errno, + char nbuf[16]); +static int ext4_remount (struct super_block * sb, int * flags, char * data); +static int ext4_statfs (struct dentry * dentry, struct kstatfs * buf); +static void ext4_unlockfs(struct super_block *sb); +static void ext4_write_super (struct super_block * sb); +static void ext4_write_super_lockfs(struct super_block *sb); + + +ext4_fsblk_t ext4_block_bitmap(struct super_block *sb, + struct ext4_group_desc *bg) +{ + return le32_to_cpu(bg->bg_block_bitmap) | + (EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT ? + (ext4_fsblk_t)le32_to_cpu(bg->bg_block_bitmap_hi) << 32 : 0); +} + +ext4_fsblk_t ext4_inode_bitmap(struct super_block *sb, + struct ext4_group_desc *bg) +{ + return le32_to_cpu(bg->bg_inode_bitmap) | + (EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT ? + (ext4_fsblk_t)le32_to_cpu(bg->bg_inode_bitmap_hi) << 32 : 0); +} + +ext4_fsblk_t ext4_inode_table(struct super_block *sb, + struct ext4_group_desc *bg) +{ + return le32_to_cpu(bg->bg_inode_table) | + (EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT ? + (ext4_fsblk_t)le32_to_cpu(bg->bg_inode_table_hi) << 32 : 0); +} + +void ext4_block_bitmap_set(struct super_block *sb, + struct ext4_group_desc *bg, ext4_fsblk_t blk) +{ + bg->bg_block_bitmap = cpu_to_le32((u32)blk); + if (EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT) + bg->bg_block_bitmap_hi = cpu_to_le32(blk >> 32); +} + +void ext4_inode_bitmap_set(struct super_block *sb, + struct ext4_group_desc *bg, ext4_fsblk_t blk) +{ + bg->bg_inode_bitmap = cpu_to_le32((u32)blk); + if (EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT) + bg->bg_inode_bitmap_hi = cpu_to_le32(blk >> 32); +} + +void ext4_inode_table_set(struct super_block *sb, + struct ext4_group_desc *bg, ext4_fsblk_t blk) +{ + bg->bg_inode_table = cpu_to_le32((u32)blk); + if (EXT4_DESC_SIZE(sb) >= EXT4_MIN_DESC_SIZE_64BIT) + bg->bg_inode_table_hi = cpu_to_le32(blk >> 32); +} + +/* + * Wrappers for jbd2_journal_start/end. + * + * The only special thing we need to do here is to make sure that all + * journal_end calls result in the superblock being marked dirty, so + * that sync() will call the filesystem's write_super callback if + * appropriate. + */ +handle_t *ext4_journal_start_sb(struct super_block *sb, int nblocks) +{ + journal_t *journal; + + if (sb->s_flags & MS_RDONLY) + return ERR_PTR(-EROFS); + + /* Special case here: if the journal has aborted behind our + * backs (eg. EIO in the commit thread), then we still need to + * take the FS itself readonly cleanly. */ + journal = EXT4_SB(sb)->s_journal; + if (is_journal_aborted(journal)) { + ext4_abort(sb, __FUNCTION__, + "Detected aborted journal"); + return ERR_PTR(-EROFS); + } + + return jbd2_journal_start(journal, nblocks); +} + +/* + * The only special thing we need to do here is to make sure that all + * jbd2_journal_stop calls result in the superblock being marked dirty, so + * that sync() will call the filesystem's write_super callback if + * appropriate. + */ +int __ext4_journal_stop(const char *where, handle_t *handle) +{ + struct super_block *sb; + int err; + int rc; + + sb = handle->h_transaction->t_journal->j_private; + err = handle->h_err; + rc = jbd2_journal_stop(handle); + + if (!err) + err = rc; + if (err) + __ext4_std_error(sb, where, err); + return err; +} + +void ext4_journal_abort_handle(const char *caller, const char *err_fn, + struct buffer_head *bh, handle_t *handle, int err) +{ + char nbuf[16]; + const char *errstr = ext4_decode_error(NULL, err, nbuf); + + if (bh) + BUFFER_TRACE(bh, "abort"); + + if (!handle->h_err) + handle->h_err = err; + + if (is_handle_aborted(handle)) + return; + + printk(KERN_ERR "%s: aborting transaction: %s in %s\n", + caller, errstr, err_fn); + + jbd2_journal_abort_handle(handle); +} + +/* Deal with the reporting of failure conditions on a filesystem such as + * inconsistencies detected or read IO failures. + * + * On ext2, we can store the error state of the filesystem in the + * superblock. That is not possible on ext4, because we may have other + * write ordering constraints on the superblock which prevent us from + * writing it out straight away; and given that the journal is about to + * be aborted, we can't rely on the current, or future, transactions to + * write out the superblock safely. + * + * We'll just use the jbd2_journal_abort() error code to record an error in + * the journal instead. On recovery, the journal will compain about + * that error until we've noted it down and cleared it. + */ + +static void ext4_handle_error(struct super_block *sb) +{ + struct ext4_super_block *es = EXT4_SB(sb)->s_es; + + EXT4_SB(sb)->s_mount_state |= EXT4_ERROR_FS; + es->s_state |= cpu_to_le16(EXT4_ERROR_FS); + + if (sb->s_flags & MS_RDONLY) + return; + + if (!test_opt (sb, ERRORS_CONT)) { + journal_t *journal = EXT4_SB(sb)->s_journal; + + EXT4_SB(sb)->s_mount_opt |= EXT4_MOUNT_ABORT; + if (journal) + jbd2_journal_abort(journal, -EIO); + } + if (test_opt (sb, ERRORS_RO)) { + printk (KERN_CRIT "Remounting filesystem read-only\n"); + sb->s_flags |= MS_RDONLY; + } + ext4_commit_super(sb, es, 1); + if (test_opt(sb, ERRORS_PANIC)) + panic("EXT4-fs (device %s): panic forced after error\n", + sb->s_id); +} + +void ext4_error (struct super_block * sb, const char * function, + const char * fmt, ...) +{ + va_list args; + + va_start(args, fmt); + printk(KERN_CRIT "EXT4-fs error (device %s): %s: ",sb->s_id, function); + vprintk(fmt, args); + printk("\n"); + va_end(args); + + ext4_handle_error(sb); +} + +static const char *ext4_decode_error(struct super_block * sb, int errno, + char nbuf[16]) +{ + char *errstr = NULL; + + switch (errno) { + case -EIO: + errstr = "IO failure"; + break; + case -ENOMEM: + errstr = "Out of memory"; + break; + case -EROFS: + if (!sb || EXT4_SB(sb)->s_journal->j_flags & JBD2_ABORT) + errstr = "Journal has aborted"; + else + errstr = "Readonly filesystem"; + break; + default: + /* If the caller passed in an extra buffer for unknown + * errors, textualise them now. Else we just return + * NULL. */ + if (nbuf) { + /* Check for truncated error codes... */ + if (snprintf(nbuf, 16, "error %d", -errno) >= 0) + errstr = nbuf; + } + break; + } + + return errstr; +} + +/* __ext4_std_error decodes expected errors from journaling functions + * automatically and invokes the appropriate error response. */ + +void __ext4_std_error (struct super_block * sb, const char * function, + int errno) +{ + char nbuf[16]; + const char *errstr; + + /* Special case: if the error is EROFS, and we're not already + * inside a transaction, then there's really no point in logging + * an error. */ + if (errno == -EROFS && journal_current_handle() == NULL && + (sb->s_flags & MS_RDONLY)) + return; + + errstr = ext4_decode_error(sb, errno, nbuf); + printk (KERN_CRIT "EXT4-fs error (device %s) in %s: %s\n", + sb->s_id, function, errstr); + + ext4_handle_error(sb); +} + +/* + * ext4_abort is a much stronger failure handler than ext4_error. The + * abort function may be used to deal with unrecoverable failures such + * as journal IO errors or ENOMEM at a critical moment in log management. + * + * We unconditionally force the filesystem into an ABORT|READONLY state, + * unless the error response on the fs has been set to panic in which + * case we take the easy way out and panic immediately. + */ + +void ext4_abort (struct super_block * sb, const char * function, + const char * fmt, ...) +{ + va_list args; + + printk (KERN_CRIT "ext4_abort called.\n"); + + va_start(args, fmt); + printk(KERN_CRIT "EXT4-fs error (device %s): %s: ",sb->s_id, function); + vprintk(fmt, args); + printk("\n"); + va_end(args); + + if (test_opt(sb, ERRORS_PANIC)) + panic("EXT4-fs panic from previous error\n"); + + if (sb->s_flags & MS_RDONLY) + return; + + printk(KERN_CRIT "Remounting filesystem read-only\n"); + EXT4_SB(sb)->s_mount_state |= EXT4_ERROR_FS; + sb->s_flags |= MS_RDONLY; + EXT4_SB(sb)->s_mount_opt |= EXT4_MOUNT_ABORT; + jbd2_journal_abort(EXT4_SB(sb)->s_journal, -EIO); +} + +void ext4_warning (struct super_block * sb, const char * function, + const char * fmt, ...) +{ + va_list args; + + va_start(args, fmt); + printk(KERN_WARNING "EXT4-fs warning (device %s): %s: ", + sb->s_id, function); + vprintk(fmt, args); + printk("\n"); + va_end(args); +} + +void ext4_update_dynamic_rev(struct super_block *sb) +{ + struct ext4_super_block *es = EXT4_SB(sb)->s_es; + + if (le32_to_cpu(es->s_rev_level) > EXT4_GOOD_OLD_REV) + return; + + ext4_warning(sb, __FUNCTION__, + "updating to rev %d because of new feature flag, " + "running e2fsck is recommended", + EXT4_DYNAMIC_REV); + + es->s_first_ino = cpu_to_le32(EXT4_GOOD_OLD_FIRST_INO); + es->s_inode_size = cpu_to_le16(EXT4_GOOD_OLD_INODE_SIZE); + es->s_rev_level = cpu_to_le32(EXT4_DYNAMIC_REV); + /* leave es->s_feature_*compat flags alone */ + /* es->s_uuid will be set by e2fsck if empty */ + + /* + * The rest of the superblock fields should be zero, and if not it + * means they are likely already in use, so leave them alone. We + * can leave it up to e2fsck to clean up any inconsistencies there. + */ +} + +/* + * Open the external journal device + */ +static struct block_device *ext4_blkdev_get(dev_t dev) +{ + struct block_device *bdev; + char b[BDEVNAME_SIZE]; + + bdev = open_by_devnum(dev, FMODE_READ|FMODE_WRITE); + if (IS_ERR(bdev)) + goto fail; + return bdev; + +fail: + printk(KERN_ERR "EXT4: failed to open journal device %s: %ld\n", + __bdevname(dev, b), PTR_ERR(bdev)); + return NULL; +} + +/* + * Release the journal device + */ +static int ext4_blkdev_put(struct block_device *bdev) +{ + bd_release(bdev); + return blkdev_put(bdev); +} + +static int ext4_blkdev_remove(struct ext4_sb_info *sbi) +{ + struct block_device *bdev; + int ret = -ENODEV; + + bdev = sbi->journal_bdev; + if (bdev) { + ret = ext4_blkdev_put(bdev); + sbi->journal_bdev = NULL; + } + return ret; +} + +static inline struct inode *orphan_list_entry(struct list_head *l) +{ + return &list_entry(l, struct ext4_inode_info, i_orphan)->vfs_inode; +} + +static void dump_orphan_list(struct super_block *sb, struct ext4_sb_info *sbi) +{ + struct list_head *l; + + printk(KERN_ERR "sb orphan head is %d\n", + le32_to_cpu(sbi->s_es->s_last_orphan)); + + printk(KERN_ERR "sb_info orphan list:\n"); + list_for_each(l, &sbi->s_orphan) { + struct inode *inode = orphan_list_entry(l); + printk(KERN_ERR " " + "inode %s:%lu at %p: mode %o, nlink %d, next %d\n", + inode->i_sb->s_id, inode->i_ino, inode, + inode->i_mode, inode->i_nlink, + NEXT_ORPHAN(inode)); + } +} + +static void ext4_put_super (struct super_block * sb) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct ext4_super_block *es = sbi->s_es; + int i; + + ext4_ext_release(sb); + ext4_xattr_put_super(sb); + jbd2_journal_destroy(sbi->s_journal); + if (!(sb->s_flags & MS_RDONLY)) { + EXT4_CLEAR_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER); + es->s_state = cpu_to_le16(sbi->s_mount_state); + BUFFER_TRACE(sbi->s_sbh, "marking dirty"); + mark_buffer_dirty(sbi->s_sbh); + ext4_commit_super(sb, es, 1); + } + + for (i = 0; i < sbi->s_gdb_count; i++) + brelse(sbi->s_group_desc[i]); + kfree(sbi->s_group_desc); + percpu_counter_destroy(&sbi->s_freeblocks_counter); + percpu_counter_destroy(&sbi->s_freeinodes_counter); + percpu_counter_destroy(&sbi->s_dirs_counter); + brelse(sbi->s_sbh); +#ifdef CONFIG_QUOTA + for (i = 0; i < MAXQUOTAS; i++) + kfree(sbi->s_qf_names[i]); +#endif + + /* Debugging code just in case the in-memory inode orphan list + * isn't empty. The on-disk one can be non-empty if we've + * detected an error and taken the fs readonly, but the + * in-memory list had better be clean by this point. */ + if (!list_empty(&sbi->s_orphan)) + dump_orphan_list(sb, sbi); + J_ASSERT(list_empty(&sbi->s_orphan)); + + invalidate_bdev(sb->s_bdev, 0); + if (sbi->journal_bdev && sbi->journal_bdev != sb->s_bdev) { + /* + * Invalidate the journal device's buffers. We don't want them + * floating about in memory - the physical journal device may + * hotswapped, and it breaks the `ro-after' testing code. + */ + sync_blockdev(sbi->journal_bdev); + invalidate_bdev(sbi->journal_bdev, 0); + ext4_blkdev_remove(sbi); + } + sb->s_fs_info = NULL; + kfree(sbi); + return; +} + +static kmem_cache_t *ext4_inode_cachep; + +/* + * Called inside transaction, so use GFP_NOFS + */ +static struct inode *ext4_alloc_inode(struct super_block *sb) +{ + struct ext4_inode_info *ei; + + ei = kmem_cache_alloc(ext4_inode_cachep, SLAB_NOFS); + if (!ei) + return NULL; +#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL + ei->i_acl = EXT4_ACL_NOT_CACHED; + ei->i_default_acl = EXT4_ACL_NOT_CACHED; +#endif + ei->i_block_alloc_info = NULL; + ei->vfs_inode.i_version = 1; + memset(&ei->i_cached_extent, 0, sizeof(struct ext4_ext_cache)); + return &ei->vfs_inode; +} + +static void ext4_destroy_inode(struct inode *inode) +{ + kmem_cache_free(ext4_inode_cachep, EXT4_I(inode)); +} + +static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags) +{ + struct ext4_inode_info *ei = (struct ext4_inode_info *) foo; + + if ((flags & (SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR)) == + SLAB_CTOR_CONSTRUCTOR) { + INIT_LIST_HEAD(&ei->i_orphan); +#ifdef CONFIG_EXT4DEV_FS_XATTR + init_rwsem(&ei->xattr_sem); +#endif + mutex_init(&ei->truncate_mutex); + inode_init_once(&ei->vfs_inode); + } +} + +static int init_inodecache(void) +{ + ext4_inode_cachep = kmem_cache_create("ext4_inode_cache", + sizeof(struct ext4_inode_info), + 0, (SLAB_RECLAIM_ACCOUNT| + SLAB_MEM_SPREAD), + init_once, NULL); + if (ext4_inode_cachep == NULL) + return -ENOMEM; + return 0; +} + +static void destroy_inodecache(void) +{ + kmem_cache_destroy(ext4_inode_cachep); +} + +static void ext4_clear_inode(struct inode *inode) +{ + struct ext4_block_alloc_info *rsv = EXT4_I(inode)->i_block_alloc_info; +#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL + if (EXT4_I(inode)->i_acl && + EXT4_I(inode)->i_acl != EXT4_ACL_NOT_CACHED) { + posix_acl_release(EXT4_I(inode)->i_acl); + EXT4_I(inode)->i_acl = EXT4_ACL_NOT_CACHED; + } + if (EXT4_I(inode)->i_default_acl && + EXT4_I(inode)->i_default_acl != EXT4_ACL_NOT_CACHED) { + posix_acl_release(EXT4_I(inode)->i_default_acl); + EXT4_I(inode)->i_default_acl = EXT4_ACL_NOT_CACHED; + } +#endif + ext4_discard_reservation(inode); + EXT4_I(inode)->i_block_alloc_info = NULL; + if (unlikely(rsv)) + kfree(rsv); +} + +static inline void ext4_show_quota_options(struct seq_file *seq, struct super_block *sb) +{ +#if defined(CONFIG_QUOTA) + struct ext4_sb_info *sbi = EXT4_SB(sb); + + if (sbi->s_jquota_fmt) + seq_printf(seq, ",jqfmt=%s", + (sbi->s_jquota_fmt == QFMT_VFS_OLD) ? "vfsold": "vfsv0"); + + if (sbi->s_qf_names[USRQUOTA]) + seq_printf(seq, ",usrjquota=%s", sbi->s_qf_names[USRQUOTA]); + + if (sbi->s_qf_names[GRPQUOTA]) + seq_printf(seq, ",grpjquota=%s", sbi->s_qf_names[GRPQUOTA]); + + if (sbi->s_mount_opt & EXT4_MOUNT_USRQUOTA) + seq_puts(seq, ",usrquota"); + + if (sbi->s_mount_opt & EXT4_MOUNT_GRPQUOTA) + seq_puts(seq, ",grpquota"); +#endif +} + +static int ext4_show_options(struct seq_file *seq, struct vfsmount *vfs) +{ + struct super_block *sb = vfs->mnt_sb; + + if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA) + seq_puts(seq, ",data=journal"); + else if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_ORDERED_DATA) + seq_puts(seq, ",data=ordered"); + else if (test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_WRITEBACK_DATA) + seq_puts(seq, ",data=writeback"); + + ext4_show_quota_options(seq, sb); + + return 0; +} + + +static struct dentry *ext4_get_dentry(struct super_block *sb, void *vobjp) +{ + __u32 *objp = vobjp; + unsigned long ino = objp[0]; + __u32 generation = objp[1]; + struct inode *inode; + struct dentry *result; + + if (ino < EXT4_FIRST_INO(sb) && ino != EXT4_ROOT_INO) + return ERR_PTR(-ESTALE); + if (ino > le32_to_cpu(EXT4_SB(sb)->s_es->s_inodes_count)) + return ERR_PTR(-ESTALE); + + /* iget isn't really right if the inode is currently unallocated!! + * + * ext4_read_inode will return a bad_inode if the inode had been + * deleted, so we should be safe. + * + * Currently we don't know the generation for parent directory, so + * a generation of 0 means "accept any" + */ + inode = iget(sb, ino); + if (inode == NULL) + return ERR_PTR(-ENOMEM); + if (is_bad_inode(inode) || + (generation && inode->i_generation != generation)) { + iput(inode); + return ERR_PTR(-ESTALE); + } + /* now to find a dentry. + * If possible, get a well-connected one + */ + result = d_alloc_anon(inode); + if (!result) { + iput(inode); + return ERR_PTR(-ENOMEM); + } + return result; +} + +#ifdef CONFIG_QUOTA +#define QTYPE2NAME(t) ((t)==USRQUOTA?"user":"group") +#define QTYPE2MOPT(on, t) ((t)==USRQUOTA?((on)##USRJQUOTA):((on)##GRPJQUOTA)) + +static int ext4_dquot_initialize(struct inode *inode, int type); +static int ext4_dquot_drop(struct inode *inode); +static int ext4_write_dquot(struct dquot *dquot); +static int ext4_acquire_dquot(struct dquot *dquot); +static int ext4_release_dquot(struct dquot *dquot); +static int ext4_mark_dquot_dirty(struct dquot *dquot); +static int ext4_write_info(struct super_block *sb, int type); +static int ext4_quota_on(struct super_block *sb, int type, int format_id, char *path); +static int ext4_quota_on_mount(struct super_block *sb, int type); +static ssize_t ext4_quota_read(struct super_block *sb, int type, char *data, + size_t len, loff_t off); +static ssize_t ext4_quota_write(struct super_block *sb, int type, + const char *data, size_t len, loff_t off); + +static struct dquot_operations ext4_quota_operations = { + .initialize = ext4_dquot_initialize, + .drop = ext4_dquot_drop, + .alloc_space = dquot_alloc_space, + .alloc_inode = dquot_alloc_inode, + .free_space = dquot_free_space, + .free_inode = dquot_free_inode, + .transfer = dquot_transfer, + .write_dquot = ext4_write_dquot, + .acquire_dquot = ext4_acquire_dquot, + .release_dquot = ext4_release_dquot, + .mark_dirty = ext4_mark_dquot_dirty, + .write_info = ext4_write_info +}; + +static struct quotactl_ops ext4_qctl_operations = { + .quota_on = ext4_quota_on, + .quota_off = vfs_quota_off, + .quota_sync = vfs_quota_sync, + .get_info = vfs_get_dqinfo, + .set_info = vfs_set_dqinfo, + .get_dqblk = vfs_get_dqblk, + .set_dqblk = vfs_set_dqblk +}; +#endif + +static struct super_operations ext4_sops = { + .alloc_inode = ext4_alloc_inode, + .destroy_inode = ext4_destroy_inode, + .read_inode = ext4_read_inode, + .write_inode = ext4_write_inode, + .dirty_inode = ext4_dirty_inode, + .delete_inode = ext4_delete_inode, + .put_super = ext4_put_super, + .write_super = ext4_write_super, + .sync_fs = ext4_sync_fs, + .write_super_lockfs = ext4_write_super_lockfs, + .unlockfs = ext4_unlockfs, + .statfs = ext4_statfs, + .remount_fs = ext4_remount, + .clear_inode = ext4_clear_inode, + .show_options = ext4_show_options, +#ifdef CONFIG_QUOTA + .quota_read = ext4_quota_read, + .quota_write = ext4_quota_write, +#endif +}; + +static struct export_operations ext4_export_ops = { + .get_parent = ext4_get_parent, + .get_dentry = ext4_get_dentry, +}; + +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_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_bh, + Opt_commit, Opt_journal_update, Opt_journal_inum, Opt_journal_dev, + Opt_abort, Opt_data_journal, Opt_data_ordered, Opt_data_writeback, + Opt_usrjquota, Opt_grpjquota, Opt_offusrjquota, Opt_offgrpjquota, + Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_quota, Opt_noquota, + Opt_ignore, Opt_barrier, Opt_err, Opt_resize, Opt_usrquota, + Opt_grpquota, Opt_extents, +}; + +static match_table_t tokens = { + {Opt_bsd_df, "bsddf"}, + {Opt_minix_df, "minixdf"}, + {Opt_grpid, "grpid"}, + {Opt_grpid, "bsdgroups"}, + {Opt_nogrpid, "nogrpid"}, + {Opt_nogrpid, "sysvgroups"}, + {Opt_resgid, "resgid=%u"}, + {Opt_resuid, "resuid=%u"}, + {Opt_sb, "sb=%u"}, + {Opt_err_cont, "errors=continue"}, + {Opt_err_panic, "errors=panic"}, + {Opt_err_ro, "errors=remount-ro"}, + {Opt_nouid32, "nouid32"}, + {Opt_nocheck, "nocheck"}, + {Opt_nocheck, "check=none"}, + {Opt_debug, "debug"}, + {Opt_oldalloc, "oldalloc"}, + {Opt_orlov, "orlov"}, + {Opt_user_xattr, "user_xattr"}, + {Opt_nouser_xattr, "nouser_xattr"}, + {Opt_acl, "acl"}, + {Opt_noacl, "noacl"}, + {Opt_reservation, "reservation"}, + {Opt_noreservation, "noreservation"}, + {Opt_noload, "noload"}, + {Opt_nobh, "nobh"}, + {Opt_bh, "bh"}, + {Opt_commit, "commit=%u"}, + {Opt_journal_update, "journal=update"}, + {Opt_journal_inum, "journal=%u"}, + {Opt_journal_dev, "journal_dev=%u"}, + {Opt_abort, "abort"}, + {Opt_data_journal, "data=journal"}, + {Opt_data_ordered, "data=ordered"}, + {Opt_data_writeback, "data=writeback"}, + {Opt_offusrjquota, "usrjquota="}, + {Opt_usrjquota, "usrjquota=%s"}, + {Opt_offgrpjquota, "grpjquota="}, + {Opt_grpjquota, "grpjquota=%s"}, + {Opt_jqfmt_vfsold, "jqfmt=vfsold"}, + {Opt_jqfmt_vfsv0, "jqfmt=vfsv0"}, + {Opt_grpquota, "grpquota"}, + {Opt_noquota, "noquota"}, + {Opt_quota, "quota"}, + {Opt_usrquota, "usrquota"}, + {Opt_barrier, "barrier=%u"}, + {Opt_extents, "extents"}, + {Opt_err, NULL}, + {Opt_resize, "resize"}, +}; + +static ext4_fsblk_t get_sb_block(void **data) +{ + ext4_fsblk_t sb_block; + char *options = (char *) *data; + + if (!options || strncmp(options, "sb=", 3) != 0) + return 1; /* Default location */ + options += 3; + /*todo: use simple_strtoll with >32bit ext4 */ + sb_block = simple_strtoul(options, &options, 0); + if (*options && *options != ',') { + printk("EXT4-fs: Invalid sb specification: %s\n", + (char *) *data); + return 1; + } + if (*options == ',') + options++; + *data = (void *) options; + return sb_block; +} + +static int parse_options (char *options, struct super_block *sb, + unsigned int *inum, unsigned long *journal_devnum, + ext4_fsblk_t *n_blocks_count, int is_remount) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + char * p; + substring_t args[MAX_OPT_ARGS]; + int data_opt = 0; + int option; +#ifdef CONFIG_QUOTA + int qtype; + char *qname; +#endif + + if (!options) + return 1; + + while ((p = strsep (&options, ",")) != NULL) { + int token; + if (!*p) + continue; + + token = match_token(p, tokens, args); + switch (token) { + case Opt_bsd_df: + clear_opt (sbi->s_mount_opt, MINIX_DF); + break; + case Opt_minix_df: + set_opt (sbi->s_mount_opt, MINIX_DF); + break; + case Opt_grpid: + set_opt (sbi->s_mount_opt, GRPID); + break; + case Opt_nogrpid: + clear_opt (sbi->s_mount_opt, GRPID); + break; + case Opt_resuid: + if (match_int(&args[0], &option)) + return 0; + sbi->s_resuid = option; + break; + case Opt_resgid: + if (match_int(&args[0], &option)) + return 0; + sbi->s_resgid = option; + break; + case Opt_sb: + /* handled by get_sb_block() instead of here */ + /* *sb_block = match_int(&args[0]); */ + break; + case Opt_err_panic: + clear_opt (sbi->s_mount_opt, ERRORS_CONT); + clear_opt (sbi->s_mount_opt, ERRORS_RO); + set_opt (sbi->s_mount_opt, ERRORS_PANIC); + break; + case Opt_err_ro: + clear_opt (sbi->s_mount_opt, ERRORS_CONT); + clear_opt (sbi->s_mount_opt, ERRORS_PANIC); + set_opt (sbi->s_mount_opt, ERRORS_RO); + break; + case Opt_err_cont: + clear_opt (sbi->s_mount_opt, ERRORS_RO); + clear_opt (sbi->s_mount_opt, ERRORS_PANIC); + set_opt (sbi->s_mount_opt, ERRORS_CONT); + break; + case Opt_nouid32: + set_opt (sbi->s_mount_opt, NO_UID32); + break; + case Opt_nocheck: + clear_opt (sbi->s_mount_opt, CHECK); + break; + case Opt_debug: + set_opt (sbi->s_mount_opt, DEBUG); + break; + case Opt_oldalloc: + set_opt (sbi->s_mount_opt, OLDALLOC); + break; + case Opt_orlov: + clear_opt (sbi->s_mount_opt, OLDALLOC); + break; +#ifdef CONFIG_EXT4DEV_FS_XATTR + case Opt_user_xattr: + set_opt (sbi->s_mount_opt, XATTR_USER); + break; + case Opt_nouser_xattr: + clear_opt (sbi->s_mount_opt, XATTR_USER); + break; +#else + case Opt_user_xattr: + case Opt_nouser_xattr: + printk("EXT4 (no)user_xattr options not supported\n"); + break; +#endif +#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL + case Opt_acl: + set_opt(sbi->s_mount_opt, POSIX_ACL); + break; + case Opt_noacl: + clear_opt(sbi->s_mount_opt, POSIX_ACL); + break; +#else + case Opt_acl: + case Opt_noacl: + printk("EXT4 (no)acl options not supported\n"); + break; +#endif + case Opt_reservation: + set_opt(sbi->s_mount_opt, RESERVATION); + break; + case Opt_noreservation: + clear_opt(sbi->s_mount_opt, RESERVATION); + break; + case Opt_journal_update: + /* @@@ FIXME */ + /* Eventually we will want to be able to create + a journal file here. For now, only allow the + user to specify an existing inode to be the + journal file. */ + if (is_remount) { + printk(KERN_ERR "EXT4-fs: cannot specify " + "journal on remount\n"); + return 0; + } + set_opt (sbi->s_mount_opt, UPDATE_JOURNAL); + break; + case Opt_journal_inum: + if (is_remount) { + printk(KERN_ERR "EXT4-fs: cannot specify " + "journal on remount\n"); + return 0; + } + if (match_int(&args[0], &option)) + return 0; + *inum = option; + break; + case Opt_journal_dev: + if (is_remount) { + printk(KERN_ERR "EXT4-fs: cannot specify " + "journal on remount\n"); + return 0; + } + if (match_int(&args[0], &option)) + return 0; + *journal_devnum = option; + break; + case Opt_noload: + set_opt (sbi->s_mount_opt, NOLOAD); + break; + case Opt_commit: + if (match_int(&args[0], &option)) + return 0; + if (option < 0) + return 0; + if (option == 0) + option = JBD_DEFAULT_MAX_COMMIT_AGE; + sbi->s_commit_interval = HZ * option; + break; + case Opt_data_journal: + data_opt = EXT4_MOUNT_JOURNAL_DATA; + goto datacheck; + case Opt_data_ordered: + data_opt = EXT4_MOUNT_ORDERED_DATA; + goto datacheck; + case Opt_data_writeback: + data_opt = EXT4_MOUNT_WRITEBACK_DATA; + datacheck: + if (is_remount) { + if ((sbi->s_mount_opt & EXT4_MOUNT_DATA_FLAGS) + != data_opt) { + printk(KERN_ERR + "EXT4-fs: cannot change data " + "mode on remount\n"); + return 0; + } + } else { + sbi->s_mount_opt &= ~EXT4_MOUNT_DATA_FLAGS; + sbi->s_mount_opt |= data_opt; + } + break; +#ifdef CONFIG_QUOTA + case Opt_usrjquota: + qtype = USRQUOTA; + goto set_qf_name; + case Opt_grpjquota: + qtype = GRPQUOTA; +set_qf_name: + if (sb_any_quota_enabled(sb)) { + printk(KERN_ERR + "EXT4-fs: Cannot change journalled " + "quota options when quota turned on.\n"); + return 0; + } + qname = match_strdup(&args[0]); + if (!qname) { + printk(KERN_ERR + "EXT4-fs: not enough memory for " + "storing quotafile name.\n"); + return 0; + } + if (sbi->s_qf_names[qtype] && + strcmp(sbi->s_qf_names[qtype], qname)) { + printk(KERN_ERR + "EXT4-fs: %s quota file already " + "specified.\n", QTYPE2NAME(qtype)); + kfree(qname); + return 0; + } + sbi->s_qf_names[qtype] = qname; + if (strchr(sbi->s_qf_names[qtype], '/')) { + printk(KERN_ERR + "EXT4-fs: quotafile must be on " + "filesystem root.\n"); + kfree(sbi->s_qf_names[qtype]); + sbi->s_qf_names[qtype] = NULL; + return 0; + } + set_opt(sbi->s_mount_opt, QUOTA); + break; + case Opt_offusrjquota: + qtype = USRQUOTA; + goto clear_qf_name; + case Opt_offgrpjquota: + qtype = GRPQUOTA; +clear_qf_name: + if (sb_any_quota_enabled(sb)) { + printk(KERN_ERR "EXT4-fs: Cannot change " + "journalled quota options when " + "quota turned on.\n"); + return 0; + } + /* + * The space will be released later when all options + * are confirmed to be correct + */ + sbi->s_qf_names[qtype] = NULL; + break; + case Opt_jqfmt_vfsold: + sbi->s_jquota_fmt = QFMT_VFS_OLD; + break; + case Opt_jqfmt_vfsv0: + sbi->s_jquota_fmt = QFMT_VFS_V0; + break; + case Opt_quota: + case Opt_usrquota: + set_opt(sbi->s_mount_opt, QUOTA); + set_opt(sbi->s_mount_opt, USRQUOTA); + break; + case Opt_grpquota: + set_opt(sbi->s_mount_opt, QUOTA); + set_opt(sbi->s_mount_opt, GRPQUOTA); + break; + case Opt_noquota: + if (sb_any_quota_enabled(sb)) { + printk(KERN_ERR "EXT4-fs: Cannot change quota " + "options when quota turned on.\n"); + return 0; + } + clear_opt(sbi->s_mount_opt, QUOTA); + clear_opt(sbi->s_mount_opt, USRQUOTA); + clear_opt(sbi->s_mount_opt, GRPQUOTA); + break; +#else + case Opt_quota: + case Opt_usrquota: + case Opt_grpquota: + case Opt_usrjquota: + case Opt_grpjquota: + case Opt_offusrjquota: + case Opt_offgrpjquota: + case Opt_jqfmt_vfsold: + case Opt_jqfmt_vfsv0: + printk(KERN_ERR + "EXT4-fs: journalled quota options not " + "supported.\n"); + break; + case Opt_noquota: + break; +#endif + case Opt_abort: + set_opt(sbi->s_mount_opt, ABORT); + break; + case Opt_barrier: + if (match_int(&args[0], &option)) + return 0; + if (option) + set_opt(sbi->s_mount_opt, BARRIER); + else + clear_opt(sbi->s_mount_opt, BARRIER); + break; + case Opt_ignore: + break; + case Opt_resize: + if (!is_remount) { + printk("EXT4-fs: resize option only available " + "for remount\n"); + return 0; + } + if (match_int(&args[0], &option) != 0) + return 0; + *n_blocks_count = option; + break; + case Opt_nobh: + set_opt(sbi->s_mount_opt, NOBH); + break; + case Opt_bh: + clear_opt(sbi->s_mount_opt, NOBH); + break; + case Opt_extents: + set_opt (sbi->s_mount_opt, EXTENTS); + break; + default: + printk (KERN_ERR + "EXT4-fs: Unrecognized mount option \"%s\" " + "or missing value\n", p); + return 0; + } + } +#ifdef CONFIG_QUOTA + if (sbi->s_qf_names[USRQUOTA] || sbi->s_qf_names[GRPQUOTA]) { + if ((sbi->s_mount_opt & EXT4_MOUNT_USRQUOTA) && + sbi->s_qf_names[USRQUOTA]) + clear_opt(sbi->s_mount_opt, USRQUOTA); + + if ((sbi->s_mount_opt & EXT4_MOUNT_GRPQUOTA) && + sbi->s_qf_names[GRPQUOTA]) + clear_opt(sbi->s_mount_opt, GRPQUOTA); + + if ((sbi->s_qf_names[USRQUOTA] && + (sbi->s_mount_opt & EXT4_MOUNT_GRPQUOTA)) || + (sbi->s_qf_names[GRPQUOTA] && + (sbi->s_mount_opt & EXT4_MOUNT_USRQUOTA))) { + printk(KERN_ERR "EXT4-fs: old and new quota " + "format mixing.\n"); + return 0; + } + + if (!sbi->s_jquota_fmt) { + printk(KERN_ERR "EXT4-fs: journalled quota format " + "not specified.\n"); + return 0; + } + } else { + if (sbi->s_jquota_fmt) { + printk(KERN_ERR "EXT4-fs: journalled quota format " + "specified with no journalling " + "enabled.\n"); + return 0; + } + } +#endif + return 1; +} + +static int ext4_setup_super(struct super_block *sb, struct ext4_super_block *es, + int read_only) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + int res = 0; + + if (le32_to_cpu(es->s_rev_level) > EXT4_MAX_SUPP_REV) { + printk (KERN_ERR "EXT4-fs warning: revision level too high, " + "forcing read-only mode\n"); + res = MS_RDONLY; + } + if (read_only) + return res; + if (!(sbi->s_mount_state & EXT4_VALID_FS)) + printk (KERN_WARNING "EXT4-fs warning: mounting unchecked fs, " + "running e2fsck is recommended\n"); + else if ((sbi->s_mount_state & EXT4_ERROR_FS)) + printk (KERN_WARNING + "EXT4-fs warning: mounting fs with errors, " + "running e2fsck is recommended\n"); + else if ((__s16) le16_to_cpu(es->s_max_mnt_count) >= 0 && + le16_to_cpu(es->s_mnt_count) >= + (unsigned short) (__s16) le16_to_cpu(es->s_max_mnt_count)) + printk (KERN_WARNING + "EXT4-fs warning: maximal mount count reached, " + "running e2fsck is recommended\n"); + else if (le32_to_cpu(es->s_checkinterval) && + (le32_to_cpu(es->s_lastcheck) + + le32_to_cpu(es->s_checkinterval) <= get_seconds())) + printk (KERN_WARNING + "EXT4-fs warning: checktime reached, " + "running e2fsck is recommended\n"); +#if 0 + /* @@@ We _will_ want to clear the valid bit if we find + * inconsistencies, to force a fsck at reboot. But for + * a plain journaled filesystem we can keep it set as + * valid forever! :) + */ + es->s_state = cpu_to_le16(le16_to_cpu(es->s_state) & ~EXT4_VALID_FS); +#endif + if (!(__s16) le16_to_cpu(es->s_max_mnt_count)) + es->s_max_mnt_count = cpu_to_le16(EXT4_DFL_MAX_MNT_COUNT); + es->s_mnt_count=cpu_to_le16(le16_to_cpu(es->s_mnt_count) + 1); + es->s_mtime = cpu_to_le32(get_seconds()); + ext4_update_dynamic_rev(sb); + EXT4_SET_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER); + + ext4_commit_super(sb, es, 1); + if (test_opt(sb, DEBUG)) + printk(KERN_INFO "[EXT4 FS bs=%lu, gc=%lu, " + "bpg=%lu, ipg=%lu, mo=%04lx]\n", + sb->s_blocksize, + sbi->s_groups_count, + EXT4_BLOCKS_PER_GROUP(sb), + EXT4_INODES_PER_GROUP(sb), + sbi->s_mount_opt); + + printk(KERN_INFO "EXT4 FS on %s, ", sb->s_id); + if (EXT4_SB(sb)->s_journal->j_inode == NULL) { + char b[BDEVNAME_SIZE]; + + printk("external journal on %s\n", + bdevname(EXT4_SB(sb)->s_journal->j_dev, b)); + } else { + printk("internal journal\n"); + } + return res; +} + +/* Called at mount-time, super-block is locked */ +static int ext4_check_descriptors (struct super_block * sb) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + ext4_fsblk_t first_block = le32_to_cpu(sbi->s_es->s_first_data_block); + ext4_fsblk_t last_block; + ext4_fsblk_t block_bitmap; + ext4_fsblk_t inode_bitmap; + ext4_fsblk_t inode_table; + struct ext4_group_desc * gdp = NULL; + int desc_block = 0; + int i; + + ext4_debug ("Checking group descriptors"); + + for (i = 0; i < sbi->s_groups_count; i++) + { + if (i == sbi->s_groups_count - 1) + last_block = ext4_blocks_count(sbi->s_es) - 1; + else + last_block = first_block + + (EXT4_BLOCKS_PER_GROUP(sb) - 1); + + if ((i % EXT4_DESC_PER_BLOCK(sb)) == 0) + gdp = (struct ext4_group_desc *) + sbi->s_group_desc[desc_block++]->b_data; + block_bitmap = ext4_block_bitmap(sb, gdp); + if (block_bitmap < first_block || block_bitmap > last_block) + { + ext4_error (sb, "ext4_check_descriptors", + "Block bitmap for group %d" + " not in group (block %llu)!", + i, block_bitmap); + return 0; + } + inode_bitmap = ext4_inode_bitmap(sb, gdp); + if (inode_bitmap < first_block || inode_bitmap > last_block) + { + ext4_error (sb, "ext4_check_descriptors", + "Inode bitmap for group %d" + " not in group (block %llu)!", + i, inode_bitmap); + return 0; + } + inode_table = ext4_inode_table(sb, gdp); + if (inode_table < first_block || + inode_table + sbi->s_itb_per_group > last_block) + { + ext4_error (sb, "ext4_check_descriptors", + "Inode table for group %d" + " not in group (block %llu)!", + i, inode_table); + return 0; + } + first_block += EXT4_BLOCKS_PER_GROUP(sb); + gdp = (struct ext4_group_desc *) + ((__u8 *)gdp + EXT4_DESC_SIZE(sb)); + } + + ext4_free_blocks_count_set(sbi->s_es, ext4_count_free_blocks(sb)); + sbi->s_es->s_free_inodes_count=cpu_to_le32(ext4_count_free_inodes(sb)); + return 1; +} + + +/* ext4_orphan_cleanup() walks a singly-linked list of inodes (starting at + * the superblock) which were deleted from all directories, but held open by + * a process at the time of a crash. We walk the list and try to delete these + * inodes at recovery time (only with a read-write filesystem). + * + * In order to keep the orphan inode chain consistent during traversal (in + * case of crash during recovery), we link each inode into the superblock + * orphan list_head and handle it the same way as an inode deletion during + * normal operation (which journals the operations for us). + * + * We only do an iget() and an iput() on each inode, which is very safe if we + * accidentally point at an in-use or already deleted inode. The worst that + * can happen in this case is that we get a "bit already cleared" message from + * ext4_free_inode(). The only reason we would point at a wrong inode is if + * e2fsck was run on this filesystem, and it must have already done the orphan + * inode cleanup for us, so we can safely abort without any further action. + */ +static void ext4_orphan_cleanup (struct super_block * sb, + struct ext4_super_block * es) +{ + unsigned int s_flags = sb->s_flags; + int nr_orphans = 0, nr_truncates = 0; +#ifdef CONFIG_QUOTA + int i; +#endif + if (!es->s_last_orphan) { + jbd_debug(4, "no orphan inodes to clean up\n"); + return; + } + + if (EXT4_SB(sb)->s_mount_state & EXT4_ERROR_FS) { + if (es->s_last_orphan) + jbd_debug(1, "Errors on filesystem, " + "clearing orphan list.\n"); + es->s_last_orphan = 0; + jbd_debug(1, "Skipping orphan recovery on fs with errors.\n"); + return; + } + + if (s_flags & MS_RDONLY) { + printk(KERN_INFO "EXT4-fs: %s: orphan cleanup on readonly fs\n", + sb->s_id); + sb->s_flags &= ~MS_RDONLY; + } +#ifdef CONFIG_QUOTA + /* Needed for iput() to work correctly and not trash data */ + sb->s_flags |= MS_ACTIVE; + /* Turn on quotas so that they are updated correctly */ + for (i = 0; i < MAXQUOTAS; i++) { + if (EXT4_SB(sb)->s_qf_names[i]) { + int ret = ext4_quota_on_mount(sb, i); + if (ret < 0) + printk(KERN_ERR + "EXT4-fs: Cannot turn on journalled " + "quota: error %d\n", ret); + } + } +#endif + + while (es->s_last_orphan) { + struct inode *inode; + + if (!(inode = + ext4_orphan_get(sb, le32_to_cpu(es->s_last_orphan)))) { + es->s_last_orphan = 0; + break; + } + + list_add(&EXT4_I(inode)->i_orphan, &EXT4_SB(sb)->s_orphan); + DQUOT_INIT(inode); + if (inode->i_nlink) { + printk(KERN_DEBUG + "%s: truncating inode %lu to %Ld bytes\n", + __FUNCTION__, inode->i_ino, inode->i_size); + jbd_debug(2, "truncating inode %lu to %Ld bytes\n", + inode->i_ino, inode->i_size); + ext4_truncate(inode); + nr_truncates++; + } else { + printk(KERN_DEBUG + "%s: deleting unreferenced inode %lu\n", + __FUNCTION__, inode->i_ino); + jbd_debug(2, "deleting unreferenced inode %lu\n", + inode->i_ino); + nr_orphans++; + } + iput(inode); /* The delete magic happens here! */ + } + +#define PLURAL(x) (x), ((x)==1) ? "" : "s" + + if (nr_orphans) + printk(KERN_INFO "EXT4-fs: %s: %d orphan inode%s deleted\n", + sb->s_id, PLURAL(nr_orphans)); + if (nr_truncates) + printk(KERN_INFO "EXT4-fs: %s: %d truncate%s cleaned up\n", + sb->s_id, PLURAL(nr_truncates)); +#ifdef CONFIG_QUOTA + /* Turn quotas off */ + for (i = 0; i < MAXQUOTAS; i++) { + if (sb_dqopt(sb)->files[i]) + vfs_quota_off(sb, i); + } +#endif + sb->s_flags = s_flags; /* Restore MS_RDONLY status */ +} + +#define log2(n) ffz(~(n)) + +/* + * Maximal file size. There is a direct, and {,double-,triple-}indirect + * block limit, and also a limit of (2^32 - 1) 512-byte sectors in i_blocks. + * We need to be 1 filesystem block less than the 2^32 sector limit. + */ +static loff_t ext4_max_size(int bits) +{ + loff_t res = EXT4_NDIR_BLOCKS; + /* This constant is calculated to be the largest file size for a + * dense, 4k-blocksize file such that the total number of + * sectors in the file, including data and all indirect blocks, + * does not exceed 2^32. */ + const loff_t upper_limit = 0x1ff7fffd000LL; + + res += 1LL << (bits-2); + res += 1LL << (2*(bits-2)); + res += 1LL << (3*(bits-2)); + res <<= bits; + if (res > upper_limit) + res = upper_limit; + return res; +} + +static ext4_fsblk_t descriptor_loc(struct super_block *sb, + ext4_fsblk_t logical_sb_block, int nr) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + unsigned long bg, first_meta_bg; + int has_super = 0; + + first_meta_bg = le32_to_cpu(sbi->s_es->s_first_meta_bg); + + if (!EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_META_BG) || + nr < first_meta_bg) + return logical_sb_block + nr + 1; + bg = sbi->s_desc_per_block * nr; + if (ext4_bg_has_super(sb, bg)) + has_super = 1; + return (has_super + ext4_group_first_block_no(sb, bg)); +} + + +static int ext4_fill_super (struct super_block *sb, void *data, int silent) +{ + struct buffer_head * bh; + struct ext4_super_block *es = NULL; + struct ext4_sb_info *sbi; + ext4_fsblk_t block; + ext4_fsblk_t sb_block = get_sb_block(&data); + ext4_fsblk_t logical_sb_block; + unsigned long offset = 0; + unsigned int journal_inum = 0; + unsigned long journal_devnum = 0; + unsigned long def_mount_opts; + struct inode *root; + int blocksize; + int hblock; + int db_count; + int i; + int needs_recovery; + __le32 features; + __u64 blocks_count; + + sbi = kzalloc(sizeof(*sbi), GFP_KERNEL); + if (!sbi) + return -ENOMEM; + sb->s_fs_info = sbi; + sbi->s_mount_opt = 0; + sbi->s_resuid = EXT4_DEF_RESUID; + sbi->s_resgid = EXT4_DEF_RESGID; + + unlock_kernel(); + + blocksize = sb_min_blocksize(sb, EXT4_MIN_BLOCK_SIZE); + if (!blocksize) { + printk(KERN_ERR "EXT4-fs: unable to set blocksize\n"); + goto out_fail; + } + + /* + * The ext4 superblock will not be buffer aligned for other than 1kB + * block sizes. We need to calculate the offset from buffer start. + */ + if (blocksize != EXT4_MIN_BLOCK_SIZE) { + logical_sb_block = sb_block * EXT4_MIN_BLOCK_SIZE; + offset = do_div(logical_sb_block, blocksize); + } else { + logical_sb_block = sb_block; + } + + if (!(bh = sb_bread(sb, logical_sb_block))) { + printk (KERN_ERR "EXT4-fs: unable to read superblock\n"); + goto out_fail; + } + /* + * Note: s_es must be initialized as soon as possible because + * some ext4 macro-instructions depend on its value + */ + es = (struct ext4_super_block *) (((char *)bh->b_data) + offset); + sbi->s_es = es; + sb->s_magic = le16_to_cpu(es->s_magic); + if (sb->s_magic != EXT4_SUPER_MAGIC) + goto cantfind_ext4; + + /* Set defaults before we parse the mount options */ + def_mount_opts = le32_to_cpu(es->s_default_mount_opts); + if (def_mount_opts & EXT4_DEFM_DEBUG) + set_opt(sbi->s_mount_opt, DEBUG); + if (def_mount_opts & EXT4_DEFM_BSDGROUPS) + set_opt(sbi->s_mount_opt, GRPID); + if (def_mount_opts & EXT4_DEFM_UID16) + set_opt(sbi->s_mount_opt, NO_UID32); + if (def_mount_opts & EXT4_DEFM_XATTR_USER) + set_opt(sbi->s_mount_opt, XATTR_USER); + if (def_mount_opts & EXT4_DEFM_ACL) + set_opt(sbi->s_mount_opt, POSIX_ACL); + if ((def_mount_opts & EXT4_DEFM_JMODE) == EXT4_DEFM_JMODE_DATA) + sbi->s_mount_opt |= EXT4_MOUNT_JOURNAL_DATA; + else if ((def_mount_opts & EXT4_DEFM_JMODE) == EXT4_DEFM_JMODE_ORDERED) + sbi->s_mount_opt |= EXT4_MOUNT_ORDERED_DATA; + else if ((def_mount_opts & EXT4_DEFM_JMODE) == EXT4_DEFM_JMODE_WBACK) + sbi->s_mount_opt |= EXT4_MOUNT_WRITEBACK_DATA; + + if (le16_to_cpu(sbi->s_es->s_errors) == EXT4_ERRORS_PANIC) + set_opt(sbi->s_mount_opt, ERRORS_PANIC); + else if (le16_to_cpu(sbi->s_es->s_errors) == EXT4_ERRORS_RO) + set_opt(sbi->s_mount_opt, ERRORS_RO); + else + set_opt(sbi->s_mount_opt, ERRORS_CONT); + + sbi->s_resuid = le16_to_cpu(es->s_def_resuid); + sbi->s_resgid = le16_to_cpu(es->s_def_resgid); + + set_opt(sbi->s_mount_opt, RESERVATION); + + if (!parse_options ((char *) data, sb, &journal_inum, &journal_devnum, + NULL, 0)) + goto failed_mount; + + sb->s_flags = (sb->s_flags & ~MS_POSIXACL) | + ((sbi->s_mount_opt & EXT4_MOUNT_POSIX_ACL) ? MS_POSIXACL : 0); + + if (le32_to_cpu(es->s_rev_level) == EXT4_GOOD_OLD_REV && + (EXT4_HAS_COMPAT_FEATURE(sb, ~0U) || + EXT4_HAS_RO_COMPAT_FEATURE(sb, ~0U) || + EXT4_HAS_INCOMPAT_FEATURE(sb, ~0U))) + printk(KERN_WARNING + "EXT4-fs warning: feature flags set on rev 0 fs, " + "running e2fsck is recommended\n"); + /* + * Check feature flags regardless of the revision level, since we + * previously didn't change the revision level when setting the flags, + * so there is a chance incompat flags are set on a rev 0 filesystem. + */ + features = EXT4_HAS_INCOMPAT_FEATURE(sb, ~EXT4_FEATURE_INCOMPAT_SUPP); + if (features) { + printk(KERN_ERR "EXT4-fs: %s: couldn't mount because of " + "unsupported optional features (%x).\n", + sb->s_id, le32_to_cpu(features)); + goto failed_mount; + } + features = EXT4_HAS_RO_COMPAT_FEATURE(sb, ~EXT4_FEATURE_RO_COMPAT_SUPP); + if (!(sb->s_flags & MS_RDONLY) && features) { + printk(KERN_ERR "EXT4-fs: %s: couldn't mount RDWR because of " + "unsupported optional features (%x).\n", + sb->s_id, le32_to_cpu(features)); + goto failed_mount; + } + blocksize = BLOCK_SIZE << le32_to_cpu(es->s_log_block_size); + + if (blocksize < EXT4_MIN_BLOCK_SIZE || + blocksize > EXT4_MAX_BLOCK_SIZE) { + printk(KERN_ERR + "EXT4-fs: Unsupported filesystem blocksize %d on %s.\n", + blocksize, sb->s_id); + goto failed_mount; + } + + hblock = bdev_hardsect_size(sb->s_bdev); + if (sb->s_blocksize != blocksize) { + /* + * Make sure the blocksize for the filesystem is larger + * than the hardware sectorsize for the machine. + */ + if (blocksize < hblock) { + printk(KERN_ERR "EXT4-fs: blocksize %d too small for " + "device blocksize %d.\n", blocksize, hblock); + goto failed_mount; + } + + brelse (bh); + sb_set_blocksize(sb, blocksize); + logical_sb_block = sb_block * EXT4_MIN_BLOCK_SIZE; + offset = do_div(logical_sb_block, blocksize); + bh = sb_bread(sb, logical_sb_block); + if (!bh) { + printk(KERN_ERR + "EXT4-fs: Can't read superblock on 2nd try.\n"); + goto failed_mount; + } + es = (struct ext4_super_block *)(((char *)bh->b_data) + offset); + sbi->s_es = es; + if (es->s_magic != cpu_to_le16(EXT4_SUPER_MAGIC)) { + printk (KERN_ERR + "EXT4-fs: Magic mismatch, very weird !\n"); + goto failed_mount; + } + } + + sb->s_maxbytes = ext4_max_size(sb->s_blocksize_bits); + + if (le32_to_cpu(es->s_rev_level) == EXT4_GOOD_OLD_REV) { + sbi->s_inode_size = EXT4_GOOD_OLD_INODE_SIZE; + sbi->s_first_ino = EXT4_GOOD_OLD_FIRST_INO; + } else { + sbi->s_inode_size = le16_to_cpu(es->s_inode_size); + sbi->s_first_ino = le32_to_cpu(es->s_first_ino); + if ((sbi->s_inode_size < EXT4_GOOD_OLD_INODE_SIZE) || + (sbi->s_inode_size & (sbi->s_inode_size - 1)) || + (sbi->s_inode_size > blocksize)) { + printk (KERN_ERR + "EXT4-fs: unsupported inode size: %d\n", + sbi->s_inode_size); + goto failed_mount; + } + } + sbi->s_frag_size = EXT4_MIN_FRAG_SIZE << + le32_to_cpu(es->s_log_frag_size); + if (blocksize != sbi->s_frag_size) { + printk(KERN_ERR + "EXT4-fs: fragsize %lu != blocksize %u (unsupported)\n", + sbi->s_frag_size, blocksize); + goto failed_mount; + } + sbi->s_desc_size = le16_to_cpu(es->s_desc_size); + if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_64BIT)) { + if (sbi->s_desc_size < EXT4_MIN_DESC_SIZE_64BIT || + sbi->s_desc_size > EXT4_MAX_DESC_SIZE || + sbi->s_desc_size & (sbi->s_desc_size - 1)) { + printk(KERN_ERR + "EXT4-fs: unsupported descriptor size %lu\n", + sbi->s_desc_size); + goto failed_mount; + } + } else + sbi->s_desc_size = EXT4_MIN_DESC_SIZE; + sbi->s_blocks_per_group = le32_to_cpu(es->s_blocks_per_group); + sbi->s_frags_per_group = le32_to_cpu(es->s_frags_per_group); + sbi->s_inodes_per_group = le32_to_cpu(es->s_inodes_per_group); + if (EXT4_INODE_SIZE(sb) == 0) + goto cantfind_ext4; + sbi->s_inodes_per_block = blocksize / EXT4_INODE_SIZE(sb); + if (sbi->s_inodes_per_block == 0) + goto cantfind_ext4; + sbi->s_itb_per_group = sbi->s_inodes_per_group / + sbi->s_inodes_per_block; + sbi->s_desc_per_block = blocksize / EXT4_DESC_SIZE(sb); + sbi->s_sbh = bh; + sbi->s_mount_state = le16_to_cpu(es->s_state); + sbi->s_addr_per_block_bits = log2(EXT4_ADDR_PER_BLOCK(sb)); + sbi->s_desc_per_block_bits = log2(EXT4_DESC_PER_BLOCK(sb)); + for (i=0; i < 4; i++) + sbi->s_hash_seed[i] = le32_to_cpu(es->s_hash_seed[i]); + sbi->s_def_hash_version = es->s_def_hash_version; + + if (sbi->s_blocks_per_group > blocksize * 8) { + printk (KERN_ERR + "EXT4-fs: #blocks per group too big: %lu\n", + sbi->s_blocks_per_group); + goto failed_mount; + } + if (sbi->s_frags_per_group > blocksize * 8) { + printk (KERN_ERR + "EXT4-fs: #fragments per group too big: %lu\n", + sbi->s_frags_per_group); + goto failed_mount; + } + if (sbi->s_inodes_per_group > blocksize * 8) { + printk (KERN_ERR + "EXT4-fs: #inodes per group too big: %lu\n", + sbi->s_inodes_per_group); + goto failed_mount; + } + + if (ext4_blocks_count(es) > + (sector_t)(~0ULL) >> (sb->s_blocksize_bits - 9)) { + printk(KERN_ERR "EXT4-fs: filesystem on %s:" + " too large to mount safely\n", sb->s_id); + if (sizeof(sector_t) < 8) + printk(KERN_WARNING "EXT4-fs: CONFIG_LBD not " + "enabled\n"); + goto failed_mount; + } + + if (EXT4_BLOCKS_PER_GROUP(sb) == 0) + goto cantfind_ext4; + blocks_count = (ext4_blocks_count(es) - + le32_to_cpu(es->s_first_data_block) + + EXT4_BLOCKS_PER_GROUP(sb) - 1); + do_div(blocks_count, EXT4_BLOCKS_PER_GROUP(sb)); + sbi->s_groups_count = blocks_count; + db_count = (sbi->s_groups_count + EXT4_DESC_PER_BLOCK(sb) - 1) / + EXT4_DESC_PER_BLOCK(sb); + sbi->s_group_desc = kmalloc(db_count * sizeof (struct buffer_head *), + GFP_KERNEL); + if (sbi->s_group_desc == NULL) { + printk (KERN_ERR "EXT4-fs: not enough memory\n"); + goto failed_mount; + } + + bgl_lock_init(&sbi->s_blockgroup_lock); + + for (i = 0; i < db_count; i++) { + block = descriptor_loc(sb, logical_sb_block, i); + sbi->s_group_desc[i] = sb_bread(sb, block); + if (!sbi->s_group_desc[i]) { + printk (KERN_ERR "EXT4-fs: " + "can't read group descriptor %d\n", i); + db_count = i; + goto failed_mount2; + } + } + if (!ext4_check_descriptors (sb)) { + printk(KERN_ERR "EXT4-fs: group descriptors corrupted!\n"); + goto failed_mount2; + } + sbi->s_gdb_count = db_count; + get_random_bytes(&sbi->s_next_generation, sizeof(u32)); + spin_lock_init(&sbi->s_next_gen_lock); + + percpu_counter_init(&sbi->s_freeblocks_counter, + ext4_count_free_blocks(sb)); + percpu_counter_init(&sbi->s_freeinodes_counter, + ext4_count_free_inodes(sb)); + percpu_counter_init(&sbi->s_dirs_counter, + ext4_count_dirs(sb)); + + /* per fileystem reservation list head & lock */ + spin_lock_init(&sbi->s_rsv_window_lock); + sbi->s_rsv_window_root = RB_ROOT; + /* Add a single, static dummy reservation to the start of the + * reservation window list --- it gives us a placeholder for + * append-at-start-of-list which makes the allocation logic + * _much_ simpler. */ + sbi->s_rsv_window_head.rsv_start = EXT4_RESERVE_WINDOW_NOT_ALLOCATED; + sbi->s_rsv_window_head.rsv_end = EXT4_RESERVE_WINDOW_NOT_ALLOCATED; + sbi->s_rsv_window_head.rsv_alloc_hit = 0; + sbi->s_rsv_window_head.rsv_goal_size = 0; + ext4_rsv_window_add(sb, &sbi->s_rsv_window_head); + + /* + * set up enough so that it can read an inode + */ + sb->s_op = &ext4_sops; + sb->s_export_op = &ext4_export_ops; + sb->s_xattr = ext4_xattr_handlers; +#ifdef CONFIG_QUOTA + sb->s_qcop = &ext4_qctl_operations; + sb->dq_op = &ext4_quota_operations; +#endif + INIT_LIST_HEAD(&sbi->s_orphan); /* unlinked but open files */ + + sb->s_root = NULL; + + needs_recovery = (es->s_last_orphan != 0 || + EXT4_HAS_INCOMPAT_FEATURE(sb, + EXT4_FEATURE_INCOMPAT_RECOVER)); + + /* + * The first inode we look at is the journal inode. Don't try + * root first: it may be modified in the journal! + */ + if (!test_opt(sb, NOLOAD) && + EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_HAS_JOURNAL)) { + if (ext4_load_journal(sb, es, journal_devnum)) + goto failed_mount3; + } else if (journal_inum) { + if (ext4_create_journal(sb, es, journal_inum)) + goto failed_mount3; + } else { + if (!silent) + printk (KERN_ERR + "ext4: No journal on filesystem on %s\n", + sb->s_id); + goto failed_mount3; + } + + /* We have now updated the journal if required, so we can + * validate the data journaling mode. */ + switch (test_opt(sb, DATA_FLAGS)) { + case 0: + /* No mode set, assume a default based on the journal + * capabilities: ORDERED_DATA if the journal can + * cope, else JOURNAL_DATA + */ + if (jbd2_journal_check_available_features + (sbi->s_journal, 0, 0, JBD2_FEATURE_INCOMPAT_REVOKE)) + set_opt(sbi->s_mount_opt, ORDERED_DATA); + else + set_opt(sbi->s_mount_opt, JOURNAL_DATA); + break; + + case EXT4_MOUNT_ORDERED_DATA: + case EXT4_MOUNT_WRITEBACK_DATA: + if (!jbd2_journal_check_available_features + (sbi->s_journal, 0, 0, JBD2_FEATURE_INCOMPAT_REVOKE)) { + printk(KERN_ERR "EXT4-fs: Journal does not support " + "requested data journaling mode\n"); + goto failed_mount4; + } + default: + break; + } + + if (test_opt(sb, NOBH)) { + if (!(test_opt(sb, DATA_FLAGS) == EXT4_MOUNT_WRITEBACK_DATA)) { + printk(KERN_WARNING "EXT4-fs: Ignoring nobh option - " + "its supported only with writeback mode\n"); + clear_opt(sbi->s_mount_opt, NOBH); + } + } + /* + * The jbd2_journal_load will have done any necessary log recovery, + * so we can safely mount the rest of the filesystem now. + */ + + root = iget(sb, EXT4_ROOT_INO); + sb->s_root = d_alloc_root(root); + if (!sb->s_root) { + printk(KERN_ERR "EXT4-fs: get root inode failed\n"); + iput(root); + goto failed_mount4; + } + if (!S_ISDIR(root->i_mode) || !root->i_blocks || !root->i_size) { + dput(sb->s_root); + sb->s_root = NULL; + printk(KERN_ERR "EXT4-fs: corrupt root inode, run e2fsck\n"); + goto failed_mount4; + } + + ext4_setup_super (sb, es, sb->s_flags & MS_RDONLY); + /* + * akpm: core read_super() calls in here with the superblock locked. + * That deadlocks, because orphan cleanup needs to lock the superblock + * in numerous places. Here we just pop the lock - it's relatively + * harmless, because we are now ready to accept write_super() requests, + * and aviro says that's the only reason for hanging onto the + * superblock lock. + */ + EXT4_SB(sb)->s_mount_state |= EXT4_ORPHAN_FS; + ext4_orphan_cleanup(sb, es); + EXT4_SB(sb)->s_mount_state &= ~EXT4_ORPHAN_FS; + if (needs_recovery) + printk (KERN_INFO "EXT4-fs: recovery complete.\n"); + ext4_mark_recovery_complete(sb, es); + printk (KERN_INFO "EXT4-fs: mounted filesystem with %s data mode.\n", + test_opt(sb,DATA_FLAGS) == EXT4_MOUNT_JOURNAL_DATA ? "journal": + test_opt(sb,DATA_FLAGS) == EXT4_MOUNT_ORDERED_DATA ? "ordered": + "writeback"); + + ext4_ext_init(sb); + + lock_kernel(); + return 0; + +cantfind_ext4: + if (!silent) + printk(KERN_ERR "VFS: Can't find ext4 filesystem on dev %s.\n", + sb->s_id); + goto failed_mount; + +failed_mount4: + jbd2_journal_destroy(sbi->s_journal); +failed_mount3: + percpu_counter_destroy(&sbi->s_freeblocks_counter); + percpu_counter_destroy(&sbi->s_freeinodes_counter); + percpu_counter_destroy(&sbi->s_dirs_counter); +failed_mount2: + for (i = 0; i < db_count; i++) + brelse(sbi->s_group_desc[i]); + kfree(sbi->s_group_desc); +failed_mount: +#ifdef CONFIG_QUOTA + for (i = 0; i < MAXQUOTAS; i++) + kfree(sbi->s_qf_names[i]); +#endif + ext4_blkdev_remove(sbi); + brelse(bh); +out_fail: + sb->s_fs_info = NULL; + kfree(sbi); + lock_kernel(); + return -EINVAL; +} + +/* + * Setup any per-fs journal parameters now. We'll do this both on + * initial mount, once the journal has been initialised but before we've + * done any recovery; and again on any subsequent remount. + */ +static void ext4_init_journal_params(struct super_block *sb, journal_t *journal) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + + if (sbi->s_commit_interval) + journal->j_commit_interval = sbi->s_commit_interval; + /* We could also set up an ext4-specific default for the commit + * interval here, but for now we'll just fall back to the jbd + * default. */ + + spin_lock(&journal->j_state_lock); + if (test_opt(sb, BARRIER)) + journal->j_flags |= JBD2_BARRIER; + else + journal->j_flags &= ~JBD2_BARRIER; + spin_unlock(&journal->j_state_lock); +} + +static journal_t *ext4_get_journal(struct super_block *sb, + unsigned int journal_inum) +{ + struct inode *journal_inode; + journal_t *journal; + + /* First, test for the existence of a valid inode on disk. Bad + * things happen if we iget() an unused inode, as the subsequent + * iput() will try to delete it. */ + + journal_inode = iget(sb, journal_inum); + if (!journal_inode) { + printk(KERN_ERR "EXT4-fs: no journal found.\n"); + return NULL; + } + if (!journal_inode->i_nlink) { + make_bad_inode(journal_inode); + iput(journal_inode); + printk(KERN_ERR "EXT4-fs: journal inode is deleted.\n"); + return NULL; + } + + jbd_debug(2, "Journal inode found at %p: %Ld bytes\n", + journal_inode, journal_inode->i_size); + if (is_bad_inode(journal_inode) || !S_ISREG(journal_inode->i_mode)) { + printk(KERN_ERR "EXT4-fs: invalid journal inode.\n"); + iput(journal_inode); + return NULL; + } + + journal = jbd2_journal_init_inode(journal_inode); + if (!journal) { + printk(KERN_ERR "EXT4-fs: Could not load journal inode\n"); + iput(journal_inode); + return NULL; + } + journal->j_private = sb; + ext4_init_journal_params(sb, journal); + return journal; +} + +static journal_t *ext4_get_dev_journal(struct super_block *sb, + dev_t j_dev) +{ + struct buffer_head * bh; + journal_t *journal; + ext4_fsblk_t start; + ext4_fsblk_t len; + int hblock, blocksize; + ext4_fsblk_t sb_block; + unsigned long offset; + struct ext4_super_block * es; + struct block_device *bdev; + + bdev = ext4_blkdev_get(j_dev); + if (bdev == NULL) + return NULL; + + if (bd_claim(bdev, sb)) { + printk(KERN_ERR + "EXT4: failed to claim external journal device.\n"); + blkdev_put(bdev); + return NULL; + } + + blocksize = sb->s_blocksize; + hblock = bdev_hardsect_size(bdev); + if (blocksize < hblock) { + printk(KERN_ERR + "EXT4-fs: blocksize too small for journal device.\n"); + goto out_bdev; + } + + sb_block = EXT4_MIN_BLOCK_SIZE / blocksize; + offset = EXT4_MIN_BLOCK_SIZE % blocksize; + set_blocksize(bdev, blocksize); + if (!(bh = __bread(bdev, sb_block, blocksize))) { + printk(KERN_ERR "EXT4-fs: couldn't read superblock of " + "external journal\n"); + goto out_bdev; + } + + es = (struct ext4_super_block *) (((char *)bh->b_data) + offset); + if ((le16_to_cpu(es->s_magic) != EXT4_SUPER_MAGIC) || + !(le32_to_cpu(es->s_feature_incompat) & + EXT4_FEATURE_INCOMPAT_JOURNAL_DEV)) { + printk(KERN_ERR "EXT4-fs: external journal has " + "bad superblock\n"); + brelse(bh); + goto out_bdev; + } + + if (memcmp(EXT4_SB(sb)->s_es->s_journal_uuid, es->s_uuid, 16)) { + printk(KERN_ERR "EXT4-fs: journal UUID does not match\n"); + brelse(bh); + goto out_bdev; + } + + len = ext4_blocks_count(es); + start = sb_block + 1; + brelse(bh); /* we're done with the superblock */ + + journal = jbd2_journal_init_dev(bdev, sb->s_bdev, + start, len, blocksize); + if (!journal) { + printk(KERN_ERR "EXT4-fs: failed to create device journal\n"); + goto out_bdev; + } + journal->j_private = sb; + ll_rw_block(READ, 1, &journal->j_sb_buffer); + wait_on_buffer(journal->j_sb_buffer); + if (!buffer_uptodate(journal->j_sb_buffer)) { + printk(KERN_ERR "EXT4-fs: I/O error on journal device\n"); + goto out_journal; + } + if (be32_to_cpu(journal->j_superblock->s_nr_users) != 1) { + printk(KERN_ERR "EXT4-fs: External journal has more than one " + "user (unsupported) - %d\n", + be32_to_cpu(journal->j_superblock->s_nr_users)); + goto out_journal; + } + EXT4_SB(sb)->journal_bdev = bdev; + ext4_init_journal_params(sb, journal); + return journal; +out_journal: + jbd2_journal_destroy(journal); +out_bdev: + ext4_blkdev_put(bdev); + return NULL; +} + +static int ext4_load_journal(struct super_block *sb, + struct ext4_super_block *es, + unsigned long journal_devnum) +{ + journal_t *journal; + unsigned int journal_inum = le32_to_cpu(es->s_journal_inum); + dev_t journal_dev; + int err = 0; + int really_read_only; + + if (journal_devnum && + journal_devnum != le32_to_cpu(es->s_journal_dev)) { + printk(KERN_INFO "EXT4-fs: external journal device major/minor " + "numbers have changed\n"); + journal_dev = new_decode_dev(journal_devnum); + } else + journal_dev = new_decode_dev(le32_to_cpu(es->s_journal_dev)); + + really_read_only = bdev_read_only(sb->s_bdev); + + /* + * Are we loading a blank journal or performing recovery after a + * crash? For recovery, we need to check in advance whether we + * can get read-write access to the device. + */ + + if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER)) { + if (sb->s_flags & MS_RDONLY) { + printk(KERN_INFO "EXT4-fs: INFO: recovery " + "required on readonly filesystem.\n"); + if (really_read_only) { + printk(KERN_ERR "EXT4-fs: write access " + "unavailable, cannot proceed.\n"); + return -EROFS; + } + printk (KERN_INFO "EXT4-fs: write access will " + "be enabled during recovery.\n"); + } + } + + if (journal_inum && journal_dev) { + printk(KERN_ERR "EXT4-fs: filesystem has both journal " + "and inode journals!\n"); + return -EINVAL; + } + + if (journal_inum) { + if (!(journal = ext4_get_journal(sb, journal_inum))) + return -EINVAL; + } else { + if (!(journal = ext4_get_dev_journal(sb, journal_dev))) + return -EINVAL; + } + + if (!really_read_only && test_opt(sb, UPDATE_JOURNAL)) { + err = jbd2_journal_update_format(journal); + if (err) { + printk(KERN_ERR "EXT4-fs: error updating journal.\n"); + jbd2_journal_destroy(journal); + return err; + } + } + + if (!EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER)) + err = jbd2_journal_wipe(journal, !really_read_only); + if (!err) + err = jbd2_journal_load(journal); + + if (err) { + printk(KERN_ERR "EXT4-fs: error loading journal.\n"); + jbd2_journal_destroy(journal); + return err; + } + + EXT4_SB(sb)->s_journal = journal; + ext4_clear_journal_err(sb, es); + + if (journal_devnum && + journal_devnum != le32_to_cpu(es->s_journal_dev)) { + es->s_journal_dev = cpu_to_le32(journal_devnum); + sb->s_dirt = 1; + + /* Make sure we flush the recovery flag to disk. */ + ext4_commit_super(sb, es, 1); + } + + return 0; +} + +static int ext4_create_journal(struct super_block * sb, + struct ext4_super_block * es, + unsigned int journal_inum) +{ + journal_t *journal; + + if (sb->s_flags & MS_RDONLY) { + printk(KERN_ERR "EXT4-fs: readonly filesystem when trying to " + "create journal.\n"); + return -EROFS; + } + + if (!(journal = ext4_get_journal(sb, journal_inum))) + return -EINVAL; + + printk(KERN_INFO "EXT4-fs: creating new journal on inode %u\n", + journal_inum); + + if (jbd2_journal_create(journal)) { + printk(KERN_ERR "EXT4-fs: error creating journal.\n"); + jbd2_journal_destroy(journal); + return -EIO; + } + + EXT4_SB(sb)->s_journal = journal; + + ext4_update_dynamic_rev(sb); + EXT4_SET_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER); + EXT4_SET_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_HAS_JOURNAL); + + es->s_journal_inum = cpu_to_le32(journal_inum); + sb->s_dirt = 1; + + /* Make sure we flush the recovery flag to disk. */ + ext4_commit_super(sb, es, 1); + + return 0; +} + +static void ext4_commit_super (struct super_block * sb, + struct ext4_super_block * es, + int sync) +{ + struct buffer_head *sbh = EXT4_SB(sb)->s_sbh; + + if (!sbh) + return; + es->s_wtime = cpu_to_le32(get_seconds()); + ext4_free_blocks_count_set(es, ext4_count_free_blocks(sb)); + es->s_free_inodes_count = cpu_to_le32(ext4_count_free_inodes(sb)); + BUFFER_TRACE(sbh, "marking dirty"); + mark_buffer_dirty(sbh); + if (sync) + sync_dirty_buffer(sbh); +} + + +/* + * Have we just finished recovery? If so, and if we are mounting (or + * remounting) the filesystem readonly, then we will end up with a + * consistent fs on disk. Record that fact. + */ +static void ext4_mark_recovery_complete(struct super_block * sb, + struct ext4_super_block * es) +{ + journal_t *journal = EXT4_SB(sb)->s_journal; + + jbd2_journal_lock_updates(journal); + jbd2_journal_flush(journal); + if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER) && + sb->s_flags & MS_RDONLY) { + EXT4_CLEAR_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER); + sb->s_dirt = 0; + ext4_commit_super(sb, es, 1); + } + jbd2_journal_unlock_updates(journal); +} + +/* + * If we are mounting (or read-write remounting) a filesystem whose journal + * has recorded an error from a previous lifetime, move that error to the + * main filesystem now. + */ +static void ext4_clear_journal_err(struct super_block * sb, + struct ext4_super_block * es) +{ + journal_t *journal; + int j_errno; + const char *errstr; + + journal = EXT4_SB(sb)->s_journal; + + /* + * Now check for any error status which may have been recorded in the + * journal by a prior ext4_error() or ext4_abort() + */ + + j_errno = jbd2_journal_errno(journal); + if (j_errno) { + char nbuf[16]; + + errstr = ext4_decode_error(sb, j_errno, nbuf); + ext4_warning(sb, __FUNCTION__, "Filesystem error recorded " + "from previous mount: %s", errstr); + ext4_warning(sb, __FUNCTION__, "Marking fs in need of " + "filesystem check."); + + EXT4_SB(sb)->s_mount_state |= EXT4_ERROR_FS; + es->s_state |= cpu_to_le16(EXT4_ERROR_FS); + ext4_commit_super (sb, es, 1); + + jbd2_journal_clear_err(journal); + } +} + +/* + * Force the running and committing transactions to commit, + * and wait on the commit. + */ +int ext4_force_commit(struct super_block *sb) +{ + journal_t *journal; + int ret; + + if (sb->s_flags & MS_RDONLY) + return 0; + + journal = EXT4_SB(sb)->s_journal; + sb->s_dirt = 0; + ret = ext4_journal_force_commit(journal); + return ret; +} + +/* + * Ext4 always journals updates to the superblock itself, so we don't + * have to propagate any other updates to the superblock on disk at this + * point. Just start an async writeback to get the buffers on their way + * to the disk. + * + * This implicitly triggers the writebehind on sync(). + */ + +static void ext4_write_super (struct super_block * sb) +{ + if (mutex_trylock(&sb->s_lock) != 0) + BUG(); + sb->s_dirt = 0; +} + +static int ext4_sync_fs(struct super_block *sb, int wait) +{ + tid_t target; + + sb->s_dirt = 0; + if (jbd2_journal_start_commit(EXT4_SB(sb)->s_journal, &target)) { + if (wait) + jbd2_log_wait_commit(EXT4_SB(sb)->s_journal, target); + } + return 0; +} + +/* + * LVM calls this function before a (read-only) snapshot is created. This + * gives us a chance to flush the journal completely and mark the fs clean. + */ +static void ext4_write_super_lockfs(struct super_block *sb) +{ + sb->s_dirt = 0; + + if (!(sb->s_flags & MS_RDONLY)) { + journal_t *journal = EXT4_SB(sb)->s_journal; + + /* Now we set up the journal barrier. */ + jbd2_journal_lock_updates(journal); + jbd2_journal_flush(journal); + + /* Journal blocked and flushed, clear needs_recovery flag. */ + EXT4_CLEAR_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER); + ext4_commit_super(sb, EXT4_SB(sb)->s_es, 1); + } +} + +/* + * Called by LVM after the snapshot is done. We need to reset the RECOVER + * flag here, even though the filesystem is not technically dirty yet. + */ +static void ext4_unlockfs(struct super_block *sb) +{ + if (!(sb->s_flags & MS_RDONLY)) { + lock_super(sb); + /* Reser the needs_recovery flag before the fs is unlocked. */ + EXT4_SET_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_RECOVER); + ext4_commit_super(sb, EXT4_SB(sb)->s_es, 1); + unlock_super(sb); + jbd2_journal_unlock_updates(EXT4_SB(sb)->s_journal); + } +} + +static int ext4_remount (struct super_block * sb, int * flags, char * data) +{ + struct ext4_super_block * es; + struct ext4_sb_info *sbi = EXT4_SB(sb); + ext4_fsblk_t n_blocks_count = 0; + unsigned long old_sb_flags; + struct ext4_mount_options old_opts; + int err; +#ifdef CONFIG_QUOTA + int i; +#endif + + /* Store the original options */ + old_sb_flags = sb->s_flags; + old_opts.s_mount_opt = sbi->s_mount_opt; + old_opts.s_resuid = sbi->s_resuid; + old_opts.s_resgid = sbi->s_resgid; + old_opts.s_commit_interval = sbi->s_commit_interval; +#ifdef CONFIG_QUOTA + old_opts.s_jquota_fmt = sbi->s_jquota_fmt; + for (i = 0; i < MAXQUOTAS; i++) + old_opts.s_qf_names[i] = sbi->s_qf_names[i]; +#endif + + /* + * Allow the "check" option to be passed as a remount option. + */ + if (!parse_options(data, sb, NULL, NULL, &n_blocks_count, 1)) { + err = -EINVAL; + goto restore_opts; + } + + if (sbi->s_mount_opt & EXT4_MOUNT_ABORT) + ext4_abort(sb, __FUNCTION__, "Abort forced by user"); + + sb->s_flags = (sb->s_flags & ~MS_POSIXACL) | + ((sbi->s_mount_opt & EXT4_MOUNT_POSIX_ACL) ? MS_POSIXACL : 0); + + es = sbi->s_es; + + ext4_init_journal_params(sb, sbi->s_journal); + + if ((*flags & MS_RDONLY) != (sb->s_flags & MS_RDONLY) || + n_blocks_count > ext4_blocks_count(es)) { + if (sbi->s_mount_opt & EXT4_MOUNT_ABORT) { + err = -EROFS; + goto restore_opts; + } + + if (*flags & MS_RDONLY) { + /* + * First of all, the unconditional stuff we have to do + * to disable replay of the journal when we next remount + */ + sb->s_flags |= MS_RDONLY; + + /* + * OK, test if we are remounting a valid rw partition + * readonly, and if so set the rdonly flag and then + * mark the partition as valid again. + */ + if (!(es->s_state & cpu_to_le16(EXT4_VALID_FS)) && + (sbi->s_mount_state & EXT4_VALID_FS)) + es->s_state = cpu_to_le16(sbi->s_mount_state); + + ext4_mark_recovery_complete(sb, es); + } else { + __le32 ret; + if ((ret = EXT4_HAS_RO_COMPAT_FEATURE(sb, + ~EXT4_FEATURE_RO_COMPAT_SUPP))) { + printk(KERN_WARNING "EXT4-fs: %s: couldn't " + "remount RDWR because of unsupported " + "optional features (%x).\n", + sb->s_id, le32_to_cpu(ret)); + err = -EROFS; + goto restore_opts; + } + /* + * Mounting a RDONLY partition read-write, so reread + * and store the current valid flag. (It may have + * been changed by e2fsck since we originally mounted + * the partition.) + */ + ext4_clear_journal_err(sb, es); + sbi->s_mount_state = le16_to_cpu(es->s_state); + if ((err = ext4_group_extend(sb, es, n_blocks_count))) + goto restore_opts; + if (!ext4_setup_super (sb, es, 0)) + sb->s_flags &= ~MS_RDONLY; + } + } +#ifdef CONFIG_QUOTA + /* Release old quota file names */ + for (i = 0; i < MAXQUOTAS; i++) + if (old_opts.s_qf_names[i] && + old_opts.s_qf_names[i] != sbi->s_qf_names[i]) + kfree(old_opts.s_qf_names[i]); +#endif + return 0; +restore_opts: + sb->s_flags = old_sb_flags; + sbi->s_mount_opt = old_opts.s_mount_opt; + sbi->s_resuid = old_opts.s_resuid; + sbi->s_resgid = old_opts.s_resgid; + sbi->s_commit_interval = old_opts.s_commit_interval; +#ifdef CONFIG_QUOTA + sbi->s_jquota_fmt = old_opts.s_jquota_fmt; + for (i = 0; i < MAXQUOTAS; i++) { + if (sbi->s_qf_names[i] && + old_opts.s_qf_names[i] != sbi->s_qf_names[i]) + kfree(sbi->s_qf_names[i]); + sbi->s_qf_names[i] = old_opts.s_qf_names[i]; + } +#endif + return err; +} + +static int ext4_statfs (struct dentry * dentry, struct kstatfs * buf) +{ + struct super_block *sb = dentry->d_sb; + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct ext4_super_block *es = sbi->s_es; + ext4_fsblk_t overhead; + int i; + + if (test_opt (sb, MINIX_DF)) + overhead = 0; + else { + unsigned long ngroups; + ngroups = EXT4_SB(sb)->s_groups_count; + smp_rmb(); + + /* + * Compute the overhead (FS structures) + */ + + /* + * All of the blocks before first_data_block are + * overhead + */ + overhead = le32_to_cpu(es->s_first_data_block); + + /* + * Add the overhead attributed to the superblock and + * block group descriptors. If the sparse superblocks + * feature is turned on, then not all groups have this. + */ + for (i = 0; i < ngroups; i++) { + overhead += ext4_bg_has_super(sb, i) + + ext4_bg_num_gdb(sb, i); + cond_resched(); + } + + /* + * Every block group has an inode bitmap, a block + * bitmap, and an inode table. + */ + overhead += (ngroups * (2 + EXT4_SB(sb)->s_itb_per_group)); + } + + buf->f_type = EXT4_SUPER_MAGIC; + buf->f_bsize = sb->s_blocksize; + buf->f_blocks = ext4_blocks_count(es) - overhead; + buf->f_bfree = percpu_counter_sum(&sbi->s_freeblocks_counter); + buf->f_bavail = buf->f_bfree - ext4_r_blocks_count(es); + if (buf->f_bfree < ext4_r_blocks_count(es)) + buf->f_bavail = 0; + buf->f_files = le32_to_cpu(es->s_inodes_count); + buf->f_ffree = percpu_counter_sum(&sbi->s_freeinodes_counter); + buf->f_namelen = EXT4_NAME_LEN; + return 0; +} + +/* Helper function for writing quotas on sync - we need to start transaction before quota file + * is locked for write. Otherwise the are possible deadlocks: + * Process 1 Process 2 + * ext4_create() quota_sync() + * jbd2_journal_start() write_dquot() + * DQUOT_INIT() down(dqio_mutex) + * down(dqio_mutex) jbd2_journal_start() + * + */ + +#ifdef CONFIG_QUOTA + +static inline struct inode *dquot_to_inode(struct dquot *dquot) +{ + return sb_dqopt(dquot->dq_sb)->files[dquot->dq_type]; +} + +static int ext4_dquot_initialize(struct inode *inode, int type) +{ + handle_t *handle; + int ret, err; + + /* We may create quota structure so we need to reserve enough blocks */ + handle = ext4_journal_start(inode, 2*EXT4_QUOTA_INIT_BLOCKS(inode->i_sb)); + if (IS_ERR(handle)) + return PTR_ERR(handle); + ret = dquot_initialize(inode, type); + err = ext4_journal_stop(handle); + if (!ret) + ret = err; + return ret; +} + +static int ext4_dquot_drop(struct inode *inode) +{ + handle_t *handle; + int ret, err; + + /* We may delete quota structure so we need to reserve enough blocks */ + handle = ext4_journal_start(inode, 2*EXT4_QUOTA_DEL_BLOCKS(inode->i_sb)); + if (IS_ERR(handle)) + return PTR_ERR(handle); + ret = dquot_drop(inode); + err = ext4_journal_stop(handle); + if (!ret) + ret = err; + return ret; +} + +static int ext4_write_dquot(struct dquot *dquot) +{ + int ret, err; + handle_t *handle; + struct inode *inode; + + inode = dquot_to_inode(dquot); + handle = ext4_journal_start(inode, + EXT4_QUOTA_TRANS_BLOCKS(dquot->dq_sb)); + if (IS_ERR(handle)) + return PTR_ERR(handle); + ret = dquot_commit(dquot); + err = ext4_journal_stop(handle); + if (!ret) + ret = err; + return ret; +} + +static int ext4_acquire_dquot(struct dquot *dquot) +{ + int ret, err; + handle_t *handle; + + handle = ext4_journal_start(dquot_to_inode(dquot), + EXT4_QUOTA_INIT_BLOCKS(dquot->dq_sb)); + if (IS_ERR(handle)) + return PTR_ERR(handle); + ret = dquot_acquire(dquot); + err = ext4_journal_stop(handle); + if (!ret) + ret = err; + return ret; +} + +static int ext4_release_dquot(struct dquot *dquot) +{ + int ret, err; + handle_t *handle; + + handle = ext4_journal_start(dquot_to_inode(dquot), + EXT4_QUOTA_DEL_BLOCKS(dquot->dq_sb)); + if (IS_ERR(handle)) + return PTR_ERR(handle); + ret = dquot_release(dquot); + err = ext4_journal_stop(handle); + if (!ret) + ret = err; + return ret; +} + +static int ext4_mark_dquot_dirty(struct dquot *dquot) +{ + /* Are we journalling quotas? */ + if (EXT4_SB(dquot->dq_sb)->s_qf_names[USRQUOTA] || + EXT4_SB(dquot->dq_sb)->s_qf_names[GRPQUOTA]) { + dquot_mark_dquot_dirty(dquot); + return ext4_write_dquot(dquot); + } else { + return dquot_mark_dquot_dirty(dquot); + } +} + +static int ext4_write_info(struct super_block *sb, int type) +{ + int ret, err; + handle_t *handle; + + /* Data block + inode block */ + handle = ext4_journal_start(sb->s_root->d_inode, 2); + if (IS_ERR(handle)) + return PTR_ERR(handle); + ret = dquot_commit_info(sb, type); + err = ext4_journal_stop(handle); + if (!ret) + ret = err; + return ret; +} + +/* + * Turn on quotas during mount time - we need to find + * the quota file and such... + */ +static int ext4_quota_on_mount(struct super_block *sb, int type) +{ + return vfs_quota_on_mount(sb, EXT4_SB(sb)->s_qf_names[type], + EXT4_SB(sb)->s_jquota_fmt, type); +} + +/* + * Standard function to be called on quota_on + */ +static int ext4_quota_on(struct super_block *sb, int type, int format_id, + char *path) +{ + int err; + struct nameidata nd; + + if (!test_opt(sb, QUOTA)) + return -EINVAL; + /* Not journalling quota? */ + if (!EXT4_SB(sb)->s_qf_names[USRQUOTA] && + !EXT4_SB(sb)->s_qf_names[GRPQUOTA]) + return vfs_quota_on(sb, type, format_id, path); + err = path_lookup(path, LOOKUP_FOLLOW, &nd); + if (err) + return err; + /* Quotafile not on the same filesystem? */ + if (nd.mnt->mnt_sb != sb) { + path_release(&nd); + return -EXDEV; + } + /* Quotafile not of fs root? */ + if (nd.dentry->d_parent->d_inode != sb->s_root->d_inode) + printk(KERN_WARNING + "EXT4-fs: Quota file not on filesystem root. " + "Journalled quota will not work.\n"); + path_release(&nd); + return vfs_quota_on(sb, type, format_id, path); +} + +/* Read data from quotafile - avoid pagecache and such because we cannot afford + * acquiring the locks... As quota files are never truncated and quota code + * itself serializes the operations (and noone else should touch the files) + * we don't have to be afraid of races */ +static ssize_t ext4_quota_read(struct super_block *sb, int type, char *data, + size_t len, loff_t off) +{ + struct inode *inode = sb_dqopt(sb)->files[type]; + sector_t blk = off >> EXT4_BLOCK_SIZE_BITS(sb); + int err = 0; + int offset = off & (sb->s_blocksize - 1); + int tocopy; + size_t toread; + struct buffer_head *bh; + loff_t i_size = i_size_read(inode); + + if (off > i_size) + return 0; + if (off+len > i_size) + len = i_size-off; + toread = len; + while (toread > 0) { + tocopy = sb->s_blocksize - offset < toread ? + sb->s_blocksize - offset : toread; + bh = ext4_bread(NULL, inode, blk, 0, &err); + if (err) + return err; + if (!bh) /* A hole? */ + memset(data, 0, tocopy); + else + memcpy(data, bh->b_data+offset, tocopy); + brelse(bh); + offset = 0; + toread -= tocopy; + data += tocopy; + blk++; + } + return len; +} + +/* Write to quotafile (we know the transaction is already started and has + * enough credits) */ +static ssize_t ext4_quota_write(struct super_block *sb, int type, + const char *data, size_t len, loff_t off) +{ + struct inode *inode = sb_dqopt(sb)->files[type]; + sector_t blk = off >> EXT4_BLOCK_SIZE_BITS(sb); + int err = 0; + int offset = off & (sb->s_blocksize - 1); + int tocopy; + int journal_quota = EXT4_SB(sb)->s_qf_names[type] != NULL; + size_t towrite = len; + struct buffer_head *bh; + handle_t *handle = journal_current_handle(); + + mutex_lock_nested(&inode->i_mutex, I_MUTEX_QUOTA); + while (towrite > 0) { + tocopy = sb->s_blocksize - offset < towrite ? + sb->s_blocksize - offset : towrite; + bh = ext4_bread(handle, inode, blk, 1, &err); + if (!bh) + goto out; + if (journal_quota) { + err = ext4_journal_get_write_access(handle, bh); + if (err) { + brelse(bh); + goto out; + } + } + lock_buffer(bh); + memcpy(bh->b_data+offset, data, tocopy); + flush_dcache_page(bh->b_page); + unlock_buffer(bh); + if (journal_quota) + err = ext4_journal_dirty_metadata(handle, bh); + else { + /* Always do at least ordered writes for quotas */ + err = ext4_journal_dirty_data(handle, bh); + mark_buffer_dirty(bh); + } + brelse(bh); + if (err) + goto out; + offset = 0; + towrite -= tocopy; + data += tocopy; + blk++; + } +out: + if (len == towrite) + return err; + if (inode->i_size < off+len-towrite) { + i_size_write(inode, off+len-towrite); + EXT4_I(inode)->i_disksize = inode->i_size; + } + inode->i_version++; + inode->i_mtime = inode->i_ctime = CURRENT_TIME; + ext4_mark_inode_dirty(handle, inode); + mutex_unlock(&inode->i_mutex); + return len - towrite; +} + +#endif + +static int ext4_get_sb(struct file_system_type *fs_type, + int flags, const char *dev_name, void *data, struct vfsmount *mnt) +{ + return get_sb_bdev(fs_type, flags, dev_name, data, ext4_fill_super, mnt); +} + +static struct file_system_type ext4dev_fs_type = { + .owner = THIS_MODULE, + .name = "ext4dev", + .get_sb = ext4_get_sb, + .kill_sb = kill_block_super, + .fs_flags = FS_REQUIRES_DEV, +}; + +static int __init init_ext4_fs(void) +{ + int err = init_ext4_xattr(); + if (err) + return err; + err = init_inodecache(); + if (err) + goto out1; + err = register_filesystem(&ext4dev_fs_type); + if (err) + goto out; + return 0; +out: + destroy_inodecache(); +out1: + exit_ext4_xattr(); + return err; +} + +static void __exit exit_ext4_fs(void) +{ + unregister_filesystem(&ext4dev_fs_type); + destroy_inodecache(); + exit_ext4_xattr(); +} + +MODULE_AUTHOR("Remy Card, Stephen Tweedie, Andrew Morton, Andreas Dilger, Theodore Ts'o and others"); +MODULE_DESCRIPTION("Fourth Extended Filesystem with extents"); +MODULE_LICENSE("GPL"); +module_init(init_ext4_fs) +module_exit(exit_ext4_fs) diff --git a/fs/ext4/symlink.c b/fs/ext4/symlink.c new file mode 100644 index 000000000000..fcf527286d75 --- /dev/null +++ b/fs/ext4/symlink.c @@ -0,0 +1,54 @@ +/* + * linux/fs/ext4/symlink.c + * + * Only fast symlinks left here - the rest is done by generic code. AV, 1999 + * + * Copyright (C) 1992, 1993, 1994, 1995 + * Remy Card (card@masi.ibp.fr) + * Laboratoire MASI - Institut Blaise Pascal + * Universite Pierre et Marie Curie (Paris VI) + * + * from + * + * linux/fs/minix/symlink.c + * + * Copyright (C) 1991, 1992 Linus Torvalds + * + * ext4 symlink handling code + */ + +#include <linux/fs.h> +#include <linux/jbd2.h> +#include <linux/ext4_fs.h> +#include <linux/namei.h> +#include "xattr.h" + +static void * ext4_follow_link(struct dentry *dentry, struct nameidata *nd) +{ + struct ext4_inode_info *ei = EXT4_I(dentry->d_inode); + nd_set_link(nd, (char*)ei->i_data); + return NULL; +} + +struct inode_operations ext4_symlink_inode_operations = { + .readlink = generic_readlink, + .follow_link = page_follow_link_light, + .put_link = page_put_link, +#ifdef CONFIG_EXT4DEV_FS_XATTR + .setxattr = generic_setxattr, + .getxattr = generic_getxattr, + .listxattr = ext4_listxattr, + .removexattr = generic_removexattr, +#endif +}; + +struct inode_operations ext4_fast_symlink_inode_operations = { + .readlink = generic_readlink, + .follow_link = ext4_follow_link, +#ifdef CONFIG_EXT4DEV_FS_XATTR + .setxattr = generic_setxattr, + .getxattr = generic_getxattr, + .listxattr = ext4_listxattr, + .removexattr = generic_removexattr, +#endif +}; diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c new file mode 100644 index 000000000000..63233cd946a7 --- /dev/null +++ b/fs/ext4/xattr.c @@ -0,0 +1,1317 @@ +/* + * linux/fs/ext4/xattr.c + * + * Copyright (C) 2001-2003 Andreas Gruenbacher, <agruen@suse.de> + * + * Fix by Harrison Xing <harrison@mountainviewdata.com>. + * Ext4 code with a lot of help from Eric Jarman <ejarman@acm.org>. + * Extended attributes for symlinks and special files added per + * suggestion of Luka Renko <luka.renko@hermes.si>. + * xattr consolidation Copyright (c) 2004 James Morris <jmorris@redhat.com>, + * Red Hat Inc. + * ea-in-inode support by Alex Tomas <alex@clusterfs.com> aka bzzz + * and Andreas Gruenbacher <agruen@suse.de>. + */ + +/* + * Extended attributes are stored directly in inodes (on file systems with + * inodes bigger than 128 bytes) and on additional disk blocks. The i_file_acl + * field contains the block number if an inode uses an additional block. All + * attributes must fit in the inode and one additional block. Blocks that + * contain the identical set of attributes may be shared among several inodes. + * Identical blocks are detected by keeping a cache of blocks that have + * recently been accessed. + * + * The attributes in inodes and on blocks have a different header; the entries + * are stored in the same format: + * + * +------------------+ + * | header | + * | entry 1 | | + * | entry 2 | | growing downwards + * | entry 3 | v + * | four null bytes | + * | . . . | + * | value 1 | ^ + * | value 3 | | growing upwards + * | value 2 | | + * +------------------+ + * + * The header is followed by multiple entry descriptors. In disk blocks, the + * entry descriptors are kept sorted. In inodes, they are unsorted. The + * attribute values are aligned to the end of the block in no specific order. + * + * Locking strategy + * ---------------- + * EXT4_I(inode)->i_file_acl is protected by EXT4_I(inode)->xattr_sem. + * EA blocks are only changed if they are exclusive to an inode, so + * holding xattr_sem also means that nothing but the EA block's reference + * count can change. Multiple writers to the same block are synchronized + * by the buffer lock. + */ + +#include <linux/init.h> +#include <linux/fs.h> +#include <linux/slab.h> +#include <linux/ext4_jbd2.h> +#include <linux/ext4_fs.h> +#include <linux/mbcache.h> +#include <linux/quotaops.h> +#include <linux/rwsem.h> +#include "xattr.h" +#include "acl.h" + +#define BHDR(bh) ((struct ext4_xattr_header *)((bh)->b_data)) +#define ENTRY(ptr) ((struct ext4_xattr_entry *)(ptr)) +#define BFIRST(bh) ENTRY(BHDR(bh)+1) +#define IS_LAST_ENTRY(entry) (*(__u32 *)(entry) == 0) + +#define IHDR(inode, raw_inode) \ + ((struct ext4_xattr_ibody_header *) \ + ((void *)raw_inode + \ + EXT4_GOOD_OLD_INODE_SIZE + \ + EXT4_I(inode)->i_extra_isize)) +#define IFIRST(hdr) ((struct ext4_xattr_entry *)((hdr)+1)) + +#ifdef EXT4_XATTR_DEBUG +# define ea_idebug(inode, f...) do { \ + printk(KERN_DEBUG "inode %s:%lu: ", \ + inode->i_sb->s_id, inode->i_ino); \ + printk(f); \ + printk("\n"); \ + } while (0) +# define ea_bdebug(bh, f...) do { \ + char b[BDEVNAME_SIZE]; \ + printk(KERN_DEBUG "block %s:%lu: ", \ + bdevname(bh->b_bdev, b), \ + (unsigned long) bh->b_blocknr); \ + printk(f); \ + printk("\n"); \ + } while (0) +#else +# define ea_idebug(f...) +# define ea_bdebug(f...) +#endif + +static void ext4_xattr_cache_insert(struct buffer_head *); +static struct buffer_head *ext4_xattr_cache_find(struct inode *, + struct ext4_xattr_header *, + struct mb_cache_entry **); +static void ext4_xattr_rehash(struct ext4_xattr_header *, + struct ext4_xattr_entry *); + +static struct mb_cache *ext4_xattr_cache; + +static struct xattr_handler *ext4_xattr_handler_map[] = { + [EXT4_XATTR_INDEX_USER] = &ext4_xattr_user_handler, +#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL + [EXT4_XATTR_INDEX_POSIX_ACL_ACCESS] = &ext4_xattr_acl_access_handler, + [EXT4_XATTR_INDEX_POSIX_ACL_DEFAULT] = &ext4_xattr_acl_default_handler, +#endif + [EXT4_XATTR_INDEX_TRUSTED] = &ext4_xattr_trusted_handler, +#ifdef CONFIG_EXT4DEV_FS_SECURITY + [EXT4_XATTR_INDEX_SECURITY] = &ext4_xattr_security_handler, +#endif +}; + +struct xattr_handler *ext4_xattr_handlers[] = { + &ext4_xattr_user_handler, + &ext4_xattr_trusted_handler, +#ifdef CONFIG_EXT4DEV_FS_POSIX_ACL + &ext4_xattr_acl_access_handler, + &ext4_xattr_acl_default_handler, +#endif +#ifdef CONFIG_EXT4DEV_FS_SECURITY + &ext4_xattr_security_handler, +#endif + NULL +}; + +static inline struct xattr_handler * +ext4_xattr_handler(int name_index) +{ + struct xattr_handler *handler = NULL; + + if (name_index > 0 && name_index < ARRAY_SIZE(ext4_xattr_handler_map)) + handler = ext4_xattr_handler_map[name_index]; + return handler; +} + +/* + * Inode operation listxattr() + * + * dentry->d_inode->i_mutex: don't care + */ +ssize_t +ext4_listxattr(struct dentry *dentry, char *buffer, size_t size) +{ + return ext4_xattr_list(dentry->d_inode, buffer, size); +} + +static int +ext4_xattr_check_names(struct ext4_xattr_entry *entry, void *end) +{ + while (!IS_LAST_ENTRY(entry)) { + struct ext4_xattr_entry *next = EXT4_XATTR_NEXT(entry); + if ((void *)next >= end) + return -EIO; + entry = next; + } + return 0; +} + +static inline int +ext4_xattr_check_block(struct buffer_head *bh) +{ + int error; + + if (BHDR(bh)->h_magic != cpu_to_le32(EXT4_XATTR_MAGIC) || + BHDR(bh)->h_blocks != cpu_to_le32(1)) + return -EIO; + error = ext4_xattr_check_names(BFIRST(bh), bh->b_data + bh->b_size); + return error; +} + +static inline int +ext4_xattr_check_entry(struct ext4_xattr_entry *entry, size_t size) +{ + size_t value_size = le32_to_cpu(entry->e_value_size); + + if (entry->e_value_block != 0 || value_size > size || + le16_to_cpu(entry->e_value_offs) + value_size > size) + return -EIO; + return 0; +} + +static int +ext4_xattr_find_entry(struct ext4_xattr_entry **pentry, int name_index, + const char *name, size_t size, int sorted) +{ + struct ext4_xattr_entry *entry; + size_t name_len; + int cmp = 1; + + if (name == NULL) + return -EINVAL; + name_len = strlen(name); + entry = *pentry; + for (; !IS_LAST_ENTRY(entry); entry = EXT4_XATTR_NEXT(entry)) { + cmp = name_index - entry->e_name_index; + if (!cmp) + cmp = name_len - entry->e_name_len; + if (!cmp) + cmp = memcmp(name, entry->e_name, name_len); + if (cmp <= 0 && (sorted || cmp == 0)) + break; + } + *pentry = entry; + if (!cmp && ext4_xattr_check_entry(entry, size)) + return -EIO; + return cmp ? -ENODATA : 0; +} + +static int +ext4_xattr_block_get(struct inode *inode, int name_index, const char *name, + void *buffer, size_t buffer_size) +{ + struct buffer_head *bh = NULL; + struct ext4_xattr_entry *entry; + size_t size; + int error; + + ea_idebug(inode, "name=%d.%s, buffer=%p, buffer_size=%ld", + name_index, name, buffer, (long)buffer_size); + + error = -ENODATA; + if (!EXT4_I(inode)->i_file_acl) + goto cleanup; + ea_idebug(inode, "reading block %u", EXT4_I(inode)->i_file_acl); + bh = sb_bread(inode->i_sb, EXT4_I(inode)->i_file_acl); + if (!bh) + goto cleanup; + ea_bdebug(bh, "b_count=%d, refcount=%d", + atomic_read(&(bh->b_count)), le32_to_cpu(BHDR(bh)->h_refcount)); + if (ext4_xattr_check_block(bh)) { +bad_block: ext4_error(inode->i_sb, __FUNCTION__, + "inode %lu: bad block %llu", inode->i_ino, + EXT4_I(inode)->i_file_acl); + error = -EIO; + goto cleanup; + } + ext4_xattr_cache_insert(bh); + entry = BFIRST(bh); + error = ext4_xattr_find_entry(&entry, name_index, name, bh->b_size, 1); + if (error == -EIO) + goto bad_block; + if (error) + goto cleanup; + size = le32_to_cpu(entry->e_value_size); + if (buffer) { + error = -ERANGE; + if (size > buffer_size) + goto cleanup; + memcpy(buffer, bh->b_data + le16_to_cpu(entry->e_value_offs), + size); + } + error = size; + +cleanup: + brelse(bh); + return error; +} + +static int +ext4_xattr_ibody_get(struct inode *inode, int name_index, const char *name, + void *buffer, size_t buffer_size) +{ + struct ext4_xattr_ibody_header *header; + struct ext4_xattr_entry *entry; + struct ext4_inode *raw_inode; + struct ext4_iloc iloc; + size_t size; + void *end; + int error; + + if (!(EXT4_I(inode)->i_state & EXT4_STATE_XATTR)) + return -ENODATA; + error = ext4_get_inode_loc(inode, &iloc); + if (error) + return error; + raw_inode = ext4_raw_inode(&iloc); + header = IHDR(inode, raw_inode); + entry = IFIRST(header); + end = (void *)raw_inode + EXT4_SB(inode->i_sb)->s_inode_size; + error = ext4_xattr_check_names(entry, end); + if (error) + goto cleanup; + error = ext4_xattr_find_entry(&entry, name_index, name, + end - (void *)entry, 0); + if (error) + goto cleanup; + size = le32_to_cpu(entry->e_value_size); + if (buffer) { + error = -ERANGE; + if (size > buffer_size) + goto cleanup; + memcpy(buffer, (void *)IFIRST(header) + + le16_to_cpu(entry->e_value_offs), size); + } + error = size; + +cleanup: + brelse(iloc.bh); + return error; +} + +/* + * ext4_xattr_get() + * + * Copy an extended attribute into the buffer + * provided, or compute the buffer size required. + * Buffer is NULL to compute the size of the buffer required. + * + * Returns a negative error number on failure, or the number of bytes + * used / required on success. + */ +int +ext4_xattr_get(struct inode *inode, int name_index, const char *name, + void *buffer, size_t buffer_size) +{ + int error; + + down_read(&EXT4_I(inode)->xattr_sem); + error = ext4_xattr_ibody_get(inode, name_index, name, buffer, + buffer_size); + if (error == -ENODATA) + error = ext4_xattr_block_get(inode, name_index, name, buffer, + buffer_size); + up_read(&EXT4_I(inode)->xattr_sem); + return error; +} + +static int +ext4_xattr_list_entries(struct inode *inode, struct ext4_xattr_entry *entry, + char *buffer, size_t buffer_size) +{ + size_t rest = buffer_size; + + for (; !IS_LAST_ENTRY(entry); entry = EXT4_XATTR_NEXT(entry)) { + struct xattr_handler *handler = + ext4_xattr_handler(entry->e_name_index); + + if (handler) { + size_t size = handler->list(inode, buffer, rest, + entry->e_name, + entry->e_name_len); + if (buffer) { + if (size > rest) + return -ERANGE; + buffer += size; + } + rest -= size; + } + } + return buffer_size - rest; +} + +static int +ext4_xattr_block_list(struct inode *inode, char *buffer, size_t buffer_size) +{ + struct buffer_head *bh = NULL; + int error; + + ea_idebug(inode, "buffer=%p, buffer_size=%ld", + buffer, (long)buffer_size); + + error = 0; + if (!EXT4_I(inode)->i_file_acl) + goto cleanup; + ea_idebug(inode, "reading block %u", EXT4_I(inode)->i_file_acl); + bh = sb_bread(inode->i_sb, EXT4_I(inode)->i_file_acl); + error = -EIO; + if (!bh) + goto cleanup; + ea_bdebug(bh, "b_count=%d, refcount=%d", + atomic_read(&(bh->b_count)), le32_to_cpu(BHDR(bh)->h_refcount)); + if (ext4_xattr_check_block(bh)) { + ext4_error(inode->i_sb, __FUNCTION__, + "inode %lu: bad block %llu", inode->i_ino, + EXT4_I(inode)->i_file_acl); + error = -EIO; + goto cleanup; + } + ext4_xattr_cache_insert(bh); + error = ext4_xattr_list_entries(inode, BFIRST(bh), buffer, buffer_size); + +cleanup: + brelse(bh); + + return error; +} + +static int +ext4_xattr_ibody_list(struct inode *inode, char *buffer, size_t buffer_size) +{ + struct ext4_xattr_ibody_header *header; + struct ext4_inode *raw_inode; + struct ext4_iloc iloc; + void *end; + int error; + + if (!(EXT4_I(inode)->i_state & EXT4_STATE_XATTR)) + return 0; + error = ext4_get_inode_loc(inode, &iloc); + if (error) + return error; + raw_inode = ext4_raw_inode(&iloc); + header = IHDR(inode, raw_inode); + end = (void *)raw_inode + EXT4_SB(inode->i_sb)->s_inode_size; + error = ext4_xattr_check_names(IFIRST(header), end); + if (error) + goto cleanup; + error = ext4_xattr_list_entries(inode, IFIRST(header), + buffer, buffer_size); + +cleanup: + brelse(iloc.bh); + return error; +} + +/* + * ext4_xattr_list() + * + * Copy a list of attribute names into the buffer + * provided, or compute the buffer size required. + * Buffer is NULL to compute the size of the buffer required. + * + * Returns a negative error number on failure, or the number of bytes + * used / required on success. + */ +int +ext4_xattr_list(struct inode *inode, char *buffer, size_t buffer_size) +{ + int i_error, b_error; + + down_read(&EXT4_I(inode)->xattr_sem); + i_error = ext4_xattr_ibody_list(inode, buffer, buffer_size); + if (i_error < 0) { + b_error = 0; + } else { + if (buffer) { + buffer += i_error; + buffer_size -= i_error; + } + b_error = ext4_xattr_block_list(inode, buffer, buffer_size); + if (b_error < 0) + i_error = 0; + } + up_read(&EXT4_I(inode)->xattr_sem); + return i_error + b_error; +} + +/* + * If the EXT4_FEATURE_COMPAT_EXT_ATTR feature of this file system is + * not set, set it. + */ +static void ext4_xattr_update_super_block(handle_t *handle, + struct super_block *sb) +{ + if (EXT4_HAS_COMPAT_FEATURE(sb, EXT4_FEATURE_COMPAT_EXT_ATTR)) + return; + + lock_super(sb); + if (ext4_journal_get_write_access(handle, EXT4_SB(sb)->s_sbh) == 0) { + EXT4_SB(sb)->s_es->s_feature_compat |= + cpu_to_le32(EXT4_FEATURE_COMPAT_EXT_ATTR); + sb->s_dirt = 1; + ext4_journal_dirty_metadata(handle, EXT4_SB(sb)->s_sbh); + } + unlock_super(sb); +} + +/* + * Release the xattr block BH: If the reference count is > 1, decrement + * it; otherwise free the block. + */ +static void +ext4_xattr_release_block(handle_t *handle, struct inode *inode, + struct buffer_head *bh) +{ + struct mb_cache_entry *ce = NULL; + + ce = mb_cache_entry_get(ext4_xattr_cache, bh->b_bdev, bh->b_blocknr); + if (BHDR(bh)->h_refcount == cpu_to_le32(1)) { + ea_bdebug(bh, "refcount now=0; freeing"); + if (ce) + mb_cache_entry_free(ce); + ext4_free_blocks(handle, inode, bh->b_blocknr, 1); + get_bh(bh); + ext4_forget(handle, 1, inode, bh, bh->b_blocknr); + } else { + if (ext4_journal_get_write_access(handle, bh) == 0) { + lock_buffer(bh); + BHDR(bh)->h_refcount = cpu_to_le32( + le32_to_cpu(BHDR(bh)->h_refcount) - 1); + ext4_journal_dirty_metadata(handle, bh); + if (IS_SYNC(inode)) + handle->h_sync = 1; + DQUOT_FREE_BLOCK(inode, 1); + unlock_buffer(bh); + ea_bdebug(bh, "refcount now=%d; releasing", + le32_to_cpu(BHDR(bh)->h_refcount)); + } + if (ce) + mb_cache_entry_release(ce); + } +} + +struct ext4_xattr_info { + int name_index; + const char *name; + const void *value; + size_t value_len; +}; + +struct ext4_xattr_search { + struct ext4_xattr_entry *first; + void *base; + void *end; + struct ext4_xattr_entry *here; + int not_found; +}; + +static int +ext4_xattr_set_entry(struct ext4_xattr_info *i, struct ext4_xattr_search *s) +{ + struct ext4_xattr_entry *last; + size_t free, min_offs = s->end - s->base, name_len = strlen(i->name); + + /* Compute min_offs and last. */ + last = s->first; + for (; !IS_LAST_ENTRY(last); last = EXT4_XATTR_NEXT(last)) { + if (!last->e_value_block && last->e_value_size) { + size_t offs = le16_to_cpu(last->e_value_offs); + if (offs < min_offs) + min_offs = offs; + } + } + free = min_offs - ((void *)last - s->base) - sizeof(__u32); + if (!s->not_found) { + if (!s->here->e_value_block && s->here->e_value_size) { + size_t size = le32_to_cpu(s->here->e_value_size); + free += EXT4_XATTR_SIZE(size); + } + free += EXT4_XATTR_LEN(name_len); + } + if (i->value) { + if (free < EXT4_XATTR_SIZE(i->value_len) || + free < EXT4_XATTR_LEN(name_len) + + EXT4_XATTR_SIZE(i->value_len)) + return -ENOSPC; + } + + if (i->value && s->not_found) { + /* Insert the new name. */ + size_t size = EXT4_XATTR_LEN(name_len); + size_t rest = (void *)last - (void *)s->here + sizeof(__u32); + memmove((void *)s->here + size, s->here, rest); + memset(s->here, 0, size); + s->here->e_name_index = i->name_index; + s->here->e_name_len = name_len; + memcpy(s->here->e_name, i->name, name_len); + } else { + if (!s->here->e_value_block && s->here->e_value_size) { + void *first_val = s->base + min_offs; + size_t offs = le16_to_cpu(s->here->e_value_offs); + void *val = s->base + offs; + size_t size = EXT4_XATTR_SIZE( + le32_to_cpu(s->here->e_value_size)); + + if (i->value && size == EXT4_XATTR_SIZE(i->value_len)) { + /* The old and the new value have the same + size. Just replace. */ + s->here->e_value_size = + cpu_to_le32(i->value_len); + memset(val + size - EXT4_XATTR_PAD, 0, + EXT4_XATTR_PAD); /* Clear pad bytes. */ + memcpy(val, i->value, i->value_len); + return 0; + } + + /* Remove the old value. */ + memmove(first_val + size, first_val, val - first_val); + memset(first_val, 0, size); + s->here->e_value_size = 0; + s->here->e_value_offs = 0; + min_offs += size; + + /* Adjust all value offsets. */ + last = s->first; + while (!IS_LAST_ENTRY(last)) { + size_t o = le16_to_cpu(last->e_value_offs); + if (!last->e_value_block && + last->e_value_size && o < offs) + last->e_value_offs = + cpu_to_le16(o + size); + last = EXT4_XATTR_NEXT(last); + } + } + if (!i->value) { + /* Remove the old name. */ + size_t size = EXT4_XATTR_LEN(name_len); + last = ENTRY((void *)last - size); + memmove(s->here, (void *)s->here + size, + (void *)last - (void *)s->here + sizeof(__u32)); + memset(last, 0, size); + } + } + + if (i->value) { + /* Insert the new value. */ + s->here->e_value_size = cpu_to_le32(i->value_len); + if (i->value_len) { + size_t size = EXT4_XATTR_SIZE(i->value_len); + void *val = s->base + min_offs - size; + s->here->e_value_offs = cpu_to_le16(min_offs - size); + memset(val + size - EXT4_XATTR_PAD, 0, + EXT4_XATTR_PAD); /* Clear the pad bytes. */ + memcpy(val, i->value, i->value_len); + } + } + return 0; +} + +struct ext4_xattr_block_find { + struct ext4_xattr_search s; + struct buffer_head *bh; +}; + +static int +ext4_xattr_block_find(struct inode *inode, struct ext4_xattr_info *i, + struct ext4_xattr_block_find *bs) +{ + struct super_block *sb = inode->i_sb; + int error; + + ea_idebug(inode, "name=%d.%s, value=%p, value_len=%ld", + i->name_index, i->name, i->value, (long)i->value_len); + + if (EXT4_I(inode)->i_file_acl) { + /* The inode already has an extended attribute block. */ + bs->bh = sb_bread(sb, EXT4_I(inode)->i_file_acl); + error = -EIO; + if (!bs->bh) + goto cleanup; + ea_bdebug(bs->bh, "b_count=%d, refcount=%d", + atomic_read(&(bs->bh->b_count)), + le32_to_cpu(BHDR(bs->bh)->h_refcount)); + if (ext4_xattr_check_block(bs->bh)) { + ext4_error(sb, __FUNCTION__, + "inode %lu: bad block %llu", inode->i_ino, + EXT4_I(inode)->i_file_acl); + error = -EIO; + goto cleanup; + } + /* Find the named attribute. */ + bs->s.base = BHDR(bs->bh); + bs->s.first = BFIRST(bs->bh); + bs->s.end = bs->bh->b_data + bs->bh->b_size; + bs->s.here = bs->s.first; + error = ext4_xattr_find_entry(&bs->s.here, i->name_index, + i->name, bs->bh->b_size, 1); + if (error && error != -ENODATA) + goto cleanup; + bs->s.not_found = error; + } + error = 0; + +cleanup: + return error; +} + +static int +ext4_xattr_block_set(handle_t *handle, struct inode *inode, + struct ext4_xattr_info *i, + struct ext4_xattr_block_find *bs) +{ + struct super_block *sb = inode->i_sb; + struct buffer_head *new_bh = NULL; + struct ext4_xattr_search *s = &bs->s; + struct mb_cache_entry *ce = NULL; + int error; + +#define header(x) ((struct ext4_xattr_header *)(x)) + + if (i->value && i->value_len > sb->s_blocksize) + return -ENOSPC; + if (s->base) { + ce = mb_cache_entry_get(ext4_xattr_cache, bs->bh->b_bdev, + bs->bh->b_blocknr); + if (header(s->base)->h_refcount == cpu_to_le32(1)) { + if (ce) { + mb_cache_entry_free(ce); + ce = NULL; + } + ea_bdebug(bs->bh, "modifying in-place"); + error = ext4_journal_get_write_access(handle, bs->bh); + if (error) + goto cleanup; + lock_buffer(bs->bh); + error = ext4_xattr_set_entry(i, s); + if (!error) { + if (!IS_LAST_ENTRY(s->first)) + ext4_xattr_rehash(header(s->base), + s->here); + ext4_xattr_cache_insert(bs->bh); + } + unlock_buffer(bs->bh); + if (error == -EIO) + goto bad_block; + if (!error) + error = ext4_journal_dirty_metadata(handle, + bs->bh); + if (error) + goto cleanup; + goto inserted; + } else { + int offset = (char *)s->here - bs->bh->b_data; + + if (ce) { + mb_cache_entry_release(ce); + ce = NULL; + } + ea_bdebug(bs->bh, "cloning"); + s->base = kmalloc(bs->bh->b_size, GFP_KERNEL); + error = -ENOMEM; + if (s->base == NULL) + goto cleanup; + memcpy(s->base, BHDR(bs->bh), bs->bh->b_size); + s->first = ENTRY(header(s->base)+1); + header(s->base)->h_refcount = cpu_to_le32(1); + s->here = ENTRY(s->base + offset); + s->end = s->base + bs->bh->b_size; + } + } else { + /* Allocate a buffer where we construct the new block. */ + s->base = kmalloc(sb->s_blocksize, GFP_KERNEL); + /* assert(header == s->base) */ + error = -ENOMEM; + if (s->base == NULL) + goto cleanup; + memset(s->base, 0, sb->s_blocksize); + header(s->base)->h_magic = cpu_to_le32(EXT4_XATTR_MAGIC); + header(s->base)->h_blocks = cpu_to_le32(1); + header(s->base)->h_refcount = cpu_to_le32(1); + s->first = ENTRY(header(s->base)+1); + s->here = ENTRY(header(s->base)+1); + s->end = s->base + sb->s_blocksize; + } + + error = ext4_xattr_set_entry(i, s); + if (error == -EIO) + goto bad_block; + if (error) + goto cleanup; + if (!IS_LAST_ENTRY(s->first)) + ext4_xattr_rehash(header(s->base), s->here); + +inserted: + if (!IS_LAST_ENTRY(s->first)) { + new_bh = ext4_xattr_cache_find(inode, header(s->base), &ce); + if (new_bh) { + /* We found an identical block in the cache. */ + if (new_bh == bs->bh) + ea_bdebug(new_bh, "keeping"); + else { + /* The old block is released after updating + the inode. */ + error = -EDQUOT; + if (DQUOT_ALLOC_BLOCK(inode, 1)) + goto cleanup; + error = ext4_journal_get_write_access(handle, + new_bh); + if (error) + goto cleanup_dquot; + lock_buffer(new_bh); + BHDR(new_bh)->h_refcount = cpu_to_le32(1 + + le32_to_cpu(BHDR(new_bh)->h_refcount)); + ea_bdebug(new_bh, "reusing; refcount now=%d", + le32_to_cpu(BHDR(new_bh)->h_refcount)); + unlock_buffer(new_bh); + error = ext4_journal_dirty_metadata(handle, + new_bh); + if (error) + goto cleanup_dquot; + } + mb_cache_entry_release(ce); + ce = NULL; + } else if (bs->bh && s->base == bs->bh->b_data) { + /* We were modifying this block in-place. */ + ea_bdebug(bs->bh, "keeping this block"); + new_bh = bs->bh; + get_bh(new_bh); + } else { + /* We need to allocate a new block */ + ext4_fsblk_t goal = le32_to_cpu( + EXT4_SB(sb)->s_es->s_first_data_block) + + (ext4_fsblk_t)EXT4_I(inode)->i_block_group * + EXT4_BLOCKS_PER_GROUP(sb); + ext4_fsblk_t block = ext4_new_block(handle, inode, + goal, &error); + if (error) + goto cleanup; + ea_idebug(inode, "creating block %d", block); + + new_bh = sb_getblk(sb, block); + if (!new_bh) { +getblk_failed: + ext4_free_blocks(handle, inode, block, 1); + error = -EIO; + goto cleanup; + } + lock_buffer(new_bh); + error = ext4_journal_get_create_access(handle, new_bh); + if (error) { + unlock_buffer(new_bh); + goto getblk_failed; + } + memcpy(new_bh->b_data, s->base, new_bh->b_size); + set_buffer_uptodate(new_bh); + unlock_buffer(new_bh); + ext4_xattr_cache_insert(new_bh); + error = ext4_journal_dirty_metadata(handle, new_bh); + if (error) + goto cleanup; + } + } + + /* Update the inode. */ + EXT4_I(inode)->i_file_acl = new_bh ? new_bh->b_blocknr : 0; + + /* Drop the previous xattr block. */ + if (bs->bh && bs->bh != new_bh) + ext4_xattr_release_block(handle, inode, bs->bh); + error = 0; + +cleanup: + if (ce) + mb_cache_entry_release(ce); + brelse(new_bh); + if (!(bs->bh && s->base == bs->bh->b_data)) + kfree(s->base); + + return error; + +cleanup_dquot: + DQUOT_FREE_BLOCK(inode, 1); + goto cleanup; + +bad_block: + ext4_error(inode->i_sb, __FUNCTION__, + "inode %lu: bad block %llu", inode->i_ino, + EXT4_I(inode)->i_file_acl); + goto cleanup; + +#undef header +} + +struct ext4_xattr_ibody_find { + struct ext4_xattr_search s; + struct ext4_iloc iloc; +}; + +static int +ext4_xattr_ibody_find(struct inode *inode, struct ext4_xattr_info *i, + struct ext4_xattr_ibody_find *is) +{ + struct ext4_xattr_ibody_header *header; + struct ext4_inode *raw_inode; + int error; + + if (EXT4_I(inode)->i_extra_isize == 0) + return 0; + raw_inode = ext4_raw_inode(&is->iloc); + header = IHDR(inode, raw_inode); + is->s.base = is->s.first = IFIRST(header); + is->s.here = is->s.first; + is->s.end = (void *)raw_inode + EXT4_SB(inode->i_sb)->s_inode_size; + if (EXT4_I(inode)->i_state & EXT4_STATE_XATTR) { + error = ext4_xattr_check_names(IFIRST(header), is->s.end); + if (error) + return error; + /* Find the named attribute. */ + error = ext4_xattr_find_entry(&is->s.here, i->name_index, + i->name, is->s.end - + (void *)is->s.base, 0); + if (error && error != -ENODATA) + return error; + is->s.not_found = error; + } + return 0; +} + +static int +ext4_xattr_ibody_set(handle_t *handle, struct inode *inode, + struct ext4_xattr_info *i, + struct ext4_xattr_ibody_find *is) +{ + struct ext4_xattr_ibody_header *header; + struct ext4_xattr_search *s = &is->s; + int error; + + if (EXT4_I(inode)->i_extra_isize == 0) + return -ENOSPC; + error = ext4_xattr_set_entry(i, s); + if (error) + return error; + header = IHDR(inode, ext4_raw_inode(&is->iloc)); + if (!IS_LAST_ENTRY(s->first)) { + header->h_magic = cpu_to_le32(EXT4_XATTR_MAGIC); + EXT4_I(inode)->i_state |= EXT4_STATE_XATTR; + } else { + header->h_magic = cpu_to_le32(0); + EXT4_I(inode)->i_state &= ~EXT4_STATE_XATTR; + } + return 0; +} + +/* + * ext4_xattr_set_handle() + * + * Create, replace or remove an extended attribute for this inode. Buffer + * is NULL to remove an existing extended attribute, and non-NULL to + * either replace an existing extended attribute, or create a new extended + * attribute. The flags XATTR_REPLACE and XATTR_CREATE + * specify that an extended attribute must exist and must not exist + * previous to the call, respectively. + * + * Returns 0, or a negative error number on failure. + */ +int +ext4_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index, + const char *name, const void *value, size_t value_len, + int flags) +{ + struct ext4_xattr_info i = { + .name_index = name_index, + .name = name, + .value = value, + .value_len = value_len, + + }; + struct ext4_xattr_ibody_find is = { + .s = { .not_found = -ENODATA, }, + }; + struct ext4_xattr_block_find bs = { + .s = { .not_found = -ENODATA, }, + }; + int error; + + if (!name) + return -EINVAL; + if (strlen(name) > 255) + return -ERANGE; + down_write(&EXT4_I(inode)->xattr_sem); + error = ext4_get_inode_loc(inode, &is.iloc); + if (error) + goto cleanup; + + if (EXT4_I(inode)->i_state & EXT4_STATE_NEW) { + struct ext4_inode *raw_inode = ext4_raw_inode(&is.iloc); + memset(raw_inode, 0, EXT4_SB(inode->i_sb)->s_inode_size); + EXT4_I(inode)->i_state &= ~EXT4_STATE_NEW; + } + + error = ext4_xattr_ibody_find(inode, &i, &is); + if (error) + goto cleanup; + if (is.s.not_found) + error = ext4_xattr_block_find(inode, &i, &bs); + if (error) + goto cleanup; + if (is.s.not_found && bs.s.not_found) { + error = -ENODATA; + if (flags & XATTR_REPLACE) + goto cleanup; + error = 0; + if (!value) + goto cleanup; + } else { + error = -EEXIST; + if (flags & XATTR_CREATE) + goto cleanup; + } + error = ext4_journal_get_write_access(handle, is.iloc.bh); + if (error) + goto cleanup; + if (!value) { + if (!is.s.not_found) + error = ext4_xattr_ibody_set(handle, inode, &i, &is); + else if (!bs.s.not_found) + error = ext4_xattr_block_set(handle, inode, &i, &bs); + } else { + error = ext4_xattr_ibody_set(handle, inode, &i, &is); + if (!error && !bs.s.not_found) { + i.value = NULL; + error = ext4_xattr_block_set(handle, inode, &i, &bs); + } else if (error == -ENOSPC) { + error = ext4_xattr_block_set(handle, inode, &i, &bs); + if (error) + goto cleanup; + if (!is.s.not_found) { + i.value = NULL; + error = ext4_xattr_ibody_set(handle, inode, &i, + &is); + } + } + } + if (!error) { + ext4_xattr_update_super_block(handle, inode->i_sb); + inode->i_ctime = CURRENT_TIME_SEC; + error = ext4_mark_iloc_dirty(handle, inode, &is.iloc); + /* + * The bh is consumed by ext4_mark_iloc_dirty, even with + * error != 0. + */ + is.iloc.bh = NULL; + if (IS_SYNC(inode)) + handle->h_sync = 1; + } + +cleanup: + brelse(is.iloc.bh); + brelse(bs.bh); + up_write(&EXT4_I(inode)->xattr_sem); + return error; +} + +/* + * ext4_xattr_set() + * + * Like ext4_xattr_set_handle, but start from an inode. This extended + * attribute modification is a filesystem transaction by itself. + * + * Returns 0, or a negative error number on failure. + */ +int +ext4_xattr_set(struct inode *inode, int name_index, const char *name, + const void *value, size_t value_len, int flags) +{ + handle_t *handle; + int error, retries = 0; + +retry: + handle = ext4_journal_start(inode, EXT4_DATA_TRANS_BLOCKS(inode->i_sb)); + if (IS_ERR(handle)) { + error = PTR_ERR(handle); + } else { + int error2; + + error = ext4_xattr_set_handle(handle, inode, name_index, name, + value, value_len, flags); + error2 = ext4_journal_stop(handle); + if (error == -ENOSPC && + ext4_should_retry_alloc(inode->i_sb, &retries)) + goto retry; + if (error == 0) + error = error2; + } + + return error; +} + +/* + * ext4_xattr_delete_inode() + * + * Free extended attribute resources associated with this inode. This + * is called immediately before an inode is freed. We have exclusive + * access to the inode. + */ +void +ext4_xattr_delete_inode(handle_t *handle, struct inode *inode) +{ + struct buffer_head *bh = NULL; + + if (!EXT4_I(inode)->i_file_acl) + goto cleanup; + bh = sb_bread(inode->i_sb, EXT4_I(inode)->i_file_acl); + if (!bh) { + ext4_error(inode->i_sb, __FUNCTION__, + "inode %lu: block %llu read error", inode->i_ino, + EXT4_I(inode)->i_file_acl); + goto cleanup; + } + if (BHDR(bh)->h_magic != cpu_to_le32(EXT4_XATTR_MAGIC) || + BHDR(bh)->h_blocks != cpu_to_le32(1)) { + ext4_error(inode->i_sb, __FUNCTION__, + "inode %lu: bad block %llu", inode->i_ino, + EXT4_I(inode)->i_file_acl); + goto cleanup; + } + ext4_xattr_release_block(handle, inode, bh); + EXT4_I(inode)->i_file_acl = 0; + +cleanup: + brelse(bh); +} + +/* + * ext4_xattr_put_super() + * + * This is called when a file system is unmounted. + */ +void +ext4_xattr_put_super(struct super_block *sb) +{ + mb_cache_shrink(sb->s_bdev); +} + +/* + * ext4_xattr_cache_insert() + * + * Create a new entry in the extended attribute cache, and insert + * it unless such an entry is already in the cache. + * + * Returns 0, or a negative error number on failure. + */ +static void +ext4_xattr_cache_insert(struct buffer_head *bh) +{ + __u32 hash = le32_to_cpu(BHDR(bh)->h_hash); + struct mb_cache_entry *ce; + int error; + + ce = mb_cache_entry_alloc(ext4_xattr_cache); + if (!ce) { + ea_bdebug(bh, "out of memory"); + return; + } + error = mb_cache_entry_insert(ce, bh->b_bdev, bh->b_blocknr, &hash); + if (error) { + mb_cache_entry_free(ce); + if (error == -EBUSY) { + ea_bdebug(bh, "already in cache"); + error = 0; + } + } else { + ea_bdebug(bh, "inserting [%x]", (int)hash); + mb_cache_entry_release(ce); + } +} + +/* + * ext4_xattr_cmp() + * + * Compare two extended attribute blocks for equality. + * + * Returns 0 if the blocks are equal, 1 if they differ, and + * a negative error number on errors. + */ +static int +ext4_xattr_cmp(struct ext4_xattr_header *header1, + struct ext4_xattr_header *header2) +{ + struct ext4_xattr_entry *entry1, *entry2; + + entry1 = ENTRY(header1+1); + entry2 = ENTRY(header2+1); + while (!IS_LAST_ENTRY(entry1)) { + if (IS_LAST_ENTRY(entry2)) + return 1; + if (entry1->e_hash != entry2->e_hash || + entry1->e_name_index != entry2->e_name_index || + entry1->e_name_len != entry2->e_name_len || + entry1->e_value_size != entry2->e_value_size || + memcmp(entry1->e_name, entry2->e_name, entry1->e_name_len)) + return 1; + if (entry1->e_value_block != 0 || entry2->e_value_block != 0) + return -EIO; + if (memcmp((char *)header1 + le16_to_cpu(entry1->e_value_offs), + (char *)header2 + le16_to_cpu(entry2->e_value_offs), + le32_to_cpu(entry1->e_value_size))) + return 1; + + entry1 = EXT4_XATTR_NEXT(entry1); + entry2 = EXT4_XATTR_NEXT(entry2); + } + if (!IS_LAST_ENTRY(entry2)) + return 1; + return 0; +} + +/* + * ext4_xattr_cache_find() + * + * Find an identical extended attribute block. + * + * Returns a pointer to the block found, or NULL if such a block was + * not found or an error occurred. + */ +static struct buffer_head * +ext4_xattr_cache_find(struct inode *inode, struct ext4_xattr_header *header, + struct mb_cache_entry **pce) +{ + __u32 hash = le32_to_cpu(header->h_hash); + struct mb_cache_entry *ce; + + if (!header->h_hash) + return NULL; /* never share */ + ea_idebug(inode, "looking for cached blocks [%x]", (int)hash); +again: + ce = mb_cache_entry_find_first(ext4_xattr_cache, 0, + inode->i_sb->s_bdev, hash); + while (ce) { + struct buffer_head *bh; + + if (IS_ERR(ce)) { + if (PTR_ERR(ce) == -EAGAIN) + goto again; + break; + } + bh = sb_bread(inode->i_sb, ce->e_block); + if (!bh) { + ext4_error(inode->i_sb, __FUNCTION__, + "inode %lu: block %lu read error", + inode->i_ino, (unsigned long) ce->e_block); + } else if (le32_to_cpu(BHDR(bh)->h_refcount) >= + EXT4_XATTR_REFCOUNT_MAX) { + ea_idebug(inode, "block %lu refcount %d>=%d", + (unsigned long) ce->e_block, + le32_to_cpu(BHDR(bh)->h_refcount), + EXT4_XATTR_REFCOUNT_MAX); + } else if (ext4_xattr_cmp(header, BHDR(bh)) == 0) { + *pce = ce; + return bh; + } + brelse(bh); + ce = mb_cache_entry_find_next(ce, 0, inode->i_sb->s_bdev, hash); + } + return NULL; +} + +#define NAME_HASH_SHIFT 5 +#define VALUE_HASH_SHIFT 16 + +/* + * ext4_xattr_hash_entry() + * + * Compute the hash of an extended attribute. + */ +static inline void ext4_xattr_hash_entry(struct ext4_xattr_header *header, + struct ext4_xattr_entry *entry) +{ + __u32 hash = 0; + char *name = entry->e_name; + int n; + + for (n=0; n < entry->e_name_len; n++) { + hash = (hash << NAME_HASH_SHIFT) ^ + (hash >> (8*sizeof(hash) - NAME_HASH_SHIFT)) ^ + *name++; + } + + if (entry->e_value_block == 0 && entry->e_value_size != 0) { + __le32 *value = (__le32 *)((char *)header + + le16_to_cpu(entry->e_value_offs)); + for (n = (le32_to_cpu(entry->e_value_size) + + EXT4_XATTR_ROUND) >> EXT4_XATTR_PAD_BITS; n; n--) { + hash = (hash << VALUE_HASH_SHIFT) ^ + (hash >> (8*sizeof(hash) - VALUE_HASH_SHIFT)) ^ + le32_to_cpu(*value++); + } + } + entry->e_hash = cpu_to_le32(hash); +} + +#undef NAME_HASH_SHIFT +#undef VALUE_HASH_SHIFT + +#define BLOCK_HASH_SHIFT 16 + +/* + * ext4_xattr_rehash() + * + * Re-compute the extended attribute hash value after an entry has changed. + */ +static void ext4_xattr_rehash(struct ext4_xattr_header *header, + struct ext4_xattr_entry *entry) +{ + struct ext4_xattr_entry *here; + __u32 hash = 0; + + ext4_xattr_hash_entry(header, entry); + here = ENTRY(header+1); + while (!IS_LAST_ENTRY(here)) { + if (!here->e_hash) { + /* Block is not shared if an entry's hash value == 0 */ + hash = 0; + break; + } + hash = (hash << BLOCK_HASH_SHIFT) ^ + (hash >> (8*sizeof(hash) - BLOCK_HASH_SHIFT)) ^ + le32_to_cpu(here->e_hash); + here = EXT4_XATTR_NEXT(here); + } + header->h_hash = cpu_to_le32(hash); +} + +#undef BLOCK_HASH_SHIFT + +int __init +init_ext4_xattr(void) +{ + ext4_xattr_cache = mb_cache_create("ext4_xattr", NULL, + sizeof(struct mb_cache_entry) + + sizeof(((struct mb_cache_entry *) 0)->e_indexes[0]), 1, 6); + if (!ext4_xattr_cache) + return -ENOMEM; + return 0; +} + +void +exit_ext4_xattr(void) +{ + if (ext4_xattr_cache) + mb_cache_destroy(ext4_xattr_cache); + ext4_xattr_cache = NULL; +} diff --git a/fs/ext4/xattr.h b/fs/ext4/xattr.h new file mode 100644 index 000000000000..79432b35398f --- /dev/null +++ b/fs/ext4/xattr.h @@ -0,0 +1,145 @@ +/* + File: fs/ext4/xattr.h + + On-disk format of extended attributes for the ext4 filesystem. + + (C) 2001 Andreas Gruenbacher, <a.gruenbacher@computer.org> +*/ + +#include <linux/xattr.h> + +/* Magic value in attribute blocks */ +#define EXT4_XATTR_MAGIC 0xEA020000 + +/* Maximum number of references to one attribute block */ +#define EXT4_XATTR_REFCOUNT_MAX 1024 + +/* Name indexes */ +#define EXT4_XATTR_INDEX_USER 1 +#define EXT4_XATTR_INDEX_POSIX_ACL_ACCESS 2 +#define EXT4_XATTR_INDEX_POSIX_ACL_DEFAULT 3 +#define EXT4_XATTR_INDEX_TRUSTED 4 +#define EXT4_XATTR_INDEX_LUSTRE 5 +#define EXT4_XATTR_INDEX_SECURITY 6 + +struct ext4_xattr_header { + __le32 h_magic; /* magic number for identification */ + __le32 h_refcount; /* reference count */ + __le32 h_blocks; /* number of disk blocks used */ + __le32 h_hash; /* hash value of all attributes */ + __u32 h_reserved[4]; /* zero right now */ +}; + +struct ext4_xattr_ibody_header { + __le32 h_magic; /* magic number for identification */ +}; + +struct ext4_xattr_entry { + __u8 e_name_len; /* length of name */ + __u8 e_name_index; /* attribute name index */ + __le16 e_value_offs; /* offset in disk block of value */ + __le32 e_value_block; /* disk block attribute is stored on (n/i) */ + __le32 e_value_size; /* size of attribute value */ + __le32 e_hash; /* hash value of name and value */ + char e_name[0]; /* attribute name */ +}; + +#define EXT4_XATTR_PAD_BITS 2 +#define EXT4_XATTR_PAD (1<<EXT4_XATTR_PAD_BITS) +#define EXT4_XATTR_ROUND (EXT4_XATTR_PAD-1) +#define EXT4_XATTR_LEN(name_len) \ + (((name_len) + EXT4_XATTR_ROUND + \ + sizeof(struct ext4_xattr_entry)) & ~EXT4_XATTR_ROUND) +#define EXT4_XATTR_NEXT(entry) \ + ( (struct ext4_xattr_entry *)( \ + (char *)(entry) + EXT4_XATTR_LEN((entry)->e_name_len)) ) +#define EXT4_XATTR_SIZE(size) \ + (((size) + EXT4_XATTR_ROUND) & ~EXT4_XATTR_ROUND) + +# ifdef CONFIG_EXT4DEV_FS_XATTR + +extern struct xattr_handler ext4_xattr_user_handler; +extern struct xattr_handler ext4_xattr_trusted_handler; +extern struct xattr_handler ext4_xattr_acl_access_handler; +extern struct xattr_handler ext4_xattr_acl_default_handler; +extern struct xattr_handler ext4_xattr_security_handler; + +extern ssize_t ext4_listxattr(struct dentry *, char *, size_t); + +extern int ext4_xattr_get(struct inode *, int, const char *, void *, size_t); +extern int ext4_xattr_list(struct inode *, char *, size_t); +extern int ext4_xattr_set(struct inode *, int, const char *, const void *, size_t, int); +extern int ext4_xattr_set_handle(handle_t *, struct inode *, int, const char *, const void *, size_t, int); + +extern void ext4_xattr_delete_inode(handle_t *, struct inode *); +extern void ext4_xattr_put_super(struct super_block *); + +extern int init_ext4_xattr(void); +extern void exit_ext4_xattr(void); + +extern struct xattr_handler *ext4_xattr_handlers[]; + +# else /* CONFIG_EXT4DEV_FS_XATTR */ + +static inline int +ext4_xattr_get(struct inode *inode, int name_index, const char *name, + void *buffer, size_t size, int flags) +{ + return -EOPNOTSUPP; +} + +static inline int +ext4_xattr_list(struct inode *inode, void *buffer, size_t size) +{ + return -EOPNOTSUPP; +} + +static inline int +ext4_xattr_set(struct inode *inode, int name_index, const char *name, + const void *value, size_t size, int flags) +{ + return -EOPNOTSUPP; +} + +static inline int +ext4_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index, + const char *name, const void *value, size_t size, int flags) +{ + return -EOPNOTSUPP; +} + +static inline void +ext4_xattr_delete_inode(handle_t *handle, struct inode *inode) +{ +} + +static inline void +ext4_xattr_put_super(struct super_block *sb) +{ +} + +static inline int +init_ext4_xattr(void) +{ + return 0; +} + +static inline void +exit_ext4_xattr(void) +{ +} + +#define ext4_xattr_handlers NULL + +# endif /* CONFIG_EXT4DEV_FS_XATTR */ + +#ifdef CONFIG_EXT4DEV_FS_SECURITY +extern int ext4_init_security(handle_t *handle, struct inode *inode, + struct inode *dir); +#else +static inline int ext4_init_security(handle_t *handle, struct inode *inode, + struct inode *dir) +{ + return 0; +} +#endif diff --git a/fs/ext4/xattr_security.c b/fs/ext4/xattr_security.c new file mode 100644 index 000000000000..b6a6861951f9 --- /dev/null +++ b/fs/ext4/xattr_security.c @@ -0,0 +1,77 @@ +/* + * linux/fs/ext4/xattr_security.c + * Handler for storing security labels as extended attributes. + */ + +#include <linux/module.h> +#include <linux/string.h> +#include <linux/fs.h> +#include <linux/smp_lock.h> +#include <linux/ext4_jbd2.h> +#include <linux/ext4_fs.h> +#include <linux/security.h> +#include "xattr.h" + +static size_t +ext4_xattr_security_list(struct inode *inode, char *list, size_t list_size, + const char *name, size_t name_len) +{ + const size_t prefix_len = sizeof(XATTR_SECURITY_PREFIX)-1; + const size_t total_len = prefix_len + name_len + 1; + + + if (list && total_len <= list_size) { + memcpy(list, XATTR_SECURITY_PREFIX, prefix_len); + memcpy(list+prefix_len, name, name_len); + list[prefix_len + name_len] = '\0'; + } + return total_len; +} + +static int +ext4_xattr_security_get(struct inode *inode, const char *name, + void *buffer, size_t size) +{ + if (strcmp(name, "") == 0) + return -EINVAL; + return ext4_xattr_get(inode, EXT4_XATTR_INDEX_SECURITY, name, + buffer, size); +} + +static int +ext4_xattr_security_set(struct inode *inode, const char *name, + const void *value, size_t size, int flags) +{ + if (strcmp(name, "") == 0) + return -EINVAL; + return ext4_xattr_set(inode, EXT4_XATTR_INDEX_SECURITY, name, + value, size, flags); +} + +int +ext4_init_security(handle_t *handle, struct inode *inode, struct inode *dir) +{ + int err; + size_t len; + void *value; + char *name; + + err = security_inode_init_security(inode, dir, &name, &value, &len); + if (err) { + if (err == -EOPNOTSUPP) + return 0; + return err; + } + err = ext4_xattr_set_handle(handle, inode, EXT4_XATTR_INDEX_SECURITY, + name, value, len, 0); + kfree(name); + kfree(value); + return err; +} + +struct xattr_handler ext4_xattr_security_handler = { + .prefix = XATTR_SECURITY_PREFIX, + .list = ext4_xattr_security_list, + .get = ext4_xattr_security_get, + .set = ext4_xattr_security_set, +}; diff --git a/fs/ext4/xattr_trusted.c b/fs/ext4/xattr_trusted.c new file mode 100644 index 000000000000..b76f2dbc82da --- /dev/null +++ b/fs/ext4/xattr_trusted.c @@ -0,0 +1,62 @@ +/* + * linux/fs/ext4/xattr_trusted.c + * Handler for trusted extended attributes. + * + * Copyright (C) 2003 by Andreas Gruenbacher, <a.gruenbacher@computer.org> + */ + +#include <linux/module.h> +#include <linux/string.h> +#include <linux/capability.h> +#include <linux/fs.h> +#include <linux/smp_lock.h> +#include <linux/ext4_jbd2.h> +#include <linux/ext4_fs.h> +#include "xattr.h" + +#define XATTR_TRUSTED_PREFIX "trusted." + +static size_t +ext4_xattr_trusted_list(struct inode *inode, char *list, size_t list_size, + const char *name, size_t name_len) +{ + const size_t prefix_len = sizeof(XATTR_TRUSTED_PREFIX)-1; + const size_t total_len = prefix_len + name_len + 1; + + if (!capable(CAP_SYS_ADMIN)) + return 0; + + if (list && total_len <= list_size) { + memcpy(list, XATTR_TRUSTED_PREFIX, prefix_len); + memcpy(list+prefix_len, name, name_len); + list[prefix_len + name_len] = '\0'; + } + return total_len; +} + +static int +ext4_xattr_trusted_get(struct inode *inode, const char *name, + void *buffer, size_t size) +{ + if (strcmp(name, "") == 0) + return -EINVAL; + return ext4_xattr_get(inode, EXT4_XATTR_INDEX_TRUSTED, name, + buffer, size); +} + +static int +ext4_xattr_trusted_set(struct inode *inode, const char *name, + const void *value, size_t size, int flags) +{ + if (strcmp(name, "") == 0) + return -EINVAL; + return ext4_xattr_set(inode, EXT4_XATTR_INDEX_TRUSTED, name, + value, size, flags); +} + +struct xattr_handler ext4_xattr_trusted_handler = { + .prefix = XATTR_TRUSTED_PREFIX, + .list = ext4_xattr_trusted_list, + .get = ext4_xattr_trusted_get, + .set = ext4_xattr_trusted_set, +}; diff --git a/fs/ext4/xattr_user.c b/fs/ext4/xattr_user.c new file mode 100644 index 000000000000..c53cded0761a --- /dev/null +++ b/fs/ext4/xattr_user.c @@ -0,0 +1,64 @@ +/* + * linux/fs/ext4/xattr_user.c + * Handler for extended user attributes. + * + * Copyright (C) 2001 by Andreas Gruenbacher, <a.gruenbacher@computer.org> + */ + +#include <linux/module.h> +#include <linux/string.h> +#include <linux/fs.h> +#include <linux/smp_lock.h> +#include <linux/ext4_jbd2.h> +#include <linux/ext4_fs.h> +#include "xattr.h" + +#define XATTR_USER_PREFIX "user." + +static size_t +ext4_xattr_user_list(struct inode *inode, char *list, size_t list_size, + const char *name, size_t name_len) +{ + const size_t prefix_len = sizeof(XATTR_USER_PREFIX)-1; + const size_t total_len = prefix_len + name_len + 1; + + if (!test_opt(inode->i_sb, XATTR_USER)) + return 0; + + if (list && total_len <= list_size) { + memcpy(list, XATTR_USER_PREFIX, prefix_len); + memcpy(list+prefix_len, name, name_len); + list[prefix_len + name_len] = '\0'; + } + return total_len; +} + +static int +ext4_xattr_user_get(struct inode *inode, const char *name, + void *buffer, size_t size) +{ + if (strcmp(name, "") == 0) + return -EINVAL; + if (!test_opt(inode->i_sb, XATTR_USER)) + return -EOPNOTSUPP; + return ext4_xattr_get(inode, EXT4_XATTR_INDEX_USER, name, buffer, size); +} + +static int +ext4_xattr_user_set(struct inode *inode, const char *name, + const void *value, size_t size, int flags) +{ + if (strcmp(name, "") == 0) + return -EINVAL; + if (!test_opt(inode->i_sb, XATTR_USER)) + return -EOPNOTSUPP; + return ext4_xattr_set(inode, EXT4_XATTR_INDEX_USER, name, + value, size, flags); +} + +struct xattr_handler ext4_xattr_user_handler = { + .prefix = XATTR_USER_PREFIX, + .list = ext4_xattr_user_list, + .get = ext4_xattr_user_get, + .set = ext4_xattr_user_set, +}; diff --git a/fs/fat/file.c b/fs/fat/file.c index f4b8f8b3fbdd..8337451e7897 100644 --- a/fs/fat/file.c +++ b/fs/fat/file.c @@ -13,6 +13,7 @@ #include <linux/smp_lock.h> #include <linux/buffer_head.h> #include <linux/writeback.h> +#include <linux/backing-dev.h> #include <linux/blkdev.h> int fat_generic_ioctl(struct inode *inode, struct file *filp, @@ -118,7 +119,7 @@ static int fat_file_release(struct inode *inode, struct file *filp) if ((filp->f_mode & FMODE_WRITE) && MSDOS_SB(inode->i_sb)->options.flush) { fat_flush_inodes(inode->i_sb, inode, NULL); - blk_congestion_wait(WRITE, HZ/10); + congestion_wait(WRITE, HZ/10); } return 0; } diff --git a/fs/fat/inode.c b/fs/fat/inode.c index 4613cb202170..78945b53b0f8 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -1472,7 +1472,7 @@ int fat_flush_inodes(struct super_block *sb, struct inode *i1, struct inode *i2) ret = writeback_inode(i1); if (!ret && i2) ret = writeback_inode(i2); - if (!ret && sb) { + if (!ret) { struct address_space *mapping = sb->s_bdev->bd_inode->i_mapping; ret = filemap_flush(mapping); } diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index 8605155db171..cfc8f81e60d0 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -138,6 +138,7 @@ static int fuse_dentry_revalidate(struct dentry *entry, struct nameidata *nd) struct fuse_entry_out outarg; struct fuse_conn *fc; struct fuse_req *req; + struct dentry *parent; /* Doesn't hurt to "reset" the validity timeout */ fuse_invalidate_entry_cache(entry); @@ -151,8 +152,10 @@ static int fuse_dentry_revalidate(struct dentry *entry, struct nameidata *nd) if (IS_ERR(req)) return 0; - fuse_lookup_init(req, entry->d_parent->d_inode, entry, &outarg); + parent = dget_parent(entry); + fuse_lookup_init(req, parent->d_inode, entry, &outarg); request_send(fc, req); + dput(parent); err = req->out.h.error; /* Zero nodeid is same as -ENOENT */ if (!err && !outarg.nodeid) @@ -163,7 +166,9 @@ static int fuse_dentry_revalidate(struct dentry *entry, struct nameidata *nd) fuse_send_forget(fc, req, outarg.nodeid, 1); return 0; } + spin_lock(&fc->lock); fi->nlookup ++; + spin_unlock(&fc->lock); } fuse_put_request(fc, req); if (err || (outarg.attr.mode ^ inode->i_mode) & S_IFMT) @@ -175,22 +180,6 @@ static int fuse_dentry_revalidate(struct dentry *entry, struct nameidata *nd) return 1; } -/* - * Check if there's already a hashed alias of this directory inode. - * If yes, then lookup and mkdir must not create a new alias. - */ -static int dir_alias(struct inode *inode) -{ - if (S_ISDIR(inode->i_mode)) { - struct dentry *alias = d_find_alias(inode); - if (alias) { - dput(alias); - return 1; - } - } - return 0; -} - static int invalid_nodeid(u64 nodeid) { return !nodeid || nodeid == FUSE_ROOT_ID; @@ -206,6 +195,24 @@ static int valid_mode(int m) S_ISBLK(m) || S_ISFIFO(m) || S_ISSOCK(m); } +/* + * Add a directory inode to a dentry, ensuring that no other dentry + * refers to this inode. Called with fc->inst_mutex. + */ +static int fuse_d_add_directory(struct dentry *entry, struct inode *inode) +{ + struct dentry *alias = d_find_alias(inode); + if (alias) { + /* This tries to shrink the subtree below alias */ + fuse_invalidate_entry(alias); + dput(alias); + if (!list_empty(&inode->i_dentry)) + return -EBUSY; + } + d_add(entry, inode); + return 0; +} + static struct dentry *fuse_lookup(struct inode *dir, struct dentry *entry, struct nameidata *nd) { @@ -241,11 +248,17 @@ static struct dentry *fuse_lookup(struct inode *dir, struct dentry *entry, if (err && err != -ENOENT) return ERR_PTR(err); - if (inode && dir_alias(inode)) { - iput(inode); - return ERR_PTR(-EIO); - } - d_add(entry, inode); + if (inode && S_ISDIR(inode->i_mode)) { + mutex_lock(&fc->inst_mutex); + err = fuse_d_add_directory(entry, inode); + mutex_unlock(&fc->inst_mutex); + if (err) { + iput(inode); + return ERR_PTR(err); + } + } else + d_add(entry, inode); + entry->d_op = &fuse_dentry_operations; if (!err) fuse_change_timeout(entry, &outarg); @@ -401,12 +414,22 @@ static int create_new_entry(struct fuse_conn *fc, struct fuse_req *req, } fuse_put_request(fc, req); - if (dir_alias(inode)) { - iput(inode); - return -EIO; - } + if (S_ISDIR(inode->i_mode)) { + struct dentry *alias; + mutex_lock(&fc->inst_mutex); + alias = d_find_alias(inode); + if (alias) { + /* New directory must have moved since mkdir */ + mutex_unlock(&fc->inst_mutex); + dput(alias); + iput(inode); + return -EBUSY; + } + d_instantiate(entry, inode); + mutex_unlock(&fc->inst_mutex); + } else + d_instantiate(entry, inode); - d_instantiate(entry, inode); fuse_change_timeout(entry, &outarg); fuse_invalidate_attr(dir); return 0; @@ -935,14 +958,30 @@ static void iattr_to_fattr(struct iattr *iattr, struct fuse_setattr_in *arg) } } +static void fuse_vmtruncate(struct inode *inode, loff_t offset) +{ + struct fuse_conn *fc = get_fuse_conn(inode); + int need_trunc; + + spin_lock(&fc->lock); + need_trunc = inode->i_size > offset; + i_size_write(inode, offset); + spin_unlock(&fc->lock); + + if (need_trunc) { + struct address_space *mapping = inode->i_mapping; + unmap_mapping_range(mapping, offset + PAGE_SIZE - 1, 0, 1); + truncate_inode_pages(mapping, offset); + } +} + /* * Set attributes, and at the same time refresh them. * * Truncation is slightly complicated, because the 'truncate' request * may fail, in which case we don't want to touch the mapping. - * vmtruncate() doesn't allow for this case. So do the rlimit - * checking by hand and call vmtruncate() only after the file has - * actually been truncated. + * vmtruncate() doesn't allow for this case, so do the rlimit checking + * and the actual truncation by hand. */ static int fuse_setattr(struct dentry *entry, struct iattr *attr) { @@ -993,12 +1032,8 @@ static int fuse_setattr(struct dentry *entry, struct iattr *attr) make_bad_inode(inode); err = -EIO; } else { - if (is_truncate) { - loff_t origsize = i_size_read(inode); - i_size_write(inode, outarg.attr.size); - if (origsize > outarg.attr.size) - vmtruncate(inode, outarg.attr.size); - } + if (is_truncate) + fuse_vmtruncate(inode, outarg.attr.size); fuse_change_attributes(inode, &outarg.attr); fi->i_time = time_to_jiffies(outarg.attr_valid, outarg.attr_valid_nsec); diff --git a/fs/fuse/file.c b/fs/fuse/file.c index 183626868eea..763a50daf1c0 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -397,14 +397,14 @@ static int fuse_readpages(struct file *file, struct address_space *mapping, err = -EIO; if (is_bad_inode(inode)) - goto clean_pages_up; + goto out; data.file = file; data.inode = inode; data.req = fuse_get_req(fc); err = PTR_ERR(data.req); if (IS_ERR(data.req)) - goto clean_pages_up; + goto out; err = read_cache_pages(mapping, pages, fuse_readpages_fill, &data); if (!err) { @@ -413,10 +413,7 @@ static int fuse_readpages(struct file *file, struct address_space *mapping, else fuse_put_request(fc, data.req); } - return err; - -clean_pages_up: - put_pages_list(pages); +out: return err; } @@ -481,8 +478,10 @@ static int fuse_commit_write(struct file *file, struct page *page, err = -EIO; if (!err) { pos += count; - if (pos > i_size_read(inode)) + spin_lock(&fc->lock); + if (pos > inode->i_size) i_size_write(inode, pos); + spin_unlock(&fc->lock); if (offset == 0 && to == PAGE_CACHE_SIZE) { clear_page_dirty(page); @@ -586,8 +585,12 @@ static ssize_t fuse_direct_io(struct file *file, const char __user *buf, } fuse_put_request(fc, req); if (res > 0) { - if (write && pos > i_size_read(inode)) - i_size_write(inode, pos); + if (write) { + spin_lock(&fc->lock); + if (pos > inode->i_size) + i_size_write(inode, pos); + spin_unlock(&fc->lock); + } *ppos = pos; } fuse_invalidate_attr(inode); diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h index 69c7750d55b8..91edb8932d90 100644 --- a/fs/fuse/fuse_i.h +++ b/fs/fuse/fuse_i.h @@ -239,6 +239,9 @@ struct fuse_conn { /** Lock protecting accessess to members of this structure */ spinlock_t lock; + /** Mutex protecting against directory alias creation */ + struct mutex inst_mutex; + /** Refcount */ atomic_t count; diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index 7d0a9aee01f2..fc4203570370 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -109,6 +109,7 @@ static int fuse_remount_fs(struct super_block *sb, int *flags, char *data) void fuse_change_attributes(struct inode *inode, struct fuse_attr *attr) { + struct fuse_conn *fc = get_fuse_conn(inode); if (S_ISREG(inode->i_mode) && i_size_read(inode) != attr->size) invalidate_inode_pages(inode->i_mapping); @@ -117,7 +118,9 @@ void fuse_change_attributes(struct inode *inode, struct fuse_attr *attr) inode->i_nlink = attr->nlink; inode->i_uid = attr->uid; inode->i_gid = attr->gid; + spin_lock(&fc->lock); i_size_write(inode, attr->size); + spin_unlock(&fc->lock); inode->i_blocks = attr->blocks; inode->i_atime.tv_sec = attr->atime; inode->i_atime.tv_nsec = attr->atimensec; @@ -130,7 +133,7 @@ void fuse_change_attributes(struct inode *inode, struct fuse_attr *attr) static void fuse_init_inode(struct inode *inode, struct fuse_attr *attr) { inode->i_mode = attr->mode & S_IFMT; - i_size_write(inode, attr->size); + inode->i_size = attr->size; if (S_ISREG(inode->i_mode)) { fuse_init_common(inode); fuse_init_file_inode(inode); @@ -169,7 +172,6 @@ struct inode *fuse_iget(struct super_block *sb, unsigned long nodeid, struct inode *inode; struct fuse_inode *fi; struct fuse_conn *fc = get_fuse_conn_super(sb); - int retried = 0; retry: inode = iget5_locked(sb, nodeid, fuse_inode_eq, fuse_inode_set, &nodeid); @@ -183,16 +185,16 @@ struct inode *fuse_iget(struct super_block *sb, unsigned long nodeid, fuse_init_inode(inode, attr); unlock_new_inode(inode); } else if ((inode->i_mode ^ attr->mode) & S_IFMT) { - BUG_ON(retried); /* Inode has changed type, any I/O on the old should fail */ make_bad_inode(inode); iput(inode); - retried = 1; goto retry; } fi = get_fuse_inode(inode); + spin_lock(&fc->lock); fi->nlookup ++; + spin_unlock(&fc->lock); fuse_change_attributes(inode, attr); return inode; } @@ -377,6 +379,7 @@ static struct fuse_conn *new_conn(void) fc = kzalloc(sizeof(*fc), GFP_KERNEL); if (fc) { spin_lock_init(&fc->lock); + mutex_init(&fc->inst_mutex); atomic_set(&fc->count, 1); init_waitqueue_head(&fc->waitq); init_waitqueue_head(&fc->blocked_waitq); @@ -396,8 +399,10 @@ static struct fuse_conn *new_conn(void) void fuse_conn_put(struct fuse_conn *fc) { - if (atomic_dec_and_test(&fc->count)) + if (atomic_dec_and_test(&fc->count)) { + mutex_destroy(&fc->inst_mutex); kfree(fc); + } } struct fuse_conn *fuse_conn_get(struct fuse_conn *fc) diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c index cc57f2ecd219..06e9a8cb45e9 100644 --- a/fs/gfs2/bmap.c +++ b/fs/gfs2/bmap.c @@ -434,8 +434,7 @@ static int lookup_block(struct gfs2_inode *ip, struct buffer_head *bh, */ static int gfs2_block_pointers(struct inode *inode, u64 lblock, int create, - struct buffer_head *bh_map, struct metapath *mp, - unsigned int maxlen) + struct buffer_head *bh_map, struct metapath *mp) { struct gfs2_inode *ip = GFS2_I(inode); struct gfs2_sbd *sdp = GFS2_SB(inode); @@ -448,6 +447,7 @@ static int gfs2_block_pointers(struct inode *inode, u64 lblock, int create, int new = 0; u64 dblock = 0; int boundary; + unsigned int maxlen = bh_map->b_size >> inode->i_blkbits; BUG_ON(maxlen == 0); @@ -541,13 +541,13 @@ static inline void bmap_unlock(struct inode *inode, int create) } int gfs2_block_map(struct inode *inode, u64 lblock, int create, - struct buffer_head *bh, unsigned int maxlen) + struct buffer_head *bh) { struct metapath mp; int ret; bmap_lock(inode, create); - ret = gfs2_block_pointers(inode, lblock, create, bh, &mp, maxlen); + ret = gfs2_block_pointers(inode, lblock, create, bh, &mp); bmap_unlock(inode, create); return ret; } @@ -555,7 +555,7 @@ int gfs2_block_map(struct inode *inode, u64 lblock, int create, int gfs2_extent_map(struct inode *inode, u64 lblock, int *new, u64 *dblock, unsigned *extlen) { struct metapath mp; - struct buffer_head bh = { .b_state = 0, .b_blocknr = 0, .b_size = 0 }; + struct buffer_head bh = { .b_state = 0, .b_blocknr = 0 }; int ret; int create = *new; @@ -563,8 +563,9 @@ int gfs2_extent_map(struct inode *inode, u64 lblock, int *new, u64 *dblock, unsi BUG_ON(!dblock); BUG_ON(!new); + bh.b_size = 1 << (inode->i_blkbits + 5); bmap_lock(inode, create); - ret = gfs2_block_pointers(inode, lblock, create, &bh, &mp, 32); + ret = gfs2_block_pointers(inode, lblock, create, &bh, &mp); bmap_unlock(inode, create); *extlen = bh.b_size >> inode->i_blkbits; *dblock = bh.b_blocknr; diff --git a/fs/gfs2/bmap.h b/fs/gfs2/bmap.h index 0fd379b4cd9e..ac2fd04370dc 100644 --- a/fs/gfs2/bmap.h +++ b/fs/gfs2/bmap.h @@ -15,7 +15,7 @@ struct gfs2_inode; struct page; int gfs2_unstuff_dinode(struct gfs2_inode *ip, struct page *page); -int gfs2_block_map(struct inode *inode, u64 lblock, int create, struct buffer_head *bh, unsigned int maxlen); +int gfs2_block_map(struct inode *inode, u64 lblock, int create, struct buffer_head *bh); int gfs2_extent_map(struct inode *inode, u64 lblock, int *new, u64 *dblock, unsigned *extlen); int gfs2_truncatei(struct gfs2_inode *ip, u64 size); diff --git a/fs/gfs2/dir.c b/fs/gfs2/dir.c index 459498cac93b..e24af28b1a12 100644 --- a/fs/gfs2/dir.c +++ b/fs/gfs2/dir.c @@ -184,7 +184,7 @@ static int gfs2_dir_write_data(struct gfs2_inode *ip, const char *buf, while (copied < size) { unsigned int amount; struct buffer_head *bh; - int new; + int new = 0; amount = size - copied; if (amount > sdp->sd_sb.sb_bsize - o) @@ -212,8 +212,6 @@ static int gfs2_dir_write_data(struct gfs2_inode *ip, const char *buf, gfs2_trans_add_bh(ip->i_gl, bh, 1); memcpy(bh->b_data + o, buf, amount); brelse(bh); - if (error) - goto fail; buf += amount; copied += amount; @@ -317,8 +315,7 @@ static int gfs2_dir_read_data(struct gfs2_inode *ip, char *buf, u64 offset, if (!ra) extlen = 1; bh = gfs2_meta_ra(ip->i_gl, dblock, extlen); - } - if (!bh) { + } else { error = gfs2_meta_read(ip->i_gl, dblock, DIO_WAIT, &bh); if (error) goto fail; @@ -332,7 +329,6 @@ static int gfs2_dir_read_data(struct gfs2_inode *ip, char *buf, u64 offset, extlen--; memcpy(buf, bh->b_data + o, amount); brelse(bh); - bh = NULL; buf += amount; copied += amount; lblock++; @@ -815,7 +811,7 @@ static struct gfs2_leaf *new_leaf(struct inode *inode, struct buffer_head **pbh, leaf = (struct gfs2_leaf *)bh->b_data; leaf->lf_depth = cpu_to_be16(depth); leaf->lf_entries = 0; - leaf->lf_dirent_format = cpu_to_be16(GFS2_FORMAT_DE); + leaf->lf_dirent_format = cpu_to_be32(GFS2_FORMAT_DE); leaf->lf_next = 0; memset(leaf->lf_reserved, 0, sizeof(leaf->lf_reserved)); dent = (struct gfs2_dirent *)(leaf+1); diff --git a/fs/gfs2/locking/dlm/mount.c b/fs/gfs2/locking/dlm/mount.c index 1f94dd35a943..cdd1694e889b 100644 --- a/fs/gfs2/locking/dlm/mount.c +++ b/fs/gfs2/locking/dlm/mount.c @@ -45,7 +45,7 @@ static struct gdlm_ls *init_gdlm(lm_callback_t cb, struct gfs2_sbd *sdp, strncpy(buf, table_name, 256); buf[255] = '\0'; - p = strstr(buf, ":"); + p = strchr(buf, ':'); if (!p) { log_info("invalid table_name \"%s\"", table_name); kfree(ls); diff --git a/fs/gfs2/log.c b/fs/gfs2/log.c index 554fe5bd1b72..0cace3da9dbb 100644 --- a/fs/gfs2/log.c +++ b/fs/gfs2/log.c @@ -312,10 +312,12 @@ void gfs2_log_release(struct gfs2_sbd *sdp, unsigned int blks) static u64 log_bmap(struct gfs2_sbd *sdp, unsigned int lbn) { + struct inode *inode = sdp->sd_jdesc->jd_inode; int error; - struct buffer_head bh_map; + struct buffer_head bh_map = { .b_state = 0, .b_blocknr = 0 }; - error = gfs2_block_map(sdp->sd_jdesc->jd_inode, lbn, 0, &bh_map, 1); + bh_map.b_size = 1 << inode->i_blkbits; + error = gfs2_block_map(inode, lbn, 0, &bh_map); if (error || !bh_map.b_blocknr) printk(KERN_INFO "error=%d, dbn=%llu lbn=%u", error, bh_map.b_blocknr, lbn); gfs2_assert_withdraw(sdp, !error && bh_map.b_blocknr); @@ -569,16 +571,15 @@ void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl) else if (sdp->sd_log_tail != current_tail(sdp) && !sdp->sd_log_idle) log_write_header(sdp, 0, PULL); lops_after_commit(sdp, ai); - sdp->sd_log_head = sdp->sd_log_flush_head; + gfs2_log_lock(sdp); + sdp->sd_log_head = sdp->sd_log_flush_head; sdp->sd_log_blks_free -= sdp->sd_log_num_hdrs; - sdp->sd_log_blks_reserved = 0; sdp->sd_log_commited_buf = 0; sdp->sd_log_num_hdrs = 0; sdp->sd_log_commited_revoke = 0; - gfs2_log_lock(sdp); if (!list_empty(&ai->ai_ail1_list)) { list_add(&ai->ai_list, &sdp->sd_ail1_list); ai = NULL; diff --git a/fs/gfs2/lops.c b/fs/gfs2/lops.c index 881e337b6a70..ab6d1115f95d 100644 --- a/fs/gfs2/lops.c +++ b/fs/gfs2/lops.c @@ -492,7 +492,7 @@ static int gfs2_check_magic(struct buffer_head *bh) ptr = kaddr + bh_offset(bh); if (*ptr == cpu_to_be32(GFS2_MAGIC)) rv = 1; - kunmap_atomic(page, KM_USER0); + kunmap_atomic(kaddr, KM_USER0); return rv; } @@ -626,7 +626,7 @@ static void databuf_lo_before_commit(struct gfs2_sbd *sdp) memcpy(bh->b_data, kaddr + bh_offset(bd2->bd_bh), sdp->sd_sb.sb_bsize); - kunmap_atomic(page, KM_USER0); + kunmap_atomic(kaddr, KM_USER0); *(__be32 *)bh->b_data = 0; } else { bh = gfs2_log_fake_buf(sdp, bd2->bd_bh); diff --git a/fs/gfs2/ops_address.c b/fs/gfs2/ops_address.c index 4fb743f4e4a4..015640b3f123 100644 --- a/fs/gfs2/ops_address.c +++ b/fs/gfs2/ops_address.c @@ -65,7 +65,7 @@ static void gfs2_page_add_databufs(struct gfs2_inode *ip, struct page *page, int gfs2_get_block(struct inode *inode, sector_t lblock, struct buffer_head *bh_result, int create) { - return gfs2_block_map(inode, lblock, create, bh_result, 32); + return gfs2_block_map(inode, lblock, create, bh_result); } /** @@ -83,7 +83,7 @@ static int gfs2_get_block_noalloc(struct inode *inode, sector_t lblock, { int error; - error = gfs2_block_map(inode, lblock, 0, bh_result, 1); + error = gfs2_block_map(inode, lblock, 0, bh_result); if (error) return error; if (bh_result->b_blocknr == 0) @@ -94,7 +94,7 @@ static int gfs2_get_block_noalloc(struct inode *inode, sector_t lblock, static int gfs2_get_block_direct(struct inode *inode, sector_t lblock, struct buffer_head *bh_result, int create) { - return gfs2_block_map(inode, lblock, 0, bh_result, 32); + return gfs2_block_map(inode, lblock, 0, bh_result); } /** @@ -162,7 +162,7 @@ static int zero_readpage(struct page *page) kaddr = kmap_atomic(page, KM_USER0); memset(kaddr, 0, PAGE_CACHE_SIZE); - kunmap_atomic(page, KM_USER0); + kunmap_atomic(kaddr, KM_USER0); SetPageUptodate(page); @@ -195,7 +195,7 @@ static int stuffed_readpage(struct gfs2_inode *ip, struct page *page) memcpy(kaddr, dibh->b_data + sizeof(struct gfs2_dinode), ip->i_di.di_size); memset(kaddr + ip->i_di.di_size, 0, PAGE_CACHE_SIZE - ip->i_di.di_size); - kunmap_atomic(page, KM_USER0); + kunmap_atomic(kaddr, KM_USER0); brelse(dibh); @@ -337,13 +337,6 @@ out: out_noerror: ret = 0; out_unlock: - /* unlock all pages, we can't do any I/O right now */ - for (page_idx = 0; page_idx < nr_pages; page_idx++) { - struct page *page = list_entry(pages->prev, struct page, lru); - list_del(&page->lru); - unlock_page(page); - page_cache_release(page); - } if (do_unlock) gfs2_holder_uninit(&gh); goto out; @@ -370,19 +363,22 @@ static int gfs2_prepare_write(struct file *file, struct page *page, loff_t pos = ((loff_t)page->index << PAGE_CACHE_SHIFT) + from; loff_t end = ((loff_t)page->index << PAGE_CACHE_SHIFT) + to; struct gfs2_alloc *al; + unsigned int write_len = to - from; + gfs2_holder_init(ip->i_gl, LM_ST_EXCLUSIVE, GL_ATIME|GL_AOP, &ip->i_gh); error = gfs2_glock_nq_m_atime(1, &ip->i_gh); if (error) goto out_uninit; - gfs2_write_calc_reserv(ip, to - from, &data_blocks, &ind_blocks); + gfs2_write_calc_reserv(ip, write_len, &data_blocks, &ind_blocks); - error = gfs2_write_alloc_required(ip, pos, from - to, &alloc_required); + error = gfs2_write_alloc_required(ip, pos, write_len, &alloc_required); if (error) goto out_unlock; + ip->i_alloc.al_requested = 0; if (alloc_required) { al = gfs2_alloc_get(ip); @@ -482,7 +478,7 @@ static int gfs2_commit_write(struct file *file, struct page *page, kaddr = kmap_atomic(page, KM_USER0); memcpy(dibh->b_data + sizeof(struct gfs2_dinode) + from, kaddr + from, to - from); - kunmap_atomic(page, KM_USER0); + kunmap_atomic(kaddr, KM_USER0); SetPageUptodate(page); diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index 178b33911843..882873a6bd69 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -794,8 +794,8 @@ static int fill_super_meta(struct super_block *sb, struct super_block *new, fs_err(sdp, "can't get root dentry\n"); error = -ENOMEM; iput(inode); - } - new->s_root->d_op = &gfs2_dops; + } else + new->s_root->d_op = &gfs2_dops; return error; } @@ -854,7 +854,6 @@ static int gfs2_get_sb_meta(struct file_system_type *fs_type, int flags, int error = 0; struct super_block *sb = NULL, *new; struct gfs2_sbd *sdp; - char *gfs2mnt = NULL; sb = get_gfs2_sb(dev_name); if (!sb) { @@ -892,8 +891,6 @@ static int gfs2_get_sb_meta(struct file_system_type *fs_type, int flags, atomic_inc(&sdp->sd_gfs2mnt->mnt_count); return simple_set_mnt(mnt, new); error: - if (gfs2mnt) - kfree(gfs2mnt); return error; } diff --git a/fs/gfs2/quota.c b/fs/gfs2/quota.c index c69b94a55588..a3deae7416c9 100644 --- a/fs/gfs2/quota.c +++ b/fs/gfs2/quota.c @@ -251,7 +251,7 @@ static int bh_get(struct gfs2_quota_data *qd) unsigned int block, offset; struct buffer_head *bh; int error; - struct buffer_head bh_map; + struct buffer_head bh_map = { .b_state = 0, .b_blocknr = 0 }; mutex_lock(&sdp->sd_quota_mutex); @@ -263,7 +263,8 @@ static int bh_get(struct gfs2_quota_data *qd) block = qd->qd_slot / sdp->sd_qc_per_block; offset = qd->qd_slot % sdp->sd_qc_per_block;; - error = gfs2_block_map(&ip->i_inode, block, 0, &bh_map, 1); + bh_map.b_size = 1 << ip->i_inode.i_blkbits; + error = gfs2_block_map(&ip->i_inode, block, 0, &bh_map); if (error) goto fail; error = gfs2_meta_read(ip->i_gl, bh_map.b_blocknr, DIO_WAIT, &bh); diff --git a/fs/gfs2/recovery.c b/fs/gfs2/recovery.c index 0a8a4b87dcc6..62cd223819b7 100644 --- a/fs/gfs2/recovery.c +++ b/fs/gfs2/recovery.c @@ -372,11 +372,12 @@ static int clean_journal(struct gfs2_jdesc *jd, struct gfs2_log_header *head) u32 hash; struct buffer_head *bh; int error; - struct buffer_head bh_map; + struct buffer_head bh_map = { .b_state = 0, .b_blocknr = 0 }; lblock = head->lh_blkno; gfs2_replay_incr_blk(sdp, &lblock); - error = gfs2_block_map(&ip->i_inode, lblock, 0, &bh_map, 1); + bh_map.b_size = 1 << ip->i_inode.i_blkbits; + error = gfs2_block_map(&ip->i_inode, lblock, 0, &bh_map); if (error) return error; if (!bh_map.b_blocknr) { diff --git a/fs/gfs2/rgrp.h b/fs/gfs2/rgrp.h index 9eedfd12bfff..b01e0cfc99b5 100644 --- a/fs/gfs2/rgrp.h +++ b/fs/gfs2/rgrp.h @@ -32,7 +32,7 @@ void gfs2_rgrp_repolish_clones(struct gfs2_rgrpd *rgd); struct gfs2_alloc *gfs2_alloc_get(struct gfs2_inode *ip); static inline void gfs2_alloc_put(struct gfs2_inode *ip) { - return; /* Se we can see where ip->i_alloc is used */ + return; /* So we can see where ip->i_alloc is used */ } int gfs2_inplace_reserve_i(struct gfs2_inode *ip, diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 5e03b2f67b93..0bea6a619e10 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -271,29 +271,27 @@ static void hugetlbfs_drop_inode(struct inode *inode) hugetlbfs_forget_inode(inode); } -/* - * h_pgoff is in HPAGE_SIZE units. - * vma->vm_pgoff is in PAGE_SIZE units. - */ static inline void -hugetlb_vmtruncate_list(struct prio_tree_root *root, unsigned long h_pgoff) +hugetlb_vmtruncate_list(struct prio_tree_root *root, pgoff_t pgoff) { struct vm_area_struct *vma; struct prio_tree_iter iter; - vma_prio_tree_foreach(vma, &iter, root, h_pgoff, ULONG_MAX) { - unsigned long h_vm_pgoff; + vma_prio_tree_foreach(vma, &iter, root, pgoff, ULONG_MAX) { unsigned long v_offset; - h_vm_pgoff = vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT); - v_offset = (h_pgoff - h_vm_pgoff) << HPAGE_SHIFT; /* - * Is this VMA fully outside the truncation point? + * Can the expression below overflow on 32-bit arches? + * No, because the prio_tree returns us only those vmas + * which overlap the truncated area starting at pgoff, + * and no vma on a 32-bit arch can span beyond the 4GB. */ - if (h_vm_pgoff >= h_pgoff) + if (vma->vm_pgoff < pgoff) + v_offset = (pgoff - vma->vm_pgoff) << PAGE_SHIFT; + else v_offset = 0; - unmap_hugepage_range(vma, + __unmap_hugepage_range(vma, vma->vm_start + v_offset, vma->vm_end); } } @@ -303,14 +301,14 @@ hugetlb_vmtruncate_list(struct prio_tree_root *root, unsigned long h_pgoff) */ static int hugetlb_vmtruncate(struct inode *inode, loff_t offset) { - unsigned long pgoff; + pgoff_t pgoff; struct address_space *mapping = inode->i_mapping; if (offset > inode->i_size) return -EINVAL; BUG_ON(offset & ~HPAGE_MASK); - pgoff = offset >> HPAGE_SHIFT; + pgoff = offset >> PAGE_SHIFT; inode->i_size = offset; spin_lock(&mapping->i_mmap_lock); @@ -624,7 +622,6 @@ hugetlbfs_parse_options(char *options, struct hugetlbfs_config *pconfig) do_div(size, 100); rest++; } - size &= HPAGE_MASK; pconfig->nr_blocks = (size >> HPAGE_SHIFT); value = rest; } else if (!strcmp(opt,"nr_inodes")) { diff --git a/fs/inode.c b/fs/inode.c index d9a21d122926..26cdb115ce67 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -1306,6 +1306,42 @@ void wake_up_inode(struct inode *inode) wake_up_bit(&inode->i_state, __I_LOCK); } +/* + * We rarely want to lock two inodes that do not have a parent/child + * relationship (such as directory, child inode) simultaneously. The + * vast majority of file systems should be able to get along fine + * without this. Do not use these functions except as a last resort. + */ +void inode_double_lock(struct inode *inode1, struct inode *inode2) +{ + if (inode1 == NULL || inode2 == NULL || inode1 == inode2) { + if (inode1) + mutex_lock(&inode1->i_mutex); + else if (inode2) + mutex_lock(&inode2->i_mutex); + return; + } + + if (inode1 < inode2) { + mutex_lock_nested(&inode1->i_mutex, I_MUTEX_PARENT); + mutex_lock_nested(&inode2->i_mutex, I_MUTEX_CHILD); + } else { + mutex_lock_nested(&inode2->i_mutex, I_MUTEX_PARENT); + mutex_lock_nested(&inode1->i_mutex, I_MUTEX_CHILD); + } +} +EXPORT_SYMBOL(inode_double_lock); + +void inode_double_unlock(struct inode *inode1, struct inode *inode2) +{ + if (inode1) + mutex_unlock(&inode1->i_mutex); + + if (inode2 && inode2 != inode1) + mutex_unlock(&inode2->i_mutex); +} +EXPORT_SYMBOL(inode_double_unlock); + static __initdata unsigned long ihash_entries; static int __init set_ihash_entries(char *str) { diff --git a/fs/ioprio.c b/fs/ioprio.c index 6dc6721d9e82..89e8da112a75 100644 --- a/fs/ioprio.c +++ b/fs/ioprio.c @@ -150,11 +150,6 @@ int ioprio_best(unsigned short aprio, unsigned short bprio) unsigned short aclass = IOPRIO_PRIO_CLASS(aprio); unsigned short bclass = IOPRIO_PRIO_CLASS(bprio); - if (!ioprio_valid(aprio)) - return bprio; - if (!ioprio_valid(bprio)) - return aprio; - if (aclass == IOPRIO_CLASS_NONE) aclass = IOPRIO_CLASS_BE; if (bclass == IOPRIO_CLASS_NONE) diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c index c518dd8fe60a..b85c686b60db 100644 --- a/fs/jbd/journal.c +++ b/fs/jbd/journal.c @@ -725,6 +725,7 @@ journal_t * journal_init_dev(struct block_device *bdev, __FUNCTION__); kfree(journal); journal = NULL; + goto out; } journal->j_dev = bdev; journal->j_fs_dev = fs_dev; @@ -735,7 +736,7 @@ journal_t * journal_init_dev(struct block_device *bdev, J_ASSERT(bh != NULL); journal->j_sb_buffer = bh; journal->j_superblock = (journal_superblock_t *)bh->b_data; - +out: return journal; } diff --git a/fs/jbd/transaction.c b/fs/jbd/transaction.c index e1b3c8af4d17..4f82bcd63e48 100644 --- a/fs/jbd/transaction.c +++ b/fs/jbd/transaction.c @@ -967,6 +967,13 @@ int journal_dirty_data(handle_t *handle, struct buffer_head *bh) */ jbd_lock_bh_state(bh); spin_lock(&journal->j_list_lock); + + /* Now that we have bh_state locked, are we really still mapped? */ + if (!buffer_mapped(bh)) { + JBUFFER_TRACE(jh, "unmapped buffer, bailing out"); + goto no_journal; + } + if (jh->b_transaction) { JBUFFER_TRACE(jh, "has transaction"); if (jh->b_transaction != handle->h_transaction) { @@ -1028,6 +1035,11 @@ int journal_dirty_data(handle_t *handle, struct buffer_head *bh) sync_dirty_buffer(bh); jbd_lock_bh_state(bh); spin_lock(&journal->j_list_lock); + /* Since we dropped the lock... */ + if (!buffer_mapped(bh)) { + JBUFFER_TRACE(jh, "buffer got unmapped"); + goto no_journal; + } /* The buffer may become locked again at any time if it is redirtied */ } @@ -1314,13 +1326,14 @@ int journal_stop(handle_t *handle) int old_handle_count, err; pid_t pid; - J_ASSERT(transaction->t_updates > 0); J_ASSERT(journal_current_handle() == handle); if (is_handle_aborted(handle)) err = -EIO; - else + else { + J_ASSERT(transaction->t_updates > 0); err = 0; + } if (--handle->h_ref > 0) { jbd_debug(4, "h_ref %d -> %d\n", handle->h_ref + 1, @@ -1823,6 +1836,7 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh) } } } else if (transaction == journal->j_committing_transaction) { + JBUFFER_TRACE(jh, "on committing transaction"); if (jh->b_jlist == BJ_Locked) { /* * The buffer is on the committing transaction's locked @@ -1837,7 +1851,6 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh) * can remove it's next_transaction pointer from the * running transaction if that is set, but nothing * else. */ - JBUFFER_TRACE(jh, "on committing transaction"); set_buffer_freed(bh); if (jh->b_next_transaction) { J_ASSERT(jh->b_next_transaction == @@ -1857,6 +1870,7 @@ static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh) * i_size already for this truncate so recovery will not * expose the disk blocks we are discarding here.) */ J_ASSERT_JH(jh, transaction == journal->j_running_transaction); + JBUFFER_TRACE(jh, "on running transaction"); may_free = __dispose_buffer(jh, transaction); } diff --git a/fs/jbd2/Makefile b/fs/jbd2/Makefile new file mode 100644 index 000000000000..802a3413872a --- /dev/null +++ b/fs/jbd2/Makefile @@ -0,0 +1,7 @@ +# +# Makefile for the linux journaling routines. +# + +obj-$(CONFIG_JBD2) += jbd2.o + +jbd2-objs := transaction.o commit.o recovery.o checkpoint.o revoke.o journal.o diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c new file mode 100644 index 000000000000..68039fa9a566 --- /dev/null +++ b/fs/jbd2/checkpoint.c @@ -0,0 +1,697 @@ +/* + * linux/fs/checkpoint.c + * + * Written by Stephen C. Tweedie <sct@redhat.com>, 1999 + * + * Copyright 1999 Red Hat Software --- All Rights Reserved + * + * This file is part of the Linux kernel and is made available under + * the terms of the GNU General Public License, version 2, or at your + * option, any later version, incorporated herein by reference. + * + * Checkpoint routines for the generic filesystem journaling code. + * Part of the ext2fs journaling system. + * + * Checkpointing is the process of ensuring that a section of the log is + * committed fully to disk, so that that portion of the log can be + * reused. + */ + +#include <linux/time.h> +#include <linux/fs.h> +#include <linux/jbd2.h> +#include <linux/errno.h> +#include <linux/slab.h> + +/* + * Unlink a buffer from a transaction checkpoint list. + * + * Called with j_list_lock held. + */ +static inline void __buffer_unlink_first(struct journal_head *jh) +{ + transaction_t *transaction = jh->b_cp_transaction; + + jh->b_cpnext->b_cpprev = jh->b_cpprev; + jh->b_cpprev->b_cpnext = jh->b_cpnext; + if (transaction->t_checkpoint_list == jh) { + transaction->t_checkpoint_list = jh->b_cpnext; + if (transaction->t_checkpoint_list == jh) + transaction->t_checkpoint_list = NULL; + } +} + +/* + * Unlink a buffer from a transaction checkpoint(io) list. + * + * Called with j_list_lock held. + */ +static inline void __buffer_unlink(struct journal_head *jh) +{ + transaction_t *transaction = jh->b_cp_transaction; + + __buffer_unlink_first(jh); + if (transaction->t_checkpoint_io_list == jh) { + transaction->t_checkpoint_io_list = jh->b_cpnext; + if (transaction->t_checkpoint_io_list == jh) + transaction->t_checkpoint_io_list = NULL; + } +} + +/* + * Move a buffer from the checkpoint list to the checkpoint io list + * + * Called with j_list_lock held + */ +static inline void __buffer_relink_io(struct journal_head *jh) +{ + transaction_t *transaction = jh->b_cp_transaction; + + __buffer_unlink_first(jh); + + if (!transaction->t_checkpoint_io_list) { + jh->b_cpnext = jh->b_cpprev = jh; + } else { + jh->b_cpnext = transaction->t_checkpoint_io_list; + jh->b_cpprev = transaction->t_checkpoint_io_list->b_cpprev; + jh->b_cpprev->b_cpnext = jh; + jh->b_cpnext->b_cpprev = jh; + } + transaction->t_checkpoint_io_list = jh; +} + +/* + * Try to release a checkpointed buffer from its transaction. + * Returns 1 if we released it and 2 if we also released the + * whole transaction. + * + * Requires j_list_lock + * Called under jbd_lock_bh_state(jh2bh(jh)), and drops it + */ +static int __try_to_free_cp_buf(struct journal_head *jh) +{ + int ret = 0; + struct buffer_head *bh = jh2bh(jh); + + if (jh->b_jlist == BJ_None && !buffer_locked(bh) && !buffer_dirty(bh)) { + JBUFFER_TRACE(jh, "remove from checkpoint list"); + ret = __jbd2_journal_remove_checkpoint(jh) + 1; + jbd_unlock_bh_state(bh); + jbd2_journal_remove_journal_head(bh); + BUFFER_TRACE(bh, "release"); + __brelse(bh); + } else { + jbd_unlock_bh_state(bh); + } + return ret; +} + +/* + * __jbd2_log_wait_for_space: wait until there is space in the journal. + * + * Called under j-state_lock *only*. It will be unlocked if we have to wait + * for a checkpoint to free up some space in the log. + */ +void __jbd2_log_wait_for_space(journal_t *journal) +{ + int nblocks; + assert_spin_locked(&journal->j_state_lock); + + nblocks = jbd_space_needed(journal); + while (__jbd2_log_space_left(journal) < nblocks) { + if (journal->j_flags & JBD2_ABORT) + return; + spin_unlock(&journal->j_state_lock); + mutex_lock(&journal->j_checkpoint_mutex); + + /* + * Test again, another process may have checkpointed while we + * were waiting for the checkpoint lock + */ + spin_lock(&journal->j_state_lock); + nblocks = jbd_space_needed(journal); + if (__jbd2_log_space_left(journal) < nblocks) { + spin_unlock(&journal->j_state_lock); + jbd2_log_do_checkpoint(journal); + spin_lock(&journal->j_state_lock); + } + mutex_unlock(&journal->j_checkpoint_mutex); + } +} + +/* + * We were unable to perform jbd_trylock_bh_state() inside j_list_lock. + * The caller must restart a list walk. Wait for someone else to run + * jbd_unlock_bh_state(). + */ +static void jbd_sync_bh(journal_t *journal, struct buffer_head *bh) + __releases(journal->j_list_lock) +{ + get_bh(bh); + spin_unlock(&journal->j_list_lock); + jbd_lock_bh_state(bh); + jbd_unlock_bh_state(bh); + put_bh(bh); +} + +/* + * Clean up transaction's list of buffers submitted for io. + * We wait for any pending IO to complete and remove any clean + * buffers. Note that we take the buffers in the opposite ordering + * from the one in which they were submitted for IO. + * + * Called with j_list_lock held. + */ +static void __wait_cp_io(journal_t *journal, transaction_t *transaction) +{ + struct journal_head *jh; + struct buffer_head *bh; + tid_t this_tid; + int released = 0; + + this_tid = transaction->t_tid; +restart: + /* Did somebody clean up the transaction in the meanwhile? */ + if (journal->j_checkpoint_transactions != transaction || + transaction->t_tid != this_tid) + return; + while (!released && transaction->t_checkpoint_io_list) { + jh = transaction->t_checkpoint_io_list; + bh = jh2bh(jh); + if (!jbd_trylock_bh_state(bh)) { + jbd_sync_bh(journal, bh); + spin_lock(&journal->j_list_lock); + goto restart; + } + if (buffer_locked(bh)) { + atomic_inc(&bh->b_count); + spin_unlock(&journal->j_list_lock); + jbd_unlock_bh_state(bh); + wait_on_buffer(bh); + /* the journal_head may have gone by now */ + BUFFER_TRACE(bh, "brelse"); + __brelse(bh); + spin_lock(&journal->j_list_lock); + goto restart; + } + /* + * Now in whatever state the buffer currently is, we know that + * it has been written out and so we can drop it from the list + */ + released = __jbd2_journal_remove_checkpoint(jh); + jbd_unlock_bh_state(bh); + jbd2_journal_remove_journal_head(bh); + __brelse(bh); + } +} + +#define NR_BATCH 64 + +static void +__flush_batch(journal_t *journal, struct buffer_head **bhs, int *batch_count) +{ + int i; + + ll_rw_block(SWRITE, *batch_count, bhs); + for (i = 0; i < *batch_count; i++) { + struct buffer_head *bh = bhs[i]; + clear_buffer_jwrite(bh); + BUFFER_TRACE(bh, "brelse"); + __brelse(bh); + } + *batch_count = 0; +} + +/* + * Try to flush one buffer from the checkpoint list to disk. + * + * Return 1 if something happened which requires us to abort the current + * scan of the checkpoint list. + * + * Called with j_list_lock held and drops it if 1 is returned + * Called under jbd_lock_bh_state(jh2bh(jh)), and drops it + */ +static int __process_buffer(journal_t *journal, struct journal_head *jh, + struct buffer_head **bhs, int *batch_count) +{ + struct buffer_head *bh = jh2bh(jh); + int ret = 0; + + if (buffer_locked(bh)) { + atomic_inc(&bh->b_count); + spin_unlock(&journal->j_list_lock); + jbd_unlock_bh_state(bh); + wait_on_buffer(bh); + /* the journal_head may have gone by now */ + BUFFER_TRACE(bh, "brelse"); + __brelse(bh); + ret = 1; + } else if (jh->b_transaction != NULL) { + transaction_t *t = jh->b_transaction; + tid_t tid = t->t_tid; + + spin_unlock(&journal->j_list_lock); + jbd_unlock_bh_state(bh); + jbd2_log_start_commit(journal, tid); + jbd2_log_wait_commit(journal, tid); + ret = 1; + } else if (!buffer_dirty(bh)) { + J_ASSERT_JH(jh, !buffer_jbddirty(bh)); + BUFFER_TRACE(bh, "remove from checkpoint"); + __jbd2_journal_remove_checkpoint(jh); + spin_unlock(&journal->j_list_lock); + jbd_unlock_bh_state(bh); + jbd2_journal_remove_journal_head(bh); + __brelse(bh); + ret = 1; + } else { + /* + * Important: we are about to write the buffer, and + * possibly block, while still holding the journal lock. + * We cannot afford to let the transaction logic start + * messing around with this buffer before we write it to + * disk, as that would break recoverability. + */ + BUFFER_TRACE(bh, "queue"); + get_bh(bh); + J_ASSERT_BH(bh, !buffer_jwrite(bh)); + set_buffer_jwrite(bh); + bhs[*batch_count] = bh; + __buffer_relink_io(jh); + jbd_unlock_bh_state(bh); + (*batch_count)++; + if (*batch_count == NR_BATCH) { + spin_unlock(&journal->j_list_lock); + __flush_batch(journal, bhs, batch_count); + ret = 1; + } + } + return ret; +} + +/* + * Perform an actual checkpoint. We take the first transaction on the + * list of transactions to be checkpointed and send all its buffers + * to disk. We submit larger chunks of data at once. + * + * The journal should be locked before calling this function. + */ +int jbd2_log_do_checkpoint(journal_t *journal) +{ + transaction_t *transaction; + tid_t this_tid; + int result; + + jbd_debug(1, "Start checkpoint\n"); + + /* + * First thing: if there are any transactions in the log which + * don't need checkpointing, just eliminate them from the + * journal straight away. + */ + result = jbd2_cleanup_journal_tail(journal); + jbd_debug(1, "cleanup_journal_tail returned %d\n", result); + if (result <= 0) + return result; + + /* + * OK, we need to start writing disk blocks. Take one transaction + * and write it. + */ + spin_lock(&journal->j_list_lock); + if (!journal->j_checkpoint_transactions) + goto out; + transaction = journal->j_checkpoint_transactions; + this_tid = transaction->t_tid; +restart: + /* + * If someone cleaned up this transaction while we slept, we're + * done (maybe it's a new transaction, but it fell at the same + * address). + */ + if (journal->j_checkpoint_transactions == transaction && + transaction->t_tid == this_tid) { + int batch_count = 0; + struct buffer_head *bhs[NR_BATCH]; + struct journal_head *jh; + int retry = 0; + + while (!retry && transaction->t_checkpoint_list) { + struct buffer_head *bh; + + jh = transaction->t_checkpoint_list; + bh = jh2bh(jh); + if (!jbd_trylock_bh_state(bh)) { + jbd_sync_bh(journal, bh); + retry = 1; + break; + } + retry = __process_buffer(journal, jh, bhs,&batch_count); + if (!retry && lock_need_resched(&journal->j_list_lock)){ + spin_unlock(&journal->j_list_lock); + retry = 1; + break; + } + } + + if (batch_count) { + if (!retry) { + spin_unlock(&journal->j_list_lock); + retry = 1; + } + __flush_batch(journal, bhs, &batch_count); + } + + if (retry) { + spin_lock(&journal->j_list_lock); + goto restart; + } + /* + * Now we have cleaned up the first transaction's checkpoint + * list. Let's clean up the second one + */ + __wait_cp_io(journal, transaction); + } +out: + spin_unlock(&journal->j_list_lock); + result = jbd2_cleanup_journal_tail(journal); + if (result < 0) + return result; + return 0; +} + +/* + * Check the list of checkpoint transactions for the journal to see if + * we have already got rid of any since the last update of the log tail + * in the journal superblock. If so, we can instantly roll the + * superblock forward to remove those transactions from the log. + * + * Return <0 on error, 0 on success, 1 if there was nothing to clean up. + * + * Called with the journal lock held. + * + * This is the only part of the journaling code which really needs to be + * aware of transaction aborts. Checkpointing involves writing to the + * main filesystem area rather than to the journal, so it can proceed + * even in abort state, but we must not update the journal superblock if + * we have an abort error outstanding. + */ + +int jbd2_cleanup_journal_tail(journal_t *journal) +{ + transaction_t * transaction; + tid_t first_tid; + unsigned long blocknr, freed; + + /* OK, work out the oldest transaction remaining in the log, and + * the log block it starts at. + * + * If the log is now empty, we need to work out which is the + * next transaction ID we will write, and where it will + * start. */ + + spin_lock(&journal->j_state_lock); + spin_lock(&journal->j_list_lock); + transaction = journal->j_checkpoint_transactions; + if (transaction) { + first_tid = transaction->t_tid; + blocknr = transaction->t_log_start; + } else if ((transaction = journal->j_committing_transaction) != NULL) { + first_tid = transaction->t_tid; + blocknr = transaction->t_log_start; + } else if ((transaction = journal->j_running_transaction) != NULL) { + first_tid = transaction->t_tid; + blocknr = journal->j_head; + } else { + first_tid = journal->j_transaction_sequence; + blocknr = journal->j_head; + } + spin_unlock(&journal->j_list_lock); + J_ASSERT(blocknr != 0); + + /* If the oldest pinned transaction is at the tail of the log + already then there's not much we can do right now. */ + if (journal->j_tail_sequence == first_tid) { + spin_unlock(&journal->j_state_lock); + return 1; + } + + /* OK, update the superblock to recover the freed space. + * Physical blocks come first: have we wrapped beyond the end of + * the log? */ + freed = blocknr - journal->j_tail; + if (blocknr < journal->j_tail) + freed = freed + journal->j_last - journal->j_first; + + jbd_debug(1, + "Cleaning journal tail from %d to %d (offset %lu), " + "freeing %lu\n", + journal->j_tail_sequence, first_tid, blocknr, freed); + + journal->j_free += freed; + journal->j_tail_sequence = first_tid; + journal->j_tail = blocknr; + spin_unlock(&journal->j_state_lock); + if (!(journal->j_flags & JBD2_ABORT)) + jbd2_journal_update_superblock(journal, 1); + return 0; +} + + +/* Checkpoint list management */ + +/* + * journal_clean_one_cp_list + * + * Find all the written-back checkpoint buffers in the given list and release them. + * + * Called with the journal locked. + * Called with j_list_lock held. + * Returns number of bufers reaped (for debug) + */ + +static int journal_clean_one_cp_list(struct journal_head *jh, int *released) +{ + struct journal_head *last_jh; + struct journal_head *next_jh = jh; + int ret, freed = 0; + + *released = 0; + if (!jh) + return 0; + + last_jh = jh->b_cpprev; + do { + jh = next_jh; + next_jh = jh->b_cpnext; + /* Use trylock because of the ranking */ + if (jbd_trylock_bh_state(jh2bh(jh))) { + ret = __try_to_free_cp_buf(jh); + if (ret) { + freed++; + if (ret == 2) { + *released = 1; + return freed; + } + } + } + /* + * This function only frees up some memory + * if possible so we dont have an obligation + * to finish processing. Bail out if preemption + * requested: + */ + if (need_resched()) + return freed; + } while (jh != last_jh); + + return freed; +} + +/* + * journal_clean_checkpoint_list + * + * Find all the written-back checkpoint buffers in the journal and release them. + * + * Called with the journal locked. + * Called with j_list_lock held. + * Returns number of buffers reaped (for debug) + */ + +int __jbd2_journal_clean_checkpoint_list(journal_t *journal) +{ + transaction_t *transaction, *last_transaction, *next_transaction; + int ret = 0; + int released; + + transaction = journal->j_checkpoint_transactions; + if (!transaction) + goto out; + + last_transaction = transaction->t_cpprev; + next_transaction = transaction; + do { + transaction = next_transaction; + next_transaction = transaction->t_cpnext; + ret += journal_clean_one_cp_list(transaction-> + t_checkpoint_list, &released); + /* + * This function only frees up some memory if possible so we + * dont have an obligation to finish processing. Bail out if + * preemption requested: + */ + if (need_resched()) + goto out; + if (released) + continue; + /* + * It is essential that we are as careful as in the case of + * t_checkpoint_list with removing the buffer from the list as + * we can possibly see not yet submitted buffers on io_list + */ + ret += journal_clean_one_cp_list(transaction-> + t_checkpoint_io_list, &released); + if (need_resched()) + goto out; + } while (transaction != last_transaction); +out: + return ret; +} + +/* + * journal_remove_checkpoint: called after a buffer has been committed + * to disk (either by being write-back flushed to disk, or being + * committed to the log). + * + * We cannot safely clean a transaction out of the log until all of the + * buffer updates committed in that transaction have safely been stored + * elsewhere on disk. To achieve this, all of the buffers in a + * transaction need to be maintained on the transaction's checkpoint + * lists until they have been rewritten, at which point this function is + * called to remove the buffer from the existing transaction's + * checkpoint lists. + * + * The function returns 1 if it frees the transaction, 0 otherwise. + * + * This function is called with the journal locked. + * This function is called with j_list_lock held. + * This function is called with jbd_lock_bh_state(jh2bh(jh)) + */ + +int __jbd2_journal_remove_checkpoint(struct journal_head *jh) +{ + transaction_t *transaction; + journal_t *journal; + int ret = 0; + + JBUFFER_TRACE(jh, "entry"); + + if ((transaction = jh->b_cp_transaction) == NULL) { + JBUFFER_TRACE(jh, "not on transaction"); + goto out; + } + journal = transaction->t_journal; + + __buffer_unlink(jh); + jh->b_cp_transaction = NULL; + + if (transaction->t_checkpoint_list != NULL || + transaction->t_checkpoint_io_list != NULL) + goto out; + JBUFFER_TRACE(jh, "transaction has no more buffers"); + + /* + * There is one special case to worry about: if we have just pulled the + * buffer off a committing transaction's forget list, then even if the + * checkpoint list is empty, the transaction obviously cannot be + * dropped! + * + * The locking here around j_committing_transaction is a bit sleazy. + * See the comment at the end of jbd2_journal_commit_transaction(). + */ + if (transaction == journal->j_committing_transaction) { + JBUFFER_TRACE(jh, "belongs to committing transaction"); + goto out; + } + + /* OK, that was the last buffer for the transaction: we can now + safely remove this transaction from the log */ + + __jbd2_journal_drop_transaction(journal, transaction); + + /* Just in case anybody was waiting for more transactions to be + checkpointed... */ + wake_up(&journal->j_wait_logspace); + ret = 1; +out: + JBUFFER_TRACE(jh, "exit"); + return ret; +} + +/* + * journal_insert_checkpoint: put a committed buffer onto a checkpoint + * list so that we know when it is safe to clean the transaction out of + * the log. + * + * Called with the journal locked. + * Called with j_list_lock held. + */ +void __jbd2_journal_insert_checkpoint(struct journal_head *jh, + transaction_t *transaction) +{ + JBUFFER_TRACE(jh, "entry"); + J_ASSERT_JH(jh, buffer_dirty(jh2bh(jh)) || buffer_jbddirty(jh2bh(jh))); + J_ASSERT_JH(jh, jh->b_cp_transaction == NULL); + + jh->b_cp_transaction = transaction; + + if (!transaction->t_checkpoint_list) { + jh->b_cpnext = jh->b_cpprev = jh; + } else { + jh->b_cpnext = transaction->t_checkpoint_list; + jh->b_cpprev = transaction->t_checkpoint_list->b_cpprev; + jh->b_cpprev->b_cpnext = jh; + jh->b_cpnext->b_cpprev = jh; + } + transaction->t_checkpoint_list = jh; +} + +/* + * We've finished with this transaction structure: adios... + * + * The transaction must have no links except for the checkpoint by this + * point. + * + * Called with the journal locked. + * Called with j_list_lock held. + */ + +void __jbd2_journal_drop_transaction(journal_t *journal, transaction_t *transaction) +{ + assert_spin_locked(&journal->j_list_lock); + if (transaction->t_cpnext) { + transaction->t_cpnext->t_cpprev = transaction->t_cpprev; + transaction->t_cpprev->t_cpnext = transaction->t_cpnext; + if (journal->j_checkpoint_transactions == transaction) + journal->j_checkpoint_transactions = + transaction->t_cpnext; + if (journal->j_checkpoint_transactions == transaction) + journal->j_checkpoint_transactions = NULL; + } + + J_ASSERT(transaction->t_state == T_FINISHED); + J_ASSERT(transaction->t_buffers == NULL); + J_ASSERT(transaction->t_sync_datalist == NULL); + J_ASSERT(transaction->t_forget == NULL); + J_ASSERT(transaction->t_iobuf_list == NULL); + J_ASSERT(transaction->t_shadow_list == NULL); + J_ASSERT(transaction->t_log_list == NULL); + J_ASSERT(transaction->t_checkpoint_list == NULL); + J_ASSERT(transaction->t_checkpoint_io_list == NULL); + J_ASSERT(transaction->t_updates == 0); + J_ASSERT(journal->j_committing_transaction != transaction); + J_ASSERT(journal->j_running_transaction != transaction); + + jbd_debug(1, "Dropping transaction %d, all done\n", transaction->t_tid); + kfree(transaction); +} diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c new file mode 100644 index 000000000000..70b2ae1ef281 --- /dev/null +++ b/fs/jbd2/commit.c @@ -0,0 +1,920 @@ +/* + * linux/fs/jbd2/commit.c + * + * Written by Stephen C. Tweedie <sct@redhat.com>, 1998 + * + * Copyright 1998 Red Hat corp --- All Rights Reserved + * + * This file is part of the Linux kernel and is made available under + * the terms of the GNU General Public License, version 2, or at your + * option, any later version, incorporated herein by reference. + * + * Journal commit routines for the generic filesystem journaling code; + * part of the ext2fs journaling system. + */ + +#include <linux/time.h> +#include <linux/fs.h> +#include <linux/jbd2.h> +#include <linux/errno.h> +#include <linux/slab.h> +#include <linux/mm.h> +#include <linux/pagemap.h> +#include <linux/smp_lock.h> + +/* + * Default IO end handler for temporary BJ_IO buffer_heads. + */ +static void journal_end_buffer_io_sync(struct buffer_head *bh, int uptodate) +{ + BUFFER_TRACE(bh, ""); + if (uptodate) + set_buffer_uptodate(bh); + else + clear_buffer_uptodate(bh); + unlock_buffer(bh); +} + +/* + * When an ext3-ordered file is truncated, it is possible that many pages are + * not sucessfully freed, because they are attached to a committing transaction. + * After the transaction commits, these pages are left on the LRU, with no + * ->mapping, and with attached buffers. These pages are trivially reclaimable + * by the VM, but their apparent absence upsets the VM accounting, and it makes + * the numbers in /proc/meminfo look odd. + * + * So here, we have a buffer which has just come off the forget list. Look to + * see if we can strip all buffers from the backing page. + * + * Called under lock_journal(), and possibly under journal_datalist_lock. The + * caller provided us with a ref against the buffer, and we drop that here. + */ +static void release_buffer_page(struct buffer_head *bh) +{ + struct page *page; + + if (buffer_dirty(bh)) + goto nope; + if (atomic_read(&bh->b_count) != 1) + goto nope; + page = bh->b_page; + if (!page) + goto nope; + if (page->mapping) + goto nope; + + /* OK, it's a truncated page */ + if (TestSetPageLocked(page)) + goto nope; + + page_cache_get(page); + __brelse(bh); + try_to_free_buffers(page); + unlock_page(page); + page_cache_release(page); + return; + +nope: + __brelse(bh); +} + +/* + * Try to acquire jbd_lock_bh_state() against the buffer, when j_list_lock is + * held. For ranking reasons we must trylock. If we lose, schedule away and + * return 0. j_list_lock is dropped in this case. + */ +static int inverted_lock(journal_t *journal, struct buffer_head *bh) +{ + if (!jbd_trylock_bh_state(bh)) { + spin_unlock(&journal->j_list_lock); + schedule(); + return 0; + } + return 1; +} + +/* Done it all: now write the commit record. We should have + * cleaned up our previous buffers by now, so if we are in abort + * mode we can now just skip the rest of the journal write + * entirely. + * + * Returns 1 if the journal needs to be aborted or 0 on success + */ +static int journal_write_commit_record(journal_t *journal, + transaction_t *commit_transaction) +{ + struct journal_head *descriptor; + struct buffer_head *bh; + int i, ret; + int barrier_done = 0; + + if (is_journal_aborted(journal)) + return 0; + + descriptor = jbd2_journal_get_descriptor_buffer(journal); + if (!descriptor) + return 1; + + bh = jh2bh(descriptor); + + /* AKPM: buglet - add `i' to tmp! */ + for (i = 0; i < bh->b_size; i += 512) { + journal_header_t *tmp = (journal_header_t*)bh->b_data; + tmp->h_magic = cpu_to_be32(JBD2_MAGIC_NUMBER); + tmp->h_blocktype = cpu_to_be32(JBD2_COMMIT_BLOCK); + tmp->h_sequence = cpu_to_be32(commit_transaction->t_tid); + } + + JBUFFER_TRACE(descriptor, "write commit block"); + set_buffer_dirty(bh); + if (journal->j_flags & JBD2_BARRIER) { + set_buffer_ordered(bh); + barrier_done = 1; + } + ret = sync_dirty_buffer(bh); + /* is it possible for another commit to fail at roughly + * the same time as this one? If so, we don't want to + * trust the barrier flag in the super, but instead want + * to remember if we sent a barrier request + */ + if (ret == -EOPNOTSUPP && barrier_done) { + char b[BDEVNAME_SIZE]; + + printk(KERN_WARNING + "JBD: barrier-based sync failed on %s - " + "disabling barriers\n", + bdevname(journal->j_dev, b)); + spin_lock(&journal->j_state_lock); + journal->j_flags &= ~JBD2_BARRIER; + spin_unlock(&journal->j_state_lock); + + /* And try again, without the barrier */ + clear_buffer_ordered(bh); + set_buffer_uptodate(bh); + set_buffer_dirty(bh); + ret = sync_dirty_buffer(bh); + } + put_bh(bh); /* One for getblk() */ + jbd2_journal_put_journal_head(descriptor); + + return (ret == -EIO); +} + +static void journal_do_submit_data(struct buffer_head **wbuf, int bufs) +{ + int i; + + for (i = 0; i < bufs; i++) { + wbuf[i]->b_end_io = end_buffer_write_sync; + /* We use-up our safety reference in submit_bh() */ + submit_bh(WRITE, wbuf[i]); + } +} + +/* + * Submit all the data buffers to disk + */ +static void journal_submit_data_buffers(journal_t *journal, + transaction_t *commit_transaction) +{ + struct journal_head *jh; + struct buffer_head *bh; + int locked; + int bufs = 0; + struct buffer_head **wbuf = journal->j_wbuf; + + /* + * Whenever we unlock the journal and sleep, things can get added + * onto ->t_sync_datalist, so we have to keep looping back to + * write_out_data until we *know* that the list is empty. + * + * Cleanup any flushed data buffers from the data list. Even in + * abort mode, we want to flush this out as soon as possible. + */ +write_out_data: + cond_resched(); + spin_lock(&journal->j_list_lock); + + while (commit_transaction->t_sync_datalist) { + jh = commit_transaction->t_sync_datalist; + bh = jh2bh(jh); + locked = 0; + + /* Get reference just to make sure buffer does not disappear + * when we are forced to drop various locks */ + get_bh(bh); + /* If the buffer is dirty, we need to submit IO and hence + * we need the buffer lock. We try to lock the buffer without + * blocking. If we fail, we need to drop j_list_lock and do + * blocking lock_buffer(). + */ + if (buffer_dirty(bh)) { + if (test_set_buffer_locked(bh)) { + BUFFER_TRACE(bh, "needs blocking lock"); + spin_unlock(&journal->j_list_lock); + /* Write out all data to prevent deadlocks */ + journal_do_submit_data(wbuf, bufs); + bufs = 0; + lock_buffer(bh); + spin_lock(&journal->j_list_lock); + } + locked = 1; + } + /* We have to get bh_state lock. Again out of order, sigh. */ + if (!inverted_lock(journal, bh)) { + jbd_lock_bh_state(bh); + spin_lock(&journal->j_list_lock); + } + /* Someone already cleaned up the buffer? */ + if (!buffer_jbd(bh) + || jh->b_transaction != commit_transaction + || jh->b_jlist != BJ_SyncData) { + jbd_unlock_bh_state(bh); + if (locked) + unlock_buffer(bh); + BUFFER_TRACE(bh, "already cleaned up"); + put_bh(bh); + continue; + } + if (locked && test_clear_buffer_dirty(bh)) { + BUFFER_TRACE(bh, "needs writeout, adding to array"); + wbuf[bufs++] = bh; + __jbd2_journal_file_buffer(jh, commit_transaction, + BJ_Locked); + jbd_unlock_bh_state(bh); + if (bufs == journal->j_wbufsize) { + spin_unlock(&journal->j_list_lock); + journal_do_submit_data(wbuf, bufs); + bufs = 0; + goto write_out_data; + } + } + else { + BUFFER_TRACE(bh, "writeout complete: unfile"); + __jbd2_journal_unfile_buffer(jh); + jbd_unlock_bh_state(bh); + if (locked) + unlock_buffer(bh); + jbd2_journal_remove_journal_head(bh); + /* Once for our safety reference, once for + * jbd2_journal_remove_journal_head() */ + put_bh(bh); + put_bh(bh); + } + + if (lock_need_resched(&journal->j_list_lock)) { + spin_unlock(&journal->j_list_lock); + goto write_out_data; + } + } + spin_unlock(&journal->j_list_lock); + journal_do_submit_data(wbuf, bufs); +} + +static inline void write_tag_block(int tag_bytes, journal_block_tag_t *tag, + unsigned long long block) +{ + tag->t_blocknr = cpu_to_be32(block & (u32)~0); + if (tag_bytes > JBD_TAG_SIZE32) + tag->t_blocknr_high = cpu_to_be32((block >> 31) >> 1); +} + +/* + * jbd2_journal_commit_transaction + * + * The primary function for committing a transaction to the log. This + * function is called by the journal thread to begin a complete commit. + */ +void jbd2_journal_commit_transaction(journal_t *journal) +{ + transaction_t *commit_transaction; + struct journal_head *jh, *new_jh, *descriptor; + struct buffer_head **wbuf = journal->j_wbuf; + int bufs; + int flags; + int err; + unsigned long long blocknr; + char *tagp = NULL; + journal_header_t *header; + journal_block_tag_t *tag = NULL; + int space_left = 0; + int first_tag = 0; + int tag_flag; + int i; + int tag_bytes = journal_tag_bytes(journal); + + /* + * First job: lock down the current transaction and wait for + * all outstanding updates to complete. + */ + +#ifdef COMMIT_STATS + spin_lock(&journal->j_list_lock); + summarise_journal_usage(journal); + spin_unlock(&journal->j_list_lock); +#endif + + /* Do we need to erase the effects of a prior jbd2_journal_flush? */ + if (journal->j_flags & JBD2_FLUSHED) { + jbd_debug(3, "super block updated\n"); + jbd2_journal_update_superblock(journal, 1); + } else { + jbd_debug(3, "superblock not updated\n"); + } + + J_ASSERT(journal->j_running_transaction != NULL); + J_ASSERT(journal->j_committing_transaction == NULL); + + commit_transaction = journal->j_running_transaction; + J_ASSERT(commit_transaction->t_state == T_RUNNING); + + jbd_debug(1, "JBD: starting commit of transaction %d\n", + commit_transaction->t_tid); + + spin_lock(&journal->j_state_lock); + commit_transaction->t_state = T_LOCKED; + + spin_lock(&commit_transaction->t_handle_lock); + while (commit_transaction->t_updates) { + DEFINE_WAIT(wait); + + prepare_to_wait(&journal->j_wait_updates, &wait, + TASK_UNINTERRUPTIBLE); + if (commit_transaction->t_updates) { + spin_unlock(&commit_transaction->t_handle_lock); + spin_unlock(&journal->j_state_lock); + schedule(); + spin_lock(&journal->j_state_lock); + spin_lock(&commit_transaction->t_handle_lock); + } + finish_wait(&journal->j_wait_updates, &wait); + } + spin_unlock(&commit_transaction->t_handle_lock); + + J_ASSERT (commit_transaction->t_outstanding_credits <= + journal->j_max_transaction_buffers); + + /* + * First thing we are allowed to do is to discard any remaining + * BJ_Reserved buffers. Note, it is _not_ permissible to assume + * that there are no such buffers: if a large filesystem + * operation like a truncate needs to split itself over multiple + * transactions, then it may try to do a jbd2_journal_restart() while + * there are still BJ_Reserved buffers outstanding. These must + * be released cleanly from the current transaction. + * + * In this case, the filesystem must still reserve write access + * again before modifying the buffer in the new transaction, but + * we do not require it to remember exactly which old buffers it + * has reserved. This is consistent with the existing behaviour + * that multiple jbd2_journal_get_write_access() calls to the same + * buffer are perfectly permissable. + */ + while (commit_transaction->t_reserved_list) { + jh = commit_transaction->t_reserved_list; + JBUFFER_TRACE(jh, "reserved, unused: refile"); + /* + * A jbd2_journal_get_undo_access()+jbd2_journal_release_buffer() may + * leave undo-committed data. + */ + if (jh->b_committed_data) { + struct buffer_head *bh = jh2bh(jh); + + jbd_lock_bh_state(bh); + jbd2_slab_free(jh->b_committed_data, bh->b_size); + jh->b_committed_data = NULL; + jbd_unlock_bh_state(bh); + } + jbd2_journal_refile_buffer(journal, jh); + } + + /* + * Now try to drop any written-back buffers from the journal's + * checkpoint lists. We do this *before* commit because it potentially + * frees some memory + */ + spin_lock(&journal->j_list_lock); + __jbd2_journal_clean_checkpoint_list(journal); + spin_unlock(&journal->j_list_lock); + + jbd_debug (3, "JBD: commit phase 1\n"); + + /* + * Switch to a new revoke table. + */ + jbd2_journal_switch_revoke_table(journal); + + commit_transaction->t_state = T_FLUSH; + journal->j_committing_transaction = commit_transaction; + journal->j_running_transaction = NULL; + commit_transaction->t_log_start = journal->j_head; + wake_up(&journal->j_wait_transaction_locked); + spin_unlock(&journal->j_state_lock); + + jbd_debug (3, "JBD: commit phase 2\n"); + + /* + * First, drop modified flag: all accesses to the buffers + * will be tracked for a new trasaction only -bzzz + */ + spin_lock(&journal->j_list_lock); + if (commit_transaction->t_buffers) { + new_jh = jh = commit_transaction->t_buffers->b_tnext; + do { + J_ASSERT_JH(new_jh, new_jh->b_modified == 1 || + new_jh->b_modified == 0); + new_jh->b_modified = 0; + new_jh = new_jh->b_tnext; + } while (new_jh != jh); + } + spin_unlock(&journal->j_list_lock); + + /* + * Now start flushing things to disk, in the order they appear + * on the transaction lists. Data blocks go first. + */ + err = 0; + journal_submit_data_buffers(journal, commit_transaction); + + /* + * Wait for all previously submitted IO to complete. + */ + spin_lock(&journal->j_list_lock); + while (commit_transaction->t_locked_list) { + struct buffer_head *bh; + + jh = commit_transaction->t_locked_list->b_tprev; + bh = jh2bh(jh); + get_bh(bh); + if (buffer_locked(bh)) { + spin_unlock(&journal->j_list_lock); + wait_on_buffer(bh); + if (unlikely(!buffer_uptodate(bh))) + err = -EIO; + spin_lock(&journal->j_list_lock); + } + if (!inverted_lock(journal, bh)) { + put_bh(bh); + spin_lock(&journal->j_list_lock); + continue; + } + if (buffer_jbd(bh) && jh->b_jlist == BJ_Locked) { + __jbd2_journal_unfile_buffer(jh); + jbd_unlock_bh_state(bh); + jbd2_journal_remove_journal_head(bh); + put_bh(bh); + } else { + jbd_unlock_bh_state(bh); + } + put_bh(bh); + cond_resched_lock(&journal->j_list_lock); + } + spin_unlock(&journal->j_list_lock); + + if (err) + __jbd2_journal_abort_hard(journal); + + jbd2_journal_write_revoke_records(journal, commit_transaction); + + jbd_debug(3, "JBD: commit phase 2\n"); + + /* + * If we found any dirty or locked buffers, then we should have + * looped back up to the write_out_data label. If there weren't + * any then journal_clean_data_list should have wiped the list + * clean by now, so check that it is in fact empty. + */ + J_ASSERT (commit_transaction->t_sync_datalist == NULL); + + jbd_debug (3, "JBD: commit phase 3\n"); + + /* + * Way to go: we have now written out all of the data for a + * transaction! Now comes the tricky part: we need to write out + * metadata. Loop over the transaction's entire buffer list: + */ + commit_transaction->t_state = T_COMMIT; + + descriptor = NULL; + bufs = 0; + while (commit_transaction->t_buffers) { + + /* Find the next buffer to be journaled... */ + + jh = commit_transaction->t_buffers; + + /* If we're in abort mode, we just un-journal the buffer and + release it for background writing. */ + + if (is_journal_aborted(journal)) { + JBUFFER_TRACE(jh, "journal is aborting: refile"); + jbd2_journal_refile_buffer(journal, jh); + /* If that was the last one, we need to clean up + * any descriptor buffers which may have been + * already allocated, even if we are now + * aborting. */ + if (!commit_transaction->t_buffers) + goto start_journal_io; + continue; + } + + /* Make sure we have a descriptor block in which to + record the metadata buffer. */ + + if (!descriptor) { + struct buffer_head *bh; + + J_ASSERT (bufs == 0); + + jbd_debug(4, "JBD: get descriptor\n"); + + descriptor = jbd2_journal_get_descriptor_buffer(journal); + if (!descriptor) { + __jbd2_journal_abort_hard(journal); + continue; + } + + bh = jh2bh(descriptor); + jbd_debug(4, "JBD: got buffer %llu (%p)\n", + (unsigned long long)bh->b_blocknr, bh->b_data); + header = (journal_header_t *)&bh->b_data[0]; + header->h_magic = cpu_to_be32(JBD2_MAGIC_NUMBER); + header->h_blocktype = cpu_to_be32(JBD2_DESCRIPTOR_BLOCK); + header->h_sequence = cpu_to_be32(commit_transaction->t_tid); + + tagp = &bh->b_data[sizeof(journal_header_t)]; + space_left = bh->b_size - sizeof(journal_header_t); + first_tag = 1; + set_buffer_jwrite(bh); + set_buffer_dirty(bh); + wbuf[bufs++] = bh; + + /* Record it so that we can wait for IO + completion later */ + BUFFER_TRACE(bh, "ph3: file as descriptor"); + jbd2_journal_file_buffer(descriptor, commit_transaction, + BJ_LogCtl); + } + + /* Where is the buffer to be written? */ + + err = jbd2_journal_next_log_block(journal, &blocknr); + /* If the block mapping failed, just abandon the buffer + and repeat this loop: we'll fall into the + refile-on-abort condition above. */ + if (err) { + __jbd2_journal_abort_hard(journal); + continue; + } + + /* + * start_this_handle() uses t_outstanding_credits to determine + * the free space in the log, but this counter is changed + * by jbd2_journal_next_log_block() also. + */ + commit_transaction->t_outstanding_credits--; + + /* Bump b_count to prevent truncate from stumbling over + the shadowed buffer! @@@ This can go if we ever get + rid of the BJ_IO/BJ_Shadow pairing of buffers. */ + atomic_inc(&jh2bh(jh)->b_count); + + /* Make a temporary IO buffer with which to write it out + (this will requeue both the metadata buffer and the + temporary IO buffer). new_bh goes on BJ_IO*/ + + set_bit(BH_JWrite, &jh2bh(jh)->b_state); + /* + * akpm: jbd2_journal_write_metadata_buffer() sets + * new_bh->b_transaction to commit_transaction. + * We need to clean this up before we release new_bh + * (which is of type BJ_IO) + */ + JBUFFER_TRACE(jh, "ph3: write metadata"); + flags = jbd2_journal_write_metadata_buffer(commit_transaction, + jh, &new_jh, blocknr); + set_bit(BH_JWrite, &jh2bh(new_jh)->b_state); + wbuf[bufs++] = jh2bh(new_jh); + + /* Record the new block's tag in the current descriptor + buffer */ + + tag_flag = 0; + if (flags & 1) + tag_flag |= JBD2_FLAG_ESCAPE; + if (!first_tag) + tag_flag |= JBD2_FLAG_SAME_UUID; + + tag = (journal_block_tag_t *) tagp; + write_tag_block(tag_bytes, tag, jh2bh(jh)->b_blocknr); + tag->t_flags = cpu_to_be32(tag_flag); + tagp += tag_bytes; + space_left -= tag_bytes; + + if (first_tag) { + memcpy (tagp, journal->j_uuid, 16); + tagp += 16; + space_left -= 16; + first_tag = 0; + } + + /* If there's no more to do, or if the descriptor is full, + let the IO rip! */ + + if (bufs == journal->j_wbufsize || + commit_transaction->t_buffers == NULL || + space_left < tag_bytes + 16) { + + jbd_debug(4, "JBD: Submit %d IOs\n", bufs); + + /* Write an end-of-descriptor marker before + submitting the IOs. "tag" still points to + the last tag we set up. */ + + tag->t_flags |= cpu_to_be32(JBD2_FLAG_LAST_TAG); + +start_journal_io: + for (i = 0; i < bufs; i++) { + struct buffer_head *bh = wbuf[i]; + lock_buffer(bh); + clear_buffer_dirty(bh); + set_buffer_uptodate(bh); + bh->b_end_io = journal_end_buffer_io_sync; + submit_bh(WRITE, bh); + } + cond_resched(); + + /* Force a new descriptor to be generated next + time round the loop. */ + descriptor = NULL; + bufs = 0; + } + } + + /* Lo and behold: we have just managed to send a transaction to + the log. Before we can commit it, wait for the IO so far to + complete. Control buffers being written are on the + transaction's t_log_list queue, and metadata buffers are on + the t_iobuf_list queue. + + Wait for the buffers in reverse order. That way we are + less likely to be woken up until all IOs have completed, and + so we incur less scheduling load. + */ + + jbd_debug(3, "JBD: commit phase 4\n"); + + /* + * akpm: these are BJ_IO, and j_list_lock is not needed. + * See __journal_try_to_free_buffer. + */ +wait_for_iobuf: + while (commit_transaction->t_iobuf_list != NULL) { + struct buffer_head *bh; + + jh = commit_transaction->t_iobuf_list->b_tprev; + bh = jh2bh(jh); + if (buffer_locked(bh)) { + wait_on_buffer(bh); + goto wait_for_iobuf; + } + if (cond_resched()) + goto wait_for_iobuf; + + if (unlikely(!buffer_uptodate(bh))) + err = -EIO; + + clear_buffer_jwrite(bh); + + JBUFFER_TRACE(jh, "ph4: unfile after journal write"); + jbd2_journal_unfile_buffer(journal, jh); + + /* + * ->t_iobuf_list should contain only dummy buffer_heads + * which were created by jbd2_journal_write_metadata_buffer(). + */ + BUFFER_TRACE(bh, "dumping temporary bh"); + jbd2_journal_put_journal_head(jh); + __brelse(bh); + J_ASSERT_BH(bh, atomic_read(&bh->b_count) == 0); + free_buffer_head(bh); + + /* We also have to unlock and free the corresponding + shadowed buffer */ + jh = commit_transaction->t_shadow_list->b_tprev; + bh = jh2bh(jh); + clear_bit(BH_JWrite, &bh->b_state); + J_ASSERT_BH(bh, buffer_jbddirty(bh)); + + /* The metadata is now released for reuse, but we need + to remember it against this transaction so that when + we finally commit, we can do any checkpointing + required. */ + JBUFFER_TRACE(jh, "file as BJ_Forget"); + jbd2_journal_file_buffer(jh, commit_transaction, BJ_Forget); + /* Wake up any transactions which were waiting for this + IO to complete */ + wake_up_bit(&bh->b_state, BH_Unshadow); + JBUFFER_TRACE(jh, "brelse shadowed buffer"); + __brelse(bh); + } + + J_ASSERT (commit_transaction->t_shadow_list == NULL); + + jbd_debug(3, "JBD: commit phase 5\n"); + + /* Here we wait for the revoke record and descriptor record buffers */ + wait_for_ctlbuf: + while (commit_transaction->t_log_list != NULL) { + struct buffer_head *bh; + + jh = commit_transaction->t_log_list->b_tprev; + bh = jh2bh(jh); + if (buffer_locked(bh)) { + wait_on_buffer(bh); + goto wait_for_ctlbuf; + } + if (cond_resched()) + goto wait_for_ctlbuf; + + if (unlikely(!buffer_uptodate(bh))) + err = -EIO; + + BUFFER_TRACE(bh, "ph5: control buffer writeout done: unfile"); + clear_buffer_jwrite(bh); + jbd2_journal_unfile_buffer(journal, jh); + jbd2_journal_put_journal_head(jh); + __brelse(bh); /* One for getblk */ + /* AKPM: bforget here */ + } + + jbd_debug(3, "JBD: commit phase 6\n"); + + if (journal_write_commit_record(journal, commit_transaction)) + err = -EIO; + + if (err) + __jbd2_journal_abort_hard(journal); + + /* End of a transaction! Finally, we can do checkpoint + processing: any buffers committed as a result of this + transaction can be removed from any checkpoint list it was on + before. */ + + jbd_debug(3, "JBD: commit phase 7\n"); + + J_ASSERT(commit_transaction->t_sync_datalist == NULL); + J_ASSERT(commit_transaction->t_buffers == NULL); + J_ASSERT(commit_transaction->t_checkpoint_list == NULL); + J_ASSERT(commit_transaction->t_iobuf_list == NULL); + J_ASSERT(commit_transaction->t_shadow_list == NULL); + J_ASSERT(commit_transaction->t_log_list == NULL); + +restart_loop: + /* + * As there are other places (journal_unmap_buffer()) adding buffers + * to this list we have to be careful and hold the j_list_lock. + */ + spin_lock(&journal->j_list_lock); + while (commit_transaction->t_forget) { + transaction_t *cp_transaction; + struct buffer_head *bh; + + jh = commit_transaction->t_forget; + spin_unlock(&journal->j_list_lock); + bh = jh2bh(jh); + jbd_lock_bh_state(bh); + J_ASSERT_JH(jh, jh->b_transaction == commit_transaction || + jh->b_transaction == journal->j_running_transaction); + + /* + * If there is undo-protected committed data against + * this buffer, then we can remove it now. If it is a + * buffer needing such protection, the old frozen_data + * field now points to a committed version of the + * buffer, so rotate that field to the new committed + * data. + * + * Otherwise, we can just throw away the frozen data now. + */ + if (jh->b_committed_data) { + jbd2_slab_free(jh->b_committed_data, bh->b_size); + jh->b_committed_data = NULL; + if (jh->b_frozen_data) { + jh->b_committed_data = jh->b_frozen_data; + jh->b_frozen_data = NULL; + } + } else if (jh->b_frozen_data) { + jbd2_slab_free(jh->b_frozen_data, bh->b_size); + jh->b_frozen_data = NULL; + } + + spin_lock(&journal->j_list_lock); + cp_transaction = jh->b_cp_transaction; + if (cp_transaction) { + JBUFFER_TRACE(jh, "remove from old cp transaction"); + __jbd2_journal_remove_checkpoint(jh); + } + + /* Only re-checkpoint the buffer_head if it is marked + * dirty. If the buffer was added to the BJ_Forget list + * by jbd2_journal_forget, it may no longer be dirty and + * there's no point in keeping a checkpoint record for + * it. */ + + /* A buffer which has been freed while still being + * journaled by a previous transaction may end up still + * being dirty here, but we want to avoid writing back + * that buffer in the future now that the last use has + * been committed. That's not only a performance gain, + * it also stops aliasing problems if the buffer is left + * behind for writeback and gets reallocated for another + * use in a different page. */ + if (buffer_freed(bh)) { + clear_buffer_freed(bh); + clear_buffer_jbddirty(bh); + } + + if (buffer_jbddirty(bh)) { + JBUFFER_TRACE(jh, "add to new checkpointing trans"); + __jbd2_journal_insert_checkpoint(jh, commit_transaction); + JBUFFER_TRACE(jh, "refile for checkpoint writeback"); + __jbd2_journal_refile_buffer(jh); + jbd_unlock_bh_state(bh); + } else { + J_ASSERT_BH(bh, !buffer_dirty(bh)); + /* The buffer on BJ_Forget list and not jbddirty means + * it has been freed by this transaction and hence it + * could not have been reallocated until this + * transaction has committed. *BUT* it could be + * reallocated once we have written all the data to + * disk and before we process the buffer on BJ_Forget + * list. */ + JBUFFER_TRACE(jh, "refile or unfile freed buffer"); + __jbd2_journal_refile_buffer(jh); + if (!jh->b_transaction) { + jbd_unlock_bh_state(bh); + /* needs a brelse */ + jbd2_journal_remove_journal_head(bh); + release_buffer_page(bh); + } else + jbd_unlock_bh_state(bh); + } + cond_resched_lock(&journal->j_list_lock); + } + spin_unlock(&journal->j_list_lock); + /* + * This is a bit sleazy. We borrow j_list_lock to protect + * journal->j_committing_transaction in __jbd2_journal_remove_checkpoint. + * Really, __jbd2_journal_remove_checkpoint should be using j_state_lock but + * it's a bit hassle to hold that across __jbd2_journal_remove_checkpoint + */ + spin_lock(&journal->j_state_lock); + spin_lock(&journal->j_list_lock); + /* + * Now recheck if some buffers did not get attached to the transaction + * while the lock was dropped... + */ + if (commit_transaction->t_forget) { + spin_unlock(&journal->j_list_lock); + spin_unlock(&journal->j_state_lock); + goto restart_loop; + } + + /* Done with this transaction! */ + + jbd_debug(3, "JBD: commit phase 8\n"); + + J_ASSERT(commit_transaction->t_state == T_COMMIT); + + commit_transaction->t_state = T_FINISHED; + J_ASSERT(commit_transaction == journal->j_committing_transaction); + journal->j_commit_sequence = commit_transaction->t_tid; + journal->j_committing_transaction = NULL; + spin_unlock(&journal->j_state_lock); + + if (commit_transaction->t_checkpoint_list == NULL) { + __jbd2_journal_drop_transaction(journal, commit_transaction); + } else { + if (journal->j_checkpoint_transactions == NULL) { + journal->j_checkpoint_transactions = commit_transaction; + commit_transaction->t_cpnext = commit_transaction; + commit_transaction->t_cpprev = commit_transaction; + } else { + commit_transaction->t_cpnext = + journal->j_checkpoint_transactions; + commit_transaction->t_cpprev = + commit_transaction->t_cpnext->t_cpprev; + commit_transaction->t_cpnext->t_cpprev = + commit_transaction; + commit_transaction->t_cpprev->t_cpnext = + commit_transaction; + } + } + spin_unlock(&journal->j_list_lock); + + jbd_debug(1, "JBD: commit %d complete, head %d\n", + journal->j_commit_sequence, journal->j_tail_sequence); + + wake_up(&journal->j_wait_done_commit); +} diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c new file mode 100644 index 000000000000..c60f378b0f76 --- /dev/null +++ b/fs/jbd2/journal.c @@ -0,0 +1,2084 @@ +/* + * linux/fs/jbd2/journal.c + * + * Written by Stephen C. Tweedie <sct@redhat.com>, 1998 + * + * Copyright 1998 Red Hat corp --- All Rights Reserved + * + * This file is part of the Linux kernel and is made available under + * the terms of the GNU General Public License, version 2, or at your + * option, any later version, incorporated herein by reference. + * + * Generic filesystem journal-writing code; part of the ext2fs + * journaling system. + * + * This file manages journals: areas of disk reserved for logging + * transactional updates. This includes the kernel journaling thread + * which is responsible for scheduling updates to the log. + * + * We do not actually manage the physical storage of the journal in this + * file: that is left to a per-journal policy function, which allows us + * to store the journal within a filesystem-specified area for ext2 + * journaling (ext2 can use a reserved inode for storing the log). + */ + +#include <linux/module.h> +#include <linux/time.h> +#include <linux/fs.h> +#include <linux/jbd2.h> +#include <linux/errno.h> +#include <linux/slab.h> +#include <linux/smp_lock.h> +#include <linux/init.h> +#include <linux/mm.h> +#include <linux/suspend.h> +#include <linux/pagemap.h> +#include <linux/kthread.h> +#include <linux/poison.h> +#include <linux/proc_fs.h> + +#include <asm/uaccess.h> +#include <asm/page.h> + +EXPORT_SYMBOL(jbd2_journal_start); +EXPORT_SYMBOL(jbd2_journal_restart); +EXPORT_SYMBOL(jbd2_journal_extend); +EXPORT_SYMBOL(jbd2_journal_stop); +EXPORT_SYMBOL(jbd2_journal_lock_updates); +EXPORT_SYMBOL(jbd2_journal_unlock_updates); +EXPORT_SYMBOL(jbd2_journal_get_write_access); +EXPORT_SYMBOL(jbd2_journal_get_create_access); +EXPORT_SYMBOL(jbd2_journal_get_undo_access); +EXPORT_SYMBOL(jbd2_journal_dirty_data); +EXPORT_SYMBOL(jbd2_journal_dirty_metadata); +EXPORT_SYMBOL(jbd2_journal_release_buffer); +EXPORT_SYMBOL(jbd2_journal_forget); +#if 0 +EXPORT_SYMBOL(journal_sync_buffer); +#endif +EXPORT_SYMBOL(jbd2_journal_flush); +EXPORT_SYMBOL(jbd2_journal_revoke); + +EXPORT_SYMBOL(jbd2_journal_init_dev); +EXPORT_SYMBOL(jbd2_journal_init_inode); +EXPORT_SYMBOL(jbd2_journal_update_format); +EXPORT_SYMBOL(jbd2_journal_check_used_features); +EXPORT_SYMBOL(jbd2_journal_check_available_features); +EXPORT_SYMBOL(jbd2_journal_set_features); +EXPORT_SYMBOL(jbd2_journal_create); +EXPORT_SYMBOL(jbd2_journal_load); +EXPORT_SYMBOL(jbd2_journal_destroy); +EXPORT_SYMBOL(jbd2_journal_update_superblock); +EXPORT_SYMBOL(jbd2_journal_abort); +EXPORT_SYMBOL(jbd2_journal_errno); +EXPORT_SYMBOL(jbd2_journal_ack_err); +EXPORT_SYMBOL(jbd2_journal_clear_err); +EXPORT_SYMBOL(jbd2_log_wait_commit); +EXPORT_SYMBOL(jbd2_journal_start_commit); +EXPORT_SYMBOL(jbd2_journal_force_commit_nested); +EXPORT_SYMBOL(jbd2_journal_wipe); +EXPORT_SYMBOL(jbd2_journal_blocks_per_page); +EXPORT_SYMBOL(jbd2_journal_invalidatepage); +EXPORT_SYMBOL(jbd2_journal_try_to_free_buffers); +EXPORT_SYMBOL(jbd2_journal_force_commit); + +static int journal_convert_superblock_v1(journal_t *, journal_superblock_t *); +static void __journal_abort_soft (journal_t *journal, int errno); +static int jbd2_journal_create_jbd_slab(size_t slab_size); + +/* + * Helper function used to manage commit timeouts + */ + +static void commit_timeout(unsigned long __data) +{ + struct task_struct * p = (struct task_struct *) __data; + + wake_up_process(p); +} + +/* + * kjournald2: The main thread function used to manage a logging device + * journal. + * + * This kernel thread is responsible for two things: + * + * 1) COMMIT: Every so often we need to commit the current state of the + * filesystem to disk. The journal thread is responsible for writing + * all of the metadata buffers to disk. + * + * 2) CHECKPOINT: We cannot reuse a used section of the log file until all + * of the data in that part of the log has been rewritten elsewhere on + * the disk. Flushing these old buffers to reclaim space in the log is + * known as checkpointing, and this thread is responsible for that job. + */ + +static int kjournald2(void *arg) +{ + journal_t *journal = arg; + transaction_t *transaction; + + /* + * Set up an interval timer which can be used to trigger a commit wakeup + * after the commit interval expires + */ + setup_timer(&journal->j_commit_timer, commit_timeout, + (unsigned long)current); + + /* Record that the journal thread is running */ + journal->j_task = current; + wake_up(&journal->j_wait_done_commit); + + printk(KERN_INFO "kjournald2 starting. Commit interval %ld seconds\n", + journal->j_commit_interval / HZ); + + /* + * And now, wait forever for commit wakeup events. + */ + spin_lock(&journal->j_state_lock); + +loop: + if (journal->j_flags & JBD2_UNMOUNT) + goto end_loop; + + jbd_debug(1, "commit_sequence=%d, commit_request=%d\n", + journal->j_commit_sequence, journal->j_commit_request); + + if (journal->j_commit_sequence != journal->j_commit_request) { + jbd_debug(1, "OK, requests differ\n"); + spin_unlock(&journal->j_state_lock); + del_timer_sync(&journal->j_commit_timer); + jbd2_journal_commit_transaction(journal); + spin_lock(&journal->j_state_lock); + goto loop; + } + + wake_up(&journal->j_wait_done_commit); + if (freezing(current)) { + /* + * The simpler the better. Flushing journal isn't a + * good idea, because that depends on threads that may + * be already stopped. + */ + jbd_debug(1, "Now suspending kjournald2\n"); + spin_unlock(&journal->j_state_lock); + refrigerator(); + spin_lock(&journal->j_state_lock); + } else { + /* + * We assume on resume that commits are already there, + * so we don't sleep + */ + DEFINE_WAIT(wait); + int should_sleep = 1; + + prepare_to_wait(&journal->j_wait_commit, &wait, + TASK_INTERRUPTIBLE); + if (journal->j_commit_sequence != journal->j_commit_request) + should_sleep = 0; + transaction = journal->j_running_transaction; + if (transaction && time_after_eq(jiffies, + transaction->t_expires)) + should_sleep = 0; + if (journal->j_flags & JBD2_UNMOUNT) + should_sleep = 0; + if (should_sleep) { + spin_unlock(&journal->j_state_lock); + schedule(); + spin_lock(&journal->j_state_lock); + } + finish_wait(&journal->j_wait_commit, &wait); + } + + jbd_debug(1, "kjournald2 wakes\n"); + + /* + * Were we woken up by a commit wakeup event? + */ + transaction = journal->j_running_transaction; + if (transaction && time_after_eq(jiffies, transaction->t_expires)) { + journal->j_commit_request = transaction->t_tid; + jbd_debug(1, "woke because of timeout\n"); + } + goto loop; + +end_loop: + spin_unlock(&journal->j_state_lock); + del_timer_sync(&journal->j_commit_timer); + journal->j_task = NULL; + wake_up(&journal->j_wait_done_commit); + jbd_debug(1, "Journal thread exiting.\n"); + return 0; +} + +static void jbd2_journal_start_thread(journal_t *journal) +{ + kthread_run(kjournald2, journal, "kjournald2"); + wait_event(journal->j_wait_done_commit, journal->j_task != 0); +} + +static void journal_kill_thread(journal_t *journal) +{ + spin_lock(&journal->j_state_lock); + journal->j_flags |= JBD2_UNMOUNT; + + while (journal->j_task) { + wake_up(&journal->j_wait_commit); + spin_unlock(&journal->j_state_lock); + wait_event(journal->j_wait_done_commit, journal->j_task == 0); + spin_lock(&journal->j_state_lock); + } + spin_unlock(&journal->j_state_lock); +} + +/* + * jbd2_journal_write_metadata_buffer: write a metadata buffer to the journal. + * + * Writes a metadata buffer to a given disk block. The actual IO is not + * performed but a new buffer_head is constructed which labels the data + * to be written with the correct destination disk block. + * + * Any magic-number escaping which needs to be done will cause a + * copy-out here. If the buffer happens to start with the + * JBD2_MAGIC_NUMBER, then we can't write it to the log directly: the + * magic number is only written to the log for descripter blocks. In + * this case, we copy the data and replace the first word with 0, and we + * return a result code which indicates that this buffer needs to be + * marked as an escaped buffer in the corresponding log descriptor + * block. The missing word can then be restored when the block is read + * during recovery. + * + * If the source buffer has already been modified by a new transaction + * since we took the last commit snapshot, we use the frozen copy of + * that data for IO. If we end up using the existing buffer_head's data + * for the write, then we *have* to lock the buffer to prevent anyone + * else from using and possibly modifying it while the IO is in + * progress. + * + * The function returns a pointer to the buffer_heads to be used for IO. + * + * We assume that the journal has already been locked in this function. + * + * Return value: + * <0: Error + * >=0: Finished OK + * + * On success: + * Bit 0 set == escape performed on the data + * Bit 1 set == buffer copy-out performed (kfree the data after IO) + */ + +int jbd2_journal_write_metadata_buffer(transaction_t *transaction, + struct journal_head *jh_in, + struct journal_head **jh_out, + unsigned long long blocknr) +{ + int need_copy_out = 0; + int done_copy_out = 0; + int do_escape = 0; + char *mapped_data; + struct buffer_head *new_bh; + struct journal_head *new_jh; + struct page *new_page; + unsigned int new_offset; + struct buffer_head *bh_in = jh2bh(jh_in); + + /* + * The buffer really shouldn't be locked: only the current committing + * transaction is allowed to write it, so nobody else is allowed + * to do any IO. + * + * akpm: except if we're journalling data, and write() output is + * also part of a shared mapping, and another thread has + * decided to launch a writepage() against this buffer. + */ + J_ASSERT_BH(bh_in, buffer_jbddirty(bh_in)); + + new_bh = alloc_buffer_head(GFP_NOFS|__GFP_NOFAIL); + + /* + * If a new transaction has already done a buffer copy-out, then + * we use that version of the data for the commit. + */ + jbd_lock_bh_state(bh_in); +repeat: + if (jh_in->b_frozen_data) { + done_copy_out = 1; + new_page = virt_to_page(jh_in->b_frozen_data); + new_offset = offset_in_page(jh_in->b_frozen_data); + } else { + new_page = jh2bh(jh_in)->b_page; + new_offset = offset_in_page(jh2bh(jh_in)->b_data); + } + + mapped_data = kmap_atomic(new_page, KM_USER0); + /* + * Check for escaping + */ + if (*((__be32 *)(mapped_data + new_offset)) == + cpu_to_be32(JBD2_MAGIC_NUMBER)) { + need_copy_out = 1; + do_escape = 1; + } + kunmap_atomic(mapped_data, KM_USER0); + + /* + * Do we need to do a data copy? + */ + if (need_copy_out && !done_copy_out) { + char *tmp; + + jbd_unlock_bh_state(bh_in); + tmp = jbd2_slab_alloc(bh_in->b_size, GFP_NOFS); + jbd_lock_bh_state(bh_in); + if (jh_in->b_frozen_data) { + jbd2_slab_free(tmp, bh_in->b_size); + goto repeat; + } + + jh_in->b_frozen_data = tmp; + mapped_data = kmap_atomic(new_page, KM_USER0); + memcpy(tmp, mapped_data + new_offset, jh2bh(jh_in)->b_size); + kunmap_atomic(mapped_data, KM_USER0); + + new_page = virt_to_page(tmp); + new_offset = offset_in_page(tmp); + done_copy_out = 1; + } + + /* + * Did we need to do an escaping? Now we've done all the + * copying, we can finally do so. + */ + if (do_escape) { + mapped_data = kmap_atomic(new_page, KM_USER0); + *((unsigned int *)(mapped_data + new_offset)) = 0; + kunmap_atomic(mapped_data, KM_USER0); + } + + /* keep subsequent assertions sane */ + new_bh->b_state = 0; + init_buffer(new_bh, NULL, NULL); + atomic_set(&new_bh->b_count, 1); + jbd_unlock_bh_state(bh_in); + + new_jh = jbd2_journal_add_journal_head(new_bh); /* This sleeps */ + + set_bh_page(new_bh, new_page, new_offset); + new_jh->b_transaction = NULL; + new_bh->b_size = jh2bh(jh_in)->b_size; + new_bh->b_bdev = transaction->t_journal->j_dev; + new_bh->b_blocknr = blocknr; + set_buffer_mapped(new_bh); + set_buffer_dirty(new_bh); + + *jh_out = new_jh; + + /* + * The to-be-written buffer needs to get moved to the io queue, + * and the original buffer whose contents we are shadowing or + * copying is moved to the transaction's shadow queue. + */ + JBUFFER_TRACE(jh_in, "file as BJ_Shadow"); + jbd2_journal_file_buffer(jh_in, transaction, BJ_Shadow); + JBUFFER_TRACE(new_jh, "file as BJ_IO"); + jbd2_journal_file_buffer(new_jh, transaction, BJ_IO); + + return do_escape | (done_copy_out << 1); +} + +/* + * Allocation code for the journal file. Manage the space left in the + * journal, so that we can begin checkpointing when appropriate. + */ + +/* + * __jbd2_log_space_left: Return the number of free blocks left in the journal. + * + * Called with the journal already locked. + * + * Called under j_state_lock + */ + +int __jbd2_log_space_left(journal_t *journal) +{ + int left = journal->j_free; + + assert_spin_locked(&journal->j_state_lock); + + /* + * Be pessimistic here about the number of those free blocks which + * might be required for log descriptor control blocks. + */ + +#define MIN_LOG_RESERVED_BLOCKS 32 /* Allow for rounding errors */ + + left -= MIN_LOG_RESERVED_BLOCKS; + + if (left <= 0) + return 0; + left -= (left >> 3); + return left; +} + +/* + * Called under j_state_lock. Returns true if a transaction was started. + */ +int __jbd2_log_start_commit(journal_t *journal, tid_t target) +{ + /* + * Are we already doing a recent enough commit? + */ + if (!tid_geq(journal->j_commit_request, target)) { + /* + * We want a new commit: OK, mark the request and wakup the + * commit thread. We do _not_ do the commit ourselves. + */ + + journal->j_commit_request = target; + jbd_debug(1, "JBD: requesting commit %d/%d\n", + journal->j_commit_request, + journal->j_commit_sequence); + wake_up(&journal->j_wait_commit); + return 1; + } + return 0; +} + +int jbd2_log_start_commit(journal_t *journal, tid_t tid) +{ + int ret; + + spin_lock(&journal->j_state_lock); + ret = __jbd2_log_start_commit(journal, tid); + spin_unlock(&journal->j_state_lock); + return ret; +} + +/* + * Force and wait upon a commit if the calling process is not within + * transaction. This is used for forcing out undo-protected data which contains + * bitmaps, when the fs is running out of space. + * + * We can only force the running transaction if we don't have an active handle; + * otherwise, we will deadlock. + * + * Returns true if a transaction was started. + */ +int jbd2_journal_force_commit_nested(journal_t *journal) +{ + transaction_t *transaction = NULL; + tid_t tid; + + spin_lock(&journal->j_state_lock); + if (journal->j_running_transaction && !current->journal_info) { + transaction = journal->j_running_transaction; + __jbd2_log_start_commit(journal, transaction->t_tid); + } else if (journal->j_committing_transaction) + transaction = journal->j_committing_transaction; + + if (!transaction) { + spin_unlock(&journal->j_state_lock); + return 0; /* Nothing to retry */ + } + + tid = transaction->t_tid; + spin_unlock(&journal->j_state_lock); + jbd2_log_wait_commit(journal, tid); + return 1; +} + +/* + * Start a commit of the current running transaction (if any). Returns true + * if a transaction was started, and fills its tid in at *ptid + */ +int jbd2_journal_start_commit(journal_t *journal, tid_t *ptid) +{ + int ret = 0; + + spin_lock(&journal->j_state_lock); + if (journal->j_running_transaction) { + tid_t tid = journal->j_running_transaction->t_tid; + + ret = __jbd2_log_start_commit(journal, tid); + if (ret && ptid) + *ptid = tid; + } else if (journal->j_committing_transaction && ptid) { + /* + * If ext3_write_super() recently started a commit, then we + * have to wait for completion of that transaction + */ + *ptid = journal->j_committing_transaction->t_tid; + ret = 1; + } + spin_unlock(&journal->j_state_lock); + return ret; +} + +/* + * Wait for a specified commit to complete. + * The caller may not hold the journal lock. + */ +int jbd2_log_wait_commit(journal_t *journal, tid_t tid) +{ + int err = 0; + +#ifdef CONFIG_JBD_DEBUG + spin_lock(&journal->j_state_lock); + if (!tid_geq(journal->j_commit_request, tid)) { + printk(KERN_EMERG + "%s: error: j_commit_request=%d, tid=%d\n", + __FUNCTION__, journal->j_commit_request, tid); + } + spin_unlock(&journal->j_state_lock); +#endif + spin_lock(&journal->j_state_lock); + while (tid_gt(tid, journal->j_commit_sequence)) { + jbd_debug(1, "JBD: want %d, j_commit_sequence=%d\n", + tid, journal->j_commit_sequence); + wake_up(&journal->j_wait_commit); + spin_unlock(&journal->j_state_lock); + wait_event(journal->j_wait_done_commit, + !tid_gt(tid, journal->j_commit_sequence)); + spin_lock(&journal->j_state_lock); + } + spin_unlock(&journal->j_state_lock); + + if (unlikely(is_journal_aborted(journal))) { + printk(KERN_EMERG "journal commit I/O error\n"); + err = -EIO; + } + return err; +} + +/* + * Log buffer allocation routines: + */ + +int jbd2_journal_next_log_block(journal_t *journal, unsigned long long *retp) +{ + unsigned long blocknr; + + spin_lock(&journal->j_state_lock); + J_ASSERT(journal->j_free > 1); + + blocknr = journal->j_head; + journal->j_head++; + journal->j_free--; + if (journal->j_head == journal->j_last) + journal->j_head = journal->j_first; + spin_unlock(&journal->j_state_lock); + return jbd2_journal_bmap(journal, blocknr, retp); +} + +/* + * Conversion of logical to physical block numbers for the journal + * + * On external journals the journal blocks are identity-mapped, so + * this is a no-op. If needed, we can use j_blk_offset - everything is + * ready. + */ +int jbd2_journal_bmap(journal_t *journal, unsigned long blocknr, + unsigned long long *retp) +{ + int err = 0; + unsigned long long ret; + + if (journal->j_inode) { + ret = bmap(journal->j_inode, blocknr); + if (ret) + *retp = ret; + else { + char b[BDEVNAME_SIZE]; + + printk(KERN_ALERT "%s: journal block not found " + "at offset %lu on %s\n", + __FUNCTION__, + blocknr, + bdevname(journal->j_dev, b)); + err = -EIO; + __journal_abort_soft(journal, err); + } + } else { + *retp = blocknr; /* +journal->j_blk_offset */ + } + return err; +} + +/* + * We play buffer_head aliasing tricks to write data/metadata blocks to + * the journal without copying their contents, but for journal + * descriptor blocks we do need to generate bona fide buffers. + * + * After the caller of jbd2_journal_get_descriptor_buffer() has finished modifying + * the buffer's contents they really should run flush_dcache_page(bh->b_page). + * But we don't bother doing that, so there will be coherency problems with + * mmaps of blockdevs which hold live JBD-controlled filesystems. + */ +struct journal_head *jbd2_journal_get_descriptor_buffer(journal_t *journal) +{ + struct buffer_head *bh; + unsigned long long blocknr; + int err; + + err = jbd2_journal_next_log_block(journal, &blocknr); + + if (err) + return NULL; + + bh = __getblk(journal->j_dev, blocknr, journal->j_blocksize); + lock_buffer(bh); + memset(bh->b_data, 0, journal->j_blocksize); + set_buffer_uptodate(bh); + unlock_buffer(bh); + BUFFER_TRACE(bh, "return this buffer"); + return jbd2_journal_add_journal_head(bh); +} + +/* + * Management for journal control blocks: functions to create and + * destroy journal_t structures, and to initialise and read existing + * journal blocks from disk. */ + +/* First: create and setup a journal_t object in memory. We initialise + * very few fields yet: that has to wait until we have created the + * journal structures from from scratch, or loaded them from disk. */ + +static journal_t * journal_init_common (void) +{ + journal_t *journal; + int err; + + journal = jbd_kmalloc(sizeof(*journal), GFP_KERNEL); + if (!journal) + goto fail; + memset(journal, 0, sizeof(*journal)); + + init_waitqueue_head(&journal->j_wait_transaction_locked); + init_waitqueue_head(&journal->j_wait_logspace); + init_waitqueue_head(&journal->j_wait_done_commit); + init_waitqueue_head(&journal->j_wait_checkpoint); + init_waitqueue_head(&journal->j_wait_commit); + init_waitqueue_head(&journal->j_wait_updates); + mutex_init(&journal->j_barrier); + mutex_init(&journal->j_checkpoint_mutex); + spin_lock_init(&journal->j_revoke_lock); + spin_lock_init(&journal->j_list_lock); + spin_lock_init(&journal->j_state_lock); + + journal->j_commit_interval = (HZ * JBD_DEFAULT_MAX_COMMIT_AGE); + + /* The journal is marked for error until we succeed with recovery! */ + journal->j_flags = JBD2_ABORT; + + /* Set up a default-sized revoke table for the new mount. */ + err = jbd2_journal_init_revoke(journal, JOURNAL_REVOKE_DEFAULT_HASH); + if (err) { + kfree(journal); + goto fail; + } + return journal; +fail: + return NULL; +} + +/* jbd2_journal_init_dev and jbd2_journal_init_inode: + * + * Create a journal structure assigned some fixed set of disk blocks to + * the journal. We don't actually touch those disk blocks yet, but we + * need to set up all of the mapping information to tell the journaling + * system where the journal blocks are. + * + */ + +/** + * journal_t * jbd2_journal_init_dev() - creates an initialises a journal structure + * @bdev: Block device on which to create the journal + * @fs_dev: Device which hold journalled filesystem for this journal. + * @start: Block nr Start of journal. + * @len: Length of the journal in blocks. + * @blocksize: blocksize of journalling device + * @returns: a newly created journal_t * + * + * jbd2_journal_init_dev creates a journal which maps a fixed contiguous + * range of blocks on an arbitrary block device. + * + */ +journal_t * jbd2_journal_init_dev(struct block_device *bdev, + struct block_device *fs_dev, + unsigned long long start, int len, int blocksize) +{ + journal_t *journal = journal_init_common(); + struct buffer_head *bh; + int n; + + if (!journal) + return NULL; + + /* journal descriptor can store up to n blocks -bzzz */ + journal->j_blocksize = blocksize; + n = journal->j_blocksize / sizeof(journal_block_tag_t); + journal->j_wbufsize = n; + journal->j_wbuf = kmalloc(n * sizeof(struct buffer_head*), GFP_KERNEL); + if (!journal->j_wbuf) { + printk(KERN_ERR "%s: Cant allocate bhs for commit thread\n", + __FUNCTION__); + kfree(journal); + journal = NULL; + goto out; + } + journal->j_dev = bdev; + journal->j_fs_dev = fs_dev; + journal->j_blk_offset = start; + journal->j_maxlen = len; + + bh = __getblk(journal->j_dev, start, journal->j_blocksize); + J_ASSERT(bh != NULL); + journal->j_sb_buffer = bh; + journal->j_superblock = (journal_superblock_t *)bh->b_data; +out: + return journal; +} + +/** + * journal_t * jbd2_journal_init_inode () - creates a journal which maps to a inode. + * @inode: An inode to create the journal in + * + * jbd2_journal_init_inode creates a journal which maps an on-disk inode as + * the journal. The inode must exist already, must support bmap() and + * must have all data blocks preallocated. + */ +journal_t * jbd2_journal_init_inode (struct inode *inode) +{ + struct buffer_head *bh; + journal_t *journal = journal_init_common(); + int err; + int n; + unsigned long long blocknr; + + if (!journal) + return NULL; + + journal->j_dev = journal->j_fs_dev = inode->i_sb->s_bdev; + journal->j_inode = inode; + jbd_debug(1, + "journal %p: inode %s/%ld, size %Ld, bits %d, blksize %ld\n", + journal, inode->i_sb->s_id, inode->i_ino, + (long long) inode->i_size, + inode->i_sb->s_blocksize_bits, inode->i_sb->s_blocksize); + + journal->j_maxlen = inode->i_size >> inode->i_sb->s_blocksize_bits; + journal->j_blocksize = inode->i_sb->s_blocksize; + + /* journal descriptor can store up to n blocks -bzzz */ + n = journal->j_blocksize / sizeof(journal_block_tag_t); + journal->j_wbufsize = n; + journal->j_wbuf = kmalloc(n * sizeof(struct buffer_head*), GFP_KERNEL); + if (!journal->j_wbuf) { + printk(KERN_ERR "%s: Cant allocate bhs for commit thread\n", + __FUNCTION__); + kfree(journal); + return NULL; + } + + err = jbd2_journal_bmap(journal, 0, &blocknr); + /* If that failed, give up */ + if (err) { + printk(KERN_ERR "%s: Cannnot locate journal superblock\n", + __FUNCTION__); + kfree(journal); + return NULL; + } + + bh = __getblk(journal->j_dev, blocknr, journal->j_blocksize); + J_ASSERT(bh != NULL); + journal->j_sb_buffer = bh; + journal->j_superblock = (journal_superblock_t *)bh->b_data; + + return journal; +} + +/* + * If the journal init or create aborts, we need to mark the journal + * superblock as being NULL to prevent the journal destroy from writing + * back a bogus superblock. + */ +static void journal_fail_superblock (journal_t *journal) +{ + struct buffer_head *bh = journal->j_sb_buffer; + brelse(bh); + journal->j_sb_buffer = NULL; +} + +/* + * Given a journal_t structure, initialise the various fields for + * startup of a new journaling session. We use this both when creating + * a journal, and after recovering an old journal to reset it for + * subsequent use. + */ + +static int journal_reset(journal_t *journal) +{ + journal_superblock_t *sb = journal->j_superblock; + unsigned long long first, last; + + first = be32_to_cpu(sb->s_first); + last = be32_to_cpu(sb->s_maxlen); + + journal->j_first = first; + journal->j_last = last; + + journal->j_head = first; + journal->j_tail = first; + journal->j_free = last - first; + + journal->j_tail_sequence = journal->j_transaction_sequence; + journal->j_commit_sequence = journal->j_transaction_sequence - 1; + journal->j_commit_request = journal->j_commit_sequence; + + journal->j_max_transaction_buffers = journal->j_maxlen / 4; + + /* Add the dynamic fields and write it to disk. */ + jbd2_journal_update_superblock(journal, 1); + jbd2_journal_start_thread(journal); + return 0; +} + +/** + * int jbd2_journal_create() - Initialise the new journal file + * @journal: Journal to create. This structure must have been initialised + * + * Given a journal_t structure which tells us which disk blocks we can + * use, create a new journal superblock and initialise all of the + * journal fields from scratch. + **/ +int jbd2_journal_create(journal_t *journal) +{ + unsigned long long blocknr; + struct buffer_head *bh; + journal_superblock_t *sb; + int i, err; + + if (journal->j_maxlen < JBD2_MIN_JOURNAL_BLOCKS) { + printk (KERN_ERR "Journal length (%d blocks) too short.\n", + journal->j_maxlen); + journal_fail_superblock(journal); + return -EINVAL; + } + + if (journal->j_inode == NULL) { + /* + * We don't know what block to start at! + */ + printk(KERN_EMERG + "%s: creation of journal on external device!\n", + __FUNCTION__); + BUG(); + } + + /* Zero out the entire journal on disk. We cannot afford to + have any blocks on disk beginning with JBD2_MAGIC_NUMBER. */ + jbd_debug(1, "JBD: Zeroing out journal blocks...\n"); + for (i = 0; i < journal->j_maxlen; i++) { + err = jbd2_journal_bmap(journal, i, &blocknr); + if (err) + return err; + bh = __getblk(journal->j_dev, blocknr, journal->j_blocksize); + lock_buffer(bh); + memset (bh->b_data, 0, journal->j_blocksize); + BUFFER_TRACE(bh, "marking dirty"); + mark_buffer_dirty(bh); + BUFFER_TRACE(bh, "marking uptodate"); + set_buffer_uptodate(bh); + unlock_buffer(bh); + __brelse(bh); + } + + sync_blockdev(journal->j_dev); + jbd_debug(1, "JBD: journal cleared.\n"); + + /* OK, fill in the initial static fields in the new superblock */ + sb = journal->j_superblock; + + sb->s_header.h_magic = cpu_to_be32(JBD2_MAGIC_NUMBER); + sb->s_header.h_blocktype = cpu_to_be32(JBD2_SUPERBLOCK_V2); + + sb->s_blocksize = cpu_to_be32(journal->j_blocksize); + sb->s_maxlen = cpu_to_be32(journal->j_maxlen); + sb->s_first = cpu_to_be32(1); + + journal->j_transaction_sequence = 1; + + journal->j_flags &= ~JBD2_ABORT; + journal->j_format_version = 2; + + return journal_reset(journal); +} + +/** + * void jbd2_journal_update_superblock() - Update journal sb on disk. + * @journal: The journal to update. + * @wait: Set to '0' if you don't want to wait for IO completion. + * + * Update a journal's dynamic superblock fields and write it to disk, + * optionally waiting for the IO to complete. + */ +void jbd2_journal_update_superblock(journal_t *journal, int wait) +{ + journal_superblock_t *sb = journal->j_superblock; + struct buffer_head *bh = journal->j_sb_buffer; + + /* + * As a special case, if the on-disk copy is already marked as needing + * no recovery (s_start == 0) and there are no outstanding transactions + * in the filesystem, then we can safely defer the superblock update + * until the next commit by setting JBD2_FLUSHED. This avoids + * attempting a write to a potential-readonly device. + */ + if (sb->s_start == 0 && journal->j_tail_sequence == + journal->j_transaction_sequence) { + jbd_debug(1,"JBD: Skipping superblock update on recovered sb " + "(start %ld, seq %d, errno %d)\n", + journal->j_tail, journal->j_tail_sequence, + journal->j_errno); + goto out; + } + + spin_lock(&journal->j_state_lock); + jbd_debug(1,"JBD: updating superblock (start %ld, seq %d, errno %d)\n", + journal->j_tail, journal->j_tail_sequence, journal->j_errno); + + sb->s_sequence = cpu_to_be32(journal->j_tail_sequence); + sb->s_start = cpu_to_be32(journal->j_tail); + sb->s_errno = cpu_to_be32(journal->j_errno); + spin_unlock(&journal->j_state_lock); + + BUFFER_TRACE(bh, "marking dirty"); + mark_buffer_dirty(bh); + if (wait) + sync_dirty_buffer(bh); + else + ll_rw_block(SWRITE, 1, &bh); + +out: + /* If we have just flushed the log (by marking s_start==0), then + * any future commit will have to be careful to update the + * superblock again to re-record the true start of the log. */ + + spin_lock(&journal->j_state_lock); + if (sb->s_start) + journal->j_flags &= ~JBD2_FLUSHED; + else + journal->j_flags |= JBD2_FLUSHED; + spin_unlock(&journal->j_state_lock); +} + +/* + * Read the superblock for a given journal, performing initial + * validation of the format. + */ + +static int journal_get_superblock(journal_t *journal) +{ + struct buffer_head *bh; + journal_superblock_t *sb; + int err = -EIO; + + bh = journal->j_sb_buffer; + + J_ASSERT(bh != NULL); + if (!buffer_uptodate(bh)) { + ll_rw_block(READ, 1, &bh); + wait_on_buffer(bh); + if (!buffer_uptodate(bh)) { + printk (KERN_ERR + "JBD: IO error reading journal superblock\n"); + goto out; + } + } + + sb = journal->j_superblock; + + err = -EINVAL; + + if (sb->s_header.h_magic != cpu_to_be32(JBD2_MAGIC_NUMBER) || + sb->s_blocksize != cpu_to_be32(journal->j_blocksize)) { + printk(KERN_WARNING "JBD: no valid journal superblock found\n"); + goto out; + } + + switch(be32_to_cpu(sb->s_header.h_blocktype)) { + case JBD2_SUPERBLOCK_V1: + journal->j_format_version = 1; + break; + case JBD2_SUPERBLOCK_V2: + journal->j_format_version = 2; + break; + default: + printk(KERN_WARNING "JBD: unrecognised superblock format ID\n"); + goto out; + } + + if (be32_to_cpu(sb->s_maxlen) < journal->j_maxlen) + journal->j_maxlen = be32_to_cpu(sb->s_maxlen); + else if (be32_to_cpu(sb->s_maxlen) > journal->j_maxlen) { + printk (KERN_WARNING "JBD: journal file too short\n"); + goto out; + } + + return 0; + +out: + journal_fail_superblock(journal); + return err; +} + +/* + * Load the on-disk journal superblock and read the key fields into the + * journal_t. + */ + +static int load_superblock(journal_t *journal) +{ + int err; + journal_superblock_t *sb; + + err = journal_get_superblock(journal); + if (err) + return err; + + sb = journal->j_superblock; + + journal->j_tail_sequence = be32_to_cpu(sb->s_sequence); + journal->j_tail = be32_to_cpu(sb->s_start); + journal->j_first = be32_to_cpu(sb->s_first); + journal->j_last = be32_to_cpu(sb->s_maxlen); + journal->j_errno = be32_to_cpu(sb->s_errno); + + return 0; +} + + +/** + * int jbd2_journal_load() - Read journal from disk. + * @journal: Journal to act on. + * + * Given a journal_t structure which tells us which disk blocks contain + * a journal, read the journal from disk to initialise the in-memory + * structures. + */ +int jbd2_journal_load(journal_t *journal) +{ + int err; + journal_superblock_t *sb; + + err = load_superblock(journal); + if (err) + return err; + + sb = journal->j_superblock; + /* If this is a V2 superblock, then we have to check the + * features flags on it. */ + + if (journal->j_format_version >= 2) { + if ((sb->s_feature_ro_compat & + ~cpu_to_be32(JBD2_KNOWN_ROCOMPAT_FEATURES)) || + (sb->s_feature_incompat & + ~cpu_to_be32(JBD2_KNOWN_INCOMPAT_FEATURES))) { + printk (KERN_WARNING + "JBD: Unrecognised features on journal\n"); + return -EINVAL; + } + } + + /* + * Create a slab for this blocksize + */ + err = jbd2_journal_create_jbd_slab(be32_to_cpu(sb->s_blocksize)); + if (err) + return err; + + /* Let the recovery code check whether it needs to recover any + * data from the journal. */ + if (jbd2_journal_recover(journal)) + goto recovery_error; + + /* OK, we've finished with the dynamic journal bits: + * reinitialise the dynamic contents of the superblock in memory + * and reset them on disk. */ + if (journal_reset(journal)) + goto recovery_error; + + journal->j_flags &= ~JBD2_ABORT; + journal->j_flags |= JBD2_LOADED; + return 0; + +recovery_error: + printk (KERN_WARNING "JBD: recovery failed\n"); + return -EIO; +} + +/** + * void jbd2_journal_destroy() - Release a journal_t structure. + * @journal: Journal to act on. + * + * Release a journal_t structure once it is no longer in use by the + * journaled object. + */ +void jbd2_journal_destroy(journal_t *journal) +{ + /* Wait for the commit thread to wake up and die. */ + journal_kill_thread(journal); + + /* Force a final log commit */ + if (journal->j_running_transaction) + jbd2_journal_commit_transaction(journal); + + /* Force any old transactions to disk */ + + /* Totally anal locking here... */ + spin_lock(&journal->j_list_lock); + while (journal->j_checkpoint_transactions != NULL) { + spin_unlock(&journal->j_list_lock); + jbd2_log_do_checkpoint(journal); + spin_lock(&journal->j_list_lock); + } + + J_ASSERT(journal->j_running_transaction == NULL); + J_ASSERT(journal->j_committing_transaction == NULL); + J_ASSERT(journal->j_checkpoint_transactions == NULL); + spin_unlock(&journal->j_list_lock); + + /* We can now mark the journal as empty. */ + journal->j_tail = 0; + journal->j_tail_sequence = ++journal->j_transaction_sequence; + if (journal->j_sb_buffer) { + jbd2_journal_update_superblock(journal, 1); + brelse(journal->j_sb_buffer); + } + + if (journal->j_inode) + iput(journal->j_inode); + if (journal->j_revoke) + jbd2_journal_destroy_revoke(journal); + kfree(journal->j_wbuf); + kfree(journal); +} + + +/** + *int jbd2_journal_check_used_features () - Check if features specified are used. + * @journal: Journal to check. + * @compat: bitmask of compatible features + * @ro: bitmask of features that force read-only mount + * @incompat: bitmask of incompatible features + * + * Check whether the journal uses all of a given set of + * features. Return true (non-zero) if it does. + **/ + +int jbd2_journal_check_used_features (journal_t *journal, unsigned long compat, + unsigned long ro, unsigned long incompat) +{ + journal_superblock_t *sb; + + if (!compat && !ro && !incompat) + return 1; + if (journal->j_format_version == 1) + return 0; + + sb = journal->j_superblock; + + if (((be32_to_cpu(sb->s_feature_compat) & compat) == compat) && + ((be32_to_cpu(sb->s_feature_ro_compat) & ro) == ro) && + ((be32_to_cpu(sb->s_feature_incompat) & incompat) == incompat)) + return 1; + + return 0; +} + +/** + * int jbd2_journal_check_available_features() - Check feature set in journalling layer + * @journal: Journal to check. + * @compat: bitmask of compatible features + * @ro: bitmask of features that force read-only mount + * @incompat: bitmask of incompatible features + * + * Check whether the journaling code supports the use of + * all of a given set of features on this journal. Return true + * (non-zero) if it can. */ + +int jbd2_journal_check_available_features (journal_t *journal, unsigned long compat, + unsigned long ro, unsigned long incompat) +{ + journal_superblock_t *sb; + + if (!compat && !ro && !incompat) + return 1; + + sb = journal->j_superblock; + + /* We can support any known requested features iff the + * superblock is in version 2. Otherwise we fail to support any + * extended sb features. */ + + if (journal->j_format_version != 2) + return 0; + + if ((compat & JBD2_KNOWN_COMPAT_FEATURES) == compat && + (ro & JBD2_KNOWN_ROCOMPAT_FEATURES) == ro && + (incompat & JBD2_KNOWN_INCOMPAT_FEATURES) == incompat) + return 1; + + return 0; +} + +/** + * int jbd2_journal_set_features () - Mark a given journal feature in the superblock + * @journal: Journal to act on. + * @compat: bitmask of compatible features + * @ro: bitmask of features that force read-only mount + * @incompat: bitmask of incompatible features + * + * Mark a given journal feature as present on the + * superblock. Returns true if the requested features could be set. + * + */ + +int jbd2_journal_set_features (journal_t *journal, unsigned long compat, + unsigned long ro, unsigned long incompat) +{ + journal_superblock_t *sb; + + if (jbd2_journal_check_used_features(journal, compat, ro, incompat)) + return 1; + + if (!jbd2_journal_check_available_features(journal, compat, ro, incompat)) + return 0; + + jbd_debug(1, "Setting new features 0x%lx/0x%lx/0x%lx\n", + compat, ro, incompat); + + sb = journal->j_superblock; + + sb->s_feature_compat |= cpu_to_be32(compat); + sb->s_feature_ro_compat |= cpu_to_be32(ro); + sb->s_feature_incompat |= cpu_to_be32(incompat); + + return 1; +} + + +/** + * int jbd2_journal_update_format () - Update on-disk journal structure. + * @journal: Journal to act on. + * + * Given an initialised but unloaded journal struct, poke about in the + * on-disk structure to update it to the most recent supported version. + */ +int jbd2_journal_update_format (journal_t *journal) +{ + journal_superblock_t *sb; + int err; + + err = journal_get_superblock(journal); + if (err) + return err; + + sb = journal->j_superblock; + + switch (be32_to_cpu(sb->s_header.h_blocktype)) { + case JBD2_SUPERBLOCK_V2: + return 0; + case JBD2_SUPERBLOCK_V1: + return journal_convert_superblock_v1(journal, sb); + default: + break; + } + return -EINVAL; +} + +static int journal_convert_superblock_v1(journal_t *journal, + journal_superblock_t *sb) +{ + int offset, blocksize; + struct buffer_head *bh; + + printk(KERN_WARNING + "JBD: Converting superblock from version 1 to 2.\n"); + + /* Pre-initialise new fields to zero */ + offset = ((char *) &(sb->s_feature_compat)) - ((char *) sb); + blocksize = be32_to_cpu(sb->s_blocksize); + memset(&sb->s_feature_compat, 0, blocksize-offset); + + sb->s_nr_users = cpu_to_be32(1); + sb->s_header.h_blocktype = cpu_to_be32(JBD2_SUPERBLOCK_V2); + journal->j_format_version = 2; + + bh = journal->j_sb_buffer; + BUFFER_TRACE(bh, "marking dirty"); + mark_buffer_dirty(bh); + sync_dirty_buffer(bh); + return 0; +} + + +/** + * int jbd2_journal_flush () - Flush journal + * @journal: Journal to act on. + * + * Flush all data for a given journal to disk and empty the journal. + * Filesystems can use this when remounting readonly to ensure that + * recovery does not need to happen on remount. + */ + +int jbd2_journal_flush(journal_t *journal) +{ + int err = 0; + transaction_t *transaction = NULL; + unsigned long old_tail; + + spin_lock(&journal->j_state_lock); + + /* Force everything buffered to the log... */ + if (journal->j_running_transaction) { + transaction = journal->j_running_transaction; + __jbd2_log_start_commit(journal, transaction->t_tid); + } else if (journal->j_committing_transaction) + transaction = journal->j_committing_transaction; + + /* Wait for the log commit to complete... */ + if (transaction) { + tid_t tid = transaction->t_tid; + + spin_unlock(&journal->j_state_lock); + jbd2_log_wait_commit(journal, tid); + } else { + spin_unlock(&journal->j_state_lock); + } + + /* ...and flush everything in the log out to disk. */ + spin_lock(&journal->j_list_lock); + while (!err && journal->j_checkpoint_transactions != NULL) { + spin_unlock(&journal->j_list_lock); + err = jbd2_log_do_checkpoint(journal); + spin_lock(&journal->j_list_lock); + } + spin_unlock(&journal->j_list_lock); + jbd2_cleanup_journal_tail(journal); + + /* Finally, mark the journal as really needing no recovery. + * This sets s_start==0 in the underlying superblock, which is + * the magic code for a fully-recovered superblock. Any future + * commits of data to the journal will restore the current + * s_start value. */ + spin_lock(&journal->j_state_lock); + old_tail = journal->j_tail; + journal->j_tail = 0; + spin_unlock(&journal->j_state_lock); + jbd2_journal_update_superblock(journal, 1); + spin_lock(&journal->j_state_lock); + journal->j_tail = old_tail; + + J_ASSERT(!journal->j_running_transaction); + J_ASSERT(!journal->j_committing_transaction); + J_ASSERT(!journal->j_checkpoint_transactions); + J_ASSERT(journal->j_head == journal->j_tail); + J_ASSERT(journal->j_tail_sequence == journal->j_transaction_sequence); + spin_unlock(&journal->j_state_lock); + return err; +} + +/** + * int jbd2_journal_wipe() - Wipe journal contents + * @journal: Journal to act on. + * @write: flag (see below) + * + * Wipe out all of the contents of a journal, safely. This will produce + * a warning if the journal contains any valid recovery information. + * Must be called between journal_init_*() and jbd2_journal_load(). + * + * If 'write' is non-zero, then we wipe out the journal on disk; otherwise + * we merely suppress recovery. + */ + +int jbd2_journal_wipe(journal_t *journal, int write) +{ + journal_superblock_t *sb; + int err = 0; + + J_ASSERT (!(journal->j_flags & JBD2_LOADED)); + + err = load_superblock(journal); + if (err) + return err; + + sb = journal->j_superblock; + + if (!journal->j_tail) + goto no_recovery; + + printk (KERN_WARNING "JBD: %s recovery information on journal\n", + write ? "Clearing" : "Ignoring"); + + err = jbd2_journal_skip_recovery(journal); + if (write) + jbd2_journal_update_superblock(journal, 1); + + no_recovery: + return err; +} + +/* + * journal_dev_name: format a character string to describe on what + * device this journal is present. + */ + +static const char *journal_dev_name(journal_t *journal, char *buffer) +{ + struct block_device *bdev; + + if (journal->j_inode) + bdev = journal->j_inode->i_sb->s_bdev; + else + bdev = journal->j_dev; + + return bdevname(bdev, buffer); +} + +/* + * Journal abort has very specific semantics, which we describe + * for journal abort. + * + * Two internal function, which provide abort to te jbd layer + * itself are here. + */ + +/* + * Quick version for internal journal use (doesn't lock the journal). + * Aborts hard --- we mark the abort as occurred, but do _nothing_ else, + * and don't attempt to make any other journal updates. + */ +void __jbd2_journal_abort_hard(journal_t *journal) +{ + transaction_t *transaction; + char b[BDEVNAME_SIZE]; + + if (journal->j_flags & JBD2_ABORT) + return; + + printk(KERN_ERR "Aborting journal on device %s.\n", + journal_dev_name(journal, b)); + + spin_lock(&journal->j_state_lock); + journal->j_flags |= JBD2_ABORT; + transaction = journal->j_running_transaction; + if (transaction) + __jbd2_log_start_commit(journal, transaction->t_tid); + spin_unlock(&journal->j_state_lock); +} + +/* Soft abort: record the abort error status in the journal superblock, + * but don't do any other IO. */ +static void __journal_abort_soft (journal_t *journal, int errno) +{ + if (journal->j_flags & JBD2_ABORT) + return; + + if (!journal->j_errno) + journal->j_errno = errno; + + __jbd2_journal_abort_hard(journal); + + if (errno) + jbd2_journal_update_superblock(journal, 1); +} + +/** + * void jbd2_journal_abort () - Shutdown the journal immediately. + * @journal: the journal to shutdown. + * @errno: an error number to record in the journal indicating + * the reason for the shutdown. + * + * Perform a complete, immediate shutdown of the ENTIRE + * journal (not of a single transaction). This operation cannot be + * undone without closing and reopening the journal. + * + * The jbd2_journal_abort function is intended to support higher level error + * recovery mechanisms such as the ext2/ext3 remount-readonly error + * mode. + * + * Journal abort has very specific semantics. Any existing dirty, + * unjournaled buffers in the main filesystem will still be written to + * disk by bdflush, but the journaling mechanism will be suspended + * immediately and no further transaction commits will be honoured. + * + * Any dirty, journaled buffers will be written back to disk without + * hitting the journal. Atomicity cannot be guaranteed on an aborted + * filesystem, but we _do_ attempt to leave as much data as possible + * behind for fsck to use for cleanup. + * + * Any attempt to get a new transaction handle on a journal which is in + * ABORT state will just result in an -EROFS error return. A + * jbd2_journal_stop on an existing handle will return -EIO if we have + * entered abort state during the update. + * + * Recursive transactions are not disturbed by journal abort until the + * final jbd2_journal_stop, which will receive the -EIO error. + * + * Finally, the jbd2_journal_abort call allows the caller to supply an errno + * which will be recorded (if possible) in the journal superblock. This + * allows a client to record failure conditions in the middle of a + * transaction without having to complete the transaction to record the + * failure to disk. ext3_error, for example, now uses this + * functionality. + * + * Errors which originate from within the journaling layer will NOT + * supply an errno; a null errno implies that absolutely no further + * writes are done to the journal (unless there are any already in + * progress). + * + */ + +void jbd2_journal_abort(journal_t *journal, int errno) +{ + __journal_abort_soft(journal, errno); +} + +/** + * int jbd2_journal_errno () - returns the journal's error state. + * @journal: journal to examine. + * + * This is the errno numbet set with jbd2_journal_abort(), the last + * time the journal was mounted - if the journal was stopped + * without calling abort this will be 0. + * + * If the journal has been aborted on this mount time -EROFS will + * be returned. + */ +int jbd2_journal_errno(journal_t *journal) +{ + int err; + + spin_lock(&journal->j_state_lock); + if (journal->j_flags & JBD2_ABORT) + err = -EROFS; + else + err = journal->j_errno; + spin_unlock(&journal->j_state_lock); + return err; +} + +/** + * int jbd2_journal_clear_err () - clears the journal's error state + * @journal: journal to act on. + * + * An error must be cleared or Acked to take a FS out of readonly + * mode. + */ +int jbd2_journal_clear_err(journal_t *journal) +{ + int err = 0; + + spin_lock(&journal->j_state_lock); + if (journal->j_flags & JBD2_ABORT) + err = -EROFS; + else + journal->j_errno = 0; + spin_unlock(&journal->j_state_lock); + return err; +} + +/** + * void jbd2_journal_ack_err() - Ack journal err. + * @journal: journal to act on. + * + * An error must be cleared or Acked to take a FS out of readonly + * mode. + */ +void jbd2_journal_ack_err(journal_t *journal) +{ + spin_lock(&journal->j_state_lock); + if (journal->j_errno) + journal->j_flags |= JBD2_ACK_ERR; + spin_unlock(&journal->j_state_lock); +} + +int jbd2_journal_blocks_per_page(struct inode *inode) +{ + return 1 << (PAGE_CACHE_SHIFT - inode->i_sb->s_blocksize_bits); +} + +/* + * helper functions to deal with 32 or 64bit block numbers. + */ +size_t journal_tag_bytes(journal_t *journal) +{ + if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_64BIT)) + return JBD_TAG_SIZE64; + else + return JBD_TAG_SIZE32; +} + +/* + * Simple support for retrying memory allocations. Introduced to help to + * debug different VM deadlock avoidance strategies. + */ +void * __jbd2_kmalloc (const char *where, size_t size, gfp_t flags, int retry) +{ + return kmalloc(size, flags | (retry ? __GFP_NOFAIL : 0)); +} + +/* + * jbd slab management: create 1k, 2k, 4k, 8k slabs as needed + * and allocate frozen and commit buffers from these slabs. + * + * Reason for doing this is to avoid, SLAB_DEBUG - since it could + * cause bh to cross page boundary. + */ + +#define JBD_MAX_SLABS 5 +#define JBD_SLAB_INDEX(size) (size >> 11) + +static kmem_cache_t *jbd_slab[JBD_MAX_SLABS]; +static const char *jbd_slab_names[JBD_MAX_SLABS] = { + "jbd2_1k", "jbd2_2k", "jbd2_4k", NULL, "jbd2_8k" +}; + +static void jbd2_journal_destroy_jbd_slabs(void) +{ + int i; + + for (i = 0; i < JBD_MAX_SLABS; i++) { + if (jbd_slab[i]) + kmem_cache_destroy(jbd_slab[i]); + jbd_slab[i] = NULL; + } +} + +static int jbd2_journal_create_jbd_slab(size_t slab_size) +{ + int i = JBD_SLAB_INDEX(slab_size); + + BUG_ON(i >= JBD_MAX_SLABS); + + /* + * Check if we already have a slab created for this size + */ + if (jbd_slab[i]) + return 0; + + /* + * Create a slab and force alignment to be same as slabsize - + * this will make sure that allocations won't cross the page + * boundary. + */ + jbd_slab[i] = kmem_cache_create(jbd_slab_names[i], + slab_size, slab_size, 0, NULL, NULL); + if (!jbd_slab[i]) { + printk(KERN_EMERG "JBD: no memory for jbd_slab cache\n"); + return -ENOMEM; + } + return 0; +} + +void * jbd2_slab_alloc(size_t size, gfp_t flags) +{ + int idx; + + idx = JBD_SLAB_INDEX(size); + BUG_ON(jbd_slab[idx] == NULL); + return kmem_cache_alloc(jbd_slab[idx], flags | __GFP_NOFAIL); +} + +void jbd2_slab_free(void *ptr, size_t size) +{ + int idx; + + idx = JBD_SLAB_INDEX(size); + BUG_ON(jbd_slab[idx] == NULL); + kmem_cache_free(jbd_slab[idx], ptr); +} + +/* + * Journal_head storage management + */ +static kmem_cache_t *jbd2_journal_head_cache; +#ifdef CONFIG_JBD_DEBUG +static atomic_t nr_journal_heads = ATOMIC_INIT(0); +#endif + +static int journal_init_jbd2_journal_head_cache(void) +{ + int retval; + + J_ASSERT(jbd2_journal_head_cache == 0); + jbd2_journal_head_cache = kmem_cache_create("jbd2_journal_head", + sizeof(struct journal_head), + 0, /* offset */ + 0, /* flags */ + NULL, /* ctor */ + NULL); /* dtor */ + retval = 0; + if (jbd2_journal_head_cache == 0) { + retval = -ENOMEM; + printk(KERN_EMERG "JBD: no memory for journal_head cache\n"); + } + return retval; +} + +static void jbd2_journal_destroy_jbd2_journal_head_cache(void) +{ + J_ASSERT(jbd2_journal_head_cache != NULL); + kmem_cache_destroy(jbd2_journal_head_cache); + jbd2_journal_head_cache = NULL; +} + +/* + * journal_head splicing and dicing + */ +static struct journal_head *journal_alloc_journal_head(void) +{ + struct journal_head *ret; + static unsigned long last_warning; + +#ifdef CONFIG_JBD_DEBUG + atomic_inc(&nr_journal_heads); +#endif + ret = kmem_cache_alloc(jbd2_journal_head_cache, GFP_NOFS); + if (ret == 0) { + jbd_debug(1, "out of memory for journal_head\n"); + if (time_after(jiffies, last_warning + 5*HZ)) { + printk(KERN_NOTICE "ENOMEM in %s, retrying.\n", + __FUNCTION__); + last_warning = jiffies; + } + while (ret == 0) { + yield(); + ret = kmem_cache_alloc(jbd2_journal_head_cache, GFP_NOFS); + } + } + return ret; +} + +static void journal_free_journal_head(struct journal_head *jh) +{ +#ifdef CONFIG_JBD_DEBUG + atomic_dec(&nr_journal_heads); + memset(jh, JBD_POISON_FREE, sizeof(*jh)); +#endif + kmem_cache_free(jbd2_journal_head_cache, jh); +} + +/* + * A journal_head is attached to a buffer_head whenever JBD has an + * interest in the buffer. + * + * Whenever a buffer has an attached journal_head, its ->b_state:BH_JBD bit + * is set. This bit is tested in core kernel code where we need to take + * JBD-specific actions. Testing the zeroness of ->b_private is not reliable + * there. + * + * When a buffer has its BH_JBD bit set, its ->b_count is elevated by one. + * + * When a buffer has its BH_JBD bit set it is immune from being released by + * core kernel code, mainly via ->b_count. + * + * A journal_head may be detached from its buffer_head when the journal_head's + * b_transaction, b_cp_transaction and b_next_transaction pointers are NULL. + * Various places in JBD call jbd2_journal_remove_journal_head() to indicate that the + * journal_head can be dropped if needed. + * + * Various places in the kernel want to attach a journal_head to a buffer_head + * _before_ attaching the journal_head to a transaction. To protect the + * journal_head in this situation, jbd2_journal_add_journal_head elevates the + * journal_head's b_jcount refcount by one. The caller must call + * jbd2_journal_put_journal_head() to undo this. + * + * So the typical usage would be: + * + * (Attach a journal_head if needed. Increments b_jcount) + * struct journal_head *jh = jbd2_journal_add_journal_head(bh); + * ... + * jh->b_transaction = xxx; + * jbd2_journal_put_journal_head(jh); + * + * Now, the journal_head's b_jcount is zero, but it is safe from being released + * because it has a non-zero b_transaction. + */ + +/* + * Give a buffer_head a journal_head. + * + * Doesn't need the journal lock. + * May sleep. + */ +struct journal_head *jbd2_journal_add_journal_head(struct buffer_head *bh) +{ + struct journal_head *jh; + struct journal_head *new_jh = NULL; + +repeat: + if (!buffer_jbd(bh)) { + new_jh = journal_alloc_journal_head(); + memset(new_jh, 0, sizeof(*new_jh)); + } + + jbd_lock_bh_journal_head(bh); + if (buffer_jbd(bh)) { + jh = bh2jh(bh); + } else { + J_ASSERT_BH(bh, + (atomic_read(&bh->b_count) > 0) || + (bh->b_page && bh->b_page->mapping)); + + if (!new_jh) { + jbd_unlock_bh_journal_head(bh); + goto repeat; + } + + jh = new_jh; + new_jh = NULL; /* We consumed it */ + set_buffer_jbd(bh); + bh->b_private = jh; + jh->b_bh = bh; + get_bh(bh); + BUFFER_TRACE(bh, "added journal_head"); + } + jh->b_jcount++; + jbd_unlock_bh_journal_head(bh); + if (new_jh) + journal_free_journal_head(new_jh); + return bh->b_private; +} + +/* + * Grab a ref against this buffer_head's journal_head. If it ended up not + * having a journal_head, return NULL + */ +struct journal_head *jbd2_journal_grab_journal_head(struct buffer_head *bh) +{ + struct journal_head *jh = NULL; + + jbd_lock_bh_journal_head(bh); + if (buffer_jbd(bh)) { + jh = bh2jh(bh); + jh->b_jcount++; + } + jbd_unlock_bh_journal_head(bh); + return jh; +} + +static void __journal_remove_journal_head(struct buffer_head *bh) +{ + struct journal_head *jh = bh2jh(bh); + + J_ASSERT_JH(jh, jh->b_jcount >= 0); + + get_bh(bh); + if (jh->b_jcount == 0) { + if (jh->b_transaction == NULL && + jh->b_next_transaction == NULL && + jh->b_cp_transaction == NULL) { + J_ASSERT_JH(jh, jh->b_jlist == BJ_None); + J_ASSERT_BH(bh, buffer_jbd(bh)); + J_ASSERT_BH(bh, jh2bh(jh) == bh); + BUFFER_TRACE(bh, "remove journal_head"); + if (jh->b_frozen_data) { + printk(KERN_WARNING "%s: freeing " + "b_frozen_data\n", + __FUNCTION__); + jbd2_slab_free(jh->b_frozen_data, bh->b_size); + } + if (jh->b_committed_data) { + printk(KERN_WARNING "%s: freeing " + "b_committed_data\n", + __FUNCTION__); + jbd2_slab_free(jh->b_committed_data, bh->b_size); + } + bh->b_private = NULL; + jh->b_bh = NULL; /* debug, really */ + clear_buffer_jbd(bh); + __brelse(bh); + journal_free_journal_head(jh); + } else { + BUFFER_TRACE(bh, "journal_head was locked"); + } + } +} + +/* + * jbd2_journal_remove_journal_head(): if the buffer isn't attached to a transaction + * and has a zero b_jcount then remove and release its journal_head. If we did + * see that the buffer is not used by any transaction we also "logically" + * decrement ->b_count. + * + * We in fact take an additional increment on ->b_count as a convenience, + * because the caller usually wants to do additional things with the bh + * after calling here. + * The caller of jbd2_journal_remove_journal_head() *must* run __brelse(bh) at some + * time. Once the caller has run __brelse(), the buffer is eligible for + * reaping by try_to_free_buffers(). + */ +void jbd2_journal_remove_journal_head(struct buffer_head *bh) +{ + jbd_lock_bh_journal_head(bh); + __journal_remove_journal_head(bh); + jbd_unlock_bh_journal_head(bh); +} + +/* + * Drop a reference on the passed journal_head. If it fell to zero then try to + * release the journal_head from the buffer_head. + */ +void jbd2_journal_put_journal_head(struct journal_head *jh) +{ + struct buffer_head *bh = jh2bh(jh); + + jbd_lock_bh_journal_head(bh); + J_ASSERT_JH(jh, jh->b_jcount > 0); + --jh->b_jcount; + if (!jh->b_jcount && !jh->b_transaction) { + __journal_remove_journal_head(bh); + __brelse(bh); + } + jbd_unlock_bh_journal_head(bh); +} + +/* + * /proc tunables + */ +#if defined(CONFIG_JBD_DEBUG) +int jbd2_journal_enable_debug; +EXPORT_SYMBOL(jbd2_journal_enable_debug); +#endif + +#if defined(CONFIG_JBD_DEBUG) && defined(CONFIG_PROC_FS) + +static struct proc_dir_entry *proc_jbd_debug; + +static int read_jbd_debug(char *page, char **start, off_t off, + int count, int *eof, void *data) +{ + int ret; + + ret = sprintf(page + off, "%d\n", jbd2_journal_enable_debug); + *eof = 1; + return ret; +} + +static int write_jbd_debug(struct file *file, const char __user *buffer, + unsigned long count, void *data) +{ + char buf[32]; + + if (count > ARRAY_SIZE(buf) - 1) + count = ARRAY_SIZE(buf) - 1; + if (copy_from_user(buf, buffer, count)) + return -EFAULT; + buf[ARRAY_SIZE(buf) - 1] = '\0'; + jbd2_journal_enable_debug = simple_strtoul(buf, NULL, 10); + return count; +} + +#define JBD_PROC_NAME "sys/fs/jbd2-debug" + +static void __init create_jbd_proc_entry(void) +{ + proc_jbd_debug = create_proc_entry(JBD_PROC_NAME, 0644, NULL); + if (proc_jbd_debug) { + /* Why is this so hard? */ + proc_jbd_debug->read_proc = read_jbd_debug; + proc_jbd_debug->write_proc = write_jbd_debug; + } +} + +static void __exit jbd2_remove_jbd_proc_entry(void) +{ + if (proc_jbd_debug) + remove_proc_entry(JBD_PROC_NAME, NULL); +} + +#else + +#define create_jbd_proc_entry() do {} while (0) +#define jbd2_remove_jbd_proc_entry() do {} while (0) + +#endif + +kmem_cache_t *jbd2_handle_cache; + +static int __init journal_init_handle_cache(void) +{ + jbd2_handle_cache = kmem_cache_create("jbd2_journal_handle", + sizeof(handle_t), + 0, /* offset */ + 0, /* flags */ + NULL, /* ctor */ + NULL); /* dtor */ + if (jbd2_handle_cache == NULL) { + printk(KERN_EMERG "JBD: failed to create handle cache\n"); + return -ENOMEM; + } + return 0; +} + +static void jbd2_journal_destroy_handle_cache(void) +{ + if (jbd2_handle_cache) + kmem_cache_destroy(jbd2_handle_cache); +} + +/* + * Module startup and shutdown + */ + +static int __init journal_init_caches(void) +{ + int ret; + + ret = jbd2_journal_init_revoke_caches(); + if (ret == 0) + ret = journal_init_jbd2_journal_head_cache(); + if (ret == 0) + ret = journal_init_handle_cache(); + return ret; +} + +static void jbd2_journal_destroy_caches(void) +{ + jbd2_journal_destroy_revoke_caches(); + jbd2_journal_destroy_jbd2_journal_head_cache(); + jbd2_journal_destroy_handle_cache(); + jbd2_journal_destroy_jbd_slabs(); +} + +static int __init journal_init(void) +{ + int ret; + + BUILD_BUG_ON(sizeof(struct journal_superblock_s) != 1024); + + ret = journal_init_caches(); + if (ret != 0) + jbd2_journal_destroy_caches(); + create_jbd_proc_entry(); + return ret; +} + +static void __exit journal_exit(void) +{ +#ifdef CONFIG_JBD_DEBUG + int n = atomic_read(&nr_journal_heads); + if (n) + printk(KERN_EMERG "JBD: leaked %d journal_heads!\n", n); +#endif + jbd2_remove_jbd_proc_entry(); + jbd2_journal_destroy_caches(); +} + +MODULE_LICENSE("GPL"); +module_init(journal_init); +module_exit(journal_exit); + diff --git a/fs/jbd2/recovery.c b/fs/jbd2/recovery.c new file mode 100644 index 000000000000..9f10acafaf70 --- /dev/null +++ b/fs/jbd2/recovery.c @@ -0,0 +1,609 @@ +/* + * linux/fs/recovery.c + * + * Written by Stephen C. Tweedie <sct@redhat.com>, 1999 + * + * Copyright 1999-2000 Red Hat Software --- All Rights Reserved + * + * This file is part of the Linux kernel and is made available under + * the terms of the GNU General Public License, version 2, or at your + * option, any later version, incorporated herein by reference. + * + * Journal recovery routines for the generic filesystem journaling code; + * part of the ext2fs journaling system. + */ + +#ifndef __KERNEL__ +#include "jfs_user.h" +#else +#include <linux/time.h> +#include <linux/fs.h> +#include <linux/jbd2.h> +#include <linux/errno.h> +#include <linux/slab.h> +#endif + +/* + * Maintain information about the progress of the recovery job, so that + * the different passes can carry information between them. + */ +struct recovery_info +{ + tid_t start_transaction; + tid_t end_transaction; + + int nr_replays; + int nr_revokes; + int nr_revoke_hits; +}; + +enum passtype {PASS_SCAN, PASS_REVOKE, PASS_REPLAY}; +static int do_one_pass(journal_t *journal, + struct recovery_info *info, enum passtype pass); +static int scan_revoke_records(journal_t *, struct buffer_head *, + tid_t, struct recovery_info *); + +#ifdef __KERNEL__ + +/* Release readahead buffers after use */ +static void journal_brelse_array(struct buffer_head *b[], int n) +{ + while (--n >= 0) + brelse (b[n]); +} + + +/* + * When reading from the journal, we are going through the block device + * layer directly and so there is no readahead being done for us. We + * need to implement any readahead ourselves if we want it to happen at + * all. Recovery is basically one long sequential read, so make sure we + * do the IO in reasonably large chunks. + * + * This is not so critical that we need to be enormously clever about + * the readahead size, though. 128K is a purely arbitrary, good-enough + * fixed value. + */ + +#define MAXBUF 8 +static int do_readahead(journal_t *journal, unsigned int start) +{ + int err; + unsigned int max, nbufs, next; + unsigned long long blocknr; + struct buffer_head *bh; + + struct buffer_head * bufs[MAXBUF]; + + /* Do up to 128K of readahead */ + max = start + (128 * 1024 / journal->j_blocksize); + if (max > journal->j_maxlen) + max = journal->j_maxlen; + + /* Do the readahead itself. We'll submit MAXBUF buffer_heads at + * a time to the block device IO layer. */ + + nbufs = 0; + + for (next = start; next < max; next++) { + err = jbd2_journal_bmap(journal, next, &blocknr); + + if (err) { + printk (KERN_ERR "JBD: bad block at offset %u\n", + next); + goto failed; + } + + bh = __getblk(journal->j_dev, blocknr, journal->j_blocksize); + if (!bh) { + err = -ENOMEM; + goto failed; + } + + if (!buffer_uptodate(bh) && !buffer_locked(bh)) { + bufs[nbufs++] = bh; + if (nbufs == MAXBUF) { + ll_rw_block(READ, nbufs, bufs); + journal_brelse_array(bufs, nbufs); + nbufs = 0; + } + } else + brelse(bh); + } + + if (nbufs) + ll_rw_block(READ, nbufs, bufs); + err = 0; + +failed: + if (nbufs) + journal_brelse_array(bufs, nbufs); + return err; +} + +#endif /* __KERNEL__ */ + + +/* + * Read a block from the journal + */ + +static int jread(struct buffer_head **bhp, journal_t *journal, + unsigned int offset) +{ + int err; + unsigned long long blocknr; + struct buffer_head *bh; + + *bhp = NULL; + + if (offset >= journal->j_maxlen) { + printk(KERN_ERR "JBD: corrupted journal superblock\n"); + return -EIO; + } + + err = jbd2_journal_bmap(journal, offset, &blocknr); + + if (err) { + printk (KERN_ERR "JBD: bad block at offset %u\n", + offset); + return err; + } + + bh = __getblk(journal->j_dev, blocknr, journal->j_blocksize); + if (!bh) + return -ENOMEM; + + if (!buffer_uptodate(bh)) { + /* If this is a brand new buffer, start readahead. + Otherwise, we assume we are already reading it. */ + if (!buffer_req(bh)) + do_readahead(journal, offset); + wait_on_buffer(bh); + } + + if (!buffer_uptodate(bh)) { + printk (KERN_ERR "JBD: Failed to read block at offset %u\n", + offset); + brelse(bh); + return -EIO; + } + + *bhp = bh; + return 0; +} + + +/* + * Count the number of in-use tags in a journal descriptor block. + */ + +static int count_tags(journal_t *journal, struct buffer_head *bh) +{ + char * tagp; + journal_block_tag_t * tag; + int nr = 0, size = journal->j_blocksize; + int tag_bytes = journal_tag_bytes(journal); + + tagp = &bh->b_data[sizeof(journal_header_t)]; + + while ((tagp - bh->b_data + tag_bytes) <= size) { + tag = (journal_block_tag_t *) tagp; + + nr++; + tagp += tag_bytes; + if (!(tag->t_flags & cpu_to_be32(JBD2_FLAG_SAME_UUID))) + tagp += 16; + + if (tag->t_flags & cpu_to_be32(JBD2_FLAG_LAST_TAG)) + break; + } + + return nr; +} + + +/* Make sure we wrap around the log correctly! */ +#define wrap(journal, var) \ +do { \ + if (var >= (journal)->j_last) \ + var -= ((journal)->j_last - (journal)->j_first); \ +} while (0) + +/** + * jbd2_journal_recover - recovers a on-disk journal + * @journal: the journal to recover + * + * The primary function for recovering the log contents when mounting a + * journaled device. + * + * Recovery is done in three passes. In the first pass, we look for the + * end of the log. In the second, we assemble the list of revoke + * blocks. In the third and final pass, we replay any un-revoked blocks + * in the log. + */ +int jbd2_journal_recover(journal_t *journal) +{ + int err; + journal_superblock_t * sb; + + struct recovery_info info; + + memset(&info, 0, sizeof(info)); + sb = journal->j_superblock; + + /* + * The journal superblock's s_start field (the current log head) + * is always zero if, and only if, the journal was cleanly + * unmounted. + */ + + if (!sb->s_start) { + jbd_debug(1, "No recovery required, last transaction %d\n", + be32_to_cpu(sb->s_sequence)); + journal->j_transaction_sequence = be32_to_cpu(sb->s_sequence) + 1; + return 0; + } + + err = do_one_pass(journal, &info, PASS_SCAN); + if (!err) + err = do_one_pass(journal, &info, PASS_REVOKE); + if (!err) + err = do_one_pass(journal, &info, PASS_REPLAY); + + jbd_debug(0, "JBD: recovery, exit status %d, " + "recovered transactions %u to %u\n", + err, info.start_transaction, info.end_transaction); + jbd_debug(0, "JBD: Replayed %d and revoked %d/%d blocks\n", + info.nr_replays, info.nr_revoke_hits, info.nr_revokes); + + /* Restart the log at the next transaction ID, thus invalidating + * any existing commit records in the log. */ + journal->j_transaction_sequence = ++info.end_transaction; + + jbd2_journal_clear_revoke(journal); + sync_blockdev(journal->j_fs_dev); + return err; +} + +/** + * jbd2_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 + * journal structures in memory to ignore it (presumably because the + * caller has evidence that it is out of date). + * This function does'nt appear to be exorted.. + * + * We perform one pass over the journal to allow us to tell the user how + * much recovery information is being erased, and to let us initialise + * the journal transaction sequence numbers to the next unused ID. + */ +int jbd2_journal_skip_recovery(journal_t *journal) +{ + int err; + journal_superblock_t * sb; + + struct recovery_info info; + + memset (&info, 0, sizeof(info)); + sb = journal->j_superblock; + + err = do_one_pass(journal, &info, PASS_SCAN); + + if (err) { + printk(KERN_ERR "JBD: error %d scanning journal\n", err); + ++journal->j_transaction_sequence; + } else { +#ifdef CONFIG_JBD_DEBUG + int dropped = info.end_transaction - be32_to_cpu(sb->s_sequence); +#endif + jbd_debug(0, + "JBD: ignoring %d transaction%s from the journal.\n", + dropped, (dropped == 1) ? "" : "s"); + journal->j_transaction_sequence = ++info.end_transaction; + } + + journal->j_tail = 0; + return err; +} + +static inline unsigned long long read_tag_block(int tag_bytes, journal_block_tag_t *tag) +{ + unsigned long long block = be32_to_cpu(tag->t_blocknr); + if (tag_bytes > JBD_TAG_SIZE32) + block |= (u64)be32_to_cpu(tag->t_blocknr_high) << 32; + return block; +} + +static int do_one_pass(journal_t *journal, + struct recovery_info *info, enum passtype pass) +{ + unsigned int first_commit_ID, next_commit_ID; + unsigned long next_log_block; + int err, success = 0; + journal_superblock_t * sb; + journal_header_t * tmp; + struct buffer_head * bh; + unsigned int sequence; + int blocktype; + int tag_bytes = journal_tag_bytes(journal); + + /* Precompute the maximum metadata descriptors in a descriptor block */ + int MAX_BLOCKS_PER_DESC; + MAX_BLOCKS_PER_DESC = ((journal->j_blocksize-sizeof(journal_header_t)) + / tag_bytes); + + /* + * First thing is to establish what we expect to find in the log + * (in terms of transaction IDs), and where (in terms of log + * block offsets): query the superblock. + */ + + sb = journal->j_superblock; + next_commit_ID = be32_to_cpu(sb->s_sequence); + next_log_block = be32_to_cpu(sb->s_start); + + first_commit_ID = next_commit_ID; + if (pass == PASS_SCAN) + info->start_transaction = first_commit_ID; + + jbd_debug(1, "Starting recovery pass %d\n", pass); + + /* + * Now we walk through the log, transaction by transaction, + * making sure that each transaction has a commit block in the + * expected place. Each complete transaction gets replayed back + * into the main filesystem. + */ + + while (1) { + int flags; + char * tagp; + journal_block_tag_t * tag; + struct buffer_head * obh; + struct buffer_head * nbh; + + cond_resched(); /* We're under lock_kernel() */ + + /* If we already know where to stop the log traversal, + * check right now that we haven't gone past the end of + * the log. */ + + if (pass != PASS_SCAN) + if (tid_geq(next_commit_ID, info->end_transaction)) + break; + + jbd_debug(2, "Scanning for sequence ID %u at %lu/%lu\n", + next_commit_ID, next_log_block, journal->j_last); + + /* Skip over each chunk of the transaction looking + * either the next descriptor block or the final commit + * record. */ + + jbd_debug(3, "JBD: checking block %ld\n", next_log_block); + err = jread(&bh, journal, next_log_block); + if (err) + goto failed; + + next_log_block++; + wrap(journal, next_log_block); + + /* What kind of buffer is it? + * + * If it is a descriptor block, check that it has the + * expected sequence number. Otherwise, we're all done + * here. */ + + tmp = (journal_header_t *)bh->b_data; + + if (tmp->h_magic != cpu_to_be32(JBD2_MAGIC_NUMBER)) { + brelse(bh); + break; + } + + blocktype = be32_to_cpu(tmp->h_blocktype); + sequence = be32_to_cpu(tmp->h_sequence); + jbd_debug(3, "Found magic %d, sequence %d\n", + blocktype, sequence); + + if (sequence != next_commit_ID) { + brelse(bh); + break; + } + + /* OK, we have a valid descriptor block which matches + * all of the sequence number checks. What are we going + * to do with it? That depends on the pass... */ + + switch(blocktype) { + case JBD2_DESCRIPTOR_BLOCK: + /* If it is a valid descriptor block, replay it + * in pass REPLAY; otherwise, just skip over the + * blocks it describes. */ + if (pass != PASS_REPLAY) { + next_log_block += count_tags(journal, bh); + wrap(journal, next_log_block); + brelse(bh); + continue; + } + + /* A descriptor block: we can now write all of + * the data blocks. Yay, useful work is finally + * getting done here! */ + + tagp = &bh->b_data[sizeof(journal_header_t)]; + while ((tagp - bh->b_data + tag_bytes) + <= journal->j_blocksize) { + unsigned long io_block; + + tag = (journal_block_tag_t *) tagp; + flags = be32_to_cpu(tag->t_flags); + + io_block = next_log_block++; + wrap(journal, next_log_block); + err = jread(&obh, journal, io_block); + if (err) { + /* Recover what we can, but + * report failure at the end. */ + success = err; + printk (KERN_ERR + "JBD: IO error %d recovering " + "block %ld in log\n", + err, io_block); + } else { + unsigned long long blocknr; + + J_ASSERT(obh != NULL); + blocknr = read_tag_block(tag_bytes, + tag); + + /* If the block has been + * revoked, then we're all done + * here. */ + if (jbd2_journal_test_revoke + (journal, blocknr, + next_commit_ID)) { + brelse(obh); + ++info->nr_revoke_hits; + goto skip_write; + } + + /* Find a buffer for the new + * data being restored */ + nbh = __getblk(journal->j_fs_dev, + blocknr, + journal->j_blocksize); + if (nbh == NULL) { + printk(KERN_ERR + "JBD: Out of memory " + "during recovery.\n"); + err = -ENOMEM; + brelse(bh); + brelse(obh); + goto failed; + } + + lock_buffer(nbh); + memcpy(nbh->b_data, obh->b_data, + journal->j_blocksize); + if (flags & JBD2_FLAG_ESCAPE) { + *((__be32 *)bh->b_data) = + cpu_to_be32(JBD2_MAGIC_NUMBER); + } + + BUFFER_TRACE(nbh, "marking dirty"); + set_buffer_uptodate(nbh); + mark_buffer_dirty(nbh); + BUFFER_TRACE(nbh, "marking uptodate"); + ++info->nr_replays; + /* ll_rw_block(WRITE, 1, &nbh); */ + unlock_buffer(nbh); + brelse(obh); + brelse(nbh); + } + + skip_write: + tagp += tag_bytes; + if (!(flags & JBD2_FLAG_SAME_UUID)) + tagp += 16; + + if (flags & JBD2_FLAG_LAST_TAG) + break; + } + + brelse(bh); + continue; + + case JBD2_COMMIT_BLOCK: + /* Found an expected commit block: not much to + * do other than move on to the next sequence + * number. */ + brelse(bh); + next_commit_ID++; + continue; + + case JBD2_REVOKE_BLOCK: + /* If we aren't in the REVOKE pass, then we can + * just skip over this block. */ + if (pass != PASS_REVOKE) { + brelse(bh); + continue; + } + + err = scan_revoke_records(journal, bh, + next_commit_ID, info); + brelse(bh); + if (err) + goto failed; + continue; + + default: + jbd_debug(3, "Unrecognised magic %d, end of scan.\n", + blocktype); + brelse(bh); + goto done; + } + } + + done: + /* + * We broke out of the log scan loop: either we came to the + * known end of the log or we found an unexpected block in the + * log. If the latter happened, then we know that the "current" + * transaction marks the end of the valid log. + */ + + if (pass == PASS_SCAN) + info->end_transaction = next_commit_ID; + else { + /* It's really bad news if different passes end up at + * different places (but possible due to IO errors). */ + if (info->end_transaction != next_commit_ID) { + printk (KERN_ERR "JBD: recovery pass %d ended at " + "transaction %u, expected %u\n", + pass, next_commit_ID, info->end_transaction); + if (!success) + success = -EIO; + } + } + + return success; + + failed: + return err; +} + + +/* Scan a revoke record, marking all blocks mentioned as revoked. */ + +static int scan_revoke_records(journal_t *journal, struct buffer_head *bh, + tid_t sequence, struct recovery_info *info) +{ + jbd2_journal_revoke_header_t *header; + int offset, max; + int record_len = 4; + + header = (jbd2_journal_revoke_header_t *) bh->b_data; + offset = sizeof(jbd2_journal_revoke_header_t); + max = be32_to_cpu(header->r_count); + + if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_64BIT)) + record_len = 8; + + while (offset + record_len <= max) { + unsigned long long blocknr; + int err; + + if (record_len == 4) + blocknr = be32_to_cpu(* ((__be32 *) (bh->b_data+offset))); + else + blocknr = be64_to_cpu(* ((__be64 *) (bh->b_data+offset))); + offset += record_len; + err = jbd2_journal_set_revoke(journal, blocknr, sequence); + if (err) + return err; + ++info->nr_revokes; + } + return 0; +} diff --git a/fs/jbd2/revoke.c b/fs/jbd2/revoke.c new file mode 100644 index 000000000000..380d19917f37 --- /dev/null +++ b/fs/jbd2/revoke.c @@ -0,0 +1,712 @@ +/* + * linux/fs/revoke.c + * + * Written by Stephen C. Tweedie <sct@redhat.com>, 2000 + * + * Copyright 2000 Red Hat corp --- All Rights Reserved + * + * This file is part of the Linux kernel and is made available under + * the terms of the GNU General Public License, version 2, or at your + * option, any later version, incorporated herein by reference. + * + * Journal revoke routines for the generic filesystem journaling code; + * part of the ext2fs journaling system. + * + * Revoke is the mechanism used to prevent old log records for deleted + * metadata from being replayed on top of newer data using the same + * blocks. The revoke mechanism is used in two separate places: + * + * + Commit: during commit we write the entire list of the current + * transaction's revoked blocks to the journal + * + * + Recovery: during recovery we record the transaction ID of all + * revoked blocks. If there are multiple revoke records in the log + * for a single block, only the last one counts, and if there is a log + * entry for a block beyond the last revoke, then that log entry still + * gets replayed. + * + * We can get interactions between revokes and new log data within a + * single transaction: + * + * Block is revoked and then journaled: + * The desired end result is the journaling of the new block, so we + * cancel the revoke before the transaction commits. + * + * Block is journaled and then revoked: + * The revoke must take precedence over the write of the block, so we + * need either to cancel the journal entry or to write the revoke + * later in the log than the log block. In this case, we choose the + * latter: journaling a block cancels any revoke record for that block + * in the current transaction, so any revoke for that block in the + * transaction must have happened after the block was journaled and so + * the revoke must take precedence. + * + * Block is revoked and then written as data: + * The data write is allowed to succeed, but the revoke is _not_ + * cancelled. We still need to prevent old log records from + * overwriting the new data. We don't even need to clear the revoke + * bit here. + * + * Revoke information on buffers is a tri-state value: + * + * RevokeValid clear: no cached revoke status, need to look it up + * RevokeValid set, Revoked clear: + * buffer has not been revoked, and cancel_revoke + * need do nothing. + * RevokeValid set, Revoked set: + * buffer has been revoked. + */ + +#ifndef __KERNEL__ +#include "jfs_user.h" +#else +#include <linux/time.h> +#include <linux/fs.h> +#include <linux/jbd2.h> +#include <linux/errno.h> +#include <linux/slab.h> +#include <linux/list.h> +#include <linux/smp_lock.h> +#include <linux/init.h> +#endif + +static kmem_cache_t *jbd2_revoke_record_cache; +static kmem_cache_t *jbd2_revoke_table_cache; + +/* Each revoke record represents one single revoked block. During + journal replay, this involves recording the transaction ID of the + last transaction to revoke this block. */ + +struct jbd2_revoke_record_s +{ + struct list_head hash; + tid_t sequence; /* Used for recovery only */ + unsigned long long blocknr; +}; + + +/* The revoke table is just a simple hash table of revoke records. */ +struct jbd2_revoke_table_s +{ + /* It is conceivable that we might want a larger hash table + * for recovery. Must be a power of two. */ + int hash_size; + int hash_shift; + struct list_head *hash_table; +}; + + +#ifdef __KERNEL__ +static void write_one_revoke_record(journal_t *, transaction_t *, + struct journal_head **, int *, + struct jbd2_revoke_record_s *); +static void flush_descriptor(journal_t *, struct journal_head *, int); +#endif + +/* Utility functions to maintain the revoke table */ + +/* Borrowed from buffer.c: this is a tried and tested block hash function */ +static inline int hash(journal_t *journal, unsigned long long block) +{ + struct jbd2_revoke_table_s *table = journal->j_revoke; + int hash_shift = table->hash_shift; + int hash = (int)block ^ (int)((block >> 31) >> 1); + + return ((hash << (hash_shift - 6)) ^ + (hash >> 13) ^ + (hash << (hash_shift - 12))) & (table->hash_size - 1); +} + +static int insert_revoke_hash(journal_t *journal, unsigned long long blocknr, + tid_t seq) +{ + struct list_head *hash_list; + struct jbd2_revoke_record_s *record; + +repeat: + record = kmem_cache_alloc(jbd2_revoke_record_cache, GFP_NOFS); + if (!record) + goto oom; + + record->sequence = seq; + record->blocknr = blocknr; + hash_list = &journal->j_revoke->hash_table[hash(journal, blocknr)]; + spin_lock(&journal->j_revoke_lock); + list_add(&record->hash, hash_list); + spin_unlock(&journal->j_revoke_lock); + return 0; + +oom: + if (!journal_oom_retry) + return -ENOMEM; + jbd_debug(1, "ENOMEM in %s, retrying\n", __FUNCTION__); + yield(); + goto repeat; +} + +/* Find a revoke record in the journal's hash table. */ + +static struct jbd2_revoke_record_s *find_revoke_record(journal_t *journal, + unsigned long long blocknr) +{ + struct list_head *hash_list; + struct jbd2_revoke_record_s *record; + + hash_list = &journal->j_revoke->hash_table[hash(journal, blocknr)]; + + spin_lock(&journal->j_revoke_lock); + record = (struct jbd2_revoke_record_s *) hash_list->next; + while (&(record->hash) != hash_list) { + if (record->blocknr == blocknr) { + spin_unlock(&journal->j_revoke_lock); + return record; + } + record = (struct jbd2_revoke_record_s *) record->hash.next; + } + spin_unlock(&journal->j_revoke_lock); + return NULL; +} + +int __init jbd2_journal_init_revoke_caches(void) +{ + jbd2_revoke_record_cache = kmem_cache_create("jbd2_revoke_record", + sizeof(struct jbd2_revoke_record_s), + 0, SLAB_HWCACHE_ALIGN, NULL, NULL); + if (jbd2_revoke_record_cache == 0) + return -ENOMEM; + + jbd2_revoke_table_cache = kmem_cache_create("jbd2_revoke_table", + sizeof(struct jbd2_revoke_table_s), + 0, 0, NULL, NULL); + if (jbd2_revoke_table_cache == 0) { + kmem_cache_destroy(jbd2_revoke_record_cache); + jbd2_revoke_record_cache = NULL; + return -ENOMEM; + } + return 0; +} + +void jbd2_journal_destroy_revoke_caches(void) +{ + kmem_cache_destroy(jbd2_revoke_record_cache); + jbd2_revoke_record_cache = NULL; + kmem_cache_destroy(jbd2_revoke_table_cache); + jbd2_revoke_table_cache = NULL; +} + +/* Initialise the revoke table for a given journal to a given size. */ + +int jbd2_journal_init_revoke(journal_t *journal, int hash_size) +{ + int shift, tmp; + + J_ASSERT (journal->j_revoke_table[0] == NULL); + + shift = 0; + tmp = hash_size; + while((tmp >>= 1UL) != 0UL) + shift++; + + journal->j_revoke_table[0] = kmem_cache_alloc(jbd2_revoke_table_cache, GFP_KERNEL); + if (!journal->j_revoke_table[0]) + return -ENOMEM; + journal->j_revoke = journal->j_revoke_table[0]; + + /* Check that the hash_size is a power of two */ + J_ASSERT ((hash_size & (hash_size-1)) == 0); + + journal->j_revoke->hash_size = hash_size; + + journal->j_revoke->hash_shift = shift; + + journal->j_revoke->hash_table = + kmalloc(hash_size * sizeof(struct list_head), GFP_KERNEL); + if (!journal->j_revoke->hash_table) { + kmem_cache_free(jbd2_revoke_table_cache, journal->j_revoke_table[0]); + journal->j_revoke = NULL; + return -ENOMEM; + } + + for (tmp = 0; tmp < hash_size; tmp++) + INIT_LIST_HEAD(&journal->j_revoke->hash_table[tmp]); + + journal->j_revoke_table[1] = kmem_cache_alloc(jbd2_revoke_table_cache, GFP_KERNEL); + if (!journal->j_revoke_table[1]) { + kfree(journal->j_revoke_table[0]->hash_table); + kmem_cache_free(jbd2_revoke_table_cache, journal->j_revoke_table[0]); + return -ENOMEM; + } + + journal->j_revoke = journal->j_revoke_table[1]; + + /* Check that the hash_size is a power of two */ + J_ASSERT ((hash_size & (hash_size-1)) == 0); + + journal->j_revoke->hash_size = hash_size; + + journal->j_revoke->hash_shift = shift; + + journal->j_revoke->hash_table = + kmalloc(hash_size * sizeof(struct list_head), GFP_KERNEL); + if (!journal->j_revoke->hash_table) { + kfree(journal->j_revoke_table[0]->hash_table); + kmem_cache_free(jbd2_revoke_table_cache, journal->j_revoke_table[0]); + kmem_cache_free(jbd2_revoke_table_cache, journal->j_revoke_table[1]); + journal->j_revoke = NULL; + return -ENOMEM; + } + + for (tmp = 0; tmp < hash_size; tmp++) + INIT_LIST_HEAD(&journal->j_revoke->hash_table[tmp]); + + spin_lock_init(&journal->j_revoke_lock); + + return 0; +} + +/* Destoy a journal's revoke table. The table must already be empty! */ + +void jbd2_journal_destroy_revoke(journal_t *journal) +{ + struct jbd2_revoke_table_s *table; + struct list_head *hash_list; + int i; + + table = journal->j_revoke_table[0]; + if (!table) + return; + + for (i=0; i<table->hash_size; i++) { + hash_list = &table->hash_table[i]; + J_ASSERT (list_empty(hash_list)); + } + + kfree(table->hash_table); + kmem_cache_free(jbd2_revoke_table_cache, table); + journal->j_revoke = NULL; + + table = journal->j_revoke_table[1]; + if (!table) + return; + + for (i=0; i<table->hash_size; i++) { + hash_list = &table->hash_table[i]; + J_ASSERT (list_empty(hash_list)); + } + + kfree(table->hash_table); + kmem_cache_free(jbd2_revoke_table_cache, table); + journal->j_revoke = NULL; +} + + +#ifdef __KERNEL__ + +/* + * jbd2_journal_revoke: revoke a given buffer_head from the journal. This + * prevents the block from being replayed during recovery if we take a + * crash after this current transaction commits. Any subsequent + * metadata writes of the buffer in this transaction cancel the + * revoke. + * + * Note that this call may block --- it is up to the caller to make + * sure that there are no further calls to journal_write_metadata + * before the revoke is complete. In ext3, this implies calling the + * revoke before clearing the block bitmap when we are deleting + * metadata. + * + * Revoke performs a jbd2_journal_forget on any buffer_head passed in as a + * parameter, but does _not_ forget the buffer_head if the bh was only + * found implicitly. + * + * bh_in may not be a journalled buffer - it may have come off + * the hash tables without an attached journal_head. + * + * If bh_in is non-zero, jbd2_journal_revoke() will decrement its b_count + * by one. + */ + +int jbd2_journal_revoke(handle_t *handle, unsigned long long blocknr, + struct buffer_head *bh_in) +{ + struct buffer_head *bh = NULL; + journal_t *journal; + struct block_device *bdev; + int err; + + might_sleep(); + if (bh_in) + BUFFER_TRACE(bh_in, "enter"); + + journal = handle->h_transaction->t_journal; + if (!jbd2_journal_set_features(journal, 0, 0, JBD2_FEATURE_INCOMPAT_REVOKE)){ + J_ASSERT (!"Cannot set revoke feature!"); + return -EINVAL; + } + + bdev = journal->j_fs_dev; + bh = bh_in; + + if (!bh) { + bh = __find_get_block(bdev, blocknr, journal->j_blocksize); + if (bh) + BUFFER_TRACE(bh, "found on hash"); + } +#ifdef JBD_EXPENSIVE_CHECKING + else { + struct buffer_head *bh2; + + /* If there is a different buffer_head lying around in + * memory anywhere... */ + bh2 = __find_get_block(bdev, blocknr, journal->j_blocksize); + if (bh2) { + /* ... and it has RevokeValid status... */ + if (bh2 != bh && buffer_revokevalid(bh2)) + /* ...then it better be revoked too, + * since it's illegal to create a revoke + * record against a buffer_head which is + * not marked revoked --- that would + * risk missing a subsequent revoke + * cancel. */ + J_ASSERT_BH(bh2, buffer_revoked(bh2)); + put_bh(bh2); + } + } +#endif + + /* We really ought not ever to revoke twice in a row without + first having the revoke cancelled: it's illegal to free a + block twice without allocating it in between! */ + if (bh) { + if (!J_EXPECT_BH(bh, !buffer_revoked(bh), + "inconsistent data on disk")) { + if (!bh_in) + brelse(bh); + return -EIO; + } + set_buffer_revoked(bh); + set_buffer_revokevalid(bh); + if (bh_in) { + BUFFER_TRACE(bh_in, "call jbd2_journal_forget"); + jbd2_journal_forget(handle, bh_in); + } else { + BUFFER_TRACE(bh, "call brelse"); + __brelse(bh); + } + } + + jbd_debug(2, "insert revoke for block %llu, bh_in=%p\n",blocknr, bh_in); + err = insert_revoke_hash(journal, blocknr, + handle->h_transaction->t_tid); + BUFFER_TRACE(bh_in, "exit"); + return err; +} + +/* + * Cancel an outstanding revoke. For use only internally by the + * journaling code (called from jbd2_journal_get_write_access). + * + * We trust buffer_revoked() on the buffer if the buffer is already + * being journaled: if there is no revoke pending on the buffer, then we + * don't do anything here. + * + * This would break if it were possible for a buffer to be revoked and + * discarded, and then reallocated within the same transaction. In such + * a case we would have lost the revoked bit, but when we arrived here + * the second time we would still have a pending revoke to cancel. So, + * do not trust the Revoked bit on buffers unless RevokeValid is also + * set. + * + * The caller must have the journal locked. + */ +int jbd2_journal_cancel_revoke(handle_t *handle, struct journal_head *jh) +{ + struct jbd2_revoke_record_s *record; + journal_t *journal = handle->h_transaction->t_journal; + int need_cancel; + int did_revoke = 0; /* akpm: debug */ + struct buffer_head *bh = jh2bh(jh); + + jbd_debug(4, "journal_head %p, cancelling revoke\n", jh); + + /* Is the existing Revoke bit valid? If so, we trust it, and + * only perform the full cancel if the revoke bit is set. If + * not, we can't trust the revoke bit, and we need to do the + * full search for a revoke record. */ + if (test_set_buffer_revokevalid(bh)) { + need_cancel = test_clear_buffer_revoked(bh); + } else { + need_cancel = 1; + clear_buffer_revoked(bh); + } + + if (need_cancel) { + record = find_revoke_record(journal, bh->b_blocknr); + if (record) { + jbd_debug(4, "cancelled existing revoke on " + "blocknr %llu\n", (unsigned long long)bh->b_blocknr); + spin_lock(&journal->j_revoke_lock); + list_del(&record->hash); + spin_unlock(&journal->j_revoke_lock); + kmem_cache_free(jbd2_revoke_record_cache, record); + did_revoke = 1; + } + } + +#ifdef JBD_EXPENSIVE_CHECKING + /* There better not be one left behind by now! */ + record = find_revoke_record(journal, bh->b_blocknr); + J_ASSERT_JH(jh, record == NULL); +#endif + + /* Finally, have we just cleared revoke on an unhashed + * buffer_head? If so, we'd better make sure we clear the + * revoked status on any hashed alias too, otherwise the revoke + * state machine will get very upset later on. */ + if (need_cancel) { + struct buffer_head *bh2; + bh2 = __find_get_block(bh->b_bdev, bh->b_blocknr, bh->b_size); + if (bh2) { + if (bh2 != bh) + clear_buffer_revoked(bh2); + __brelse(bh2); + } + } + return did_revoke; +} + +/* journal_switch_revoke table select j_revoke for next transaction + * we do not want to suspend any processing until all revokes are + * written -bzzz + */ +void jbd2_journal_switch_revoke_table(journal_t *journal) +{ + int i; + + if (journal->j_revoke == journal->j_revoke_table[0]) + journal->j_revoke = journal->j_revoke_table[1]; + else + journal->j_revoke = journal->j_revoke_table[0]; + + for (i = 0; i < journal->j_revoke->hash_size; i++) + INIT_LIST_HEAD(&journal->j_revoke->hash_table[i]); +} + +/* + * Write revoke records to the journal for all entries in the current + * revoke hash, deleting the entries as we go. + * + * Called with the journal lock held. + */ + +void jbd2_journal_write_revoke_records(journal_t *journal, + transaction_t *transaction) +{ + struct journal_head *descriptor; + struct jbd2_revoke_record_s *record; + struct jbd2_revoke_table_s *revoke; + struct list_head *hash_list; + int i, offset, count; + + descriptor = NULL; + offset = 0; + count = 0; + + /* select revoke table for committing transaction */ + revoke = journal->j_revoke == journal->j_revoke_table[0] ? + journal->j_revoke_table[1] : journal->j_revoke_table[0]; + + for (i = 0; i < revoke->hash_size; i++) { + hash_list = &revoke->hash_table[i]; + + while (!list_empty(hash_list)) { + record = (struct jbd2_revoke_record_s *) + hash_list->next; + write_one_revoke_record(journal, transaction, + &descriptor, &offset, + record); + count++; + list_del(&record->hash); + kmem_cache_free(jbd2_revoke_record_cache, record); + } + } + if (descriptor) + flush_descriptor(journal, descriptor, offset); + jbd_debug(1, "Wrote %d revoke records\n", count); +} + +/* + * Write out one revoke record. We need to create a new descriptor + * block if the old one is full or if we have not already created one. + */ + +static void write_one_revoke_record(journal_t *journal, + transaction_t *transaction, + struct journal_head **descriptorp, + int *offsetp, + struct jbd2_revoke_record_s *record) +{ + struct journal_head *descriptor; + int offset; + journal_header_t *header; + + /* If we are already aborting, this all becomes a noop. We + still need to go round the loop in + jbd2_journal_write_revoke_records in order to free all of the + revoke records: only the IO to the journal is omitted. */ + if (is_journal_aborted(journal)) + return; + + descriptor = *descriptorp; + offset = *offsetp; + + /* Make sure we have a descriptor with space left for the record */ + if (descriptor) { + if (offset == journal->j_blocksize) { + flush_descriptor(journal, descriptor, offset); + descriptor = NULL; + } + } + + if (!descriptor) { + descriptor = jbd2_journal_get_descriptor_buffer(journal); + if (!descriptor) + return; + header = (journal_header_t *) &jh2bh(descriptor)->b_data[0]; + header->h_magic = cpu_to_be32(JBD2_MAGIC_NUMBER); + header->h_blocktype = cpu_to_be32(JBD2_REVOKE_BLOCK); + header->h_sequence = cpu_to_be32(transaction->t_tid); + + /* Record it so that we can wait for IO completion later */ + JBUFFER_TRACE(descriptor, "file as BJ_LogCtl"); + jbd2_journal_file_buffer(descriptor, transaction, BJ_LogCtl); + + offset = sizeof(jbd2_journal_revoke_header_t); + *descriptorp = descriptor; + } + + if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_64BIT)) { + * ((__be64 *)(&jh2bh(descriptor)->b_data[offset])) = + cpu_to_be64(record->blocknr); + offset += 8; + + } else { + * ((__be32 *)(&jh2bh(descriptor)->b_data[offset])) = + cpu_to_be32(record->blocknr); + offset += 4; + } + + *offsetp = offset; +} + +/* + * Flush a revoke descriptor out to the journal. If we are aborting, + * this is a noop; otherwise we are generating a buffer which needs to + * be waited for during commit, so it has to go onto the appropriate + * journal buffer list. + */ + +static void flush_descriptor(journal_t *journal, + struct journal_head *descriptor, + int offset) +{ + jbd2_journal_revoke_header_t *header; + struct buffer_head *bh = jh2bh(descriptor); + + if (is_journal_aborted(journal)) { + put_bh(bh); + return; + } + + header = (jbd2_journal_revoke_header_t *) jh2bh(descriptor)->b_data; + header->r_count = cpu_to_be32(offset); + set_buffer_jwrite(bh); + BUFFER_TRACE(bh, "write"); + set_buffer_dirty(bh); + ll_rw_block(SWRITE, 1, &bh); +} +#endif + +/* + * Revoke support for recovery. + * + * Recovery needs to be able to: + * + * record all revoke records, including the tid of the latest instance + * of each revoke in the journal + * + * check whether a given block in a given transaction should be replayed + * (ie. has not been revoked by a revoke record in that or a subsequent + * transaction) + * + * empty the revoke table after recovery. + */ + +/* + * First, setting revoke records. We create a new revoke record for + * every block ever revoked in the log as we scan it for recovery, and + * we update the existing records if we find multiple revokes for a + * single block. + */ + +int jbd2_journal_set_revoke(journal_t *journal, + unsigned long long blocknr, + tid_t sequence) +{ + struct jbd2_revoke_record_s *record; + + record = find_revoke_record(journal, blocknr); + if (record) { + /* If we have multiple occurrences, only record the + * latest sequence number in the hashed record */ + if (tid_gt(sequence, record->sequence)) + record->sequence = sequence; + return 0; + } + return insert_revoke_hash(journal, blocknr, sequence); +} + +/* + * Test revoke records. For a given block referenced in the log, has + * that block been revoked? A revoke record with a given transaction + * sequence number revokes all blocks in that transaction and earlier + * ones, but later transactions still need replayed. + */ + +int jbd2_journal_test_revoke(journal_t *journal, + unsigned long long blocknr, + tid_t sequence) +{ + struct jbd2_revoke_record_s *record; + + record = find_revoke_record(journal, blocknr); + if (!record) + return 0; + if (tid_gt(sequence, record->sequence)) + return 0; + return 1; +} + +/* + * Finally, once recovery is over, we need to clear the revoke table so + * that it can be reused by the running filesystem. + */ + +void jbd2_journal_clear_revoke(journal_t *journal) +{ + int i; + struct list_head *hash_list; + struct jbd2_revoke_record_s *record; + struct jbd2_revoke_table_s *revoke; + + revoke = journal->j_revoke; + + for (i = 0; i < revoke->hash_size; i++) { + hash_list = &revoke->hash_table[i]; + while (!list_empty(hash_list)) { + record = (struct jbd2_revoke_record_s*) hash_list->next; + list_del(&record->hash); + kmem_cache_free(jbd2_revoke_record_cache, record); + } + } +} diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c new file mode 100644 index 000000000000..c051a94c8a97 --- /dev/null +++ b/fs/jbd2/transaction.c @@ -0,0 +1,2094 @@ +/* + * linux/fs/transaction.c + * + * Written by Stephen C. Tweedie <sct@redhat.com>, 1998 + * + * Copyright 1998 Red Hat corp --- All Rights Reserved + * + * This file is part of the Linux kernel and is made available under + * the terms of the GNU General Public License, version 2, or at your + * option, any later version, incorporated herein by reference. + * + * Generic filesystem transaction handling code; part of the ext2fs + * journaling system. + * + * This file manages transactions (compound commits managed by the + * journaling code) and handles (individual atomic operations by the + * filesystem). + */ + +#include <linux/time.h> +#include <linux/fs.h> +#include <linux/jbd2.h> +#include <linux/errno.h> +#include <linux/slab.h> +#include <linux/timer.h> +#include <linux/smp_lock.h> +#include <linux/mm.h> +#include <linux/highmem.h> + +/* + * jbd2_get_transaction: obtain a new transaction_t object. + * + * Simply allocate and initialise a new transaction. Create it in + * RUNNING state and add it to the current journal (which should not + * have an existing running transaction: we only make a new transaction + * once we have started to commit the old one). + * + * Preconditions: + * The journal MUST be locked. We don't perform atomic mallocs on the + * new transaction and we can't block without protecting against other + * processes trying to touch the journal while it is in transition. + * + * Called under j_state_lock + */ + +static transaction_t * +jbd2_get_transaction(journal_t *journal, transaction_t *transaction) +{ + transaction->t_journal = journal; + transaction->t_state = T_RUNNING; + transaction->t_tid = journal->j_transaction_sequence++; + transaction->t_expires = jiffies + journal->j_commit_interval; + spin_lock_init(&transaction->t_handle_lock); + + /* Set up the commit timer for the new transaction. */ + journal->j_commit_timer.expires = transaction->t_expires; + add_timer(&journal->j_commit_timer); + + J_ASSERT(journal->j_running_transaction == NULL); + journal->j_running_transaction = transaction; + + return transaction; +} + +/* + * Handle management. + * + * A handle_t is an object which represents a single atomic update to a + * filesystem, and which tracks all of the modifications which form part + * of that one update. + */ + +/* + * start_this_handle: Given a handle, deal with any locking or stalling + * needed to make sure that there is enough journal space for the handle + * to begin. Attach the handle to a transaction and set up the + * transaction's buffer credits. + */ + +static int start_this_handle(journal_t *journal, handle_t *handle) +{ + transaction_t *transaction; + int needed; + int nblocks = handle->h_buffer_credits; + transaction_t *new_transaction = NULL; + int ret = 0; + + if (nblocks > journal->j_max_transaction_buffers) { + printk(KERN_ERR "JBD: %s wants too many credits (%d > %d)\n", + current->comm, nblocks, + journal->j_max_transaction_buffers); + ret = -ENOSPC; + goto out; + } + +alloc_transaction: + if (!journal->j_running_transaction) { + new_transaction = jbd_kmalloc(sizeof(*new_transaction), + GFP_NOFS); + if (!new_transaction) { + ret = -ENOMEM; + goto out; + } + memset(new_transaction, 0, sizeof(*new_transaction)); + } + + jbd_debug(3, "New handle %p going live.\n", handle); + +repeat: + + /* + * We need to hold j_state_lock until t_updates has been incremented, + * for proper journal barrier handling + */ + spin_lock(&journal->j_state_lock); +repeat_locked: + if (is_journal_aborted(journal) || + (journal->j_errno != 0 && !(journal->j_flags & JBD2_ACK_ERR))) { + spin_unlock(&journal->j_state_lock); + ret = -EROFS; + goto out; + } + + /* Wait on the journal's transaction barrier if necessary */ + if (journal->j_barrier_count) { + spin_unlock(&journal->j_state_lock); + wait_event(journal->j_wait_transaction_locked, + journal->j_barrier_count == 0); + goto repeat; + } + + if (!journal->j_running_transaction) { + if (!new_transaction) { + spin_unlock(&journal->j_state_lock); + goto alloc_transaction; + } + jbd2_get_transaction(journal, new_transaction); + new_transaction = NULL; + } + + transaction = journal->j_running_transaction; + + /* + * If the current transaction is locked down for commit, wait for the + * lock to be released. + */ + if (transaction->t_state == T_LOCKED) { + DEFINE_WAIT(wait); + + prepare_to_wait(&journal->j_wait_transaction_locked, + &wait, TASK_UNINTERRUPTIBLE); + spin_unlock(&journal->j_state_lock); + schedule(); + finish_wait(&journal->j_wait_transaction_locked, &wait); + goto repeat; + } + + /* + * If there is not enough space left in the log to write all potential + * buffers requested by this operation, we need to stall pending a log + * checkpoint to free some more log space. + */ + spin_lock(&transaction->t_handle_lock); + needed = transaction->t_outstanding_credits + nblocks; + + if (needed > journal->j_max_transaction_buffers) { + /* + * If the current transaction is already too large, then start + * to commit it: we can then go back and attach this handle to + * a new transaction. + */ + DEFINE_WAIT(wait); + + jbd_debug(2, "Handle %p starting new commit...\n", handle); + spin_unlock(&transaction->t_handle_lock); + prepare_to_wait(&journal->j_wait_transaction_locked, &wait, + TASK_UNINTERRUPTIBLE); + __jbd2_log_start_commit(journal, transaction->t_tid); + spin_unlock(&journal->j_state_lock); + schedule(); + finish_wait(&journal->j_wait_transaction_locked, &wait); + goto repeat; + } + + /* + * The commit code assumes that it can get enough log space + * without forcing a checkpoint. This is *critical* for + * correctness: a checkpoint of a buffer which is also + * associated with a committing transaction creates a deadlock, + * so commit simply cannot force through checkpoints. + * + * We must therefore ensure the necessary space in the journal + * *before* starting to dirty potentially checkpointed buffers + * in the new transaction. + * + * The worst part is, any transaction currently committing can + * reduce the free space arbitrarily. Be careful to account for + * those buffers when checkpointing. + */ + + /* + * @@@ AKPM: This seems rather over-defensive. We're giving commit + * a _lot_ of headroom: 1/4 of the journal plus the size of + * the committing transaction. Really, we only need to give it + * committing_transaction->t_outstanding_credits plus "enough" for + * the log control blocks. + * Also, this test is inconsitent with the matching one in + * jbd2_journal_extend(). + */ + if (__jbd2_log_space_left(journal) < jbd_space_needed(journal)) { + jbd_debug(2, "Handle %p waiting for checkpoint...\n", handle); + spin_unlock(&transaction->t_handle_lock); + __jbd2_log_wait_for_space(journal); + goto repeat_locked; + } + + /* OK, account for the buffers that this operation expects to + * use and add the handle to the running transaction. */ + + handle->h_transaction = transaction; + transaction->t_outstanding_credits += nblocks; + transaction->t_updates++; + transaction->t_handle_count++; + jbd_debug(4, "Handle %p given %d credits (total %d, free %d)\n", + handle, nblocks, transaction->t_outstanding_credits, + __jbd2_log_space_left(journal)); + spin_unlock(&transaction->t_handle_lock); + spin_unlock(&journal->j_state_lock); +out: + if (unlikely(new_transaction)) /* It's usually NULL */ + kfree(new_transaction); + return ret; +} + +/* Allocate a new handle. This should probably be in a slab... */ +static handle_t *new_handle(int nblocks) +{ + handle_t *handle = jbd_alloc_handle(GFP_NOFS); + if (!handle) + return NULL; + memset(handle, 0, sizeof(*handle)); + handle->h_buffer_credits = nblocks; + handle->h_ref = 1; + + return handle; +} + +/** + * handle_t *jbd2_journal_start() - Obtain a new handle. + * @journal: Journal to start transaction on. + * @nblocks: number of block buffer we might modify + * + * We make sure that the transaction can guarantee at least nblocks of + * modified buffers in the log. We block until the log can guarantee + * that much space. + * + * This function is visible to journal users (like ext3fs), so is not + * called with the journal already locked. + * + * Return a pointer to a newly allocated handle, or NULL on failure + */ +handle_t *jbd2_journal_start(journal_t *journal, int nblocks) +{ + handle_t *handle = journal_current_handle(); + int err; + + if (!journal) + return ERR_PTR(-EROFS); + + if (handle) { + J_ASSERT(handle->h_transaction->t_journal == journal); + handle->h_ref++; + return handle; + } + + handle = new_handle(nblocks); + if (!handle) + return ERR_PTR(-ENOMEM); + + current->journal_info = handle; + + err = start_this_handle(journal, handle); + if (err < 0) { + jbd_free_handle(handle); + current->journal_info = NULL; + handle = ERR_PTR(err); + } + return handle; +} + +/** + * int jbd2_journal_extend() - extend buffer credits. + * @handle: handle to 'extend' + * @nblocks: nr blocks to try to extend by. + * + * Some transactions, such as large extends and truncates, can be done + * atomically all at once or in several stages. The operation requests + * a credit for a number of buffer modications in advance, but can + * extend its credit if it needs more. + * + * jbd2_journal_extend tries to give the running handle more buffer credits. + * It does not guarantee that allocation - this is a best-effort only. + * The calling process MUST be able to deal cleanly with a failure to + * extend here. + * + * Return 0 on success, non-zero on failure. + * + * return code < 0 implies an error + * return code > 0 implies normal transaction-full status. + */ +int jbd2_journal_extend(handle_t *handle, int nblocks) +{ + transaction_t *transaction = handle->h_transaction; + journal_t *journal = transaction->t_journal; + int result; + int wanted; + + result = -EIO; + if (is_handle_aborted(handle)) + goto out; + + result = 1; + + spin_lock(&journal->j_state_lock); + + /* Don't extend a locked-down transaction! */ + if (handle->h_transaction->t_state != T_RUNNING) { + jbd_debug(3, "denied handle %p %d blocks: " + "transaction not running\n", handle, nblocks); + goto error_out; + } + + spin_lock(&transaction->t_handle_lock); + wanted = transaction->t_outstanding_credits + nblocks; + + if (wanted > journal->j_max_transaction_buffers) { + jbd_debug(3, "denied handle %p %d blocks: " + "transaction too large\n", handle, nblocks); + goto unlock; + } + + if (wanted > __jbd2_log_space_left(journal)) { + jbd_debug(3, "denied handle %p %d blocks: " + "insufficient log space\n", handle, nblocks); + goto unlock; + } + + handle->h_buffer_credits += nblocks; + transaction->t_outstanding_credits += nblocks; + result = 0; + + jbd_debug(3, "extended handle %p by %d\n", handle, nblocks); +unlock: + spin_unlock(&transaction->t_handle_lock); +error_out: + spin_unlock(&journal->j_state_lock); +out: + return result; +} + + +/** + * int jbd2_journal_restart() - restart a handle . + * @handle: handle to restart + * @nblocks: nr credits requested + * + * Restart a handle for a multi-transaction filesystem + * operation. + * + * If the jbd2_journal_extend() call above fails to grant new buffer credits + * to a running handle, a call to jbd2_journal_restart will commit the + * handle's transaction so far and reattach the handle to a new + * transaction capabable of guaranteeing the requested number of + * credits. + */ + +int jbd2_journal_restart(handle_t *handle, int nblocks) +{ + transaction_t *transaction = handle->h_transaction; + journal_t *journal = transaction->t_journal; + int ret; + + /* If we've had an abort of any type, don't even think about + * actually doing the restart! */ + if (is_handle_aborted(handle)) + return 0; + + /* + * First unlink the handle from its current transaction, and start the + * commit on that. + */ + J_ASSERT(transaction->t_updates > 0); + J_ASSERT(journal_current_handle() == handle); + + spin_lock(&journal->j_state_lock); + spin_lock(&transaction->t_handle_lock); + transaction->t_outstanding_credits -= handle->h_buffer_credits; + transaction->t_updates--; + + if (!transaction->t_updates) + wake_up(&journal->j_wait_updates); + spin_unlock(&transaction->t_handle_lock); + + jbd_debug(2, "restarting handle %p\n", handle); + __jbd2_log_start_commit(journal, transaction->t_tid); + spin_unlock(&journal->j_state_lock); + + handle->h_buffer_credits = nblocks; + ret = start_this_handle(journal, handle); + return ret; +} + + +/** + * void jbd2_journal_lock_updates () - establish a transaction barrier. + * @journal: Journal to establish a barrier on. + * + * This locks out any further updates from being started, and blocks + * until all existing updates have completed, returning only once the + * journal is in a quiescent state with no updates running. + * + * The journal lock should not be held on entry. + */ +void jbd2_journal_lock_updates(journal_t *journal) +{ + DEFINE_WAIT(wait); + + spin_lock(&journal->j_state_lock); + ++journal->j_barrier_count; + + /* Wait until there are no running updates */ + while (1) { + transaction_t *transaction = journal->j_running_transaction; + + if (!transaction) + break; + + spin_lock(&transaction->t_handle_lock); + if (!transaction->t_updates) { + spin_unlock(&transaction->t_handle_lock); + break; + } + prepare_to_wait(&journal->j_wait_updates, &wait, + TASK_UNINTERRUPTIBLE); + spin_unlock(&transaction->t_handle_lock); + spin_unlock(&journal->j_state_lock); + schedule(); + finish_wait(&journal->j_wait_updates, &wait); + spin_lock(&journal->j_state_lock); + } + spin_unlock(&journal->j_state_lock); + + /* + * We have now established a barrier against other normal updates, but + * we also need to barrier against other jbd2_journal_lock_updates() calls + * to make sure that we serialise special journal-locked operations + * too. + */ + mutex_lock(&journal->j_barrier); +} + +/** + * void jbd2_journal_unlock_updates (journal_t* journal) - release barrier + * @journal: Journal to release the barrier on. + * + * Release a transaction barrier obtained with jbd2_journal_lock_updates(). + * + * Should be called without the journal lock held. + */ +void jbd2_journal_unlock_updates (journal_t *journal) +{ + J_ASSERT(journal->j_barrier_count != 0); + + mutex_unlock(&journal->j_barrier); + spin_lock(&journal->j_state_lock); + --journal->j_barrier_count; + spin_unlock(&journal->j_state_lock); + wake_up(&journal->j_wait_transaction_locked); +} + +/* + * Report any unexpected dirty buffers which turn up. Normally those + * indicate an error, but they can occur if the user is running (say) + * tune2fs to modify the live filesystem, so we need the option of + * continuing as gracefully as possible. # + * + * The caller should already hold the journal lock and + * j_list_lock spinlock: most callers will need those anyway + * in order to probe the buffer's journaling state safely. + */ +static void jbd_unexpected_dirty_buffer(struct journal_head *jh) +{ + int jlist; + + /* If this buffer is one which might reasonably be dirty + * --- ie. data, or not part of this journal --- then + * we're OK to leave it alone, but otherwise we need to + * move the dirty bit to the journal's own internal + * JBDDirty bit. */ + jlist = jh->b_jlist; + + if (jlist == BJ_Metadata || jlist == BJ_Reserved || + jlist == BJ_Shadow || jlist == BJ_Forget) { + struct buffer_head *bh = jh2bh(jh); + + if (test_clear_buffer_dirty(bh)) + set_buffer_jbddirty(bh); + } +} + +/* + * If the buffer is already part of the current transaction, then there + * is nothing we need to do. If it is already part of a prior + * transaction which we are still committing to disk, then we need to + * make sure that we do not overwrite the old copy: we do copy-out to + * preserve the copy going to disk. We also account the buffer against + * the handle's metadata buffer credits (unless the buffer is already + * part of the transaction, that is). + * + */ +static int +do_get_write_access(handle_t *handle, struct journal_head *jh, + int force_copy) +{ + struct buffer_head *bh; + transaction_t *transaction; + journal_t *journal; + int error; + char *frozen_buffer = NULL; + int need_copy = 0; + + if (is_handle_aborted(handle)) + return -EROFS; + + transaction = handle->h_transaction; + journal = transaction->t_journal; + + jbd_debug(5, "buffer_head %p, force_copy %d\n", jh, force_copy); + + JBUFFER_TRACE(jh, "entry"); +repeat: + bh = jh2bh(jh); + + /* @@@ Need to check for errors here at some point. */ + + lock_buffer(bh); + jbd_lock_bh_state(bh); + + /* We now hold the buffer lock so it is safe to query the buffer + * state. Is the buffer dirty? + * + * If so, there are two possibilities. The buffer may be + * non-journaled, and undergoing a quite legitimate writeback. + * Otherwise, it is journaled, and we don't expect dirty buffers + * in that state (the buffers should be marked JBD_Dirty + * instead.) So either the IO is being done under our own + * control and this is a bug, or it's a third party IO such as + * dump(8) (which may leave the buffer scheduled for read --- + * ie. locked but not dirty) or tune2fs (which may actually have + * the buffer dirtied, ugh.) */ + + if (buffer_dirty(bh)) { + /* + * First question: is this buffer already part of the current + * transaction or the existing committing transaction? + */ + if (jh->b_transaction) { + J_ASSERT_JH(jh, + jh->b_transaction == transaction || + jh->b_transaction == + journal->j_committing_transaction); + if (jh->b_next_transaction) + J_ASSERT_JH(jh, jh->b_next_transaction == + transaction); + } + /* + * In any case we need to clean the dirty flag and we must + * do it under the buffer lock to be sure we don't race + * with running write-out. + */ + JBUFFER_TRACE(jh, "Unexpected dirty buffer"); + jbd_unexpected_dirty_buffer(jh); + } + + unlock_buffer(bh); + + error = -EROFS; + if (is_handle_aborted(handle)) { + jbd_unlock_bh_state(bh); + goto out; + } + error = 0; + + /* + * The buffer is already part of this transaction if b_transaction or + * b_next_transaction points to it + */ + if (jh->b_transaction == transaction || + jh->b_next_transaction == transaction) + goto done; + + /* + * If there is already a copy-out version of this buffer, then we don't + * need to make another one + */ + if (jh->b_frozen_data) { + JBUFFER_TRACE(jh, "has frozen data"); + J_ASSERT_JH(jh, jh->b_next_transaction == NULL); + jh->b_next_transaction = transaction; + goto done; + } + + /* Is there data here we need to preserve? */ + + if (jh->b_transaction && jh->b_transaction != transaction) { + JBUFFER_TRACE(jh, "owned by older transaction"); + J_ASSERT_JH(jh, jh->b_next_transaction == NULL); + J_ASSERT_JH(jh, jh->b_transaction == + journal->j_committing_transaction); + + /* There is one case we have to be very careful about. + * If the committing transaction is currently writing + * this buffer out to disk and has NOT made a copy-out, + * then we cannot modify the buffer contents at all + * right now. The essence of copy-out is that it is the + * extra copy, not the primary copy, which gets + * journaled. If the primary copy is already going to + * disk then we cannot do copy-out here. */ + + if (jh->b_jlist == BJ_Shadow) { + DEFINE_WAIT_BIT(wait, &bh->b_state, BH_Unshadow); + wait_queue_head_t *wqh; + + wqh = bit_waitqueue(&bh->b_state, BH_Unshadow); + + JBUFFER_TRACE(jh, "on shadow: sleep"); + jbd_unlock_bh_state(bh); + /* commit wakes up all shadow buffers after IO */ + for ( ; ; ) { + prepare_to_wait(wqh, &wait.wait, + TASK_UNINTERRUPTIBLE); + if (jh->b_jlist != BJ_Shadow) + break; + schedule(); + } + finish_wait(wqh, &wait.wait); + goto repeat; + } + + /* Only do the copy if the currently-owning transaction + * still needs it. If it is on the Forget list, the + * committing transaction is past that stage. The + * buffer had better remain locked during the kmalloc, + * but that should be true --- we hold the journal lock + * still and the buffer is already on the BUF_JOURNAL + * list so won't be flushed. + * + * Subtle point, though: if this is a get_undo_access, + * then we will be relying on the frozen_data to contain + * the new value of the committed_data record after the + * transaction, so we HAVE to force the frozen_data copy + * in that case. */ + + if (jh->b_jlist != BJ_Forget || force_copy) { + JBUFFER_TRACE(jh, "generate frozen data"); + if (!frozen_buffer) { + JBUFFER_TRACE(jh, "allocate memory for buffer"); + jbd_unlock_bh_state(bh); + frozen_buffer = + jbd2_slab_alloc(jh2bh(jh)->b_size, + GFP_NOFS); + if (!frozen_buffer) { + printk(KERN_EMERG + "%s: OOM for frozen_buffer\n", + __FUNCTION__); + JBUFFER_TRACE(jh, "oom!"); + error = -ENOMEM; + jbd_lock_bh_state(bh); + goto done; + } + goto repeat; + } + jh->b_frozen_data = frozen_buffer; + frozen_buffer = NULL; + need_copy = 1; + } + jh->b_next_transaction = transaction; + } + + + /* + * Finally, if the buffer is not journaled right now, we need to make + * sure it doesn't get written to disk before the caller actually + * commits the new data + */ + if (!jh->b_transaction) { + JBUFFER_TRACE(jh, "no transaction"); + J_ASSERT_JH(jh, !jh->b_next_transaction); + jh->b_transaction = transaction; + JBUFFER_TRACE(jh, "file as BJ_Reserved"); + spin_lock(&journal->j_list_lock); + __jbd2_journal_file_buffer(jh, transaction, BJ_Reserved); + spin_unlock(&journal->j_list_lock); + } + +done: + if (need_copy) { + struct page *page; + int offset; + char *source; + + J_EXPECT_JH(jh, buffer_uptodate(jh2bh(jh)), + "Possible IO failure.\n"); + page = jh2bh(jh)->b_page; + offset = ((unsigned long) jh2bh(jh)->b_data) & ~PAGE_MASK; + source = kmap_atomic(page, KM_USER0); + memcpy(jh->b_frozen_data, source+offset, jh2bh(jh)->b_size); + kunmap_atomic(source, KM_USER0); + } + jbd_unlock_bh_state(bh); + + /* + * If we are about to journal a buffer, then any revoke pending on it is + * no longer valid + */ + jbd2_journal_cancel_revoke(handle, jh); + +out: + if (unlikely(frozen_buffer)) /* It's usually NULL */ + jbd2_slab_free(frozen_buffer, bh->b_size); + + JBUFFER_TRACE(jh, "exit"); + return error; +} + +/** + * int jbd2_journal_get_write_access() - notify intent to modify a buffer for metadata (not data) update. + * @handle: transaction to add buffer modifications to + * @bh: bh to be used for metadata writes + * @credits: variable that will receive credits for the buffer + * + * Returns an error code or 0 on success. + * + * In full data journalling mode the buffer may be of type BJ_AsyncData, + * because we're write()ing a buffer which is also part of a shared mapping. + */ + +int jbd2_journal_get_write_access(handle_t *handle, struct buffer_head *bh) +{ + struct journal_head *jh = jbd2_journal_add_journal_head(bh); + int rc; + + /* We do not want to get caught playing with fields which the + * log thread also manipulates. Make sure that the buffer + * completes any outstanding IO before proceeding. */ + rc = do_get_write_access(handle, jh, 0); + jbd2_journal_put_journal_head(jh); + return rc; +} + + +/* + * When the user wants to journal a newly created buffer_head + * (ie. getblk() returned a new buffer and we are going to populate it + * manually rather than reading off disk), then we need to keep the + * buffer_head locked until it has been completely filled with new + * data. In this case, we should be able to make the assertion that + * the bh is not already part of an existing transaction. + * + * The buffer should already be locked by the caller by this point. + * There is no lock ranking violation: it was a newly created, + * unlocked buffer beforehand. */ + +/** + * int jbd2_journal_get_create_access () - notify intent to use newly created bh + * @handle: transaction to new buffer to + * @bh: new buffer. + * + * Call this if you create a new bh. + */ +int jbd2_journal_get_create_access(handle_t *handle, struct buffer_head *bh) +{ + transaction_t *transaction = handle->h_transaction; + journal_t *journal = transaction->t_journal; + struct journal_head *jh = jbd2_journal_add_journal_head(bh); + int err; + + jbd_debug(5, "journal_head %p\n", jh); + err = -EROFS; + if (is_handle_aborted(handle)) + goto out; + err = 0; + + JBUFFER_TRACE(jh, "entry"); + /* + * The buffer may already belong to this transaction due to pre-zeroing + * in the filesystem's new_block code. It may also be on the previous, + * committing transaction's lists, but it HAS to be in Forget state in + * that case: the transaction must have deleted the buffer for it to be + * reused here. + */ + jbd_lock_bh_state(bh); + spin_lock(&journal->j_list_lock); + J_ASSERT_JH(jh, (jh->b_transaction == transaction || + jh->b_transaction == NULL || + (jh->b_transaction == journal->j_committing_transaction && + jh->b_jlist == BJ_Forget))); + + J_ASSERT_JH(jh, jh->b_next_transaction == NULL); + J_ASSERT_JH(jh, buffer_locked(jh2bh(jh))); + + if (jh->b_transaction == NULL) { + jh->b_transaction = transaction; + JBUFFER_TRACE(jh, "file as BJ_Reserved"); + __jbd2_journal_file_buffer(jh, transaction, BJ_Reserved); + } else if (jh->b_transaction == journal->j_committing_transaction) { + JBUFFER_TRACE(jh, "set next transaction"); + jh->b_next_transaction = transaction; + } + spin_unlock(&journal->j_list_lock); + jbd_unlock_bh_state(bh); + + /* + * akpm: I added this. ext3_alloc_branch can pick up new indirect + * blocks which contain freed but then revoked metadata. We need + * to cancel the revoke in case we end up freeing it yet again + * and the reallocating as data - this would cause a second revoke, + * which hits an assertion error. + */ + JBUFFER_TRACE(jh, "cancelling revoke"); + jbd2_journal_cancel_revoke(handle, jh); + jbd2_journal_put_journal_head(jh); +out: + return err; +} + +/** + * int jbd2_journal_get_undo_access() - Notify intent to modify metadata with + * non-rewindable consequences + * @handle: transaction + * @bh: buffer to undo + * @credits: store the number of taken credits here (if not NULL) + * + * Sometimes there is a need to distinguish between metadata which has + * been committed to disk and that which has not. The ext3fs code uses + * this for freeing and allocating space, we have to make sure that we + * do not reuse freed space until the deallocation has been committed, + * since if we overwrote that space we would make the delete + * un-rewindable in case of a crash. + * + * To deal with that, jbd2_journal_get_undo_access requests write access to a + * buffer for parts of non-rewindable operations such as delete + * operations on the bitmaps. The journaling code must keep a copy of + * the buffer's contents prior to the undo_access call until such time + * as we know that the buffer has definitely been committed to disk. + * + * We never need to know which transaction the committed data is part + * of, buffers touched here are guaranteed to be dirtied later and so + * will be committed to a new transaction in due course, at which point + * we can discard the old committed data pointer. + * + * Returns error number or 0 on success. + */ +int jbd2_journal_get_undo_access(handle_t *handle, struct buffer_head *bh) +{ + int err; + struct journal_head *jh = jbd2_journal_add_journal_head(bh); + char *committed_data = NULL; + + JBUFFER_TRACE(jh, "entry"); + + /* + * Do this first --- it can drop the journal lock, so we want to + * make sure that obtaining the committed_data is done + * atomically wrt. completion of any outstanding commits. + */ + err = do_get_write_access(handle, jh, 1); + if (err) + goto out; + +repeat: + if (!jh->b_committed_data) { + committed_data = jbd2_slab_alloc(jh2bh(jh)->b_size, GFP_NOFS); + if (!committed_data) { + printk(KERN_EMERG "%s: No memory for committed data\n", + __FUNCTION__); + err = -ENOMEM; + goto out; + } + } + + jbd_lock_bh_state(bh); + if (!jh->b_committed_data) { + /* Copy out the current buffer contents into the + * preserved, committed copy. */ + JBUFFER_TRACE(jh, "generate b_committed data"); + if (!committed_data) { + jbd_unlock_bh_state(bh); + goto repeat; + } + + jh->b_committed_data = committed_data; + committed_data = NULL; + memcpy(jh->b_committed_data, bh->b_data, bh->b_size); + } + jbd_unlock_bh_state(bh); +out: + jbd2_journal_put_journal_head(jh); + if (unlikely(committed_data)) + jbd2_slab_free(committed_data, bh->b_size); + return err; +} + +/** + * int jbd2_journal_dirty_data() - mark a buffer as containing dirty data which + * needs to be flushed before we can commit the + * current transaction. + * @handle: transaction + * @bh: bufferhead to mark + * + * The buffer is placed on the transaction's data list and is marked as + * belonging to the transaction. + * + * Returns error number or 0 on success. + * + * jbd2_journal_dirty_data() can be called via page_launder->ext3_writepage + * by kswapd. + */ +int jbd2_journal_dirty_data(handle_t *handle, struct buffer_head *bh) +{ + journal_t *journal = handle->h_transaction->t_journal; + int need_brelse = 0; + struct journal_head *jh; + + if (is_handle_aborted(handle)) + return 0; + + jh = jbd2_journal_add_journal_head(bh); + JBUFFER_TRACE(jh, "entry"); + + /* + * The buffer could *already* be dirty. Writeout can start + * at any time. + */ + jbd_debug(4, "jh: %p, tid:%d\n", jh, handle->h_transaction->t_tid); + + /* + * What if the buffer is already part of a running transaction? + * + * There are two cases: + * 1) It is part of the current running transaction. Refile it, + * just in case we have allocated it as metadata, deallocated + * it, then reallocated it as data. + * 2) It is part of the previous, still-committing transaction. + * If all we want to do is to guarantee that the buffer will be + * written to disk before this new transaction commits, then + * being sure that the *previous* transaction has this same + * property is sufficient for us! Just leave it on its old + * transaction. + * + * In case (2), the buffer must not already exist as metadata + * --- that would violate write ordering (a transaction is free + * to write its data at any point, even before the previous + * committing transaction has committed). The caller must + * never, ever allow this to happen: there's nothing we can do + * about it in this layer. + */ + jbd_lock_bh_state(bh); + spin_lock(&journal->j_list_lock); + + /* Now that we have bh_state locked, are we really still mapped? */ + if (!buffer_mapped(bh)) { + JBUFFER_TRACE(jh, "unmapped buffer, bailing out"); + goto no_journal; + } + + if (jh->b_transaction) { + JBUFFER_TRACE(jh, "has transaction"); + if (jh->b_transaction != handle->h_transaction) { + JBUFFER_TRACE(jh, "belongs to older transaction"); + J_ASSERT_JH(jh, jh->b_transaction == + journal->j_committing_transaction); + + /* @@@ IS THIS TRUE ? */ + /* + * Not any more. Scenario: someone does a write() + * in data=journal mode. The buffer's transaction has + * moved into commit. Then someone does another + * write() to the file. We do the frozen data copyout + * and set b_next_transaction to point to j_running_t. + * And while we're in that state, someone does a + * writepage() in an attempt to pageout the same area + * of the file via a shared mapping. At present that + * calls jbd2_journal_dirty_data(), and we get right here. + * It may be too late to journal the data. Simply + * falling through to the next test will suffice: the + * data will be dirty and wil be checkpointed. The + * ordering comments in the next comment block still + * apply. + */ + //J_ASSERT_JH(jh, jh->b_next_transaction == NULL); + + /* + * If we're journalling data, and this buffer was + * subject to a write(), it could be metadata, forget + * or shadow against the committing transaction. Now, + * someone has dirtied the same darn page via a mapping + * and it is being writepage()'d. + * We *could* just steal the page from commit, with some + * fancy locking there. Instead, we just skip it - + * don't tie the page's buffers to the new transaction + * at all. + * Implication: if we crash before the writepage() data + * is written into the filesystem, recovery will replay + * the write() data. + */ + if (jh->b_jlist != BJ_None && + jh->b_jlist != BJ_SyncData && + jh->b_jlist != BJ_Locked) { + JBUFFER_TRACE(jh, "Not stealing"); + goto no_journal; + } + + /* + * This buffer may be undergoing writeout in commit. We + * can't return from here and let the caller dirty it + * again because that can cause the write-out loop in + * commit to never terminate. + */ + if (buffer_dirty(bh)) { + get_bh(bh); + spin_unlock(&journal->j_list_lock); + jbd_unlock_bh_state(bh); + need_brelse = 1; + sync_dirty_buffer(bh); + jbd_lock_bh_state(bh); + spin_lock(&journal->j_list_lock); + /* Since we dropped the lock... */ + if (!buffer_mapped(bh)) { + JBUFFER_TRACE(jh, "buffer got unmapped"); + goto no_journal; + } + /* The buffer may become locked again at any + time if it is redirtied */ + } + + /* journal_clean_data_list() may have got there first */ + if (jh->b_transaction != NULL) { + JBUFFER_TRACE(jh, "unfile from commit"); + __jbd2_journal_temp_unlink_buffer(jh); + /* It still points to the committing + * transaction; move it to this one so + * that the refile assert checks are + * happy. */ + jh->b_transaction = handle->h_transaction; + } + /* The buffer will be refiled below */ + + } + /* + * Special case --- the buffer might actually have been + * allocated and then immediately deallocated in the previous, + * committing transaction, so might still be left on that + * transaction's metadata lists. + */ + if (jh->b_jlist != BJ_SyncData && jh->b_jlist != BJ_Locked) { + JBUFFER_TRACE(jh, "not on correct data list: unfile"); + J_ASSERT_JH(jh, jh->b_jlist != BJ_Shadow); + __jbd2_journal_temp_unlink_buffer(jh); + jh->b_transaction = handle->h_transaction; + JBUFFER_TRACE(jh, "file as data"); + __jbd2_journal_file_buffer(jh, handle->h_transaction, + BJ_SyncData); + } + } else { + JBUFFER_TRACE(jh, "not on a transaction"); + __jbd2_journal_file_buffer(jh, handle->h_transaction, BJ_SyncData); + } +no_journal: + spin_unlock(&journal->j_list_lock); + jbd_unlock_bh_state(bh); + if (need_brelse) { + BUFFER_TRACE(bh, "brelse"); + __brelse(bh); + } + JBUFFER_TRACE(jh, "exit"); + jbd2_journal_put_journal_head(jh); + return 0; +} + +/** + * int jbd2_journal_dirty_metadata() - mark a buffer as containing dirty metadata + * @handle: transaction to add buffer to. + * @bh: buffer to mark + * + * mark dirty metadata which needs to be journaled as part of the current + * transaction. + * + * The buffer is placed on the transaction's metadata list and is marked + * as belonging to the transaction. + * + * Returns error number or 0 on success. + * + * Special care needs to be taken if the buffer already belongs to the + * current committing transaction (in which case we should have frozen + * data present for that commit). In that case, we don't relink the + * buffer: that only gets done when the old transaction finally + * completes its commit. + */ +int jbd2_journal_dirty_metadata(handle_t *handle, struct buffer_head *bh) +{ + transaction_t *transaction = handle->h_transaction; + journal_t *journal = transaction->t_journal; + struct journal_head *jh = bh2jh(bh); + + jbd_debug(5, "journal_head %p\n", jh); + JBUFFER_TRACE(jh, "entry"); + if (is_handle_aborted(handle)) + goto out; + + jbd_lock_bh_state(bh); + + if (jh->b_modified == 0) { + /* + * This buffer's got modified and becoming part + * of the transaction. This needs to be done + * once a transaction -bzzz + */ + jh->b_modified = 1; + J_ASSERT_JH(jh, handle->h_buffer_credits > 0); + handle->h_buffer_credits--; + } + + /* + * fastpath, to avoid expensive locking. If this buffer is already + * on the running transaction's metadata list there is nothing to do. + * Nobody can take it off again because there is a handle open. + * I _think_ we're OK here with SMP barriers - a mistaken decision will + * result in this test being false, so we go in and take the locks. + */ + if (jh->b_transaction == transaction && jh->b_jlist == BJ_Metadata) { + JBUFFER_TRACE(jh, "fastpath"); + J_ASSERT_JH(jh, jh->b_transaction == + journal->j_running_transaction); + goto out_unlock_bh; + } + + set_buffer_jbddirty(bh); + + /* + * Metadata already on the current transaction list doesn't + * need to be filed. Metadata on another transaction's list must + * be committing, and will be refiled once the commit completes: + * leave it alone for now. + */ + if (jh->b_transaction != transaction) { + JBUFFER_TRACE(jh, "already on other transaction"); + J_ASSERT_JH(jh, jh->b_transaction == + journal->j_committing_transaction); + J_ASSERT_JH(jh, jh->b_next_transaction == transaction); + /* And this case is illegal: we can't reuse another + * transaction's data buffer, ever. */ + goto out_unlock_bh; + } + + /* That test should have eliminated the following case: */ + J_ASSERT_JH(jh, jh->b_frozen_data == 0); + + JBUFFER_TRACE(jh, "file as BJ_Metadata"); + spin_lock(&journal->j_list_lock); + __jbd2_journal_file_buffer(jh, handle->h_transaction, BJ_Metadata); + spin_unlock(&journal->j_list_lock); +out_unlock_bh: + jbd_unlock_bh_state(bh); +out: + JBUFFER_TRACE(jh, "exit"); + return 0; +} + +/* + * jbd2_journal_release_buffer: undo a get_write_access without any buffer + * updates, if the update decided in the end that it didn't need access. + * + */ +void +jbd2_journal_release_buffer(handle_t *handle, struct buffer_head *bh) +{ + BUFFER_TRACE(bh, "entry"); +} + +/** + * void jbd2_journal_forget() - bforget() for potentially-journaled buffers. + * @handle: transaction handle + * @bh: bh to 'forget' + * + * We can only do the bforget if there are no commits pending against the + * buffer. If the buffer is dirty in the current running transaction we + * can safely unlink it. + * + * bh may not be a journalled buffer at all - it may be a non-JBD + * buffer which came off the hashtable. Check for this. + * + * Decrements bh->b_count by one. + * + * Allow this call even if the handle has aborted --- it may be part of + * the caller's cleanup after an abort. + */ +int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh) +{ + transaction_t *transaction = handle->h_transaction; + journal_t *journal = transaction->t_journal; + struct journal_head *jh; + int drop_reserve = 0; + int err = 0; + + BUFFER_TRACE(bh, "entry"); + + jbd_lock_bh_state(bh); + spin_lock(&journal->j_list_lock); + + if (!buffer_jbd(bh)) + goto not_jbd; + jh = bh2jh(bh); + + /* Critical error: attempting to delete a bitmap buffer, maybe? + * Don't do any jbd operations, and return an error. */ + if (!J_EXPECT_JH(jh, !jh->b_committed_data, + "inconsistent data on disk")) { + err = -EIO; + goto not_jbd; + } + + /* + * The buffer's going from the transaction, we must drop + * all references -bzzz + */ + jh->b_modified = 0; + + if (jh->b_transaction == handle->h_transaction) { + J_ASSERT_JH(jh, !jh->b_frozen_data); + + /* If we are forgetting a buffer which is already part + * of this transaction, then we can just drop it from + * the transaction immediately. */ + clear_buffer_dirty(bh); + clear_buffer_jbddirty(bh); + + JBUFFER_TRACE(jh, "belongs to current transaction: unfile"); + + drop_reserve = 1; + + /* + * We are no longer going to journal this buffer. + * However, the commit of this transaction is still + * important to the buffer: the delete that we are now + * processing might obsolete an old log entry, so by + * committing, we can satisfy the buffer's checkpoint. + * + * So, if we have a checkpoint on the buffer, we should + * now refile the buffer on our BJ_Forget list so that + * we know to remove the checkpoint after we commit. + */ + + if (jh->b_cp_transaction) { + __jbd2_journal_temp_unlink_buffer(jh); + __jbd2_journal_file_buffer(jh, transaction, BJ_Forget); + } else { + __jbd2_journal_unfile_buffer(jh); + jbd2_journal_remove_journal_head(bh); + __brelse(bh); + if (!buffer_jbd(bh)) { + spin_unlock(&journal->j_list_lock); + jbd_unlock_bh_state(bh); + __bforget(bh); + goto drop; + } + } + } else if (jh->b_transaction) { + J_ASSERT_JH(jh, (jh->b_transaction == + journal->j_committing_transaction)); + /* However, if the buffer is still owned by a prior + * (committing) transaction, we can't drop it yet... */ + JBUFFER_TRACE(jh, "belongs to older transaction"); + /* ... but we CAN drop it from the new transaction if we + * have also modified it since the original commit. */ + + if (jh->b_next_transaction) { + J_ASSERT(jh->b_next_transaction == transaction); + jh->b_next_transaction = NULL; + drop_reserve = 1; + } + } + +not_jbd: + spin_unlock(&journal->j_list_lock); + jbd_unlock_bh_state(bh); + __brelse(bh); +drop: + if (drop_reserve) { + /* no need to reserve log space for this block -bzzz */ + handle->h_buffer_credits++; + } + return err; +} + +/** + * int jbd2_journal_stop() - complete a transaction + * @handle: tranaction to complete. + * + * All done for a particular handle. + * + * There is not much action needed here. We just return any remaining + * buffer credits to the transaction and remove the handle. The only + * complication is that we need to start a commit operation if the + * filesystem is marked for synchronous update. + * + * jbd2_journal_stop itself will not usually return an error, but it may + * do so in unusual circumstances. In particular, expect it to + * return -EIO if a jbd2_journal_abort has been executed since the + * transaction began. + */ +int jbd2_journal_stop(handle_t *handle) +{ + transaction_t *transaction = handle->h_transaction; + journal_t *journal = transaction->t_journal; + int old_handle_count, err; + pid_t pid; + + J_ASSERT(journal_current_handle() == handle); + + if (is_handle_aborted(handle)) + err = -EIO; + else { + J_ASSERT(transaction->t_updates > 0); + err = 0; + } + + if (--handle->h_ref > 0) { + jbd_debug(4, "h_ref %d -> %d\n", handle->h_ref + 1, + handle->h_ref); + return err; + } + + jbd_debug(4, "Handle %p going down\n", handle); + + /* + * Implement synchronous transaction batching. If the handle + * was synchronous, don't force a commit immediately. Let's + * yield and let another thread piggyback onto this transaction. + * Keep doing that while new threads continue to arrive. + * It doesn't cost much - we're about to run a commit and sleep + * on IO anyway. Speeds up many-threaded, many-dir operations + * by 30x or more... + * + * But don't do this if this process was the most recent one to + * perform a synchronous write. We do this to detect the case where a + * single process is doing a stream of sync writes. No point in waiting + * for joiners in that case. + */ + pid = current->pid; + if (handle->h_sync && journal->j_last_sync_writer != pid) { + journal->j_last_sync_writer = pid; + do { + old_handle_count = transaction->t_handle_count; + schedule_timeout_uninterruptible(1); + } while (old_handle_count != transaction->t_handle_count); + } + + current->journal_info = NULL; + spin_lock(&journal->j_state_lock); + spin_lock(&transaction->t_handle_lock); + transaction->t_outstanding_credits -= handle->h_buffer_credits; + transaction->t_updates--; + if (!transaction->t_updates) { + wake_up(&journal->j_wait_updates); + if (journal->j_barrier_count) + wake_up(&journal->j_wait_transaction_locked); + } + + /* + * If the handle is marked SYNC, we need to set another commit + * going! We also want to force a commit if the current + * transaction is occupying too much of the log, or if the + * transaction is too old now. + */ + if (handle->h_sync || + transaction->t_outstanding_credits > + journal->j_max_transaction_buffers || + time_after_eq(jiffies, transaction->t_expires)) { + /* Do this even for aborted journals: an abort still + * completes the commit thread, it just doesn't write + * anything to disk. */ + tid_t tid = transaction->t_tid; + + spin_unlock(&transaction->t_handle_lock); + jbd_debug(2, "transaction too old, requesting commit for " + "handle %p\n", handle); + /* This is non-blocking */ + __jbd2_log_start_commit(journal, transaction->t_tid); + spin_unlock(&journal->j_state_lock); + + /* + * Special case: JBD2_SYNC synchronous updates require us + * to wait for the commit to complete. + */ + if (handle->h_sync && !(current->flags & PF_MEMALLOC)) + err = jbd2_log_wait_commit(journal, tid); + } else { + spin_unlock(&transaction->t_handle_lock); + spin_unlock(&journal->j_state_lock); + } + + jbd_free_handle(handle); + return err; +} + +/**int jbd2_journal_force_commit() - force any uncommitted transactions + * @journal: journal to force + * + * For synchronous operations: force any uncommitted transactions + * to disk. May seem kludgy, but it reuses all the handle batching + * code in a very simple manner. + */ +int jbd2_journal_force_commit(journal_t *journal) +{ + handle_t *handle; + int ret; + + handle = jbd2_journal_start(journal, 1); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + } else { + handle->h_sync = 1; + ret = jbd2_journal_stop(handle); + } + return ret; +} + +/* + * + * List management code snippets: various functions for manipulating the + * transaction buffer lists. + * + */ + +/* + * Append a buffer to a transaction list, given the transaction's list head + * pointer. + * + * j_list_lock is held. + * + * jbd_lock_bh_state(jh2bh(jh)) is held. + */ + +static inline void +__blist_add_buffer(struct journal_head **list, struct journal_head *jh) +{ + if (!*list) { + jh->b_tnext = jh->b_tprev = jh; + *list = jh; + } else { + /* Insert at the tail of the list to preserve order */ + struct journal_head *first = *list, *last = first->b_tprev; + jh->b_tprev = last; + jh->b_tnext = first; + last->b_tnext = first->b_tprev = jh; + } +} + +/* + * Remove a buffer from a transaction list, given the transaction's list + * head pointer. + * + * Called with j_list_lock held, and the journal may not be locked. + * + * jbd_lock_bh_state(jh2bh(jh)) is held. + */ + +static inline void +__blist_del_buffer(struct journal_head **list, struct journal_head *jh) +{ + if (*list == jh) { + *list = jh->b_tnext; + if (*list == jh) + *list = NULL; + } + jh->b_tprev->b_tnext = jh->b_tnext; + jh->b_tnext->b_tprev = jh->b_tprev; +} + +/* + * Remove a buffer from the appropriate transaction list. + * + * Note that this function can *change* the value of + * bh->b_transaction->t_sync_datalist, t_buffers, t_forget, + * t_iobuf_list, t_shadow_list, t_log_list or t_reserved_list. If the caller + * is holding onto a copy of one of thee pointers, it could go bad. + * Generally the caller needs to re-read the pointer from the transaction_t. + * + * Called under j_list_lock. The journal may not be locked. + */ +void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh) +{ + struct journal_head **list = NULL; + transaction_t *transaction; + struct buffer_head *bh = jh2bh(jh); + + J_ASSERT_JH(jh, jbd_is_locked_bh_state(bh)); + transaction = jh->b_transaction; + if (transaction) + assert_spin_locked(&transaction->t_journal->j_list_lock); + + J_ASSERT_JH(jh, jh->b_jlist < BJ_Types); + if (jh->b_jlist != BJ_None) + J_ASSERT_JH(jh, transaction != 0); + + switch (jh->b_jlist) { + case BJ_None: + return; + case BJ_SyncData: + list = &transaction->t_sync_datalist; + break; + case BJ_Metadata: + transaction->t_nr_buffers--; + J_ASSERT_JH(jh, transaction->t_nr_buffers >= 0); + list = &transaction->t_buffers; + break; + case BJ_Forget: + list = &transaction->t_forget; + break; + case BJ_IO: + list = &transaction->t_iobuf_list; + break; + case BJ_Shadow: + list = &transaction->t_shadow_list; + break; + case BJ_LogCtl: + list = &transaction->t_log_list; + break; + case BJ_Reserved: + list = &transaction->t_reserved_list; + break; + case BJ_Locked: + list = &transaction->t_locked_list; + break; + } + + __blist_del_buffer(list, jh); + jh->b_jlist = BJ_None; + if (test_clear_buffer_jbddirty(bh)) + mark_buffer_dirty(bh); /* Expose it to the VM */ +} + +void __jbd2_journal_unfile_buffer(struct journal_head *jh) +{ + __jbd2_journal_temp_unlink_buffer(jh); + jh->b_transaction = NULL; +} + +void jbd2_journal_unfile_buffer(journal_t *journal, struct journal_head *jh) +{ + jbd_lock_bh_state(jh2bh(jh)); + spin_lock(&journal->j_list_lock); + __jbd2_journal_unfile_buffer(jh); + spin_unlock(&journal->j_list_lock); + jbd_unlock_bh_state(jh2bh(jh)); +} + +/* + * Called from jbd2_journal_try_to_free_buffers(). + * + * Called under jbd_lock_bh_state(bh) + */ +static void +__journal_try_to_free_buffer(journal_t *journal, struct buffer_head *bh) +{ + struct journal_head *jh; + + jh = bh2jh(bh); + + if (buffer_locked(bh) || buffer_dirty(bh)) + goto out; + + if (jh->b_next_transaction != 0) + goto out; + + spin_lock(&journal->j_list_lock); + if (jh->b_transaction != 0 && jh->b_cp_transaction == 0) { + if (jh->b_jlist == BJ_SyncData || jh->b_jlist == BJ_Locked) { + /* A written-back ordered data buffer */ + JBUFFER_TRACE(jh, "release data"); + __jbd2_journal_unfile_buffer(jh); + jbd2_journal_remove_journal_head(bh); + __brelse(bh); + } + } else if (jh->b_cp_transaction != 0 && jh->b_transaction == 0) { + /* written-back checkpointed metadata buffer */ + if (jh->b_jlist == BJ_None) { + JBUFFER_TRACE(jh, "remove from checkpoint list"); + __jbd2_journal_remove_checkpoint(jh); + jbd2_journal_remove_journal_head(bh); + __brelse(bh); + } + } + spin_unlock(&journal->j_list_lock); +out: + return; +} + + +/** + * int jbd2_journal_try_to_free_buffers() - try to free page buffers. + * @journal: journal for operation + * @page: to try and free + * @unused_gfp_mask: unused + * + * + * For all the buffers on this page, + * if they are fully written out ordered data, move them onto BUF_CLEAN + * so try_to_free_buffers() can reap them. + * + * This function returns non-zero if we wish try_to_free_buffers() + * to be called. We do this if the page is releasable by try_to_free_buffers(). + * We also do it if the page has locked or dirty buffers and the caller wants + * us to perform sync or async writeout. + * + * This complicates JBD locking somewhat. We aren't protected by the + * BKL here. We wish to remove the buffer from its committing or + * running transaction's ->t_datalist via __jbd2_journal_unfile_buffer. + * + * This may *change* the value of transaction_t->t_datalist, so anyone + * who looks at t_datalist needs to lock against this function. + * + * Even worse, someone may be doing a jbd2_journal_dirty_data on this + * buffer. So we need to lock against that. jbd2_journal_dirty_data() + * will come out of the lock with the buffer dirty, which makes it + * ineligible for release here. + * + * Who else is affected by this? hmm... Really the only contender + * is do_get_write_access() - it could be looking at the buffer while + * journal_try_to_free_buffer() is changing its state. But that + * cannot happen because we never reallocate freed data as metadata + * while the data is part of a transaction. Yes? + */ +int jbd2_journal_try_to_free_buffers(journal_t *journal, + struct page *page, gfp_t unused_gfp_mask) +{ + struct buffer_head *head; + struct buffer_head *bh; + int ret = 0; + + J_ASSERT(PageLocked(page)); + + head = page_buffers(page); + bh = head; + do { + struct journal_head *jh; + + /* + * We take our own ref against the journal_head here to avoid + * having to add tons of locking around each instance of + * jbd2_journal_remove_journal_head() and jbd2_journal_put_journal_head(). + */ + jh = jbd2_journal_grab_journal_head(bh); + if (!jh) + continue; + + jbd_lock_bh_state(bh); + __journal_try_to_free_buffer(journal, bh); + jbd2_journal_put_journal_head(jh); + jbd_unlock_bh_state(bh); + if (buffer_jbd(bh)) + goto busy; + } while ((bh = bh->b_this_page) != head); + ret = try_to_free_buffers(page); +busy: + return ret; +} + +/* + * This buffer is no longer needed. If it is on an older transaction's + * checkpoint list we need to record it on this transaction's forget list + * to pin this buffer (and hence its checkpointing transaction) down until + * this transaction commits. If the buffer isn't on a checkpoint list, we + * release it. + * Returns non-zero if JBD no longer has an interest in the buffer. + * + * Called under j_list_lock. + * + * Called under jbd_lock_bh_state(bh). + */ +static int __dispose_buffer(struct journal_head *jh, transaction_t *transaction) +{ + int may_free = 1; + struct buffer_head *bh = jh2bh(jh); + + __jbd2_journal_unfile_buffer(jh); + + if (jh->b_cp_transaction) { + JBUFFER_TRACE(jh, "on running+cp transaction"); + __jbd2_journal_file_buffer(jh, transaction, BJ_Forget); + clear_buffer_jbddirty(bh); + may_free = 0; + } else { + JBUFFER_TRACE(jh, "on running transaction"); + jbd2_journal_remove_journal_head(bh); + __brelse(bh); + } + return may_free; +} + +/* + * jbd2_journal_invalidatepage + * + * This code is tricky. It has a number of cases to deal with. + * + * There are two invariants which this code relies on: + * + * i_size must be updated on disk before we start calling invalidatepage on the + * data. + * + * This is done in ext3 by defining an ext3_setattr method which + * updates i_size before truncate gets going. By maintaining this + * invariant, we can be sure that it is safe to throw away any buffers + * attached to the current transaction: once the transaction commits, + * we know that the data will not be needed. + * + * Note however that we can *not* throw away data belonging to the + * previous, committing transaction! + * + * Any disk blocks which *are* part of the previous, committing + * transaction (and which therefore cannot be discarded immediately) are + * not going to be reused in the new running transaction + * + * The bitmap committed_data images guarantee this: any block which is + * allocated in one transaction and removed in the next will be marked + * as in-use in the committed_data bitmap, so cannot be reused until + * the next transaction to delete the block commits. This means that + * leaving committing buffers dirty is quite safe: the disk blocks + * cannot be reallocated to a different file and so buffer aliasing is + * not possible. + * + * + * The above applies mainly to ordered data mode. In writeback mode we + * don't make guarantees about the order in which data hits disk --- in + * particular we don't guarantee that new dirty data is flushed before + * transaction commit --- so it is always safe just to discard data + * immediately in that mode. --sct + */ + +/* + * The journal_unmap_buffer helper function returns zero if the buffer + * concerned remains pinned as an anonymous buffer belonging to an older + * transaction. + * + * We're outside-transaction here. Either or both of j_running_transaction + * and j_committing_transaction may be NULL. + */ +static int journal_unmap_buffer(journal_t *journal, struct buffer_head *bh) +{ + transaction_t *transaction; + struct journal_head *jh; + int may_free = 1; + int ret; + + BUFFER_TRACE(bh, "entry"); + + /* + * It is safe to proceed here without the j_list_lock because the + * buffers cannot be stolen by try_to_free_buffers as long as we are + * holding the page lock. --sct + */ + + if (!buffer_jbd(bh)) + goto zap_buffer_unlocked; + + spin_lock(&journal->j_state_lock); + jbd_lock_bh_state(bh); + spin_lock(&journal->j_list_lock); + + jh = jbd2_journal_grab_journal_head(bh); + if (!jh) + goto zap_buffer_no_jh; + + transaction = jh->b_transaction; + if (transaction == NULL) { + /* First case: not on any transaction. If it + * has no checkpoint link, then we can zap it: + * it's a writeback-mode buffer so we don't care + * if it hits disk safely. */ + if (!jh->b_cp_transaction) { + JBUFFER_TRACE(jh, "not on any transaction: zap"); + goto zap_buffer; + } + + if (!buffer_dirty(bh)) { + /* bdflush has written it. We can drop it now */ + goto zap_buffer; + } + + /* OK, it must be in the journal but still not + * written fully to disk: it's metadata or + * journaled data... */ + + if (journal->j_running_transaction) { + /* ... and once the current transaction has + * committed, the buffer won't be needed any + * longer. */ + JBUFFER_TRACE(jh, "checkpointed: add to BJ_Forget"); + ret = __dispose_buffer(jh, + journal->j_running_transaction); + jbd2_journal_put_journal_head(jh); + spin_unlock(&journal->j_list_lock); + jbd_unlock_bh_state(bh); + spin_unlock(&journal->j_state_lock); + return ret; + } else { + /* There is no currently-running transaction. So the + * orphan record which we wrote for this file must have + * passed into commit. We must attach this buffer to + * the committing transaction, if it exists. */ + if (journal->j_committing_transaction) { + JBUFFER_TRACE(jh, "give to committing trans"); + ret = __dispose_buffer(jh, + journal->j_committing_transaction); + jbd2_journal_put_journal_head(jh); + spin_unlock(&journal->j_list_lock); + jbd_unlock_bh_state(bh); + spin_unlock(&journal->j_state_lock); + return ret; + } else { + /* The orphan record's transaction has + * committed. We can cleanse this buffer */ + clear_buffer_jbddirty(bh); + goto zap_buffer; + } + } + } else if (transaction == journal->j_committing_transaction) { + JBUFFER_TRACE(jh, "on committing transaction"); + if (jh->b_jlist == BJ_Locked) { + /* + * The buffer is on the committing transaction's locked + * list. We have the buffer locked, so I/O has + * completed. So we can nail the buffer now. + */ + may_free = __dispose_buffer(jh, transaction); + goto zap_buffer; + } + /* + * If it is committing, we simply cannot touch it. We + * can remove it's next_transaction pointer from the + * running transaction if that is set, but nothing + * else. */ + set_buffer_freed(bh); + if (jh->b_next_transaction) { + J_ASSERT(jh->b_next_transaction == + journal->j_running_transaction); + jh->b_next_transaction = NULL; + } + jbd2_journal_put_journal_head(jh); + spin_unlock(&journal->j_list_lock); + jbd_unlock_bh_state(bh); + spin_unlock(&journal->j_state_lock); + return 0; + } else { + /* Good, the buffer belongs to the running transaction. + * We are writing our own transaction's data, not any + * previous one's, so it is safe to throw it away + * (remember that we expect the filesystem to have set + * i_size already for this truncate so recovery will not + * expose the disk blocks we are discarding here.) */ + J_ASSERT_JH(jh, transaction == journal->j_running_transaction); + JBUFFER_TRACE(jh, "on running transaction"); + may_free = __dispose_buffer(jh, transaction); + } + +zap_buffer: + jbd2_journal_put_journal_head(jh); +zap_buffer_no_jh: + spin_unlock(&journal->j_list_lock); + jbd_unlock_bh_state(bh); + spin_unlock(&journal->j_state_lock); +zap_buffer_unlocked: + clear_buffer_dirty(bh); + J_ASSERT_BH(bh, !buffer_jbddirty(bh)); + clear_buffer_mapped(bh); + clear_buffer_req(bh); + clear_buffer_new(bh); + bh->b_bdev = NULL; + return may_free; +} + +/** + * void jbd2_journal_invalidatepage() + * @journal: journal to use for flush... + * @page: page to flush + * @offset: length of page to invalidate. + * + * Reap page buffers containing data after offset in page. + * + */ +void jbd2_journal_invalidatepage(journal_t *journal, + struct page *page, + unsigned long offset) +{ + struct buffer_head *head, *bh, *next; + unsigned int curr_off = 0; + int may_free = 1; + + if (!PageLocked(page)) + BUG(); + if (!page_has_buffers(page)) + return; + + /* We will potentially be playing with lists other than just the + * data lists (especially for journaled data mode), so be + * cautious in our locking. */ + + head = bh = page_buffers(page); + do { + unsigned int next_off = curr_off + bh->b_size; + next = bh->b_this_page; + + if (offset <= curr_off) { + /* This block is wholly outside the truncation point */ + lock_buffer(bh); + may_free &= journal_unmap_buffer(journal, bh); + unlock_buffer(bh); + } + curr_off = next_off; + bh = next; + + } while (bh != head); + + if (!offset) { + if (may_free && try_to_free_buffers(page)) + J_ASSERT(!page_has_buffers(page)); + } +} + +/* + * File a buffer on the given transaction list. + */ +void __jbd2_journal_file_buffer(struct journal_head *jh, + transaction_t *transaction, int jlist) +{ + struct journal_head **list = NULL; + int was_dirty = 0; + struct buffer_head *bh = jh2bh(jh); + + J_ASSERT_JH(jh, jbd_is_locked_bh_state(bh)); + assert_spin_locked(&transaction->t_journal->j_list_lock); + + J_ASSERT_JH(jh, jh->b_jlist < BJ_Types); + J_ASSERT_JH(jh, jh->b_transaction == transaction || + jh->b_transaction == 0); + + if (jh->b_transaction && jh->b_jlist == jlist) + return; + + /* The following list of buffer states needs to be consistent + * with __jbd_unexpected_dirty_buffer()'s handling of dirty + * state. */ + + if (jlist == BJ_Metadata || jlist == BJ_Reserved || + jlist == BJ_Shadow || jlist == BJ_Forget) { + if (test_clear_buffer_dirty(bh) || + test_clear_buffer_jbddirty(bh)) + was_dirty = 1; + } + + if (jh->b_transaction) + __jbd2_journal_temp_unlink_buffer(jh); + jh->b_transaction = transaction; + + switch (jlist) { + case BJ_None: + J_ASSERT_JH(jh, !jh->b_committed_data); + J_ASSERT_JH(jh, !jh->b_frozen_data); + return; + case BJ_SyncData: + list = &transaction->t_sync_datalist; + break; + case BJ_Metadata: + transaction->t_nr_buffers++; + list = &transaction->t_buffers; + break; + case BJ_Forget: + list = &transaction->t_forget; + break; + case BJ_IO: + list = &transaction->t_iobuf_list; + break; + case BJ_Shadow: + list = &transaction->t_shadow_list; + break; + case BJ_LogCtl: + list = &transaction->t_log_list; + break; + case BJ_Reserved: + list = &transaction->t_reserved_list; + break; + case BJ_Locked: + list = &transaction->t_locked_list; + break; + } + + __blist_add_buffer(list, jh); + jh->b_jlist = jlist; + + if (was_dirty) + set_buffer_jbddirty(bh); +} + +void jbd2_journal_file_buffer(struct journal_head *jh, + transaction_t *transaction, int jlist) +{ + jbd_lock_bh_state(jh2bh(jh)); + spin_lock(&transaction->t_journal->j_list_lock); + __jbd2_journal_file_buffer(jh, transaction, jlist); + spin_unlock(&transaction->t_journal->j_list_lock); + jbd_unlock_bh_state(jh2bh(jh)); +} + +/* + * Remove a buffer from its current buffer list in preparation for + * dropping it from its current transaction entirely. If the buffer has + * already started to be used by a subsequent transaction, refile the + * buffer on that transaction's metadata list. + * + * Called under journal->j_list_lock + * + * Called under jbd_lock_bh_state(jh2bh(jh)) + */ +void __jbd2_journal_refile_buffer(struct journal_head *jh) +{ + int was_dirty; + struct buffer_head *bh = jh2bh(jh); + + J_ASSERT_JH(jh, jbd_is_locked_bh_state(bh)); + if (jh->b_transaction) + assert_spin_locked(&jh->b_transaction->t_journal->j_list_lock); + + /* If the buffer is now unused, just drop it. */ + if (jh->b_next_transaction == NULL) { + __jbd2_journal_unfile_buffer(jh); + return; + } + + /* + * It has been modified by a later transaction: add it to the new + * transaction's metadata list. + */ + + was_dirty = test_clear_buffer_jbddirty(bh); + __jbd2_journal_temp_unlink_buffer(jh); + jh->b_transaction = jh->b_next_transaction; + jh->b_next_transaction = NULL; + __jbd2_journal_file_buffer(jh, jh->b_transaction, + was_dirty ? BJ_Metadata : BJ_Reserved); + J_ASSERT_JH(jh, jh->b_transaction->t_state == T_RUNNING); + + if (was_dirty) + set_buffer_jbddirty(bh); +} + +/* + * For the unlocked version of this call, also make sure that any + * hanging journal_head is cleaned up if necessary. + * + * __jbd2_journal_refile_buffer is usually called as part of a single locked + * operation on a buffer_head, in which the caller is probably going to + * be hooking the journal_head onto other lists. In that case it is up + * to the caller to remove the journal_head if necessary. For the + * unlocked jbd2_journal_refile_buffer call, the caller isn't going to be + * doing anything else to the buffer so we need to do the cleanup + * ourselves to avoid a jh leak. + * + * *** The journal_head may be freed by this call! *** + */ +void jbd2_journal_refile_buffer(journal_t *journal, struct journal_head *jh) +{ + struct buffer_head *bh = jh2bh(jh); + + jbd_lock_bh_state(bh); + spin_lock(&journal->j_list_lock); + + __jbd2_journal_refile_buffer(jh); + jbd_unlock_bh_state(bh); + jbd2_journal_remove_journal_head(bh); + + spin_unlock(&journal->j_list_lock); + __brelse(bh); +} diff --git a/fs/jffs2/super.c b/fs/jffs2/super.c index 6de374513c01..bc4b8106a490 100644 --- a/fs/jffs2/super.c +++ b/fs/jffs2/super.c @@ -334,10 +334,10 @@ static int __init init_jffs2_fs(void) which means just 'no padding', without the alignment thing. But GCC doesn't have that -- we have to just hope the structs are the right sizes, instead. */ - BUG_ON(sizeof(struct jffs2_unknown_node) != 12); - BUG_ON(sizeof(struct jffs2_raw_dirent) != 40); - BUG_ON(sizeof(struct jffs2_raw_inode) != 68); - BUG_ON(sizeof(struct jffs2_raw_summary) != 32); + BUILD_BUG_ON(sizeof(struct jffs2_unknown_node) != 12); + BUILD_BUG_ON(sizeof(struct jffs2_raw_dirent) != 40); + BUILD_BUG_ON(sizeof(struct jffs2_raw_inode) != 68); + BUILD_BUG_ON(sizeof(struct jffs2_raw_summary) != 32); printk(KERN_INFO "JFFS2 version 2.2." #ifdef CONFIG_JFFS2_FS_WRITEBUFFER diff --git a/fs/jfs/file.c b/fs/jfs/file.c index 34181b8f5a0a..aa9132d04920 100644 --- a/fs/jfs/file.c +++ b/fs/jfs/file.c @@ -109,6 +109,8 @@ const struct file_operations jfs_file_operations = { .aio_write = generic_file_aio_write, .mmap = generic_file_mmap, .sendfile = generic_file_sendfile, + .splice_read = generic_file_splice_read, + .splice_write = generic_file_splice_write, .fsync = jfs_fsync, .release = jfs_release, .ioctl = jfs_ioctl, diff --git a/fs/jfs/jfs_imap.c b/fs/jfs/jfs_imap.c index 489a3d63002d..ee9b473b7b80 100644 --- a/fs/jfs/jfs_imap.c +++ b/fs/jfs/jfs_imap.c @@ -318,7 +318,7 @@ int diRead(struct inode *ip) struct inomap *imap; int block_offset; int inodes_left; - uint pageno; + unsigned long pageno; int rel_inode; jfs_info("diRead: ino = %ld", ip->i_ino); @@ -606,7 +606,7 @@ int diWrite(tid_t tid, struct inode *ip) int block_offset; int inodes_left; struct metapage *mp; - uint pageno; + unsigned long pageno; int rel_inode; int dioffset; struct inode *ipimap; diff --git a/fs/jfs/xattr.c b/fs/jfs/xattr.c index 4c7985ebca92..b753ba216450 100644 --- a/fs/jfs/xattr.c +++ b/fs/jfs/xattr.c @@ -756,6 +756,11 @@ static int can_set_system_xattr(struct inode *inode, const char *name, return -EOPNOTSUPP; } +/* + * Most of the permission checking is done by xattr_permission in the vfs. + * The local file system is responsible for handling the system.* namespace. + * We also need to verify that this is a namespace that we recognize. + */ static int can_set_xattr(struct inode *inode, const char *name, const void *value, size_t value_len) { @@ -771,10 +776,6 @@ static int can_set_xattr(struct inode *inode, const char *name, strncmp(name, XATTR_OS2_PREFIX, XATTR_OS2_PREFIX_LEN)) return -EOPNOTSUPP; - if (!S_ISREG(inode->i_mode) && - (!S_ISDIR(inode->i_mode) || inode->i_mode &S_ISVTX)) - return -EPERM; - return 0; } diff --git a/fs/lockd/clntlock.c b/fs/lockd/clntlock.c index e8c7765419e8..b85a0ad2cfb6 100644 --- a/fs/lockd/clntlock.c +++ b/fs/lockd/clntlock.c @@ -100,12 +100,12 @@ int nlmclnt_block(struct nlm_wait *block, struct nlm_rqst *req, long timeout) /* * The server lockd has called us back to tell us the lock was granted */ -u32 nlmclnt_grant(const struct sockaddr_in *addr, const struct nlm_lock *lock) +__be32 nlmclnt_grant(const struct sockaddr_in *addr, const struct nlm_lock *lock) { const struct file_lock *fl = &lock->fl; const struct nfs_fh *fh = &lock->fh; struct nlm_wait *block; - u32 res = nlm_lck_denied; + __be32 res = nlm_lck_denied; /* * Look up blocked request based on arguments. diff --git a/fs/lockd/mon.c b/fs/lockd/mon.c index e0179f8c327f..eb243edf8932 100644 --- a/fs/lockd/mon.c +++ b/fs/lockd/mon.c @@ -148,8 +148,8 @@ nsm_create(void) * XDR functions for NSM. */ -static u32 * -xdr_encode_common(struct rpc_rqst *rqstp, u32 *p, struct nsm_args *argp) +static __be32 * +xdr_encode_common(struct rpc_rqst *rqstp, __be32 *p, struct nsm_args *argp) { char buffer[20], *name; @@ -176,7 +176,7 @@ xdr_encode_common(struct rpc_rqst *rqstp, u32 *p, struct nsm_args *argp) } static int -xdr_encode_mon(struct rpc_rqst *rqstp, u32 *p, struct nsm_args *argp) +xdr_encode_mon(struct rpc_rqst *rqstp, __be32 *p, struct nsm_args *argp) { p = xdr_encode_common(rqstp, p, argp); if (IS_ERR(p)) @@ -192,7 +192,7 @@ xdr_encode_mon(struct rpc_rqst *rqstp, u32 *p, struct nsm_args *argp) } static int -xdr_encode_unmon(struct rpc_rqst *rqstp, u32 *p, struct nsm_args *argp) +xdr_encode_unmon(struct rpc_rqst *rqstp, __be32 *p, struct nsm_args *argp) { p = xdr_encode_common(rqstp, p, argp); if (IS_ERR(p)) @@ -202,7 +202,7 @@ xdr_encode_unmon(struct rpc_rqst *rqstp, u32 *p, struct nsm_args *argp) } static int -xdr_decode_stat_res(struct rpc_rqst *rqstp, u32 *p, struct nsm_res *resp) +xdr_decode_stat_res(struct rpc_rqst *rqstp, __be32 *p, struct nsm_res *resp) { resp->status = ntohl(*p++); resp->state = ntohl(*p++); @@ -212,7 +212,7 @@ xdr_decode_stat_res(struct rpc_rqst *rqstp, u32 *p, struct nsm_res *resp) } static int -xdr_decode_stat(struct rpc_rqst *rqstp, u32 *p, struct nsm_res *resp) +xdr_decode_stat(struct rpc_rqst *rqstp, __be32 *p, struct nsm_res *resp) { resp->state = ntohl(*p++); return 0; diff --git a/fs/lockd/svc4proc.c b/fs/lockd/svc4proc.c index fa370f6eb07b..0ce5c81ff507 100644 --- a/fs/lockd/svc4proc.c +++ b/fs/lockd/svc4proc.c @@ -24,14 +24,14 @@ /* * Obtain client and file from arguments */ -static u32 +static __be32 nlm4svc_retrieve_args(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_host **hostp, struct nlm_file **filp) { struct nlm_host *host = NULL; struct nlm_file *file = NULL; struct nlm_lock *lock = &argp->lock; - u32 error = 0; + __be32 error = 0; /* nfsd callbacks must have been installed for this procedure */ if (!nlmsvc_ops) @@ -68,7 +68,7 @@ no_locks: /* * NULL: Test for presence of service */ -static int +static __be32 nlm4svc_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) { dprintk("lockd: NULL called\n"); @@ -78,7 +78,7 @@ nlm4svc_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) /* * TEST: Check for conflicting lock */ -static int +static __be32 nlm4svc_proc_test(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -96,7 +96,7 @@ nlm4svc_proc_test(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlm4svc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; /* Now check for conflicting locks */ resp->status = nlmsvc_testlock(file, &argp->lock, &resp->lock); @@ -107,7 +107,7 @@ nlm4svc_proc_test(struct svc_rqst *rqstp, struct nlm_args *argp, return rpc_success; } -static int +static __be32 nlm4svc_proc_lock(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -126,7 +126,7 @@ nlm4svc_proc_lock(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlm4svc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; #if 0 /* If supplied state doesn't match current state, we assume it's @@ -150,7 +150,7 @@ nlm4svc_proc_lock(struct svc_rqst *rqstp, struct nlm_args *argp, return rpc_success; } -static int +static __be32 nlm4svc_proc_cancel(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -169,7 +169,7 @@ nlm4svc_proc_cancel(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlm4svc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; /* Try to cancel request. */ resp->status = nlmsvc_cancel_blocked(file, &argp->lock); @@ -183,7 +183,7 @@ nlm4svc_proc_cancel(struct svc_rqst *rqstp, struct nlm_args *argp, /* * UNLOCK: release a lock */ -static int +static __be32 nlm4svc_proc_unlock(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -202,7 +202,7 @@ nlm4svc_proc_unlock(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlm4svc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; /* Now try to remove the lock */ resp->status = nlmsvc_unlock(file, &argp->lock); @@ -217,7 +217,7 @@ nlm4svc_proc_unlock(struct svc_rqst *rqstp, struct nlm_args *argp, * GRANTED: A server calls us to tell that a process' lock request * was granted */ -static int +static __be32 nlm4svc_proc_granted(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -253,12 +253,12 @@ static const struct rpc_call_ops nlm4svc_callback_ops = { * because we send the callback before the reply proper. I hope this * doesn't break any clients. */ -static int nlm4svc_callback(struct svc_rqst *rqstp, u32 proc, struct nlm_args *argp, - int (*func)(struct svc_rqst *, struct nlm_args *, struct nlm_res *)) +static __be32 nlm4svc_callback(struct svc_rqst *rqstp, u32 proc, struct nlm_args *argp, + __be32 (*func)(struct svc_rqst *, struct nlm_args *, struct nlm_res *)) { struct nlm_host *host; struct nlm_rqst *call; - int stat; + __be32 stat; host = nlmsvc_lookup_host(rqstp, argp->lock.caller, @@ -282,35 +282,35 @@ static int nlm4svc_callback(struct svc_rqst *rqstp, u32 proc, struct nlm_args *a return rpc_success; } -static int nlm4svc_proc_test_msg(struct svc_rqst *rqstp, struct nlm_args *argp, +static __be32 nlm4svc_proc_test_msg(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { dprintk("lockd: TEST_MSG called\n"); return nlm4svc_callback(rqstp, NLMPROC_TEST_RES, argp, nlm4svc_proc_test); } -static int nlm4svc_proc_lock_msg(struct svc_rqst *rqstp, struct nlm_args *argp, +static __be32 nlm4svc_proc_lock_msg(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { dprintk("lockd: LOCK_MSG called\n"); return nlm4svc_callback(rqstp, NLMPROC_LOCK_RES, argp, nlm4svc_proc_lock); } -static int nlm4svc_proc_cancel_msg(struct svc_rqst *rqstp, struct nlm_args *argp, +static __be32 nlm4svc_proc_cancel_msg(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { dprintk("lockd: CANCEL_MSG called\n"); return nlm4svc_callback(rqstp, NLMPROC_CANCEL_RES, argp, nlm4svc_proc_cancel); } -static int nlm4svc_proc_unlock_msg(struct svc_rqst *rqstp, struct nlm_args *argp, +static __be32 nlm4svc_proc_unlock_msg(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { dprintk("lockd: UNLOCK_MSG called\n"); return nlm4svc_callback(rqstp, NLMPROC_UNLOCK_RES, argp, nlm4svc_proc_unlock); } -static int nlm4svc_proc_granted_msg(struct svc_rqst *rqstp, struct nlm_args *argp, +static __be32 nlm4svc_proc_granted_msg(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { dprintk("lockd: GRANTED_MSG called\n"); @@ -320,7 +320,7 @@ static int nlm4svc_proc_granted_msg(struct svc_rqst *rqstp, struct nlm_args *arg /* * SHARE: create a DOS share or alter existing share. */ -static int +static __be32 nlm4svc_proc_share(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -339,7 +339,7 @@ nlm4svc_proc_share(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlm4svc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; /* Now try to create the share */ resp->status = nlmsvc_share_file(host, file, argp); @@ -353,7 +353,7 @@ nlm4svc_proc_share(struct svc_rqst *rqstp, struct nlm_args *argp, /* * UNSHARE: Release a DOS share. */ -static int +static __be32 nlm4svc_proc_unshare(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -372,7 +372,7 @@ nlm4svc_proc_unshare(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlm4svc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; /* Now try to lock the file */ resp->status = nlmsvc_unshare_file(host, file, argp); @@ -386,7 +386,7 @@ nlm4svc_proc_unshare(struct svc_rqst *rqstp, struct nlm_args *argp, /* * NM_LOCK: Create an unmonitored lock */ -static int +static __be32 nlm4svc_proc_nm_lock(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -399,7 +399,7 @@ nlm4svc_proc_nm_lock(struct svc_rqst *rqstp, struct nlm_args *argp, /* * FREE_ALL: Release all locks and shares held by client */ -static int +static __be32 nlm4svc_proc_free_all(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { @@ -417,7 +417,7 @@ nlm4svc_proc_free_all(struct svc_rqst *rqstp, struct nlm_args *argp, /* * SM_NOTIFY: private callback from statd (not part of official NLM proto) */ -static int +static __be32 nlm4svc_proc_sm_notify(struct svc_rqst *rqstp, struct nlm_reboot *argp, void *resp) { @@ -446,7 +446,7 @@ nlm4svc_proc_sm_notify(struct svc_rqst *rqstp, struct nlm_reboot *argp, /* * client sent a GRANTED_RES, let's remove the associated block */ -static int +static __be32 nlm4svc_proc_granted_res(struct svc_rqst *rqstp, struct nlm_res *argp, void *resp) { diff --git a/fs/lockd/svclock.c b/fs/lockd/svclock.c index 814c6064c9e0..7e219b938552 100644 --- a/fs/lockd/svclock.c +++ b/fs/lockd/svclock.c @@ -334,13 +334,13 @@ static void nlmsvc_freegrantargs(struct nlm_rqst *call) * Attempt to establish a lock, and if it can't be granted, block it * if required. */ -u32 +__be32 nlmsvc_lock(struct svc_rqst *rqstp, struct nlm_file *file, struct nlm_lock *lock, int wait, struct nlm_cookie *cookie) { struct nlm_block *block, *newblock = NULL; int error; - u32 ret; + __be32 ret; dprintk("lockd: nlmsvc_lock(%s/%ld, ty=%d, pi=%d, %Ld-%Ld, bl=%d)\n", file->f_file->f_dentry->d_inode->i_sb->s_id, @@ -415,7 +415,7 @@ out: /* * Test for presence of a conflicting lock. */ -u32 +__be32 nlmsvc_testlock(struct nlm_file *file, struct nlm_lock *lock, struct nlm_lock *conflock) { @@ -448,7 +448,7 @@ nlmsvc_testlock(struct nlm_file *file, struct nlm_lock *lock, * afterwards. In this case the block will still be there, and hence * must be removed. */ -u32 +__be32 nlmsvc_unlock(struct nlm_file *file, struct nlm_lock *lock) { int error; @@ -476,7 +476,7 @@ nlmsvc_unlock(struct nlm_file *file, struct nlm_lock *lock) * be in progress. * The calling procedure must check whether the file can be closed. */ -u32 +__be32 nlmsvc_cancel_blocked(struct nlm_file *file, struct nlm_lock *lock) { struct nlm_block *block; diff --git a/fs/lockd/svcproc.c b/fs/lockd/svcproc.c index 75b2c81bcb93..32e99a6e8dca 100644 --- a/fs/lockd/svcproc.c +++ b/fs/lockd/svcproc.c @@ -22,8 +22,8 @@ #define NLMDBG_FACILITY NLMDBG_CLIENT #ifdef CONFIG_LOCKD_V4 -static u32 -cast_to_nlm(u32 status, u32 vers) +static __be32 +cast_to_nlm(__be32 status, u32 vers) { /* Note: status is assumed to be in network byte order !!! */ if (vers != 4){ @@ -52,14 +52,14 @@ cast_to_nlm(u32 status, u32 vers) /* * Obtain client and file from arguments */ -static u32 +static __be32 nlmsvc_retrieve_args(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_host **hostp, struct nlm_file **filp) { struct nlm_host *host = NULL; struct nlm_file *file = NULL; struct nlm_lock *lock = &argp->lock; - u32 error; + __be32 error = 0; /* nfsd callbacks must have been installed for this procedure */ if (!nlmsvc_ops) @@ -88,13 +88,15 @@ nlmsvc_retrieve_args(struct svc_rqst *rqstp, struct nlm_args *argp, no_locks: if (host) nlm_release_host(host); + if (error) + return error; return nlm_lck_denied_nolocks; } /* * NULL: Test for presence of service */ -static int +static __be32 nlmsvc_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) { dprintk("lockd: NULL called\n"); @@ -104,7 +106,7 @@ nlmsvc_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) /* * TEST: Check for conflicting lock */ -static int +static __be32 nlmsvc_proc_test(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -122,7 +124,7 @@ nlmsvc_proc_test(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlmsvc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; /* Now check for conflicting locks */ resp->status = cast_status(nlmsvc_testlock(file, &argp->lock, &resp->lock)); @@ -134,7 +136,7 @@ nlmsvc_proc_test(struct svc_rqst *rqstp, struct nlm_args *argp, return rpc_success; } -static int +static __be32 nlmsvc_proc_lock(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -153,7 +155,7 @@ nlmsvc_proc_lock(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlmsvc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; #if 0 /* If supplied state doesn't match current state, we assume it's @@ -177,7 +179,7 @@ nlmsvc_proc_lock(struct svc_rqst *rqstp, struct nlm_args *argp, return rpc_success; } -static int +static __be32 nlmsvc_proc_cancel(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -196,7 +198,7 @@ nlmsvc_proc_cancel(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlmsvc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; /* Try to cancel request. */ resp->status = cast_status(nlmsvc_cancel_blocked(file, &argp->lock)); @@ -210,7 +212,7 @@ nlmsvc_proc_cancel(struct svc_rqst *rqstp, struct nlm_args *argp, /* * UNLOCK: release a lock */ -static int +static __be32 nlmsvc_proc_unlock(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -229,7 +231,7 @@ nlmsvc_proc_unlock(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlmsvc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; /* Now try to remove the lock */ resp->status = cast_status(nlmsvc_unlock(file, &argp->lock)); @@ -244,7 +246,7 @@ nlmsvc_proc_unlock(struct svc_rqst *rqstp, struct nlm_args *argp, * GRANTED: A server calls us to tell that a process' lock request * was granted */ -static int +static __be32 nlmsvc_proc_granted(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -280,12 +282,12 @@ static const struct rpc_call_ops nlmsvc_callback_ops = { * because we send the callback before the reply proper. I hope this * doesn't break any clients. */ -static int nlmsvc_callback(struct svc_rqst *rqstp, u32 proc, struct nlm_args *argp, - int (*func)(struct svc_rqst *, struct nlm_args *, struct nlm_res *)) +static __be32 nlmsvc_callback(struct svc_rqst *rqstp, u32 proc, struct nlm_args *argp, + __be32 (*func)(struct svc_rqst *, struct nlm_args *, struct nlm_res *)) { struct nlm_host *host; struct nlm_rqst *call; - int stat; + __be32 stat; host = nlmsvc_lookup_host(rqstp, argp->lock.caller, @@ -309,28 +311,28 @@ static int nlmsvc_callback(struct svc_rqst *rqstp, u32 proc, struct nlm_args *ar return rpc_success; } -static int nlmsvc_proc_test_msg(struct svc_rqst *rqstp, struct nlm_args *argp, +static __be32 nlmsvc_proc_test_msg(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { dprintk("lockd: TEST_MSG called\n"); return nlmsvc_callback(rqstp, NLMPROC_TEST_RES, argp, nlmsvc_proc_test); } -static int nlmsvc_proc_lock_msg(struct svc_rqst *rqstp, struct nlm_args *argp, +static __be32 nlmsvc_proc_lock_msg(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { dprintk("lockd: LOCK_MSG called\n"); return nlmsvc_callback(rqstp, NLMPROC_LOCK_RES, argp, nlmsvc_proc_lock); } -static int nlmsvc_proc_cancel_msg(struct svc_rqst *rqstp, struct nlm_args *argp, +static __be32 nlmsvc_proc_cancel_msg(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { dprintk("lockd: CANCEL_MSG called\n"); return nlmsvc_callback(rqstp, NLMPROC_CANCEL_RES, argp, nlmsvc_proc_cancel); } -static int +static __be32 nlmsvc_proc_unlock_msg(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { @@ -338,7 +340,7 @@ nlmsvc_proc_unlock_msg(struct svc_rqst *rqstp, struct nlm_args *argp, return nlmsvc_callback(rqstp, NLMPROC_UNLOCK_RES, argp, nlmsvc_proc_unlock); } -static int +static __be32 nlmsvc_proc_granted_msg(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { @@ -349,7 +351,7 @@ nlmsvc_proc_granted_msg(struct svc_rqst *rqstp, struct nlm_args *argp, /* * SHARE: create a DOS share or alter existing share. */ -static int +static __be32 nlmsvc_proc_share(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -368,7 +370,7 @@ nlmsvc_proc_share(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlmsvc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; /* Now try to create the share */ resp->status = cast_status(nlmsvc_share_file(host, file, argp)); @@ -382,7 +384,7 @@ nlmsvc_proc_share(struct svc_rqst *rqstp, struct nlm_args *argp, /* * UNSHARE: Release a DOS share. */ -static int +static __be32 nlmsvc_proc_unshare(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -401,7 +403,7 @@ nlmsvc_proc_unshare(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain client and file */ if ((resp->status = nlmsvc_retrieve_args(rqstp, argp, &host, &file))) - return rpc_success; + return resp->status == nlm_drop_reply ? rpc_drop_reply :rpc_success; /* Now try to unshare the file */ resp->status = cast_status(nlmsvc_unshare_file(host, file, argp)); @@ -415,7 +417,7 @@ nlmsvc_proc_unshare(struct svc_rqst *rqstp, struct nlm_args *argp, /* * NM_LOCK: Create an unmonitored lock */ -static int +static __be32 nlmsvc_proc_nm_lock(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_res *resp) { @@ -428,7 +430,7 @@ nlmsvc_proc_nm_lock(struct svc_rqst *rqstp, struct nlm_args *argp, /* * FREE_ALL: Release all locks and shares held by client */ -static int +static __be32 nlmsvc_proc_free_all(struct svc_rqst *rqstp, struct nlm_args *argp, void *resp) { @@ -446,7 +448,7 @@ nlmsvc_proc_free_all(struct svc_rqst *rqstp, struct nlm_args *argp, /* * SM_NOTIFY: private callback from statd (not part of official NLM proto) */ -static int +static __be32 nlmsvc_proc_sm_notify(struct svc_rqst *rqstp, struct nlm_reboot *argp, void *resp) { @@ -475,7 +477,7 @@ nlmsvc_proc_sm_notify(struct svc_rqst *rqstp, struct nlm_reboot *argp, /* * client sent a GRANTED_RES, let's remove the associated block */ -static int +static __be32 nlmsvc_proc_granted_res(struct svc_rqst *rqstp, struct nlm_res *argp, void *resp) { diff --git a/fs/lockd/svcshare.c b/fs/lockd/svcshare.c index b9926ce8782e..6220dc2a3f2c 100644 --- a/fs/lockd/svcshare.c +++ b/fs/lockd/svcshare.c @@ -23,7 +23,7 @@ nlm_cmp_owner(struct nlm_share *share, struct xdr_netobj *oh) && !memcmp(share->s_owner.data, oh->data, oh->len); } -u32 +__be32 nlmsvc_share_file(struct nlm_host *host, struct nlm_file *file, struct nlm_args *argp) { @@ -64,7 +64,7 @@ update: /* * Delete a share. */ -u32 +__be32 nlmsvc_unshare_file(struct nlm_host *host, struct nlm_file *file, struct nlm_args *argp) { diff --git a/fs/lockd/svcsubs.c b/fs/lockd/svcsubs.c index 514f5f20701e..e83024e16042 100644 --- a/fs/lockd/svcsubs.c +++ b/fs/lockd/svcsubs.c @@ -78,14 +78,14 @@ static inline unsigned int file_hash(struct nfs_fh *f) * This is not quite right, but for now, we assume the client performs * the proper R/W checking. */ -u32 +__be32 nlm_lookup_file(struct svc_rqst *rqstp, struct nlm_file **result, struct nfs_fh *f) { struct hlist_node *pos; struct nlm_file *file; unsigned int hash; - u32 nfserr; + __be32 nfserr; nlm_debug_print_fh("nlm_file_lookup", f); @@ -135,12 +135,6 @@ out_unlock: out_free: kfree(file); -#ifdef CONFIG_LOCKD_V4 - if (nfserr == 1) - nfserr = nlm4_stale_fh; - else -#endif - nfserr = nlm_lck_denied; goto out_unlock; } @@ -324,7 +318,16 @@ nlmsvc_same_host(struct nlm_host *host, struct nlm_host *other) static int nlmsvc_is_client(struct nlm_host *host, struct nlm_host *dummy) { - return host->h_server; + if (host->h_server) { + /* we are destroying locks even though the client + * hasn't asked us too, so don't unmonitor the + * client + */ + if (host->h_nsmhandle) + host->h_nsmhandle->sm_sticky = 1; + return 1; + } else + return 0; } /* diff --git a/fs/lockd/xdr.c b/fs/lockd/xdr.c index 61c46facf257..b7c949256e5a 100644 --- a/fs/lockd/xdr.c +++ b/fs/lockd/xdr.c @@ -43,7 +43,7 @@ loff_t_to_s32(loff_t offset) /* * XDR functions for basic NLM types */ -static u32 *nlm_decode_cookie(u32 *p, struct nlm_cookie *c) +static __be32 *nlm_decode_cookie(__be32 *p, struct nlm_cookie *c) { unsigned int len; @@ -69,8 +69,8 @@ static u32 *nlm_decode_cookie(u32 *p, struct nlm_cookie *c) return p; } -static inline u32 * -nlm_encode_cookie(u32 *p, struct nlm_cookie *c) +static inline __be32 * +nlm_encode_cookie(__be32 *p, struct nlm_cookie *c) { *p++ = htonl(c->len); memcpy(p, c->data, c->len); @@ -78,8 +78,8 @@ nlm_encode_cookie(u32 *p, struct nlm_cookie *c) return p; } -static u32 * -nlm_decode_fh(u32 *p, struct nfs_fh *f) +static __be32 * +nlm_decode_fh(__be32 *p, struct nfs_fh *f) { unsigned int len; @@ -95,8 +95,8 @@ nlm_decode_fh(u32 *p, struct nfs_fh *f) return p + XDR_QUADLEN(NFS2_FHSIZE); } -static inline u32 * -nlm_encode_fh(u32 *p, struct nfs_fh *f) +static inline __be32 * +nlm_encode_fh(__be32 *p, struct nfs_fh *f) { *p++ = htonl(NFS2_FHSIZE); memcpy(p, f->data, NFS2_FHSIZE); @@ -106,20 +106,20 @@ nlm_encode_fh(u32 *p, struct nfs_fh *f) /* * Encode and decode owner handle */ -static inline u32 * -nlm_decode_oh(u32 *p, struct xdr_netobj *oh) +static inline __be32 * +nlm_decode_oh(__be32 *p, struct xdr_netobj *oh) { return xdr_decode_netobj(p, oh); } -static inline u32 * -nlm_encode_oh(u32 *p, struct xdr_netobj *oh) +static inline __be32 * +nlm_encode_oh(__be32 *p, struct xdr_netobj *oh) { return xdr_encode_netobj(p, oh); } -static u32 * -nlm_decode_lock(u32 *p, struct nlm_lock *lock) +static __be32 * +nlm_decode_lock(__be32 *p, struct nlm_lock *lock) { struct file_lock *fl = &lock->fl; s32 start, len, end; @@ -153,8 +153,8 @@ nlm_decode_lock(u32 *p, struct nlm_lock *lock) /* * Encode a lock as part of an NLM call */ -static u32 * -nlm_encode_lock(u32 *p, struct nlm_lock *lock) +static __be32 * +nlm_encode_lock(__be32 *p, struct nlm_lock *lock) { struct file_lock *fl = &lock->fl; __s32 start, len; @@ -184,8 +184,8 @@ nlm_encode_lock(u32 *p, struct nlm_lock *lock) /* * Encode result of a TEST/TEST_MSG call */ -static u32 * -nlm_encode_testres(u32 *p, struct nlm_res *resp) +static __be32 * +nlm_encode_testres(__be32 *p, struct nlm_res *resp) { s32 start, len; @@ -221,7 +221,7 @@ nlm_encode_testres(u32 *p, struct nlm_res *resp) * First, the server side XDR functions */ int -nlmsvc_decode_testargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) +nlmsvc_decode_testargs(struct svc_rqst *rqstp, __be32 *p, nlm_args *argp) { u32 exclusive; @@ -238,7 +238,7 @@ nlmsvc_decode_testargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) } int -nlmsvc_encode_testres(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) +nlmsvc_encode_testres(struct svc_rqst *rqstp, __be32 *p, struct nlm_res *resp) { if (!(p = nlm_encode_testres(p, resp))) return 0; @@ -246,7 +246,7 @@ nlmsvc_encode_testres(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) } int -nlmsvc_decode_lockargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) +nlmsvc_decode_lockargs(struct svc_rqst *rqstp, __be32 *p, nlm_args *argp) { u32 exclusive; @@ -266,7 +266,7 @@ nlmsvc_decode_lockargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) } int -nlmsvc_decode_cancargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) +nlmsvc_decode_cancargs(struct svc_rqst *rqstp, __be32 *p, nlm_args *argp) { u32 exclusive; @@ -282,7 +282,7 @@ nlmsvc_decode_cancargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) } int -nlmsvc_decode_unlockargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) +nlmsvc_decode_unlockargs(struct svc_rqst *rqstp, __be32 *p, nlm_args *argp) { if (!(p = nlm_decode_cookie(p, &argp->cookie)) || !(p = nlm_decode_lock(p, &argp->lock))) @@ -292,7 +292,7 @@ nlmsvc_decode_unlockargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) } int -nlmsvc_decode_shareargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) +nlmsvc_decode_shareargs(struct svc_rqst *rqstp, __be32 *p, nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -313,7 +313,7 @@ nlmsvc_decode_shareargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) } int -nlmsvc_encode_shareres(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) +nlmsvc_encode_shareres(struct svc_rqst *rqstp, __be32 *p, struct nlm_res *resp) { if (!(p = nlm_encode_cookie(p, &resp->cookie))) return 0; @@ -323,7 +323,7 @@ nlmsvc_encode_shareres(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) } int -nlmsvc_encode_res(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) +nlmsvc_encode_res(struct svc_rqst *rqstp, __be32 *p, struct nlm_res *resp) { if (!(p = nlm_encode_cookie(p, &resp->cookie))) return 0; @@ -332,7 +332,7 @@ nlmsvc_encode_res(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) } int -nlmsvc_decode_notify(struct svc_rqst *rqstp, u32 *p, struct nlm_args *argp) +nlmsvc_decode_notify(struct svc_rqst *rqstp, __be32 *p, struct nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -344,7 +344,7 @@ nlmsvc_decode_notify(struct svc_rqst *rqstp, u32 *p, struct nlm_args *argp) } int -nlmsvc_decode_reboot(struct svc_rqst *rqstp, u32 *p, struct nlm_reboot *argp) +nlmsvc_decode_reboot(struct svc_rqst *rqstp, __be32 *p, struct nlm_reboot *argp) { if (!(p = xdr_decode_string_inplace(p, &argp->mon, &argp->len, SM_MAXSTRLEN))) return 0; @@ -357,7 +357,7 @@ nlmsvc_decode_reboot(struct svc_rqst *rqstp, u32 *p, struct nlm_reboot *argp) } int -nlmsvc_decode_res(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) +nlmsvc_decode_res(struct svc_rqst *rqstp, __be32 *p, struct nlm_res *resp) { if (!(p = nlm_decode_cookie(p, &resp->cookie))) return 0; @@ -366,13 +366,13 @@ nlmsvc_decode_res(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) } int -nlmsvc_decode_void(struct svc_rqst *rqstp, u32 *p, void *dummy) +nlmsvc_decode_void(struct svc_rqst *rqstp, __be32 *p, void *dummy) { return xdr_argsize_check(rqstp, p); } int -nlmsvc_encode_void(struct svc_rqst *rqstp, u32 *p, void *dummy) +nlmsvc_encode_void(struct svc_rqst *rqstp, __be32 *p, void *dummy) { return xdr_ressize_check(rqstp, p); } @@ -389,7 +389,7 @@ nlmclt_decode_void(struct rpc_rqst *req, u32 *p, void *ptr) #endif static int -nlmclt_encode_testargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) +nlmclt_encode_testargs(struct rpc_rqst *req, __be32 *p, nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -403,7 +403,7 @@ nlmclt_encode_testargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) } static int -nlmclt_decode_testres(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) +nlmclt_decode_testres(struct rpc_rqst *req, __be32 *p, struct nlm_res *resp) { if (!(p = nlm_decode_cookie(p, &resp->cookie))) return -EIO; @@ -438,7 +438,7 @@ nlmclt_decode_testres(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) static int -nlmclt_encode_lockargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) +nlmclt_encode_lockargs(struct rpc_rqst *req, __be32 *p, nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -455,7 +455,7 @@ nlmclt_encode_lockargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) } static int -nlmclt_encode_cancargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) +nlmclt_encode_cancargs(struct rpc_rqst *req, __be32 *p, nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -470,7 +470,7 @@ nlmclt_encode_cancargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) } static int -nlmclt_encode_unlockargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) +nlmclt_encode_unlockargs(struct rpc_rqst *req, __be32 *p, nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -483,7 +483,7 @@ nlmclt_encode_unlockargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) } static int -nlmclt_encode_res(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) +nlmclt_encode_res(struct rpc_rqst *req, __be32 *p, struct nlm_res *resp) { if (!(p = nlm_encode_cookie(p, &resp->cookie))) return -EIO; @@ -493,7 +493,7 @@ nlmclt_encode_res(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) } static int -nlmclt_encode_testres(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) +nlmclt_encode_testres(struct rpc_rqst *req, __be32 *p, struct nlm_res *resp) { if (!(p = nlm_encode_testres(p, resp))) return -EIO; @@ -502,7 +502,7 @@ nlmclt_encode_testres(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) } static int -nlmclt_decode_res(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) +nlmclt_decode_res(struct rpc_rqst *req, __be32 *p, struct nlm_res *resp) { if (!(p = nlm_decode_cookie(p, &resp->cookie))) return -EIO; diff --git a/fs/lockd/xdr4.c b/fs/lockd/xdr4.c index 36eb175ec335..f4c0b2b9f75a 100644 --- a/fs/lockd/xdr4.c +++ b/fs/lockd/xdr4.c @@ -44,8 +44,8 @@ loff_t_to_s64(loff_t offset) /* * XDR functions for basic NLM types */ -static u32 * -nlm4_decode_cookie(u32 *p, struct nlm_cookie *c) +static __be32 * +nlm4_decode_cookie(__be32 *p, struct nlm_cookie *c) { unsigned int len; @@ -71,8 +71,8 @@ nlm4_decode_cookie(u32 *p, struct nlm_cookie *c) return p; } -static u32 * -nlm4_encode_cookie(u32 *p, struct nlm_cookie *c) +static __be32 * +nlm4_encode_cookie(__be32 *p, struct nlm_cookie *c) { *p++ = htonl(c->len); memcpy(p, c->data, c->len); @@ -80,8 +80,8 @@ nlm4_encode_cookie(u32 *p, struct nlm_cookie *c) return p; } -static u32 * -nlm4_decode_fh(u32 *p, struct nfs_fh *f) +static __be32 * +nlm4_decode_fh(__be32 *p, struct nfs_fh *f) { memset(f->data, 0, sizeof(f->data)); f->size = ntohl(*p++); @@ -95,8 +95,8 @@ nlm4_decode_fh(u32 *p, struct nfs_fh *f) return p + XDR_QUADLEN(f->size); } -static u32 * -nlm4_encode_fh(u32 *p, struct nfs_fh *f) +static __be32 * +nlm4_encode_fh(__be32 *p, struct nfs_fh *f) { *p++ = htonl(f->size); if (f->size) p[XDR_QUADLEN(f->size)-1] = 0; /* don't leak anything */ @@ -107,20 +107,20 @@ nlm4_encode_fh(u32 *p, struct nfs_fh *f) /* * Encode and decode owner handle */ -static u32 * -nlm4_decode_oh(u32 *p, struct xdr_netobj *oh) +static __be32 * +nlm4_decode_oh(__be32 *p, struct xdr_netobj *oh) { return xdr_decode_netobj(p, oh); } -static u32 * -nlm4_encode_oh(u32 *p, struct xdr_netobj *oh) +static __be32 * +nlm4_encode_oh(__be32 *p, struct xdr_netobj *oh) { return xdr_encode_netobj(p, oh); } -static u32 * -nlm4_decode_lock(u32 *p, struct nlm_lock *lock) +static __be32 * +nlm4_decode_lock(__be32 *p, struct nlm_lock *lock) { struct file_lock *fl = &lock->fl; __s64 len, start, end; @@ -153,8 +153,8 @@ nlm4_decode_lock(u32 *p, struct nlm_lock *lock) /* * Encode a lock as part of an NLM call */ -static u32 * -nlm4_encode_lock(u32 *p, struct nlm_lock *lock) +static __be32 * +nlm4_encode_lock(__be32 *p, struct nlm_lock *lock) { struct file_lock *fl = &lock->fl; __s64 start, len; @@ -185,8 +185,8 @@ nlm4_encode_lock(u32 *p, struct nlm_lock *lock) /* * Encode result of a TEST/TEST_MSG call */ -static u32 * -nlm4_encode_testres(u32 *p, struct nlm_res *resp) +static __be32 * +nlm4_encode_testres(__be32 *p, struct nlm_res *resp) { s64 start, len; @@ -227,7 +227,7 @@ nlm4_encode_testres(u32 *p, struct nlm_res *resp) * First, the server side XDR functions */ int -nlm4svc_decode_testargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) +nlm4svc_decode_testargs(struct svc_rqst *rqstp, __be32 *p, nlm_args *argp) { u32 exclusive; @@ -244,7 +244,7 @@ nlm4svc_decode_testargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) } int -nlm4svc_encode_testres(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) +nlm4svc_encode_testres(struct svc_rqst *rqstp, __be32 *p, struct nlm_res *resp) { if (!(p = nlm4_encode_testres(p, resp))) return 0; @@ -252,7 +252,7 @@ nlm4svc_encode_testres(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) } int -nlm4svc_decode_lockargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) +nlm4svc_decode_lockargs(struct svc_rqst *rqstp, __be32 *p, nlm_args *argp) { u32 exclusive; @@ -272,7 +272,7 @@ nlm4svc_decode_lockargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) } int -nlm4svc_decode_cancargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) +nlm4svc_decode_cancargs(struct svc_rqst *rqstp, __be32 *p, nlm_args *argp) { u32 exclusive; @@ -288,7 +288,7 @@ nlm4svc_decode_cancargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) } int -nlm4svc_decode_unlockargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) +nlm4svc_decode_unlockargs(struct svc_rqst *rqstp, __be32 *p, nlm_args *argp) { if (!(p = nlm4_decode_cookie(p, &argp->cookie)) || !(p = nlm4_decode_lock(p, &argp->lock))) @@ -298,7 +298,7 @@ nlm4svc_decode_unlockargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) } int -nlm4svc_decode_shareargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) +nlm4svc_decode_shareargs(struct svc_rqst *rqstp, __be32 *p, nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -319,7 +319,7 @@ nlm4svc_decode_shareargs(struct svc_rqst *rqstp, u32 *p, nlm_args *argp) } int -nlm4svc_encode_shareres(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) +nlm4svc_encode_shareres(struct svc_rqst *rqstp, __be32 *p, struct nlm_res *resp) { if (!(p = nlm4_encode_cookie(p, &resp->cookie))) return 0; @@ -329,7 +329,7 @@ nlm4svc_encode_shareres(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) } int -nlm4svc_encode_res(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) +nlm4svc_encode_res(struct svc_rqst *rqstp, __be32 *p, struct nlm_res *resp) { if (!(p = nlm4_encode_cookie(p, &resp->cookie))) return 0; @@ -338,7 +338,7 @@ nlm4svc_encode_res(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) } int -nlm4svc_decode_notify(struct svc_rqst *rqstp, u32 *p, struct nlm_args *argp) +nlm4svc_decode_notify(struct svc_rqst *rqstp, __be32 *p, struct nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -350,7 +350,7 @@ nlm4svc_decode_notify(struct svc_rqst *rqstp, u32 *p, struct nlm_args *argp) } int -nlm4svc_decode_reboot(struct svc_rqst *rqstp, u32 *p, struct nlm_reboot *argp) +nlm4svc_decode_reboot(struct svc_rqst *rqstp, __be32 *p, struct nlm_reboot *argp) { if (!(p = xdr_decode_string_inplace(p, &argp->mon, &argp->len, SM_MAXSTRLEN))) return 0; @@ -363,7 +363,7 @@ nlm4svc_decode_reboot(struct svc_rqst *rqstp, u32 *p, struct nlm_reboot *argp) } int -nlm4svc_decode_res(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) +nlm4svc_decode_res(struct svc_rqst *rqstp, __be32 *p, struct nlm_res *resp) { if (!(p = nlm4_decode_cookie(p, &resp->cookie))) return 0; @@ -372,13 +372,13 @@ nlm4svc_decode_res(struct svc_rqst *rqstp, u32 *p, struct nlm_res *resp) } int -nlm4svc_decode_void(struct svc_rqst *rqstp, u32 *p, void *dummy) +nlm4svc_decode_void(struct svc_rqst *rqstp, __be32 *p, void *dummy) { return xdr_argsize_check(rqstp, p); } int -nlm4svc_encode_void(struct svc_rqst *rqstp, u32 *p, void *dummy) +nlm4svc_encode_void(struct svc_rqst *rqstp, __be32 *p, void *dummy) { return xdr_ressize_check(rqstp, p); } @@ -388,14 +388,14 @@ nlm4svc_encode_void(struct svc_rqst *rqstp, u32 *p, void *dummy) */ #ifdef NLMCLNT_SUPPORT_SHARES static int -nlm4clt_decode_void(struct rpc_rqst *req, u32 *p, void *ptr) +nlm4clt_decode_void(struct rpc_rqst *req, __be32 *p, void *ptr) { return 0; } #endif static int -nlm4clt_encode_testargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) +nlm4clt_encode_testargs(struct rpc_rqst *req, __be32 *p, nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -409,7 +409,7 @@ nlm4clt_encode_testargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) } static int -nlm4clt_decode_testres(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) +nlm4clt_decode_testres(struct rpc_rqst *req, __be32 *p, struct nlm_res *resp) { if (!(p = nlm4_decode_cookie(p, &resp->cookie))) return -EIO; @@ -444,7 +444,7 @@ nlm4clt_decode_testres(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) static int -nlm4clt_encode_lockargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) +nlm4clt_encode_lockargs(struct rpc_rqst *req, __be32 *p, nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -461,7 +461,7 @@ nlm4clt_encode_lockargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) } static int -nlm4clt_encode_cancargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) +nlm4clt_encode_cancargs(struct rpc_rqst *req, __be32 *p, nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -476,7 +476,7 @@ nlm4clt_encode_cancargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) } static int -nlm4clt_encode_unlockargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) +nlm4clt_encode_unlockargs(struct rpc_rqst *req, __be32 *p, nlm_args *argp) { struct nlm_lock *lock = &argp->lock; @@ -489,7 +489,7 @@ nlm4clt_encode_unlockargs(struct rpc_rqst *req, u32 *p, nlm_args *argp) } static int -nlm4clt_encode_res(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) +nlm4clt_encode_res(struct rpc_rqst *req, __be32 *p, struct nlm_res *resp) { if (!(p = nlm4_encode_cookie(p, &resp->cookie))) return -EIO; @@ -499,7 +499,7 @@ nlm4clt_encode_res(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) } static int -nlm4clt_encode_testres(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) +nlm4clt_encode_testres(struct rpc_rqst *req, __be32 *p, struct nlm_res *resp) { if (!(p = nlm4_encode_testres(p, resp))) return -EIO; @@ -508,7 +508,7 @@ nlm4clt_encode_testres(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) } static int -nlm4clt_decode_res(struct rpc_rqst *req, u32 *p, struct nlm_res *resp) +nlm4clt_decode_res(struct rpc_rqst *req, __be32 *p, struct nlm_res *resp) { if (!(p = nlm4_decode_cookie(p, &resp->cookie))) return -EIO; diff --git a/fs/minix/inode.c b/fs/minix/inode.c index c11a4b9fb863..1e36bae4d0eb 100644 --- a/fs/minix/inode.c +++ b/fs/minix/inode.c @@ -149,12 +149,8 @@ static int minix_fill_super(struct super_block *s, void *data, int silent) return -ENOMEM; s->s_fs_info = sbi; - /* N.B. These should be compile-time tests. - Unfortunately that is impossible. */ - if (32 != sizeof (struct minix_inode)) - panic("bad V1 i-node size"); - if (64 != sizeof(struct minix2_inode)) - panic("bad V2 i-node size"); + BUILD_BUG_ON(32 != sizeof (struct minix_inode)); + BUILD_BUG_ON(64 != sizeof(struct minix2_inode)); if (!sb_set_blocksize(s, BLOCK_SIZE)) goto out_bad_hblock; diff --git a/fs/nfs/callback.h b/fs/nfs/callback.h index 5676163d26e8..db3d7919c601 100644 --- a/fs/nfs/callback.h +++ b/fs/nfs/callback.h @@ -31,10 +31,10 @@ struct cb_compound_hdr_arg { }; struct cb_compound_hdr_res { - uint32_t *status; + __be32 *status; int taglen; const char *tag; - uint32_t *nops; + __be32 *nops; }; struct cb_getattrargs { @@ -44,7 +44,7 @@ struct cb_getattrargs { }; struct cb_getattrres { - uint32_t status; + __be32 status; uint32_t bitmap[2]; uint64_t size; uint64_t change_attr; @@ -59,8 +59,8 @@ struct cb_recallargs { uint32_t truncate; }; -extern unsigned nfs4_callback_getattr(struct cb_getattrargs *args, struct cb_getattrres *res); -extern unsigned nfs4_callback_recall(struct cb_recallargs *args, void *dummy); +extern __be32 nfs4_callback_getattr(struct cb_getattrargs *args, struct cb_getattrres *res); +extern __be32 nfs4_callback_recall(struct cb_recallargs *args, void *dummy); #ifdef CONFIG_NFS_V4 extern int nfs_callback_up(void); diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c index 97cf8f71451f..72e55d83756d 100644 --- a/fs/nfs/callback_proc.c +++ b/fs/nfs/callback_proc.c @@ -14,7 +14,7 @@ #define NFSDBG_FACILITY NFSDBG_CALLBACK -unsigned nfs4_callback_getattr(struct cb_getattrargs *args, struct cb_getattrres *res) +__be32 nfs4_callback_getattr(struct cb_getattrargs *args, struct cb_getattrres *res) { struct nfs_client *clp; struct nfs_delegation *delegation; @@ -55,11 +55,11 @@ out: return res->status; } -unsigned nfs4_callback_recall(struct cb_recallargs *args, void *dummy) +__be32 nfs4_callback_recall(struct cb_recallargs *args, void *dummy) { struct nfs_client *clp; struct inode *inode; - unsigned res; + __be32 res; res = htonl(NFS4ERR_BADHANDLE); clp = nfs_find_client(args->addr, 4); diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c index 29f932192054..f8ea1f51f590 100644 --- a/fs/nfs/callback_xdr.c +++ b/fs/nfs/callback_xdr.c @@ -22,9 +22,9 @@ #define NFSDBG_FACILITY NFSDBG_CALLBACK -typedef unsigned (*callback_process_op_t)(void *, void *); -typedef unsigned (*callback_decode_arg_t)(struct svc_rqst *, struct xdr_stream *, void *); -typedef unsigned (*callback_encode_res_t)(struct svc_rqst *, struct xdr_stream *, void *); +typedef __be32 (*callback_process_op_t)(void *, void *); +typedef __be32 (*callback_decode_arg_t)(struct svc_rqst *, struct xdr_stream *, void *); +typedef __be32 (*callback_encode_res_t)(struct svc_rqst *, struct xdr_stream *, void *); struct callback_op { @@ -36,24 +36,24 @@ struct callback_op { static struct callback_op callback_ops[]; -static int nfs4_callback_null(struct svc_rqst *rqstp, void *argp, void *resp) +static __be32 nfs4_callback_null(struct svc_rqst *rqstp, void *argp, void *resp) { return htonl(NFS4_OK); } -static int nfs4_decode_void(struct svc_rqst *rqstp, uint32_t *p, void *dummy) +static int nfs4_decode_void(struct svc_rqst *rqstp, __be32 *p, void *dummy) { return xdr_argsize_check(rqstp, p); } -static int nfs4_encode_void(struct svc_rqst *rqstp, uint32_t *p, void *dummy) +static int nfs4_encode_void(struct svc_rqst *rqstp, __be32 *p, void *dummy) { return xdr_ressize_check(rqstp, p); } -static uint32_t *read_buf(struct xdr_stream *xdr, int nbytes) +static __be32 *read_buf(struct xdr_stream *xdr, int nbytes) { - uint32_t *p; + __be32 *p; p = xdr_inline_decode(xdr, nbytes); if (unlikely(p == NULL)) @@ -61,9 +61,9 @@ static uint32_t *read_buf(struct xdr_stream *xdr, int nbytes) return p; } -static unsigned decode_string(struct xdr_stream *xdr, unsigned int *len, const char **str) +static __be32 decode_string(struct xdr_stream *xdr, unsigned int *len, const char **str) { - uint32_t *p; + __be32 *p; p = read_buf(xdr, 4); if (unlikely(p == NULL)) @@ -81,9 +81,9 @@ static unsigned decode_string(struct xdr_stream *xdr, unsigned int *len, const c return 0; } -static unsigned decode_fh(struct xdr_stream *xdr, struct nfs_fh *fh) +static __be32 decode_fh(struct xdr_stream *xdr, struct nfs_fh *fh) { - uint32_t *p; + __be32 *p; p = read_buf(xdr, 4); if (unlikely(p == NULL)) @@ -99,9 +99,9 @@ static unsigned decode_fh(struct xdr_stream *xdr, struct nfs_fh *fh) return 0; } -static unsigned decode_bitmap(struct xdr_stream *xdr, uint32_t *bitmap) +static __be32 decode_bitmap(struct xdr_stream *xdr, uint32_t *bitmap) { - uint32_t *p; + __be32 *p; unsigned int attrlen; p = read_buf(xdr, 4); @@ -118,9 +118,9 @@ static unsigned decode_bitmap(struct xdr_stream *xdr, uint32_t *bitmap) return 0; } -static unsigned decode_stateid(struct xdr_stream *xdr, nfs4_stateid *stateid) +static __be32 decode_stateid(struct xdr_stream *xdr, nfs4_stateid *stateid) { - uint32_t *p; + __be32 *p; p = read_buf(xdr, 16); if (unlikely(p == NULL)) @@ -129,11 +129,11 @@ static unsigned decode_stateid(struct xdr_stream *xdr, nfs4_stateid *stateid) return 0; } -static unsigned decode_compound_hdr_arg(struct xdr_stream *xdr, struct cb_compound_hdr_arg *hdr) +static __be32 decode_compound_hdr_arg(struct xdr_stream *xdr, struct cb_compound_hdr_arg *hdr) { - uint32_t *p; + __be32 *p; unsigned int minor_version; - unsigned status; + __be32 status; status = decode_string(xdr, &hdr->taglen, &hdr->tag); if (unlikely(status != 0)) @@ -159,9 +159,9 @@ static unsigned decode_compound_hdr_arg(struct xdr_stream *xdr, struct cb_compou return 0; } -static unsigned decode_op_hdr(struct xdr_stream *xdr, unsigned int *op) +static __be32 decode_op_hdr(struct xdr_stream *xdr, unsigned int *op) { - uint32_t *p; + __be32 *p; p = read_buf(xdr, 4); if (unlikely(p == NULL)) return htonl(NFS4ERR_RESOURCE); @@ -169,9 +169,9 @@ static unsigned decode_op_hdr(struct xdr_stream *xdr, unsigned int *op) return 0; } -static unsigned decode_getattr_args(struct svc_rqst *rqstp, struct xdr_stream *xdr, struct cb_getattrargs *args) +static __be32 decode_getattr_args(struct svc_rqst *rqstp, struct xdr_stream *xdr, struct cb_getattrargs *args) { - unsigned status; + __be32 status; status = decode_fh(xdr, &args->fh); if (unlikely(status != 0)) @@ -183,10 +183,10 @@ out: return status; } -static unsigned decode_recall_args(struct svc_rqst *rqstp, struct xdr_stream *xdr, struct cb_recallargs *args) +static __be32 decode_recall_args(struct svc_rqst *rqstp, struct xdr_stream *xdr, struct cb_recallargs *args) { - uint32_t *p; - unsigned status; + __be32 *p; + __be32 status; args->addr = &rqstp->rq_addr; status = decode_stateid(xdr, &args->stateid); @@ -204,9 +204,9 @@ out: return status; } -static unsigned encode_string(struct xdr_stream *xdr, unsigned int len, const char *str) +static __be32 encode_string(struct xdr_stream *xdr, unsigned int len, const char *str) { - uint32_t *p; + __be32 *p; p = xdr_reserve_space(xdr, 4 + len); if (unlikely(p == NULL)) @@ -217,10 +217,10 @@ static unsigned encode_string(struct xdr_stream *xdr, unsigned int len, const ch #define CB_SUPPORTED_ATTR0 (FATTR4_WORD0_CHANGE|FATTR4_WORD0_SIZE) #define CB_SUPPORTED_ATTR1 (FATTR4_WORD1_TIME_METADATA|FATTR4_WORD1_TIME_MODIFY) -static unsigned encode_attr_bitmap(struct xdr_stream *xdr, const uint32_t *bitmap, uint32_t **savep) +static __be32 encode_attr_bitmap(struct xdr_stream *xdr, const uint32_t *bitmap, __be32 **savep) { - uint32_t bm[2]; - uint32_t *p; + __be32 bm[2]; + __be32 *p; bm[0] = htonl(bitmap[0] & CB_SUPPORTED_ATTR0); bm[1] = htonl(bitmap[1] & CB_SUPPORTED_ATTR1); @@ -247,9 +247,9 @@ static unsigned encode_attr_bitmap(struct xdr_stream *xdr, const uint32_t *bitma return 0; } -static unsigned encode_attr_change(struct xdr_stream *xdr, const uint32_t *bitmap, uint64_t change) +static __be32 encode_attr_change(struct xdr_stream *xdr, const uint32_t *bitmap, uint64_t change) { - uint32_t *p; + __be32 *p; if (!(bitmap[0] & FATTR4_WORD0_CHANGE)) return 0; @@ -260,9 +260,9 @@ static unsigned encode_attr_change(struct xdr_stream *xdr, const uint32_t *bitma return 0; } -static unsigned encode_attr_size(struct xdr_stream *xdr, const uint32_t *bitmap, uint64_t size) +static __be32 encode_attr_size(struct xdr_stream *xdr, const uint32_t *bitmap, uint64_t size) { - uint32_t *p; + __be32 *p; if (!(bitmap[0] & FATTR4_WORD0_SIZE)) return 0; @@ -273,9 +273,9 @@ static unsigned encode_attr_size(struct xdr_stream *xdr, const uint32_t *bitmap, return 0; } -static unsigned encode_attr_time(struct xdr_stream *xdr, const struct timespec *time) +static __be32 encode_attr_time(struct xdr_stream *xdr, const struct timespec *time) { - uint32_t *p; + __be32 *p; p = xdr_reserve_space(xdr, 12); if (unlikely(p == 0)) @@ -285,23 +285,23 @@ static unsigned encode_attr_time(struct xdr_stream *xdr, const struct timespec * return 0; } -static unsigned encode_attr_ctime(struct xdr_stream *xdr, const uint32_t *bitmap, const struct timespec *time) +static __be32 encode_attr_ctime(struct xdr_stream *xdr, const uint32_t *bitmap, const struct timespec *time) { if (!(bitmap[1] & FATTR4_WORD1_TIME_METADATA)) return 0; return encode_attr_time(xdr,time); } -static unsigned encode_attr_mtime(struct xdr_stream *xdr, const uint32_t *bitmap, const struct timespec *time) +static __be32 encode_attr_mtime(struct xdr_stream *xdr, const uint32_t *bitmap, const struct timespec *time) { if (!(bitmap[1] & FATTR4_WORD1_TIME_MODIFY)) return 0; return encode_attr_time(xdr,time); } -static unsigned encode_compound_hdr_res(struct xdr_stream *xdr, struct cb_compound_hdr_res *hdr) +static __be32 encode_compound_hdr_res(struct xdr_stream *xdr, struct cb_compound_hdr_res *hdr) { - unsigned status; + __be32 status; hdr->status = xdr_reserve_space(xdr, 4); if (unlikely(hdr->status == NULL)) @@ -315,9 +315,9 @@ static unsigned encode_compound_hdr_res(struct xdr_stream *xdr, struct cb_compou return 0; } -static unsigned encode_op_hdr(struct xdr_stream *xdr, uint32_t op, uint32_t res) +static __be32 encode_op_hdr(struct xdr_stream *xdr, uint32_t op, __be32 res) { - uint32_t *p; + __be32 *p; p = xdr_reserve_space(xdr, 8); if (unlikely(p == NULL)) @@ -327,10 +327,10 @@ static unsigned encode_op_hdr(struct xdr_stream *xdr, uint32_t op, uint32_t res) return 0; } -static unsigned encode_getattr_res(struct svc_rqst *rqstp, struct xdr_stream *xdr, const struct cb_getattrres *res) +static __be32 encode_getattr_res(struct svc_rqst *rqstp, struct xdr_stream *xdr, const struct cb_getattrres *res) { - uint32_t *savep = NULL; - unsigned status = res->status; + __be32 *savep = NULL; + __be32 status = res->status; if (unlikely(status != 0)) goto out; @@ -353,15 +353,15 @@ out: return status; } -static unsigned process_op(struct svc_rqst *rqstp, +static __be32 process_op(struct svc_rqst *rqstp, struct xdr_stream *xdr_in, void *argp, struct xdr_stream *xdr_out, void *resp) { struct callback_op *op = &callback_ops[0]; unsigned int op_nr = OP_CB_ILLEGAL; - unsigned int status = 0; + __be32 status = 0; long maxlen; - unsigned res; + __be32 res; dprintk("%s: start\n", __FUNCTION__); status = decode_op_hdr(xdr_in, &op_nr); @@ -399,20 +399,20 @@ static unsigned process_op(struct svc_rqst *rqstp, /* * Decode, process and encode a COMPOUND */ -static int nfs4_callback_compound(struct svc_rqst *rqstp, void *argp, void *resp) +static __be32 nfs4_callback_compound(struct svc_rqst *rqstp, void *argp, void *resp) { struct cb_compound_hdr_arg hdr_arg; struct cb_compound_hdr_res hdr_res; struct xdr_stream xdr_in, xdr_out; - uint32_t *p; - unsigned int status; + __be32 *p; + __be32 status; unsigned int nops = 1; dprintk("%s: start\n", __FUNCTION__); xdr_init_decode(&xdr_in, &rqstp->rq_arg, rqstp->rq_arg.head[0].iov_base); - p = (uint32_t*)((char *)rqstp->rq_res.head[0].iov_base + rqstp->rq_res.head[0].iov_len); + p = (__be32*)((char *)rqstp->rq_res.head[0].iov_base + rqstp->rq_res.head[0].iov_len); xdr_init_encode(&xdr_out, &rqstp->rq_res, p); decode_compound_hdr_arg(&xdr_in, &hdr_arg); diff --git a/fs/nfs/client.c b/fs/nfs/client.c index 34c3996bd0f5..5fea638743e4 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -232,11 +232,15 @@ void nfs_put_client(struct nfs_client *clp) * Find a client by address * - caller must hold nfs_client_lock */ -static struct nfs_client *__nfs_find_client(const struct sockaddr_in *addr, int nfsversion) +static struct nfs_client *__nfs_find_client(const struct sockaddr_in *addr, int nfsversion, int match_port) { struct nfs_client *clp; list_for_each_entry(clp, &nfs_client_list, cl_share_link) { + /* Don't match clients that failed to initialise properly */ + if (clp->cl_cons_state < 0) + continue; + /* Different NFS versions cannot share the same nfs_client */ if (clp->cl_nfsversion != nfsversion) continue; @@ -245,7 +249,7 @@ static struct nfs_client *__nfs_find_client(const struct sockaddr_in *addr, int sizeof(clp->cl_addr.sin_addr)) != 0) continue; - if (clp->cl_addr.sin_port == addr->sin_port) + if (!match_port || clp->cl_addr.sin_port == addr->sin_port) goto found; } @@ -265,11 +269,12 @@ struct nfs_client *nfs_find_client(const struct sockaddr_in *addr, int nfsversio struct nfs_client *clp; spin_lock(&nfs_client_lock); - clp = __nfs_find_client(addr, nfsversion); + clp = __nfs_find_client(addr, nfsversion, 0); spin_unlock(&nfs_client_lock); - - BUG_ON(clp && clp->cl_cons_state == 0); - + if (clp != NULL && clp->cl_cons_state != NFS_CS_READY) { + nfs_put_client(clp); + clp = NULL; + } return clp; } @@ -292,7 +297,7 @@ static struct nfs_client *nfs_get_client(const char *hostname, do { spin_lock(&nfs_client_lock); - clp = __nfs_find_client(addr, nfsversion); + clp = __nfs_find_client(addr, nfsversion, 1); if (clp) goto found_client; if (new) @@ -849,6 +854,7 @@ error: */ static int nfs4_init_client(struct nfs_client *clp, int proto, int timeo, int retrans, + const char *ip_addr, rpc_authflavor_t authflavour) { int error; @@ -865,6 +871,7 @@ static int nfs4_init_client(struct nfs_client *clp, error = nfs_create_rpc_client(clp, proto, timeo, retrans, authflavour); if (error < 0) goto error; + memcpy(clp->cl_ipaddr, ip_addr, sizeof(clp->cl_ipaddr)); error = nfs_idmap_new(clp); if (error < 0) { @@ -888,6 +895,7 @@ error: */ static int nfs4_set_client(struct nfs_server *server, const char *hostname, const struct sockaddr_in *addr, + const char *ip_addr, rpc_authflavor_t authflavour, int proto, int timeo, int retrans) { @@ -902,7 +910,7 @@ static int nfs4_set_client(struct nfs_server *server, error = PTR_ERR(clp); goto error; } - error = nfs4_init_client(clp, proto, timeo, retrans, authflavour); + error = nfs4_init_client(clp, proto, timeo, retrans, ip_addr, authflavour); if (error < 0) goto error_put; @@ -971,7 +979,7 @@ struct nfs_server *nfs4_create_server(const struct nfs4_mount_data *data, return ERR_PTR(-ENOMEM); /* Get a client record */ - error = nfs4_set_client(server, hostname, addr, authflavour, + error = nfs4_set_client(server, hostname, addr, ip_addr, authflavour, data->proto, data->timeo, data->retrans); if (error < 0) goto error; @@ -1041,6 +1049,7 @@ struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *data, /* Get a client representation. * Note: NFSv4 always uses TCP, */ error = nfs4_set_client(server, data->hostname, data->addr, + parent_client->cl_ipaddr, data->authflavor, parent_server->client->cl_xprt->prot, parent_client->retrans_timeo, diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 481f8892a919..b34cd16f472f 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -142,12 +142,12 @@ nfs_opendir(struct inode *inode, struct file *filp) return res; } -typedef u32 * (*decode_dirent_t)(u32 *, struct nfs_entry *, int); +typedef __be32 * (*decode_dirent_t)(__be32 *, struct nfs_entry *, int); typedef struct { struct file *file; struct page *page; unsigned long page_index; - u32 *ptr; + __be32 *ptr; u64 *dir_cookie; loff_t current_index; struct nfs_entry *entry; @@ -203,8 +203,10 @@ int nfs_readdir_filler(nfs_readdir_descriptor_t *desc, struct page *page) * Note: assumes we have exclusive access to this mapping either * through inode->i_mutex or some other mechanism. */ - if (page->index == 0) - invalidate_inode_pages2_range(inode->i_mapping, PAGE_CACHE_SIZE, -1); + if (page->index == 0 && invalidate_inode_pages2_range(inode->i_mapping, PAGE_CACHE_SIZE, -1) < 0) { + /* Should never happen */ + nfs_zap_mapping(inode, inode->i_mapping); + } unlock_page(page); return 0; error: @@ -218,7 +220,7 @@ int nfs_readdir_filler(nfs_readdir_descriptor_t *desc, struct page *page) static inline int dir_decode(nfs_readdir_descriptor_t *desc) { - u32 *p = desc->ptr; + __be32 *p = desc->ptr; p = desc->decode(p, desc->entry, desc->plus); if (IS_ERR(p)) return PTR_ERR(p); @@ -933,8 +935,17 @@ static struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, stru no_entry: res = d_materialise_unique(dentry, inode); - if (res != NULL) + if (res != NULL) { + struct dentry *parent; + if (IS_ERR(res)) + goto out_unlock; + /* Was a directory renamed! */ + parent = dget_parent(res); + if (!IS_ROOT(parent)) + nfs_mark_for_revalidate(parent->d_inode); + dput(parent); dentry = res; + } nfs_renew_times(dentry); nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); out_unlock: @@ -1130,6 +1141,8 @@ static struct dentry *nfs_readdir_lookup(nfs_readdir_descriptor_t *desc) alias = d_materialise_unique(dentry, inode); if (alias != NULL) { dput(dentry); + if (IS_ERR(alias)) + return NULL; dentry = alias; } @@ -1517,8 +1530,8 @@ static int nfs_symlink(struct inode *dir, struct dentry *dentry, const char *sym pagevec_init(&lru_pvec, 0); if (!add_to_page_cache(page, dentry->d_inode->i_mapping, 0, GFP_KERNEL)) { - if (!pagevec_add(&lru_pvec, page)) - __pagevec_lru_add(&lru_pvec); + pagevec_add(&lru_pvec, page); + pagevec_lru_add(&lru_pvec); SetPageUptodate(page); unlock_page(page); } else diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index 9f7f8b9ea1e2..bdfabf854a51 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -497,6 +497,7 @@ static void nfs_direct_write_complete(struct nfs_direct_req *dreq, struct inode if (dreq->commit_data != NULL) nfs_commit_free(dreq->commit_data); nfs_direct_free_writedata(dreq); + nfs_zap_mapping(inode, inode->i_mapping); nfs_direct_complete(dreq); } } @@ -517,6 +518,7 @@ static void nfs_direct_write_complete(struct nfs_direct_req *dreq, struct inode { nfs_end_data_update(inode); nfs_direct_free_writedata(dreq); + nfs_zap_mapping(inode, inode->i_mapping); nfs_direct_complete(dreq); } #endif @@ -532,10 +534,12 @@ static void nfs_direct_write_result(struct rpc_task *task, void *calldata) spin_lock(&dreq->lock); - if (likely(status >= 0)) - dreq->count += data->res.count; - else - dreq->error = task->tk_status; + if (unlikely(status < 0)) { + dreq->error = status; + goto out_unlock; + } + + dreq->count += data->res.count; if (data->res.verf->committed != NFS_FILE_SYNC) { switch (dreq->flags) { @@ -550,7 +554,7 @@ static void nfs_direct_write_result(struct rpc_task *task, void *calldata) } } } - +out_unlock: spin_unlock(&dreq->lock); } @@ -828,17 +832,6 @@ ssize_t nfs_file_direct_write(struct kiocb *iocb, const struct iovec *iov, retval = nfs_direct_write(iocb, (unsigned long) buf, count, pos); - /* - * XXX: nfs_end_data_update() already ensures this file's - * cached data is subsequently invalidated. Do we really - * need to call invalidate_inode_pages2() again here? - * - * For aio writes, this invalidation will almost certainly - * occur before the writes complete. Kind of racey. - */ - if (mapping->nrpages) - invalidate_inode_pages2(mapping); - if (retval > 0) iocb->ki_pos = pos + retval; diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index bc9376ca86cd..08cc4c5919ab 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -131,6 +131,15 @@ void nfs_zap_caches(struct inode *inode) spin_unlock(&inode->i_lock); } +void nfs_zap_mapping(struct inode *inode, struct address_space *mapping) +{ + if (mapping->nrpages != 0) { + spin_lock(&inode->i_lock); + NFS_I(inode)->cache_validity |= NFS_INO_INVALID_DATA; + spin_unlock(&inode->i_lock); + } +} + static void nfs_zap_acl_cache(struct inode *inode) { void (*clear_acl_cache)(struct inode *); @@ -574,7 +583,7 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) nfs_inc_stats(inode, NFSIOS_INODEREVALIDATE); lock_kernel(); - if (!inode || is_bad_inode(inode)) + if (is_bad_inode(inode)) goto out_nowait; if (NFS_STALE(inode)) goto out_nowait; @@ -671,13 +680,20 @@ int nfs_revalidate_mapping(struct inode *inode, struct address_space *mapping) if ((nfsi->cache_validity & NFS_INO_REVAL_PAGECACHE) || nfs_attribute_timeout(inode)) ret = __nfs_revalidate_inode(NFS_SERVER(inode), inode); + if (ret < 0) + goto out; if (nfsi->cache_validity & NFS_INO_INVALID_DATA) { - nfs_inc_stats(inode, NFSIOS_DATAINVALIDATE); - if (S_ISREG(inode->i_mode)) - nfs_sync_mapping(mapping); - invalidate_inode_pages2(mapping); - + if (mapping->nrpages != 0) { + if (S_ISREG(inode->i_mode)) { + ret = nfs_sync_mapping(mapping); + if (ret < 0) + goto out; + } + ret = invalidate_inode_pages2(mapping); + if (ret < 0) + goto out; + } spin_lock(&inode->i_lock); nfsi->cache_validity &= ~NFS_INO_INVALID_DATA; if (S_ISDIR(inode->i_mode)) { @@ -687,10 +703,12 @@ int nfs_revalidate_mapping(struct inode *inode, struct address_space *mapping) } spin_unlock(&inode->i_lock); + nfs_inc_stats(inode, NFSIOS_DATAINVALIDATE); dfprintk(PAGECACHE, "NFS: (%s/%Ld) data cache invalidated\n", inode->i_sb->s_id, (long long)NFS_FILEID(inode)); } +out: return ret; } diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index bea0b016bd70..d205466233f6 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -93,15 +93,15 @@ extern void nfs_destroy_directcache(void); /* nfs2xdr.c */ extern int nfs_stat_to_errno(int); extern struct rpc_procinfo nfs_procedures[]; -extern u32 * nfs_decode_dirent(u32 *, struct nfs_entry *, int); +extern __be32 * nfs_decode_dirent(__be32 *, struct nfs_entry *, int); /* nfs3xdr.c */ extern struct rpc_procinfo nfs3_procedures[]; -extern u32 *nfs3_decode_dirent(u32 *, struct nfs_entry *, int); +extern __be32 *nfs3_decode_dirent(__be32 *, struct nfs_entry *, int); /* nfs4xdr.c */ #ifdef CONFIG_NFS_V4 -extern u32 *nfs4_decode_dirent(u32 *p, struct nfs_entry *entry, int plus); +extern __be32 *nfs4_decode_dirent(__be32 *p, struct nfs_entry *entry, int plus); #endif /* nfs4proc.c */ diff --git a/fs/nfs/mount_clnt.c b/fs/nfs/mount_clnt.c index d507b021207f..f75fe72b4160 100644 --- a/fs/nfs/mount_clnt.c +++ b/fs/nfs/mount_clnt.c @@ -95,7 +95,7 @@ mnt_create(char *hostname, struct sockaddr_in *srvaddr, int version, * XDR encode/decode functions for MOUNT */ static int -xdr_encode_dirpath(struct rpc_rqst *req, u32 *p, const char *path) +xdr_encode_dirpath(struct rpc_rqst *req, __be32 *p, const char *path) { p = xdr_encode_string(p, path); @@ -104,7 +104,7 @@ xdr_encode_dirpath(struct rpc_rqst *req, u32 *p, const char *path) } static int -xdr_decode_fhstatus(struct rpc_rqst *req, u32 *p, struct mnt_fhstatus *res) +xdr_decode_fhstatus(struct rpc_rqst *req, __be32 *p, struct mnt_fhstatus *res) { struct nfs_fh *fh = res->fh; @@ -116,7 +116,7 @@ xdr_decode_fhstatus(struct rpc_rqst *req, u32 *p, struct mnt_fhstatus *res) } static int -xdr_decode_fhstatus3(struct rpc_rqst *req, u32 *p, struct mnt_fhstatus *res) +xdr_decode_fhstatus3(struct rpc_rqst *req, __be32 *p, struct mnt_fhstatus *res) { struct nfs_fh *fh = res->fh; diff --git a/fs/nfs/nfs2xdr.c b/fs/nfs/nfs2xdr.c index b49501fc0a79..3be4e72a0227 100644 --- a/fs/nfs/nfs2xdr.c +++ b/fs/nfs/nfs2xdr.c @@ -66,15 +66,15 @@ /* * Common NFS XDR functions as inlines */ -static inline u32 * -xdr_encode_fhandle(u32 *p, struct nfs_fh *fhandle) +static inline __be32 * +xdr_encode_fhandle(__be32 *p, struct nfs_fh *fhandle) { memcpy(p, fhandle->data, NFS2_FHSIZE); return p + XDR_QUADLEN(NFS2_FHSIZE); } -static inline u32 * -xdr_decode_fhandle(u32 *p, struct nfs_fh *fhandle) +static inline __be32 * +xdr_decode_fhandle(__be32 *p, struct nfs_fh *fhandle) { /* NFSv2 handles have a fixed length */ fhandle->size = NFS2_FHSIZE; @@ -82,8 +82,8 @@ xdr_decode_fhandle(u32 *p, struct nfs_fh *fhandle) return p + XDR_QUADLEN(NFS2_FHSIZE); } -static inline u32* -xdr_encode_time(u32 *p, struct timespec *timep) +static inline __be32* +xdr_encode_time(__be32 *p, struct timespec *timep) { *p++ = htonl(timep->tv_sec); /* Convert nanoseconds into microseconds */ @@ -91,8 +91,8 @@ xdr_encode_time(u32 *p, struct timespec *timep) return p; } -static inline u32* -xdr_encode_current_server_time(u32 *p, struct timespec *timep) +static inline __be32* +xdr_encode_current_server_time(__be32 *p, struct timespec *timep) { /* * Passing the invalid value useconds=1000000 is a @@ -108,8 +108,8 @@ xdr_encode_current_server_time(u32 *p, struct timespec *timep) return p; } -static inline u32* -xdr_decode_time(u32 *p, struct timespec *timep) +static inline __be32* +xdr_decode_time(__be32 *p, struct timespec *timep) { timep->tv_sec = ntohl(*p++); /* Convert microseconds into nanoseconds */ @@ -117,8 +117,8 @@ xdr_decode_time(u32 *p, struct timespec *timep) return p; } -static u32 * -xdr_decode_fattr(u32 *p, struct nfs_fattr *fattr) +static __be32 * +xdr_decode_fattr(__be32 *p, struct nfs_fattr *fattr) { u32 rdev; fattr->type = (enum nfs_ftype) ntohl(*p++); @@ -146,10 +146,10 @@ xdr_decode_fattr(u32 *p, struct nfs_fattr *fattr) return p; } -static inline u32 * -xdr_encode_sattr(u32 *p, struct iattr *attr) +static inline __be32 * +xdr_encode_sattr(__be32 *p, struct iattr *attr) { - const u32 not_set = __constant_htonl(0xFFFFFFFF); + const __be32 not_set = __constant_htonl(0xFFFFFFFF); *p++ = (attr->ia_valid & ATTR_MODE) ? htonl(attr->ia_mode) : not_set; *p++ = (attr->ia_valid & ATTR_UID) ? htonl(attr->ia_uid) : not_set; @@ -184,7 +184,7 @@ xdr_encode_sattr(u32 *p, struct iattr *attr) * GETATTR, READLINK, STATFS */ static int -nfs_xdr_fhandle(struct rpc_rqst *req, u32 *p, struct nfs_fh *fh) +nfs_xdr_fhandle(struct rpc_rqst *req, __be32 *p, struct nfs_fh *fh) { p = xdr_encode_fhandle(p, fh); req->rq_slen = xdr_adjust_iovec(req->rq_svec, p); @@ -195,7 +195,7 @@ nfs_xdr_fhandle(struct rpc_rqst *req, u32 *p, struct nfs_fh *fh) * Encode SETATTR arguments */ static int -nfs_xdr_sattrargs(struct rpc_rqst *req, u32 *p, struct nfs_sattrargs *args) +nfs_xdr_sattrargs(struct rpc_rqst *req, __be32 *p, struct nfs_sattrargs *args) { p = xdr_encode_fhandle(p, args->fh); p = xdr_encode_sattr(p, args->sattr); @@ -208,7 +208,7 @@ nfs_xdr_sattrargs(struct rpc_rqst *req, u32 *p, struct nfs_sattrargs *args) * LOOKUP, REMOVE, RMDIR */ static int -nfs_xdr_diropargs(struct rpc_rqst *req, u32 *p, struct nfs_diropargs *args) +nfs_xdr_diropargs(struct rpc_rqst *req, __be32 *p, struct nfs_diropargs *args) { p = xdr_encode_fhandle(p, args->fh); p = xdr_encode_array(p, args->name, args->len); @@ -222,7 +222,7 @@ nfs_xdr_diropargs(struct rpc_rqst *req, u32 *p, struct nfs_diropargs *args) * exactly to the page we want to fetch. */ static int -nfs_xdr_readargs(struct rpc_rqst *req, u32 *p, struct nfs_readargs *args) +nfs_xdr_readargs(struct rpc_rqst *req, __be32 *p, struct nfs_readargs *args) { struct rpc_auth *auth = req->rq_task->tk_auth; unsigned int replen; @@ -246,7 +246,7 @@ nfs_xdr_readargs(struct rpc_rqst *req, u32 *p, struct nfs_readargs *args) * Decode READ reply */ static int -nfs_xdr_readres(struct rpc_rqst *req, u32 *p, struct nfs_readres *res) +nfs_xdr_readres(struct rpc_rqst *req, __be32 *p, struct nfs_readres *res) { struct kvec *iov = req->rq_rcv_buf.head; int status, count, recvd, hdrlen; @@ -286,7 +286,7 @@ nfs_xdr_readres(struct rpc_rqst *req, u32 *p, struct nfs_readres *res) * Write arguments. Splice the buffer to be written into the iovec. */ static int -nfs_xdr_writeargs(struct rpc_rqst *req, u32 *p, struct nfs_writeargs *args) +nfs_xdr_writeargs(struct rpc_rqst *req, __be32 *p, struct nfs_writeargs *args) { struct xdr_buf *sndbuf = &req->rq_snd_buf; u32 offset = (u32)args->offset; @@ -309,7 +309,7 @@ nfs_xdr_writeargs(struct rpc_rqst *req, u32 *p, struct nfs_writeargs *args) * CREATE, MKDIR */ static int -nfs_xdr_createargs(struct rpc_rqst *req, u32 *p, struct nfs_createargs *args) +nfs_xdr_createargs(struct rpc_rqst *req, __be32 *p, struct nfs_createargs *args) { p = xdr_encode_fhandle(p, args->fh); p = xdr_encode_array(p, args->name, args->len); @@ -322,7 +322,7 @@ nfs_xdr_createargs(struct rpc_rqst *req, u32 *p, struct nfs_createargs *args) * Encode RENAME arguments */ static int -nfs_xdr_renameargs(struct rpc_rqst *req, u32 *p, struct nfs_renameargs *args) +nfs_xdr_renameargs(struct rpc_rqst *req, __be32 *p, struct nfs_renameargs *args) { p = xdr_encode_fhandle(p, args->fromfh); p = xdr_encode_array(p, args->fromname, args->fromlen); @@ -336,7 +336,7 @@ nfs_xdr_renameargs(struct rpc_rqst *req, u32 *p, struct nfs_renameargs *args) * Encode LINK arguments */ static int -nfs_xdr_linkargs(struct rpc_rqst *req, u32 *p, struct nfs_linkargs *args) +nfs_xdr_linkargs(struct rpc_rqst *req, __be32 *p, struct nfs_linkargs *args) { p = xdr_encode_fhandle(p, args->fromfh); p = xdr_encode_fhandle(p, args->tofh); @@ -349,7 +349,7 @@ nfs_xdr_linkargs(struct rpc_rqst *req, u32 *p, struct nfs_linkargs *args) * Encode SYMLINK arguments */ static int -nfs_xdr_symlinkargs(struct rpc_rqst *req, u32 *p, struct nfs_symlinkargs *args) +nfs_xdr_symlinkargs(struct rpc_rqst *req, __be32 *p, struct nfs_symlinkargs *args) { struct xdr_buf *sndbuf = &req->rq_snd_buf; size_t pad; @@ -378,7 +378,7 @@ nfs_xdr_symlinkargs(struct rpc_rqst *req, u32 *p, struct nfs_symlinkargs *args) * Encode arguments to readdir call */ static int -nfs_xdr_readdirargs(struct rpc_rqst *req, u32 *p, struct nfs_readdirargs *args) +nfs_xdr_readdirargs(struct rpc_rqst *req, __be32 *p, struct nfs_readdirargs *args) { struct rpc_task *task = req->rq_task; struct rpc_auth *auth = task->tk_auth; @@ -404,7 +404,7 @@ nfs_xdr_readdirargs(struct rpc_rqst *req, u32 *p, struct nfs_readdirargs *args) * from nfs_readdir for each entry. */ static int -nfs_xdr_readdirres(struct rpc_rqst *req, u32 *p, void *dummy) +nfs_xdr_readdirres(struct rpc_rqst *req, __be32 *p, void *dummy) { struct xdr_buf *rcvbuf = &req->rq_rcv_buf; struct kvec *iov = rcvbuf->head; @@ -412,7 +412,7 @@ nfs_xdr_readdirres(struct rpc_rqst *req, u32 *p, void *dummy) int hdrlen, recvd; int status, nr; unsigned int len, pglen; - u32 *end, *entry, *kaddr; + __be32 *end, *entry, *kaddr; if ((status = ntohl(*p++))) return -nfs_stat_to_errno(status); @@ -432,8 +432,8 @@ nfs_xdr_readdirres(struct rpc_rqst *req, u32 *p, void *dummy) if (pglen > recvd) pglen = recvd; page = rcvbuf->pages; - kaddr = p = (u32 *)kmap_atomic(*page, KM_USER0); - end = (u32 *)((char *)p + pglen); + kaddr = p = kmap_atomic(*page, KM_USER0); + end = (__be32 *)((char *)p + pglen); entry = p; for (nr = 0; *p++; nr++) { if (p + 2 > end) @@ -468,8 +468,8 @@ err_unmap: goto out; } -u32 * -nfs_decode_dirent(u32 *p, struct nfs_entry *entry, int plus) +__be32 * +nfs_decode_dirent(__be32 *p, struct nfs_entry *entry, int plus) { if (!*p++) { if (!*p) @@ -496,7 +496,7 @@ nfs_decode_dirent(u32 *p, struct nfs_entry *entry, int plus) * Decode simple status reply */ static int -nfs_xdr_stat(struct rpc_rqst *req, u32 *p, void *dummy) +nfs_xdr_stat(struct rpc_rqst *req, __be32 *p, void *dummy) { int status; @@ -510,7 +510,7 @@ nfs_xdr_stat(struct rpc_rqst *req, u32 *p, void *dummy) * GETATTR, SETATTR, WRITE */ static int -nfs_xdr_attrstat(struct rpc_rqst *req, u32 *p, struct nfs_fattr *fattr) +nfs_xdr_attrstat(struct rpc_rqst *req, __be32 *p, struct nfs_fattr *fattr) { int status; @@ -525,7 +525,7 @@ nfs_xdr_attrstat(struct rpc_rqst *req, u32 *p, struct nfs_fattr *fattr) * LOOKUP, CREATE, MKDIR */ static int -nfs_xdr_diropres(struct rpc_rqst *req, u32 *p, struct nfs_diropok *res) +nfs_xdr_diropres(struct rpc_rqst *req, __be32 *p, struct nfs_diropok *res) { int status; @@ -540,7 +540,7 @@ nfs_xdr_diropres(struct rpc_rqst *req, u32 *p, struct nfs_diropok *res) * Encode READLINK args */ static int -nfs_xdr_readlinkargs(struct rpc_rqst *req, u32 *p, struct nfs_readlinkargs *args) +nfs_xdr_readlinkargs(struct rpc_rqst *req, __be32 *p, struct nfs_readlinkargs *args) { struct rpc_auth *auth = req->rq_task->tk_auth; unsigned int replen; @@ -558,7 +558,7 @@ nfs_xdr_readlinkargs(struct rpc_rqst *req, u32 *p, struct nfs_readlinkargs *args * Decode READLINK reply */ static int -nfs_xdr_readlinkres(struct rpc_rqst *req, u32 *p, void *dummy) +nfs_xdr_readlinkres(struct rpc_rqst *req, __be32 *p, void *dummy) { struct xdr_buf *rcvbuf = &req->rq_rcv_buf; struct kvec *iov = rcvbuf->head; @@ -601,7 +601,7 @@ nfs_xdr_readlinkres(struct rpc_rqst *req, u32 *p, void *dummy) * Decode WRITE reply */ static int -nfs_xdr_writeres(struct rpc_rqst *req, u32 *p, struct nfs_writeres *res) +nfs_xdr_writeres(struct rpc_rqst *req, __be32 *p, struct nfs_writeres *res) { res->verf->committed = NFS_FILE_SYNC; return nfs_xdr_attrstat(req, p, res->fattr); @@ -611,7 +611,7 @@ nfs_xdr_writeres(struct rpc_rqst *req, u32 *p, struct nfs_writeres *res) * Decode STATFS reply */ static int -nfs_xdr_statfsres(struct rpc_rqst *req, u32 *p, struct nfs2_fsstat *res) +nfs_xdr_statfsres(struct rpc_rqst *req, __be32 *p, struct nfs2_fsstat *res) { int status; diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c index 3b234d4601e7..e5f128ffc32d 100644 --- a/fs/nfs/nfs3proc.c +++ b/fs/nfs/nfs3proc.c @@ -668,7 +668,7 @@ nfs3_proc_readdir(struct dentry *dentry, struct rpc_cred *cred, { struct inode *dir = dentry->d_inode; struct nfs_fattr dir_attr; - u32 *verf = NFS_COOKIEVERF(dir); + __be32 *verf = NFS_COOKIEVERF(dir); struct nfs3_readdirargs arg = { .fh = NFS_FH(dir), .cookie = cookie, diff --git a/fs/nfs/nfs3xdr.c b/fs/nfs/nfs3xdr.c index 16556fa4effb..0ace092d126f 100644 --- a/fs/nfs/nfs3xdr.c +++ b/fs/nfs/nfs3xdr.c @@ -105,14 +105,14 @@ static struct { /* * Common NFS XDR functions as inlines */ -static inline u32 * -xdr_encode_fhandle(u32 *p, struct nfs_fh *fh) +static inline __be32 * +xdr_encode_fhandle(__be32 *p, struct nfs_fh *fh) { return xdr_encode_array(p, fh->data, fh->size); } -static inline u32 * -xdr_decode_fhandle(u32 *p, struct nfs_fh *fh) +static inline __be32 * +xdr_decode_fhandle(__be32 *p, struct nfs_fh *fh) { if ((fh->size = ntohl(*p++)) <= NFS3_FHSIZE) { memcpy(fh->data, p, fh->size); @@ -124,24 +124,24 @@ xdr_decode_fhandle(u32 *p, struct nfs_fh *fh) /* * Encode/decode time. */ -static inline u32 * -xdr_encode_time3(u32 *p, struct timespec *timep) +static inline __be32 * +xdr_encode_time3(__be32 *p, struct timespec *timep) { *p++ = htonl(timep->tv_sec); *p++ = htonl(timep->tv_nsec); return p; } -static inline u32 * -xdr_decode_time3(u32 *p, struct timespec *timep) +static inline __be32 * +xdr_decode_time3(__be32 *p, struct timespec *timep) { timep->tv_sec = ntohl(*p++); timep->tv_nsec = ntohl(*p++); return p; } -static u32 * -xdr_decode_fattr(u32 *p, struct nfs_fattr *fattr) +static __be32 * +xdr_decode_fattr(__be32 *p, struct nfs_fattr *fattr) { unsigned int type, major, minor; int fmode; @@ -177,8 +177,8 @@ xdr_decode_fattr(u32 *p, struct nfs_fattr *fattr) return p; } -static inline u32 * -xdr_encode_sattr(u32 *p, struct iattr *attr) +static inline __be32 * +xdr_encode_sattr(__be32 *p, struct iattr *attr) { if (attr->ia_valid & ATTR_MODE) { *p++ = xdr_one; @@ -223,8 +223,8 @@ xdr_encode_sattr(u32 *p, struct iattr *attr) return p; } -static inline u32 * -xdr_decode_wcc_attr(u32 *p, struct nfs_fattr *fattr) +static inline __be32 * +xdr_decode_wcc_attr(__be32 *p, struct nfs_fattr *fattr) { p = xdr_decode_hyper(p, &fattr->pre_size); p = xdr_decode_time3(p, &fattr->pre_mtime); @@ -233,16 +233,16 @@ xdr_decode_wcc_attr(u32 *p, struct nfs_fattr *fattr) return p; } -static inline u32 * -xdr_decode_post_op_attr(u32 *p, struct nfs_fattr *fattr) +static inline __be32 * +xdr_decode_post_op_attr(__be32 *p, struct nfs_fattr *fattr) { if (*p++) p = xdr_decode_fattr(p, fattr); return p; } -static inline u32 * -xdr_decode_pre_op_attr(u32 *p, struct nfs_fattr *fattr) +static inline __be32 * +xdr_decode_pre_op_attr(__be32 *p, struct nfs_fattr *fattr) { if (*p++) return xdr_decode_wcc_attr(p, fattr); @@ -250,8 +250,8 @@ xdr_decode_pre_op_attr(u32 *p, struct nfs_fattr *fattr) } -static inline u32 * -xdr_decode_wcc_data(u32 *p, struct nfs_fattr *fattr) +static inline __be32 * +xdr_decode_wcc_data(__be32 *p, struct nfs_fattr *fattr) { p = xdr_decode_pre_op_attr(p, fattr); return xdr_decode_post_op_attr(p, fattr); @@ -265,7 +265,7 @@ xdr_decode_wcc_data(u32 *p, struct nfs_fattr *fattr) * Encode file handle argument */ static int -nfs3_xdr_fhandle(struct rpc_rqst *req, u32 *p, struct nfs_fh *fh) +nfs3_xdr_fhandle(struct rpc_rqst *req, __be32 *p, struct nfs_fh *fh) { p = xdr_encode_fhandle(p, fh); req->rq_slen = xdr_adjust_iovec(req->rq_svec, p); @@ -276,7 +276,7 @@ nfs3_xdr_fhandle(struct rpc_rqst *req, u32 *p, struct nfs_fh *fh) * Encode SETATTR arguments */ static int -nfs3_xdr_sattrargs(struct rpc_rqst *req, u32 *p, struct nfs3_sattrargs *args) +nfs3_xdr_sattrargs(struct rpc_rqst *req, __be32 *p, struct nfs3_sattrargs *args) { p = xdr_encode_fhandle(p, args->fh); p = xdr_encode_sattr(p, args->sattr); @@ -291,7 +291,7 @@ nfs3_xdr_sattrargs(struct rpc_rqst *req, u32 *p, struct nfs3_sattrargs *args) * Encode directory ops argument */ static int -nfs3_xdr_diropargs(struct rpc_rqst *req, u32 *p, struct nfs3_diropargs *args) +nfs3_xdr_diropargs(struct rpc_rqst *req, __be32 *p, struct nfs3_diropargs *args) { p = xdr_encode_fhandle(p, args->fh); p = xdr_encode_array(p, args->name, args->len); @@ -303,7 +303,7 @@ nfs3_xdr_diropargs(struct rpc_rqst *req, u32 *p, struct nfs3_diropargs *args) * Encode access() argument */ static int -nfs3_xdr_accessargs(struct rpc_rqst *req, u32 *p, struct nfs3_accessargs *args) +nfs3_xdr_accessargs(struct rpc_rqst *req, __be32 *p, struct nfs3_accessargs *args) { p = xdr_encode_fhandle(p, args->fh); *p++ = htonl(args->access); @@ -317,7 +317,7 @@ nfs3_xdr_accessargs(struct rpc_rqst *req, u32 *p, struct nfs3_accessargs *args) * exactly to the page we want to fetch. */ static int -nfs3_xdr_readargs(struct rpc_rqst *req, u32 *p, struct nfs_readargs *args) +nfs3_xdr_readargs(struct rpc_rqst *req, __be32 *p, struct nfs_readargs *args) { struct rpc_auth *auth = req->rq_task->tk_auth; unsigned int replen; @@ -339,7 +339,7 @@ nfs3_xdr_readargs(struct rpc_rqst *req, u32 *p, struct nfs_readargs *args) * Write arguments. Splice the buffer to be written into the iovec. */ static int -nfs3_xdr_writeargs(struct rpc_rqst *req, u32 *p, struct nfs_writeargs *args) +nfs3_xdr_writeargs(struct rpc_rqst *req, __be32 *p, struct nfs_writeargs *args) { struct xdr_buf *sndbuf = &req->rq_snd_buf; u32 count = args->count; @@ -360,7 +360,7 @@ nfs3_xdr_writeargs(struct rpc_rqst *req, u32 *p, struct nfs_writeargs *args) * Encode CREATE arguments */ static int -nfs3_xdr_createargs(struct rpc_rqst *req, u32 *p, struct nfs3_createargs *args) +nfs3_xdr_createargs(struct rpc_rqst *req, __be32 *p, struct nfs3_createargs *args) { p = xdr_encode_fhandle(p, args->fh); p = xdr_encode_array(p, args->name, args->len); @@ -380,7 +380,7 @@ nfs3_xdr_createargs(struct rpc_rqst *req, u32 *p, struct nfs3_createargs *args) * Encode MKDIR arguments */ static int -nfs3_xdr_mkdirargs(struct rpc_rqst *req, u32 *p, struct nfs3_mkdirargs *args) +nfs3_xdr_mkdirargs(struct rpc_rqst *req, __be32 *p, struct nfs3_mkdirargs *args) { p = xdr_encode_fhandle(p, args->fh); p = xdr_encode_array(p, args->name, args->len); @@ -393,7 +393,7 @@ nfs3_xdr_mkdirargs(struct rpc_rqst *req, u32 *p, struct nfs3_mkdirargs *args) * Encode SYMLINK arguments */ static int -nfs3_xdr_symlinkargs(struct rpc_rqst *req, u32 *p, struct nfs3_symlinkargs *args) +nfs3_xdr_symlinkargs(struct rpc_rqst *req, __be32 *p, struct nfs3_symlinkargs *args) { p = xdr_encode_fhandle(p, args->fromfh); p = xdr_encode_array(p, args->fromname, args->fromlen); @@ -410,7 +410,7 @@ nfs3_xdr_symlinkargs(struct rpc_rqst *req, u32 *p, struct nfs3_symlinkargs *args * Encode MKNOD arguments */ static int -nfs3_xdr_mknodargs(struct rpc_rqst *req, u32 *p, struct nfs3_mknodargs *args) +nfs3_xdr_mknodargs(struct rpc_rqst *req, __be32 *p, struct nfs3_mknodargs *args) { p = xdr_encode_fhandle(p, args->fh); p = xdr_encode_array(p, args->name, args->len); @@ -429,7 +429,7 @@ nfs3_xdr_mknodargs(struct rpc_rqst *req, u32 *p, struct nfs3_mknodargs *args) * Encode RENAME arguments */ static int -nfs3_xdr_renameargs(struct rpc_rqst *req, u32 *p, struct nfs3_renameargs *args) +nfs3_xdr_renameargs(struct rpc_rqst *req, __be32 *p, struct nfs3_renameargs *args) { p = xdr_encode_fhandle(p, args->fromfh); p = xdr_encode_array(p, args->fromname, args->fromlen); @@ -443,7 +443,7 @@ nfs3_xdr_renameargs(struct rpc_rqst *req, u32 *p, struct nfs3_renameargs *args) * Encode LINK arguments */ static int -nfs3_xdr_linkargs(struct rpc_rqst *req, u32 *p, struct nfs3_linkargs *args) +nfs3_xdr_linkargs(struct rpc_rqst *req, __be32 *p, struct nfs3_linkargs *args) { p = xdr_encode_fhandle(p, args->fromfh); p = xdr_encode_fhandle(p, args->tofh); @@ -456,7 +456,7 @@ nfs3_xdr_linkargs(struct rpc_rqst *req, u32 *p, struct nfs3_linkargs *args) * Encode arguments to readdir call */ static int -nfs3_xdr_readdirargs(struct rpc_rqst *req, u32 *p, struct nfs3_readdirargs *args) +nfs3_xdr_readdirargs(struct rpc_rqst *req, __be32 *p, struct nfs3_readdirargs *args) { struct rpc_auth *auth = req->rq_task->tk_auth; unsigned int replen; @@ -485,7 +485,7 @@ nfs3_xdr_readdirargs(struct rpc_rqst *req, u32 *p, struct nfs3_readdirargs *args * We just check for syntactical correctness. */ static int -nfs3_xdr_readdirres(struct rpc_rqst *req, u32 *p, struct nfs3_readdirres *res) +nfs3_xdr_readdirres(struct rpc_rqst *req, __be32 *p, struct nfs3_readdirres *res) { struct xdr_buf *rcvbuf = &req->rq_rcv_buf; struct kvec *iov = rcvbuf->head; @@ -493,7 +493,7 @@ nfs3_xdr_readdirres(struct rpc_rqst *req, u32 *p, struct nfs3_readdirres *res) int hdrlen, recvd; int status, nr; unsigned int len, pglen; - u32 *entry, *end, *kaddr; + __be32 *entry, *end, *kaddr; status = ntohl(*p++); /* Decode post_op_attrs */ @@ -523,8 +523,8 @@ nfs3_xdr_readdirres(struct rpc_rqst *req, u32 *p, struct nfs3_readdirres *res) if (pglen > recvd) pglen = recvd; page = rcvbuf->pages; - kaddr = p = (u32 *)kmap_atomic(*page, KM_USER0); - end = (u32 *)((char *)p + pglen); + kaddr = p = kmap_atomic(*page, KM_USER0); + end = (__be32 *)((char *)p + pglen); entry = p; for (nr = 0; *p++; nr++) { if (p + 3 > end) @@ -583,8 +583,8 @@ err_unmap: goto out; } -u32 * -nfs3_decode_dirent(u32 *p, struct nfs_entry *entry, int plus) +__be32 * +nfs3_decode_dirent(__be32 *p, struct nfs_entry *entry, int plus) { struct nfs_entry old = *entry; @@ -626,7 +626,7 @@ nfs3_decode_dirent(u32 *p, struct nfs_entry *entry, int plus) * Encode COMMIT arguments */ static int -nfs3_xdr_commitargs(struct rpc_rqst *req, u32 *p, struct nfs_writeargs *args) +nfs3_xdr_commitargs(struct rpc_rqst *req, __be32 *p, struct nfs_writeargs *args) { p = xdr_encode_fhandle(p, args->fh); p = xdr_encode_hyper(p, args->offset); @@ -640,7 +640,7 @@ nfs3_xdr_commitargs(struct rpc_rqst *req, u32 *p, struct nfs_writeargs *args) * Encode GETACL arguments */ static int -nfs3_xdr_getaclargs(struct rpc_rqst *req, u32 *p, +nfs3_xdr_getaclargs(struct rpc_rqst *req, __be32 *p, struct nfs3_getaclargs *args) { struct rpc_auth *auth = req->rq_task->tk_auth; @@ -664,7 +664,7 @@ nfs3_xdr_getaclargs(struct rpc_rqst *req, u32 *p, * Encode SETACL arguments */ static int -nfs3_xdr_setaclargs(struct rpc_rqst *req, u32 *p, +nfs3_xdr_setaclargs(struct rpc_rqst *req, __be32 *p, struct nfs3_setaclargs *args) { struct xdr_buf *buf = &req->rq_snd_buf; @@ -711,7 +711,7 @@ nfs3_xdr_setaclargs(struct rpc_rqst *req, u32 *p, * Decode attrstat reply. */ static int -nfs3_xdr_attrstat(struct rpc_rqst *req, u32 *p, struct nfs_fattr *fattr) +nfs3_xdr_attrstat(struct rpc_rqst *req, __be32 *p, struct nfs_fattr *fattr) { int status; @@ -726,7 +726,7 @@ nfs3_xdr_attrstat(struct rpc_rqst *req, u32 *p, struct nfs_fattr *fattr) * SATTR, REMOVE, RMDIR */ static int -nfs3_xdr_wccstat(struct rpc_rqst *req, u32 *p, struct nfs_fattr *fattr) +nfs3_xdr_wccstat(struct rpc_rqst *req, __be32 *p, struct nfs_fattr *fattr) { int status; @@ -740,7 +740,7 @@ nfs3_xdr_wccstat(struct rpc_rqst *req, u32 *p, struct nfs_fattr *fattr) * Decode LOOKUP reply */ static int -nfs3_xdr_lookupres(struct rpc_rqst *req, u32 *p, struct nfs3_diropres *res) +nfs3_xdr_lookupres(struct rpc_rqst *req, __be32 *p, struct nfs3_diropres *res) { int status; @@ -759,7 +759,7 @@ nfs3_xdr_lookupres(struct rpc_rqst *req, u32 *p, struct nfs3_diropres *res) * Decode ACCESS reply */ static int -nfs3_xdr_accessres(struct rpc_rqst *req, u32 *p, struct nfs3_accessres *res) +nfs3_xdr_accessres(struct rpc_rqst *req, __be32 *p, struct nfs3_accessres *res) { int status = ntohl(*p++); @@ -771,7 +771,7 @@ nfs3_xdr_accessres(struct rpc_rqst *req, u32 *p, struct nfs3_accessres *res) } static int -nfs3_xdr_readlinkargs(struct rpc_rqst *req, u32 *p, struct nfs3_readlinkargs *args) +nfs3_xdr_readlinkargs(struct rpc_rqst *req, __be32 *p, struct nfs3_readlinkargs *args) { struct rpc_auth *auth = req->rq_task->tk_auth; unsigned int replen; @@ -789,7 +789,7 @@ nfs3_xdr_readlinkargs(struct rpc_rqst *req, u32 *p, struct nfs3_readlinkargs *ar * Decode READLINK reply */ static int -nfs3_xdr_readlinkres(struct rpc_rqst *req, u32 *p, struct nfs_fattr *fattr) +nfs3_xdr_readlinkres(struct rpc_rqst *req, __be32 *p, struct nfs_fattr *fattr) { struct xdr_buf *rcvbuf = &req->rq_rcv_buf; struct kvec *iov = rcvbuf->head; @@ -837,7 +837,7 @@ nfs3_xdr_readlinkres(struct rpc_rqst *req, u32 *p, struct nfs_fattr *fattr) * Decode READ reply */ static int -nfs3_xdr_readres(struct rpc_rqst *req, u32 *p, struct nfs_readres *res) +nfs3_xdr_readres(struct rpc_rqst *req, __be32 *p, struct nfs_readres *res) { struct kvec *iov = req->rq_rcv_buf.head; int status, count, ocount, recvd, hdrlen; @@ -888,7 +888,7 @@ nfs3_xdr_readres(struct rpc_rqst *req, u32 *p, struct nfs_readres *res) * Decode WRITE response */ static int -nfs3_xdr_writeres(struct rpc_rqst *req, u32 *p, struct nfs_writeres *res) +nfs3_xdr_writeres(struct rpc_rqst *req, __be32 *p, struct nfs_writeres *res) { int status; @@ -910,7 +910,7 @@ nfs3_xdr_writeres(struct rpc_rqst *req, u32 *p, struct nfs_writeres *res) * Decode a CREATE response */ static int -nfs3_xdr_createres(struct rpc_rqst *req, u32 *p, struct nfs3_diropres *res) +nfs3_xdr_createres(struct rpc_rqst *req, __be32 *p, struct nfs3_diropres *res) { int status; @@ -937,7 +937,7 @@ nfs3_xdr_createres(struct rpc_rqst *req, u32 *p, struct nfs3_diropres *res) * Decode RENAME reply */ static int -nfs3_xdr_renameres(struct rpc_rqst *req, u32 *p, struct nfs3_renameres *res) +nfs3_xdr_renameres(struct rpc_rqst *req, __be32 *p, struct nfs3_renameres *res) { int status; @@ -952,7 +952,7 @@ nfs3_xdr_renameres(struct rpc_rqst *req, u32 *p, struct nfs3_renameres *res) * Decode LINK reply */ static int -nfs3_xdr_linkres(struct rpc_rqst *req, u32 *p, struct nfs3_linkres *res) +nfs3_xdr_linkres(struct rpc_rqst *req, __be32 *p, struct nfs3_linkres *res) { int status; @@ -967,7 +967,7 @@ nfs3_xdr_linkres(struct rpc_rqst *req, u32 *p, struct nfs3_linkres *res) * Decode FSSTAT reply */ static int -nfs3_xdr_fsstatres(struct rpc_rqst *req, u32 *p, struct nfs_fsstat *res) +nfs3_xdr_fsstatres(struct rpc_rqst *req, __be32 *p, struct nfs_fsstat *res) { int status; @@ -992,7 +992,7 @@ nfs3_xdr_fsstatres(struct rpc_rqst *req, u32 *p, struct nfs_fsstat *res) * Decode FSINFO reply */ static int -nfs3_xdr_fsinfores(struct rpc_rqst *req, u32 *p, struct nfs_fsinfo *res) +nfs3_xdr_fsinfores(struct rpc_rqst *req, __be32 *p, struct nfs_fsinfo *res) { int status; @@ -1020,7 +1020,7 @@ nfs3_xdr_fsinfores(struct rpc_rqst *req, u32 *p, struct nfs_fsinfo *res) * Decode PATHCONF reply */ static int -nfs3_xdr_pathconfres(struct rpc_rqst *req, u32 *p, struct nfs_pathconf *res) +nfs3_xdr_pathconfres(struct rpc_rqst *req, __be32 *p, struct nfs_pathconf *res) { int status; @@ -1040,7 +1040,7 @@ nfs3_xdr_pathconfres(struct rpc_rqst *req, u32 *p, struct nfs_pathconf *res) * Decode COMMIT reply */ static int -nfs3_xdr_commitres(struct rpc_rqst *req, u32 *p, struct nfs_writeres *res) +nfs3_xdr_commitres(struct rpc_rqst *req, __be32 *p, struct nfs_writeres *res) { int status; @@ -1059,7 +1059,7 @@ nfs3_xdr_commitres(struct rpc_rqst *req, u32 *p, struct nfs_writeres *res) * Decode GETACL reply */ static int -nfs3_xdr_getaclres(struct rpc_rqst *req, u32 *p, +nfs3_xdr_getaclres(struct rpc_rqst *req, __be32 *p, struct nfs3_getaclres *res) { struct xdr_buf *buf = &req->rq_rcv_buf; @@ -1091,7 +1091,7 @@ nfs3_xdr_getaclres(struct rpc_rqst *req, u32 *p, * Decode setacl reply. */ static int -nfs3_xdr_setaclres(struct rpc_rqst *req, u32 *p, struct nfs_fattr *fattr) +nfs3_xdr_setaclres(struct rpc_rqst *req, __be32 *p, struct nfs_fattr *fattr) { int status = ntohl(*p++); diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h index 61095fe4b5ca..6f346677332d 100644 --- a/fs/nfs/nfs4_fs.h +++ b/fs/nfs/nfs4_fs.h @@ -212,7 +212,7 @@ extern void nfs_free_seqid(struct nfs_seqid *seqid); extern const nfs4_stateid zero_stateid; /* nfs4xdr.c */ -extern uint32_t *nfs4_decode_dirent(uint32_t *p, struct nfs_entry *entry, int plus); +extern __be32 *nfs4_decode_dirent(__be32 *p, struct nfs_entry *entry, int plus); extern struct rpc_procinfo nfs4_procedures[]; struct nfs4_mount_data; diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 47c7e6e3910d..8118036cc449 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -138,10 +138,10 @@ const u32 nfs4_fs_locations_bitmap[2] = { | FATTR4_WORD1_MOUNTED_ON_FILEID }; -static void nfs4_setup_readdir(u64 cookie, u32 *verifier, struct dentry *dentry, +static void nfs4_setup_readdir(u64 cookie, __be32 *verifier, struct dentry *dentry, struct nfs4_readdir_arg *readdir) { - u32 *start, *p; + __be32 *start, *p; BUG_ON(readdir->count < 80); if (cookie > 2) { @@ -162,7 +162,7 @@ static void nfs4_setup_readdir(u64 cookie, u32 *verifier, struct dentry *dentry, * when talking to the server, we always send cookie 0 * instead of 1 or 2. */ - start = p = (u32 *)kmap_atomic(*readdir->pages, KM_USER0); + start = p = kmap_atomic(*readdir->pages, KM_USER0); if (cookie == 0) { *p++ = xdr_one; /* next */ @@ -1314,11 +1314,9 @@ nfs4_open_revalidate(struct inode *dir, struct dentry *dentry, int openflags, st case -EROFS: lookup_instantiate_filp(nd, (struct dentry *)state, NULL); return 1; - case -ENOENT: - if (dentry->d_inode == NULL) - return 1; + default: + goto out_drop; } - goto out_drop; } if (state->inode == dentry->d_inode) { nfs4_intent_set_file(nd, dentry, state); @@ -2917,11 +2915,11 @@ int nfs4_proc_setclientid(struct nfs_client *clp, u32 program, unsigned short po .rpc_resp = clp, .rpc_cred = cred, }; - u32 *p; + __be32 *p; int loop = 0; int status; - p = (u32*)sc_verifier.data; + p = (__be32*)sc_verifier.data; *p++ = htonl((u32)clp->cl_boot_time.tv_sec); *p = htonl((u32)clp->cl_boot_time.tv_nsec); diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 3dd413f52da1..0cf3fa312a33 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -471,7 +471,7 @@ struct compound_hdr { static void encode_string(struct xdr_stream *xdr, unsigned int len, const char *str) { - uint32_t *p; + __be32 *p; p = xdr_reserve_space(xdr, 4 + len); BUG_ON(p == NULL); @@ -480,7 +480,7 @@ static void encode_string(struct xdr_stream *xdr, unsigned int len, const char * static int encode_compound_hdr(struct xdr_stream *xdr, struct compound_hdr *hdr) { - uint32_t *p; + __be32 *p; dprintk("encode_compound: tag=%.*s\n", (int)hdr->taglen, hdr->tag); BUG_ON(hdr->taglen > NFS4_MAXTAGLEN); @@ -494,7 +494,7 @@ static int encode_compound_hdr(struct xdr_stream *xdr, struct compound_hdr *hdr) static void encode_nfs4_verifier(struct xdr_stream *xdr, const nfs4_verifier *verf) { - uint32_t *p; + __be32 *p; p = xdr_reserve_space(xdr, NFS4_VERIFIER_SIZE); BUG_ON(p == NULL); @@ -507,8 +507,8 @@ static int encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, const s char owner_group[IDMAP_NAMESZ]; int owner_namelen = 0; int owner_grouplen = 0; - uint32_t *p; - uint32_t *q; + __be32 *p; + __be32 *q; int len; uint32_t bmval0 = 0; uint32_t bmval1 = 0; @@ -630,7 +630,7 @@ static int encode_attrs(struct xdr_stream *xdr, const struct iattr *iap, const s static int encode_access(struct xdr_stream *xdr, u32 access) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(8); WRITE32(OP_ACCESS); @@ -641,7 +641,7 @@ static int encode_access(struct xdr_stream *xdr, u32 access) static int encode_close(struct xdr_stream *xdr, const struct nfs_closeargs *arg) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(8+sizeof(arg->stateid->data)); WRITE32(OP_CLOSE); @@ -653,7 +653,7 @@ static int encode_close(struct xdr_stream *xdr, const struct nfs_closeargs *arg) static int encode_commit(struct xdr_stream *xdr, const struct nfs_writeargs *args) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(16); WRITE32(OP_COMMIT); @@ -665,7 +665,7 @@ static int encode_commit(struct xdr_stream *xdr, const struct nfs_writeargs *arg static int encode_create(struct xdr_stream *xdr, const struct nfs4_create_arg *create) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(8); WRITE32(OP_CREATE); @@ -697,7 +697,7 @@ static int encode_create(struct xdr_stream *xdr, const struct nfs4_create_arg *c static int encode_getattr_one(struct xdr_stream *xdr, uint32_t bitmap) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(12); WRITE32(OP_GETATTR); @@ -708,7 +708,7 @@ static int encode_getattr_one(struct xdr_stream *xdr, uint32_t bitmap) static int encode_getattr_two(struct xdr_stream *xdr, uint32_t bm0, uint32_t bm1) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(16); WRITE32(OP_GETATTR); @@ -740,7 +740,7 @@ static int encode_fs_locations(struct xdr_stream *xdr, const u32* bitmask) static int encode_getfh(struct xdr_stream *xdr) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(4); WRITE32(OP_GETFH); @@ -750,7 +750,7 @@ static int encode_getfh(struct xdr_stream *xdr) static int encode_link(struct xdr_stream *xdr, const struct qstr *name) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(8 + name->len); WRITE32(OP_LINK); @@ -780,7 +780,7 @@ static inline uint64_t nfs4_lock_length(struct file_lock *fl) */ static int encode_lock(struct xdr_stream *xdr, const struct nfs_lock_args *args) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(32); WRITE32(OP_LOCK); @@ -809,7 +809,7 @@ static int encode_lock(struct xdr_stream *xdr, const struct nfs_lock_args *args) static int encode_lockt(struct xdr_stream *xdr, const struct nfs_lockt_args *args) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(40); WRITE32(OP_LOCKT); @@ -825,7 +825,7 @@ static int encode_lockt(struct xdr_stream *xdr, const struct nfs_lockt_args *arg static int encode_locku(struct xdr_stream *xdr, const struct nfs_locku_args *args) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(44); WRITE32(OP_LOCKU); @@ -841,7 +841,7 @@ static int encode_locku(struct xdr_stream *xdr, const struct nfs_locku_args *arg static int encode_lookup(struct xdr_stream *xdr, const struct qstr *name) { int len = name->len; - uint32_t *p; + __be32 *p; RESERVE_SPACE(8 + len); WRITE32(OP_LOOKUP); @@ -853,7 +853,7 @@ static int encode_lookup(struct xdr_stream *xdr, const struct qstr *name) static void encode_share_access(struct xdr_stream *xdr, int open_flags) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(8); switch (open_flags & (FMODE_READ|FMODE_WRITE)) { @@ -874,7 +874,7 @@ static void encode_share_access(struct xdr_stream *xdr, int open_flags) static inline void encode_openhdr(struct xdr_stream *xdr, const struct nfs_openargs *arg) { - uint32_t *p; + __be32 *p; /* * opcode 4, seqid 4, share_access 4, share_deny 4, clientid 8, ownerlen 4, * owner 4 = 32 @@ -891,7 +891,7 @@ static inline void encode_openhdr(struct xdr_stream *xdr, const struct nfs_opena static inline void encode_createmode(struct xdr_stream *xdr, const struct nfs_openargs *arg) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(4); switch(arg->open_flags & O_EXCL) { @@ -907,7 +907,7 @@ static inline void encode_createmode(struct xdr_stream *xdr, const struct nfs_op static void encode_opentype(struct xdr_stream *xdr, const struct nfs_openargs *arg) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(4); switch (arg->open_flags & O_CREAT) { @@ -923,7 +923,7 @@ static void encode_opentype(struct xdr_stream *xdr, const struct nfs_openargs *a static inline void encode_delegation_type(struct xdr_stream *xdr, int delegation_type) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(4); switch (delegation_type) { @@ -943,7 +943,7 @@ static inline void encode_delegation_type(struct xdr_stream *xdr, int delegation static inline void encode_claim_null(struct xdr_stream *xdr, const struct qstr *name) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(4); WRITE32(NFS4_OPEN_CLAIM_NULL); @@ -952,7 +952,7 @@ static inline void encode_claim_null(struct xdr_stream *xdr, const struct qstr * static inline void encode_claim_previous(struct xdr_stream *xdr, int type) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(4); WRITE32(NFS4_OPEN_CLAIM_PREVIOUS); @@ -961,7 +961,7 @@ static inline void encode_claim_previous(struct xdr_stream *xdr, int type) static inline void encode_claim_delegate_cur(struct xdr_stream *xdr, const struct qstr *name, const nfs4_stateid *stateid) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(4+sizeof(stateid->data)); WRITE32(NFS4_OPEN_CLAIM_DELEGATE_CUR); @@ -991,7 +991,7 @@ static int encode_open(struct xdr_stream *xdr, const struct nfs_openargs *arg) static int encode_open_confirm(struct xdr_stream *xdr, const struct nfs_open_confirmargs *arg) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(8+sizeof(arg->stateid->data)); WRITE32(OP_OPEN_CONFIRM); @@ -1003,7 +1003,7 @@ static int encode_open_confirm(struct xdr_stream *xdr, const struct nfs_open_con static int encode_open_downgrade(struct xdr_stream *xdr, const struct nfs_closeargs *arg) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(8+sizeof(arg->stateid->data)); WRITE32(OP_OPEN_DOWNGRADE); @@ -1017,7 +1017,7 @@ static int encode_putfh(struct xdr_stream *xdr, const struct nfs_fh *fh) { int len = fh->size; - uint32_t *p; + __be32 *p; RESERVE_SPACE(8 + len); WRITE32(OP_PUTFH); @@ -1029,7 +1029,7 @@ encode_putfh(struct xdr_stream *xdr, const struct nfs_fh *fh) static int encode_putrootfh(struct xdr_stream *xdr) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(4); WRITE32(OP_PUTROOTFH); @@ -1040,7 +1040,7 @@ static int encode_putrootfh(struct xdr_stream *xdr) static void encode_stateid(struct xdr_stream *xdr, const struct nfs_open_context *ctx) { nfs4_stateid stateid; - uint32_t *p; + __be32 *p; RESERVE_SPACE(16); if (ctx->state != NULL) { @@ -1052,7 +1052,7 @@ static void encode_stateid(struct xdr_stream *xdr, const struct nfs_open_context static int encode_read(struct xdr_stream *xdr, const struct nfs_readargs *args) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(4); WRITE32(OP_READ); @@ -1074,7 +1074,7 @@ static int encode_readdir(struct xdr_stream *xdr, const struct nfs4_readdir_arg FATTR4_WORD1_MOUNTED_ON_FILEID, }; int replen; - uint32_t *p; + __be32 *p; RESERVE_SPACE(32+sizeof(nfs4_verifier)); WRITE32(OP_READDIR); @@ -1116,7 +1116,7 @@ static int encode_readlink(struct xdr_stream *xdr, const struct nfs4_readlink *r { struct rpc_auth *auth = req->rq_task->tk_auth; unsigned int replen; - uint32_t *p; + __be32 *p; RESERVE_SPACE(4); WRITE32(OP_READLINK); @@ -1134,7 +1134,7 @@ static int encode_readlink(struct xdr_stream *xdr, const struct nfs4_readlink *r static int encode_remove(struct xdr_stream *xdr, const struct qstr *name) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(8 + name->len); WRITE32(OP_REMOVE); @@ -1146,7 +1146,7 @@ static int encode_remove(struct xdr_stream *xdr, const struct qstr *name) static int encode_rename(struct xdr_stream *xdr, const struct qstr *oldname, const struct qstr *newname) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(8 + oldname->len); WRITE32(OP_RENAME); @@ -1162,7 +1162,7 @@ static int encode_rename(struct xdr_stream *xdr, const struct qstr *oldname, con static int encode_renew(struct xdr_stream *xdr, const struct nfs_client *client_stateid) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(12); WRITE32(OP_RENEW); @@ -1174,7 +1174,7 @@ static int encode_renew(struct xdr_stream *xdr, const struct nfs_client *client_ static int encode_restorefh(struct xdr_stream *xdr) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(4); WRITE32(OP_RESTOREFH); @@ -1185,7 +1185,7 @@ encode_restorefh(struct xdr_stream *xdr) static int encode_setacl(struct xdr_stream *xdr, struct nfs_setaclargs *arg) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(4+sizeof(zero_stateid.data)); WRITE32(OP_SETATTR); @@ -1204,7 +1204,7 @@ encode_setacl(struct xdr_stream *xdr, struct nfs_setaclargs *arg) static int encode_savefh(struct xdr_stream *xdr) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(4); WRITE32(OP_SAVEFH); @@ -1215,7 +1215,7 @@ encode_savefh(struct xdr_stream *xdr) static int encode_setattr(struct xdr_stream *xdr, const struct nfs_setattrargs *arg, const struct nfs_server *server) { int status; - uint32_t *p; + __be32 *p; RESERVE_SPACE(4+sizeof(arg->stateid.data)); WRITE32(OP_SETATTR); @@ -1229,7 +1229,7 @@ static int encode_setattr(struct xdr_stream *xdr, const struct nfs_setattrargs * static int encode_setclientid(struct xdr_stream *xdr, const struct nfs4_setclientid *setclientid) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(4 + sizeof(setclientid->sc_verifier->data)); WRITE32(OP_SETCLIENTID); @@ -1248,7 +1248,7 @@ static int encode_setclientid(struct xdr_stream *xdr, const struct nfs4_setclien static int encode_setclientid_confirm(struct xdr_stream *xdr, const struct nfs_client *client_state) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(12 + sizeof(client_state->cl_confirm.data)); WRITE32(OP_SETCLIENTID_CONFIRM); @@ -1260,7 +1260,7 @@ static int encode_setclientid_confirm(struct xdr_stream *xdr, const struct nfs_c static int encode_write(struct xdr_stream *xdr, const struct nfs_writeargs *args) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(4); WRITE32(OP_WRITE); @@ -1279,7 +1279,7 @@ static int encode_write(struct xdr_stream *xdr, const struct nfs_writeargs *args static int encode_delegreturn(struct xdr_stream *xdr, const nfs4_stateid *stateid) { - uint32_t *p; + __be32 *p; RESERVE_SPACE(20); @@ -1295,7 +1295,7 @@ static int encode_delegreturn(struct xdr_stream *xdr, const nfs4_stateid *statei /* * Encode an ACCESS request */ -static int nfs4_xdr_enc_access(struct rpc_rqst *req, uint32_t *p, const struct nfs4_accessargs *args) +static int nfs4_xdr_enc_access(struct rpc_rqst *req, __be32 *p, const struct nfs4_accessargs *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1313,7 +1313,7 @@ static int nfs4_xdr_enc_access(struct rpc_rqst *req, uint32_t *p, const struct n /* * Encode LOOKUP request */ -static int nfs4_xdr_enc_lookup(struct rpc_rqst *req, uint32_t *p, const struct nfs4_lookup_arg *args) +static int nfs4_xdr_enc_lookup(struct rpc_rqst *req, __be32 *p, const struct nfs4_lookup_arg *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1337,7 +1337,7 @@ out: /* * Encode LOOKUP_ROOT request */ -static int nfs4_xdr_enc_lookup_root(struct rpc_rqst *req, uint32_t *p, const struct nfs4_lookup_root_arg *args) +static int nfs4_xdr_enc_lookup_root(struct rpc_rqst *req, __be32 *p, const struct nfs4_lookup_root_arg *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1358,7 +1358,7 @@ out: /* * Encode REMOVE request */ -static int nfs4_xdr_enc_remove(struct rpc_rqst *req, uint32_t *p, const struct nfs4_remove_arg *args) +static int nfs4_xdr_enc_remove(struct rpc_rqst *req, __be32 *p, const struct nfs4_remove_arg *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1380,7 +1380,7 @@ out: /* * Encode RENAME request */ -static int nfs4_xdr_enc_rename(struct rpc_rqst *req, uint32_t *p, const struct nfs4_rename_arg *args) +static int nfs4_xdr_enc_rename(struct rpc_rqst *req, __be32 *p, const struct nfs4_rename_arg *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1410,7 +1410,7 @@ out: /* * Encode LINK request */ -static int nfs4_xdr_enc_link(struct rpc_rqst *req, uint32_t *p, const struct nfs4_link_arg *args) +static int nfs4_xdr_enc_link(struct rpc_rqst *req, __be32 *p, const struct nfs4_link_arg *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1440,7 +1440,7 @@ out: /* * Encode CREATE request */ -static int nfs4_xdr_enc_create(struct rpc_rqst *req, uint32_t *p, const struct nfs4_create_arg *args) +static int nfs4_xdr_enc_create(struct rpc_rqst *req, __be32 *p, const struct nfs4_create_arg *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1470,7 +1470,7 @@ out: /* * Encode SYMLINK request */ -static int nfs4_xdr_enc_symlink(struct rpc_rqst *req, uint32_t *p, const struct nfs4_create_arg *args) +static int nfs4_xdr_enc_symlink(struct rpc_rqst *req, __be32 *p, const struct nfs4_create_arg *args) { return nfs4_xdr_enc_create(req, p, args); } @@ -1478,7 +1478,7 @@ static int nfs4_xdr_enc_symlink(struct rpc_rqst *req, uint32_t *p, const struct /* * Encode GETATTR request */ -static int nfs4_xdr_enc_getattr(struct rpc_rqst *req, uint32_t *p, const struct nfs4_getattr_arg *args) +static int nfs4_xdr_enc_getattr(struct rpc_rqst *req, __be32 *p, const struct nfs4_getattr_arg *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1496,7 +1496,7 @@ static int nfs4_xdr_enc_getattr(struct rpc_rqst *req, uint32_t *p, const struct /* * Encode a CLOSE request */ -static int nfs4_xdr_enc_close(struct rpc_rqst *req, uint32_t *p, struct nfs_closeargs *args) +static int nfs4_xdr_enc_close(struct rpc_rqst *req, __be32 *p, struct nfs_closeargs *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1520,7 +1520,7 @@ out: /* * Encode an OPEN request */ -static int nfs4_xdr_enc_open(struct rpc_rqst *req, uint32_t *p, struct nfs_openargs *args) +static int nfs4_xdr_enc_open(struct rpc_rqst *req, __be32 *p, struct nfs_openargs *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1556,7 +1556,7 @@ out: /* * Encode an OPEN_CONFIRM request */ -static int nfs4_xdr_enc_open_confirm(struct rpc_rqst *req, uint32_t *p, struct nfs_open_confirmargs *args) +static int nfs4_xdr_enc_open_confirm(struct rpc_rqst *req, __be32 *p, struct nfs_open_confirmargs *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1577,7 +1577,7 @@ out: /* * Encode an OPEN request with no attributes. */ -static int nfs4_xdr_enc_open_noattr(struct rpc_rqst *req, uint32_t *p, struct nfs_openargs *args) +static int nfs4_xdr_enc_open_noattr(struct rpc_rqst *req, __be32 *p, struct nfs_openargs *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1601,7 +1601,7 @@ out: /* * Encode an OPEN_DOWNGRADE request */ -static int nfs4_xdr_enc_open_downgrade(struct rpc_rqst *req, uint32_t *p, struct nfs_closeargs *args) +static int nfs4_xdr_enc_open_downgrade(struct rpc_rqst *req, __be32 *p, struct nfs_closeargs *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1625,7 +1625,7 @@ out: /* * Encode a LOCK request */ -static int nfs4_xdr_enc_lock(struct rpc_rqst *req, uint32_t *p, struct nfs_lock_args *args) +static int nfs4_xdr_enc_lock(struct rpc_rqst *req, __be32 *p, struct nfs_lock_args *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1646,7 +1646,7 @@ out: /* * Encode a LOCKT request */ -static int nfs4_xdr_enc_lockt(struct rpc_rqst *req, uint32_t *p, struct nfs_lockt_args *args) +static int nfs4_xdr_enc_lockt(struct rpc_rqst *req, __be32 *p, struct nfs_lockt_args *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1667,7 +1667,7 @@ out: /* * Encode a LOCKU request */ -static int nfs4_xdr_enc_locku(struct rpc_rqst *req, uint32_t *p, struct nfs_locku_args *args) +static int nfs4_xdr_enc_locku(struct rpc_rqst *req, __be32 *p, struct nfs_locku_args *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1688,7 +1688,7 @@ out: /* * Encode a READLINK request */ -static int nfs4_xdr_enc_readlink(struct rpc_rqst *req, uint32_t *p, const struct nfs4_readlink *args) +static int nfs4_xdr_enc_readlink(struct rpc_rqst *req, __be32 *p, const struct nfs4_readlink *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1709,7 +1709,7 @@ out: /* * Encode a READDIR request */ -static int nfs4_xdr_enc_readdir(struct rpc_rqst *req, uint32_t *p, const struct nfs4_readdir_arg *args) +static int nfs4_xdr_enc_readdir(struct rpc_rqst *req, __be32 *p, const struct nfs4_readdir_arg *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1730,7 +1730,7 @@ out: /* * Encode a READ request */ -static int nfs4_xdr_enc_read(struct rpc_rqst *req, uint32_t *p, struct nfs_readargs *args) +static int nfs4_xdr_enc_read(struct rpc_rqst *req, __be32 *p, struct nfs_readargs *args) { struct rpc_auth *auth = req->rq_task->tk_auth; struct xdr_stream xdr; @@ -1762,7 +1762,7 @@ out: /* * Encode an SETATTR request */ -static int nfs4_xdr_enc_setattr(struct rpc_rqst *req, uint32_t *p, struct nfs_setattrargs *args) +static int nfs4_xdr_enc_setattr(struct rpc_rqst *req, __be32 *p, struct nfs_setattrargs *args) { struct xdr_stream xdr; @@ -1788,7 +1788,7 @@ out: * Encode a GETACL request */ static int -nfs4_xdr_enc_getacl(struct rpc_rqst *req, uint32_t *p, +nfs4_xdr_enc_getacl(struct rpc_rqst *req, __be32 *p, struct nfs_getaclargs *args) { struct xdr_stream xdr; @@ -1815,7 +1815,7 @@ out: /* * Encode a WRITE request */ -static int nfs4_xdr_enc_write(struct rpc_rqst *req, uint32_t *p, struct nfs_writeargs *args) +static int nfs4_xdr_enc_write(struct rpc_rqst *req, __be32 *p, struct nfs_writeargs *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1839,7 +1839,7 @@ out: /* * a COMMIT request */ -static int nfs4_xdr_enc_commit(struct rpc_rqst *req, uint32_t *p, struct nfs_writeargs *args) +static int nfs4_xdr_enc_commit(struct rpc_rqst *req, __be32 *p, struct nfs_writeargs *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1863,7 +1863,7 @@ out: /* * FSINFO request */ -static int nfs4_xdr_enc_fsinfo(struct rpc_rqst *req, uint32_t *p, struct nfs4_fsinfo_arg *args) +static int nfs4_xdr_enc_fsinfo(struct rpc_rqst *req, __be32 *p, struct nfs4_fsinfo_arg *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1882,7 +1882,7 @@ static int nfs4_xdr_enc_fsinfo(struct rpc_rqst *req, uint32_t *p, struct nfs4_fs /* * a PATHCONF request */ -static int nfs4_xdr_enc_pathconf(struct rpc_rqst *req, uint32_t *p, const struct nfs4_pathconf_arg *args) +static int nfs4_xdr_enc_pathconf(struct rpc_rqst *req, __be32 *p, const struct nfs4_pathconf_arg *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1902,7 +1902,7 @@ static int nfs4_xdr_enc_pathconf(struct rpc_rqst *req, uint32_t *p, const struct /* * a STATFS request */ -static int nfs4_xdr_enc_statfs(struct rpc_rqst *req, uint32_t *p, const struct nfs4_statfs_arg *args) +static int nfs4_xdr_enc_statfs(struct rpc_rqst *req, __be32 *p, const struct nfs4_statfs_arg *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1923,7 +1923,7 @@ static int nfs4_xdr_enc_statfs(struct rpc_rqst *req, uint32_t *p, const struct n /* * GETATTR_BITMAP request */ -static int nfs4_xdr_enc_server_caps(struct rpc_rqst *req, uint32_t *p, const struct nfs_fh *fhandle) +static int nfs4_xdr_enc_server_caps(struct rpc_rqst *req, __be32 *p, const struct nfs_fh *fhandle) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1945,7 +1945,7 @@ static int nfs4_xdr_enc_server_caps(struct rpc_rqst *req, uint32_t *p, const str /* * a RENEW request */ -static int nfs4_xdr_enc_renew(struct rpc_rqst *req, uint32_t *p, struct nfs_client *clp) +static int nfs4_xdr_enc_renew(struct rpc_rqst *req, __be32 *p, struct nfs_client *clp) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1960,7 +1960,7 @@ static int nfs4_xdr_enc_renew(struct rpc_rqst *req, uint32_t *p, struct nfs_clie /* * a SETCLIENTID request */ -static int nfs4_xdr_enc_setclientid(struct rpc_rqst *req, uint32_t *p, struct nfs4_setclientid *sc) +static int nfs4_xdr_enc_setclientid(struct rpc_rqst *req, __be32 *p, struct nfs4_setclientid *sc) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1975,7 +1975,7 @@ static int nfs4_xdr_enc_setclientid(struct rpc_rqst *req, uint32_t *p, struct nf /* * a SETCLIENTID_CONFIRM request */ -static int nfs4_xdr_enc_setclientid_confirm(struct rpc_rqst *req, uint32_t *p, struct nfs_client *clp) +static int nfs4_xdr_enc_setclientid_confirm(struct rpc_rqst *req, __be32 *p, struct nfs_client *clp) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -1997,7 +1997,7 @@ static int nfs4_xdr_enc_setclientid_confirm(struct rpc_rqst *req, uint32_t *p, s /* * DELEGRETURN request */ -static int nfs4_xdr_enc_delegreturn(struct rpc_rqst *req, uint32_t *p, const struct nfs4_delegreturnargs *args) +static int nfs4_xdr_enc_delegreturn(struct rpc_rqst *req, __be32 *p, const struct nfs4_delegreturnargs *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -2021,7 +2021,7 @@ out: /* * Encode FS_LOCATIONS request */ -static int nfs4_xdr_enc_fs_locations(struct rpc_rqst *req, uint32_t *p, struct nfs4_fs_locations_arg *args) +static int nfs4_xdr_enc_fs_locations(struct rpc_rqst *req, __be32 *p, struct nfs4_fs_locations_arg *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -2086,7 +2086,7 @@ out: static int decode_opaque_inline(struct xdr_stream *xdr, unsigned int *len, char **string) { - uint32_t *p; + __be32 *p; READ_BUF(4); READ32(*len); @@ -2097,7 +2097,7 @@ static int decode_opaque_inline(struct xdr_stream *xdr, unsigned int *len, char static int decode_compound_hdr(struct xdr_stream *xdr, struct compound_hdr *hdr) { - uint32_t *p; + __be32 *p; READ_BUF(8); READ32(hdr->status); @@ -2112,7 +2112,7 @@ static int decode_compound_hdr(struct xdr_stream *xdr, struct compound_hdr *hdr) static int decode_op_hdr(struct xdr_stream *xdr, enum nfs_opnum4 expected) { - uint32_t *p; + __be32 *p; uint32_t opnum; int32_t nfserr; @@ -2134,7 +2134,7 @@ static int decode_op_hdr(struct xdr_stream *xdr, enum nfs_opnum4 expected) /* Dummy routine */ static int decode_ace(struct xdr_stream *xdr, void *ace, struct nfs_client *clp) { - uint32_t *p; + __be32 *p; unsigned int strlen; char *str; @@ -2144,7 +2144,8 @@ static int decode_ace(struct xdr_stream *xdr, void *ace, struct nfs_client *clp) static int decode_attr_bitmap(struct xdr_stream *xdr, uint32_t *bitmap) { - uint32_t bmlen, *p; + uint32_t bmlen; + __be32 *p; READ_BUF(4); READ32(bmlen); @@ -2159,9 +2160,9 @@ static int decode_attr_bitmap(struct xdr_stream *xdr, uint32_t *bitmap) return 0; } -static inline int decode_attr_length(struct xdr_stream *xdr, uint32_t *attrlen, uint32_t **savep) +static inline int decode_attr_length(struct xdr_stream *xdr, uint32_t *attrlen, __be32 **savep) { - uint32_t *p; + __be32 *p; READ_BUF(4); READ32(*attrlen); @@ -2182,7 +2183,7 @@ static int decode_attr_supported(struct xdr_stream *xdr, uint32_t *bitmap, uint3 static int decode_attr_type(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *type) { - uint32_t *p; + __be32 *p; *type = 0; if (unlikely(bitmap[0] & (FATTR4_WORD0_TYPE - 1U))) @@ -2202,7 +2203,7 @@ static int decode_attr_type(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t * static int decode_attr_change(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *change) { - uint32_t *p; + __be32 *p; *change = 0; if (unlikely(bitmap[0] & (FATTR4_WORD0_CHANGE - 1U))) @@ -2219,7 +2220,7 @@ static int decode_attr_change(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t static int decode_attr_size(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *size) { - uint32_t *p; + __be32 *p; *size = 0; if (unlikely(bitmap[0] & (FATTR4_WORD0_SIZE - 1U))) @@ -2235,7 +2236,7 @@ static int decode_attr_size(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t * static int decode_attr_link_support(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *res) { - uint32_t *p; + __be32 *p; *res = 0; if (unlikely(bitmap[0] & (FATTR4_WORD0_LINK_SUPPORT - 1U))) @@ -2251,7 +2252,7 @@ static int decode_attr_link_support(struct xdr_stream *xdr, uint32_t *bitmap, ui static int decode_attr_symlink_support(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *res) { - uint32_t *p; + __be32 *p; *res = 0; if (unlikely(bitmap[0] & (FATTR4_WORD0_SYMLINK_SUPPORT - 1U))) @@ -2267,7 +2268,7 @@ static int decode_attr_symlink_support(struct xdr_stream *xdr, uint32_t *bitmap, static int decode_attr_fsid(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs_fsid *fsid) { - uint32_t *p; + __be32 *p; fsid->major = 0; fsid->minor = 0; @@ -2287,7 +2288,7 @@ static int decode_attr_fsid(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs static int decode_attr_lease_time(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *res) { - uint32_t *p; + __be32 *p; *res = 60; if (unlikely(bitmap[0] & (FATTR4_WORD0_LEASE_TIME - 1U))) @@ -2303,7 +2304,7 @@ static int decode_attr_lease_time(struct xdr_stream *xdr, uint32_t *bitmap, uint static int decode_attr_aclsupport(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *res) { - uint32_t *p; + __be32 *p; *res = ACL4_SUPPORT_ALLOW_ACL|ACL4_SUPPORT_DENY_ACL; if (unlikely(bitmap[0] & (FATTR4_WORD0_ACLSUPPORT - 1U))) @@ -2319,7 +2320,7 @@ static int decode_attr_aclsupport(struct xdr_stream *xdr, uint32_t *bitmap, uint static int decode_attr_fileid(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *fileid) { - uint32_t *p; + __be32 *p; *fileid = 0; if (unlikely(bitmap[0] & (FATTR4_WORD0_FILEID - 1U))) @@ -2335,7 +2336,7 @@ static int decode_attr_fileid(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t static int decode_attr_mounted_on_fileid(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *fileid) { - uint32_t *p; + __be32 *p; *fileid = 0; if (unlikely(bitmap[1] & (FATTR4_WORD1_MOUNTED_ON_FILEID - 1U))) @@ -2351,7 +2352,7 @@ static int decode_attr_mounted_on_fileid(struct xdr_stream *xdr, uint32_t *bitma static int decode_attr_files_avail(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *res) { - uint32_t *p; + __be32 *p; int status = 0; *res = 0; @@ -2368,7 +2369,7 @@ static int decode_attr_files_avail(struct xdr_stream *xdr, uint32_t *bitmap, uin static int decode_attr_files_free(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *res) { - uint32_t *p; + __be32 *p; int status = 0; *res = 0; @@ -2385,7 +2386,7 @@ static int decode_attr_files_free(struct xdr_stream *xdr, uint32_t *bitmap, uint static int decode_attr_files_total(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *res) { - uint32_t *p; + __be32 *p; int status = 0; *res = 0; @@ -2403,7 +2404,7 @@ static int decode_attr_files_total(struct xdr_stream *xdr, uint32_t *bitmap, uin static int decode_pathname(struct xdr_stream *xdr, struct nfs4_pathname *path) { int n; - uint32_t *p; + __be32 *p; int status = 0; READ_BUF(4); @@ -2448,7 +2449,7 @@ out_eio: static int decode_attr_fs_locations(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs4_fs_locations *res) { int n; - uint32_t *p; + __be32 *p; int status = -EIO; if (unlikely(bitmap[0] & (FATTR4_WORD0_FS_LOCATIONS -1U))) @@ -2512,7 +2513,7 @@ out_eio: static int decode_attr_maxfilesize(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *res) { - uint32_t *p; + __be32 *p; int status = 0; *res = 0; @@ -2529,7 +2530,7 @@ static int decode_attr_maxfilesize(struct xdr_stream *xdr, uint32_t *bitmap, uin static int decode_attr_maxlink(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *maxlink) { - uint32_t *p; + __be32 *p; int status = 0; *maxlink = 1; @@ -2546,7 +2547,7 @@ static int decode_attr_maxlink(struct xdr_stream *xdr, uint32_t *bitmap, uint32_ static int decode_attr_maxname(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *maxname) { - uint32_t *p; + __be32 *p; int status = 0; *maxname = 1024; @@ -2563,7 +2564,7 @@ static int decode_attr_maxname(struct xdr_stream *xdr, uint32_t *bitmap, uint32_ static int decode_attr_maxread(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *res) { - uint32_t *p; + __be32 *p; int status = 0; *res = 1024; @@ -2584,7 +2585,7 @@ static int decode_attr_maxread(struct xdr_stream *xdr, uint32_t *bitmap, uint32_ static int decode_attr_maxwrite(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *res) { - uint32_t *p; + __be32 *p; int status = 0; *res = 1024; @@ -2605,7 +2606,7 @@ static int decode_attr_maxwrite(struct xdr_stream *xdr, uint32_t *bitmap, uint32 static int decode_attr_mode(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *mode) { - uint32_t *p; + __be32 *p; *mode = 0; if (unlikely(bitmap[1] & (FATTR4_WORD1_MODE - 1U))) @@ -2622,7 +2623,7 @@ static int decode_attr_mode(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t * static int decode_attr_nlink(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t *nlink) { - uint32_t *p; + __be32 *p; *nlink = 1; if (unlikely(bitmap[1] & (FATTR4_WORD1_NUMLINKS - 1U))) @@ -2638,7 +2639,8 @@ static int decode_attr_nlink(struct xdr_stream *xdr, uint32_t *bitmap, uint32_t static int decode_attr_owner(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs_client *clp, int32_t *uid) { - uint32_t len, *p; + uint32_t len; + __be32 *p; *uid = -2; if (unlikely(bitmap[1] & (FATTR4_WORD1_OWNER - 1U))) @@ -2662,7 +2664,8 @@ static int decode_attr_owner(struct xdr_stream *xdr, uint32_t *bitmap, struct nf static int decode_attr_group(struct xdr_stream *xdr, uint32_t *bitmap, struct nfs_client *clp, int32_t *gid) { - uint32_t len, *p; + uint32_t len; + __be32 *p; *gid = -2; if (unlikely(bitmap[1] & (FATTR4_WORD1_OWNER_GROUP - 1U))) @@ -2686,7 +2689,8 @@ static int decode_attr_group(struct xdr_stream *xdr, uint32_t *bitmap, struct nf static int decode_attr_rdev(struct xdr_stream *xdr, uint32_t *bitmap, dev_t *rdev) { - uint32_t major = 0, minor = 0, *p; + uint32_t major = 0, minor = 0; + __be32 *p; *rdev = MKDEV(0,0); if (unlikely(bitmap[1] & (FATTR4_WORD1_RAWDEV - 1U))) @@ -2708,7 +2712,7 @@ static int decode_attr_rdev(struct xdr_stream *xdr, uint32_t *bitmap, dev_t *rde static int decode_attr_space_avail(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *res) { - uint32_t *p; + __be32 *p; int status = 0; *res = 0; @@ -2725,7 +2729,7 @@ static int decode_attr_space_avail(struct xdr_stream *xdr, uint32_t *bitmap, uin static int decode_attr_space_free(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *res) { - uint32_t *p; + __be32 *p; int status = 0; *res = 0; @@ -2742,7 +2746,7 @@ static int decode_attr_space_free(struct xdr_stream *xdr, uint32_t *bitmap, uint static int decode_attr_space_total(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *res) { - uint32_t *p; + __be32 *p; int status = 0; *res = 0; @@ -2759,7 +2763,7 @@ static int decode_attr_space_total(struct xdr_stream *xdr, uint32_t *bitmap, uin static int decode_attr_space_used(struct xdr_stream *xdr, uint32_t *bitmap, uint64_t *used) { - uint32_t *p; + __be32 *p; *used = 0; if (unlikely(bitmap[1] & (FATTR4_WORD1_SPACE_USED - 1U))) @@ -2776,7 +2780,7 @@ static int decode_attr_space_used(struct xdr_stream *xdr, uint32_t *bitmap, uint static int decode_attr_time(struct xdr_stream *xdr, struct timespec *time) { - uint32_t *p; + __be32 *p; uint64_t sec; uint32_t nsec; @@ -2836,7 +2840,7 @@ static int decode_attr_time_modify(struct xdr_stream *xdr, uint32_t *bitmap, str return status; } -static int verify_attr_len(struct xdr_stream *xdr, uint32_t *savep, uint32_t attrlen) +static int verify_attr_len(struct xdr_stream *xdr, __be32 *savep, uint32_t attrlen) { unsigned int attrwords = XDR_QUADLEN(attrlen); unsigned int nwords = xdr->p - savep; @@ -2854,7 +2858,7 @@ static int verify_attr_len(struct xdr_stream *xdr, uint32_t *savep, uint32_t att static int decode_change_info(struct xdr_stream *xdr, struct nfs4_change_info *cinfo) { - uint32_t *p; + __be32 *p; READ_BUF(20); READ32(cinfo->atomic); @@ -2865,7 +2869,7 @@ static int decode_change_info(struct xdr_stream *xdr, struct nfs4_change_info *c static int decode_access(struct xdr_stream *xdr, struct nfs4_accessres *access) { - uint32_t *p; + __be32 *p; uint32_t supp, acc; int status; @@ -2882,7 +2886,7 @@ static int decode_access(struct xdr_stream *xdr, struct nfs4_accessres *access) static int decode_close(struct xdr_stream *xdr, struct nfs_closeres *res) { - uint32_t *p; + __be32 *p; int status; status = decode_op_hdr(xdr, OP_CLOSE); @@ -2895,7 +2899,7 @@ static int decode_close(struct xdr_stream *xdr, struct nfs_closeres *res) static int decode_commit(struct xdr_stream *xdr, struct nfs_writeres *res) { - uint32_t *p; + __be32 *p; int status; status = decode_op_hdr(xdr, OP_COMMIT); @@ -2908,7 +2912,7 @@ static int decode_commit(struct xdr_stream *xdr, struct nfs_writeres *res) static int decode_create(struct xdr_stream *xdr, struct nfs4_change_info *cinfo) { - uint32_t *p; + __be32 *p; uint32_t bmlen; int status; @@ -2925,7 +2929,7 @@ static int decode_create(struct xdr_stream *xdr, struct nfs4_change_info *cinfo) static int decode_server_caps(struct xdr_stream *xdr, struct nfs4_server_caps_res *res) { - uint32_t *savep; + __be32 *savep; uint32_t attrlen, bitmap[2] = {0}; int status; @@ -2952,7 +2956,7 @@ xdr_error: static int decode_statfs(struct xdr_stream *xdr, struct nfs_fsstat *fsstat) { - uint32_t *savep; + __be32 *savep; uint32_t attrlen, bitmap[2] = {0}; int status; @@ -2985,7 +2989,7 @@ xdr_error: static int decode_pathconf(struct xdr_stream *xdr, struct nfs_pathconf *pathconf) { - uint32_t *savep; + __be32 *savep; uint32_t attrlen, bitmap[2] = {0}; int status; @@ -3010,7 +3014,7 @@ xdr_error: static int decode_getfattr(struct xdr_stream *xdr, struct nfs_fattr *fattr, const struct nfs_server *server) { - uint32_t *savep; + __be32 *savep; uint32_t attrlen, bitmap[2] = {0}, type; @@ -3079,7 +3083,7 @@ xdr_error: static int decode_fsinfo(struct xdr_stream *xdr, struct nfs_fsinfo *fsinfo) { - uint32_t *savep; + __be32 *savep; uint32_t attrlen, bitmap[2]; int status; @@ -3111,7 +3115,7 @@ xdr_error: static int decode_getfh(struct xdr_stream *xdr, struct nfs_fh *fh) { - uint32_t *p; + __be32 *p; uint32_t len; int status; @@ -3147,7 +3151,7 @@ static int decode_link(struct xdr_stream *xdr, struct nfs4_change_info *cinfo) static int decode_lock_denied (struct xdr_stream *xdr, struct file_lock *fl) { uint64_t offset, length, clientid; - uint32_t *p; + __be32 *p; uint32_t namelen, type; READ_BUF(32); @@ -3172,7 +3176,7 @@ static int decode_lock_denied (struct xdr_stream *xdr, struct file_lock *fl) static int decode_lock(struct xdr_stream *xdr, struct nfs_lock_res *res) { - uint32_t *p; + __be32 *p; int status; status = decode_op_hdr(xdr, OP_LOCK); @@ -3195,7 +3199,7 @@ static int decode_lockt(struct xdr_stream *xdr, struct nfs_lockt_res *res) static int decode_locku(struct xdr_stream *xdr, struct nfs_locku_res *res) { - uint32_t *p; + __be32 *p; int status; status = decode_op_hdr(xdr, OP_LOCKU); @@ -3214,7 +3218,7 @@ static int decode_lookup(struct xdr_stream *xdr) /* This is too sick! */ static int decode_space_limit(struct xdr_stream *xdr, u64 *maxsize) { - uint32_t *p; + __be32 *p; uint32_t limit_type, nblocks, blocksize; READ_BUF(12); @@ -3233,7 +3237,7 @@ static int decode_space_limit(struct xdr_stream *xdr, u64 *maxsize) static int decode_delegation(struct xdr_stream *xdr, struct nfs_openres *res) { - uint32_t *p; + __be32 *p; uint32_t delegation_type; READ_BUF(4); @@ -3259,7 +3263,7 @@ static int decode_delegation(struct xdr_stream *xdr, struct nfs_openres *res) static int decode_open(struct xdr_stream *xdr, struct nfs_openres *res) { - uint32_t *p; + __be32 *p; uint32_t bmlen; int status; @@ -3287,7 +3291,7 @@ xdr_error: static int decode_open_confirm(struct xdr_stream *xdr, struct nfs_open_confirmres *res) { - uint32_t *p; + __be32 *p; int status; status = decode_op_hdr(xdr, OP_OPEN_CONFIRM); @@ -3300,7 +3304,7 @@ static int decode_open_confirm(struct xdr_stream *xdr, struct nfs_open_confirmre static int decode_open_downgrade(struct xdr_stream *xdr, struct nfs_closeres *res) { - uint32_t *p; + __be32 *p; int status; status = decode_op_hdr(xdr, OP_OPEN_DOWNGRADE); @@ -3324,7 +3328,7 @@ static int decode_putrootfh(struct xdr_stream *xdr) static int decode_read(struct xdr_stream *xdr, struct rpc_rqst *req, struct nfs_readres *res) { struct kvec *iov = req->rq_rcv_buf.head; - uint32_t *p; + __be32 *p; uint32_t count, eof, recvd, hdrlen; int status; @@ -3354,7 +3358,7 @@ static int decode_readdir(struct xdr_stream *xdr, struct rpc_rqst *req, struct n struct page *page = *rcvbuf->pages; struct kvec *iov = rcvbuf->head; unsigned int nr, pglen = rcvbuf->page_len; - uint32_t *end, *entry, *p, *kaddr; + __be32 *end, *entry, *p, *kaddr; uint32_t len, attrlen, xlen; int hdrlen, recvd, status; @@ -3376,7 +3380,7 @@ static int decode_readdir(struct xdr_stream *xdr, struct rpc_rqst *req, struct n xdr_read_pages(xdr, pglen); BUG_ON(pglen + readdir->pgbase > PAGE_CACHE_SIZE); - kaddr = p = (uint32_t *) kmap_atomic(page, KM_USER0); + kaddr = p = kmap_atomic(page, KM_USER0); end = p + ((pglen + readdir->pgbase) >> 2); entry = p; for (nr = 0; *p++; nr++) { @@ -3428,7 +3432,7 @@ static int decode_readlink(struct xdr_stream *xdr, struct rpc_rqst *req) struct xdr_buf *rcvbuf = &req->rq_rcv_buf; struct kvec *iov = rcvbuf->head; int hdrlen, len, recvd; - uint32_t *p; + __be32 *p; char *kaddr; int status; @@ -3505,7 +3509,7 @@ decode_restorefh(struct xdr_stream *xdr) static int decode_getacl(struct xdr_stream *xdr, struct rpc_rqst *req, size_t *acl_len) { - uint32_t *savep; + __be32 *savep; uint32_t attrlen, bitmap[2] = {0}; struct kvec *iov = req->rq_rcv_buf.head; @@ -3551,7 +3555,7 @@ decode_savefh(struct xdr_stream *xdr) static int decode_setattr(struct xdr_stream *xdr, struct nfs_setattrres *res) { - uint32_t *p; + __be32 *p; uint32_t bmlen; int status; @@ -3567,7 +3571,7 @@ static int decode_setattr(struct xdr_stream *xdr, struct nfs_setattrres *res) static int decode_setclientid(struct xdr_stream *xdr, struct nfs_client *clp) { - uint32_t *p; + __be32 *p; uint32_t opnum; int32_t nfserr; @@ -3610,7 +3614,7 @@ static int decode_setclientid_confirm(struct xdr_stream *xdr) static int decode_write(struct xdr_stream *xdr, struct nfs_writeres *res) { - uint32_t *p; + __be32 *p; int status; status = decode_op_hdr(xdr, OP_WRITE); @@ -3632,7 +3636,7 @@ static int decode_delegreturn(struct xdr_stream *xdr) /* * Decode OPEN_DOWNGRADE response */ -static int nfs4_xdr_dec_open_downgrade(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_closeres *res) +static int nfs4_xdr_dec_open_downgrade(struct rpc_rqst *rqstp, __be32 *p, struct nfs_closeres *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -3660,7 +3664,7 @@ out: /* * Decode ACCESS response */ -static int nfs4_xdr_dec_access(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_accessres *res) +static int nfs4_xdr_dec_access(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_accessres *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -3678,7 +3682,7 @@ out: /* * Decode LOOKUP response */ -static int nfs4_xdr_dec_lookup(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_lookup_res *res) +static int nfs4_xdr_dec_lookup(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_lookup_res *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -3701,7 +3705,7 @@ out: /* * Decode LOOKUP_ROOT response */ -static int nfs4_xdr_dec_lookup_root(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_lookup_res *res) +static int nfs4_xdr_dec_lookup_root(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_lookup_res *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -3721,7 +3725,7 @@ out: /* * Decode REMOVE response */ -static int nfs4_xdr_dec_remove(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_remove_res *res) +static int nfs4_xdr_dec_remove(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_remove_res *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -3742,7 +3746,7 @@ out: /* * Decode RENAME response */ -static int nfs4_xdr_dec_rename(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_rename_res *res) +static int nfs4_xdr_dec_rename(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_rename_res *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -3772,7 +3776,7 @@ out: /* * Decode LINK response */ -static int nfs4_xdr_dec_link(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_link_res *res) +static int nfs4_xdr_dec_link(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_link_res *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -3805,7 +3809,7 @@ out: /* * Decode CREATE response */ -static int nfs4_xdr_dec_create(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_create_res *res) +static int nfs4_xdr_dec_create(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_create_res *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -3834,7 +3838,7 @@ out: /* * Decode SYMLINK response */ -static int nfs4_xdr_dec_symlink(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_create_res *res) +static int nfs4_xdr_dec_symlink(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_create_res *res) { return nfs4_xdr_dec_create(rqstp, p, res); } @@ -3842,7 +3846,7 @@ static int nfs4_xdr_dec_symlink(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4 /* * Decode GETATTR response */ -static int nfs4_xdr_dec_getattr(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_getattr_res *res) +static int nfs4_xdr_dec_getattr(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_getattr_res *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -3865,7 +3869,7 @@ out: * Encode an SETACL request */ static int -nfs4_xdr_enc_setacl(struct rpc_rqst *req, uint32_t *p, struct nfs_setaclargs *args) +nfs4_xdr_enc_setacl(struct rpc_rqst *req, __be32 *p, struct nfs_setaclargs *args) { struct xdr_stream xdr; struct compound_hdr hdr = { @@ -3886,7 +3890,7 @@ out: * Decode SETACL response */ static int -nfs4_xdr_dec_setacl(struct rpc_rqst *rqstp, uint32_t *p, void *res) +nfs4_xdr_dec_setacl(struct rpc_rqst *rqstp, __be32 *p, void *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -3908,7 +3912,7 @@ out: * Decode GETACL response */ static int -nfs4_xdr_dec_getacl(struct rpc_rqst *rqstp, uint32_t *p, size_t *acl_len) +nfs4_xdr_dec_getacl(struct rpc_rqst *rqstp, __be32 *p, size_t *acl_len) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -3930,7 +3934,7 @@ out: /* * Decode CLOSE response */ -static int nfs4_xdr_dec_close(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_closeres *res) +static int nfs4_xdr_dec_close(struct rpc_rqst *rqstp, __be32 *p, struct nfs_closeres *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -3960,7 +3964,7 @@ out: /* * Decode OPEN response */ -static int nfs4_xdr_dec_open(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_openres *res) +static int nfs4_xdr_dec_open(struct rpc_rqst *rqstp, __be32 *p, struct nfs_openres *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -3994,7 +3998,7 @@ out: /* * Decode OPEN_CONFIRM response */ -static int nfs4_xdr_dec_open_confirm(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_open_confirmres *res) +static int nfs4_xdr_dec_open_confirm(struct rpc_rqst *rqstp, __be32 *p, struct nfs_open_confirmres *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4015,7 +4019,7 @@ out: /* * Decode OPEN response */ -static int nfs4_xdr_dec_open_noattr(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_openres *res) +static int nfs4_xdr_dec_open_noattr(struct rpc_rqst *rqstp, __be32 *p, struct nfs_openres *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4039,7 +4043,7 @@ out: /* * Decode SETATTR response */ -static int nfs4_xdr_dec_setattr(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_setattrres *res) +static int nfs4_xdr_dec_setattr(struct rpc_rqst *rqstp, __be32 *p, struct nfs_setattrres *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4065,7 +4069,7 @@ out: /* * Decode LOCK response */ -static int nfs4_xdr_dec_lock(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_lock_res *res) +static int nfs4_xdr_dec_lock(struct rpc_rqst *rqstp, __be32 *p, struct nfs_lock_res *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4086,7 +4090,7 @@ out: /* * Decode LOCKT response */ -static int nfs4_xdr_dec_lockt(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_lockt_res *res) +static int nfs4_xdr_dec_lockt(struct rpc_rqst *rqstp, __be32 *p, struct nfs_lockt_res *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4107,7 +4111,7 @@ out: /* * Decode LOCKU response */ -static int nfs4_xdr_dec_locku(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_locku_res *res) +static int nfs4_xdr_dec_locku(struct rpc_rqst *rqstp, __be32 *p, struct nfs_locku_res *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4128,7 +4132,7 @@ out: /* * Decode READLINK response */ -static int nfs4_xdr_dec_readlink(struct rpc_rqst *rqstp, uint32_t *p, void *res) +static int nfs4_xdr_dec_readlink(struct rpc_rqst *rqstp, __be32 *p, void *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4149,7 +4153,7 @@ out: /* * Decode READDIR response */ -static int nfs4_xdr_dec_readdir(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_readdir_res *res) +static int nfs4_xdr_dec_readdir(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_readdir_res *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4170,7 +4174,7 @@ out: /* * Decode Read response */ -static int nfs4_xdr_dec_read(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_readres *res) +static int nfs4_xdr_dec_read(struct rpc_rqst *rqstp, __be32 *p, struct nfs_readres *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4193,7 +4197,7 @@ out: /* * Decode WRITE response */ -static int nfs4_xdr_dec_write(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_writeres *res) +static int nfs4_xdr_dec_write(struct rpc_rqst *rqstp, __be32 *p, struct nfs_writeres *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4219,7 +4223,7 @@ out: /* * Decode COMMIT response */ -static int nfs4_xdr_dec_commit(struct rpc_rqst *rqstp, uint32_t *p, struct nfs_writeres *res) +static int nfs4_xdr_dec_commit(struct rpc_rqst *rqstp, __be32 *p, struct nfs_writeres *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4243,7 +4247,7 @@ out: /* * FSINFO request */ -static int nfs4_xdr_dec_fsinfo(struct rpc_rqst *req, uint32_t *p, struct nfs_fsinfo *fsinfo) +static int nfs4_xdr_dec_fsinfo(struct rpc_rqst *req, __be32 *p, struct nfs_fsinfo *fsinfo) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4263,7 +4267,7 @@ static int nfs4_xdr_dec_fsinfo(struct rpc_rqst *req, uint32_t *p, struct nfs_fsi /* * PATHCONF request */ -static int nfs4_xdr_dec_pathconf(struct rpc_rqst *req, uint32_t *p, struct nfs_pathconf *pathconf) +static int nfs4_xdr_dec_pathconf(struct rpc_rqst *req, __be32 *p, struct nfs_pathconf *pathconf) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4281,7 +4285,7 @@ static int nfs4_xdr_dec_pathconf(struct rpc_rqst *req, uint32_t *p, struct nfs_p /* * STATFS request */ -static int nfs4_xdr_dec_statfs(struct rpc_rqst *req, uint32_t *p, struct nfs_fsstat *fsstat) +static int nfs4_xdr_dec_statfs(struct rpc_rqst *req, __be32 *p, struct nfs_fsstat *fsstat) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4299,7 +4303,7 @@ static int nfs4_xdr_dec_statfs(struct rpc_rqst *req, uint32_t *p, struct nfs_fss /* * GETATTR_BITMAP request */ -static int nfs4_xdr_dec_server_caps(struct rpc_rqst *req, uint32_t *p, struct nfs4_server_caps_res *res) +static int nfs4_xdr_dec_server_caps(struct rpc_rqst *req, __be32 *p, struct nfs4_server_caps_res *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4318,7 +4322,7 @@ out: /* * Decode RENEW response */ -static int nfs4_xdr_dec_renew(struct rpc_rqst *rqstp, uint32_t *p, void *dummy) +static int nfs4_xdr_dec_renew(struct rpc_rqst *rqstp, __be32 *p, void *dummy) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4334,7 +4338,7 @@ static int nfs4_xdr_dec_renew(struct rpc_rqst *rqstp, uint32_t *p, void *dummy) /* * a SETCLIENTID request */ -static int nfs4_xdr_dec_setclientid(struct rpc_rqst *req, uint32_t *p, +static int nfs4_xdr_dec_setclientid(struct rpc_rqst *req, __be32 *p, struct nfs_client *clp) { struct xdr_stream xdr; @@ -4353,7 +4357,7 @@ static int nfs4_xdr_dec_setclientid(struct rpc_rqst *req, uint32_t *p, /* * a SETCLIENTID_CONFIRM request */ -static int nfs4_xdr_dec_setclientid_confirm(struct rpc_rqst *req, uint32_t *p, struct nfs_fsinfo *fsinfo) +static int nfs4_xdr_dec_setclientid_confirm(struct rpc_rqst *req, __be32 *p, struct nfs_fsinfo *fsinfo) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4375,7 +4379,7 @@ static int nfs4_xdr_dec_setclientid_confirm(struct rpc_rqst *req, uint32_t *p, s /* * DELEGRETURN request */ -static int nfs4_xdr_dec_delegreturn(struct rpc_rqst *rqstp, uint32_t *p, struct nfs4_delegreturnres *res) +static int nfs4_xdr_dec_delegreturn(struct rpc_rqst *rqstp, __be32 *p, struct nfs4_delegreturnres *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4397,7 +4401,7 @@ out: /* * FS_LOCATIONS request */ -static int nfs4_xdr_dec_fs_locations(struct rpc_rqst *req, uint32_t *p, struct nfs4_fs_locations *res) +static int nfs4_xdr_dec_fs_locations(struct rpc_rqst *req, __be32 *p, struct nfs4_fs_locations *res) { struct xdr_stream xdr; struct compound_hdr hdr; @@ -4417,7 +4421,7 @@ out: return status; } -uint32_t *nfs4_decode_dirent(uint32_t *p, struct nfs_entry *entry, int plus) +__be32 *nfs4_decode_dirent(__be32 *p, struct nfs_entry *entry, int plus) { uint32_t bitmap[2] = {0}; uint32_t len; diff --git a/fs/nfs/super.c b/fs/nfs/super.c index 28659a919d6e..28108c82b887 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -834,7 +834,7 @@ static int nfs4_get_sb(struct file_system_type *fs_type, } /* RFC3530: The default port for NFS is 2049 */ if (addr.sin_port == 0) - addr.sin_port = NFS_PORT; + addr.sin_port = htons(NFS_PORT); /* Grab the authentication type */ authflavour = RPC_AUTH_UNIX; diff --git a/fs/nfs/write.c b/fs/nfs/write.c index f6675d2c386c..883dd4a1c157 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -57,6 +57,8 @@ #include <linux/nfs_fs.h> #include <linux/nfs_mount.h> #include <linux/nfs_page.h> +#include <linux/backing-dev.h> + #include <asm/uaccess.h> #include <linux/smp_lock.h> @@ -395,7 +397,7 @@ int nfs_writepages(struct address_space *mapping, struct writeback_control *wbc) out: clear_bit(BDI_write_congested, &bdi->state); wake_up_all(&nfs_write_congestion); - writeback_congestion_end(); + congestion_end(WRITE); return err; } @@ -588,10 +590,10 @@ static void nfs_cancel_commit_list(struct list_head *head) while(!list_empty(head)) { req = nfs_list_entry(head->next); + dec_zone_page_state(req->wb_page, NR_UNSTABLE_NFS); nfs_list_remove_request(req); nfs_inode_remove_request(req); - dec_zone_page_state(req->wb_page, NR_UNSTABLE_NFS); - nfs_clear_page_writeback(req); + nfs_unlock_request(req); } } diff --git a/fs/nfs_common/nfsacl.c b/fs/nfs_common/nfsacl.c index 0c2be8c0307d..c11f5375d7c1 100644 --- a/fs/nfs_common/nfsacl.c +++ b/fs/nfs_common/nfsacl.c @@ -46,7 +46,7 @@ xdr_nfsace_encode(struct xdr_array2_desc *desc, void *elem) { struct nfsacl_encode_desc *nfsacl_desc = (struct nfsacl_encode_desc *) desc; - u32 *p = (u32 *) elem; + __be32 *p = elem; struct posix_acl_entry *entry = &nfsacl_desc->acl->a_entries[nfsacl_desc->count++]; @@ -127,7 +127,7 @@ xdr_nfsace_decode(struct xdr_array2_desc *desc, void *elem) { struct nfsacl_decode_desc *nfsacl_desc = (struct nfsacl_decode_desc *) desc; - u32 *p = (u32 *) elem; + __be32 *p = elem; struct posix_acl_entry *entry; if (!nfsacl_desc->acl) { diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c index e13fa23bd108..f37df46d2eaa 100644 --- a/fs/nfsd/export.c +++ b/fs/nfsd/export.c @@ -1148,12 +1148,12 @@ exp_find(struct auth_domain *clp, int fsid_type, u32 *fsidv, * for a given NFSv4 client. The root is defined to be the * export point with fsid==0 */ -int +__be32 exp_pseudoroot(struct auth_domain *clp, struct svc_fh *fhp, struct cache_req *creq) { struct svc_export *exp; - int rv; + __be32 rv; u32 fsidv[2]; mk_fsid_v1(fsidv, 0); diff --git a/fs/nfsd/lockd.c b/fs/nfsd/lockd.c index 7b889ff15ae6..11fdaf7721b4 100644 --- a/fs/nfsd/lockd.c +++ b/fs/nfsd/lockd.c @@ -25,7 +25,7 @@ static u32 nlm_fopen(struct svc_rqst *rqstp, struct nfs_fh *f, struct file **filp) { - u32 nfserr; + __be32 nfserr; struct svc_fh fh; /* must initialize before using! but maxsize doesn't matter */ @@ -39,18 +39,20 @@ nlm_fopen(struct svc_rqst *rqstp, struct nfs_fh *f, struct file **filp) fh_put(&fh); rqstp->rq_client = NULL; exp_readunlock(); - /* nlm and nfsd don't share error codes. - * we invent: 0 = no error - * 1 = stale file handle - * 2 = other error + /* We return nlm error codes as nlm doesn't know + * about nfsd, but nfsd does know about nlm.. */ switch (nfserr) { case nfs_ok: return 0; + case nfserr_dropit: + return nlm_drop_reply; +#ifdef CONFIG_LOCKD_V4 case nfserr_stale: - return 1; + return nlm4_stale_fh; +#endif default: - return 2; + return nlm_lck_denied; } } diff --git a/fs/nfsd/nfs2acl.c b/fs/nfsd/nfs2acl.c index 9187755661df..e3eca0816986 100644 --- a/fs/nfsd/nfs2acl.c +++ b/fs/nfsd/nfs2acl.c @@ -21,7 +21,7 @@ /* * NULL call. */ -static int +static __be32 nfsacld_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) { return nfs_ok; @@ -30,12 +30,12 @@ nfsacld_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) /* * Get the Access and/or Default ACL of a file. */ -static int nfsacld_proc_getacl(struct svc_rqst * rqstp, +static __be32 nfsacld_proc_getacl(struct svc_rqst * rqstp, struct nfsd3_getaclargs *argp, struct nfsd3_getaclres *resp) { svc_fh *fh; struct posix_acl *acl; - int nfserr = 0; + __be32 nfserr = 0; dprintk("nfsd: GETACL(2acl) %s\n", SVCFH_fmt(&argp->fh)); @@ -97,12 +97,12 @@ fail: /* * Set the Access and/or Default ACL of a file. */ -static int nfsacld_proc_setacl(struct svc_rqst * rqstp, +static __be32 nfsacld_proc_setacl(struct svc_rqst * rqstp, struct nfsd3_setaclargs *argp, struct nfsd_attrstat *resp) { svc_fh *fh; - int nfserr = 0; + __be32 nfserr = 0; dprintk("nfsd: SETACL(2acl) %s\n", SVCFH_fmt(&argp->fh)); @@ -128,7 +128,7 @@ static int nfsacld_proc_setacl(struct svc_rqst * rqstp, /* * Check file attributes */ -static int nfsacld_proc_getattr(struct svc_rqst * rqstp, +static __be32 nfsacld_proc_getattr(struct svc_rqst * rqstp, struct nfsd_fhandle *argp, struct nfsd_attrstat *resp) { dprintk("nfsd: GETATTR %s\n", SVCFH_fmt(&argp->fh)); @@ -140,10 +140,10 @@ static int nfsacld_proc_getattr(struct svc_rqst * rqstp, /* * Check file access */ -static int nfsacld_proc_access(struct svc_rqst *rqstp, struct nfsd3_accessargs *argp, +static __be32 nfsacld_proc_access(struct svc_rqst *rqstp, struct nfsd3_accessargs *argp, struct nfsd3_accessres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: ACCESS(2acl) %s 0x%x\n", SVCFH_fmt(&argp->fh), @@ -158,7 +158,7 @@ static int nfsacld_proc_access(struct svc_rqst *rqstp, struct nfsd3_accessargs * /* * XDR decode functions */ -static int nfsaclsvc_decode_getaclargs(struct svc_rqst *rqstp, u32 *p, +static int nfsaclsvc_decode_getaclargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_getaclargs *argp) { if (!(p = nfs2svc_decode_fh(p, &argp->fh))) @@ -169,7 +169,7 @@ static int nfsaclsvc_decode_getaclargs(struct svc_rqst *rqstp, u32 *p, } -static int nfsaclsvc_decode_setaclargs(struct svc_rqst *rqstp, u32 *p, +static int nfsaclsvc_decode_setaclargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_setaclargs *argp) { struct kvec *head = rqstp->rq_arg.head; @@ -194,7 +194,7 @@ static int nfsaclsvc_decode_setaclargs(struct svc_rqst *rqstp, u32 *p, return (n > 0); } -static int nfsaclsvc_decode_fhandleargs(struct svc_rqst *rqstp, u32 *p, +static int nfsaclsvc_decode_fhandleargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_fhandle *argp) { if (!(p = nfs2svc_decode_fh(p, &argp->fh))) @@ -202,7 +202,7 @@ static int nfsaclsvc_decode_fhandleargs(struct svc_rqst *rqstp, u32 *p, return xdr_argsize_check(rqstp, p); } -static int nfsaclsvc_decode_accessargs(struct svc_rqst *rqstp, u32 *p, +static int nfsaclsvc_decode_accessargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_accessargs *argp) { if (!(p = nfs2svc_decode_fh(p, &argp->fh))) @@ -217,7 +217,7 @@ static int nfsaclsvc_decode_accessargs(struct svc_rqst *rqstp, u32 *p, */ /* GETACL */ -static int nfsaclsvc_encode_getaclres(struct svc_rqst *rqstp, u32 *p, +static int nfsaclsvc_encode_getaclres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_getaclres *resp) { struct dentry *dentry = resp->fh.fh_dentry; @@ -259,7 +259,7 @@ static int nfsaclsvc_encode_getaclres(struct svc_rqst *rqstp, u32 *p, return 1; } -static int nfsaclsvc_encode_attrstatres(struct svc_rqst *rqstp, u32 *p, +static int nfsaclsvc_encode_attrstatres(struct svc_rqst *rqstp, __be32 *p, struct nfsd_attrstat *resp) { p = nfs2svc_encode_fattr(rqstp, p, &resp->fh); @@ -267,7 +267,7 @@ static int nfsaclsvc_encode_attrstatres(struct svc_rqst *rqstp, u32 *p, } /* ACCESS */ -static int nfsaclsvc_encode_accessres(struct svc_rqst *rqstp, u32 *p, +static int nfsaclsvc_encode_accessres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_accessres *resp) { p = nfs2svc_encode_fattr(rqstp, p, &resp->fh); @@ -278,7 +278,7 @@ static int nfsaclsvc_encode_accessres(struct svc_rqst *rqstp, u32 *p, /* * XDR release functions */ -static int nfsaclsvc_release_getacl(struct svc_rqst *rqstp, u32 *p, +static int nfsaclsvc_release_getacl(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_getaclres *resp) { fh_put(&resp->fh); @@ -287,7 +287,7 @@ static int nfsaclsvc_release_getacl(struct svc_rqst *rqstp, u32 *p, return 1; } -static int nfsaclsvc_release_fhandle(struct svc_rqst *rqstp, u32 *p, +static int nfsaclsvc_release_fhandle(struct svc_rqst *rqstp, __be32 *p, struct nfsd_fhandle *resp) { fh_put(&resp->fh); diff --git a/fs/nfsd/nfs3acl.c b/fs/nfsd/nfs3acl.c index d4bdc00c1169..fcad2895ddb0 100644 --- a/fs/nfsd/nfs3acl.c +++ b/fs/nfsd/nfs3acl.c @@ -19,7 +19,7 @@ /* * NULL call. */ -static int +static __be32 nfsd3_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) { return nfs_ok; @@ -28,12 +28,12 @@ nfsd3_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) /* * Get the Access and/or Default ACL of a file. */ -static int nfsd3_proc_getacl(struct svc_rqst * rqstp, +static __be32 nfsd3_proc_getacl(struct svc_rqst * rqstp, struct nfsd3_getaclargs *argp, struct nfsd3_getaclres *resp) { svc_fh *fh; struct posix_acl *acl; - int nfserr = 0; + __be32 nfserr = 0; fh = fh_copy(&resp->fh, &argp->fh); if ((nfserr = fh_verify(rqstp, &resp->fh, 0, MAY_NOP))) @@ -93,12 +93,12 @@ fail: /* * Set the Access and/or Default ACL of a file. */ -static int nfsd3_proc_setacl(struct svc_rqst * rqstp, +static __be32 nfsd3_proc_setacl(struct svc_rqst * rqstp, struct nfsd3_setaclargs *argp, struct nfsd3_attrstat *resp) { svc_fh *fh; - int nfserr = 0; + __be32 nfserr = 0; fh = fh_copy(&resp->fh, &argp->fh); nfserr = fh_verify(rqstp, &resp->fh, 0, MAY_SATTR); @@ -122,7 +122,7 @@ static int nfsd3_proc_setacl(struct svc_rqst * rqstp, /* * XDR decode functions */ -static int nfs3svc_decode_getaclargs(struct svc_rqst *rqstp, u32 *p, +static int nfs3svc_decode_getaclargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_getaclargs *args) { if (!(p = nfs3svc_decode_fh(p, &args->fh))) @@ -133,7 +133,7 @@ static int nfs3svc_decode_getaclargs(struct svc_rqst *rqstp, u32 *p, } -static int nfs3svc_decode_setaclargs(struct svc_rqst *rqstp, u32 *p, +static int nfs3svc_decode_setaclargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_setaclargs *args) { struct kvec *head = rqstp->rq_arg.head; @@ -163,7 +163,7 @@ static int nfs3svc_decode_setaclargs(struct svc_rqst *rqstp, u32 *p, */ /* GETACL */ -static int nfs3svc_encode_getaclres(struct svc_rqst *rqstp, u32 *p, +static int nfs3svc_encode_getaclres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_getaclres *resp) { struct dentry *dentry = resp->fh.fh_dentry; @@ -208,7 +208,7 @@ static int nfs3svc_encode_getaclres(struct svc_rqst *rqstp, u32 *p, } /* SETACL */ -static int nfs3svc_encode_setaclres(struct svc_rqst *rqstp, u32 *p, +static int nfs3svc_encode_setaclres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_attrstat *resp) { p = nfs3svc_encode_post_op_attr(rqstp, p, &resp->fh); @@ -219,7 +219,7 @@ static int nfs3svc_encode_setaclres(struct svc_rqst *rqstp, u32 *p, /* * XDR release functions */ -static int nfs3svc_release_getacl(struct svc_rqst *rqstp, u32 *p, +static int nfs3svc_release_getacl(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_getaclres *resp) { fh_put(&resp->fh); diff --git a/fs/nfsd/nfs3proc.c b/fs/nfsd/nfs3proc.c index a5ebc7dbb384..64db601c2bd2 100644 --- a/fs/nfsd/nfs3proc.c +++ b/fs/nfsd/nfs3proc.c @@ -43,7 +43,7 @@ static int nfs3_ftypes[] = { /* * NULL call. */ -static int +static __be32 nfsd3_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) { return nfs_ok; @@ -52,11 +52,12 @@ nfsd3_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) /* * Get a file's attributes */ -static int +static __be32 nfsd3_proc_getattr(struct svc_rqst *rqstp, struct nfsd_fhandle *argp, struct nfsd3_attrstat *resp) { - int err, nfserr; + int err; + __be32 nfserr; dprintk("nfsd: GETATTR(3) %s\n", SVCFH_fmt(&argp->fh)); @@ -76,11 +77,11 @@ nfsd3_proc_getattr(struct svc_rqst *rqstp, struct nfsd_fhandle *argp, /* * Set a file's attributes */ -static int +static __be32 nfsd3_proc_setattr(struct svc_rqst *rqstp, struct nfsd3_sattrargs *argp, struct nfsd3_attrstat *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: SETATTR(3) %s\n", SVCFH_fmt(&argp->fh)); @@ -94,11 +95,11 @@ nfsd3_proc_setattr(struct svc_rqst *rqstp, struct nfsd3_sattrargs *argp, /* * Look up a path name component */ -static int +static __be32 nfsd3_proc_lookup(struct svc_rqst *rqstp, struct nfsd3_diropargs *argp, struct nfsd3_diropres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: LOOKUP(3) %s %.*s\n", SVCFH_fmt(&argp->fh), @@ -118,11 +119,11 @@ nfsd3_proc_lookup(struct svc_rqst *rqstp, struct nfsd3_diropargs *argp, /* * Check file access */ -static int +static __be32 nfsd3_proc_access(struct svc_rqst *rqstp, struct nfsd3_accessargs *argp, struct nfsd3_accessres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: ACCESS(3) %s 0x%x\n", SVCFH_fmt(&argp->fh), @@ -137,11 +138,11 @@ nfsd3_proc_access(struct svc_rqst *rqstp, struct nfsd3_accessargs *argp, /* * Read a symlink. */ -static int +static __be32 nfsd3_proc_readlink(struct svc_rqst *rqstp, struct nfsd3_readlinkargs *argp, struct nfsd3_readlinkres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: READLINK(3) %s\n", SVCFH_fmt(&argp->fh)); @@ -155,11 +156,11 @@ nfsd3_proc_readlink(struct svc_rqst *rqstp, struct nfsd3_readlinkargs *argp, /* * Read a portion of a file. */ -static int +static __be32 nfsd3_proc_read(struct svc_rqst *rqstp, struct nfsd3_readargs *argp, struct nfsd3_readres *resp) { - int nfserr; + __be32 nfserr; u32 max_blocksize = svc_max_payload(rqstp); dprintk("nfsd: READ(3) %s %lu bytes at %lu\n", @@ -195,11 +196,11 @@ nfsd3_proc_read(struct svc_rqst *rqstp, struct nfsd3_readargs *argp, /* * Write data to a file */ -static int +static __be32 nfsd3_proc_write(struct svc_rqst *rqstp, struct nfsd3_writeargs *argp, struct nfsd3_writeres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: WRITE(3) %s %d bytes at %ld%s\n", SVCFH_fmt(&argp->fh), @@ -223,13 +224,13 @@ nfsd3_proc_write(struct svc_rqst *rqstp, struct nfsd3_writeargs *argp, * At least in theory; we'll see how it fares in practice when the * first reports about SunOS compatibility problems start to pour in... */ -static int +static __be32 nfsd3_proc_create(struct svc_rqst *rqstp, struct nfsd3_createargs *argp, struct nfsd3_diropres *resp) { svc_fh *dirfhp, *newfhp = NULL; struct iattr *attr; - u32 nfserr; + __be32 nfserr; dprintk("nfsd: CREATE(3) %s %.*s\n", SVCFH_fmt(&argp->fh), @@ -265,11 +266,11 @@ nfsd3_proc_create(struct svc_rqst *rqstp, struct nfsd3_createargs *argp, /* * Make directory. This operation is not idempotent. */ -static int +static __be32 nfsd3_proc_mkdir(struct svc_rqst *rqstp, struct nfsd3_createargs *argp, struct nfsd3_diropres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: MKDIR(3) %s %.*s\n", SVCFH_fmt(&argp->fh), @@ -285,11 +286,11 @@ nfsd3_proc_mkdir(struct svc_rqst *rqstp, struct nfsd3_createargs *argp, RETURN_STATUS(nfserr); } -static int +static __be32 nfsd3_proc_symlink(struct svc_rqst *rqstp, struct nfsd3_symlinkargs *argp, struct nfsd3_diropres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: SYMLINK(3) %s %.*s -> %.*s\n", SVCFH_fmt(&argp->ffh), @@ -307,11 +308,12 @@ nfsd3_proc_symlink(struct svc_rqst *rqstp, struct nfsd3_symlinkargs *argp, /* * Make socket/fifo/device. */ -static int +static __be32 nfsd3_proc_mknod(struct svc_rqst *rqstp, struct nfsd3_mknodargs *argp, struct nfsd3_diropres *resp) { - int nfserr, type; + __be32 nfserr; + int type; dev_t rdev = 0; dprintk("nfsd: MKNOD(3) %s %.*s\n", @@ -343,11 +345,11 @@ nfsd3_proc_mknod(struct svc_rqst *rqstp, struct nfsd3_mknodargs *argp, /* * Remove file/fifo/socket etc. */ -static int +static __be32 nfsd3_proc_remove(struct svc_rqst *rqstp, struct nfsd3_diropargs *argp, struct nfsd3_attrstat *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: REMOVE(3) %s %.*s\n", SVCFH_fmt(&argp->fh), @@ -363,11 +365,11 @@ nfsd3_proc_remove(struct svc_rqst *rqstp, struct nfsd3_diropargs *argp, /* * Remove a directory */ -static int +static __be32 nfsd3_proc_rmdir(struct svc_rqst *rqstp, struct nfsd3_diropargs *argp, struct nfsd3_attrstat *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: RMDIR(3) %s %.*s\n", SVCFH_fmt(&argp->fh), @@ -379,11 +381,11 @@ nfsd3_proc_rmdir(struct svc_rqst *rqstp, struct nfsd3_diropargs *argp, RETURN_STATUS(nfserr); } -static int +static __be32 nfsd3_proc_rename(struct svc_rqst *rqstp, struct nfsd3_renameargs *argp, struct nfsd3_renameres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: RENAME(3) %s %.*s ->\n", SVCFH_fmt(&argp->ffh), @@ -401,11 +403,11 @@ nfsd3_proc_rename(struct svc_rqst *rqstp, struct nfsd3_renameargs *argp, RETURN_STATUS(nfserr); } -static int +static __be32 nfsd3_proc_link(struct svc_rqst *rqstp, struct nfsd3_linkargs *argp, struct nfsd3_linkres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: LINK(3) %s ->\n", SVCFH_fmt(&argp->ffh)); @@ -424,11 +426,12 @@ nfsd3_proc_link(struct svc_rqst *rqstp, struct nfsd3_linkargs *argp, /* * Read a portion of a directory. */ -static int +static __be32 nfsd3_proc_readdir(struct svc_rqst *rqstp, struct nfsd3_readdirargs *argp, struct nfsd3_readdirres *resp) { - int nfserr, count; + __be32 nfserr; + int count; dprintk("nfsd: READDIR(3) %s %d bytes at %d\n", SVCFH_fmt(&argp->fh), @@ -459,11 +462,12 @@ nfsd3_proc_readdir(struct svc_rqst *rqstp, struct nfsd3_readdirargs *argp, * Read a portion of a directory, including file handles and attrs. * For now, we choose to ignore the dircount parameter. */ -static int +static __be32 nfsd3_proc_readdirplus(struct svc_rqst *rqstp, struct nfsd3_readdirargs *argp, struct nfsd3_readdirres *resp) { - int nfserr, count = 0; + __be32 nfserr; + int count = 0; loff_t offset; int i; caddr_t page_addr = NULL; @@ -517,11 +521,11 @@ nfsd3_proc_readdirplus(struct svc_rqst *rqstp, struct nfsd3_readdirargs *argp, /* * Get file system stats */ -static int +static __be32 nfsd3_proc_fsstat(struct svc_rqst * rqstp, struct nfsd_fhandle *argp, struct nfsd3_fsstatres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: FSSTAT(3) %s\n", SVCFH_fmt(&argp->fh)); @@ -534,11 +538,11 @@ nfsd3_proc_fsstat(struct svc_rqst * rqstp, struct nfsd_fhandle *argp, /* * Get file system info */ -static int +static __be32 nfsd3_proc_fsinfo(struct svc_rqst * rqstp, struct nfsd_fhandle *argp, struct nfsd3_fsinfores *resp) { - int nfserr; + __be32 nfserr; u32 max_blocksize = svc_max_payload(rqstp); dprintk("nfsd: FSINFO(3) %s\n", @@ -576,11 +580,11 @@ nfsd3_proc_fsinfo(struct svc_rqst * rqstp, struct nfsd_fhandle *argp, /* * Get pathconf info for the specified file */ -static int +static __be32 nfsd3_proc_pathconf(struct svc_rqst * rqstp, struct nfsd_fhandle *argp, struct nfsd3_pathconfres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: PATHCONF(3) %s\n", SVCFH_fmt(&argp->fh)); @@ -619,11 +623,11 @@ nfsd3_proc_pathconf(struct svc_rqst * rqstp, struct nfsd_fhandle *argp, /* * Commit a file (range) to stable storage. */ -static int +static __be32 nfsd3_proc_commit(struct svc_rqst * rqstp, struct nfsd3_commitargs *argp, struct nfsd3_commitres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: COMMIT(3) %s %u@%Lu\n", SVCFH_fmt(&argp->fh), diff --git a/fs/nfsd/nfs3xdr.c b/fs/nfsd/nfs3xdr.c index 247d518248bf..b4baca3053c3 100644 --- a/fs/nfsd/nfs3xdr.c +++ b/fs/nfsd/nfs3xdr.c @@ -42,23 +42,23 @@ static u32 nfs3_ftypes[] = { /* * XDR functions for basic NFS types */ -static inline u32 * -encode_time3(u32 *p, struct timespec *time) +static inline __be32 * +encode_time3(__be32 *p, struct timespec *time) { *p++ = htonl((u32) time->tv_sec); *p++ = htonl(time->tv_nsec); return p; } -static inline u32 * -decode_time3(u32 *p, struct timespec *time) +static inline __be32 * +decode_time3(__be32 *p, struct timespec *time) { time->tv_sec = ntohl(*p++); time->tv_nsec = ntohl(*p++); return p; } -static inline u32 * -decode_fh(u32 *p, struct svc_fh *fhp) +static inline __be32 * +decode_fh(__be32 *p, struct svc_fh *fhp) { unsigned int size; fh_init(fhp, NFS3_FHSIZE); @@ -72,13 +72,13 @@ decode_fh(u32 *p, struct svc_fh *fhp) } /* Helper function for NFSv3 ACL code */ -u32 *nfs3svc_decode_fh(u32 *p, struct svc_fh *fhp) +__be32 *nfs3svc_decode_fh(__be32 *p, struct svc_fh *fhp) { return decode_fh(p, fhp); } -static inline u32 * -encode_fh(u32 *p, struct svc_fh *fhp) +static inline __be32 * +encode_fh(__be32 *p, struct svc_fh *fhp) { unsigned int size = fhp->fh_handle.fh_size; *p++ = htonl(size); @@ -91,8 +91,8 @@ encode_fh(u32 *p, struct svc_fh *fhp) * Decode a file name and make sure that the path contains * no slashes or null bytes. */ -static inline u32 * -decode_filename(u32 *p, char **namp, int *lenp) +static inline __be32 * +decode_filename(__be32 *p, char **namp, int *lenp) { char *name; int i; @@ -107,8 +107,8 @@ decode_filename(u32 *p, char **namp, int *lenp) return p; } -static inline u32 * -decode_sattr3(u32 *p, struct iattr *iap) +static inline __be32 * +decode_sattr3(__be32 *p, struct iattr *iap) { u32 tmp; @@ -153,8 +153,8 @@ decode_sattr3(u32 *p, struct iattr *iap) return p; } -static inline u32 * -encode_fattr3(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp, +static inline __be32 * +encode_fattr3(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp, struct kstat *stat) { struct dentry *dentry = fhp->fh_dentry; @@ -186,8 +186,8 @@ encode_fattr3(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp, return p; } -static inline u32 * -encode_saved_post_attr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) +static inline __be32 * +encode_saved_post_attr(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp) { struct inode *inode = fhp->fh_dentry->d_inode; @@ -224,8 +224,8 @@ encode_saved_post_attr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) * The inode may be NULL if the call failed because of a stale file * handle. In this case, no attributes are returned. */ -static u32 * -encode_post_op_attr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) +static __be32 * +encode_post_op_attr(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp) { struct dentry *dentry = fhp->fh_dentry; if (dentry && dentry->d_inode != NULL) { @@ -243,8 +243,8 @@ encode_post_op_attr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) } /* Helper for NFSv3 ACLs */ -u32 * -nfs3svc_encode_post_op_attr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) +__be32 * +nfs3svc_encode_post_op_attr(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp) { return encode_post_op_attr(rqstp, p, fhp); } @@ -252,8 +252,8 @@ nfs3svc_encode_post_op_attr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) /* * Enocde weak cache consistency data */ -static u32 * -encode_wcc_data(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) +static __be32 * +encode_wcc_data(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp) { struct dentry *dentry = fhp->fh_dentry; @@ -278,7 +278,7 @@ encode_wcc_data(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) * XDR decode functions */ int -nfs3svc_decode_fhandle(struct svc_rqst *rqstp, u32 *p, struct nfsd_fhandle *args) +nfs3svc_decode_fhandle(struct svc_rqst *rqstp, __be32 *p, struct nfsd_fhandle *args) { if (!(p = decode_fh(p, &args->fh))) return 0; @@ -286,7 +286,7 @@ nfs3svc_decode_fhandle(struct svc_rqst *rqstp, u32 *p, struct nfsd_fhandle *args } int -nfs3svc_decode_sattrargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_sattrargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_sattrargs *args) { if (!(p = decode_fh(p, &args->fh)) @@ -303,7 +303,7 @@ nfs3svc_decode_sattrargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_diropargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_diropargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_diropargs *args) { if (!(p = decode_fh(p, &args->fh)) @@ -314,7 +314,7 @@ nfs3svc_decode_diropargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_accessargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_accessargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_accessargs *args) { if (!(p = decode_fh(p, &args->fh))) @@ -325,7 +325,7 @@ nfs3svc_decode_accessargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_readargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_readargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_readargs *args) { unsigned int len; @@ -355,7 +355,7 @@ nfs3svc_decode_readargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_writeargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_writeargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_writeargs *args) { unsigned int len, v, hdr; @@ -393,7 +393,7 @@ nfs3svc_decode_writeargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_createargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_createargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_createargs *args) { if (!(p = decode_fh(p, &args->fh)) @@ -417,7 +417,7 @@ nfs3svc_decode_createargs(struct svc_rqst *rqstp, u32 *p, return xdr_argsize_check(rqstp, p); } int -nfs3svc_decode_mkdirargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_mkdirargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_createargs *args) { if (!(p = decode_fh(p, &args->fh)) @@ -429,7 +429,7 @@ nfs3svc_decode_mkdirargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_symlinkargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_symlinkargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_symlinkargs *args) { unsigned int len; @@ -481,7 +481,7 @@ nfs3svc_decode_symlinkargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_mknodargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_mknodargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_mknodargs *args) { if (!(p = decode_fh(p, &args->fh)) @@ -505,7 +505,7 @@ nfs3svc_decode_mknodargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_renameargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_renameargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_renameargs *args) { if (!(p = decode_fh(p, &args->ffh)) @@ -518,7 +518,7 @@ nfs3svc_decode_renameargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_readlinkargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_readlinkargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_readlinkargs *args) { if (!(p = decode_fh(p, &args->fh))) @@ -530,7 +530,7 @@ nfs3svc_decode_readlinkargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_linkargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_linkargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_linkargs *args) { if (!(p = decode_fh(p, &args->ffh)) @@ -542,7 +542,7 @@ nfs3svc_decode_linkargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_readdirargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_readdirargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_readdirargs *args) { if (!(p = decode_fh(p, &args->fh))) @@ -562,7 +562,7 @@ nfs3svc_decode_readdirargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_readdirplusargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_readdirplusargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_readdirargs *args) { int len, pn; @@ -590,7 +590,7 @@ nfs3svc_decode_readdirplusargs(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_decode_commitargs(struct svc_rqst *rqstp, u32 *p, +nfs3svc_decode_commitargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_commitargs *args) { if (!(p = decode_fh(p, &args->fh))) @@ -609,14 +609,14 @@ nfs3svc_decode_commitargs(struct svc_rqst *rqstp, u32 *p, * will work properly. */ int -nfs3svc_encode_voidres(struct svc_rqst *rqstp, u32 *p, void *dummy) +nfs3svc_encode_voidres(struct svc_rqst *rqstp, __be32 *p, void *dummy) { return xdr_ressize_check(rqstp, p); } /* GETATTR */ int -nfs3svc_encode_attrstat(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_attrstat(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_attrstat *resp) { if (resp->status == 0) @@ -626,7 +626,7 @@ nfs3svc_encode_attrstat(struct svc_rqst *rqstp, u32 *p, /* SETATTR, REMOVE, RMDIR */ int -nfs3svc_encode_wccstat(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_wccstat(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_attrstat *resp) { p = encode_wcc_data(rqstp, p, &resp->fh); @@ -635,7 +635,7 @@ nfs3svc_encode_wccstat(struct svc_rqst *rqstp, u32 *p, /* LOOKUP */ int -nfs3svc_encode_diropres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_diropres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_diropres *resp) { if (resp->status == 0) { @@ -648,7 +648,7 @@ nfs3svc_encode_diropres(struct svc_rqst *rqstp, u32 *p, /* ACCESS */ int -nfs3svc_encode_accessres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_accessres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_accessres *resp) { p = encode_post_op_attr(rqstp, p, &resp->fh); @@ -659,7 +659,7 @@ nfs3svc_encode_accessres(struct svc_rqst *rqstp, u32 *p, /* READLINK */ int -nfs3svc_encode_readlinkres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_readlinkres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_readlinkres *resp) { p = encode_post_op_attr(rqstp, p, &resp->fh); @@ -680,7 +680,7 @@ nfs3svc_encode_readlinkres(struct svc_rqst *rqstp, u32 *p, /* READ */ int -nfs3svc_encode_readres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_readres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_readres *resp) { p = encode_post_op_attr(rqstp, p, &resp->fh); @@ -704,7 +704,7 @@ nfs3svc_encode_readres(struct svc_rqst *rqstp, u32 *p, /* WRITE */ int -nfs3svc_encode_writeres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_writeres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_writeres *resp) { p = encode_wcc_data(rqstp, p, &resp->fh); @@ -719,7 +719,7 @@ nfs3svc_encode_writeres(struct svc_rqst *rqstp, u32 *p, /* CREATE, MKDIR, SYMLINK, MKNOD */ int -nfs3svc_encode_createres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_createres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_diropres *resp) { if (resp->status == 0) { @@ -733,7 +733,7 @@ nfs3svc_encode_createres(struct svc_rqst *rqstp, u32 *p, /* RENAME */ int -nfs3svc_encode_renameres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_renameres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_renameres *resp) { p = encode_wcc_data(rqstp, p, &resp->ffh); @@ -743,7 +743,7 @@ nfs3svc_encode_renameres(struct svc_rqst *rqstp, u32 *p, /* LINK */ int -nfs3svc_encode_linkres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_linkres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_linkres *resp) { p = encode_post_op_attr(rqstp, p, &resp->fh); @@ -753,7 +753,7 @@ nfs3svc_encode_linkres(struct svc_rqst *rqstp, u32 *p, /* READDIR */ int -nfs3svc_encode_readdirres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_readdirres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_readdirres *resp) { p = encode_post_op_attr(rqstp, p, &resp->fh); @@ -776,8 +776,8 @@ nfs3svc_encode_readdirres(struct svc_rqst *rqstp, u32 *p, return xdr_ressize_check(rqstp, p); } -static inline u32 * -encode_entry_baggage(struct nfsd3_readdirres *cd, u32 *p, const char *name, +static inline __be32 * +encode_entry_baggage(struct nfsd3_readdirres *cd, __be32 *p, const char *name, int namlen, ino_t ino) { *p++ = xdr_one; /* mark entry present */ @@ -790,8 +790,8 @@ encode_entry_baggage(struct nfsd3_readdirres *cd, u32 *p, const char *name, return p; } -static inline u32 * -encode_entryplus_baggage(struct nfsd3_readdirres *cd, u32 *p, +static inline __be32 * +encode_entryplus_baggage(struct nfsd3_readdirres *cd, __be32 *p, struct svc_fh *fhp) { p = encode_post_op_attr(cd->rqstp, p, fhp); @@ -853,7 +853,7 @@ encode_entry(struct readdir_cd *ccd, const char *name, { struct nfsd3_readdirres *cd = container_of(ccd, struct nfsd3_readdirres, common); - u32 *p = cd->buffer; + __be32 *p = cd->buffer; caddr_t curr_page_addr = NULL; int pn; /* current page number */ int slen; /* string (name) length */ @@ -919,7 +919,7 @@ encode_entry(struct readdir_cd *ccd, const char *name, } else if (cd->rqstp->rq_respages[pn+1] != NULL) { /* temporarily encode entry into next page, then move back to * current and next page in rq_respages[] */ - u32 *p1, *tmp; + __be32 *p1, *tmp; int len1, len2; /* grab next page for temporary storage of entry */ @@ -1009,7 +1009,7 @@ nfs3svc_encode_entry_plus(struct readdir_cd *cd, const char *name, /* FSSTAT */ int -nfs3svc_encode_fsstatres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_fsstatres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_fsstatres *resp) { struct kstatfs *s = &resp->stats; @@ -1031,7 +1031,7 @@ nfs3svc_encode_fsstatres(struct svc_rqst *rqstp, u32 *p, /* FSINFO */ int -nfs3svc_encode_fsinfores(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_fsinfores(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_fsinfores *resp) { *p++ = xdr_zero; /* no post_op_attr */ @@ -1055,7 +1055,7 @@ nfs3svc_encode_fsinfores(struct svc_rqst *rqstp, u32 *p, /* PATHCONF */ int -nfs3svc_encode_pathconfres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_pathconfres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_pathconfres *resp) { *p++ = xdr_zero; /* no post_op_attr */ @@ -1074,7 +1074,7 @@ nfs3svc_encode_pathconfres(struct svc_rqst *rqstp, u32 *p, /* COMMIT */ int -nfs3svc_encode_commitres(struct svc_rqst *rqstp, u32 *p, +nfs3svc_encode_commitres(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_commitres *resp) { p = encode_wcc_data(rqstp, p, &resp->fh); @@ -1090,7 +1090,7 @@ nfs3svc_encode_commitres(struct svc_rqst *rqstp, u32 *p, * XDR release functions */ int -nfs3svc_release_fhandle(struct svc_rqst *rqstp, u32 *p, +nfs3svc_release_fhandle(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_attrstat *resp) { fh_put(&resp->fh); @@ -1098,7 +1098,7 @@ nfs3svc_release_fhandle(struct svc_rqst *rqstp, u32 *p, } int -nfs3svc_release_fhandle2(struct svc_rqst *rqstp, u32 *p, +nfs3svc_release_fhandle2(struct svc_rqst *rqstp, __be32 *p, struct nfsd3_fhandle_pair *resp) { fh_put(&resp->fh1); diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c index f6ca9fb3fc63..f57655a7a2b6 100644 --- a/fs/nfsd/nfs4callback.c +++ b/fs/nfsd/nfs4callback.c @@ -85,8 +85,8 @@ enum nfs_cb_opnum4 { /* * Generic encode routines from fs/nfs/nfs4xdr.c */ -static inline u32 * -xdr_writemem(u32 *p, const void *ptr, int nbytes) +static inline __be32 * +xdr_writemem(__be32 *p, const void *ptr, int nbytes) { int tmp = XDR_QUADLEN(nbytes); if (!tmp) @@ -205,7 +205,7 @@ nfs_cb_stat_to_errno(int stat) static int encode_cb_compound_hdr(struct xdr_stream *xdr, struct nfs4_cb_compound_hdr *hdr) { - u32 * p; + __be32 * p; RESERVE_SPACE(16); WRITE32(0); /* tag length is always 0 */ @@ -218,7 +218,7 @@ encode_cb_compound_hdr(struct xdr_stream *xdr, struct nfs4_cb_compound_hdr *hdr) static int encode_cb_recall(struct xdr_stream *xdr, struct nfs4_cb_recall *cb_rec) { - u32 *p; + __be32 *p; int len = cb_rec->cbr_fhlen; RESERVE_SPACE(12+sizeof(cb_rec->cbr_stateid) + len); @@ -231,7 +231,7 @@ encode_cb_recall(struct xdr_stream *xdr, struct nfs4_cb_recall *cb_rec) } static int -nfs4_xdr_enc_cb_null(struct rpc_rqst *req, u32 *p) +nfs4_xdr_enc_cb_null(struct rpc_rqst *req, __be32 *p) { struct xdr_stream xdrs, *xdr = &xdrs; @@ -241,7 +241,7 @@ nfs4_xdr_enc_cb_null(struct rpc_rqst *req, u32 *p) } static int -nfs4_xdr_enc_cb_recall(struct rpc_rqst *req, u32 *p, struct nfs4_cb_recall *args) +nfs4_xdr_enc_cb_recall(struct rpc_rqst *req, __be32 *p, struct nfs4_cb_recall *args) { struct xdr_stream xdr; struct nfs4_cb_compound_hdr hdr = { @@ -257,7 +257,7 @@ nfs4_xdr_enc_cb_recall(struct rpc_rqst *req, u32 *p, struct nfs4_cb_recall *args static int decode_cb_compound_hdr(struct xdr_stream *xdr, struct nfs4_cb_compound_hdr *hdr){ - u32 *p; + __be32 *p; READ_BUF(8); READ32(hdr->status); @@ -272,7 +272,7 @@ decode_cb_compound_hdr(struct xdr_stream *xdr, struct nfs4_cb_compound_hdr *hdr) static int decode_cb_op_hdr(struct xdr_stream *xdr, enum nfs_opnum4 expected) { - u32 *p; + __be32 *p; u32 op; int32_t nfserr; @@ -291,13 +291,13 @@ decode_cb_op_hdr(struct xdr_stream *xdr, enum nfs_opnum4 expected) } static int -nfs4_xdr_dec_cb_null(struct rpc_rqst *req, u32 *p) +nfs4_xdr_dec_cb_null(struct rpc_rqst *req, __be32 *p) { return 0; } static int -nfs4_xdr_dec_cb_recall(struct rpc_rqst *rqstp, u32 *p) +nfs4_xdr_dec_cb_recall(struct rpc_rqst *rqstp, __be32 *p) { struct xdr_stream xdr; struct nfs4_cb_compound_hdr hdr; @@ -421,7 +421,7 @@ nfsd4_probe_callback(struct nfs4_client *clp) /* Create RPC client */ cb->cb_client = rpc_create(&args); - if (!cb->cb_client) { + if (IS_ERR(cb->cb_client)) { dprintk("NFSD: couldn't create callback client\n"); goto out_err; } @@ -448,10 +448,10 @@ nfsd4_probe_callback(struct nfs4_client *clp) out_rpciod: atomic_dec(&clp->cl_count); rpciod_down(); - cb->cb_client = NULL; out_clnt: rpc_shutdown_client(cb->cb_client); out_err: + cb->cb_client = NULL; dprintk("NFSD: warning: no callback path to client %.*s\n", (int)clp->cl_name.len, clp->cl_name.data); } @@ -461,7 +461,7 @@ nfs4_cb_null(struct rpc_task *task, void *dummy) { struct nfs4_client *clp = (struct nfs4_client *)task->tk_msg.rpc_argp; struct nfs4_callback *cb = &clp->cl_callback; - u32 addr = htonl(cb->cb_addr); + __be32 addr = htonl(cb->cb_addr); dprintk("NFSD: nfs4_cb_null task->tk_status %d\n", task->tk_status); diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index 8333db12caca..0a7bbdc4a10a 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -67,32 +67,32 @@ fh_dup2(struct svc_fh *dst, struct svc_fh *src) *dst = *src; } -static int -do_open_permission(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open *open) +static __be32 +do_open_permission(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open *open, int accmode) { - int accmode, status; + __be32 status; if (open->op_truncate && !(open->op_share_access & NFS4_SHARE_ACCESS_WRITE)) return nfserr_inval; - accmode = MAY_NOP; if (open->op_share_access & NFS4_SHARE_ACCESS_READ) - accmode = MAY_READ; - if (open->op_share_deny & NFS4_SHARE_ACCESS_WRITE) + accmode |= MAY_READ; + if (open->op_share_access & NFS4_SHARE_ACCESS_WRITE) accmode |= (MAY_WRITE | MAY_TRUNC); - accmode |= MAY_OWNER_OVERRIDE; + if (open->op_share_deny & NFS4_SHARE_DENY_WRITE) + accmode |= MAY_WRITE; status = fh_verify(rqstp, current_fh, S_IFREG, accmode); return status; } -static int +static __be32 do_open_lookup(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open *open) { struct svc_fh resfh; - int status; + __be32 status; fh_init(&resfh, NFS4_FHSIZE); open->op_truncate = 0; @@ -124,17 +124,17 @@ do_open_lookup(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_o &resfh.fh_handle.fh_base, resfh.fh_handle.fh_size); - status = do_open_permission(rqstp, current_fh, open); + status = do_open_permission(rqstp, current_fh, open, MAY_NOP); } fh_put(&resfh); return status; } -static int +static __be32 do_open_fhandle(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open *open) { - int status; + __be32 status; /* Only reclaims from previously confirmed clients are valid */ if ((status = nfs4_check_open_reclaim(&open->op_clientid))) @@ -155,16 +155,16 @@ do_open_fhandle(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_ open->op_truncate = (open->op_iattr.ia_valid & ATTR_SIZE) && (open->op_iattr.ia_size == 0); - status = do_open_permission(rqstp, current_fh, open); + status = do_open_permission(rqstp, current_fh, open, MAY_OWNER_OVERRIDE); return status; } -static inline int +static inline __be32 nfsd4_open(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open *open, struct nfs4_stateowner **replay_owner) { - int status; + __be32 status; dprintk("NFSD: nfsd4_open filename %.*s op_stateowner %p\n", (int)open->op_fname.len, open->op_fname.data, open->op_stateowner); @@ -177,7 +177,7 @@ nfsd4_open(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open /* check seqid for replay. set nfs4_owner */ status = nfsd4_process_open1(open); - if (status == NFSERR_REPLAY_ME) { + if (status == nfserr_replay_me) { struct nfs4_replay *rp = &open->op_stateowner->so_replay; fh_put(current_fh); current_fh->fh_handle.fh_size = rp->rp_openfh_len; @@ -188,7 +188,7 @@ nfsd4_open(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open dprintk("nfsd4_open: replay failed" " restoring previous filehandle\n"); else - status = NFSERR_REPLAY_ME; + status = nfserr_replay_me; } if (status) goto out; @@ -261,7 +261,7 @@ out: /* * filehandle-manipulating ops. */ -static inline int +static inline __be32 nfsd4_getfh(struct svc_fh *current_fh, struct svc_fh **getfh) { if (!current_fh->fh_dentry) @@ -271,7 +271,7 @@ nfsd4_getfh(struct svc_fh *current_fh, struct svc_fh **getfh) return nfs_ok; } -static inline int +static inline __be32 nfsd4_putfh(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_putfh *putfh) { fh_put(current_fh); @@ -280,10 +280,10 @@ nfsd4_putfh(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_putf return fh_verify(rqstp, current_fh, 0, MAY_NOP); } -static inline int +static inline __be32 nfsd4_putrootfh(struct svc_rqst *rqstp, struct svc_fh *current_fh) { - int status; + __be32 status; fh_put(current_fh); status = exp_pseudoroot(rqstp->rq_client, current_fh, @@ -291,7 +291,7 @@ nfsd4_putrootfh(struct svc_rqst *rqstp, struct svc_fh *current_fh) return status; } -static inline int +static inline __be32 nfsd4_restorefh(struct svc_fh *current_fh, struct svc_fh *save_fh) { if (!save_fh->fh_dentry) @@ -301,7 +301,7 @@ nfsd4_restorefh(struct svc_fh *current_fh, struct svc_fh *save_fh) return nfs_ok; } -static inline int +static inline __be32 nfsd4_savefh(struct svc_fh *current_fh, struct svc_fh *save_fh) { if (!current_fh->fh_dentry) @@ -314,7 +314,7 @@ nfsd4_savefh(struct svc_fh *current_fh, struct svc_fh *save_fh) /* * misc nfsv4 ops */ -static inline int +static inline __be32 nfsd4_access(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_access *access) { if (access->ac_req_access & ~NFS3_ACCESS_FULL) @@ -324,10 +324,10 @@ nfsd4_access(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_acc return nfsd_access(rqstp, current_fh, &access->ac_resp_access, &access->ac_supported); } -static inline int +static inline __be32 nfsd4_commit(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_commit *commit) { - int status; + __be32 status; u32 *p = (u32 *)commit->co_verf.data; *p++ = nfssvc_boot.tv_sec; @@ -339,11 +339,11 @@ nfsd4_commit(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_com return status; } -static int +static __be32 nfsd4_create(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_create *create) { struct svc_fh resfh; - int status; + __be32 status; dev_t rdev; fh_init(&resfh, NFS4_FHSIZE); @@ -423,10 +423,10 @@ nfsd4_create(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_cre return status; } -static inline int +static inline __be32 nfsd4_getattr(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_getattr *getattr) { - int status; + __be32 status; status = fh_verify(rqstp, current_fh, 0, MAY_NOP); if (status) @@ -442,11 +442,11 @@ nfsd4_getattr(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_ge return nfs_ok; } -static inline int +static inline __be32 nfsd4_link(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct svc_fh *save_fh, struct nfsd4_link *link) { - int status = nfserr_nofilehandle; + __be32 status = nfserr_nofilehandle; if (!save_fh->fh_dentry) return status; @@ -456,11 +456,11 @@ nfsd4_link(struct svc_rqst *rqstp, struct svc_fh *current_fh, return status; } -static int +static __be32 nfsd4_lookupp(struct svc_rqst *rqstp, struct svc_fh *current_fh) { struct svc_fh tmp_fh; - int ret; + __be32 ret; fh_init(&tmp_fh, NFS4_FHSIZE); if((ret = exp_pseudoroot(rqstp->rq_client, &tmp_fh, @@ -474,16 +474,16 @@ nfsd4_lookupp(struct svc_rqst *rqstp, struct svc_fh *current_fh) return nfsd_lookup(rqstp, current_fh, "..", 2, current_fh); } -static inline int +static inline __be32 nfsd4_lookup(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lookup *lookup) { return nfsd_lookup(rqstp, current_fh, lookup->lo_name, lookup->lo_len, current_fh); } -static inline int +static inline __be32 nfsd4_read(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_read *read) { - int status; + __be32 status; /* no need to check permission - this will be done in nfsd_read() */ @@ -508,7 +508,7 @@ out: return status; } -static inline int +static inline __be32 nfsd4_readdir(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_readdir *readdir) { u64 cookie = readdir->rd_cookie; @@ -531,7 +531,7 @@ nfsd4_readdir(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_re return nfs_ok; } -static inline int +static inline __be32 nfsd4_readlink(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_readlink *readlink) { readlink->rl_rqstp = rqstp; @@ -539,10 +539,10 @@ nfsd4_readlink(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_r return nfs_ok; } -static inline int +static inline __be32 nfsd4_remove(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_remove *remove) { - int status; + __be32 status; if (nfs4_in_grace()) return nfserr_grace; @@ -556,11 +556,11 @@ nfsd4_remove(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_rem return status; } -static inline int +static inline __be32 nfsd4_rename(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct svc_fh *save_fh, struct nfsd4_rename *rename) { - int status = nfserr_nofilehandle; + __be32 status = nfserr_nofilehandle; if (!save_fh->fh_dentry) return status; @@ -589,10 +589,10 @@ nfsd4_rename(struct svc_rqst *rqstp, struct svc_fh *current_fh, return status; } -static inline int +static inline __be32 nfsd4_setattr(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_setattr *setattr) { - int status = nfs_ok; + __be32 status = nfs_ok; if (setattr->sa_iattr.ia_valid & ATTR_SIZE) { nfs4_lock_state(); @@ -614,13 +614,13 @@ nfsd4_setattr(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_se return status; } -static inline int +static inline __be32 nfsd4_write(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_write *write) { stateid_t *stateid = &write->wr_stateid; struct file *filp = NULL; u32 *p; - int status = nfs_ok; + __be32 status = nfs_ok; /* no need to check permission - this will be done in nfsd_write() */ @@ -661,12 +661,12 @@ nfsd4_write(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_writ * attributes matched. VERIFY is implemented by mapping NFSERR_SAME * to NFS_OK after the call; NVERIFY by mapping NFSERR_NOT_SAME to NFS_OK. */ -static int +static __be32 nfsd4_verify(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_verify *verify) { - u32 *buf, *p; + __be32 *buf, *p; int count; - int status; + __be32 status; status = fh_verify(rqstp, current_fh, 0, MAY_NOP); if (status) @@ -715,7 +715,7 @@ out_kfree: /* * NULL call. */ -static int +static __be32 nfsd4_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) { return nfs_ok; @@ -731,7 +731,7 @@ static inline void nfsd4_increment_op_stats(u32 opnum) /* * COMPOUND call. */ -static int +static __be32 nfsd4_proc_compound(struct svc_rqst *rqstp, struct nfsd4_compoundargs *args, struct nfsd4_compoundres *resp) @@ -741,7 +741,7 @@ nfsd4_proc_compound(struct svc_rqst *rqstp, struct svc_fh *save_fh = NULL; struct nfs4_stateowner *replay_owner = NULL; int slack_space; /* in words, not bytes! */ - int status; + __be32 status; status = nfserr_resource; current_fh = kmalloc(sizeof(*current_fh), GFP_KERNEL); @@ -937,7 +937,7 @@ nfsd4_proc_compound(struct svc_rqst *rqstp, } encode_op: - if (op->status == NFSERR_REPLAY_ME) { + if (op->status == nfserr_replay_me) { op->replay = &replay_owner->so_replay; nfsd4_encode_replay(resp, op); status = op->status = op->replay->rp_status; diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c index 1cbd2e4ee122..81b8565d3837 100644 --- a/fs/nfsd/nfs4recover.c +++ b/fs/nfsd/nfs4recover.c @@ -83,13 +83,13 @@ md5_to_hex(char *out, char *md5) *out = '\0'; } -int +__be32 nfs4_make_rec_clidname(char *dname, struct xdr_netobj *clname) { struct xdr_netobj cksum; struct hash_desc desc; struct scatterlist sg[1]; - int status = nfserr_resource; + __be32 status = nfserr_resource; dprintk("NFSD: nfs4_make_rec_clidname for %.*s\n", clname->len, clname->data); @@ -193,7 +193,7 @@ nfsd4_build_dentrylist(void *arg, const char *name, int namlen, struct dentry_list *child; if (name && isdotent(name, namlen)) - return nfs_ok; + return 0; dentry = lookup_one_len(name, parent, namlen); if (IS_ERR(dentry)) return PTR_ERR(dentry); @@ -274,7 +274,7 @@ nfsd4_clear_clid_dir(struct dentry *dir, struct dentry *dentry) * any regular files anyway, just in case the directory was created by * a kernel from the future.... */ nfsd4_list_rec_dir(dentry, nfsd4_remove_clid_file); - mutex_lock(&dir->d_inode->i_mutex); + mutex_lock_nested(&dir->d_inode->i_mutex, I_MUTEX_PARENT); status = vfs_rmdir(dir->d_inode, dentry); mutex_unlock(&dir->d_inode->i_mutex); return status; @@ -333,14 +333,14 @@ purge_old(struct dentry *parent, struct dentry *child) int status; if (nfs4_has_reclaimed_state(child->d_name.name)) - return nfs_ok; + return 0; status = nfsd4_clear_clid_dir(parent, child); if (status) printk("failed to remove client recovery directory %s\n", child->d_name.name); /* Keep trying, success or failure: */ - return nfs_ok; + return 0; } void @@ -365,10 +365,10 @@ load_recdir(struct dentry *parent, struct dentry *child) printk("nfsd4: illegal name %s in recovery directory\n", child->d_name.name); /* Keep trying; maybe the others are OK: */ - return nfs_ok; + return 0; } nfs4_client_to_reclaim(child->d_name.name); - return nfs_ok; + return 0; } int diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index ebcf226a9e4a..293b6495829f 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -710,10 +710,10 @@ out_err: * as described above. * */ -int +__be32 nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_setclientid *setclid) { - u32 ip_addr = rqstp->rq_addr.sin_addr.s_addr; + __be32 ip_addr = rqstp->rq_addr.sin_addr.s_addr; struct xdr_netobj clname = { .len = setclid->se_namelen, .data = setclid->se_name, @@ -721,7 +721,7 @@ nfsd4_setclientid(struct svc_rqst *rqstp, struct nfsd4_setclientid *setclid) nfs4_verifier clverifier = setclid->se_verf; unsigned int strhashval; struct nfs4_client *conf, *unconf, *new; - int status; + __be32 status; char dname[HEXDIR_LEN]; if (!check_name(clname)) @@ -875,14 +875,14 @@ out: * * NOTE: callback information will be processed here in a future patch */ -int +__be32 nfsd4_setclientid_confirm(struct svc_rqst *rqstp, struct nfsd4_setclientid_confirm *setclientid_confirm) { - u32 ip_addr = rqstp->rq_addr.sin_addr.s_addr; + __be32 ip_addr = rqstp->rq_addr.sin_addr.s_addr; struct nfs4_client *conf, *unconf; nfs4_verifier confirm = setclientid_confirm->sc_confirm; clientid_t * clid = &setclientid_confirm->sc_clientid; - int status; + __be32 status; if (STALE_CLIENTID(clid)) return nfserr_stale_clientid; @@ -1280,13 +1280,13 @@ test_share(struct nfs4_stateid *stp, struct nfsd4_open *open) { * Called to check deny when READ with all zero stateid or * WRITE with all zero or all one stateid */ -static int +static __be32 nfs4_share_conflict(struct svc_fh *current_fh, unsigned int deny_type) { struct inode *ino = current_fh->fh_dentry->d_inode; struct nfs4_file *fp; struct nfs4_stateid *stp; - int ret; + __be32 ret; dprintk("NFSD: nfs4_share_conflict\n"); @@ -1444,7 +1444,7 @@ static struct lock_manager_operations nfsd_lease_mng_ops = { }; -int +__be32 nfsd4_process_open1(struct nfsd4_open *open) { clientid_t *clientid = &open->op_clientid; @@ -1477,7 +1477,7 @@ nfsd4_process_open1(struct nfsd4_open *open) } if (open->op_seqid == sop->so_seqid - 1) { if (sop->so_replay.rp_buflen) - return NFSERR_REPLAY_ME; + return nfserr_replay_me; /* The original OPEN failed so spectacularly * that we don't even have replay data saved! * Therefore, we have no choice but to continue @@ -1501,7 +1501,7 @@ renew: return nfs_ok; } -static inline int +static inline __be32 nfs4_check_delegmode(struct nfs4_delegation *dp, int flags) { if ((flags & WR_STATE) && (dp->dl_type == NFS4_OPEN_DELEGATE_READ)) @@ -1522,12 +1522,12 @@ find_delegation_file(struct nfs4_file *fp, stateid_t *stid) return NULL; } -static int +static __be32 nfs4_check_deleg(struct nfs4_file *fp, struct nfsd4_open *open, struct nfs4_delegation **dp) { int flags; - int status = nfserr_bad_stateid; + __be32 status = nfserr_bad_stateid; *dp = find_delegation_file(fp, &open->op_delegate_stateid); if (*dp == NULL) @@ -1546,11 +1546,11 @@ out: return nfs_ok; } -static int +static __be32 nfs4_check_open(struct nfs4_file *fp, struct nfsd4_open *open, struct nfs4_stateid **stpp) { struct nfs4_stateid *local; - int status = nfserr_share_denied; + __be32 status = nfserr_share_denied; struct nfs4_stateowner *sop = open->op_stateowner; list_for_each_entry(local, &fp->fi_stateids, st_perfile) { @@ -1575,7 +1575,7 @@ nfs4_alloc_stateid(void) return kmem_cache_alloc(stateid_slab, GFP_KERNEL); } -static int +static __be32 nfs4_new_open(struct svc_rqst *rqstp, struct nfs4_stateid **stpp, struct nfs4_delegation *dp, struct svc_fh *cur_fh, int flags) @@ -1590,7 +1590,7 @@ nfs4_new_open(struct svc_rqst *rqstp, struct nfs4_stateid **stpp, get_file(dp->dl_vfs_file); stp->st_vfs_file = dp->dl_vfs_file; } else { - int status; + __be32 status; status = nfsd_open(rqstp, cur_fh, S_IFREG, flags, &stp->st_vfs_file); if (status) { @@ -1604,7 +1604,7 @@ nfs4_new_open(struct svc_rqst *rqstp, struct nfs4_stateid **stpp, return 0; } -static inline int +static inline __be32 nfsd4_truncate(struct svc_rqst *rqstp, struct svc_fh *fh, struct nfsd4_open *open) { @@ -1619,22 +1619,22 @@ nfsd4_truncate(struct svc_rqst *rqstp, struct svc_fh *fh, return nfsd_setattr(rqstp, fh, &iattr, 0, (time_t)0); } -static int +static __be32 nfs4_upgrade_open(struct svc_rqst *rqstp, struct svc_fh *cur_fh, struct nfs4_stateid *stp, struct nfsd4_open *open) { struct file *filp = stp->st_vfs_file; struct inode *inode = filp->f_dentry->d_inode; unsigned int share_access, new_writer; - int status; + __be32 status; set_access(&share_access, stp->st_access_bmap); new_writer = (~share_access) & open->op_share_access & NFS4_SHARE_ACCESS_WRITE; if (new_writer) { - status = get_write_access(inode); - if (status) - return nfserrno(status); + int err = get_write_access(inode); + if (err) + return nfserrno(err); } status = nfsd4_truncate(rqstp, cur_fh, open); if (status) { @@ -1738,14 +1738,14 @@ out: /* * called with nfs4_lock_state() held. */ -int +__be32 nfsd4_process_open2(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open *open) { struct nfs4_file *fp = NULL; struct inode *ino = current_fh->fh_dentry->d_inode; struct nfs4_stateid *stp = NULL; struct nfs4_delegation *dp = NULL; - int status; + __be32 status; status = nfserr_inval; if (!access_valid(open->op_share_access) @@ -1833,11 +1833,11 @@ static struct work_struct laundromat_work; static void laundromat_main(void *); static DECLARE_WORK(laundromat_work, laundromat_main, NULL); -int +__be32 nfsd4_renew(clientid_t *clid) { struct nfs4_client *clp; - int status; + __be32 status; nfs4_lock_state(); dprintk("process_renew(%08x/%08x): starting\n", @@ -1996,9 +1996,9 @@ access_permit_write(unsigned long access_bmap) } static -int nfs4_check_openmode(struct nfs4_stateid *stp, int flags) +__be32 nfs4_check_openmode(struct nfs4_stateid *stp, int flags) { - int status = nfserr_openmode; + __be32 status = nfserr_openmode; if ((flags & WR_STATE) && (!access_permit_write(stp->st_access_bmap))) goto out; @@ -2009,7 +2009,7 @@ out: return status; } -static inline int +static inline __be32 check_special_stateids(svc_fh *current_fh, stateid_t *stateid, int flags) { /* Trying to call delegreturn with a special stateid? Yuch: */ @@ -2043,14 +2043,14 @@ io_during_grace_disallowed(struct inode *inode, int flags) /* * Checks for stateid operations */ -int +__be32 nfs4_preprocess_stateid_op(struct svc_fh *current_fh, stateid_t *stateid, int flags, struct file **filpp) { struct nfs4_stateid *stp = NULL; struct nfs4_delegation *dp = NULL; stateid_t *stidp; struct inode *ino = current_fh->fh_dentry->d_inode; - int status; + __be32 status; dprintk("NFSD: preprocess_stateid_op: stateid = (%08x/%08x/%08x/%08x)\n", stateid->si_boot, stateid->si_stateownerid, @@ -2125,7 +2125,7 @@ setlkflg (int type) /* * Checks for sequence id mutating operations. */ -static int +static __be32 nfs4_preprocess_seqid_op(struct svc_fh *current_fh, u32 seqid, stateid_t *stateid, int flags, struct nfs4_stateowner **sopp, struct nfs4_stateid **stpp, struct nfsd4_lock *lock) { struct nfs4_stateid *stp; @@ -2169,7 +2169,7 @@ nfs4_preprocess_seqid_op(struct svc_fh *current_fh, u32 seqid, stateid_t *statei clientid_t *lockclid = &lock->v.new.clientid; struct nfs4_client *clp = sop->so_client; int lkflg = 0; - int status; + __be32 status; lkflg = setlkflg(lock->lk_type); @@ -2233,7 +2233,7 @@ check_replay: if (seqid == sop->so_seqid - 1) { dprintk("NFSD: preprocess_seqid_op: retransmission?\n"); /* indicate replay to calling function */ - return NFSERR_REPLAY_ME; + return nfserr_replay_me; } printk("NFSD: preprocess_seqid_op: bad seqid (expected %d, got %d)\n", sop->so_seqid, seqid); @@ -2241,10 +2241,10 @@ check_replay: return nfserr_bad_seqid; } -int +__be32 nfsd4_open_confirm(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open_confirm *oc, struct nfs4_stateowner **replay_owner) { - int status; + __be32 status; struct nfs4_stateowner *sop; struct nfs4_stateid *stp; @@ -2310,10 +2310,10 @@ reset_union_bmap_deny(unsigned long deny, unsigned long *bmap) } } -int +__be32 nfsd4_open_downgrade(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_open_downgrade *od, struct nfs4_stateowner **replay_owner) { - int status; + __be32 status; struct nfs4_stateid *stp; unsigned int share_access; @@ -2365,10 +2365,10 @@ out: /* * nfs4_unlock_state() called after encode */ -int +__be32 nfsd4_close(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_close *close, struct nfs4_stateowner **replay_owner) { - int status; + __be32 status; struct nfs4_stateid *stp; dprintk("NFSD: nfsd4_close on file %.*s\n", @@ -2404,10 +2404,10 @@ out: return status; } -int +__be32 nfsd4_delegreturn(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_delegreturn *dr) { - int status; + __be32 status; if ((status = fh_verify(rqstp, current_fh, S_IFREG, 0))) goto out; @@ -2635,7 +2635,7 @@ check_lock_length(u64 offset, u64 length) /* * LOCK operation */ -int +__be32 nfsd4_lock(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lock *lock, struct nfs4_stateowner **replay_owner) { struct nfs4_stateowner *open_sop = NULL; @@ -2644,8 +2644,9 @@ nfsd4_lock(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lock struct file *filp; struct file_lock file_lock; struct file_lock conflock; - int status = 0; + __be32 status = 0; unsigned int strhashval; + int err; dprintk("NFSD: nfsd4_lock: start=%Ld length=%Ld\n", (long long) lock->lk_offset, @@ -2758,13 +2759,14 @@ nfsd4_lock(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lock * locks_copy_lock: */ conflock.fl_ops = NULL; conflock.fl_lmops = NULL; - status = posix_lock_file_conf(filp, &file_lock, &conflock); + err = posix_lock_file_conf(filp, &file_lock, &conflock); dprintk("NFSD: nfsd4_lock: posix_lock_file_conf status %d\n",status); - switch (-status) { + switch (-err) { case 0: /* success! */ update_stateid(&lock_stp->st_stateid); memcpy(&lock->lk_resp_stateid, &lock_stp->st_stateid, sizeof(stateid_t)); + status = 0; break; case (EAGAIN): /* conflock holds conflicting lock */ status = nfserr_denied; @@ -2775,7 +2777,7 @@ nfsd4_lock(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lock status = nfserr_deadlock; break; default: - dprintk("NFSD: nfsd4_lock: posix_lock_file_conf() failed! status %d\n",status); + dprintk("NFSD: nfsd4_lock: posix_lock_file_conf() failed! status %d\n",err); status = nfserr_resource; break; } @@ -2793,14 +2795,14 @@ out: /* * LOCKT operation */ -int +__be32 nfsd4_lockt(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lockt *lockt) { struct inode *inode; struct file file; struct file_lock file_lock; struct file_lock conflock; - int status; + __be32 status; if (nfs4_in_grace()) return nfserr_grace; @@ -2873,13 +2875,14 @@ out: return status; } -int +__be32 nfsd4_locku(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_locku *locku, struct nfs4_stateowner **replay_owner) { struct nfs4_stateid *stp; struct file *filp = NULL; struct file_lock file_lock; - int status; + __be32 status; + int err; dprintk("NFSD: nfsd4_locku: start=%Ld length=%Ld\n", (long long) locku->lu_offset, @@ -2917,8 +2920,8 @@ nfsd4_locku(struct svc_rqst *rqstp, struct svc_fh *current_fh, struct nfsd4_lock /* * Try to unlock the file in the VFS. */ - status = posix_lock_file(filp, &file_lock); - if (status) { + err = posix_lock_file(filp, &file_lock); + if (err) { dprintk("NFSD: nfs4_locku: posix_lock_file failed!\n"); goto out_nfserr; } @@ -2937,7 +2940,7 @@ out: return status; out_nfserr: - status = nfserrno(status); + status = nfserrno(err); goto out; } @@ -2965,7 +2968,7 @@ out: return status; } -int +__be32 nfsd4_release_lockowner(struct svc_rqst *rqstp, struct nfsd4_release_lockowner *rlockowner) { clientid_t *clid = &rlockowner->rl_clientid; @@ -2974,7 +2977,7 @@ nfsd4_release_lockowner(struct svc_rqst *rqstp, struct nfsd4_release_lockowner * struct xdr_netobj *owner = &rlockowner->rl_owner; struct list_head matches; int i; - int status; + __be32 status; dprintk("nfsd4_release_lockowner clientid: (%08x/%08x):\n", clid->cl_boot, clid->cl_id); @@ -3111,7 +3114,7 @@ nfs4_find_reclaim_client(clientid_t *clid) /* * Called from OPEN. Look for clientid in reclaim list. */ -int +__be32 nfs4_check_open_reclaim(clientid_t *clid) { return nfs4_find_reclaim_client(clid) ? nfs_ok : nfserr_reclaim_bad; diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index 41fc241b729a..f3f239db04bb 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -68,8 +68,8 @@ #define NFS4_REFERRAL_FSID_MAJOR 0x8000000ULL #define NFS4_REFERRAL_FSID_MINOR 0x8000000ULL -static int -check_filename(char *str, int len, int err) +static __be32 +check_filename(char *str, int len, __be32 err) { int i; @@ -94,8 +94,8 @@ check_filename(char *str, int len, int err) * consistent with the style used in NFSv2/v3... */ #define DECODE_HEAD \ - u32 *p; \ - int status + __be32 *p; \ + __be32 status #define DECODE_TAIL \ status = 0; \ out: \ @@ -144,13 +144,13 @@ xdr_error: \ } \ } while (0) -static u32 *read_buf(struct nfsd4_compoundargs *argp, int nbytes) +static __be32 *read_buf(struct nfsd4_compoundargs *argp, int nbytes) { /* We want more bytes than seem to be available. * Maybe we need a new page, maybe we have just run out */ int avail = (char*)argp->end - (char*)argp->p; - u32 *p; + __be32 *p; if (avail + argp->pagelen < nbytes) return NULL; if (avail + PAGE_SIZE < nbytes) /* need more than a page !! */ @@ -197,7 +197,7 @@ defer_free(struct nfsd4_compoundargs *argp, return 0; } -static char *savemem(struct nfsd4_compoundargs *argp, u32 *p, int nbytes) +static char *savemem(struct nfsd4_compoundargs *argp, __be32 *p, int nbytes) { void *new = NULL; if (p == argp->tmp) { @@ -217,7 +217,7 @@ static char *savemem(struct nfsd4_compoundargs *argp, u32 *p, int nbytes) } -static int +static __be32 nfsd4_decode_bitmap(struct nfsd4_compoundargs *argp, u32 *bmval) { u32 bmlen; @@ -240,13 +240,14 @@ nfsd4_decode_bitmap(struct nfsd4_compoundargs *argp, u32 *bmval) DECODE_TAIL; } -static int +static __be32 nfsd4_decode_fattr(struct nfsd4_compoundargs *argp, u32 *bmval, struct iattr *iattr, struct nfs4_acl **acl) { int expected_len, len = 0; u32 dummy32; char *buf; + int host_err; DECODE_HEAD; iattr->ia_valid = 0; @@ -280,7 +281,7 @@ nfsd4_decode_fattr(struct nfsd4_compoundargs *argp, u32 *bmval, struct iattr *ia *acl = nfs4_acl_new(); if (*acl == NULL) { - status = -ENOMEM; + host_err = -ENOMEM; goto out_nfserr; } defer_free(argp, (void (*)(const void *))nfs4_acl_free, *acl); @@ -295,20 +296,20 @@ nfsd4_decode_fattr(struct nfsd4_compoundargs *argp, u32 *bmval, struct iattr *ia len += XDR_QUADLEN(dummy32) << 2; READMEM(buf, dummy32); ace.whotype = nfs4_acl_get_whotype(buf, dummy32); - status = 0; + host_err = 0; if (ace.whotype != NFS4_ACL_WHO_NAMED) ace.who = 0; else if (ace.flag & NFS4_ACE_IDENTIFIER_GROUP) - status = nfsd_map_name_to_gid(argp->rqstp, + host_err = nfsd_map_name_to_gid(argp->rqstp, buf, dummy32, &ace.who); else - status = nfsd_map_name_to_uid(argp->rqstp, + host_err = nfsd_map_name_to_uid(argp->rqstp, buf, dummy32, &ace.who); - if (status) + if (host_err) goto out_nfserr; - status = nfs4_acl_add_ace(*acl, ace.type, ace.flag, + host_err = nfs4_acl_add_ace(*acl, ace.type, ace.flag, ace.access_mask, ace.whotype, ace.who); - if (status) + if (host_err) goto out_nfserr; } } else @@ -327,7 +328,7 @@ nfsd4_decode_fattr(struct nfsd4_compoundargs *argp, u32 *bmval, struct iattr *ia READ_BUF(dummy32); len += (XDR_QUADLEN(dummy32) << 2); READMEM(buf, dummy32); - if ((status = nfsd_map_name_to_uid(argp->rqstp, buf, dummy32, &iattr->ia_uid))) + if ((host_err = nfsd_map_name_to_uid(argp->rqstp, buf, dummy32, &iattr->ia_uid))) goto out_nfserr; iattr->ia_valid |= ATTR_UID; } @@ -338,7 +339,7 @@ nfsd4_decode_fattr(struct nfsd4_compoundargs *argp, u32 *bmval, struct iattr *ia READ_BUF(dummy32); len += (XDR_QUADLEN(dummy32) << 2); READMEM(buf, dummy32); - if ((status = nfsd_map_name_to_gid(argp->rqstp, buf, dummy32, &iattr->ia_gid))) + if ((host_err = nfsd_map_name_to_gid(argp->rqstp, buf, dummy32, &iattr->ia_gid))) goto out_nfserr; iattr->ia_valid |= ATTR_GID; } @@ -414,11 +415,11 @@ nfsd4_decode_fattr(struct nfsd4_compoundargs *argp, u32 *bmval, struct iattr *ia DECODE_TAIL; out_nfserr: - status = nfserrno(status); + status = nfserrno(host_err); goto out; } -static int +static __be32 nfsd4_decode_access(struct nfsd4_compoundargs *argp, struct nfsd4_access *access) { DECODE_HEAD; @@ -429,7 +430,7 @@ nfsd4_decode_access(struct nfsd4_compoundargs *argp, struct nfsd4_access *access DECODE_TAIL; } -static int +static __be32 nfsd4_decode_close(struct nfsd4_compoundargs *argp, struct nfsd4_close *close) { DECODE_HEAD; @@ -444,7 +445,7 @@ nfsd4_decode_close(struct nfsd4_compoundargs *argp, struct nfsd4_close *close) } -static int +static __be32 nfsd4_decode_commit(struct nfsd4_compoundargs *argp, struct nfsd4_commit *commit) { DECODE_HEAD; @@ -456,7 +457,7 @@ nfsd4_decode_commit(struct nfsd4_compoundargs *argp, struct nfsd4_commit *commit DECODE_TAIL; } -static int +static __be32 nfsd4_decode_create(struct nfsd4_compoundargs *argp, struct nfsd4_create *create) { DECODE_HEAD; @@ -496,7 +497,7 @@ nfsd4_decode_create(struct nfsd4_compoundargs *argp, struct nfsd4_create *create DECODE_TAIL; } -static inline int +static inline __be32 nfsd4_decode_delegreturn(struct nfsd4_compoundargs *argp, struct nfsd4_delegreturn *dr) { DECODE_HEAD; @@ -508,13 +509,13 @@ nfsd4_decode_delegreturn(struct nfsd4_compoundargs *argp, struct nfsd4_delegretu DECODE_TAIL; } -static inline int +static inline __be32 nfsd4_decode_getattr(struct nfsd4_compoundargs *argp, struct nfsd4_getattr *getattr) { return nfsd4_decode_bitmap(argp, getattr->ga_bmval); } -static int +static __be32 nfsd4_decode_link(struct nfsd4_compoundargs *argp, struct nfsd4_link *link) { DECODE_HEAD; @@ -529,7 +530,7 @@ nfsd4_decode_link(struct nfsd4_compoundargs *argp, struct nfsd4_link *link) DECODE_TAIL; } -static int +static __be32 nfsd4_decode_lock(struct nfsd4_compoundargs *argp, struct nfsd4_lock *lock) { DECODE_HEAD; @@ -568,7 +569,7 @@ nfsd4_decode_lock(struct nfsd4_compoundargs *argp, struct nfsd4_lock *lock) DECODE_TAIL; } -static int +static __be32 nfsd4_decode_lockt(struct nfsd4_compoundargs *argp, struct nfsd4_lockt *lockt) { DECODE_HEAD; @@ -587,7 +588,7 @@ nfsd4_decode_lockt(struct nfsd4_compoundargs *argp, struct nfsd4_lockt *lockt) DECODE_TAIL; } -static int +static __be32 nfsd4_decode_locku(struct nfsd4_compoundargs *argp, struct nfsd4_locku *locku) { DECODE_HEAD; @@ -606,7 +607,7 @@ nfsd4_decode_locku(struct nfsd4_compoundargs *argp, struct nfsd4_locku *locku) DECODE_TAIL; } -static int +static __be32 nfsd4_decode_lookup(struct nfsd4_compoundargs *argp, struct nfsd4_lookup *lookup) { DECODE_HEAD; @@ -621,7 +622,7 @@ nfsd4_decode_lookup(struct nfsd4_compoundargs *argp, struct nfsd4_lookup *lookup DECODE_TAIL; } -static int +static __be32 nfsd4_decode_open(struct nfsd4_compoundargs *argp, struct nfsd4_open *open) { DECODE_HEAD; @@ -699,7 +700,7 @@ nfsd4_decode_open(struct nfsd4_compoundargs *argp, struct nfsd4_open *open) DECODE_TAIL; } -static int +static __be32 nfsd4_decode_open_confirm(struct nfsd4_compoundargs *argp, struct nfsd4_open_confirm *open_conf) { DECODE_HEAD; @@ -713,7 +714,7 @@ nfsd4_decode_open_confirm(struct nfsd4_compoundargs *argp, struct nfsd4_open_con DECODE_TAIL; } -static int +static __be32 nfsd4_decode_open_downgrade(struct nfsd4_compoundargs *argp, struct nfsd4_open_downgrade *open_down) { DECODE_HEAD; @@ -729,7 +730,7 @@ nfsd4_decode_open_downgrade(struct nfsd4_compoundargs *argp, struct nfsd4_open_d DECODE_TAIL; } -static int +static __be32 nfsd4_decode_putfh(struct nfsd4_compoundargs *argp, struct nfsd4_putfh *putfh) { DECODE_HEAD; @@ -744,7 +745,7 @@ nfsd4_decode_putfh(struct nfsd4_compoundargs *argp, struct nfsd4_putfh *putfh) DECODE_TAIL; } -static int +static __be32 nfsd4_decode_read(struct nfsd4_compoundargs *argp, struct nfsd4_read *read) { DECODE_HEAD; @@ -758,7 +759,7 @@ nfsd4_decode_read(struct nfsd4_compoundargs *argp, struct nfsd4_read *read) DECODE_TAIL; } -static int +static __be32 nfsd4_decode_readdir(struct nfsd4_compoundargs *argp, struct nfsd4_readdir *readdir) { DECODE_HEAD; @@ -774,7 +775,7 @@ nfsd4_decode_readdir(struct nfsd4_compoundargs *argp, struct nfsd4_readdir *read DECODE_TAIL; } -static int +static __be32 nfsd4_decode_remove(struct nfsd4_compoundargs *argp, struct nfsd4_remove *remove) { DECODE_HEAD; @@ -789,7 +790,7 @@ nfsd4_decode_remove(struct nfsd4_compoundargs *argp, struct nfsd4_remove *remove DECODE_TAIL; } -static int +static __be32 nfsd4_decode_rename(struct nfsd4_compoundargs *argp, struct nfsd4_rename *rename) { DECODE_HEAD; @@ -809,7 +810,7 @@ nfsd4_decode_rename(struct nfsd4_compoundargs *argp, struct nfsd4_rename *rename DECODE_TAIL; } -static int +static __be32 nfsd4_decode_renew(struct nfsd4_compoundargs *argp, clientid_t *clientid) { DECODE_HEAD; @@ -820,7 +821,7 @@ nfsd4_decode_renew(struct nfsd4_compoundargs *argp, clientid_t *clientid) DECODE_TAIL; } -static int +static __be32 nfsd4_decode_setattr(struct nfsd4_compoundargs *argp, struct nfsd4_setattr *setattr) { DECODE_HEAD; @@ -834,7 +835,7 @@ nfsd4_decode_setattr(struct nfsd4_compoundargs *argp, struct nfsd4_setattr *seta DECODE_TAIL; } -static int +static __be32 nfsd4_decode_setclientid(struct nfsd4_compoundargs *argp, struct nfsd4_setclientid *setclientid) { DECODE_HEAD; @@ -859,7 +860,7 @@ nfsd4_decode_setclientid(struct nfsd4_compoundargs *argp, struct nfsd4_setclient DECODE_TAIL; } -static int +static __be32 nfsd4_decode_setclientid_confirm(struct nfsd4_compoundargs *argp, struct nfsd4_setclientid_confirm *scd_c) { DECODE_HEAD; @@ -872,7 +873,7 @@ nfsd4_decode_setclientid_confirm(struct nfsd4_compoundargs *argp, struct nfsd4_s } /* Also used for NVERIFY */ -static int +static __be32 nfsd4_decode_verify(struct nfsd4_compoundargs *argp, struct nfsd4_verify *verify) { #if 0 @@ -908,7 +909,7 @@ nfsd4_decode_verify(struct nfsd4_compoundargs *argp, struct nfsd4_verify *verify DECODE_TAIL; } -static int +static __be32 nfsd4_decode_write(struct nfsd4_compoundargs *argp, struct nfsd4_write *write) { int avail; @@ -951,15 +952,15 @@ nfsd4_decode_write(struct nfsd4_compoundargs *argp, struct nfsd4_write *write) argp->pagelen -= len; } } - argp->end = (u32*) (argp->rqstp->rq_vec[v].iov_base + argp->rqstp->rq_vec[v].iov_len); - argp->p = (u32*) (argp->rqstp->rq_vec[v].iov_base + (XDR_QUADLEN(len) << 2)); + argp->end = (__be32*) (argp->rqstp->rq_vec[v].iov_base + argp->rqstp->rq_vec[v].iov_len); + argp->p = (__be32*) (argp->rqstp->rq_vec[v].iov_base + (XDR_QUADLEN(len) << 2)); argp->rqstp->rq_vec[v].iov_len = len; write->wr_vlen = v+1; DECODE_TAIL; } -static int +static __be32 nfsd4_decode_release_lockowner(struct nfsd4_compoundargs *argp, struct nfsd4_release_lockowner *rlockowner) { DECODE_HEAD; @@ -973,7 +974,7 @@ nfsd4_decode_release_lockowner(struct nfsd4_compoundargs *argp, struct nfsd4_rel DECODE_TAIL; } -static int +static __be32 nfsd4_decode_compound(struct nfsd4_compoundargs *argp) { DECODE_HEAD; @@ -1179,7 +1180,7 @@ nfsd4_decode_compound(struct nfsd4_compoundargs *argp) * task to translate them into Linux-specific versions which are more * consistent with the style used in NFSv2/v3... */ -#define ENCODE_HEAD u32 *p +#define ENCODE_HEAD __be32 *p #define WRITE32(n) *p++ = htonl(n) #define WRITE64(n) do { \ @@ -1209,8 +1210,8 @@ nfsd4_decode_compound(struct nfsd4_compoundargs *argp) * Header routine to setup seqid operation replay cache */ #define ENCODE_SEQID_OP_HEAD \ - u32 *p; \ - u32 *save; \ + __be32 *p; \ + __be32 *save; \ \ save = resp->p; @@ -1234,11 +1235,11 @@ nfsd4_decode_compound(struct nfsd4_compoundargs *argp) /* Encode as an array of strings the string given with components * seperated @sep. */ -static int nfsd4_encode_components(char sep, char *components, - u32 **pp, int *buflen) +static __be32 nfsd4_encode_components(char sep, char *components, + __be32 **pp, int *buflen) { - u32 *p = *pp; - u32 *countp = p; + __be32 *p = *pp; + __be32 *countp = p; int strlen, count=0; char *str, *end; @@ -1271,11 +1272,11 @@ static int nfsd4_encode_components(char sep, char *components, /* * encode a location element of a fs_locations structure */ -static int nfsd4_encode_fs_location4(struct nfsd4_fs_location *location, - u32 **pp, int *buflen) +static __be32 nfsd4_encode_fs_location4(struct nfsd4_fs_location *location, + __be32 **pp, int *buflen) { - int status; - u32 *p = *pp; + __be32 status; + __be32 *p = *pp; status = nfsd4_encode_components(':', location->hosts, &p, buflen); if (status) @@ -1292,16 +1293,15 @@ static int nfsd4_encode_fs_location4(struct nfsd4_fs_location *location, * Returned string is safe to use as long as the caller holds a reference * to @exp. */ -static char *nfsd4_path(struct svc_rqst *rqstp, struct svc_export *exp) +static char *nfsd4_path(struct svc_rqst *rqstp, struct svc_export *exp, __be32 *stat) { struct svc_fh tmp_fh; char *path, *rootpath; - int stat; fh_init(&tmp_fh, NFS4_FHSIZE); - stat = exp_pseudoroot(rqstp->rq_client, &tmp_fh, &rqstp->rq_chandle); - if (stat) - return ERR_PTR(stat); + *stat = exp_pseudoroot(rqstp->rq_client, &tmp_fh, &rqstp->rq_chandle); + if (*stat) + return NULL; rootpath = tmp_fh.fh_export->ex_path; path = exp->ex_path; @@ -1309,7 +1309,8 @@ static char *nfsd4_path(struct svc_rqst *rqstp, struct svc_export *exp) if (strncmp(path, rootpath, strlen(rootpath))) { printk("nfsd: fs_locations failed;" "%s is not contained in %s\n", path, rootpath); - return ERR_PTR(-EOPNOTSUPP); + *stat = nfserr_notsupp; + return NULL; } return path + strlen(rootpath); @@ -1318,17 +1319,18 @@ static char *nfsd4_path(struct svc_rqst *rqstp, struct svc_export *exp) /* * encode a fs_locations structure */ -static int nfsd4_encode_fs_locations(struct svc_rqst *rqstp, +static __be32 nfsd4_encode_fs_locations(struct svc_rqst *rqstp, struct svc_export *exp, - u32 **pp, int *buflen) + __be32 **pp, int *buflen) { - int status, i; - u32 *p = *pp; + __be32 status; + int i; + __be32 *p = *pp; struct nfsd4_fs_locations *fslocs = &exp->ex_fslocs; - char *root = nfsd4_path(rqstp, exp); + char *root = nfsd4_path(rqstp, exp, &status); - if (IS_ERR(root)) - return PTR_ERR(root); + if (status) + return status; status = nfsd4_encode_components('/', root, &p, buflen); if (status) return status; @@ -1352,9 +1354,9 @@ static u32 nfs4_ftypes[16] = { NF4SOCK, NF4BAD, NF4LNK, NF4BAD, }; -static int +static __be32 nfsd4_encode_name(struct svc_rqst *rqstp, int whotype, uid_t id, int group, - u32 **p, int *buflen) + __be32 **p, int *buflen) { int status; @@ -1374,21 +1376,21 @@ nfsd4_encode_name(struct svc_rqst *rqstp, int whotype, uid_t id, int group, return 0; } -static inline int -nfsd4_encode_user(struct svc_rqst *rqstp, uid_t uid, u32 **p, int *buflen) +static inline __be32 +nfsd4_encode_user(struct svc_rqst *rqstp, uid_t uid, __be32 **p, int *buflen) { return nfsd4_encode_name(rqstp, NFS4_ACL_WHO_NAMED, uid, 0, p, buflen); } -static inline int -nfsd4_encode_group(struct svc_rqst *rqstp, uid_t gid, u32 **p, int *buflen) +static inline __be32 +nfsd4_encode_group(struct svc_rqst *rqstp, uid_t gid, __be32 **p, int *buflen) { return nfsd4_encode_name(rqstp, NFS4_ACL_WHO_NAMED, gid, 1, p, buflen); } -static inline int +static inline __be32 nfsd4_encode_aclname(struct svc_rqst *rqstp, int whotype, uid_t id, int group, - u32 **p, int *buflen) + __be32 **p, int *buflen) { return nfsd4_encode_name(rqstp, whotype, id, group, p, buflen); } @@ -1397,7 +1399,7 @@ nfsd4_encode_aclname(struct svc_rqst *rqstp, int whotype, uid_t id, int group, FATTR4_WORD0_RDATTR_ERROR) #define WORD1_ABSENT_FS_ATTRS FATTR4_WORD1_MOUNTED_ON_FILEID -static int fattr_handle_absent_fs(u32 *bmval0, u32 *bmval1, u32 *rdattr_err) +static __be32 fattr_handle_absent_fs(u32 *bmval0, u32 *bmval1, u32 *rdattr_err) { /* As per referral draft: */ if (*bmval0 & ~WORD0_ABSENT_FS_ATTRS || @@ -1420,9 +1422,9 @@ static int fattr_handle_absent_fs(u32 *bmval0, u32 *bmval1, u32 *rdattr_err) * @countp is the buffer size in _words_; upon successful return this becomes * replaced with the number of words written. */ -int +__be32 nfsd4_encode_fattr(struct svc_fh *fhp, struct svc_export *exp, - struct dentry *dentry, u32 *buffer, int *countp, u32 *bmval, + struct dentry *dentry, __be32 *buffer, int *countp, u32 *bmval, struct svc_rqst *rqstp) { u32 bmval0 = bmval[0]; @@ -1431,12 +1433,13 @@ nfsd4_encode_fattr(struct svc_fh *fhp, struct svc_export *exp, struct svc_fh tempfh; struct kstatfs statfs; int buflen = *countp << 2; - u32 *attrlenp; + __be32 *attrlenp; u32 dummy; u64 dummy64; u32 rdattr_err = 0; - u32 *p = buffer; - int status; + __be32 *p = buffer; + __be32 status; + int err; int aclsupport = 0; struct nfs4_acl *acl = NULL; @@ -1450,14 +1453,14 @@ nfsd4_encode_fattr(struct svc_fh *fhp, struct svc_export *exp, goto out; } - status = vfs_getattr(exp->ex_mnt, dentry, &stat); - if (status) + err = vfs_getattr(exp->ex_mnt, dentry, &stat); + if (err) goto out_nfserr; if ((bmval0 & (FATTR4_WORD0_FILES_FREE | FATTR4_WORD0_FILES_TOTAL)) || (bmval1 & (FATTR4_WORD1_SPACE_AVAIL | FATTR4_WORD1_SPACE_FREE | FATTR4_WORD1_SPACE_TOTAL))) { - status = vfs_statfs(dentry, &statfs); - if (status) + err = vfs_statfs(dentry, &statfs); + if (err) goto out_nfserr; } if ((bmval0 & (FATTR4_WORD0_FILEHANDLE | FATTR4_WORD0_FSID)) && !fhp) { @@ -1469,15 +1472,15 @@ nfsd4_encode_fattr(struct svc_fh *fhp, struct svc_export *exp, } if (bmval0 & (FATTR4_WORD0_ACL | FATTR4_WORD0_ACLSUPPORT | FATTR4_WORD0_SUPPORTED_ATTRS)) { - status = nfsd4_get_nfs4_acl(rqstp, dentry, &acl); - aclsupport = (status == 0); + err = nfsd4_get_nfs4_acl(rqstp, dentry, &acl); + aclsupport = (err == 0); if (bmval0 & FATTR4_WORD0_ACL) { - if (status == -EOPNOTSUPP) + if (err == -EOPNOTSUPP) bmval0 &= ~FATTR4_WORD0_ACL; - else if (status == -EINVAL) { + else if (err == -EINVAL) { status = nfserr_attrnotsupp; goto out; - } else if (status != 0) + } else if (err != 0) goto out_nfserr; } } @@ -1817,7 +1820,7 @@ out: fh_put(&tempfh); return status; out_nfserr: - status = nfserrno(status); + status = nfserrno(err); goto out; out_resource: *countp = 0; @@ -1828,13 +1831,13 @@ out_serverfault: goto out; } -static int +static __be32 nfsd4_encode_dirent_fattr(struct nfsd4_readdir *cd, - const char *name, int namlen, u32 *p, int *buflen) + const char *name, int namlen, __be32 *p, int *buflen) { struct svc_export *exp = cd->rd_fhp->fh_export; struct dentry *dentry; - int nfserr; + __be32 nfserr; dentry = lookup_one_len(name, cd->rd_fhp->fh_dentry, namlen); if (IS_ERR(dentry)) @@ -1863,10 +1866,10 @@ out_put: return nfserr; } -static u32 * -nfsd4_encode_rdattr_error(u32 *p, int buflen, int nfserr) +static __be32 * +nfsd4_encode_rdattr_error(__be32 *p, int buflen, __be32 nfserr) { - u32 *attrlenp; + __be32 *attrlenp; if (buflen < 6) return NULL; @@ -1886,8 +1889,8 @@ nfsd4_encode_dirent(struct readdir_cd *ccd, const char *name, int namlen, { struct nfsd4_readdir *cd = container_of(ccd, struct nfsd4_readdir, common); int buflen; - u32 *p = cd->buffer; - int nfserr = nfserr_toosmall; + __be32 *p = cd->buffer; + __be32 nfserr = nfserr_toosmall; /* In nfsv4, "." and ".." never make it onto the wire.. */ if (name && isdotent(name, namlen)) { @@ -1943,7 +1946,7 @@ fail: } static void -nfsd4_encode_access(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_access *access) +nfsd4_encode_access(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_access *access) { ENCODE_HEAD; @@ -1956,7 +1959,7 @@ nfsd4_encode_access(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_acc } static void -nfsd4_encode_close(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_close *close) +nfsd4_encode_close(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_close *close) { ENCODE_SEQID_OP_HEAD; @@ -1971,7 +1974,7 @@ nfsd4_encode_close(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_clos static void -nfsd4_encode_commit(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_commit *commit) +nfsd4_encode_commit(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_commit *commit) { ENCODE_HEAD; @@ -1983,7 +1986,7 @@ nfsd4_encode_commit(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_com } static void -nfsd4_encode_create(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_create *create) +nfsd4_encode_create(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_create *create) { ENCODE_HEAD; @@ -1997,8 +2000,8 @@ nfsd4_encode_create(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_cre } } -static int -nfsd4_encode_getattr(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_getattr *getattr) +static __be32 +nfsd4_encode_getattr(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_getattr *getattr) { struct svc_fh *fhp = getattr->ga_fhp; int buflen; @@ -2016,7 +2019,7 @@ nfsd4_encode_getattr(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_ge } static void -nfsd4_encode_getfh(struct nfsd4_compoundres *resp, int nfserr, struct svc_fh *fhp) +nfsd4_encode_getfh(struct nfsd4_compoundres *resp, __be32 nfserr, struct svc_fh *fhp) { unsigned int len; ENCODE_HEAD; @@ -2056,7 +2059,7 @@ nfsd4_encode_lock_denied(struct nfsd4_compoundres *resp, struct nfsd4_lock_denie } static void -nfsd4_encode_lock(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_lock *lock) +nfsd4_encode_lock(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_lock *lock) { ENCODE_SEQID_OP_HEAD; @@ -2072,14 +2075,14 @@ nfsd4_encode_lock(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_lock } static void -nfsd4_encode_lockt(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_lockt *lockt) +nfsd4_encode_lockt(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_lockt *lockt) { if (nfserr == nfserr_denied) nfsd4_encode_lock_denied(resp, &lockt->lt_denied); } static void -nfsd4_encode_locku(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_locku *locku) +nfsd4_encode_locku(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_locku *locku) { ENCODE_SEQID_OP_HEAD; @@ -2095,7 +2098,7 @@ nfsd4_encode_locku(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_lock static void -nfsd4_encode_link(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_link *link) +nfsd4_encode_link(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_link *link) { ENCODE_HEAD; @@ -2108,7 +2111,7 @@ nfsd4_encode_link(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_link static void -nfsd4_encode_open(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_open *open) +nfsd4_encode_open(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_open *open) { ENCODE_SEQID_OP_HEAD; @@ -2173,7 +2176,7 @@ out: } static void -nfsd4_encode_open_confirm(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_open_confirm *oc) +nfsd4_encode_open_confirm(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_open_confirm *oc) { ENCODE_SEQID_OP_HEAD; @@ -2188,7 +2191,7 @@ nfsd4_encode_open_confirm(struct nfsd4_compoundres *resp, int nfserr, struct nfs } static void -nfsd4_encode_open_downgrade(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_open_downgrade *od) +nfsd4_encode_open_downgrade(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_open_downgrade *od) { ENCODE_SEQID_OP_HEAD; @@ -2202,8 +2205,8 @@ nfsd4_encode_open_downgrade(struct nfsd4_compoundres *resp, int nfserr, struct n ENCODE_SEQID_OP_TAIL(od->od_stateowner); } -static int -nfsd4_encode_read(struct nfsd4_compoundres *resp, int nfserr, +static __be32 +nfsd4_encode_read(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_read *read) { u32 eof; @@ -2267,8 +2270,8 @@ nfsd4_encode_read(struct nfsd4_compoundres *resp, int nfserr, return 0; } -static int -nfsd4_encode_readlink(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_readlink *readlink) +static __be32 +nfsd4_encode_readlink(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_readlink *readlink) { int maxcount; char *page; @@ -2315,12 +2318,12 @@ nfsd4_encode_readlink(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_r return 0; } -static int -nfsd4_encode_readdir(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_readdir *readdir) +static __be32 +nfsd4_encode_readdir(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_readdir *readdir) { int maxcount; loff_t offset; - u32 *page, *savep, *tailbase; + __be32 *page, *savep, *tailbase; ENCODE_HEAD; if (nfserr) @@ -2395,7 +2398,7 @@ err_no_verf: } static void -nfsd4_encode_remove(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_remove *remove) +nfsd4_encode_remove(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_remove *remove) { ENCODE_HEAD; @@ -2407,7 +2410,7 @@ nfsd4_encode_remove(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_rem } static void -nfsd4_encode_rename(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_rename *rename) +nfsd4_encode_rename(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_rename *rename) { ENCODE_HEAD; @@ -2424,7 +2427,7 @@ nfsd4_encode_rename(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_ren * regardless of the error status. */ static void -nfsd4_encode_setattr(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_setattr *setattr) +nfsd4_encode_setattr(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_setattr *setattr) { ENCODE_HEAD; @@ -2443,7 +2446,7 @@ nfsd4_encode_setattr(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_se } static void -nfsd4_encode_setclientid(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_setclientid *scd) +nfsd4_encode_setclientid(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_setclientid *scd) { ENCODE_HEAD; @@ -2462,7 +2465,7 @@ nfsd4_encode_setclientid(struct nfsd4_compoundres *resp, int nfserr, struct nfsd } static void -nfsd4_encode_write(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_write *write) +nfsd4_encode_write(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_write *write) { ENCODE_HEAD; @@ -2478,7 +2481,7 @@ nfsd4_encode_write(struct nfsd4_compoundres *resp, int nfserr, struct nfsd4_writ void nfsd4_encode_operation(struct nfsd4_compoundres *resp, struct nfsd4_op *op) { - u32 *statp; + __be32 *statp; ENCODE_HEAD; RESERVE_SPACE(8); @@ -2616,7 +2619,7 @@ nfsd4_encode_replay(struct nfsd4_compoundres *resp, struct nfsd4_op *op) */ int -nfs4svc_encode_voidres(struct svc_rqst *rqstp, u32 *p, void *dummy) +nfs4svc_encode_voidres(struct svc_rqst *rqstp, __be32 *p, void *dummy) { return xdr_ressize_check(rqstp, p); } @@ -2638,9 +2641,9 @@ void nfsd4_release_compoundargs(struct nfsd4_compoundargs *args) } int -nfs4svc_decode_compoundargs(struct svc_rqst *rqstp, u32 *p, struct nfsd4_compoundargs *args) +nfs4svc_decode_compoundargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd4_compoundargs *args) { - int status; + __be32 status; args->p = p; args->end = rqstp->rq_arg.head[0].iov_base + rqstp->rq_arg.head[0].iov_len; @@ -2659,7 +2662,7 @@ nfs4svc_decode_compoundargs(struct svc_rqst *rqstp, u32 *p, struct nfsd4_compoun } int -nfs4svc_encode_compoundres(struct svc_rqst *rqstp, u32 *p, struct nfsd4_compoundres *resp) +nfs4svc_encode_compoundres(struct svc_rqst *rqstp, __be32 *p, struct nfsd4_compoundres *resp) { /* * All that remains is to write the tag and operation count... diff --git a/fs/nfsd/nfscache.c b/fs/nfsd/nfscache.c index fdf7cf3dfadc..6100bbe27432 100644 --- a/fs/nfsd/nfscache.c +++ b/fs/nfsd/nfscache.c @@ -29,7 +29,7 @@ */ #define CACHESIZE 1024 #define HASHSIZE 64 -#define REQHASH(xid) ((((xid) >> 24) ^ (xid)) & (HASHSIZE-1)) +#define REQHASH(xid) (((((__force __u32)xid) >> 24) ^ ((__force __u32)xid)) & (HASHSIZE-1)) static struct hlist_head * hash_list; static struct list_head lru_head; @@ -127,8 +127,8 @@ nfsd_cache_lookup(struct svc_rqst *rqstp, int type) struct hlist_node *hn; struct hlist_head *rh; struct svc_cacherep *rp; - u32 xid = rqstp->rq_xid, - proto = rqstp->rq_prot, + __be32 xid = rqstp->rq_xid; + u32 proto = rqstp->rq_prot, vers = rqstp->rq_vers, proc = rqstp->rq_proc; unsigned long age; @@ -258,7 +258,7 @@ found_entry: * In this case, nfsd_cache_update is called with statp == NULL. */ void -nfsd_cache_update(struct svc_rqst *rqstp, int cachetype, u32 *statp) +nfsd_cache_update(struct svc_rqst *rqstp, int cachetype, __be32 *statp) { struct svc_cacherep *rp; struct kvec *resv = &rqstp->rq_res.head[0], *cachv; diff --git a/fs/nfsd/nfsfh.c b/fs/nfsd/nfsfh.c index 501d83884530..727ab3bd450d 100644 --- a/fs/nfsd/nfsfh.c +++ b/fs/nfsd/nfsfh.c @@ -76,7 +76,7 @@ static int nfsd_acceptable(void *expv, struct dentry *dentry) * comment in the NFSv3 spec says this is incorrect (implementation notes for * the write call). */ -static inline int +static inline __be32 nfsd_mode_check(struct svc_rqst *rqstp, umode_t mode, int type) { /* Type can be negative when creating hardlinks - not to a dir */ @@ -110,13 +110,13 @@ nfsd_mode_check(struct svc_rqst *rqstp, umode_t mode, int type) * This is only called at the start of an nfsproc call, so fhp points to * a svc_fh which is all 0 except for the over-the-wire file handle. */ -u32 +__be32 fh_verify(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, int access) { struct knfsd_fh *fh = &fhp->fh_handle; struct svc_export *exp = NULL; struct dentry *dentry; - u32 error = 0; + __be32 error = 0; dprintk("nfsd: fh_verify(%s)\n", SVCFH_fmt(fhp)); @@ -315,7 +315,7 @@ static inline void _fh_update_old(struct dentry *dentry, fh->ofh_dirino = 0; } -int +__be32 fh_compose(struct svc_fh *fhp, struct svc_export *exp, struct dentry *dentry, struct svc_fh *ref_fh) { /* ref_fh is a reference file handle. @@ -451,7 +451,7 @@ fh_compose(struct svc_fh *fhp, struct svc_export *exp, struct dentry *dentry, st * Update file handle information after changing a dentry. * This is only called by nfsd_create, nfsd_create_v3 and nfsd_proc_create */ -int +__be32 fh_update(struct svc_fh *fhp) { struct dentry *dentry; diff --git a/fs/nfsd/nfsproc.c b/fs/nfsd/nfsproc.c index 9ee1dab5d44a..ec983b777680 100644 --- a/fs/nfsd/nfsproc.c +++ b/fs/nfsd/nfsproc.c @@ -30,22 +30,22 @@ typedef struct svc_buf svc_buf; #define NFSDDBG_FACILITY NFSDDBG_PROC -static int +static __be32 nfsd_proc_null(struct svc_rqst *rqstp, void *argp, void *resp) { return nfs_ok; } -static int -nfsd_return_attrs(int err, struct nfsd_attrstat *resp) +static __be32 +nfsd_return_attrs(__be32 err, struct nfsd_attrstat *resp) { if (err) return err; return nfserrno(vfs_getattr(resp->fh.fh_export->ex_mnt, resp->fh.fh_dentry, &resp->stat)); } -static int -nfsd_return_dirop(int err, struct nfsd_diropres *resp) +static __be32 +nfsd_return_dirop(__be32 err, struct nfsd_diropres *resp) { if (err) return err; return nfserrno(vfs_getattr(resp->fh.fh_export->ex_mnt, @@ -56,11 +56,11 @@ nfsd_return_dirop(int err, struct nfsd_diropres *resp) * Get a file's attributes * N.B. After this call resp->fh needs an fh_put */ -static int +static __be32 nfsd_proc_getattr(struct svc_rqst *rqstp, struct nfsd_fhandle *argp, struct nfsd_attrstat *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: GETATTR %s\n", SVCFH_fmt(&argp->fh)); fh_copy(&resp->fh, &argp->fh); @@ -72,11 +72,11 @@ nfsd_proc_getattr(struct svc_rqst *rqstp, struct nfsd_fhandle *argp, * Set a file's attributes * N.B. After this call resp->fh needs an fh_put */ -static int +static __be32 nfsd_proc_setattr(struct svc_rqst *rqstp, struct nfsd_sattrargs *argp, struct nfsd_attrstat *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: SETATTR %s, valid=%x, size=%ld\n", SVCFH_fmt(&argp->fh), argp->attrs.ia_valid, (long) argp->attrs.ia_size); @@ -92,11 +92,11 @@ nfsd_proc_setattr(struct svc_rqst *rqstp, struct nfsd_sattrargs *argp, * doesn't exist yet. * N.B. After this call resp->fh needs an fh_put */ -static int +static __be32 nfsd_proc_lookup(struct svc_rqst *rqstp, struct nfsd_diropargs *argp, struct nfsd_diropres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: LOOKUP %s %.*s\n", SVCFH_fmt(&argp->fh), argp->len, argp->name); @@ -112,11 +112,11 @@ nfsd_proc_lookup(struct svc_rqst *rqstp, struct nfsd_diropargs *argp, /* * Read a symlink. */ -static int +static __be32 nfsd_proc_readlink(struct svc_rqst *rqstp, struct nfsd_readlinkargs *argp, struct nfsd_readlinkres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: READLINK %s\n", SVCFH_fmt(&argp->fh)); @@ -132,11 +132,11 @@ nfsd_proc_readlink(struct svc_rqst *rqstp, struct nfsd_readlinkargs *argp, * Read a portion of a file. * N.B. After this call resp->fh needs an fh_put */ -static int +static __be32 nfsd_proc_read(struct svc_rqst *rqstp, struct nfsd_readargs *argp, struct nfsd_readres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: READ %s %d bytes at %d\n", SVCFH_fmt(&argp->fh), @@ -172,11 +172,11 @@ nfsd_proc_read(struct svc_rqst *rqstp, struct nfsd_readargs *argp, * Write data to a file * N.B. After this call resp->fh needs an fh_put */ -static int +static __be32 nfsd_proc_write(struct svc_rqst *rqstp, struct nfsd_writeargs *argp, struct nfsd_attrstat *resp) { - int nfserr; + __be32 nfserr; int stable = 1; dprintk("nfsd: WRITE %s %d bytes at %d\n", @@ -197,7 +197,7 @@ nfsd_proc_write(struct svc_rqst *rqstp, struct nfsd_writeargs *argp, * and the actual create() call in compliance with VFS protocols. * N.B. After this call _both_ argp->fh and resp->fh need an fh_put */ -static int +static __be32 nfsd_proc_create(struct svc_rqst *rqstp, struct nfsd_createargs *argp, struct nfsd_diropres *resp) { @@ -206,7 +206,8 @@ nfsd_proc_create(struct svc_rqst *rqstp, struct nfsd_createargs *argp, struct iattr *attr = &argp->attrs; struct inode *inode; struct dentry *dchild; - int nfserr, type, mode; + int type, mode; + __be32 nfserr; dev_t rdev = 0, wanted = new_decode_dev(attr->ia_size); dprintk("nfsd: CREATE %s %.*s\n", @@ -348,11 +349,11 @@ done: return nfsd_return_dirop(nfserr, resp); } -static int +static __be32 nfsd_proc_remove(struct svc_rqst *rqstp, struct nfsd_diropargs *argp, void *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: REMOVE %s %.*s\n", SVCFH_fmt(&argp->fh), argp->len, argp->name); @@ -363,11 +364,11 @@ nfsd_proc_remove(struct svc_rqst *rqstp, struct nfsd_diropargs *argp, return nfserr; } -static int +static __be32 nfsd_proc_rename(struct svc_rqst *rqstp, struct nfsd_renameargs *argp, void *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: RENAME %s %.*s -> \n", SVCFH_fmt(&argp->ffh), argp->flen, argp->fname); @@ -381,11 +382,11 @@ nfsd_proc_rename(struct svc_rqst *rqstp, struct nfsd_renameargs *argp, return nfserr; } -static int +static __be32 nfsd_proc_link(struct svc_rqst *rqstp, struct nfsd_linkargs *argp, void *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: LINK %s ->\n", SVCFH_fmt(&argp->ffh)); @@ -401,12 +402,12 @@ nfsd_proc_link(struct svc_rqst *rqstp, struct nfsd_linkargs *argp, return nfserr; } -static int +static __be32 nfsd_proc_symlink(struct svc_rqst *rqstp, struct nfsd_symlinkargs *argp, void *resp) { struct svc_fh newfh; - int nfserr; + __be32 nfserr; dprintk("nfsd: SYMLINK %s %.*s -> %.*s\n", SVCFH_fmt(&argp->ffh), argp->flen, argp->fname, @@ -430,11 +431,11 @@ nfsd_proc_symlink(struct svc_rqst *rqstp, struct nfsd_symlinkargs *argp, * Make directory. This operation is not idempotent. * N.B. After this call resp->fh needs an fh_put */ -static int +static __be32 nfsd_proc_mkdir(struct svc_rqst *rqstp, struct nfsd_createargs *argp, struct nfsd_diropres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: MKDIR %s %.*s\n", SVCFH_fmt(&argp->fh), argp->len, argp->name); @@ -454,11 +455,11 @@ nfsd_proc_mkdir(struct svc_rqst *rqstp, struct nfsd_createargs *argp, /* * Remove a directory */ -static int +static __be32 nfsd_proc_rmdir(struct svc_rqst *rqstp, struct nfsd_diropargs *argp, void *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: RMDIR %s %.*s\n", SVCFH_fmt(&argp->fh), argp->len, argp->name); @@ -470,11 +471,12 @@ nfsd_proc_rmdir(struct svc_rqst *rqstp, struct nfsd_diropargs *argp, /* * Read a portion of a directory. */ -static int +static __be32 nfsd_proc_readdir(struct svc_rqst *rqstp, struct nfsd_readdirargs *argp, struct nfsd_readdirres *resp) { - int nfserr, count; + int count; + __be32 nfserr; loff_t offset; dprintk("nfsd: READDIR %s %d bytes at %d\n", @@ -509,11 +511,11 @@ nfsd_proc_readdir(struct svc_rqst *rqstp, struct nfsd_readdirargs *argp, /* * Get file system info */ -static int +static __be32 nfsd_proc_statfs(struct svc_rqst * rqstp, struct nfsd_fhandle *argp, struct nfsd_statfsres *resp) { - int nfserr; + __be32 nfserr; dprintk("nfsd: STATFS %s\n", SVCFH_fmt(&argp->fh)); @@ -579,11 +581,11 @@ struct svc_version nfsd_version2 = { /* * Map errnos to NFS errnos. */ -int +__be32 nfserrno (int errno) { static struct { - int nfserr; + __be32 nfserr; int syserr; } nfs_errtbl[] = { { nfs_ok, 0 }, @@ -615,11 +617,10 @@ nfserrno (int errno) { nfserr_badname, -ESRCH }, { nfserr_io, -ETXTBSY }, { nfserr_notsupp, -EOPNOTSUPP }, - { -1, -EIO } }; int i; - for (i = 0; nfs_errtbl[i].nfserr != -1; i++) { + for (i = 0; i < ARRAY_SIZE(nfs_errtbl); i++) { if (nfs_errtbl[i].syserr == errno) return nfs_errtbl[i].nfserr; } diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c index 013b38996e64..0aaccb03bf76 100644 --- a/fs/nfsd/nfssvc.c +++ b/fs/nfsd/nfssvc.c @@ -491,12 +491,12 @@ out: } int -nfsd_dispatch(struct svc_rqst *rqstp, u32 *statp) +nfsd_dispatch(struct svc_rqst *rqstp, __be32 *statp) { struct svc_procedure *proc; kxdrproc_t xdr; - u32 nfserr; - u32 *nfserrp; + __be32 nfserr; + __be32 *nfserrp; dprintk("nfsd_dispatch: vers %d proc %d\n", rqstp->rq_vers, rqstp->rq_proc); @@ -515,7 +515,7 @@ nfsd_dispatch(struct svc_rqst *rqstp, u32 *statp) /* Decode arguments */ xdr = proc->pc_decode; - if (xdr && !xdr(rqstp, (u32*)rqstp->rq_arg.head[0].iov_base, + if (xdr && !xdr(rqstp, (__be32*)rqstp->rq_arg.head[0].iov_base, rqstp->rq_argp)) { dprintk("nfsd: failed to decode arguments!\n"); nfsd_cache_update(rqstp, RC_NOCACHE, NULL); @@ -528,7 +528,7 @@ nfsd_dispatch(struct svc_rqst *rqstp, u32 *statp) */ nfserrp = rqstp->rq_res.head[0].iov_base + rqstp->rq_res.head[0].iov_len; - rqstp->rq_res.head[0].iov_len += sizeof(u32); + rqstp->rq_res.head[0].iov_len += sizeof(__be32); /* Now call the procedure handler, and encode NFS status. */ nfserr = proc->pc_func(rqstp, rqstp->rq_argp, rqstp->rq_resp); diff --git a/fs/nfsd/nfsxdr.c b/fs/nfsd/nfsxdr.c index 1135c0d14557..56ebb1443e0e 100644 --- a/fs/nfsd/nfsxdr.c +++ b/fs/nfsd/nfsxdr.c @@ -37,8 +37,8 @@ static u32 nfs_ftypes[] = { /* * XDR functions for basic NFS types */ -static u32 * -decode_fh(u32 *p, struct svc_fh *fhp) +static __be32 * +decode_fh(__be32 *p, struct svc_fh *fhp) { fh_init(fhp, NFS_FHSIZE); memcpy(&fhp->fh_handle.fh_base, p, NFS_FHSIZE); @@ -50,13 +50,13 @@ decode_fh(u32 *p, struct svc_fh *fhp) } /* Helper function for NFSv2 ACL code */ -u32 *nfs2svc_decode_fh(u32 *p, struct svc_fh *fhp) +__be32 *nfs2svc_decode_fh(__be32 *p, struct svc_fh *fhp) { return decode_fh(p, fhp); } -static inline u32 * -encode_fh(u32 *p, struct svc_fh *fhp) +static inline __be32 * +encode_fh(__be32 *p, struct svc_fh *fhp) { memcpy(p, &fhp->fh_handle.fh_base, NFS_FHSIZE); return p + (NFS_FHSIZE>> 2); @@ -66,8 +66,8 @@ encode_fh(u32 *p, struct svc_fh *fhp) * Decode a file name and make sure that the path contains * no slashes or null bytes. */ -static inline u32 * -decode_filename(u32 *p, char **namp, int *lenp) +static inline __be32 * +decode_filename(__be32 *p, char **namp, int *lenp) { char *name; int i; @@ -82,8 +82,8 @@ decode_filename(u32 *p, char **namp, int *lenp) return p; } -static inline u32 * -decode_pathname(u32 *p, char **namp, int *lenp) +static inline __be32 * +decode_pathname(__be32 *p, char **namp, int *lenp) { char *name; int i; @@ -98,8 +98,8 @@ decode_pathname(u32 *p, char **namp, int *lenp) return p; } -static inline u32 * -decode_sattr(u32 *p, struct iattr *iap) +static inline __be32 * +decode_sattr(__be32 *p, struct iattr *iap) { u32 tmp, tmp1; @@ -151,8 +151,8 @@ decode_sattr(u32 *p, struct iattr *iap) return p; } -static u32 * -encode_fattr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp, +static __be32 * +encode_fattr(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp, struct kstat *stat) { struct dentry *dentry = fhp->fh_dentry; @@ -195,7 +195,7 @@ encode_fattr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp, } /* Helper function for NFSv2 ACL code */ -u32 *nfs2svc_encode_fattr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) +__be32 *nfs2svc_encode_fattr(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp) { struct kstat stat; vfs_getattr(fhp->fh_export->ex_mnt, fhp->fh_dentry, &stat); @@ -206,13 +206,13 @@ u32 *nfs2svc_encode_fattr(struct svc_rqst *rqstp, u32 *p, struct svc_fh *fhp) * XDR decode functions */ int -nfssvc_decode_void(struct svc_rqst *rqstp, u32 *p, void *dummy) +nfssvc_decode_void(struct svc_rqst *rqstp, __be32 *p, void *dummy) { return xdr_argsize_check(rqstp, p); } int -nfssvc_decode_fhandle(struct svc_rqst *rqstp, u32 *p, struct nfsd_fhandle *args) +nfssvc_decode_fhandle(struct svc_rqst *rqstp, __be32 *p, struct nfsd_fhandle *args) { if (!(p = decode_fh(p, &args->fh))) return 0; @@ -220,7 +220,7 @@ nfssvc_decode_fhandle(struct svc_rqst *rqstp, u32 *p, struct nfsd_fhandle *args) } int -nfssvc_decode_sattrargs(struct svc_rqst *rqstp, u32 *p, +nfssvc_decode_sattrargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_sattrargs *args) { if (!(p = decode_fh(p, &args->fh)) @@ -231,7 +231,7 @@ nfssvc_decode_sattrargs(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_decode_diropargs(struct svc_rqst *rqstp, u32 *p, +nfssvc_decode_diropargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_diropargs *args) { if (!(p = decode_fh(p, &args->fh)) @@ -242,7 +242,7 @@ nfssvc_decode_diropargs(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_decode_readargs(struct svc_rqst *rqstp, u32 *p, +nfssvc_decode_readargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_readargs *args) { unsigned int len; @@ -273,7 +273,7 @@ nfssvc_decode_readargs(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_decode_writeargs(struct svc_rqst *rqstp, u32 *p, +nfssvc_decode_writeargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_writeargs *args) { unsigned int len; @@ -303,7 +303,7 @@ nfssvc_decode_writeargs(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_decode_createargs(struct svc_rqst *rqstp, u32 *p, +nfssvc_decode_createargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_createargs *args) { if (!(p = decode_fh(p, &args->fh)) @@ -315,7 +315,7 @@ nfssvc_decode_createargs(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_decode_renameargs(struct svc_rqst *rqstp, u32 *p, +nfssvc_decode_renameargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_renameargs *args) { if (!(p = decode_fh(p, &args->ffh)) @@ -328,7 +328,7 @@ nfssvc_decode_renameargs(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_decode_readlinkargs(struct svc_rqst *rqstp, u32 *p, struct nfsd_readlinkargs *args) +nfssvc_decode_readlinkargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_readlinkargs *args) { if (!(p = decode_fh(p, &args->fh))) return 0; @@ -338,7 +338,7 @@ nfssvc_decode_readlinkargs(struct svc_rqst *rqstp, u32 *p, struct nfsd_readlinka } int -nfssvc_decode_linkargs(struct svc_rqst *rqstp, u32 *p, +nfssvc_decode_linkargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_linkargs *args) { if (!(p = decode_fh(p, &args->ffh)) @@ -350,7 +350,7 @@ nfssvc_decode_linkargs(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_decode_symlinkargs(struct svc_rqst *rqstp, u32 *p, +nfssvc_decode_symlinkargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_symlinkargs *args) { if (!(p = decode_fh(p, &args->ffh)) @@ -363,7 +363,7 @@ nfssvc_decode_symlinkargs(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_decode_readdirargs(struct svc_rqst *rqstp, u32 *p, +nfssvc_decode_readdirargs(struct svc_rqst *rqstp, __be32 *p, struct nfsd_readdirargs *args) { if (!(p = decode_fh(p, &args->fh))) @@ -382,13 +382,13 @@ nfssvc_decode_readdirargs(struct svc_rqst *rqstp, u32 *p, * XDR encode functions */ int -nfssvc_encode_void(struct svc_rqst *rqstp, u32 *p, void *dummy) +nfssvc_encode_void(struct svc_rqst *rqstp, __be32 *p, void *dummy) { return xdr_ressize_check(rqstp, p); } int -nfssvc_encode_attrstat(struct svc_rqst *rqstp, u32 *p, +nfssvc_encode_attrstat(struct svc_rqst *rqstp, __be32 *p, struct nfsd_attrstat *resp) { p = encode_fattr(rqstp, p, &resp->fh, &resp->stat); @@ -396,7 +396,7 @@ nfssvc_encode_attrstat(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_encode_diropres(struct svc_rqst *rqstp, u32 *p, +nfssvc_encode_diropres(struct svc_rqst *rqstp, __be32 *p, struct nfsd_diropres *resp) { p = encode_fh(p, &resp->fh); @@ -405,7 +405,7 @@ nfssvc_encode_diropres(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_encode_readlinkres(struct svc_rqst *rqstp, u32 *p, +nfssvc_encode_readlinkres(struct svc_rqst *rqstp, __be32 *p, struct nfsd_readlinkres *resp) { *p++ = htonl(resp->len); @@ -421,7 +421,7 @@ nfssvc_encode_readlinkres(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_encode_readres(struct svc_rqst *rqstp, u32 *p, +nfssvc_encode_readres(struct svc_rqst *rqstp, __be32 *p, struct nfsd_readres *resp) { p = encode_fattr(rqstp, p, &resp->fh, &resp->stat); @@ -440,7 +440,7 @@ nfssvc_encode_readres(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_encode_readdirres(struct svc_rqst *rqstp, u32 *p, +nfssvc_encode_readdirres(struct svc_rqst *rqstp, __be32 *p, struct nfsd_readdirres *resp) { xdr_ressize_check(rqstp, p); @@ -453,7 +453,7 @@ nfssvc_encode_readdirres(struct svc_rqst *rqstp, u32 *p, } int -nfssvc_encode_statfsres(struct svc_rqst *rqstp, u32 *p, +nfssvc_encode_statfsres(struct svc_rqst *rqstp, __be32 *p, struct nfsd_statfsres *resp) { struct kstatfs *stat = &resp->stats; @@ -471,7 +471,7 @@ nfssvc_encode_entry(struct readdir_cd *ccd, const char *name, int namlen, loff_t offset, ino_t ino, unsigned int d_type) { struct nfsd_readdirres *cd = container_of(ccd, struct nfsd_readdirres, common); - u32 *p = cd->buffer; + __be32 *p = cd->buffer; int buflen, slen; /* @@ -497,7 +497,7 @@ nfssvc_encode_entry(struct readdir_cd *ccd, const char *name, *p++ = htonl((u32) ino); /* file id */ p = xdr_encode_array(p, name, namlen);/* name length & name */ cd->offset = p; /* remember pointer */ - *p++ = ~(u32) 0; /* offset of next entry */ + *p++ = htonl(~0U); /* offset of next entry */ cd->buflen = buflen; cd->buffer = p; @@ -509,7 +509,7 @@ nfssvc_encode_entry(struct readdir_cd *ccd, const char *name, * XDR release functions */ int -nfssvc_release_fhandle(struct svc_rqst *rqstp, u32 *p, +nfssvc_release_fhandle(struct svc_rqst *rqstp, __be32 *p, struct nfsd_fhandle *resp) { fh_put(&resp->fh); diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 1141bd29e4e3..f21e917bb8ed 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -110,7 +110,7 @@ nfsd_cross_mnt(struct svc_rqst *rqstp, struct dentry **dpp, struct dentry *dentry = *dpp; struct vfsmount *mnt = mntget(exp->ex_mnt); struct dentry *mounts = dget(dentry); - int err = nfs_ok; + int err = 0; while (follow_down(&mnt,&mounts)&&d_mountpoint(mounts)); @@ -148,14 +148,15 @@ out: * clients and is explicitly disallowed for NFSv3 * NeilBrown <neilb@cse.unsw.edu.au> */ -int +__be32 nfsd_lookup(struct svc_rqst *rqstp, struct svc_fh *fhp, const char *name, int len, struct svc_fh *resfh) { struct svc_export *exp; struct dentry *dparent; struct dentry *dentry; - int err; + __be32 err; + int host_err; dprintk("nfsd: nfsd_lookup(fh %s, %.*s)\n", SVCFH_fmt(fhp), len,name); @@ -193,7 +194,7 @@ nfsd_lookup(struct svc_rqst *rqstp, struct svc_fh *fhp, const char *name, exp2 = exp_parent(exp->ex_client, mnt, dentry, &rqstp->rq_chandle); if (IS_ERR(exp2)) { - err = PTR_ERR(exp2); + host_err = PTR_ERR(exp2); dput(dentry); mntput(mnt); goto out_nfserr; @@ -210,14 +211,14 @@ nfsd_lookup(struct svc_rqst *rqstp, struct svc_fh *fhp, const char *name, } else { fh_lock(fhp); dentry = lookup_one_len(name, dparent, len); - err = PTR_ERR(dentry); + host_err = PTR_ERR(dentry); if (IS_ERR(dentry)) goto out_nfserr; /* * check if we have crossed a mount point ... */ if (d_mountpoint(dentry)) { - if ((err = nfsd_cross_mnt(rqstp, &dentry, &exp))) { + if ((host_err = nfsd_cross_mnt(rqstp, &dentry, &exp))) { dput(dentry); goto out_nfserr; } @@ -236,7 +237,7 @@ out: return err; out_nfserr: - err = nfserrno(err); + err = nfserrno(host_err); goto out; } @@ -244,7 +245,7 @@ out_nfserr: * Set various file attributes. * N.B. After this call fhp needs an fh_put */ -int +__be32 nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap, int check_guard, time_t guardtime) { @@ -253,7 +254,8 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap, int accmode = MAY_SATTR; int ftype = 0; int imode; - int err; + __be32 err; + int host_err; int size_change = 0; if (iap->ia_valid & (ATTR_ATIME | ATTR_MTIME | ATTR_SIZE)) @@ -319,19 +321,19 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap, * If we are changing the size of the file, then * we need to break all leases. */ - err = break_lease(inode, FMODE_WRITE | O_NONBLOCK); - if (err == -EWOULDBLOCK) - err = -ETIMEDOUT; - if (err) /* ENOMEM or EWOULDBLOCK */ + host_err = break_lease(inode, FMODE_WRITE | O_NONBLOCK); + if (host_err == -EWOULDBLOCK) + host_err = -ETIMEDOUT; + if (host_err) /* ENOMEM or EWOULDBLOCK */ goto out_nfserr; - err = get_write_access(inode); - if (err) + host_err = get_write_access(inode); + if (host_err) goto out_nfserr; size_change = 1; - err = locks_verify_truncate(inode, NULL, iap->ia_size); - if (err) { + host_err = locks_verify_truncate(inode, NULL, iap->ia_size); + if (host_err) { put_write_access(inode); goto out_nfserr; } @@ -357,8 +359,8 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap, err = nfserr_notsync; if (!check_guard || guardtime == inode->i_ctime.tv_sec) { fh_lock(fhp); - err = notify_change(dentry, iap); - err = nfserrno(err); + host_err = notify_change(dentry, iap); + err = nfserrno(host_err); fh_unlock(fhp); } if (size_change) @@ -370,7 +372,7 @@ out: return err; out_nfserr: - err = nfserrno(err); + err = nfserrno(host_err); goto out; } @@ -420,11 +422,12 @@ out: return error; } -int +__be32 nfsd4_set_nfs4_acl(struct svc_rqst *rqstp, struct svc_fh *fhp, struct nfs4_acl *acl) { - int error; + __be32 error; + int host_error; struct dentry *dentry; struct inode *inode; struct posix_acl *pacl = NULL, *dpacl = NULL; @@ -440,20 +443,20 @@ nfsd4_set_nfs4_acl(struct svc_rqst *rqstp, struct svc_fh *fhp, if (S_ISDIR(inode->i_mode)) flags = NFS4_ACL_DIR; - error = nfs4_acl_nfsv4_to_posix(acl, &pacl, &dpacl, flags); - if (error == -EINVAL) { + host_error = nfs4_acl_nfsv4_to_posix(acl, &pacl, &dpacl, flags); + if (host_error == -EINVAL) { error = nfserr_attrnotsupp; goto out; - } else if (error < 0) + } else if (host_error < 0) goto out_nfserr; - error = set_nfsv4_acl_one(dentry, pacl, POSIX_ACL_XATTR_ACCESS); - if (error < 0) + host_error = set_nfsv4_acl_one(dentry, pacl, POSIX_ACL_XATTR_ACCESS); + if (host_error < 0) goto out_nfserr; if (S_ISDIR(inode->i_mode)) { - error = set_nfsv4_acl_one(dentry, dpacl, POSIX_ACL_XATTR_DEFAULT); - if (error < 0) + host_error = set_nfsv4_acl_one(dentry, dpacl, POSIX_ACL_XATTR_DEFAULT); + if (host_error < 0) goto out_nfserr; } @@ -464,7 +467,7 @@ out: posix_acl_release(dpacl); return (error); out_nfserr: - error = nfserrno(error); + error = nfserrno(host_error); goto out; } @@ -571,14 +574,14 @@ static struct accessmap nfs3_anyaccess[] = { { 0, 0 } }; -int +__be32 nfsd_access(struct svc_rqst *rqstp, struct svc_fh *fhp, u32 *access, u32 *supported) { struct accessmap *map; struct svc_export *export; struct dentry *dentry; u32 query, result = 0, sresult = 0; - unsigned int error; + __be32 error; error = fh_verify(rqstp, fhp, 0, MAY_NOP); if (error) @@ -598,7 +601,7 @@ nfsd_access(struct svc_rqst *rqstp, struct svc_fh *fhp, u32 *access, u32 *suppor query = *access; for (; map->access; map++) { if (map->access & query) { - unsigned int err2; + __be32 err2; sresult |= map->access; @@ -637,13 +640,15 @@ nfsd_access(struct svc_rqst *rqstp, struct svc_fh *fhp, u32 *access, u32 *suppor * The access argument indicates the type of open (read/write/lock) * N.B. After this call fhp needs an fh_put */ -int +__be32 nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, int access, struct file **filp) { struct dentry *dentry; struct inode *inode; - int flags = O_RDONLY|O_LARGEFILE, err; + int flags = O_RDONLY|O_LARGEFILE; + __be32 err; + int host_err; /* * If we get here, then the client has already done an "open", @@ -673,10 +678,10 @@ nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, * Check to see if there are any leases on this file. * This may block while leases are broken. */ - err = break_lease(inode, O_NONBLOCK | ((access & MAY_WRITE) ? FMODE_WRITE : 0)); - if (err == -EWOULDBLOCK) - err = -ETIMEDOUT; - if (err) /* NOMEM or WOULDBLOCK */ + host_err = break_lease(inode, O_NONBLOCK | ((access & MAY_WRITE) ? FMODE_WRITE : 0)); + if (host_err == -EWOULDBLOCK) + host_err = -ETIMEDOUT; + if (host_err) /* NOMEM or WOULDBLOCK */ goto out_nfserr; if (access & MAY_WRITE) { @@ -689,10 +694,9 @@ nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, } *filp = dentry_open(dget(dentry), mntget(fhp->fh_export->ex_mnt), flags); if (IS_ERR(*filp)) - err = PTR_ERR(*filp); + host_err = PTR_ERR(*filp); out_nfserr: - if (err) - err = nfserrno(err); + err = nfserrno(host_err); out: return err; } @@ -830,14 +834,15 @@ nfsd_read_actor(read_descriptor_t *desc, struct page *page, unsigned long offset return size; } -static int +static __be32 nfsd_vfs_read(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, loff_t offset, struct kvec *vec, int vlen, unsigned long *count) { struct inode *inode; struct raparms *ra; mm_segment_t oldfs; - int err; + __be32 err; + int host_err; err = nfserr_perm; inode = file->f_dentry->d_inode; @@ -855,12 +860,12 @@ nfsd_vfs_read(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, if (file->f_op->sendfile && rqstp->rq_sendfile_ok) { rqstp->rq_resused = 1; - err = file->f_op->sendfile(file, &offset, *count, + host_err = file->f_op->sendfile(file, &offset, *count, nfsd_read_actor, rqstp); } else { oldfs = get_fs(); set_fs(KERNEL_DS); - err = vfs_readv(file, (struct iovec __user *)vec, vlen, &offset); + host_err = vfs_readv(file, (struct iovec __user *)vec, vlen, &offset); set_fs(oldfs); } @@ -874,13 +879,13 @@ nfsd_vfs_read(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, spin_unlock(&rab->pb_lock); } - if (err >= 0) { - nfsdstats.io_read += err; - *count = err; + if (host_err >= 0) { + nfsdstats.io_read += host_err; + *count = host_err; err = 0; fsnotify_access(file->f_dentry); } else - err = nfserrno(err); + err = nfserrno(host_err); out: return err; } @@ -895,7 +900,7 @@ static void kill_suid(struct dentry *dentry) mutex_unlock(&dentry->d_inode->i_mutex); } -static int +static __be32 nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, loff_t offset, struct kvec *vec, int vlen, unsigned long cnt, int *stablep) @@ -904,7 +909,8 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, struct dentry *dentry; struct inode *inode; mm_segment_t oldfs; - int err = 0; + __be32 err = 0; + int host_err; int stable = *stablep; #ifdef MSNFS @@ -940,18 +946,18 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, /* Write the data. */ oldfs = get_fs(); set_fs(KERNEL_DS); - err = vfs_writev(file, (struct iovec __user *)vec, vlen, &offset); + host_err = vfs_writev(file, (struct iovec __user *)vec, vlen, &offset); set_fs(oldfs); - if (err >= 0) { + if (host_err >= 0) { nfsdstats.io_write += cnt; fsnotify_modify(file->f_dentry); } /* clear setuid/setgid flag after write */ - if (err >= 0 && (inode->i_mode & (S_ISUID | S_ISGID))) + if (host_err >= 0 && (inode->i_mode & (S_ISUID | S_ISGID))) kill_suid(dentry); - if (err >= 0 && stable) { + if (host_err >= 0 && stable) { static ino_t last_ino; static dev_t last_dev; @@ -977,7 +983,7 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, if (inode->i_state & I_DIRTY) { dprintk("nfsd: write sync %d\n", current->pid); - err=nfsd_sync(file); + host_err=nfsd_sync(file); } #if 0 wake_up(&inode->i_wait); @@ -987,11 +993,11 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, last_dev = inode->i_sb->s_dev; } - dprintk("nfsd: write complete err=%d\n", err); - if (err >= 0) + dprintk("nfsd: write complete host_err=%d\n", host_err); + if (host_err >= 0) err = 0; else - err = nfserrno(err); + err = nfserrno(host_err); out: return err; } @@ -1001,12 +1007,12 @@ out: * on entry. On return, *count contains the number of bytes actually read. * N.B. After this call fhp needs an fh_put */ -int +__be32 nfsd_read(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, loff_t offset, struct kvec *vec, int vlen, unsigned long *count) { - int err; + __be32 err; if (file) { err = nfsd_permission(fhp->fh_export, fhp->fh_dentry, @@ -1030,12 +1036,12 @@ out: * The stable flag requests synchronous writes. * N.B. After this call fhp needs an fh_put */ -int +__be32 nfsd_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file *file, loff_t offset, struct kvec *vec, int vlen, unsigned long cnt, int *stablep) { - int err = 0; + __be32 err = 0; if (file) { err = nfsd_permission(fhp->fh_export, fhp->fh_dentry, @@ -1067,12 +1073,12 @@ out: * Unfortunately we cannot lock the file to make sure we return full WCC * data to the client, as locking happens lower down in the filesystem. */ -int +__be32 nfsd_commit(struct svc_rqst *rqstp, struct svc_fh *fhp, loff_t offset, unsigned long count) { struct file *file; - int err; + __be32 err; if ((u64)count > ~(u64)offset) return nfserr_inval; @@ -1100,14 +1106,15 @@ nfsd_commit(struct svc_rqst *rqstp, struct svc_fh *fhp, * * N.B. Every call to nfsd_create needs an fh_put for _both_ fhp and resfhp */ -int +__be32 nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, char *fname, int flen, struct iattr *iap, int type, dev_t rdev, struct svc_fh *resfhp) { struct dentry *dentry, *dchild = NULL; struct inode *dirp; - int err; + __be32 err; + int host_err; err = nfserr_perm; if (!flen) @@ -1134,7 +1141,7 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, /* called from nfsd_proc_mkdir, or possibly nfsd3_proc_create */ fh_lock_nested(fhp, I_MUTEX_PARENT); dchild = lookup_one_len(fname, dentry, flen); - err = PTR_ERR(dchild); + host_err = PTR_ERR(dchild); if (IS_ERR(dchild)) goto out_nfserr; err = fh_compose(resfhp, fhp->fh_export, dchild, fhp); @@ -1173,22 +1180,22 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, err = nfserr_perm; switch (type) { case S_IFREG: - err = vfs_create(dirp, dchild, iap->ia_mode, NULL); + host_err = vfs_create(dirp, dchild, iap->ia_mode, NULL); break; case S_IFDIR: - err = vfs_mkdir(dirp, dchild, iap->ia_mode); + host_err = vfs_mkdir(dirp, dchild, iap->ia_mode); break; case S_IFCHR: case S_IFBLK: case S_IFIFO: case S_IFSOCK: - err = vfs_mknod(dirp, dchild, iap->ia_mode, rdev); + host_err = vfs_mknod(dirp, dchild, iap->ia_mode, rdev); break; default: printk("nfsd: bad file type %o in nfsd_create\n", type); - err = -EINVAL; + host_err = -EINVAL; } - if (err < 0) + if (host_err < 0) goto out_nfserr; if (EX_ISSYNC(fhp->fh_export)) { @@ -1203,7 +1210,7 @@ nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, * directories via NFS. */ if ((iap->ia_valid &= ~(ATTR_UID|ATTR_GID|ATTR_MODE)) != 0) { - int err2 = nfsd_setattr(rqstp, resfhp, iap, 0, (time_t)0); + __be32 err2 = nfsd_setattr(rqstp, resfhp, iap, 0, (time_t)0); if (err2) err = err2; } @@ -1218,7 +1225,7 @@ out: return err; out_nfserr: - err = nfserrno(err); + err = nfserrno(host_err); goto out; } @@ -1226,7 +1233,7 @@ out_nfserr: /* * NFSv3 version of nfsd_create */ -int +__be32 nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, char *fname, int flen, struct iattr *iap, struct svc_fh *resfhp, int createmode, u32 *verifier, @@ -1234,7 +1241,8 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, { struct dentry *dentry, *dchild = NULL; struct inode *dirp; - int err; + __be32 err; + int host_err; __u32 v_mtime=0, v_atime=0; int v_mode=0; @@ -1264,7 +1272,7 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, * Compose the response file handle. */ dchild = lookup_one_len(fname, dentry, flen); - err = PTR_ERR(dchild); + host_err = PTR_ERR(dchild); if (IS_ERR(dchild)) goto out_nfserr; @@ -1320,8 +1328,8 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, goto out; } - err = vfs_create(dirp, dchild, iap->ia_mode, NULL); - if (err < 0) + host_err = vfs_create(dirp, dchild, iap->ia_mode, NULL); + if (host_err < 0) goto out_nfserr; if (EX_ISSYNC(fhp->fh_export)) { @@ -1350,7 +1358,7 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, */ set_attr: if ((iap->ia_valid &= ~(ATTR_UID|ATTR_GID)) != 0) { - int err2 = nfsd_setattr(rqstp, resfhp, iap, 0, (time_t)0); + __be32 err2 = nfsd_setattr(rqstp, resfhp, iap, 0, (time_t)0); if (err2) err = err2; } @@ -1368,7 +1376,7 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, return err; out_nfserr: - err = nfserrno(err); + err = nfserrno(host_err); goto out; } #endif /* CONFIG_NFSD_V3 */ @@ -1378,13 +1386,14 @@ nfsd_create_v3(struct svc_rqst *rqstp, struct svc_fh *fhp, * fits into the buffer. On return, it contains the true length. * N.B. After this call fhp needs an fh_put */ -int +__be32 nfsd_readlink(struct svc_rqst *rqstp, struct svc_fh *fhp, char *buf, int *lenp) { struct dentry *dentry; struct inode *inode; mm_segment_t oldfs; - int err; + __be32 err; + int host_err; err = fh_verify(rqstp, fhp, S_IFLNK, MAY_NOP); if (err) @@ -1403,18 +1412,18 @@ nfsd_readlink(struct svc_rqst *rqstp, struct svc_fh *fhp, char *buf, int *lenp) */ oldfs = get_fs(); set_fs(KERNEL_DS); - err = inode->i_op->readlink(dentry, buf, *lenp); + host_err = inode->i_op->readlink(dentry, buf, *lenp); set_fs(oldfs); - if (err < 0) + if (host_err < 0) goto out_nfserr; - *lenp = err; + *lenp = host_err; err = 0; out: return err; out_nfserr: - err = nfserrno(err); + err = nfserrno(host_err); goto out; } @@ -1422,7 +1431,7 @@ out_nfserr: * Create a symlink and look up its inode * N.B. After this call _both_ fhp and resfhp need an fh_put */ -int +__be32 nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp, char *fname, int flen, char *path, int plen, @@ -1430,7 +1439,8 @@ nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap) { struct dentry *dentry, *dnew; - int err, cerr; + __be32 err, cerr; + int host_err; umode_t mode; err = nfserr_noent; @@ -1446,7 +1456,7 @@ nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp, fh_lock(fhp); dentry = fhp->fh_dentry; dnew = lookup_one_len(fname, dentry, flen); - err = PTR_ERR(dnew); + host_err = PTR_ERR(dnew); if (IS_ERR(dnew)) goto out_nfserr; @@ -1458,21 +1468,21 @@ nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp, if (unlikely(path[plen] != 0)) { char *path_alloced = kmalloc(plen+1, GFP_KERNEL); if (path_alloced == NULL) - err = -ENOMEM; + host_err = -ENOMEM; else { strncpy(path_alloced, path, plen); path_alloced[plen] = 0; - err = vfs_symlink(dentry->d_inode, dnew, path_alloced, mode); + host_err = vfs_symlink(dentry->d_inode, dnew, path_alloced, mode); kfree(path_alloced); } } else - err = vfs_symlink(dentry->d_inode, dnew, path, mode); + host_err = vfs_symlink(dentry->d_inode, dnew, path, mode); - if (!err) + if (!host_err) { if (EX_ISSYNC(fhp->fh_export)) - err = nfsd_sync_dir(dentry); - if (err) - err = nfserrno(err); + host_err = nfsd_sync_dir(dentry); + } + err = nfserrno(host_err); fh_unlock(fhp); cerr = fh_compose(resfhp, fhp->fh_export, dnew, fhp); @@ -1482,7 +1492,7 @@ out: return err; out_nfserr: - err = nfserrno(err); + err = nfserrno(host_err); goto out; } @@ -1490,13 +1500,14 @@ out_nfserr: * Create a hardlink * N.B. After this call _both_ ffhp and tfhp need an fh_put */ -int +__be32 nfsd_link(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *name, int len, struct svc_fh *tfhp) { struct dentry *ddir, *dnew, *dold; struct inode *dirp, *dest; - int err; + __be32 err; + int host_err; err = fh_verify(rqstp, ffhp, S_IFDIR, MAY_CREATE); if (err) @@ -1517,24 +1528,25 @@ nfsd_link(struct svc_rqst *rqstp, struct svc_fh *ffhp, dirp = ddir->d_inode; dnew = lookup_one_len(name, ddir, len); - err = PTR_ERR(dnew); + host_err = PTR_ERR(dnew); if (IS_ERR(dnew)) goto out_nfserr; dold = tfhp->fh_dentry; dest = dold->d_inode; - err = vfs_link(dold, dirp, dnew); - if (!err) { + host_err = vfs_link(dold, dirp, dnew); + if (!host_err) { if (EX_ISSYNC(ffhp->fh_export)) { err = nfserrno(nfsd_sync_dir(ddir)); write_inode_now(dest, 1); } + err = 0; } else { - if (err == -EXDEV && rqstp->rq_vers == 2) + if (host_err == -EXDEV && rqstp->rq_vers == 2) err = nfserr_acces; else - err = nfserrno(err); + err = nfserrno(host_err); } dput(dnew); @@ -1544,7 +1556,7 @@ out: return err; out_nfserr: - err = nfserrno(err); + err = nfserrno(host_err); goto out_unlock; } @@ -1552,13 +1564,14 @@ out_nfserr: * Rename a file * N.B. After this call _both_ ffhp and tfhp need an fh_put */ -int +__be32 nfsd_rename(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *fname, int flen, struct svc_fh *tfhp, char *tname, int tlen) { struct dentry *fdentry, *tdentry, *odentry, *ndentry, *trap; struct inode *fdir, *tdir; - int err; + __be32 err; + int host_err; err = fh_verify(rqstp, ffhp, S_IFDIR, MAY_REMOVE); if (err) @@ -1589,22 +1602,22 @@ nfsd_rename(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *fname, int flen, fill_pre_wcc(tfhp); odentry = lookup_one_len(fname, fdentry, flen); - err = PTR_ERR(odentry); + host_err = PTR_ERR(odentry); if (IS_ERR(odentry)) goto out_nfserr; - err = -ENOENT; + host_err = -ENOENT; if (!odentry->d_inode) goto out_dput_old; - err = -EINVAL; + host_err = -EINVAL; if (odentry == trap) goto out_dput_old; ndentry = lookup_one_len(tname, tdentry, tlen); - err = PTR_ERR(ndentry); + host_err = PTR_ERR(ndentry); if (IS_ERR(ndentry)) goto out_dput_old; - err = -ENOTEMPTY; + host_err = -ENOTEMPTY; if (ndentry == trap) goto out_dput_new; @@ -1612,14 +1625,14 @@ nfsd_rename(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *fname, int flen, if ((ffhp->fh_export->ex_flags & NFSEXP_MSNFS) && ((atomic_read(&odentry->d_count) > 1) || (atomic_read(&ndentry->d_count) > 1))) { - err = -EPERM; + host_err = -EPERM; } else #endif - err = vfs_rename(fdir, odentry, tdir, ndentry); - if (!err && EX_ISSYNC(tfhp->fh_export)) { - err = nfsd_sync_dir(tdentry); - if (!err) - err = nfsd_sync_dir(fdentry); + host_err = vfs_rename(fdir, odentry, tdir, ndentry); + if (!host_err && EX_ISSYNC(tfhp->fh_export)) { + host_err = nfsd_sync_dir(tdentry); + if (!host_err) + host_err = nfsd_sync_dir(fdentry); } out_dput_new: @@ -1627,8 +1640,7 @@ nfsd_rename(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *fname, int flen, out_dput_old: dput(odentry); out_nfserr: - if (err) - err = nfserrno(err); + err = nfserrno(host_err); /* we cannot reply on fh_unlock on the two filehandles, * as that would do the wrong thing if the two directories @@ -1647,13 +1659,14 @@ out: * Unlink a file or directory * N.B. After this call fhp needs an fh_put */ -int +__be32 nfsd_unlink(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, char *fname, int flen) { struct dentry *dentry, *rdentry; struct inode *dirp; - int err; + __be32 err; + int host_err; err = nfserr_acces; if (!flen || isdotent(fname, flen)) @@ -1667,7 +1680,7 @@ nfsd_unlink(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, dirp = dentry->d_inode; rdentry = lookup_one_len(fname, dentry, flen); - err = PTR_ERR(rdentry); + host_err = PTR_ERR(rdentry); if (IS_ERR(rdentry)) goto out_nfserr; @@ -1684,22 +1697,23 @@ nfsd_unlink(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, #ifdef MSNFS if ((fhp->fh_export->ex_flags & NFSEXP_MSNFS) && (atomic_read(&rdentry->d_count) > 1)) { - err = -EPERM; + host_err = -EPERM; } else #endif - err = vfs_unlink(dirp, rdentry); + host_err = vfs_unlink(dirp, rdentry); } else { /* It's RMDIR */ - err = vfs_rmdir(dirp, rdentry); + host_err = vfs_rmdir(dirp, rdentry); } dput(rdentry); - if (err == 0 && - EX_ISSYNC(fhp->fh_export)) - err = nfsd_sync_dir(dentry); + if (host_err) + goto out_nfserr; + if (EX_ISSYNC(fhp->fh_export)) + host_err = nfsd_sync_dir(dentry); out_nfserr: - err = nfserrno(err); + err = nfserrno(host_err); out: return err; } @@ -1708,11 +1722,12 @@ out: * Read entries from a directory. * The NFSv3/4 verifier we ignore for now. */ -int +__be32 nfsd_readdir(struct svc_rqst *rqstp, struct svc_fh *fhp, loff_t *offsetp, struct readdir_cd *cdp, encode_dent_fn func) { - int err; + __be32 err; + int host_err; struct file *file; loff_t offset = *offsetp; @@ -1734,10 +1749,10 @@ nfsd_readdir(struct svc_rqst *rqstp, struct svc_fh *fhp, loff_t *offsetp, do { cdp->err = nfserr_eof; /* will be cleared on successful read */ - err = vfs_readdir(file, (filldir_t) func, cdp); - } while (err >=0 && cdp->err == nfs_ok); - if (err) - err = nfserrno(err); + host_err = vfs_readdir(file, (filldir_t) func, cdp); + } while (host_err >=0 && cdp->err == nfs_ok); + if (host_err) + err = nfserrno(host_err); else err = cdp->err; *offsetp = vfs_llseek(file, 0, 1); @@ -1754,10 +1769,10 @@ out: * Get file system stats * N.B. After this call fhp needs an fh_put */ -int +__be32 nfsd_statfs(struct svc_rqst *rqstp, struct svc_fh *fhp, struct kstatfs *stat) { - int err = fh_verify(rqstp, fhp, 0, MAY_NOP); + __be32 err = fh_verify(rqstp, fhp, 0, MAY_NOP); if (!err && vfs_statfs(fhp->fh_dentry,stat)) err = nfserr_io; return err; @@ -1766,7 +1781,7 @@ nfsd_statfs(struct svc_rqst *rqstp, struct svc_fh *fhp, struct kstatfs *stat) /* * Check for a user's access permissions to this inode. */ -int +__be32 nfsd_permission(struct svc_export *exp, struct dentry *dentry, int acc) { struct inode *inode = dentry->d_inode; diff --git a/fs/ocfs2/cluster/nodemanager.c b/fs/ocfs2/cluster/nodemanager.c index e1fceb8aa32d..d11753c50bc1 100644 --- a/fs/ocfs2/cluster/nodemanager.c +++ b/fs/ocfs2/cluster/nodemanager.c @@ -152,14 +152,16 @@ static struct o2nm_node *o2nm_node_ip_tree_lookup(struct o2nm_cluster *cluster, struct o2nm_node *node, *ret = NULL; while (*p) { + int cmp; + parent = *p; node = rb_entry(parent, struct o2nm_node, nd_ip_node); - if (memcmp(&ip_needle, &node->nd_ipv4_address, - sizeof(ip_needle)) < 0) + cmp = memcmp(&ip_needle, &node->nd_ipv4_address, + sizeof(ip_needle)); + if (cmp < 0) p = &(*p)->rb_left; - else if (memcmp(&ip_needle, &node->nd_ipv4_address, - sizeof(ip_needle)) > 0) + else if (cmp > 0) p = &(*p)->rb_right; else { ret = node; diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index d9ba0a931a03..1be74c4e7814 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -30,6 +30,7 @@ #include <linux/highmem.h> #include <linux/pagemap.h> #include <linux/uio.h> +#include <linux/sched.h> #define MLOG_MASK_PREFIX ML_INODE #include <cluster/masklog.h> @@ -691,6 +692,12 @@ static int ocfs2_zero_extend(struct inode *inode, } start_off += sb->s_blocksize; + + /* + * Very large extends have the potential to lock up + * the cpu for extended periods of time. + */ + cond_resched(); } out: @@ -728,31 +735,36 @@ static int ocfs2_extend_file(struct inode *inode, clusters_to_add = ocfs2_clusters_for_bytes(inode->i_sb, new_i_size) - OCFS2_I(inode)->ip_clusters; - if (clusters_to_add) { - /* - * protect the pages that ocfs2_zero_extend is going to - * be pulling into the page cache.. we do this before the - * metadata extend so that we don't get into the situation - * where we've extended the metadata but can't get the data - * lock to zero. - */ - ret = ocfs2_data_lock(inode, 1); - if (ret < 0) { - mlog_errno(ret); - goto out; - } + /* + * protect the pages that ocfs2_zero_extend is going to be + * pulling into the page cache.. we do this before the + * metadata extend so that we don't get into the situation + * where we've extended the metadata but can't get the data + * lock to zero. + */ + ret = ocfs2_data_lock(inode, 1); + if (ret < 0) { + mlog_errno(ret); + goto out; + } + if (clusters_to_add) { ret = ocfs2_extend_allocation(inode, clusters_to_add); if (ret < 0) { mlog_errno(ret); goto out_unlock; } + } - ret = ocfs2_zero_extend(inode, (u64)new_i_size - tail_to_skip); - if (ret < 0) { - mlog_errno(ret); - goto out_unlock; - } + /* + * Call this even if we don't add any clusters to the tree. We + * still need to zero the area between the old i_size and the + * new i_size. + */ + ret = ocfs2_zero_extend(inode, (u64)new_i_size - tail_to_skip); + if (ret < 0) { + mlog_errno(ret); + goto out_unlock; } if (!tail_to_skip) { @@ -764,8 +776,7 @@ static int ocfs2_extend_file(struct inode *inode, } out_unlock: - if (clusters_to_add) /* this is the only case in which we lock */ - ocfs2_data_unlock(inode, 1); + ocfs2_data_unlock(inode, 1); out: return ret; diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c index 259155f0eb2e..a57b751d4f40 100644 --- a/fs/ocfs2/namei.c +++ b/fs/ocfs2/namei.c @@ -1085,14 +1085,6 @@ static int ocfs2_rename(struct inode *old_dir, BUG(); } - if (atomic_read(&old_dentry->d_count) > 2) { - shrink_dcache_parent(old_dentry); - if (atomic_read(&old_dentry->d_count) > 2) { - status = -EBUSY; - goto bail; - } - } - /* Assume a directory heirarchy thusly: * a/b/c * a/d diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index 4c29cd7cc8e6..76b46ebbb10c 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -339,7 +339,7 @@ static unsigned long long ocfs2_max_file_offset(unsigned int blockshift) #if BITS_PER_LONG == 32 # if defined(CONFIG_LBD) - BUG_ON(sizeof(sector_t) != 8); + BUILD_BUG_ON(sizeof(sector_t) != 8); pagefactor = PAGE_CACHE_SIZE; bitshift = BITS_PER_LONG; # else diff --git a/fs/partitions/check.c b/fs/partitions/check.c index 51c6a748df49..6fb4b6150d77 100644 --- a/fs/partitions/check.c +++ b/fs/partitions/check.c @@ -376,18 +376,48 @@ static char *make_block_name(struct gendisk *disk) return name; } -static void disk_sysfs_symlinks(struct gendisk *disk) +static int disk_sysfs_symlinks(struct gendisk *disk) { struct device *target = get_device(disk->driverfs_dev); + int err; + char *disk_name = NULL; + if (target) { - char *disk_name = make_block_name(disk); - sysfs_create_link(&disk->kobj,&target->kobj,"device"); - if (disk_name) { - sysfs_create_link(&target->kobj,&disk->kobj,disk_name); - kfree(disk_name); + disk_name = make_block_name(disk); + if (!disk_name) { + err = -ENOMEM; + goto err_out; } + + err = sysfs_create_link(&disk->kobj, &target->kobj, "device"); + if (err) + goto err_out_disk_name; + + err = sysfs_create_link(&target->kobj, &disk->kobj, disk_name); + if (err) + goto err_out_dev_link; } - sysfs_create_link(&disk->kobj, &block_subsys.kset.kobj, "subsystem"); + + err = sysfs_create_link(&disk->kobj, &block_subsys.kset.kobj, + "subsystem"); + if (err) + goto err_out_disk_name_lnk; + + kfree(disk_name); + + return 0; + +err_out_disk_name_lnk: + if (target) { + sysfs_remove_link(&target->kobj, disk_name); +err_out_dev_link: + sysfs_remove_link(&disk->kobj, "device"); +err_out_disk_name: + kfree(disk_name); +err_out: + put_device(target); + } + return err; } /* Not exported, helper to add_disk(). */ @@ -406,7 +436,11 @@ void register_disk(struct gendisk *disk) *s = '!'; if ((err = kobject_add(&disk->kobj))) return; - disk_sysfs_symlinks(disk); + err = disk_sysfs_symlinks(disk); + if (err) { + kobject_del(&disk->kobj); + return; + } disk_sysfs_add_subdirs(disk); /* No minors to use for partitions */ diff --git a/fs/proc/base.c b/fs/proc/base.c index 82da55b5cffe..8df27401d292 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -72,6 +72,7 @@ #include <linux/audit.h> #include <linux/poll.h> #include <linux/nsproxy.h> +#include <linux/oom.h> #include "internal.h" /* NOTE: @@ -86,7 +87,7 @@ /* Worst case buffer size needed for holding an integer. */ -#define PROC_NUMBUF 10 +#define PROC_NUMBUF 13 struct pid_entry { int len; @@ -689,7 +690,8 @@ static ssize_t oom_adjust_write(struct file *file, const char __user *buf, if (copy_from_user(buffer, buf, count)) return -EFAULT; oom_adjust = simple_strtol(buffer, &end, 0); - if ((oom_adjust < -16 || oom_adjust > 15) && oom_adjust != OOM_DISABLE) + if ((oom_adjust < OOM_ADJUST_MIN || oom_adjust > OOM_ADJUST_MAX) && + oom_adjust != OOM_DISABLE) return -EINVAL; if (*end == '\n') end++; diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index ad8cbc49883a..85ce23268302 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -53,6 +53,7 @@ #include <linux/workqueue.h> #include <linux/writeback.h> #include <linux/blkdev.h> +#include <linux/backing-dev.h> /* gets a struct reiserfs_journal_list * from a list head */ #define JOURNAL_LIST_ENTRY(h) (list_entry((h), struct reiserfs_journal_list, \ @@ -970,7 +971,7 @@ int reiserfs_async_progress_wait(struct super_block *s) DEFINE_WAIT(wait); struct reiserfs_journal *j = SB_JOURNAL(s); if (atomic_read(&j->j_async_throttle)) - blk_congestion_wait(WRITE, HZ / 10); + congestion_wait(WRITE, HZ / 10); return 0; } diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index c89aa2338191..17249994110f 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -430,20 +430,29 @@ int remove_save_link(struct inode *inode, int truncate) return journal_end(&th, inode->i_sb, JOURNAL_PER_BALANCE_CNT); } -static void reiserfs_put_super(struct super_block *s) +static void reiserfs_kill_sb(struct super_block *s) { - struct reiserfs_transaction_handle th; - th.t_trans_id = 0; + if (REISERFS_SB(s)) { + if (REISERFS_SB(s)->xattr_root) { + d_invalidate(REISERFS_SB(s)->xattr_root); + dput(REISERFS_SB(s)->xattr_root); + REISERFS_SB(s)->xattr_root = NULL; + } - if (REISERFS_SB(s)->xattr_root) { - d_invalidate(REISERFS_SB(s)->xattr_root); - dput(REISERFS_SB(s)->xattr_root); + if (REISERFS_SB(s)->priv_root) { + d_invalidate(REISERFS_SB(s)->priv_root); + dput(REISERFS_SB(s)->priv_root); + REISERFS_SB(s)->priv_root = NULL; + } } - if (REISERFS_SB(s)->priv_root) { - d_invalidate(REISERFS_SB(s)->priv_root); - dput(REISERFS_SB(s)->priv_root); - } + kill_block_super(s); +} + +static void reiserfs_put_super(struct super_block *s) +{ + struct reiserfs_transaction_handle th; + th.t_trans_id = 0; /* change file system state to current state if it was mounted with read-write permissions */ if (!(s->s_flags & MS_RDONLY)) { @@ -1610,6 +1619,7 @@ static int reiserfs_fill_super(struct super_block *s, void *data, int silent) "jmacd-8: reiserfs_fill_super: unable to read bitmap"); goto error; } + errval = -EINVAL; #ifdef CONFIG_REISERFS_CHECK SWARN(silent, s, "CONFIG_REISERFS_CHECK is set ON"); SWARN(silent, s, "- it is slow mode for debugging."); @@ -2156,7 +2166,7 @@ struct file_system_type reiserfs_fs_type = { .owner = THIS_MODULE, .name = "reiserfs", .get_sb = get_super_block, - .kill_sb = kill_block_super, + .kill_sb = reiserfs_kill_sb, .fs_flags = FS_REQUIRES_DEV, }; diff --git a/fs/splice.c b/fs/splice.c index 13e92dd19fbb..da74583a00ee 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -74,7 +74,7 @@ static int page_cache_pipe_buf_steal(struct pipe_inode_info *pipe, wait_on_page_writeback(page); if (PagePrivate(page)) - try_to_release_page(page, mapping_gfp_mask(mapping)); + try_to_release_page(page, GFP_KERNEL); /* * If we succeeded in removing the mapping, set LRU flag @@ -333,7 +333,7 @@ __generic_file_splice_read(struct file *in, loff_t *ppos, break; error = add_to_page_cache_lru(page, mapping, index, - mapping_gfp_mask(mapping)); + GFP_KERNEL); if (unlikely(error)) { page_cache_release(page); if (error == -EEXIST) @@ -557,7 +557,6 @@ static int pipe_to_file(struct pipe_inode_info *pipe, struct pipe_buffer *buf, { struct file *file = sd->file; struct address_space *mapping = file->f_mapping; - gfp_t gfp_mask = mapping_gfp_mask(mapping); unsigned int offset, this_len; struct page *page; pgoff_t index; @@ -591,7 +590,7 @@ static int pipe_to_file(struct pipe_inode_info *pipe, struct pipe_buffer *buf, goto find_page; page = buf->page; - if (add_to_page_cache(page, mapping, index, gfp_mask)) { + if (add_to_page_cache(page, mapping, index, GFP_KERNEL)) { unlock_page(page); goto find_page; } @@ -607,13 +606,13 @@ find_page: ret = -ENOMEM; page = page_cache_alloc_cold(mapping); if (unlikely(!page)) - goto out_nomem; + goto out_ret; /* * This will also lock the page */ ret = add_to_page_cache_lru(page, mapping, index, - gfp_mask); + GFP_KERNEL); if (unlikely(ret)) goto out; } @@ -666,7 +665,7 @@ find_page: if (sd->pos + this_len > isize) vmtruncate(mapping->host, isize); - goto out; + goto out_ret; } if (buf->page != page) { @@ -698,7 +697,7 @@ find_page: out: page_cache_release(page); unlock_page(page); -out_nomem: +out_ret: return ret; } @@ -707,9 +706,9 @@ out_nomem: * key here is the 'actor' worker passed in that actually moves the data * to the wanted destination. See pipe_to_file/pipe_to_sendpage above. */ -ssize_t splice_from_pipe(struct pipe_inode_info *pipe, struct file *out, - loff_t *ppos, size_t len, unsigned int flags, - splice_actor *actor) +static ssize_t __splice_from_pipe(struct pipe_inode_info *pipe, + struct file *out, loff_t *ppos, size_t len, + unsigned int flags, splice_actor *actor) { int ret, do_wakeup, err; struct splice_desc sd; @@ -722,9 +721,6 @@ ssize_t splice_from_pipe(struct pipe_inode_info *pipe, struct file *out, sd.file = out; sd.pos = *ppos; - if (pipe->inode) - mutex_lock(&pipe->inode->i_mutex); - for (;;) { if (pipe->nrbufs) { struct pipe_buffer *buf = pipe->bufs + pipe->curbuf; @@ -797,9 +793,6 @@ ssize_t splice_from_pipe(struct pipe_inode_info *pipe, struct file *out, pipe_wait(pipe); } - if (pipe->inode) - mutex_unlock(&pipe->inode->i_mutex); - if (do_wakeup) { smp_mb(); if (waitqueue_active(&pipe->wait)) @@ -810,6 +803,73 @@ ssize_t splice_from_pipe(struct pipe_inode_info *pipe, struct file *out, return ret; } +ssize_t splice_from_pipe(struct pipe_inode_info *pipe, struct file *out, + loff_t *ppos, size_t len, unsigned int flags, + splice_actor *actor) +{ + ssize_t ret; + struct inode *inode = out->f_mapping->host; + + /* + * The actor worker might be calling ->prepare_write and + * ->commit_write. Most of the time, these expect i_mutex to + * be held. Since this may result in an ABBA deadlock with + * pipe->inode, we have to order lock acquiry here. + */ + inode_double_lock(inode, pipe->inode); + ret = __splice_from_pipe(pipe, out, ppos, len, flags, actor); + inode_double_unlock(inode, pipe->inode); + + return ret; +} + +/** + * generic_file_splice_write_nolock - generic_file_splice_write without mutexes + * @pipe: pipe info + * @out: file to write to + * @len: number of bytes to splice + * @flags: splice modifier flags + * + * Will either move or copy pages (determined by @flags options) from + * the given pipe inode to the given file. The caller is responsible + * for acquiring i_mutex on both inodes. + * + */ +ssize_t +generic_file_splice_write_nolock(struct pipe_inode_info *pipe, struct file *out, + loff_t *ppos, size_t len, unsigned int flags) +{ + struct address_space *mapping = out->f_mapping; + struct inode *inode = mapping->host; + ssize_t ret; + int err; + + err = remove_suid(out->f_dentry); + if (unlikely(err)) + return err; + + ret = __splice_from_pipe(pipe, out, ppos, len, flags, pipe_to_file); + if (ret > 0) { + *ppos += ret; + + /* + * If file or inode is SYNC and we actually wrote some data, + * sync it. + */ + if (unlikely((out->f_flags & O_SYNC) || IS_SYNC(inode))) { + err = generic_osync_inode(inode, mapping, + OSYNC_METADATA|OSYNC_DATA); + + if (err) + ret = err; + } + } + + return ret; +} + +EXPORT_SYMBOL(generic_file_splice_write_nolock); + /** * generic_file_splice_write - splice data from a pipe to a file * @pipe: pipe info @@ -826,12 +886,21 @@ generic_file_splice_write(struct pipe_inode_info *pipe, struct file *out, loff_t *ppos, size_t len, unsigned int flags) { struct address_space *mapping = out->f_mapping; + struct inode *inode = mapping->host; ssize_t ret; + int err; + + err = should_remove_suid(out->f_dentry); + if (unlikely(err)) { + mutex_lock(&inode->i_mutex); + err = __remove_suid(out->f_dentry, err); + mutex_unlock(&inode->i_mutex); + if (err) + return err; + } ret = splice_from_pipe(pipe, out, ppos, len, flags, pipe_to_file); if (ret > 0) { - struct inode *inode = mapping->host; - *ppos += ret; /* @@ -839,8 +908,6 @@ generic_file_splice_write(struct pipe_inode_info *pipe, struct file *out, * sync it. */ if (unlikely((out->f_flags & O_SYNC) || IS_SYNC(inode))) { - int err; - mutex_lock(&inode->i_mutex); err = generic_osync_inode(inode, mapping, OSYNC_METADATA|OSYNC_DATA); @@ -1042,6 +1109,19 @@ out_release: EXPORT_SYMBOL(do_splice_direct); /* + * After the inode slimming patch, i_pipe/i_bdev/i_cdev share the same + * location, so checking ->i_pipe is not enough to verify that this is a + * pipe. + */ +static inline struct pipe_inode_info *pipe_info(struct inode *inode) +{ + if (S_ISFIFO(inode->i_mode)) + return inode->i_pipe; + + return NULL; +} + +/* * Determine where to splice to/from. */ static long do_splice(struct file *in, loff_t __user *off_in, @@ -1052,7 +1132,7 @@ static long do_splice(struct file *in, loff_t __user *off_in, loff_t offset, *off; long ret; - pipe = in->f_dentry->d_inode->i_pipe; + pipe = pipe_info(in->f_dentry->d_inode); if (pipe) { if (off_in) return -ESPIPE; @@ -1073,7 +1153,7 @@ static long do_splice(struct file *in, loff_t __user *off_in, return ret; } - pipe = out->f_dentry->d_inode->i_pipe; + pipe = pipe_info(out->f_dentry->d_inode); if (pipe) { if (off_out) return -ESPIPE; @@ -1231,7 +1311,7 @@ static int get_iovec_page_array(const struct iovec __user *iov, static long do_vmsplice(struct file *file, const struct iovec __user *iov, unsigned long nr_segs, unsigned int flags) { - struct pipe_inode_info *pipe = file->f_dentry->d_inode->i_pipe; + struct pipe_inode_info *pipe; struct page *pages[PIPE_BUFFERS]; struct partial_page partial[PIPE_BUFFERS]; struct splice_pipe_desc spd = { @@ -1241,7 +1321,8 @@ static long do_vmsplice(struct file *file, const struct iovec __user *iov, .ops = &user_page_pipe_buf_ops, }; - if (unlikely(!pipe)) + pipe = pipe_info(file->f_dentry->d_inode); + if (!pipe) return -EBADF; if (unlikely(nr_segs > UIO_MAXIOV)) return -EINVAL; @@ -1400,13 +1481,7 @@ static int link_pipe(struct pipe_inode_info *ipipe, * grabbing by inode address. Otherwise two different processes * could deadlock (one doing tee from A -> B, the other from B -> A). */ - if (ipipe->inode < opipe->inode) { - mutex_lock_nested(&ipipe->inode->i_mutex, I_MUTEX_PARENT); - mutex_lock_nested(&opipe->inode->i_mutex, I_MUTEX_CHILD); - } else { - mutex_lock_nested(&opipe->inode->i_mutex, I_MUTEX_PARENT); - mutex_lock_nested(&ipipe->inode->i_mutex, I_MUTEX_CHILD); - } + inode_double_lock(ipipe->inode, opipe->inode); do { if (!opipe->readers) { @@ -1450,8 +1525,7 @@ static int link_pipe(struct pipe_inode_info *ipipe, i++; } while (len); - mutex_unlock(&ipipe->inode->i_mutex); - mutex_unlock(&opipe->inode->i_mutex); + inode_double_unlock(ipipe->inode, opipe->inode); /* * If we put data in the output pipe, wakeup any potential readers. @@ -1475,8 +1549,8 @@ static int link_pipe(struct pipe_inode_info *ipipe, static long do_tee(struct file *in, struct file *out, size_t len, unsigned int flags) { - struct pipe_inode_info *ipipe = in->f_dentry->d_inode->i_pipe; - struct pipe_inode_info *opipe = out->f_dentry->d_inode->i_pipe; + struct pipe_inode_info *ipipe = pipe_info(in->f_dentry->d_inode); + struct pipe_inode_info *opipe = pipe_info(out->f_dentry->d_inode); int ret = -EINVAL; /* diff --git a/fs/super.c b/fs/super.c index aec99ddbe53f..47e554c12e76 100644 --- a/fs/super.c +++ b/fs/super.c @@ -260,17 +260,17 @@ int fsync_super(struct super_block *sb) * that need destruction out of superblock, call generic_shutdown_super() * and release aforementioned objects. Note: dentries and inodes _are_ * taken care of and do not need specific handling. + * + * Upon calling this function, the filesystem may no longer alter or + * rearrange the set of dentries belonging to this super_block, nor may it + * change the attachments of dentries to inodes. */ void generic_shutdown_super(struct super_block *sb) { - struct dentry *root = sb->s_root; struct super_operations *sop = sb->s_op; - if (root) { - sb->s_root = NULL; - shrink_dcache_parent(root); - shrink_dcache_sb(sb); - dput(root); + if (sb->s_root) { + shrink_dcache_for_umount(sb); fsync_super(sb); lock_super(sb); sb->s_flags &= ~MS_ACTIVE; diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c index 146f1dedec84..298303b5a716 100644 --- a/fs/sysfs/file.c +++ b/fs/sysfs/file.c @@ -483,17 +483,12 @@ int sysfs_update_file(struct kobject * kobj, const struct attribute * attr) (victim->d_parent->d_inode == dir->d_inode)) { victim->d_inode->i_mtime = CURRENT_TIME; fsnotify_modify(victim); - - /** - * Drop reference from initial sysfs_get_dentry(). - */ - dput(victim); res = 0; } else d_drop(victim); /** - * Drop the reference acquired from sysfs_get_dentry() above. + * Drop the reference acquired from lookup_one_len() above. */ dput(victim); } diff --git a/fs/sysv/super.c b/fs/sysv/super.c index 350cba5d6803..dc9e7dc07fb7 100644 --- a/fs/sysv/super.c +++ b/fs/sysv/super.c @@ -358,16 +358,11 @@ static int sysv_fill_super(struct super_block *sb, void *data, int silent) unsigned long blocknr; int size = 0, i; - if (1024 != sizeof (struct xenix_super_block)) - panic("Xenix FS: bad superblock size"); - if (512 != sizeof (struct sysv4_super_block)) - panic("SystemV FS: bad superblock size"); - if (512 != sizeof (struct sysv2_super_block)) - panic("SystemV FS: bad superblock size"); - if (500 != sizeof (struct coh_super_block)) - panic("Coherent FS: bad superblock size"); - if (64 != sizeof (struct sysv_inode)) - panic("sysv fs: bad inode size"); + BUILD_BUG_ON(1024 != sizeof (struct xenix_super_block)); + BUILD_BUG_ON(512 != sizeof (struct sysv4_super_block)); + BUILD_BUG_ON(512 != sizeof (struct sysv2_super_block)); + BUILD_BUG_ON(500 != sizeof (struct coh_super_block)); + BUILD_BUG_ON(64 != sizeof (struct sysv_inode)); sbi = kzalloc(sizeof(struct sysv_sb_info), GFP_KERNEL); if (!sbi) diff --git a/fs/xattr.c b/fs/xattr.c index 395635100f77..0901bdc2ce24 100644 --- a/fs/xattr.c +++ b/fs/xattr.c @@ -48,14 +48,21 @@ xattr_permission(struct inode *inode, const char *name, int mask) return 0; /* - * The trusted.* namespace can only accessed by a privilegued user. + * The trusted.* namespace can only be accessed by a privileged user. */ if (!strncmp(name, XATTR_TRUSTED_PREFIX, XATTR_TRUSTED_PREFIX_LEN)) return (capable(CAP_SYS_ADMIN) ? 0 : -EPERM); + /* In user.* namespace, only regular files and directories can have + * extended attributes. For sticky directories, only the owner and + * privileged user can write attributes. + */ if (!strncmp(name, XATTR_USER_PREFIX, XATTR_USER_PREFIX_LEN)) { - if (!S_ISREG(inode->i_mode) && - (!S_ISDIR(inode->i_mode) || inode->i_mode & S_ISVTX)) + if (!S_ISREG(inode->i_mode) && !S_ISDIR(inode->i_mode)) + return -EPERM; + if (S_ISDIR(inode->i_mode) && (inode->i_mode & S_ISVTX) && + (mask & MAY_WRITE) && (current->fsuid != inode->i_uid) && + !capable(CAP_FOWNER)) return -EPERM; } diff --git a/fs/xfs/linux-2.6/kmem.c b/fs/xfs/linux-2.6/kmem.c index d59737589815..004baf600611 100644 --- a/fs/xfs/linux-2.6/kmem.c +++ b/fs/xfs/linux-2.6/kmem.c @@ -21,6 +21,7 @@ #include <linux/highmem.h> #include <linux/swap.h> #include <linux/blkdev.h> +#include <linux/backing-dev.h> #include "time.h" #include "kmem.h" @@ -53,7 +54,7 @@ kmem_alloc(size_t size, unsigned int __nocast flags) printk(KERN_ERR "XFS: possible memory allocation " "deadlock in %s (mode:0x%x)\n", __FUNCTION__, lflags); - blk_congestion_wait(WRITE, HZ/50); + congestion_wait(WRITE, HZ/50); } while (1); } @@ -131,7 +132,7 @@ kmem_zone_alloc(kmem_zone_t *zone, unsigned int __nocast flags) printk(KERN_ERR "XFS: possible memory allocation " "deadlock in %s (mode:0x%x)\n", __FUNCTION__, lflags); - blk_congestion_wait(WRITE, HZ/50); + congestion_wait(WRITE, HZ/50); } while (1); } diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c index 9bbadafdcb00..db5f5a3608ca 100644 --- a/fs/xfs/linux-2.6/xfs_buf.c +++ b/fs/xfs/linux-2.6/xfs_buf.c @@ -30,6 +30,7 @@ #include <linux/hash.h> #include <linux/kthread.h> #include <linux/migrate.h> +#include <linux/backing-dev.h> #include "xfs_linux.h" STATIC kmem_zone_t *xfs_buf_zone; @@ -395,7 +396,7 @@ _xfs_buf_lookup_pages( XFS_STATS_INC(xb_page_retries); xfsbufd_wakeup(0, gfp_mask); - blk_congestion_wait(WRITE, HZ/50); + congestion_wait(WRITE, HZ/50); goto retry; } |
