diff --git a/fs/inode.c b/fs/inode.c index 5aab80dc008cf681e64f47a32759055bc013b13f..73920d555c8890b2bd0fd208ef2b469af142e358 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -143,6 +143,7 @@ int inode_init_always(struct super_block *sb, struct inode *inode) inode->i_op = &empty_iops; inode->i_fop = &empty_fops; inode->i_nlink = 1; + inode->i_opflags = 0; inode->i_uid = 0; inode->i_gid = 0; atomic_set(&inode->i_writecount, 0); diff --git a/fs/namei.c b/fs/namei.c index 3d607bd80e09f683fac132d6387f152c1df11f4c..4a98bf154d88421fc3148d96133843a4da01a828 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -308,6 +308,26 @@ int generic_permission(struct inode *inode, int mask) return -EACCES; } +/* + * We _really_ want to just do "generic_permission()" without + * even looking at the inode->i_op values. So we keep a cache + * flag in inode->i_opflags, that says "this has not special + * permission function, use the fast case". + */ +static inline int do_inode_permission(struct inode *inode, int mask) +{ + if (unlikely(!(inode->i_opflags & IOP_FASTPERM))) { + if (likely(inode->i_op->permission)) + return inode->i_op->permission(inode, mask); + + /* This gets set once for the inode lifetime */ + spin_lock(&inode->i_lock); + inode->i_opflags |= IOP_FASTPERM; + spin_unlock(&inode->i_lock); + } + return generic_permission(inode, mask); +} + /** * inode_permission - check for access rights to a given inode * @inode: inode to check permission on @@ -322,7 +342,7 @@ int inode_permission(struct inode *inode, int mask) { int retval; - if (mask & MAY_WRITE) { + if (unlikely(mask & MAY_WRITE)) { umode_t mode = inode->i_mode; /* @@ -339,11 +359,7 @@ int inode_permission(struct inode *inode, int mask) return -EACCES; } - if (inode->i_op->permission) - retval = inode->i_op->permission(inode, mask); - else - retval = generic_permission(inode, mask); - + retval = do_inode_permission(inode, mask); if (retval) return retval; @@ -1245,6 +1261,26 @@ static void terminate_walk(struct nameidata *nd) } } +/* + * Do we need to follow links? We _really_ want to be able + * to do this check without having to look at inode->i_op, + * so we keep a cache of "no, this doesn't need follow_link" + * for the common case. + */ +static inline int do_follow_link(struct inode *inode, int follow) +{ + if (unlikely(!(inode->i_opflags & IOP_NOFOLLOW))) { + if (likely(inode->i_op->follow_link)) + return follow; + + /* This gets set once for the inode lifetime */ + spin_lock(&inode->i_lock); + inode->i_opflags |= IOP_NOFOLLOW; + spin_unlock(&inode->i_lock); + } + return 0; +} + static inline int walk_component(struct nameidata *nd, struct path *path, struct qstr *name, int type, int follow) { @@ -1267,7 +1303,7 @@ static inline int walk_component(struct nameidata *nd, struct path *path, terminate_walk(nd); return -ENOENT; } - if (unlikely(inode->i_op->follow_link) && follow) { + if (do_follow_link(inode, follow)) { if (nd->flags & LOOKUP_RCU) { if (unlikely(unlazy_walk(nd, path->dentry))) { terminate_walk(nd); @@ -1319,6 +1355,26 @@ static inline int nested_symlink(struct path *path, struct nameidata *nd) return res; } +/* + * We really don't want to look at inode->i_op->lookup + * when we don't have to. So we keep a cache bit in + * the inode ->i_opflags field that says "yes, we can + * do lookup on this inode". + */ +static inline int can_lookup(struct inode *inode) +{ + if (likely(inode->i_opflags & IOP_LOOKUP)) + return 1; + if (likely(!inode->i_op->lookup)) + return 0; + + /* We do this once for the lifetime of the inode */ + spin_lock(&inode->i_lock); + inode->i_opflags |= IOP_LOOKUP; + spin_unlock(&inode->i_lock); + return 1; +} + /* * Name resolution. * This is the basic name resolution function, turning a pathname into @@ -1398,10 +1454,10 @@ static int link_path_walk(const char *name, struct nameidata *nd) if (err) return err; } + if (can_lookup(nd->inode)) + continue; err = -ENOTDIR; - if (!nd->inode->i_op->lookup) - break; - continue; + break; /* here ends the main loop */ last_component: diff --git a/fs/stat.c b/fs/stat.c index 961039121cb8cbde185bf1b8399c6e4ccc71cdf5..ba5316ffac617679a4beae086c1048886d4ae209 100644 --- a/fs/stat.c +++ b/fs/stat.c @@ -27,12 +27,12 @@ void generic_fillattr(struct inode *inode, struct kstat *stat) stat->uid = inode->i_uid; stat->gid = inode->i_gid; stat->rdev = inode->i_rdev; + stat->size = i_size_read(inode); stat->atime = inode->i_atime; stat->mtime = inode->i_mtime; stat->ctime = inode->i_ctime; - stat->size = i_size_read(inode); - stat->blocks = inode->i_blocks; stat->blksize = (1 << inode->i_blkbits); + stat->blocks = inode->i_blocks; } EXPORT_SYMBOL(generic_fillattr); diff --git a/include/linux/fs.h b/include/linux/fs.h index 786b3b1113cfb1acd5bf59848d04fe2e460e2de0..178cdb4f1d4afe81a66e631872de4b587e95d1c4 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -738,22 +738,54 @@ static inline int mapping_writably_mapped(struct address_space *mapping) struct posix_acl; #define ACL_NOT_CACHED ((void *)(-1)) +#define IOP_FASTPERM 0x0001 +#define IOP_LOOKUP 0x0002 +#define IOP_NOFOLLOW 0x0004 + +/* + * Keep mostly read-only and often accessed (especially for + * the RCU path lookup and 'stat' data) fields at the beginning + * of the 'struct inode' + */ struct inode { - /* RCU path lookup touches following: */ umode_t i_mode; + unsigned short i_opflags; uid_t i_uid; gid_t i_gid; + unsigned int i_flags; + +#ifdef CONFIG_FS_POSIX_ACL + struct posix_acl *i_acl; + struct posix_acl *i_default_acl; +#endif + const struct inode_operations *i_op; struct super_block *i_sb; + struct address_space *i_mapping; - spinlock_t i_lock; /* i_blocks, i_bytes, maybe i_size */ - unsigned int i_flags; - unsigned long i_state; #ifdef CONFIG_SECURITY void *i_security; #endif - struct mutex i_mutex; + /* Stat data, not accessed from path walking */ + unsigned long i_ino; + unsigned int i_nlink; + dev_t i_rdev; + loff_t i_size; + struct timespec i_atime; + struct timespec i_mtime; + struct timespec i_ctime; + unsigned int i_blkbits; + blkcnt_t i_blocks; + +#ifdef __NEED_I_SIZE_ORDERED + seqcount_t i_size_seqcount; +#endif + + /* Misc */ + unsigned long i_state; + spinlock_t i_lock; /* i_blocks, i_bytes, maybe i_size */ + struct mutex i_mutex; unsigned long dirtied_when; /* jiffies of first dirtying */ @@ -765,25 +797,12 @@ struct inode { struct list_head i_dentry; struct rcu_head i_rcu; }; - unsigned long i_ino; atomic_t i_count; - unsigned int i_nlink; - dev_t i_rdev; - unsigned int i_blkbits; u64 i_version; - loff_t i_size; -#ifdef __NEED_I_SIZE_ORDERED - seqcount_t i_size_seqcount; -#endif - struct timespec i_atime; - struct timespec i_mtime; - struct timespec i_ctime; - blkcnt_t i_blocks; unsigned short i_bytes; atomic_t i_dio_count; const struct file_operations *i_fop; /* former ->i_op->default_file_ops */ struct file_lock *i_flock; - struct address_space *i_mapping; struct address_space i_data; #ifdef CONFIG_QUOTA struct dquot *i_dquot[MAXQUOTAS]; @@ -806,10 +825,6 @@ struct inode { atomic_t i_readcount; /* struct files open RO */ #endif atomic_t i_writecount; -#ifdef CONFIG_FS_POSIX_ACL - struct posix_acl *i_acl; - struct posix_acl *i_default_acl; -#endif void *i_private; /* fs or device private pointer */ };