diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2013-07-03 09:10:19 -0700 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2013-07-03 09:10:19 -0700 |
commit | 790eac5640abf7a57fa3a644386df330e18c11b0 (patch) | |
tree | 08de20bde44f59e51b91ff473a71047c2957e8c9 | |
parent | 0b0585c3e192967cb2ef0ac0816eb8a8c8d99840 (diff) | |
parent | 48bde8d3620f5f3c6ae9ff599eb404055ae51664 (diff) | |
download | blackbird-op-linux-790eac5640abf7a57fa3a644386df330e18c11b0.tar.gz blackbird-op-linux-790eac5640abf7a57fa3a644386df330e18c11b0.zip |
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs
Pull second set of VFS changes from Al Viro:
"Assorted f_pos race fixes, making do_splice_direct() safe to call with
i_mutex on parent, O_TMPFILE support, Jeff's locks.c series,
->d_hash/->d_compare calling conventions changes from Linus, misc
stuff all over the place."
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (63 commits)
Document ->tmpfile()
ext4: ->tmpfile() support
vfs: export lseek_execute() to modules
lseek_execute() doesn't need an inode passed to it
block_dev: switch to fixed_size_llseek()
cpqphp_sysfs: switch to fixed_size_llseek()
tile-srom: switch to fixed_size_llseek()
proc_powerpc: switch to fixed_size_llseek()
ubi/cdev: switch to fixed_size_llseek()
pci/proc: switch to fixed_size_llseek()
isapnp: switch to fixed_size_llseek()
lpfc: switch to fixed_size_llseek()
locks: give the blocked_hash its own spinlock
locks: add a new "lm_owner_key" lock operation
locks: turn the blocked_list into a hashtable
locks: convert fl_link to a hlist_node
locks: avoid taking global lock if possible when waking up blocked waiters
locks: protect most of the file_lock handling with i_lock
locks: encapsulate the fl_link list handling
locks: make "added" in __posix_lock_file a bool
...
98 files changed, 979 insertions, 986 deletions
diff --git a/Documentation/filesystems/Locking b/Documentation/filesystems/Locking index 9858f337529c..fe7afe225381 100644 --- a/Documentation/filesystems/Locking +++ b/Documentation/filesystems/Locking @@ -11,10 +11,8 @@ be able to use diff(1). prototypes: int (*d_revalidate)(struct dentry *, unsigned int); int (*d_weak_revalidate)(struct dentry *, unsigned int); - int (*d_hash)(const struct dentry *, const struct inode *, - struct qstr *); - int (*d_compare)(const struct dentry *, const struct inode *, - const struct dentry *, const struct inode *, + int (*d_hash)(const struct dentry *, struct qstr *); + int (*d_compare)(const struct dentry *, const struct dentry *, unsigned int, const char *, const struct qstr *); int (*d_delete)(struct dentry *); void (*d_release)(struct dentry *); @@ -66,6 +64,7 @@ prototypes: int (*atomic_open)(struct inode *, struct dentry *, struct file *, unsigned open_flag, umode_t create_mode, int *opened); + int (*tmpfile) (struct inode *, struct dentry *, umode_t); locking rules: all may block @@ -93,6 +92,7 @@ removexattr: yes fiemap: no update_time: no atomic_open: yes +tmpfile: no Additionally, ->rmdir(), ->unlink() and ->rename() have ->i_mutex on victim. @@ -344,25 +344,38 @@ prototypes: locking rules: - file_lock_lock may block + inode->i_lock may block fl_copy_lock: yes no fl_release_private: maybe no ----------------------- lock_manager_operations --------------------------- prototypes: int (*lm_compare_owner)(struct file_lock *, struct file_lock *); + unsigned long (*lm_owner_key)(struct file_lock *); void (*lm_notify)(struct file_lock *); /* unblock callback */ int (*lm_grant)(struct file_lock *, struct file_lock *, int); void (*lm_break)(struct file_lock *); /* break_lease callback */ int (*lm_change)(struct file_lock **, int); locking rules: - file_lock_lock may block -lm_compare_owner: yes no -lm_notify: yes no -lm_grant: no no -lm_break: yes no -lm_change yes no + + inode->i_lock blocked_lock_lock may block +lm_compare_owner: yes[1] maybe no +lm_owner_key yes[1] yes no +lm_notify: yes yes no +lm_grant: no no no +lm_break: yes no no +lm_change yes no no + +[1]: ->lm_compare_owner and ->lm_owner_key are generally called with +*an* inode->i_lock held. It may not be the i_lock of the inode +associated with either file_lock argument! This is the case with deadlock +detection, since the code has to chase down the owners of locks that may +be entirely unrelated to the one on which the lock is being acquired. +For deadlock detection however, the blocked_lock_lock is also held. The +fact that these locks are held ensures that the file_locks do not +disappear out from under you while doing the comparison or generating an +owner key. --------------------------- buffer_head ----------------------------------- prototypes: diff --git a/Documentation/filesystems/vfs.txt b/Documentation/filesystems/vfs.txt index e6bd1ffd821e..1f0ba30ae47e 100644 --- a/Documentation/filesystems/vfs.txt +++ b/Documentation/filesystems/vfs.txt @@ -360,6 +360,8 @@ struct inode_operations { int (*removexattr) (struct dentry *, const char *); void (*update_time)(struct inode *, struct timespec *, int); int (*atomic_open)(struct inode *, struct dentry *, + int (*tmpfile) (struct inode *, struct dentry *, umode_t); +} ____cacheline_aligned; struct file *, unsigned open_flag, umode_t create_mode, int *opened); }; @@ -472,6 +474,9 @@ otherwise noted. component is negative or needs lookup. Cached positive dentries are still handled by f_op->open(). + tmpfile: called in the end of O_TMPFILE open(). Optional, equivalent to + atomically creating, opening and unlinking a file in given directory. + The Address Space Object ======================== @@ -901,10 +906,8 @@ defined: struct dentry_operations { int (*d_revalidate)(struct dentry *, unsigned int); int (*d_weak_revalidate)(struct dentry *, unsigned int); - int (*d_hash)(const struct dentry *, const struct inode *, - struct qstr *); - int (*d_compare)(const struct dentry *, const struct inode *, - const struct dentry *, const struct inode *, + int (*d_hash)(const struct dentry *, struct qstr *); + int (*d_compare)(const struct dentry *, const struct dentry *, unsigned int, const char *, const struct qstr *); int (*d_delete)(const struct dentry *); void (*d_release)(struct dentry *); @@ -949,25 +952,24 @@ struct dentry_operations { d_hash: called when the VFS adds a dentry to the hash table. The first dentry passed to d_hash is the parent directory that the name is - to be hashed into. The inode is the dentry's inode. + to be hashed into. Same locking and synchronisation rules as d_compare regarding what is safe to dereference etc. d_compare: called to compare a dentry name with a given name. The first dentry is the parent of the dentry to be compared, the second is - the parent's inode, then the dentry and inode (may be NULL) of the - child dentry. len and name string are properties of the dentry to be - compared. qstr is the name to compare it with. + the child dentry. len and name string are properties of the dentry + to be compared. qstr is the name to compare it with. Must be constant and idempotent, and should not take locks if - possible, and should not or store into the dentry or inodes. - Should not dereference pointers outside the dentry or inodes without + possible, and should not or store into the dentry. + Should not dereference pointers outside the dentry without lots of care (eg. d_parent, d_inode, d_name should not be used). However, our vfsmount is pinned, and RCU held, so the dentries and inodes won't disappear, neither will our sb or filesystem module. - ->i_sb and ->d_sb may be used. + ->d_sb may be used. It is a tricky calling convention because it needs to be called under "rcu-walk", ie. without any locks or references on things. diff --git a/arch/alpha/include/uapi/asm/fcntl.h b/arch/alpha/include/uapi/asm/fcntl.h index 6d9e805f18a7..dfdadb0b4bef 100644 --- a/arch/alpha/include/uapi/asm/fcntl.h +++ b/arch/alpha/include/uapi/asm/fcntl.h @@ -32,6 +32,7 @@ #define O_SYNC (__O_SYNC|O_DSYNC) #define O_PATH 040000000 +#define O_TMPFILE 0100000000 #define F_GETLK 7 #define F_SETLK 8 diff --git a/arch/arc/kernel/troubleshoot.c b/arch/arc/kernel/troubleshoot.c index 11c301b81c92..a03528ecd276 100644 --- a/arch/arc/kernel/troubleshoot.c +++ b/arch/arc/kernel/troubleshoot.c @@ -101,7 +101,7 @@ static void show_faulting_vma(unsigned long address, char *buf) if (file) { struct path *path = &file->f_path; nm = d_path(path, buf, PAGE_SIZE - 1); - inode = vma->vm_file->f_path.dentry->d_inode; + inode = file_inode(vma->vm_file); dev = inode->i_sb->s_dev; ino = inode->i_ino; } diff --git a/arch/parisc/include/uapi/asm/fcntl.h b/arch/parisc/include/uapi/asm/fcntl.h index 0304b92ccfea..cc61c475f277 100644 --- a/arch/parisc/include/uapi/asm/fcntl.h +++ b/arch/parisc/include/uapi/asm/fcntl.h @@ -20,6 +20,7 @@ #define O_INVISIBLE 004000000 /* invisible I/O, for DMAPI/XDSM */ #define O_PATH 020000000 +#define O_TMPFILE 040000000 #define F_GETLK64 8 #define F_SETLK64 9 diff --git a/arch/powerpc/kernel/proc_powerpc.c b/arch/powerpc/kernel/proc_powerpc.c index feb8580fdc84..c30612aad68e 100644 --- a/arch/powerpc/kernel/proc_powerpc.c +++ b/arch/powerpc/kernel/proc_powerpc.c @@ -29,25 +29,9 @@ #ifdef CONFIG_PPC64 -static loff_t page_map_seek( struct file *file, loff_t off, int whence) +static loff_t page_map_seek(struct file *file, loff_t off, int whence) { - loff_t new; - switch(whence) { - case 0: - new = off; - break; - case 1: - new = file->f_pos + off; - break; - case 2: - new = PAGE_SIZE + off; - break; - default: - return -EINVAL; - } - if ( new < 0 || new > PAGE_SIZE ) - return -EINVAL; - return (file->f_pos = new); + return fixed_size_llseek(file, off, whence, PAGE_SIZE); } static ssize_t page_map_read( struct file *file, char __user *buf, size_t nbytes, diff --git a/arch/sparc/include/uapi/asm/fcntl.h b/arch/sparc/include/uapi/asm/fcntl.h index d0b83f66f356..d73e5e008b0d 100644 --- a/arch/sparc/include/uapi/asm/fcntl.h +++ b/arch/sparc/include/uapi/asm/fcntl.h @@ -35,6 +35,7 @@ #define O_SYNC (__O_SYNC|O_DSYNC) #define O_PATH 0x1000000 +#define O_TMPFILE 0x2000000 #define F_GETOWN 5 /* for sockets. */ #define F_SETOWN 6 /* for sockets. */ diff --git a/drivers/char/ps3flash.c b/drivers/char/ps3flash.c index 8cafa9ccd43f..0b311fa277ef 100644 --- a/drivers/char/ps3flash.c +++ b/drivers/char/ps3flash.c @@ -98,32 +98,8 @@ static int ps3flash_fetch(struct ps3_storage_device *dev, u64 start_sector) static loff_t ps3flash_llseek(struct file *file, loff_t offset, int origin) { struct ps3_storage_device *dev = ps3flash_dev; - loff_t res; - - mutex_lock(&file->f_mapping->host->i_mutex); - switch (origin) { - case 0: - break; - case 1: - offset += file->f_pos; - break; - case 2: - offset += dev->regions[dev->region_idx].size*dev->blk_size; - break; - default: - offset = -1; - } - if (offset < 0) { - res = -EINVAL; - goto out; - } - - file->f_pos = offset; - res = file->f_pos; - -out: - mutex_unlock(&file->f_mapping->host->i_mutex); - return res; + return generic_file_llseek_size(file, offset, origin, MAX_LFS_FILESIZE, + dev->regions[dev->region_idx].size*dev->blk_size); } static ssize_t ps3flash_read(char __user *userbuf, void *kernelbuf, diff --git a/drivers/char/tile-srom.c b/drivers/char/tile-srom.c index 2e2036e940fc..7faeb1cde97d 100644 --- a/drivers/char/tile-srom.c +++ b/drivers/char/tile-srom.c @@ -273,32 +273,10 @@ static ssize_t srom_write(struct file *filp, const char __user *buf, } /* Provide our own implementation so we can use srom->total_size. */ -loff_t srom_llseek(struct file *filp, loff_t offset, int origin) +loff_t srom_llseek(struct file *file, loff_t offset, int origin) { - struct srom_dev *srom = filp->private_data; - - if (mutex_lock_interruptible(&srom->lock)) - return -ERESTARTSYS; - - switch (origin) { - case SEEK_END: - offset += srom->total_size; - break; - case SEEK_CUR: - offset += filp->f_pos; - break; - } - - if (offset < 0 || offset > srom->total_size) { - offset = -EINVAL; - } else { - filp->f_pos = offset; - filp->f_version = 0; - } - - mutex_unlock(&srom->lock); - - return offset; + struct srom_dev *srom = file->private_data; + return fixed_size_llseek(file, offset, origin, srom->total_size); } static ssize_t total_show(struct device *dev, diff --git a/drivers/mtd/mtdchar.c b/drivers/mtd/mtdchar.c index c719879284bd..684bfa39e4ee 100644 --- a/drivers/mtd/mtdchar.c +++ b/drivers/mtd/mtdchar.c @@ -55,25 +55,7 @@ struct mtd_file_info { static loff_t mtdchar_lseek(struct file *file, loff_t offset, int orig) { struct mtd_file_info *mfi = file->private_data; - struct mtd_info *mtd = mfi->mtd; - - switch (orig) { - case SEEK_SET: - break; - case SEEK_CUR: - offset += file->f_pos; - break; - case SEEK_END: - offset += mtd->size; - break; - default: - return -EINVAL; - } - - if (offset >= 0 && offset <= mtd->size) - return file->f_pos = offset; - - return -EINVAL; + return fixed_size_llseek(file, offset, orig, mfi->mtd->size); } static int count; diff --git a/drivers/mtd/ubi/cdev.c b/drivers/mtd/ubi/cdev.c index 4f02848bb2bc..8ca49f2043e4 100644 --- a/drivers/mtd/ubi/cdev.c +++ b/drivers/mtd/ubi/cdev.c @@ -155,7 +155,6 @@ static loff_t vol_cdev_llseek(struct file *file, loff_t offset, int origin) { struct ubi_volume_desc *desc = file->private_data; struct ubi_volume *vol = desc->vol; - loff_t new_offset; if (vol->updating) { /* Update is in progress, seeking is prohibited */ @@ -163,30 +162,7 @@ static loff_t vol_cdev_llseek(struct file *file, loff_t offset, int origin) return -EBUSY; } - switch (origin) { - case 0: /* SEEK_SET */ - new_offset = offset; - break; - case 1: /* SEEK_CUR */ - new_offset = file->f_pos + offset; - break; - case 2: /* SEEK_END */ - new_offset = vol->used_bytes + offset; - break; - default: - return -EINVAL; - } - - if (new_offset < 0 || new_offset > vol->used_bytes) { - ubi_err("bad seek %lld", new_offset); - return -EINVAL; - } - - dbg_gen("seek volume %d, offset %lld, origin %d, new offset %lld", - vol->vol_id, offset, origin, new_offset); - - file->f_pos = new_offset; - return new_offset; + return fixed_size_llseek(file, offset, origin, vol->used_bytes); } static int vol_cdev_fsync(struct file *file, loff_t start, loff_t end, diff --git a/drivers/net/ethernet/brocade/bna/bnad_debugfs.c b/drivers/net/ethernet/brocade/bna/bnad_debugfs.c index 94d957d203a6..7d6aa8c87df8 100644 --- a/drivers/net/ethernet/brocade/bna/bnad_debugfs.c +++ b/drivers/net/ethernet/brocade/bna/bnad_debugfs.c @@ -230,32 +230,12 @@ bnad_debugfs_open_drvinfo(struct inode *inode, struct file *file) static loff_t bnad_debugfs_lseek(struct file *file, loff_t offset, int orig) { - loff_t pos = file->f_pos; struct bnad_debug_info *debug = file->private_data; if (!debug) return -EINVAL; - switch (orig) { - case 0: - file->f_pos = offset; - break; - case 1: - file->f_pos += offset; - break; - case 2: - file->f_pos = debug->buffer_len + offset; - break; - default: - return -EINVAL; - } - - if (file->f_pos < 0 || file->f_pos > debug->buffer_len) { - file->f_pos = pos; - return -EINVAL; - } - - return file->f_pos; + return fixed_size_llseek(file, offset, orig, debug->buffer_len); } static ssize_t diff --git a/drivers/net/wireless/ti/wlcore/debugfs.c b/drivers/net/wireless/ti/wlcore/debugfs.c index c3e1f79c7856..e17630c2a849 100644 --- a/drivers/net/wireless/ti/wlcore/debugfs.c +++ b/drivers/net/wireless/ti/wlcore/debugfs.c @@ -1056,7 +1056,7 @@ static ssize_t dev_mem_read(struct file *file, return -EINVAL; memset(&part, 0, sizeof(part)); - part.mem.start = file->f_pos; + part.mem.start = *ppos; part.mem.size = bytes; buf = kmalloc(bytes, GFP_KERNEL); @@ -1137,7 +1137,7 @@ static ssize_t dev_mem_write(struct file *file, const char __user *user_buf, return -EINVAL; memset(&part, 0, sizeof(part)); - part.mem.start = file->f_pos; + part.mem.start = *ppos; part.mem.size = bytes; buf = kmalloc(bytes, GFP_KERNEL); diff --git a/drivers/parisc/eisa_eeprom.c b/drivers/parisc/eisa_eeprom.c index af212c6a6158..783906fe659a 100644 --- a/drivers/parisc/eisa_eeprom.c +++ b/drivers/parisc/eisa_eeprom.c @@ -31,20 +31,9 @@ #define EISA_EEPROM_MINOR 241 -static loff_t eisa_eeprom_llseek(struct file *file, loff_t offset, int origin ) +static loff_t eisa_eeprom_llseek(struct file *file, loff_t offset, int origin) { - switch (origin) { - case 0: - /* nothing to do */ - break; - case 1: - offset += file->f_pos; - break; - case 2: - offset += HPEE_MAX_LENGTH; - break; - } - return (offset >= 0 && offset < HPEE_MAX_LENGTH) ? (file->f_pos = offset) : -EINVAL; + return fixed_size_llseek(file, offset, origin, HPEE_MAX_LENGTH); } static ssize_t eisa_eeprom_read(struct file * file, diff --git a/drivers/pci/hotplug/cpqphp_sysfs.c b/drivers/pci/hotplug/cpqphp_sysfs.c index 4cb30447a486..17c1f36315d1 100644 --- a/drivers/pci/hotplug/cpqphp_sysfs.c +++ b/drivers/pci/hotplug/cpqphp_sysfs.c @@ -167,26 +167,8 @@ exit: static loff_t lseek(struct file *file, loff_t off, int whence) { - struct ctrl_dbg *dbg; - loff_t new = -1; - - mutex_lock(&cpqphp_mutex); - dbg = file->private_data; - - switch (whence) { - case 0: - new = off; - break; - case 1: - new = file->f_pos + off; - break; - } - if (new < 0 || new > dbg->size) { - mutex_unlock(&cpqphp_mutex); - return -EINVAL; - } - mutex_unlock(&cpqphp_mutex); - return (file->f_pos = new); + struct ctrl_dbg *dbg = file->private_data; + return fixed_size_llseek(file, off, whence, dbg->size); } static ssize_t read(struct file *file, char __user *buf, diff --git a/drivers/pci/proc.c b/drivers/pci/proc.c index 08126087ec31..cdc7836d7e3d 100644 --- a/drivers/pci/proc.c +++ b/drivers/pci/proc.c @@ -20,27 +20,8 @@ static int proc_initialized; /* = 0 */ static loff_t proc_bus_pci_lseek(struct file *file, loff_t off, int whence) { - loff_t new = -1; - struct inode *inode = file_inode(file); - - mutex_lock(&inode->i_mutex); - switch (whence) { - case 0: - new = off; - break; - case 1: - new = file->f_pos + off; - break; - case 2: - new = inode->i_size + off; - break; - } - if (new < 0 || new > inode->i_size) - new = -EINVAL; - else - file->f_pos = new; - mutex_unlock(&inode->i_mutex); - return new; + struct pci_dev *dev = PDE_DATA(file_inode(file)); + return fixed_size_llseek(file, off, whence, dev->cfg_size); } static ssize_t diff --git a/drivers/pnp/isapnp/proc.c b/drivers/pnp/isapnp/proc.c index 2365ef37ae24..5edee645d890 100644 --- a/drivers/pnp/isapnp/proc.c +++ b/drivers/pnp/isapnp/proc.c @@ -29,27 +29,7 @@ static struct proc_dir_entry *isapnp_proc_bus_dir = NULL; static loff_t isapnp_proc_bus_lseek(struct file *file, loff_t off, int whence) { - loff_t new = -1; - struct inode *inode = file_inode(file); - - mutex_lock(&inode->i_mutex); - switch (whence) { - case 0: - new = off; - break; - case 1: - new = file->f_pos + off; - break; - case 2: - new = 256 + off; - break; - } - if (new < 0 || new > 256) - new = -EINVAL; - else - file->f_pos = new; - mutex_unlock(&inode->i_mutex); - return new; + return fixed_size_llseek(file, off, whence, 256); } static ssize_t isapnp_proc_bus_read(struct file *file, char __user * buf, diff --git a/drivers/scsi/bfa/bfad_debugfs.c b/drivers/scsi/bfa/bfad_debugfs.c index b63d534192e3..8e83d0474fe7 100644 --- a/drivers/scsi/bfa/bfad_debugfs.c +++ b/drivers/scsi/bfa/bfad_debugfs.c @@ -173,31 +173,9 @@ bfad_debugfs_open_reg(struct inode *inode, struct file *file) static loff_t bfad_debugfs_lseek(struct file *file, loff_t offset, int orig) { - struct bfad_debug_info *debug; - loff_t pos = file->f_pos; - - debug = file->private_data; - - switch (orig) { - case 0: - file->f_pos = offset; - break; - case 1: - file->f_pos += offset; - break; - case 2: - file->f_pos = debug->buffer_len + offset; - break; - default: - return -EINVAL; - } - - if (file->f_pos < 0 || file->f_pos > debug->buffer_len) { - file->f_pos = pos; - return -EINVAL; - } - - return file->f_pos; + struct bfad_debug_info *debug = file->private_data; + return fixed_size_llseek(file, offset, orig, + debug->buffer_len); } static ssize_t diff --git a/drivers/scsi/fnic/fnic_debugfs.c b/drivers/scsi/fnic/fnic_debugfs.c index 85e1ffd0e5c5..cbcb0121c84d 100644 --- a/drivers/scsi/fnic/fnic_debugfs.c +++ b/drivers/scsi/fnic/fnic_debugfs.c @@ -164,20 +164,8 @@ static loff_t fnic_trace_debugfs_lseek(struct file *file, int howto) { fnic_dbgfs_t *fnic_dbg_prt = file->private_data; - loff_t pos = -1; - - switch (howto) { - case 0: - pos = offset; - break; - case 1: - pos = file->f_pos + offset; - break; - case 2: - pos = fnic_dbg_prt->buffer_len + offset; - } - return (pos < 0 || pos > fnic_dbg_prt->buffer_len) ? - -EINVAL : (file->f_pos = pos); + return fixed_size_llseek(file, offset, howto, + fnic_dbg_prt->buffer_len); } /* diff --git a/drivers/scsi/lpfc/lpfc_debugfs.c b/drivers/scsi/lpfc/lpfc_debugfs.c index f525ecb7a9c6..60084e6ad2f2 100644 --- a/drivers/scsi/lpfc/lpfc_debugfs.c +++ b/drivers/scsi/lpfc/lpfc_debugfs.c @@ -1165,22 +1165,8 @@ out: static loff_t lpfc_debugfs_lseek(struct file *file, loff_t off, int whence) { - struct lpfc_debug *debug; - loff_t pos = -1; - - debug = file->private_data; - - switch (whence) { - case 0: - pos = off; - break; - case 1: - pos = file->f_pos + off; - break; - case 2: - pos = debug->len + off; - } - return (pos < 0 || pos > debug->len) ? -EINVAL : (file->f_pos = pos); + struct lpfc_debug *debug = file->private_data; + return fixed_size_llseek(file, off, whence, debug->len); } /** diff --git a/drivers/staging/android/logger.c b/drivers/staging/android/logger.c index 9bd874789ce5..080abf2faf97 100644 --- a/drivers/staging/android/logger.c +++ b/drivers/staging/android/logger.c @@ -696,7 +696,7 @@ static long logger_ioctl(struct file *file, unsigned int cmd, unsigned long arg) ret = -EBADF; break; } - if (!(in_egroup_p(file->f_dentry->d_inode->i_gid) || + if (!(in_egroup_p(file_inode(file)->i_gid) || capable(CAP_SYSLOG))) { ret = -EPERM; break; diff --git a/drivers/staging/comedi/comedi_fops.c b/drivers/staging/comedi/comedi_fops.c index 0794aacc928a..8647518259f6 100644 --- a/drivers/staging/comedi/comedi_fops.c +++ b/drivers/staging/comedi/comedi_fops.c @@ -2329,9 +2329,6 @@ static int comedi_close(struct inode *inode, struct file *file) mutex_unlock(&dev->mutex); - if (file->f_flags & FASYNC) - comedi_fasync(-1, file, 0); - return 0; } diff --git a/drivers/tty/vt/vc_screen.c b/drivers/tty/vt/vc_screen.c index d7799deacb21..14a2b5f11bca 100644 --- a/drivers/tty/vt/vc_screen.c +++ b/drivers/tty/vt/vc_screen.c @@ -188,22 +188,7 @@ static loff_t vcs_lseek(struct file *file, loff_t offset, int orig) console_unlock(); if (size < 0) return size; - switch (orig) { - default: - return -EINVAL; - case 2: - offset += size; - break; - case 1: - offset += file->f_pos; - case 0: - break; - } - if (offset < 0 || offset > size) { - return -EINVAL; - } - file->f_pos = offset; - return file->f_pos; + return fixed_size_llseek(file, offset, orig, size); } diff --git a/drivers/zorro/proc.c b/drivers/zorro/proc.c index 1c15ee7456b6..ea1ce822a8e0 100644 --- a/drivers/zorro/proc.c +++ b/drivers/zorro/proc.c @@ -21,27 +21,7 @@ static loff_t proc_bus_zorro_lseek(struct file *file, loff_t off, int whence) { - loff_t new = -1; - struct inode *inode = file_inode(file); - - mutex_lock(&inode->i_mutex); - switch (whence) { - case 0: - new = off; - break; - case 1: - new = file->f_pos + off; - break; - case 2: - new = sizeof(struct ConfigDev) + off; - break; - } - if (new < 0 || new > sizeof(struct ConfigDev)) - new = -EINVAL; - else - file->f_pos = new; - mutex_unlock(&inode->i_mutex); - return new; + return fixed_size_llseek(file, off, whence, sizeof(struct ConfigDev)); } static ssize_t diff --git a/fs/adfs/dir.c b/fs/adfs/dir.c index ade28bb058e3..0d138c0de293 100644 --- a/fs/adfs/dir.c +++ b/fs/adfs/dir.c @@ -191,8 +191,7 @@ const struct file_operations adfs_dir_operations = { }; static int -adfs_hash(const struct dentry *parent, const struct inode *inode, - struct qstr *qstr) +adfs_hash(const struct dentry *parent, struct qstr *qstr) { const unsigned int name_len = ADFS_SB(parent->d_sb)->s_namelen; const unsigned char *name; @@ -228,8 +227,7 @@ adfs_hash(const struct dentry *parent, const struct inode *inode, * requirements of the underlying filesystem. */ static int -adfs_compare(const struct dentry *parent, const struct inode *pinode, - const struct dentry *dentry, const struct inode *inode, +adfs_compare(const struct dentry *parent, const struct dentry *dentry, unsigned int len, const char *str, const struct qstr *name) { int i; diff --git a/fs/affs/namei.c b/fs/affs/namei.c index ff65884a7839..c36cbb4537a2 100644 --- a/fs/affs/namei.c +++ b/fs/affs/namei.c @@ -13,18 +13,12 @@ typedef int (*toupper_t)(int); static int affs_toupper(int ch); -static int affs_hash_dentry(const struct dentry *, - const struct inode *, struct qstr *); -static int affs_compare_dentry(const struct dentry *parent, - const struct inode *pinode, - const struct dentry *dentry, const struct inode *inode, +static int affs_hash_dentry(const struct dentry *, struct qstr *); +static int affs_compare_dentry(const struct dentry *parent, const struct dentry *dentry, unsigned int len, const char *str, const struct qstr *name); static int affs_intl_toupper(int ch); -static int affs_intl_hash_dentry(const struct dentry *, - const struct inode *, struct qstr *); -static int affs_intl_compare_dentry(const struct dentry *parent, - const struct inode *pinode, - const struct dentry *dentry, const struct inode *inode, +static int affs_intl_hash_dentry(const struct dentry *, struct qstr *); +static int affs_intl_compare_dentry(const struct dentry *parent, const struct dentry *dentry, unsigned int len, const char *str, const struct qstr *name); const struct dentry_operations affs_dentry_operations = { @@ -86,14 +80,12 @@ __affs_hash_dentry(struct qstr *qstr, toupper_t toupper) } static int -affs_hash_dentry(const struct dentry *dentry, const struct inode *inode, - struct qstr *qstr) +affs_hash_dentry(const struct dentry *dentry, struct qstr *qstr) { return __affs_hash_dentry(qstr, affs_toupper); } static int -affs_intl_hash_dentry(const struct dentry *dentry, const struct inode *inode, - struct qstr *qstr) +affs_intl_hash_dentry(const struct dentry *dentry, struct qstr *qstr) { return __affs_hash_dentry(qstr, affs_intl_toupper); } @@ -131,15 +123,13 @@ static inline int __affs_compare_dentry(unsigned int len, } static int -affs_compare_dentry(const struct dentry *parent, const struct inode *pinode, - const struct dentry *dentry, const struct inode *inode, +affs_compare_dentry(const struct dentry *parent, const struct dentry *dentry, unsigned int len, const char *str, const struct qstr *name) { return __affs_compare_dentry(len, str, name, affs_toupper); } static int -affs_intl_compare_dentry(const struct dentry *parent,const struct inode *pinode, - const struct dentry *dentry, const struct inode *inode, +affs_intl_compare_dentry(const struct dentry *parent, const struct dentry *dentry, unsigned int len, const char *str, const struct qstr *name) { return __affs_compare_dentry(len, str, name, affs_intl_toupper); diff --git a/fs/afs/flock.c b/fs/afs/flock.c index 2497bf306c70..a8cf2cff836c 100644 --- a/fs/afs/flock.c +++ b/fs/afs/flock.c @@ -252,7 +252,8 @@ static void afs_defer_unlock(struct afs_vnode *vnode, struct key *key) */ static int afs_do_setlk(struct file *file, struct file_lock *fl) { - struct afs_vnode *vnode = AFS_FS_I(file->f_mapping->host); + struct inode *inode = file_inode(file); + struct afs_vnode *vnode = AFS_FS_I(inode); afs_lock_type_t type; struct key *key = file->private_data; int ret; @@ -273,7 +274,7 @@ static int afs_do_setlk(struct file *file, struct file_lock *fl) type = (fl->fl_type == F_RDLCK) ? AFS_LOCK_READ : AFS_LOCK_WRITE; - lock_flocks(); + spin_lock(&inode->i_lock); /* make sure we've got a callback on this file and that our view of the * data version is up to date */ @@ -420,7 +421,7 @@ given_lock: afs_vnode_fetch_status(vnode, NULL, key); error: - unlock_flocks(); + spin_unlock(&inode->i_lock); _leave(" = %d", ret); return ret; @@ -39,6 +39,8 @@ #include <asm/kmap_types.h> #include <asm/uaccess.h> +#include "internal.h" + #define AIO_RING_MAGIC 0xa10a10a1 #define AIO_RING_COMPAT_FEATURES 1 #define AIO_RING_INCOMPAT_FEATURES 0 diff --git a/fs/block_dev.c b/fs/block_dev.c index 2091db8cdd78..431b6a04ebfd 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -325,31 +325,10 @@ static int blkdev_write_end(struct file *file, struct address_space *mapping, static loff_t block_llseek(struct file *file, loff_t offset, int whence) { struct inode *bd_inode = file->f_mapping->host; - loff_t size; loff_t retval; mutex_lock(&bd_inode->i_mutex); - size = i_size_read(bd_inode); - - retval = -EINVAL; - switch (whence) { - case SEEK_END: - offset += size; - break; - case SEEK_CUR: - offset += file->f_pos; - case SEEK_SET: - break; - default: - goto out; - } - if (offset >= 0 && offset <= size) { - if (offset != file->f_pos) { - file->f_pos = offset; - } - retval = offset; - } -out: + retval = fixed_size_llseek(file, offset, whence, i_size_read(bd_inode)); mutex_unlock(&bd_inode->i_mutex); return retval; } diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 4205ba752d40..89da56a58b63 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -2425,20 +2425,7 @@ static loff_t btrfs_file_llseek(struct file *file, loff_t offset, int whence) } } - if (offset < 0 && !(file->f_mode & FMODE_UNSIGNED_OFFSET)) { - offset = -EINVAL; - goto out; - } - if (offset > inode->i_sb->s_maxbytes) { - offset = -EINVAL; - goto out; - } - - /* Special lock needed here? */ - if (offset != file->f_pos) { - file->f_pos = offset; - file->f_version = 0; - } + offset = vfs_setpos(file, offset, inode->i_sb->s_maxbytes); out: mutex_unlock(&inode->i_mutex); return offset; diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 0f81d67cdc8d..cd7e96c73cb7 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -3881,7 +3881,7 @@ drop_write: static long btrfs_ioctl_quota_rescan(struct file *file, void __user *arg) { - struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root; + struct btrfs_root *root = BTRFS_I(file_inode(file))->root; struct btrfs_ioctl_quota_rescan_args *qsa; int ret; @@ -3914,7 +3914,7 @@ drop_write: static long btrfs_ioctl_quota_rescan_status(struct file *file, void __user *arg) { - struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root; + struct btrfs_root *root = BTRFS_I(file_inode(file))->root; struct btrfs_ioctl_quota_rescan_args *qsa; int ret = 0; @@ -4020,7 +4020,7 @@ out: static int btrfs_ioctl_get_fslabel(struct file *file, void __user *arg) { - struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root; + struct btrfs_root *root = BTRFS_I(file_inode(file))->root; const char *label = root->fs_info->super_copy->label; size_t len = strnlen(label, BTRFS_LABEL_SIZE); int ret; @@ -4039,7 +4039,7 @@ static int btrfs_ioctl_get_fslabel(struct file *file, void __user *arg) static int btrfs_ioctl_set_fslabel(struct file *file, void __user *arg) { - struct btrfs_root *root = BTRFS_I(fdentry(file)->d_inode)->root; + struct btrfs_root *root = BTRFS_I(file_inode(file))->root; struct btrfs_super_block *super_block = root->fs_info->super_copy; struct btrfs_trans_handle *trans; char label[BTRFS_LABEL_SIZE]; diff --git a/fs/ceph/file.c b/fs/ceph/file.c index 656e16907430..16c989d3e23c 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -866,16 +866,7 @@ static loff_t ceph_llseek(struct file *file, loff_t offset, int whence) break; } - if (offset < 0 || offset > inode->i_sb->s_maxbytes) { - offset = -EINVAL; - goto out; - } - - /* Special lock needed here? */ - if (offset != file->f_pos) { - file->f_pos = offset; - file->f_version = 0; - } + offset = vfs_setpos(file, offset, inode->i_sb->s_maxbytes); out: mutex_unlock(&inode->i_mutex); diff --git a/fs/ceph/locks.c b/fs/ceph/locks.c index ebbf680378e2..690f73f42425 100644 --- a/fs/ceph/locks.c +++ b/fs/ceph/locks.c @@ -192,7 +192,7 @@ void ceph_count_locks(struct inode *inode, int *fcntl_count, int *flock_count) /** * Encode the flock and fcntl locks for the given inode into the ceph_filelock - * array. Must be called with lock_flocks() already held. + * array. Must be called with inode->i_lock already held. * If we encounter more of a specific lock type than expected, return -ENOSPC. */ int ceph_encode_locks_to_buffer(struct inode *inode, diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c index 4d2920304be8..74fd2898b2ab 100644 --- a/fs/ceph/mds_client.c +++ b/fs/ceph/mds_client.c @@ -2481,20 +2481,20 @@ static int encode_caps_cb(struct inode *inode, struct ceph_cap *cap, struct ceph_filelock *flocks; encode_again: - lock_flocks(); + spin_lock(&inode->i_lock); ceph_count_locks(inode, &num_fcntl_locks, &num_flock_locks); - unlock_flocks(); + spin_unlock(&inode->i_lock); flocks = kmalloc((num_fcntl_locks+num_flock_locks) * sizeof(struct ceph_filelock), GFP_NOFS); if (!flocks) { err = -ENOMEM; goto out_free; } - lock_flocks(); + spin_lock(&inode->i_lock); err = ceph_encode_locks_to_buffer(inode, flocks, num_fcntl_locks, num_flock_locks); - unlock_flocks(); + spin_unlock(&inode->i_lock); if (err) { kfree(flocks); if (err == -ENOSPC) diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 540c1ccfcdb2..a445e71746fa 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -765,7 +765,7 @@ static loff_t cifs_llseek(struct file *file, loff_t offset, int whence) static int cifs_setlease(struct file *file, long arg, struct file_lock **lease) { - /* note that this is called by vfs setlease with lock_flocks held + /* note that this is called by vfs setlease with i_lock held to protect *lease from going away */ struct inode *inode = file_inode(file); struct cifsFileInfo *cfile = file->private_data; diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c index 5699b5036ed8..5175aebf6737 100644 --- a/fs/cifs/dir.c +++ b/fs/cifs/dir.c @@ -822,8 +822,7 @@ const struct dentry_operations cifs_dentry_ops = { /* d_delete: cifs_d_delete, */ /* not needed except for debugging */ }; -static int cifs_ci_hash(const struct dentry *dentry, const struct inode *inode, - struct qstr *q) +static int cifs_ci_hash(const struct dentry *dentry, struct qstr *q) { struct nls_table *codepage = CIFS_SB(dentry->d_sb)->local_nls; unsigned long hash; @@ -838,12 +837,10 @@ static int cifs_ci_hash(const struct dentry *dentry, const struct inode *inode, return 0; } -static int cifs_ci_compare(const struct dentry *parent, - const struct inode *pinode, - const struct dentry *dentry, const struct inode *inode, +static int cifs_ci_compare(const struct dentry *parent, const struct dentry *dentry, unsigned int len, const char *str, const struct qstr *name) { - struct nls_table *codepage = CIFS_SB(pinode->i_sb)->local_nls; + struct nls_table *codepage = CIFS_SB(parent->d_sb)->local_nls; if ((name->len == len) && (nls_strnicmp(codepage, name->name, str, len) == 0)) diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 4d8ba8d491e5..91d8629e69a2 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -999,7 +999,7 @@ try_again: rc = wait_event_interruptible(flock->fl_wait, !flock->fl_next); if (!rc) goto try_again; - locks_delete_block(flock); + posix_unblock_lock(flock); } return rc; } @@ -1092,6 +1092,7 @@ struct lock_to_push { static int cifs_push_posix_locks(struct cifsFileInfo *cfile) { + struct inode *inode = cfile->dentry->d_inode; struct cifs_tcon *tcon = tlink_tcon(cfile->tlink); struct file_lock *flock, **before; unsigned int count = 0, i = 0; @@ -1102,12 +1103,12 @@ cifs_push_posix_locks(struct cifsFileInfo *cfile) xid = get_xid(); - lock_flocks(); - cifs_for_each_lock(cfile->dentry->d_inode, before) { + spin_lock(&inode->i_lock); + cifs_for_each_lock(inode, before) { if ((*before)->fl_flags & FL_POSIX) count++; } - unlock_flocks(); + spin_unlock(&inode->i_lock); INIT_LIST_HEAD(&locks_to_send); @@ -1126,8 +1127,8 @@ cifs_push_posix_locks(struct cifsFileInfo *cfile) } el = locks_to_send.next; - lock_flocks(); - cifs_for_each_lock(cfile->dentry->d_inode, before) { + spin_lock(&inode->i_lock); + cifs_for_each_lock(inode, before) { flock = *before; if ((flock->fl_flags & FL_POSIX) == 0) continue; @@ -1152,7 +1153,7 @@ cifs_push_posix_locks(struct cifsFileInfo *cfile) lck->offset = flock->fl_start; el = el->next; } - unlock_flocks(); + spin_unlock(&inode->i_lock); list_for_each_entry_safe(lck, tmp, &locks_to_send, llist) { int stored_rc; diff --git a/fs/coda/dir.c b/fs/coda/dir.c index 87e0ee9f4465..14a14808320c 100644 --- a/fs/coda/dir.c +++ b/fs/coda/dir.c @@ -487,13 +487,7 @@ static int coda_venus_readdir(struct file *coda_file, struct dir_context *ctx) /* skip null entries */ if (vdir->d_fileno && name.len) { - /* try to look up this entry in the dcache, that way - * userspace doesn't have to worry about breaking - * getcwd by having mismatched inode numbers for - * internal volume mountpoints. */ - ino = find_inode_number(de, &name); - if (!ino) ino = vdir->d_fileno; - + ino = vdir->d_fileno; type = CDT2DT(vdir->d_type); if (!dir_emit(ctx, name.name, name.len, ino, type)) break; diff --git a/fs/dcache.c b/fs/dcache.c index 5a23073138df..87bdb5329c3c 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -1730,7 +1730,7 @@ EXPORT_SYMBOL(d_add_ci); * Do the slow-case of the dentry name compare. * * Unlike the dentry_cmp() function, we need to atomically - * load the name, length and inode information, so that the + * load the name and length information, so that the * filesystem can rely on them, and can use the 'name' and * 'len' information without worrying about walking off the * end of memory etc. @@ -1748,22 +1748,18 @@ enum slow_d_compare { static noinline enum slow_d_compare slow_dentry_cmp( const struct dentry *parent, - struct inode *inode, struct dentry *dentry, unsigned int seq, const struct qstr *name) { int tlen = dentry->d_name.len; const char *tname = dentry->d_name.name; - struct inode *i = dentry->d_inode; if (read_seqcount_retry(&dentry->d_seq, seq)) { cpu_relax(); return D_COMP_SEQRETRY; } - if (parent->d_op->d_compare(parent, inode, - dentry, i, - tlen, tname, name)) + if (parent->d_op->d_compare(parent, dentry, tlen, tname, name)) return D_COMP_NOMATCH; return D_COMP_OK; } @@ -1773,7 +1769,6 @@ static noinline enum slow_d_compare slow_dentry_cmp( * @parent: parent dentry * @name: qstr of name we wish to find * @seqp: returns d_seq value at the point where the dentry was found - * @inode: returns dentry->d_inode when the inode was found valid. * Returns: dentry, or NULL * * __d_lookup_rcu is the dcache lookup function for rcu-walk name @@ -1800,7 +1795,7 @@ static noinline enum slow_d_compare slow_dentry_cmp( */ struct dentry *__d_lookup_rcu(const struct dentry *parent, const struct qstr *name, - unsigned *seqp, struct inode *inode) + unsigned *seqp) { u64 hashlen = name->hash_len; const unsigned char *str = name->name; @@ -1834,11 +1829,10 @@ struct dentry *__d_lookup_rcu(const struct dentry *parent, seqretry: /* * The dentry sequence count protects us from concurrent - * renames, and thus protects inode, parent and name fields. + * renames, and thus protects parent and name fields. * * The caller must perform a seqcount check in order - * to do anything useful with the returned dentry, - * including using the 'd_inode' pointer. + * to do anything useful with the returned dentry. * * NOTE! We do a "raw" seqcount_begin here. That means that * we don't wait for the sequence count to stabilize if it @@ -1852,12 +1846,12 @@ seqretry: continue; if (d_unhashed(dentry)) continue; - *seqp = seq; if (unlikely(parent->d_flags & DCACHE_OP_COMPARE)) { if (dentry->d_name.hash != hashlen_hash(hashlen)) continue; - switch (slow_dentry_cmp(parent, inode, dentry, seq, name)) { + *seqp = seq; + switch (slow_dentry_cmp(parent, dentry, seq, name)) { case D_COMP_OK: return dentry; case D_COMP_NOMATCH: @@ -1869,6 +1863,7 @@ seqretry: if (dentry->d_name.hash_len != hashlen) continue; + *seqp = seq; if (!dentry_cmp(dentry, str, hashlen_len(hashlen))) return dentry; } @@ -1966,9 +1961,7 @@ struct dentry *__d_lookup(const struct dentry *parent, const struct qstr *name) if (parent->d_flags & DCACHE_OP_COMPARE) { int tlen = dentry->d_name.len; const char *tname = dentry->d_name.name; - if (parent->d_op->d_compare(parent, parent->d_inode, - dentry, dentry->d_inode, - tlen, tname, name)) + if (parent->d_op->d_compare(parent, dentry, tlen, tname, name)) goto next; } else { if (dentry->d_name.len != len) @@ -2005,7 +1998,7 @@ struct dentry *d_hash_and_lookup(struct dentry *dir, struct qstr *name) */ name->hash = full_name_hash(name->name, name->len); if (dir->d_flags & DCACHE_OP_HASH) { - int err = dir->d_op->d_hash(dir, dir->d_inode, name); + int err = dir->d_op->d_hash(dir, name); if (unlikely(err < 0)) return ERR_PTR(err); } @@ -2975,34 +2968,21 @@ rename_retry: goto again; } -/** - * find_inode_number - check for dentry with name - * @dir: directory to check - * @name: Name to find. - * - * Check whether a dentry already exists for the given name, - * and return the inode number if it has an inode. Otherwise - * 0 is returned. - * - * This routine is used to post-process directory listings for - * filesystems using synthetic inode numbers, and is necessary - * to keep getcwd() working. - */ - -ino_t find_inode_number(struct dentry *dir, struct qstr *name) +void d_tmpfile(struct dentry *dentry, struct inode *inode) { - struct dentry * dentry; - ino_t ino = 0; - - dentry = d_hash_and_lookup(dir, name); - if (!IS_ERR_OR_NULL(dentry)) { - if (dentry->d_inode) - ino = dentry->d_inode->i_ino; - dput(dentry); - } - return ino; + inode_dec_link_count(inode); + BUG_ON(dentry->d_name.name != dentry->d_iname || + !hlist_unhashed(&dentry->d_alias) || + !d_unlinked(dentry)); + spin_lock(&dentry->d_parent->d_lock); + spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); + dentry->d_name.len = sprintf(dentry->d_iname, "#%llu", + (unsigned long long)inode->i_ino); + spin_unlock(&dentry->d_lock); + spin_unlock(&dentry->d_parent->d_lock); + d_instantiate(dentry, inode); } -EXPORT_SYMBOL(find_inode_number); +EXPORT_SYMBOL(d_tmpfile); static __initdata unsigned long dhash_entries; static int __init set_dhash_entries(char *str) diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c index f71ec125290d..cfa109a4d5a2 100644 --- a/fs/ecryptfs/crypto.c +++ b/fs/ecryptfs/crypto.c @@ -2243,12 +2243,11 @@ out: */ int ecryptfs_decode_and_decrypt_filename(char **plaintext_name, size_t *plaintext_name_size, - struct dentry *ecryptfs_dir_dentry, + struct super_block *sb, const char *name, size_t name_size) { struct ecryptfs_mount_crypt_stat *mount_crypt_stat = - &ecryptfs_superblock_to_private( - ecryptfs_dir_dentry->d_sb)->mount_crypt_stat; + &ecryptfs_superblock_to_private(sb)->mount_crypt_stat; char *decoded_name; size_t decoded_name_size; size_t packet_size; diff --git a/fs/ecryptfs/ecryptfs_kernel.h b/fs/ecryptfs/ecryptfs_kernel.h index f622a733f7ad..df19d34a033b 100644 --- a/fs/ecryptfs/ecryptfs_kernel.h +++ b/fs/ecryptfs/ecryptfs_kernel.h @@ -575,7 +575,7 @@ int ecryptfs_initialize_file(struct dentry *ecryptfs_dentry, struct inode *ecryptfs_inode); int ecryptfs_decode_and_decrypt_filename(char **decrypted_name, size_t *decrypted_name_size, - struct dentry *ecryptfs_dentry, + struct super_block *sb, const char *name, size_t name_size); int ecryptfs_fill_zeros(struct file *file, loff_t new_length); int ecryptfs_encrypt_and_encode_filename( diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c index 9aa05e08060b..24f1105fda3a 100644 --- a/fs/ecryptfs/file.c +++ b/fs/ecryptfs/file.c @@ -70,7 +70,7 @@ static ssize_t ecryptfs_read_update_atime(struct kiocb *iocb, struct ecryptfs_getdents_callback { struct dir_context ctx; struct dir_context *caller; - struct dentry *dentry; + struct super_block *sb; int filldir_called; int entries_written; }; @@ -88,7 +88,7 @@ ecryptfs_filldir(void *dirent, const char *lower_name, int lower_namelen, buf->filldir_called++; rc = ecryptfs_decode_and_decrypt_filename(&name, &name_size, - buf->dentry, lower_name, + buf->sb, lower_name, lower_namelen); if (rc) { printk(KERN_ERR "%s: Error attempting to decode and decrypt " @@ -114,15 +114,14 @@ static int ecryptfs_readdir(struct file *file, struct dir_context *ctx) { int rc; struct file *lower_file; - struct inode *inode; + struct inode *inode = file_inode(file); struct ecryptfs_getdents_callback buf = { .ctx.actor = ecryptfs_filldir, .caller = ctx, - .dentry = file->f_path.dentry + .sb = inode->i_sb, }; lower_file = ecryptfs_file_to_lower(file); lower_file->f_pos = ctx->pos; - inode = file_inode(file); rc = iterate_dir(lower_file, &buf.ctx); ctx->pos = buf.ctx.pos; if (rc < 0) diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index 5eab400e2590..a2f2bb2c256d 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -679,7 +679,7 @@ static int ecryptfs_readlink_lower(struct dentry *dentry, char **buf, set_fs(old_fs); if (rc < 0) goto out; - rc = ecryptfs_decode_and_decrypt_filename(buf, bufsiz, dentry, + rc = ecryptfs_decode_and_decrypt_filename(buf, bufsiz, dentry->d_sb, lower_buf, rc); out: kfree(lower_buf); diff --git a/fs/efivarfs/super.c b/fs/efivarfs/super.c index 141aee31884f..a8766b880c07 100644 --- a/fs/efivarfs/super.c +++ b/fs/efivarfs/super.c @@ -45,8 +45,8 @@ static struct super_block *efivarfs_sb; * So we need to perform a case-sensitive match on part 1 and a * case-insensitive match on part 2. */ -static int efivarfs_d_compare(const struct dentry *parent, const struct inode *pinode, - const struct dentry *dentry, const struct inode *inode, +static int efivarfs_d_compare(const struct dentry *parent, + const struct dentry *dentry, unsigned int len, const char *str, const struct qstr *name) { @@ -63,8 +63,7 @@ static int efivarfs_d_compare(const struct dentry *parent, const struct inode *p return strncasecmp(name->name + guid, str + guid, EFI_VARIABLE_GUID_LEN); } -static int efivarfs_d_hash(const struct dentry *dentry, - const struct inode *inode, struct qstr *qstr) +static int efivarfs_d_hash(const struct dentry *dentry, struct qstr *qstr) { unsigned long hash = init_name_hash(); const unsigned char *s = qstr->name; @@ -108,7 +107,7 @@ static struct dentry *efivarfs_alloc_dentry(struct dentry *parent, char *name) q.name = name; q.len = strlen(name); - err = efivarfs_d_hash(NULL, NULL, &q); + err = efivarfs_d_hash(NULL, &q); if (err) return ERR_PTR(err); diff --git a/fs/exec.c b/fs/exec.c index ffd7a813ad3d..03b907cfd765 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -110,13 +110,14 @@ SYSCALL_DEFINE1(uselib, const char __user *, library) static const struct open_flags uselib_flags = { .open_flag = O_LARGEFILE | O_RDONLY | __FMODE_EXEC, .acc_mode = MAY_READ | MAY_EXEC | MAY_OPEN, - .intent = LOOKUP_OPEN + .intent = LOOKUP_OPEN, + .lookup_flags = LOOKUP_FOLLOW, }; if (IS_ERR(tmp)) goto out; - file = do_filp_open(AT_FDCWD, tmp, &uselib_flags, LOOKUP_FOLLOW); + file = do_filp_open(AT_FDCWD, tmp, &uselib_flags); putname(tmp); error = PTR_ERR(file); if (IS_ERR(file)) @@ -756,10 +757,11 @@ struct file *open_exec(const char *name) static const struct open_flags open_exec_flags = { .open_flag = O_LARGEFILE | O_RDONLY | __FMODE_EXEC, .acc_mode = MAY_EXEC | MAY_OPEN, - .intent = LOOKUP_OPEN + .intent = LOOKUP_OPEN, + .lookup_flags = LOOKUP_FOLLOW, }; - file = do_filp_open(AT_FDCWD, &tmp, &open_exec_flags, LOOKUP_FOLLOW); + file = do_filp_open(AT_FDCWD, &tmp, &open_exec_flags); if (IS_ERR(file)) goto out; diff --git a/fs/ext2/namei.c b/fs/ext2/namei.c index 73b0d9519836..256dd5f4c1c4 100644 --- a/fs/ext2/namei.c +++ b/fs/ext2/namei.c @@ -119,6 +119,29 @@ static int ext2_create (struct inode * dir, struct dentry * dentry, umode_t mode return ext2_add_nondir(dentry, inode); } +static int ext2_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode) +{ + struct inode *inode = ext2_new_inode(dir, mode, NULL); + if (IS_ERR(inode)) + return PTR_ERR(inode); + + inode->i_op = &ext2_file_inode_operations; + if (ext2_use_xip(inode->i_sb)) { + inode->i_mapping->a_ops = &ext2_aops_xip; + inode->i_fop = &ext2_xip_file_operations; + } else if (test_opt(inode->i_sb, NOBH)) { + inode->i_mapping->a_ops = &ext2_nobh_aops; + inode->i_fop = &ext2_file_operations; + } else { + inode->i_mapping->a_ops = &ext2_aops; + inode->i_fop = &ext2_file_operations; + } + mark_inode_dirty(inode); + d_tmpfile(dentry, inode); + unlock_new_inode(inode); + return 0; +} + static int ext2_mknod (struct inode * dir, struct dentry *dentry, umode_t mode, dev_t rdev) { struct inode * inode; @@ -398,6 +421,7 @@ const struct inode_operations ext2_dir_inode_operations = { #endif .setattr = ext2_setattr, .get_acl = ext2_get_acl, + .tmpfile = ext2_tmpfile, }; const struct inode_operations ext2_special_inode_operations = { diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c index cea8ecf3e76e..998ea111e537 100644 --- a/fs/ext3/namei.c +++ b/fs/ext3/namei.c @@ -1759,6 +1759,45 @@ retry: return err; } +static int ext3_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode) +{ + handle_t *handle; + struct inode *inode; + int err, retries = 0; + + dquot_initialize(dir); + +retry: + handle = ext3_journal_start(dir, EXT3_MAXQUOTAS_INIT_BLOCKS(dir->i_sb) + + 4 + EXT3_XATTR_TRANS_BLOCKS); + + if (IS_ERR(handle)) + return PTR_ERR(handle); + + inode = ext3_new_inode (handle, dir, NULL, mode); + err = PTR_ERR(inode); + if (!IS_ERR(inode)) { + inode->i_op = &ext3_file_inode_operations; + inode->i_fop = &ext3_file_operations; + ext3_set_aops(inode); + err = ext3_orphan_add(handle, inode); + if (err) + goto err_drop_inode; + mark_inode_dirty(inode); + d_tmpfile(dentry, inode); + unlock_new_inode(inode); + } + ext3_journal_stop(handle); + if (err == -ENOSPC && ext3_should_retry_alloc(dir->i_sb, &retries)) + goto retry; + return err; +err_drop_inode: + ext3_journal_stop(handle); + unlock_new_inode(inode); + iput(inode); + return err; +} + static int ext3_mkdir(struct inode * dir, struct dentry * dentry, umode_t mode) { handle_t *handle; @@ -2300,7 +2339,7 @@ static int ext3_link (struct dentry * old_dentry, retry: handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) + - EXT3_INDEX_EXTRA_TRANS_BLOCKS); + EXT3_INDEX_EXTRA_TRANS_BLOCKS + 1); if (IS_ERR(handle)) return PTR_ERR(handle); @@ -2314,6 +2353,11 @@ retry: err = ext3_add_entry(handle, dentry, inode); if (!err) { ext3_mark_inode_dirty(handle, inode); + /* this can happen only for tmpfile being + * linked the first time + */ + if (inode->i_nlink == 1) + ext3_orphan_del(handle, inode); d_instantiate(dentry, inode); } else { drop_nlink(inode); @@ -2516,6 +2560,7 @@ const struct inode_operations ext3_dir_inode_operations = { .mkdir = ext3_mkdir, .rmdir = ext3_rmdir, .mknod = ext3_mknod, + .tmpfile = ext3_tmpfile, .rename = ext3_rename, .setattr = ext3_setattr, #ifdef CONFIG_EXT3_FS_XATTR diff --git a/fs/ext4/file.c b/fs/ext4/file.c index b19f0a457f32..6f4cc567c382 100644 --- a/fs/ext4/file.c +++ b/fs/ext4/file.c @@ -494,17 +494,7 @@ static loff_t ext4_seek_data(struct file *file, loff_t offset, loff_t maxsize) if (dataoff > isize) return -ENXIO; - if (dataoff < 0 && !(file->f_mode & FMODE_UNSIGNED_OFFSET)) - return -EINVAL; - if (dataoff > maxsize) - return -EINVAL; - - if (dataoff != file->f_pos) { - file->f_pos = dataoff; - file->f_version = 0; - } - - return dataoff; + return vfs_setpos(file, dataoff, maxsize); } /* @@ -580,17 +570,7 @@ static loff_t ext4_seek_hole(struct file *file, loff_t offset, loff_t maxsize) if (holeoff > isize) holeoff = isize; - if (holeoff < 0 && !(file->f_mode & FMODE_UNSIGNED_OFFSET)) - return -EINVAL; - if (holeoff > maxsize) - return -EINVAL; - - if (holeoff != file->f_pos) { - file->f_pos = holeoff; - file->f_version = 0; - } - - return holeoff; + return vfs_setpos(file, holeoff, maxsize); } /* diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index ab2f6dc44b3a..234b834d5a97 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -2296,6 +2296,45 @@ retry: return err; } +static int ext4_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode) +{ + handle_t *handle; + struct inode *inode; + int err, retries = 0; + + dquot_initialize(dir); + +retry: + inode = ext4_new_inode_start_handle(dir, mode, + NULL, 0, NULL, + EXT4_HT_DIR, + EXT4_MAXQUOTAS_INIT_BLOCKS(dir->i_sb) + + 4 + EXT4_XATTR_TRANS_BLOCKS); + handle = ext4_journal_current_handle(); + 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_orphan_add(handle, inode); + if (err) + goto err_drop_inode; + mark_inode_dirty(inode); + d_tmpfile(dentry, inode); + unlock_new_inode(inode); + } + if (handle) + ext4_journal_stop(handle); + if (err == -ENOSPC && ext4_should_retry_alloc(dir->i_sb, &retries)) + goto retry; + return err; +err_drop_inode: + ext4_journal_stop(handle); + unlock_new_inode(inode); + iput(inode); + return err; +} + struct ext4_dir_entry_2 *ext4_init_dot_dotdot(struct inode *inode, struct ext4_dir_entry_2 *de, int blocksize, int csum_size, @@ -2903,7 +2942,7 @@ static int ext4_link(struct dentry *old_dentry, retry: handle = ext4_journal_start(dir, EXT4_HT_DIR, (EXT4_DATA_TRANS_BLOCKS(dir->i_sb) + - EXT4_INDEX_EXTRA_TRANS_BLOCKS)); + EXT4_INDEX_EXTRA_TRANS_BLOCKS) + 1); if (IS_ERR(handle)) return PTR_ERR(handle); @@ -2917,6 +2956,11 @@ retry: err = ext4_add_entry(handle, dentry, inode); if (!err) { ext4_mark_inode_dirty(handle, inode); + /* this can happen only for tmpfile being + * linked the first time + */ + if (inode->i_nlink == 1) + ext4_orphan_del(handle, inode); d_instantiate(dentry, inode); } else { drop_nlink(inode); @@ -3169,6 +3213,7 @@ const struct inode_operations ext4_dir_inode_operations = { .mkdir = ext4_mkdir, .rmdir = ext4_rmdir, .mknod = ext4_mknod, + .tmpfile = ext4_tmpfile, .rename = ext4_rename, .setattr = ext4_setattr, .setxattr = generic_setxattr, diff --git a/fs/fat/namei_msdos.c b/fs/fat/namei_msdos.c index 081b759cff83..a783b0e1272a 100644 --- a/fs/fat/namei_msdos.c +++ b/fs/fat/namei_msdos.c @@ -148,8 +148,7 @@ static int msdos_find(struct inode *dir, const unsigned char *name, int len, * that the existing dentry can be used. The msdos fs routines will * return ENOENT or EINVAL as appropriate. */ -static int msdos_hash(const struct dentry *dentry, const struct inode *inode, - struct qstr *qstr) +static int msdos_hash(const struct dentry *dentry, struct qstr *qstr) { struct fat_mount_options *options = &MSDOS_SB(dentry->d_sb)->options; unsigned char msdos_name[MSDOS_NAME]; @@ -165,8 +164,7 @@ static int msdos_hash(const struct dentry *dentry, const struct inode *inode, * Compare two msdos names. If either of the names are invalid, * we fall back to doing the standard name comparison. */ -static int msdos_cmp(const struct dentry *parent, const struct inode *pinode, - const struct dentry *dentry, const struct inode *inode, +static int msdos_cmp(const struct dentry *parent, const struct dentry *dentry, unsigned int len, const char *str, const struct qstr *name) { struct fat_mount_options *options = &MSDOS_SB(parent->d_sb)->options; diff --git a/fs/fat/namei_vfat.c b/fs/fat/namei_vfat.c index 2da952036a3d..6df8d3d885e5 100644 --- a/fs/fat/namei_vfat.c +++ b/fs/fat/namei_vfat.c @@ -107,8 +107,7 @@ static unsigned int vfat_striptail_len(const struct qstr *qstr) * that the existing dentry can be used. The vfat fs routines will * return ENOENT or EINVAL as appropriate. */ -static int vfat_hash(const struct dentry *dentry, const struct inode *inode, - struct qstr *qstr) +static int vfat_hash(const struct dentry *dentry, struct qstr *qstr) { qstr->hash = full_name_hash(qstr->name, vfat_striptail_len(qstr)); return 0; @@ -120,8 +119,7 @@ static int vfat_hash(const struct dentry *dentry, const struct inode *inode, * that the existing dentry can be used. The vfat fs routines will * return ENOENT or EINVAL as appropriate. */ -static int vfat_hashi(const struct dentry *dentry, const struct inode *inode, - struct qstr *qstr) +static int vfat_hashi(const struct dentry *dentry, struct qstr *qstr) { struct nls_table *t = MSDOS_SB(dentry->d_sb)->nls_io; const unsigned char *name; @@ -142,8 +140,7 @@ static int vfat_hashi(const struct dentry *dentry, const struct inode *inode, /* * Case insensitive compare of two vfat names. */ -static int vfat_cmpi(const struct dentry *parent, const struct inode *pinode, - const struct dentry *dentry, const struct inode *inode, +static int vfat_cmpi(const struct dentry *parent, const struct dentry *dentry, unsigned int len, const char *str, const struct qstr *name) { struct nls_table *t = MSDOS_SB(parent->d_sb)->nls_io; @@ -162,8 +159,7 @@ static int vfat_cmpi(const struct dentry *parent, const struct inode *pinode, /* * Case sensitive compare of two vfat names. */ -static int vfat_cmp(const struct dentry *parent, const struct inode *pinode, - const struct dentry *dentry, const struct inode *inode, +static int vfat_cmp(const struct dentry *parent, const struct dentry *dentry, unsigned int len, const char *str, const struct qstr *name) { unsigned int alen, blen; diff --git a/fs/file_table.c b/fs/file_table.c index 485dc0eddd67..08e719b884ca 100644 --- a/fs/file_table.c +++ b/fs/file_table.c @@ -227,7 +227,7 @@ static void __fput(struct file *file) { struct dentry *dentry = file->f_path.dentry; struct vfsmount *mnt = file->f_path.mnt; - struct inode *inode = dentry->d_inode; + struct inode *inode = file->f_inode; might_sleep(); diff --git a/fs/fuse/file.c b/fs/fuse/file.c index 35f281033142..5c121fe19c5f 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -548,8 +548,7 @@ static void fuse_aio_complete(struct fuse_io_priv *io, int err, ssize_t pos) res = io->bytes < 0 ? io->size : io->bytes; if (!is_sync_kiocb(io->iocb)) { - struct path *path = &io->iocb->ki_filp->f_path; - struct inode *inode = path->dentry->d_inode; + struct inode *inode = file_inode(io->iocb->ki_filp); struct fuse_conn *fc = get_fuse_conn(inode); struct fuse_inode *fi = get_fuse_inode(inode); diff --git a/fs/gfs2/dentry.c b/fs/gfs2/dentry.c index 4fddb3c22d25..f2448ab2aac5 100644 --- a/fs/gfs2/dentry.c +++ b/fs/gfs2/dentry.c @@ -109,8 +109,7 @@ fail: return 0; } -static int gfs2_dhash(const struct dentry *dentry, const struct inode *inode, - struct qstr *str) +static int gfs2_dhash(const struct dentry *dentry, struct qstr *str) { str->hash = gfs2_disk_hash(str->name, str->len); return 0; diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c index f99f9e8a325f..72c3866a7320 100644 --- a/fs/gfs2/file.c +++ b/fs/gfs2/file.c @@ -912,7 +912,7 @@ out_uninit: * cluster; until we do, disable leases (by just returning -EINVAL), * unless the administrator has requested purely local locking. * - * Locking: called under lock_flocks + * Locking: called under i_lock * * Returns: errno */ diff --git a/fs/hfs/hfs_fs.h b/fs/hfs/hfs_fs.h index a73b11839a41..0524cda47a6e 100644 --- a/fs/hfs/hfs_fs.h +++ b/fs/hfs/hfs_fs.h @@ -229,13 +229,10 @@ extern int hfs_part_find(struct super_block *, sector_t *, sector_t *); /* string.c */ extern const struct dentry_operations hfs_dentry_operations; -extern int hfs_hash_dentry(const struct dentry *, const struct inode *, - struct qstr *); +extern int hfs_hash_dentry(const struct dentry *, struct qstr *); extern int hfs_strcmp(const unsigned char *, unsigned int, const unsigned char *, unsigned int); -extern int hfs_compare_dentry(const struct dentry *parent, - const struct inode *pinode, - const struct dentry *dentry, const struct inode *inode, +extern int hfs_compare_dentry(const struct dentry *parent, const struct dentry *dentry, unsigned int len, const char *str, const struct qstr *name); /* trans.c */ diff --git a/fs/hfs/string.c b/fs/hfs/string.c index 495a976a3cc9..85b610c3909f 100644 --- a/fs/hfs/string.c +++ b/fs/hfs/string.c @@ -51,8 +51,7 @@ static unsigned char caseorder[256] = { /* * Hash a string to an integer in a case-independent way */ -int hfs_hash_dentry(const struct dentry *dentry, const struct inode *inode, - struct qstr *this) +int hfs_hash_dentry(const struct dentry *dentry, struct qstr *this) { const unsigned char *name = this->name; unsigned int hash, len = this->len; @@ -93,8 +92,7 @@ int hfs_strcmp(const unsigned char *s1, unsigned int len1, * Test for equality of two strings in the HFS filename character ordering. * return 1 on failure and 0 on success */ -int hfs_compare_dentry(const struct dentry *parent, const struct inode *pinode, - const struct dentry *dentry, const struct inode *inode, +int hfs_compare_dentry(const struct dentry *parent, const struct dentry *dentry, unsigned int len, const char *str, const struct qstr *name) { const unsigned char *n1, *n2; diff --git a/fs/hfsplus/hfsplus_fs.h b/fs/hfsplus/hfsplus_fs.h index 60b0a3388b26..ede79317cfb8 100644 --- a/fs/hfsplus/hfsplus_fs.h +++ b/fs/hfsplus/hfsplus_fs.h @@ -495,11 +495,8 @@ int hfsplus_uni2asc(struct super_block *, const struct hfsplus_unistr *, char *, int *); int hfsplus_asc2uni(struct super_block *, struct hfsplus_unistr *, int, const char *, int); -int hfsplus_hash_dentry(const struct dentry *dentry, - const struct inode *inode, struct qstr *str); -int hfsplus_compare_dentry(const struct dentry *parent, - const struct inode *pinode, - const struct dentry *dentry, const struct inode *inode, +int hfsplus_hash_dentry(const struct dentry *dentry, struct qstr *str); +int hfsplus_compare_dentry(const struct dentry *parent, const struct dentry *dentry, unsigned int len, const char *str, const struct qstr *name); /* wrapper.c */ diff --git a/fs/hfsplus/unicode.c b/fs/hfsplus/unicode.c index 2c2e47dcfdd8..e8ef121a4d8b 100644 --- a/fs/hfsplus/unicode.c +++ b/fs/hfsplus/unicode.c @@ -334,8 +334,7 @@ int hfsplus_asc2uni(struct super_block *sb, * Composed unicode characters are decomposed and case-folding is performed * if the appropriate bits are (un)set on the superblock. */ -int hfsplus_hash_dentry(const struct dentry *dentry, const struct inode *inode, - struct qstr *str) +int hfsplus_hash_dentry(const struct dentry *dentry, struct qstr *str) { struct super_block *sb = dentry->d_sb; const char *astr; @@ -386,9 +385,7 @@ int hfsplus_hash_dentry(const struct dentry *dentry, const struct inode *inode, * Composed unicode characters are decomposed and case-folding is performed * if the appropriate bits are (un)set on the superblock. */ -int hfsplus_compare_dentry(const struct dentry *parent, - const struct inode *pinode, - const struct dentry *dentry, const struct inode *inode, +int hfsplus_compare_dentry(const struct dentry *parent, const struct dentry *dentry, unsigned int len, const char *str, const struct qstr *name) { struct super_block *sb = parent->d_sb; diff --git a/fs/hpfs/dentry.c b/fs/hpfs/dentry.c index 05d4816e4e77..fa27980f2229 100644 --- a/fs/hpfs/dentry.c +++ b/fs/hpfs/dentry.c @@ -12,8 +12,7 @@ * Note: the dentry argument is the parent dentry. */ -static int hpfs_hash_dentry(const struct dentry *dentry, const struct inode *inode, - struct qstr *qstr) +static int hpfs_hash_dentry(const struct dentry *dentry, struct qstr *qstr) { unsigned long hash; int i; @@ -35,9 +34,7 @@ static int hpfs_hash_dentry(const struct dentry *dentry, const struct inode *ino return 0; } -static int hpfs_compare_dentry(const struct dentry *parent, - const struct inode *pinode, - const struct dentry *dentry, const struct inode *inode, +static int hpfs_compare_dentry(const struct dentry *parent, const struct dentry *dentry, unsigned int len, const char *str, const struct qstr *name) { unsigned al = len; diff --git a/fs/inode.c b/fs/inode.c index 00d5fc3b86e1..d6dfb09c8280 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -333,8 +333,10 @@ EXPORT_SYMBOL(set_nlink); */ void inc_nlink(struct inode *inode) { - if (WARN_ON(inode->i_nlink == 0)) + if (unlikely(inode->i_nlink == 0)) { + WARN_ON(!(inode->i_state & I_LINKABLE)); atomic_long_dec(&inode->i_sb->s_remove_count); + } inode->__i_nlink++; } diff --git a/fs/internal.h b/fs/internal.h index 68121584ae37..7c5f01cf619d 100644 --- a/fs/internal.h +++ b/fs/internal.h @@ -96,11 +96,12 @@ struct open_flags { umode_t mode; int acc_mode; int intent; + int lookup_flags; }; extern struct file *do_filp_open(int dfd, struct filename *pathname, - const struct open_flags *op, int flags); + const struct open_flags *op); extern struct file *do_file_open_root(struct dentry *, struct vfsmount *, - const char *, const struct open_flags *, int lookup_flags); + const char *, const struct open_flags *); extern long do_handle_open(int mountdirfd, struct file_handle __user *ufh, int open_flag); @@ -130,6 +131,7 @@ extern struct dentry *__d_alloc(struct super_block *, const struct qstr *); * read_write.c */ extern ssize_t __kernel_write(struct file *, const char *, size_t, loff_t *); +extern int rw_verify_area(int, struct file *, const loff_t *, size_t); /* * splice.c diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c index d9b8aebdeb22..c348d6d88624 100644 --- a/fs/isofs/inode.c +++ b/fs/isofs/inode.c @@ -28,31 +28,23 @@ #define BEQUIET -static int isofs_hashi(const struct dentry *parent, const struct inode *inode, - struct qstr *qstr); -static int isofs_hash(const struct dentry *parent, const struct inode *inode, - struct qstr *qstr); +static int isofs_hashi(const struct dentry *parent, struct qstr *qstr); +static int isofs_hash(const struct dentry *parent, struct qstr *qstr); static int isofs_dentry_cmpi(const struct dentry *parent, - const struct inode *pinode, - const struct dentry *dentry, const struct inode *inode, + const struct dentry *dentry, unsigned int len, const char *str, const struct qstr *name); static int isofs_dentry_cmp(const struct dentry *parent, - const struct inode *pinode, - const struct dentry *dentry, const struct inode *inode, + const struct dentry *dentry, unsigned int len, const char *str, const struct qstr *name); #ifdef CONFIG_JOLIET -static int isofs_hashi_ms(const struct dentry *parent, const struct inode *inode, - struct qstr *qstr); -static int isofs_hash_ms(const struct dentry *parent, const struct inode *inode, - struct qstr *qstr); +static int isofs_hashi_ms(const struct dentry *parent, struct qstr *qstr); +static int isofs_hash_ms(const struct dentry *parent, struct qstr *qstr); static int isofs_dentry_cmpi_ms(const struct dentry *parent, - const struct inode *pinode, - const struct dentry *dentry, const struct inode *inode, + const struct dentry *dentry, unsigned int len, const char *str, const struct qstr *name); static int isofs_dentry_cmp_ms(const struct dentry *parent, - const struct inode *pinode, - const struct dentry *dentry, const struct inode *inode, + const struct dentry *dentry, unsigned int len, const char *str, const struct qstr *name); #endif @@ -265,30 +257,26 @@ static int isofs_dentry_cmp_common( } static int -isofs_hash(const struct dentry *dentry, const struct inode *inode, - struct qstr *qstr) +isofs_hash(const struct dentry *dentry, struct qstr *qstr) { return isofs_hash_common(dentry, qstr, 0); } static int -isofs_hashi(const struct dentry *dentry, const struct inode *inode, - struct qstr *qstr) +isofs_hashi(const struct dentry *dentry, struct qstr *qstr) { return isofs_hashi_common(dentry, qstr, 0); } static int -isofs_dentry_cmp(const struct dentry *parent, const struct inode *pinode, - const struct dentry *dentry, const struct inode *inode, +isofs_dentry_cmp(const struct dentry *parent, const struct dentry *dentry, unsigned int len, const char *str, const struct qstr *name) { return isofs_dentry_cmp_common(len, str, name, 0, 0); } static int -isofs_dentry_cmpi(const struct dentry *parent, const struct inode *pinode, - const struct dentry *dentry, const struct inode *inode, +isofs_dentry_cmpi(const struct dentry *parent, const struct dentry *dentry, unsigned int len, const char *str, const struct qstr *name) { return isofs_dentry_cmp_common(len, str, name, 0, 1); @@ -296,30 +284,26 @@ isofs_dentry_cmpi(const struct dentry *parent, const struct inode *pinode, #ifdef CONFIG_JOLIET static int -isofs_hash_ms(const struct dentry *dentry, const struct inode *inode, - struct qstr *qstr) +isofs_hash_ms(const struct dentry *dentry, struct qstr *qstr) { return isofs_hash_common(dentry, qstr, 1); } static int -isofs_hashi_ms(const struct dentry *dentry, const struct inode *inode, - struct qstr *qstr) +isofs_hashi_ms(const struct dentry *dentry, struct qstr *qstr) { return isofs_hashi_common(dentry, qstr, 1); } static int -isofs_dentry_cmp_ms(const struct dentry *parent, const struct inode *pinode, - const struct dentry *dentry, const struct inode *inode, +isofs_dentry_cmp_ms(const struct dentry *parent, const struct dentry *dentry, unsigned int len, const char *str, const struct qstr *name) { return isofs_dentry_cmp_common(len, str, name, 1, 0); } static int -isofs_dentry_cmpi_ms(const struct dentry *parent, const struct inode *pinode, - const struct dentry *dentry, const struct inode *inode, +isofs_dentry_cmpi_ms(const struct dentry *parent, const struct dentry *dentry, unsigned int len, const char *str, const struct qstr *name) { return isofs_dentry_cmp_common(len, str, name, 1, 1); diff --git a/fs/isofs/namei.c b/fs/isofs/namei.c index c167028844ed..95295640d9c8 100644 --- a/fs/isofs/namei.c +++ b/fs/isofs/namei.c @@ -37,8 +37,7 @@ isofs_cmp(struct dentry *dentry, const char *compare, int dlen) qstr.name = compare; qstr.len = dlen; - return dentry->d_op->d_compare(NULL, NULL, NULL, NULL, - dentry->d_name.len, dentry->d_name.name, &qstr); + return dentry->d_op->d_compare(NULL, NULL, dentry->d_name.len, dentry->d_name.name, &qstr); } /* diff --git a/fs/jfs/namei.c b/fs/jfs/namei.c index 89186b7b9002..8b19027291d6 100644 --- a/fs/jfs/namei.c +++ b/fs/jfs/namei.c @@ -1538,8 +1538,7 @@ const struct file_operations jfs_dir_operations = { .llseek = generic_file_llseek, }; -static int jfs_ci_hash(const struct dentry *dir, const struct inode *inode, - struct qstr *this) +static int jfs_ci_hash(const struct dentry *dir, struct qstr *this) { unsigned long hash; int i; @@ -1552,9 +1551,7 @@ static int jfs_ci_hash(const struct dentry *dir, const struct inode *inode, return 0; } -static int jfs_ci_compare(const struct dentry *parent, - const struct inode *pinode, - const struct dentry *dentry, const struct inode *inode, +static int jfs_ci_compare(const struct dentry *parent, const struct dentry *dentry, unsigned int len, const char *str, const struct qstr *name) { int i, result = 1; diff --git a/fs/lockd/svclock.c b/fs/lockd/svclock.c index e703318c41df..067778b0ccc9 100644 --- a/fs/lockd/svclock.c +++ b/fs/lockd/svclock.c @@ -276,7 +276,7 @@ static int nlmsvc_unlink_block(struct nlm_block *block) dprintk("lockd: unlinking block %p...\n", block); /* Remove block from list */ - status = posix_unblock_lock(block->b_file->f_file, &block->b_call->a_args.lock.fl); + status = posix_unblock_lock(&block->b_call->a_args.lock.fl); nlmsvc_remove_block(block); return status; } @@ -744,8 +744,20 @@ static int nlmsvc_same_owner(struct file_lock *fl1, struct file_lock *fl2) return fl1->fl_owner == fl2->fl_owner && fl1->fl_pid == fl2->fl_pid; } +/* + * Since NLM uses two "keys" for tracking locks, we need to hash them down + * to one for the blocked_hash. Here, we're just xor'ing the host address + * with the pid in order to create a key value for picking a hash bucket. + */ +static unsigned long +nlmsvc_owner_key(struct file_lock *fl) +{ + return (unsigned long)fl->fl_owner ^ (unsigned long)fl->fl_pid; +} + const struct lock_manager_operations nlmsvc_lock_operations = { .lm_compare_owner = nlmsvc_same_owner, + .lm_owner_key = nlmsvc_owner_key, .lm_notify = nlmsvc_notify_blocked, .lm_grant = nlmsvc_grant_deferred, }; diff --git a/fs/lockd/svcsubs.c b/fs/lockd/svcsubs.c index 97e87415b145..dc5c75930f0f 100644 --- a/fs/lockd/svcsubs.c +++ b/fs/lockd/svcsubs.c @@ -169,7 +169,7 @@ nlm_traverse_locks(struct nlm_host *host, struct nlm_file *file, again: file->f_locks = 0; - lock_flocks(); /* protects i_flock list */ + spin_lock(&inode->i_lock); for (fl = inode->i_flock; fl; fl = fl->fl_next) { if (fl->fl_lmops != &nlmsvc_lock_operations) continue; @@ -181,7 +181,7 @@ again: if (match(lockhost, host)) { struct file_lock lock = *fl; - unlock_flocks(); + spin_unlock(&inode->i_lock); lock.fl_type = F_UNLCK; lock.fl_start = 0; lock.fl_end = OFFSET_MAX; @@ -193,7 +193,7 @@ again: goto again; } } - unlock_flocks(); + spin_unlock(&inode->i_lock); return 0; } @@ -228,14 +228,14 @@ nlm_file_inuse(struct nlm_file *file) if (file->f_count || !list_empty(&file->f_blocks) || file->f_shares) return 1; - lock_flocks(); + spin_lock(&inode->i_lock); for (fl = inode->i_flock; fl; fl = fl->fl_next) { if (fl->fl_lmops == &nlmsvc_lock_operations) { - unlock_flocks(); + spin_unlock(&inode->i_lock); return 1; } } - unlock_flocks(); + spin_unlock(&inode->i_lock); file->f_locks = 0; return 0; } diff --git a/fs/locks.c b/fs/locks.c index cb424a4fed71..04e2c1fdb157 100644 --- a/fs/locks.c +++ b/fs/locks.c @@ -126,6 +126,7 @@ #include <linux/time.h> #include <linux/rcupdate.h> #include <linux/pid_namespace.h> +#include <linux/hashtable.h> #include <asm/uaccess.h> @@ -153,30 +154,51 @@ int lease_break_time = 45; #define for_each_lock(inode, lockp) \ for (lockp = &inode->i_flock; *lockp != NULL; lockp = &(*lockp)->fl_next) -static LIST_HEAD(file_lock_list); -static LIST_HEAD(blocked_list); +/* + * The global file_lock_list is only used for displaying /proc/locks. Protected + * by the file_lock_lock. + */ +static HLIST_HEAD(file_lock_list); static DEFINE_SPINLOCK(file_lock_lock); /* - * Protects the two list heads above, plus the inode->i_flock list + * The blocked_hash is used to find POSIX lock loops for deadlock detection. + * It is protected by blocked_lock_lock. + * + * We hash locks by lockowner in order to optimize searching for the lock a + * particular lockowner is waiting on. + * + * FIXME: make this value scale via some heuristic? We generally will want more + * buckets when we have more lockowners holding locks, but that's a little + * difficult to determine without knowing what the workload will look like. */ -void lock_flocks(void) -{ - spin_lock(&file_lock_lock); -} -EXPORT_SYMBOL_GPL(lock_flocks); +#define BLOCKED_HASH_BITS 7 +static DEFINE_HASHTABLE(blocked_hash, BLOCKED_HASH_BITS); -void unlock_flocks(void) -{ - spin_unlock(&file_lock_lock); -} -EXPORT_SYMBOL_GPL(unlock_flocks); +/* + * This lock protects the blocked_hash. Generally, if you're accessing it, you + * want to be holding this lock. + * + * In addition, it also protects the fl->fl_block list, and the fl->fl_next + * pointer for file_lock structures that are acting as lock requests (in + * contrast to those that are acting as records of acquired locks). + * + * Note that when we acquire this lock in order to change the above fields, + * we often hold the i_lock as well. In certain cases, when reading the fields + * protected by this lock, we can skip acquiring it iff we already hold the + * i_lock. + * + * In particular, adding an entry to the fl_block list requires that you hold + * both the i_lock and the blocked_lock_lock (acquired in that order). Deleting + * an entry from the list however only requires the file_lock_lock. + */ +static DEFINE_SPINLOCK(blocked_lock_lock); static struct kmem_cache *filelock_cache __read_mostly; static void locks_init_lock_heads(struct file_lock *fl) { - INIT_LIST_HEAD(&fl->fl_link); + INIT_HLIST_NODE(&fl->fl_link); INIT_LIST_HEAD(&fl->fl_block); init_waitqueue_head(&fl->fl_wait); } @@ -210,7 +232,7 @@ void locks_free_lock(struct file_lock *fl) { BUG_ON(waitqueue_active(&fl->fl_wait)); BUG_ON(!list_empty(&fl->fl_block)); - BUG_ON(!list_empty(&fl->fl_link)); + BUG_ON(!hlist_unhashed(&fl->fl_link)); locks_release_private(fl); kmem_cache_free(filelock_cache, fl); @@ -484,47 +506,108 @@ static int posix_same_owner(struct file_lock *fl1, struct file_lock *fl2) return fl1->fl_owner == fl2->fl_owner; } +static inline void +locks_insert_global_locks(struct file_lock *fl) +{ + spin_lock(&file_lock_lock); + hlist_add_head(&fl->fl_link, &file_lock_list); + spin_unlock(&file_lock_lock); +} + +static inline void +locks_delete_global_locks(struct file_lock *fl) +{ + spin_lock(&file_lock_lock); + hlist_del_init(&fl->fl_link); + spin_unlock(&file_lock_lock); +} + +static unsigned long +posix_owner_key(struct file_lock *fl) +{ + if (fl->fl_lmops && fl->fl_lmops->lm_owner_key) + return fl->fl_lmops->lm_owner_key(fl); + return (unsigned long)fl->fl_owner; +} + +static inline void +locks_insert_global_blocked(struct file_lock *waiter) +{ + hash_add(blocked_hash, &waiter->fl_link, posix_owner_key(waiter)); +} + +static inline void +locks_delete_global_blocked(struct file_lock *waiter) +{ + hash_del(&waiter->fl_link); +} + /* Remove waiter from blocker's block list. * When blocker ends up pointing to itself then the list is empty. + * + * Must be called with blocked_lock_lock held. */ static void __locks_delete_block(struct file_lock *waiter) { + locks_delete_global_blocked(waiter); list_del_init(&waiter->fl_block); - list_del_init(&waiter->fl_link); waiter->fl_next = NULL; } -/* - */ -void locks_delete_block(struct file_lock *waiter) +static void locks_delete_block(struct file_lock *waiter) { - lock_flocks(); + spin_lock(&blocked_lock_lock); __locks_delete_block(waiter); - unlock_flocks(); + spin_unlock(&blocked_lock_lock); } -EXPORT_SYMBOL(locks_delete_block); /* Insert waiter into blocker's block list. * We use a circular list so that processes can be easily woken up in * the order they blocked. The documentation doesn't require this but * it seems like the reasonable thing to do. + * + * Must be called with both the i_lock and blocked_lock_lock held. The fl_block + * list itself is protected by the file_lock_list, but by ensuring that the + * i_lock is also held on insertions we can avoid taking the blocked_lock_lock + * in some cases when we see that the fl_block list is empty. */ -static void locks_insert_block(struct file_lock *blocker, - struct file_lock *waiter) +static void __locks_insert_block(struct file_lock *blocker, + struct file_lock *waiter) { BUG_ON(!list_empty(&waiter->fl_block)); - list_add_tail(&waiter->fl_block, &blocker->fl_block); waiter->fl_next = blocker; + list_add_tail(&waiter->fl_block, &blocker->fl_block); if (IS_POSIX(blocker)) - list_add(&waiter->fl_link, &blocked_list); + locks_insert_global_blocked(waiter); } -/* Wake up processes blocked waiting for blocker. - * If told to wait then schedule the processes until the block list - * is empty, otherwise empty the block list ourselves. +/* Must be called with i_lock held. */ +static void locks_insert_block(struct file_lock *blocker, + struct file_lock *waiter) +{ + spin_lock(&blocked_lock_lock); + __locks_insert_block(blocker, waiter); + spin_unlock(&blocked_lock_lock); +} + +/* + * Wake up processes blocked waiting for blocker. + * + * Must be called with the inode->i_lock held! */ static void locks_wake_up_blocks(struct file_lock *blocker) { + /* + * Avoid taking global lock if list is empty. This is safe since new + * blocked requests are only added to the list under the i_lock, and + * the i_lock is always held here. Note that removal from the fl_block + * list does not require the i_lock, so we must recheck list_empty() + * after acquiring the blocked_lock_lock. + */ + if (list_empty(&blocker->fl_block)) + return; + + spin_lock(&blocked_lock_lock); while (!list_empty(&blocker->fl_block)) { struct file_lock *waiter; @@ -536,20 +619,23 @@ static void locks_wake_up_blocks(struct file_lock *blocker) else wake_up(&waiter->fl_wait); } + spin_unlock(&blocked_lock_lock); } /* Insert file lock fl into an inode's lock list at the position indicated * by pos. At the same time add the lock to the global file lock list. + * + * Must be called with the i_lock held! */ static void locks_insert_lock(struct file_lock **pos, struct file_lock *fl) { - list_add(&fl->fl_link, &file_lock_list); - fl->fl_nspid = get_pid(task_tgid(current)); /* insert into file's list */ fl->fl_next = *pos; *pos = fl; + + locks_insert_global_locks(fl); } /* @@ -557,14 +643,17 @@ static void locks_insert_lock(struct file_lock **pos, struct file_lock *fl) * Wake up processes that are blocked waiting for this lock, * notify the FS that the lock has been cleared and * finally free the lock. + * + * Must be called with the i_lock held! */ static void locks_delete_lock(struct file_lock **thisfl_p) { struct file_lock *fl = *thisfl_p; + locks_delete_global_locks(fl); + *thisfl_p = fl->fl_next; fl->fl_next = NULL; - list_del_init(&fl->fl_link); if (fl->fl_nspid) { put_pid(fl->fl_nspid); @@ -625,8 +714,9 @@ void posix_test_lock(struct file *filp, struct file_lock *fl) { struct file_lock *cfl; + struct inode *inode = file_inode(filp); - lock_flocks(); + spin_lock(&inode->i_lock); for (cfl = file_inode(filp)->i_flock; cfl; cfl = cfl->fl_next) { if (!IS_POSIX(cfl)) continue; @@ -639,7 +729,7 @@ posix_test_lock(struct file *filp, struct file_lock *fl) fl->fl_pid = pid_vnr(cfl->fl_nspid); } else fl->fl_type = F_UNLCK; - unlock_flocks(); + spin_unlock(&inode->i_lock); return; } EXPORT_SYMBOL(posix_test_lock); @@ -676,13 +766,14 @@ static struct file_lock *what_owner_is_waiting_for(struct file_lock *block_fl) { struct file_lock *fl; - list_for_each_entry(fl, &blocked_list, fl_link) { + hash_for_each_possible(blocked_hash, fl, fl_link, posix_owner_key(block_fl)) { if (posix_same_owner(fl, block_fl)) return fl->fl_next; } return NULL; } +/* Must be called with the blocked_lock_lock held! */ static int posix_locks_deadlock(struct file_lock *caller_fl, struct file_lock *block_fl) { @@ -718,7 +809,7 @@ static int flock_lock_file(struct file *filp, struct file_lock *request) return -ENOMEM; } - lock_flocks(); + spin_lock(&inode->i_lock); if (request->fl_flags & FL_ACCESS) goto find_conflict; @@ -748,9 +839,9 @@ static int flock_lock_file(struct file *filp, struct file_lock *request) * give it the opportunity to lock the file. */ if (found) { - unlock_flocks(); + spin_unlock(&inode->i_lock); cond_resched(); - lock_flocks(); + spin_lock(&inode->i_lock); } find_conflict: @@ -777,7 +868,7 @@ find_conflict: error = 0; out: - unlock_flocks(); + spin_unlock(&inode->i_lock); if (new_fl) locks_free_lock(new_fl); return error; @@ -791,7 +882,8 @@ static int __posix_lock_file(struct inode *inode, struct file_lock *request, str struct file_lock *left = NULL; struct file_lock *right = NULL; struct file_lock **before; - int error, added = 0; + int error; + bool added = false; /* * We may need two file_lock structures for this operation, @@ -806,7 +898,12 @@ static int __posix_lock_file(struct inode *inode, struct file_lock *request, str new_fl2 = locks_alloc_lock(); } - lock_flocks(); + spin_lock(&inode->i_lock); + /* + * New lock request. Walk all POSIX locks and look for conflicts. If + * there are any, either return error or put the request on the + * blocker's list of waiters and the global blocked_hash. + */ if (request->fl_type != F_UNLCK) { for_each_lock(inode, before) { fl = *before; @@ -819,11 +916,17 @@ static int __posix_lock_file(struct inode *inode, struct file_lock *request, str error = -EAGAIN; if (!(request->fl_flags & FL_SLEEP)) goto out; + /* + * Deadlock detection and insertion into the blocked + * locks list must be done while holding the same lock! + */ error = -EDEADLK; - if (posix_locks_deadlock(request, fl)) - goto out; - error = FILE_LOCK_DEFERRED; - locks_insert_block(fl, request); + spin_lock(&blocked_lock_lock); + if (likely(!posix_locks_deadlock(request, fl))) { + error = FILE_LOCK_DEFERRED; + __locks_insert_block(fl, request); + } + spin_unlock(&blocked_lock_lock); goto out; } } @@ -845,7 +948,7 @@ static int __posix_lock_file(struct inode *inode, struct file_lock *request, str before = &fl->fl_next; } - /* Process locks with this owner. */ + /* Process locks with this owner. */ while ((fl = *before) && posix_same_owner(request, fl)) { /* Detect adjacent or overlapping regions (if same lock type) */ @@ -880,7 +983,7 @@ static int __posix_lock_file(struct inode *inode, struct file_lock *request, str continue; } request = fl; - added = 1; + added = true; } else { /* Processing for different lock types is a bit @@ -891,7 +994,7 @@ static int __posix_lock_file(struct inode *inode, struct file_lock *request, str if (fl->fl_start > request->fl_end) break; if (request->fl_type == F_UNLCK) - added = 1; + added = true; if (fl->fl_start < request->fl_start) left = fl; /* If the next lock in the list has a higher end @@ -921,7 +1024,7 @@ static int __posix_lock_file(struct inode *inode, struct file_lock *request, str locks_release_private(fl); locks_copy_private(fl, request); request = fl; - added = 1; + added = true; } } /* Go on to next lock. @@ -931,10 +1034,9 @@ static int __posix_lock_file(struct inode *inode, struct file_lock *request, str } /* - * The above code only modifies existing locks in case of - * merging or replacing. If new lock(s) need to be inserted - * all modifications are done bellow this, so it's safe yet to - * bail out. + * The above code only modifies existing locks in case of merging or + * replacing. If new lock(s) need to be inserted all modifications are + * done below this, so it's safe yet to bail out. */ error = -ENOLCK; /* "no luck" */ if (right && left == right && !new_fl2) @@ -974,7 +1076,7 @@ static int __posix_lock_file(struct inode *inode, struct file_lock *request, str locks_wake_up_blocks(left); } out: - unlock_flocks(); + spin_unlock(&inode->i_lock); /* * Free any unused locks. */ @@ -1049,14 +1151,14 @@ int locks_mandatory_locked(struct inode *inode) /* * Search the lock list for this inode for any POSIX locks. */ - lock_flocks(); + spin_lock(&inode->i_lock); for (fl = inode->i_flock; fl != NULL; fl = fl->fl_next) { if (!IS_POSIX(fl)) continue; if (fl->fl_owner != owner) break; } - unlock_flocks(); + spin_unlock(&inode->i_lock); return fl ? -EAGAIN : 0; } @@ -1199,7 +1301,7 @@ int __break_lease(struct inode *inode, unsigned int mode) if (IS_ERR(new_fl)) return PTR_ERR(new_fl); - lock_flocks(); + spin_lock(&inode->i_lock); time_out_leases(inode); @@ -1249,11 +1351,11 @@ restart: break_time++; } locks_insert_block(flock, new_fl); - unlock_flocks(); + spin_unlock(&inode->i_lock); error = wait_event_interruptible_timeout(new_fl->fl_wait, !new_fl->fl_next, break_time); - lock_flocks(); - __locks_delete_block(new_fl); + spin_lock(&inode->i_lock); + locks_delete_block(new_fl); if (error >= 0) { if (error == 0) time_out_leases(inode); @@ -1270,7 +1372,7 @@ restart: } out: - unlock_flocks(); + spin_unlock(&inode->i_lock); locks_free_lock(new_fl); return error; } @@ -1323,9 +1425,10 @@ EXPORT_SYMBOL(lease_get_mtime); int fcntl_getlease(struct file *filp) { struct file_lock *fl; + struct inode *inode = file_inode(filp); int type = F_UNLCK; - lock_flocks(); + spin_lock(&inode->i_lock); time_out_leases(file_inode(filp)); for (fl = file_inode(filp)->i_flock; fl && IS_LEASE(fl); fl = fl->fl_next) { @@ -1334,11 +1437,11 @@ int fcntl_getlease(struct file *filp) break; } } - unlock_flocks(); + spin_unlock(&inode->i_lock); return type; } -int generic_add_lease(struct file *filp, long arg, struct file_lock **flp) +static int generic_add_lease(struct file *filp, long arg, struct file_lock **flp) { struct file_lock *fl, **before, **my_before = NULL, *lease; struct dentry *dentry = filp->f_path.dentry; @@ -1403,7 +1506,7 @@ out: return error; } -int generic_delete_lease(struct file *filp, struct file_lock **flp) +static int generic_delete_lease(struct file *filp, struct file_lock **flp) { struct file_lock *fl, **before; struct dentry *dentry = filp->f_path.dentry; @@ -1428,7 +1531,7 @@ int generic_delete_lease(struct file *filp, struct file_lock **flp) * The (input) flp->fl_lmops->lm_break function is required * by break_lease(). * - * Called with file_lock_lock held. + * Called with inode->i_lock held. */ int generic_setlease(struct file *filp, long arg, struct file_lock **flp) { @@ -1497,11 +1600,12 @@ static int __vfs_setlease(struct file *filp, long arg, struct file_lock **lease) int vfs_setlease(struct file *filp, long arg, struct file_lock **lease) { + struct inode *inode = file_inode(filp); int error; - lock_flocks(); + spin_lock(&inode->i_lock); error = __vfs_setlease(filp, arg, lease); - unlock_flocks(); + spin_unlock(&inode->i_lock); return error; } @@ -1519,6 +1623,7 @@ static int do_fcntl_delete_lease(struct file *filp) static int do_fcntl_add_lease(unsigned int fd, struct file *filp, long arg) { struct file_lock *fl, *ret; + struct inode *inode = file_inode(filp); struct fasync_struct *new; int error; @@ -1532,10 +1637,10 @@ static int do_fcntl_add_lease(unsigned int fd, struct file *filp, long arg) return -ENOMEM; } ret = fl; - lock_flocks(); + spin_lock(&inode->i_lock); error = __vfs_setlease(filp, arg, &ret); if (error) { - unlock_flocks(); + spin_unlock(&inode->i_lock); locks_free_lock(fl); goto out_free_fasync; } @@ -1552,7 +1657,7 @@ static int do_fcntl_add_lease(unsigned int fd, struct file *filp, long arg) new = NULL; error = __f_setown(filp, task_pid(current), PIDTYPE_PID, 0); - unlock_flocks(); + spin_unlock(&inode->i_lock); out_free_fasync: if (new) @@ -2076,7 +2181,7 @@ void locks_remove_flock(struct file *filp) fl.fl_ops->fl_release_private(&fl); } - lock_flocks(); + spin_lock(&inode->i_lock); before = &inode->i_flock; while ((fl = *before) != NULL) { @@ -2094,30 +2199,28 @@ void locks_remove_flock(struct file *filp) } before = &fl->fl_next; } - unlock_flocks(); + spin_unlock(&inode->i_lock); } /** * posix_unblock_lock - stop waiting for a file lock - * @filp: how the file was opened * @waiter: the lock which was waiting * * lockd needs to block waiting for locks. */ int -posix_unblock_lock(struct file *filp, struct file_lock *waiter) +posix_unblock_lock(struct file_lock *waiter) { int status = 0; - lock_flocks(); + spin_lock(&blocked_lock_lock); if (waiter->fl_next) __locks_delete_block(waiter); else status = -ENOENT; - unlock_flocks(); + spin_unlock(&blocked_lock_lock); return status; } - EXPORT_SYMBOL(posix_unblock_lock); /** @@ -2215,7 +2318,7 @@ static int locks_show(struct seq_file *f, void *v) { struct file_lock *fl, *bfl; - fl = list_entry(v, struct file_lock, fl_link); + fl = hlist_entry(v, struct file_lock, fl_link); lock_get_status(f, fl, *((loff_t *)f->private), ""); @@ -2229,21 +2332,23 @@ static void *locks_start(struct seq_file *f, loff_t *pos) { loff_t *p = f->private; - lock_flocks(); + spin_lock(&file_lock_lock); + spin_lock(&blocked_lock_lock); *p = (*pos + 1); - return seq_list_start(&file_lock_list, *pos); + return seq_hlist_start(&file_lock_list, *pos); } static void *locks_next(struct seq_file *f, void *v, loff_t *pos) { loff_t *p = f->private; ++*p; - return seq_list_next(v, &file_lock_list, pos); + return seq_hlist_next(v, &file_lock_list, pos); } static void locks_stop(struct seq_file *f, void *v) { - unlock_flocks(); + spin_unlock(&blocked_lock_lock); + spin_unlock(&file_lock_lock); } static const struct seq_operations locks_seq_operations = { @@ -2290,7 +2395,8 @@ int lock_may_read(struct inode *inode, loff_t start, unsigned long len) { struct file_lock *fl; int result = 1; - lock_flocks(); + + spin_lock(&inode->i_lock); for (fl = inode->i_flock; fl != NULL; fl = fl->fl_next) { if (IS_POSIX(fl)) { if (fl->fl_type == F_RDLCK) @@ -2307,7 +2413,7 @@ int lock_may_read(struct inode *inode, loff_t start, unsigned long len) result = 0; break; } - unlock_flocks(); + spin_unlock(&inode->i_lock); return result; } @@ -2330,7 +2436,8 @@ int lock_may_write(struct inode *inode, loff_t start, unsigned long len) { struct file_lock *fl; int result = 1; - lock_flocks(); + + spin_lock(&inode->i_lock); for (fl = inode->i_flock; fl != NULL; fl = fl->fl_next) { if (IS_POSIX(fl)) { if ((fl->fl_end < start) || (fl->fl_start > (start + len))) @@ -2345,7 +2452,7 @@ int lock_may_write(struct inode *inode, loff_t start, unsigned long len) result = 0; break; } - unlock_flocks(); + spin_unlock(&inode->i_lock); return result; } diff --git a/fs/minix/dir.c b/fs/minix/dir.c index 08c442902fcd..dfaf6fa9b7b5 100644 --- a/fs/minix/dir.c +++ b/fs/minix/dir.c @@ -93,7 +93,7 @@ static int minix_readdir(struct file *file, struct dir_context *ctx) unsigned offset; unsigned long n; - ctx->pos = pos = (pos + chunk_size-1) & ~(chunk_size-1); + ctx->pos = pos = ALIGN(pos, chunk_size); if (pos >= inode->i_size) return 0; diff --git a/fs/minix/namei.c b/fs/minix/namei.c index 0db73d9dd668..cd950e2331b6 100644 --- a/fs/minix/namei.c +++ b/fs/minix/namei.c @@ -54,6 +54,18 @@ static int minix_mknod(struct inode * dir, struct dentry *dentry, umode_t mode, return error; } +static int minix_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode) +{ + int error; + struct inode *inode = minix_new_inode(dir, mode, &error); + if (inode) { + minix_set_inode(inode, 0); + mark_inode_dirty(inode); + d_tmpfile(dentry, inode); + } + return error; +} + static int minix_create(struct inode *dir, struct dentry *dentry, umode_t mode, bool excl) { @@ -254,4 +266,5 @@ const struct inode_operations minix_dir_inode_operations = { .mknod = minix_mknod, .rename = minix_rename, .getattr = minix_getattr, + .tmpfile = minix_tmpfile, }; diff --git a/fs/namei.c b/fs/namei.c index 9ed9361223c0..b2beee7a733f 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -1352,7 +1352,7 @@ static int lookup_fast(struct nameidata *nd, */ if (nd->flags & LOOKUP_RCU) { unsigned seq; - dentry = __d_lookup_rcu(parent, &nd->last, &seq, nd->inode); + dentry = __d_lookup_rcu(parent, &nd->last, &seq); if (!dentry) goto unlazy; @@ -1787,8 +1787,7 @@ static int link_path_walk(const char *name, struct nameidata *nd) struct dentry *parent = nd->path.dentry; nd->flags &= ~LOOKUP_JUMPED; if (unlikely(parent->d_flags & DCACHE_OP_HASH)) { - err = parent->d_op->d_hash(parent, nd->inode, - &this); + err = parent->d_op->d_hash(parent, &this); if (err < 0) break; } @@ -2121,7 +2120,7 @@ struct dentry *lookup_one_len(const char *name, struct dentry *base, int len) * to use its own hash.. */ if (base->d_flags & DCACHE_OP_HASH) { - int err = base->d_op->d_hash(base, base->d_inode, &this); + int err = base->d_op->d_hash(base, &this); if (err < 0) return ERR_PTR(err); } @@ -2690,28 +2689,10 @@ static int do_last(struct nameidata *nd, struct path *path, nd->flags &= ~LOOKUP_PARENT; nd->flags |= op->intent; - switch (nd->last_type) { - case LAST_DOTDOT: - case LAST_DOT: + if (nd->last_type != LAST_NORM) { error = handle_dots(nd, nd->last_type); if (error) return error; - /* fallthrough */ - case LAST_ROOT: - error = complete_walk(nd); - if (error) - return error; - audit_inode(name, nd->path.dentry, 0); - if (open_flag & O_CREAT) { - error = -EISDIR; - goto out; - } - goto finish_open; - case LAST_BIND: - error = complete_walk(nd); - if (error) - return error; - audit_inode(name, dir, 0); goto finish_open; } @@ -2841,19 +2822,19 @@ finish_lookup: } nd->inode = inode; /* Why this, you ask? _Now_ we might have grown LOOKUP_JUMPED... */ +finish_open: error = complete_walk(nd); if (error) { path_put(&save_parent); return error; } + audit_inode(name, nd->path.dentry, 0); error = -EISDIR; if ((open_flag & O_CREAT) && S_ISDIR(nd->inode->i_mode)) goto out; error = -ENOTDIR; if ((nd->flags & LOOKUP_DIRECTORY) && !can_lookup(nd->inode)) goto out; - audit_inode(name, nd->path.dentry, 0); -finish_open: if (!S_ISREG(nd->inode->i_mode)) will_truncate = false; @@ -2920,6 +2901,67 @@ stale_open: goto retry_lookup; } +static int do_tmpfile(int dfd, struct filename *pathname, + struct nameidata *nd, int flags, + const struct open_flags *op, + struct file *file, int *opened) +{ + static const struct qstr name = QSTR_INIT("/", 1); + struct dentry *dentry, *child; + struct inode *dir; + int error = path_lookupat(dfd, pathname->name, + flags | LOOKUP_DIRECTORY, nd); + if (unlikely(error)) + return error; + error = mnt_want_write(nd->path.mnt); + if (unlikely(error)) + goto out; + /* we want directory to be writable */ + error = inode_permission(nd->inode, MAY_WRITE | MAY_EXEC); + if (error) + goto out2; + dentry = nd->path.dentry; + dir = dentry->d_inode; + if (!dir->i_op->tmpfile) { + error = -EOPNOTSUPP; + goto out2; + } + child = d_alloc(dentry, &name); + if (unlikely(!child)) { + error = -ENOMEM; + goto out2; + } + nd->flags &= ~LOOKUP_DIRECTORY; + nd->flags |= op->intent; + dput(nd->path.dentry); + nd->path.dentry = child; + error = dir->i_op->tmpfile(dir, nd->path.dentry, op->mode); + if (error) + goto out2; + audit_inode(pathname, nd->path.dentry, 0); + error = may_open(&nd->path, op->acc_mode, op->open_flag); + if (error) + goto out2; + file->f_path.mnt = nd->path.mnt; + error = finish_open(file, nd->path.dentry, NULL, opened); + if (error) + goto out2; + error = open_check_o_direct(file); + if (error) { + fput(file); + } else if (!(op->open_flag & O_EXCL)) { + struct inode *inode = file_inode(file); + spin_lock(&inode->i_lock); + inode->i_state |= I_LINKABLE; + spin_unlock(&inode->i_lock); + } +out2: + mnt_drop_write(nd->path.mnt); +out: + path_put(&nd->path); + return error; +} + static struct file *path_openat(int dfd, struct filename *pathname, struct nameidata *nd, const struct open_flags *op, int flags) { @@ -2935,6 +2977,11 @@ static struct file *path_openat(int dfd, struct filename *pathname, file->f_flags = op->open_flag; + if (unlikely(file->f_flags & O_TMPFILE)) { + error = do_tmpfile(dfd, pathname, nd, flags, op, file, &opened); + goto out; + } + error = path_init(dfd, pathname->name, flags | LOOKUP_PARENT, nd, &base); if (unlikely(error)) goto out; @@ -2987,9 +3034,10 @@ out: } struct file *do_filp_open(int dfd, struct filename *pathname, - const struct open_flags *op, int flags) + const struct open_flags *op) { struct nameidata nd; + int flags = op->lookup_flags; struct file *filp; filp = path_openat(dfd, pathname, &nd, op, flags | LOOKUP_RCU); @@ -3001,17 +3049,16 @@ struct file *do_filp_open(int dfd, struct filename *pathname, } struct file *do_file_open_root(struct dentry *dentry, struct vfsmount *mnt, - const char *name, const struct open_flags *op, int flags) + const char *name, const struct open_flags *op) { struct nameidata nd; struct file *file; struct filename filename = { .name = name }; + int flags = op->lookup_flags | LOOKUP_ROOT; nd.root.mnt = mnt; nd.root.dentry = dentry; - flags |= LOOKUP_ROOT; - if (dentry->d_inode->i_op->follow_link && op->intent & LOOKUP_OPEN) return ERR_PTR(-ELOOP); @@ -3586,12 +3633,18 @@ int vfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *new_de mutex_lock(&inode->i_mutex); /* Make sure we don't allow creating hardlink to an unlinked file */ - if (inode->i_nlink == 0) + if (inode->i_nlink == 0 && !(inode->i_state & I_LINKABLE)) error = -ENOENT; else if (max_links && inode->i_nlink >= max_links) error = -EMLINK; else error = dir->i_op->link(old_dentry, dir, new_dentry); + + if (!error && (inode->i_state & I_LINKABLE)) { + spin_lock(&inode->i_lock); + inode->i_state &= ~I_LINKABLE; + spin_unlock(&inode->i_lock); + } mutex_unlock(&inode->i_mutex); if (!error) fsnotify_link(dir, inode, new_dentry); diff --git a/fs/ncpfs/dir.c b/fs/ncpfs/dir.c index 0e7f00298213..3be047474bfc 100644 --- a/fs/ncpfs/dir.c +++ b/fs/ncpfs/dir.c @@ -73,10 +73,8 @@ const struct inode_operations ncp_dir_inode_operations = * Dentry operations routines */ static int ncp_lookup_validate(struct dentry *, unsigned int); -static int ncp_hash_dentry(const struct dentry *, const struct inode *, - struct qstr *); -static int ncp_compare_dentry(const struct dentry *, const struct inode *, - const struct dentry *, const struct inode *, +static int ncp_hash_dentry(const struct dentry *, struct qstr *); +static int ncp_compare_dentry(const struct dentry *, const struct dentry *, unsigned int, const char *, const struct qstr *); static int ncp_delete_dentry(const struct dentry *); @@ -119,11 +117,19 @@ static inline int ncp_case_sensitive(const struct inode *i) /* * Note: leave the hash unchanged if the directory * is case-sensitive. + * + * Accessing the parent inode can be racy under RCU pathwalking. + * Use ACCESS_ONCE() to make sure we use _one_ particular inode, + * the callers will handle races. */ static int -ncp_hash_dentry(const struct dentry *dentry, const struct inode *inode, - struct qstr *this) +ncp_hash_dentry(const struct dentry *dentry, struct qstr *this) { + struct inode *inode = ACCESS_ONCE(dentry->d_inode); + + if (!inode) + return 0; + if (!ncp_case_sensitive(inode)) { struct super_block *sb = dentry->d_sb; struct nls_table *t; @@ -140,14 +146,24 @@ ncp_hash_dentry(const struct dentry *dentry, const struct inode *inode, return 0; } +/* + * Accessing the parent inode can be racy under RCU pathwalking. + * Use ACCESS_ONCE() to make sure we use _one_ particular inode, + * the callers will handle races. + */ static int -ncp_compare_dentry(const struct dentry *parent, const struct inode *pinode, - const struct dentry *dentry, const struct inode *inode, +ncp_compare_dentry(const struct dentry *parent, const struct dentry *dentry, unsigned int len, const char *str, const struct qstr *name) { + struct inode *pinode; + if (len != name->len) return 1; + pinode = ACCESS_ONCE(parent->d_inode); + if (!pinode) + return 1; + if (ncp_case_sensitive(pinode)) return strncmp(str, name->name, len); @@ -660,8 +676,6 @@ end_advance: ctl.valid = 0; if (!ctl.filled && (ctl.fpos == ctx->pos)) { if (!ino) - ino = find_inode_number(dentry, &qname); - if (!ino) ino = iunique(dir->i_sb, 2); ctl.filled = !dir_emit(ctx, qname.name, qname.len, ino, DT_UNKNOWN); @@ -1123,17 +1137,6 @@ static int ncp_rename(struct inode *old_dir, struct dentry *old_dentry, old_dentry->d_parent->d_name.name, old_dentry->d_name.name, new_dentry->d_parent->d_name.name, new_dentry->d_name.name); - if (new_dentry->d_inode && S_ISDIR(new_dentry->d_inode->i_mode)) { - /* - * fail with EBUSY if there are still references to this - * directory. - */ - dentry_unhash(new_dentry); - error = -EBUSY; - if (!d_unhashed(new_dentry)) - goto out; - } - ncp_age_dentry(server, old_dentry); ncp_age_dentry(server, new_dentry); diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c index 26910c8154da..0765ad12c382 100644 --- a/fs/ncpfs/inode.c +++ b/fs/ncpfs/inode.c @@ -891,6 +891,10 @@ int ncp_notify_change(struct dentry *dentry, struct iattr *attr) if (!server) /* How this could happen? */ goto out; + result = -EPERM; + if (IS_DEADDIR(dentry->d_inode)) + goto out; + /* ageing the dentry to force validation */ ncp_age_dentry(server, dentry); diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c index 57db3244f4d9..7ec4814e298d 100644 --- a/fs/nfs/delegation.c +++ b/fs/nfs/delegation.c @@ -73,20 +73,20 @@ static int nfs_delegation_claim_locks(struct nfs_open_context *ctx, struct nfs4_ if (inode->i_flock == NULL) goto out; - /* Protect inode->i_flock using the file locks lock */ - lock_flocks(); + /* Protect inode->i_flock using the i_lock */ + spin_lock(&inode->i_lock); for (fl = inode->i_flock; fl != NULL; fl = fl->fl_next) { if (!(fl->fl_flags & (FL_POSIX|FL_FLOCK))) continue; if (nfs_file_open_context(fl->fl_file) != ctx) continue; - unlock_flocks(); + spin_unlock(&inode->i_lock); status = nfs4_lock_delegation_recall(fl, state, stateid); if (status < 0) goto out; - lock_flocks(); + spin_lock(&inode->i_lock); } - unlock_flocks(); + spin_unlock(&inode->i_lock); out: return status; } diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index 1fab140764c4..ff10b4aa534c 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -1373,13 +1373,13 @@ static int nfs4_reclaim_locks(struct nfs4_state *state, const struct nfs4_state_ /* Guard against delegation returns and new lock/unlock calls */ down_write(&nfsi->rwsem); /* Protect inode->i_flock using the BKL */ - lock_flocks(); + spin_lock(&inode->i_lock); for (fl = inode->i_flock; fl != NULL; fl = fl->fl_next) { if (!(fl->fl_flags & (FL_POSIX|FL_FLOCK))) continue; if (nfs_file_open_context(fl->fl_file)->state != state) continue; - unlock_flocks(); + spin_unlock(&inode->i_lock); status = ops->recover_lock(state, fl); switch (status) { case 0: @@ -1406,9 +1406,9 @@ static int nfs4_reclaim_locks(struct nfs4_state *state, const struct nfs4_state_ /* kill_proc(fl->fl_pid, SIGLOST, 1); */ status = 0; } - lock_flocks(); + spin_lock(&inode->i_lock); } - unlock_flocks(); + spin_unlock(&inode->i_lock); out: up_write(&nfsi->rwsem); return status; diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 316ec843dec2..f17051838b41 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -2645,13 +2645,13 @@ static void nfsd_break_one_deleg(struct nfs4_delegation *dp) list_add_tail(&dp->dl_recall_lru, &nn->del_recall_lru); - /* only place dl_time is set. protected by lock_flocks*/ + /* Only place dl_time is set; protected by i_lock: */ dp->dl_time = get_seconds(); nfsd4_cb_recall(dp); } -/* Called from break_lease() with lock_flocks() held. */ +/* Called from break_lease() with i_lock held. */ static void nfsd_break_deleg_cb(struct file_lock *fl) { struct nfs4_file *fp = (struct nfs4_file *)fl->fl_owner; @@ -4520,7 +4520,7 @@ check_for_locks(struct nfs4_file *filp, struct nfs4_lockowner *lowner) struct inode *inode = filp->fi_inode; int status = 0; - lock_flocks(); + spin_lock(&inode->i_lock); for (flpp = &inode->i_flock; *flpp != NULL; flpp = &(*flpp)->fl_next) { if ((*flpp)->fl_owner == (fl_owner_t)lowner) { status = 1; @@ -4528,7 +4528,7 @@ check_for_locks(struct nfs4_file *filp, struct nfs4_lockowner *lowner) } } out: - unlock_flocks(); + spin_unlock(&inode->i_lock); return status; } diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c index 6c80083a984f..1ea52f7c031f 100644 --- a/fs/notify/fanotify/fanotify_user.c +++ b/fs/notify/fanotify/fanotify_user.c @@ -399,9 +399,6 @@ static int fanotify_release(struct inode *ignored, struct file *file) wake_up(&group->fanotify_data.access_waitq); #endif - if (file->f_flags & FASYNC) - fsnotify_fasync(-1, file, 0); - /* matches the fanotify_init->fsnotify_alloc_group */ fsnotify_destroy_group(group); diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index 8a38714f1d92..41000f223ca4 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -2646,17 +2646,7 @@ static loff_t ocfs2_file_llseek(struct file *file, loff_t offset, int whence) goto out; } - if (offset < 0 && !(file->f_mode & FMODE_UNSIGNED_OFFSET)) - ret = -EINVAL; - if (!ret && offset > inode->i_sb->s_maxbytes) - ret = -EINVAL; - if (ret) - goto out; - - if (offset != file->f_pos) { - file->f_pos = offset; - file->f_version = 0; - } + offset = vfs_setpos(file, offset, inode->i_sb->s_maxbytes); out: mutex_unlock(&inode->i_mutex); diff --git a/fs/open.c b/fs/open.c index 8c741002f947..fca72c4d3f17 100644 --- a/fs/open.c +++ b/fs/open.c @@ -840,11 +840,15 @@ static inline int build_open_flags(int flags, umode_t mode, struct open_flags *o if (flags & __O_SYNC) flags |= O_DSYNC; - /* - * If we have O_PATH in the open flag. Then we - * cannot have anything other than the below set of flags - */ - if (flags & O_PATH) { + if (flags & O_TMPFILE) { + if (!(flags & O_CREAT)) + return -EINVAL; + acc_mode = MAY_OPEN | ACC_MODE(flags); + } else if (flags & O_PATH) { + /* + * If we have O_PATH in the open flag. Then we + * cannot have anything other than the below set of flags + */ flags &= O_DIRECTORY | O_NOFOLLOW | O_PATH; acc_mode = 0; } else { @@ -876,7 +880,8 @@ static inline int build_open_flags(int flags, umode_t mode, struct open_flags *o lookup_flags |= LOOKUP_DIRECTORY; if (!(flags & O_NOFOLLOW)) lookup_flags |= LOOKUP_FOLLOW; - return lookup_flags; + op->lookup_flags = lookup_flags; + return 0; } /** @@ -893,8 +898,8 @@ static inline int build_open_flags(int flags, umode_t mode, struct open_flags *o struct file *file_open_name(struct filename *name, int flags, umode_t mode) { struct open_flags op; - int lookup = build_open_flags(flags, mode, &op); - return do_filp_open(AT_FDCWD, name, &op, lookup); + int err = build_open_flags(flags, mode, &op); + return err ? ERR_PTR(err) : do_filp_open(AT_FDCWD, name, &op); } /** @@ -919,37 +924,43 @@ struct file *file_open_root(struct dentry *dentry, struct vfsmount *mnt, const char *filename, int flags) { struct open_flags op; - int lookup = build_open_flags(flags, 0, &op); + int err = build_open_flags(flags, 0, &op); + if (err) + return ERR_PTR(err); if (flags & O_CREAT) return ERR_PTR(-EINVAL); if (!filename && (flags & O_DIRECTORY)) if (!dentry->d_inode->i_op->lookup) return ERR_PTR(-ENOTDIR); - return do_file_open_root(dentry, mnt, filename, &op, lookup); + return do_file_open_root(dentry, mnt, filename, &op); } EXPORT_SYMBOL(file_open_root); long do_sys_open(int dfd, const char __user *filename, int flags, umode_t mode) { struct open_flags op; - int lookup = build_open_flags(flags, mode, &op); - struct filename *tmp = getname(filename); - int fd = PTR_ERR(tmp); - - if (!IS_ERR(tmp)) { - fd = get_unused_fd_flags(flags); - if (fd >= 0) { - struct file *f = do_filp_open(dfd, tmp, &op, lookup); - if (IS_ERR(f)) { - put_unused_fd(fd); - fd = PTR_ERR(f); - } else { - fsnotify_open(f); - fd_install(fd, f); - } + int fd = build_open_flags(flags, mode, &op); + struct filename *tmp; + + if (fd) + return fd; + + tmp = getname(filename); + if (IS_ERR(tmp)) + return PTR_ERR(tmp); + + fd = get_unused_fd_flags(flags); + if (fd >= 0) { + struct file *f = do_filp_open(dfd, tmp, &op); + if (IS_ERR(f)) { + put_unused_fd(fd); + fd = PTR_ERR(f); + } else { + fsnotify_open(f); + fd_install(fd, f); } - putname(tmp); } + putname(tmp); return fd; } diff --git a/fs/proc/base.c b/fs/proc/base.c index 0016350ad95e..1485e38daaa3 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -1686,41 +1686,29 @@ bool proc_fill_cache(struct file *file, struct dir_context *ctx, instantiate_t instantiate, struct task_struct *task, const void *ptr) { struct dentry *child, *dir = file->f_path.dentry; + struct qstr qname = QSTR_INIT(name, len); struct inode *inode; - struct qstr qname; - ino_t ino = 0; - unsigned type = DT_UNKNOWN; + unsigned type; + ino_t ino; - qname.name = name; - qname.len = len; - qname.hash = full_name_hash(name, len); - - child = d_lookup(dir, &qname); + child = d_hash_and_lookup(dir, &qname); if (!child) { - struct dentry *new; - new = d_alloc(dir, &qname); - if (new) { - child = instantiate(dir->d_inode, new, task, ptr); - if (child) - dput(new); - else - child = new; + child = d_alloc(dir, &qname); + if (!child) + goto end_instantiate; + if (instantiate(dir->d_inode, child, task, ptr) < 0) { + dput(child); + goto end_instantiate; } } - if (!child || IS_ERR(child) || !child->d_inode) - goto end_instantiate; inode = child->d_inode; - if (inode) { - ino = inode->i_ino; - type = inode->i_mode >> 12; - } + ino = inode->i_ino; + type = inode->i_mode >> 12; dput(child); -end_instantiate: - if (!ino) - ino = find_inode_number(dir, &qname); - if (!ino) - ino = 1; return dir_emit(ctx, name, len, ino, type); + +end_instantiate: + return dir_emit(ctx, name, len, 1, DT_UNKNOWN); } #ifdef CONFIG_CHECKPOINT_RESTORE @@ -1846,7 +1834,7 @@ struct map_files_info { unsigned char name[4*sizeof(long)+2]; /* max: %lx-%lx\0 */ }; -static struct dentry * +static int proc_map_files_instantiate(struct inode *dir, struct dentry *dentry, struct task_struct *task, const void *ptr) { @@ -1856,7 +1844,7 @@ proc_map_files_instantiate(struct inode *dir, struct dentry *dentry, inode = proc_pid_make_inode(dir->i_sb, task); if (!inode) - return ERR_PTR(-ENOENT); + return -ENOENT; ei = PROC_I(inode); ei->op.proc_get_link = proc_map_files_get_link; @@ -1873,7 +1861,7 @@ proc_map_files_instantiate(struct inode *dir, struct dentry *dentry, d_set_d_op(dentry, &tid_map_files_dentry_operations); d_add(dentry, inode); - return NULL; + return 0; } static struct dentry *proc_map_files_lookup(struct inode *dir, @@ -1882,23 +1870,23 @@ static struct dentry *proc_map_files_lookup(struct inode *dir, unsigned long vm_start, vm_end; struct vm_area_struct *vma; struct task_struct *task; - struct dentry *result; + int result; struct mm_struct *mm; - result = ERR_PTR(-EPERM); + result = -EPERM; if (!capable(CAP_SYS_ADMIN)) goto out; - result = ERR_PTR(-ENOENT); + result = -ENOENT; task = get_proc_task(dir); if (!task) goto out; - result = ERR_PTR(-EACCES); + result = -EACCES; if (!ptrace_may_access(task, PTRACE_MODE_READ)) goto out_put_task; - result = ERR_PTR(-ENOENT); + result = -ENOENT; if (dname_to_vma_addr(dentry, &vm_start, &vm_end)) goto out_put_task; @@ -1921,7 +1909,7 @@ out_no_vma: out_put_task: put_task_struct(task); out: - return result; + return ERR_PTR(result); } static const struct inode_operations proc_map_files_inode_operations = { @@ -2135,13 +2123,12 @@ static const struct file_operations proc_timers_operations = { }; #endif /* CONFIG_CHECKPOINT_RESTORE */ -static struct dentry *proc_pident_instantiate(struct inode *dir, +static int proc_pident_instantiate(struct inode *dir, struct dentry *dentry, struct task_struct *task, const void *ptr) { const struct pid_entry *p = ptr; struct inode *inode; struct proc_inode *ei; - struct dentry *error = ERR_PTR(-ENOENT); inode = proc_pid_make_inode(dir->i_sb, task); if (!inode) @@ -2160,9 +2147,9 @@ static struct dentry *proc_pident_instantiate(struct inode *dir, d_add(dentry, inode); /* Close the race of the process dying before we return the dentry */ if (pid_revalidate(dentry, 0)) - error = NULL; + return 0; out: - return error; + return -ENOENT; } static struct dentry *proc_pident_lookup(struct inode *dir, @@ -2170,11 +2157,11 @@ static struct dentry *proc_pident_lookup(struct inode *dir, const struct pid_entry *ents, unsigned int nents) { - struct dentry *error; + int error; struct task_struct *task = get_proc_task(dir); const struct pid_entry *p, *last; - error = ERR_PTR(-ENOENT); + error = -ENOENT; if (!task) goto out_no_task; @@ -2197,7 +2184,7 @@ static struct dentry *proc_pident_lookup(struct inode *dir, out: put_task_struct(task); out_no_task: - return error; + return ERR_PTR(error); } static int proc_pident_readdir(struct file *file, struct dir_context *ctx, @@ -2780,11 +2767,10 @@ void proc_flush_task(struct task_struct *task) } } -static struct dentry *proc_pid_instantiate(struct inode *dir, - struct dentry * dentry, - struct task_struct *task, const void *ptr) +static int proc_pid_instantiate(struct inode *dir, + struct dentry * dentry, + struct task_struct *task, const void *ptr) { - struct dentry *error = ERR_PTR(-ENOENT); struct inode *inode; inode = proc_pid_make_inode(dir->i_sb, task); @@ -2804,14 +2790,14 @@ static struct dentry *proc_pid_instantiate(struct inode *dir, d_add(dentry, inode); /* Close the race of the process dying before we return the dentry */ if (pid_revalidate(dentry, 0)) - error = NULL; + return 0; out: - return error; + return -ENOENT; } struct dentry *proc_pid_lookup(struct inode *dir, struct dentry * dentry, unsigned int flags) { - struct dentry *result = NULL; + int result = 0; struct task_struct *task; unsigned tgid; struct pid_namespace *ns; @@ -2832,7 +2818,7 @@ struct dentry *proc_pid_lookup(struct inode *dir, struct dentry * dentry, unsign result = proc_pid_instantiate(dir, dentry, task, NULL); put_task_struct(task); out: - return result; + return ERR_PTR(result); } /* @@ -2884,21 +2870,21 @@ retry: int proc_pid_readdir(struct file *file, struct dir_context *ctx) { struct tgid_iter iter; - struct pid_namespace *ns; + struct pid_namespace *ns = file->f_dentry->d_sb->s_fs_info; loff_t pos = ctx->pos; if (pos >= PID_MAX_LIMIT + TGID_OFFSET) return 0; if (pos == TGID_OFFSET - 1) { - if (!proc_fill_cache(file, ctx, "self", 4, NULL, NULL, NULL)) + struct inode *inode = ns->proc_self->d_inode; + if (!dir_emit(ctx, "self", 4, inode->i_ino, DT_LNK)) return 0; iter.tgid = 0; } else { iter.tgid = pos - TGID_OFFSET; } iter.task = NULL; - ns = file->f_dentry->d_sb->s_fs_info; for (iter = next_tgid(ns, iter); iter.task; iter.tgid += 1, iter = next_tgid(ns, iter)) { @@ -3027,10 +3013,9 @@ static const struct inode_operations proc_tid_base_inode_operations = { .setattr = proc_setattr, }; -static struct dentry *proc_task_instantiate(struct inode *dir, +static int proc_task_instantiate(struct inode *dir, struct dentry *dentry, struct task_struct *task, const void *ptr) { - struct dentry *error = ERR_PTR(-ENOENT); struct inode *inode; inode = proc_pid_make_inode(dir->i_sb, task); @@ -3049,14 +3034,14 @@ static struct dentry *proc_task_instantiate(struct inode *dir, d_add(dentry, inode); /* Close the race of the process dying before we return the dentry */ if (pid_revalidate(dentry, 0)) - error = NULL; + return 0; out: - return error; + return -ENOENT; } static struct dentry *proc_task_lookup(struct inode *dir, struct dentry * dentry, unsigned int flags) { - struct dentry *result = ERR_PTR(-ENOENT); + int result = -ENOENT; struct task_struct *task; struct task_struct *leader = get_proc_task(dir); unsigned tid; @@ -3086,7 +3071,7 @@ out_drop_task: out: put_task_struct(leader); out_no_task: - return result; + return ERR_PTR(result); } /* diff --git a/fs/proc/fd.c b/fs/proc/fd.c index 1441f143c43b..75f2890abbd8 100644 --- a/fs/proc/fd.c +++ b/fs/proc/fd.c @@ -167,11 +167,10 @@ static int proc_fd_link(struct dentry *dentry, struct path *path) return ret; } -static struct dentry * +static int proc_fd_instantiate(struct inode *dir, struct dentry *dentry, struct task_struct *task, const void *ptr) { - struct dentry *error = ERR_PTR(-ENOENT); unsigned fd = (unsigned long)ptr; struct proc_inode *ei; struct inode *inode; @@ -194,9 +193,9 @@ proc_fd_instantiate(struct inode *dir, struct dentry *dentry, /* Close the race of the process dying before we return the dentry */ if (tid_fd_revalidate(dentry, 0)) - error = NULL; + return 0; out: - return error; + return -ENOENT; } static struct dentry *proc_lookupfd_common(struct inode *dir, @@ -204,7 +203,7 @@ static struct dentry *proc_lookupfd_common(struct inode *dir, instantiate_t instantiate) { struct task_struct *task = get_proc_task(dir); - struct dentry *result = ERR_PTR(-ENOENT); + int result = -ENOENT; unsigned fd = name_to_int(dentry); if (!task) @@ -216,7 +215,7 @@ static struct dentry *proc_lookupfd_common(struct inode *dir, out: put_task_struct(task); out_no_task: - return result; + return ERR_PTR(result); } static int proc_readfd_common(struct file *file, struct dir_context *ctx, @@ -300,11 +299,10 @@ const struct inode_operations proc_fd_inode_operations = { .setattr = proc_setattr, }; -static struct dentry * +static int proc_fdinfo_instantiate(struct inode *dir, struct dentry *dentry, struct task_struct *task, const void *ptr) { - struct dentry *error = ERR_PTR(-ENOENT); unsigned fd = (unsigned long)ptr; struct proc_inode *ei; struct inode *inode; @@ -324,9 +322,9 @@ proc_fdinfo_instantiate(struct inode *dir, struct dentry *dentry, /* Close the race of the process dying before we return the dentry */ if (tid_fd_revalidate(dentry, 0)) - error = NULL; + return 0; out: - return error; + return -ENOENT; } static struct dentry * diff --git a/fs/proc/internal.h b/fs/proc/internal.h index 4eae2e149f31..651d09a11dde 100644 --- a/fs/proc/internal.h +++ b/fs/proc/internal.h @@ -170,7 +170,7 @@ extern struct dentry *proc_pid_lookup(struct inode *, struct dentry *, unsigned extern loff_t mem_lseek(struct file *, loff_t, int); /* Lookups */ -typedef struct dentry *instantiate_t(struct inode *, struct dentry *, +typedef int instantiate_t(struct inode *, struct dentry *, struct task_struct *, const void *); extern bool proc_fill_cache(struct file *, struct dir_context *, const char *, int, instantiate_t, struct task_struct *, const void *); diff --git a/fs/proc/namespaces.c b/fs/proc/namespaces.c index f6abbbbfad8a..49a7fff2e83a 100644 --- a/fs/proc/namespaces.c +++ b/fs/proc/namespaces.c @@ -187,13 +187,12 @@ static const struct inode_operations proc_ns_link_inode_operations = { .setattr = proc_setattr, }; -static struct dentry *proc_ns_instantiate(struct inode *dir, +static int proc_ns_instantiate(struct inode *dir, struct dentry *dentry, struct task_struct *task, const void *ptr) { const struct proc_ns_operations *ns_ops = ptr; struct inode *inode; struct proc_inode *ei; - struct dentry *error = ERR_PTR(-ENOENT); inode = proc_pid_make_inode(dir->i_sb, task); if (!inode) @@ -208,9 +207,9 @@ static struct dentry *proc_ns_instantiate(struct inode *dir, d_add(dentry, inode); /* Close the race of the process dying before we return the dentry */ if (pid_revalidate(dentry, 0)) - error = NULL; + return 0; out: - return error; + return -ENOENT; } static int proc_ns_dir_readdir(struct file *file, struct dir_context *ctx) @@ -248,12 +247,12 @@ const struct file_operations proc_ns_dir_operations = { static struct dentry *proc_ns_dir_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) { - struct dentry *error; + int error; struct task_struct *task = get_proc_task(dir); const struct proc_ns_operations **entry, **last; unsigned int len = dentry->d_name.len; - error = ERR_PTR(-ENOENT); + error = -ENOENT; if (!task) goto out_no_task; @@ -272,7 +271,7 @@ static struct dentry *proc_ns_dir_lookup(struct inode *dir, out: put_task_struct(task); out_no_task: - return error; + return ERR_PTR(error); } const struct inode_operations proc_ns_dir_inode_operations = { diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c index f3a570e7c257..71290463a1d3 100644 --- a/fs/proc/proc_sysctl.c +++ b/fs/proc/proc_sysctl.c @@ -796,15 +796,16 @@ static int sysctl_is_seen(struct ctl_table_header *p) return res; } -static int proc_sys_compare(const struct dentry *parent, - const struct inode *pinode, - const struct dentry *dentry, const struct inode *inode, +static int proc_sys_compare(const struct dentry *parent, const struct dentry *dentry, unsigned int len, const char *str, const struct qstr *name) { struct ctl_table_header *head; + struct inode *inode; + /* Although proc doesn't have negative dentries, rcu-walk means * that inode here can be NULL */ /* AV: can it, indeed? */ + inode = ACCESS_ONCE(dentry->d_inode); if (!inode) return 1; if (name->len != len) diff --git a/fs/read_write.c b/fs/read_write.c index 2cefa417be34..122a3846d9e1 100644 --- a/fs/read_write.c +++ b/fs/read_write.c @@ -41,8 +41,19 @@ static inline int unsigned_offsets(struct file *file) return file->f_mode & FMODE_UNSIGNED_OFFSET; } -static loff_t lseek_execute(struct file *file, struct inode *inode, - loff_t offset, loff_t maxsize) +/** + * vfs_setpos - update the file offset for lseek + * @file: file structure in question + * @offset: file offset to seek to + * @maxsize: maximum file size + * + * This is a low-level filesystem helper for updating the file offset to + * the value specified by @offset if the given offset is valid and it is + * not equal to the current file offset. + * + * Return the specified offset on success and -EINVAL on invalid offset. + */ +loff_t vfs_setpos(struct file *file, loff_t offset, loff_t maxsize) { if (offset < 0 && !unsigned_offsets(file)) return -EINVAL; @@ -55,6 +66,7 @@ static loff_t lseek_execute(struct file *file, struct inode *inode, } return offset; } +EXPORT_SYMBOL(vfs_setpos); /** * generic_file_llseek_size - generic llseek implementation for regular files @@ -76,8 +88,6 @@ loff_t generic_file_llseek_size(struct file *file, loff_t offset, int whence, loff_t maxsize, loff_t eof) { - struct inode *inode = file->f_mapping->host; - switch (whence) { case SEEK_END: offset += eof; @@ -97,8 +107,7 @@ generic_file_llseek_size(struct file *file, loff_t offset, int whence, * like SEEK_SET. */ spin_lock(&file->f_lock); - offset = lseek_execute(file, inode, file->f_pos + offset, - maxsize); + offset = vfs_setpos(file, file->f_pos + offset, maxsize); spin_unlock(&file->f_lock); return offset; case SEEK_DATA: @@ -120,7 +129,7 @@ generic_file_llseek_size(struct file *file, loff_t offset, int whence, break; } - return lseek_execute(file, inode, offset, maxsize); + return vfs_setpos(file, offset, maxsize); } EXPORT_SYMBOL(generic_file_llseek_size); @@ -145,6 +154,26 @@ loff_t generic_file_llseek(struct file *file, loff_t offset, int whence) EXPORT_SYMBOL(generic_file_llseek); /** + * fixed_size_llseek - llseek implementation for fixed-sized devices + * @file: file structure to seek on + * @offset: file offset to seek to + * @whence: type of seek + * @size: size of the file + * + */ +loff_t fixed_size_llseek(struct file *file, loff_t offset, int whence, loff_t size) +{ + switch (whence) { + case SEEK_SET: case SEEK_CUR: case SEEK_END: + return generic_file_llseek_size(file, offset, whence, + size, size); + default: + return -EINVAL; + } +} +EXPORT_SYMBOL(fixed_size_llseek); + +/** * noop_llseek - No Operation Performed llseek implementation * @file: file structure to seek on * @offset: file offset to seek to @@ -296,7 +325,7 @@ out_putf: * them to something that fits in "int" so that others * won't have to do range checks all the time. */ -int rw_verify_area(int read_write, struct file *file, loff_t *ppos, size_t count) +int rw_verify_area(int read_write, struct file *file, const loff_t *ppos, size_t count) { struct inode *inode; loff_t pos; @@ -477,7 +506,8 @@ SYSCALL_DEFINE3(read, unsigned int, fd, char __user *, buf, size_t, count) if (f.file) { loff_t pos = file_pos_read(f.file); ret = vfs_read(f.file, buf, count, &pos); - file_pos_write(f.file, pos); + if (ret >= 0) + file_pos_write(f.file, pos); fdput(f); } return ret; @@ -492,7 +522,8 @@ SYSCALL_DEFINE3(write, unsigned int, fd, const char __user *, buf, if (f.file) { loff_t pos = file_pos_read(f.file); ret = vfs_write(f.file, buf, count, &pos); - file_pos_write(f.file, pos); + if (ret >= 0) + file_pos_write(f.file, pos); fdput(f); } @@ -780,7 +811,8 @@ SYSCALL_DEFINE3(readv, unsigned long, fd, const struct iovec __user *, vec, if (f.file) { loff_t pos = file_pos_read(f.file); ret = vfs_readv(f.file, vec, vlen, &pos); - file_pos_write(f.file, pos); + if (ret >= 0) + file_pos_write(f.file, pos); fdput(f); } @@ -799,7 +831,8 @@ SYSCALL_DEFINE3(writev, unsigned long, fd, const struct iovec __user *, vec, if (f.file) { loff_t pos = file_pos_read(f.file); ret = vfs_writev(f.file, vec, vlen, &pos); - file_pos_write(f.file, pos); + if (ret >= 0) + file_pos_write(f.file, pos); fdput(f); } @@ -959,7 +992,8 @@ COMPAT_SYSCALL_DEFINE3(readv, unsigned long, fd, return -EBADF; pos = f.file->f_pos; ret = compat_readv(f.file, vec, vlen, &pos); - f.file->f_pos = pos; + if (ret >= 0) + f.file->f_pos = pos; fdput(f); return ret; } @@ -1025,7 +1059,8 @@ COMPAT_SYSCALL_DEFINE3(writev, unsigned long, fd, return -EBADF; pos = f.file->f_pos; ret = compat_writev(f.file, vec, vlen, &pos); - f.file->f_pos = pos; + if (ret >= 0) + f.file->f_pos = pos; fdput(f); return ret; } @@ -1129,7 +1164,9 @@ static ssize_t do_sendfile(int out_fd, int in_fd, loff_t *ppos, if (in.file->f_flags & O_NONBLOCK) fl = SPLICE_F_NONBLOCK; #endif + file_start_write(out.file); retval = do_splice_direct(in.file, &pos, out.file, &out_pos, count, fl); + file_end_write(out.file); if (retval > 0) { add_rchar(current, retval); diff --git a/fs/splice.c b/fs/splice.c index d37431dd60a1..3b7ee656f3aa 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -1098,27 +1098,13 @@ static long do_splice_from(struct pipe_inode_info *pipe, struct file *out, { ssize_t (*splice_write)(struct pipe_inode_info *, struct file *, loff_t *, size_t, unsigned int); - int ret; - - if (unlikely(!(out->f_mode & FMODE_WRITE))) - return -EBADF; - - if (unlikely(out->f_flags & O_APPEND)) - return -EINVAL; - - ret = rw_verify_area(WRITE, out, ppos, len); - if (unlikely(ret < 0)) - return ret; if (out->f_op && out->f_op->splice_write) splice_write = out->f_op->splice_write; else splice_write = default_file_splice_write; - file_start_write(out); - ret = splice_write(pipe, out, ppos, len, flags); - file_end_write(out); - return ret; + return splice_write(pipe, out, ppos, len, flags); } /* @@ -1307,6 +1293,16 @@ long do_splice_direct(struct file *in, loff_t *ppos, struct file *out, }; long ret; + if (unlikely(!(out->f_mode & FMODE_WRITE))) + return -EBADF; + + if (unlikely(out->f_flags & O_APPEND)) + return -EINVAL; + + ret = rw_verify_area(WRITE, out, opos, len); + if (unlikely(ret < 0)) + return ret; + ret = splice_direct_to_actor(in, &sd, direct_splice_actor); if (ret > 0) *ppos = sd.pos; @@ -1362,7 +1358,19 @@ static long do_splice(struct file *in, loff_t __user *off_in, offset = out->f_pos; } + if (unlikely(!(out->f_mode & FMODE_WRITE))) + return -EBADF; + + if (unlikely(out->f_flags & O_APPEND)) + return -EINVAL; + + ret = rw_verify_area(WRITE, out, &offset, len); + if (unlikely(ret < 0)) + return ret; + + file_start_write(out); ret = do_splice_from(ipipe, out, &offset, len, flags); + file_end_write(out); if (!off_out) out->f_pos = offset; diff --git a/fs/sysv/namei.c b/fs/sysv/namei.c index 1c0d5f264767..731b2bbcaab3 100644 --- a/fs/sysv/namei.c +++ b/fs/sysv/namei.c @@ -27,8 +27,7 @@ static int add_nondir(struct dentry *dentry, struct inode *inode) return err; } -static int sysv_hash(const struct dentry *dentry, const struct inode *inode, - struct qstr *qstr) +static int sysv_hash(const struct dentry *dentry, struct qstr *qstr) { /* Truncate the name in place, avoids having to define a compare function. */ diff --git a/fs/udf/namei.c b/fs/udf/namei.c index 102c072c6bbf..5f6fc17d6bc5 100644 --- a/fs/udf/namei.c +++ b/fs/udf/namei.c @@ -594,6 +594,29 @@ static int udf_create(struct inode *dir, struct dentry *dentry, umode_t mode, return 0; } +static int udf_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode) +{ + struct inode *inode; + struct udf_inode_info *iinfo; + int err; + + inode = udf_new_inode(dir, mode, &err); + if (!inode) + return err; + + iinfo = UDF_I(inode); + if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB) + inode->i_data.a_ops = &udf_adinicb_aops; + else + inode->i_data.a_ops = &udf_aops; + inode->i_op = &udf_file_inode_operations; + inode->i_fop = &udf_file_operations; + mark_inode_dirty(inode); + + d_tmpfile(dentry, inode); + return 0; +} + static int udf_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t rdev) { @@ -1311,6 +1334,7 @@ const struct inode_operations udf_dir_inode_operations = { .rmdir = udf_rmdir, .mknod = udf_mknod, .rename = udf_rename, + .tmpfile = udf_tmpfile, }; const struct inode_operations udf_symlink_inode_operations = { .readlink = generic_readlink, diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index 0ad2b95fca12..de3dc98f4e8f 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -1268,8 +1268,7 @@ xfs_seek_data( } out: - if (offset != file->f_pos) - file->f_pos = offset; + offset = vfs_setpos(file, offset, inode->i_sb->s_maxbytes); out_unlock: xfs_iunlock_map_shared(ip, lock); @@ -1377,8 +1376,7 @@ out: * situation in particular. */ offset = min_t(loff_t, offset, isize); - if (offset != file->f_pos) - file->f_pos = offset; + offset = vfs_setpos(file, offset, inode->i_sb->s_maxbytes); out_unlock: xfs_iunlock_map_shared(ip, lock); diff --git a/include/linux/dcache.h b/include/linux/dcache.h index 1a6bb81f0fe5..f42dbe145479 100644 --- a/include/linux/dcache.h +++ b/include/linux/dcache.h @@ -146,10 +146,8 @@ enum dentry_d_lock_class struct dentry_operations { int (*d_revalidate)(struct dentry *, unsigned int); int (*d_weak_revalidate)(struct dentry *, unsigned int); - int (*d_hash)(const struct dentry *, const struct inode *, - struct qstr *); - int (*d_compare)(const struct dentry *, const struct inode *, - const struct dentry *, const struct inode *, + int (*d_hash)(const struct dentry *, struct qstr *); + int (*d_compare)(const struct dentry *, const struct dentry *, unsigned int, const char *, const struct qstr *); int (*d_delete)(const struct dentry *); void (*d_release)(struct dentry *); @@ -246,6 +244,8 @@ extern struct dentry * d_make_root(struct inode *); /* <clickety>-<click> the ramfs-type tree */ extern void d_genocide(struct dentry *); +extern void d_tmpfile(struct dentry *, struct inode *); + extern struct dentry *d_find_alias(struct inode *); extern void d_prune_aliases(struct inode *); @@ -300,8 +300,7 @@ extern struct dentry *d_lookup(const struct dentry *, const struct qstr *); extern struct dentry *d_hash_and_lookup(struct dentry *, struct qstr *); extern struct dentry *__d_lookup(const struct dentry *, const struct qstr *); extern struct dentry *__d_lookup_rcu(const struct dentry *parent, - const struct qstr *name, - unsigned *seq, struct inode *inode); + const struct qstr *name, unsigned *seq); /** * __d_rcu_to_refcount - take a refcount on dentry if sequence check is ok diff --git a/include/linux/fs.h b/include/linux/fs.h index f8a5240541b7..2b82c8041490 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -908,6 +908,7 @@ struct file_lock_operations { struct lock_manager_operations { int (*lm_compare_owner)(struct file_lock *, struct file_lock *); + unsigned long (*lm_owner_key)(struct file_lock *); void (*lm_notify)(struct file_lock *); /* unblock callback */ int (*lm_grant)(struct file_lock *, struct file_lock *, int); void (*lm_break)(struct file_lock *); @@ -926,9 +927,27 @@ int locks_in_grace(struct net *); /* that will die - we need it for nfs_lock_info */ #include <linux/nfs_fs_i.h> +/* + * struct file_lock represents a generic "file lock". It's used to represent + * POSIX byte range locks, BSD (flock) locks, and leases. It's important to + * note that the same struct is used to represent both a request for a lock and + * the lock itself, but the same object is never used for both. + * + * FIXME: should we create a separate "struct lock_request" to help distinguish + * these two uses? + * + * The i_flock list is ordered by: + * + * 1) lock type -- FL_LEASEs first, then FL_FLOCK, and finally FL_POSIX + * 2) lock owner + * 3) lock range start + * 4) lock range end + * + * Obviously, the last two criteria only matter for POSIX locks. + */ struct file_lock { struct file_lock *fl_next; /* singly linked list for this inode */ - struct list_head fl_link; /* doubly linked list of all locks */ + struct hlist_node fl_link; /* node in global lists */ struct list_head fl_block; /* circular list of blocked processes */ fl_owner_t fl_owner; unsigned int fl_flags; @@ -994,7 +1013,7 @@ extern void locks_release_private(struct file_lock *); extern void posix_test_lock(struct file *, struct file_lock *); extern int posix_lock_file(struct file *, struct file_lock *, struct file_lock *); extern int posix_lock_file_wait(struct file *, struct file_lock *); -extern int posix_unblock_lock(struct file *, struct file_lock *); +extern int posix_unblock_lock(struct file_lock *); extern int vfs_test_lock(struct file *, struct file_lock *); extern int vfs_lock_file(struct file *, unsigned int, struct file_lock *, struct file_lock *); extern int vfs_cancel_lock(struct file *filp, struct file_lock *fl); @@ -1006,9 +1025,6 @@ extern int vfs_setlease(struct file *, long, struct file_lock **); extern int lease_modify(struct file_lock **, int); extern int lock_may_read(struct inode *, loff_t start, unsigned long count); extern int lock_may_write(struct inode *, loff_t start, unsigned long count); -extern void locks_delete_block(struct file_lock *waiter); -extern void lock_flocks(void); -extern void unlock_flocks(void); #else /* !CONFIG_FILE_LOCKING */ static inline int fcntl_getlk(struct file *file, struct flock __user *user) { @@ -1084,8 +1100,7 @@ static inline int posix_lock_file_wait(struct file *filp, struct file_lock *fl) return -ENOLCK; } -static inline int posix_unblock_lock(struct file *filp, - struct file_lock *waiter) +static inline int posix_unblock_lock(struct file_lock *waiter) { return -ENOENT; } @@ -1150,19 +1165,6 @@ static inline int lock_may_write(struct inode *inode, loff_t start, { return 1; } - -static inline void locks_delete_block(struct file_lock *waiter) -{ -} - -static inline void lock_flocks(void) -{ -} - -static inline void unlock_flocks(void) -{ -} - #endif /* !CONFIG_FILE_LOCKING */ @@ -1580,6 +1582,7 @@ struct inode_operations { int (*atomic_open)(struct inode *, struct dentry *, struct file *, unsigned open_flag, umode_t create_mode, int *opened); + int (*tmpfile) (struct inode *, struct dentry *, umode_t); } ____cacheline_aligned; ssize_t rw_copy_check_uvector(int type, const struct iovec __user * uvector, @@ -1743,6 +1746,7 @@ struct super_operations { #define I_REFERENCED (1 << 8) #define __I_DIO_WAKEUP 9 #define I_DIO_WAKEUP (1 << I_DIO_WAKEUP) +#define I_LINKABLE (1 << 10) #define I_DIRTY (I_DIRTY_SYNC | I_DIRTY_DATASYNC | I_DIRTY_PAGES) @@ -1896,7 +1900,6 @@ extern int current_umask(void); extern struct kobject *fs_kobj; #define MAX_RW_COUNT (INT_MAX & PAGE_CACHE_MASK) -extern int rw_verify_area(int, struct file *, loff_t *, size_t); #define FLOCK_VERIFY_READ 1 #define FLOCK_VERIFY_WRITE 2 @@ -2309,7 +2312,6 @@ extern struct file * open_exec(const char *); /* fs/dcache.c -- generic fs support functions */ extern int is_subdir(struct dentry *, struct dentry *); extern int path_is_under(struct path *, struct path *); -extern ino_t find_inode_number(struct dentry *, struct qstr *); #include <linux/err.h> @@ -2424,9 +2426,12 @@ extern void file_ra_state_init(struct file_ra_state *ra, struct address_space *mapping); extern loff_t noop_llseek(struct file *file, loff_t offset, int whence); extern loff_t no_llseek(struct file *file, loff_t offset, int whence); +extern loff_t vfs_setpos(struct file *file, loff_t offset, loff_t maxsize); extern loff_t generic_file_llseek(struct file *file, loff_t offset, int whence); extern loff_t generic_file_llseek_size(struct file *file, loff_t offset, int whence, loff_t maxsize, loff_t eof); +extern loff_t fixed_size_llseek(struct file *file, loff_t offset, + int whence, loff_t size); extern int generic_file_open(struct inode * inode, struct file * filp); extern int nonseekable_open(struct inode * inode, struct file * filp); diff --git a/include/linux/fsnotify.h b/include/linux/fsnotify.h index a78680a92dba..1c804b057fb1 100644 --- a/include/linux/fsnotify.h +++ b/include/linux/fsnotify.h @@ -38,7 +38,7 @@ static inline int fsnotify_parent(struct path *path, struct dentry *dentry, __u3 static inline int fsnotify_perm(struct file *file, int mask) { struct path *path = &file->f_path; - struct inode *inode = path->dentry->d_inode; + struct inode *inode = file_inode(file); __u32 fsnotify_mask = 0; int ret; @@ -192,7 +192,7 @@ static inline void fsnotify_mkdir(struct inode *inode, struct dentry *dentry) static inline void fsnotify_access(struct file *file) { struct path *path = &file->f_path; - struct inode *inode = path->dentry->d_inode; + struct inode *inode = file_inode(file); __u32 mask = FS_ACCESS; if (S_ISDIR(inode->i_mode)) @@ -210,7 +210,7 @@ static inline void fsnotify_access(struct file *file) static inline void fsnotify_modify(struct file *file) { struct path *path = &file->f_path; - struct inode *inode = path->dentry->d_inode; + struct inode *inode = file_inode(file); __u32 mask = FS_MODIFY; if (S_ISDIR(inode->i_mode)) @@ -228,7 +228,7 @@ static inline void fsnotify_modify(struct file *file) static inline void fsnotify_open(struct file *file) { struct path *path = &file->f_path; - struct inode *inode = path->dentry->d_inode; + struct inode *inode = file_inode(file); __u32 mask = FS_OPEN; if (S_ISDIR(inode->i_mode)) diff --git a/include/uapi/asm-generic/fcntl.h b/include/uapi/asm-generic/fcntl.h index a48937d4a5ea..06632beaa6d5 100644 --- a/include/uapi/asm-generic/fcntl.h +++ b/include/uapi/asm-generic/fcntl.h @@ -84,6 +84,10 @@ #define O_PATH 010000000 #endif +#ifndef O_TMPFILE +#define O_TMPFILE 020000000 +#endif + #ifndef O_NDELAY #define O_NDELAY O_NONBLOCK #endif diff --git a/mm/shmem.c b/mm/shmem.c index 5e6a8422658b..118dfa4952f4 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -1798,10 +1798,7 @@ static loff_t shmem_file_llseek(struct file *file, loff_t offset, int whence) } } - if (offset >= 0 && offset != file->f_pos) { - file->f_pos = offset; - file->f_version = 0; - } + offset = vfs_setpos(file, offset, MAX_LFS_FILESIZE); mutex_unlock(&inode->i_mutex); return offset; } @@ -1965,6 +1962,37 @@ shmem_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t dev) return error; } +static int +shmem_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode) +{ + struct inode *inode; + int error = -ENOSPC; + + inode = shmem_get_inode(dir->i_sb, dir, mode, 0, VM_NORESERVE); + if (inode) { + error = security_inode_init_security(inode, dir, + NULL, + shmem_initxattrs, NULL); + if (error) { + if (error != -EOPNOTSUPP) { + iput(inode); + return error; + } + } +#ifdef CONFIG_TMPFS_POSIX_ACL + error = generic_acl_init(inode, dir); + if (error) { + iput(inode); + return error; + } +#else + error = 0; +#endif + d_tmpfile(dentry, inode); + } + return error; +} + static int shmem_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { int error; @@ -2723,6 +2751,7 @@ static const struct inode_operations shmem_dir_inode_operations = { .rmdir = shmem_rmdir, .mknod = shmem_mknod, .rename = shmem_rename, + .tmpfile = shmem_tmpfile, #endif #ifdef CONFIG_TMPFS_XATTR .setxattr = shmem_setxattr, diff --git a/net/sunrpc/auth_gss/svcauth_gss.c b/net/sunrpc/auth_gss/svcauth_gss.c index 29b4ba93ab3c..b05ace4c5f12 100644 --- a/net/sunrpc/auth_gss/svcauth_gss.c +++ b/net/sunrpc/auth_gss/svcauth_gss.c @@ -1330,7 +1330,7 @@ static int wait_for_gss_proxy(struct net *net, struct file *file) static ssize_t write_gssp(struct file *file, const char __user *buf, size_t count, loff_t *ppos) { - struct net *net = PDE_DATA(file->f_path.dentry->d_inode); + struct net *net = PDE_DATA(file_inode(file)); char tbuf[20]; unsigned long i; int res; @@ -1358,7 +1358,7 @@ static ssize_t write_gssp(struct file *file, const char __user *buf, static ssize_t read_gssp(struct file *file, char __user *buf, size_t count, loff_t *ppos) { - struct net *net = PDE_DATA(file->f_path.dentry->d_inode); + struct net *net = PDE_DATA(file_inode(file)); unsigned long p = *ppos; char tbuf[10]; size_t len; diff --git a/security/integrity/ima/ima_main.c b/security/integrity/ima/ima_main.c index 6c491a63128e..e9508d5bbfcf 100644 --- a/security/integrity/ima/ima_main.c +++ b/security/integrity/ima/ima_main.c @@ -57,7 +57,7 @@ __setup("ima_hash=", hash_setup); static void ima_rdwr_violation_check(struct file *file) { struct dentry *dentry = file->f_path.dentry; - struct inode *inode = dentry->d_inode; + struct inode *inode = file_inode(file); fmode_t mode = file->f_mode; int must_measure; bool send_tomtou = false, send_writers = false; diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index 5c6f2cd2d095..db1fca990a24 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -1547,6 +1547,18 @@ static inline int path_has_perm(const struct cred *cred, return inode_has_perm(cred, inode, av, &ad, 0); } +/* Same as path_has_perm, but uses the inode from the file struct. */ +static inline int file_path_has_perm(const struct cred *cred, + struct file *file, + u32 av) +{ + struct common_audit_data ad; + + ad.type = LSM_AUDIT_DATA_PATH; + ad.u.path = file->f_path; + return inode_has_perm(cred, file_inode(file), av, &ad, 0); +} + /* Check whether a task can use an open file descriptor to access an inode in a given way. Check access to the descriptor itself, and then use dentry_has_perm to @@ -2141,14 +2153,14 @@ static inline void flush_unauthorized_files(const struct cred *cred, struct tty_file_private *file_priv; /* Revalidate access to controlling tty. - Use path_has_perm on the tty path directly rather - than using file_has_perm, as this particular open - file may belong to another process and we are only - interested in the inode-based check here. */ + Use file_path_has_perm on the tty path directly + rather than using file_has_perm, as this particular + open file may belong to another process and we are + only interested in the inode-based check here. */ file_priv = list_first_entry(&tty->tty_files, struct tty_file_private, list); file = file_priv->file; - if (path_has_perm(cred, &file->f_path, FILE__READ | FILE__WRITE)) + if (file_path_has_perm(cred, file, FILE__READ | FILE__WRITE)) drop_tty = 1; } spin_unlock(&tty_files_lock); @@ -3259,7 +3271,7 @@ static int selinux_file_open(struct file *file, const struct cred *cred) * new inode label or new policy. * This check is not redundant - do not remove. */ - return path_has_perm(cred, &file->f_path, open_file_to_av(file)); + return file_path_has_perm(cred, file, open_file_to_av(file)); } /* task security operations */ diff --git a/sound/core/pcm_native.c b/sound/core/pcm_native.c index f92818155958..a68d4c6d702c 100644 --- a/sound/core/pcm_native.c +++ b/sound/core/pcm_native.c @@ -1589,29 +1589,16 @@ static int snd_pcm_drop(struct snd_pcm_substream *substream) } -/* WARNING: Don't forget to fput back the file */ -static struct file *snd_pcm_file_fd(int fd, int *fput_needed) +static bool is_pcm_file(struct file *file) { - struct file *file; - struct inode *inode; + struct inode *inode = file_inode(file); unsigned int minor; - file = fget_light(fd, fput_needed); - if (!file) - return NULL; - inode = file_inode(file); - if (!S_ISCHR(inode->i_mode) || - imajor(inode) != snd_major) { - fput_light(file, *fput_needed); - return NULL; - } + if (!S_ISCHR(inode->i_mode) || imajor(inode) != snd_major) + return false; minor = iminor(inode); - if (!snd_lookup_minor_data(minor, SNDRV_DEVICE_TYPE_PCM_PLAYBACK) && - !snd_lookup_minor_data(minor, SNDRV_DEVICE_TYPE_PCM_CAPTURE)) { - fput_light(file, *fput_needed); - return NULL; - } - return file; + return snd_lookup_minor_data(minor, SNDRV_DEVICE_TYPE_PCM_PLAYBACK) || + snd_lookup_minor_data(minor, SNDRV_DEVICE_TYPE_PCM_CAPTURE); } /* @@ -1620,16 +1607,18 @@ static struct file *snd_pcm_file_fd(int fd, int *fput_needed) static int snd_pcm_link(struct snd_pcm_substream *substream, int fd) { int res = 0; - struct file *file; struct snd_pcm_file *pcm_file; struct snd_pcm_substream *substream1; struct snd_pcm_group *group; - int fput_needed; + struct fd f = fdget(fd); - file = snd_pcm_file_fd(fd, &fput_needed); - if (!file) + if (!f.file) return -EBADFD; - pcm_file = file->private_data; + if (!is_pcm_file(f.file)) { + res = -EBADFD; + goto _badf; + } + pcm_file = f.file->private_data; substream1 = pcm_file->substream; group = kmalloc(sizeof(*group), GFP_KERNEL); if (!group) { @@ -1663,8 +1652,9 @@ static int snd_pcm_link(struct snd_pcm_substream *substream, int fd) up_write(&snd_pcm_link_rwsem); _nolock: snd_card_unref(substream1->pcm->card); - fput_light(file, fput_needed); kfree(group); + _badf: + fdput(f); return res; } |