diff options
Diffstat (limited to 'fs')
297 files changed, 6259 insertions, 3115 deletions
diff --git a/fs/9p/acl.c b/fs/9p/acl.c index 12d602351db..6e58c4ca1e6 100644 --- a/fs/9p/acl.c +++ b/fs/9p/acl.c @@ -91,11 +91,14 @@ static struct posix_acl *v9fs_get_cached_acl(struct inode *inode, int type) return acl; } -int v9fs_check_acl(struct inode *inode, int mask) +int v9fs_check_acl(struct inode *inode, int mask, unsigned int flags) { struct posix_acl *acl; struct v9fs_session_info *v9ses; + if (flags & IPERM_FLAG_RCU) + return -ECHILD; + v9ses = v9fs_inode2v9ses(inode); if ((v9ses->flags & V9FS_ACCESS_MASK) != V9FS_ACCESS_CLIENT) { /* diff --git a/fs/9p/acl.h b/fs/9p/acl.h index 59e18c2e8c7..7ef3ac9f6d9 100644 --- a/fs/9p/acl.h +++ b/fs/9p/acl.h @@ -16,7 +16,7 @@ #ifdef CONFIG_9P_FS_POSIX_ACL extern int v9fs_get_acl(struct inode *, struct p9_fid *); -extern int v9fs_check_acl(struct inode *inode, int mask); +extern int v9fs_check_acl(struct inode *inode, int mask, unsigned int flags); extern int v9fs_acl_chmod(struct dentry *); extern int v9fs_set_create_acl(struct dentry *, struct posix_acl *, struct posix_acl *); diff --git a/fs/9p/vfs_dentry.c b/fs/9p/vfs_dentry.c index cbf4e50f393..466d2a4fc5c 100644 --- a/fs/9p/vfs_dentry.c +++ b/fs/9p/vfs_dentry.c @@ -51,7 +51,7 @@ * */ -static int v9fs_dentry_delete(struct dentry *dentry) +static int v9fs_dentry_delete(const struct dentry *dentry) { P9_DPRINTK(P9_DEBUG_VFS, " dentry: %s (%p)\n", dentry->d_name.name, dentry); @@ -68,7 +68,7 @@ static int v9fs_dentry_delete(struct dentry *dentry) * */ -static int v9fs_cached_dentry_delete(struct dentry *dentry) +static int v9fs_cached_dentry_delete(const struct dentry *dentry) { struct inode *inode = dentry->d_inode; P9_DPRINTK(P9_DEBUG_VFS, " dentry: %s (%p)\n", dentry->d_name.name, diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index 34bf71b5654..59782981b22 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -237,10 +237,17 @@ struct inode *v9fs_alloc_inode(struct super_block *sb) * */ -void v9fs_destroy_inode(struct inode *inode) +static void v9fs_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(vcookie_cache, v9fs_inode2cookie(inode)); } + +void v9fs_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, v9fs_i_callback); +} #endif /** @@ -270,11 +277,11 @@ static struct dentry *v9fs_dentry_from_dir_inode(struct inode *inode) { struct dentry *dentry; - spin_lock(&dcache_lock); + spin_lock(&inode->i_lock); /* Directory should have only one entry. */ BUG_ON(S_ISDIR(inode->i_mode) && !list_is_singular(&inode->i_dentry)); dentry = list_entry(inode->i_dentry.next, struct dentry, d_alias); - spin_unlock(&dcache_lock); + spin_unlock(&inode->i_lock); return dentry; } @@ -628,9 +635,9 @@ v9fs_create(struct v9fs_session_info *v9ses, struct inode *dir, } if (v9ses->cache) - dentry->d_op = &v9fs_cached_dentry_operations; + d_set_d_op(dentry, &v9fs_cached_dentry_operations); else - dentry->d_op = &v9fs_dentry_operations; + d_set_d_op(dentry, &v9fs_dentry_operations); d_instantiate(dentry, inode); err = v9fs_fid_add(dentry, fid); @@ -742,7 +749,7 @@ v9fs_vfs_create_dotl(struct inode *dir, struct dentry *dentry, int omode, err); goto error; } - dentry->d_op = &v9fs_cached_dentry_operations; + d_set_d_op(dentry, &v9fs_cached_dentry_operations); d_instantiate(dentry, inode); err = v9fs_fid_add(dentry, fid); if (err < 0) @@ -760,7 +767,7 @@ v9fs_vfs_create_dotl(struct inode *dir, struct dentry *dentry, int omode, err = PTR_ERR(inode); goto error; } - dentry->d_op = &v9fs_dentry_operations; + d_set_d_op(dentry, &v9fs_dentry_operations); d_instantiate(dentry, inode); } /* Now set the ACL based on the default value */ @@ -949,7 +956,7 @@ static int v9fs_vfs_mkdir_dotl(struct inode *dir, err); goto error; } - dentry->d_op = &v9fs_cached_dentry_operations; + d_set_d_op(dentry, &v9fs_cached_dentry_operations); d_instantiate(dentry, inode); err = v9fs_fid_add(dentry, fid); if (err < 0) @@ -966,7 +973,7 @@ static int v9fs_vfs_mkdir_dotl(struct inode *dir, err = PTR_ERR(inode); goto error; } - dentry->d_op = &v9fs_dentry_operations; + d_set_d_op(dentry, &v9fs_dentry_operations); d_instantiate(dentry, inode); } /* Now set the ACL based on the default value */ @@ -1034,9 +1041,9 @@ static struct dentry *v9fs_vfs_lookup(struct inode *dir, struct dentry *dentry, inst_out: if (v9ses->cache) - dentry->d_op = &v9fs_cached_dentry_operations; + d_set_d_op(dentry, &v9fs_cached_dentry_operations); else - dentry->d_op = &v9fs_dentry_operations; + d_set_d_op(dentry, &v9fs_dentry_operations); d_add(dentry, inode); return NULL; @@ -1702,7 +1709,7 @@ v9fs_vfs_symlink_dotl(struct inode *dir, struct dentry *dentry, err); goto error; } - dentry->d_op = &v9fs_cached_dentry_operations; + d_set_d_op(dentry, &v9fs_cached_dentry_operations); d_instantiate(dentry, inode); err = v9fs_fid_add(dentry, fid); if (err < 0) @@ -1715,7 +1722,7 @@ v9fs_vfs_symlink_dotl(struct inode *dir, struct dentry *dentry, err = PTR_ERR(inode); goto error; } - dentry->d_op = &v9fs_dentry_operations; + d_set_d_op(dentry, &v9fs_dentry_operations); d_instantiate(dentry, inode); } @@ -1849,7 +1856,7 @@ v9fs_vfs_link_dotl(struct dentry *old_dentry, struct inode *dir, ihold(old_dentry->d_inode); } - dentry->d_op = old_dentry->d_op; + d_set_d_op(dentry, old_dentry->d_op); d_instantiate(dentry, old_dentry->d_inode); return err; @@ -1973,7 +1980,7 @@ v9fs_vfs_mknod_dotl(struct inode *dir, struct dentry *dentry, int omode, err); goto error; } - dentry->d_op = &v9fs_cached_dentry_operations; + d_set_d_op(dentry, &v9fs_cached_dentry_operations); d_instantiate(dentry, inode); err = v9fs_fid_add(dentry, fid); if (err < 0) @@ -1989,7 +1996,7 @@ v9fs_vfs_mknod_dotl(struct inode *dir, struct dentry *dentry, int omode, err = PTR_ERR(inode); goto error; } - dentry->d_op = &v9fs_dentry_operations; + d_set_d_op(dentry, &v9fs_dentry_operations); d_instantiate(dentry, inode); } /* Now set the ACL based on the default value */ diff --git a/fs/adfs/dir.c b/fs/adfs/dir.c index f4287e4de74..bf7693c384f 100644 --- a/fs/adfs/dir.c +++ b/fs/adfs/dir.c @@ -201,7 +201,8 @@ const struct file_operations adfs_dir_operations = { }; static int -adfs_hash(struct dentry *parent, struct qstr *qstr) +adfs_hash(const struct dentry *parent, const struct inode *inode, + struct qstr *qstr) { const unsigned int name_len = ADFS_SB(parent->d_sb)->s_namelen; const unsigned char *name; @@ -237,17 +238,19 @@ adfs_hash(struct dentry *parent, struct qstr *qstr) * requirements of the underlying filesystem. */ static int -adfs_compare(struct dentry *parent, struct qstr *entry, struct qstr *name) +adfs_compare(const struct dentry *parent, const struct inode *pinode, + const struct dentry *dentry, const struct inode *inode, + unsigned int len, const char *str, const struct qstr *name) { int i; - if (entry->len != name->len) + if (len != name->len) return 1; for (i = 0; i < name->len; i++) { char a, b; - a = entry->name[i]; + a = str[i]; b = name->name[i]; if (a >= 'A' && a <= 'Z') @@ -273,7 +276,7 @@ adfs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) struct object_info obj; int error; - dentry->d_op = &adfs_dentry_operations; + d_set_d_op(dentry, &adfs_dentry_operations); lock_kernel(); error = adfs_dir_lookup_byname(dir, &dentry->d_name, &obj); if (error == 0) { diff --git a/fs/adfs/super.c b/fs/adfs/super.c index 959dbff2d42..a4041b52fbc 100644 --- a/fs/adfs/super.c +++ b/fs/adfs/super.c @@ -240,11 +240,18 @@ static struct inode *adfs_alloc_inode(struct super_block *sb) return &ei->vfs_inode; } -static void adfs_destroy_inode(struct inode *inode) +static void adfs_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(adfs_inode_cachep, ADFS_I(inode)); } +static void adfs_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, adfs_i_callback); +} + static void init_once(void *foo) { struct adfs_inode_info *ei = (struct adfs_inode_info *) foo; @@ -477,7 +484,7 @@ static int adfs_fill_super(struct super_block *sb, void *data, int silent) adfs_error(sb, "get root inode failed\n"); goto error; } else - sb->s_root->d_op = &adfs_dentry_operations; + d_set_d_op(sb->s_root, &adfs_dentry_operations); unlock_kernel(); return 0; diff --git a/fs/affs/amigaffs.c b/fs/affs/amigaffs.c index 7d0f0a30f7a..3a4557e8325 100644 --- a/fs/affs/amigaffs.c +++ b/fs/affs/amigaffs.c @@ -128,7 +128,7 @@ affs_fix_dcache(struct dentry *dentry, u32 entry_ino) void *data = dentry->d_fsdata; struct list_head *head, *next; - spin_lock(&dcache_lock); + spin_lock(&inode->i_lock); head = &inode->i_dentry; next = head->next; while (next != head) { @@ -139,7 +139,7 @@ affs_fix_dcache(struct dentry *dentry, u32 entry_ino) } next = next->next; } - spin_unlock(&dcache_lock); + spin_unlock(&inode->i_lock); } diff --git a/fs/affs/namei.c b/fs/affs/namei.c index 914d1c0bc07..944a4042fb6 100644 --- a/fs/affs/namei.c +++ b/fs/affs/namei.c @@ -13,11 +13,19 @@ typedef int (*toupper_t)(int); static int affs_toupper(int ch); -static int affs_hash_dentry(struct dentry *, struct qstr *); -static int affs_compare_dentry(struct dentry *, struct qstr *, struct qstr *); +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, + unsigned int len, const char *str, const struct qstr *name); static int affs_intl_toupper(int ch); -static int affs_intl_hash_dentry(struct dentry *, struct qstr *); -static int affs_intl_compare_dentry(struct dentry *, struct qstr *, struct qstr *); +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, + unsigned int len, const char *str, const struct qstr *name); const struct dentry_operations affs_dentry_operations = { .d_hash = affs_hash_dentry, @@ -58,13 +66,13 @@ affs_get_toupper(struct super_block *sb) * Note: the dentry argument is the parent dentry. */ static inline int -__affs_hash_dentry(struct dentry *dentry, struct qstr *qstr, toupper_t toupper) +__affs_hash_dentry(struct qstr *qstr, toupper_t toupper) { const u8 *name = qstr->name; unsigned long hash; int i; - i = affs_check_name(qstr->name,qstr->len); + i = affs_check_name(qstr->name, qstr->len); if (i) return i; @@ -78,39 +86,41 @@ __affs_hash_dentry(struct dentry *dentry, struct qstr *qstr, toupper_t toupper) } static int -affs_hash_dentry(struct dentry *dentry, struct qstr *qstr) +affs_hash_dentry(const struct dentry *dentry, const struct inode *inode, + struct qstr *qstr) { - return __affs_hash_dentry(dentry, qstr, affs_toupper); + return __affs_hash_dentry(qstr, affs_toupper); } static int -affs_intl_hash_dentry(struct dentry *dentry, struct qstr *qstr) +affs_intl_hash_dentry(const struct dentry *dentry, const struct inode *inode, + struct qstr *qstr) { - return __affs_hash_dentry(dentry, qstr, affs_intl_toupper); + return __affs_hash_dentry(qstr, affs_intl_toupper); } -static inline int -__affs_compare_dentry(struct dentry *dentry, struct qstr *a, struct qstr *b, toupper_t toupper) +static inline int __affs_compare_dentry(unsigned int len, + const char *str, const struct qstr *name, toupper_t toupper) { - const u8 *aname = a->name; - const u8 *bname = b->name; - int len; + const u8 *aname = str; + const u8 *bname = name->name; - /* 'a' is the qstr of an already existing dentry, so the name - * must be valid. 'b' must be validated first. + /* + * 'str' is the name of an already existing dentry, so the name + * must be valid. 'name' must be validated first. */ - if (affs_check_name(b->name,b->len)) + if (affs_check_name(name->name, name->len)) return 1; - /* If the names are longer than the allowed 30 chars, + /* + * If the names are longer than the allowed 30 chars, * the excess is ignored, so their length may differ. */ - len = a->len; if (len >= 30) { - if (b->len < 30) + if (name->len < 30) return 1; len = 30; - } else if (len != b->len) + } else if (len != name->len) return 1; for (; len > 0; len--) @@ -121,14 +131,18 @@ __affs_compare_dentry(struct dentry *dentry, struct qstr *a, struct qstr *b, tou } static int -affs_compare_dentry(struct dentry *dentry, struct qstr *a, struct qstr *b) +affs_compare_dentry(const struct dentry *parent, const struct inode *pinode, + const struct dentry *dentry, const struct inode *inode, + unsigned int len, const char *str, const struct qstr *name) { - return __affs_compare_dentry(dentry, a, b, affs_toupper); + return __affs_compare_dentry(len, str, name, affs_toupper); } static int -affs_intl_compare_dentry(struct dentry *dentry, struct qstr *a, struct qstr *b) +affs_intl_compare_dentry(const struct dentry *parent,const struct inode *pinode, + const struct dentry *dentry, const struct inode *inode, + unsigned int len, const char *str, const struct qstr *name) { - return __affs_compare_dentry(dentry, a, b, affs_intl_toupper); + return __affs_compare_dentry(len, str, name, affs_intl_toupper); } /* @@ -226,7 +240,7 @@ affs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) if (IS_ERR(inode)) return ERR_CAST(inode); } - dentry->d_op = AFFS_SB(sb)->s_flags & SF_INTL ? &affs_intl_dentry_operations : &affs_dentry_operations; + d_set_d_op(dentry, AFFS_SB(sb)->s_flags & SF_INTL ? &affs_intl_dentry_operations : &affs_dentry_operations); d_add(dentry, inode); return NULL; } diff --git a/fs/affs/super.c b/fs/affs/super.c index 0cf7f4384cb..d39081bbe7c 100644 --- a/fs/affs/super.c +++ b/fs/affs/super.c @@ -95,11 +95,18 @@ static struct inode *affs_alloc_inode(struct super_block *sb) return &i->vfs_inode; } -static void affs_destroy_inode(struct inode *inode) +static void affs_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(affs_inode_cachep, AFFS_I(inode)); } +static void affs_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, affs_i_callback); +} + static void init_once(void *foo) { struct affs_inode_info *ei = (struct affs_inode_info *) foo; @@ -475,7 +482,7 @@ got_root: printk(KERN_ERR "AFFS: Get root inode failed\n"); goto out_error; } - sb->s_root->d_op = &affs_dentry_operations; + d_set_d_op(sb->s_root, &affs_dentry_operations); pr_debug("AFFS: s_flags=%lX\n",sb->s_flags); return 0; diff --git a/fs/afs/dir.c b/fs/afs/dir.c index 5439e1bc9a8..34a3263d60a 100644 --- a/fs/afs/dir.c +++ b/fs/afs/dir.c @@ -13,6 +13,7 @@ #include <linux/module.h> #include <linux/init.h> #include <linux/fs.h> +#include <linux/namei.h> #include <linux/pagemap.h> #include <linux/ctype.h> #include <linux/sched.h> @@ -23,7 +24,7 @@ static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, static int afs_dir_open(struct inode *inode, struct file *file); static int afs_readdir(struct file *file, void *dirent, filldir_t filldir); static int afs_d_revalidate(struct dentry *dentry, struct nameidata *nd); -static int afs_d_delete(struct dentry *dentry); +static int afs_d_delete(const struct dentry *dentry); static void afs_d_release(struct dentry *dentry); static int afs_lookup_filldir(void *_cookie, const char *name, int nlen, loff_t fpos, u64 ino, unsigned dtype); @@ -581,7 +582,7 @@ static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, } success: - dentry->d_op = &afs_fs_dentry_operations; + d_set_d_op(dentry, &afs_fs_dentry_operations); d_add(dentry, inode); _leave(" = 0 { vn=%u u=%u } -> { ino=%lu v=%llu }", @@ -607,6 +608,9 @@ static int afs_d_revalidate(struct dentry *dentry, struct nameidata *nd) void *dir_version; int ret; + if (nd->flags & LOOKUP_RCU) + return -ECHILD; + vnode = AFS_FS_I(dentry->d_inode); if (dentry->d_inode) @@ -730,7 +734,7 @@ out_bad: * - called from dput() when d_count is going to 0. * - return 1 to request dentry be unhashed, 0 otherwise */ -static int afs_d_delete(struct dentry *dentry) +static int afs_d_delete(const struct dentry *dentry) { _enter("%s", dentry->d_name.name); diff --git a/fs/afs/internal.h b/fs/afs/internal.h index cca8eef736f..6d4bc1c8ff6 100644 --- a/fs/afs/internal.h +++ b/fs/afs/internal.h @@ -624,7 +624,7 @@ extern void afs_clear_permits(struct afs_vnode *); extern void afs_cache_permit(struct afs_vnode *, struct key *, long); extern void afs_zap_permits(struct rcu_head *); extern struct key *afs_request_key(struct afs_cell *); -extern int afs_permission(struct inode *, int); +extern int afs_permission(struct inode *, int, unsigned int); /* * server.c diff --git a/fs/afs/security.c b/fs/afs/security.c index bb4ed144d0e..f44b9d35537 100644 --- a/fs/afs/security.c +++ b/fs/afs/security.c @@ -285,13 +285,16 @@ static int afs_check_permit(struct afs_vnode *vnode, struct key *key, * - AFS ACLs are attached to directories only, and a file is controlled by its * parent directory's ACL */ -int afs_permission(struct inode *inode, int mask) +int afs_permission(struct inode *inode, int mask, unsigned int flags) { struct afs_vnode *vnode = AFS_FS_I(inode); afs_access_t uninitialized_var(access); struct key *key; int ret; + if (flags & IPERM_FLAG_RCU) + return -ECHILD; + _enter("{{%x:%u},%lx},%x,", vnode->fid.vid, vnode->fid.vnode, vnode->flags, mask); @@ -347,7 +350,7 @@ int afs_permission(struct inode *inode, int mask) } key_put(key); - ret = generic_permission(inode, mask, NULL); + ret = generic_permission(inode, mask, flags, NULL); _leave(" = %d", ret); return ret; diff --git a/fs/afs/super.c b/fs/afs/super.c index 27201cffece..f901a9d7c11 100644 --- a/fs/afs/super.c +++ b/fs/afs/super.c @@ -498,6 +498,14 @@ static struct inode *afs_alloc_inode(struct super_block *sb) return &vnode->vfs_inode; } +static void afs_i_callback(struct rcu_head *head) +{ + struct inode *inode = container_of(head, struct inode, i_rcu); + struct afs_vnode *vnode = AFS_FS_I(inode); + INIT_LIST_HEAD(&inode->i_dentry); + kmem_cache_free(afs_inode_cachep, vnode); +} + /* * destroy an AFS inode struct */ @@ -511,7 +519,7 @@ static void afs_destroy_inode(struct inode *inode) ASSERTCMP(vnode->server, ==, NULL); - kmem_cache_free(afs_inode_cachep, vnode); + call_rcu(&inode->i_rcu, afs_i_callback); atomic_dec(&afs_count_active_inodes); } diff --git a/fs/anon_inodes.c b/fs/anon_inodes.c index 57ce55b2564..5fd38112a6c 100644 --- a/fs/anon_inodes.c +++ b/fs/anon_inodes.c @@ -102,7 +102,7 @@ struct file *anon_inode_getfile(const char *name, this.name = name; this.len = strlen(name); this.hash = 0; - path.dentry = d_alloc(anon_inode_mnt->mnt_sb->s_root, &this); + path.dentry = d_alloc_pseudo(anon_inode_mnt->mnt_sb, &this); if (!path.dentry) goto err_module; @@ -113,7 +113,7 @@ struct file *anon_inode_getfile(const char *name, */ ihold(anon_inode_inode); - path.dentry->d_op = &anon_inodefs_dentry_operations; + d_set_d_op(path.dentry, &anon_inodefs_dentry_operations); d_instantiate(path.dentry, anon_inode_inode); error = -ENFILE; @@ -232,7 +232,7 @@ static int __init anon_inode_init(void) return 0; err_mntput: - mntput(anon_inode_mnt); + mntput_long(anon_inode_mnt); err_unregister_filesystem: unregister_filesystem(&anon_inode_fs_type); err_exit: diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h index 3d283abf67d..0fffe1c24ce 100644 --- a/fs/autofs4/autofs_i.h +++ b/fs/autofs4/autofs_i.h @@ -16,6 +16,7 @@ #include <linux/auto_fs4.h> #include <linux/auto_dev-ioctl.h> #include <linux/mutex.h> +#include <linux/spinlock.h> #include <linux/list.h> /* This is the range of ioctl() numbers we claim as ours */ @@ -60,6 +61,8 @@ do { \ current->pid, __func__, ##args); \ } while (0) +extern spinlock_t autofs4_lock; + /* Unified info structure. This is pointed to by both the dentry and inode structures. Each file in the filesystem has an instance of this structure. It holds a reference to the dentry, so dentries are never @@ -254,17 +257,15 @@ static inline int simple_positive(struct dentry *dentry) return dentry->d_inode && !d_unhashed(dentry); } -static inline int __simple_empty(struct dentry *dentry) +static inline void __autofs4_add_expiring(struct dentry *dentry) { - struct dentry *child; - int ret = 0; - - list_for_each_entry(child, &dentry->d_subdirs, d_u.d_child) - if (simple_positive(child)) - goto out; - ret = 1; -out: - return ret; + struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb); + struct autofs_info *ino = autofs4_dentry_ino(dentry); + if (ino) { + if (list_empty(&ino->expiring)) + list_add(&ino->expiring, &sbi->expiring_list); + } + return; } static inline void autofs4_add_expiring(struct dentry *dentry) diff --git a/fs/autofs4/expire.c b/fs/autofs4/expire.c index a796c9417fb..cc1d0136590 100644 --- a/fs/autofs4/expire.c +++ b/fs/autofs4/expire.c @@ -91,24 +91,64 @@ done: } /* - * Calculate next entry in top down tree traversal. - * From next_mnt in namespace.c - elegant. + * Calculate and dget next entry in top down tree traversal. */ -static struct dentry *next_dentry(struct dentry *p, struct dentry *root) +static struct dentry *get_next_positive_dentry(struct dentry *prev, + struct dentry *root) { - struct list_head *next = p->d_subdirs.next; + struct list_head *next; + struct dentry *p, *ret; + + if (prev == NULL) + return dget(prev); + spin_lock(&autofs4_lock); +relock: + p = prev; + spin_lock(&p->d_lock); +again: + next = p->d_subdirs.next; if (next == &p->d_subdirs) { while (1) { - if (p == root) + struct dentry *parent; + + if (p == root) { + spin_unlock(&p->d_lock); + spin_unlock(&autofs4_lock); + dput(prev); return NULL; + } + + parent = p->d_parent; + if (!spin_trylock(&parent->d_lock)) { + spin_unlock(&p->d_lock); + cpu_relax(); + goto relock; + } + spin_unlock(&p->d_lock); next = p->d_u.d_child.next; - if (next != &p->d_parent->d_subdirs) + p = parent; + if (next != &parent->d_subdirs) break; - p = p->d_parent; } } - return list_entry(next, struct dentry, d_u.d_child); + ret = list_entry(next, struct dentry, d_u.d_child); + + spin_lock_nested(&ret->d_lock, DENTRY_D_LOCK_NESTED); + /* Negative dentry - try next */ + if (!simple_positive(ret)) { + spin_unlock(&ret->d_lock); + p = ret; + goto again; + } + dget_dlock(ret); + spin_unlock(&ret->d_lock); + spin_unlock(&p->d_lock); + spin_unlock(&autofs4_lock); + + dput(prev); + + return ret; } /* @@ -158,18 +198,11 @@ static int autofs4_tree_busy(struct vfsmount *mnt, if (!simple_positive(top)) return 1; - spin_lock(&dcache_lock); - for (p = top; p; p = next_dentry(p, top)) { - /* Negative dentry - give up */ - if (!simple_positive(p)) - continue; - + p = NULL; + while ((p = get_next_positive_dentry(p, top))) { DPRINTK("dentry %p %.*s", p, (int) p->d_name.len, p->d_name.name); - p = dget(p); - spin_unlock(&dcache_lock); - /* * Is someone visiting anywhere in the subtree ? * If there's no mount we need to check the usage @@ -198,16 +231,13 @@ static int autofs4_tree_busy(struct vfsmount *mnt, else ino_count++; - if (atomic_read(&p->d_count) > ino_count) { + if (p->d_count > ino_count) { top_ino->last_used = jiffies; dput(p); return 1; } } - dput(p); - spin_lock(&dcache_lock); } - spin_unlock(&dcache_lock); /* Timeout of a tree mount is ultimately determined by its top dentry */ if (!autofs4_can_expire(top, timeout, do_now)) @@ -226,32 +256,21 @@ static struct dentry *autofs4_check_leaves(struct vfsmount *mnt, DPRINTK("parent %p %.*s", parent, (int)parent->d_name.len, parent->d_name.name); - spin_lock(&dcache_lock); - for (p = parent; p; p = next_dentry(p, parent)) { - /* Negative dentry - give up */ - if (!simple_positive(p)) - continue; - + p = NULL; + while ((p = get_next_positive_dentry(p, parent))) { DPRINTK("dentry %p %.*s", p, (int) p->d_name.len, p->d_name.name); - p = dget(p); - spin_unlock(&dcache_lock); - if (d_mountpoint(p)) { /* Can we umount this guy */ if (autofs4_mount_busy(mnt, p)) - goto cont; + continue; /* Can we expire this guy */ if (autofs4_can_expire(p, timeout, do_now)) return p; } -cont: - dput(p); - spin_lock(&dcache_lock); } - spin_unlock(&dcache_lock); return NULL; } @@ -276,7 +295,9 @@ struct dentry *autofs4_expire_direct(struct super_block *sb, struct autofs_info *ino = autofs4_dentry_ino(root); if (d_mountpoint(root)) { ino->flags |= AUTOFS_INF_MOUNTPOINT; - root->d_mounted--; + spin_lock(&root->d_lock); + root->d_flags &= ~DCACHE_MOUNTED; + spin_unlock(&root->d_lock); } ino->flags |= AUTOFS_INF_EXPIRING; init_completion(&ino->expire_complete); @@ -302,8 +323,8 @@ struct dentry *autofs4_expire_indirect(struct super_block *sb, { unsigned long timeout; struct dentry *root = sb->s_root; + struct dentry *dentry; struct dentry *expired = NULL; - struct list_head *next; int do_now = how & AUTOFS_EXP_IMMEDIATE; int exp_leaves = how & AUTOFS_EXP_LEAVES; struct autofs_info *ino; @@ -315,23 +336,8 @@ struct dentry *autofs4_expire_indirect(struct super_block *sb, now = jiffies; timeout = sbi->exp_timeout; - spin_lock(&dcache_lock); - next = root->d_subdirs.next; - - /* On exit from the loop expire is set to a dgot dentry - * to expire or it's NULL */ - while ( next != &root->d_subdirs ) { - struct dentry *dentry = list_entry(next, struct dentry, d_u.d_child); - - /* Negative dentry - give up */ - if (!simple_positive(dentry)) { - next = next->next; - continue; - } - - dentry = dget(dentry); - spin_unlock(&dcache_lock); - + dentry = NULL; + while ((dentry = get_next_positive_dentry(dentry, root))) { spin_lock(&sbi->fs_lock); ino = autofs4_dentry_ino(dentry); @@ -347,7 +353,7 @@ struct dentry *autofs4_expire_indirect(struct super_block *sb, /* Path walk currently on this dentry? */ ino_count = atomic_read(&ino->count) + 2; - if (atomic_read(&dentry->d_count) > ino_count) + if (dentry->d_count > ino_count) goto next; /* Can we umount this guy */ @@ -369,7 +375,7 @@ struct dentry *autofs4_expire_indirect(struct super_block *sb, if (!exp_leaves) { /* Path walk currently on this dentry? */ ino_count = atomic_read(&ino->count) + 1; - if (atomic_read(&dentry->d_count) > ino_count) + if (dentry->d_count > ino_count) goto next; if (!autofs4_tree_busy(mnt, dentry, timeout, do_now)) { @@ -383,7 +389,7 @@ struct dentry *autofs4_expire_indirect(struct super_block *sb, } else { /* Path walk currently on this dentry? */ ino_count = atomic_read(&ino->count) + 1; - if (atomic_read(&dentry->d_count) > ino_count) + if (dentry->d_count > ino_count) goto next; expired = autofs4_check_leaves(mnt, dentry, timeout, do_now); @@ -394,11 +400,7 @@ struct dentry *autofs4_expire_indirect(struct super_block *sb, } next: spin_unlock(&sbi->fs_lock); - dput(dentry); - spin_lock(&dcache_lock); - next = next->next; } - spin_unlock(&dcache_lock); return NULL; found: @@ -408,9 +410,13 @@ found: ino->flags |= AUTOFS_INF_EXPIRING; init_completion(&ino->expire_complete); spin_unlock(&sbi->fs_lock); - spin_lock(&dcache_lock); + spin_lock(&autofs4_lock); + spin_lock(&expired->d_parent->d_lock); + spin_lock_nested(&expired->d_lock, DENTRY_D_LOCK_NESTED); list_move(&expired->d_parent->d_subdirs, &expired->d_u.d_child); - spin_unlock(&dcache_lock); + spin_unlock(&expired->d_lock); + spin_unlock(&expired->d_parent->d_lock); + spin_unlock(&autofs4_lock); return expired; } @@ -499,7 +505,14 @@ int autofs4_do_expire_multi(struct super_block *sb, struct vfsmount *mnt, spin_lock(&sbi->fs_lock); if (ino->flags & AUTOFS_INF_MOUNTPOINT) { - sb->s_root->d_mounted++; + spin_lock(&sb->s_root->d_lock); + /* + * If we haven't been expired away, then reset + * mounted status. + */ + if (mnt->mnt_parent != mnt) + sb->s_root->d_flags |= DCACHE_MOUNTED; + spin_unlock(&sb->s_root->d_lock); ino->flags &= ~AUTOFS_INF_MOUNTPOINT; } ino->flags &= ~AUTOFS_INF_EXPIRING; diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c index ac87e49fa70..a7bdb9dcac8 100644 --- a/fs/autofs4/inode.c +++ b/fs/autofs4/inode.c @@ -309,7 +309,7 @@ int autofs4_fill_super(struct super_block *s, void *data, int silent) goto fail_iput; pipe = NULL; - root->d_op = &autofs4_sb_dentry_operations; + d_set_d_op(root, &autofs4_sb_dentry_operations); root->d_fsdata = ino; /* Can this call block? */ diff --git a/fs/autofs4/root.c b/fs/autofs4/root.c index d5c1401f003..651e4ef563b 100644 --- a/fs/autofs4/root.c +++ b/fs/autofs4/root.c @@ -23,6 +23,8 @@ #include "autofs_i.h" +DEFINE_SPINLOCK(autofs4_lock); + static int autofs4_dir_symlink(struct inode *,struct dentry *,const char *); static int autofs4_dir_unlink(struct inode *,struct dentry *); static int autofs4_dir_rmdir(struct inode *,struct dentry *); @@ -142,12 +144,15 @@ static int autofs4_dir_open(struct inode *inode, struct file *file) * autofs file system so just let the libfs routines handle * it. */ - spin_lock(&dcache_lock); + spin_lock(&autofs4_lock); + spin_lock(&dentry->d_lock); if (!d_mountpoint(dentry) && list_empty(&dentry->d_subdirs)) { - spin_unlock(&dcache_lock); + spin_unlock(&dentry->d_lock); + spin_unlock(&autofs4_lock); return -ENOENT; } - spin_unlock(&dcache_lock); + spin_unlock(&dentry->d_lock); + spin_unlock(&autofs4_lock); out: return dcache_dir_open(inode, file); @@ -252,9 +257,11 @@ static void *autofs4_follow_link(struct dentry *dentry, struct nameidata *nd) /* We trigger a mount for almost all flags */ lookup_type = autofs4_need_mount(nd->flags); spin_lock(&sbi->fs_lock); - spin_lock(&dcache_lock); + spin_lock(&autofs4_lock); + spin_lock(&dentry->d_lock); if (!(lookup_type || ino->flags & AUTOFS_INF_PENDING)) { - spin_unlock(&dcache_lock); + spin_unlock(&dentry->d_lock); + spin_unlock(&autofs4_lock); spin_unlock(&sbi->fs_lock); goto follow; } @@ -266,7 +273,8 @@ static void *autofs4_follow_link(struct dentry *dentry, struct nameidata *nd) */ if (ino->flags & AUTOFS_INF_PENDING || (!d_mountpoint(dentry) && list_empty(&dentry->d_subdirs))) { - spin_unlock(&dcache_lock); + spin_unlock(&dentry->d_lock); + spin_unlock(&autofs4_lock); spin_unlock(&sbi->fs_lock); status = try_to_fill_dentry(dentry, nd->flags); @@ -275,7 +283,8 @@ static void *autofs4_follow_link(struct dentry *dentry, struct nameidata *nd) goto follow; } - spin_unlock(&dcache_lock); + spin_unlock(&dentry->d_lock); + spin_unlock(&autofs4_lock); spin_unlock(&sbi->fs_lock); follow: /* @@ -306,12 +315,19 @@ out_error: */ static int autofs4_revalidate(struct dentry *dentry, struct nameidata *nd) { - struct inode *dir = dentry->d_parent->d_inode; - struct autofs_sb_info *sbi = autofs4_sbi(dir->i_sb); - int oz_mode = autofs4_oz_mode(sbi); + struct inode *dir; + struct autofs_sb_info *sbi; + int oz_mode; int flags = nd ? nd->flags : 0; int status = 1; + if (flags & LOOKUP_RCU) + return -ECHILD; + + dir = dentry->d_parent->d_inode; + sbi = autofs4_sbi(dir->i_sb); + oz_mode = autofs4_oz_mode(sbi); + /* Pending dentry */ spin_lock(&sbi->fs_lock); if (autofs4_ispending(dentry)) { @@ -346,12 +362,14 @@ static int autofs4_revalidate(struct dentry *dentry, struct nameidata *nd) return 0; /* Check for a non-mountpoint directory with no contents */ - spin_lock(&dcache_lock); + spin_lock(&autofs4_lock); + spin_lock(&dentry->d_lock); if (S_ISDIR(dentry->d_inode->i_mode) && !d_mountpoint(dentry) && list_empty(&dentry->d_subdirs)) { DPRINTK("dentry=%p %.*s, emptydir", dentry, dentry->d_name.len, dentry->d_name.name); - spin_unlock(&dcache_lock); + spin_unlock(&dentry->d_lock); + spin_unlock(&autofs4_lock); /* The daemon never causes a mount to trigger */ if (oz_mode) @@ -367,7 +385,8 @@ static int autofs4_revalidate(struct dentry *dentry, struct nameidata *nd) return status; } - spin_unlock(&dcache_lock); + spin_unlock(&dentry->d_lock); + spin_unlock(&autofs4_lock); return 1; } @@ -422,7 +441,7 @@ static struct dentry *autofs4_lookup_active(struct dentry *dentry) const unsigned char *str = name->name; struct list_head *p, *head; - spin_lock(&dcache_lock); + spin_lock(&autofs4_lock); spin_lock(&sbi->lookup_lock); head = &sbi->active_list; list_for_each(p, head) { @@ -436,7 +455,7 @@ static struct dentry *autofs4_lookup_active(struct dentry *dentry) spin_lock(&active->d_lock); /* Already gone? */ - if (atomic_read(&active->d_count) == 0) + if (active->d_count == 0) goto next; qstr = &active->d_name; @@ -452,17 +471,17 @@ static struct dentry *autofs4_lookup_active(struct dentry *dentry) goto next; if (d_unhashed(active)) { - dget(active); + dget_dlock(active); spin_unlock(&active->d_lock); spin_unlock(&sbi->lookup_lock); - spin_unlock(&dcache_lock); + spin_unlock(&autofs4_lock); return active; } next: spin_unlock(&active->d_lock); } spin_unlock(&sbi->lookup_lock); - spin_unlock(&dcache_lock); + spin_unlock(&autofs4_lock); return NULL; } @@ -477,7 +496,7 @@ static struct dentry *autofs4_lookup_expiring(struct dentry *dentry) const unsigned char *str = name->name; struct list_head *p, *head; - spin_lock(&dcache_lock); + spin_lock(&autofs4_lock); spin_lock(&sbi->lookup_lock); head = &sbi->expiring_list; list_for_each(p, head) { @@ -507,17 +526,17 @@ static struct dentry *autofs4_lookup_expiring(struct dentry *dentry) goto next; if (d_unhashed(expiring)) { - dget(expiring); + dget_dlock(expiring); spin_unlock(&expiring->d_lock); spin_unlock(&sbi->lookup_lock); - spin_unlock(&dcache_lock); + spin_unlock(&autofs4_lock); return expiring; } next: spin_unlock(&expiring->d_lock); } spin_unlock(&sbi->lookup_lock); - spin_unlock(&dcache_lock); + spin_unlock(&autofs4_lock); return NULL; } @@ -559,7 +578,7 @@ static struct dentry *autofs4_lookup(struct inode *dir, struct dentry *dentry, s * we check for the hashed dentry and return the newly * hashed dentry. */ - dentry->d_op = &autofs4_root_dentry_operations; + d_set_d_op(dentry, &autofs4_root_dentry_operations); /* * And we need to ensure that the same dentry is used for @@ -698,9 +717,9 @@ static int autofs4_dir_symlink(struct inode *dir, d_add(dentry, inode); if (dir == dir->i_sb->s_root->d_inode) - dentry->d_op = &autofs4_root_dentry_operations; + d_set_d_op(dentry, &autofs4_root_dentry_operations); else - dentry->d_op = &autofs4_dentry_operations; + d_set_d_op(dentry, &autofs4_dentry_operations); dentry->d_fsdata = ino; ino->dentry = dget(dentry); @@ -753,12 +772,12 @@ static int autofs4_dir_unlink(struct inode *dir, struct dentry *dentry) dir->i_mtime = CURRENT_TIME; - spin_lock(&dcache_lock); + spin_lock(&autofs4_lock); autofs4_add_expiring(dentry); spin_lock(&dentry->d_lock); __d_drop(dentry); spin_unlock(&dentry->d_lock); - spin_unlock(&dcache_lock); + spin_unlock(&autofs4_lock); return 0; } @@ -775,16 +794,20 @@ static int autofs4_dir_rmdir(struct inode *dir, struct dentry *dentry) if (!autofs4_oz_mode(sbi)) return -EACCES; - spin_lock(&dcache_lock); + spin_lock(&autofs4_lock); + spin_lock(&sbi->lookup_lock); + spin_lock(&dentry->d_lock); if (!list_empty(&dentry->d_subdirs)) { - spin_unlock(&dcache_lock); + spin_unlock(&dentry->d_lock); + spin_unlock(&sbi->lookup_lock); + spin_unlock(&autofs4_lock); return -ENOTEMPTY; } - autofs4_add_expiring(dentry); - spin_lock(&dentry->d_lock); + __autofs4_add_expiring(dentry); + spin_unlock(&sbi->lookup_lock); __d_drop(dentry); spin_unlock(&dentry->d_lock); - spin_unlock(&dcache_lock); + spin_unlock(&autofs4_lock); if (atomic_dec_and_test(&ino->count)) { p_ino = autofs4_dentry_ino(dentry->d_parent); @@ -829,9 +852,9 @@ static int autofs4_dir_mkdir(struct inode *dir, struct dentry *dentry, int mode) d_add(dentry, inode); if (dir == dir->i_sb->s_root->d_inode) - dentry->d_op = &autofs4_root_dentry_operations; + d_set_d_op(dentry, &autofs4_root_dentry_operations); else - dentry->d_op = &autofs4_dentry_operations; + d_set_d_op(dentry, &autofs4_dentry_operations); dentry->d_fsdata = ino; ino->dentry = dget(dentry); @@ -980,19 +1003,11 @@ static int autofs4_root_ioctl_unlocked(struct inode *inode, struct file *filp, } } -static DEFINE_MUTEX(autofs4_ioctl_mutex); - static long autofs4_root_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { - long ret; struct inode *inode = filp->f_dentry->d_inode; - - mutex_lock(&autofs4_ioctl_mutex); - ret = autofs4_root_ioctl_unlocked(inode, filp, cmd, arg); - mutex_unlock(&autofs4_ioctl_mutex); - - return ret; + return autofs4_root_ioctl_unlocked(inode, filp, cmd, arg); } #ifdef CONFIG_COMPAT @@ -1002,13 +1017,11 @@ static long autofs4_root_compat_ioctl(struct file *filp, struct inode *inode = filp->f_path.dentry->d_inode; int ret; - mutex_lock(&autofs4_ioctl_mutex); if (cmd == AUTOFS_IOC_READY || cmd == AUTOFS_IOC_FAIL) ret = autofs4_root_ioctl_unlocked(inode, filp, cmd, arg); else ret = autofs4_root_ioctl_unlocked(inode, filp, cmd, (unsigned long)compat_ptr(arg)); - mutex_unlock(&autofs4_ioctl_mutex); return ret; } diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c index 2341375386f..c5f8459c905 100644 --- a/fs/autofs4/waitq.c +++ b/fs/autofs4/waitq.c @@ -186,16 +186,26 @@ static int autofs4_getpath(struct autofs_sb_info *sbi, { struct dentry *root = sbi->sb->s_root; struct dentry *tmp; - char *buf = *name; + char *buf; char *p; - int len = 0; + int len; + unsigned seq; - spin_lock(&dcache_lock); +rename_retry: + buf = *name; + len = 0; + + seq = read_seqbegin(&rename_lock); + rcu_read_lock(); + spin_lock(&autofs4_lock); for (tmp = dentry ; tmp != root ; tmp = tmp->d_parent) len += tmp->d_name.len + 1; if (!len || --len > NAME_MAX) { - spin_unlock(&dcache_lock); + spin_unlock(&autofs4_lock); + rcu_read_unlock(); + if (read_seqretry(&rename_lock, seq)) + goto rename_retry; return 0; } @@ -208,7 +218,10 @@ static int autofs4_getpath(struct autofs_sb_info *sbi, p -= tmp->d_name.len; strncpy(p, tmp->d_name.name, tmp->d_name.len); } - spin_unlock(&dcache_lock); + spin_unlock(&autofs4_lock); + rcu_read_unlock(); + if (read_seqretry(&rename_lock, seq)) + goto rename_retry; return len; } diff --git a/fs/bad_inode.c b/fs/bad_inode.c index f024d8aadde..9ad2369d9e3 100644 --- a/fs/bad_inode.c +++ b/fs/bad_inode.c @@ -229,8 +229,11 @@ static int bad_inode_readlink(struct dentry *dentry, char __user *buffer, return -EIO; } -static int bad_inode_permission(struct inode *inode, int mask) +static int bad_inode_permission(struct inode *inode, int mask, unsigned int flags) { + if (flags & IPERM_FLAG_RCU) + return -ECHILD; + return -EIO; } diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c index aa4e7c7ae3c..de93581b79a 100644 --- a/fs/befs/linuxvfs.c +++ b/fs/befs/linuxvfs.c @@ -284,12 +284,18 @@ befs_alloc_inode(struct super_block *sb) return &bi->vfs_inode; } -static void -befs_destroy_inode(struct inode *inode) +static void befs_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(befs_inode_cachep, BEFS_I(inode)); } +static void befs_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, befs_i_callback); +} + static void init_once(void *foo) { struct befs_inode_info *bi = (struct befs_inode_info *) foo; diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c index 76db6d7d49b..a8e37f81d09 100644 --- a/fs/bfs/inode.c +++ b/fs/bfs/inode.c @@ -248,11 +248,18 @@ static struct inode *bfs_alloc_inode(struct super_block *sb) return &bi->vfs_inode; } -static void bfs_destroy_inode(struct inode *inode) +static void bfs_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(bfs_inode_cachep, BFS_I(inode)); } +static void bfs_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, bfs_i_callback); +} + static void init_once(void *foo) { struct bfs_inode_info *bi = foo; @@ -370,6 +370,9 @@ struct bio *bio_kmalloc(gfp_t gfp_mask, int nr_iovecs) { struct bio *bio; + if (nr_iovecs > UIO_MAXIOV) + return NULL; + bio = kmalloc(sizeof(struct bio) + nr_iovecs * sizeof(struct bio_vec), gfp_mask); if (unlikely(!bio)) @@ -697,8 +700,12 @@ static void bio_free_map_data(struct bio_map_data *bmd) static struct bio_map_data *bio_alloc_map_data(int nr_segs, int iov_count, gfp_t gfp_mask) { - struct bio_map_data *bmd = kmalloc(sizeof(*bmd), gfp_mask); + struct bio_map_data *bmd; + if (iov_count > UIO_MAXIOV) + return NULL; + + bmd = kmalloc(sizeof(*bmd), gfp_mask); if (!bmd) return NULL; @@ -827,6 +834,12 @@ struct bio *bio_copy_user_iov(struct request_queue *q, end = (uaddr + iov[i].iov_len + PAGE_SIZE - 1) >> PAGE_SHIFT; start = uaddr >> PAGE_SHIFT; + /* + * Overflow, abort + */ + if (end < start) + return ERR_PTR(-EINVAL); + nr_pages += end - start; len += iov[i].iov_len; } @@ -955,6 +968,12 @@ static struct bio *__bio_map_user_iov(struct request_queue *q, unsigned long end = (uaddr + len + PAGE_SIZE - 1) >> PAGE_SHIFT; unsigned long start = uaddr >> PAGE_SHIFT; + /* + * Overflow, abort + */ + if (end < start) + return ERR_PTR(-EINVAL); + nr_pages += end - start; /* * buffer must be aligned to at least hardsector size for now @@ -982,7 +1001,7 @@ static struct bio *__bio_map_user_iov(struct request_queue *q, unsigned long start = uaddr >> PAGE_SHIFT; const int local_nr_pages = end - start; const int page_limit = cur_page + local_nr_pages; - + ret = get_user_pages_fast(uaddr, local_nr_pages, write_to_vm, &pages[cur_page]); if (ret < local_nr_pages) { diff --git a/fs/block_dev.c b/fs/block_dev.c index 06e8ff12b97..771f2352701 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -11,7 +11,6 @@ #include <linux/slab.h> #include <linux/kmod.h> #include <linux/major.h> -#include <linux/smp_lock.h> #include <linux/device_cgroup.h> #include <linux/highmem.h> #include <linux/blkdev.h> @@ -410,13 +409,20 @@ static struct inode *bdev_alloc_inode(struct super_block *sb) return &ei->vfs_inode; } -static void bdev_destroy_inode(struct inode *inode) +static void bdev_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); struct bdev_inode *bdi = BDEV_I(inode); + INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(bdev_cachep, bdi); } +static void bdev_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, bdev_i_callback); +} + static void init_once(void *foo) { struct bdev_inode *ei = (struct bdev_inode *) foo; diff --git a/fs/btrfs/acl.c b/fs/btrfs/acl.c index 2222d161c7b..6ae2c8cac9d 100644 --- a/fs/btrfs/acl.c +++ b/fs/btrfs/acl.c @@ -185,18 +185,23 @@ static int btrfs_xattr_acl_set(struct dentry *dentry, const char *name, return ret; } -int btrfs_check_acl(struct inode *inode, int mask) +int btrfs_check_acl(struct inode *inode, int mask, unsigned int flags) { - struct posix_acl *acl; int error = -EAGAIN; - acl = btrfs_get_acl(inode, ACL_TYPE_ACCESS); + if (flags & IPERM_FLAG_RCU) { + if (!negative_cached_acl(inode, ACL_TYPE_ACCESS)) + error = -ECHILD; - if (IS_ERR(acl)) - return PTR_ERR(acl); - if (acl) { - error = posix_acl_permission(inode, acl, mask); - posix_acl_release(acl); + } else { + struct posix_acl *acl; + acl = btrfs_get_acl(inode, ACL_TYPE_ACCESS); + if (IS_ERR(acl)) + return PTR_ERR(acl); + if (acl) { + error = posix_acl_permission(inode, acl, mask); + posix_acl_release(acl); + } } return error; diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c index 7845d1f7d1d..b50bc4bd5c5 100644 --- a/fs/btrfs/compression.c +++ b/fs/btrfs/compression.c @@ -91,23 +91,10 @@ static inline int compressed_bio_size(struct btrfs_root *root, static struct bio *compressed_bio_alloc(struct block_device *bdev, u64 first_byte, gfp_t gfp_flags) { - struct bio *bio; int nr_vecs; nr_vecs = bio_get_nr_vecs(bdev); - bio = bio_alloc(gfp_flags, nr_vecs); - - if (bio == NULL && (current->flags & PF_MEMALLOC)) { - while (!bio && (nr_vecs /= 2)) - bio = bio_alloc(gfp_flags, nr_vecs); - } - - if (bio) { - bio->bi_size = 0; - bio->bi_bdev = bdev; - bio->bi_sector = first_byte >> 9; - } - return bio; + return btrfs_bio_alloc(bdev, first_byte >> 9, nr_vecs, gfp_flags); } static int check_compressed_csum(struct inode *inode, diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index 8db9234f6b4..a142d204b52 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -808,9 +808,9 @@ struct btrfs_block_group_cache { int extents_thresh; int free_extents; int total_bitmaps; - int ro:1; - int dirty:1; - int iref:1; + unsigned int ro:1; + unsigned int dirty:1; + unsigned int iref:1; int disk_cache_state; @@ -2544,7 +2544,7 @@ int btrfs_sync_fs(struct super_block *sb, int wait); /* acl.c */ #ifdef CONFIG_BTRFS_FS_POSIX_ACL -int btrfs_check_acl(struct inode *inode, int mask); +int btrfs_check_acl(struct inode *inode, int mask, unsigned int flags); #else #define btrfs_check_acl NULL #endif diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index fb827d0d718..51d2e4de34e 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -28,6 +28,7 @@ #include <linux/freezer.h> #include <linux/crc32c.h> #include <linux/slab.h> +#include <linux/migrate.h> #include "compat.h" #include "ctree.h" #include "disk-io.h" @@ -355,6 +356,8 @@ static int csum_dirty_buffer(struct btrfs_root *root, struct page *page) ret = btree_read_extent_buffer_pages(root, eb, start + PAGE_CACHE_SIZE, btrfs_header_generation(eb)); BUG_ON(ret); + WARN_ON(!btrfs_header_flag(eb, BTRFS_HEADER_FLAG_WRITTEN)); + found_start = btrfs_header_bytenr(eb); if (found_start != start) { WARN_ON(1); @@ -693,6 +696,27 @@ static int btree_submit_bio_hook(struct inode *inode, int rw, struct bio *bio, __btree_submit_bio_done); } +#ifdef CONFIG_MIGRATION +static int btree_migratepage(struct address_space *mapping, + struct page *newpage, struct page *page) +{ + /* + * we can't safely write a btree page from here, + * we haven't done the locking hook + */ + if (PageDirty(page)) + return -EAGAIN; + /* + * Buffers may be managed in a filesystem specific way. + * We must have no buffers or drop them. + */ + if (page_has_private(page) && + !try_to_release_page(page, GFP_KERNEL)) + return -EAGAIN; + return migrate_page(mapping, newpage, page); +} +#endif + static int btree_writepage(struct page *page, struct writeback_control *wbc) { struct extent_io_tree *tree; @@ -707,8 +731,7 @@ static int btree_writepage(struct page *page, struct writeback_control *wbc) } redirty_page_for_writepage(wbc, page); - eb = btrfs_find_tree_block(root, page_offset(page), - PAGE_CACHE_SIZE); + eb = btrfs_find_tree_block(root, page_offset(page), PAGE_CACHE_SIZE); WARN_ON(!eb); was_dirty = test_and_set_bit(EXTENT_BUFFER_DIRTY, &eb->bflags); @@ -799,6 +822,9 @@ static const struct address_space_operations btree_aops = { .releasepage = btree_releasepage, .invalidatepage = btree_invalidatepage, .sync_page = block_sync_page, +#ifdef CONFIG_MIGRATION + .migratepage = btree_migratepage, +#endif }; int readahead_tree_block(struct btrfs_root *root, u64 bytenr, u32 blocksize, @@ -981,7 +1007,10 @@ static int find_and_setup_root(struct btrfs_root *tree_root, blocksize = btrfs_level_size(root, btrfs_root_level(&root->root_item)); root->node = read_tree_block(root, btrfs_root_bytenr(&root->root_item), blocksize, generation); - BUG_ON(!root->node); + if (!root->node || !btrfs_buffer_uptodate(root->node, generation)) { + free_extent_buffer(root->node); + return -EIO; + } root->commit_root = btrfs_root_node(root); return 0; } @@ -1538,10 +1567,8 @@ struct btrfs_root *open_ctree(struct super_block *sb, GFP_NOFS); struct btrfs_root *csum_root = kzalloc(sizeof(struct btrfs_root), GFP_NOFS); - struct btrfs_root *tree_root = kzalloc(sizeof(struct btrfs_root), - GFP_NOFS); - struct btrfs_fs_info *fs_info = kzalloc(sizeof(*fs_info), - GFP_NOFS); + struct btrfs_root *tree_root = btrfs_sb(sb); + struct btrfs_fs_info *fs_info = tree_root->fs_info; struct btrfs_root *chunk_root = kzalloc(sizeof(struct btrfs_root), GFP_NOFS); struct btrfs_root *dev_root = kzalloc(sizeof(struct btrfs_root), diff --git a/fs/btrfs/export.c b/fs/btrfs/export.c index 951ef09b82f..0ccf9a8afcd 100644 --- a/fs/btrfs/export.c +++ b/fs/btrfs/export.c @@ -110,7 +110,7 @@ static struct dentry *btrfs_get_dentry(struct super_block *sb, u64 objectid, dentry = d_obtain_alias(inode); if (!IS_ERR(dentry)) - dentry->d_op = &btrfs_dentry_operations; + d_set_d_op(dentry, &btrfs_dentry_operations); return dentry; fail: srcu_read_unlock(&fs_info->subvol_srcu, index); @@ -166,7 +166,7 @@ static struct dentry *btrfs_fh_to_dentry(struct super_block *sb, struct fid *fh, static struct dentry *btrfs_get_parent(struct dentry *child) { struct inode *dir = child->d_inode; - static struct dentry *dentry; + struct dentry *dentry; struct btrfs_root *root = BTRFS_I(dir)->root; struct btrfs_path *path; struct extent_buffer *leaf; @@ -225,16 +225,92 @@ static struct dentry *btrfs_get_parent(struct dentry *child) key.offset = 0; dentry = d_obtain_alias(btrfs_iget(root->fs_info->sb, &key, root, NULL)); if (!IS_ERR(dentry)) - dentry->d_op = &btrfs_dentry_operations; + d_set_d_op(dentry, &btrfs_dentry_operations); return dentry; fail: btrfs_free_path(path); return ERR_PTR(ret); } +static int btrfs_get_name(struct dentry *parent, char *name, + struct dentry *child) +{ + struct inode *inode = child->d_inode; + struct inode *dir = parent->d_inode; + struct btrfs_path *path; + struct btrfs_root *root = BTRFS_I(dir)->root; + struct btrfs_inode_ref *iref; + struct btrfs_root_ref *rref; + struct extent_buffer *leaf; + unsigned long name_ptr; + struct btrfs_key key; + int name_len; + int ret; + + if (!dir || !inode) + return -EINVAL; + + if (!S_ISDIR(dir->i_mode)) + return -EINVAL; + + path = btrfs_alloc_path(); + if (!path) + return -ENOMEM; + path->leave_spinning = 1; + + if (inode->i_ino == BTRFS_FIRST_FREE_OBJECTID) { + key.objectid = BTRFS_I(inode)->root->root_key.objectid; + key.type = BTRFS_ROOT_BACKREF_KEY; + key.offset = (u64)-1; + root = root->fs_info->tree_root; + } else { + key.objectid = inode->i_ino; + key.offset = dir->i_ino; + key.type = BTRFS_INODE_REF_KEY; + } + + ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); + if (ret < 0) { + btrfs_free_path(path); + return ret; + } else if (ret > 0) { + if (inode->i_ino == BTRFS_FIRST_FREE_OBJECTID) { + path->slots[0]--; + } else { + btrfs_free_path(path); + return -ENOENT; + } + } + leaf = path->nodes[0]; + + if (inode->i_ino == BTRFS_FIRST_FREE_OBJECTID) { + rref = btrfs_item_ptr(leaf, path->slots[0], + struct btrfs_root_ref); + name_ptr = (unsigned long)(rref + 1); + name_len = btrfs_root_ref_name_len(leaf, rref); + } else { + iref = btrfs_item_ptr(leaf, path->slots[0], + struct btrfs_inode_ref); + name_ptr = (unsigned long)(iref + 1); + name_len = btrfs_inode_ref_name_len(leaf, iref); + } + + read_extent_buffer(leaf, name, name_ptr, name_len); + btrfs_free_path(path); + + /* + * have to add the null termination to make sure that reconnect_path + * gets the right len for strlen + */ + name[name_len] = '\0'; + + return 0; +} + const struct export_operations btrfs_export_ops = { .encode_fh = btrfs_encode_fh, .fh_to_dentry = btrfs_fh_to_dentry, .fh_to_parent = btrfs_fh_to_parent, .get_parent = btrfs_get_parent, + .get_name = btrfs_get_name, }; diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 0c097f3aec4..227e5815d83 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -429,6 +429,7 @@ err: static int cache_block_group(struct btrfs_block_group_cache *cache, struct btrfs_trans_handle *trans, + struct btrfs_root *root, int load_cache_only) { struct btrfs_fs_info *fs_info = cache->fs_info; @@ -442,9 +443,12 @@ static int cache_block_group(struct btrfs_block_group_cache *cache, /* * We can't do the read from on-disk cache during a commit since we need - * to have the normal tree locking. + * to have the normal tree locking. Also if we are currently trying to + * allocate blocks for the tree root we can't do the fast caching since + * we likely hold important locks. */ - if (!trans->transaction->in_commit) { + if (!trans->transaction->in_commit && + (root && root != root->fs_info->tree_root)) { spin_lock(&cache->lock); if (cache->cached != BTRFS_CACHE_NO) { spin_unlock(&cache->lock); @@ -2741,6 +2745,7 @@ static int cache_save_setup(struct btrfs_block_group_cache *block_group, struct btrfs_root *root = block_group->fs_info->tree_root; struct inode *inode = NULL; u64 alloc_hint = 0; + int dcs = BTRFS_DC_ERROR; int num_pages = 0; int retries = 0; int ret = 0; @@ -2795,6 +2800,8 @@ again: spin_lock(&block_group->lock); if (block_group->cached != BTRFS_CACHE_FINISHED) { + /* We're not cached, don't bother trying to write stuff out */ + dcs = BTRFS_DC_WRITTEN; spin_unlock(&block_group->lock); goto out_put; } @@ -2821,6 +2828,8 @@ again: ret = btrfs_prealloc_file_range_trans(inode, trans, 0, 0, num_pages, num_pages, num_pages, &alloc_hint); + if (!ret) + dcs = BTRFS_DC_SETUP; btrfs_free_reserved_data_space(inode, num_pages); out_put: iput(inode); @@ -2828,10 +2837,7 @@ out_free: btrfs_release_path(root, path); out: spin_lock(&block_group->lock); - if (ret) - block_group->disk_cache_state = BTRFS_DC_ERROR; - else - block_group->disk_cache_state = BTRFS_DC_SETUP; + block_group->disk_cache_state = dcs; spin_unlock(&block_group->lock); return ret; @@ -3037,7 +3043,13 @@ static void set_avail_alloc_bits(struct btrfs_fs_info *fs_info, u64 flags) u64 btrfs_reduce_alloc_profile(struct btrfs_root *root, u64 flags) { - u64 num_devices = root->fs_info->fs_devices->rw_devices; + /* + * we add in the count of missing devices because we want + * to make sure that any RAID levels on a degraded FS + * continue to be honored. + */ + u64 num_devices = root->fs_info->fs_devices->rw_devices + + root->fs_info->fs_devices->missing_devices; if (num_devices == 1) flags &= ~(BTRFS_BLOCK_GROUP_RAID1 | BTRFS_BLOCK_GROUP_RAID0); @@ -3412,7 +3424,7 @@ again: * our reservation. */ if (unused <= space_info->total_bytes) { - unused -= space_info->total_bytes; + unused = space_info->total_bytes - unused; if (unused >= num_bytes) { if (!reserved) space_info->bytes_reserved += orig_bytes; @@ -4080,7 +4092,7 @@ static int update_block_group(struct btrfs_trans_handle *trans, * space back to the block group, otherwise we will leak space. */ if (!alloc && cache->cached == BTRFS_CACHE_NO) - cache_block_group(cache, trans, 1); + cache_block_group(cache, trans, NULL, 1); byte_in_group = bytenr - cache->key.objectid; WARN_ON(byte_in_group > cache->key.offset); @@ -4930,11 +4942,31 @@ search: btrfs_get_block_group(block_group); search_start = block_group->key.objectid; + /* + * this can happen if we end up cycling through all the + * raid types, but we want to make sure we only allocate + * for the proper type. + */ + if (!block_group_bits(block_group, data)) { + u64 extra = BTRFS_BLOCK_GROUP_DUP | + BTRFS_BLOCK_GROUP_RAID1 | + BTRFS_BLOCK_GROUP_RAID10; + + /* + * if they asked for extra copies and this block group + * doesn't provide them, bail. This does allow us to + * fill raid0 from raid1. + */ + if ((data & extra) && !(block_group->flags & extra)) + goto loop; + } + have_block_group: if (unlikely(block_group->cached == BTRFS_CACHE_NO)) { u64 free_percent; - ret = cache_block_group(block_group, trans, 1); + ret = cache_block_group(block_group, trans, + orig_root, 1); if (block_group->cached == BTRFS_CACHE_FINISHED) goto have_block_group; @@ -4958,7 +4990,8 @@ have_block_group: if (loop > LOOP_CACHING_NOWAIT || (loop > LOOP_FIND_IDEAL && atomic_read(&space_info->caching_threads) < 2)) { - ret = cache_block_group(block_group, trans, 0); + ret = cache_block_group(block_group, trans, + orig_root, 0); BUG_ON(ret); } found_uncached_bg = true; @@ -5515,7 +5548,7 @@ int btrfs_alloc_logged_file_extent(struct btrfs_trans_handle *trans, u64 num_bytes = ins->offset; block_group = btrfs_lookup_block_group(root->fs_info, ins->objectid); - cache_block_group(block_group, trans, 0); + cache_block_group(block_group, trans, NULL, 0); caching_ctl = get_caching_control(block_group); if (!caching_ctl) { @@ -6300,9 +6333,13 @@ int btrfs_drop_snapshot(struct btrfs_root *root, NULL, NULL); BUG_ON(ret < 0); if (ret > 0) { - ret = btrfs_del_orphan_item(trans, tree_root, - root->root_key.objectid); - BUG_ON(ret); + /* if we fail to delete the orphan item this time + * around, it'll get picked up the next time. + * + * The most common failure here is just -ENOENT. + */ + btrfs_del_orphan_item(trans, tree_root, + root->root_key.objectid); } } @@ -7878,7 +7915,14 @@ static u64 update_block_group_flags(struct btrfs_root *root, u64 flags) u64 stripped = BTRFS_BLOCK_GROUP_RAID0 | BTRFS_BLOCK_GROUP_RAID1 | BTRFS_BLOCK_GROUP_RAID10; - num_devices = root->fs_info->fs_devices->rw_devices; + /* + * we add in the count of missing devices because we want + * to make sure that any RAID levels on a degraded FS + * continue to be honored. + */ + num_devices = root->fs_info->fs_devices->rw_devices + + root->fs_info->fs_devices->missing_devices; + if (num_devices == 1) { stripped |= BTRFS_BLOCK_GROUP_DUP; stripped = flags & ~stripped; @@ -8247,7 +8291,6 @@ int btrfs_read_block_groups(struct btrfs_root *root) break; if (ret != 0) goto error; - leaf = path->nodes[0]; btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]); cache = kzalloc(sizeof(*cache), GFP_NOFS); diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index eac10e3260a..3e86b9f3650 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -1828,9 +1828,9 @@ static void end_bio_extent_preparewrite(struct bio *bio, int err) bio_put(bio); } -static struct bio * -extent_bio_alloc(struct block_device *bdev, u64 first_sector, int nr_vecs, - gfp_t gfp_flags) +struct bio * +btrfs_bio_alloc(struct block_device *bdev, u64 first_sector, int nr_vecs, + gfp_t gfp_flags) { struct bio *bio; @@ -1919,7 +1919,7 @@ static int submit_extent_page(int rw, struct extent_io_tree *tree, else nr = bio_get_nr_vecs(bdev); - bio = extent_bio_alloc(bdev, sector, nr, GFP_NOFS | __GFP_HIGH); + bio = btrfs_bio_alloc(bdev, sector, nr, GFP_NOFS | __GFP_HIGH); bio_add_page(bio, page, page_size, offset); bio->bi_end_io = end_io_func; @@ -2901,21 +2901,53 @@ out: int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, __u64 start, __u64 len, get_extent_t *get_extent) { - int ret; + int ret = 0; u64 off = start; u64 max = start + len; u32 flags = 0; + u32 found_type; + u64 last; u64 disko = 0; + struct btrfs_key found_key; struct extent_map *em = NULL; struct extent_state *cached_state = NULL; + struct btrfs_path *path; + struct btrfs_file_extent_item *item; int end = 0; u64 em_start = 0, em_len = 0; unsigned long emflags; - ret = 0; + int hole = 0; if (len == 0) return -EINVAL; + path = btrfs_alloc_path(); + if (!path) + return -ENOMEM; + path->leave_spinning = 1; + + ret = btrfs_lookup_file_extent(NULL, BTRFS_I(inode)->root, + path, inode->i_ino, -1, 0); + if (ret < 0) { + btrfs_free_path(path); + return ret; + } + WARN_ON(!ret); + path->slots[0]--; + item = btrfs_item_ptr(path->nodes[0], path->slots[0], + struct btrfs_file_extent_item); + btrfs_item_key_to_cpu(path->nodes[0], &found_key, path->slots[0]); + found_type = btrfs_key_type(&found_key); + + /* No extents, just return */ + if (found_key.objectid != inode->i_ino || + found_type != BTRFS_EXTENT_DATA_KEY) { + btrfs_free_path(path); + return 0; + } + last = found_key.offset; + btrfs_free_path(path); + lock_extent_bits(&BTRFS_I(inode)->io_tree, start, start + len, 0, &cached_state, GFP_NOFS); em = get_extent(inode, NULL, 0, off, max - off, 0); @@ -2925,11 +2957,18 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, ret = PTR_ERR(em); goto out; } + while (!end) { + hole = 0; off = em->start + em->len; if (off >= max) end = 1; + if (em->block_start == EXTENT_MAP_HOLE) { + hole = 1; + goto next; + } + em_start = em->start; em_len = em->len; @@ -2939,8 +2978,6 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, if (em->block_start == EXTENT_MAP_LAST_BYTE) { end = 1; flags |= FIEMAP_EXTENT_LAST; - } else if (em->block_start == EXTENT_MAP_HOLE) { - flags |= FIEMAP_EXTENT_UNWRITTEN; } else if (em->block_start == EXTENT_MAP_INLINE) { flags |= (FIEMAP_EXTENT_DATA_INLINE | FIEMAP_EXTENT_NOT_ALIGNED); @@ -2953,10 +2990,10 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, if (test_bit(EXTENT_FLAG_COMPRESSED, &em->flags)) flags |= FIEMAP_EXTENT_ENCODED; +next: emflags = em->flags; free_extent_map(em); em = NULL; - if (!end) { em = get_extent(inode, NULL, 0, off, max - off, 0); if (!em) @@ -2967,15 +3004,23 @@ int extent_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, } emflags = em->flags; } + if (test_bit(EXTENT_FLAG_VACANCY, &emflags)) { flags |= FIEMAP_EXTENT_LAST; end = 1; } - ret = fiemap_fill_next_extent(fieinfo, em_start, disko, - em_len, flags); - if (ret) - goto out_free; + if (em_start == last) { + flags |= FIEMAP_EXTENT_LAST; + end = 1; + } + + if (!hole) { + ret = fiemap_fill_next_extent(fieinfo, em_start, disko, + em_len, flags); + if (ret) + goto out_free; + } } out_free: free_extent_map(em); @@ -3836,8 +3881,10 @@ int try_release_extent_buffer(struct extent_io_tree *tree, struct page *page) spin_lock(&tree->buffer_lock); eb = radix_tree_lookup(&tree->buffer, start >> PAGE_CACHE_SHIFT); - if (!eb) - goto out; + if (!eb) { + spin_unlock(&tree->buffer_lock); + return ret; + } if (test_bit(EXTENT_BUFFER_DIRTY, &eb->bflags)) { ret = 0; diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h index 1c6d4f342ef..4183c8178f0 100644 --- a/fs/btrfs/extent_io.h +++ b/fs/btrfs/extent_io.h @@ -310,4 +310,7 @@ int extent_clear_unlock_delalloc(struct inode *inode, struct extent_io_tree *tree, u64 start, u64 end, struct page *locked_page, unsigned long op); +struct bio * +btrfs_bio_alloc(struct block_device *bdev, u64 first_sector, int nr_vecs, + gfp_t gfp_flags); #endif diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index e354c33df08..66836d85763 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -48,30 +48,34 @@ static noinline int btrfs_copy_from_user(loff_t pos, int num_pages, struct page **prepared_pages, struct iov_iter *i) { - size_t copied; + size_t copied = 0; int pg = 0; int offset = pos & (PAGE_CACHE_SIZE - 1); + int total_copied = 0; while (write_bytes > 0) { size_t count = min_t(size_t, PAGE_CACHE_SIZE - offset, write_bytes); struct page *page = prepared_pages[pg]; -again: - if (unlikely(iov_iter_fault_in_readable(i, count))) - return -EFAULT; - - /* Copy data from userspace to the current page */ - copied = iov_iter_copy_from_user(page, i, offset, count); + /* + * Copy data from userspace to the current page + * + * Disable pagefault to avoid recursive lock since + * the pages are already locked + */ + pagefault_disable(); + copied = iov_iter_copy_from_user_atomic(page, i, offset, count); + pagefault_enable(); /* Flush processor's dcache for this page */ flush_dcache_page(page); iov_iter_advance(i, copied); write_bytes -= copied; + total_copied += copied; + /* Return to btrfs_file_aio_write to fault page */ if (unlikely(copied == 0)) { - count = min_t(size_t, PAGE_CACHE_SIZE - offset, - iov_iter_single_seg_count(i)); - goto again; + break; } if (unlikely(copied < PAGE_CACHE_SIZE - offset)) { @@ -81,7 +85,7 @@ again: offset = 0; } } - return 0; + return total_copied; } /* @@ -854,6 +858,8 @@ static ssize_t btrfs_file_aio_write(struct kiocb *iocb, unsigned long last_index; int will_write; int buffered = 0; + int copied = 0; + int dirty_pages = 0; will_write = ((file->f_flags & O_DSYNC) || IS_SYNC(inode) || (file->f_flags & O_DIRECT)); @@ -970,7 +976,17 @@ static ssize_t btrfs_file_aio_write(struct kiocb *iocb, WARN_ON(num_pages > nrptrs); memset(pages, 0, sizeof(struct page *) * nrptrs); - ret = btrfs_delalloc_reserve_space(inode, write_bytes); + /* + * Fault pages before locking them in prepare_pages + * to avoid recursive lock + */ + if (unlikely(iov_iter_fault_in_readable(&i, write_bytes))) { + ret = -EFAULT; + goto out; + } + + ret = btrfs_delalloc_reserve_space(inode, + num_pages << PAGE_CACHE_SHIFT); if (ret) goto out; @@ -978,37 +994,49 @@ static ssize_t btrfs_file_aio_write(struct kiocb *iocb, pos, first_index, last_index, write_bytes); if (ret) { - btrfs_delalloc_release_space(inode, write_bytes); + btrfs_delalloc_release_space(inode, + num_pages << PAGE_CACHE_SHIFT); goto out; } - ret = btrfs_copy_from_user(pos, num_pages, + copied = btrfs_copy_from_user(pos, num_pages, write_bytes, pages, &i); - if (ret == 0) { + dirty_pages = (copied + PAGE_CACHE_SIZE - 1) >> + PAGE_CACHE_SHIFT; + + if (num_pages > dirty_pages) { + if (copied > 0) + atomic_inc( + &BTRFS_I(inode)->outstanding_extents); + btrfs_delalloc_release_space(inode, + (num_pages - dirty_pages) << + PAGE_CACHE_SHIFT); + } + + if (copied > 0) { dirty_and_release_pages(NULL, root, file, pages, - num_pages, pos, write_bytes); + dirty_pages, pos, copied); } btrfs_drop_pages(pages, num_pages); - if (ret) { - btrfs_delalloc_release_space(inode, write_bytes); - goto out; - } - if (will_write) { - filemap_fdatawrite_range(inode->i_mapping, pos, - pos + write_bytes - 1); - } else { - balance_dirty_pages_ratelimited_nr(inode->i_mapping, - num_pages); - if (num_pages < - (root->leafsize >> PAGE_CACHE_SHIFT) + 1) - btrfs_btree_balance_dirty(root, 1); - btrfs_throttle(root); + if (copied > 0) { + if (will_write) { + filemap_fdatawrite_range(inode->i_mapping, pos, + pos + copied - 1); + } else { + balance_dirty_pages_ratelimited_nr( + inode->i_mapping, + dirty_pages); + if (dirty_pages < + (root->leafsize >> PAGE_CACHE_SHIFT) + 1) + btrfs_btree_balance_dirty(root, 1); + btrfs_throttle(root); + } } - pos += write_bytes; - num_written += write_bytes; + pos += copied; + num_written += copied; cond_resched(); } @@ -1047,8 +1075,14 @@ out: if ((file->f_flags & O_DSYNC) || IS_SYNC(inode)) { trans = btrfs_start_transaction(root, 0); + if (IS_ERR(trans)) { + num_written = PTR_ERR(trans); + goto done; + } + mutex_lock(&inode->i_mutex); ret = btrfs_log_dentry_safe(trans, root, file->f_dentry); + mutex_unlock(&inode->i_mutex); if (ret == 0) { ret = btrfs_sync_log(trans, root); if (ret == 0) @@ -1067,6 +1101,7 @@ out: (start_pos + num_written - 1) >> PAGE_CACHE_SHIFT); } } +done: current->backing_dev_info = NULL; return num_written ? num_written : err; } diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c index 22ee0dc2e6b..60d68426695 100644 --- a/fs/btrfs/free-space-cache.c +++ b/fs/btrfs/free-space-cache.c @@ -290,7 +290,7 @@ int load_free_space_cache(struct btrfs_fs_info *fs_info, (unsigned long long)BTRFS_I(inode)->generation, (unsigned long long)generation, (unsigned long long)block_group->key.objectid); - goto out; + goto free_cache; } if (!num_entries) @@ -524,6 +524,12 @@ int btrfs_write_out_cache(struct btrfs_root *root, return 0; } + node = rb_first(&block_group->free_space_offset); + if (!node) { + iput(inode); + return 0; + } + last_index = (i_size_read(inode) - 1) >> PAGE_CACHE_SHIFT; filemap_write_and_wait(inode->i_mapping); btrfs_wait_ordered_range(inode, inode->i_size & @@ -543,10 +549,6 @@ int btrfs_write_out_cache(struct btrfs_root *root, */ first_page_offset = (sizeof(u32) * num_checksums) + sizeof(u64); - node = rb_first(&block_group->free_space_offset); - if (!node) - goto out_free; - /* * Lock all pages first so we can lock the extent safely. * diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 558cac2dfa5..a0ff46a4789 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -495,7 +495,7 @@ again: add_async_extent(async_cow, start, num_bytes, total_compressed, pages, nr_pages_ret); - if (start + num_bytes < end && start + num_bytes < actual_end) { + if (start + num_bytes < end) { start += num_bytes; pages = NULL; cond_resched(); @@ -4084,7 +4084,7 @@ struct inode *btrfs_lookup_dentry(struct inode *dir, struct dentry *dentry) int index; int ret; - dentry->d_op = &btrfs_dentry_operations; + d_set_d_op(dentry, &btrfs_dentry_operations); if (dentry->d_name.len > BTRFS_NAME_LEN) return ERR_PTR(-ENAMETOOLONG); @@ -4127,7 +4127,7 @@ struct inode *btrfs_lookup_dentry(struct inode *dir, struct dentry *dentry) return inode; } -static int btrfs_dentry_delete(struct dentry *dentry) +static int btrfs_dentry_delete(const struct dentry *dentry) { struct btrfs_root *root; @@ -4501,6 +4501,7 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans, BTRFS_I(inode)->index_cnt = 2; BTRFS_I(inode)->root = root; BTRFS_I(inode)->generation = trans->transid; + inode->i_generation = BTRFS_I(inode)->generation; btrfs_set_inode_space_info(root, inode); if (mode & S_IFDIR) @@ -4622,12 +4623,12 @@ int btrfs_add_link(struct btrfs_trans_handle *trans, } static int btrfs_add_nondir(struct btrfs_trans_handle *trans, - struct dentry *dentry, struct inode *inode, - int backref, u64 index) + struct inode *dir, struct dentry *dentry, + struct inode *inode, int backref, u64 index) { - int err = btrfs_add_link(trans, dentry->d_parent->d_inode, - inode, dentry->d_name.name, - dentry->d_name.len, backref, index); + int err = btrfs_add_link(trans, dir, inode, + dentry->d_name.name, dentry->d_name.len, + backref, index); if (!err) { d_instantiate(dentry, inode); return 0; @@ -4668,8 +4669,7 @@ static int btrfs_mknod(struct inode *dir, struct dentry *dentry, btrfs_set_trans_block_group(trans, dir); inode = btrfs_new_inode(trans, root, dir, dentry->d_name.name, - dentry->d_name.len, - dentry->d_parent->d_inode->i_ino, objectid, + dentry->d_name.len, dir->i_ino, objectid, BTRFS_I(dir)->block_group, mode, &index); err = PTR_ERR(inode); if (IS_ERR(inode)) @@ -4682,7 +4682,7 @@ static int btrfs_mknod(struct inode *dir, struct dentry *dentry, } btrfs_set_trans_block_group(trans, inode); - err = btrfs_add_nondir(trans, dentry, inode, 0, index); + err = btrfs_add_nondir(trans, dir, dentry, inode, 0, index); if (err) drop_inode = 1; else { @@ -4730,10 +4730,8 @@ static int btrfs_create(struct inode *dir, struct dentry *dentry, btrfs_set_trans_block_group(trans, dir); inode = btrfs_new_inode(trans, root, dir, dentry->d_name.name, - dentry->d_name.len, - dentry->d_parent->d_inode->i_ino, - objectid, BTRFS_I(dir)->block_group, mode, - &index); + dentry->d_name.len, dir->i_ino, objectid, + BTRFS_I(dir)->block_group, mode, &index); err = PTR_ERR(inode); if (IS_ERR(inode)) goto out_unlock; @@ -4745,7 +4743,7 @@ static int btrfs_create(struct inode *dir, struct dentry *dentry, } btrfs_set_trans_block_group(trans, inode); - err = btrfs_add_nondir(trans, dentry, inode, 0, index); + err = btrfs_add_nondir(trans, dir, dentry, inode, 0, index); if (err) drop_inode = 1; else { @@ -4787,6 +4785,7 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir, return -EPERM; btrfs_inc_nlink(inode); + inode->i_ctime = CURRENT_TIME; err = btrfs_set_inode_index(dir, &index); if (err) @@ -4805,15 +4804,17 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir, btrfs_set_trans_block_group(trans, dir); ihold(inode); - err = btrfs_add_nondir(trans, dentry, inode, 1, index); + err = btrfs_add_nondir(trans, dir, dentry, inode, 1, index); if (err) { drop_inode = 1; } else { + struct dentry *parent = dget_parent(dentry); btrfs_update_inode_block_group(trans, dir); err = btrfs_update_inode(trans, root, inode); BUG_ON(err); - btrfs_log_new_name(trans, inode, NULL, dentry->d_parent); + btrfs_log_new_name(trans, inode, NULL, parent); + dput(parent); } nr = trans->blocks_used; @@ -4853,8 +4854,7 @@ static int btrfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) btrfs_set_trans_block_group(trans, dir); inode = btrfs_new_inode(trans, root, dir, dentry->d_name.name, - dentry->d_name.len, - dentry->d_parent->d_inode->i_ino, objectid, + dentry->d_name.len, dir->i_ino, objectid, BTRFS_I(dir)->block_group, S_IFDIR | mode, &index); if (IS_ERR(inode)) { @@ -4877,9 +4877,8 @@ static int btrfs_mkdir(struct inode *dir, struct dentry *dentry, int mode) if (err) goto out_fail; - err = btrfs_add_link(trans, dentry->d_parent->d_inode, - inode, dentry->d_name.name, - dentry->d_name.len, 0, index); + err = btrfs_add_link(trans, dir, inode, dentry->d_name.name, + dentry->d_name.len, 0, index); if (err) goto out_fail; @@ -5535,13 +5534,21 @@ struct btrfs_dio_private { u64 bytes; u32 *csums; void *private; + + /* number of bios pending for this dio */ + atomic_t pending_bios; + + /* IO errors */ + int errors; + + struct bio *orig_bio; }; static void btrfs_endio_direct_read(struct bio *bio, int err) { + struct btrfs_dio_private *dip = bio->bi_private; struct bio_vec *bvec_end = bio->bi_io_vec + bio->bi_vcnt - 1; struct bio_vec *bvec = bio->bi_io_vec; - struct btrfs_dio_private *dip = bio->bi_private; struct inode *inode = dip->inode; struct btrfs_root *root = BTRFS_I(inode)->root; u64 start; @@ -5595,15 +5602,18 @@ static void btrfs_endio_direct_write(struct bio *bio, int err) struct btrfs_trans_handle *trans; struct btrfs_ordered_extent *ordered = NULL; struct extent_state *cached_state = NULL; + u64 ordered_offset = dip->logical_offset; + u64 ordered_bytes = dip->bytes; int ret; if (err) goto out_done; - - ret = btrfs_dec_test_ordered_pending(inode, &ordered, - dip->logical_offset, dip->bytes); +again: + ret = btrfs_dec_test_first_ordered_pending(inode, &ordered, + &ordered_offset, + ordered_bytes); if (!ret) - goto out_done; + goto out_test; BUG_ON(!ordered); @@ -5663,8 +5673,20 @@ out_unlock: out: btrfs_delalloc_release_metadata(inode, ordered->len); btrfs_end_transaction(trans, root); + ordered_offset = ordered->file_offset + ordered->len; btrfs_put_ordered_extent(ordered); btrfs_put_ordered_extent(ordered); + +out_test: + /* + * our bio might span multiple ordered extents. If we haven't + * completed the accounting for the whole dio, go back and try again + */ + if (ordered_offset < dip->logical_offset + dip->bytes) { + ordered_bytes = dip->logical_offset + dip->bytes - + ordered_offset; + goto again; + } out_done: bio->bi_private = dip->private; @@ -5684,6 +5706,176 @@ static int __btrfs_submit_bio_start_direct_io(struct inode *inode, int rw, return 0; } +static void btrfs_end_dio_bio(struct bio *bio, int err) +{ + struct btrfs_dio_private *dip = bio->bi_private; + + if (err) { + printk(KERN_ERR "btrfs direct IO failed ino %lu rw %lu " + "sector %#Lx len %u err no %d\n", + dip->inode->i_ino, bio->bi_rw, + (unsigned long long)bio->bi_sector, bio->bi_size, err); + dip->errors = 1; + + /* + * before atomic variable goto zero, we must make sure + * dip->errors is perceived to be set. + */ + smp_mb__before_atomic_dec(); + } + + /* if there are more bios still pending for this dio, just exit */ + if (!atomic_dec_and_test(&dip->pending_bios)) + goto out; + + if (dip->errors) + bio_io_error(dip->orig_bio); + else { + set_bit(BIO_UPTODATE, &dip->orig_bio->bi_flags); + bio_endio(dip->orig_bio, 0); + } +out: + bio_put(bio); +} + +static struct bio *btrfs_dio_bio_alloc(struct block_device *bdev, + u64 first_sector, gfp_t gfp_flags) +{ + int nr_vecs = bio_get_nr_vecs(bdev); + return btrfs_bio_alloc(bdev, first_sector, nr_vecs, gfp_flags); +} + +static inline int __btrfs_submit_dio_bio(struct bio *bio, struct inode *inode, + int rw, u64 file_offset, int skip_sum, + u32 *csums) +{ + int write = rw & REQ_WRITE; + struct btrfs_root *root = BTRFS_I(inode)->root; + int ret; + + bio_get(bio); + ret = btrfs_bio_wq_end_io(root->fs_info, bio, 0); + if (ret) + goto err; + + if (write && !skip_sum) { + ret = btrfs_wq_submit_bio(root->fs_info, + inode, rw, bio, 0, 0, + file_offset, + __btrfs_submit_bio_start_direct_io, + __btrfs_submit_bio_done); + goto err; + } else if (!skip_sum) + btrfs_lookup_bio_sums_dio(root, inode, bio, + file_offset, csums); + + ret = btrfs_map_bio(root, rw, bio, 0, 1); +err: + bio_put(bio); + return ret; +} + +static int btrfs_submit_direct_hook(int rw, struct btrfs_dio_private *dip, + int skip_sum) +{ + struct inode *inode = dip->inode; + struct btrfs_root *root = BTRFS_I(inode)->root; + struct btrfs_mapping_tree *map_tree = &root->fs_info->mapping_tree; + struct bio *bio; + struct bio *orig_bio = dip->orig_bio; + struct bio_vec *bvec = orig_bio->bi_io_vec; + u64 start_sector = orig_bio->bi_sector; + u64 file_offset = dip->logical_offset; + u64 submit_len = 0; + u64 map_length; + int nr_pages = 0; + u32 *csums = dip->csums; + int ret = 0; + + bio = btrfs_dio_bio_alloc(orig_bio->bi_bdev, start_sector, GFP_NOFS); + if (!bio) + return -ENOMEM; + bio->bi_private = dip; + bio->bi_end_io = btrfs_end_dio_bio; + atomic_inc(&dip->pending_bios); + + map_length = orig_bio->bi_size; + ret = btrfs_map_block(map_tree, READ, start_sector << 9, + &map_length, NULL, 0); + if (ret) { + bio_put(bio); + return -EIO; + } + + while (bvec <= (orig_bio->bi_io_vec + orig_bio->bi_vcnt - 1)) { + if (unlikely(map_length < submit_len + bvec->bv_len || + bio_add_page(bio, bvec->bv_page, bvec->bv_len, + bvec->bv_offset) < bvec->bv_len)) { + /* + * inc the count before we submit the bio so + * we know the end IO handler won't happen before + * we inc the count. Otherwise, the dip might get freed + * before we're done setting it up + */ + atomic_inc(&dip->pending_bios); + ret = __btrfs_submit_dio_bio(bio, inode, rw, + file_offset, skip_sum, + csums); + if (ret) { + bio_put(bio); + atomic_dec(&dip->pending_bios); + goto out_err; + } + + if (!skip_sum) + csums = csums + nr_pages; + start_sector += submit_len >> 9; + file_offset += submit_len; + + submit_len = 0; + nr_pages = 0; + + bio = btrfs_dio_bio_alloc(orig_bio->bi_bdev, + start_sector, GFP_NOFS); + if (!bio) + goto out_err; + bio->bi_private = dip; + bio->bi_end_io = btrfs_end_dio_bio; + + map_length = orig_bio->bi_size; + ret = btrfs_map_block(map_tree, READ, start_sector << 9, + &map_length, NULL, 0); + if (ret) { + bio_put(bio); + goto out_err; + } + } else { + submit_len += bvec->bv_len; + nr_pages ++; + bvec++; + } + } + + ret = __btrfs_submit_dio_bio(bio, inode, rw, file_offset, skip_sum, + csums); + if (!ret) + return 0; + + bio_put(bio); +out_err: + dip->errors = 1; + /* + * before atomic variable goto zero, we must + * make sure dip->errors is perceived to be set. + */ + smp_mb__before_atomic_dec(); + if (atomic_dec_and_test(&dip->pending_bios)) + bio_io_error(dip->orig_bio); + + /* bio_end_io() will handle error, so we needn't return it */ + return 0; +} + static void btrfs_submit_direct(int rw, struct bio *bio, struct inode *inode, loff_t file_offset) { @@ -5723,36 +5915,18 @@ static void btrfs_submit_direct(int rw, struct bio *bio, struct inode *inode, dip->disk_bytenr = (u64)bio->bi_sector << 9; bio->bi_private = dip; + dip->errors = 0; + dip->orig_bio = bio; + atomic_set(&dip->pending_bios, 0); if (write) bio->bi_end_io = btrfs_endio_direct_write; else bio->bi_end_io = btrfs_endio_direct_read; - ret = btrfs_bio_wq_end_io(root->fs_info, bio, 0); - if (ret) - goto out_err; - - if (write && !skip_sum) { - ret = btrfs_wq_submit_bio(BTRFS_I(inode)->root->fs_info, - inode, rw, bio, 0, 0, - dip->logical_offset, - __btrfs_submit_bio_start_direct_io, - __btrfs_submit_bio_done); - if (ret) - goto out_err; + ret = btrfs_submit_direct_hook(rw, dip, skip_sum); + if (!ret) return; - } else if (!skip_sum) - btrfs_lookup_bio_sums_dio(root, inode, bio, - dip->logical_offset, dip->csums); - - ret = btrfs_map_bio(root, rw, bio, 0, 1); - if (ret) - goto out_err; - return; -out_err: - kfree(dip->csums); - kfree(dip); free_ordered: /* * If this is a write, we need to clean up the reserved space and kill @@ -5760,8 +5934,7 @@ free_ordered: */ if (write) { struct btrfs_ordered_extent *ordered; - ordered = btrfs_lookup_ordered_extent(inode, - dip->logical_offset); + ordered = btrfs_lookup_ordered_extent(inode, file_offset); if (!test_bit(BTRFS_ORDERED_PREALLOC, &ordered->flags) && !test_bit(BTRFS_ORDERED_NOCOW, &ordered->flags)) btrfs_free_reserved_extent(root, ordered->start, @@ -6322,6 +6495,13 @@ struct inode *btrfs_alloc_inode(struct super_block *sb) return inode; } +static void btrfs_i_callback(struct rcu_head *head) +{ + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); + kmem_cache_free(btrfs_inode_cachep, BTRFS_I(inode)); +} + void btrfs_destroy_inode(struct inode *inode) { struct btrfs_ordered_extent *ordered; @@ -6391,7 +6571,7 @@ void btrfs_destroy_inode(struct inode *inode) inode_tree_del(inode); btrfs_drop_extent_cache(inode, 0, (u64)-1, 0); free: - kmem_cache_free(btrfs_inode_cachep, BTRFS_I(inode)); + call_rcu(&inode->i_rcu, btrfs_i_callback); } int btrfs_drop_inode(struct inode *inode) @@ -6607,8 +6787,9 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry, BUG_ON(ret); if (old_inode->i_ino != BTRFS_FIRST_FREE_OBJECTID) { - btrfs_log_new_name(trans, old_inode, old_dir, - new_dentry->d_parent); + struct dentry *parent = dget_parent(new_dentry); + btrfs_log_new_name(trans, old_inode, old_dir, parent); + dput(parent); btrfs_end_log_trans(root); } out_fail: @@ -6758,8 +6939,7 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry, btrfs_set_trans_block_group(trans, dir); inode = btrfs_new_inode(trans, root, dir, dentry->d_name.name, - dentry->d_name.len, - dentry->d_parent->d_inode->i_ino, objectid, + dentry->d_name.len, dir->i_ino, objectid, BTRFS_I(dir)->block_group, S_IFLNK|S_IRWXUGO, &index); err = PTR_ERR(inode); @@ -6773,7 +6953,7 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry, } btrfs_set_trans_block_group(trans, inode); - err = btrfs_add_nondir(trans, dentry, inode, 0, index); + err = btrfs_add_nondir(trans, dir, dentry, inode, 0, index); if (err) drop_inode = 1; else { @@ -6844,6 +7024,7 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode, struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_key ins; u64 cur_offset = start; + u64 i_size; int ret = 0; bool own_trans = true; @@ -6885,11 +7066,11 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode, (actual_len > inode->i_size) && (cur_offset > inode->i_size)) { if (cur_offset > actual_len) - i_size_write(inode, actual_len); + i_size = actual_len; else - i_size_write(inode, cur_offset); - i_size_write(inode, cur_offset); - btrfs_ordered_update_i_size(inode, cur_offset, NULL); + i_size = cur_offset; + i_size_write(inode, i_size); + btrfs_ordered_update_i_size(inode, i_size, NULL); } ret = btrfs_update_inode(trans, root, inode); @@ -6943,6 +7124,10 @@ static long btrfs_fallocate(struct inode *inode, int mode, btrfs_wait_ordered_range(inode, alloc_start, alloc_end - alloc_start); mutex_lock(&inode->i_mutex); + ret = inode_newsize_ok(inode, alloc_end); + if (ret) + goto out; + if (alloc_start > inode->i_size) { ret = btrfs_cont_expand(inode, alloc_start); if (ret) @@ -7026,11 +7211,11 @@ static int btrfs_set_page_dirty(struct page *page) return __set_page_dirty_nobuffers(page); } -static int btrfs_permission(struct inode *inode, int mask) +static int btrfs_permission(struct inode *inode, int mask, unsigned int flags) { if ((BTRFS_I(inode)->flags & BTRFS_INODE_READONLY) && (mask & MAY_WRITE)) return -EACCES; - return generic_permission(inode, mask, btrfs_check_acl); + return generic_permission(inode, mask, flags, btrfs_check_acl); } static const struct inode_operations btrfs_dir_inode_operations = { @@ -7139,6 +7324,7 @@ static const struct inode_operations btrfs_symlink_inode_operations = { .readlink = generic_readlink, .follow_link = page_follow_link_light, .put_link = page_put_link, + .getattr = btrfs_getattr, .permission = btrfs_permission, .setxattr = btrfs_setxattr, .getxattr = btrfs_getxattr, diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 463d91b4dd3..f87552a1d7e 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -233,7 +233,8 @@ static noinline int create_subvol(struct btrfs_root *root, struct btrfs_inode_item *inode_item; struct extent_buffer *leaf; struct btrfs_root *new_root; - struct inode *dir = dentry->d_parent->d_inode; + struct dentry *parent = dget_parent(dentry); + struct inode *dir; int ret; int err; u64 objectid; @@ -242,8 +243,13 @@ static noinline int create_subvol(struct btrfs_root *root, ret = btrfs_find_free_objectid(NULL, root->fs_info->tree_root, 0, &objectid); - if (ret) + if (ret) { + dput(parent); return ret; + } + + dir = parent->d_inode; + /* * 1 - inode item * 2 - refs @@ -251,8 +257,10 @@ static noinline int create_subvol(struct btrfs_root *root, * 2 - dir items */ trans = btrfs_start_transaction(root, 6); - if (IS_ERR(trans)) + if (IS_ERR(trans)) { + dput(parent); return PTR_ERR(trans); + } leaf = btrfs_alloc_free_block(trans, root, root->leafsize, 0, objectid, NULL, 0, 0, 0); @@ -339,6 +347,7 @@ static noinline int create_subvol(struct btrfs_root *root, d_instantiate(dentry, btrfs_lookup_dentry(dir, dentry)); fail: + dput(parent); if (async_transid) { *async_transid = trans->transid; err = btrfs_commit_transaction_async(trans, root, 1); @@ -354,6 +363,7 @@ static int create_snapshot(struct btrfs_root *root, struct dentry *dentry, char *name, int namelen, u64 *async_transid) { struct inode *inode; + struct dentry *parent; struct btrfs_pending_snapshot *pending_snapshot; struct btrfs_trans_handle *trans; int ret; @@ -396,7 +406,9 @@ static int create_snapshot(struct btrfs_root *root, struct dentry *dentry, btrfs_orphan_cleanup(pending_snapshot->snap); - inode = btrfs_lookup_dentry(dentry->d_parent->d_inode, dentry); + parent = dget_parent(dentry); + inode = btrfs_lookup_dentry(parent->d_inode, dentry); + dput(parent); if (IS_ERR(inode)) { ret = PTR_ERR(inode); goto fail; @@ -935,23 +947,42 @@ out: static noinline int btrfs_ioctl_snap_create(struct file *file, void __user *arg, int subvol, - int async) + int v2) { struct btrfs_ioctl_vol_args *vol_args = NULL; - struct btrfs_ioctl_async_vol_args *async_vol_args = NULL; + struct btrfs_ioctl_vol_args_v2 *vol_args_v2 = NULL; char *name; u64 fd; - u64 transid = 0; int ret; - if (async) { - async_vol_args = memdup_user(arg, sizeof(*async_vol_args)); - if (IS_ERR(async_vol_args)) - return PTR_ERR(async_vol_args); + if (v2) { + u64 transid = 0; + u64 *ptr = NULL; + + vol_args_v2 = memdup_user(arg, sizeof(*vol_args_v2)); + if (IS_ERR(vol_args_v2)) + return PTR_ERR(vol_args_v2); + + if (vol_args_v2->flags & ~BTRFS_SUBVOL_CREATE_ASYNC) { + ret = -EINVAL; + goto out; + } + + name = vol_args_v2->name; + fd = vol_args_v2->fd; + vol_args_v2->name[BTRFS_SUBVOL_NAME_MAX] = '\0'; + + if (vol_args_v2->flags & BTRFS_SUBVOL_CREATE_ASYNC) + ptr = &transid; + + ret = btrfs_ioctl_snap_create_transid(file, name, fd, + subvol, ptr); - name = async_vol_args->name; - fd = async_vol_args->fd; - async_vol_args->name[BTRFS_SNAPSHOT_NAME_MAX] = '\0'; + if (ret == 0 && ptr && + copy_to_user(arg + + offsetof(struct btrfs_ioctl_vol_args_v2, + transid), ptr, sizeof(*ptr))) + ret = -EFAULT; } else { vol_args = memdup_user(arg, sizeof(*vol_args)); if (IS_ERR(vol_args)) @@ -959,20 +990,13 @@ static noinline int btrfs_ioctl_snap_create(struct file *file, name = vol_args->name; fd = vol_args->fd; vol_args->name[BTRFS_PATH_NAME_MAX] = '\0'; - } - - ret = btrfs_ioctl_snap_create_transid(file, name, fd, - subvol, &transid); - if (!ret && async) { - if (copy_to_user(arg + - offsetof(struct btrfs_ioctl_async_vol_args, - transid), &transid, sizeof(transid))) - return -EFAULT; + ret = btrfs_ioctl_snap_create_transid(file, name, fd, + subvol, NULL); } - +out: kfree(vol_args); - kfree(async_vol_args); + kfree(vol_args_v2); return ret; } @@ -1669,12 +1693,11 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd, olen = len = src->i_size - off; /* if we extend to eof, continue to block boundary */ if (off + len == src->i_size) - len = ((src->i_size + bs-1) & ~(bs-1)) - - off; + len = ALIGN(src->i_size, bs) - off; /* verify the end result is block aligned */ - if ((off & (bs-1)) || - ((off + len) & (bs-1))) + if (!IS_ALIGNED(off, bs) || !IS_ALIGNED(off + len, bs) || + !IS_ALIGNED(destoff, bs)) goto out_unlock; /* do any pending delalloc/csum calc on src, one way or @@ -1874,8 +1897,8 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd, * but shouldn't round up the file size */ endoff = new_key.offset + datal; - if (endoff > off+olen) - endoff = off+olen; + if (endoff > destoff+olen) + endoff = destoff+olen; if (endoff > inode->i_size) btrfs_i_size_write(inode, endoff); @@ -2235,7 +2258,7 @@ long btrfs_ioctl(struct file *file, unsigned int return btrfs_ioctl_getversion(file, argp); case BTRFS_IOC_SNAP_CREATE: return btrfs_ioctl_snap_create(file, argp, 0, 0); - case BTRFS_IOC_SNAP_CREATE_ASYNC: + case BTRFS_IOC_SNAP_CREATE_V2: return btrfs_ioctl_snap_create(file, argp, 0, 1); case BTRFS_IOC_SUBVOL_CREATE: return btrfs_ioctl_snap_create(file, argp, 1, 0); diff --git a/fs/btrfs/ioctl.h b/fs/btrfs/ioctl.h index 17c99ebdf96..c344d12c646 100644 --- a/fs/btrfs/ioctl.h +++ b/fs/btrfs/ioctl.h @@ -30,11 +30,15 @@ struct btrfs_ioctl_vol_args { char name[BTRFS_PATH_NAME_MAX + 1]; }; -#define BTRFS_SNAPSHOT_NAME_MAX 4079 -struct btrfs_ioctl_async_vol_args { +#define BTRFS_SUBVOL_CREATE_ASYNC (1ULL << 0) + +#define BTRFS_SUBVOL_NAME_MAX 4039 +struct btrfs_ioctl_vol_args_v2 { __s64 fd; __u64 transid; - char name[BTRFS_SNAPSHOT_NAME_MAX + 1]; + __u64 flags; + __u64 unused[4]; + char name[BTRFS_SUBVOL_NAME_MAX + 1]; }; #define BTRFS_INO_LOOKUP_PATH_MAX 4080 @@ -187,6 +191,6 @@ struct btrfs_ioctl_space_args { struct btrfs_ioctl_space_args) #define BTRFS_IOC_START_SYNC _IOR(BTRFS_IOCTL_MAGIC, 24, __u64) #define BTRFS_IOC_WAIT_SYNC _IOW(BTRFS_IOCTL_MAGIC, 22, __u64) -#define BTRFS_IOC_SNAP_CREATE_ASYNC _IOW(BTRFS_IOCTL_MAGIC, 23, \ - struct btrfs_ioctl_async_vol_args) +#define BTRFS_IOC_SNAP_CREATE_V2 _IOW(BTRFS_IOCTL_MAGIC, 23, \ + struct btrfs_ioctl_vol_args_v2) #endif diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c index f4621f6deca..ae7737e352c 100644 --- a/fs/btrfs/ordered-data.c +++ b/fs/btrfs/ordered-data.c @@ -250,6 +250,73 @@ int btrfs_add_ordered_sum(struct inode *inode, /* * this is used to account for finished IO across a given range + * of the file. The IO may span ordered extents. If + * a given ordered_extent is completely done, 1 is returned, otherwise + * 0. + * + * test_and_set_bit on a flag in the struct btrfs_ordered_extent is used + * to make sure this function only returns 1 once for a given ordered extent. + * + * file_offset is updated to one byte past the range that is recorded as + * complete. This allows you to walk forward in the file. + */ +int btrfs_dec_test_first_ordered_pending(struct inode *inode, + struct btrfs_ordered_extent **cached, + u64 *file_offset, u64 io_size) +{ + struct btrfs_ordered_inode_tree *tree; + struct rb_node *node; + struct btrfs_ordered_extent *entry = NULL; + int ret; + u64 dec_end; + u64 dec_start; + u64 to_dec; + + tree = &BTRFS_I(inode)->ordered_tree; + spin_lock(&tree->lock); + node = tree_search(tree, *file_offset); + if (!node) { + ret = 1; + goto out; + } + + entry = rb_entry(node, struct btrfs_ordered_extent, rb_node); + if (!offset_in_entry(entry, *file_offset)) { + ret = 1; + goto out; + } + + dec_start = max(*file_offset, entry->file_offset); + dec_end = min(*file_offset + io_size, entry->file_offset + + entry->len); + *file_offset = dec_end; + if (dec_start > dec_end) { + printk(KERN_CRIT "bad ordering dec_start %llu end %llu\n", + (unsigned long long)dec_start, + (unsigned long long)dec_end); + } + to_dec = dec_end - dec_start; + if (to_dec > entry->bytes_left) { + printk(KERN_CRIT "bad ordered accounting left %llu size %llu\n", + (unsigned long long)entry->bytes_left, + (unsigned long long)to_dec); + } + entry->bytes_left -= to_dec; + if (entry->bytes_left == 0) + ret = test_and_set_bit(BTRFS_ORDERED_IO_DONE, &entry->flags); + else + ret = 1; +out: + if (!ret && cached && entry) { + *cached = entry; + atomic_inc(&entry->refs); + } + spin_unlock(&tree->lock); + return ret == 0; +} + +/* + * this is used to account for finished IO across a given range * of the file. The IO should not span ordered extents. If * a given ordered_extent is completely done, 1 is returned, otherwise * 0. diff --git a/fs/btrfs/ordered-data.h b/fs/btrfs/ordered-data.h index 8ac365492a3..61dca83119d 100644 --- a/fs/btrfs/ordered-data.h +++ b/fs/btrfs/ordered-data.h @@ -141,6 +141,9 @@ int btrfs_remove_ordered_extent(struct inode *inode, int btrfs_dec_test_ordered_pending(struct inode *inode, struct btrfs_ordered_extent **cached, u64 file_offset, u64 io_size); +int btrfs_dec_test_first_ordered_pending(struct inode *inode, + struct btrfs_ordered_extent **cached, + u64 *file_offset, u64 io_size); int btrfs_add_ordered_extent(struct inode *inode, u64 file_offset, u64 start, u64 len, u64 disk_len, int type); int btrfs_add_ordered_extent_dio(struct inode *inode, u64 file_offset, diff --git a/fs/btrfs/orphan.c b/fs/btrfs/orphan.c index 79cba5fbc28..f8be250963a 100644 --- a/fs/btrfs/orphan.c +++ b/fs/btrfs/orphan.c @@ -56,8 +56,12 @@ int btrfs_del_orphan_item(struct btrfs_trans_handle *trans, return -ENOMEM; ret = btrfs_search_slot(trans, root, &key, path, -1, 1); - if (ret) + if (ret < 0) goto out; + if (ret) { + ret = -ENOENT; + goto out; + } ret = btrfs_del_item(trans, root, path); diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index 8299a25ffc8..883c6fa1367 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -244,6 +244,7 @@ int btrfs_parse_options(struct btrfs_root *root, char *options) case Opt_space_cache: printk(KERN_INFO "btrfs: enabling disk space caching\n"); btrfs_set_opt(info->mount_opt, SPACE_CACHE); + break; case Opt_clear_cache: printk(KERN_INFO "btrfs: force clearing of disk cache\n"); btrfs_set_opt(info->mount_opt, CLEAR_CACHE); @@ -562,12 +563,26 @@ static int btrfs_show_options(struct seq_file *seq, struct vfsmount *vfs) static int btrfs_test_super(struct super_block *s, void *data) { - struct btrfs_fs_devices *test_fs_devices = data; + struct btrfs_root *test_root = data; struct btrfs_root *root = btrfs_sb(s); - return root->fs_info->fs_devices == test_fs_devices; + /* + * If this super block is going away, return false as it + * can't match as an existing super block. + */ + if (!atomic_read(&s->s_active)) + return 0; + return root->fs_info->fs_devices == test_root->fs_info->fs_devices; +} + +static int btrfs_set_super(struct super_block *s, void *data) +{ + s->s_fs_info = data; + + return set_anon_super(s, data); } + /* * Find a superblock for the given device / mount point. * @@ -581,6 +596,8 @@ static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags, struct super_block *s; struct dentry *root; struct btrfs_fs_devices *fs_devices = NULL; + struct btrfs_root *tree_root = NULL; + struct btrfs_fs_info *fs_info = NULL; fmode_t mode = FMODE_READ; char *subvol_name = NULL; u64 subvol_objectid = 0; @@ -608,8 +625,24 @@ static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags, goto error_close_devices; } + /* + * Setup a dummy root and fs_info for test/set super. This is because + * we don't actually fill this stuff out until open_ctree, but we need + * it for searching for existing supers, so this lets us do that and + * then open_ctree will properly initialize everything later. + */ + fs_info = kzalloc(sizeof(struct btrfs_fs_info), GFP_NOFS); + tree_root = kzalloc(sizeof(struct btrfs_root), GFP_NOFS); + if (!fs_info || !tree_root) { + error = -ENOMEM; + goto error_close_devices; + } + fs_info->tree_root = tree_root; + fs_info->fs_devices = fs_devices; + tree_root->fs_info = fs_info; + bdev = fs_devices->latest_bdev; - s = sget(fs_type, btrfs_test_super, set_anon_super, fs_devices); + s = sget(fs_type, btrfs_test_super, btrfs_set_super, tree_root); if (IS_ERR(s)) goto error_s; @@ -652,9 +685,9 @@ static struct dentry *btrfs_mount(struct file_system_type *fs_type, int flags, mutex_unlock(&root->d_inode->i_mutex); if (IS_ERR(new_root)) { + dput(root); deactivate_locked_super(s); error = PTR_ERR(new_root); - dput(root); goto error_free_subvol_name; } if (!new_root->d_inode) { @@ -675,6 +708,8 @@ error_s: error = PTR_ERR(s); error_close_devices: btrfs_close_devices(fs_devices); + kfree(fs_info); + kfree(tree_root); error_free_subvol_name: kfree(subvol_name); return ERR_PTR(error); diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index 1fffbc017bd..f50e931fc21 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -902,6 +902,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, struct btrfs_root *root = pending->root; struct btrfs_root *parent_root; struct inode *parent_inode; + struct dentry *parent; struct dentry *dentry; struct extent_buffer *tmp; struct extent_buffer *old; @@ -941,7 +942,8 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, trans->block_rsv = &pending->block_rsv; dentry = pending->dentry; - parent_inode = dentry->d_parent->d_inode; + parent = dget_parent(dentry); + parent_inode = parent->d_inode; parent_root = BTRFS_I(parent_inode)->root; record_root_in_trans(trans, parent_root); @@ -989,6 +991,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, parent_inode->i_ino, index, dentry->d_name.name, dentry->d_name.len); BUG_ON(ret); + dput(parent); key.offset = (u64)-1; pending->snap = btrfs_read_fs_root_no_name(root->fs_info, &key); diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index a29f19384a2..054744ac571 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -2869,6 +2869,7 @@ static noinline int check_parent_dirs_for_sync(struct btrfs_trans_handle *trans, { int ret = 0; struct btrfs_root *root; + struct dentry *old_parent = NULL; /* * for regular files, if its inode is already on disk, we don't @@ -2910,10 +2911,13 @@ static noinline int check_parent_dirs_for_sync(struct btrfs_trans_handle *trans, if (IS_ROOT(parent)) break; - parent = parent->d_parent; + parent = dget_parent(parent); + dput(old_parent); + old_parent = parent; inode = parent->d_inode; } + dput(old_parent); out: return ret; } @@ -2945,6 +2949,7 @@ int btrfs_log_inode_parent(struct btrfs_trans_handle *trans, { int inode_only = exists_only ? LOG_INODE_EXISTS : LOG_INODE_ALL; struct super_block *sb; + struct dentry *old_parent = NULL; int ret = 0; u64 last_committed = root->fs_info->last_trans_committed; @@ -3016,10 +3021,13 @@ int btrfs_log_inode_parent(struct btrfs_trans_handle *trans, if (IS_ROOT(parent)) break; - parent = parent->d_parent; + parent = dget_parent(parent); + dput(old_parent); + old_parent = parent; } ret = 0; end_trans: + dput(old_parent); if (ret < 0) { BUG_ON(ret != -ENOSPC); root->fs_info->last_trans_log_full_commit = trans->transid; @@ -3039,8 +3047,13 @@ end_no_trans: int btrfs_log_dentry_safe(struct btrfs_trans_handle *trans, struct btrfs_root *root, struct dentry *dentry) { - return btrfs_log_inode_parent(trans, root, dentry->d_inode, - dentry->d_parent, 0); + struct dentry *parent = dget_parent(dentry); + int ret; + + ret = btrfs_log_inode_parent(trans, root, dentry->d_inode, parent, 0); + dput(parent); + + return ret; } /* diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index cc04dc1445d..6b988450783 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -412,12 +412,16 @@ static noinline int device_list_add(const char *path, device->fs_devices = fs_devices; fs_devices->num_devices++; - } else if (strcmp(device->name, path)) { + } else if (!device->name || strcmp(device->name, path)) { name = kstrdup(path, GFP_NOFS); if (!name) return -ENOMEM; kfree(device->name); device->name = name; + if (device->missing) { + fs_devices->missing_devices--; + device->missing = 0; + } } if (found_transid > fs_devices->latest_trans) { @@ -1236,6 +1240,9 @@ int btrfs_rm_device(struct btrfs_root *root, char *device_path) device->fs_devices->num_devices--; + if (device->missing) + root->fs_info->fs_devices->missing_devices--; + next_device = list_entry(root->fs_info->fs_devices->devices.next, struct btrfs_device, dev_list); if (device->bdev == root->fs_info->sb->s_bdev) @@ -3080,7 +3087,9 @@ static struct btrfs_device *add_missing_dev(struct btrfs_root *root, device->devid = devid; device->work.func = pending_bios_fn; device->fs_devices = fs_devices; + device->missing = 1; fs_devices->num_devices++; + fs_devices->missing_devices++; spin_lock_init(&device->io_lock); INIT_LIST_HEAD(&device->dev_alloc_list); memcpy(device->uuid, dev_uuid, BTRFS_UUID_SIZE); @@ -3278,6 +3287,15 @@ static int read_one_dev(struct btrfs_root *root, device = add_missing_dev(root, devid, dev_uuid); if (!device) return -ENOMEM; + } else if (!device->missing) { + /* + * this happens when a device that was properly setup + * in the device info lists suddenly goes bad. + * device->bdev is NULL, and so we have to set + * device->missing to one here + */ + root->fs_info->fs_devices->missing_devices++; + device->missing = 1; } } diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h index 2b638b6e4ee..2740db49eb0 100644 --- a/fs/btrfs/volumes.h +++ b/fs/btrfs/volumes.h @@ -44,6 +44,7 @@ struct btrfs_device { int writeable; int in_fs_metadata; + int missing; spinlock_t io_lock; @@ -93,6 +94,7 @@ struct btrfs_fs_devices { u64 num_devices; u64 open_devices; u64 rw_devices; + u64 missing_devices; u64 total_rw_bytes; struct block_device *latest_bdev; diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index e9c874abc9e..561438b6a50 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -204,7 +204,7 @@ static int readpage_nounlock(struct file *filp, struct page *page) err = ceph_osdc_readpages(osdc, ceph_vino(inode), &ci->i_layout, page->index << PAGE_CACHE_SHIFT, &len, ci->i_truncate_seq, ci->i_truncate_size, - &page, 1); + &page, 1, 0); if (err == -ENOENT) err = 0; if (err < 0) { @@ -287,7 +287,7 @@ static int ceph_readpages(struct file *file, struct address_space *mapping, rc = ceph_osdc_readpages(osdc, ceph_vino(inode), &ci->i_layout, offset, &len, ci->i_truncate_seq, ci->i_truncate_size, - pages, nr_pages); + pages, nr_pages, 0); if (rc == -ENOENT) rc = 0; if (rc < 0) @@ -774,7 +774,7 @@ get_more_pages: snapc, do_sync, ci->i_truncate_seq, ci->i_truncate_size, - &inode->i_mtime, true, 1); + &inode->i_mtime, true, 1, 0); max_pages = req->r_num_pages; alloc_page_vec(fsc, req); diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c index 98ab13e2b71..60d27bc9eb8 100644 --- a/fs/ceph/caps.c +++ b/fs/ceph/caps.c @@ -1430,8 +1430,8 @@ static int try_nonblocking_invalidate(struct inode *inode) invalidating_gen == ci->i_rdcache_gen) { /* success. */ dout("try_nonblocking_invalidate %p success\n", inode); - ci->i_rdcache_gen = 0; - ci->i_rdcache_revoking = 0; + /* save any racing async invalidate some trouble */ + ci->i_rdcache_revoking = ci->i_rdcache_gen - 1; return 0; } dout("try_nonblocking_invalidate %p failed\n", inode); @@ -2273,8 +2273,7 @@ static void handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant, { struct ceph_inode_info *ci = ceph_inode(inode); int mds = session->s_mds; - unsigned seq = le32_to_cpu(grant->seq); - unsigned issue_seq = le32_to_cpu(grant->issue_seq); + int seq = le32_to_cpu(grant->seq); int newcaps = le32_to_cpu(grant->caps); int issued, implemented, used, wanted, dirty; u64 size = le64_to_cpu(grant->size); @@ -2286,8 +2285,8 @@ static void handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant, int revoked_rdcache = 0; int queue_invalidate = 0; - dout("handle_cap_grant inode %p cap %p mds%d seq %u/%u %s\n", - inode, cap, mds, seq, issue_seq, ceph_cap_string(newcaps)); + dout("handle_cap_grant inode %p cap %p mds%d seq %d %s\n", + inode, cap, mds, seq, ceph_cap_string(newcaps)); dout(" size %llu max_size %llu, i_size %llu\n", size, max_size, inode->i_size); @@ -2383,7 +2382,6 @@ static void handle_cap_grant(struct inode *inode, struct ceph_mds_caps *grant, } cap->seq = seq; - cap->issue_seq = issue_seq; /* file layout may have changed */ ci->i_layout = grant->layout; @@ -2691,6 +2689,11 @@ static void handle_cap_import(struct ceph_mds_client *mdsc, NULL /* no caps context */); try_flush_caps(inode, session, NULL); up_read(&mdsc->snap_rwsem); + + /* make sure we re-request max_size, if necessary */ + spin_lock(&inode->i_lock); + ci->i_requested_max_size = 0; + spin_unlock(&inode->i_lock); } /* diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c index e0a2dc6fcaf..fa7ca04ee81 100644 --- a/fs/ceph/dir.c +++ b/fs/ceph/dir.c @@ -40,12 +40,13 @@ int ceph_init_dentry(struct dentry *dentry) if (dentry->d_fsdata) return 0; - if (ceph_snap(dentry->d_parent->d_inode) == CEPH_NOSNAP) - dentry->d_op = &ceph_dentry_ops; + if (dentry->d_parent == NULL || /* nfs fh_to_dentry */ + ceph_snap(dentry->d_parent->d_inode) == CEPH_NOSNAP) + d_set_d_op(dentry, &ceph_dentry_ops); else if (ceph_snap(dentry->d_parent->d_inode) == CEPH_SNAPDIR) - dentry->d_op = &ceph_snapdir_dentry_ops; + d_set_d_op(dentry, &ceph_snapdir_dentry_ops); else - dentry->d_op = &ceph_snap_dentry_ops; + d_set_d_op(dentry, &ceph_snap_dentry_ops); di = kmem_cache_alloc(ceph_dentry_cachep, GFP_NOFS | __GFP_ZERO); if (!di) @@ -111,11 +112,11 @@ static int __dcache_readdir(struct file *filp, dout("__dcache_readdir %p at %llu (last %p)\n", dir, filp->f_pos, last); - spin_lock(&dcache_lock); + spin_lock(&parent->d_lock); /* start at beginning? */ - if (filp->f_pos == 2 || (last && - filp->f_pos < ceph_dentry(last)->offset)) { + if (filp->f_pos == 2 || last == NULL || + filp->f_pos < ceph_dentry(last)->offset) { if (list_empty(&parent->d_subdirs)) goto out_unlock; p = parent->d_subdirs.prev; @@ -135,6 +136,7 @@ more: fi->at_end = 1; goto out_unlock; } + spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); if (!d_unhashed(dentry) && dentry->d_inode && ceph_snap(dentry->d_inode) != CEPH_SNAPDIR && ceph_ino(dentry->d_inode) != CEPH_INO_CEPH && @@ -144,13 +146,15 @@ more: dentry->d_name.len, dentry->d_name.name, di->offset, filp->f_pos, d_unhashed(dentry) ? " unhashed" : "", !dentry->d_inode ? " null" : ""); + spin_unlock(&dentry->d_lock); p = p->prev; dentry = list_entry(p, struct dentry, d_u.d_child); di = ceph_dentry(dentry); } - atomic_inc(&dentry->d_count); - spin_unlock(&dcache_lock); + dget_dlock(dentry); + spin_unlock(&dentry->d_lock); + spin_unlock(&parent->d_lock); dout(" %llu (%llu) dentry %p %.*s %p\n", di->offset, filp->f_pos, dentry, dentry->d_name.len, dentry->d_name.name, dentry->d_inode); @@ -176,19 +180,19 @@ more: filp->f_pos++; - /* make sure a dentry wasn't dropped while we didn't have dcache_lock */ + /* make sure a dentry wasn't dropped while we didn't have parent lock */ if (!ceph_i_test(dir, CEPH_I_COMPLETE)) { dout(" lost I_COMPLETE on %p; falling back to mds\n", dir); err = -EAGAIN; goto out; } - spin_lock(&dcache_lock); + spin_lock(&parent->d_lock); p = p->prev; /* advance to next dentry */ goto more; out_unlock: - spin_unlock(&dcache_lock); + spin_unlock(&parent->d_lock); out: if (last) dput(last); @@ -336,7 +340,10 @@ more: if (req->r_reply_info.dir_end) { kfree(fi->last_name); fi->last_name = NULL; - fi->next_offset = 2; + if (ceph_frag_is_rightmost(frag)) + fi->next_offset = 2; + else + fi->next_offset = 0; } else { rinfo = &req->r_reply_info; err = note_last_dentry(fi, @@ -355,18 +362,22 @@ more: u64 pos = ceph_make_fpos(frag, off); struct ceph_mds_reply_inode *in = rinfo->dir_in[off - fi->offset].in; + struct ceph_vino vino; + ino_t ino; + dout("readdir off %d (%d/%d) -> %lld '%.*s' %p\n", off, off - fi->offset, rinfo->dir_nr, pos, rinfo->dir_dname_len[off - fi->offset], rinfo->dir_dname[off - fi->offset], in); BUG_ON(!in); ftype = le32_to_cpu(in->mode) >> 12; + vino.ino = le64_to_cpu(in->ino); + vino.snap = le64_to_cpu(in->snapid); + ino = ceph_vino_to_ino(vino); if (filldir(dirent, rinfo->dir_dname[off - fi->offset], rinfo->dir_dname_len[off - fi->offset], - pos, - le64_to_cpu(in->ino), - ftype) < 0) { + pos, ino, ftype) < 0) { dout("filldir stopping us...\n"); return 0; } @@ -414,6 +425,7 @@ static void reset_readdir(struct ceph_file_info *fi) fi->last_readdir = NULL; } kfree(fi->last_name); + fi->last_name = NULL; fi->next_offset = 2; /* compensate for . and .. */ if (fi->dentry) { dput(fi->dentry); @@ -978,7 +990,12 @@ static int dir_lease_is_valid(struct inode *dir, struct dentry *dentry) */ static int ceph_d_revalidate(struct dentry *dentry, struct nameidata *nd) { - struct inode *dir = dentry->d_parent->d_inode; + struct inode *dir; + + if (nd->flags & LOOKUP_RCU) + return -ECHILD; + + dir = dentry->d_parent->d_inode; dout("d_revalidate %p '%.*s' inode %p offset %lld\n", dentry, dentry->d_name.len, dentry->d_name.name, dentry->d_inode, diff --git a/fs/ceph/file.c b/fs/ceph/file.c index e77c28cf369..7d0e4a82d89 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -154,11 +154,13 @@ int ceph_open(struct inode *inode, struct file *file) } /* - * No need to block if we have any caps. Update wanted set + * No need to block if we have caps on the auth MDS (for + * write) or any MDS (for read). Update wanted set * asynchronously. */ spin_lock(&inode->i_lock); - if (__ceph_is_any_real_caps(ci)) { + if (__ceph_is_any_real_caps(ci) && + (((fmode & CEPH_FILE_MODE_WR) == 0) || ci->i_auth_cap)) { int mds_wanted = __ceph_caps_mds_wanted(ci); int issued = __ceph_caps_issued(ci, NULL); @@ -280,11 +282,13 @@ int ceph_release(struct inode *inode, struct file *file) static int striped_read(struct inode *inode, u64 off, u64 len, struct page **pages, int num_pages, - int *checkeof) + int *checkeof, bool align_to_pages, + unsigned long buf_align) { struct ceph_fs_client *fsc = ceph_inode_to_client(inode); struct ceph_inode_info *ci = ceph_inode(inode); u64 pos, this_len; + int io_align, page_align; int page_off = off & ~PAGE_CACHE_MASK; /* first byte's offset in page */ int left, pages_left; int read; @@ -300,14 +304,19 @@ static int striped_read(struct inode *inode, page_pos = pages; pages_left = num_pages; read = 0; + io_align = off & ~PAGE_MASK; more: + if (align_to_pages) + page_align = (pos - io_align + buf_align) & ~PAGE_MASK; + else + page_align = pos & ~PAGE_MASK; this_len = left; ret = ceph_osdc_readpages(&fsc->client->osdc, ceph_vino(inode), &ci->i_layout, pos, &this_len, ci->i_truncate_seq, ci->i_truncate_size, - page_pos, pages_left); + page_pos, pages_left, page_align); hit_stripe = this_len < left; was_short = ret >= 0 && ret < this_len; if (ret == -ENOENT) @@ -368,32 +377,34 @@ static ssize_t ceph_sync_read(struct file *file, char __user *data, struct inode *inode = file->f_dentry->d_inode; struct page **pages; u64 off = *poff; - int num_pages = calc_pages_for(off, len); - int ret; + int num_pages, ret; dout("sync_read on file %p %llu~%u %s\n", file, off, len, (file->f_flags & O_DIRECT) ? "O_DIRECT" : ""); if (file->f_flags & O_DIRECT) { - pages = ceph_get_direct_page_vector(data, num_pages, off, len); - - /* - * flush any page cache pages in this range. this - * will make concurrent normal and O_DIRECT io slow, - * but it will at least behave sensibly when they are - * in sequence. - */ + num_pages = calc_pages_for((unsigned long)data, len); + pages = ceph_get_direct_page_vector(data, num_pages, true); } else { + num_pages = calc_pages_for(off, len); pages = ceph_alloc_page_vector(num_pages, GFP_NOFS); } if (IS_ERR(pages)) return PTR_ERR(pages); + /* + * flush any page cache pages in this range. this + * will make concurrent normal and sync io slow, + * but it will at least behave sensibly when they are + * in sequence. + */ ret = filemap_write_and_wait(inode->i_mapping); if (ret < 0) goto done; - ret = striped_read(inode, off, len, pages, num_pages, checkeof); + ret = striped_read(inode, off, len, pages, num_pages, checkeof, + file->f_flags & O_DIRECT, + (unsigned long)data & ~PAGE_MASK); if (ret >= 0 && (file->f_flags & O_DIRECT) == 0) ret = ceph_copy_page_vector_to_user(pages, data, off, ret); @@ -402,7 +413,7 @@ static ssize_t ceph_sync_read(struct file *file, char __user *data, done: if (file->f_flags & O_DIRECT) - ceph_put_page_vector(pages, num_pages); + ceph_put_page_vector(pages, num_pages, true); else ceph_release_page_vector(pages, num_pages); dout("sync_read result %d\n", ret); @@ -448,6 +459,8 @@ static ssize_t ceph_sync_write(struct file *file, const char __user *data, int flags; int do_sync = 0; int check_caps = 0; + int page_align, io_align; + unsigned long buf_align; int ret; struct timespec mtime = CURRENT_TIME; @@ -462,6 +475,9 @@ static ssize_t ceph_sync_write(struct file *file, const char __user *data, else pos = *offset; + io_align = pos & ~PAGE_MASK; + buf_align = (unsigned long)data & ~PAGE_MASK; + ret = filemap_write_and_wait_range(inode->i_mapping, pos, pos + left); if (ret < 0) return ret; @@ -486,20 +502,27 @@ static ssize_t ceph_sync_write(struct file *file, const char __user *data, */ more: len = left; + if (file->f_flags & O_DIRECT) { + /* write from beginning of first page, regardless of + io alignment */ + page_align = (pos - io_align + buf_align) & ~PAGE_MASK; + num_pages = calc_pages_for((unsigned long)data, len); + } else { + page_align = pos & ~PAGE_MASK; + num_pages = calc_pages_for(pos, len); + } req = ceph_osdc_new_request(&fsc->client->osdc, &ci->i_layout, ceph_vino(inode), pos, &len, CEPH_OSD_OP_WRITE, flags, ci->i_snap_realm->cached_context, do_sync, ci->i_truncate_seq, ci->i_truncate_size, - &mtime, false, 2); + &mtime, false, 2, page_align); if (!req) return -ENOMEM; - num_pages = calc_pages_for(pos, len); - if (file->f_flags & O_DIRECT) { - pages = ceph_get_direct_page_vector(data, num_pages, pos, len); + pages = ceph_get_direct_page_vector(data, num_pages, false); if (IS_ERR(pages)) { ret = PTR_ERR(pages); goto out; @@ -549,7 +572,7 @@ more: } if (file->f_flags & O_DIRECT) - ceph_put_page_vector(pages, num_pages); + ceph_put_page_vector(pages, num_pages, false); else if (file->f_flags & O_SYNC) ceph_release_page_vector(pages, num_pages); diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c index 1d6a45b5a04..e61de4f7b99 100644 --- a/fs/ceph/inode.c +++ b/fs/ceph/inode.c @@ -2,7 +2,6 @@ #include <linux/module.h> #include <linux/fs.h> -#include <linux/smp_lock.h> #include <linux/slab.h> #include <linux/string.h> #include <linux/uaccess.h> @@ -369,6 +368,15 @@ struct inode *ceph_alloc_inode(struct super_block *sb) return &ci->vfs_inode; } +static void ceph_i_callback(struct rcu_head *head) +{ + struct inode *inode = container_of(head, struct inode, i_rcu); + struct ceph_inode_info *ci = ceph_inode(inode); + + INIT_LIST_HEAD(&inode->i_dentry); + kmem_cache_free(ceph_inode_cachep, ci); +} + void ceph_destroy_inode(struct inode *inode) { struct ceph_inode_info *ci = ceph_inode(inode); @@ -408,7 +416,7 @@ void ceph_destroy_inode(struct inode *inode) if (ci->i_xattrs.prealloc_blob) ceph_buffer_put(ci->i_xattrs.prealloc_blob); - kmem_cache_free(ceph_inode_cachep, ci); + call_rcu(&inode->i_rcu, ceph_i_callback); } @@ -471,7 +479,9 @@ void ceph_fill_file_time(struct inode *inode, int issued, if (issued & (CEPH_CAP_FILE_EXCL| CEPH_CAP_FILE_WR| - CEPH_CAP_FILE_BUFFER)) { + CEPH_CAP_FILE_BUFFER| + CEPH_CAP_AUTH_EXCL| + CEPH_CAP_XATTR_EXCL)) { if (timespec_compare(ctime, &inode->i_ctime) > 0) { dout("ctime %ld.%09ld -> %ld.%09ld inc w/ cap\n", inode->i_ctime.tv_sec, inode->i_ctime.tv_nsec, @@ -511,7 +521,7 @@ void ceph_fill_file_time(struct inode *inode, int issued, warn = 1; } } else { - /* we have no write caps; whatever the MDS says is true */ + /* we have no write|excl caps; whatever the MDS says is true */ if (ceph_seq_cmp(time_warp_seq, ci->i_time_warp_seq) >= 0) { inode->i_ctime = *ctime; inode->i_mtime = *mtime; @@ -567,12 +577,17 @@ static int fill_inode(struct inode *inode, /* * provided version will be odd if inode value is projected, - * even if stable. skip the update if we have a newer info - * (e.g., due to inode info racing form multiple MDSs), or if - * we are getting projected (unstable) inode info. + * even if stable. skip the update if we have newer stable + * info (ours>=theirs, e.g. due to racing mds replies), unless + * we are getting projected (unstable) info (in which case the + * version is odd, and we want ours>theirs). + * us them + * 2 2 skip + * 3 2 skip + * 3 3 update */ if (le64_to_cpu(info->version) > 0 && - (ci->i_version & ~1) > le64_to_cpu(info->version)) + (ci->i_version & ~1) >= le64_to_cpu(info->version)) goto no_change; issued = __ceph_caps_issued(ci, &implemented); @@ -606,7 +621,14 @@ static int fill_inode(struct inode *inode, le32_to_cpu(info->time_warp_seq), &ctime, &mtime, &atime); - ci->i_max_size = le64_to_cpu(info->max_size); + /* only update max_size on auth cap */ + if ((info->cap.flags & CEPH_CAP_FLAG_AUTH) && + ci->i_max_size != le64_to_cpu(info->max_size)) { + dout("max_size %lld -> %llu\n", ci->i_max_size, + le64_to_cpu(info->max_size)); + ci->i_max_size = le64_to_cpu(info->max_size); + } + ci->i_layout = info->layout; inode->i_blkbits = fls(le32_to_cpu(info->layout.fl_stripe_unit)) - 1; @@ -828,13 +850,13 @@ static void ceph_set_dentry_offset(struct dentry *dn) di->offset = ceph_inode(inode)->i_max_offset++; spin_unlock(&inode->i_lock); - spin_lock(&dcache_lock); - spin_lock(&dn->d_lock); + spin_lock(&dir->d_lock); + spin_lock_nested(&dn->d_lock, DENTRY_D_LOCK_NESTED); list_move(&dn->d_u.d_child, &dir->d_subdirs); dout("set_dentry_offset %p %lld (%p %p)\n", dn, di->offset, dn->d_u.d_child.prev, dn->d_u.d_child.next); spin_unlock(&dn->d_lock); - spin_unlock(&dcache_lock); + spin_unlock(&dir->d_lock); } /* @@ -866,8 +888,8 @@ static struct dentry *splice_dentry(struct dentry *dn, struct inode *in, } else if (realdn) { dout("dn %p (%d) spliced with %p (%d) " "inode %p ino %llx.%llx\n", - dn, atomic_read(&dn->d_count), - realdn, atomic_read(&realdn->d_count), + dn, dn->d_count, + realdn, realdn->d_count, realdn->d_inode, ceph_vinop(realdn->d_inode)); dput(dn); dn = realdn; @@ -1055,7 +1077,8 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req, ininfo = rinfo->targeti.in; vino.ino = le64_to_cpu(ininfo->ino); vino.snap = le64_to_cpu(ininfo->snapid); - if (!dn->d_inode) { + in = dn->d_inode; + if (!in) { in = ceph_get_inode(sb, vino); if (IS_ERR(in)) { pr_err("fill_trace bad get_inode " @@ -1217,11 +1240,11 @@ retry_lookup: goto retry_lookup; } else { /* reorder parent's d_subdirs */ - spin_lock(&dcache_lock); - spin_lock(&dn->d_lock); + spin_lock(&parent->d_lock); + spin_lock_nested(&dn->d_lock, DENTRY_D_LOCK_NESTED); list_move(&dn->d_u.d_child, &parent->d_subdirs); spin_unlock(&dn->d_lock); - spin_unlock(&dcache_lock); + spin_unlock(&parent->d_lock); } di = dn->d_fsdata; @@ -1386,11 +1409,8 @@ static void ceph_invalidate_work(struct work_struct *work) spin_lock(&inode->i_lock); dout("invalidate_pages %p gen %d revoking %d\n", inode, ci->i_rdcache_gen, ci->i_rdcache_revoking); - if (ci->i_rdcache_gen == 0 || - ci->i_rdcache_revoking != ci->i_rdcache_gen) { - BUG_ON(ci->i_rdcache_revoking > ci->i_rdcache_gen); + if (ci->i_rdcache_revoking != ci->i_rdcache_gen) { /* nevermind! */ - ci->i_rdcache_revoking = 0; spin_unlock(&inode->i_lock); goto out; } @@ -1400,15 +1420,16 @@ static void ceph_invalidate_work(struct work_struct *work) ceph_invalidate_nondirty_pages(inode->i_mapping); spin_lock(&inode->i_lock); - if (orig_gen == ci->i_rdcache_gen) { + if (orig_gen == ci->i_rdcache_gen && + orig_gen == ci->i_rdcache_revoking) { dout("invalidate_pages %p gen %d successful\n", inode, ci->i_rdcache_gen); - ci->i_rdcache_gen = 0; - ci->i_rdcache_revoking = 0; + ci->i_rdcache_revoking--; check = 1; } else { - dout("invalidate_pages %p gen %d raced, gen now %d\n", - inode, orig_gen, ci->i_rdcache_gen); + dout("invalidate_pages %p gen %d raced, now %d revoking %d\n", + inode, orig_gen, ci->i_rdcache_gen, + ci->i_rdcache_revoking); } spin_unlock(&inode->i_lock); @@ -1739,7 +1760,7 @@ int ceph_do_getattr(struct inode *inode, int mask) return 0; } - dout("do_getattr inode %p mask %s\n", inode, ceph_cap_string(mask)); + dout("do_getattr inode %p mask %s mode 0%o\n", inode, ceph_cap_string(mask), inode->i_mode); if (ceph_caps_issued_mask(ceph_inode(inode), mask, 1)) return 0; @@ -1760,12 +1781,17 @@ int ceph_do_getattr(struct inode *inode, int mask) * Check inode permissions. We verify we have a valid value for * the AUTH cap, then call the generic handler. */ -int ceph_permission(struct inode *inode, int mask) +int ceph_permission(struct inode *inode, int mask, unsigned int flags) { - int err = ceph_do_getattr(inode, CEPH_CAP_AUTH_SHARED); + int err; + + if (flags & IPERM_FLAG_RCU) + return -ECHILD; + + err = ceph_do_getattr(inode, CEPH_CAP_AUTH_SHARED); if (!err) - err = generic_permission(inode, mask, NULL); + err = generic_permission(inode, mask, flags, NULL); return err; } diff --git a/fs/ceph/ioctl.h b/fs/ceph/ioctl.h index a6ce54e94eb..52e8fd74d45 100644 --- a/fs/ceph/ioctl.h +++ b/fs/ceph/ioctl.h @@ -4,7 +4,7 @@ #include <linux/ioctl.h> #include <linux/types.h> -#define CEPH_IOCTL_MAGIC 0x98 +#define CEPH_IOCTL_MAGIC 0x97 /* just use u64 to align sanely on all archs */ struct ceph_ioctl_layout { diff --git a/fs/ceph/locks.c b/fs/ceph/locks.c index 40abde93c34..476b329867d 100644 --- a/fs/ceph/locks.c +++ b/fs/ceph/locks.c @@ -11,40 +11,68 @@ * Implement fcntl and flock locking functions. */ static int ceph_lock_message(u8 lock_type, u16 operation, struct file *file, - u64 pid, u64 pid_ns, - int cmd, u64 start, u64 length, u8 wait) + int cmd, u8 wait, struct file_lock *fl) { struct inode *inode = file->f_dentry->d_inode; struct ceph_mds_client *mdsc = ceph_sb_to_client(inode->i_sb)->mdsc; struct ceph_mds_request *req; int err; + u64 length = 0; req = ceph_mdsc_create_request(mdsc, operation, USE_AUTH_MDS); if (IS_ERR(req)) return PTR_ERR(req); req->r_inode = igrab(inode); + /* mds requires start and length rather than start and end */ + if (LLONG_MAX == fl->fl_end) + length = 0; + else + length = fl->fl_end - fl->fl_start + 1; + dout("ceph_lock_message: rule: %d, op: %d, pid: %llu, start: %llu, " "length: %llu, wait: %d, type`: %d", (int)lock_type, - (int)operation, pid, start, length, wait, cmd); + (int)operation, (u64)fl->fl_pid, fl->fl_start, + length, wait, fl->fl_type); + req->r_args.filelock_change.rule = lock_type; req->r_args.filelock_change.type = cmd; - req->r_args.filelock_change.pid = cpu_to_le64(pid); + req->r_args.filelock_change.pid = cpu_to_le64((u64)fl->fl_pid); /* This should be adjusted, but I'm not sure if namespaces actually get id numbers*/ req->r_args.filelock_change.pid_namespace = - cpu_to_le64((u64)pid_ns); - req->r_args.filelock_change.start = cpu_to_le64(start); + cpu_to_le64((u64)(unsigned long)fl->fl_nspid); + req->r_args.filelock_change.start = cpu_to_le64(fl->fl_start); req->r_args.filelock_change.length = cpu_to_le64(length); req->r_args.filelock_change.wait = wait; err = ceph_mdsc_do_request(mdsc, inode, req); + + if ( operation == CEPH_MDS_OP_GETFILELOCK){ + fl->fl_pid = le64_to_cpu(req->r_reply_info.filelock_reply->pid); + if (CEPH_LOCK_SHARED == req->r_reply_info.filelock_reply->type) + fl->fl_type = F_RDLCK; + else if (CEPH_LOCK_EXCL == req->r_reply_info.filelock_reply->type) + fl->fl_type = F_WRLCK; + else + fl->fl_type = F_UNLCK; + + fl->fl_start = le64_to_cpu(req->r_reply_info.filelock_reply->start); + length = le64_to_cpu(req->r_reply_info.filelock_reply->start) + + le64_to_cpu(req->r_reply_info.filelock_reply->length); + if (length >= 1) + fl->fl_end = length -1; + else + fl->fl_end = 0; + + } ceph_mdsc_put_request(req); dout("ceph_lock_message: rule: %d, op: %d, pid: %llu, start: %llu, " - "length: %llu, wait: %d, type`: %d err code %d", (int)lock_type, - (int)operation, pid, start, length, wait, cmd, err); + "length: %llu, wait: %d, type`: %d, err code %d", (int)lock_type, + (int)operation, (u64)fl->fl_pid, fl->fl_start, + length, wait, fl->fl_type, err); return err; } @@ -54,7 +82,6 @@ static int ceph_lock_message(u8 lock_type, u16 operation, struct file *file, */ int ceph_lock(struct file *file, int cmd, struct file_lock *fl) { - u64 length; u8 lock_cmd; int err; u8 wait = 0; @@ -76,29 +103,20 @@ int ceph_lock(struct file *file, int cmd, struct file_lock *fl) else lock_cmd = CEPH_LOCK_UNLOCK; - if (LLONG_MAX == fl->fl_end) - length = 0; - else - length = fl->fl_end - fl->fl_start + 1; - - err = ceph_lock_message(CEPH_LOCK_FCNTL, op, file, - (u64)fl->fl_pid, - (u64)(unsigned long)fl->fl_nspid, - lock_cmd, fl->fl_start, - length, wait); + err = ceph_lock_message(CEPH_LOCK_FCNTL, op, file, lock_cmd, wait, fl); if (!err) { - dout("mds locked, locking locally"); - err = posix_lock_file(file, fl, NULL); - if (err && (CEPH_MDS_OP_SETFILELOCK == op)) { - /* undo! This should only happen if the kernel detects - * local deadlock. */ - ceph_lock_message(CEPH_LOCK_FCNTL, op, file, - (u64)fl->fl_pid, - (u64)(unsigned long)fl->fl_nspid, - CEPH_LOCK_UNLOCK, fl->fl_start, - length, 0); - dout("got %d on posix_lock_file, undid lock", err); + if ( op != CEPH_MDS_OP_GETFILELOCK ){ + dout("mds locked, locking locally"); + err = posix_lock_file(file, fl, NULL); + if (err && (CEPH_MDS_OP_SETFILELOCK == op)) { + /* undo! This should only happen if the kernel detects + * local deadlock. */ + ceph_lock_message(CEPH_LOCK_FCNTL, op, file, + CEPH_LOCK_UNLOCK, 0, fl); + dout("got %d on posix_lock_file, undid lock", err); + } } + } else { dout("mds returned error code %d", err); } @@ -107,7 +125,6 @@ int ceph_lock(struct file *file, int cmd, struct file_lock *fl) int ceph_flock(struct file *file, int cmd, struct file_lock *fl) { - u64 length; u8 lock_cmd; int err; u8 wait = 1; @@ -127,26 +144,15 @@ int ceph_flock(struct file *file, int cmd, struct file_lock *fl) lock_cmd = CEPH_LOCK_EXCL; else lock_cmd = CEPH_LOCK_UNLOCK; - /* mds requires start and length rather than start and end */ - if (LLONG_MAX == fl->fl_end) - length = 0; - else - length = fl->fl_end - fl->fl_start + 1; err = ceph_lock_message(CEPH_LOCK_FLOCK, CEPH_MDS_OP_SETFILELOCK, - file, (u64)fl->fl_pid, - (u64)(unsigned long)fl->fl_nspid, - lock_cmd, fl->fl_start, - length, wait); + file, lock_cmd, wait, fl); if (!err) { err = flock_lock_file_wait(file, fl); if (err) { ceph_lock_message(CEPH_LOCK_FLOCK, CEPH_MDS_OP_SETFILELOCK, - file, (u64)fl->fl_pid, - (u64)(unsigned long)fl->fl_nspid, - CEPH_LOCK_UNLOCK, fl->fl_start, - length, 0); + file, CEPH_LOCK_UNLOCK, 0, fl); dout("got %d on flock_lock_file_wait, undid lock", err); } } else { diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c index 3142b15940c..a50fca1e03b 100644 --- a/fs/ceph/mds_client.c +++ b/fs/ceph/mds_client.c @@ -6,7 +6,6 @@ #include <linux/sched.h> #include <linux/debugfs.h> #include <linux/seq_file.h> -#include <linux/smp_lock.h> #include "super.h" #include "mds_client.h" @@ -203,6 +202,38 @@ out_bad: } /* + * parse fcntl F_GETLK results + */ +static int parse_reply_info_filelock(void **p, void *end, + struct ceph_mds_reply_info_parsed *info) +{ + if (*p + sizeof(*info->filelock_reply) > end) + goto bad; + + info->filelock_reply = *p; + *p += sizeof(*info->filelock_reply); + + if (unlikely(*p != end)) + goto bad; + return 0; + +bad: + return -EIO; +} + +/* + * parse extra results + */ +static int parse_reply_info_extra(void **p, void *end, + struct ceph_mds_reply_info_parsed *info) +{ + if (info->head->op == CEPH_MDS_OP_GETFILELOCK) + return parse_reply_info_filelock(p, end, info); + else + return parse_reply_info_dir(p, end, info); +} + +/* * parse entire mds reply */ static int parse_reply_info(struct ceph_msg *msg, @@ -224,10 +255,10 @@ static int parse_reply_info(struct ceph_msg *msg, goto out_bad; } - /* dir content */ + /* extra */ ceph_decode_32_safe(&p, end, len, bad); if (len > 0) { - err = parse_reply_info_dir(&p, p+len, info); + err = parse_reply_info_extra(&p, p+len, info); if (err < 0) goto out_bad; } @@ -529,6 +560,9 @@ static void __register_request(struct ceph_mds_client *mdsc, ceph_mdsc_get_request(req); __insert_request(mdsc, req); + req->r_uid = current_fsuid(); + req->r_gid = current_fsgid(); + if (dir) { struct ceph_inode_info *ci = ceph_inode(dir); @@ -1452,7 +1486,7 @@ retry: *base = ceph_ino(temp->d_inode); *plen = len; dout("build_path on %p %d built %llx '%.*s'\n", - dentry, atomic_read(&dentry->d_count), *base, len, path); + dentry, dentry->d_count, *base, len, path); return path; } @@ -1588,8 +1622,8 @@ static struct ceph_msg *create_request_message(struct ceph_mds_client *mdsc, head->mdsmap_epoch = cpu_to_le32(mdsc->mdsmap->m_epoch); head->op = cpu_to_le32(req->r_op); - head->caller_uid = cpu_to_le32(current_fsuid()); - head->caller_gid = cpu_to_le32(current_fsgid()); + head->caller_uid = cpu_to_le32(req->r_uid); + head->caller_gid = cpu_to_le32(req->r_gid); head->args = req->r_args; ceph_encode_filepath(&p, end, ino1, path1); @@ -2072,7 +2106,7 @@ static void handle_reply(struct ceph_mds_session *session, struct ceph_msg *msg) mutex_lock(&session->s_mutex); if (err < 0) { - pr_err("mdsc_handle_reply got corrupt reply mds%d\n", mds); + pr_err("mdsc_handle_reply got corrupt reply mds%d(tid:%lld)\n", mds, tid); ceph_msg_dump(msg); goto out_err; } @@ -2092,7 +2126,8 @@ static void handle_reply(struct ceph_mds_session *session, struct ceph_msg *msg) mutex_lock(&req->r_fill_mutex); err = ceph_fill_trace(mdsc->fsc->sb, req, req->r_session); if (err == 0) { - if (result == 0 && rinfo->dir_nr) + if (result == 0 && req->r_op != CEPH_MDS_OP_GETFILELOCK && + rinfo->dir_nr) ceph_readdir_prepopulate(req, req->r_session); ceph_unreserve_caps(mdsc, &req->r_caps_reservation); } diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h index d66d63c7235..aabe563b54d 100644 --- a/fs/ceph/mds_client.h +++ b/fs/ceph/mds_client.h @@ -42,26 +42,37 @@ struct ceph_mds_reply_info_in { }; /* - * parsed info about an mds reply, including information about the - * target inode and/or its parent directory and dentry, and directory - * contents (for readdir results). + * parsed info about an mds reply, including information about + * either: 1) the target inode and/or its parent directory and dentry, + * and directory contents (for readdir results), or + * 2) the file range lock info (for fcntl F_GETLK results). */ struct ceph_mds_reply_info_parsed { struct ceph_mds_reply_head *head; + /* trace */ struct ceph_mds_reply_info_in diri, targeti; struct ceph_mds_reply_dirfrag *dirfrag; char *dname; u32 dname_len; struct ceph_mds_reply_lease *dlease; - struct ceph_mds_reply_dirfrag *dir_dir; - int dir_nr; - char **dir_dname; - u32 *dir_dname_len; - struct ceph_mds_reply_lease **dir_dlease; - struct ceph_mds_reply_info_in *dir_in; - u8 dir_complete, dir_end; + /* extra */ + union { + /* for fcntl F_GETLK results */ + struct ceph_filelock *filelock_reply; + + /* for readdir results */ + struct { + struct ceph_mds_reply_dirfrag *dir_dir; + int dir_nr; + char **dir_dname; + u32 *dir_dname_len; + struct ceph_mds_reply_lease **dir_dlease; + struct ceph_mds_reply_info_in *dir_in; + u8 dir_complete, dir_end; + }; + }; /* encoded blob describing snapshot contexts for certain operations (e.g., open) */ @@ -170,6 +181,8 @@ struct ceph_mds_request { union ceph_mds_request_args r_args; int r_fmode; /* file mode, if expecting cap */ + uid_t r_uid; + gid_t r_gid; /* for choosing which mds to send this request to */ int r_direct_mode; diff --git a/fs/ceph/super.h b/fs/ceph/super.h index 1886294e12f..4553d8829ed 100644 --- a/fs/ceph/super.h +++ b/fs/ceph/super.h @@ -293,9 +293,7 @@ struct ceph_inode_info { int i_rd_ref, i_rdcache_ref, i_wr_ref; int i_wrbuffer_ref, i_wrbuffer_ref_head; u32 i_shared_gen; /* increment each time we get FILE_SHARED */ - u32 i_rdcache_gen; /* we increment this each time we get - FILE_CACHE. If it's non-zero, we - _may_ have cached pages. */ + u32 i_rdcache_gen; /* incremented each time we get FILE_CACHE. */ u32 i_rdcache_revoking; /* RDCACHE gen to async invalidate, if any */ struct list_head i_unsafe_writes; /* uncommitted sync writes */ @@ -667,7 +665,7 @@ extern void ceph_queue_invalidate(struct inode *inode); extern void ceph_queue_writeback(struct inode *inode); extern int ceph_do_getattr(struct inode *inode, int mask); -extern int ceph_permission(struct inode *inode, int mask); +extern int ceph_permission(struct inode *inode, int mask, unsigned int flags); extern int ceph_setattr(struct dentry *dentry, struct iattr *attr); extern int ceph_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat); diff --git a/fs/cifs/Kconfig b/fs/cifs/Kconfig index 0ed213970ce..ee45648b0d1 100644 --- a/fs/cifs/Kconfig +++ b/fs/cifs/Kconfig @@ -4,6 +4,7 @@ config CIFS select NLS select CRYPTO select CRYPTO_MD5 + select CRYPTO_HMAC select CRYPTO_ARC4 help This is the client VFS module for the Common Internet File System @@ -143,6 +144,13 @@ config CIFS_FSCACHE to be cached locally on disk through the general filesystem cache manager. If unsure, say N. +config CIFS_ACL + bool "Provide CIFS ACL support (EXPERIMENTAL)" + depends on EXPERIMENTAL && CIFS_XATTR + help + Allows to fetch CIFS/NTFS ACL from the server. The DACL blob + is handed over to the application/caller. + config CIFS_EXPERIMENTAL bool "CIFS Experimental Features (EXPERIMENTAL)" depends on CIFS && EXPERIMENTAL diff --git a/fs/cifs/Makefile b/fs/cifs/Makefile index adefa60a9bd..43b19dd3919 100644 --- a/fs/cifs/Makefile +++ b/fs/cifs/Makefile @@ -6,7 +6,9 @@ obj-$(CONFIG_CIFS) += cifs.o cifs-y := cifsfs.o cifssmb.o cifs_debug.o connect.o dir.o file.o inode.o \ link.o misc.o netmisc.o smbdes.o smbencrypt.o transport.o asn1.o \ md4.o md5.o cifs_unicode.o nterr.o xattr.o cifsencrypt.o \ - readdir.o ioctl.o sess.o export.o cifsacl.o + readdir.o ioctl.o sess.o export.o + +cifs-$(CONFIG_CIFS_ACL) += cifsacl.o cifs-$(CONFIG_CIFS_UPCALL) += cifs_spnego.o diff --git a/fs/cifs/README b/fs/cifs/README index ee68d103654..46af99ab361 100644 --- a/fs/cifs/README +++ b/fs/cifs/README @@ -337,6 +337,15 @@ A partial list of the supported mount options follows: wsize default write size (default 57344) maximum wsize currently allowed by CIFS is 57344 (fourteen 4096 byte pages) + actimeo=n attribute cache timeout in seconds (default 1 second). + After this timeout, the cifs client requests fresh attribute + information from the server. This option allows to tune the + attribute cache timeout to suit the workload needs. Shorter + timeouts mean better the cache coherency, but increased number + of calls to the server. Longer timeouts mean reduced number + of calls to the server at the expense of less stricter cache + coherency checks (i.e. incorrect attribute cache for a short + period of time). rw mount the network share read-write (note that the server may still consider the share read-only) ro mount network share read-only diff --git a/fs/cifs/TODO b/fs/cifs/TODO index 5aff46c61e5..355abcdcda9 100644 --- a/fs/cifs/TODO +++ b/fs/cifs/TODO @@ -81,7 +81,7 @@ u) DOS attrs - returned as pseudo-xattr in Samba format (check VFAT and NTFS for v) mount check for unmatched uids -w) Add support for new vfs entry points for setlease and fallocate +w) Add support for new vfs entry point for fallocate x) Fix Samba 3 server to handle Linux kernel aio so dbench with lots of processes can proceed better in parallel (on the server) diff --git a/fs/cifs/cifs_fs_sb.h b/fs/cifs/cifs_fs_sb.h index 525ba59a410..7852cd67705 100644 --- a/fs/cifs/cifs_fs_sb.h +++ b/fs/cifs/cifs_fs_sb.h @@ -15,7 +15,7 @@ * the GNU Lesser General Public License for more details. * */ -#include <linux/radix-tree.h> +#include <linux/rbtree.h> #ifndef _CIFS_FS_SB_H #define _CIFS_FS_SB_H @@ -42,12 +42,13 @@ #define CIFS_MOUNT_MULTIUSER 0x20000 /* multiuser mount */ struct cifs_sb_info { - struct radix_tree_root tlink_tree; -#define CIFS_TLINK_MASTER_TAG 0 /* is "master" (mount) tcon */ + struct rb_root tlink_tree; spinlock_t tlink_tree_lock; + struct tcon_link *master_tlink; struct nls_table *local_nls; unsigned int rsize; unsigned int wsize; + unsigned long actimeo; /* attribute cache timeout (jiffies) */ atomic_t active; uid_t mnt_uid; gid_t mnt_gid; diff --git a/fs/cifs/cifsacl.c b/fs/cifs/cifsacl.c index c9b4792ae82..a437ec391a0 100644 --- a/fs/cifs/cifsacl.c +++ b/fs/cifs/cifsacl.c @@ -30,8 +30,6 @@ #include "cifs_debug.h" -#ifdef CONFIG_CIFS_EXPERIMENTAL - static struct cifs_wksid wksidarr[NUM_WK_SIDS] = { {{1, 0, {0, 0, 0, 0, 0, 0}, {0, 0, 0, 0, 0} }, "null user"}, {{1, 1, {0, 0, 0, 0, 0, 1}, {0, 0, 0, 0, 0} }, "nobody"}, @@ -560,7 +558,7 @@ static struct cifs_ntsd *get_cifs_acl_by_fid(struct cifs_sb_info *cifs_sb, struct tcon_link *tlink = cifs_sb_tlink(cifs_sb); if (IS_ERR(tlink)) - return NULL; + return ERR_CAST(tlink); xid = GetXid(); rc = CIFSSMBGetCIFSACL(xid, tlink_tcon(tlink), fid, &pntsd, pacllen); @@ -568,7 +566,9 @@ static struct cifs_ntsd *get_cifs_acl_by_fid(struct cifs_sb_info *cifs_sb, cifs_put_tlink(tlink); - cFYI(1, "GetCIFSACL rc = %d ACL len %d", rc, *pacllen); + cFYI(1, "%s: rc = %d ACL len %d", __func__, rc, *pacllen); + if (rc) + return ERR_PTR(rc); return pntsd; } @@ -583,7 +583,7 @@ static struct cifs_ntsd *get_cifs_acl_by_path(struct cifs_sb_info *cifs_sb, struct tcon_link *tlink = cifs_sb_tlink(cifs_sb); if (IS_ERR(tlink)) - return NULL; + return ERR_CAST(tlink); tcon = tlink_tcon(tlink); xid = GetXid(); @@ -591,23 +591,22 @@ static struct cifs_ntsd *get_cifs_acl_by_path(struct cifs_sb_info *cifs_sb, rc = CIFSSMBOpen(xid, tcon, path, FILE_OPEN, READ_CONTROL, 0, &fid, &oplock, NULL, cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); - if (rc) { - cERROR(1, "Unable to open file to get ACL"); - goto out; + if (!rc) { + rc = CIFSSMBGetCIFSACL(xid, tcon, fid, &pntsd, pacllen); + CIFSSMBClose(xid, tcon, fid); } - rc = CIFSSMBGetCIFSACL(xid, tcon, fid, &pntsd, pacllen); - cFYI(1, "GetCIFSACL rc = %d ACL len %d", rc, *pacllen); - - CIFSSMBClose(xid, tcon, fid); - out: cifs_put_tlink(tlink); FreeXid(xid); + + cFYI(1, "%s: rc = %d ACL len %d", __func__, rc, *pacllen); + if (rc) + return ERR_PTR(rc); return pntsd; } /* Retrieve an ACL from the server */ -static struct cifs_ntsd *get_cifs_acl(struct cifs_sb_info *cifs_sb, +struct cifs_ntsd *get_cifs_acl(struct cifs_sb_info *cifs_sb, struct inode *inode, const char *path, u32 *pacllen) { @@ -695,7 +694,7 @@ static int set_cifs_acl(struct cifs_ntsd *pnntsd, __u32 acllen, } /* Translate the CIFS ACL (simlar to NTFS ACL) for a file into mode bits */ -void +int cifs_acl_to_fattr(struct cifs_sb_info *cifs_sb, struct cifs_fattr *fattr, struct inode *inode, const char *path, const __u16 *pfid) { @@ -711,17 +710,21 @@ cifs_acl_to_fattr(struct cifs_sb_info *cifs_sb, struct cifs_fattr *fattr, pntsd = get_cifs_acl(cifs_sb, inode, path, &acllen); /* if we can retrieve the ACL, now parse Access Control Entries, ACEs */ - if (pntsd) + if (IS_ERR(pntsd)) { + rc = PTR_ERR(pntsd); + cERROR(1, "%s: error %d getting sec desc", __func__, rc); + } else { rc = parse_sec_desc(pntsd, acllen, fattr); - if (rc) - cFYI(1, "parse sec desc failed rc = %d", rc); + kfree(pntsd); + if (rc) + cERROR(1, "parse sec desc failed rc = %d", rc); + } - kfree(pntsd); - return; + return rc; } /* Convert mode bits to an ACL so we can update the ACL on the server */ -int mode_to_acl(struct inode *inode, const char *path, __u64 nmode) +int mode_to_cifs_acl(struct inode *inode, const char *path, __u64 nmode) { int rc = 0; __u32 secdesclen = 0; @@ -736,7 +739,10 @@ int mode_to_acl(struct inode *inode, const char *path, __u64 nmode) /* Add three ACEs for owner, group, everyone getting rid of other ACEs as chmod disables ACEs and set the security descriptor */ - if (pntsd) { + if (IS_ERR(pntsd)) { + rc = PTR_ERR(pntsd); + cERROR(1, "%s: error %d getting sec desc", __func__, rc); + } else { /* allocate memory for the smb header, set security descriptor request security descriptor parameters, and secuirty descriptor itself */ @@ -766,4 +772,3 @@ int mode_to_acl(struct inode *inode, const char *path, __u64 nmode) return rc; } -#endif /* CONFIG_CIFS_EXPERIMENTAL */ diff --git a/fs/cifs/cifsacl.h b/fs/cifs/cifsacl.h index 6c8096cf515..c4ae7d03656 100644 --- a/fs/cifs/cifsacl.h +++ b/fs/cifs/cifsacl.h @@ -74,11 +74,7 @@ struct cifs_wksid { char sidname[SIDNAMELENGTH]; } __attribute__((packed)); -#ifdef CONFIG_CIFS_EXPERIMENTAL - extern int match_sid(struct cifs_sid *); extern int compare_sids(const struct cifs_sid *, const struct cifs_sid *); -#endif /* CONFIG_CIFS_EXPERIMENTAL */ - #endif /* _CIFSACL_H */ diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 75c4eaa7958..8e21e0fe65d 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -116,7 +116,7 @@ cifs_read_super(struct super_block *sb, void *data, return -ENOMEM; spin_lock_init(&cifs_sb->tlink_tree_lock); - INIT_RADIX_TREE(&cifs_sb->tlink_tree, GFP_KERNEL); + cifs_sb->tlink_tree = RB_ROOT; rc = bdi_setup_and_register(&cifs_sb->bdi, "cifs", BDI_CAP_MAP_COPY); if (rc) { @@ -283,10 +283,13 @@ cifs_statfs(struct dentry *dentry, struct kstatfs *buf) return 0; } -static int cifs_permission(struct inode *inode, int mask) +static int cifs_permission(struct inode *inode, int mask, unsigned int flags) { struct cifs_sb_info *cifs_sb; + if (flags & IPERM_FLAG_RCU) + return -ECHILD; + cifs_sb = CIFS_SB(inode->i_sb); if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_PERM) { @@ -298,7 +301,7 @@ static int cifs_permission(struct inode *inode, int mask) on the client (above and beyond ACL on servers) for servers which do not support setting and viewing mode bits, so allowing client to check permissions is useful */ - return generic_permission(inode, mask, NULL); + return generic_permission(inode, mask, flags, NULL); } static struct kmem_cache *cifs_inode_cachep; @@ -321,8 +324,7 @@ cifs_alloc_inode(struct super_block *sb) /* Until the file is open and we have gotten oplock info back from the server, can not assume caching of file data or metadata */ - cifs_inode->clientCanCacheRead = false; - cifs_inode->clientCanCacheAll = false; + cifs_set_oplock_level(cifs_inode, 0); cifs_inode->delete_pending = false; cifs_inode->invalid_mapping = false; cifs_inode->vfs_inode.i_blkbits = 14; /* 2**14 = CIFS_MAX_MSGSIZE */ @@ -335,10 +337,17 @@ cifs_alloc_inode(struct super_block *sb) return &cifs_inode->vfs_inode; } +static void cifs_i_callback(struct rcu_head *head) +{ + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); + kmem_cache_free(cifs_inode_cachep, CIFS_I(inode)); +} + static void cifs_destroy_inode(struct inode *inode) { - kmem_cache_free(cifs_inode_cachep, CIFS_I(inode)); + call_rcu(&inode->i_rcu, cifs_i_callback); } static void @@ -459,9 +468,13 @@ cifs_show_options(struct seq_file *s, struct vfsmount *m) seq_printf(s, ",acl"); if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MF_SYMLINKS) seq_printf(s, ",mfsymlinks"); + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_FSCACHE) + seq_printf(s, ",fsc"); seq_printf(s, ",rsize=%d", cifs_sb->rsize); seq_printf(s, ",wsize=%d", cifs_sb->wsize); + /* convert actimeo and display it in seconds */ + seq_printf(s, ",actimeo=%lu", cifs_sb->actimeo / HZ); return 0; } @@ -934,7 +947,6 @@ init_cifs(void) GlobalCurrentXid = 0; GlobalTotalActiveXid = 0; GlobalMaxActiveXid = 0; - memset(Local_System_Name, 0, 15); spin_lock_init(&cifs_tcp_ses_lock); spin_lock_init(&cifs_file_list_lock); spin_lock_init(&GlobalMid_Lock); diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index f259e4d7612..7136c0c3e2f 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -45,6 +45,16 @@ #define CIFS_MIN_RCV_POOL 4 /* + * default attribute cache timeout (jiffies) + */ +#define CIFS_DEF_ACTIMEO (1 * HZ) + +/* + * max attribute cache timeout (jiffies) - 2^30 + */ +#define CIFS_MAX_ACTIMEO (1 << 30) + +/* * MAX_REQ is the maximum number of requests that WE will send * on one socket concurrently. It also matches the most common * value of max multiplex returned by servers. We may @@ -336,7 +346,8 @@ struct cifsTconInfo { * "get" on the container. */ struct tcon_link { - unsigned long tl_index; + struct rb_node tl_rbnode; + uid_t tl_uid; unsigned long tl_flags; #define TCON_LINK_MASTER 0 #define TCON_LINK_PENDING 1 @@ -745,8 +756,6 @@ GLOBAL_EXTERN unsigned int GlobalTotalActiveXid; /* prot by GlobalMid_Sem */ GLOBAL_EXTERN unsigned int GlobalMaxActiveXid; /* prot by GlobalMid_Sem */ GLOBAL_EXTERN spinlock_t GlobalMid_Lock; /* protects above & list operations */ /* on midQ entries */ -GLOBAL_EXTERN char Local_System_Name[15]; - /* * Global counters, updated atomically */ diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index edb6d90efdf..e6d1481b16c 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -54,7 +54,8 @@ do { \ __func__, curr_xid, (int)rc); \ } while (0) extern char *build_path_from_dentry(struct dentry *); -extern char *cifs_build_path_to_root(struct cifs_sb_info *cifs_sb); +extern char *cifs_build_path_to_root(struct cifs_sb_info *cifs_sb, + struct cifsTconInfo *tcon); extern char *build_wildcard_path_from_dentry(struct dentry *direntry); extern char *cifs_compose_mount_options(const char *sb_mountdata, const char *fullpath, const struct dfs_info3_param *ref, @@ -79,9 +80,7 @@ extern bool is_valid_oplock_break(struct smb_hdr *smb, struct TCP_Server_Info *); extern bool is_size_safe_to_change(struct cifsInodeInfo *, __u64 eof); extern struct cifsFileInfo *find_writable_file(struct cifsInodeInfo *, bool); -#ifdef CONFIG_CIFS_EXPERIMENTAL extern struct cifsFileInfo *find_readable_file(struct cifsInodeInfo *, bool); -#endif extern unsigned int smbCalcSize(struct smb_hdr *ptr); extern unsigned int smbCalcSize_LE(struct smb_hdr *ptr); extern int decode_negTokenInit(unsigned char *security_blob, int length, @@ -104,6 +103,7 @@ extern struct timespec cifs_NTtimeToUnix(__le64 utc_nanoseconds_since_1601); extern u64 cifs_UnixTimeToNT(struct timespec); extern struct timespec cnvrtDosUnixTm(__le16 le_date, __le16 le_time, int offset); +extern void cifs_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock); extern struct cifsFileInfo *cifs_new_fileinfo(__u16 fileHandle, struct file *file, struct tcon_link *tlink, @@ -129,10 +129,12 @@ extern int cifs_get_file_info_unix(struct file *filp); extern int cifs_get_inode_info_unix(struct inode **pinode, const unsigned char *search_path, struct super_block *sb, int xid); -extern void cifs_acl_to_fattr(struct cifs_sb_info *cifs_sb, +extern int cifs_acl_to_fattr(struct cifs_sb_info *cifs_sb, struct cifs_fattr *fattr, struct inode *inode, const char *path, const __u16 *pfid); -extern int mode_to_acl(struct inode *inode, const char *path, __u64); +extern int mode_to_cifs_acl(struct inode *inode, const char *path, __u64); +extern struct cifs_ntsd *get_cifs_acl(struct cifs_sb_info *, struct inode *, + const char *, u32 *); extern int cifs_mount(struct super_block *, struct cifs_sb_info *, char *, const char *); diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 2f2632b6df5..67acfb3acad 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -2478,95 +2478,6 @@ querySymLinkRetry: } #ifdef CONFIG_CIFS_EXPERIMENTAL -/* Initialize NT TRANSACT SMB into small smb request buffer. - This assumes that all NT TRANSACTS that we init here have - total parm and data under about 400 bytes (to fit in small cifs - buffer size), which is the case so far, it easily fits. NB: - Setup words themselves and ByteCount - MaxSetupCount (size of returned setup area) and - MaxParameterCount (returned parms size) must be set by caller */ -static int -smb_init_nttransact(const __u16 sub_command, const int setup_count, - const int parm_len, struct cifsTconInfo *tcon, - void **ret_buf) -{ - int rc; - __u32 temp_offset; - struct smb_com_ntransact_req *pSMB; - - rc = small_smb_init(SMB_COM_NT_TRANSACT, 19 + setup_count, tcon, - (void **)&pSMB); - if (rc) - return rc; - *ret_buf = (void *)pSMB; - pSMB->Reserved = 0; - pSMB->TotalParameterCount = cpu_to_le32(parm_len); - pSMB->TotalDataCount = 0; - pSMB->MaxDataCount = cpu_to_le32((tcon->ses->server->maxBuf - - MAX_CIFS_HDR_SIZE) & 0xFFFFFF00); - pSMB->ParameterCount = pSMB->TotalParameterCount; - pSMB->DataCount = pSMB->TotalDataCount; - temp_offset = offsetof(struct smb_com_ntransact_req, Parms) + - (setup_count * 2) - 4 /* for rfc1001 length itself */; - pSMB->ParameterOffset = cpu_to_le32(temp_offset); - pSMB->DataOffset = cpu_to_le32(temp_offset + parm_len); - pSMB->SetupCount = setup_count; /* no need to le convert byte fields */ - pSMB->SubCommand = cpu_to_le16(sub_command); - return 0; -} - -static int -validate_ntransact(char *buf, char **ppparm, char **ppdata, - __u32 *pparmlen, __u32 *pdatalen) -{ - char *end_of_smb; - __u32 data_count, data_offset, parm_count, parm_offset; - struct smb_com_ntransact_rsp *pSMBr; - - *pdatalen = 0; - *pparmlen = 0; - - if (buf == NULL) - return -EINVAL; - - pSMBr = (struct smb_com_ntransact_rsp *)buf; - - /* ByteCount was converted from little endian in SendReceive */ - end_of_smb = 2 /* sizeof byte count */ + pSMBr->ByteCount + - (char *)&pSMBr->ByteCount; - - data_offset = le32_to_cpu(pSMBr->DataOffset); - data_count = le32_to_cpu(pSMBr->DataCount); - parm_offset = le32_to_cpu(pSMBr->ParameterOffset); - parm_count = le32_to_cpu(pSMBr->ParameterCount); - - *ppparm = (char *)&pSMBr->hdr.Protocol + parm_offset; - *ppdata = (char *)&pSMBr->hdr.Protocol + data_offset; - - /* should we also check that parm and data areas do not overlap? */ - if (*ppparm > end_of_smb) { - cFYI(1, "parms start after end of smb"); - return -EINVAL; - } else if (parm_count + *ppparm > end_of_smb) { - cFYI(1, "parm end after end of smb"); - return -EINVAL; - } else if (*ppdata > end_of_smb) { - cFYI(1, "data starts after end of smb"); - return -EINVAL; - } else if (data_count + *ppdata > end_of_smb) { - cFYI(1, "data %p + count %d (%p) past smb end %p start %p", - *ppdata, data_count, (data_count + *ppdata), - end_of_smb, pSMBr); - return -EINVAL; - } else if (parm_count + data_count > pSMBr->ByteCount) { - cFYI(1, "parm count and data count larger than SMB"); - return -EINVAL; - } - *pdatalen = data_count; - *pparmlen = parm_count; - return 0; -} - int CIFSSMBQueryReparseLinkInfo(const int xid, struct cifsTconInfo *tcon, const unsigned char *searchName, @@ -3056,7 +2967,97 @@ GetExtAttrOut: #endif /* CONFIG_POSIX */ -#ifdef CONFIG_CIFS_EXPERIMENTAL +#ifdef CONFIG_CIFS_ACL +/* + * Initialize NT TRANSACT SMB into small smb request buffer. This assumes that + * all NT TRANSACTS that we init here have total parm and data under about 400 + * bytes (to fit in small cifs buffer size), which is the case so far, it + * easily fits. NB: Setup words themselves and ByteCount MaxSetupCount (size of + * returned setup area) and MaxParameterCount (returned parms size) must be set + * by caller + */ +static int +smb_init_nttransact(const __u16 sub_command, const int setup_count, + const int parm_len, struct cifsTconInfo *tcon, + void **ret_buf) +{ + int rc; + __u32 temp_offset; + struct smb_com_ntransact_req *pSMB; + + rc = small_smb_init(SMB_COM_NT_TRANSACT, 19 + setup_count, tcon, + (void **)&pSMB); + if (rc) + return rc; + *ret_buf = (void *)pSMB; + pSMB->Reserved = 0; + pSMB->TotalParameterCount = cpu_to_le32(parm_len); + pSMB->TotalDataCount = 0; + pSMB->MaxDataCount = cpu_to_le32((tcon->ses->server->maxBuf - + MAX_CIFS_HDR_SIZE) & 0xFFFFFF00); + pSMB->ParameterCount = pSMB->TotalParameterCount; + pSMB->DataCount = pSMB->TotalDataCount; + temp_offset = offsetof(struct smb_com_ntransact_req, Parms) + + (setup_count * 2) - 4 /* for rfc1001 length itself */; + pSMB->ParameterOffset = cpu_to_le32(temp_offset); + pSMB->DataOffset = cpu_to_le32(temp_offset + parm_len); + pSMB->SetupCount = setup_count; /* no need to le convert byte fields */ + pSMB->SubCommand = cpu_to_le16(sub_command); + return 0; +} + +static int +validate_ntransact(char *buf, char **ppparm, char **ppdata, + __u32 *pparmlen, __u32 *pdatalen) +{ + char *end_of_smb; + __u32 data_count, data_offset, parm_count, parm_offset; + struct smb_com_ntransact_rsp *pSMBr; + + *pdatalen = 0; + *pparmlen = 0; + + if (buf == NULL) + return -EINVAL; + + pSMBr = (struct smb_com_ntransact_rsp *)buf; + + /* ByteCount was converted from little endian in SendReceive */ + end_of_smb = 2 /* sizeof byte count */ + pSMBr->ByteCount + + (char *)&pSMBr->ByteCount; + + data_offset = le32_to_cpu(pSMBr->DataOffset); + data_count = le32_to_cpu(pSMBr->DataCount); + parm_offset = le32_to_cpu(pSMBr->ParameterOffset); + parm_count = le32_to_cpu(pSMBr->ParameterCount); + + *ppparm = (char *)&pSMBr->hdr.Protocol + parm_offset; + *ppdata = (char *)&pSMBr->hdr.Protocol + data_offset; + + /* should we also check that parm and data areas do not overlap? */ + if (*ppparm > end_of_smb) { + cFYI(1, "parms start after end of smb"); + return -EINVAL; + } else if (parm_count + *ppparm > end_of_smb) { + cFYI(1, "parm end after end of smb"); + return -EINVAL; + } else if (*ppdata > end_of_smb) { + cFYI(1, "data starts after end of smb"); + return -EINVAL; + } else if (data_count + *ppdata > end_of_smb) { + cFYI(1, "data %p + count %d (%p) past smb end %p start %p", + *ppdata, data_count, (data_count + *ppdata), + end_of_smb, pSMBr); + return -EINVAL; + } else if (parm_count + data_count > pSMBr->ByteCount) { + cFYI(1, "parm count and data count larger than SMB"); + return -EINVAL; + } + *pdatalen = data_count; + *pparmlen = parm_count; + return 0; +} + /* Get Security Descriptor (by handle) from remote server for a file or dir */ int CIFSSMBGetCIFSACL(const int xid, struct cifsTconInfo *tcon, __u16 fid, @@ -3214,7 +3215,7 @@ setCifsAclRetry: return (rc); } -#endif /* CONFIG_CIFS_EXPERIMENTAL */ +#endif /* CONFIG_CIFS_ACL */ /* Legacy Query Path Information call for lookup to old servers such as Win9x/WinME */ diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 9eb327defa1..cc1a8604a79 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -105,6 +105,7 @@ struct smb_vol { unsigned int wsize; bool sockopt_tcp_nodelay:1; unsigned short int port; + unsigned long actimeo; /* attribute cache timeout (jiffies) */ char *prepath; struct sockaddr_storage srcaddr; /* allow binding to a local IP */ struct nls_table *local_nls; @@ -116,6 +117,7 @@ struct smb_vol { static int ipv4_connect(struct TCP_Server_Info *server); static int ipv6_connect(struct TCP_Server_Info *server); +static void tlink_rb_insert(struct rb_root *root, struct tcon_link *new_tlink); static void cifs_prune_tlinks(struct work_struct *work); /* @@ -805,23 +807,20 @@ cifs_parse_mount_options(char *options, const char *devname, short int override_gid = -1; bool uid_specified = false; bool gid_specified = false; + char *nodename = utsname()->nodename; separator[0] = ','; separator[1] = 0; - if (Local_System_Name[0] != 0) - memcpy(vol->source_rfc1001_name, Local_System_Name, 15); - else { - char *nodename = utsname()->nodename; - int n = strnlen(nodename, 15); - memset(vol->source_rfc1001_name, 0x20, 15); - for (i = 0; i < n; i++) { - /* does not have to be perfect mapping since field is - informational, only used for servers that do not support - port 445 and it can be overridden at mount time */ - vol->source_rfc1001_name[i] = toupper(nodename[i]); - } - } + /* + * does not have to be perfect mapping since field is + * informational, only used for servers that do not support + * port 445 and it can be overridden at mount time + */ + memset(vol->source_rfc1001_name, 0x20, 15); + for (i = 0; i < strnlen(nodename, 15); i++) + vol->source_rfc1001_name[i] = toupper(nodename[i]); + vol->source_rfc1001_name[15] = 0; /* null target name indicates to use *SMBSERVR default called name if we end up sending RFC1001 session initialize */ @@ -839,6 +838,8 @@ cifs_parse_mount_options(char *options, const char *devname, /* default to using server inode numbers where available */ vol->server_ino = 1; + vol->actimeo = CIFS_DEF_ACTIMEO; + if (!options) return 1; @@ -1213,6 +1214,16 @@ cifs_parse_mount_options(char *options, const char *devname, printk(KERN_WARNING "CIFS: server net" "biosname longer than 15 truncated.\n"); } + } else if (strnicmp(data, "actimeo", 7) == 0) { + if (value && *value) { + vol->actimeo = HZ * simple_strtoul(value, + &value, 0); + if (vol->actimeo > CIFS_MAX_ACTIMEO) { + cERROR(1, "CIFS: attribute cache" + "timeout too large"); + return 1; + } + } } else if (strnicmp(data, "credentials", 4) == 0) { /* ignore */ } else if (strnicmp(data, "version", 3) == 0) { @@ -1351,6 +1362,11 @@ cifs_parse_mount_options(char *options, const char *devname, "supported. Instead set " "/proc/fs/cifs/LookupCacheEnabled to 0\n"); } else if (strnicmp(data, "fsc", 3) == 0) { +#ifndef CONFIG_CIFS_FSCACHE + cERROR(1, "FS-Cache support needs CONFIG_CIFS_FSCACHE" + "kernel config option set"); + return 1; +#endif vol->fsc = true; } else if (strnicmp(data, "mfsymlinks", 10) == 0) { vol->mfsymlinks = true; @@ -2565,6 +2581,8 @@ static void setup_cifs_sb(struct smb_vol *pvolume_info, cFYI(1, "file mode: 0x%x dir mode: 0x%x", cifs_sb->mnt_file_mode, cifs_sb->mnt_dir_mode); + cifs_sb->actimeo = pvolume_info->actimeo; + if (pvolume_info->noperm) cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_NO_PERM; if (pvolume_info->setuids) @@ -2815,13 +2833,13 @@ remote_path_check: /* check if a whole path (including prepath) is not remote */ if (!rc && cifs_sb->prepathlen && tcon) { /* build_path_to_root works only when we have a valid tcon */ - full_path = cifs_build_path_to_root(cifs_sb); + full_path = cifs_build_path_to_root(cifs_sb, tcon); if (full_path == NULL) { rc = -ENOMEM; goto mount_fail_check; } rc = is_path_accessible(xid, tcon, cifs_sb, full_path); - if (rc != -EREMOTE) { + if (rc != 0 && rc != -EREMOTE) { kfree(full_path); goto mount_fail_check; } @@ -2900,24 +2918,16 @@ remote_path_check: goto mount_fail_check; } - tlink->tl_index = pSesInfo->linux_uid; + tlink->tl_uid = pSesInfo->linux_uid; tlink->tl_tcon = tcon; tlink->tl_time = jiffies; set_bit(TCON_LINK_MASTER, &tlink->tl_flags); set_bit(TCON_LINK_IN_TREE, &tlink->tl_flags); - rc = radix_tree_preload(GFP_KERNEL); - if (rc == -ENOMEM) { - kfree(tlink); - goto mount_fail_check; - } - + cifs_sb->master_tlink = tlink; spin_lock(&cifs_sb->tlink_tree_lock); - radix_tree_insert(&cifs_sb->tlink_tree, pSesInfo->linux_uid, tlink); - radix_tree_tag_set(&cifs_sb->tlink_tree, pSesInfo->linux_uid, - CIFS_TLINK_MASTER_TAG); + tlink_rb_insert(&cifs_sb->tlink_tree, tlink); spin_unlock(&cifs_sb->tlink_tree_lock); - radix_tree_preload_end(); queue_delayed_work(system_nrt_wq, &cifs_sb->prune_tlinks, TLINK_IDLE_EXPIRE); @@ -3107,32 +3117,25 @@ CIFSTCon(unsigned int xid, struct cifsSesInfo *ses, int cifs_umount(struct super_block *sb, struct cifs_sb_info *cifs_sb) { - int i, ret; + struct rb_root *root = &cifs_sb->tlink_tree; + struct rb_node *node; + struct tcon_link *tlink; char *tmp; - struct tcon_link *tlink[8]; - unsigned long index = 0; cancel_delayed_work_sync(&cifs_sb->prune_tlinks); - do { - spin_lock(&cifs_sb->tlink_tree_lock); - ret = radix_tree_gang_lookup(&cifs_sb->tlink_tree, - (void **)tlink, index, - ARRAY_SIZE(tlink)); - /* increment index for next pass */ - if (ret > 0) - index = tlink[ret - 1]->tl_index + 1; - for (i = 0; i < ret; i++) { - cifs_get_tlink(tlink[i]); - clear_bit(TCON_LINK_IN_TREE, &tlink[i]->tl_flags); - radix_tree_delete(&cifs_sb->tlink_tree, - tlink[i]->tl_index); - } - spin_unlock(&cifs_sb->tlink_tree_lock); + spin_lock(&cifs_sb->tlink_tree_lock); + while ((node = rb_first(root))) { + tlink = rb_entry(node, struct tcon_link, tl_rbnode); + cifs_get_tlink(tlink); + clear_bit(TCON_LINK_IN_TREE, &tlink->tl_flags); + rb_erase(node, root); - for (i = 0; i < ret; i++) - cifs_put_tlink(tlink[i]); - } while (ret != 0); + spin_unlock(&cifs_sb->tlink_tree_lock); + cifs_put_tlink(tlink); + spin_lock(&cifs_sb->tlink_tree_lock); + } + spin_unlock(&cifs_sb->tlink_tree_lock); tmp = cifs_sb->prepath; cifs_sb->prepathlen = 0; @@ -3271,22 +3274,10 @@ out: return tcon; } -static struct tcon_link * +static inline struct tcon_link * cifs_sb_master_tlink(struct cifs_sb_info *cifs_sb) { - struct tcon_link *tlink; - unsigned int ret; - - spin_lock(&cifs_sb->tlink_tree_lock); - ret = radix_tree_gang_lookup_tag(&cifs_sb->tlink_tree, (void **)&tlink, - 0, 1, CIFS_TLINK_MASTER_TAG); - spin_unlock(&cifs_sb->tlink_tree_lock); - - /* the master tcon should always be present */ - if (ret == 0) - BUG(); - - return tlink; + return cifs_sb->master_tlink; } struct cifsTconInfo * @@ -3302,6 +3293,47 @@ cifs_sb_tcon_pending_wait(void *unused) return signal_pending(current) ? -ERESTARTSYS : 0; } +/* find and return a tlink with given uid */ +static struct tcon_link * +tlink_rb_search(struct rb_root *root, uid_t uid) +{ + struct rb_node *node = root->rb_node; + struct tcon_link *tlink; + + while (node) { + tlink = rb_entry(node, struct tcon_link, tl_rbnode); + + if (tlink->tl_uid > uid) + node = node->rb_left; + else if (tlink->tl_uid < uid) + node = node->rb_right; + else + return tlink; + } + return NULL; +} + +/* insert a tcon_link into the tree */ +static void +tlink_rb_insert(struct rb_root *root, struct tcon_link *new_tlink) +{ + struct rb_node **new = &(root->rb_node), *parent = NULL; + struct tcon_link *tlink; + + while (*new) { + tlink = rb_entry(*new, struct tcon_link, tl_rbnode); + parent = *new; + + if (tlink->tl_uid > new_tlink->tl_uid) + new = &((*new)->rb_left); + else + new = &((*new)->rb_right); + } + + rb_link_node(&new_tlink->tl_rbnode, parent, new); + rb_insert_color(&new_tlink->tl_rbnode, root); +} + /* * Find or construct an appropriate tcon given a cifs_sb and the fsuid of the * current task. @@ -3309,7 +3341,7 @@ cifs_sb_tcon_pending_wait(void *unused) * If the superblock doesn't refer to a multiuser mount, then just return * the master tcon for the mount. * - * First, search the radix tree for an existing tcon for this fsuid. If one + * First, search the rbtree for an existing tcon for this fsuid. If one * exists, then check to see if it's pending construction. If it is then wait * for construction to complete. Once it's no longer pending, check to see if * it failed and either return an error or retry construction, depending on @@ -3322,14 +3354,14 @@ struct tcon_link * cifs_sb_tlink(struct cifs_sb_info *cifs_sb) { int ret; - unsigned long fsuid = (unsigned long) current_fsuid(); + uid_t fsuid = current_fsuid(); struct tcon_link *tlink, *newtlink; if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MULTIUSER)) return cifs_get_tlink(cifs_sb_master_tlink(cifs_sb)); spin_lock(&cifs_sb->tlink_tree_lock); - tlink = radix_tree_lookup(&cifs_sb->tlink_tree, fsuid); + tlink = tlink_rb_search(&cifs_sb->tlink_tree, fsuid); if (tlink) cifs_get_tlink(tlink); spin_unlock(&cifs_sb->tlink_tree_lock); @@ -3338,36 +3370,24 @@ cifs_sb_tlink(struct cifs_sb_info *cifs_sb) newtlink = kzalloc(sizeof(*tlink), GFP_KERNEL); if (newtlink == NULL) return ERR_PTR(-ENOMEM); - newtlink->tl_index = fsuid; + newtlink->tl_uid = fsuid; newtlink->tl_tcon = ERR_PTR(-EACCES); set_bit(TCON_LINK_PENDING, &newtlink->tl_flags); set_bit(TCON_LINK_IN_TREE, &newtlink->tl_flags); cifs_get_tlink(newtlink); - ret = radix_tree_preload(GFP_KERNEL); - if (ret != 0) { - kfree(newtlink); - return ERR_PTR(ret); - } - spin_lock(&cifs_sb->tlink_tree_lock); /* was one inserted after previous search? */ - tlink = radix_tree_lookup(&cifs_sb->tlink_tree, fsuid); + tlink = tlink_rb_search(&cifs_sb->tlink_tree, fsuid); if (tlink) { cifs_get_tlink(tlink); spin_unlock(&cifs_sb->tlink_tree_lock); - radix_tree_preload_end(); kfree(newtlink); goto wait_for_construction; } - ret = radix_tree_insert(&cifs_sb->tlink_tree, fsuid, newtlink); - spin_unlock(&cifs_sb->tlink_tree_lock); - radix_tree_preload_end(); - if (ret) { - kfree(newtlink); - return ERR_PTR(ret); - } tlink = newtlink; + tlink_rb_insert(&cifs_sb->tlink_tree, tlink); + spin_unlock(&cifs_sb->tlink_tree_lock); } else { wait_for_construction: ret = wait_on_bit(&tlink->tl_flags, TCON_LINK_PENDING, @@ -3413,39 +3433,39 @@ cifs_prune_tlinks(struct work_struct *work) { struct cifs_sb_info *cifs_sb = container_of(work, struct cifs_sb_info, prune_tlinks.work); - struct tcon_link *tlink[8]; - unsigned long now = jiffies; - unsigned long index = 0; - int i, ret; + struct rb_root *root = &cifs_sb->tlink_tree; + struct rb_node *node = rb_first(root); + struct rb_node *tmp; + struct tcon_link *tlink; - do { - spin_lock(&cifs_sb->tlink_tree_lock); - ret = radix_tree_gang_lookup(&cifs_sb->tlink_tree, - (void **)tlink, index, - ARRAY_SIZE(tlink)); - /* increment index for next pass */ - if (ret > 0) - index = tlink[ret - 1]->tl_index + 1; - for (i = 0; i < ret; i++) { - if (test_bit(TCON_LINK_MASTER, &tlink[i]->tl_flags) || - atomic_read(&tlink[i]->tl_count) != 0 || - time_after(tlink[i]->tl_time + TLINK_IDLE_EXPIRE, - now)) { - tlink[i] = NULL; - continue; - } - cifs_get_tlink(tlink[i]); - clear_bit(TCON_LINK_IN_TREE, &tlink[i]->tl_flags); - radix_tree_delete(&cifs_sb->tlink_tree, - tlink[i]->tl_index); - } - spin_unlock(&cifs_sb->tlink_tree_lock); + /* + * Because we drop the spinlock in the loop in order to put the tlink + * it's not guarded against removal of links from the tree. The only + * places that remove entries from the tree are this function and + * umounts. Because this function is non-reentrant and is canceled + * before umount can proceed, this is safe. + */ + spin_lock(&cifs_sb->tlink_tree_lock); + node = rb_first(root); + while (node != NULL) { + tmp = node; + node = rb_next(tmp); + tlink = rb_entry(tmp, struct tcon_link, tl_rbnode); + + if (test_bit(TCON_LINK_MASTER, &tlink->tl_flags) || + atomic_read(&tlink->tl_count) != 0 || + time_after(tlink->tl_time + TLINK_IDLE_EXPIRE, jiffies)) + continue; - for (i = 0; i < ret; i++) { - if (tlink[i] != NULL) - cifs_put_tlink(tlink[i]); - } - } while (ret != 0); + cifs_get_tlink(tlink); + clear_bit(TCON_LINK_IN_TREE, &tlink->tl_flags); + rb_erase(tmp, root); + + spin_unlock(&cifs_sb->tlink_tree_lock); + cifs_put_tlink(tlink); + spin_lock(&cifs_sb->tlink_tree_lock); + } + spin_unlock(&cifs_sb->tlink_tree_lock); queue_delayed_work(system_nrt_wq, &cifs_sb->prune_tlinks, TLINK_IDLE_EXPIRE); diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c index 3840eddbfb7..db2a58c00f7 100644 --- a/fs/cifs/dir.c +++ b/fs/cifs/dir.c @@ -135,9 +135,9 @@ static void setup_cifs_dentry(struct cifsTconInfo *tcon, struct inode *newinode) { if (tcon->nocase) - direntry->d_op = &cifs_ci_dentry_ops; + d_set_d_op(direntry, &cifs_ci_dentry_ops); else - direntry->d_op = &cifs_dentry_ops; + d_set_d_op(direntry, &cifs_dentry_ops); d_instantiate(direntry, newinode); } @@ -421,9 +421,9 @@ int cifs_mknod(struct inode *inode, struct dentry *direntry, int mode, rc = cifs_get_inode_info_unix(&newinode, full_path, inode->i_sb, xid); if (pTcon->nocase) - direntry->d_op = &cifs_ci_dentry_ops; + d_set_d_op(direntry, &cifs_ci_dentry_ops); else - direntry->d_op = &cifs_dentry_ops; + d_set_d_op(direntry, &cifs_dentry_ops); if (rc == 0) d_instantiate(direntry, newinode); @@ -604,9 +604,9 @@ cifs_lookup(struct inode *parent_dir_inode, struct dentry *direntry, if ((rc == 0) && (newInode != NULL)) { if (pTcon->nocase) - direntry->d_op = &cifs_ci_dentry_ops; + d_set_d_op(direntry, &cifs_ci_dentry_ops); else - direntry->d_op = &cifs_dentry_ops; + d_set_d_op(direntry, &cifs_dentry_ops); d_add(direntry, newInode); if (posix_open) { filp = lookup_instantiate_filp(nd, direntry, @@ -634,9 +634,9 @@ cifs_lookup(struct inode *parent_dir_inode, struct dentry *direntry, rc = 0; direntry->d_time = jiffies; if (pTcon->nocase) - direntry->d_op = &cifs_ci_dentry_ops; + d_set_d_op(direntry, &cifs_ci_dentry_ops); else - direntry->d_op = &cifs_dentry_ops; + d_set_d_op(direntry, &cifs_dentry_ops); d_add(direntry, NULL); /* if it was once a directory (but how can we tell?) we could do shrink_dcache_parent(direntry); */ @@ -656,22 +656,37 @@ lookup_out: static int cifs_d_revalidate(struct dentry *direntry, struct nameidata *nd) { - int isValid = 1; + if (nd->flags & LOOKUP_RCU) + return -ECHILD; if (direntry->d_inode) { if (cifs_revalidate_dentry(direntry)) return 0; - } else { - cFYI(1, "neg dentry 0x%p name = %s", - direntry, direntry->d_name.name); - if (time_after(jiffies, direntry->d_time + HZ) || - !lookupCacheEnabled) { - d_drop(direntry); - isValid = 0; - } + else + return 1; } - return isValid; + /* + * This may be nfsd (or something), anyway, we can't see the + * intent of this. So, since this can be for creation, drop it. + */ + if (!nd) + return 0; + + /* + * Drop the negative dentry, in order to make sure to use the + * case sensitive name which is specified by user if this is + * for creation. + */ + if (!(nd->flags & (LOOKUP_CONTINUE | LOOKUP_PARENT))) { + if (nd->flags & (LOOKUP_CREATE | LOOKUP_RENAME_TARGET)) + return 0; + } + + if (time_after(jiffies, direntry->d_time + HZ) || !lookupCacheEnabled) + return 0; + + return 1; } /* static int cifs_d_delete(struct dentry *direntry) @@ -688,9 +703,10 @@ const struct dentry_operations cifs_dentry_ops = { /* d_delete: cifs_d_delete, */ /* not needed except for debugging */ }; -static int cifs_ci_hash(struct dentry *dentry, struct qstr *q) +static int cifs_ci_hash(const struct dentry *dentry, const struct inode *inode, + struct qstr *q) { - struct nls_table *codepage = CIFS_SB(dentry->d_inode->i_sb)->local_nls; + struct nls_table *codepage = CIFS_SB(dentry->d_sb)->local_nls; unsigned long hash; int i; @@ -703,21 +719,16 @@ static int cifs_ci_hash(struct dentry *dentry, struct qstr *q) return 0; } -static int cifs_ci_compare(struct dentry *dentry, struct qstr *a, - struct qstr *b) +static int cifs_ci_compare(const struct dentry *parent, + const struct inode *pinode, + const struct dentry *dentry, const struct inode *inode, + unsigned int len, const char *str, const struct qstr *name) { - struct nls_table *codepage = CIFS_SB(dentry->d_inode->i_sb)->local_nls; - - if ((a->len == b->len) && - (nls_strnicmp(codepage, a->name, b->name, a->len) == 0)) { - /* - * To preserve case, don't let an existing negative dentry's - * case take precedence. If a is not a negative dentry, this - * should have no side effects - */ - memcpy((void *)a->name, b->name, a->len); + struct nls_table *codepage = CIFS_SB(pinode->i_sb)->local_nls; + + if ((name->len == len) && + (nls_strnicmp(codepage, name->name, str, len) == 0)) return 0; - } return 1; } diff --git a/fs/cifs/dns_resolve.c b/fs/cifs/dns_resolve.c index 0eb87026cad..548f06230a6 100644 --- a/fs/cifs/dns_resolve.c +++ b/fs/cifs/dns_resolve.c @@ -66,7 +66,7 @@ dns_resolve_server_name_to_ip(const char *unc, char **ip_addr) /* Search for server name delimiter */ sep = memchr(hostname, '\\', len); if (sep) - len = sep - unc; + len = sep - hostname; else cFYI(1, "%s: probably server name is whole unc: %s", __func__, unc); diff --git a/fs/cifs/file.c b/fs/cifs/file.c index ae82159cf7f..5a28660ca2b 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -146,12 +146,7 @@ client_can_cache: rc = cifs_get_inode_info(&inode, full_path, buf, inode->i_sb, xid, NULL); - if ((oplock & 0xF) == OPLOCK_EXCLUSIVE) { - pCifsInode->clientCanCacheAll = true; - pCifsInode->clientCanCacheRead = true; - cFYI(1, "Exclusive Oplock granted on inode %p", inode); - } else if ((oplock & 0xF) == OPLOCK_READ) - pCifsInode->clientCanCacheRead = true; + cifs_set_oplock_level(pCifsInode, oplock); return rc; } @@ -253,12 +248,7 @@ cifs_new_fileinfo(__u16 fileHandle, struct file *file, list_add_tail(&pCifsFile->flist, &pCifsInode->openFileList); spin_unlock(&cifs_file_list_lock); - if ((oplock & 0xF) == OPLOCK_EXCLUSIVE) { - pCifsInode->clientCanCacheAll = true; - pCifsInode->clientCanCacheRead = true; - cFYI(1, "Exclusive Oplock inode %p", inode); - } else if ((oplock & 0xF) == OPLOCK_READ) - pCifsInode->clientCanCacheRead = true; + cifs_set_oplock_level(pCifsInode, oplock); file->private_data = pCifsFile; return pCifsFile; @@ -271,8 +261,9 @@ cifs_new_fileinfo(__u16 fileHandle, struct file *file, */ void cifsFileInfo_put(struct cifsFileInfo *cifs_file) { + struct inode *inode = cifs_file->dentry->d_inode; struct cifsTconInfo *tcon = tlink_tcon(cifs_file->tlink); - struct cifsInodeInfo *cifsi = CIFS_I(cifs_file->dentry->d_inode); + struct cifsInodeInfo *cifsi = CIFS_I(inode); struct cifsLockInfo *li, *tmp; spin_lock(&cifs_file_list_lock); @@ -288,8 +279,7 @@ void cifsFileInfo_put(struct cifsFileInfo *cifs_file) if (list_empty(&cifsi->openFileList)) { cFYI(1, "closing last open instance for inode %p", cifs_file->dentry->d_inode); - cifsi->clientCanCacheRead = false; - cifsi->clientCanCacheAll = false; + cifs_set_oplock_level(cifsi, 0); } spin_unlock(&cifs_file_list_lock); @@ -607,8 +597,6 @@ reopen_success: rc = filemap_write_and_wait(inode->i_mapping); mapping_set_error(inode->i_mapping, rc); - pCifsInode->clientCanCacheAll = false; - pCifsInode->clientCanCacheRead = false; if (tcon->unix_ext) rc = cifs_get_inode_info_unix(&inode, full_path, inode->i_sb, xid); @@ -622,18 +610,9 @@ reopen_success: invalidate the current end of file on the server we can not go to the server to get the new inod info */ - if ((oplock & 0xF) == OPLOCK_EXCLUSIVE) { - pCifsInode->clientCanCacheAll = true; - pCifsInode->clientCanCacheRead = true; - cFYI(1, "Exclusive Oplock granted on inode %p", - pCifsFile->dentry->d_inode); - } else if ((oplock & 0xF) == OPLOCK_READ) { - pCifsInode->clientCanCacheRead = true; - pCifsInode->clientCanCacheAll = false; - } else { - pCifsInode->clientCanCacheRead = false; - pCifsInode->clientCanCacheAll = false; - } + + cifs_set_oplock_level(pCifsInode, oplock); + cifs_relock_file(pCifsFile); reopen_error_exit: @@ -775,12 +754,6 @@ int cifs_lock(struct file *file, int cmd, struct file_lock *pfLock) cifs_sb = CIFS_SB(file->f_path.dentry->d_sb); tcon = tlink_tcon(((struct cifsFileInfo *)file->private_data)->tlink); - - if (file->private_data == NULL) { - rc = -EBADF; - FreeXid(xid); - return rc; - } netfid = ((struct cifsFileInfo *)file->private_data)->netfid; if ((tcon->ses->capabilities & CAP_UNIX) && @@ -956,6 +929,7 @@ cifs_update_eof(struct cifsInodeInfo *cifsi, loff_t offset, ssize_t cifs_user_write(struct file *file, const char __user *write_data, size_t write_size, loff_t *poffset) { + struct inode *inode = file->f_path.dentry->d_inode; int rc = 0; unsigned int bytes_written = 0; unsigned int total_written; @@ -963,7 +937,7 @@ ssize_t cifs_user_write(struct file *file, const char __user *write_data, struct cifsTconInfo *pTcon; int xid, long_op; struct cifsFileInfo *open_file; - struct cifsInodeInfo *cifsi = CIFS_I(file->f_path.dentry->d_inode); + struct cifsInodeInfo *cifsi = CIFS_I(inode); cifs_sb = CIFS_SB(file->f_path.dentry->d_sb); @@ -1029,21 +1003,17 @@ ssize_t cifs_user_write(struct file *file, const char __user *write_data, cifs_stats_bytes_written(pTcon, total_written); - /* since the write may have blocked check these pointers again */ - if ((file->f_path.dentry) && (file->f_path.dentry->d_inode)) { - struct inode *inode = file->f_path.dentry->d_inode; /* Do not update local mtime - server will set its actual value on write - * inode->i_ctime = inode->i_mtime = - * current_fs_time(inode->i_sb);*/ - if (total_written > 0) { - spin_lock(&inode->i_lock); - if (*poffset > file->f_path.dentry->d_inode->i_size) - i_size_write(file->f_path.dentry->d_inode, - *poffset); - spin_unlock(&inode->i_lock); - } - mark_inode_dirty_sync(file->f_path.dentry->d_inode); + * inode->i_ctime = inode->i_mtime = + * current_fs_time(inode->i_sb);*/ + if (total_written > 0) { + spin_lock(&inode->i_lock); + if (*poffset > inode->i_size) + i_size_write(inode, *poffset); + spin_unlock(&inode->i_lock); } + mark_inode_dirty_sync(inode); + FreeXid(xid); return total_written; } @@ -1138,7 +1108,6 @@ static ssize_t cifs_write(struct cifsFileInfo *open_file, return total_written; } -#ifdef CONFIG_CIFS_EXPERIMENTAL struct cifsFileInfo *find_readable_file(struct cifsInodeInfo *cifs_inode, bool fsuid_only) { @@ -1172,13 +1141,12 @@ struct cifsFileInfo *find_readable_file(struct cifsInodeInfo *cifs_inode, spin_unlock(&cifs_file_list_lock); return NULL; } -#endif struct cifsFileInfo *find_writable_file(struct cifsInodeInfo *cifs_inode, bool fsuid_only) { struct cifsFileInfo *open_file; - struct cifs_sb_info *cifs_sb = CIFS_SB(cifs_inode->vfs_inode.i_sb); + struct cifs_sb_info *cifs_sb; bool any_available = false; int rc; @@ -1192,6 +1160,8 @@ struct cifsFileInfo *find_writable_file(struct cifsInodeInfo *cifs_inode, return NULL; } + cifs_sb = CIFS_SB(cifs_inode->vfs_inode.i_sb); + /* only filter by fsuid on multiuser mounts */ if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MULTIUSER)) fsuid_only = false; @@ -2299,8 +2269,10 @@ void cifs_oplock_break_get(struct cifsFileInfo *cfile) void cifs_oplock_break_put(struct cifsFileInfo *cfile) { + struct super_block *sb = cfile->dentry->d_sb; + cifsFileInfo_put(cfile); - cifs_sb_deactive(cfile->dentry->d_sb); + cifs_sb_deactive(sb); } const struct address_space_operations cifs_addr_ops = { diff --git a/fs/cifs/fscache.c b/fs/cifs/fscache.c index a2ad94efcfe..297a43d0ff7 100644 --- a/fs/cifs/fscache.c +++ b/fs/cifs/fscache.c @@ -2,7 +2,7 @@ * fs/cifs/fscache.c - CIFS filesystem cache interface * * Copyright (c) 2010 Novell, Inc. - * Author(s): Suresh Jayaraman (sjayaraman@suse.de> + * Author(s): Suresh Jayaraman <sjayaraman@suse.de> * * This library is free software; you can redistribute it and/or modify * it under the terms of the GNU Lesser General Public License as published @@ -67,10 +67,12 @@ static void cifs_fscache_enable_inode_cookie(struct inode *inode) if (cifsi->fscache) return; - cifsi->fscache = fscache_acquire_cookie(tcon->fscache, + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_FSCACHE) { + cifsi->fscache = fscache_acquire_cookie(tcon->fscache, &cifs_fscache_inode_object_def, cifsi); - cFYI(1, "CIFS: got FH cookie (0x%p/0x%p)", tcon->fscache, + cFYI(1, "CIFS: got FH cookie (0x%p/0x%p)", tcon->fscache, cifsi->fscache); + } } void cifs_fscache_release_inode_cookie(struct inode *inode) @@ -101,10 +103,8 @@ void cifs_fscache_set_inode_cookie(struct inode *inode, struct file *filp) { if ((filp->f_flags & O_ACCMODE) != O_RDONLY) cifs_fscache_disable_inode_cookie(inode); - else { + else cifs_fscache_enable_inode_cookie(inode); - cFYI(1, "CIFS: fscache inode cookie set"); - } } void cifs_fscache_reset_inode_cookie(struct inode *inode) diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index 39869c3c3ef..a853a89857a 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -686,13 +686,18 @@ int cifs_get_inode_info(struct inode **pinode, cFYI(1, "cifs_sfu_type failed: %d", tmprc); } -#ifdef CONFIG_CIFS_EXPERIMENTAL +#ifdef CONFIG_CIFS_ACL /* fill in 0777 bits from ACL */ if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL) { - cFYI(1, "Getting mode bits from ACL"); - cifs_acl_to_fattr(cifs_sb, &fattr, *pinode, full_path, pfid); + rc = cifs_acl_to_fattr(cifs_sb, &fattr, *pinode, full_path, + pfid); + if (rc) { + cFYI(1, "%s: Getting ACL failed with error: %d", + __func__, rc); + goto cgii_exit; + } } -#endif +#endif /* CONFIG_CIFS_ACL */ /* fill in remaining high mode bits e.g. SUID, VTX */ if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UNX_EMUL) @@ -723,12 +728,12 @@ static const struct inode_operations cifs_ipc_inode_ops = { .lookup = cifs_lookup, }; -char *cifs_build_path_to_root(struct cifs_sb_info *cifs_sb) +char *cifs_build_path_to_root(struct cifs_sb_info *cifs_sb, + struct cifsTconInfo *tcon) { int pplen = cifs_sb->prepathlen; int dfsplen; char *full_path = NULL; - struct cifsTconInfo *tcon = cifs_sb_master_tcon(cifs_sb); /* if no prefix path, simply set path to the root of share to "" */ if (pplen == 0) { @@ -804,14 +809,14 @@ inode_has_hashed_dentries(struct inode *inode) { struct dentry *dentry; - spin_lock(&dcache_lock); + spin_lock(&inode->i_lock); list_for_each_entry(dentry, &inode->i_dentry, d_alias) { if (!d_unhashed(dentry) || IS_ROOT(dentry)) { - spin_unlock(&dcache_lock); + spin_unlock(&inode->i_lock); return true; } } - spin_unlock(&dcache_lock); + spin_unlock(&inode->i_lock); return false; } @@ -870,7 +875,7 @@ struct inode *cifs_root_iget(struct super_block *sb, unsigned long ino) char *full_path; struct cifsTconInfo *tcon = cifs_sb_master_tcon(cifs_sb); - full_path = cifs_build_path_to_root(cifs_sb); + full_path = cifs_build_path_to_root(cifs_sb, tcon); if (full_path == NULL) return ERR_PTR(-ENOMEM); @@ -881,8 +886,10 @@ struct inode *cifs_root_iget(struct super_block *sb, unsigned long ino) rc = cifs_get_inode_info(&inode, full_path, NULL, sb, xid, NULL); - if (!inode) - return ERR_PTR(rc); + if (!inode) { + inode = ERR_PTR(rc); + goto out; + } #ifdef CONFIG_CIFS_FSCACHE /* populate tcon->resource_id */ @@ -898,13 +905,11 @@ struct inode *cifs_root_iget(struct super_block *sb, unsigned long ino) inode->i_uid = cifs_sb->mnt_uid; inode->i_gid = cifs_sb->mnt_gid; } else if (rc) { - kfree(full_path); - _FreeXid(xid); iget_failed(inode); - return ERR_PTR(rc); + inode = ERR_PTR(rc); } - +out: kfree(full_path); /* can not call macro FreeXid here since in a void func * TODO: This is no longer true @@ -1314,9 +1319,9 @@ int cifs_mkdir(struct inode *inode, struct dentry *direntry, int mode) to set uid/gid */ inc_nlink(inode); if (pTcon->nocase) - direntry->d_op = &cifs_ci_dentry_ops; + d_set_d_op(direntry, &cifs_ci_dentry_ops); else - direntry->d_op = &cifs_dentry_ops; + d_set_d_op(direntry, &cifs_dentry_ops); cifs_unix_basic_to_fattr(&fattr, pInfo, cifs_sb); cifs_fill_uniqueid(inode->i_sb, &fattr); @@ -1358,9 +1363,9 @@ mkdir_get_info: inode->i_sb, xid, NULL); if (pTcon->nocase) - direntry->d_op = &cifs_ci_dentry_ops; + d_set_d_op(direntry, &cifs_ci_dentry_ops); else - direntry->d_op = &cifs_dentry_ops; + d_set_d_op(direntry, &cifs_dentry_ops); d_instantiate(direntry, newinode); /* setting nlink not necessary except in cases where we * failed to get it from the server or was set bogus */ @@ -1648,6 +1653,7 @@ static bool cifs_inode_needs_reval(struct inode *inode) { struct cifsInodeInfo *cifs_i = CIFS_I(inode); + struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); if (cifs_i->clientCanCacheRead) return false; @@ -1658,19 +1664,21 @@ cifs_inode_needs_reval(struct inode *inode) if (cifs_i->time == 0) return true; - /* FIXME: the actimeo should be tunable */ - if (time_after_eq(jiffies, cifs_i->time + HZ)) + if (!time_in_range(jiffies, cifs_i->time, + cifs_i->time + cifs_sb->actimeo)) return true; /* hardlinked files w/ noserverino get "special" treatment */ - if (!(CIFS_SB(inode->i_sb)->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM) && + if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM) && S_ISREG(inode->i_mode) && inode->i_nlink != 1) return true; return false; } -/* check invalid_mapping flag and zap the cache if it's set */ +/* + * Zap the cache. Called when invalid_mapping flag is set. + */ static void cifs_invalidate_mapping(struct inode *inode) { @@ -2114,11 +2122,16 @@ cifs_setattr_nounix(struct dentry *direntry, struct iattr *attrs) if (attrs->ia_valid & ATTR_MODE) { rc = 0; -#ifdef CONFIG_CIFS_EXPERIMENTAL - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL) - rc = mode_to_acl(inode, full_path, mode); - else -#endif +#ifdef CONFIG_CIFS_ACL + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL) { + rc = mode_to_cifs_acl(inode, full_path, mode); + if (rc) { + cFYI(1, "%s: Setting ACL failed with error: %d", + __func__, rc); + goto cifs_setattr_exit; + } + } else +#endif /* CONFIG_CIFS_ACL */ if (((mode & S_IWUGO) == 0) && (cifsInode->cifsAttrs & ATTR_READONLY) == 0) { @@ -2177,7 +2190,6 @@ cifs_setattr_nounix(struct dentry *direntry, struct iattr *attrs) setattr_copy(inode, attrs); mark_inode_dirty(inode); - return 0; cifs_setattr_exit: kfree(full_path); diff --git a/fs/cifs/ioctl.c b/fs/cifs/ioctl.c index 077bf756f34..0c98672d012 100644 --- a/fs/cifs/ioctl.c +++ b/fs/cifs/ioctl.c @@ -38,10 +38,10 @@ long cifs_ioctl(struct file *filep, unsigned int command, unsigned long arg) struct cifs_sb_info *cifs_sb; #ifdef CONFIG_CIFS_POSIX struct cifsFileInfo *pSMBFile = filep->private_data; - struct cifsTconInfo *tcon = tlink_tcon(pSMBFile->tlink); + struct cifsTconInfo *tcon; __u64 ExtAttrBits = 0; __u64 ExtAttrMask = 0; - __u64 caps = le64_to_cpu(tcon->fsUnixInfo.Capability); + __u64 caps; #endif /* CONFIG_CIFS_POSIX */ xid = GetXid(); @@ -62,9 +62,11 @@ long cifs_ioctl(struct file *filep, unsigned int command, unsigned long arg) break; #ifdef CONFIG_CIFS_POSIX case FS_IOC_GETFLAGS: + if (pSMBFile == NULL) + break; + tcon = tlink_tcon(pSMBFile->tlink); + caps = le64_to_cpu(tcon->fsUnixInfo.Capability); if (CIFS_UNIX_EXTATTR_CAP & caps) { - if (pSMBFile == NULL) - break; rc = CIFSGetExtAttr(xid, tcon, pSMBFile->netfid, &ExtAttrBits, &ExtAttrMask); if (rc == 0) @@ -75,13 +77,15 @@ long cifs_ioctl(struct file *filep, unsigned int command, unsigned long arg) break; case FS_IOC_SETFLAGS: + if (pSMBFile == NULL) + break; + tcon = tlink_tcon(pSMBFile->tlink); + caps = le64_to_cpu(tcon->fsUnixInfo.Capability); if (CIFS_UNIX_EXTATTR_CAP & caps) { if (get_user(ExtAttrBits, (int __user *)arg)) { rc = -EFAULT; break; } - if (pSMBFile == NULL) - break; /* rc= CIFSGetExtAttr(xid,tcon,pSMBFile->netfid, extAttrBits, &ExtAttrMask);*/ } diff --git a/fs/cifs/link.c b/fs/cifs/link.c index 85cdbf831e7..fe2f6a93c49 100644 --- a/fs/cifs/link.c +++ b/fs/cifs/link.c @@ -525,9 +525,9 @@ cifs_symlink(struct inode *inode, struct dentry *direntry, const char *symname) rc); } else { if (pTcon->nocase) - direntry->d_op = &cifs_ci_dentry_ops; + d_set_d_op(direntry, &cifs_ci_dentry_ops); else - direntry->d_op = &cifs_dentry_ops; + d_set_d_op(direntry, &cifs_dentry_ops); d_instantiate(direntry, newinode); } } diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index c4e296fe351..43f10281bc1 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -569,10 +569,9 @@ is_valid_oplock_break(struct smb_hdr *buf, struct TCP_Server_Info *srv) cFYI(1, "file id match, oplock break"); pCifsInode = CIFS_I(netfile->dentry->d_inode); - pCifsInode->clientCanCacheAll = false; - if (pSMB->OplockLevel == 0) - pCifsInode->clientCanCacheRead = false; + cifs_set_oplock_level(pCifsInode, + pSMB->OplockLevel); /* * cifs_oplock_break_put() can't be called * from here. Get reference after queueing @@ -722,3 +721,23 @@ cifs_autodisable_serverino(struct cifs_sb_info *cifs_sb) cifs_sb_master_tcon(cifs_sb)->treeName); } } + +void cifs_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock) +{ + oplock &= 0xF; + + if (oplock == OPLOCK_EXCLUSIVE) { + cinode->clientCanCacheAll = true; + cinode->clientCanCacheRead = true; + cFYI(1, "Exclusive Oplock granted on inode %p", + &cinode->vfs_inode); + } else if (oplock == OPLOCK_READ) { + cinode->clientCanCacheAll = false; + cinode->clientCanCacheRead = true; + cFYI(1, "Level II Oplock granted on inode %p", + &cinode->vfs_inode); + } else { + cinode->clientCanCacheAll = false; + cinode->clientCanCacheRead = false; + } +} diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c index ef7bb7b50f5..ec5b68e3b92 100644 --- a/fs/cifs/readdir.c +++ b/fs/cifs/readdir.c @@ -79,7 +79,7 @@ cifs_readdir_lookup(struct dentry *parent, struct qstr *name, cFYI(1, "For %s", name->name); if (parent->d_op && parent->d_op->d_hash) - parent->d_op->d_hash(parent, name); + parent->d_op->d_hash(parent, parent->d_inode, name); else name->hash = full_name_hash(name->name, name->len); @@ -103,9 +103,9 @@ cifs_readdir_lookup(struct dentry *parent, struct qstr *name, } if (cifs_sb_master_tcon(CIFS_SB(sb))->nocase) - dentry->d_op = &cifs_ci_dentry_ops; + d_set_d_op(dentry, &cifs_ci_dentry_ops); else - dentry->d_op = &cifs_dentry_ops; + d_set_d_op(dentry, &cifs_dentry_ops); alias = d_materialise_unique(dentry, inode); if (alias != NULL) { @@ -226,26 +226,29 @@ static int initiate_cifs_search(const int xid, struct file *file) char *full_path = NULL; struct cifsFileInfo *cifsFile; struct cifs_sb_info *cifs_sb = CIFS_SB(file->f_path.dentry->d_sb); - struct tcon_link *tlink; + struct tcon_link *tlink = NULL; struct cifsTconInfo *pTcon; - tlink = cifs_sb_tlink(cifs_sb); - if (IS_ERR(tlink)) - return PTR_ERR(tlink); - pTcon = tlink_tcon(tlink); - - if (file->private_data == NULL) - file->private_data = - kzalloc(sizeof(struct cifsFileInfo), GFP_KERNEL); if (file->private_data == NULL) { - rc = -ENOMEM; - goto error_exit; + tlink = cifs_sb_tlink(cifs_sb); + if (IS_ERR(tlink)) + return PTR_ERR(tlink); + + cifsFile = kzalloc(sizeof(struct cifsFileInfo), GFP_KERNEL); + if (cifsFile == NULL) { + rc = -ENOMEM; + goto error_exit; + } + file->private_data = cifsFile; + cifsFile->tlink = cifs_get_tlink(tlink); + pTcon = tlink_tcon(tlink); + } else { + cifsFile = file->private_data; + pTcon = tlink_tcon(cifsFile->tlink); } - cifsFile = file->private_data; cifsFile->invalidHandle = true; cifsFile->srch_inf.endOfSearch = false; - cifsFile->tlink = cifs_get_tlink(tlink); full_path = build_path_from_dentry(file->f_path.dentry); if (full_path == NULL) { @@ -756,18 +759,6 @@ static int cifs_filldir(char *pfindEntry, struct file *file, filldir_t filldir, rc = filldir(direntry, qstring.name, qstring.len, file->f_pos, ino, fattr.cf_dtype); - /* - * we can not return filldir errors to the caller since they are - * "normal" when the stat blocksize is too small - we return remapped - * error instead - * - * FIXME: This looks bogus. filldir returns -EOVERFLOW in the above - * case already. Why should we be clobbering other errors from it? - */ - if (rc) { - cFYI(1, "filldir rc = %d", rc); - rc = -EOVERFLOW; - } dput(tmp_dentry); return rc; } diff --git a/fs/cifs/xattr.c b/fs/cifs/xattr.c index a264b744bb4..eae2a149160 100644 --- a/fs/cifs/xattr.c +++ b/fs/cifs/xattr.c @@ -30,10 +30,11 @@ #define MAX_EA_VALUE_SIZE 65535 #define CIFS_XATTR_DOS_ATTRIB "user.DosAttrib" +#define CIFS_XATTR_CIFS_ACL "system.cifs_acl" #define CIFS_XATTR_USER_PREFIX "user." #define CIFS_XATTR_SYSTEM_PREFIX "system." #define CIFS_XATTR_OS2_PREFIX "os2." -#define CIFS_XATTR_SECURITY_PREFIX ".security" +#define CIFS_XATTR_SECURITY_PREFIX "security." #define CIFS_XATTR_TRUSTED_PREFIX "trusted." #define XATTR_TRUSTED_PREFIX_LEN 8 #define XATTR_SECURITY_PREFIX_LEN 9 @@ -277,29 +278,8 @@ ssize_t cifs_getxattr(struct dentry *direntry, const char *ea_name, cifs_sb->local_nls, cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); -#ifdef CONFIG_CIFS_EXPERIMENTAL - else if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_CIFS_ACL) { - __u16 fid; - int oplock = 0; - struct cifs_ntsd *pacl = NULL; - __u32 buflen = 0; - if (experimEnabled) - rc = CIFSSMBOpen(xid, pTcon, full_path, - FILE_OPEN, GENERIC_READ, 0, &fid, - &oplock, NULL, cifs_sb->local_nls, - cifs_sb->mnt_cifs_flags & - CIFS_MOUNT_MAP_SPECIAL_CHR); - /* else rc is EOPNOTSUPP from above */ - - if (rc == 0) { - rc = CIFSSMBGetCIFSACL(xid, pTcon, fid, &pacl, - &buflen); - CIFSSMBClose(xid, pTcon, fid); - } - } -#endif /* EXPERIMENTAL */ #else - cFYI(1, "query POSIX ACL not supported yet"); + cFYI(1, "Query POSIX ACL not supported yet"); #endif /* CONFIG_CIFS_POSIX */ } else if (strncmp(ea_name, POSIX_ACL_XATTR_DEFAULT, strlen(POSIX_ACL_XATTR_DEFAULT)) == 0) { @@ -311,8 +291,33 @@ ssize_t cifs_getxattr(struct dentry *direntry, const char *ea_name, cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MAP_SPECIAL_CHR); #else - cFYI(1, "query POSIX default ACL not supported yet"); -#endif + cFYI(1, "Query POSIX default ACL not supported yet"); +#endif /* CONFIG_CIFS_POSIX */ + } else if (strncmp(ea_name, CIFS_XATTR_CIFS_ACL, + strlen(CIFS_XATTR_CIFS_ACL)) == 0) { +#ifdef CONFIG_CIFS_ACL + u32 acllen; + struct cifs_ntsd *pacl; + + pacl = get_cifs_acl(cifs_sb, direntry->d_inode, + full_path, &acllen); + if (IS_ERR(pacl)) { + rc = PTR_ERR(pacl); + cERROR(1, "%s: error %zd getting sec desc", + __func__, rc); + } else { + if (ea_value) { + if (acllen > buf_size) + acllen = -ERANGE; + else + memcpy(ea_value, pacl, acllen); + } + rc = acllen; + kfree(pacl); + } +#else + cFYI(1, "Query CIFS ACL not supported yet"); +#endif /* CONFIG_CIFS_ACL */ } else if (strncmp(ea_name, CIFS_XATTR_TRUSTED_PREFIX, XATTR_TRUSTED_PREFIX_LEN) == 0) { cFYI(1, "Trusted xattr namespace not supported yet"); diff --git a/fs/coda/cache.c b/fs/coda/cache.c index 9060f08e70c..5525e1c660f 100644 --- a/fs/coda/cache.c +++ b/fs/coda/cache.c @@ -93,7 +93,7 @@ static void coda_flag_children(struct dentry *parent, int flag) struct list_head *child; struct dentry *de; - spin_lock(&dcache_lock); + spin_lock(&parent->d_lock); list_for_each(child, &parent->d_subdirs) { de = list_entry(child, struct dentry, d_u.d_child); @@ -102,7 +102,7 @@ static void coda_flag_children(struct dentry *parent, int flag) continue; coda_flag_inode(de->d_inode, flag); } - spin_unlock(&dcache_lock); + spin_unlock(&parent->d_lock); return; } diff --git a/fs/coda/dir.c b/fs/coda/dir.c index 5d8b3553960..29badd91360 100644 --- a/fs/coda/dir.c +++ b/fs/coda/dir.c @@ -18,6 +18,7 @@ #include <linux/errno.h> #include <linux/string.h> #include <linux/spinlock.h> +#include <linux/namei.h> #include <asm/uaccess.h> @@ -47,7 +48,7 @@ static int coda_readdir(struct file *file, void *buf, filldir_t filldir); /* dentry ops */ static int coda_dentry_revalidate(struct dentry *de, struct nameidata *nd); -static int coda_dentry_delete(struct dentry *); +static int coda_dentry_delete(const struct dentry *); /* support routines */ static int coda_venus_readdir(struct file *coda_file, void *buf, @@ -125,7 +126,7 @@ static struct dentry *coda_lookup(struct inode *dir, struct dentry *entry, struc return ERR_PTR(error); exit: - entry->d_op = &coda_dentry_operations; + d_set_d_op(entry, &coda_dentry_operations); if (inode && (type & CODA_NOCACHE)) coda_flag_inode(inode, C_VATTR | C_PURGE); @@ -134,10 +135,13 @@ exit: } -int coda_permission(struct inode *inode, int mask) +int coda_permission(struct inode *inode, int mask, unsigned int flags) { int error; + if (flags & IPERM_FLAG_RCU) + return -ECHILD; + mask &= MAY_READ | MAY_WRITE | MAY_EXEC; if (!mask) @@ -541,9 +545,13 @@ out: /* called when a cache lookup succeeds */ static int coda_dentry_revalidate(struct dentry *de, struct nameidata *nd) { - struct inode *inode = de->d_inode; + struct inode *inode; struct coda_inode_info *cii; + if (nd->flags & LOOKUP_RCU) + return -ECHILD; + + inode = de->d_inode; if (!inode || coda_isroot(inode)) goto out; if (is_bad_inode(inode)) @@ -559,7 +567,7 @@ static int coda_dentry_revalidate(struct dentry *de, struct nameidata *nd) if (cii->c_flags & C_FLUSH) coda_flag_inode_children(inode, C_FLUSH); - if (atomic_read(&de->d_count) > 1) + if (de->d_count > 1) /* pretend it's valid, but don't change the flags */ goto out; @@ -577,7 +585,7 @@ out: * This is the callback from dput() when d_count is going to 0. * We use this to unhash dentries with bad inodes. */ -static int coda_dentry_delete(struct dentry * dentry) +static int coda_dentry_delete(const struct dentry * dentry) { int flags; diff --git a/fs/coda/inode.c b/fs/coda/inode.c index 5ea57c8c7f9..50dc7d189f5 100644 --- a/fs/coda/inode.c +++ b/fs/coda/inode.c @@ -56,11 +56,18 @@ static struct inode *coda_alloc_inode(struct super_block *sb) return &ei->vfs_inode; } -static void coda_destroy_inode(struct inode *inode) +static void coda_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(coda_inode_cachep, ITOC(inode)); } +static void coda_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, coda_i_callback); +} + static void init_once(void *foo) { struct coda_inode_info *ei = (struct coda_inode_info *) foo; diff --git a/fs/coda/pioctl.c b/fs/coda/pioctl.c index 2fd89b5c5c7..741f0bd0391 100644 --- a/fs/coda/pioctl.c +++ b/fs/coda/pioctl.c @@ -24,7 +24,7 @@ #include <linux/coda_psdev.h> /* pioctl ops */ -static int coda_ioctl_permission(struct inode *inode, int mask); +static int coda_ioctl_permission(struct inode *inode, int mask, unsigned int flags); static long coda_pioctl(struct file *filp, unsigned int cmd, unsigned long user_data); @@ -41,8 +41,10 @@ const struct file_operations coda_ioctl_operations = { }; /* the coda pioctl inode ops */ -static int coda_ioctl_permission(struct inode *inode, int mask) +static int coda_ioctl_permission(struct inode *inode, int mask, unsigned int flags) { + if (flags & IPERM_FLAG_RCU) + return -ECHILD; return (mask & MAY_EXEC) ? -EACCES : 0; } diff --git a/fs/compat.c b/fs/compat.c index c580c322fa6..eb1740ac8c0 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -1350,6 +1350,10 @@ static int compat_count(compat_uptr_t __user *argv, int max) argv++; if (i++ >= max) return -E2BIG; + + if (fatal_signal_pending(current)) + return -ERESTARTNOHAND; + cond_resched(); } } return i; @@ -1391,6 +1395,12 @@ static int compat_copy_strings(int argc, compat_uptr_t __user *argv, while (len > 0) { int offset, bytes_to_copy; + if (fatal_signal_pending(current)) { + ret = -ERESTARTNOHAND; + goto out; + } + cond_resched(); + offset = pos % PAGE_SIZE; if (offset == 0) offset = PAGE_SIZE; @@ -1407,18 +1417,8 @@ static int compat_copy_strings(int argc, compat_uptr_t __user *argv, if (!kmapped_page || kpos != (pos & PAGE_MASK)) { struct page *page; -#ifdef CONFIG_STACK_GROWSUP - ret = expand_stack_downwards(bprm->vma, pos); - if (ret < 0) { - /* We've exceed the stack rlimit. */ - ret = -E2BIG; - goto out; - } -#endif - ret = get_user_pages(current, bprm->mm, pos, - 1, 1, 1, &page, NULL); - if (ret <= 0) { - /* We've exceed the stack rlimit. */ + page = get_arg_page(bprm, pos, 1); + if (!page) { ret = -E2BIG; goto out; } @@ -1539,8 +1539,10 @@ int compat_do_execve(char * filename, return retval; out: - if (bprm->mm) + if (bprm->mm) { + acct_arg_size(bprm, 0); mmput(bprm->mm); + } out_file: if (bprm->file) { diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c index 410ed188faa..61abb638b4b 100644 --- a/fs/compat_ioctl.c +++ b/fs/compat_ioctl.c @@ -19,7 +19,6 @@ #include <linux/compiler.h> #include <linux/sched.h> #include <linux/smp.h> -#include <linux/smp_lock.h> #include <linux/ioctl.h> #include <linux/if.h> #include <linux/if_bridge.h> @@ -43,7 +42,7 @@ #include <linux/tty.h> #include <linux/vt_kern.h> #include <linux/fb.h> -#include <linux/videodev.h> +#include <linux/videodev2.h> #include <linux/netdevice.h> #include <linux/raw.h> #include <linux/blkdev.h> @@ -837,6 +836,7 @@ COMPATIBLE_IOCTL(TCSETSW) COMPATIBLE_IOCTL(TCSETSF) COMPATIBLE_IOCTL(TIOCLINUX) COMPATIBLE_IOCTL(TIOCSBRK) +COMPATIBLE_IOCTL(TIOCGDEV) COMPATIBLE_IOCTL(TIOCCBRK) COMPATIBLE_IOCTL(TIOCGSID) COMPATIBLE_IOCTL(TIOCGICOUNT) diff --git a/fs/configfs/configfs_internal.h b/fs/configfs/configfs_internal.h index da6061a6df4..026cf68553a 100644 --- a/fs/configfs/configfs_internal.h +++ b/fs/configfs/configfs_internal.h @@ -120,7 +120,7 @@ static inline struct config_item *configfs_get_config_item(struct dentry *dentry { struct config_item * item = NULL; - spin_lock(&dcache_lock); + spin_lock(&dentry->d_lock); if (!d_unhashed(dentry)) { struct configfs_dirent * sd = dentry->d_fsdata; if (sd->s_type & CONFIGFS_ITEM_LINK) { @@ -129,7 +129,7 @@ static inline struct config_item *configfs_get_config_item(struct dentry *dentry } else item = config_item_get(sd->s_element); } - spin_unlock(&dcache_lock); + spin_unlock(&dentry->d_lock); return item; } diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c index 0b502f80c69..36637a8c1ed 100644 --- a/fs/configfs/dir.c +++ b/fs/configfs/dir.c @@ -67,7 +67,7 @@ static void configfs_d_iput(struct dentry * dentry, * We _must_ delete our dentries on last dput, as the chain-to-parent * behavior is required to clear the parents of default_groups. */ -static int configfs_d_delete(struct dentry *dentry) +static int configfs_d_delete(const struct dentry *dentry) { return 1; } @@ -232,10 +232,8 @@ int configfs_make_dirent(struct configfs_dirent * parent_sd, sd->s_mode = mode; sd->s_dentry = dentry; - if (dentry) { + if (dentry) dentry->d_fsdata = configfs_get(sd); - dentry->d_op = &configfs_dentry_ops; - } return 0; } @@ -278,7 +276,6 @@ static int create_dir(struct config_item * k, struct dentry * p, error = configfs_create(d, mode, init_dir); if (!error) { inc_nlink(p->d_inode); - (d)->d_op = &configfs_dentry_ops; } else { struct configfs_dirent *sd = d->d_fsdata; if (sd) { @@ -371,9 +368,7 @@ int configfs_create_link(struct configfs_symlink *sl, CONFIGFS_ITEM_LINK); if (!err) { err = configfs_create(dentry, mode, init_symlink); - if (!err) - dentry->d_op = &configfs_dentry_ops; - else { + if (err) { struct configfs_dirent *sd = dentry->d_fsdata; if (sd) { spin_lock(&configfs_dirent_lock); @@ -399,8 +394,7 @@ static void remove_dir(struct dentry * d) if (d->d_inode) simple_rmdir(parent->d_inode,d); - pr_debug(" o %s removing done (%d)\n",d->d_name.name, - atomic_read(&d->d_count)); + pr_debug(" o %s removing done (%d)\n",d->d_name.name, d->d_count); dput(parent); } @@ -448,7 +442,7 @@ static int configfs_attach_attr(struct configfs_dirent * sd, struct dentry * den return error; } - dentry->d_op = &configfs_dentry_ops; + d_set_d_op(dentry, &configfs_dentry_ops); d_rehash(dentry); return 0; @@ -493,7 +487,11 @@ static struct dentry * configfs_lookup(struct inode *dir, * If it doesn't exist and it isn't a NOT_PINNED item, * it must be negative. */ - return simple_lookup(dir, dentry, nd); + if (dentry->d_name.len > NAME_MAX) + return ERR_PTR(-ENAMETOOLONG); + d_set_d_op(dentry, &configfs_dentry_ops); + d_add(dentry, NULL); + return NULL; } out: @@ -685,6 +683,7 @@ static int create_default_group(struct config_group *parent_group, ret = -ENOMEM; child = d_alloc(parent, &name); if (child) { + d_set_d_op(child, &configfs_dentry_ops); d_add(child, NULL); ret = configfs_attach_group(&parent_group->cg_item, @@ -1682,6 +1681,7 @@ int configfs_register_subsystem(struct configfs_subsystem *subsys) err = -ENOMEM; dentry = d_alloc(configfs_sb->s_root, &name); if (dentry) { + d_set_d_op(dentry, &configfs_dentry_ops); d_add(dentry, NULL); err = configfs_attach_group(sd->s_element, &group->cg_item, diff --git a/fs/configfs/inode.c b/fs/configfs/inode.c index 253476d78ed..c83f4768eea 100644 --- a/fs/configfs/inode.c +++ b/fs/configfs/inode.c @@ -250,18 +250,14 @@ void configfs_drop_dentry(struct configfs_dirent * sd, struct dentry * parent) struct dentry * dentry = sd->s_dentry; if (dentry) { - spin_lock(&dcache_lock); spin_lock(&dentry->d_lock); if (!(d_unhashed(dentry) && dentry->d_inode)) { - dget_locked(dentry); + dget_dlock(dentry); __d_drop(dentry); spin_unlock(&dentry->d_lock); - spin_unlock(&dcache_lock); simple_unlink(parent->d_inode, dentry); - } else { + } else spin_unlock(&dentry->d_lock); - spin_unlock(&dcache_lock); - } } } diff --git a/fs/dcache.c b/fs/dcache.c index 23702a9d4e6..5699d4c027c 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -33,20 +33,58 @@ #include <linux/bootmem.h> #include <linux/fs_struct.h> #include <linux/hardirq.h> +#include <linux/bit_spinlock.h> +#include <linux/rculist_bl.h> #include "internal.h" +/* + * Usage: + * dcache->d_inode->i_lock protects: + * - i_dentry, d_alias, d_inode of aliases + * dcache_hash_bucket lock protects: + * - the dcache hash table + * s_anon bl list spinlock protects: + * - the s_anon list (see __d_drop) + * dcache_lru_lock protects: + * - the dcache lru lists and counters + * d_lock protects: + * - d_flags + * - d_name + * - d_lru + * - d_count + * - d_unhashed() + * - d_parent and d_subdirs + * - childrens' d_child and d_parent + * - d_alias, d_inode + * + * Ordering: + * dentry->d_inode->i_lock + * dentry->d_lock + * dcache_lru_lock + * dcache_hash_bucket lock + * s_anon lock + * + * If there is an ancestor relationship: + * dentry->d_parent->...->d_parent->d_lock + * ... + * dentry->d_parent->d_lock + * dentry->d_lock + * + * If no ancestor relationship: + * if (dentry1 < dentry2) + * dentry1->d_lock + * dentry2->d_lock + */ int sysctl_vfs_cache_pressure __read_mostly = 100; EXPORT_SYMBOL_GPL(sysctl_vfs_cache_pressure); - __cacheline_aligned_in_smp DEFINE_SPINLOCK(dcache_lock); +static __cacheline_aligned_in_smp DEFINE_SPINLOCK(dcache_lru_lock); __cacheline_aligned_in_smp DEFINE_SEQLOCK(rename_lock); -EXPORT_SYMBOL(dcache_lock); +EXPORT_SYMBOL(rename_lock); static struct kmem_cache *dentry_cache __read_mostly; -#define DNAME_INLINE_LEN (sizeof(struct dentry)-offsetof(struct dentry,d_iname)) - /* * This is the single most critical data structure when it comes * to the dcache: the hashtable for lookups. Somebody should try @@ -60,22 +98,51 @@ static struct kmem_cache *dentry_cache __read_mostly; static unsigned int d_hash_mask __read_mostly; static unsigned int d_hash_shift __read_mostly; -static struct hlist_head *dentry_hashtable __read_mostly; + +struct dcache_hash_bucket { + struct hlist_bl_head head; +}; +static struct dcache_hash_bucket *dentry_hashtable __read_mostly; + +static inline struct dcache_hash_bucket *d_hash(struct dentry *parent, + unsigned long hash) +{ + hash += ((unsigned long) parent ^ GOLDEN_RATIO_PRIME) / L1_CACHE_BYTES; + hash = hash ^ ((hash ^ GOLDEN_RATIO_PRIME) >> D_HASHBITS); + return dentry_hashtable + (hash & D_HASHMASK); +} + +static inline void spin_lock_bucket(struct dcache_hash_bucket *b) +{ + bit_spin_lock(0, (unsigned long *)&b->head.first); +} + +static inline void spin_unlock_bucket(struct dcache_hash_bucket *b) +{ + __bit_spin_unlock(0, (unsigned long *)&b->head.first); +} /* Statistics gathering. */ struct dentry_stat_t dentry_stat = { .age_limit = 45, }; -static struct percpu_counter nr_dentry __cacheline_aligned_in_smp; -static struct percpu_counter nr_dentry_unused __cacheline_aligned_in_smp; +static DEFINE_PER_CPU(unsigned int, nr_dentry); #if defined(CONFIG_SYSCTL) && defined(CONFIG_PROC_FS) +static int get_nr_dentry(void) +{ + int i; + int sum = 0; + for_each_possible_cpu(i) + sum += per_cpu(nr_dentry, i); + return sum < 0 ? 0 : sum; +} + int proc_nr_dentry(ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { - dentry_stat.nr_dentry = percpu_counter_sum_positive(&nr_dentry); - dentry_stat.nr_unused = percpu_counter_sum_positive(&nr_dentry_unused); + dentry_stat.nr_dentry = get_nr_dentry(); return proc_dointvec(table, write, buffer, lenp, ppos); } #endif @@ -91,35 +158,50 @@ static void __d_free(struct rcu_head *head) } /* - * no dcache_lock, please. + * no locks, please. */ static void d_free(struct dentry *dentry) { - percpu_counter_dec(&nr_dentry); + BUG_ON(dentry->d_count); + this_cpu_dec(nr_dentry); if (dentry->d_op && dentry->d_op->d_release) dentry->d_op->d_release(dentry); /* if dentry was never inserted into hash, immediate free is OK */ - if (hlist_unhashed(&dentry->d_hash)) + if (hlist_bl_unhashed(&dentry->d_hash)) __d_free(&dentry->d_u.d_rcu); else call_rcu(&dentry->d_u.d_rcu, __d_free); } +/** + * dentry_rcuwalk_barrier - invalidate in-progress rcu-walk lookups + * After this call, in-progress rcu-walk path lookup will fail. This + * should be called after unhashing, and after changing d_inode (if + * the dentry has not already been unhashed). + */ +static inline void dentry_rcuwalk_barrier(struct dentry *dentry) +{ + assert_spin_locked(&dentry->d_lock); + /* Go through a barrier */ + write_seqcount_barrier(&dentry->d_seq); +} + /* * Release the dentry's inode, using the filesystem - * d_iput() operation if defined. + * d_iput() operation if defined. Dentry has no refcount + * and is unhashed. */ static void dentry_iput(struct dentry * dentry) __releases(dentry->d_lock) - __releases(dcache_lock) + __releases(dentry->d_inode->i_lock) { struct inode *inode = dentry->d_inode; if (inode) { dentry->d_inode = NULL; list_del_init(&dentry->d_alias); spin_unlock(&dentry->d_lock); - spin_unlock(&dcache_lock); + spin_unlock(&inode->i_lock); if (!inode->i_nlink) fsnotify_inoderemove(inode); if (dentry->d_op && dentry->d_op->d_iput) @@ -128,40 +210,72 @@ static void dentry_iput(struct dentry * dentry) iput(inode); } else { spin_unlock(&dentry->d_lock); - spin_unlock(&dcache_lock); } } /* - * dentry_lru_(add|del|move_tail) must be called with dcache_lock held. + * Release the dentry's inode, using the filesystem + * d_iput() operation if defined. dentry remains in-use. + */ +static void dentry_unlink_inode(struct dentry * dentry) + __releases(dentry->d_lock) + __releases(dentry->d_inode->i_lock) +{ + struct inode *inode = dentry->d_inode; + dentry->d_inode = NULL; + list_del_init(&dentry->d_alias); + dentry_rcuwalk_barrier(dentry); + spin_unlock(&dentry->d_lock); + spin_unlock(&inode->i_lock); + if (!inode->i_nlink) + fsnotify_inoderemove(inode); + if (dentry->d_op && dentry->d_op->d_iput) + dentry->d_op->d_iput(dentry, inode); + else + iput(inode); +} + +/* + * dentry_lru_(add|del|move_tail) must be called with d_lock held. */ static void dentry_lru_add(struct dentry *dentry) { if (list_empty(&dentry->d_lru)) { + spin_lock(&dcache_lru_lock); list_add(&dentry->d_lru, &dentry->d_sb->s_dentry_lru); dentry->d_sb->s_nr_dentry_unused++; - percpu_counter_inc(&nr_dentry_unused); + dentry_stat.nr_unused++; + spin_unlock(&dcache_lru_lock); } } +static void __dentry_lru_del(struct dentry *dentry) +{ + list_del_init(&dentry->d_lru); + dentry->d_sb->s_nr_dentry_unused--; + dentry_stat.nr_unused--; +} + static void dentry_lru_del(struct dentry *dentry) { if (!list_empty(&dentry->d_lru)) { - list_del_init(&dentry->d_lru); - dentry->d_sb->s_nr_dentry_unused--; - percpu_counter_dec(&nr_dentry_unused); + spin_lock(&dcache_lru_lock); + __dentry_lru_del(dentry); + spin_unlock(&dcache_lru_lock); } } static void dentry_lru_move_tail(struct dentry *dentry) { + spin_lock(&dcache_lru_lock); if (list_empty(&dentry->d_lru)) { list_add_tail(&dentry->d_lru, &dentry->d_sb->s_dentry_lru); dentry->d_sb->s_nr_dentry_unused++; - percpu_counter_inc(&nr_dentry_unused); + dentry_stat.nr_unused++; } else { list_move_tail(&dentry->d_lru, &dentry->d_sb->s_dentry_lru); } + spin_unlock(&dcache_lru_lock); } /** @@ -171,22 +285,115 @@ static void dentry_lru_move_tail(struct dentry *dentry) * The dentry must already be unhashed and removed from the LRU. * * If this is the root of the dentry tree, return NULL. + * + * dentry->d_lock and parent->d_lock must be held by caller, and are dropped by + * d_kill. */ -static struct dentry *d_kill(struct dentry *dentry) +static struct dentry *d_kill(struct dentry *dentry, struct dentry *parent) __releases(dentry->d_lock) - __releases(dcache_lock) + __releases(parent->d_lock) + __releases(dentry->d_inode->i_lock) { - struct dentry *parent; - + dentry->d_parent = NULL; list_del(&dentry->d_u.d_child); - /*drops the locks, at that point nobody can reach this dentry */ + if (parent) + spin_unlock(&parent->d_lock); dentry_iput(dentry); + /* + * dentry_iput drops the locks, at which point nobody (except + * transient RCU lookups) can reach this dentry. + */ + d_free(dentry); + return parent; +} + +/** + * d_drop - drop a dentry + * @dentry: dentry to drop + * + * d_drop() unhashes the entry from the parent dentry hashes, so that it won't + * be found through a VFS lookup any more. Note that this is different from + * deleting the dentry - d_delete will try to mark the dentry negative if + * possible, giving a successful _negative_ lookup, while d_drop will + * just make the cache lookup fail. + * + * d_drop() is used mainly for stuff that wants to invalidate a dentry for some + * reason (NFS timeouts or autofs deletes). + * + * __d_drop requires dentry->d_lock. + */ +void __d_drop(struct dentry *dentry) +{ + if (!(dentry->d_flags & DCACHE_UNHASHED)) { + if (unlikely(dentry->d_flags & DCACHE_DISCONNECTED)) { + bit_spin_lock(0, + (unsigned long *)&dentry->d_sb->s_anon.first); + dentry->d_flags |= DCACHE_UNHASHED; + hlist_bl_del_init(&dentry->d_hash); + __bit_spin_unlock(0, + (unsigned long *)&dentry->d_sb->s_anon.first); + } else { + struct dcache_hash_bucket *b; + b = d_hash(dentry->d_parent, dentry->d_name.hash); + spin_lock_bucket(b); + /* + * We may not actually need to put DCACHE_UNHASHED + * manipulations under the hash lock, but follow + * the principle of least surprise. + */ + dentry->d_flags |= DCACHE_UNHASHED; + hlist_bl_del_rcu(&dentry->d_hash); + spin_unlock_bucket(b); + dentry_rcuwalk_barrier(dentry); + } + } +} +EXPORT_SYMBOL(__d_drop); + +void d_drop(struct dentry *dentry) +{ + spin_lock(&dentry->d_lock); + __d_drop(dentry); + spin_unlock(&dentry->d_lock); +} +EXPORT_SYMBOL(d_drop); + +/* + * Finish off a dentry we've decided to kill. + * dentry->d_lock must be held, returns with it unlocked. + * If ref is non-zero, then decrement the refcount too. + * Returns dentry requiring refcount drop, or NULL if we're done. + */ +static inline struct dentry *dentry_kill(struct dentry *dentry, int ref) + __releases(dentry->d_lock) +{ + struct inode *inode; + struct dentry *parent; + + inode = dentry->d_inode; + if (inode && !spin_trylock(&inode->i_lock)) { +relock: + spin_unlock(&dentry->d_lock); + cpu_relax(); + return dentry; /* try again with same dentry */ + } if (IS_ROOT(dentry)) parent = NULL; else parent = dentry->d_parent; - d_free(dentry); - return parent; + if (parent && !spin_trylock(&parent->d_lock)) { + if (inode) + spin_unlock(&inode->i_lock); + goto relock; + } + + if (ref) + dentry->d_count--; + /* if dentry was on the d_lru list delete it from there */ + dentry_lru_del(dentry); + /* if it was on the hash then remove it */ + __d_drop(dentry); + return d_kill(dentry, parent); } /* @@ -214,34 +421,26 @@ static struct dentry *d_kill(struct dentry *dentry) * call the dentry unlink method as well as removing it from the queues and * releasing its resources. If the parent dentries were scheduled for release * they too may now get deleted. - * - * no dcache lock, please. */ - void dput(struct dentry *dentry) { if (!dentry) return; repeat: - if (atomic_read(&dentry->d_count) == 1) + if (dentry->d_count == 1) might_sleep(); - if (!atomic_dec_and_lock(&dentry->d_count, &dcache_lock)) - return; - spin_lock(&dentry->d_lock); - if (atomic_read(&dentry->d_count)) { + BUG_ON(!dentry->d_count); + if (dentry->d_count > 1) { + dentry->d_count--; spin_unlock(&dentry->d_lock); - spin_unlock(&dcache_lock); return; } - /* - * AV: ->d_delete() is _NOT_ allowed to block now. - */ - if (dentry->d_op && dentry->d_op->d_delete) { + if (dentry->d_flags & DCACHE_OP_DELETE) { if (dentry->d_op->d_delete(dentry)) - goto unhash_it; + goto kill_it; } /* Unreachable? Get rid of it */ @@ -252,16 +451,12 @@ repeat: dentry->d_flags |= DCACHE_REFERENCED; dentry_lru_add(dentry); - spin_unlock(&dentry->d_lock); - spin_unlock(&dcache_lock); + dentry->d_count--; + spin_unlock(&dentry->d_lock); return; -unhash_it: - __d_drop(dentry); kill_it: - /* if dentry was on the d_lru list delete it from there */ - dentry_lru_del(dentry); - dentry = d_kill(dentry); + dentry = dentry_kill(dentry, 1); if (dentry) goto repeat; } @@ -284,9 +479,9 @@ int d_invalidate(struct dentry * dentry) /* * If it's already been dropped, return OK. */ - spin_lock(&dcache_lock); + spin_lock(&dentry->d_lock); if (d_unhashed(dentry)) { - spin_unlock(&dcache_lock); + spin_unlock(&dentry->d_lock); return 0; } /* @@ -294,9 +489,9 @@ int d_invalidate(struct dentry * dentry) * to get rid of unused child entries. */ if (!list_empty(&dentry->d_subdirs)) { - spin_unlock(&dcache_lock); + spin_unlock(&dentry->d_lock); shrink_dcache_parent(dentry); - spin_lock(&dcache_lock); + spin_lock(&dentry->d_lock); } /* @@ -309,35 +504,61 @@ int d_invalidate(struct dentry * dentry) * we might still populate it if it was a * working directory or similar). */ - spin_lock(&dentry->d_lock); - if (atomic_read(&dentry->d_count) > 1) { + if (dentry->d_count > 1) { if (dentry->d_inode && S_ISDIR(dentry->d_inode->i_mode)) { spin_unlock(&dentry->d_lock); - spin_unlock(&dcache_lock); return -EBUSY; } } __d_drop(dentry); spin_unlock(&dentry->d_lock); - spin_unlock(&dcache_lock); return 0; } EXPORT_SYMBOL(d_invalidate); -/* This should be called _only_ with dcache_lock held */ -static inline struct dentry * __dget_locked(struct dentry *dentry) +/* This must be called with d_lock held */ +static inline void __dget_dlock(struct dentry *dentry) { - atomic_inc(&dentry->d_count); - dentry_lru_del(dentry); - return dentry; + dentry->d_count++; } -struct dentry * dget_locked(struct dentry *dentry) +static inline void __dget(struct dentry *dentry) { - return __dget_locked(dentry); + spin_lock(&dentry->d_lock); + __dget_dlock(dentry); + spin_unlock(&dentry->d_lock); +} + +struct dentry *dget_parent(struct dentry *dentry) +{ + struct dentry *ret; + +repeat: + /* + * Don't need rcu_dereference because we re-check it was correct under + * the lock. + */ + rcu_read_lock(); + ret = dentry->d_parent; + if (!ret) { + rcu_read_unlock(); + goto out; + } + spin_lock(&ret->d_lock); + if (unlikely(ret != dentry->d_parent)) { + spin_unlock(&ret->d_lock); + rcu_read_unlock(); + goto repeat; + } + rcu_read_unlock(); + BUG_ON(!ret->d_count); + ret->d_count++; + spin_unlock(&ret->d_lock); +out: + return ret; } -EXPORT_SYMBOL(dget_locked); +EXPORT_SYMBOL(dget_parent); /** * d_find_alias - grab a hashed alias of inode @@ -355,42 +576,51 @@ EXPORT_SYMBOL(dget_locked); * any other hashed alias over that one unless @want_discon is set, * in which case only return an IS_ROOT, DCACHE_DISCONNECTED alias. */ - -static struct dentry * __d_find_alias(struct inode *inode, int want_discon) +static struct dentry *__d_find_alias(struct inode *inode, int want_discon) { - struct list_head *head, *next, *tmp; - struct dentry *alias, *discon_alias=NULL; + struct dentry *alias, *discon_alias; - head = &inode->i_dentry; - next = inode->i_dentry.next; - while (next != head) { - tmp = next; - next = tmp->next; - prefetch(next); - alias = list_entry(tmp, struct dentry, d_alias); +again: + discon_alias = NULL; + list_for_each_entry(alias, &inode->i_dentry, d_alias) { + spin_lock(&alias->d_lock); if (S_ISDIR(inode->i_mode) || !d_unhashed(alias)) { if (IS_ROOT(alias) && - (alias->d_flags & DCACHE_DISCONNECTED)) + (alias->d_flags & DCACHE_DISCONNECTED)) { discon_alias = alias; - else if (!want_discon) { - __dget_locked(alias); + } else if (!want_discon) { + __dget_dlock(alias); + spin_unlock(&alias->d_lock); + return alias; + } + } + spin_unlock(&alias->d_lock); + } + if (discon_alias) { + alias = discon_alias; + spin_lock(&alias->d_lock); + if (S_ISDIR(inode->i_mode) || !d_unhashed(alias)) { + if (IS_ROOT(alias) && + (alias->d_flags & DCACHE_DISCONNECTED)) { + __dget_dlock(alias); + spin_unlock(&alias->d_lock); return alias; } } + spin_unlock(&alias->d_lock); + goto again; } - if (discon_alias) - __dget_locked(discon_alias); - return discon_alias; + return NULL; } -struct dentry * d_find_alias(struct inode *inode) +struct dentry *d_find_alias(struct inode *inode) { struct dentry *de = NULL; if (!list_empty(&inode->i_dentry)) { - spin_lock(&dcache_lock); + spin_lock(&inode->i_lock); de = __d_find_alias(inode, 0); - spin_unlock(&dcache_lock); + spin_unlock(&inode->i_lock); } return de; } @@ -404,54 +634,61 @@ void d_prune_aliases(struct inode *inode) { struct dentry *dentry; restart: - spin_lock(&dcache_lock); + spin_lock(&inode->i_lock); list_for_each_entry(dentry, &inode->i_dentry, d_alias) { spin_lock(&dentry->d_lock); - if (!atomic_read(&dentry->d_count)) { - __dget_locked(dentry); + if (!dentry->d_count) { + __dget_dlock(dentry); __d_drop(dentry); spin_unlock(&dentry->d_lock); - spin_unlock(&dcache_lock); + spin_unlock(&inode->i_lock); dput(dentry); goto restart; } spin_unlock(&dentry->d_lock); } - spin_unlock(&dcache_lock); + spin_unlock(&inode->i_lock); } EXPORT_SYMBOL(d_prune_aliases); /* - * Throw away a dentry - free the inode, dput the parent. This requires that - * the LRU list has already been removed. + * Try to throw away a dentry - free the inode, dput the parent. + * Requires dentry->d_lock is held, and dentry->d_count == 0. + * Releases dentry->d_lock. * - * Try to prune ancestors as well. This is necessary to prevent - * quadratic behavior of shrink_dcache_parent(), but is also expected - * to be beneficial in reducing dentry cache fragmentation. + * This may fail if locks cannot be acquired no problem, just try again. */ -static void prune_one_dentry(struct dentry * dentry) +static void try_prune_one_dentry(struct dentry *dentry) __releases(dentry->d_lock) - __releases(dcache_lock) - __acquires(dcache_lock) { - __d_drop(dentry); - dentry = d_kill(dentry); + struct dentry *parent; + parent = dentry_kill(dentry, 0); /* - * Prune ancestors. Locking is simpler than in dput(), - * because dcache_lock needs to be taken anyway. + * If dentry_kill returns NULL, we have nothing more to do. + * if it returns the same dentry, trylocks failed. In either + * case, just loop again. + * + * Otherwise, we need to prune ancestors too. This is necessary + * to prevent quadratic behavior of shrink_dcache_parent(), but + * is also expected to be beneficial in reducing dentry cache + * fragmentation. */ - spin_lock(&dcache_lock); + if (!parent) + return; + if (parent == dentry) + return; + + /* Prune ancestors. */ + dentry = parent; while (dentry) { - if (!atomic_dec_and_lock(&dentry->d_count, &dentry->d_lock)) + spin_lock(&dentry->d_lock); + if (dentry->d_count > 1) { + dentry->d_count--; + spin_unlock(&dentry->d_lock); return; - - if (dentry->d_op && dentry->d_op->d_delete) - dentry->d_op->d_delete(dentry); - dentry_lru_del(dentry); - __d_drop(dentry); - dentry = d_kill(dentry); - spin_lock(&dcache_lock); + } + dentry = dentry_kill(dentry, 1); } } @@ -459,24 +696,35 @@ static void shrink_dentry_list(struct list_head *list) { struct dentry *dentry; - while (!list_empty(list)) { - dentry = list_entry(list->prev, struct dentry, d_lru); - dentry_lru_del(dentry); + rcu_read_lock(); + for (;;) { + dentry = list_entry_rcu(list->prev, struct dentry, d_lru); + if (&dentry->d_lru == list) + break; /* empty */ + spin_lock(&dentry->d_lock); + if (dentry != list_entry(list->prev, struct dentry, d_lru)) { + spin_unlock(&dentry->d_lock); + continue; + } /* * We found an inuse dentry which was not removed from * the LRU because of laziness during lookup. Do not free * it - just keep it off the LRU list. */ - spin_lock(&dentry->d_lock); - if (atomic_read(&dentry->d_count)) { + if (dentry->d_count) { + dentry_lru_del(dentry); spin_unlock(&dentry->d_lock); continue; } - prune_one_dentry(dentry); - /* dentry->d_lock was dropped in prune_one_dentry() */ - cond_resched_lock(&dcache_lock); + + rcu_read_unlock(); + + try_prune_one_dentry(dentry); + + rcu_read_lock(); } + rcu_read_unlock(); } /** @@ -495,42 +743,44 @@ static void __shrink_dcache_sb(struct super_block *sb, int *count, int flags) LIST_HEAD(tmp); int cnt = *count; - spin_lock(&dcache_lock); +relock: + spin_lock(&dcache_lru_lock); while (!list_empty(&sb->s_dentry_lru)) { dentry = list_entry(sb->s_dentry_lru.prev, struct dentry, d_lru); BUG_ON(dentry->d_sb != sb); + if (!spin_trylock(&dentry->d_lock)) { + spin_unlock(&dcache_lru_lock); + cpu_relax(); + goto relock; + } + /* * If we are honouring the DCACHE_REFERENCED flag and the * dentry has this flag set, don't free it. Clear the flag * and put it back on the LRU. */ - if (flags & DCACHE_REFERENCED) { - spin_lock(&dentry->d_lock); - if (dentry->d_flags & DCACHE_REFERENCED) { - dentry->d_flags &= ~DCACHE_REFERENCED; - list_move(&dentry->d_lru, &referenced); - spin_unlock(&dentry->d_lock); - cond_resched_lock(&dcache_lock); - continue; - } + if (flags & DCACHE_REFERENCED && + dentry->d_flags & DCACHE_REFERENCED) { + dentry->d_flags &= ~DCACHE_REFERENCED; + list_move(&dentry->d_lru, &referenced); spin_unlock(&dentry->d_lock); + } else { + list_move_tail(&dentry->d_lru, &tmp); + spin_unlock(&dentry->d_lock); + if (!--cnt) + break; } - - list_move_tail(&dentry->d_lru, &tmp); - if (!--cnt) - break; - cond_resched_lock(&dcache_lock); + cond_resched_lock(&dcache_lru_lock); } - - *count = cnt; - shrink_dentry_list(&tmp); - if (!list_empty(&referenced)) list_splice(&referenced, &sb->s_dentry_lru); - spin_unlock(&dcache_lock); + spin_unlock(&dcache_lru_lock); + shrink_dentry_list(&tmp); + + *count = cnt; } /** @@ -546,13 +796,12 @@ static void prune_dcache(int count) { struct super_block *sb, *p = NULL; int w_count; - int unused = percpu_counter_sum_positive(&nr_dentry_unused); + int unused = dentry_stat.nr_unused; int prune_ratio; int pruned; if (unused == 0 || count == 0) return; - spin_lock(&dcache_lock); if (count >= unused) prune_ratio = 1; else @@ -589,11 +838,9 @@ static void prune_dcache(int count) if (down_read_trylock(&sb->s_umount)) { if ((sb->s_root != NULL) && (!list_empty(&sb->s_dentry_lru))) { - spin_unlock(&dcache_lock); __shrink_dcache_sb(sb, &w_count, DCACHE_REFERENCED); pruned -= w_count; - spin_lock(&dcache_lock); } up_read(&sb->s_umount); } @@ -609,7 +856,6 @@ static void prune_dcache(int count) if (p) __put_super(p); spin_unlock(&sb_lock); - spin_unlock(&dcache_lock); } /** @@ -623,12 +869,14 @@ void shrink_dcache_sb(struct super_block *sb) { LIST_HEAD(tmp); - spin_lock(&dcache_lock); + spin_lock(&dcache_lru_lock); while (!list_empty(&sb->s_dentry_lru)) { list_splice_init(&sb->s_dentry_lru, &tmp); + spin_unlock(&dcache_lru_lock); shrink_dentry_list(&tmp); + spin_lock(&dcache_lru_lock); } - spin_unlock(&dcache_lock); + spin_unlock(&dcache_lru_lock); } EXPORT_SYMBOL(shrink_dcache_sb); @@ -645,10 +893,10 @@ static void shrink_dcache_for_umount_subtree(struct dentry *dentry) BUG_ON(!IS_ROOT(dentry)); /* detach this root from the system */ - spin_lock(&dcache_lock); + spin_lock(&dentry->d_lock); dentry_lru_del(dentry); __d_drop(dentry); - spin_unlock(&dcache_lock); + spin_unlock(&dentry->d_lock); for (;;) { /* descend to the first leaf in the current subtree */ @@ -657,14 +905,16 @@ static void shrink_dcache_for_umount_subtree(struct dentry *dentry) /* this is a branch with children - detach all of them * from the system in one go */ - spin_lock(&dcache_lock); + spin_lock(&dentry->d_lock); list_for_each_entry(loop, &dentry->d_subdirs, d_u.d_child) { + spin_lock_nested(&loop->d_lock, + DENTRY_D_LOCK_NESTED); dentry_lru_del(loop); __d_drop(loop); - cond_resched_lock(&dcache_lock); + spin_unlock(&loop->d_lock); } - spin_unlock(&dcache_lock); + spin_unlock(&dentry->d_lock); /* move to the first child */ dentry = list_entry(dentry->d_subdirs.next, @@ -676,7 +926,7 @@ static void shrink_dcache_for_umount_subtree(struct dentry *dentry) do { struct inode *inode; - if (atomic_read(&dentry->d_count) != 0) { + if (dentry->d_count != 0) { printk(KERN_ERR "BUG: Dentry %p{i=%lx,n=%s}" " still in use (%d)" @@ -685,20 +935,23 @@ static void shrink_dcache_for_umount_subtree(struct dentry *dentry) dentry->d_inode ? dentry->d_inode->i_ino : 0UL, dentry->d_name.name, - atomic_read(&dentry->d_count), + dentry->d_count, dentry->d_sb->s_type->name, dentry->d_sb->s_id); BUG(); } - if (IS_ROOT(dentry)) + if (IS_ROOT(dentry)) { parent = NULL; - else { + list_del(&dentry->d_u.d_child); + } else { parent = dentry->d_parent; - atomic_dec(&parent->d_count); + spin_lock(&parent->d_lock); + parent->d_count--; + list_del(&dentry->d_u.d_child); + spin_unlock(&parent->d_lock); } - list_del(&dentry->d_u.d_child); detached++; inode = dentry->d_inode; @@ -728,8 +981,7 @@ static void shrink_dcache_for_umount_subtree(struct dentry *dentry) /* * destroy the dentries attached to a superblock on unmounting - * - we don't need to use dentry->d_lock, and only need dcache_lock when - * removing the dentry from the system lists and hashes because: + * - we don't need to use dentry->d_lock because: * - the superblock is detached from all mountings and open files, so the * dentry trees will not be rearranged by the VFS * - s_umount is write-locked, so the memory pressure shrinker will ignore @@ -746,11 +998,13 @@ void shrink_dcache_for_umount(struct super_block *sb) dentry = sb->s_root; sb->s_root = NULL; - atomic_dec(&dentry->d_count); + spin_lock(&dentry->d_lock); + dentry->d_count--; + spin_unlock(&dentry->d_lock); shrink_dcache_for_umount_subtree(dentry); - while (!hlist_empty(&sb->s_anon)) { - dentry = hlist_entry(sb->s_anon.first, struct dentry, d_hash); + while (!hlist_bl_empty(&sb->s_anon)) { + dentry = hlist_bl_entry(hlist_bl_first(&sb->s_anon), struct dentry, d_hash); shrink_dcache_for_umount_subtree(dentry); } } @@ -768,15 +1022,20 @@ void shrink_dcache_for_umount(struct super_block *sb) * Return true if the parent or its subdirectories contain * a mount point */ - int have_submounts(struct dentry *parent) { - struct dentry *this_parent = parent; + struct dentry *this_parent; struct list_head *next; + unsigned seq; + int locked = 0; + + seq = read_seqbegin(&rename_lock); +again: + this_parent = parent; - spin_lock(&dcache_lock); if (d_mountpoint(parent)) goto positive; + spin_lock(&this_parent->d_lock); repeat: next = this_parent->d_subdirs.next; resume: @@ -784,27 +1043,65 @@ resume: struct list_head *tmp = next; struct dentry *dentry = list_entry(tmp, struct dentry, d_u.d_child); next = tmp->next; + + spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); /* Have we found a mount point ? */ - if (d_mountpoint(dentry)) + if (d_mountpoint(dentry)) { + spin_unlock(&dentry->d_lock); + spin_unlock(&this_parent->d_lock); goto positive; + } if (!list_empty(&dentry->d_subdirs)) { + spin_unlock(&this_parent->d_lock); + spin_release(&dentry->d_lock.dep_map, 1, _RET_IP_); this_parent = dentry; + spin_acquire(&this_parent->d_lock.dep_map, 0, 1, _RET_IP_); goto repeat; } + spin_unlock(&dentry->d_lock); } /* * All done at this level ... ascend and resume the search. */ if (this_parent != parent) { - next = this_parent->d_u.d_child.next; - this_parent = this_parent->d_parent; + struct dentry *tmp; + struct dentry *child; + + tmp = this_parent->d_parent; + rcu_read_lock(); + spin_unlock(&this_parent->d_lock); + child = this_parent; + this_parent = tmp; + spin_lock(&this_parent->d_lock); + /* might go back up the wrong parent if we have had a rename + * or deletion */ + if (this_parent != child->d_parent || + (!locked && read_seqretry(&rename_lock, seq))) { + spin_unlock(&this_parent->d_lock); + rcu_read_unlock(); + goto rename_retry; + } + rcu_read_unlock(); + next = child->d_u.d_child.next; goto resume; } - spin_unlock(&dcache_lock); + spin_unlock(&this_parent->d_lock); + if (!locked && read_seqretry(&rename_lock, seq)) + goto rename_retry; + if (locked) + write_sequnlock(&rename_lock); return 0; /* No mount points found in tree */ positive: - spin_unlock(&dcache_lock); + if (!locked && read_seqretry(&rename_lock, seq)) + goto rename_retry; + if (locked) + write_sequnlock(&rename_lock); return 1; + +rename_retry: + locked = 1; + write_seqlock(&rename_lock); + goto again; } EXPORT_SYMBOL(have_submounts); @@ -824,11 +1121,16 @@ EXPORT_SYMBOL(have_submounts); */ static int select_parent(struct dentry * parent) { - struct dentry *this_parent = parent; + struct dentry *this_parent; struct list_head *next; + unsigned seq; int found = 0; + int locked = 0; - spin_lock(&dcache_lock); + seq = read_seqbegin(&rename_lock); +again: + this_parent = parent; + spin_lock(&this_parent->d_lock); repeat: next = this_parent->d_subdirs.next; resume: @@ -837,11 +1139,13 @@ resume: struct dentry *dentry = list_entry(tmp, struct dentry, d_u.d_child); next = tmp->next; + spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); + /* * move only zero ref count dentries to the end * of the unused list for prune_dcache */ - if (!atomic_read(&dentry->d_count)) { + if (!dentry->d_count) { dentry_lru_move_tail(dentry); found++; } else { @@ -853,28 +1157,63 @@ resume: * ensures forward progress). We'll be coming back to find * the rest. */ - if (found && need_resched()) + if (found && need_resched()) { + spin_unlock(&dentry->d_lock); goto out; + } /* * Descend a level if the d_subdirs list is non-empty. */ if (!list_empty(&dentry->d_subdirs)) { + spin_unlock(&this_parent->d_lock); + spin_release(&dentry->d_lock.dep_map, 1, _RET_IP_); this_parent = dentry; + spin_acquire(&this_parent->d_lock.dep_map, 0, 1, _RET_IP_); goto repeat; } + + spin_unlock(&dentry->d_lock); } /* * All done at this level ... ascend and resume the search. */ if (this_parent != parent) { - next = this_parent->d_u.d_child.next; - this_parent = this_parent->d_parent; + struct dentry *tmp; + struct dentry *child; + + tmp = this_parent->d_parent; + rcu_read_lock(); + spin_unlock(&this_parent->d_lock); + child = this_parent; + this_parent = tmp; + spin_lock(&this_parent->d_lock); + /* might go back up the wrong parent if we have had a rename + * or deletion */ + if (this_parent != child->d_parent || + (!locked && read_seqretry(&rename_lock, seq))) { + spin_unlock(&this_parent->d_lock); + rcu_read_unlock(); + goto rename_retry; + } + rcu_read_unlock(); + next = child->d_u.d_child.next; goto resume; } out: - spin_unlock(&dcache_lock); + spin_unlock(&this_parent->d_lock); + if (!locked && read_seqretry(&rename_lock, seq)) + goto rename_retry; + if (locked) + write_sequnlock(&rename_lock); return found; + +rename_retry: + if (found) + return found; + locked = 1; + write_seqlock(&rename_lock); + goto again; } /** @@ -908,16 +1247,13 @@ EXPORT_SYMBOL(shrink_dcache_parent); */ static int shrink_dcache_memory(struct shrinker *shrink, int nr, gfp_t gfp_mask) { - int nr_unused; - if (nr) { if (!(gfp_mask & __GFP_FS)) return -1; prune_dcache(nr); } - nr_unused = percpu_counter_sum_positive(&nr_dentry_unused); - return (nr_unused / 100) * sysctl_vfs_cache_pressure; + return (dentry_stat.nr_unused / 100) * sysctl_vfs_cache_pressure; } static struct shrinker dcache_shrinker = { @@ -960,38 +1296,52 @@ struct dentry *d_alloc(struct dentry * parent, const struct qstr *name) memcpy(dname, name->name, name->len); dname[name->len] = 0; - atomic_set(&dentry->d_count, 1); + dentry->d_count = 1; dentry->d_flags = DCACHE_UNHASHED; spin_lock_init(&dentry->d_lock); + seqcount_init(&dentry->d_seq); dentry->d_inode = NULL; dentry->d_parent = NULL; dentry->d_sb = NULL; dentry->d_op = NULL; dentry->d_fsdata = NULL; - dentry->d_mounted = 0; - INIT_HLIST_NODE(&dentry->d_hash); + INIT_HLIST_BL_NODE(&dentry->d_hash); INIT_LIST_HEAD(&dentry->d_lru); INIT_LIST_HEAD(&dentry->d_subdirs); INIT_LIST_HEAD(&dentry->d_alias); + INIT_LIST_HEAD(&dentry->d_u.d_child); if (parent) { - dentry->d_parent = dget(parent); + spin_lock(&parent->d_lock); + /* + * don't need child lock because it is not subject + * to concurrency here + */ + __dget_dlock(parent); + dentry->d_parent = parent; dentry->d_sb = parent->d_sb; - } else { - INIT_LIST_HEAD(&dentry->d_u.d_child); - } - - spin_lock(&dcache_lock); - if (parent) list_add(&dentry->d_u.d_child, &parent->d_subdirs); - spin_unlock(&dcache_lock); + spin_unlock(&parent->d_lock); + } - percpu_counter_inc(&nr_dentry); + this_cpu_inc(nr_dentry); return dentry; } EXPORT_SYMBOL(d_alloc); +struct dentry *d_alloc_pseudo(struct super_block *sb, const struct qstr *name) +{ + struct dentry *dentry = d_alloc(NULL, name); + if (dentry) { + dentry->d_sb = sb; + dentry->d_parent = dentry; + dentry->d_flags |= DCACHE_DISCONNECTED; + } + return dentry; +} +EXPORT_SYMBOL(d_alloc_pseudo); + struct dentry *d_alloc_name(struct dentry *parent, const char *name) { struct qstr q; @@ -1003,12 +1353,36 @@ struct dentry *d_alloc_name(struct dentry *parent, const char *name) } EXPORT_SYMBOL(d_alloc_name); -/* the caller must hold dcache_lock */ +void d_set_d_op(struct dentry *dentry, const struct dentry_operations *op) +{ + BUG_ON(dentry->d_op); + BUG_ON(dentry->d_flags & (DCACHE_OP_HASH | + DCACHE_OP_COMPARE | + DCACHE_OP_REVALIDATE | + DCACHE_OP_DELETE )); + dentry->d_op = op; + if (!op) + return; + if (op->d_hash) + dentry->d_flags |= DCACHE_OP_HASH; + if (op->d_compare) + dentry->d_flags |= DCACHE_OP_COMPARE; + if (op->d_revalidate) + dentry->d_flags |= DCACHE_OP_REVALIDATE; + if (op->d_delete) + dentry->d_flags |= DCACHE_OP_DELETE; + +} +EXPORT_SYMBOL(d_set_d_op); + static void __d_instantiate(struct dentry *dentry, struct inode *inode) { + spin_lock(&dentry->d_lock); if (inode) list_add(&dentry->d_alias, &inode->i_dentry); dentry->d_inode = inode; + dentry_rcuwalk_barrier(dentry); + spin_unlock(&dentry->d_lock); fsnotify_d_instantiate(dentry, inode); } @@ -1030,9 +1404,11 @@ static void __d_instantiate(struct dentry *dentry, struct inode *inode) void d_instantiate(struct dentry *entry, struct inode * inode) { BUG_ON(!list_empty(&entry->d_alias)); - spin_lock(&dcache_lock); + if (inode) + spin_lock(&inode->i_lock); __d_instantiate(entry, inode); - spin_unlock(&dcache_lock); + if (inode) + spin_unlock(&inode->i_lock); security_d_instantiate(entry, inode); } EXPORT_SYMBOL(d_instantiate); @@ -1069,15 +1445,18 @@ static struct dentry *__d_instantiate_unique(struct dentry *entry, list_for_each_entry(alias, &inode->i_dentry, d_alias) { struct qstr *qstr = &alias->d_name; + /* + * Don't need alias->d_lock here, because aliases with + * d_parent == entry->d_parent are not subject to name or + * parent changes, because the parent inode i_mutex is held. + */ if (qstr->hash != hash) continue; if (alias->d_parent != entry->d_parent) continue; - if (qstr->len != len) + if (dentry_cmp(qstr->name, qstr->len, name, len)) continue; - if (memcmp(qstr->name, name, len)) - continue; - dget_locked(alias); + __dget(alias); return alias; } @@ -1091,9 +1470,11 @@ struct dentry *d_instantiate_unique(struct dentry *entry, struct inode *inode) BUG_ON(!list_empty(&entry->d_alias)); - spin_lock(&dcache_lock); + if (inode) + spin_lock(&inode->i_lock); result = __d_instantiate_unique(entry, inode); - spin_unlock(&dcache_lock); + if (inode) + spin_unlock(&inode->i_lock); if (!result) { security_d_instantiate(entry, inode); @@ -1134,14 +1515,6 @@ struct dentry * d_alloc_root(struct inode * root_inode) } EXPORT_SYMBOL(d_alloc_root); -static inline struct hlist_head *d_hash(struct dentry *parent, - unsigned long hash) -{ - hash += ((unsigned long) parent ^ GOLDEN_RATIO_PRIME) / L1_CACHE_BYTES; - hash = hash ^ ((hash ^ GOLDEN_RATIO_PRIME) >> D_HASHBITS); - return dentry_hashtable + (hash & D_HASHMASK); -} - /** * d_obtain_alias - find or allocate a dentry for a given inode * @inode: inode to allocate the dentry for @@ -1182,10 +1555,11 @@ struct dentry *d_obtain_alias(struct inode *inode) } tmp->d_parent = tmp; /* make sure dput doesn't croak */ - spin_lock(&dcache_lock); + + spin_lock(&inode->i_lock); res = __d_find_alias(inode, 0); if (res) { - spin_unlock(&dcache_lock); + spin_unlock(&inode->i_lock); dput(tmp); goto out_iput; } @@ -1195,12 +1569,14 @@ struct dentry *d_obtain_alias(struct inode *inode) tmp->d_sb = inode->i_sb; tmp->d_inode = inode; tmp->d_flags |= DCACHE_DISCONNECTED; - tmp->d_flags &= ~DCACHE_UNHASHED; list_add(&tmp->d_alias, &inode->i_dentry); - hlist_add_head(&tmp->d_hash, &inode->i_sb->s_anon); + bit_spin_lock(0, (unsigned long *)&tmp->d_sb->s_anon.first); + tmp->d_flags &= ~DCACHE_UNHASHED; + hlist_bl_add_head(&tmp->d_hash, &tmp->d_sb->s_anon); + __bit_spin_unlock(0, (unsigned long *)&tmp->d_sb->s_anon.first); spin_unlock(&tmp->d_lock); + spin_unlock(&inode->i_lock); - spin_unlock(&dcache_lock); return tmp; out_iput: @@ -1230,18 +1606,18 @@ struct dentry *d_splice_alias(struct inode *inode, struct dentry *dentry) struct dentry *new = NULL; if (inode && S_ISDIR(inode->i_mode)) { - spin_lock(&dcache_lock); + spin_lock(&inode->i_lock); new = __d_find_alias(inode, 1); if (new) { BUG_ON(!(new->d_flags & DCACHE_DISCONNECTED)); - spin_unlock(&dcache_lock); + spin_unlock(&inode->i_lock); security_d_instantiate(new, inode); d_move(new, dentry); iput(inode); } else { - /* already taking dcache_lock, so d_add() by hand */ + /* already taking inode->i_lock, so d_add() by hand */ __d_instantiate(dentry, inode); - spin_unlock(&dcache_lock); + spin_unlock(&inode->i_lock); security_d_instantiate(dentry, inode); d_rehash(dentry); } @@ -1314,10 +1690,10 @@ struct dentry *d_add_ci(struct dentry *dentry, struct inode *inode, * Negative dentry: instantiate it unless the inode is a directory and * already has a dentry. */ - spin_lock(&dcache_lock); + spin_lock(&inode->i_lock); if (!S_ISDIR(inode->i_mode) || list_empty(&inode->i_dentry)) { __d_instantiate(found, inode); - spin_unlock(&dcache_lock); + spin_unlock(&inode->i_lock); security_d_instantiate(found, inode); return found; } @@ -1327,8 +1703,8 @@ struct dentry *d_add_ci(struct dentry *dentry, struct inode *inode, * reference to it, move it in place and use it. */ new = list_entry(inode->i_dentry.next, struct dentry, d_alias); - dget_locked(new); - spin_unlock(&dcache_lock); + __dget(new); + spin_unlock(&inode->i_lock); security_d_instantiate(found, inode); d_move(new, found); iput(inode); @@ -1342,6 +1718,112 @@ err_out: EXPORT_SYMBOL(d_add_ci); /** + * __d_lookup_rcu - search for a dentry (racy, store-free) + * @parent: parent dentry + * @name: qstr of name we wish to find + * @seq: 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 + * resolution (store-free path walking) design described in + * Documentation/filesystems/path-lookup.txt. + * + * This is not to be used outside core vfs. + * + * __d_lookup_rcu must only be used in rcu-walk mode, ie. with vfsmount lock + * held, and rcu_read_lock held. The returned dentry must not be stored into + * without taking d_lock and checking d_seq sequence count against @seq + * returned here. + * + * A refcount may be taken on the found dentry with the __d_rcu_to_refcount + * function. + * + * Alternatively, __d_lookup_rcu may be called again to look up the child of + * the returned dentry, so long as its parent's seqlock is checked after the + * child is looked up. Thus, an interlocking stepping of sequence lock checks + * is formed, giving integrity down the path walk. + */ +struct dentry *__d_lookup_rcu(struct dentry *parent, struct qstr *name, + unsigned *seq, struct inode **inode) +{ + unsigned int len = name->len; + unsigned int hash = name->hash; + const unsigned char *str = name->name; + struct dcache_hash_bucket *b = d_hash(parent, hash); + struct hlist_bl_node *node; + struct dentry *dentry; + + /* + * Note: There is significant duplication with __d_lookup_rcu which is + * required to prevent single threaded performance regressions + * especially on architectures where smp_rmb (in seqcounts) are costly. + * Keep the two functions in sync. + */ + + /* + * The hash list is protected using RCU. + * + * Carefully use d_seq when comparing a candidate dentry, to avoid + * races with d_move(). + * + * It is possible that concurrent renames can mess up our list + * walk here and result in missing our dentry, resulting in the + * false-negative result. d_lookup() protects against concurrent + * renames using rename_lock seqlock. + * + * See Documentation/vfs/dcache-locking.txt for more details. + */ + hlist_bl_for_each_entry_rcu(dentry, node, &b->head, d_hash) { + struct inode *i; + const char *tname; + int tlen; + + if (dentry->d_name.hash != hash) + continue; + +seqretry: + *seq = read_seqcount_begin(&dentry->d_seq); + if (dentry->d_parent != parent) + continue; + if (d_unhashed(dentry)) + continue; + tlen = dentry->d_name.len; + tname = dentry->d_name.name; + i = dentry->d_inode; + prefetch(tname); + if (i) + prefetch(i); + /* + * This seqcount check is required to ensure name and + * len are loaded atomically, so as not to walk off the + * edge of memory when walking. If we could load this + * atomically some other way, we could drop this check. + */ + if (read_seqcount_retry(&dentry->d_seq, *seq)) + goto seqretry; + if (parent->d_flags & DCACHE_OP_COMPARE) { + if (parent->d_op->d_compare(parent, *inode, + dentry, i, + tlen, tname, name)) + continue; + } else { + if (dentry_cmp(tname, tlen, str, len)) + continue; + } + /* + * No extra seqcount check is required after the name + * compare. The caller must perform a seqcount check in + * order to do anything useful with the returned dentry + * anyway. + */ + *inode = i; + return dentry; + } + return NULL; +} + +/** * d_lookup - search for a dentry * @parent: parent dentry * @name: qstr of name we wish to find @@ -1352,10 +1834,10 @@ EXPORT_SYMBOL(d_add_ci); * dentry is returned. The caller must use dput to free the entry when it has * finished using it. %NULL is returned if the dentry does not exist. */ -struct dentry * d_lookup(struct dentry * parent, struct qstr * name) +struct dentry *d_lookup(struct dentry *parent, struct qstr *name) { - struct dentry * dentry = NULL; - unsigned long seq; + struct dentry *dentry; + unsigned seq; do { seq = read_seqbegin(&rename_lock); @@ -1367,7 +1849,7 @@ struct dentry * d_lookup(struct dentry * parent, struct qstr * name) } EXPORT_SYMBOL(d_lookup); -/* +/** * __d_lookup - search for a dentry (racy) * @parent: parent dentry * @name: qstr of name we wish to find @@ -1382,17 +1864,24 @@ EXPORT_SYMBOL(d_lookup); * * __d_lookup callers must be commented. */ -struct dentry * __d_lookup(struct dentry * parent, struct qstr * name) +struct dentry *__d_lookup(struct dentry *parent, struct qstr *name) { unsigned int len = name->len; unsigned int hash = name->hash; const unsigned char *str = name->name; - struct hlist_head *head = d_hash(parent,hash); + struct dcache_hash_bucket *b = d_hash(parent, hash); + struct hlist_bl_node *node; struct dentry *found = NULL; - struct hlist_node *node; struct dentry *dentry; /* + * Note: There is significant duplication with __d_lookup_rcu which is + * required to prevent single threaded performance regressions + * especially on architectures where smp_rmb (in seqcounts) are costly. + * Keep the two functions in sync. + */ + + /* * The hash list is protected using RCU. * * Take d_lock when comparing a candidate dentry, to avoid races @@ -1407,25 +1896,16 @@ struct dentry * __d_lookup(struct dentry * parent, struct qstr * name) */ rcu_read_lock(); - hlist_for_each_entry_rcu(dentry, node, head, d_hash) { - struct qstr *qstr; + hlist_bl_for_each_entry_rcu(dentry, node, &b->head, d_hash) { + const char *tname; + int tlen; if (dentry->d_name.hash != hash) continue; - if (dentry->d_parent != parent) - continue; spin_lock(&dentry->d_lock); - - /* - * Recheck the dentry after taking the lock - d_move may have - * changed things. Don't bother checking the hash because - * we're about to compare the whole name anyway. - */ if (dentry->d_parent != parent) goto next; - - /* non-existing due to RCU? */ if (d_unhashed(dentry)) goto next; @@ -1433,18 +1913,19 @@ struct dentry * __d_lookup(struct dentry * parent, struct qstr * name) * It is safe to compare names since d_move() cannot * change the qstr (protected by d_lock). */ - qstr = &dentry->d_name; - if (parent->d_op && parent->d_op->d_compare) { - if (parent->d_op->d_compare(parent, qstr, name)) + tlen = dentry->d_name.len; + tname = dentry->d_name.name; + if (parent->d_flags & DCACHE_OP_COMPARE) { + if (parent->d_op->d_compare(parent, parent->d_inode, + dentry, dentry->d_inode, + tlen, tname, name)) goto next; } else { - if (qstr->len != len) - goto next; - if (memcmp(qstr->name, str, len)) + if (dentry_cmp(tname, tlen, str, len)) goto next; } - atomic_inc(&dentry->d_count); + dentry->d_count++; found = dentry; spin_unlock(&dentry->d_lock); break; @@ -1473,8 +1954,8 @@ struct dentry *d_hash_and_lookup(struct dentry *dir, struct qstr *name) * routine may choose to leave the hash value unchanged. */ name->hash = full_name_hash(name->name, name->len); - if (dir->d_op && dir->d_op->d_hash) { - if (dir->d_op->d_hash(dir, name) < 0) + if (dir->d_flags & DCACHE_OP_HASH) { + if (dir->d_op->d_hash(dir, dir->d_inode, name) < 0) goto out; } dentry = d_lookup(dir, name); @@ -1483,34 +1964,32 @@ out: } /** - * d_validate - verify dentry provided from insecure source + * d_validate - verify dentry provided from insecure source (deprecated) * @dentry: The dentry alleged to be valid child of @dparent * @dparent: The parent dentry (known to be valid) * * An insecure source has sent us a dentry, here we verify it and dget() it. * This is used by ncpfs in its readdir implementation. * Zero is returned in the dentry is invalid. + * + * This function is slow for big directories, and deprecated, do not use it. */ -int d_validate(struct dentry *dentry, struct dentry *parent) +int d_validate(struct dentry *dentry, struct dentry *dparent) { - struct hlist_head *head = d_hash(parent, dentry->d_name.hash); - struct hlist_node *node; - struct dentry *d; - - /* Check whether the ptr might be valid at all.. */ - if (!kmem_ptr_validate(dentry_cache, dentry)) - return 0; - if (dentry->d_parent != parent) - return 0; + struct dentry *child; - rcu_read_lock(); - hlist_for_each_entry_rcu(d, node, head, d_hash) { - if (d == dentry) { - dget(dentry); + spin_lock(&dparent->d_lock); + list_for_each_entry(child, &dparent->d_subdirs, d_u.d_child) { + if (dentry == child) { + spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); + __dget_dlock(dentry); + spin_unlock(&dentry->d_lock); + spin_unlock(&dparent->d_lock); return 1; } } - rcu_read_unlock(); + spin_unlock(&dparent->d_lock); + return 0; } EXPORT_SYMBOL(d_validate); @@ -1538,16 +2017,23 @@ EXPORT_SYMBOL(d_validate); void d_delete(struct dentry * dentry) { + struct inode *inode; int isdir = 0; /* * Are we the only user? */ - spin_lock(&dcache_lock); +again: spin_lock(&dentry->d_lock); - isdir = S_ISDIR(dentry->d_inode->i_mode); - if (atomic_read(&dentry->d_count) == 1) { + inode = dentry->d_inode; + isdir = S_ISDIR(inode->i_mode); + if (dentry->d_count == 1) { + if (inode && !spin_trylock(&inode->i_lock)) { + spin_unlock(&dentry->d_lock); + cpu_relax(); + goto again; + } dentry->d_flags &= ~DCACHE_CANT_MOUNT; - dentry_iput(dentry); + dentry_unlink_inode(dentry); fsnotify_nameremove(dentry, isdir); return; } @@ -1556,17 +2042,18 @@ void d_delete(struct dentry * dentry) __d_drop(dentry); spin_unlock(&dentry->d_lock); - spin_unlock(&dcache_lock); fsnotify_nameremove(dentry, isdir); } EXPORT_SYMBOL(d_delete); -static void __d_rehash(struct dentry * entry, struct hlist_head *list) +static void __d_rehash(struct dentry * entry, struct dcache_hash_bucket *b) { - + BUG_ON(!d_unhashed(entry)); + spin_lock_bucket(b); entry->d_flags &= ~DCACHE_UNHASHED; - hlist_add_head_rcu(&entry->d_hash, list); + hlist_bl_add_head_rcu(&entry->d_hash, &b->head); + spin_unlock_bucket(b); } static void _d_rehash(struct dentry * entry) @@ -1583,25 +2070,39 @@ static void _d_rehash(struct dentry * entry) void d_rehash(struct dentry * entry) { - spin_lock(&dcache_lock); spin_lock(&entry->d_lock); _d_rehash(entry); spin_unlock(&entry->d_lock); - spin_unlock(&dcache_lock); } EXPORT_SYMBOL(d_rehash); -/* - * When switching names, the actual string doesn't strictly have to - * be preserved in the target - because we're dropping the target - * anyway. As such, we can just do a simple memcpy() to copy over - * the new name before we switch. +/** + * dentry_update_name_case - update case insensitive dentry with a new name + * @dentry: dentry to be updated + * @name: new name * - * Note that we have to be a lot more careful about getting the hash - * switched - we have to switch the hash value properly even if it - * then no longer matches the actual (corrupted) string of the target. - * The hash value has to match the hash queue that the dentry is on.. + * Update a case insensitive dentry with new case of name. + * + * dentry must have been returned by d_lookup with name @name. Old and new + * name lengths must match (ie. no d_compare which allows mismatched name + * lengths). + * + * Parent inode i_mutex must be held over d_lookup and into this call (to + * keep renames and concurrent inserts, and readdir(2) away). */ +void dentry_update_name_case(struct dentry *dentry, struct qstr *name) +{ + BUG_ON(!mutex_is_locked(&dentry->d_inode->i_mutex)); + BUG_ON(dentry->d_name.len != name->len); /* d_lookup gives this */ + + spin_lock(&dentry->d_lock); + write_seqcount_begin(&dentry->d_seq); + memcpy((unsigned char *)dentry->d_name.name, name->name, name->len); + write_seqcount_end(&dentry->d_seq); + spin_unlock(&dentry->d_lock); +} +EXPORT_SYMBOL(dentry_update_name_case); + static void switch_names(struct dentry *dentry, struct dentry *target) { if (dname_external(target)) { @@ -1643,54 +2144,84 @@ static void switch_names(struct dentry *dentry, struct dentry *target) swap(dentry->d_name.len, target->d_name.len); } +static void dentry_lock_for_move(struct dentry *dentry, struct dentry *target) +{ + /* + * XXXX: do we really need to take target->d_lock? + */ + if (IS_ROOT(dentry) || dentry->d_parent == target->d_parent) + spin_lock(&target->d_parent->d_lock); + else { + if (d_ancestor(dentry->d_parent, target->d_parent)) { + spin_lock(&dentry->d_parent->d_lock); + spin_lock_nested(&target->d_parent->d_lock, + DENTRY_D_LOCK_NESTED); + } else { + spin_lock(&target->d_parent->d_lock); + spin_lock_nested(&dentry->d_parent->d_lock, + DENTRY_D_LOCK_NESTED); + } + } + if (target < dentry) { + spin_lock_nested(&target->d_lock, 2); + spin_lock_nested(&dentry->d_lock, 3); + } else { + spin_lock_nested(&dentry->d_lock, 2); + spin_lock_nested(&target->d_lock, 3); + } +} + +static void dentry_unlock_parents_for_move(struct dentry *dentry, + struct dentry *target) +{ + if (target->d_parent != dentry->d_parent) + spin_unlock(&dentry->d_parent->d_lock); + if (target->d_parent != target) + spin_unlock(&target->d_parent->d_lock); +} + /* - * We cannibalize "target" when moving dentry on top of it, - * because it's going to be thrown away anyway. We could be more - * polite about it, though. - * - * This forceful removal will result in ugly /proc output if - * somebody holds a file open that got deleted due to a rename. - * We could be nicer about the deleted file, and let it show - * up under the name it had before it was deleted rather than - * under the original name of the file that was moved on top of it. + * When switching names, the actual string doesn't strictly have to + * be preserved in the target - because we're dropping the target + * anyway. As such, we can just do a simple memcpy() to copy over + * the new name before we switch. + * + * Note that we have to be a lot more careful about getting the hash + * switched - we have to switch the hash value properly even if it + * then no longer matches the actual (corrupted) string of the target. + * The hash value has to match the hash queue that the dentry is on.. */ - /* - * d_move_locked - move a dentry + * d_move - move a dentry * @dentry: entry to move * @target: new dentry * * Update the dcache to reflect the move of a file name. Negative * dcache entries should not be moved in this way. */ -static void d_move_locked(struct dentry * dentry, struct dentry * target) +void d_move(struct dentry * dentry, struct dentry * target) { - struct hlist_head *list; - if (!dentry->d_inode) printk(KERN_WARNING "VFS: moving negative dcache entry\n"); + BUG_ON(d_ancestor(dentry, target)); + BUG_ON(d_ancestor(target, dentry)); + write_seqlock(&rename_lock); - /* - * XXXX: do we really need to take target->d_lock? - */ - if (target < dentry) { - spin_lock(&target->d_lock); - spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); - } else { - spin_lock(&dentry->d_lock); - spin_lock_nested(&target->d_lock, DENTRY_D_LOCK_NESTED); - } - /* Move the dentry to the target hash queue, if on different bucket */ - if (d_unhashed(dentry)) - goto already_unhashed; + dentry_lock_for_move(dentry, target); - hlist_del_rcu(&dentry->d_hash); + write_seqcount_begin(&dentry->d_seq); + write_seqcount_begin(&target->d_seq); -already_unhashed: - list = d_hash(target->d_parent, target->d_name.hash); - __d_rehash(dentry, list); + /* __d_drop does write_seqcount_barrier, but they're OK to nest. */ + + /* + * Move the dentry to the target hash queue. Don't bother checking + * for the same hash queue because of how unlikely it is. + */ + __d_drop(dentry); + __d_rehash(dentry, d_hash(target->d_parent, target->d_name.hash)); /* Unhash the target: dput() will then get rid of it */ __d_drop(target); @@ -1715,27 +2246,16 @@ already_unhashed: } list_add(&dentry->d_u.d_child, &dentry->d_parent->d_subdirs); + + write_seqcount_end(&target->d_seq); + write_seqcount_end(&dentry->d_seq); + + dentry_unlock_parents_for_move(dentry, target); spin_unlock(&target->d_lock); fsnotify_d_move(dentry); spin_unlock(&dentry->d_lock); write_sequnlock(&rename_lock); } - -/** - * d_move - move a dentry - * @dentry: entry to move - * @target: new dentry - * - * Update the dcache to reflect the move of a file name. Negative - * dcache entries should not be moved in this way. - */ - -void d_move(struct dentry * dentry, struct dentry * target) -{ - spin_lock(&dcache_lock); - d_move_locked(dentry, target); - spin_unlock(&dcache_lock); -} EXPORT_SYMBOL(d_move); /** @@ -1761,13 +2281,13 @@ struct dentry *d_ancestor(struct dentry *p1, struct dentry *p2) * This helper attempts to cope with remotely renamed directories * * It assumes that the caller is already holding - * dentry->d_parent->d_inode->i_mutex and the dcache_lock + * dentry->d_parent->d_inode->i_mutex and the inode->i_lock * * Note: If ever the locking in lock_rename() changes, then please * remember to update this too... */ -static struct dentry *__d_unalias(struct dentry *dentry, struct dentry *alias) - __releases(dcache_lock) +static struct dentry *__d_unalias(struct inode *inode, + struct dentry *dentry, struct dentry *alias) { struct mutex *m1 = NULL, *m2 = NULL; struct dentry *ret; @@ -1790,10 +2310,10 @@ static struct dentry *__d_unalias(struct dentry *dentry, struct dentry *alias) goto out_err; m2 = &alias->d_parent->d_inode->i_mutex; out_unalias: - d_move_locked(alias, dentry); + d_move(alias, dentry); ret = alias; out_err: - spin_unlock(&dcache_lock); + spin_unlock(&inode->i_lock); if (m2) mutex_unlock(m2); if (m1) @@ -1804,17 +2324,23 @@ out_err: /* * Prepare an anonymous dentry for life in the superblock's dentry tree as a * named dentry in place of the dentry to be replaced. + * returns with anon->d_lock held! */ static void __d_materialise_dentry(struct dentry *dentry, struct dentry *anon) { struct dentry *dparent, *aparent; - switch_names(dentry, anon); - swap(dentry->d_name.hash, anon->d_name.hash); + dentry_lock_for_move(anon, dentry); + + write_seqcount_begin(&dentry->d_seq); + write_seqcount_begin(&anon->d_seq); dparent = dentry->d_parent; aparent = anon->d_parent; + switch_names(dentry, anon); + swap(dentry->d_name.hash, anon->d_name.hash); + dentry->d_parent = (aparent == anon) ? dentry : aparent; list_del(&dentry->d_u.d_child); if (!IS_ROOT(dentry)) @@ -1829,6 +2355,13 @@ static void __d_materialise_dentry(struct dentry *dentry, struct dentry *anon) else INIT_LIST_HEAD(&anon->d_u.d_child); + write_seqcount_end(&dentry->d_seq); + write_seqcount_end(&anon->d_seq); + + dentry_unlock_parents_for_move(anon, dentry); + spin_unlock(&dentry->d_lock); + + /* anon->d_lock still locked, returns locked */ anon->d_flags &= ~DCACHE_DISCONNECTED; } @@ -1846,14 +2379,15 @@ struct dentry *d_materialise_unique(struct dentry *dentry, struct inode *inode) BUG_ON(!d_unhashed(dentry)); - spin_lock(&dcache_lock); - if (!inode) { actual = dentry; __d_instantiate(dentry, NULL); - goto found_lock; + d_rehash(actual); + goto out_nolock; } + spin_lock(&inode->i_lock); + if (S_ISDIR(inode->i_mode)) { struct dentry *alias; @@ -1864,13 +2398,12 @@ struct dentry *d_materialise_unique(struct dentry *dentry, struct inode *inode) /* Is this an anonymous mountpoint that we could splice * into our tree? */ if (IS_ROOT(alias)) { - spin_lock(&alias->d_lock); __d_materialise_dentry(dentry, alias); __d_drop(alias); goto found; } /* Nope, but we must(!) avoid directory aliasing */ - actual = __d_unalias(dentry, alias); + actual = __d_unalias(inode, dentry, alias); if (IS_ERR(actual)) dput(alias); goto out_nolock; @@ -1881,15 +2414,14 @@ struct dentry *d_materialise_unique(struct dentry *dentry, struct inode *inode) actual = __d_instantiate_unique(dentry, inode); if (!actual) actual = dentry; - else if (unlikely(!d_unhashed(actual))) - goto shouldnt_be_hashed; + else + BUG_ON(!d_unhashed(actual)); -found_lock: spin_lock(&actual->d_lock); found: _d_rehash(actual); spin_unlock(&actual->d_lock); - spin_unlock(&dcache_lock); + spin_unlock(&inode->i_lock); out_nolock: if (actual == dentry) { security_d_instantiate(dentry, inode); @@ -1898,10 +2430,6 @@ out_nolock: iput(inode); return actual; - -shouldnt_be_hashed: - spin_unlock(&dcache_lock); - BUG(); } EXPORT_SYMBOL_GPL(d_materialise_unique); @@ -1928,7 +2456,7 @@ static int prepend_name(char **buffer, int *buflen, struct qstr *name) * @buffer: pointer to the end of the buffer * @buflen: pointer to buffer length * - * Caller holds the dcache_lock. + * Caller holds the rename_lock. * * If path is not reachable from the supplied root, then the value of * root is changed (without modifying refcounts). @@ -1956,7 +2484,9 @@ static int prepend_path(const struct path *path, struct path *root, } parent = dentry->d_parent; prefetch(parent); + spin_lock(&dentry->d_lock); error = prepend_name(buffer, buflen, &dentry->d_name); + spin_unlock(&dentry->d_lock); if (!error) error = prepend(buffer, buflen, "/", 1); if (error) @@ -2012,9 +2542,9 @@ char *__d_path(const struct path *path, struct path *root, int error; prepend(&res, &buflen, "\0", 1); - spin_lock(&dcache_lock); + write_seqlock(&rename_lock); error = prepend_path(path, root, &res, &buflen); - spin_unlock(&dcache_lock); + write_sequnlock(&rename_lock); if (error) return ERR_PTR(error); @@ -2076,12 +2606,12 @@ char *d_path(const struct path *path, char *buf, int buflen) return path->dentry->d_op->d_dname(path->dentry, buf, buflen); get_fs_root(current->fs, &root); - spin_lock(&dcache_lock); + write_seqlock(&rename_lock); tmp = root; error = path_with_deleted(path, &tmp, &res, &buflen); if (error) res = ERR_PTR(error); - spin_unlock(&dcache_lock); + write_sequnlock(&rename_lock); path_put(&root); return res; } @@ -2107,12 +2637,12 @@ char *d_path_with_unreachable(const struct path *path, char *buf, int buflen) return path->dentry->d_op->d_dname(path->dentry, buf, buflen); get_fs_root(current->fs, &root); - spin_lock(&dcache_lock); + write_seqlock(&rename_lock); tmp = root; error = path_with_deleted(path, &tmp, &res, &buflen); if (!error && !path_equal(&tmp, &root)) error = prepend_unreachable(&res, &buflen); - spin_unlock(&dcache_lock); + write_sequnlock(&rename_lock); path_put(&root); if (error) res = ERR_PTR(error); @@ -2144,7 +2674,7 @@ char *dynamic_dname(struct dentry *dentry, char *buffer, int buflen, /* * Write full pathname from the root of the filesystem into the buffer. */ -char *__dentry_path(struct dentry *dentry, char *buf, int buflen) +static char *__dentry_path(struct dentry *dentry, char *buf, int buflen) { char *end = buf + buflen; char *retval; @@ -2158,10 +2688,13 @@ char *__dentry_path(struct dentry *dentry, char *buf, int buflen) while (!IS_ROOT(dentry)) { struct dentry *parent = dentry->d_parent; + int error; prefetch(parent); - if ((prepend_name(&end, &buflen, &dentry->d_name) != 0) || - (prepend(&end, &buflen, "/", 1) != 0)) + spin_lock(&dentry->d_lock); + error = prepend_name(&end, &buflen, &dentry->d_name); + spin_unlock(&dentry->d_lock); + if (error != 0 || prepend(&end, &buflen, "/", 1) != 0) goto Elong; retval = end; @@ -2171,14 +2704,25 @@ char *__dentry_path(struct dentry *dentry, char *buf, int buflen) Elong: return ERR_PTR(-ENAMETOOLONG); } -EXPORT_SYMBOL(__dentry_path); + +char *dentry_path_raw(struct dentry *dentry, char *buf, int buflen) +{ + char *retval; + + write_seqlock(&rename_lock); + retval = __dentry_path(dentry, buf, buflen); + write_sequnlock(&rename_lock); + + return retval; +} +EXPORT_SYMBOL(dentry_path_raw); char *dentry_path(struct dentry *dentry, char *buf, int buflen) { char *p = NULL; char *retval; - spin_lock(&dcache_lock); + write_seqlock(&rename_lock); if (d_unlinked(dentry)) { p = buf + buflen; if (prepend(&p, &buflen, "//deleted", 10) != 0) @@ -2186,12 +2730,11 @@ char *dentry_path(struct dentry *dentry, char *buf, int buflen) buflen++; } retval = __dentry_path(dentry, buf, buflen); - spin_unlock(&dcache_lock); + write_sequnlock(&rename_lock); if (!IS_ERR(retval) && p) *p = '/'; /* restore '/' overriden with '\0' */ return retval; Elong: - spin_unlock(&dcache_lock); return ERR_PTR(-ENAMETOOLONG); } @@ -2225,7 +2768,7 @@ SYSCALL_DEFINE2(getcwd, char __user *, buf, unsigned long, size) get_fs_root_and_pwd(current->fs, &root, &pwd); error = -ENOENT; - spin_lock(&dcache_lock); + write_seqlock(&rename_lock); if (!d_unlinked(pwd.dentry)) { unsigned long len; struct path tmp = root; @@ -2234,7 +2777,7 @@ SYSCALL_DEFINE2(getcwd, char __user *, buf, unsigned long, size) prepend(&cwd, &buflen, "\0", 1); error = prepend_path(&pwd, &tmp, &cwd, &buflen); - spin_unlock(&dcache_lock); + write_sequnlock(&rename_lock); if (error) goto out; @@ -2253,8 +2796,9 @@ SYSCALL_DEFINE2(getcwd, char __user *, buf, unsigned long, size) if (copy_to_user(buf, cwd, len)) error = -EFAULT; } - } else - spin_unlock(&dcache_lock); + } else { + write_sequnlock(&rename_lock); + } out: path_put(&pwd); @@ -2282,25 +2826,25 @@ out: int is_subdir(struct dentry *new_dentry, struct dentry *old_dentry) { int result; - unsigned long seq; + unsigned seq; if (new_dentry == old_dentry) return 1; - /* - * Need rcu_readlock to protect against the d_parent trashing - * due to d_move - */ - rcu_read_lock(); do { /* for restarting inner loop in case of seq retry */ seq = read_seqbegin(&rename_lock); + /* + * Need rcu_readlock to protect against the d_parent trashing + * due to d_move + */ + rcu_read_lock(); if (d_ancestor(old_dentry, new_dentry)) result = 1; else result = 0; + rcu_read_unlock(); } while (read_seqretry(&rename_lock, seq)); - rcu_read_unlock(); return result; } @@ -2332,10 +2876,15 @@ EXPORT_SYMBOL(path_is_under); void d_genocide(struct dentry *root) { - struct dentry *this_parent = root; + struct dentry *this_parent; struct list_head *next; + unsigned seq; + int locked = 0; - spin_lock(&dcache_lock); + seq = read_seqbegin(&rename_lock); +again: + this_parent = root; + spin_lock(&this_parent->d_lock); repeat: next = this_parent->d_subdirs.next; resume: @@ -2343,21 +2892,62 @@ resume: struct list_head *tmp = next; struct dentry *dentry = list_entry(tmp, struct dentry, d_u.d_child); next = tmp->next; - if (d_unhashed(dentry)||!dentry->d_inode) + + spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); + if (d_unhashed(dentry) || !dentry->d_inode) { + spin_unlock(&dentry->d_lock); continue; + } if (!list_empty(&dentry->d_subdirs)) { + spin_unlock(&this_parent->d_lock); + spin_release(&dentry->d_lock.dep_map, 1, _RET_IP_); this_parent = dentry; + spin_acquire(&this_parent->d_lock.dep_map, 0, 1, _RET_IP_); goto repeat; } - atomic_dec(&dentry->d_count); + if (!(dentry->d_flags & DCACHE_GENOCIDE)) { + dentry->d_flags |= DCACHE_GENOCIDE; + dentry->d_count--; + } + spin_unlock(&dentry->d_lock); } if (this_parent != root) { - next = this_parent->d_u.d_child.next; - atomic_dec(&this_parent->d_count); - this_parent = this_parent->d_parent; + struct dentry *tmp; + struct dentry *child; + + tmp = this_parent->d_parent; + if (!(this_parent->d_flags & DCACHE_GENOCIDE)) { + this_parent->d_flags |= DCACHE_GENOCIDE; + this_parent->d_count--; + } + rcu_read_lock(); + spin_unlock(&this_parent->d_lock); + child = this_parent; + this_parent = tmp; + spin_lock(&this_parent->d_lock); + /* might go back up the wrong parent if we have had a rename + * or deletion */ + if (this_parent != child->d_parent || + (!locked && read_seqretry(&rename_lock, seq))) { + spin_unlock(&this_parent->d_lock); + rcu_read_unlock(); + goto rename_retry; + } + rcu_read_unlock(); + next = child->d_u.d_child.next; goto resume; } - spin_unlock(&dcache_lock); + spin_unlock(&this_parent->d_lock); + if (!locked && read_seqretry(&rename_lock, seq)) + goto rename_retry; + if (locked) + write_sequnlock(&rename_lock); + return; + +rename_retry: + locked = 1; + write_seqlock(&rename_lock); + goto again; } /** @@ -2411,7 +3001,7 @@ static void __init dcache_init_early(void) dentry_hashtable = alloc_large_system_hash("Dentry cache", - sizeof(struct hlist_head), + sizeof(struct dcache_hash_bucket), dhash_entries, 13, HASH_EARLY, @@ -2420,16 +3010,13 @@ static void __init dcache_init_early(void) 0); for (loop = 0; loop < (1 << d_hash_shift); loop++) - INIT_HLIST_HEAD(&dentry_hashtable[loop]); + INIT_HLIST_BL_HEAD(&dentry_hashtable[loop].head); } static void __init dcache_init(void) { int loop; - percpu_counter_init(&nr_dentry, 0); - percpu_counter_init(&nr_dentry_unused, 0); - /* * A constructor could be added for stable state like the lists, * but it is probably not worth it because of the cache nature @@ -2446,7 +3033,7 @@ static void __init dcache_init(void) dentry_hashtable = alloc_large_system_hash("Dentry cache", - sizeof(struct hlist_head), + sizeof(struct dcache_hash_bucket), dhash_entries, 13, 0, @@ -2455,7 +3042,7 @@ static void __init dcache_init(void) 0); for (loop = 0; loop < (1 << d_hash_shift); loop++) - INIT_HLIST_HEAD(&dentry_hashtable[loop]); + INIT_HLIST_BL_HEAD(&dentry_hashtable[loop].head); } /* SLAB cache for __getname() consumers */ diff --git a/fs/ecryptfs/dentry.c b/fs/ecryptfs/dentry.c index 906e803f7f7..6fc4f319b55 100644 --- a/fs/ecryptfs/dentry.c +++ b/fs/ecryptfs/dentry.c @@ -44,12 +44,17 @@ */ static int ecryptfs_d_revalidate(struct dentry *dentry, struct nameidata *nd) { - struct dentry *lower_dentry = ecryptfs_dentry_to_lower(dentry); - struct vfsmount *lower_mnt = ecryptfs_dentry_to_lower_mnt(dentry); + struct dentry *lower_dentry; + struct vfsmount *lower_mnt; struct dentry *dentry_save; struct vfsmount *vfsmount_save; int rc = 1; + if (nd->flags & LOOKUP_RCU) + return -ECHILD; + + lower_dentry = ecryptfs_dentry_to_lower(dentry); + lower_mnt = ecryptfs_dentry_to_lower_mnt(dentry); if (!lower_dentry->d_op || !lower_dentry->d_op->d_revalidate) goto out; dentry_save = nd->path.dentry; diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index 9d1a22d6276..337352a9475 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -260,7 +260,7 @@ int ecryptfs_lookup_and_interpose_lower(struct dentry *ecryptfs_dentry, ecryptfs_dentry->d_parent)); lower_inode = lower_dentry->d_inode; fsstack_copy_attr_atime(ecryptfs_dir_inode, lower_dir_dentry->d_inode); - BUG_ON(!atomic_read(&lower_dentry->d_count)); + BUG_ON(!lower_dentry->d_count); ecryptfs_set_dentry_private(ecryptfs_dentry, kmem_cache_alloc(ecryptfs_dentry_info_cache, GFP_KERNEL)); @@ -441,7 +441,7 @@ static struct dentry *ecryptfs_lookup(struct inode *ecryptfs_dir_inode, struct qstr lower_name; int rc = 0; - ecryptfs_dentry->d_op = &ecryptfs_dops; + d_set_d_op(ecryptfs_dentry, &ecryptfs_dops); if ((ecryptfs_dentry->d_name.len == 1 && !strcmp(ecryptfs_dentry->d_name.name, ".")) || (ecryptfs_dentry->d_name.len == 2 @@ -454,7 +454,7 @@ static struct dentry *ecryptfs_lookup(struct inode *ecryptfs_dir_inode, lower_name.hash = ecryptfs_dentry->d_name.hash; if (lower_dir_dentry->d_op && lower_dir_dentry->d_op->d_hash) { rc = lower_dir_dentry->d_op->d_hash(lower_dir_dentry, - &lower_name); + lower_dir_dentry->d_inode, &lower_name); if (rc < 0) goto out_d_drop; } @@ -489,7 +489,7 @@ static struct dentry *ecryptfs_lookup(struct inode *ecryptfs_dir_inode, lower_name.hash = full_name_hash(lower_name.name, lower_name.len); if (lower_dir_dentry->d_op && lower_dir_dentry->d_op->d_hash) { rc = lower_dir_dentry->d_op->d_hash(lower_dir_dentry, - &lower_name); + lower_dir_dentry->d_inode, &lower_name); if (rc < 0) goto out_d_drop; } @@ -980,8 +980,10 @@ int ecryptfs_truncate(struct dentry *dentry, loff_t new_length) } static int -ecryptfs_permission(struct inode *inode, int mask) +ecryptfs_permission(struct inode *inode, int mask, unsigned int flags) { + if (flags & IPERM_FLAG_RCU) + return -ECHILD; return inode_permission(ecryptfs_inode_to_lower(inode), mask); } diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c index a9dbd62518e..35103867537 100644 --- a/fs/ecryptfs/main.c +++ b/fs/ecryptfs/main.c @@ -189,7 +189,7 @@ int ecryptfs_interpose(struct dentry *lower_dentry, struct dentry *dentry, if (special_file(lower_inode->i_mode)) init_special_inode(inode, lower_inode->i_mode, lower_inode->i_rdev); - dentry->d_op = &ecryptfs_dops; + d_set_d_op(dentry, &ecryptfs_dops); fsstack_copy_attr_all(inode, lower_inode); /* This size will be overwritten for real files w/ headers and * other metadata */ @@ -594,7 +594,7 @@ static struct dentry *ecryptfs_mount(struct file_system_type *fs_type, int flags deactivate_locked_super(s); goto out; } - s->s_root->d_op = &ecryptfs_dops; + d_set_d_op(s->s_root, &ecryptfs_dops); s->s_root->d_sb = s; s->s_root->d_parent = s->s_root; diff --git a/fs/ecryptfs/super.c b/fs/ecryptfs/super.c index 253732382d3..3042fe123a3 100644 --- a/fs/ecryptfs/super.c +++ b/fs/ecryptfs/super.c @@ -28,7 +28,6 @@ #include <linux/key.h> #include <linux/slab.h> #include <linux/seq_file.h> -#include <linux/smp_lock.h> #include <linux/file.h> #include <linux/crypto.h> #include "ecryptfs_kernel.h" @@ -63,6 +62,16 @@ out: return inode; } +static void ecryptfs_i_callback(struct rcu_head *head) +{ + struct inode *inode = container_of(head, struct inode, i_rcu); + struct ecryptfs_inode_info *inode_info; + inode_info = ecryptfs_inode_to_private(inode); + + INIT_LIST_HEAD(&inode->i_dentry); + kmem_cache_free(ecryptfs_inode_info_cache, inode_info); +} + /** * ecryptfs_destroy_inode * @inode: The ecryptfs inode @@ -89,7 +98,7 @@ static void ecryptfs_destroy_inode(struct inode *inode) } } ecryptfs_destroy_crypt_stat(&inode_info->crypt_stat); - kmem_cache_free(ecryptfs_inode_info_cache, inode_info); + call_rcu(&inode->i_rcu, ecryptfs_i_callback); } /** diff --git a/fs/efs/super.c b/fs/efs/super.c index 5073a07652c..0f31acb0131 100644 --- a/fs/efs/super.c +++ b/fs/efs/super.c @@ -65,11 +65,18 @@ static struct inode *efs_alloc_inode(struct super_block *sb) return &ei->vfs_inode; } -static void efs_destroy_inode(struct inode *inode) +static void efs_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(efs_inode_cachep, INODE_INFO(inode)); } +static void efs_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, efs_i_callback); +} + static void init_once(void *foo) { struct efs_inode_info *ei = (struct efs_inode_info *) foo; diff --git a/fs/exec.c b/fs/exec.c index 99d33a1371e..c62efcb959c 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -164,7 +164,26 @@ out: #ifdef CONFIG_MMU -static struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos, +void acct_arg_size(struct linux_binprm *bprm, unsigned long pages) +{ + struct mm_struct *mm = current->mm; + long diff = (long)(pages - bprm->vma_pages); + + if (!mm || !diff) + return; + + bprm->vma_pages = pages; + +#ifdef SPLIT_RSS_COUNTING + add_mm_counter(mm, MM_ANONPAGES, diff); +#else + spin_lock(&mm->page_table_lock); + add_mm_counter(mm, MM_ANONPAGES, diff); + spin_unlock(&mm->page_table_lock); +#endif +} + +struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos, int write) { struct page *page; @@ -186,6 +205,8 @@ static struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos, unsigned long size = bprm->vma->vm_end - bprm->vma->vm_start; struct rlimit *rlim; + acct_arg_size(bprm, size / PAGE_SIZE); + /* * We've historically supported up to 32 pages (ARG_MAX) * of argument strings even with small stacks @@ -254,6 +275,11 @@ static int __bprm_mm_init(struct linux_binprm *bprm) vma->vm_flags = VM_STACK_FLAGS | VM_STACK_INCOMPLETE_SETUP; vma->vm_page_prot = vm_get_page_prot(vma->vm_flags); INIT_LIST_HEAD(&vma->anon_vma_chain); + + err = security_file_mmap(NULL, 0, 0, 0, vma->vm_start, 1); + if (err) + goto err; + err = insert_vm_struct(mm, vma); if (err) goto err; @@ -276,7 +302,11 @@ static bool valid_arg_len(struct linux_binprm *bprm, long len) #else -static struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos, +void acct_arg_size(struct linux_binprm *bprm, unsigned long pages) +{ +} + +struct page *get_arg_page(struct linux_binprm *bprm, unsigned long pos, int write) { struct page *page; @@ -1003,6 +1033,7 @@ int flush_old_exec(struct linux_binprm * bprm) /* * Release all of the old mmap stuff */ + acct_arg_size(bprm, 0); retval = exec_mmap(bprm->mm); if (retval) goto out; @@ -1426,8 +1457,10 @@ int do_execve(const char * filename, return retval; out: - if (bprm->mm) - mmput (bprm->mm); + if (bprm->mm) { + acct_arg_size(bprm, 0); + mmput(bprm->mm); + } out_file: if (bprm->file) { diff --git a/fs/exofs/super.c b/fs/exofs/super.c index 79c3ae6e045..8c6c4669b38 100644 --- a/fs/exofs/super.c +++ b/fs/exofs/super.c @@ -150,12 +150,19 @@ static struct inode *exofs_alloc_inode(struct super_block *sb) return &oi->vfs_inode; } +static void exofs_i_callback(struct rcu_head *head) +{ + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); + kmem_cache_free(exofs_inode_cachep, exofs_i(inode)); +} + /* * Remove an inode from the cache */ static void exofs_destroy_inode(struct inode *inode) { - kmem_cache_free(exofs_inode_cachep, exofs_i(inode)); + call_rcu(&inode->i_rcu, exofs_i_callback); } /* diff --git a/fs/exportfs/expfs.c b/fs/exportfs/expfs.c index 51b304056f1..4b6825740dd 100644 --- a/fs/exportfs/expfs.c +++ b/fs/exportfs/expfs.c @@ -43,24 +43,26 @@ find_acceptable_alias(struct dentry *result, void *context) { struct dentry *dentry, *toput = NULL; + struct inode *inode; if (acceptable(context, result)) return result; - spin_lock(&dcache_lock); - list_for_each_entry(dentry, &result->d_inode->i_dentry, d_alias) { - dget_locked(dentry); - spin_unlock(&dcache_lock); + inode = result->d_inode; + spin_lock(&inode->i_lock); + list_for_each_entry(dentry, &inode->i_dentry, d_alias) { + dget(dentry); + spin_unlock(&inode->i_lock); if (toput) dput(toput); if (dentry != result && acceptable(context, dentry)) { dput(result); return dentry; } - spin_lock(&dcache_lock); + spin_lock(&inode->i_lock); toput = dentry; } - spin_unlock(&dcache_lock); + spin_unlock(&inode->i_lock); if (toput) dput(toput); diff --git a/fs/ext2/acl.c b/fs/ext2/acl.c index 2bcc0431bad..7b4180554a6 100644 --- a/fs/ext2/acl.c +++ b/fs/ext2/acl.c @@ -232,10 +232,17 @@ ext2_set_acl(struct inode *inode, int type, struct posix_acl *acl) } int -ext2_check_acl(struct inode *inode, int mask) +ext2_check_acl(struct inode *inode, int mask, unsigned int flags) { - struct posix_acl *acl = ext2_get_acl(inode, ACL_TYPE_ACCESS); + struct posix_acl *acl; + + if (flags & IPERM_FLAG_RCU) { + if (!negative_cached_acl(inode, ACL_TYPE_ACCESS)) + return -ECHILD; + return -EAGAIN; + } + acl = ext2_get_acl(inode, ACL_TYPE_ACCESS); if (IS_ERR(acl)) return PTR_ERR(acl); if (acl) { diff --git a/fs/ext2/acl.h b/fs/ext2/acl.h index 3ff6cbb9ac4..c939b7b1209 100644 --- a/fs/ext2/acl.h +++ b/fs/ext2/acl.h @@ -54,7 +54,7 @@ static inline int ext2_acl_count(size_t size) #ifdef CONFIG_EXT2_FS_POSIX_ACL /* acl.c */ -extern int ext2_check_acl (struct inode *, int); +extern int ext2_check_acl (struct inode *, int, unsigned int); extern int ext2_acl_chmod (struct inode *); extern int ext2_init_acl (struct inode *, struct inode *); diff --git a/fs/ext2/super.c b/fs/ext2/super.c index d89e0b6a2d7..e0c6380ff99 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -161,11 +161,18 @@ static struct inode *ext2_alloc_inode(struct super_block *sb) return &ei->vfs_inode; } -static void ext2_destroy_inode(struct inode *inode) +static void ext2_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(ext2_inode_cachep, EXT2_I(inode)); } +static void ext2_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, ext2_i_callback); +} + static void init_once(void *foo) { struct ext2_inode_info *ei = (struct ext2_inode_info *) foo; diff --git a/fs/ext3/acl.c b/fs/ext3/acl.c index 8a11fe21218..e4fa49e6c53 100644 --- a/fs/ext3/acl.c +++ b/fs/ext3/acl.c @@ -240,10 +240,17 @@ ext3_set_acl(handle_t *handle, struct inode *inode, int type, } int -ext3_check_acl(struct inode *inode, int mask) +ext3_check_acl(struct inode *inode, int mask, unsigned int flags) { - struct posix_acl *acl = ext3_get_acl(inode, ACL_TYPE_ACCESS); + struct posix_acl *acl; + + if (flags & IPERM_FLAG_RCU) { + if (!negative_cached_acl(inode, ACL_TYPE_ACCESS)) + return -ECHILD; + return -EAGAIN; + } + acl = ext3_get_acl(inode, ACL_TYPE_ACCESS); if (IS_ERR(acl)) return PTR_ERR(acl); if (acl) { diff --git a/fs/ext3/acl.h b/fs/ext3/acl.h index 597334626de..5faf8048e90 100644 --- a/fs/ext3/acl.h +++ b/fs/ext3/acl.h @@ -54,7 +54,7 @@ static inline int ext3_acl_count(size_t size) #ifdef CONFIG_EXT3_FS_POSIX_ACL /* acl.c */ -extern int ext3_check_acl (struct inode *, int); +extern int ext3_check_acl (struct inode *, int, unsigned int); extern int ext3_acl_chmod (struct inode *); extern int ext3_init_acl (handle_t *, struct inode *, struct inode *); diff --git a/fs/ext3/super.c b/fs/ext3/super.c index 2fedaf8b501..77ce1616f72 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -27,7 +27,6 @@ #include <linux/init.h> #include <linux/blkdev.h> #include <linux/parser.h> -#include <linux/smp_lock.h> #include <linux/buffer_head.h> #include <linux/exportfs.h> #include <linux/vfs.h> @@ -480,6 +479,13 @@ static struct inode *ext3_alloc_inode(struct super_block *sb) return &ei->vfs_inode; } +static void ext3_i_callback(struct rcu_head *head) +{ + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); + kmem_cache_free(ext3_inode_cachep, EXT3_I(inode)); +} + static void ext3_destroy_inode(struct inode *inode) { if (!list_empty(&(EXT3_I(inode)->i_orphan))) { @@ -490,7 +496,7 @@ static void ext3_destroy_inode(struct inode *inode) false); dump_stack(); } - kmem_cache_free(ext3_inode_cachep, EXT3_I(inode)); + call_rcu(&inode->i_rcu, ext3_i_callback); } static void init_once(void *foo) diff --git a/fs/ext4/acl.c b/fs/ext4/acl.c index 5e2ed4504ea..e0270d1f8d8 100644 --- a/fs/ext4/acl.c +++ b/fs/ext4/acl.c @@ -238,10 +238,17 @@ ext4_set_acl(handle_t *handle, struct inode *inode, int type, } int -ext4_check_acl(struct inode *inode, int mask) +ext4_check_acl(struct inode *inode, int mask, unsigned int flags) { - struct posix_acl *acl = ext4_get_acl(inode, ACL_TYPE_ACCESS); + struct posix_acl *acl; + + if (flags & IPERM_FLAG_RCU) { + if (!negative_cached_acl(inode, ACL_TYPE_ACCESS)) + return -ECHILD; + return -EAGAIN; + } + acl = ext4_get_acl(inode, ACL_TYPE_ACCESS); if (IS_ERR(acl)) return PTR_ERR(acl); if (acl) { diff --git a/fs/ext4/acl.h b/fs/ext4/acl.h index 9d843d5deac..dec821168fd 100644 --- a/fs/ext4/acl.h +++ b/fs/ext4/acl.h @@ -54,7 +54,7 @@ static inline int ext4_acl_count(size_t size) #ifdef CONFIG_EXT4_FS_POSIX_ACL /* acl.c */ -extern int ext4_check_acl(struct inode *, int); +extern int ext4_check_acl(struct inode *, int, unsigned int); extern int ext4_acl_chmod(struct inode *); extern int ext4_init_acl(handle_t *, struct inode *, struct inode *); diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index 8b5dd6369f8..94ce3d7a1c4 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -177,7 +177,7 @@ struct mpage_da_data { struct ext4_io_page { struct page *p_page; - int p_count; + atomic_t p_count; }; #define MAX_IO_PAGES 128 @@ -858,6 +858,7 @@ struct ext4_inode_info { spinlock_t i_completed_io_lock; /* current io_end structure for async DIO write*/ ext4_io_end_t *cur_aio_dio; + atomic_t i_ioend_count; /* Number of outstanding io_end structs */ /* * Transactions that contain inode's metadata needed to complete @@ -909,6 +910,7 @@ struct ext4_inode_info { #define EXT4_MOUNT_JOURNAL_CHECKSUM 0x800000 /* Journal checksums */ #define EXT4_MOUNT_JOURNAL_ASYNC_COMMIT 0x1000000 /* Journal Async Commit */ #define EXT4_MOUNT_I_VERSION 0x2000000 /* i_version support */ +#define EXT4_MOUNT_MBLK_IO_SUBMIT 0x4000000 /* multi-block io submits */ #define EXT4_MOUNT_DELALLOC 0x8000000 /* Delalloc support */ #define EXT4_MOUNT_DATA_ERR_ABORT 0x10000000 /* Abort on file data write */ #define EXT4_MOUNT_BLOCK_VALIDITY 0x20000000 /* Block validity checking */ @@ -2060,6 +2062,7 @@ extern int ext4_move_extents(struct file *o_filp, struct file *d_filp, /* page-io.c */ extern int __init ext4_init_pageio(void); extern void ext4_exit_pageio(void); +extern void ext4_ioend_wait(struct inode *); extern void ext4_free_io_end(ext4_io_end_t *io); extern ext4_io_end_t *ext4_init_io_end(struct inode *inode, gfp_t flags); extern int ext4_end_io_nolock(ext4_io_end_t *io); diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 19161647046..e659597b690 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -53,6 +53,7 @@ static inline int ext4_begin_ordered_truncate(struct inode *inode, loff_t new_size) { + trace_ext4_begin_ordered_truncate(inode, new_size); return jbd2_journal_begin_ordered_truncate( EXT4_SB(inode->i_sb)->s_journal, &EXT4_I(inode)->jinode, @@ -178,6 +179,7 @@ void ext4_evict_inode(struct inode *inode) handle_t *handle; int err; + trace_ext4_evict_inode(inode); if (inode->i_nlink) { truncate_inode_pages(&inode->i_data, 0); goto no_delete; @@ -2123,9 +2125,12 @@ static int mpage_da_submit_io(struct mpage_da_data *mpd, */ if (unlikely(journal_data && PageChecked(page))) err = __ext4_journalled_writepage(page, len); - else + else if (test_opt(inode->i_sb, MBLK_IO_SUBMIT)) err = ext4_bio_write_page(&io_submit, page, len, mpd->wbc); + else + err = block_write_full_page(page, + noalloc_get_block_write, mpd->wbc); if (!err) mpd->pages_written++; @@ -5410,9 +5415,7 @@ int ext4_getattr(struct vfsmount *mnt, struct dentry *dentry, * will return the blocks that include the delayed allocation * blocks for this file. */ - spin_lock(&EXT4_I(inode)->i_block_reservation_lock); delalloc_blocks = EXT4_I(inode)->i_reserved_data_blocks; - spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); stat->blocks += (delalloc_blocks << inode->i_sb->s_blocksize_bits)>>9; return 0; @@ -5649,6 +5652,7 @@ int ext4_mark_inode_dirty(handle_t *handle, struct inode *inode) int err, ret; might_sleep(); + trace_ext4_mark_inode_dirty(inode, _RET_IP_); err = ext4_reserve_inode_write(handle, inode, &iloc); if (ext4_handle_valid(handle) && EXT4_I(inode)->i_extra_isize < sbi->s_want_extra_isize && diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c index bf5ae883b1b..eb3bc2fe647 100644 --- a/fs/ext4/ioctl.c +++ b/fs/ext4/ioctl.c @@ -331,6 +331,30 @@ mext_out: return err; } + case FITRIM: + { + struct super_block *sb = inode->i_sb; + struct fstrim_range range; + int ret = 0; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + if (copy_from_user(&range, (struct fstrim_range *)arg, + sizeof(range))) + return -EFAULT; + + ret = ext4_trim_fs(sb, &range); + if (ret < 0) + return ret; + + if (copy_to_user((struct fstrim_range *)arg, &range, + sizeof(range))) + return -EFAULT; + + return 0; + } + default: return -ENOTTY; } diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index c58eba34724..5b4d4e3a4d5 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -4640,8 +4640,6 @@ do_more: * with group lock held. generate_buddy look at * them with group lock_held */ - if (test_opt(sb, DISCARD)) - ext4_issue_discard(sb, block_group, bit, count); ext4_lock_group(sb, block_group); mb_clear_bits(bitmap_bh->b_data, bit, count); mb_free_blocks(inode, &e4b, bit, count); diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index 92203b8a099..dc40e75cba8 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -872,7 +872,7 @@ static struct buffer_head * ext4_find_entry (struct inode *dir, if (namelen > EXT4_NAME_LEN) return NULL; if ((namelen <= 2) && (name[0] == '.') && - (name[1] == '.' || name[1] == '0')) { + (name[1] == '.' || name[1] == '\0')) { /* * "." or ".." will only be in the first block * NFS may look up ".."; "." should be handled by the VFS diff --git a/fs/ext4/page-io.c b/fs/ext4/page-io.c index 46a7d6a9d97..beacce11ac5 100644 --- a/fs/ext4/page-io.c +++ b/fs/ext4/page-io.c @@ -32,8 +32,14 @@ static struct kmem_cache *io_page_cachep, *io_end_cachep; +#define WQ_HASH_SZ 37 +#define to_ioend_wq(v) (&ioend_wq[((unsigned long)v) % WQ_HASH_SZ]) +static wait_queue_head_t ioend_wq[WQ_HASH_SZ]; + int __init ext4_init_pageio(void) { + int i; + io_page_cachep = KMEM_CACHE(ext4_io_page, SLAB_RECLAIM_ACCOUNT); if (io_page_cachep == NULL) return -ENOMEM; @@ -42,6 +48,8 @@ int __init ext4_init_pageio(void) kmem_cache_destroy(io_page_cachep); return -ENOMEM; } + for (i = 0; i < WQ_HASH_SZ; i++) + init_waitqueue_head(&ioend_wq[i]); return 0; } @@ -52,24 +60,37 @@ void ext4_exit_pageio(void) kmem_cache_destroy(io_page_cachep); } +void ext4_ioend_wait(struct inode *inode) +{ + wait_queue_head_t *wq = to_ioend_wq(inode); + + wait_event(*wq, (atomic_read(&EXT4_I(inode)->i_ioend_count) == 0)); +} + +static void put_io_page(struct ext4_io_page *io_page) +{ + if (atomic_dec_and_test(&io_page->p_count)) { + end_page_writeback(io_page->p_page); + put_page(io_page->p_page); + kmem_cache_free(io_page_cachep, io_page); + } +} + void ext4_free_io_end(ext4_io_end_t *io) { int i; + wait_queue_head_t *wq; BUG_ON(!io); if (io->page) put_page(io->page); - for (i = 0; i < io->num_io_pages; i++) { - if (--io->pages[i]->p_count == 0) { - struct page *page = io->pages[i]->p_page; - - end_page_writeback(page); - put_page(page); - kmem_cache_free(io_page_cachep, io->pages[i]); - } - } + for (i = 0; i < io->num_io_pages; i++) + put_io_page(io->pages[i]); io->num_io_pages = 0; - iput(io->inode); + wq = to_ioend_wq(io->inode); + if (atomic_dec_and_test(&EXT4_I(io->inode)->i_ioend_count) && + waitqueue_active(wq)) + wake_up_all(wq); kmem_cache_free(io_end_cachep, io); } @@ -142,8 +163,8 @@ ext4_io_end_t *ext4_init_io_end(struct inode *inode, gfp_t flags) io = kmem_cache_alloc(io_end_cachep, flags); if (io) { memset(io, 0, sizeof(*io)); - io->inode = igrab(inode); - BUG_ON(!io->inode); + atomic_inc(&EXT4_I(inode)->i_ioend_count); + io->inode = inode; INIT_WORK(&io->work, ext4_end_io_work); INIT_LIST_HEAD(&io->list); } @@ -171,35 +192,15 @@ static void ext4_end_bio(struct bio *bio, int error) struct workqueue_struct *wq; struct inode *inode; unsigned long flags; - ext4_fsblk_t err_block; int i; BUG_ON(!io_end); - inode = io_end->inode; bio->bi_private = NULL; bio->bi_end_io = NULL; if (test_bit(BIO_UPTODATE, &bio->bi_flags)) error = 0; - err_block = bio->bi_sector >> (inode->i_blkbits - 9); bio_put(bio); - if (!(inode->i_sb->s_flags & MS_ACTIVE)) { - pr_err("sb umounted, discard end_io request for inode %lu\n", - io_end->inode->i_ino); - ext4_free_io_end(io_end); - return; - } - - if (error) { - io_end->flag |= EXT4_IO_END_ERROR; - ext4_warning(inode->i_sb, "I/O error writing to inode %lu " - "(offset %llu size %ld starting block %llu)", - inode->i_ino, - (unsigned long long) io_end->offset, - (long) io_end->size, - (unsigned long long) err_block); - } - for (i = 0; i < io_end->num_io_pages; i++) { struct page *page = io_end->pages[i]->p_page; struct buffer_head *bh, *head; @@ -236,14 +237,6 @@ static void ext4_end_bio(struct bio *bio, int error) } while (bh != head); } - if (--io_end->pages[i]->p_count == 0) { - struct page *page = io_end->pages[i]->p_page; - - end_page_writeback(page); - put_page(page); - kmem_cache_free(io_page_cachep, io_end->pages[i]); - } - /* * If this is a partial write which happened to make * all buffers uptodate then we can optimize away a @@ -253,9 +246,22 @@ static void ext4_end_bio(struct bio *bio, int error) */ if (!partial_write) SetPageUptodate(page); - } + put_io_page(io_end->pages[i]); + } io_end->num_io_pages = 0; + inode = io_end->inode; + + if (error) { + io_end->flag |= EXT4_IO_END_ERROR; + ext4_warning(inode->i_sb, "I/O error writing to inode %lu " + "(offset %llu size %ld starting block %llu)", + inode->i_ino, + (unsigned long long) io_end->offset, + (long) io_end->size, + (unsigned long long) + bio->bi_sector >> (inode->i_blkbits - 9)); + } /* Add the io_end to per-inode completed io list*/ spin_lock_irqsave(&EXT4_I(inode)->i_completed_io_lock, flags); @@ -305,7 +311,6 @@ static int io_submit_init(struct ext4_io_submit *io, bio->bi_private = io->io_end = io_end; bio->bi_end_io = ext4_end_bio; - io_end->inode = inode; io_end->offset = (page->index << PAGE_CACHE_SHIFT) + bh_offset(bh); io->io_bio = bio; @@ -360,7 +365,7 @@ submit_and_retry: if ((io_end->num_io_pages == 0) || (io_end->pages[io_end->num_io_pages-1] != io_page)) { io_end->pages[io_end->num_io_pages++] = io_page; - io_page->p_count++; + atomic_inc(&io_page->p_count); } return 0; } @@ -389,7 +394,7 @@ int ext4_bio_write_page(struct ext4_io_submit *io, return -ENOMEM; } io_page->p_page = page; - io_page->p_count = 0; + atomic_set(&io_page->p_count, 1); get_page(page); for (bh = head = page_buffers(page), block_start = 0; @@ -421,10 +426,6 @@ int ext4_bio_write_page(struct ext4_io_submit *io, * PageWriteback bit from the page to prevent the system from * wedging later on. */ - if (io_page->p_count == 0) { - put_page(page); - end_page_writeback(page); - kmem_cache_free(io_page_cachep, io_page); - } + put_io_page(io_page); return ret; } diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c index dc963929de6..981c8477ada 100644 --- a/fs/ext4/resize.c +++ b/fs/ext4/resize.c @@ -232,6 +232,8 @@ static int setup_new_group_blocks(struct super_block *sb, GFP_NOFS); if (err) goto exit_bh; + for (i = 0, bit = gdblocks + 1; i < reserved_gdb; i++, bit++) + ext4_set_bit(bit, bh->b_data); ext4_debug("mark block bitmap %#04llx (+%llu)\n", input->block_bitmap, input->block_bitmap - start); @@ -247,6 +249,9 @@ static int setup_new_group_blocks(struct super_block *sb, err = sb_issue_zeroout(sb, block, sbi->s_itb_per_group, GFP_NOFS); if (err) goto exit_bh; + for (i = 0, bit = input->inode_table - start; + i < sbi->s_itb_per_group; i++, bit++) + ext4_set_bit(bit, bh->b_data); if ((err = extend_or_restart_transaction(handle, 2, bh))) goto exit_bh; diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 40131b777af..cd37f9d5e44 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -828,12 +828,29 @@ static struct inode *ext4_alloc_inode(struct super_block *sb) ei->cur_aio_dio = NULL; ei->i_sync_tid = 0; ei->i_datasync_tid = 0; + atomic_set(&ei->i_ioend_count, 0); return &ei->vfs_inode; } +static int ext4_drop_inode(struct inode *inode) +{ + int drop = generic_drop_inode(inode); + + trace_ext4_drop_inode(inode, drop); + return drop; +} + +static void ext4_i_callback(struct rcu_head *head) +{ + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); + kmem_cache_free(ext4_inode_cachep, EXT4_I(inode)); +} + static void ext4_destroy_inode(struct inode *inode) { + ext4_ioend_wait(inode); if (!list_empty(&(EXT4_I(inode)->i_orphan))) { ext4_msg(inode->i_sb, KERN_ERR, "Inode %lu (%p): orphan list check failed!", @@ -843,7 +860,7 @@ static void ext4_destroy_inode(struct inode *inode) true); dump_stack(); } - kmem_cache_free(ext4_inode_cachep, EXT4_I(inode)); + call_rcu(&inode->i_rcu, ext4_i_callback); } static void init_once(void *foo) @@ -1016,6 +1033,8 @@ static int ext4_show_options(struct seq_file *seq, struct vfsmount *vfs) !(def_mount_opts & EXT4_DEFM_NODELALLOC)) seq_puts(seq, ",nodelalloc"); + if (test_opt(sb, MBLK_IO_SUBMIT)) + seq_puts(seq, ",mblk_io_submit"); if (sbi->s_stripe) seq_printf(seq, ",stripe=%lu", sbi->s_stripe); /* @@ -1173,6 +1192,7 @@ static const struct super_operations ext4_sops = { .destroy_inode = ext4_destroy_inode, .write_inode = ext4_write_inode, .dirty_inode = ext4_dirty_inode, + .drop_inode = ext4_drop_inode, .evict_inode = ext4_evict_inode, .put_super = ext4_put_super, .sync_fs = ext4_sync_fs, @@ -1186,7 +1206,6 @@ static const struct super_operations ext4_sops = { .quota_write = ext4_quota_write, #endif .bdev_try_to_free_page = bdev_try_to_free_page, - .trim_fs = ext4_trim_fs }; static const struct super_operations ext4_nojournal_sops = { @@ -1194,6 +1213,7 @@ static const struct super_operations ext4_nojournal_sops = { .destroy_inode = ext4_destroy_inode, .write_inode = ext4_write_inode, .dirty_inode = ext4_dirty_inode, + .drop_inode = ext4_drop_inode, .evict_inode = ext4_evict_inode, .write_super = ext4_write_super, .put_super = ext4_put_super, @@ -1228,8 +1248,8 @@ enum { Opt_jqfmt_vfsold, Opt_jqfmt_vfsv0, Opt_jqfmt_vfsv1, Opt_quota, Opt_noquota, Opt_ignore, Opt_barrier, Opt_nobarrier, Opt_err, Opt_resize, Opt_usrquota, Opt_grpquota, Opt_i_version, - Opt_stripe, Opt_delalloc, Opt_nodelalloc, - Opt_block_validity, Opt_noblock_validity, + Opt_stripe, Opt_delalloc, Opt_nodelalloc, Opt_mblk_io_submit, + Opt_nomblk_io_submit, Opt_block_validity, Opt_noblock_validity, Opt_inode_readahead_blks, Opt_journal_ioprio, Opt_dioread_nolock, Opt_dioread_lock, Opt_discard, Opt_nodiscard, @@ -1293,6 +1313,8 @@ static const match_table_t tokens = { {Opt_resize, "resize"}, {Opt_delalloc, "delalloc"}, {Opt_nodelalloc, "nodelalloc"}, + {Opt_mblk_io_submit, "mblk_io_submit"}, + {Opt_nomblk_io_submit, "nomblk_io_submit"}, {Opt_block_validity, "block_validity"}, {Opt_noblock_validity, "noblock_validity"}, {Opt_inode_readahead_blks, "inode_readahead_blks=%u"}, @@ -1714,6 +1736,12 @@ set_qf_format: case Opt_nodelalloc: clear_opt(sbi->s_mount_opt, DELALLOC); break; + case Opt_mblk_io_submit: + set_opt(sbi->s_mount_opt, MBLK_IO_SUBMIT); + break; + case Opt_nomblk_io_submit: + clear_opt(sbi->s_mount_opt, MBLK_IO_SUBMIT); + break; case Opt_stripe: if (match_int(&args[0], &option)) return 0; @@ -2699,7 +2727,6 @@ static int ext4_lazyinit_thread(void *arg) struct ext4_li_request *elr; unsigned long next_wakeup; DEFINE_WAIT(wait); - int ret; BUG_ON(NULL == eli); @@ -2723,13 +2750,12 @@ cont_thread: elr = list_entry(pos, struct ext4_li_request, lr_request); - if (time_after_eq(jiffies, elr->lr_next_sched)) - ret = ext4_run_li_request(elr); - - if (ret) { - ret = 0; - ext4_remove_li_request(elr); - continue; + if (time_after_eq(jiffies, elr->lr_next_sched)) { + if (ext4_run_li_request(elr) != 0) { + /* error, remove the lazy_init job */ + ext4_remove_li_request(elr); + continue; + } } if (time_before(elr->lr_next_sched, next_wakeup)) @@ -2740,7 +2766,8 @@ cont_thread: if (freezing(current)) refrigerator(); - if (time_after_eq(jiffies, next_wakeup)) { + if ((time_after_eq(jiffies, next_wakeup)) || + (MAX_JIFFY_OFFSET == next_wakeup)) { cond_resched(); continue; } @@ -2788,9 +2815,6 @@ static void ext4_clear_request_list(void) struct ext4_li_request *elr; mutex_lock(&ext4_li_info->li_list_mtx); - if (list_empty(&ext4_li_info->li_request_list)) - return; - list_for_each_safe(pos, n, &ext4_li_info->li_request_list) { elr = list_entry(pos, struct ext4_li_request, lr_request); @@ -3257,13 +3281,14 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) * Test whether we have more sectors than will fit in sector_t, * and whether the max offset is addressable by the page cache. */ - ret = generic_check_addressable(sb->s_blocksize_bits, + err = generic_check_addressable(sb->s_blocksize_bits, ext4_blocks_count(es)); - if (ret) { + if (err) { ext4_msg(sb, KERN_ERR, "filesystem" " too large to mount safely on this system"); if (sizeof(sector_t) < 8) ext4_msg(sb, KERN_WARNING, "CONFIG_LBDAF not enabled"); + ret = err; goto failed_mount; } @@ -3348,6 +3373,24 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) get_random_bytes(&sbi->s_next_generation, sizeof(u32)); spin_lock_init(&sbi->s_next_gen_lock); + err = percpu_counter_init(&sbi->s_freeblocks_counter, + ext4_count_free_blocks(sb)); + if (!err) { + err = percpu_counter_init(&sbi->s_freeinodes_counter, + ext4_count_free_inodes(sb)); + } + if (!err) { + err = percpu_counter_init(&sbi->s_dirs_counter, + ext4_count_dirs(sb)); + } + if (!err) { + err = percpu_counter_init(&sbi->s_dirtyblocks_counter, 0); + } + if (err) { + ext4_msg(sb, KERN_ERR, "insufficient memory"); + goto failed_mount3; + } + sbi->s_stripe = ext4_get_stripe_size(sbi); sbi->s_max_writeback_mb_bump = 128; @@ -3446,22 +3489,19 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) } set_task_ioprio(sbi->s_journal->j_task, journal_ioprio); -no_journal: - err = percpu_counter_init(&sbi->s_freeblocks_counter, - ext4_count_free_blocks(sb)); - if (!err) - err = percpu_counter_init(&sbi->s_freeinodes_counter, - ext4_count_free_inodes(sb)); - if (!err) - err = percpu_counter_init(&sbi->s_dirs_counter, - ext4_count_dirs(sb)); - if (!err) - err = percpu_counter_init(&sbi->s_dirtyblocks_counter, 0); - if (err) { - ext4_msg(sb, KERN_ERR, "insufficient memory"); - goto failed_mount_wq; - } + /* + * The journal may have updated the bg summary counts, so we + * need to update the global counters. + */ + percpu_counter_set(&sbi->s_freeblocks_counter, + ext4_count_free_blocks(sb)); + percpu_counter_set(&sbi->s_freeinodes_counter, + ext4_count_free_inodes(sb)); + percpu_counter_set(&sbi->s_dirs_counter, + ext4_count_dirs(sb)); + percpu_counter_set(&sbi->s_dirtyblocks_counter, 0); +no_journal: EXT4_SB(sb)->dio_unwritten_wq = create_workqueue("ext4-dio-unwritten"); if (!EXT4_SB(sb)->dio_unwritten_wq) { printk(KERN_ERR "EXT4-fs: failed to create DIO workqueue\n"); @@ -3611,10 +3651,6 @@ failed_mount_wq: jbd2_journal_destroy(sbi->s_journal); sbi->s_journal = NULL; } - percpu_counter_destroy(&sbi->s_freeblocks_counter); - percpu_counter_destroy(&sbi->s_freeinodes_counter); - percpu_counter_destroy(&sbi->s_dirs_counter); - percpu_counter_destroy(&sbi->s_dirtyblocks_counter); failed_mount3: if (sbi->s_flex_groups) { if (is_vmalloc_addr(sbi->s_flex_groups)) @@ -3622,6 +3658,10 @@ failed_mount3: else kfree(sbi->s_flex_groups); } + percpu_counter_destroy(&sbi->s_freeblocks_counter); + percpu_counter_destroy(&sbi->s_freeinodes_counter); + percpu_counter_destroy(&sbi->s_dirs_counter); + percpu_counter_destroy(&sbi->s_dirtyblocks_counter); failed_mount2: for (i = 0; i < db_count; i++) brelse(sbi->s_group_desc[i]); @@ -3949,13 +3989,11 @@ static int ext4_commit_super(struct super_block *sb, int sync) else es->s_kbytes_written = cpu_to_le64(EXT4_SB(sb)->s_kbytes_written); - if (percpu_counter_initialized(&EXT4_SB(sb)->s_freeblocks_counter)) - ext4_free_blocks_count_set(es, percpu_counter_sum_positive( - &EXT4_SB(sb)->s_freeblocks_counter)); - if (percpu_counter_initialized(&EXT4_SB(sb)->s_freeinodes_counter)) - es->s_free_inodes_count = - cpu_to_le32(percpu_counter_sum_positive( - &EXT4_SB(sb)->s_freeinodes_counter)); + ext4_free_blocks_count_set(es, percpu_counter_sum_positive( + &EXT4_SB(sb)->s_freeblocks_counter)); + es->s_free_inodes_count = + cpu_to_le32(percpu_counter_sum_positive( + &EXT4_SB(sb)->s_freeinodes_counter)); sb->s_dirt = 0; BUFFER_TRACE(sbh, "marking dirty"); mark_buffer_dirty(sbh); @@ -4556,12 +4594,10 @@ static int ext4_quota_on(struct super_block *sb, int type, int format_id, static int ext4_quota_off(struct super_block *sb, int type) { - /* Force all delayed allocation blocks to be allocated */ - if (test_opt(sb, DELALLOC)) { - down_read(&sb->s_umount); + /* Force all delayed allocation blocks to be allocated. + * Caller already holds s_umount sem */ + if (test_opt(sb, DELALLOC)) sync_filesystem(sb); - up_read(&sb->s_umount); - } return dquot_quota_off(sb, type); } diff --git a/fs/fat/inode.c b/fs/fat/inode.c index ad6998a92c3..206351af7c5 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -514,11 +514,18 @@ static struct inode *fat_alloc_inode(struct super_block *sb) return &ei->vfs_inode; } -static void fat_destroy_inode(struct inode *inode) +static void fat_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(fat_inode_cachep, MSDOS_I(inode)); } +static void fat_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, fat_i_callback); +} + static void init_once(void *foo) { struct msdos_inode_info *ei = (struct msdos_inode_info *)foo; @@ -743,7 +750,7 @@ static struct dentry *fat_fh_to_dentry(struct super_block *sb, */ result = d_obtain_alias(inode); if (!IS_ERR(result)) - result->d_op = sb->s_root->d_op; + d_set_d_op(result, sb->s_root->d_op); return result; } @@ -793,7 +800,7 @@ static struct dentry *fat_get_parent(struct dentry *child) parent = d_obtain_alias(inode); if (!IS_ERR(parent)) - parent->d_op = sb->s_root->d_op; + d_set_d_op(parent, sb->s_root->d_op); out: unlock_super(sb); diff --git a/fs/fat/namei_msdos.c b/fs/fat/namei_msdos.c index 3345aabd1dd..35ffe43afa4 100644 --- a/fs/fat/namei_msdos.c +++ b/fs/fat/namei_msdos.c @@ -148,7 +148,8 @@ 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(struct dentry *dentry, struct qstr *qstr) +static int msdos_hash(const struct dentry *dentry, const struct inode *inode, + struct qstr *qstr) { struct fat_mount_options *options = &MSDOS_SB(dentry->d_sb)->options; unsigned char msdos_name[MSDOS_NAME]; @@ -164,16 +165,18 @@ static int msdos_hash(struct dentry *dentry, struct qstr *qstr) * Compare two msdos names. If either of the names are invalid, * we fall back to doing the standard name comparison. */ -static int msdos_cmp(struct dentry *dentry, struct qstr *a, struct qstr *b) +static int msdos_cmp(const struct dentry *parent, const struct inode *pinode, + const struct dentry *dentry, const struct inode *inode, + unsigned int len, const char *str, const struct qstr *name) { - struct fat_mount_options *options = &MSDOS_SB(dentry->d_sb)->options; + struct fat_mount_options *options = &MSDOS_SB(parent->d_sb)->options; unsigned char a_msdos_name[MSDOS_NAME], b_msdos_name[MSDOS_NAME]; int error; - error = msdos_format_name(a->name, a->len, a_msdos_name, options); + error = msdos_format_name(name->name, name->len, a_msdos_name, options); if (error) goto old_compare; - error = msdos_format_name(b->name, b->len, b_msdos_name, options); + error = msdos_format_name(str, len, b_msdos_name, options); if (error) goto old_compare; error = memcmp(a_msdos_name, b_msdos_name, MSDOS_NAME); @@ -182,8 +185,8 @@ out: old_compare: error = 1; - if (a->len == b->len) - error = memcmp(a->name, b->name, a->len); + if (name->len == len) + error = memcmp(name->name, str, len); goto out; } @@ -224,10 +227,10 @@ static struct dentry *msdos_lookup(struct inode *dir, struct dentry *dentry, } out: unlock_super(sb); - dentry->d_op = &msdos_dentry_operations; + d_set_d_op(dentry, &msdos_dentry_operations); dentry = d_splice_alias(inode, dentry); if (dentry) - dentry->d_op = &msdos_dentry_operations; + d_set_d_op(dentry, &msdos_dentry_operations); return dentry; error: @@ -670,7 +673,7 @@ static int msdos_fill_super(struct super_block *sb, void *data, int silent) } sb->s_flags |= MS_NOATIME; - sb->s_root->d_op = &msdos_dentry_operations; + d_set_d_op(sb->s_root, &msdos_dentry_operations); unlock_super(sb); return 0; } diff --git a/fs/fat/namei_vfat.c b/fs/fat/namei_vfat.c index b936703b892..e3ffc5e1233 100644 --- a/fs/fat/namei_vfat.c +++ b/fs/fat/namei_vfat.c @@ -43,6 +43,9 @@ static int vfat_revalidate_shortname(struct dentry *dentry) static int vfat_revalidate(struct dentry *dentry, struct nameidata *nd) { + if (nd->flags & LOOKUP_RCU) + return -ECHILD; + /* This is not negative dentry. Always valid. */ if (dentry->d_inode) return 1; @@ -51,6 +54,9 @@ static int vfat_revalidate(struct dentry *dentry, struct nameidata *nd) static int vfat_revalidate_ci(struct dentry *dentry, struct nameidata *nd) { + if (nd->flags & LOOKUP_RCU) + return -ECHILD; + /* * This is not negative dentry. Always valid. * @@ -85,22 +91,26 @@ static int vfat_revalidate_ci(struct dentry *dentry, struct nameidata *nd) } /* returns the length of a struct qstr, ignoring trailing dots */ -static unsigned int vfat_striptail_len(struct qstr *qstr) +static unsigned int __vfat_striptail_len(unsigned int len, const char *name) { - unsigned int len = qstr->len; - - while (len && qstr->name[len - 1] == '.') + while (len && name[len - 1] == '.') len--; return len; } +static unsigned int vfat_striptail_len(const struct qstr *qstr) +{ + return __vfat_striptail_len(qstr->len, qstr->name); +} + /* * Compute the hash for the vfat name corresponding to the dentry. * Note: if the name is invalid, we leave the hash code unchanged so * that the existing dentry can be used. The vfat fs routines will * return ENOENT or EINVAL as appropriate. */ -static int vfat_hash(struct dentry *dentry, struct qstr *qstr) +static int vfat_hash(const struct dentry *dentry, const struct inode *inode, + struct qstr *qstr) { qstr->hash = full_name_hash(qstr->name, vfat_striptail_len(qstr)); return 0; @@ -112,9 +122,10 @@ static int vfat_hash(struct dentry *dentry, struct qstr *qstr) * that the existing dentry can be used. The vfat fs routines will * return ENOENT or EINVAL as appropriate. */ -static int vfat_hashi(struct dentry *dentry, struct qstr *qstr) +static int vfat_hashi(const struct dentry *dentry, const struct inode *inode, + struct qstr *qstr) { - struct nls_table *t = MSDOS_SB(dentry->d_inode->i_sb)->nls_io; + struct nls_table *t = MSDOS_SB(dentry->d_sb)->nls_io; const unsigned char *name; unsigned int len; unsigned long hash; @@ -133,16 +144,18 @@ static int vfat_hashi(struct dentry *dentry, struct qstr *qstr) /* * Case insensitive compare of two vfat names. */ -static int vfat_cmpi(struct dentry *dentry, struct qstr *a, struct qstr *b) +static int vfat_cmpi(const struct dentry *parent, const struct inode *pinode, + const struct dentry *dentry, const struct inode *inode, + unsigned int len, const char *str, const struct qstr *name) { - struct nls_table *t = MSDOS_SB(dentry->d_inode->i_sb)->nls_io; + struct nls_table *t = MSDOS_SB(parent->d_sb)->nls_io; unsigned int alen, blen; /* A filename cannot end in '.' or we treat it like it has none */ - alen = vfat_striptail_len(a); - blen = vfat_striptail_len(b); + alen = vfat_striptail_len(name); + blen = __vfat_striptail_len(len, str); if (alen == blen) { - if (nls_strnicmp(t, a->name, b->name, alen) == 0) + if (nls_strnicmp(t, name->name, str, alen) == 0) return 0; } return 1; @@ -151,15 +164,17 @@ static int vfat_cmpi(struct dentry *dentry, struct qstr *a, struct qstr *b) /* * Case sensitive compare of two vfat names. */ -static int vfat_cmp(struct dentry *dentry, struct qstr *a, struct qstr *b) +static int vfat_cmp(const struct dentry *parent, const struct inode *pinode, + const struct dentry *dentry, const struct inode *inode, + unsigned int len, const char *str, const struct qstr *name) { unsigned int alen, blen; /* A filename cannot end in '.' or we treat it like it has none */ - alen = vfat_striptail_len(a); - blen = vfat_striptail_len(b); + alen = vfat_striptail_len(name); + blen = __vfat_striptail_len(len, str); if (alen == blen) { - if (strncmp(a->name, b->name, alen) == 0) + if (strncmp(name->name, str, alen) == 0) return 0; } return 1; @@ -757,11 +772,11 @@ static struct dentry *vfat_lookup(struct inode *dir, struct dentry *dentry, out: unlock_super(sb); - dentry->d_op = sb->s_root->d_op; + d_set_d_op(dentry, sb->s_root->d_op); dentry->d_time = dentry->d_parent->d_inode->i_version; dentry = d_splice_alias(inode, dentry); if (dentry) { - dentry->d_op = sb->s_root->d_op; + d_set_d_op(dentry, sb->s_root->d_op); dentry->d_time = dentry->d_parent->d_inode->i_version; } return dentry; @@ -1063,9 +1078,9 @@ static int vfat_fill_super(struct super_block *sb, void *data, int silent) } if (MSDOS_SB(sb)->options.name_check != 's') - sb->s_root->d_op = &vfat_ci_dentry_ops; + d_set_d_op(sb->s_root, &vfat_ci_dentry_ops); else - sb->s_root->d_op = &vfat_dentry_ops; + d_set_d_op(sb->s_root, &vfat_dentry_ops); unlock_super(sb); return 0; diff --git a/fs/filesystems.c b/fs/filesystems.c index 68ba492d8ee..751d6b255a1 100644 --- a/fs/filesystems.c +++ b/fs/filesystems.c @@ -115,6 +115,9 @@ int unregister_filesystem(struct file_system_type * fs) tmp = &(*tmp)->next; } write_unlock(&file_systems_lock); + + synchronize_rcu(); + return -EINVAL; } diff --git a/fs/freevxfs/vxfs_inode.c b/fs/freevxfs/vxfs_inode.c index 8c04eac5079..2ba6719ac61 100644 --- a/fs/freevxfs/vxfs_inode.c +++ b/fs/freevxfs/vxfs_inode.c @@ -337,6 +337,13 @@ vxfs_iget(struct super_block *sbp, ino_t ino) return ip; } +static void vxfs_i_callback(struct rcu_head *head) +{ + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); + kmem_cache_free(vxfs_inode_cachep, inode->i_private); +} + /** * vxfs_evict_inode - remove inode from main memory * @ip: inode to discard. @@ -350,5 +357,5 @@ vxfs_evict_inode(struct inode *ip) { truncate_inode_pages(&ip->i_data, 0); end_writeback(ip); - kmem_cache_free(vxfs_inode_cachep, ip->i_private); + call_rcu(&ip->i_rcu, vxfs_i_callback); } diff --git a/fs/fs_struct.c b/fs/fs_struct.c index ed45a9cf5f3..68ca487bedb 100644 --- a/fs/fs_struct.c +++ b/fs/fs_struct.c @@ -14,12 +14,14 @@ void set_fs_root(struct fs_struct *fs, struct path *path) struct path old_root; spin_lock(&fs->lock); + write_seqcount_begin(&fs->seq); old_root = fs->root; fs->root = *path; - path_get(path); + path_get_long(path); + write_seqcount_end(&fs->seq); spin_unlock(&fs->lock); if (old_root.dentry) - path_put(&old_root); + path_put_long(&old_root); } /* @@ -31,13 +33,15 @@ void set_fs_pwd(struct fs_struct *fs, struct path *path) struct path old_pwd; spin_lock(&fs->lock); + write_seqcount_begin(&fs->seq); old_pwd = fs->pwd; fs->pwd = *path; - path_get(path); + path_get_long(path); + write_seqcount_end(&fs->seq); spin_unlock(&fs->lock); if (old_pwd.dentry) - path_put(&old_pwd); + path_put_long(&old_pwd); } void chroot_fs_refs(struct path *old_root, struct path *new_root) @@ -52,31 +56,33 @@ void chroot_fs_refs(struct path *old_root, struct path *new_root) fs = p->fs; if (fs) { spin_lock(&fs->lock); + write_seqcount_begin(&fs->seq); if (fs->root.dentry == old_root->dentry && fs->root.mnt == old_root->mnt) { - path_get(new_root); + path_get_long(new_root); fs->root = *new_root; count++; } if (fs->pwd.dentry == old_root->dentry && fs->pwd.mnt == old_root->mnt) { - path_get(new_root); + path_get_long(new_root); fs->pwd = *new_root; count++; } + write_seqcount_end(&fs->seq); spin_unlock(&fs->lock); } task_unlock(p); } while_each_thread(g, p); read_unlock(&tasklist_lock); while (count--) - path_put(old_root); + path_put_long(old_root); } void free_fs_struct(struct fs_struct *fs) { - path_put(&fs->root); - path_put(&fs->pwd); + path_put_long(&fs->root); + path_put_long(&fs->pwd); kmem_cache_free(fs_cachep, fs); } @@ -88,8 +94,10 @@ void exit_fs(struct task_struct *tsk) int kill; task_lock(tsk); spin_lock(&fs->lock); + write_seqcount_begin(&fs->seq); tsk->fs = NULL; kill = !--fs->users; + write_seqcount_end(&fs->seq); spin_unlock(&fs->lock); task_unlock(tsk); if (kill) @@ -105,8 +113,15 @@ struct fs_struct *copy_fs_struct(struct fs_struct *old) fs->users = 1; fs->in_exec = 0; spin_lock_init(&fs->lock); + seqcount_init(&fs->seq); fs->umask = old->umask; - get_fs_root_and_pwd(old, &fs->root, &fs->pwd); + + spin_lock(&old->lock); + fs->root = old->root; + path_get_long(&fs->root); + fs->pwd = old->pwd; + path_get_long(&fs->pwd); + spin_unlock(&old->lock); } return fs; } @@ -144,6 +159,7 @@ EXPORT_SYMBOL(current_umask); struct fs_struct init_fs = { .users = 1, .lock = __SPIN_LOCK_UNLOCKED(init_fs.lock), + .seq = SEQCNT_ZERO, .umask = 0022, }; diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index c9627c95482..f738599fd8c 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -156,8 +156,12 @@ u64 fuse_get_attr_version(struct fuse_conn *fc) */ static int fuse_dentry_revalidate(struct dentry *entry, struct nameidata *nd) { - struct inode *inode = entry->d_inode; + struct inode *inode; + if (nd->flags & LOOKUP_RCU) + return -ECHILD; + + inode = entry->d_inode; if (inode && is_bad_inode(inode)) return 0; else if (fuse_dentry_time(entry) < get_jiffies_64()) { @@ -347,7 +351,7 @@ static struct dentry *fuse_lookup(struct inode *dir, struct dentry *entry, } entry = newent ? newent : entry; - entry->d_op = &fuse_dentry_operations; + d_set_d_op(entry, &fuse_dentry_operations); if (outarg_valid) fuse_change_entry_timeout(entry, &outarg); else @@ -981,12 +985,15 @@ static int fuse_access(struct inode *inode, int mask) * access request is sent. Execute permission is still checked * locally based on file mode. */ -static int fuse_permission(struct inode *inode, int mask) +static int fuse_permission(struct inode *inode, int mask, unsigned int flags) { struct fuse_conn *fc = get_fuse_conn(inode); bool refreshed = false; int err = 0; + if (flags & IPERM_FLAG_RCU) + return -ECHILD; + if (!fuse_allow_task(fc, current)) return -EACCES; @@ -1001,7 +1008,7 @@ static int fuse_permission(struct inode *inode, int mask) } if (fc->flags & FUSE_DEFAULT_PERMISSIONS) { - err = generic_permission(inode, mask, NULL); + err = generic_permission(inode, mask, flags, NULL); /* If permission is denied, try to refresh file attributes. This is also needed, because the root @@ -1009,7 +1016,8 @@ static int fuse_permission(struct inode *inode, int mask) if (err == -EACCES && !refreshed) { err = fuse_do_getattr(inode, NULL, NULL); if (!err) - err = generic_permission(inode, mask, NULL); + err = generic_permission(inode, mask, + flags, NULL); } /* Note: the opposite of the above test does not diff --git a/fs/fuse/file.c b/fs/fuse/file.c index c8224587123..8b984a2cebb 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -13,6 +13,7 @@ #include <linux/kernel.h> #include <linux/sched.h> #include <linux/module.h> +#include <linux/compat.h> static const struct file_operations fuse_direct_io_file_operations; @@ -134,6 +135,7 @@ EXPORT_SYMBOL_GPL(fuse_do_open); void fuse_finish_open(struct inode *inode, struct file *file) { struct fuse_file *ff = file->private_data; + struct fuse_conn *fc = get_fuse_conn(inode); if (ff->open_flags & FOPEN_DIRECT_IO) file->f_op = &fuse_direct_io_file_operations; @@ -141,6 +143,15 @@ void fuse_finish_open(struct inode *inode, struct file *file) invalidate_inode_pages2(inode->i_mapping); if (ff->open_flags & FOPEN_NONSEEKABLE) nonseekable_open(inode, file); + if (fc->atomic_o_trunc && (file->f_flags & O_TRUNC)) { + struct fuse_inode *fi = get_fuse_inode(inode); + + spin_lock(&fc->lock); + fi->attr_version = ++fc->attr_version; + i_size_write(inode, 0); + spin_unlock(&fc->lock); + fuse_invalidate_attr(inode); + } } int fuse_open_common(struct inode *inode, struct file *file, bool isdir) @@ -1618,6 +1629,58 @@ static int fuse_ioctl_copy_user(struct page **pages, struct iovec *iov, } /* + * CUSE servers compiled on 32bit broke on 64bit kernels because the + * ABI was defined to be 'struct iovec' which is different on 32bit + * and 64bit. Fortunately we can determine which structure the server + * used from the size of the reply. + */ +static int fuse_copy_ioctl_iovec(struct iovec *dst, void *src, + size_t transferred, unsigned count, + bool is_compat) +{ +#ifdef CONFIG_COMPAT + if (count * sizeof(struct compat_iovec) == transferred) { + struct compat_iovec *ciov = src; + unsigned i; + + /* + * With this interface a 32bit server cannot support + * non-compat (i.e. ones coming from 64bit apps) ioctl + * requests + */ + if (!is_compat) + return -EINVAL; + + for (i = 0; i < count; i++) { + dst[i].iov_base = compat_ptr(ciov[i].iov_base); + dst[i].iov_len = ciov[i].iov_len; + } + return 0; + } +#endif + + if (count * sizeof(struct iovec) != transferred) + return -EIO; + + memcpy(dst, src, transferred); + return 0; +} + +/* Make sure iov_length() won't overflow */ +static int fuse_verify_ioctl_iov(struct iovec *iov, size_t count) +{ + size_t n; + u32 max = FUSE_MAX_PAGES_PER_REQ << PAGE_SHIFT; + + for (n = 0; n < count; n++) { + if (iov->iov_len > (size_t) max) + return -ENOMEM; + max -= iov->iov_len; + } + return 0; +} + +/* * For ioctls, there is no generic way to determine how much memory * needs to be read and/or written. Furthermore, ioctls are allowed * to dereference the passed pointer, so the parameter requires deep @@ -1798,18 +1861,25 @@ long fuse_do_ioctl(struct file *file, unsigned int cmd, unsigned long arg, in_iovs + out_iovs > FUSE_IOCTL_MAX_IOV) goto out; - err = -EIO; - if ((in_iovs + out_iovs) * sizeof(struct iovec) != transferred) - goto out; - - /* okay, copy in iovs and retry */ vaddr = kmap_atomic(pages[0], KM_USER0); - memcpy(page_address(iov_page), vaddr, transferred); + err = fuse_copy_ioctl_iovec(page_address(iov_page), vaddr, + transferred, in_iovs + out_iovs, + (flags & FUSE_IOCTL_COMPAT) != 0); kunmap_atomic(vaddr, KM_USER0); + if (err) + goto out; in_iov = page_address(iov_page); out_iov = in_iov + in_iovs; + err = fuse_verify_ioctl_iov(in_iov, in_iovs); + if (err) + goto out; + + err = fuse_verify_ioctl_iov(out_iov, out_iovs); + if (err) + goto out; + goto retry; } diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index cfce3ad86a9..a8b31da19b9 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -99,6 +99,13 @@ static struct inode *fuse_alloc_inode(struct super_block *sb) return inode; } +static void fuse_i_callback(struct rcu_head *head) +{ + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); + kmem_cache_free(fuse_inode_cachep, inode); +} + static void fuse_destroy_inode(struct inode *inode) { struct fuse_inode *fi = get_fuse_inode(inode); @@ -106,7 +113,7 @@ static void fuse_destroy_inode(struct inode *inode) BUG_ON(!list_empty(&fi->queued_writes)); if (fi->forget_req) fuse_request_free(fi->forget_req); - kmem_cache_free(fuse_inode_cachep, inode); + call_rcu(&inode->i_rcu, fuse_i_callback); } void fuse_send_forget(struct fuse_conn *fc, struct fuse_req *req, @@ -619,7 +626,7 @@ static struct dentry *fuse_get_dentry(struct super_block *sb, entry = d_obtain_alias(inode); if (!IS_ERR(entry) && get_node_id(inode) != FUSE_ROOT_ID) { - entry->d_op = &fuse_dentry_operations; + d_set_d_op(entry, &fuse_dentry_operations); fuse_invalidate_entry_cache(entry); } @@ -721,7 +728,7 @@ static struct dentry *fuse_get_parent(struct dentry *child) parent = d_obtain_alias(inode); if (!IS_ERR(parent) && get_node_id(inode) != FUSE_ROOT_ID) { - parent->d_op = &fuse_dentry_operations; + d_set_d_op(parent, &fuse_dentry_operations); fuse_invalidate_entry_cache(parent); } diff --git a/fs/generic_acl.c b/fs/generic_acl.c index 6bc9e3a5a69..06c48a89183 100644 --- a/fs/generic_acl.c +++ b/fs/generic_acl.c @@ -190,14 +190,20 @@ generic_acl_chmod(struct inode *inode) } int -generic_check_acl(struct inode *inode, int mask) +generic_check_acl(struct inode *inode, int mask, unsigned int flags) { - struct posix_acl *acl = get_cached_acl(inode, ACL_TYPE_ACCESS); - - if (acl) { - int error = posix_acl_permission(inode, acl, mask); - posix_acl_release(acl); - return error; + if (flags & IPERM_FLAG_RCU) { + if (!negative_cached_acl(inode, ACL_TYPE_ACCESS)) + return -ECHILD; + } else { + struct posix_acl *acl; + + acl = get_cached_acl(inode, ACL_TYPE_ACCESS); + if (acl) { + int error = posix_acl_permission(inode, acl, mask); + posix_acl_release(acl); + return error; + } } return -EAGAIN; } diff --git a/fs/gfs2/acl.c b/fs/gfs2/acl.c index 48171f4c943..7118f1a780a 100644 --- a/fs/gfs2/acl.c +++ b/fs/gfs2/acl.c @@ -75,11 +75,14 @@ static struct posix_acl *gfs2_acl_get(struct gfs2_inode *ip, int type) * Returns: errno */ -int gfs2_check_acl(struct inode *inode, int mask) +int gfs2_check_acl(struct inode *inode, int mask, unsigned int flags) { struct posix_acl *acl; int error; + if (flags & IPERM_FLAG_RCU) + return -ECHILD; + acl = gfs2_acl_get(GFS2_I(inode), ACL_TYPE_ACCESS); if (IS_ERR(acl)) return PTR_ERR(acl); diff --git a/fs/gfs2/acl.h b/fs/gfs2/acl.h index b522b0cb39e..a93907c8159 100644 --- a/fs/gfs2/acl.h +++ b/fs/gfs2/acl.h @@ -16,7 +16,7 @@ #define GFS2_POSIX_ACL_DEFAULT "posix_acl_default" #define GFS2_ACL_MAX_ENTRIES 25 -extern int gfs2_check_acl(struct inode *inode, int mask); +extern int gfs2_check_acl(struct inode *inode, int mask, unsigned int); extern int gfs2_acl_create(struct gfs2_inode *dip, struct inode *inode); extern int gfs2_acl_chmod(struct gfs2_inode *ip, struct iattr *attr); extern const struct xattr_handler gfs2_xattr_system_handler; diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c index 5476c066d4e..3c4039d5eef 100644 --- a/fs/gfs2/bmap.c +++ b/fs/gfs2/bmap.c @@ -763,7 +763,7 @@ static int do_strip(struct gfs2_inode *ip, struct buffer_head *dibh, int metadata; unsigned int revokes = 0; int x; - int error; + int error = 0; if (!*top) sm->sm_first = 0; @@ -780,7 +780,11 @@ static int do_strip(struct gfs2_inode *ip, struct buffer_head *dibh, if (metadata) revokes = (height) ? sdp->sd_inptrs : sdp->sd_diptrs; - error = gfs2_rindex_hold(sdp, &ip->i_alloc->al_ri_gh); + if (ip != GFS2_I(sdp->sd_rindex)) + error = gfs2_rindex_hold(sdp, &ip->i_alloc->al_ri_gh); + else if (!sdp->sd_rgrps) + error = gfs2_ri_update(ip); + if (error) return error; @@ -879,7 +883,8 @@ out_rg_gunlock: out_rlist: gfs2_rlist_free(&rlist); out: - gfs2_glock_dq_uninit(&ip->i_alloc->al_ri_gh); + if (ip != GFS2_I(sdp->sd_rindex)) + gfs2_glock_dq_uninit(&ip->i_alloc->al_ri_gh); return error; } diff --git a/fs/gfs2/dentry.c b/fs/gfs2/dentry.c index 6798755b385..4a456338b87 100644 --- a/fs/gfs2/dentry.c +++ b/fs/gfs2/dentry.c @@ -11,6 +11,7 @@ #include <linux/completion.h> #include <linux/buffer_head.h> #include <linux/gfs2_ondisk.h> +#include <linux/namei.h> #include <linux/crc32.h> #include "gfs2.h" @@ -34,15 +35,23 @@ static int gfs2_drevalidate(struct dentry *dentry, struct nameidata *nd) { - struct dentry *parent = dget_parent(dentry); - struct gfs2_sbd *sdp = GFS2_SB(parent->d_inode); - struct gfs2_inode *dip = GFS2_I(parent->d_inode); - struct inode *inode = dentry->d_inode; + struct dentry *parent; + struct gfs2_sbd *sdp; + struct gfs2_inode *dip; + struct inode *inode; struct gfs2_holder d_gh; struct gfs2_inode *ip = NULL; int error; int had_lock = 0; + if (nd->flags & LOOKUP_RCU) + return -ECHILD; + + parent = dget_parent(dentry); + sdp = GFS2_SB(parent->d_inode); + dip = GFS2_I(parent->d_inode); + inode = dentry->d_inode; + if (inode) { if (is_bad_inode(inode)) goto invalid; @@ -100,13 +109,14 @@ fail: return 0; } -static int gfs2_dhash(struct dentry *dentry, struct qstr *str) +static int gfs2_dhash(const struct dentry *dentry, const struct inode *inode, + struct qstr *str) { str->hash = gfs2_disk_hash(str->name, str->len); return 0; } -static int gfs2_dentry_delete(struct dentry *dentry) +static int gfs2_dentry_delete(const struct dentry *dentry) { struct gfs2_inode *ginode; diff --git a/fs/gfs2/export.c b/fs/gfs2/export.c index 06d582732d3..97012ecff56 100644 --- a/fs/gfs2/export.c +++ b/fs/gfs2/export.c @@ -130,7 +130,7 @@ static struct dentry *gfs2_get_parent(struct dentry *child) dentry = d_obtain_alias(gfs2_lookupi(child->d_inode, &gfs2_qdotdot, 1)); if (!IS_ERR(dentry)) - dentry->d_op = &gfs2_dops; + d_set_d_op(dentry, &gfs2_dops); return dentry; } @@ -138,10 +138,8 @@ static struct dentry *gfs2_get_dentry(struct super_block *sb, struct gfs2_inum_host *inum) { struct gfs2_sbd *sdp = sb->s_fs_info; - struct gfs2_holder i_gh; struct inode *inode; struct dentry *dentry; - int error; inode = gfs2_ilookup(sb, inum->no_addr); if (inode) { @@ -152,52 +150,16 @@ static struct dentry *gfs2_get_dentry(struct super_block *sb, goto out_inode; } - error = gfs2_glock_nq_num(sdp, inum->no_addr, &gfs2_inode_glops, - LM_ST_SHARED, LM_FLAG_ANY, &i_gh); - if (error) - return ERR_PTR(error); - - error = gfs2_check_blk_type(sdp, inum->no_addr, GFS2_BLKST_DINODE); - if (error) - goto fail; - - inode = gfs2_inode_lookup(sb, DT_UNKNOWN, inum->no_addr, 0); - if (IS_ERR(inode)) { - error = PTR_ERR(inode); - goto fail; - } - - error = gfs2_inode_refresh(GFS2_I(inode)); - if (error) { - iput(inode); - goto fail; - } - - /* Pick up the works we bypass in gfs2_inode_lookup */ - if (inode->i_state & I_NEW) - gfs2_set_iop(inode); - - if (GFS2_I(inode)->i_no_formal_ino != inum->no_formal_ino) { - iput(inode); - goto fail; - } - - error = -EIO; - if (GFS2_I(inode)->i_diskflags & GFS2_DIF_SYSTEM) { - iput(inode); - goto fail; - } - - gfs2_glock_dq_uninit(&i_gh); + inode = gfs2_lookup_by_inum(sdp, inum->no_addr, &inum->no_formal_ino, + GFS2_BLKST_DINODE); + if (IS_ERR(inode)) + return ERR_CAST(inode); out_inode: dentry = d_obtain_alias(inode); if (!IS_ERR(dentry)) - dentry->d_op = &gfs2_dops; + d_set_d_op(dentry, &gfs2_dops); return dentry; -fail: - gfs2_glock_dq_uninit(&i_gh); - return ERR_PTR(error); } static struct dentry *gfs2_fh_to_dentry(struct super_block *sb, struct fid *fid, diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c index aa996471ec5..fca6689e12e 100644 --- a/fs/gfs2/file.c +++ b/fs/gfs2/file.c @@ -241,7 +241,7 @@ static int do_gfs2_set_flags(struct file *filp, u32 reqflags, u32 mask) !capable(CAP_LINUX_IMMUTABLE)) goto out; if (!IS_IMMUTABLE(inode)) { - error = gfs2_permission(inode, MAY_WRITE); + error = gfs2_permission(inode, MAY_WRITE, 0); if (error) goto out; } diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index 87778857f09..08a8beb152e 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -541,21 +541,6 @@ out_locked: spin_unlock(&gl->gl_spin); } -static unsigned int gfs2_lm_lock(struct gfs2_sbd *sdp, void *lock, - unsigned int req_state, - unsigned int flags) -{ - int ret = LM_OUT_ERROR; - - if (!sdp->sd_lockstruct.ls_ops->lm_lock) - return req_state == LM_ST_UNLOCKED ? 0 : req_state; - - if (likely(!test_bit(SDF_SHUTDOWN, &sdp->sd_flags))) - ret = sdp->sd_lockstruct.ls_ops->lm_lock(lock, - req_state, flags); - return ret; -} - /** * do_xmote - Calls the DLM to change the state of a lock * @gl: The lock state @@ -575,13 +560,14 @@ __acquires(&gl->gl_spin) lck_flags &= (LM_FLAG_TRY | LM_FLAG_TRY_1CB | LM_FLAG_NOEXP | LM_FLAG_PRIORITY); - BUG_ON(gl->gl_state == target); - BUG_ON(gl->gl_state == gl->gl_target); + GLOCK_BUG_ON(gl, gl->gl_state == target); + GLOCK_BUG_ON(gl, gl->gl_state == gl->gl_target); if ((target == LM_ST_UNLOCKED || target == LM_ST_DEFERRED) && glops->go_inval) { set_bit(GLF_INVALIDATE_IN_PROGRESS, &gl->gl_flags); do_error(gl, 0); /* Fail queued try locks */ } + gl->gl_req = target; spin_unlock(&gl->gl_spin); if (glops->go_xmote_th) glops->go_xmote_th(gl); @@ -594,15 +580,17 @@ __acquires(&gl->gl_spin) gl->gl_state == LM_ST_DEFERRED) && !(lck_flags & (LM_FLAG_TRY | LM_FLAG_TRY_1CB))) lck_flags |= LM_FLAG_TRY_1CB; - ret = gfs2_lm_lock(sdp, gl, target, lck_flags); - if (!(ret & LM_OUT_ASYNC)) { - finish_xmote(gl, ret); + if (sdp->sd_lockstruct.ls_ops->lm_lock) { + /* lock_dlm */ + ret = sdp->sd_lockstruct.ls_ops->lm_lock(gl, target, lck_flags); + GLOCK_BUG_ON(gl, ret); + } else { /* lock_nolock */ + finish_xmote(gl, target); if (queue_delayed_work(glock_workqueue, &gl->gl_work, 0) == 0) gfs2_glock_put(gl); - } else { - GLOCK_BUG_ON(gl, ret != LM_OUT_ASYNC); } + spin_lock(&gl->gl_spin); } @@ -686,21 +674,20 @@ static void delete_work_func(struct work_struct *work) { struct gfs2_glock *gl = container_of(work, struct gfs2_glock, gl_delete); struct gfs2_sbd *sdp = gl->gl_sbd; - struct gfs2_inode *ip = NULL; + struct gfs2_inode *ip; struct inode *inode; - u64 no_addr = 0; + u64 no_addr = gl->gl_name.ln_number; + + ip = gl->gl_object; + /* Note: Unsafe to dereference ip as we don't hold right refs/locks */ - spin_lock(&gl->gl_spin); - ip = (struct gfs2_inode *)gl->gl_object; if (ip) - no_addr = ip->i_no_addr; - spin_unlock(&gl->gl_spin); - if (ip) { inode = gfs2_ilookup(sdp->sd_vfs, no_addr); - if (inode) { - d_prune_aliases(inode); - iput(inode); - } + else + inode = gfs2_lookup_by_inum(sdp, no_addr, NULL, GFS2_BLKST_UNLINKED); + if (inode && !IS_ERR(inode)) { + d_prune_aliases(inode); + iput(inode); } gfs2_glock_put(gl); } @@ -952,17 +939,22 @@ int gfs2_glock_wait(struct gfs2_holder *gh) void gfs2_print_dbg(struct seq_file *seq, const char *fmt, ...) { + struct va_format vaf; va_list args; va_start(args, fmt); + if (seq) { struct gfs2_glock_iter *gi = seq->private; vsprintf(gi->string, fmt, args); seq_printf(seq, gi->string); } else { - printk(KERN_ERR " "); - vprintk(fmt, args); + vaf.fmt = fmt; + vaf.va = &args; + + printk(KERN_ERR " %pV", &vaf); } + va_end(args); } @@ -1362,24 +1354,28 @@ static int gfs2_should_freeze(const struct gfs2_glock *gl) * @gl: Pointer to the glock * @ret: The return value from the dlm * + * The gl_reply field is under the gl_spin lock so that it is ok + * to use a bitfield shared with other glock state fields. */ void gfs2_glock_complete(struct gfs2_glock *gl, int ret) { struct lm_lockstruct *ls = &gl->gl_sbd->sd_lockstruct; + spin_lock(&gl->gl_spin); gl->gl_reply = ret; if (unlikely(test_bit(DFL_BLOCK_LOCKS, &ls->ls_flags))) { - spin_lock(&gl->gl_spin); if (gfs2_should_freeze(gl)) { set_bit(GLF_FROZEN, &gl->gl_flags); spin_unlock(&gl->gl_spin); return; } - spin_unlock(&gl->gl_spin); } + + spin_unlock(&gl->gl_spin); set_bit(GLF_REPLY_PENDING, &gl->gl_flags); + smp_wmb(); gfs2_glock_hold(gl); if (queue_delayed_work(glock_workqueue, &gl->gl_work, 0) == 0) gfs2_glock_put(gl); @@ -1627,18 +1623,17 @@ static const char *hflags2str(char *buf, unsigned flags, unsigned long iflags) static int dump_holder(struct seq_file *seq, const struct gfs2_holder *gh) { struct task_struct *gh_owner = NULL; - char buffer[KSYM_SYMBOL_LEN]; char flags_buf[32]; - sprint_symbol(buffer, gh->gh_ip); if (gh->gh_owner_pid) gh_owner = pid_task(gh->gh_owner_pid, PIDTYPE_PID); - gfs2_print_dbg(seq, " H: s:%s f:%s e:%d p:%ld [%s] %s\n", - state2str(gh->gh_state), - hflags2str(flags_buf, gh->gh_flags, gh->gh_iflags), - gh->gh_error, - gh->gh_owner_pid ? (long)pid_nr(gh->gh_owner_pid) : -1, - gh_owner ? gh_owner->comm : "(ended)", buffer); + gfs2_print_dbg(seq, " H: s:%s f:%s e:%d p:%ld [%s] %pS\n", + state2str(gh->gh_state), + hflags2str(flags_buf, gh->gh_flags, gh->gh_iflags), + gh->gh_error, + gh->gh_owner_pid ? (long)pid_nr(gh->gh_owner_pid) : -1, + gh_owner ? gh_owner->comm : "(ended)", + (void *)gh->gh_ip); return 0; } @@ -1783,12 +1778,13 @@ int __init gfs2_glock_init(void) } #endif - glock_workqueue = alloc_workqueue("glock_workqueue", WQ_RESCUER | + glock_workqueue = alloc_workqueue("glock_workqueue", WQ_MEM_RECLAIM | WQ_HIGHPRI | WQ_FREEZEABLE, 0); if (IS_ERR(glock_workqueue)) return PTR_ERR(glock_workqueue); - gfs2_delete_workqueue = alloc_workqueue("delete_workqueue", WQ_RESCUER | - WQ_FREEZEABLE, 0); + gfs2_delete_workqueue = alloc_workqueue("delete_workqueue", + WQ_MEM_RECLAIM | WQ_FREEZEABLE, + 0); if (IS_ERR(gfs2_delete_workqueue)) { destroy_workqueue(glock_workqueue); return PTR_ERR(gfs2_delete_workqueue); diff --git a/fs/gfs2/glock.h b/fs/gfs2/glock.h index db1c26d6d22..691851ceb61 100644 --- a/fs/gfs2/glock.h +++ b/fs/gfs2/glock.h @@ -87,11 +87,10 @@ enum { #define GL_ASYNC 0x00000040 #define GL_EXACT 0x00000080 #define GL_SKIP 0x00000100 -#define GL_ATIME 0x00000200 #define GL_NOCACHE 0x00000400 /* - * lm_lock() and lm_async_cb return flags + * lm_async_cb return flags * * LM_OUT_ST_MASK * Masks the lower two bits of lock state in the returned value. @@ -99,15 +98,11 @@ enum { * LM_OUT_CANCELED * The lock request was canceled. * - * LM_OUT_ASYNC - * The result of the request will be returned in an LM_CB_ASYNC callback. - * */ #define LM_OUT_ST_MASK 0x00000003 #define LM_OUT_CANCELED 0x00000008 -#define LM_OUT_ASYNC 0x00000080 -#define LM_OUT_ERROR 0x00000100 +#define LM_OUT_ERROR 0x00000004 /* * lm_recovery_done() messages @@ -124,25 +119,12 @@ struct lm_lockops { void (*lm_unmount) (struct gfs2_sbd *sdp); void (*lm_withdraw) (struct gfs2_sbd *sdp); void (*lm_put_lock) (struct kmem_cache *cachep, struct gfs2_glock *gl); - unsigned int (*lm_lock) (struct gfs2_glock *gl, - unsigned int req_state, unsigned int flags); + int (*lm_lock) (struct gfs2_glock *gl, unsigned int req_state, + unsigned int flags); void (*lm_cancel) (struct gfs2_glock *gl); const match_table_t *lm_tokens; }; -#define LM_FLAG_TRY 0x00000001 -#define LM_FLAG_TRY_1CB 0x00000002 -#define LM_FLAG_NOEXP 0x00000004 -#define LM_FLAG_ANY 0x00000008 -#define LM_FLAG_PRIORITY 0x00000010 - -#define GL_ASYNC 0x00000040 -#define GL_EXACT 0x00000080 -#define GL_SKIP 0x00000100 -#define GL_NOCACHE 0x00000400 - -#define GLR_TRYFAILED 13 - extern struct workqueue_struct *gfs2_delete_workqueue; static inline struct gfs2_holder *gfs2_glock_is_locked_by_me(struct gfs2_glock *gl) { @@ -212,6 +194,8 @@ int gfs2_glock_nq_num(struct gfs2_sbd *sdp, int gfs2_glock_nq_m(unsigned int num_gh, struct gfs2_holder *ghs); void gfs2_glock_dq_m(unsigned int num_gh, struct gfs2_holder *ghs); void gfs2_glock_dq_uninit_m(unsigned int num_gh, struct gfs2_holder *ghs); + +__attribute__ ((format(printf, 2, 3))) void gfs2_print_dbg(struct seq_file *seq, const char *fmt, ...); /** diff --git a/fs/gfs2/glops.c b/fs/gfs2/glops.c index 0d149dcc04e..263561bf1a5 100644 --- a/fs/gfs2/glops.c +++ b/fs/gfs2/glops.c @@ -325,7 +325,6 @@ static void trans_go_sync(struct gfs2_glock *gl) if (gl->gl_state != LM_ST_UNLOCKED && test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags)) { - flush_workqueue(gfs2_delete_workqueue); gfs2_meta_syncfs(sdp); gfs2_log_shutdown(sdp); } diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h index 764fbb49efc..8d3d2b4a0a7 100644 --- a/fs/gfs2/incore.h +++ b/fs/gfs2/incore.h @@ -207,12 +207,14 @@ struct gfs2_glock { spinlock_t gl_spin; - unsigned int gl_state; - unsigned int gl_target; - unsigned int gl_reply; + /* State fields protected by gl_spin */ + unsigned int gl_state:2, /* Current state */ + gl_target:2, /* Target state */ + gl_demote_state:2, /* State requested by remote node */ + gl_req:2, /* State in last dlm request */ + gl_reply:8; /* Last reply from the dlm */ + unsigned int gl_hash; - unsigned int gl_req; - unsigned int gl_demote_state; /* state requested by remote node */ unsigned long gl_demote_time; /* time of first demote request */ struct list_head gl_holders; diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index 06370f8bd8c..2232b3c780b 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -73,49 +73,6 @@ static struct inode *gfs2_iget(struct super_block *sb, u64 no_addr) return iget5_locked(sb, hash, iget_test, iget_set, &no_addr); } -struct gfs2_skip_data { - u64 no_addr; - int skipped; -}; - -static int iget_skip_test(struct inode *inode, void *opaque) -{ - struct gfs2_inode *ip = GFS2_I(inode); - struct gfs2_skip_data *data = opaque; - - if (ip->i_no_addr == data->no_addr) { - if (inode->i_state & (I_FREEING|I_WILL_FREE)){ - data->skipped = 1; - return 0; - } - return 1; - } - return 0; -} - -static int iget_skip_set(struct inode *inode, void *opaque) -{ - struct gfs2_inode *ip = GFS2_I(inode); - struct gfs2_skip_data *data = opaque; - - if (data->skipped) - return 1; - inode->i_ino = (unsigned long)(data->no_addr); - ip->i_no_addr = data->no_addr; - return 0; -} - -static struct inode *gfs2_iget_skip(struct super_block *sb, - u64 no_addr) -{ - struct gfs2_skip_data data; - unsigned long hash = (unsigned long)no_addr; - - data.no_addr = no_addr; - data.skipped = 0; - return iget5_locked(sb, hash, iget_skip_test, iget_skip_set, &data); -} - /** * GFS2 lookup code fills in vfs inode contents based on info obtained * from directory entry inside gfs2_inode_lookup(). This has caused issues @@ -243,93 +200,54 @@ fail: return ERR_PTR(error); } -/** - * gfs2_process_unlinked_inode - Lookup an unlinked inode for reclamation - * and try to reclaim it by doing iput. - * - * This function assumes no rgrp locks are currently held. - * - * @sb: The super block - * no_addr: The inode number - * - */ - -void gfs2_process_unlinked_inode(struct super_block *sb, u64 no_addr) +struct inode *gfs2_lookup_by_inum(struct gfs2_sbd *sdp, u64 no_addr, + u64 *no_formal_ino, unsigned int blktype) { - struct gfs2_sbd *sdp; - struct gfs2_inode *ip; - struct gfs2_glock *io_gl = NULL; - int error; - struct gfs2_holder gh; + struct super_block *sb = sdp->sd_vfs; + struct gfs2_holder i_gh; struct inode *inode; + int error; - inode = gfs2_iget_skip(sb, no_addr); - - if (!inode) - return; - - /* If it's not a new inode, someone's using it, so leave it alone. */ - if (!(inode->i_state & I_NEW)) { - iput(inode); - return; - } - - ip = GFS2_I(inode); - sdp = GFS2_SB(inode); - ip->i_no_formal_ino = -1; + error = gfs2_glock_nq_num(sdp, no_addr, &gfs2_inode_glops, + LM_ST_SHARED, LM_FLAG_ANY, &i_gh); + if (error) + return ERR_PTR(error); - error = gfs2_glock_get(sdp, no_addr, &gfs2_inode_glops, CREATE, &ip->i_gl); - if (unlikely(error)) + error = gfs2_check_blk_type(sdp, no_addr, blktype); + if (error) goto fail; - ip->i_gl->gl_object = ip; - - error = gfs2_glock_get(sdp, no_addr, &gfs2_iopen_glops, CREATE, &io_gl); - if (unlikely(error)) - goto fail_put; - set_bit(GIF_INVALID, &ip->i_flags); - error = gfs2_glock_nq_init(io_gl, LM_ST_SHARED, LM_FLAG_TRY | GL_EXACT, - &ip->i_iopen_gh); - if (unlikely(error)) - goto fail_iopen; + inode = gfs2_inode_lookup(sb, DT_UNKNOWN, no_addr, 0); + if (IS_ERR(inode)) + goto fail; - ip->i_iopen_gh.gh_gl->gl_object = ip; - gfs2_glock_put(io_gl); - io_gl = NULL; + error = gfs2_inode_refresh(GFS2_I(inode)); + if (error) + goto fail_iput; - inode->i_mode = DT2IF(DT_UNKNOWN); + /* Pick up the works we bypass in gfs2_inode_lookup */ + if (inode->i_state & I_NEW) + gfs2_set_iop(inode); - /* - * We must read the inode in order to work out its type in - * this case. Note that this doesn't happen often as we normally - * know the type beforehand. This code path only occurs during - * unlinked inode recovery (where it is safe to do this glock, - * which is not true in the general case). - */ - error = gfs2_glock_nq_init(ip->i_gl, LM_ST_EXCLUSIVE, LM_FLAG_TRY, - &gh); - if (unlikely(error)) - goto fail_glock; + /* Two extra checks for NFS only */ + if (no_formal_ino) { + error = -ESTALE; + if (GFS2_I(inode)->i_no_formal_ino != *no_formal_ino) + goto fail_iput; - /* Inode is now uptodate */ - gfs2_glock_dq_uninit(&gh); - gfs2_set_iop(inode); + error = -EIO; + if (GFS2_I(inode)->i_diskflags & GFS2_DIF_SYSTEM) + goto fail_iput; - /* The iput will cause it to be deleted. */ - iput(inode); - return; + error = 0; + } -fail_glock: - gfs2_glock_dq(&ip->i_iopen_gh); -fail_iopen: - if (io_gl) - gfs2_glock_put(io_gl); -fail_put: - ip->i_gl->gl_object = NULL; - gfs2_glock_put(ip->i_gl); fail: - iget_failed(inode); - return; + gfs2_glock_dq_uninit(&i_gh); + return error ? ERR_PTR(error) : inode; +fail_iput: + iput(inode); + goto fail; } static int gfs2_dinode_in(struct gfs2_inode *ip, const void *buf) @@ -591,7 +509,7 @@ struct inode *gfs2_lookupi(struct inode *dir, const struct qstr *name, } if (!is_root) { - error = gfs2_permission(dir, MAY_EXEC); + error = gfs2_permission(dir, MAY_EXEC, 0); if (error) goto out; } @@ -621,7 +539,7 @@ static int create_ok(struct gfs2_inode *dip, const struct qstr *name, { int error; - error = gfs2_permission(&dip->i_inode, MAY_WRITE | MAY_EXEC); + error = gfs2_permission(&dip->i_inode, MAY_WRITE | MAY_EXEC, 0); if (error) return error; @@ -998,17 +916,8 @@ static int __gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr) if (error) return error; - if ((attr->ia_valid & ATTR_SIZE) && - attr->ia_size != i_size_read(inode)) { - error = vmtruncate(inode, attr->ia_size); - if (error) - return error; - } - setattr_copy(inode, attr); mark_inode_dirty(inode); - - gfs2_assert_warn(GFS2_SB(inode), !error); gfs2_trans_add_bh(ip->i_gl, dibh, 1); gfs2_dinode_out(ip, dibh->b_data); brelse(dibh); diff --git a/fs/gfs2/inode.h b/fs/gfs2/inode.h index 6720d7d5fbc..732a183efdb 100644 --- a/fs/gfs2/inode.h +++ b/fs/gfs2/inode.h @@ -99,7 +99,9 @@ err: extern void gfs2_set_iop(struct inode *inode); extern struct inode *gfs2_inode_lookup(struct super_block *sb, unsigned type, u64 no_addr, u64 no_formal_ino); -extern void gfs2_process_unlinked_inode(struct super_block *sb, u64 no_addr); +extern struct inode *gfs2_lookup_by_inum(struct gfs2_sbd *sdp, u64 no_addr, + u64 *no_formal_ino, + unsigned int blktype); extern struct inode *gfs2_ilookup(struct super_block *sb, u64 no_addr); extern int gfs2_inode_refresh(struct gfs2_inode *ip); @@ -111,7 +113,7 @@ extern struct inode *gfs2_lookupi(struct inode *dir, const struct qstr *name, extern struct inode *gfs2_createi(struct gfs2_holder *ghs, const struct qstr *name, unsigned int mode, dev_t dev); -extern int gfs2_permission(struct inode *inode, int mask); +extern int gfs2_permission(struct inode *inode, int mask, unsigned int flags); extern int gfs2_setattr_simple(struct gfs2_inode *ip, struct iattr *attr); extern struct inode *gfs2_lookup_simple(struct inode *dip, const char *name); extern void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf); diff --git a/fs/gfs2/lock_dlm.c b/fs/gfs2/lock_dlm.c index 1c09425b45f..6e493aee28f 100644 --- a/fs/gfs2/lock_dlm.c +++ b/fs/gfs2/lock_dlm.c @@ -146,15 +146,13 @@ static u32 make_flags(const u32 lkid, const unsigned int gfs_flags, return lkf; } -static unsigned int gdlm_lock(struct gfs2_glock *gl, - unsigned int req_state, unsigned int flags) +static int gdlm_lock(struct gfs2_glock *gl, unsigned int req_state, + unsigned int flags) { struct lm_lockstruct *ls = &gl->gl_sbd->sd_lockstruct; - int error; int req; u32 lkf; - gl->gl_req = req_state; req = make_mode(req_state); lkf = make_flags(gl->gl_lksb.sb_lkid, flags, req); @@ -162,13 +160,8 @@ static unsigned int gdlm_lock(struct gfs2_glock *gl, * Submit the actual lock request. */ - error = dlm_lock(ls->ls_dlm, req, &gl->gl_lksb, lkf, gl->gl_strname, - GDLM_STRNAME_BYTES - 1, 0, gdlm_ast, gl, gdlm_bast); - if (error == -EAGAIN) - return 0; - if (error) - return LM_OUT_ERROR; - return LM_OUT_ASYNC; + return dlm_lock(ls->ls_dlm, req, &gl->gl_lksb, lkf, gl->gl_strname, + GDLM_STRNAME_BYTES - 1, 0, gdlm_ast, gl, gdlm_bast); } static void gdlm_put_lock(struct kmem_cache *cachep, struct gfs2_glock *gl) diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index 3eb1393f7b8..2aeabd4218c 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -440,7 +440,7 @@ static int gfs2_lookup_root(struct super_block *sb, struct dentry **dptr, iput(inode); return -ENOMEM; } - dentry->d_op = &gfs2_dops; + d_set_d_op(dentry, &gfs2_dops); *dptr = dentry; return 0; } diff --git a/fs/gfs2/ops_inode.c b/fs/gfs2/ops_inode.c index 12cbea7502c..1501db4f0e6 100644 --- a/fs/gfs2/ops_inode.c +++ b/fs/gfs2/ops_inode.c @@ -106,7 +106,7 @@ static struct dentry *gfs2_lookup(struct inode *dir, struct dentry *dentry, { struct inode *inode = NULL; - dentry->d_op = &gfs2_dops; + d_set_d_op(dentry, &gfs2_dops); inode = gfs2_lookupi(dir, &dentry->d_name, 0); if (inode && IS_ERR(inode)) @@ -166,7 +166,7 @@ static int gfs2_link(struct dentry *old_dentry, struct inode *dir, if (error) goto out_child; - error = gfs2_permission(dir, MAY_WRITE | MAY_EXEC); + error = gfs2_permission(dir, MAY_WRITE | MAY_EXEC, 0); if (error) goto out_gunlock; @@ -289,7 +289,7 @@ static int gfs2_unlink_ok(struct gfs2_inode *dip, const struct qstr *name, if (IS_APPEND(&dip->i_inode)) return -EPERM; - error = gfs2_permission(&dip->i_inode, MAY_WRITE | MAY_EXEC); + error = gfs2_permission(&dip->i_inode, MAY_WRITE | MAY_EXEC, 0); if (error) return error; @@ -822,7 +822,7 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, } } } else { - error = gfs2_permission(ndir, MAY_WRITE | MAY_EXEC); + error = gfs2_permission(ndir, MAY_WRITE | MAY_EXEC, 0); if (error) goto out_gunlock; @@ -857,7 +857,7 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, /* Check out the dir to be renamed */ if (dir_rename) { - error = gfs2_permission(odentry->d_inode, MAY_WRITE); + error = gfs2_permission(odentry->d_inode, MAY_WRITE, 0); if (error) goto out_gunlock; } @@ -1041,13 +1041,17 @@ static void gfs2_put_link(struct dentry *dentry, struct nameidata *nd, void *p) * Returns: errno */ -int gfs2_permission(struct inode *inode, int mask) +int gfs2_permission(struct inode *inode, int mask, unsigned int flags) { - struct gfs2_inode *ip = GFS2_I(inode); + struct gfs2_inode *ip; struct gfs2_holder i_gh; int error; int unlock = 0; + if (flags & IPERM_FLAG_RCU) + return -ECHILD; + + ip = GFS2_I(inode); if (gfs2_glock_is_locked_by_me(ip->i_gl) == NULL) { error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &i_gh); if (error) @@ -1058,7 +1062,7 @@ int gfs2_permission(struct inode *inode, int mask) if ((mask & MAY_WRITE) && IS_IMMUTABLE(inode)) error = -EACCES; else - error = generic_permission(inode, mask, gfs2_check_acl); + error = generic_permission(inode, mask, flags, gfs2_check_acl); if (unlock) gfs2_glock_dq_uninit(&i_gh); @@ -1069,7 +1073,6 @@ static int setattr_chown(struct inode *inode, struct iattr *attr) { struct gfs2_inode *ip = GFS2_I(inode); struct gfs2_sbd *sdp = GFS2_SB(inode); - struct buffer_head *dibh; u32 ouid, ogid, nuid, ngid; int error; @@ -1100,25 +1103,10 @@ static int setattr_chown(struct inode *inode, struct iattr *attr) if (error) goto out_gunlock_q; - error = gfs2_meta_inode_buffer(ip, &dibh); + error = gfs2_setattr_simple(ip, attr); if (error) goto out_end_trans; - if ((attr->ia_valid & ATTR_SIZE) && - attr->ia_size != i_size_read(inode)) { - int error; - - error = vmtruncate(inode, attr->ia_size); - gfs2_assert_warn(sdp, !error); - } - - setattr_copy(inode, attr); - mark_inode_dirty(inode); - - gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(ip, dibh->b_data); - brelse(dibh); - if (ouid != NO_QUOTA_CHANGE || ogid != NO_QUOTA_CHANGE) { u64 blocks = gfs2_get_inode_blocks(&ip->i_inode); gfs2_quota_change(ip, -blocks, ouid, ogid); diff --git a/fs/gfs2/quota.c b/fs/gfs2/quota.c index 58a9b9998b4..a689901963d 100644 --- a/fs/gfs2/quota.c +++ b/fs/gfs2/quota.c @@ -631,6 +631,7 @@ static int gfs2_adjust_quota(struct gfs2_inode *ip, loff_t loc, struct fs_disk_quota *fdq) { struct inode *inode = &ip->i_inode; + struct gfs2_sbd *sdp = GFS2_SB(inode); struct address_space *mapping = inode->i_mapping; unsigned long index = loc >> PAGE_CACHE_SHIFT; unsigned offset = loc & (PAGE_CACHE_SIZE - 1); @@ -658,13 +659,17 @@ static int gfs2_adjust_quota(struct gfs2_inode *ip, loff_t loc, qd->qd_qb.qb_value = qp->qu_value; if (fdq) { if (fdq->d_fieldmask & FS_DQ_BSOFT) { - qp->qu_warn = cpu_to_be64(fdq->d_blk_softlimit); + qp->qu_warn = cpu_to_be64(fdq->d_blk_softlimit >> sdp->sd_fsb2bb_shift); qd->qd_qb.qb_warn = qp->qu_warn; } if (fdq->d_fieldmask & FS_DQ_BHARD) { - qp->qu_limit = cpu_to_be64(fdq->d_blk_hardlimit); + qp->qu_limit = cpu_to_be64(fdq->d_blk_hardlimit >> sdp->sd_fsb2bb_shift); qd->qd_qb.qb_limit = qp->qu_limit; } + if (fdq->d_fieldmask & FS_DQ_BCOUNT) { + qp->qu_value = cpu_to_be64(fdq->d_bcount >> sdp->sd_fsb2bb_shift); + qd->qd_qb.qb_value = qp->qu_value; + } } /* Write the quota into the quota file on disk */ @@ -1497,9 +1502,9 @@ static int gfs2_get_dqblk(struct super_block *sb, int type, qid_t id, fdq->d_version = FS_DQUOT_VERSION; fdq->d_flags = (type == QUOTA_USER) ? FS_USER_QUOTA : FS_GROUP_QUOTA; fdq->d_id = id; - fdq->d_blk_hardlimit = be64_to_cpu(qlvb->qb_limit); - fdq->d_blk_softlimit = be64_to_cpu(qlvb->qb_warn); - fdq->d_bcount = be64_to_cpu(qlvb->qb_value); + fdq->d_blk_hardlimit = be64_to_cpu(qlvb->qb_limit) << sdp->sd_fsb2bb_shift; + fdq->d_blk_softlimit = be64_to_cpu(qlvb->qb_warn) << sdp->sd_fsb2bb_shift; + fdq->d_bcount = be64_to_cpu(qlvb->qb_value) << sdp->sd_fsb2bb_shift; gfs2_glock_dq_uninit(&q_gh); out: @@ -1508,7 +1513,7 @@ out: } /* GFS2 only supports a subset of the XFS fields */ -#define GFS2_FIELDMASK (FS_DQ_BSOFT|FS_DQ_BHARD) +#define GFS2_FIELDMASK (FS_DQ_BSOFT|FS_DQ_BHARD|FS_DQ_BCOUNT) static int gfs2_set_dqblk(struct super_block *sb, int type, qid_t id, struct fs_disk_quota *fdq) @@ -1566,11 +1571,17 @@ static int gfs2_set_dqblk(struct super_block *sb, int type, qid_t id, /* If nothing has changed, this is a no-op */ if ((fdq->d_fieldmask & FS_DQ_BSOFT) && - (fdq->d_blk_softlimit == be64_to_cpu(qd->qd_qb.qb_warn))) + ((fdq->d_blk_softlimit >> sdp->sd_fsb2bb_shift) == be64_to_cpu(qd->qd_qb.qb_warn))) fdq->d_fieldmask ^= FS_DQ_BSOFT; + if ((fdq->d_fieldmask & FS_DQ_BHARD) && - (fdq->d_blk_hardlimit == be64_to_cpu(qd->qd_qb.qb_limit))) + ((fdq->d_blk_hardlimit >> sdp->sd_fsb2bb_shift) == be64_to_cpu(qd->qd_qb.qb_limit))) fdq->d_fieldmask ^= FS_DQ_BHARD; + + if ((fdq->d_fieldmask & FS_DQ_BCOUNT) && + ((fdq->d_bcount >> sdp->sd_fsb2bb_shift) == be64_to_cpu(qd->qd_qb.qb_value))) + fdq->d_fieldmask ^= FS_DQ_BCOUNT; + if (fdq->d_fieldmask == 0) goto out_i; @@ -1619,4 +1630,3 @@ const struct quotactl_ops gfs2_quotactl_ops = { .get_dqblk = gfs2_get_dqblk, .set_dqblk = gfs2_set_dqblk, }; - diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c index bef3ab6cf5c..7293ea27020 100644 --- a/fs/gfs2/rgrp.c +++ b/fs/gfs2/rgrp.c @@ -500,7 +500,7 @@ u64 gfs2_ri_total(struct gfs2_sbd *sdp) for (rgrps = 0;; rgrps++) { loff_t pos = rgrps * sizeof(struct gfs2_rindex); - if (pos + sizeof(struct gfs2_rindex) >= i_size_read(inode)) + if (pos + sizeof(struct gfs2_rindex) > i_size_read(inode)) break; error = gfs2_internal_read(ip, &ra_state, buf, &pos, sizeof(struct gfs2_rindex)); @@ -583,7 +583,7 @@ static int read_rindex_entry(struct gfs2_inode *ip, * Returns: 0 on successful update, error code otherwise */ -static int gfs2_ri_update(struct gfs2_inode *ip) +int gfs2_ri_update(struct gfs2_inode *ip) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); struct inode *inode = &ip->i_inode; @@ -614,46 +614,6 @@ static int gfs2_ri_update(struct gfs2_inode *ip) } /** - * gfs2_ri_update_special - Pull in a new resource index from the disk - * - * This is a special version that's safe to call from gfs2_inplace_reserve_i. - * In this case we know that we don't have any resource groups in memory yet. - * - * @ip: pointer to the rindex inode - * - * Returns: 0 on successful update, error code otherwise - */ -static int gfs2_ri_update_special(struct gfs2_inode *ip) -{ - struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); - struct inode *inode = &ip->i_inode; - struct file_ra_state ra_state; - struct gfs2_rgrpd *rgd; - unsigned int max_data = 0; - int error; - - file_ra_state_init(&ra_state, inode->i_mapping); - for (sdp->sd_rgrps = 0;; sdp->sd_rgrps++) { - /* Ignore partials */ - if ((sdp->sd_rgrps + 1) * sizeof(struct gfs2_rindex) > - i_size_read(inode)) - break; - error = read_rindex_entry(ip, &ra_state); - if (error) { - clear_rgrpdi(sdp); - return error; - } - } - list_for_each_entry(rgd, &sdp->sd_rindex_list, rd_list) - if (rgd->rd_data > max_data) - max_data = rgd->rd_data; - sdp->sd_max_rg_data = max_data; - - sdp->sd_rindex_uptodate = 1; - return 0; -} - -/** * gfs2_rindex_hold - Grab a lock on the rindex * @sdp: The GFS2 superblock * @ri_gh: the glock holder @@ -963,17 +923,18 @@ static int try_rgrp_fit(struct gfs2_rgrpd *rgd, struct gfs2_alloc *al) * The inode, if one has been found, in inode. */ -static u64 try_rgrp_unlink(struct gfs2_rgrpd *rgd, u64 *last_unlinked, - u64 skip) +static void try_rgrp_unlink(struct gfs2_rgrpd *rgd, u64 *last_unlinked, u64 skip) { u32 goal = 0, block; u64 no_addr; struct gfs2_sbd *sdp = rgd->rd_sbd; unsigned int n; + struct gfs2_glock *gl; + struct gfs2_inode *ip; + int error; + int found = 0; - for(;;) { - if (goal >= rgd->rd_data) - break; + while (goal < rgd->rd_data) { down_write(&sdp->sd_log_flush_lock); n = 1; block = rgblk_search(rgd, goal, GFS2_BLKST_UNLINKED, @@ -990,11 +951,32 @@ static u64 try_rgrp_unlink(struct gfs2_rgrpd *rgd, u64 *last_unlinked, if (no_addr == skip) continue; *last_unlinked = no_addr; - return no_addr; + + error = gfs2_glock_get(sdp, no_addr, &gfs2_inode_glops, CREATE, &gl); + if (error) + continue; + + /* If the inode is already in cache, we can ignore it here + * because the existing inode disposal code will deal with + * it when all refs have gone away. Accessing gl_object like + * this is not safe in general. Here it is ok because we do + * not dereference the pointer, and we only need an approx + * answer to whether it is NULL or not. + */ + ip = gl->gl_object; + + if (ip || queue_work(gfs2_delete_workqueue, &gl->gl_delete) == 0) + gfs2_glock_put(gl); + else + found++; + + /* Limit reclaim to sensible number of tasks */ + if (found > 2*NR_CPUS) + return; } rgd->rd_flags &= ~GFS2_RDF_CHECK; - return 0; + return; } /** @@ -1075,11 +1057,9 @@ static void forward_rgrp_set(struct gfs2_sbd *sdp, struct gfs2_rgrpd *rgd) * Try to acquire rgrp in way which avoids contending with others. * * Returns: errno - * unlinked: the block address of an unlinked block to be reclaimed */ -static int get_local_rgrp(struct gfs2_inode *ip, u64 *unlinked, - u64 *last_unlinked) +static int get_local_rgrp(struct gfs2_inode *ip, u64 *last_unlinked) { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); struct gfs2_rgrpd *rgd, *begin = NULL; @@ -1089,7 +1069,6 @@ static int get_local_rgrp(struct gfs2_inode *ip, u64 *unlinked, int loops = 0; int error, rg_locked; - *unlinked = 0; rgd = gfs2_blk2rgrpd(sdp, ip->i_goal); while (rgd) { @@ -1106,17 +1085,10 @@ static int get_local_rgrp(struct gfs2_inode *ip, u64 *unlinked, case 0: if (try_rgrp_fit(rgd, al)) goto out; - /* If the rg came in already locked, there's no - way we can recover from a failed try_rgrp_unlink - because that would require an iput which can only - happen after the rgrp is unlocked. */ - if (!rg_locked && rgd->rd_flags & GFS2_RDF_CHECK) - *unlinked = try_rgrp_unlink(rgd, last_unlinked, - ip->i_no_addr); + if (rgd->rd_flags & GFS2_RDF_CHECK) + try_rgrp_unlink(rgd, last_unlinked, ip->i_no_addr); if (!rg_locked) gfs2_glock_dq_uninit(&al->al_rgd_gh); - if (*unlinked) - return -EAGAIN; /* fall through */ case GLR_TRYFAILED: rgd = recent_rgrp_next(rgd); @@ -1145,13 +1117,10 @@ static int get_local_rgrp(struct gfs2_inode *ip, u64 *unlinked, case 0: if (try_rgrp_fit(rgd, al)) goto out; - if (!rg_locked && rgd->rd_flags & GFS2_RDF_CHECK) - *unlinked = try_rgrp_unlink(rgd, last_unlinked, - ip->i_no_addr); + if (rgd->rd_flags & GFS2_RDF_CHECK) + try_rgrp_unlink(rgd, last_unlinked, ip->i_no_addr); if (!rg_locked) gfs2_glock_dq_uninit(&al->al_rgd_gh); - if (*unlinked) - return -EAGAIN; break; case GLR_TRYFAILED: @@ -1204,12 +1173,12 @@ int gfs2_inplace_reserve_i(struct gfs2_inode *ip, int hold_rindex, struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); struct gfs2_alloc *al = ip->i_alloc; int error = 0; - u64 last_unlinked = NO_BLOCK, unlinked; + u64 last_unlinked = NO_BLOCK; + int tries = 0; if (gfs2_assert_warn(sdp, al->al_requested)) return -EINVAL; -try_again: if (hold_rindex) { /* We need to hold the rindex unless the inode we're using is the rindex itself, in which case it's already held. */ @@ -1217,32 +1186,33 @@ try_again: error = gfs2_rindex_hold(sdp, &al->al_ri_gh); else if (!sdp->sd_rgrps) /* We may not have the rindex read in, so: */ - error = gfs2_ri_update_special(ip); + error = gfs2_ri_update(ip); + if (error) + return error; } - if (error) - return error; +try_again: + do { + error = get_local_rgrp(ip, &last_unlinked); + /* If there is no space, flushing the log may release some */ + if (error) { + if (ip == GFS2_I(sdp->sd_rindex) && + !sdp->sd_rindex_uptodate) { + error = gfs2_ri_update(ip); + if (error) + return error; + goto try_again; + } + gfs2_log_flush(sdp, NULL); + } + } while (error && tries++ < 3); - /* Find an rgrp suitable for allocation. If it encounters any unlinked - dinodes along the way, error will equal -EAGAIN and unlinked will - contains it block address. We then need to look up that inode and - try to free it, and try the allocation again. */ - error = get_local_rgrp(ip, &unlinked, &last_unlinked); if (error) { if (hold_rindex && ip != GFS2_I(sdp->sd_rindex)) gfs2_glock_dq_uninit(&al->al_ri_gh); - if (error != -EAGAIN) - return error; - - gfs2_process_unlinked_inode(ip->i_inode.i_sb, unlinked); - /* regardless of whether or not gfs2_process_unlinked_inode - was successful, we don't want to repeat it again. */ - last_unlinked = unlinked; - gfs2_log_flush(sdp, NULL); - error = 0; - - goto try_again; + return error; } + /* no error, so we have the rgrp set in the inode's allocation. */ al->al_file = file; al->al_line = line; diff --git a/fs/gfs2/rgrp.h b/fs/gfs2/rgrp.h index 0e35c0466f9..50c2bb04369 100644 --- a/fs/gfs2/rgrp.h +++ b/fs/gfs2/rgrp.h @@ -48,6 +48,7 @@ extern int gfs2_inplace_reserve_i(struct gfs2_inode *ip, int hold_rindex, extern void gfs2_inplace_release(struct gfs2_inode *ip); +extern int gfs2_ri_update(struct gfs2_inode *ip); extern int gfs2_alloc_block(struct gfs2_inode *ip, u64 *bn, unsigned int *n); extern int gfs2_alloc_di(struct gfs2_inode *ip, u64 *bn, u64 *generation); diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c index 2b2c4997430..16c2ecac7eb 100644 --- a/fs/gfs2/super.c +++ b/fs/gfs2/super.c @@ -1405,11 +1405,18 @@ static struct inode *gfs2_alloc_inode(struct super_block *sb) return &ip->i_inode; } -static void gfs2_destroy_inode(struct inode *inode) +static void gfs2_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(gfs2_inode_cachep, inode); } +static void gfs2_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, gfs2_i_callback); +} + const struct super_operations gfs2_super_ops = { .alloc_inode = gfs2_alloc_inode, .destroy_inode = gfs2_destroy_inode, diff --git a/fs/gfs2/xattr.c b/fs/gfs2/xattr.c index 30b58f07c8a..439b61c0326 100644 --- a/fs/gfs2/xattr.c +++ b/fs/gfs2/xattr.c @@ -1296,10 +1296,8 @@ fail: int gfs2_xattr_acl_chmod(struct gfs2_inode *ip, struct iattr *attr, char *data) { - struct inode *inode = &ip->i_inode; struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); struct gfs2_ea_location el; - struct buffer_head *dibh; int error; error = gfs2_ea_find(ip, GFS2_EATYPE_SYS, GFS2_POSIX_ACL_ACCESS, &el); @@ -1321,26 +1319,7 @@ int gfs2_xattr_acl_chmod(struct gfs2_inode *ip, struct iattr *attr, char *data) if (error) return error; - error = gfs2_meta_inode_buffer(ip, &dibh); - if (error) - goto out_trans_end; - - if ((attr->ia_valid & ATTR_SIZE) && - attr->ia_size != i_size_read(inode)) { - int error; - - error = vmtruncate(inode, attr->ia_size); - gfs2_assert_warn(GFS2_SB(inode), !error); - } - - setattr_copy(inode, attr); - mark_inode_dirty(inode); - - gfs2_trans_add_bh(ip->i_gl, dibh, 1); - gfs2_dinode_out(ip, dibh->b_data); - brelse(dibh); - -out_trans_end: + error = gfs2_setattr_simple(ip, attr); gfs2_trans_end(sdp); return error; } diff --git a/fs/hfs/dir.c b/fs/hfs/dir.c index 2b3b8611b41..ea4aefe7c65 100644 --- a/fs/hfs/dir.c +++ b/fs/hfs/dir.c @@ -25,7 +25,7 @@ static struct dentry *hfs_lookup(struct inode *dir, struct dentry *dentry, struct inode *inode = NULL; int res; - dentry->d_op = &hfs_dentry_operations; + d_set_d_op(dentry, &hfs_dentry_operations); hfs_find_init(HFS_SB(dir->i_sb)->cat_tree, &fd); hfs_cat_build_key(dir->i_sb, fd.search_key, dir->i_ino, &dentry->d_name); diff --git a/fs/hfs/hfs_fs.h b/fs/hfs/hfs_fs.h index c8cffb81e84..ad97c2d5828 100644 --- a/fs/hfs/hfs_fs.h +++ b/fs/hfs/hfs_fs.h @@ -213,10 +213,14 @@ 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(struct dentry *, struct qstr *); +extern int hfs_hash_dentry(const struct dentry *, const struct inode *, + struct qstr *); extern int hfs_strcmp(const unsigned char *, unsigned int, const unsigned char *, unsigned int); -extern int hfs_compare_dentry(struct dentry *, struct qstr *, struct qstr *); +extern int hfs_compare_dentry(const struct dentry *parent, + const struct inode *pinode, + const struct dentry *dentry, const struct inode *inode, + unsigned int len, const char *str, const struct qstr *name); /* trans.c */ extern void hfs_asc2mac(struct super_block *, struct hfs_name *, struct qstr *); diff --git a/fs/hfs/string.c b/fs/hfs/string.c index 927a5af7942..495a976a3cc 100644 --- a/fs/hfs/string.c +++ b/fs/hfs/string.c @@ -51,7 +51,8 @@ static unsigned char caseorder[256] = { /* * Hash a string to an integer in a case-independent way */ -int hfs_hash_dentry(struct dentry *dentry, struct qstr *this) +int hfs_hash_dentry(const struct dentry *dentry, const struct inode *inode, + struct qstr *this) { const unsigned char *name = this->name; unsigned int hash, len = this->len; @@ -92,21 +93,21 @@ 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(struct dentry *dentry, struct qstr *s1, struct qstr *s2) +int hfs_compare_dentry(const struct dentry *parent, const struct inode *pinode, + const struct dentry *dentry, const struct inode *inode, + unsigned int len, const char *str, const struct qstr *name) { const unsigned char *n1, *n2; - int len; - len = s1->len; if (len >= HFS_NAMELEN) { - if (s2->len < HFS_NAMELEN) + if (name->len < HFS_NAMELEN) return 1; len = HFS_NAMELEN; - } else if (len != s2->len) + } else if (len != name->len) return 1; - n1 = s1->name; - n2 = s2->name; + n1 = str; + n2 = name->name; while (len--) { if (caseorder[*n1++] != caseorder[*n2++]) return 1; diff --git a/fs/hfs/super.c b/fs/hfs/super.c index 4824c27cebb..0bef62aa4f4 100644 --- a/fs/hfs/super.c +++ b/fs/hfs/super.c @@ -167,11 +167,18 @@ static struct inode *hfs_alloc_inode(struct super_block *sb) return i ? &i->vfs_inode : NULL; } -static void hfs_destroy_inode(struct inode *inode) +static void hfs_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(hfs_inode_cachep, HFS_I(inode)); } +static void hfs_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, hfs_i_callback); +} + static const struct super_operations hfs_super_operations = { .alloc_inode = hfs_alloc_inode, .destroy_inode = hfs_destroy_inode, @@ -427,7 +434,7 @@ static int hfs_fill_super(struct super_block *sb, void *data, int silent) if (!sb->s_root) goto bail_iput; - sb->s_root->d_op = &hfs_dentry_operations; + d_set_d_op(sb->s_root, &hfs_dentry_operations); /* everything's okay */ return 0; diff --git a/fs/hfs/sysdep.c b/fs/hfs/sysdep.c index 7478f5c219a..19cf291eb91 100644 --- a/fs/hfs/sysdep.c +++ b/fs/hfs/sysdep.c @@ -8,15 +8,20 @@ * This file contains the code to do various system dependent things. */ +#include <linux/namei.h> #include "hfs_fs.h" /* dentry case-handling: just lowercase everything */ static int hfs_revalidate_dentry(struct dentry *dentry, struct nameidata *nd) { - struct inode *inode = dentry->d_inode; + struct inode *inode; int diff; + if (nd->flags & LOOKUP_RCU) + return -ECHILD; + + inode = dentry->d_inode; if(!inode) return 1; diff --git a/fs/hfsplus/dir.c b/fs/hfsplus/dir.c index 9d59c0571f5..ccab87145f7 100644 --- a/fs/hfsplus/dir.c +++ b/fs/hfsplus/dir.c @@ -37,7 +37,7 @@ static struct dentry *hfsplus_lookup(struct inode *dir, struct dentry *dentry, sb = dir->i_sb; - dentry->d_op = &hfsplus_dentry_operations; + d_set_d_op(dentry, &hfsplus_dentry_operations); dentry->d_fsdata = NULL; hfs_find_init(HFSPLUS_SB(sb)->cat_tree, &fd); hfsplus_cat_build_key(sb, fd.search_key, dir->i_ino, &dentry->d_name); diff --git a/fs/hfsplus/hfsplus_fs.h b/fs/hfsplus/hfsplus_fs.h index cb3653efb57..a5308f491e3 100644 --- a/fs/hfsplus/hfsplus_fs.h +++ b/fs/hfsplus/hfsplus_fs.h @@ -379,8 +379,12 @@ int hfsplus_strcasecmp(const struct hfsplus_unistr *, const struct hfsplus_unist int hfsplus_strcmp(const struct hfsplus_unistr *, const struct hfsplus_unistr *); int hfsplus_uni2asc(struct super_block *, const struct hfsplus_unistr *, char *, int *); int hfsplus_asc2uni(struct super_block *, struct hfsplus_unistr *, const char *, int); -int hfsplus_hash_dentry(struct dentry *dentry, struct qstr *str); -int hfsplus_compare_dentry(struct dentry *dentry, struct qstr *s1, struct qstr *s2); +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, + unsigned int len, const char *str, const struct qstr *name); /* wrapper.c */ int hfsplus_read_wrapper(struct super_block *); diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c index 52cc746d3ba..ddf712e4700 100644 --- a/fs/hfsplus/super.c +++ b/fs/hfsplus/super.c @@ -419,7 +419,7 @@ static int hfsplus_fill_super(struct super_block *sb, void *data, int silent) err = -ENOMEM; goto cleanup; } - sb->s_root->d_op = &hfsplus_dentry_operations; + d_set_d_op(sb->s_root, &hfsplus_dentry_operations); str.len = sizeof(HFSP_HIDDENDIR_NAME) - 1; str.name = HFSP_HIDDENDIR_NAME; @@ -488,11 +488,19 @@ static struct inode *hfsplus_alloc_inode(struct super_block *sb) return i ? &i->vfs_inode : NULL; } -static void hfsplus_destroy_inode(struct inode *inode) +static void hfsplus_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); + + INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(hfsplus_inode_cachep, HFSPLUS_I(inode)); } +static void hfsplus_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, hfsplus_i_callback); +} + #define HFSPLUS_INODE_SIZE sizeof(struct hfsplus_inode_info) static struct dentry *hfsplus_mount(struct file_system_type *fs_type, diff --git a/fs/hfsplus/unicode.c b/fs/hfsplus/unicode.c index b66d67de882..d800aa0f2c8 100644 --- a/fs/hfsplus/unicode.c +++ b/fs/hfsplus/unicode.c @@ -320,7 +320,8 @@ int hfsplus_asc2uni(struct super_block *sb, struct hfsplus_unistr *ustr, * Composed unicode characters are decomposed and case-folding is performed * if the appropriate bits are (un)set on the superblock. */ -int hfsplus_hash_dentry(struct dentry *dentry, struct qstr *str) +int hfsplus_hash_dentry(const struct dentry *dentry, const struct inode *inode, + struct qstr *str) { struct super_block *sb = dentry->d_sb; const char *astr; @@ -363,9 +364,12 @@ int hfsplus_hash_dentry(struct dentry *dentry, struct qstr *str) * Composed unicode characters are decomposed and case-folding is performed * if the appropriate bits are (un)set on the superblock. */ -int hfsplus_compare_dentry(struct dentry *dentry, struct qstr *s1, struct qstr *s2) +int hfsplus_compare_dentry(const struct dentry *parent, + const struct inode *pinode, + const struct dentry *dentry, const struct inode *inode, + unsigned int len, const char *str, const struct qstr *name) { - struct super_block *sb = dentry->d_sb; + struct super_block *sb = parent->d_sb; int casefold, decompose, size; int dsize1, dsize2, len1, len2; const u16 *dstr1, *dstr2; @@ -375,10 +379,10 @@ int hfsplus_compare_dentry(struct dentry *dentry, struct qstr *s1, struct qstr * casefold = test_bit(HFSPLUS_SB_CASEFOLD, &HFSPLUS_SB(sb)->flags); decompose = !test_bit(HFSPLUS_SB_NODECOMPOSE, &HFSPLUS_SB(sb)->flags); - astr1 = s1->name; - len1 = s1->len; - astr2 = s2->name; - len2 = s2->len; + astr1 = str; + len1 = len; + astr2 = name->name; + len2 = name->len; dsize1 = dsize2 = 0; dstr1 = dstr2 = NULL; diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c index 2c0f148a49e..d3244d949a4 100644 --- a/fs/hostfs/hostfs_kern.c +++ b/fs/hostfs/hostfs_kern.c @@ -32,7 +32,7 @@ static inline struct hostfs_inode_info *HOSTFS_I(struct inode *inode) #define FILE_HOSTFS_I(file) HOSTFS_I((file)->f_path.dentry->d_inode) -static int hostfs_d_delete(struct dentry *dentry) +static int hostfs_d_delete(const struct dentry *dentry) { return 1; } @@ -92,12 +92,10 @@ __uml_setup("hostfs=", hostfs_args, static char *__dentry_name(struct dentry *dentry, char *name) { - char *p = __dentry_path(dentry, name, PATH_MAX); + char *p = dentry_path_raw(dentry, name, PATH_MAX); char *root; size_t len; - spin_unlock(&dcache_lock); - root = dentry->d_sb->s_fs_info; len = strlen(root); if (IS_ERR(p)) { @@ -123,25 +121,23 @@ static char *dentry_name(struct dentry *dentry) if (!name) return NULL; - spin_lock(&dcache_lock); return __dentry_name(dentry, name); /* will unlock */ } static char *inode_name(struct inode *ino) { struct dentry *dentry; - char *name = __getname(); - if (!name) - return NULL; + char *name; - spin_lock(&dcache_lock); - if (list_empty(&ino->i_dentry)) { - spin_unlock(&dcache_lock); - __putname(name); + dentry = d_find_alias(ino); + if (!dentry) return NULL; - } - dentry = list_first_entry(&ino->i_dentry, struct dentry, d_alias); - return __dentry_name(dentry, name); /* will unlock */ + + name = dentry_name(dentry); + + dput(dentry); + + return name; } static char *follow_link(char *link) @@ -251,11 +247,18 @@ static void hostfs_evict_inode(struct inode *inode) } } -static void hostfs_destroy_inode(struct inode *inode) +static void hostfs_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); kfree(HOSTFS_I(inode)); } +static void hostfs_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, hostfs_i_callback); +} + static int hostfs_show_options(struct seq_file *seq, struct vfsmount *vfs) { const char *root_path = vfs->mnt_sb->s_fs_info; @@ -609,7 +612,7 @@ struct dentry *hostfs_lookup(struct inode *ino, struct dentry *dentry, goto out_put; d_add(dentry, inode); - dentry->d_op = &hostfs_dentry_ops; + d_set_d_op(dentry, &hostfs_dentry_ops); return NULL; out_put: @@ -746,11 +749,14 @@ int hostfs_rename(struct inode *from_ino, struct dentry *from, return err; } -int hostfs_permission(struct inode *ino, int desired) +int hostfs_permission(struct inode *ino, int desired, unsigned int flags) { char *name; int r = 0, w = 0, x = 0, err; + if (flags & IPERM_FLAG_RCU) + return -ECHILD; + if (desired & MAY_READ) r = 1; if (desired & MAY_WRITE) w = 1; if (desired & MAY_EXEC) x = 1; @@ -765,7 +771,7 @@ int hostfs_permission(struct inode *ino, int desired) err = access_file(name, r, w, x); __putname(name); if (!err) - err = generic_permission(ino, desired, NULL); + err = generic_permission(ino, desired, flags, NULL); return err; } diff --git a/fs/hpfs/dentry.c b/fs/hpfs/dentry.c index 67d9d36b3d5..32c13a94e1e 100644 --- a/fs/hpfs/dentry.c +++ b/fs/hpfs/dentry.c @@ -12,7 +12,8 @@ * Note: the dentry argument is the parent dentry. */ -static int hpfs_hash_dentry(struct dentry *dentry, struct qstr *qstr) +static int hpfs_hash_dentry(const struct dentry *dentry, const struct inode *inode, + struct qstr *qstr) { unsigned long hash; int i; @@ -34,19 +35,25 @@ static int hpfs_hash_dentry(struct dentry *dentry, struct qstr *qstr) return 0; } -static int hpfs_compare_dentry(struct dentry *dentry, struct qstr *a, struct qstr *b) +static int hpfs_compare_dentry(const struct dentry *parent, + const struct inode *pinode, + const struct dentry *dentry, const struct inode *inode, + unsigned int len, const char *str, const struct qstr *name) { - unsigned al=a->len; - unsigned bl=b->len; - hpfs_adjust_length(a->name, &al); + unsigned al = len; + unsigned bl = name->len; + + hpfs_adjust_length(str, &al); /*hpfs_adjust_length(b->name, &bl);*/ - /* 'a' is the qstr of an already existing dentry, so the name - * must be valid. 'b' must be validated first. + + /* + * 'str' is the nane of an already existing dentry, so the name + * must be valid. 'name' must be validated first. */ - if (hpfs_chk_name(b->name, &bl)) + if (hpfs_chk_name(name->name, &bl)) return 1; - if (hpfs_compare_names(dentry->d_sb, a->name, al, b->name, bl, 0)) + if (hpfs_compare_names(parent->d_sb, str, al, name->name, bl, 0)) return 1; return 0; } @@ -58,5 +65,5 @@ static const struct dentry_operations hpfs_dentry_operations = { void hpfs_set_dentry_operations(struct dentry *dentry) { - dentry->d_op = &hpfs_dentry_operations; + d_set_d_op(dentry, &hpfs_dentry_operations); } diff --git a/fs/hpfs/namei.c b/fs/hpfs/namei.c index 11c2b4080f6..f4ad9e31ddc 100644 --- a/fs/hpfs/namei.c +++ b/fs/hpfs/namei.c @@ -419,7 +419,7 @@ again: unlock_kernel(); return -ENOSPC; } - if (generic_permission(inode, MAY_WRITE, NULL) || + if (generic_permission(inode, MAY_WRITE, 0, NULL) || !S_ISREG(inode->i_mode) || get_write_access(inode)) { d_rehash(dentry); diff --git a/fs/hpfs/super.c b/fs/hpfs/super.c index 6c5f01597c3..49935ba78db 100644 --- a/fs/hpfs/super.c +++ b/fs/hpfs/super.c @@ -177,11 +177,18 @@ static struct inode *hpfs_alloc_inode(struct super_block *sb) return &ei->vfs_inode; } -static void hpfs_destroy_inode(struct inode *inode) +static void hpfs_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(hpfs_inode_cachep, hpfs_i(inode)); } +static void hpfs_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, hpfs_i_callback); +} + static void init_once(void *foo) { struct hpfs_inode_info *ei = (struct hpfs_inode_info *) foo; diff --git a/fs/hppfs/hppfs.c b/fs/hppfs/hppfs.c index f702b5f713f..87ed48e0343 100644 --- a/fs/hppfs/hppfs.c +++ b/fs/hppfs/hppfs.c @@ -632,11 +632,18 @@ void hppfs_evict_inode(struct inode *ino) mntput(ino->i_sb->s_fs_info); } -static void hppfs_destroy_inode(struct inode *inode) +static void hppfs_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); kfree(HPPFS_I(inode)); } +static void hppfs_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, hppfs_i_callback); +} + static const struct super_operations hppfs_sbops = { .alloc_inode = hppfs_alloc_inode, .destroy_inode = hppfs_destroy_inode, diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index d6cfac1f0a4..9885082b470 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -663,11 +663,18 @@ static struct inode *hugetlbfs_alloc_inode(struct super_block *sb) return &p->vfs_inode; } +static void hugetlbfs_i_callback(struct rcu_head *head) +{ + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); + kmem_cache_free(hugetlbfs_inode_cachep, HUGETLBFS_I(inode)); +} + static void hugetlbfs_destroy_inode(struct inode *inode) { hugetlbfs_inc_free_inodes(HUGETLBFS_SB(inode->i_sb)); mpol_free_shared_policy(&HUGETLBFS_I(inode)->policy); - kmem_cache_free(hugetlbfs_inode_cachep, HUGETLBFS_I(inode)); + call_rcu(&inode->i_rcu, hugetlbfs_i_callback); } static const struct address_space_operations hugetlbfs_aops = { @@ -932,8 +939,7 @@ struct file *hugetlb_file_setup(const char *name, size_t size, int acctflag, if (creat_flags == HUGETLB_SHMFS_INODE && !can_do_hugetlb_shm()) { *user = current_user(); if (user_shm_lock(size, *user)) { - WARN_ONCE(1, - "Using mlock ulimits for SHM_HUGETLB deprecated\n"); + printk_once(KERN_WARNING "Using mlock ulimits for SHM_HUGETLB is deprecated\n"); } else { *user = NULL; return ERR_PTR(-EPERM); diff --git a/fs/inode.c b/fs/inode.c index ae2727ab0c3..da85e56378f 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -102,26 +102,29 @@ static DECLARE_RWSEM(iprune_sem); */ struct inodes_stat_t inodes_stat; -static struct percpu_counter nr_inodes __cacheline_aligned_in_smp; -static struct percpu_counter nr_inodes_unused __cacheline_aligned_in_smp; +static DEFINE_PER_CPU(unsigned int, nr_inodes); static struct kmem_cache *inode_cachep __read_mostly; -static inline int get_nr_inodes(void) +static int get_nr_inodes(void) { - return percpu_counter_sum_positive(&nr_inodes); + int i; + int sum = 0; + for_each_possible_cpu(i) + sum += per_cpu(nr_inodes, i); + return sum < 0 ? 0 : sum; } static inline int get_nr_inodes_unused(void) { - return percpu_counter_sum_positive(&nr_inodes_unused); + return inodes_stat.nr_unused; } int get_nr_dirty_inodes(void) { + /* not actually dirty inodes, but a wild approximation */ int nr_dirty = get_nr_inodes() - get_nr_inodes_unused(); return nr_dirty > 0 ? nr_dirty : 0; - } /* @@ -132,7 +135,6 @@ int proc_nr_inodes(ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { inodes_stat.nr_inodes = get_nr_inodes(); - inodes_stat.nr_unused = get_nr_inodes_unused(); return proc_dointvec(table, write, buffer, lenp, ppos); } #endif @@ -224,7 +226,7 @@ int inode_init_always(struct super_block *sb, struct inode *inode) inode->i_fsnotify_mask = 0; #endif - percpu_counter_inc(&nr_inodes); + this_cpu_inc(nr_inodes); return 0; out: @@ -255,6 +257,12 @@ static struct inode *alloc_inode(struct super_block *sb) return inode; } +void free_inode_nonrcu(struct inode *inode) +{ + kmem_cache_free(inode_cachep, inode); +} +EXPORT_SYMBOL(free_inode_nonrcu); + void __destroy_inode(struct inode *inode) { BUG_ON(inode_has_buffers(inode)); @@ -266,10 +274,17 @@ void __destroy_inode(struct inode *inode) if (inode->i_default_acl && inode->i_default_acl != ACL_NOT_CACHED) posix_acl_release(inode->i_default_acl); #endif - percpu_counter_dec(&nr_inodes); + this_cpu_dec(nr_inodes); } EXPORT_SYMBOL(__destroy_inode); +static void i_callback(struct rcu_head *head) +{ + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); + kmem_cache_free(inode_cachep, inode); +} + static void destroy_inode(struct inode *inode) { BUG_ON(!list_empty(&inode->i_lru)); @@ -277,7 +292,7 @@ static void destroy_inode(struct inode *inode) if (inode->i_sb->s_op->destroy_inode) inode->i_sb->s_op->destroy_inode(inode); else - kmem_cache_free(inode_cachep, (inode)); + call_rcu(&inode->i_rcu, i_callback); } /* @@ -335,7 +350,7 @@ static void inode_lru_list_add(struct inode *inode) { if (list_empty(&inode->i_lru)) { list_add(&inode->i_lru, &inode_lru); - percpu_counter_inc(&nr_inodes_unused); + inodes_stat.nr_unused++; } } @@ -343,7 +358,7 @@ static void inode_lru_list_del(struct inode *inode) { if (!list_empty(&inode->i_lru)) { list_del_init(&inode->i_lru); - percpu_counter_dec(&nr_inodes_unused); + inodes_stat.nr_unused--; } } @@ -430,6 +445,7 @@ void end_writeback(struct inode *inode) BUG_ON(!(inode->i_state & I_FREEING)); BUG_ON(inode->i_state & I_CLEAR); inode_sync_wait(inode); + /* don't need i_lock here, no concurrent mods to i_state */ inode->i_state = I_FREEING | I_CLEAR; } EXPORT_SYMBOL(end_writeback); @@ -513,7 +529,7 @@ void evict_inodes(struct super_block *sb) list_move(&inode->i_lru, &dispose); list_del_init(&inode->i_wb_list); if (!(inode->i_state & (I_DIRTY | I_SYNC))) - percpu_counter_dec(&nr_inodes_unused); + inodes_stat.nr_unused--; } spin_unlock(&inode_lock); @@ -554,7 +570,7 @@ int invalidate_inodes(struct super_block *sb) list_move(&inode->i_lru, &dispose); list_del_init(&inode->i_wb_list); if (!(inode->i_state & (I_DIRTY | I_SYNC))) - percpu_counter_dec(&nr_inodes_unused); + inodes_stat.nr_unused--; } spin_unlock(&inode_lock); @@ -616,7 +632,7 @@ static void prune_icache(int nr_to_scan) if (atomic_read(&inode->i_count) || (inode->i_state & ~I_REFERENCED)) { list_del_init(&inode->i_lru); - percpu_counter_dec(&nr_inodes_unused); + inodes_stat.nr_unused--; continue; } @@ -650,7 +666,7 @@ static void prune_icache(int nr_to_scan) */ list_move(&inode->i_lru, &freeable); list_del_init(&inode->i_wb_list); - percpu_counter_dec(&nr_inodes_unused); + inodes_stat.nr_unused--; } if (current_is_kswapd()) __count_vm_events(KSWAPD_INODESTEAL, reap); @@ -1648,8 +1664,6 @@ void __init inode_init(void) SLAB_MEM_SPREAD), init_once); register_shrinker(&icache_shrinker); - percpu_counter_init(&nr_inodes, 0); - percpu_counter_init(&nr_inodes_unused, 0); /* Hash may have been set up in inode_init_early */ if (!hashdist) diff --git a/fs/internal.h b/fs/internal.h index e43b9a4dbf4..9687c2ee273 100644 --- a/fs/internal.h +++ b/fs/internal.h @@ -63,6 +63,7 @@ extern int copy_mount_string(const void __user *, char **); extern void free_vfsmnt(struct vfsmount *); extern struct vfsmount *alloc_vfsmnt(const char *); +extern unsigned int mnt_get_count(struct vfsmount *mnt); extern struct vfsmount *__lookup_mnt(struct vfsmount *, struct dentry *, int); extern void mnt_set_mountpoint(struct vfsmount *, struct dentry *, struct vfsmount *); diff --git a/fs/ioctl.c b/fs/ioctl.c index e92fdbb3bc3..d6cc1647662 100644 --- a/fs/ioctl.c +++ b/fs/ioctl.c @@ -6,7 +6,6 @@ #include <linux/syscalls.h> #include <linux/mm.h> -#include <linux/smp_lock.h> #include <linux/capability.h> #include <linux/file.h> #include <linux/fs.h> @@ -530,41 +529,6 @@ static int ioctl_fsthaw(struct file *filp) return thaw_super(sb); } -static int ioctl_fstrim(struct file *filp, void __user *argp) -{ - struct super_block *sb = filp->f_path.dentry->d_inode->i_sb; - struct fstrim_range range; - int ret = 0; - - if (!capable(CAP_SYS_ADMIN)) - return -EPERM; - - /* If filesystem doesn't support trim feature, return. */ - if (sb->s_op->trim_fs == NULL) - return -EOPNOTSUPP; - - /* If a blockdevice-backed filesystem isn't specified, return EINVAL. */ - if (sb->s_bdev == NULL) - return -EINVAL; - - if (argp == NULL) { - range.start = 0; - range.len = ULLONG_MAX; - range.minlen = 0; - } else if (copy_from_user(&range, argp, sizeof(range))) - return -EFAULT; - - ret = sb->s_op->trim_fs(sb, &range); - if (ret < 0) - return ret; - - if ((argp != NULL) && - (copy_to_user(argp, &range, sizeof(range)))) - return -EFAULT; - - return 0; -} - /* * When you add any new common ioctls to the switches above and below * please update compat_sys_ioctl() too. @@ -615,10 +579,6 @@ int do_vfs_ioctl(struct file *filp, unsigned int fd, unsigned int cmd, error = ioctl_fsthaw(filp); break; - case FITRIM: - error = ioctl_fstrim(filp, argp); - break; - case FS_IOC_FIEMAP: return ioctl_fiemap(filp, arg); diff --git a/fs/ioprio.c b/fs/ioprio.c index 748cfb92dcc..7da2a06508e 100644 --- a/fs/ioprio.c +++ b/fs/ioprio.c @@ -103,12 +103,7 @@ SYSCALL_DEFINE3(ioprio_set, int, which, int, who, int, ioprio) } ret = -ESRCH; - /* - * We want IOPRIO_WHO_PGRP/IOPRIO_WHO_USER to be "atomic", - * so we can't use rcu_read_lock(). See re-copy of ->ioprio - * in copy_process(). - */ - read_lock(&tasklist_lock); + rcu_read_lock(); switch (which) { case IOPRIO_WHO_PROCESS: if (!who) @@ -153,7 +148,7 @@ free_uid: ret = -EINVAL; } - read_unlock(&tasklist_lock); + rcu_read_unlock(); return ret; } @@ -197,7 +192,7 @@ SYSCALL_DEFINE2(ioprio_get, int, which, int, who) int ret = -ESRCH; int tmpio; - read_lock(&tasklist_lock); + rcu_read_lock(); switch (which) { case IOPRIO_WHO_PROCESS: if (!who) @@ -250,6 +245,6 @@ SYSCALL_DEFINE2(ioprio_get, int, which, int, who) ret = -EINVAL; } - read_unlock(&tasklist_lock); + rcu_read_unlock(); return ret; } diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c index bfdeb82a53b..844a7903c72 100644 --- a/fs/isofs/inode.c +++ b/fs/isofs/inode.c @@ -26,16 +26,32 @@ #define BEQUIET -static int isofs_hashi(struct dentry *parent, struct qstr *qstr); -static int isofs_hash(struct dentry *parent, struct qstr *qstr); -static int isofs_dentry_cmpi(struct dentry *dentry, struct qstr *a, struct qstr *b); -static int isofs_dentry_cmp(struct dentry *dentry, struct qstr *a, struct qstr *b); +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_dentry_cmpi(const struct dentry *parent, + const struct inode *pinode, + const struct dentry *dentry, const struct inode *inode, + 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, + unsigned int len, const char *str, const struct qstr *name); #ifdef CONFIG_JOLIET -static int isofs_hashi_ms(struct dentry *parent, struct qstr *qstr); -static int isofs_hash_ms(struct dentry *parent, struct qstr *qstr); -static int isofs_dentry_cmpi_ms(struct dentry *dentry, struct qstr *a, struct qstr *b); -static int isofs_dentry_cmp_ms(struct dentry *dentry, struct qstr *a, struct qstr *b); +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_dentry_cmpi_ms(const struct dentry *parent, + const struct inode *pinode, + const struct dentry *dentry, const struct inode *inode, + 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, + unsigned int len, const char *str, const struct qstr *name); #endif static void isofs_put_super(struct super_block *sb) @@ -65,11 +81,18 @@ static struct inode *isofs_alloc_inode(struct super_block *sb) return &ei->vfs_inode; } -static void isofs_destroy_inode(struct inode *inode) +static void isofs_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(isofs_inode_cachep, ISOFS_I(inode)); } +static void isofs_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, isofs_i_callback); +} + static void init_once(void *foo) { struct iso_inode_info *ei = foo; @@ -160,7 +183,7 @@ struct iso9660_options{ * Compute the hash for the isofs name corresponding to the dentry. */ static int -isofs_hash_common(struct dentry *dentry, struct qstr *qstr, int ms) +isofs_hash_common(const struct dentry *dentry, struct qstr *qstr, int ms) { const char *name; int len; @@ -181,7 +204,7 @@ isofs_hash_common(struct dentry *dentry, struct qstr *qstr, int ms) * Compute the hash for the isofs name corresponding to the dentry. */ static int -isofs_hashi_common(struct dentry *dentry, struct qstr *qstr, int ms) +isofs_hashi_common(const struct dentry *dentry, struct qstr *qstr, int ms) { const char *name; int len; @@ -206,100 +229,94 @@ isofs_hashi_common(struct dentry *dentry, struct qstr *qstr, int ms) } /* - * Case insensitive compare of two isofs names. - */ -static int isofs_dentry_cmpi_common(struct dentry *dentry, struct qstr *a, - struct qstr *b, int ms) -{ - int alen, blen; - - /* A filename cannot end in '.' or we treat it like it has none */ - alen = a->len; - blen = b->len; - if (ms) { - while (alen && a->name[alen-1] == '.') - alen--; - while (blen && b->name[blen-1] == '.') - blen--; - } - if (alen == blen) { - if (strnicmp(a->name, b->name, alen) == 0) - return 0; - } - return 1; -} - -/* - * Case sensitive compare of two isofs names. + * Compare of two isofs names. */ -static int isofs_dentry_cmp_common(struct dentry *dentry, struct qstr *a, - struct qstr *b, int ms) +static int isofs_dentry_cmp_common( + unsigned int len, const char *str, + const struct qstr *name, int ms, int ci) { int alen, blen; /* A filename cannot end in '.' or we treat it like it has none */ - alen = a->len; - blen = b->len; + alen = name->len; + blen = len; if (ms) { - while (alen && a->name[alen-1] == '.') + while (alen && name->name[alen-1] == '.') alen--; - while (blen && b->name[blen-1] == '.') + while (blen && str[blen-1] == '.') blen--; } if (alen == blen) { - if (strncmp(a->name, b->name, alen) == 0) - return 0; + if (ci) { + if (strnicmp(name->name, str, alen) == 0) + return 0; + } else { + if (strncmp(name->name, str, alen) == 0) + return 0; + } } return 1; } static int -isofs_hash(struct dentry *dentry, struct qstr *qstr) +isofs_hash(const struct dentry *dentry, const struct inode *inode, + struct qstr *qstr) { return isofs_hash_common(dentry, qstr, 0); } static int -isofs_hashi(struct dentry *dentry, struct qstr *qstr) +isofs_hashi(const struct dentry *dentry, const struct inode *inode, + struct qstr *qstr) { return isofs_hashi_common(dentry, qstr, 0); } static int -isofs_dentry_cmp(struct dentry *dentry,struct qstr *a,struct qstr *b) +isofs_dentry_cmp(const struct dentry *parent, const struct inode *pinode, + const struct dentry *dentry, const struct inode *inode, + unsigned int len, const char *str, const struct qstr *name) { - return isofs_dentry_cmp_common(dentry, a, b, 0); + return isofs_dentry_cmp_common(len, str, name, 0, 0); } static int -isofs_dentry_cmpi(struct dentry *dentry,struct qstr *a,struct qstr *b) +isofs_dentry_cmpi(const struct dentry *parent, const struct inode *pinode, + const struct dentry *dentry, const struct inode *inode, + unsigned int len, const char *str, const struct qstr *name) { - return isofs_dentry_cmpi_common(dentry, a, b, 0); + return isofs_dentry_cmp_common(len, str, name, 0, 1); } #ifdef CONFIG_JOLIET static int -isofs_hash_ms(struct dentry *dentry, struct qstr *qstr) +isofs_hash_ms(const struct dentry *dentry, const struct inode *inode, + struct qstr *qstr) { return isofs_hash_common(dentry, qstr, 1); } static int -isofs_hashi_ms(struct dentry *dentry, struct qstr *qstr) +isofs_hashi_ms(const struct dentry *dentry, const struct inode *inode, + struct qstr *qstr) { return isofs_hashi_common(dentry, qstr, 1); } static int -isofs_dentry_cmp_ms(struct dentry *dentry,struct qstr *a,struct qstr *b) +isofs_dentry_cmp_ms(const struct dentry *parent, const struct inode *pinode, + const struct dentry *dentry, const struct inode *inode, + unsigned int len, const char *str, const struct qstr *name) { - return isofs_dentry_cmp_common(dentry, a, b, 1); + return isofs_dentry_cmp_common(len, str, name, 1, 0); } static int -isofs_dentry_cmpi_ms(struct dentry *dentry,struct qstr *a,struct qstr *b) +isofs_dentry_cmpi_ms(const struct dentry *parent, const struct inode *pinode, + const struct dentry *dentry, const struct inode *inode, + unsigned int len, const char *str, const struct qstr *name) { - return isofs_dentry_cmpi_common(dentry, a, b, 1); + return isofs_dentry_cmp_common(len, str, name, 1, 1); } #endif @@ -932,7 +949,7 @@ root_found: table += 2; if (opt.check == 'r') table++; - s->s_root->d_op = &isofs_dentry_ops[table]; + d_set_d_op(s->s_root, &isofs_dentry_ops[table]); kfree(opt.iocharset); diff --git a/fs/isofs/namei.c b/fs/isofs/namei.c index 0d23abfd428..679a849c3b2 100644 --- a/fs/isofs/namei.c +++ b/fs/isofs/namei.c @@ -37,7 +37,8 @@ isofs_cmp(struct dentry *dentry, const char *compare, int dlen) qstr.name = compare; qstr.len = dlen; - return dentry->d_op->d_compare(dentry, &dentry->d_name, &qstr); + return dentry->d_op->d_compare(NULL, NULL, NULL, NULL, + dentry->d_name.len, dentry->d_name.name, &qstr); } /* @@ -171,7 +172,7 @@ struct dentry *isofs_lookup(struct inode *dir, struct dentry *dentry, struct nam struct inode *inode; struct page *page; - dentry->d_op = dir->i_sb->s_root->d_op; + d_set_d_op(dentry, dir->i_sb->s_root->d_op); page = alloc_page(GFP_USER); if (!page) diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index c590d155c09..f837ba95352 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -899,6 +899,14 @@ journal_t * jbd2_journal_init_dev(struct block_device *bdev, /* journal descriptor can store up to n blocks -bzzz */ journal->j_blocksize = blocksize; + journal->j_dev = bdev; + journal->j_fs_dev = fs_dev; + journal->j_blk_offset = start; + journal->j_maxlen = len; + bdevname(journal->j_dev, journal->j_devname); + p = journal->j_devname; + while ((p = strchr(p, '/'))) + *p = '!'; jbd2_stats_proc_init(journal); n = journal->j_blocksize / sizeof(journal_block_tag_t); journal->j_wbufsize = n; @@ -908,14 +916,6 @@ journal_t * jbd2_journal_init_dev(struct block_device *bdev, __func__); goto out_err; } - journal->j_dev = bdev; - journal->j_fs_dev = fs_dev; - journal->j_blk_offset = start; - journal->j_maxlen = len; - bdevname(journal->j_dev, journal->j_devname); - p = journal->j_devname; - while ((p = strchr(p, '/'))) - *p = '!'; bh = __getblk(journal->j_dev, start, journal->j_blocksize); if (!bh) { diff --git a/fs/jffs2/acl.c b/fs/jffs2/acl.c index 54a92fd02bb..95b79672150 100644 --- a/fs/jffs2/acl.c +++ b/fs/jffs2/acl.c @@ -259,11 +259,14 @@ static int jffs2_set_acl(struct inode *inode, int type, struct posix_acl *acl) return rc; } -int jffs2_check_acl(struct inode *inode, int mask) +int jffs2_check_acl(struct inode *inode, int mask, unsigned int flags) { struct posix_acl *acl; int rc; + if (flags & IPERM_FLAG_RCU) + return -ECHILD; + acl = jffs2_get_acl(inode, ACL_TYPE_ACCESS); if (IS_ERR(acl)) return PTR_ERR(acl); diff --git a/fs/jffs2/acl.h b/fs/jffs2/acl.h index 5e42de8d954..3119f59253d 100644 --- a/fs/jffs2/acl.h +++ b/fs/jffs2/acl.h @@ -26,7 +26,7 @@ struct jffs2_acl_header { #ifdef CONFIG_JFFS2_FS_POSIX_ACL -extern int jffs2_check_acl(struct inode *, int); +extern int jffs2_check_acl(struct inode *, int, unsigned int); extern int jffs2_acl_chmod(struct inode *); extern int jffs2_init_acl_pre(struct inode *, struct inode *, int *); extern int jffs2_init_acl_post(struct inode *); diff --git a/fs/jffs2/super.c b/fs/jffs2/super.c index c86041b866a..853b8e30008 100644 --- a/fs/jffs2/super.c +++ b/fs/jffs2/super.c @@ -40,11 +40,18 @@ static struct inode *jffs2_alloc_inode(struct super_block *sb) return &f->vfs_inode; } -static void jffs2_destroy_inode(struct inode *inode) +static void jffs2_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(jffs2_inode_cachep, JFFS2_INODE_INFO(inode)); } +static void jffs2_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, jffs2_i_callback); +} + static void jffs2_i_init_once(void *foo) { struct jffs2_inode_info *f = foo; diff --git a/fs/jfs/acl.c b/fs/jfs/acl.c index 1057a4998e4..e5de9422fa3 100644 --- a/fs/jfs/acl.c +++ b/fs/jfs/acl.c @@ -114,10 +114,14 @@ out: return rc; } -int jfs_check_acl(struct inode *inode, int mask) +int jfs_check_acl(struct inode *inode, int mask, unsigned int flags) { - struct posix_acl *acl = jfs_get_acl(inode, ACL_TYPE_ACCESS); + struct posix_acl *acl; + + if (flags & IPERM_FLAG_RCU) + return -ECHILD; + acl = jfs_get_acl(inode, ACL_TYPE_ACCESS); if (IS_ERR(acl)) return PTR_ERR(acl); if (acl) { diff --git a/fs/jfs/jfs_acl.h b/fs/jfs/jfs_acl.h index 54e07559878..f9285c4900f 100644 --- a/fs/jfs/jfs_acl.h +++ b/fs/jfs/jfs_acl.h @@ -20,7 +20,7 @@ #ifdef CONFIG_JFS_POSIX_ACL -int jfs_check_acl(struct inode *, int); +int jfs_check_acl(struct inode *, int, unsigned int flags); int jfs_init_acl(tid_t, struct inode *, struct inode *); int jfs_acl_chmod(struct inode *inode); diff --git a/fs/jfs/namei.c b/fs/jfs/namei.c index 231ca4af9bc..4414e3a4226 100644 --- a/fs/jfs/namei.c +++ b/fs/jfs/namei.c @@ -18,6 +18,7 @@ */ #include <linux/fs.h> +#include <linux/namei.h> #include <linux/ctype.h> #include <linux/quotaops.h> #include <linux/exportfs.h> @@ -1465,7 +1466,7 @@ static struct dentry *jfs_lookup(struct inode *dip, struct dentry *dentry, struc jfs_info("jfs_lookup: name = %s", name); if (JFS_SBI(dip->i_sb)->mntflag & JFS_OS2) - dentry->d_op = &jfs_ci_dentry_operations; + d_set_d_op(dentry, &jfs_ci_dentry_operations); if ((name[0] == '.') && (len == 1)) inum = dip->i_ino; @@ -1494,7 +1495,7 @@ static struct dentry *jfs_lookup(struct inode *dip, struct dentry *dentry, struc dentry = d_splice_alias(ip, dentry); if (dentry && (JFS_SBI(dip->i_sb)->mntflag & JFS_OS2)) - dentry->d_op = &jfs_ci_dentry_operations; + d_set_d_op(dentry, &jfs_ci_dentry_operations); return dentry; } @@ -1573,7 +1574,8 @@ const struct file_operations jfs_dir_operations = { .llseek = generic_file_llseek, }; -static int jfs_ci_hash(struct dentry *dir, struct qstr *this) +static int jfs_ci_hash(const struct dentry *dir, const struct inode *inode, + struct qstr *this) { unsigned long hash; int i; @@ -1586,32 +1588,63 @@ static int jfs_ci_hash(struct dentry *dir, struct qstr *this) return 0; } -static int jfs_ci_compare(struct dentry *dir, struct qstr *a, struct qstr *b) +static int jfs_ci_compare(const struct dentry *parent, + const struct inode *pinode, + const struct dentry *dentry, const struct inode *inode, + unsigned int len, const char *str, const struct qstr *name) { int i, result = 1; - if (a->len != b->len) + if (len != name->len) goto out; - for (i=0; i < a->len; i++) { - if (tolower(a->name[i]) != tolower(b->name[i])) + for (i=0; i < len; i++) { + if (tolower(str[i]) != tolower(name->name[i])) goto out; } result = 0; +out: + return result; +} +static int jfs_ci_revalidate(struct dentry *dentry, struct nameidata *nd) +{ + if (nd->flags & LOOKUP_RCU) + return -ECHILD; /* - * We want creates to preserve case. A negative dentry, a, that - * has a different case than b may cause a new entry to be created - * with the wrong case. Since we can't tell if a comes from a negative - * dentry, we blindly replace it with b. This should be harmless if - * a is not a negative dentry. + * This is not negative dentry. Always valid. + * + * Note, rename() to existing directory entry will have ->d_inode, + * and will use existing name which isn't specified name by user. + * + * We may be able to drop this positive dentry here. But dropping + * positive dentry isn't good idea. So it's unsupported like + * rename("filename", "FILENAME") for now. */ - memcpy((unsigned char *)a->name, b->name, a->len); -out: - return result; + if (dentry->d_inode) + return 1; + + /* + * This may be nfsd (or something), anyway, we can't see the + * intent of this. So, since this can be for creation, drop it. + */ + if (!nd) + return 0; + + /* + * Drop the negative dentry, in order to make sure to use the + * case sensitive name which is specified by user if this is + * for creation. + */ + if (!(nd->flags & (LOOKUP_CONTINUE | LOOKUP_PARENT))) { + if (nd->flags & (LOOKUP_CREATE | LOOKUP_RENAME_TARGET)) + return 0; + } + return 1; } const struct dentry_operations jfs_ci_dentry_operations = { .d_hash = jfs_ci_hash, .d_compare = jfs_ci_compare, + .d_revalidate = jfs_ci_revalidate, }; diff --git a/fs/jfs/super.c b/fs/jfs/super.c index 0669fc1cc3b..3150d766e0d 100644 --- a/fs/jfs/super.c +++ b/fs/jfs/super.c @@ -115,6 +115,14 @@ static struct inode *jfs_alloc_inode(struct super_block *sb) return &jfs_inode->vfs_inode; } +static void jfs_i_callback(struct rcu_head *head) +{ + struct inode *inode = container_of(head, struct inode, i_rcu); + struct jfs_inode_info *ji = JFS_IP(inode); + INIT_LIST_HEAD(&inode->i_dentry); + kmem_cache_free(jfs_inode_cachep, ji); +} + static void jfs_destroy_inode(struct inode *inode) { struct jfs_inode_info *ji = JFS_IP(inode); @@ -128,7 +136,7 @@ static void jfs_destroy_inode(struct inode *inode) ji->active_ag = -1; } spin_unlock_irq(&ji->ag_lock); - kmem_cache_free(jfs_inode_cachep, ji); + call_rcu(&inode->i_rcu, jfs_i_callback); } static int jfs_statfs(struct dentry *dentry, struct kstatfs *buf) @@ -517,7 +525,7 @@ static int jfs_fill_super(struct super_block *sb, void *data, int silent) goto out_no_root; if (sbi->mntflag & JFS_OS2) - sb->s_root->d_op = &jfs_ci_dentry_operations; + d_set_d_op(sb->s_root, &jfs_ci_dentry_operations); /* logical blocks are represented by 40 bits in pxd_t, etc. */ sb->s_maxbytes = ((u64) sb->s_blocksize) << 40; diff --git a/fs/libfs.c b/fs/libfs.c index a3accdf528a..889311e3d06 100644 --- a/fs/libfs.c +++ b/fs/libfs.c @@ -16,6 +16,11 @@ #include <asm/uaccess.h> +static inline int simple_positive(struct dentry *dentry) +{ + return dentry->d_inode && !d_unhashed(dentry); +} + int simple_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) { @@ -37,7 +42,7 @@ int simple_statfs(struct dentry *dentry, struct kstatfs *buf) * Retaining negative dentries for an in-memory filesystem just wastes * memory and lookup time: arrange for them to be deleted immediately. */ -static int simple_delete_dentry(struct dentry *dentry) +static int simple_delete_dentry(const struct dentry *dentry) { return 1; } @@ -54,7 +59,7 @@ struct dentry *simple_lookup(struct inode *dir, struct dentry *dentry, struct na if (dentry->d_name.len > NAME_MAX) return ERR_PTR(-ENAMETOOLONG); - dentry->d_op = &simple_dentry_operations; + d_set_d_op(dentry, &simple_dentry_operations); d_add(dentry, NULL); return NULL; } @@ -76,7 +81,8 @@ int dcache_dir_close(struct inode *inode, struct file *file) loff_t dcache_dir_lseek(struct file *file, loff_t offset, int origin) { - mutex_lock(&file->f_path.dentry->d_inode->i_mutex); + struct dentry *dentry = file->f_path.dentry; + mutex_lock(&dentry->d_inode->i_mutex); switch (origin) { case 1: offset += file->f_pos; @@ -84,7 +90,7 @@ loff_t dcache_dir_lseek(struct file *file, loff_t offset, int origin) if (offset >= 0) break; default: - mutex_unlock(&file->f_path.dentry->d_inode->i_mutex); + mutex_unlock(&dentry->d_inode->i_mutex); return -EINVAL; } if (offset != file->f_pos) { @@ -94,21 +100,24 @@ loff_t dcache_dir_lseek(struct file *file, loff_t offset, int origin) struct dentry *cursor = file->private_data; loff_t n = file->f_pos - 2; - spin_lock(&dcache_lock); + spin_lock(&dentry->d_lock); + /* d_lock not required for cursor */ list_del(&cursor->d_u.d_child); - p = file->f_path.dentry->d_subdirs.next; - while (n && p != &file->f_path.dentry->d_subdirs) { + p = dentry->d_subdirs.next; + while (n && p != &dentry->d_subdirs) { struct dentry *next; next = list_entry(p, struct dentry, d_u.d_child); - if (!d_unhashed(next) && next->d_inode) + spin_lock_nested(&next->d_lock, DENTRY_D_LOCK_NESTED); + if (simple_positive(next)) n--; + spin_unlock(&next->d_lock); p = p->next; } list_add_tail(&cursor->d_u.d_child, p); - spin_unlock(&dcache_lock); + spin_unlock(&dentry->d_lock); } } - mutex_unlock(&file->f_path.dentry->d_inode->i_mutex); + mutex_unlock(&dentry->d_inode->i_mutex); return offset; } @@ -148,29 +157,35 @@ int dcache_readdir(struct file * filp, void * dirent, filldir_t filldir) i++; /* fallthrough */ default: - spin_lock(&dcache_lock); + spin_lock(&dentry->d_lock); if (filp->f_pos == 2) list_move(q, &dentry->d_subdirs); for (p=q->next; p != &dentry->d_subdirs; p=p->next) { struct dentry *next; next = list_entry(p, struct dentry, d_u.d_child); - if (d_unhashed(next) || !next->d_inode) + spin_lock_nested(&next->d_lock, DENTRY_D_LOCK_NESTED); + if (!simple_positive(next)) { + spin_unlock(&next->d_lock); continue; + } - spin_unlock(&dcache_lock); + spin_unlock(&next->d_lock); + spin_unlock(&dentry->d_lock); if (filldir(dirent, next->d_name.name, next->d_name.len, filp->f_pos, next->d_inode->i_ino, dt_type(next->d_inode)) < 0) return 0; - spin_lock(&dcache_lock); + spin_lock(&dentry->d_lock); + spin_lock_nested(&next->d_lock, DENTRY_D_LOCK_NESTED); /* next is still alive */ list_move(q, p); + spin_unlock(&next->d_lock); p = q; filp->f_pos++; } - spin_unlock(&dcache_lock); + spin_unlock(&dentry->d_lock); } return 0; } @@ -259,23 +274,23 @@ int simple_link(struct dentry *old_dentry, struct inode *dir, struct dentry *den return 0; } -static inline int simple_positive(struct dentry *dentry) -{ - return dentry->d_inode && !d_unhashed(dentry); -} - int simple_empty(struct dentry *dentry) { struct dentry *child; int ret = 0; - spin_lock(&dcache_lock); - list_for_each_entry(child, &dentry->d_subdirs, d_u.d_child) - if (simple_positive(child)) + spin_lock(&dentry->d_lock); + list_for_each_entry(child, &dentry->d_subdirs, d_u.d_child) { + spin_lock_nested(&child->d_lock, DENTRY_D_LOCK_NESTED); + if (simple_positive(child)) { + spin_unlock(&child->d_lock); goto out; + } + spin_unlock(&child->d_lock); + } ret = 1; out: - spin_unlock(&dcache_lock); + spin_unlock(&dentry->d_lock); return ret; } diff --git a/fs/lockd/clntlock.c b/fs/lockd/clntlock.c index d5bb86866e6..25509eb28fd 100644 --- a/fs/lockd/clntlock.c +++ b/fs/lockd/clntlock.c @@ -14,7 +14,6 @@ #include <linux/sunrpc/clnt.h> #include <linux/sunrpc/svc.h> #include <linux/lockd/lockd.h> -#include <linux/smp_lock.h> #include <linux/kthread.h> #define NLMDBG_FACILITY NLMDBG_CLIENT diff --git a/fs/lockd/clntproc.c b/fs/lockd/clntproc.c index 47ea1e1925b..332c54cf75e 100644 --- a/fs/lockd/clntproc.c +++ b/fs/lockd/clntproc.c @@ -7,7 +7,6 @@ */ #include <linux/module.h> -#include <linux/smp_lock.h> #include <linux/slab.h> #include <linux/types.h> #include <linux/errno.h> diff --git a/fs/lockd/host.c b/fs/lockd/host.c index 25e21e4023b..ed0c59fe23c 100644 --- a/fs/lockd/host.c +++ b/fs/lockd/host.c @@ -124,7 +124,7 @@ static struct nlm_host *nlm_lookup_host(struct nlm_lookup_host_info *ni) continue; if (host->h_server != ni->server) continue; - if (ni->server && + if (ni->server && ni->src_len != 0 && !rpc_cmp_addr(nlm_srcaddr(host), ni->src_sap)) continue; @@ -167,6 +167,7 @@ static struct nlm_host *nlm_lookup_host(struct nlm_lookup_host_info *ni) host->h_addrlen = ni->salen; rpc_set_port(nlm_addr(host), 0); memcpy(nlm_srcaddr(host), ni->src_sap, ni->src_len); + host->h_srcaddrlen = ni->src_len; host->h_version = ni->version; host->h_proto = ni->protocol; host->h_rpcclnt = NULL; @@ -238,9 +239,6 @@ struct nlm_host *nlmclnt_lookup_host(const struct sockaddr *sap, const char *hostname, int noresvport) { - const struct sockaddr source = { - .sa_family = AF_UNSPEC, - }; struct nlm_lookup_host_info ni = { .server = 0, .sap = sap, @@ -249,8 +247,6 @@ struct nlm_host *nlmclnt_lookup_host(const struct sockaddr *sap, .version = version, .hostname = hostname, .hostname_len = strlen(hostname), - .src_sap = &source, - .src_len = sizeof(source), .noresvport = noresvport, }; @@ -357,7 +353,6 @@ nlm_bind_host(struct nlm_host *host) .protocol = host->h_proto, .address = nlm_addr(host), .addrsize = host->h_addrlen, - .saddress = nlm_srcaddr(host), .timeout = &timeparms, .servername = host->h_name, .program = &nlm_program, @@ -376,6 +371,8 @@ nlm_bind_host(struct nlm_host *host) args.flags |= RPC_CLNT_CREATE_HARDRTRY; if (host->h_noresvport) args.flags |= RPC_CLNT_CREATE_NONPRIVPORT; + if (host->h_srcaddrlen) + args.saddress = nlm_srcaddr(host); clnt = rpc_create(&args); if (!IS_ERR(clnt)) diff --git a/fs/lockd/svc4proc.c b/fs/lockd/svc4proc.c index a336e832475..38d26119245 100644 --- a/fs/lockd/svc4proc.c +++ b/fs/lockd/svc4proc.c @@ -9,7 +9,6 @@ #include <linux/types.h> #include <linux/time.h> -#include <linux/smp_lock.h> #include <linux/lockd/lockd.h> #include <linux/lockd/share.h> diff --git a/fs/lockd/svclock.c b/fs/lockd/svclock.c index c462d346acb..ef5659b211e 100644 --- a/fs/lockd/svclock.c +++ b/fs/lockd/svclock.c @@ -25,7 +25,6 @@ #include <linux/errno.h> #include <linux/kernel.h> #include <linux/sched.h> -#include <linux/smp_lock.h> #include <linux/sunrpc/clnt.h> #include <linux/sunrpc/svc.h> #include <linux/lockd/nlm.h> diff --git a/fs/lockd/svcproc.c b/fs/lockd/svcproc.c index c3069f38d60..0caea5310ac 100644 --- a/fs/lockd/svcproc.c +++ b/fs/lockd/svcproc.c @@ -9,7 +9,6 @@ #include <linux/types.h> #include <linux/time.h> -#include <linux/smp_lock.h> #include <linux/lockd/lockd.h> #include <linux/lockd/share.h> diff --git a/fs/locks.c b/fs/locks.c index 65765cb6afe..08415b2a6d3 100644 --- a/fs/locks.c +++ b/fs/locks.c @@ -122,7 +122,6 @@ #include <linux/module.h> #include <linux/security.h> #include <linux/slab.h> -#include <linux/smp_lock.h> #include <linux/syscalls.h> #include <linux/time.h> #include <linux/rcupdate.h> @@ -1390,7 +1389,7 @@ int generic_setlease(struct file *filp, long arg, struct file_lock **flp) if ((arg == F_RDLCK) && (atomic_read(&inode->i_writecount) > 0)) goto out; if ((arg == F_WRLCK) - && ((atomic_read(&dentry->d_count) > 1) + && ((dentry->d_count > 1) || (atomic_read(&inode->i_count) > 1))) goto out; } @@ -1504,9 +1503,8 @@ 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; + struct file_lock *fl, *ret; struct fasync_struct *new; - struct inode *inode = filp->f_path.dentry->d_inode; int error; fl = lease_alloc(filp, arg); @@ -1518,13 +1516,16 @@ static int do_fcntl_add_lease(unsigned int fd, struct file *filp, long arg) locks_free_lock(fl); return -ENOMEM; } + ret = fl; lock_flocks(); - error = __vfs_setlease(filp, arg, &fl); + error = __vfs_setlease(filp, arg, &ret); if (error) { unlock_flocks(); locks_free_lock(fl); goto out_free_fasync; } + if (ret != fl) + locks_free_lock(fl); /* * fasync_insert_entry() returns the old entry if any. @@ -1532,17 +1533,10 @@ static int do_fcntl_add_lease(unsigned int fd, struct file *filp, long arg) * inserted it into the fasync list. Clear new so that * we don't release it here. */ - if (!fasync_insert_entry(fd, filp, &fl->fl_fasync, new)) + if (!fasync_insert_entry(fd, filp, &ret->fl_fasync, new)) new = NULL; - if (error < 0) { - /* remove lease just inserted by setlease */ - fl->fl_type = F_UNLCK | F_INPROGRESS; - fl->fl_break_time = jiffies - 10; - time_out_leases(inode); - } else { - error = __f_setown(filp, task_pid(current), PIDTYPE_PID, 0); - } + error = __f_setown(filp, task_pid(current), PIDTYPE_PID, 0); unlock_flocks(); out_free_fasync: diff --git a/fs/logfs/dir.c b/fs/logfs/dir.c index 409dfd65e9a..f9ddf0c388c 100644 --- a/fs/logfs/dir.c +++ b/fs/logfs/dir.c @@ -555,9 +555,11 @@ static int logfs_symlink(struct inode *dir, struct dentry *dentry, return __logfs_create(dir, dentry, inode, target, destlen); } -static int logfs_permission(struct inode *inode, int mask) +static int logfs_permission(struct inode *inode, int mask, unsigned int flags) { - return generic_permission(inode, mask, NULL); + if (flags & IPERM_FLAG_RCU) + return -ECHILD; + return generic_permission(inode, mask, flags, NULL); } static int logfs_link(struct dentry *old_dentry, struct inode *dir, diff --git a/fs/logfs/inode.c b/fs/logfs/inode.c index d8c71ece098..03b8c240aed 100644 --- a/fs/logfs/inode.c +++ b/fs/logfs/inode.c @@ -141,13 +141,20 @@ struct inode *logfs_safe_iget(struct super_block *sb, ino_t ino, int *is_cached) return __logfs_iget(sb, ino); } +static void logfs_i_callback(struct rcu_head *head) +{ + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); + kmem_cache_free(logfs_inode_cache, logfs_inode(inode)); +} + static void __logfs_destroy_inode(struct inode *inode) { struct logfs_inode *li = logfs_inode(inode); BUG_ON(li->li_block); list_del(&li->li_freeing_list); - kmem_cache_free(logfs_inode_cache, li); + call_rcu(&inode->i_rcu, logfs_i_callback); } static void logfs_destroy_inode(struct inode *inode) diff --git a/fs/logfs/journal.c b/fs/logfs/journal.c index f46ee8b0e13..9da29706f91 100644 --- a/fs/logfs/journal.c +++ b/fs/logfs/journal.c @@ -828,7 +828,7 @@ void do_logfs_journal_wl_pass(struct super_block *sb) super->s_journal_seg[i] = segno; super->s_journal_ec[i] = ec; logfs_set_segment_reserved(sb, segno); - err = btree_insert32(head, segno, (void *)1, GFP_KERNEL); + err = btree_insert32(head, segno, (void *)1, GFP_NOFS); BUG_ON(err); /* mempool should prevent this */ err = logfs_erase_segment(sb, segno, 1); BUG_ON(err); /* FIXME: remount-ro would be nicer */ diff --git a/fs/logfs/logfs.h b/fs/logfs/logfs.h index cd51a36b37f..57afd4a6fab 100644 --- a/fs/logfs/logfs.h +++ b/fs/logfs/logfs.h @@ -486,7 +486,7 @@ static inline int logfs_get_sb_bdev(struct logfs_super *s, /* dev_mtd.c */ #ifdef CONFIG_MTD -int logfs_get_sb_mtd(struct logfs_super *s, int mtdnr) +int logfs_get_sb_mtd(struct logfs_super *s, int mtdnr); #else static inline int logfs_get_sb_mtd(struct logfs_super *s, int mtdnr) { diff --git a/fs/logfs/readwrite.c b/fs/logfs/readwrite.c index 6127baf0e18..ee99a9f5dfd 100644 --- a/fs/logfs/readwrite.c +++ b/fs/logfs/readwrite.c @@ -1994,6 +1994,9 @@ static int do_write_inode(struct inode *inode) /* FIXME: transaction is part of logfs_block now. Is that enough? */ err = logfs_write_buf(master_inode, page, 0); + if (err) + move_page_to_inode(inode, page); + logfs_put_write_page(page); return err; } diff --git a/fs/minix/inode.c b/fs/minix/inode.c index fb2020858a3..ae0b83f476a 100644 --- a/fs/minix/inode.c +++ b/fs/minix/inode.c @@ -68,11 +68,18 @@ static struct inode *minix_alloc_inode(struct super_block *sb) return &ei->vfs_inode; } -static void minix_destroy_inode(struct inode *inode) +static void minix_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(minix_inode_cachep, minix_i(inode)); } +static void minix_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, minix_i_callback); +} + static void init_once(void *foo) { struct minix_inode_info *ei = (struct minix_inode_info *) foo; diff --git a/fs/minix/namei.c b/fs/minix/namei.c index c0d35a3acce..1b9e07728a9 100644 --- a/fs/minix/namei.c +++ b/fs/minix/namei.c @@ -23,7 +23,7 @@ static struct dentry *minix_lookup(struct inode * dir, struct dentry *dentry, st struct inode * inode = NULL; ino_t ino; - dentry->d_op = dir->i_sb->s_root->d_op; + d_set_d_op(dentry, dir->i_sb->s_root->d_op); if (dentry->d_name.len > minix_sb(dir->i_sb)->s_namelen) return ERR_PTR(-ENAMETOOLONG); diff --git a/fs/namei.c b/fs/namei.c index 5362af9b737..19433cdba01 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -169,8 +169,8 @@ EXPORT_SYMBOL(putname); /* * This does basic POSIX ACL permission checking */ -static int acl_permission_check(struct inode *inode, int mask, - int (*check_acl)(struct inode *inode, int mask)) +static int acl_permission_check(struct inode *inode, int mask, unsigned int flags, + int (*check_acl)(struct inode *inode, int mask, unsigned int flags)) { umode_t mode = inode->i_mode; @@ -180,7 +180,7 @@ static int acl_permission_check(struct inode *inode, int mask, mode >>= 6; else { if (IS_POSIXACL(inode) && (mode & S_IRWXG) && check_acl) { - int error = check_acl(inode, mask); + int error = check_acl(inode, mask, flags); if (error != -EAGAIN) return error; } @@ -198,25 +198,30 @@ static int acl_permission_check(struct inode *inode, int mask, } /** - * generic_permission - check for access rights on a Posix-like filesystem + * generic_permission - check for access rights on a Posix-like filesystem * @inode: inode to check access rights for * @mask: right to check for (%MAY_READ, %MAY_WRITE, %MAY_EXEC) * @check_acl: optional callback to check for Posix ACLs + * @flags IPERM_FLAG_ flags. * * Used to check for read/write/execute permissions on a file. * We use "fsuid" for this, letting us set arbitrary permissions * for filesystem access without changing the "normal" uids which - * are used for other things.. + * are used for other things. + * + * generic_permission is rcu-walk aware. It returns -ECHILD in case an rcu-walk + * request cannot be satisfied (eg. requires blocking or too much complexity). + * It would then be called again in ref-walk mode. */ -int generic_permission(struct inode *inode, int mask, - int (*check_acl)(struct inode *inode, int mask)) +int generic_permission(struct inode *inode, int mask, unsigned int flags, + int (*check_acl)(struct inode *inode, int mask, unsigned int flags)) { int ret; /* * Do the basic POSIX ACL permission checks. */ - ret = acl_permission_check(inode, mask, check_acl); + ret = acl_permission_check(inode, mask, flags, check_acl); if (ret != -EACCES) return ret; @@ -271,9 +276,10 @@ int inode_permission(struct inode *inode, int mask) } if (inode->i_op->permission) - retval = inode->i_op->permission(inode, mask); + retval = inode->i_op->permission(inode, mask, 0); else - retval = generic_permission(inode, mask, inode->i_op->check_acl); + retval = generic_permission(inode, mask, 0, + inode->i_op->check_acl); if (retval) return retval; @@ -362,6 +368,18 @@ void path_get(struct path *path) EXPORT_SYMBOL(path_get); /** + * path_get_long - get a long reference to a path + * @path: path to get the reference to + * + * Given a path increment the reference count to the dentry and the vfsmount. + */ +void path_get_long(struct path *path) +{ + mntget_long(path->mnt); + dget(path->dentry); +} + +/** * path_put - put a reference to a path * @path: path to put the reference to * @@ -375,6 +393,185 @@ void path_put(struct path *path) EXPORT_SYMBOL(path_put); /** + * path_put_long - put a long reference to a path + * @path: path to put the reference to + * + * Given a path decrement the reference count to the dentry and the vfsmount. + */ +void path_put_long(struct path *path) +{ + dput(path->dentry); + mntput_long(path->mnt); +} + +/** + * nameidata_drop_rcu - drop this nameidata out of rcu-walk + * @nd: nameidata pathwalk data to drop + * @Returns: 0 on success, -ECHLID on failure + * + * Path walking has 2 modes, rcu-walk and ref-walk (see + * Documentation/filesystems/path-lookup.txt). __drop_rcu* functions attempt + * to drop out of rcu-walk mode and take normal reference counts on dentries + * and vfsmounts to transition to rcu-walk mode. __drop_rcu* functions take + * refcounts at the last known good point before rcu-walk got stuck, so + * ref-walk may continue from there. If this is not successful (eg. a seqcount + * has changed), then failure is returned and path walk restarts from the + * beginning in ref-walk mode. + * + * nameidata_drop_rcu attempts to drop the current nd->path and nd->root into + * ref-walk. Must be called from rcu-walk context. + */ +static int nameidata_drop_rcu(struct nameidata *nd) +{ + struct fs_struct *fs = current->fs; + struct dentry *dentry = nd->path.dentry; + + BUG_ON(!(nd->flags & LOOKUP_RCU)); + if (nd->root.mnt) { + spin_lock(&fs->lock); + if (nd->root.mnt != fs->root.mnt || + nd->root.dentry != fs->root.dentry) + goto err_root; + } + spin_lock(&dentry->d_lock); + if (!__d_rcu_to_refcount(dentry, nd->seq)) + goto err; + BUG_ON(nd->inode != dentry->d_inode); + spin_unlock(&dentry->d_lock); + if (nd->root.mnt) { + path_get(&nd->root); + spin_unlock(&fs->lock); + } + mntget(nd->path.mnt); + + rcu_read_unlock(); + br_read_unlock(vfsmount_lock); + nd->flags &= ~LOOKUP_RCU; + return 0; +err: + spin_unlock(&dentry->d_lock); +err_root: + if (nd->root.mnt) + spin_unlock(&fs->lock); + return -ECHILD; +} + +/* Try to drop out of rcu-walk mode if we were in it, otherwise do nothing. */ +static inline int nameidata_drop_rcu_maybe(struct nameidata *nd) +{ + if (nd->flags & LOOKUP_RCU) + return nameidata_drop_rcu(nd); + return 0; +} + +/** + * nameidata_dentry_drop_rcu - drop nameidata and dentry out of rcu-walk + * @nd: nameidata pathwalk data to drop + * @dentry: dentry to drop + * @Returns: 0 on success, -ECHLID on failure + * + * nameidata_dentry_drop_rcu attempts to drop the current nd->path and nd->root, + * and dentry into ref-walk. @dentry must be a path found by a do_lookup call on + * @nd. Must be called from rcu-walk context. + */ +static int nameidata_dentry_drop_rcu(struct nameidata *nd, struct dentry *dentry) +{ + struct fs_struct *fs = current->fs; + struct dentry *parent = nd->path.dentry; + + BUG_ON(!(nd->flags & LOOKUP_RCU)); + if (nd->root.mnt) { + spin_lock(&fs->lock); + if (nd->root.mnt != fs->root.mnt || + nd->root.dentry != fs->root.dentry) + goto err_root; + } + spin_lock(&parent->d_lock); + spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED); + if (!__d_rcu_to_refcount(dentry, nd->seq)) + goto err; + /* + * If the sequence check on the child dentry passed, then the child has + * not been removed from its parent. This means the parent dentry must + * be valid and able to take a reference at this point. + */ + BUG_ON(!IS_ROOT(dentry) && dentry->d_parent != parent); + BUG_ON(!parent->d_count); + parent->d_count++; + spin_unlock(&dentry->d_lock); + spin_unlock(&parent->d_lock); + if (nd->root.mnt) { + path_get(&nd->root); + spin_unlock(&fs->lock); + } + mntget(nd->path.mnt); + + rcu_read_unlock(); + br_read_unlock(vfsmount_lock); + nd->flags &= ~LOOKUP_RCU; + return 0; +err: + spin_unlock(&dentry->d_lock); + spin_unlock(&parent->d_lock); +err_root: + if (nd->root.mnt) + spin_unlock(&fs->lock); + return -ECHILD; +} + +/* Try to drop out of rcu-walk mode if we were in it, otherwise do nothing. */ +static inline int nameidata_dentry_drop_rcu_maybe(struct nameidata *nd, struct dentry *dentry) +{ + if (nd->flags & LOOKUP_RCU) + return nameidata_dentry_drop_rcu(nd, dentry); + return 0; +} + +/** + * nameidata_drop_rcu_last - drop nameidata ending path walk out of rcu-walk + * @nd: nameidata pathwalk data to drop + * @Returns: 0 on success, -ECHLID on failure + * + * nameidata_drop_rcu_last attempts to drop the current nd->path into ref-walk. + * nd->path should be the final element of the lookup, so nd->root is discarded. + * Must be called from rcu-walk context. + */ +static int nameidata_drop_rcu_last(struct nameidata *nd) +{ + struct dentry *dentry = nd->path.dentry; + + BUG_ON(!(nd->flags & LOOKUP_RCU)); + nd->flags &= ~LOOKUP_RCU; + nd->root.mnt = NULL; + spin_lock(&dentry->d_lock); + if (!__d_rcu_to_refcount(dentry, nd->seq)) + goto err_unlock; + BUG_ON(nd->inode != dentry->d_inode); + spin_unlock(&dentry->d_lock); + + mntget(nd->path.mnt); + + rcu_read_unlock(); + br_read_unlock(vfsmount_lock); + + return 0; + +err_unlock: + spin_unlock(&dentry->d_lock); + rcu_read_unlock(); + br_read_unlock(vfsmount_lock); + return -ECHILD; +} + +/* Try to drop out of rcu-walk mode if we were in it, otherwise do nothing. */ +static inline int nameidata_drop_rcu_last_maybe(struct nameidata *nd) +{ + if (likely(nd->flags & LOOKUP_RCU)) + return nameidata_drop_rcu_last(nd); + return 0; +} + +/** * release_open_intent - free up open intent resources * @nd: pointer to nameidata */ @@ -386,10 +583,26 @@ void release_open_intent(struct nameidata *nd) fput(nd->intent.open.file); } +static int d_revalidate(struct dentry *dentry, struct nameidata *nd) +{ + int status; + + status = dentry->d_op->d_revalidate(dentry, nd); + if (status == -ECHILD) { + if (nameidata_dentry_drop_rcu(nd, dentry)) + return status; + status = dentry->d_op->d_revalidate(dentry, nd); + } + + return status; +} + static inline struct dentry * do_revalidate(struct dentry *dentry, struct nameidata *nd) { - int status = dentry->d_op->d_revalidate(dentry, nd); + int status; + + status = d_revalidate(dentry, nd); if (unlikely(status <= 0)) { /* * The dentry failed validation. @@ -397,19 +610,36 @@ do_revalidate(struct dentry *dentry, struct nameidata *nd) * the dentry otherwise d_revalidate is asking us * to return a fail status. */ - if (!status) { + if (status < 0) { + /* If we're in rcu-walk, we don't have a ref */ + if (!(nd->flags & LOOKUP_RCU)) + dput(dentry); + dentry = ERR_PTR(status); + + } else { + /* Don't d_invalidate in rcu-walk mode */ + if (nameidata_dentry_drop_rcu_maybe(nd, dentry)) + return ERR_PTR(-ECHILD); if (!d_invalidate(dentry)) { dput(dentry); dentry = NULL; } - } else { - dput(dentry); - dentry = ERR_PTR(status); } } return dentry; } +static inline int need_reval_dot(struct dentry *dentry) +{ + if (likely(!(dentry->d_flags & DCACHE_OP_REVALIDATE))) + return 0; + + if (likely(!(dentry->d_sb->s_type->fs_flags & FS_REVAL_DOT))) + return 0; + + return 1; +} + /* * force_reval_path - force revalidation of a dentry * @@ -433,13 +663,12 @@ force_reval_path(struct path *path, struct nameidata *nd) /* * only check on filesystems where it's possible for the dentry to - * become stale. It's assumed that if this flag is set then the - * d_revalidate op will also be defined. + * become stale. */ - if (!(dentry->d_sb->s_type->fs_flags & FS_REVAL_DOT)) + if (!need_reval_dot(dentry)) return 0; - status = dentry->d_op->d_revalidate(dentry, nd); + status = d_revalidate(dentry, nd); if (status > 0) return 0; @@ -459,26 +688,27 @@ force_reval_path(struct path *path, struct nameidata *nd) * short-cut DAC fails, then call ->permission() to do more * complete permission check. */ -static int exec_permission(struct inode *inode) +static inline int exec_permission(struct inode *inode, unsigned int flags) { int ret; if (inode->i_op->permission) { - ret = inode->i_op->permission(inode, MAY_EXEC); - if (!ret) - goto ok; - return ret; + ret = inode->i_op->permission(inode, MAY_EXEC, flags); + } else { + ret = acl_permission_check(inode, MAY_EXEC, flags, + inode->i_op->check_acl); } - ret = acl_permission_check(inode, MAY_EXEC, inode->i_op->check_acl); - if (!ret) + if (likely(!ret)) goto ok; + if (ret == -ECHILD) + return ret; if (capable(CAP_DAC_OVERRIDE) || capable(CAP_DAC_READ_SEARCH)) goto ok; return ret; ok: - return security_inode_permission(inode, MAY_EXEC); + return security_inode_exec_permission(inode, flags); } static __always_inline void set_root(struct nameidata *nd) @@ -489,8 +719,23 @@ static __always_inline void set_root(struct nameidata *nd) static int link_path_walk(const char *, struct nameidata *); +static __always_inline void set_root_rcu(struct nameidata *nd) +{ + if (!nd->root.mnt) { + struct fs_struct *fs = current->fs; + unsigned seq; + + do { + seq = read_seqcount_begin(&fs->seq); + nd->root = fs->root; + } while (read_seqcount_retry(&fs->seq, seq)); + } +} + static __always_inline int __vfs_follow_link(struct nameidata *nd, const char *link) { + int ret; + if (IS_ERR(link)) goto fail; @@ -500,8 +745,10 @@ static __always_inline int __vfs_follow_link(struct nameidata *nd, const char *l nd->path = nd->root; path_get(&nd->root); } + nd->inode = nd->path.dentry->d_inode; - return link_path_walk(link, nd); + ret = link_path_walk(link, nd); + return ret; fail: path_put(&nd->path); return PTR_ERR(link); @@ -516,11 +763,12 @@ static void path_put_conditional(struct path *path, struct nameidata *nd) static inline void path_to_nameidata(struct path *path, struct nameidata *nd) { - dput(nd->path.dentry); - if (nd->path.mnt != path->mnt) { - mntput(nd->path.mnt); - nd->path.mnt = path->mnt; + if (!(nd->flags & LOOKUP_RCU)) { + dput(nd->path.dentry); + if (nd->path.mnt != path->mnt) + mntput(nd->path.mnt); } + nd->path.mnt = path->mnt; nd->path.dentry = path->dentry; } @@ -535,9 +783,11 @@ __do_follow_link(struct path *path, struct nameidata *nd, void **p) if (path->mnt != nd->path.mnt) { path_to_nameidata(path, nd); + nd->inode = nd->path.dentry->d_inode; dget(dentry); } mntget(path->mnt); + nd->last_type = LAST_BIND; *p = dentry->d_inode->i_op->follow_link(dentry, nd); error = PTR_ERR(*p); @@ -591,6 +841,20 @@ loop: return err; } +static int follow_up_rcu(struct path *path) +{ + struct vfsmount *parent; + struct dentry *mountpoint; + + parent = path->mnt->mnt_parent; + if (parent == path->mnt) + return 0; + mountpoint = path->mnt->mnt_mountpoint; + path->dentry = mountpoint; + path->mnt = parent; + return 1; +} + int follow_up(struct path *path) { struct vfsmount *parent; @@ -612,9 +876,24 @@ int follow_up(struct path *path) return 1; } -/* no need for dcache_lock, as serialization is taken care in - * namespace.c +/* + * serialization is taken care of in namespace.c */ +static void __follow_mount_rcu(struct nameidata *nd, struct path *path, + struct inode **inode) +{ + while (d_mountpoint(path->dentry)) { + struct vfsmount *mounted; + mounted = __lookup_mnt(path->mnt, path->dentry, 1); + if (!mounted) + return; + path->mnt = mounted; + path->dentry = mounted->mnt_root; + nd->seq = read_seqcount_begin(&path->dentry->d_seq); + *inode = path->dentry->d_inode; + } +} + static int __follow_mount(struct path *path) { int res = 0; @@ -645,9 +924,6 @@ static void follow_mount(struct path *path) } } -/* no need for dcache_lock, as serialization is taken care in - * namespace.c - */ int follow_down(struct path *path) { struct vfsmount *mounted; @@ -663,7 +939,42 @@ int follow_down(struct path *path) return 0; } -static __always_inline void follow_dotdot(struct nameidata *nd) +static int follow_dotdot_rcu(struct nameidata *nd) +{ + struct inode *inode = nd->inode; + + set_root_rcu(nd); + + while(1) { + if (nd->path.dentry == nd->root.dentry && + nd->path.mnt == nd->root.mnt) { + break; + } + if (nd->path.dentry != nd->path.mnt->mnt_root) { + struct dentry *old = nd->path.dentry; + struct dentry *parent = old->d_parent; + unsigned seq; + + seq = read_seqcount_begin(&parent->d_seq); + if (read_seqcount_retry(&old->d_seq, nd->seq)) + return -ECHILD; + inode = parent->d_inode; + nd->path.dentry = parent; + nd->seq = seq; + break; + } + if (!follow_up_rcu(&nd->path)) + break; + nd->seq = read_seqcount_begin(&nd->path.dentry->d_seq); + inode = nd->path.dentry->d_inode; + } + __follow_mount_rcu(nd, &nd->path, &inode); + nd->inode = inode; + + return 0; +} + +static void follow_dotdot(struct nameidata *nd) { set_root(nd); @@ -684,6 +995,7 @@ static __always_inline void follow_dotdot(struct nameidata *nd) break; } follow_mount(&nd->path); + nd->inode = nd->path.dentry->d_inode; } /* @@ -721,17 +1033,17 @@ static struct dentry *d_alloc_and_lookup(struct dentry *parent, * It _is_ time-critical. */ static int do_lookup(struct nameidata *nd, struct qstr *name, - struct path *path) + struct path *path, struct inode **inode) { struct vfsmount *mnt = nd->path.mnt; - struct dentry *dentry, *parent; + struct dentry *dentry, *parent = nd->path.dentry; struct inode *dir; /* * See if the low-level filesystem might want * to use its own hash.. */ - if (nd->path.dentry->d_op && nd->path.dentry->d_op->d_hash) { - int err = nd->path.dentry->d_op->d_hash(nd->path.dentry, name); + if (unlikely(parent->d_flags & DCACHE_OP_HASH)) { + int err = parent->d_op->d_hash(parent, nd->inode, name); if (err < 0) return err; } @@ -741,21 +1053,44 @@ static int do_lookup(struct nameidata *nd, struct qstr *name, * of a false negative due to a concurrent rename, we're going to * do the non-racy lookup, below. */ - dentry = __d_lookup(nd->path.dentry, name); - if (!dentry) - goto need_lookup; + if (nd->flags & LOOKUP_RCU) { + unsigned seq; + + *inode = nd->inode; + dentry = __d_lookup_rcu(parent, name, &seq, inode); + if (!dentry) { + if (nameidata_drop_rcu(nd)) + return -ECHILD; + goto need_lookup; + } + /* Memory barrier in read_seqcount_begin of child is enough */ + if (__read_seqcount_retry(&parent->d_seq, nd->seq)) + return -ECHILD; + + nd->seq = seq; + if (dentry->d_flags & DCACHE_OP_REVALIDATE) + goto need_revalidate; + path->mnt = mnt; + path->dentry = dentry; + __follow_mount_rcu(nd, path, inode); + } else { + dentry = __d_lookup(parent, name); + if (!dentry) + goto need_lookup; found: - if (dentry->d_op && dentry->d_op->d_revalidate) - goto need_revalidate; + if (dentry->d_flags & DCACHE_OP_REVALIDATE) + goto need_revalidate; done: - path->mnt = mnt; - path->dentry = dentry; - __follow_mount(path); + path->mnt = mnt; + path->dentry = dentry; + __follow_mount(path); + *inode = path->dentry->d_inode; + } return 0; need_lookup: - parent = nd->path.dentry; dir = parent->d_inode; + BUG_ON(nd->inode != dir); mutex_lock(&dir->i_mutex); /* @@ -817,7 +1152,6 @@ static inline int follow_on_final(struct inode *inode, unsigned lookup_flags) static int link_path_walk(const char *name, struct nameidata *nd) { struct path next; - struct inode *inode; int err; unsigned int lookup_flags = nd->flags; @@ -826,18 +1160,28 @@ static int link_path_walk(const char *name, struct nameidata *nd) if (!*name) goto return_reval; - inode = nd->path.dentry->d_inode; if (nd->depth) lookup_flags = LOOKUP_FOLLOW | (nd->flags & LOOKUP_CONTINUE); /* At this point we know we have a real path component. */ for(;;) { + struct inode *inode; unsigned long hash; struct qstr this; unsigned int c; nd->flags |= LOOKUP_CONTINUE; - err = exec_permission(inode); + if (nd->flags & LOOKUP_RCU) { + err = exec_permission(nd->inode, IPERM_FLAG_RCU); + if (err == -ECHILD) { + if (nameidata_drop_rcu(nd)) + return -ECHILD; + goto exec_again; + } + } else { +exec_again: + err = exec_permission(nd->inode, 0); + } if (err) break; @@ -868,37 +1212,44 @@ static int link_path_walk(const char *name, struct nameidata *nd) if (this.name[0] == '.') switch (this.len) { default: break; - case 2: + case 2: if (this.name[1] != '.') break; - follow_dotdot(nd); - inode = nd->path.dentry->d_inode; + if (nd->flags & LOOKUP_RCU) { + if (follow_dotdot_rcu(nd)) + return -ECHILD; + } else + follow_dotdot(nd); /* fallthrough */ case 1: continue; } /* This does the actual lookups.. */ - err = do_lookup(nd, &this, &next); + err = do_lookup(nd, &this, &next, &inode); if (err) break; - err = -ENOENT; - inode = next.dentry->d_inode; if (!inode) goto out_dput; if (inode->i_op->follow_link) { + /* We commonly drop rcu-walk here */ + if (nameidata_dentry_drop_rcu_maybe(nd, next.dentry)) + return -ECHILD; + BUG_ON(inode != next.dentry->d_inode); err = do_follow_link(&next, nd); if (err) goto return_err; + nd->inode = nd->path.dentry->d_inode; err = -ENOENT; - inode = nd->path.dentry->d_inode; - if (!inode) + if (!nd->inode) break; - } else + } else { path_to_nameidata(&next, nd); + nd->inode = inode; + } err = -ENOTDIR; - if (!inode->i_op->lookup) + if (!nd->inode->i_op->lookup) break; continue; /* here ends the main loop */ @@ -913,32 +1264,39 @@ last_component: if (this.name[0] == '.') switch (this.len) { default: break; - case 2: + case 2: if (this.name[1] != '.') break; - follow_dotdot(nd); - inode = nd->path.dentry->d_inode; + if (nd->flags & LOOKUP_RCU) { + if (follow_dotdot_rcu(nd)) + return -ECHILD; + } else + follow_dotdot(nd); /* fallthrough */ case 1: goto return_reval; } - err = do_lookup(nd, &this, &next); + err = do_lookup(nd, &this, &next, &inode); if (err) break; - inode = next.dentry->d_inode; if (follow_on_final(inode, lookup_flags)) { + if (nameidata_dentry_drop_rcu_maybe(nd, next.dentry)) + return -ECHILD; + BUG_ON(inode != next.dentry->d_inode); err = do_follow_link(&next, nd); if (err) goto return_err; - inode = nd->path.dentry->d_inode; - } else + nd->inode = nd->path.dentry->d_inode; + } else { path_to_nameidata(&next, nd); + nd->inode = inode; + } err = -ENOENT; - if (!inode) + if (!nd->inode) break; if (lookup_flags & LOOKUP_DIRECTORY) { err = -ENOTDIR; - if (!inode->i_op->lookup) + if (!nd->inode->i_op->lookup) break; } goto return_base; @@ -958,25 +1316,43 @@ return_reval: * We bypassed the ordinary revalidation routines. * We may need to check the cached dentry for staleness. */ - if (nd->path.dentry && nd->path.dentry->d_sb && - (nd->path.dentry->d_sb->s_type->fs_flags & FS_REVAL_DOT)) { - err = -ESTALE; + if (need_reval_dot(nd->path.dentry)) { /* Note: we do not d_invalidate() */ - if (!nd->path.dentry->d_op->d_revalidate( - nd->path.dentry, nd)) + err = d_revalidate(nd->path.dentry, nd); + if (!err) + err = -ESTALE; + if (err < 0) break; } return_base: + if (nameidata_drop_rcu_last_maybe(nd)) + return -ECHILD; return 0; out_dput: - path_put_conditional(&next, nd); + if (!(nd->flags & LOOKUP_RCU)) + path_put_conditional(&next, nd); break; } - path_put(&nd->path); + if (!(nd->flags & LOOKUP_RCU)) + path_put(&nd->path); return_err: return err; } +static inline int path_walk_rcu(const char *name, struct nameidata *nd) +{ + current->total_link_count = 0; + + return link_path_walk(name, nd); +} + +static inline int path_walk_simple(const char *name, struct nameidata *nd) +{ + current->total_link_count = 0; + + return link_path_walk(name, nd); +} + static int path_walk(const char *name, struct nameidata *nd) { struct path save = nd->path; @@ -1002,6 +1378,93 @@ static int path_walk(const char *name, struct nameidata *nd) return result; } +static void path_finish_rcu(struct nameidata *nd) +{ + if (nd->flags & LOOKUP_RCU) { + /* RCU dangling. Cancel it. */ + nd->flags &= ~LOOKUP_RCU; + nd->root.mnt = NULL; + rcu_read_unlock(); + br_read_unlock(vfsmount_lock); + } + if (nd->file) + fput(nd->file); +} + +static int path_init_rcu(int dfd, const char *name, unsigned int flags, struct nameidata *nd) +{ + int retval = 0; + int fput_needed; + struct file *file; + + nd->last_type = LAST_ROOT; /* if there are only slashes... */ + nd->flags = flags | LOOKUP_RCU; + nd->depth = 0; + nd->root.mnt = NULL; + nd->file = NULL; + + if (*name=='/') { + struct fs_struct *fs = current->fs; + unsigned seq; + + br_read_lock(vfsmount_lock); + rcu_read_lock(); + + do { + seq = read_seqcount_begin(&fs->seq); + nd->root = fs->root; + nd->path = nd->root; + nd->seq = __read_seqcount_begin(&nd->path.dentry->d_seq); + } while (read_seqcount_retry(&fs->seq, seq)); + + } else if (dfd == AT_FDCWD) { + struct fs_struct *fs = current->fs; + unsigned seq; + + br_read_lock(vfsmount_lock); + rcu_read_lock(); + + do { + seq = read_seqcount_begin(&fs->seq); + nd->path = fs->pwd; + nd->seq = __read_seqcount_begin(&nd->path.dentry->d_seq); + } while (read_seqcount_retry(&fs->seq, seq)); + + } else { + struct dentry *dentry; + + file = fget_light(dfd, &fput_needed); + retval = -EBADF; + if (!file) + goto out_fail; + + dentry = file->f_path.dentry; + + retval = -ENOTDIR; + if (!S_ISDIR(dentry->d_inode->i_mode)) + goto fput_fail; + + retval = file_permission(file, MAY_EXEC); + if (retval) + goto fput_fail; + + nd->path = file->f_path; + if (fput_needed) + nd->file = file; + + nd->seq = __read_seqcount_begin(&nd->path.dentry->d_seq); + br_read_lock(vfsmount_lock); + rcu_read_lock(); + } + nd->inode = nd->path.dentry->d_inode; + return 0; + +fput_fail: + fput_light(file, fput_needed); +out_fail: + return retval; +} + static int path_init(int dfd, const char *name, unsigned int flags, struct nameidata *nd) { int retval = 0; @@ -1042,6 +1505,7 @@ static int path_init(int dfd, const char *name, unsigned int flags, struct namei fput_light(file, fput_needed); } + nd->inode = nd->path.dentry->d_inode; return 0; fput_fail: @@ -1054,16 +1518,53 @@ out_fail: static int do_path_lookup(int dfd, const char *name, unsigned int flags, struct nameidata *nd) { - int retval = path_init(dfd, name, flags, nd); - if (!retval) - retval = path_walk(name, nd); - if (unlikely(!retval && !audit_dummy_context() && nd->path.dentry && - nd->path.dentry->d_inode)) - audit_inode(name, nd->path.dentry); + int retval; + + /* + * Path walking is largely split up into 2 different synchronisation + * schemes, rcu-walk and ref-walk (explained in + * Documentation/filesystems/path-lookup.txt). These share much of the + * path walk code, but some things particularly setup, cleanup, and + * following mounts are sufficiently divergent that functions are + * duplicated. Typically there is a function foo(), and its RCU + * analogue, foo_rcu(). + * + * -ECHILD is the error number of choice (just to avoid clashes) that + * is returned if some aspect of an rcu-walk fails. Such an error must + * be handled by restarting a traditional ref-walk (which will always + * be able to complete). + */ + retval = path_init_rcu(dfd, name, flags, nd); + if (unlikely(retval)) + return retval; + retval = path_walk_rcu(name, nd); + path_finish_rcu(nd); if (nd->root.mnt) { path_put(&nd->root); nd->root.mnt = NULL; } + + if (unlikely(retval == -ECHILD || retval == -ESTALE)) { + /* slower, locked walk */ + if (retval == -ESTALE) + flags |= LOOKUP_REVAL; + retval = path_init(dfd, name, flags, nd); + if (unlikely(retval)) + return retval; + retval = path_walk(name, nd); + if (nd->root.mnt) { + path_put(&nd->root); + nd->root.mnt = NULL; + } + } + + if (likely(!retval)) { + if (unlikely(!audit_dummy_context())) { + if (nd->path.dentry && nd->inode) + audit_inode(name, nd->path.dentry); + } + } + return retval; } @@ -1106,10 +1607,11 @@ int vfs_path_lookup(struct dentry *dentry, struct vfsmount *mnt, path_get(&nd->path); nd->root = nd->path; path_get(&nd->root); + nd->inode = nd->path.dentry->d_inode; retval = path_walk(name, nd); if (unlikely(!retval && !audit_dummy_context() && nd->path.dentry && - nd->path.dentry->d_inode)) + nd->inode)) audit_inode(name, nd->path.dentry); path_put(&nd->root); @@ -1125,7 +1627,7 @@ static struct dentry *__lookup_hash(struct qstr *name, struct dentry *dentry; int err; - err = exec_permission(inode); + err = exec_permission(inode, 0); if (err) return ERR_PTR(err); @@ -1133,8 +1635,8 @@ static struct dentry *__lookup_hash(struct qstr *name, * See if the low-level filesystem might want * to use its own hash.. */ - if (base->d_op && base->d_op->d_hash) { - err = base->d_op->d_hash(base, name); + if (base->d_flags & DCACHE_OP_HASH) { + err = base->d_op->d_hash(base, inode, name); dentry = ERR_PTR(err); if (err < 0) goto out; @@ -1147,7 +1649,7 @@ static struct dentry *__lookup_hash(struct qstr *name, */ dentry = d_lookup(base, name); - if (dentry && dentry->d_op && dentry->d_op->d_revalidate) + if (dentry && (dentry->d_flags & DCACHE_OP_REVALIDATE)) dentry = do_revalidate(dentry, nd); if (!dentry) @@ -1490,6 +1992,7 @@ out_unlock: mutex_unlock(&dir->d_inode->i_mutex); dput(nd->path.dentry); nd->path.dentry = path->dentry; + if (error) return error; /* Don't check for write permission, don't truncate */ @@ -1584,6 +2087,9 @@ exit: return ERR_PTR(error); } +/* + * Handle O_CREAT case for do_filp_open + */ static struct file *do_last(struct nameidata *nd, struct path *path, int open_flag, int acc_mode, int mode, const char *pathname) @@ -1597,50 +2103,25 @@ static struct file *do_last(struct nameidata *nd, struct path *path, follow_dotdot(nd); dir = nd->path.dentry; case LAST_DOT: - if (nd->path.mnt->mnt_sb->s_type->fs_flags & FS_REVAL_DOT) { - if (!dir->d_op->d_revalidate(dir, nd)) { + if (need_reval_dot(dir)) { + error = d_revalidate(nd->path.dentry, nd); + if (!error) error = -ESTALE; + if (error < 0) goto exit; - } } /* fallthrough */ case LAST_ROOT: - if (open_flag & O_CREAT) - goto exit; - /* fallthrough */ + goto exit; case LAST_BIND: audit_inode(pathname, dir); goto ok; } /* trailing slashes? */ - if (nd->last.name[nd->last.len]) { - if (open_flag & O_CREAT) - goto exit; - nd->flags |= LOOKUP_DIRECTORY | LOOKUP_FOLLOW; - } - - /* just plain open? */ - if (!(open_flag & O_CREAT)) { - error = do_lookup(nd, &nd->last, path); - if (error) - goto exit; - error = -ENOENT; - if (!path->dentry->d_inode) - goto exit_dput; - if (path->dentry->d_inode->i_op->follow_link) - return NULL; - error = -ENOTDIR; - if (nd->flags & LOOKUP_DIRECTORY) { - if (!path->dentry->d_inode->i_op->lookup) - goto exit_dput; - } - path_to_nameidata(path, nd); - audit_inode(pathname, nd->path.dentry); - goto ok; - } + if (nd->last.name[nd->last.len]) + goto exit; - /* OK, it's O_CREAT */ mutex_lock(&dir->d_inode->i_mutex); path->dentry = lookup_hash(nd); @@ -1711,8 +2192,9 @@ static struct file *do_last(struct nameidata *nd, struct path *path, return NULL; path_to_nameidata(path, nd); + nd->inode = path->dentry->d_inode; error = -EISDIR; - if (S_ISDIR(path->dentry->d_inode->i_mode)) + if (S_ISDIR(nd->inode->i_mode)) goto exit; ok: filp = finish_open(nd, open_flag, acc_mode); @@ -1743,11 +2225,14 @@ struct file *do_filp_open(int dfd, const char *pathname, struct path path; int count = 0; int flag = open_to_namei_flags(open_flag); - int force_reval = 0; + int flags; if (!(open_flag & O_CREAT)) mode = 0; + /* Must never be set by userspace */ + open_flag &= ~FMODE_NONOTIFY; + /* * O_SYNC is implemented as __O_SYNC|O_DSYNC. As many places only * check for O_DSYNC if the need any syncing at all we enforce it's @@ -1769,54 +2254,84 @@ struct file *do_filp_open(int dfd, const char *pathname, if (open_flag & O_APPEND) acc_mode |= MAY_APPEND; - /* find the parent */ -reval: - error = path_init(dfd, pathname, LOOKUP_PARENT, &nd); + flags = LOOKUP_OPEN; + if (open_flag & O_CREAT) { + flags |= LOOKUP_CREATE; + if (open_flag & O_EXCL) + flags |= LOOKUP_EXCL; + } + if (open_flag & O_DIRECTORY) + flags |= LOOKUP_DIRECTORY; + if (!(open_flag & O_NOFOLLOW)) + flags |= LOOKUP_FOLLOW; + + filp = get_empty_filp(); + if (!filp) + return ERR_PTR(-ENFILE); + + filp->f_flags = open_flag; + nd.intent.open.file = filp; + nd.intent.open.flags = flag; + nd.intent.open.create_mode = mode; + + if (open_flag & O_CREAT) + goto creat; + + /* !O_CREAT, simple open */ + error = do_path_lookup(dfd, pathname, flags, &nd); + if (unlikely(error)) + goto out_filp; + error = -ELOOP; + if (!(nd.flags & LOOKUP_FOLLOW)) { + if (nd.inode->i_op->follow_link) + goto out_path; + } + error = -ENOTDIR; + if (nd.flags & LOOKUP_DIRECTORY) { + if (!nd.inode->i_op->lookup) + goto out_path; + } + audit_inode(pathname, nd.path.dentry); + filp = finish_open(&nd, open_flag, acc_mode); + return filp; + +creat: + /* OK, have to create the file. Find the parent. */ + error = path_init_rcu(dfd, pathname, + LOOKUP_PARENT | (flags & LOOKUP_REVAL), &nd); if (error) - return ERR_PTR(error); - if (force_reval) - nd.flags |= LOOKUP_REVAL; + goto out_filp; + error = path_walk_rcu(pathname, &nd); + path_finish_rcu(&nd); + if (unlikely(error == -ECHILD || error == -ESTALE)) { + /* slower, locked walk */ + if (error == -ESTALE) { +reval: + flags |= LOOKUP_REVAL; + } + error = path_init(dfd, pathname, + LOOKUP_PARENT | (flags & LOOKUP_REVAL), &nd); + if (error) + goto out_filp; - current->total_link_count = 0; - error = link_path_walk(pathname, &nd); - if (error) { - filp = ERR_PTR(error); - goto out; + error = path_walk_simple(pathname, &nd); } - if (unlikely(!audit_dummy_context()) && (open_flag & O_CREAT)) + if (unlikely(error)) + goto out_filp; + if (unlikely(!audit_dummy_context())) audit_inode(pathname, nd.path.dentry); /* * We have the parent and last component. */ - - error = -ENFILE; - filp = get_empty_filp(); - if (filp == NULL) - goto exit_parent; - nd.intent.open.file = filp; - filp->f_flags = open_flag; - nd.intent.open.flags = flag; - nd.intent.open.create_mode = mode; - nd.flags &= ~LOOKUP_PARENT; - nd.flags |= LOOKUP_OPEN; - if (open_flag & O_CREAT) { - nd.flags |= LOOKUP_CREATE; - if (open_flag & O_EXCL) - nd.flags |= LOOKUP_EXCL; - } - if (open_flag & O_DIRECTORY) - nd.flags |= LOOKUP_DIRECTORY; - if (!(open_flag & O_NOFOLLOW)) - nd.flags |= LOOKUP_FOLLOW; + nd.flags = flags; filp = do_last(&nd, &path, open_flag, acc_mode, mode, pathname); while (unlikely(!filp)) { /* trailing symlink */ struct path holder; - struct inode *inode = path.dentry->d_inode; void *cookie; error = -ELOOP; /* S_ISDIR part is a temporary automount kludge */ - if (!(nd.flags & LOOKUP_FOLLOW) && !S_ISDIR(inode->i_mode)) + if (!(nd.flags & LOOKUP_FOLLOW) && !S_ISDIR(nd.inode->i_mode)) goto exit_dput; if (count++ == 32) goto exit_dput; @@ -1837,36 +2352,33 @@ reval: goto exit_dput; error = __do_follow_link(&path, &nd, &cookie); if (unlikely(error)) { + if (!IS_ERR(cookie) && nd.inode->i_op->put_link) + nd.inode->i_op->put_link(path.dentry, &nd, cookie); /* nd.path had been dropped */ - if (!IS_ERR(cookie) && inode->i_op->put_link) - inode->i_op->put_link(path.dentry, &nd, cookie); - path_put(&path); - release_open_intent(&nd); - filp = ERR_PTR(error); - goto out; + nd.path = path; + goto out_path; } holder = path; nd.flags &= ~LOOKUP_PARENT; filp = do_last(&nd, &path, open_flag, acc_mode, mode, pathname); - if (inode->i_op->put_link) - inode->i_op->put_link(holder.dentry, &nd, cookie); + if (nd.inode->i_op->put_link) + nd.inode->i_op->put_link(holder.dentry, &nd, cookie); path_put(&holder); } out: if (nd.root.mnt) path_put(&nd.root); - if (filp == ERR_PTR(-ESTALE) && !force_reval) { - force_reval = 1; + if (filp == ERR_PTR(-ESTALE) && !(flags & LOOKUP_REVAL)) goto reval; - } return filp; exit_dput: path_put_conditional(&path, &nd); +out_path: + path_put(&nd.path); +out_filp: if (!IS_ERR(nd.intent.open.file)) release_open_intent(&nd); -exit_parent: - path_put(&nd.path); filp = ERR_PTR(error); goto out; } @@ -2127,12 +2639,10 @@ void dentry_unhash(struct dentry *dentry) { dget(dentry); shrink_dcache_parent(dentry); - spin_lock(&dcache_lock); spin_lock(&dentry->d_lock); - if (atomic_read(&dentry->d_count) == 2) + if (dentry->d_count == 2) __d_drop(dentry); spin_unlock(&dentry->d_lock); - spin_unlock(&dcache_lock); } int vfs_rmdir(struct inode *dir, struct dentry *dentry) diff --git a/fs/namespace.c b/fs/namespace.c index 8a415c9c5e5..3ddfd9046c4 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -13,7 +13,6 @@ #include <linux/sched.h> #include <linux/spinlock.h> #include <linux/percpu.h> -#include <linux/smp_lock.h> #include <linux/init.h> #include <linux/kernel.h> #include <linux/acct.h> @@ -139,6 +138,64 @@ void mnt_release_group_id(struct vfsmount *mnt) mnt->mnt_group_id = 0; } +/* + * vfsmount lock must be held for read + */ +static inline void mnt_add_count(struct vfsmount *mnt, int n) +{ +#ifdef CONFIG_SMP + this_cpu_add(mnt->mnt_pcp->mnt_count, n); +#else + preempt_disable(); + mnt->mnt_count += n; + preempt_enable(); +#endif +} + +static inline void mnt_set_count(struct vfsmount *mnt, int n) +{ +#ifdef CONFIG_SMP + this_cpu_write(mnt->mnt_pcp->mnt_count, n); +#else + mnt->mnt_count = n; +#endif +} + +/* + * vfsmount lock must be held for read + */ +static inline void mnt_inc_count(struct vfsmount *mnt) +{ + mnt_add_count(mnt, 1); +} + +/* + * vfsmount lock must be held for read + */ +static inline void mnt_dec_count(struct vfsmount *mnt) +{ + mnt_add_count(mnt, -1); +} + +/* + * vfsmount lock must be held for write + */ +unsigned int mnt_get_count(struct vfsmount *mnt) +{ +#ifdef CONFIG_SMP + unsigned int count = atomic_read(&mnt->mnt_longrefs); + int cpu; + + for_each_possible_cpu(cpu) { + count += per_cpu_ptr(mnt->mnt_pcp, cpu)->mnt_count; + } + + return count; +#else + return mnt->mnt_count; +#endif +} + struct vfsmount *alloc_vfsmnt(const char *name) { struct vfsmount *mnt = kmem_cache_zalloc(mnt_cache, GFP_KERNEL); @@ -155,7 +212,17 @@ struct vfsmount *alloc_vfsmnt(const char *name) goto out_free_id; } - atomic_set(&mnt->mnt_count, 1); +#ifdef CONFIG_SMP + mnt->mnt_pcp = alloc_percpu(struct mnt_pcp); + if (!mnt->mnt_pcp) + goto out_free_devname; + + atomic_set(&mnt->mnt_longrefs, 1); +#else + mnt->mnt_count = 1; + mnt->mnt_writers = 0; +#endif + INIT_LIST_HEAD(&mnt->mnt_hash); INIT_LIST_HEAD(&mnt->mnt_child); INIT_LIST_HEAD(&mnt->mnt_mounts); @@ -167,13 +234,6 @@ struct vfsmount *alloc_vfsmnt(const char *name) #ifdef CONFIG_FSNOTIFY INIT_HLIST_HEAD(&mnt->mnt_fsnotify_marks); #endif -#ifdef CONFIG_SMP - mnt->mnt_writers = alloc_percpu(int); - if (!mnt->mnt_writers) - goto out_free_devname; -#else - mnt->mnt_writers = 0; -#endif } return mnt; @@ -217,32 +277,32 @@ int __mnt_is_readonly(struct vfsmount *mnt) } EXPORT_SYMBOL_GPL(__mnt_is_readonly); -static inline void inc_mnt_writers(struct vfsmount *mnt) +static inline void mnt_inc_writers(struct vfsmount *mnt) { #ifdef CONFIG_SMP - (*per_cpu_ptr(mnt->mnt_writers, smp_processor_id()))++; + this_cpu_inc(mnt->mnt_pcp->mnt_writers); #else mnt->mnt_writers++; #endif } -static inline void dec_mnt_writers(struct vfsmount *mnt) +static inline void mnt_dec_writers(struct vfsmount *mnt) { #ifdef CONFIG_SMP - (*per_cpu_ptr(mnt->mnt_writers, smp_processor_id()))--; + this_cpu_dec(mnt->mnt_pcp->mnt_writers); #else mnt->mnt_writers--; #endif } -static unsigned int count_mnt_writers(struct vfsmount *mnt) +static unsigned int mnt_get_writers(struct vfsmount *mnt) { #ifdef CONFIG_SMP unsigned int count = 0; int cpu; for_each_possible_cpu(cpu) { - count += *per_cpu_ptr(mnt->mnt_writers, cpu); + count += per_cpu_ptr(mnt->mnt_pcp, cpu)->mnt_writers; } return count; @@ -274,9 +334,9 @@ int mnt_want_write(struct vfsmount *mnt) int ret = 0; preempt_disable(); - inc_mnt_writers(mnt); + mnt_inc_writers(mnt); /* - * The store to inc_mnt_writers must be visible before we pass + * The store to mnt_inc_writers must be visible before we pass * MNT_WRITE_HOLD loop below, so that the slowpath can see our * incremented count after it has set MNT_WRITE_HOLD. */ @@ -290,7 +350,7 @@ int mnt_want_write(struct vfsmount *mnt) */ smp_rmb(); if (__mnt_is_readonly(mnt)) { - dec_mnt_writers(mnt); + mnt_dec_writers(mnt); ret = -EROFS; goto out; } @@ -318,7 +378,7 @@ int mnt_clone_write(struct vfsmount *mnt) if (__mnt_is_readonly(mnt)) return -EROFS; preempt_disable(); - inc_mnt_writers(mnt); + mnt_inc_writers(mnt); preempt_enable(); return 0; } @@ -352,7 +412,7 @@ EXPORT_SYMBOL_GPL(mnt_want_write_file); void mnt_drop_write(struct vfsmount *mnt) { preempt_disable(); - dec_mnt_writers(mnt); + mnt_dec_writers(mnt); preempt_enable(); } EXPORT_SYMBOL_GPL(mnt_drop_write); @@ -385,7 +445,7 @@ static int mnt_make_readonly(struct vfsmount *mnt) * MNT_WRITE_HOLD, so it can't be decremented by another CPU while * we're counting up here. */ - if (count_mnt_writers(mnt) > 0) + if (mnt_get_writers(mnt) > 0) ret = -EBUSY; else mnt->mnt_flags |= MNT_READONLY; @@ -419,7 +479,7 @@ void free_vfsmnt(struct vfsmount *mnt) kfree(mnt->mnt_devname); mnt_free_id(mnt); #ifdef CONFIG_SMP - free_percpu(mnt->mnt_writers); + free_percpu(mnt->mnt_pcp); #endif kmem_cache_free(mnt_cache, mnt); } @@ -493,6 +553,27 @@ static void __touch_mnt_namespace(struct mnt_namespace *ns) } /* + * Clear dentry's mounted state if it has no remaining mounts. + * vfsmount_lock must be held for write. + */ +static void dentry_reset_mounted(struct vfsmount *mnt, struct dentry *dentry) +{ + unsigned u; + + for (u = 0; u < HASH_SIZE; u++) { + struct vfsmount *p; + + list_for_each_entry(p, &mount_hashtable[u], mnt_hash) { + if (p->mnt_mountpoint == dentry) + return; + } + } + spin_lock(&dentry->d_lock); + dentry->d_flags &= ~DCACHE_MOUNTED; + spin_unlock(&dentry->d_lock); +} + +/* * vfsmount lock must be held for write */ static void detach_mnt(struct vfsmount *mnt, struct path *old_path) @@ -503,7 +584,7 @@ static void detach_mnt(struct vfsmount *mnt, struct path *old_path) mnt->mnt_mountpoint = mnt->mnt_root; list_del_init(&mnt->mnt_child); list_del_init(&mnt->mnt_hash); - old_path->dentry->d_mounted--; + dentry_reset_mounted(old_path->mnt, old_path->dentry); } /* @@ -514,7 +595,9 @@ void mnt_set_mountpoint(struct vfsmount *mnt, struct dentry *dentry, { child_mnt->mnt_parent = mntget(mnt); child_mnt->mnt_mountpoint = dget(dentry); - dentry->d_mounted++; + spin_lock(&dentry->d_lock); + dentry->d_flags |= DCACHE_MOUNTED; + spin_unlock(&dentry->d_lock); } /* @@ -630,9 +713,10 @@ static struct vfsmount *clone_mnt(struct vfsmount *old, struct dentry *root, return NULL; } -static inline void __mntput(struct vfsmount *mnt) +static inline void mntfree(struct vfsmount *mnt) { struct super_block *sb = mnt->mnt_sb; + /* * This probably indicates that somebody messed * up a mnt_want/drop_write() pair. If this @@ -640,38 +724,123 @@ static inline void __mntput(struct vfsmount *mnt) * to make r/w->r/o transitions. */ /* - * atomic_dec_and_lock() used to deal with ->mnt_count decrements - * provides barriers, so count_mnt_writers() below is safe. AV + * The locking used to deal with mnt_count decrement provides barriers, + * so mnt_get_writers() below is safe. */ - WARN_ON(count_mnt_writers(mnt)); + WARN_ON(mnt_get_writers(mnt)); fsnotify_vfsmount_delete(mnt); dput(mnt->mnt_root); free_vfsmnt(mnt); deactivate_super(sb); } -void mntput_no_expire(struct vfsmount *mnt) -{ -repeat: - if (atomic_add_unless(&mnt->mnt_count, -1, 1)) - return; +#ifdef CONFIG_SMP +static inline void __mntput(struct vfsmount *mnt, int longrefs) +{ + if (!longrefs) { +put_again: + br_read_lock(vfsmount_lock); + if (likely(atomic_read(&mnt->mnt_longrefs))) { + mnt_dec_count(mnt); + br_read_unlock(vfsmount_lock); + return; + } + br_read_unlock(vfsmount_lock); + } else { + BUG_ON(!atomic_read(&mnt->mnt_longrefs)); + if (atomic_add_unless(&mnt->mnt_longrefs, -1, 1)) + return; + } + br_write_lock(vfsmount_lock); - if (!atomic_dec_and_test(&mnt->mnt_count)) { + if (!longrefs) + mnt_dec_count(mnt); + else + atomic_dec(&mnt->mnt_longrefs); + if (mnt_get_count(mnt)) { br_write_unlock(vfsmount_lock); return; } - if (likely(!mnt->mnt_pinned)) { + if (unlikely(mnt->mnt_pinned)) { + mnt_add_count(mnt, mnt->mnt_pinned + 1); + mnt->mnt_pinned = 0; br_write_unlock(vfsmount_lock); - __mntput(mnt); + acct_auto_close_mnt(mnt); + goto put_again; + } + br_write_unlock(vfsmount_lock); + mntfree(mnt); +} +#else +static inline void __mntput(struct vfsmount *mnt, int longrefs) +{ +put_again: + mnt_dec_count(mnt); + if (likely(mnt_get_count(mnt))) return; + br_write_lock(vfsmount_lock); + if (unlikely(mnt->mnt_pinned)) { + mnt_add_count(mnt, mnt->mnt_pinned + 1); + mnt->mnt_pinned = 0; + br_write_unlock(vfsmount_lock); + acct_auto_close_mnt(mnt); + goto put_again; } - atomic_add(mnt->mnt_pinned + 1, &mnt->mnt_count); - mnt->mnt_pinned = 0; br_write_unlock(vfsmount_lock); - acct_auto_close_mnt(mnt); - goto repeat; + mntfree(mnt); +} +#endif + +static void mntput_no_expire(struct vfsmount *mnt) +{ + __mntput(mnt, 0); +} + +void mntput(struct vfsmount *mnt) +{ + if (mnt) { + /* avoid cacheline pingpong, hope gcc doesn't get "smart" */ + if (unlikely(mnt->mnt_expiry_mark)) + mnt->mnt_expiry_mark = 0; + __mntput(mnt, 0); + } +} +EXPORT_SYMBOL(mntput); + +struct vfsmount *mntget(struct vfsmount *mnt) +{ + if (mnt) + mnt_inc_count(mnt); + return mnt; } -EXPORT_SYMBOL(mntput_no_expire); +EXPORT_SYMBOL(mntget); + +void mntput_long(struct vfsmount *mnt) +{ +#ifdef CONFIG_SMP + if (mnt) { + /* avoid cacheline pingpong, hope gcc doesn't get "smart" */ + if (unlikely(mnt->mnt_expiry_mark)) + mnt->mnt_expiry_mark = 0; + __mntput(mnt, 1); + } +#else + mntput(mnt); +#endif +} +EXPORT_SYMBOL(mntput_long); + +struct vfsmount *mntget_long(struct vfsmount *mnt) +{ +#ifdef CONFIG_SMP + if (mnt) + atomic_inc(&mnt->mnt_longrefs); + return mnt; +#else + return mntget(mnt); +#endif +} +EXPORT_SYMBOL(mntget_long); void mnt_pin(struct vfsmount *mnt) { @@ -679,19 +848,17 @@ void mnt_pin(struct vfsmount *mnt) mnt->mnt_pinned++; br_write_unlock(vfsmount_lock); } - EXPORT_SYMBOL(mnt_pin); void mnt_unpin(struct vfsmount *mnt) { br_write_lock(vfsmount_lock); if (mnt->mnt_pinned) { - atomic_inc(&mnt->mnt_count); + mnt_inc_count(mnt); mnt->mnt_pinned--; } br_write_unlock(vfsmount_lock); } - EXPORT_SYMBOL(mnt_unpin); static inline void mangle(struct seq_file *m, const char *s) @@ -986,12 +1153,13 @@ int may_umount_tree(struct vfsmount *mnt) int minimum_refs = 0; struct vfsmount *p; - br_read_lock(vfsmount_lock); + /* write lock needed for mnt_get_count */ + br_write_lock(vfsmount_lock); for (p = mnt; p; p = next_mnt(p, mnt)) { - actual_refs += atomic_read(&p->mnt_count); + actual_refs += mnt_get_count(p); minimum_refs += 2; } - br_read_unlock(vfsmount_lock); + br_write_unlock(vfsmount_lock); if (actual_refs > minimum_refs) return 0; @@ -1018,10 +1186,10 @@ int may_umount(struct vfsmount *mnt) { int ret = 1; down_read(&namespace_sem); - br_read_lock(vfsmount_lock); + br_write_lock(vfsmount_lock); if (propagate_mount_busy(mnt, 2)) ret = 0; - br_read_unlock(vfsmount_lock); + br_write_unlock(vfsmount_lock); up_read(&namespace_sem); return ret; } @@ -1048,7 +1216,7 @@ void release_mounts(struct list_head *head) dput(dentry); mntput(m); } - mntput(mnt); + mntput_long(mnt); } } @@ -1074,7 +1242,7 @@ void umount_tree(struct vfsmount *mnt, int propagate, struct list_head *kill) list_del_init(&p->mnt_child); if (p->mnt_parent != p) { p->mnt_parent->mnt_ghosts++; - p->mnt_mountpoint->d_mounted--; + dentry_reset_mounted(p->mnt_parent, p->mnt_mountpoint); } change_mnt_propagation(p, MS_PRIVATE); } @@ -1103,8 +1271,16 @@ static int do_umount(struct vfsmount *mnt, int flags) flags & (MNT_FORCE | MNT_DETACH)) return -EINVAL; - if (atomic_read(&mnt->mnt_count) != 2) + /* + * probably don't strictly need the lock here if we examined + * all race cases, but it's a slowpath. + */ + br_write_lock(vfsmount_lock); + if (mnt_get_count(mnt) != 2) { + br_write_lock(vfsmount_lock); return -EBUSY; + } + br_write_unlock(vfsmount_lock); if (!xchg(&mnt->mnt_expiry_mark, 1)) return -EAGAIN; @@ -1793,7 +1969,7 @@ int do_add_mount(struct vfsmount *newmnt, struct path *path, unlock: up_write(&namespace_sem); - mntput(newmnt); + mntput_long(newmnt); return err; } @@ -2126,11 +2302,11 @@ static struct mnt_namespace *dup_mnt_ns(struct mnt_namespace *mnt_ns, if (fs) { if (p == fs->root.mnt) { rootmnt = p; - fs->root.mnt = mntget(q); + fs->root.mnt = mntget_long(q); } if (p == fs->pwd.mnt) { pwdmnt = p; - fs->pwd.mnt = mntget(q); + fs->pwd.mnt = mntget_long(q); } } p = next_mnt(p, mnt_ns->root); @@ -2139,9 +2315,9 @@ static struct mnt_namespace *dup_mnt_ns(struct mnt_namespace *mnt_ns, up_write(&namespace_sem); if (rootmnt) - mntput(rootmnt); + mntput_long(rootmnt); if (pwdmnt) - mntput(pwdmnt); + mntput_long(pwdmnt); return new_ns; } @@ -2328,6 +2504,7 @@ SYSCALL_DEFINE2(pivot_root, const char __user *, new_root, touch_mnt_namespace(current->nsproxy->mnt_ns); br_write_unlock(vfsmount_lock); chroot_fs_refs(&root, &new); + error = 0; path_put(&root_parent); path_put(&parent_path); @@ -2354,6 +2531,7 @@ static void __init init_mount_tree(void) mnt = do_kern_mount("rootfs", 0, "rootfs", NULL); if (IS_ERR(mnt)) panic("Can't create rootfs"); + ns = create_mnt_ns(mnt); if (IS_ERR(ns)) panic("Can't allocate initial namespace"); diff --git a/fs/ncpfs/dir.c b/fs/ncpfs/dir.c index aac8832e919..28f136d4aae 100644 --- a/fs/ncpfs/dir.c +++ b/fs/ncpfs/dir.c @@ -17,9 +17,9 @@ #include <linux/kernel.h> #include <linux/vmalloc.h> #include <linux/mm.h> +#include <linux/namei.h> #include <asm/uaccess.h> #include <asm/byteorder.h> -#include <linux/smp_lock.h> #include <linux/ncp_fs.h> @@ -75,9 +75,12 @@ const struct inode_operations ncp_dir_inode_operations = * Dentry operations routines */ static int ncp_lookup_validate(struct dentry *, struct nameidata *); -static int ncp_hash_dentry(struct dentry *, struct qstr *); -static int ncp_compare_dentry (struct dentry *, struct qstr *, struct qstr *); -static int ncp_delete_dentry(struct dentry *); +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 *, + unsigned int, const char *, const struct qstr *); +static int ncp_delete_dentry(const struct dentry *); static const struct dentry_operations ncp_dentry_operations = { @@ -114,10 +117,10 @@ static inline int ncp_preserve_entry_case(struct inode *i, __u32 nscreator) #define ncp_preserve_case(i) (ncp_namespace(i) != NW_NS_DOS) -static inline int ncp_case_sensitive(struct dentry *dentry) +static inline int ncp_case_sensitive(const struct inode *i) { #ifdef CONFIG_NCPFS_NFS_NS - return ncp_namespace(dentry->d_inode) == NW_NS_NFS; + return ncp_namespace(i) == NW_NS_NFS; #else return 0; #endif /* CONFIG_NCPFS_NFS_NS */ @@ -128,14 +131,16 @@ static inline int ncp_case_sensitive(struct dentry *dentry) * is case-sensitive. */ static int -ncp_hash_dentry(struct dentry *dentry, struct qstr *this) +ncp_hash_dentry(const struct dentry *dentry, const struct inode *inode, + struct qstr *this) { - if (!ncp_case_sensitive(dentry)) { + if (!ncp_case_sensitive(inode)) { + struct super_block *sb = dentry->d_sb; struct nls_table *t; unsigned long hash; int i; - t = NCP_IO_TABLE(dentry); + t = NCP_IO_TABLE(sb); hash = init_name_hash(); for (i=0; i<this->len ; i++) hash = partial_name_hash(ncp_tolower(t, this->name[i]), @@ -146,15 +151,17 @@ ncp_hash_dentry(struct dentry *dentry, struct qstr *this) } static int -ncp_compare_dentry(struct dentry *dentry, struct qstr *a, struct qstr *b) +ncp_compare_dentry(const struct dentry *parent, const struct inode *pinode, + const struct dentry *dentry, const struct inode *inode, + unsigned int len, const char *str, const struct qstr *name) { - if (a->len != b->len) + if (len != name->len) return 1; - if (ncp_case_sensitive(dentry)) - return strncmp(a->name, b->name, a->len); + if (ncp_case_sensitive(pinode)) + return strncmp(str, name->name, len); - return ncp_strnicmp(NCP_IO_TABLE(dentry), a->name, b->name, a->len); + return ncp_strnicmp(NCP_IO_TABLE(pinode->i_sb), str, name->name, len); } /* @@ -163,7 +170,7 @@ ncp_compare_dentry(struct dentry *dentry, struct qstr *a, struct qstr *b) * Closing files can be safely postponed until iput() - it's done there anyway. */ static int -ncp_delete_dentry(struct dentry * dentry) +ncp_delete_dentry(const struct dentry * dentry) { struct inode *inode = dentry->d_inode; @@ -302,6 +309,9 @@ ncp_lookup_validate(struct dentry *dentry, struct nameidata *nd) int res, val = 0, len; __u8 __name[NCP_MAXPATHLEN + 1]; + if (nd->flags & LOOKUP_RCU) + return -ECHILD; + parent = dget_parent(dentry); dir = parent->d_inode; @@ -385,21 +395,21 @@ ncp_dget_fpos(struct dentry *dentry, struct dentry *parent, unsigned long fpos) } /* If a pointer is invalid, we search the dentry. */ - spin_lock(&dcache_lock); + spin_lock(&parent->d_lock); next = parent->d_subdirs.next; while (next != &parent->d_subdirs) { dent = list_entry(next, struct dentry, d_u.d_child); if ((unsigned long)dent->d_fsdata == fpos) { if (dent->d_inode) - dget_locked(dent); + dget(dent); else dent = NULL; - spin_unlock(&dcache_lock); + spin_unlock(&parent->d_lock); goto out; } next = next->next; } - spin_unlock(&dcache_lock); + spin_unlock(&parent->d_lock); return NULL; out: @@ -593,7 +603,7 @@ ncp_fill_cache(struct file *filp, void *dirent, filldir_t filldir, qname.hash = full_name_hash(qname.name, qname.len); if (dentry->d_op && dentry->d_op->d_hash) - if (dentry->d_op->d_hash(dentry, &qname) != 0) + if (dentry->d_op->d_hash(dentry, dentry->d_inode, &qname) != 0) goto end_advance; newdent = d_lookup(dentry, &qname); @@ -612,35 +622,12 @@ ncp_fill_cache(struct file *filp, void *dirent, filldir_t filldir, shrink_dcache_parent(newdent); /* - * It is not as dangerous as it looks. NetWare's OS2 namespace is - * case preserving yet case insensitive. So we update dentry's name - * as received from server. We found dentry via d_lookup with our - * hash, so we know that hash does not change, and so replacing name - * should be reasonably safe. + * NetWare's OS2 namespace is case preserving yet case + * insensitive. So we update dentry's name as received from + * server. Parent dir's i_mutex is locked because we're in + * readdir. */ - if (qname.len == newdent->d_name.len && - memcmp(newdent->d_name.name, qname.name, newdent->d_name.len)) { - struct inode *inode = newdent->d_inode; - - /* - * Inside ncpfs all uses of d_name are either for debugging, - * or on functions which acquire inode mutex (mknod, creat, - * lookup). So grab i_mutex here, to be sure. d_path - * uses dcache_lock when generating path, so we should too. - * And finally d_compare is protected by dentry's d_lock, so - * here we go. - */ - if (inode) - mutex_lock(&inode->i_mutex); - spin_lock(&dcache_lock); - spin_lock(&newdent->d_lock); - memcpy((char *) newdent->d_name.name, qname.name, - newdent->d_name.len); - spin_unlock(&newdent->d_lock); - spin_unlock(&dcache_lock); - if (inode) - mutex_unlock(&inode->i_mutex); - } + dentry_update_name_case(newdent, &qname); } if (!newdent->d_inode) { @@ -650,7 +637,7 @@ ncp_fill_cache(struct file *filp, void *dirent, filldir_t filldir, entry->ino = iunique(dir->i_sb, 2); inode = ncp_iget(dir->i_sb, entry); if (inode) { - newdent->d_op = &ncp_dentry_operations; + d_set_d_op(newdent, &ncp_dentry_operations); d_instantiate(newdent, inode); if (!hashed) d_rehash(newdent); @@ -658,7 +645,7 @@ ncp_fill_cache(struct file *filp, void *dirent, filldir_t filldir, } else { struct inode *inode = newdent->d_inode; - mutex_lock(&inode->i_mutex); + mutex_lock_nested(&inode->i_mutex, I_MUTEX_CHILD); ncp_update_inode2(inode, entry); mutex_unlock(&inode->i_mutex); } @@ -906,7 +893,7 @@ static struct dentry *ncp_lookup(struct inode *dir, struct dentry *dentry, struc if (inode) { ncp_new_dentry(dentry); add_entry: - dentry->d_op = &ncp_dentry_operations; + d_set_d_op(dentry, &ncp_dentry_operations); d_add(dentry, inode); error = 0; } diff --git a/fs/ncpfs/file.c b/fs/ncpfs/file.c index 6c754f70c52..cb50aaf981d 100644 --- a/fs/ncpfs/file.c +++ b/fs/ncpfs/file.c @@ -17,7 +17,6 @@ #include <linux/mm.h> #include <linux/vmalloc.h> #include <linux/sched.h> -#include <linux/smp_lock.h> #include <linux/ncp_fs.h> #include "ncplib_kernel.h" diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c index d290545aa0c..9531c052d7a 100644 --- a/fs/ncpfs/inode.c +++ b/fs/ncpfs/inode.c @@ -26,10 +26,10 @@ #include <linux/slab.h> #include <linux/vmalloc.h> #include <linux/init.h> -#include <linux/smp_lock.h> #include <linux/vfs.h> #include <linux/mount.h> #include <linux/seq_file.h> +#include <linux/namei.h> #include <linux/ncp_fs.h> @@ -59,11 +59,18 @@ static struct inode *ncp_alloc_inode(struct super_block *sb) return &ei->vfs_inode; } -static void ncp_destroy_inode(struct inode *inode) +static void ncp_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(ncp_inode_cachep, NCP_FINFO(inode)); } +static void ncp_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, ncp_i_callback); +} + static void init_once(void *foo) { struct ncp_inode_info *ei = (struct ncp_inode_info *) foo; @@ -711,7 +718,7 @@ static int ncp_fill_super(struct super_block *sb, void *raw_data, int silent) sb->s_root = d_alloc_root(root_inode); if (!sb->s_root) goto out_no_root; - sb->s_root->d_op = &ncp_root_dentry_operations; + d_set_d_op(sb->s_root, &ncp_root_dentry_operations); return 0; out_no_root: diff --git a/fs/ncpfs/ioctl.c b/fs/ncpfs/ioctl.c index c2a1f9a155c..d40a547e337 100644 --- a/fs/ncpfs/ioctl.c +++ b/fs/ncpfs/ioctl.c @@ -17,7 +17,6 @@ #include <linux/mount.h> #include <linux/slab.h> #include <linux/highuid.h> -#include <linux/smp_lock.h> #include <linux/vmalloc.h> #include <linux/sched.h> diff --git a/fs/ncpfs/ncplib_kernel.h b/fs/ncpfs/ncplib_kernel.h index 3c57eca634c..1220df75ff2 100644 --- a/fs/ncpfs/ncplib_kernel.h +++ b/fs/ncpfs/ncplib_kernel.h @@ -135,7 +135,7 @@ int ncp__vol2io(struct ncp_server *, unsigned char *, unsigned int *, const unsigned char *, unsigned int, int); #define NCP_ESC ':' -#define NCP_IO_TABLE(dentry) (NCP_SERVER((dentry)->d_inode)->nls_io) +#define NCP_IO_TABLE(sb) (NCP_SBP(sb)->nls_io) #define ncp_tolower(t, c) nls_tolower(t, c) #define ncp_toupper(t, c) nls_toupper(t, c) #define ncp_strnicmp(t, s1, s2, len) \ @@ -150,15 +150,15 @@ int ncp__io2vol(unsigned char *, unsigned int *, int ncp__vol2io(unsigned char *, unsigned int *, const unsigned char *, unsigned int, int); -#define NCP_IO_TABLE(dentry) NULL +#define NCP_IO_TABLE(sb) NULL #define ncp_tolower(t, c) tolower(c) #define ncp_toupper(t, c) toupper(c) #define ncp_io2vol(S,m,i,n,k,U) ncp__io2vol(m,i,n,k,U) #define ncp_vol2io(S,m,i,n,k,U) ncp__vol2io(m,i,n,k,U) -static inline int ncp_strnicmp(struct nls_table *t, const unsigned char *s1, - const unsigned char *s2, int len) +static inline int ncp_strnicmp(const struct nls_table *t, + const unsigned char *s1, const unsigned char *s2, int len) { while (len--) { if (tolower(*s1++) != tolower(*s2++)) @@ -193,7 +193,7 @@ ncp_renew_dentries(struct dentry *parent) struct list_head *next; struct dentry *dentry; - spin_lock(&dcache_lock); + spin_lock(&parent->d_lock); next = parent->d_subdirs.next; while (next != &parent->d_subdirs) { dentry = list_entry(next, struct dentry, d_u.d_child); @@ -205,7 +205,7 @@ ncp_renew_dentries(struct dentry *parent) next = next->next; } - spin_unlock(&dcache_lock); + spin_unlock(&parent->d_lock); } static inline void @@ -215,7 +215,7 @@ ncp_invalidate_dircache_entries(struct dentry *parent) struct list_head *next; struct dentry *dentry; - spin_lock(&dcache_lock); + spin_lock(&parent->d_lock); next = parent->d_subdirs.next; while (next != &parent->d_subdirs) { dentry = list_entry(next, struct dentry, d_u.d_child); @@ -223,7 +223,7 @@ ncp_invalidate_dircache_entries(struct dentry *parent) ncp_age_dentry(server, dentry); next = next->next; } - spin_unlock(&dcache_lock); + spin_unlock(&parent->d_lock); } struct ncp_cache_head { diff --git a/fs/nfs/callback.c b/fs/nfs/callback.c index aeec017fe81..93a8b3bd69e 100644 --- a/fs/nfs/callback.c +++ b/fs/nfs/callback.c @@ -9,7 +9,6 @@ #include <linux/completion.h> #include <linux/ip.h> #include <linux/module.h> -#include <linux/smp_lock.h> #include <linux/sunrpc/svc.h> #include <linux/sunrpc/svcsock.h> #include <linux/nfs_fs.h> diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c index 232a7eead33..1fd62fc49be 100644 --- a/fs/nfs/delegation.c +++ b/fs/nfs/delegation.c @@ -11,7 +11,6 @@ #include <linux/module.h> #include <linux/sched.h> #include <linux/slab.h> -#include <linux/smp_lock.h> #include <linux/spinlock.h> #include <linux/nfs4.h> diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 07ac3847e56..d33da530097 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -34,6 +34,7 @@ #include <linux/mount.h> #include <linux/sched.h> #include <linux/vmalloc.h> +#include <linux/kmemleak.h> #include "delegation.h" #include "iostat.h" @@ -56,7 +57,7 @@ static int nfs_rename(struct inode *, struct dentry *, struct inode *, struct dentry *); static int nfs_fsync_dir(struct file *, int); static loff_t nfs_llseek_dir(struct file *, loff_t, int); -static int nfs_readdir_clear_array(struct page*, gfp_t); +static void nfs_readdir_clear_array(struct page*); const struct file_operations nfs_dir_operations = { .llseek = nfs_llseek_dir, @@ -82,8 +83,8 @@ const struct inode_operations nfs_dir_inode_operations = { .setattr = nfs_setattr, }; -const struct address_space_operations nfs_dir_addr_space_ops = { - .releasepage = nfs_readdir_clear_array, +const struct address_space_operations nfs_dir_aops = { + .freepage = nfs_readdir_clear_array, }; #ifdef CONFIG_NFS_V3 @@ -161,6 +162,7 @@ struct nfs_cache_array_entry { u64 cookie; u64 ino; struct qstr string; + unsigned char d_type; }; struct nfs_cache_array { @@ -170,14 +172,13 @@ struct nfs_cache_array { struct nfs_cache_array_entry array[0]; }; -#define MAX_READDIR_ARRAY ((PAGE_SIZE - sizeof(struct nfs_cache_array)) / sizeof(struct nfs_cache_array_entry)) - typedef __be32 * (*decode_dirent_t)(struct xdr_stream *, struct nfs_entry *, struct nfs_server *, int); typedef struct { struct file *file; struct page *page; unsigned long page_index; u64 *dir_cookie; + u64 last_cookie; loff_t current_index; decode_dirent_t decode; @@ -194,9 +195,13 @@ typedef struct { static struct nfs_cache_array *nfs_readdir_get_array(struct page *page) { + void *ptr; if (page == NULL) return ERR_PTR(-EIO); - return (struct nfs_cache_array *)kmap(page); + ptr = kmap(page); + if (ptr == NULL) + return ERR_PTR(-ENOMEM); + return ptr; } static @@ -209,14 +214,15 @@ void nfs_readdir_release_array(struct page *page) * we are freeing strings created by nfs_add_to_readdir_array() */ static -int nfs_readdir_clear_array(struct page *page, gfp_t mask) +void nfs_readdir_clear_array(struct page *page) { - struct nfs_cache_array *array = nfs_readdir_get_array(page); + struct nfs_cache_array *array; int i; + + array = kmap_atomic(page, KM_USER0); for (i = 0; i < array->size; i++) kfree(array->array[i].string.name); - nfs_readdir_release_array(page); - return 0; + kunmap_atomic(array, KM_USER0); } /* @@ -231,6 +237,11 @@ int nfs_readdir_make_qstr(struct qstr *string, const char *name, unsigned int le string->name = kmemdup(name, len, GFP_KERNEL); if (string->name == NULL) return -ENOMEM; + /* + * Avoid a kmemleak false positive. The pointer to the name is stored + * in a page cache page which kmemleak does not scan. + */ + kmemleak_not_leak(string->name); string->hash = full_name_hash(name, len); return 0; } @@ -244,20 +255,24 @@ int nfs_readdir_add_to_array(struct nfs_entry *entry, struct page *page) if (IS_ERR(array)) return PTR_ERR(array); - ret = -EIO; - if (array->size >= MAX_READDIR_ARRAY) - goto out; cache_entry = &array->array[array->size]; + + /* Check that this entry lies within the page bounds */ + ret = -ENOSPC; + if ((char *)&cache_entry[1] - (char *)page_address(page) > PAGE_SIZE) + goto out; + cache_entry->cookie = entry->prev_cookie; cache_entry->ino = entry->ino; + cache_entry->d_type = entry->d_type; ret = nfs_readdir_make_qstr(&cache_entry->string, entry->name, entry->len); if (ret) goto out; array->last_cookie = entry->cookie; - if (entry->eof == 1) - array->eof_index = array->size; array->size++; + if (entry->eof != 0) + array->eof_index = array->size; out: nfs_readdir_release_array(page); return ret; @@ -272,7 +287,7 @@ int nfs_readdir_search_for_pos(struct nfs_cache_array *array, nfs_readdir_descri if (diff < 0) goto out_eof; if (diff >= array->size) { - if (array->eof_index > 0) + if (array->eof_index >= 0) goto out_eof; desc->current_index += array->size; return -EAGAIN; @@ -281,8 +296,6 @@ int nfs_readdir_search_for_pos(struct nfs_cache_array *array, nfs_readdir_descri index = (unsigned int)diff; *desc->dir_cookie = array->array[index].cookie; desc->cache_entry_index = index; - if (index == array->eof_index) - desc->eof = 1; return 0; out_eof: desc->eof = 1; @@ -296,17 +309,16 @@ int nfs_readdir_search_for_cookie(struct nfs_cache_array *array, nfs_readdir_des int status = -EAGAIN; for (i = 0; i < array->size; i++) { - if (i == array->eof_index) { - desc->eof = 1; - status = -EBADCOOKIE; - } if (array->array[i].cookie == *desc->dir_cookie) { desc->cache_entry_index = i; - status = 0; - break; + return 0; } } - + if (array->eof_index >= 0) { + status = -EBADCOOKIE; + if (*desc->dir_cookie == array->last_cookie) + desc->eof = 1; + } return status; } @@ -314,10 +326,7 @@ static int nfs_readdir_search_array(nfs_readdir_descriptor_t *desc) { struct nfs_cache_array *array; - int status = -EBADCOOKIE; - - if (desc->dir_cookie == NULL) - goto out; + int status; array = nfs_readdir_get_array(desc->page); if (IS_ERR(array)) { @@ -330,6 +339,10 @@ int nfs_readdir_search_array(nfs_readdir_descriptor_t *desc) else status = nfs_readdir_search_for_cookie(array, desc); + if (status == -EAGAIN) { + desc->last_cookie = array->last_cookie; + desc->page_index++; + } nfs_readdir_release_array(desc->page); out: return status; @@ -381,13 +394,9 @@ int xdr_decode(nfs_readdir_descriptor_t *desc, struct nfs_entry *entry, struct x static int nfs_same_file(struct dentry *dentry, struct nfs_entry *entry) { - struct nfs_inode *node; if (dentry->d_inode == NULL) goto different; - node = NFS_I(dentry->d_inode); - if (node->fh.size != entry->fh->size) - goto different; - if (strncmp(node->fh.data, entry->fh->data, node->fh.size) != 0) + if (nfs_compare_fh(entry->fh, NFS_FH(dentry->d_inode)) != 0) goto different; return 1; different: @@ -429,7 +438,7 @@ void nfs_prime_dcache(struct dentry *parent, struct nfs_entry *entry) if (dentry == NULL) return; - dentry->d_op = NFS_PROTO(dir)->dentry_ops; + d_set_d_op(dentry, NFS_PROTO(dir)->dentry_ops); inode = nfs_fhget(dentry->d_sb, entry->fh, entry->fattr); if (IS_ERR(inode)) goto out; @@ -449,14 +458,15 @@ out: /* Perform conversion from xdr to cache array */ static -void nfs_readdir_page_filler(nfs_readdir_descriptor_t *desc, struct nfs_entry *entry, +int nfs_readdir_page_filler(nfs_readdir_descriptor_t *desc, struct nfs_entry *entry, void *xdr_page, struct page *page, unsigned int buflen) { struct xdr_stream stream; struct xdr_buf buf; __be32 *ptr = xdr_page; - int status; struct nfs_cache_array *array; + unsigned int count = 0; + int status; buf.head->iov_base = xdr_page; buf.head->iov_len = buflen; @@ -471,21 +481,32 @@ void nfs_readdir_page_filler(nfs_readdir_descriptor_t *desc, struct nfs_entry *e do { status = xdr_decode(desc, entry, &stream); - if (status != 0) + if (status != 0) { + if (status == -EAGAIN) + status = 0; break; + } - if (nfs_readdir_add_to_array(entry, page) == -1) - break; - if (desc->plus == 1) + count++; + + if (desc->plus != 0) nfs_prime_dcache(desc->file->f_path.dentry, entry); + + status = nfs_readdir_add_to_array(entry, page); + if (status != 0) + break; } while (!entry->eof); - if (status == -EBADCOOKIE && entry->eof) { + if (count == 0 || (status == -EBADCOOKIE && entry->eof != 0)) { array = nfs_readdir_get_array(page); - array->eof_index = array->size - 1; - status = 0; - nfs_readdir_release_array(page); + if (!IS_ERR(array)) { + array->eof_index = array->size; + status = 0; + nfs_readdir_release_array(page); + } else + status = PTR_ERR(array); } + return status; } static @@ -537,11 +558,11 @@ int nfs_readdir_xdr_to_array(nfs_readdir_descriptor_t *desc, struct page *page, struct nfs_entry entry; struct file *file = desc->file; struct nfs_cache_array *array; - int status = 0; + int status = -ENOMEM; unsigned int array_size = ARRAY_SIZE(pages); entry.prev_cookie = 0; - entry.cookie = *desc->dir_cookie; + entry.cookie = desc->last_cookie; entry.eof = 0; entry.fh = nfs_alloc_fhandle(); entry.fattr = nfs_alloc_fattr(); @@ -549,6 +570,10 @@ int nfs_readdir_xdr_to_array(nfs_readdir_descriptor_t *desc, struct page *page, goto out; array = nfs_readdir_get_array(page); + if (IS_ERR(array)) { + status = PTR_ERR(array); + goto out; + } memset(array, 0, sizeof(struct nfs_cache_array)); array->eof_index = -1; @@ -556,12 +581,19 @@ int nfs_readdir_xdr_to_array(nfs_readdir_descriptor_t *desc, struct page *page, if (!pages_ptr) goto out_release_array; do { + unsigned int pglen; status = nfs_readdir_xdr_filler(pages, desc, &entry, file, inode); if (status < 0) break; - nfs_readdir_page_filler(desc, &entry, pages_ptr, page, array_size * PAGE_SIZE); - } while (array->eof_index < 0 && array->size < MAX_READDIR_ARRAY); + pglen = status; + status = nfs_readdir_page_filler(desc, &entry, pages_ptr, page, pglen); + if (status < 0) { + if (status == -ENOSPC) + status = 0; + break; + } + } while (array->eof_index < 0); nfs_readdir_free_large_page(pages_ptr, pages, array_size); out_release_array: @@ -582,8 +614,10 @@ static int nfs_readdir_filler(nfs_readdir_descriptor_t *desc, struct page* page) { struct inode *inode = desc->file->f_path.dentry->d_inode; + int ret; - if (nfs_readdir_xdr_to_array(desc, page, inode) < 0) + ret = nfs_readdir_xdr_to_array(desc, page, inode); + if (ret < 0) goto error; SetPageUptodate(page); @@ -595,12 +629,14 @@ int nfs_readdir_filler(nfs_readdir_descriptor_t *desc, struct page* page) return 0; error: unlock_page(page); - return -EIO; + return ret; } static void cache_page_release(nfs_readdir_descriptor_t *desc) { + if (!desc->page->mapping) + nfs_readdir_clear_array(desc->page); page_cache_release(desc->page); desc->page = NULL; } @@ -608,12 +644,8 @@ void cache_page_release(nfs_readdir_descriptor_t *desc) static struct page *get_cache_page(nfs_readdir_descriptor_t *desc) { - struct page *page; - page = read_cache_page(desc->file->f_path.dentry->d_inode->i_mapping, + return read_cache_page(desc->file->f_path.dentry->d_inode->i_mapping, desc->page_index, (filler_t *)nfs_readdir_filler, desc); - if (IS_ERR(page)) - desc->eof = 1; - return page; } /* @@ -629,9 +661,8 @@ int find_cache_page(nfs_readdir_descriptor_t *desc) return PTR_ERR(desc->page); res = nfs_readdir_search_array(desc); - if (res == 0) - return 0; - cache_page_release(desc); + if (res != 0) + cache_page_release(desc); return res; } @@ -639,22 +670,18 @@ int find_cache_page(nfs_readdir_descriptor_t *desc) static inline int readdir_search_pagecache(nfs_readdir_descriptor_t *desc) { - int res = -EAGAIN; + int res; - while (1) { - res = find_cache_page(desc); - if (res != -EAGAIN) - break; - desc->page_index++; + if (desc->page_index == 0) { + desc->current_index = 0; + desc->last_cookie = 0; } + do { + res = find_cache_page(desc); + } while (res == -EAGAIN); return res; } -static inline unsigned int dt_type(struct inode *inode) -{ - return (inode->i_mode >> 12) & 15; -} - /* * Once we've found the start of the dirent within a page: fill 'er up... */ @@ -666,35 +693,35 @@ int nfs_do_filldir(nfs_readdir_descriptor_t *desc, void *dirent, int i = 0; int res = 0; struct nfs_cache_array *array = NULL; - unsigned int d_type = DT_UNKNOWN; - struct dentry *dentry = NULL; array = nfs_readdir_get_array(desc->page); + if (IS_ERR(array)) { + res = PTR_ERR(array); + goto out; + } for (i = desc->cache_entry_index; i < array->size; i++) { - d_type = DT_UNKNOWN; + struct nfs_cache_array_entry *ent; - res = filldir(dirent, array->array[i].string.name, - array->array[i].string.len, file->f_pos, - nfs_compat_user_ino64(array->array[i].ino), d_type); - if (res < 0) + ent = &array->array[i]; + if (filldir(dirent, ent->string.name, ent->string.len, + file->f_pos, nfs_compat_user_ino64(ent->ino), + ent->d_type) < 0) { + desc->eof = 1; break; + } file->f_pos++; - desc->cache_entry_index = i; if (i < (array->size-1)) *desc->dir_cookie = array->array[i+1].cookie; else *desc->dir_cookie = array->last_cookie; - if (i == array->eof_index) { - desc->eof = 1; - break; - } } + if (array->eof_index >= 0) + desc->eof = 1; nfs_readdir_release_array(desc->page); +out: cache_page_release(desc); - if (dentry != NULL) - dput(dentry); dfprintk(DIRCACHE, "NFS: nfs_do_filldir() filling ended @ cookie %Lu; returning = %d\n", (unsigned long long)*desc->dir_cookie, res); return res; @@ -729,13 +756,14 @@ int uncached_readdir(nfs_readdir_descriptor_t *desc, void *dirent, goto out; } - if (nfs_readdir_xdr_to_array(desc, page, inode) == -1) { - status = -EIO; - goto out_release; - } - desc->page_index = 0; + desc->last_cookie = *desc->dir_cookie; desc->page = page; + + status = nfs_readdir_xdr_to_array(desc, page, inode); + if (status < 0) + goto out_release; + status = nfs_do_filldir(desc, dirent, filldir); out: @@ -757,7 +785,7 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir) struct inode *inode = dentry->d_inode; nfs_readdir_descriptor_t my_desc, *desc = &my_desc; - int res = -ENOMEM; + int res; dfprintk(FILE, "NFS: readdir(%s/%s) starting at cookie %llu\n", dentry->d_parent->d_name.name, dentry->d_name.name, @@ -782,18 +810,18 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir) if (res < 0) goto out; - while (desc->eof != 1) { + do { res = readdir_search_pagecache(desc); if (res == -EBADCOOKIE) { + res = 0; /* This means either end of directory */ if (*desc->dir_cookie && desc->eof == 0) { /* Or that the server has 'lost' a cookie */ res = uncached_readdir(desc, dirent, filldir); - if (res >= 0) + if (res == 0) continue; } - res = 0; break; } if (res == -ETOOSMALL && desc->plus) { @@ -808,11 +836,9 @@ static int nfs_readdir(struct file *filp, void *dirent, filldir_t filldir) break; res = nfs_do_filldir(desc, dirent, filldir); - if (res < 0) { - res = 0; + if (res < 0) break; - } - } + } while (!desc->eof); out: nfs_unblock_sillyrename(dentry); if (res > 0) @@ -912,7 +938,8 @@ static int nfs_check_verifier(struct inode *dir, struct dentry *dentry) * component of the path. * We check for this using LOOKUP_CONTINUE and LOOKUP_PARENT. */ -static inline unsigned int nfs_lookup_check_intent(struct nameidata *nd, unsigned int mask) +static inline unsigned int nfs_lookup_check_intent(struct nameidata *nd, + unsigned int mask) { if (nd->flags & (LOOKUP_CONTINUE|LOOKUP_PARENT)) return 0; @@ -992,7 +1019,7 @@ int nfs_neg_need_reval(struct inode *dir, struct dentry *dentry, * If the parent directory is seen to have changed, we throw out the * cached dentry and do a new lookup. */ -static int nfs_lookup_revalidate(struct dentry * dentry, struct nameidata *nd) +static int nfs_lookup_revalidate(struct dentry *dentry, struct nameidata *nd) { struct inode *dir; struct inode *inode; @@ -1001,6 +1028,9 @@ static int nfs_lookup_revalidate(struct dentry * dentry, struct nameidata *nd) struct nfs_fattr *fattr = NULL; int error; + if (nd->flags & LOOKUP_RCU) + return -ECHILD; + parent = dget_parent(dentry); dir = parent->d_inode; nfs_inc_stats(dir, NFSIOS_DENTRYREVALIDATE); @@ -1091,7 +1121,7 @@ out_error: /* * This is called from dput() when d_count is going to 0. */ -static int nfs_dentry_delete(struct dentry *dentry) +static int nfs_dentry_delete(const struct dentry *dentry) { dfprintk(VFS, "NFS: dentry_delete(%s/%s, %x)\n", dentry->d_parent->d_name.name, dentry->d_name.name, @@ -1162,7 +1192,7 @@ static struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, stru if (dentry->d_name.len > NFS_SERVER(dir)->namelen) goto out; - dentry->d_op = NFS_PROTO(dir)->dentry_ops; + d_set_d_op(dentry, NFS_PROTO(dir)->dentry_ops); /* * If we're doing an exclusive create, optimize away the lookup @@ -1307,7 +1337,7 @@ static struct dentry *nfs_atomic_lookup(struct inode *dir, struct dentry *dentry res = ERR_PTR(-ENAMETOOLONG); goto out; } - dentry->d_op = NFS_PROTO(dir)->dentry_ops; + d_set_d_op(dentry, NFS_PROTO(dir)->dentry_ops); /* Let vfs_create() deal with O_EXCL. Instantiate, but don't hash * the dentry. */ @@ -1345,12 +1375,12 @@ static struct dentry *nfs_atomic_lookup(struct inode *dir, struct dentry *dentry res = NULL; goto out; /* This turned out not to be a regular file */ - case -EISDIR: case -ENOTDIR: goto no_open; case -ELOOP: if (!(nd->intent.open.flags & O_NOFOLLOW)) goto no_open; + /* case -EISDIR: */ /* case -EINVAL: */ default: res = ERR_CAST(inode); @@ -1692,11 +1722,9 @@ static int nfs_unlink(struct inode *dir, struct dentry *dentry) dfprintk(VFS, "NFS: unlink(%s/%ld, %s)\n", dir->i_sb->s_id, dir->i_ino, dentry->d_name.name); - spin_lock(&dcache_lock); spin_lock(&dentry->d_lock); - if (atomic_read(&dentry->d_count) > 1) { + if (dentry->d_count > 1) { spin_unlock(&dentry->d_lock); - spin_unlock(&dcache_lock); /* Start asynchronous writeout of the inode */ write_inode_now(dentry->d_inode, 0); error = nfs_sillyrename(dir, dentry); @@ -1707,7 +1735,6 @@ static int nfs_unlink(struct inode *dir, struct dentry *dentry) need_rehash = 1; } spin_unlock(&dentry->d_lock); - spin_unlock(&dcache_lock); error = nfs_safe_remove(dentry); if (!error || error == -ENOENT) { nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); @@ -1842,7 +1869,7 @@ static int nfs_rename(struct inode *old_dir, struct dentry *old_dentry, dfprintk(VFS, "NFS: rename(%s/%s -> %s/%s, ct=%d)\n", old_dentry->d_parent->d_name.name, old_dentry->d_name.name, new_dentry->d_parent->d_name.name, new_dentry->d_name.name, - atomic_read(&new_dentry->d_count)); + new_dentry->d_count); /* * For non-directories, check whether the target is busy and if so, @@ -1860,7 +1887,7 @@ static int nfs_rename(struct inode *old_dir, struct dentry *old_dentry, rehash = new_dentry; } - if (atomic_read(&new_dentry->d_count) > 2) { + if (new_dentry->d_count > 2) { int err; /* copy the target dentry's name */ @@ -2162,11 +2189,14 @@ int nfs_may_open(struct inode *inode, struct rpc_cred *cred, int openflags) return nfs_do_access(inode, cred, nfs_open_permission_mask(openflags)); } -int nfs_permission(struct inode *inode, int mask) +int nfs_permission(struct inode *inode, int mask, unsigned int flags) { struct rpc_cred *cred; int res = 0; + if (flags & IPERM_FLAG_RCU) + return -ECHILD; + nfs_inc_stats(inode, NFSIOS_VFSACCESS); if ((mask & (MAY_READ | MAY_WRITE | MAY_EXEC)) == 0) @@ -2214,7 +2244,7 @@ out: out_notsup: res = nfs_revalidate_inode(NFS_SERVER(inode), inode); if (res == 0) - res = generic_permission(inode, mask, NULL); + res = generic_permission(inode, mask, flags, NULL); goto out; } diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index 84d3c8b9020..e6ace0d93c7 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -867,7 +867,7 @@ static ssize_t nfs_direct_write(struct kiocb *iocb, const struct iovec *iov, goto out; nfs_alloc_commit_data(dreq); - if (dreq->commit_data == NULL || count < wsize) + if (dreq->commit_data == NULL || count <= wsize) sync = NFS_FILE_SYNC; dreq->inode = inode; diff --git a/fs/nfs/file.c b/fs/nfs/file.c index 60677f9f131..7bf029ef408 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -693,6 +693,7 @@ do_getlk(struct file *filp, int cmd, struct file_lock *fl, int is_local) { struct inode *inode = filp->f_mapping->host; int status = 0; + unsigned int saved_type = fl->fl_type; /* Try local locking first */ posix_test_lock(filp, fl); @@ -700,6 +701,7 @@ do_getlk(struct file *filp, int cmd, struct file_lock *fl, int is_local) /* found a conflict */ goto out; } + fl->fl_type = saved_type; if (nfs_have_delegation(inode, FMODE_READ)) goto out_noconflict; diff --git a/fs/nfs/getroot.c b/fs/nfs/getroot.c index ac7b814ce16..5596c6a2881 100644 --- a/fs/nfs/getroot.c +++ b/fs/nfs/getroot.c @@ -63,9 +63,11 @@ static int nfs_superblock_set_dummy_root(struct super_block *sb, struct inode *i * This again causes shrink_dcache_for_umount_subtree() to * Oops, since the test for IS_ROOT() will fail. */ - spin_lock(&dcache_lock); + spin_lock(&sb->s_root->d_inode->i_lock); + spin_lock(&sb->s_root->d_lock); list_del_init(&sb->s_root->d_alias); - spin_unlock(&dcache_lock); + spin_unlock(&sb->s_root->d_lock); + spin_unlock(&sb->s_root->d_inode->i_lock); } return 0; } @@ -119,7 +121,7 @@ struct dentry *nfs_get_root(struct super_block *sb, struct nfs_fh *mntfh) security_d_instantiate(ret, inode); if (ret->d_op == NULL) - ret->d_op = server->nfs_client->rpc_ops->dentry_ops; + d_set_d_op(ret, server->nfs_client->rpc_ops->dentry_ops); out: nfs_free_fattr(fsinfo.fattr); return ret; @@ -226,7 +228,7 @@ struct dentry *nfs4_get_root(struct super_block *sb, struct nfs_fh *mntfh) security_d_instantiate(ret, inode); if (ret->d_op == NULL) - ret->d_op = server->nfs_client->rpc_ops->dentry_ops; + d_set_d_op(ret, server->nfs_client->rpc_ops->dentry_ops); out: nfs_free_fattr(fattr); diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 314f5716460..017daa3bed3 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -289,6 +289,7 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr) } else if (S_ISDIR(inode->i_mode)) { inode->i_op = NFS_SB(sb)->nfs_client->rpc_ops->dir_inode_ops; inode->i_fop = &nfs_dir_operations; + inode->i_data.a_ops = &nfs_dir_aops; if (nfs_server_capable(inode, NFS_CAP_READDIRPLUS)) set_bit(NFS_INO_ADVISE_RDPLUS, &NFS_I(inode)->flags); /* Deal with crossing mountpoints */ @@ -1437,11 +1438,18 @@ struct inode *nfs_alloc_inode(struct super_block *sb) return &nfsi->vfs_inode; } -void nfs_destroy_inode(struct inode *inode) +static void nfs_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(nfs_inode_cachep, NFS_I(inode)); } +void nfs_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, nfs_i_callback); +} + static inline void nfs4_init_once(struct nfs_inode *nfsi) { #ifdef CONFIG_NFS_V4 diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index db08ff3ff45..e6356b750b7 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -362,6 +362,15 @@ unsigned int nfs_page_length(struct page *page) } /* + * Convert a umode to a dirent->d_type + */ +static inline +unsigned char nfs_umode_to_dtype(umode_t mode) +{ + return (mode >> 12) & 15; +} + +/* * Determine the number of pages in an array of length 'len' and * with a base offset of 'base' */ diff --git a/fs/nfs/mount_clnt.c b/fs/nfs/mount_clnt.c index eceafe74f47..4f981f1f668 100644 --- a/fs/nfs/mount_clnt.c +++ b/fs/nfs/mount_clnt.c @@ -505,13 +505,13 @@ static struct rpc_procinfo mnt3_procedures[] = { static struct rpc_version mnt_version1 = { .number = 1, - .nrprocs = 2, + .nrprocs = ARRAY_SIZE(mnt_procedures), .procs = mnt_procedures, }; static struct rpc_version mnt_version3 = { .number = 3, - .nrprocs = 2, + .nrprocs = ARRAY_SIZE(mnt3_procedures), .procs = mnt3_procedures, }; diff --git a/fs/nfs/namespace.c b/fs/nfs/namespace.c index db6aa3673cf..74aaf3963c1 100644 --- a/fs/nfs/namespace.c +++ b/fs/nfs/namespace.c @@ -49,12 +49,17 @@ char *nfs_path(const char *base, const struct dentry *dentry, char *buffer, ssize_t buflen) { - char *end = buffer+buflen; + char *end; int namelen; + unsigned seq; +rename_retry: + end = buffer+buflen; *--end = '\0'; buflen--; - spin_lock(&dcache_lock); + + seq = read_seqbegin(&rename_lock); + rcu_read_lock(); while (!IS_ROOT(dentry) && dentry != droot) { namelen = dentry->d_name.len; buflen -= namelen + 1; @@ -65,7 +70,9 @@ char *nfs_path(const char *base, *--end = '/'; dentry = dentry->d_parent; } - spin_unlock(&dcache_lock); + rcu_read_unlock(); + if (read_seqretry(&rename_lock, seq)) + goto rename_retry; if (*end != '/') { if (--buflen < 0) goto Elong; @@ -82,7 +89,9 @@ char *nfs_path(const char *base, memcpy(end, base, namelen); return end; Elong_unlock: - spin_unlock(&dcache_lock); + rcu_read_unlock(); + if (read_seqretry(&rename_lock, seq)) + goto rename_retry; Elong: return ERR_PTR(-ENAMETOOLONG); } diff --git a/fs/nfs/nfs2xdr.c b/fs/nfs/nfs2xdr.c index e6bf45710cc..5914a1911c9 100644 --- a/fs/nfs/nfs2xdr.c +++ b/fs/nfs/nfs2xdr.c @@ -423,7 +423,7 @@ nfs_xdr_readdirres(struct rpc_rqst *req, __be32 *p, void *dummy) struct page **page; size_t hdrlen; unsigned int pglen, recvd; - int status, nr = 0; + int status; if ((status = ntohl(*p++))) return nfs_stat_to_errno(status); @@ -443,7 +443,7 @@ nfs_xdr_readdirres(struct rpc_rqst *req, __be32 *p, void *dummy) if (pglen > recvd) pglen = recvd; page = rcvbuf->pages; - return nr; + return pglen; } static void print_overflow_msg(const char *func, const struct xdr_stream *xdr) @@ -485,6 +485,8 @@ nfs_decode_dirent(struct xdr_stream *xdr, struct nfs_entry *entry, struct nfs_se entry->prev_cookie = entry->cookie; entry->cookie = ntohl(*p++); + entry->d_type = DT_UNKNOWN; + p = xdr_inline_peek(xdr, 8); if (p != NULL) entry->eof = !p[0] && p[1]; @@ -495,7 +497,7 @@ nfs_decode_dirent(struct xdr_stream *xdr, struct nfs_entry *entry, struct nfs_se out_overflow: print_overflow_msg(__func__, xdr); - return ERR_PTR(-EIO); + return ERR_PTR(-EAGAIN); } /* diff --git a/fs/nfs/nfs3xdr.c b/fs/nfs/nfs3xdr.c index d9a5e832c25..f6cc60f06da 100644 --- a/fs/nfs/nfs3xdr.c +++ b/fs/nfs/nfs3xdr.c @@ -555,7 +555,7 @@ nfs3_xdr_readdirres(struct rpc_rqst *req, __be32 *p, struct nfs3_readdirres *res struct page **page; size_t hdrlen; u32 recvd, pglen; - int status, nr = 0; + int status; status = ntohl(*p++); /* Decode post_op_attrs */ @@ -586,7 +586,7 @@ nfs3_xdr_readdirres(struct rpc_rqst *req, __be32 *p, struct nfs3_readdirres *res pglen = recvd; page = rcvbuf->pages; - return nr; + return pglen; } __be32 * @@ -622,11 +622,13 @@ nfs3_decode_dirent(struct xdr_stream *xdr, struct nfs_entry *entry, struct nfs_s entry->prev_cookie = entry->cookie; p = xdr_decode_hyper(p, &entry->cookie); + entry->d_type = DT_UNKNOWN; if (plus) { entry->fattr->valid = 0; p = xdr_decode_post_op_attr_stream(xdr, entry->fattr); if (IS_ERR(p)) goto out_overflow_exit; + entry->d_type = nfs_umode_to_dtype(entry->fattr->mode); /* In fact, a post_op_fh3: */ p = xdr_inline_decode(xdr, 4); if (unlikely(!p)) @@ -656,7 +658,7 @@ nfs3_decode_dirent(struct xdr_stream *xdr, struct nfs_entry *entry, struct nfs_s out_overflow: print_overflow_msg(__func__, xdr); out_overflow_exit: - return ERR_PTR(-EIO); + return ERR_PTR(-EAGAIN); } /* diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 0f24cdf2cb1..4435e5e1f90 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -2852,8 +2852,10 @@ static int _nfs4_proc_readdir(struct dentry *dentry, struct rpc_cred *cred, nfs4_setup_readdir(cookie, NFS_COOKIEVERF(dir), dentry, &args); res.pgbase = args.pgbase; status = nfs4_call_sync(NFS_SERVER(dir), &msg, &args, &res, 0); - if (status == 0) + if (status >= 0) { memcpy(NFS_COOKIEVERF(dir), res.verifier.data, NFS4_VERIFIER_SIZE); + status += args.pgbase; + } nfs_invalidate_atime(dir); @@ -3359,6 +3361,8 @@ static ssize_t nfs4_proc_get_acl(struct inode *inode, void *buf, size_t buflen) ret = nfs_revalidate_inode(server, inode); if (ret < 0) return ret; + if (NFS_I(inode)->cache_validity & NFS_INO_INVALID_ACL) + nfs_zap_acl_cache(inode); ret = nfs4_read_cached_acl(inode, buf, buflen); if (ret != -ENOENT) return ret; @@ -3387,6 +3391,13 @@ static int __nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t bufl nfs_inode_return_delegation(inode); buf_to_pages(buf, buflen, arg.acl_pages, &arg.acl_pgbase); ret = nfs4_call_sync(server, &msg, &arg, &res, 1); + /* + * Acl update can result in inode attribute update. + * so mark the attribute cache invalid. + */ + spin_lock(&inode->i_lock); + NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ATTR; + spin_unlock(&inode->i_lock); nfs_access_zap_cache(inode); nfs_zap_acl_cache(inode); return ret; diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index f313c4cce7e..9f1826b012e 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -4518,7 +4518,7 @@ static int decode_readdir(struct xdr_stream *xdr, struct rpc_rqst *req, struct n xdr_read_pages(xdr, pglen); - return 0; + return pglen; } static int decode_readlink(struct xdr_stream *xdr, struct rpc_rqst *req) @@ -6208,6 +6208,10 @@ __be32 *nfs4_decode_dirent(struct xdr_stream *xdr, struct nfs_entry *entry, if (entry->fattr->valid & NFS_ATTR_FATTR_FILEID) entry->ino = entry->fattr->fileid; + entry->d_type = DT_UNKNOWN; + if (entry->fattr->valid & NFS_ATTR_FATTR_TYPE) + entry->d_type = nfs_umode_to_dtype(entry->fattr->mode); + if (verify_attr_len(xdr, p, len) < 0) goto out_overflow; @@ -6221,7 +6225,7 @@ __be32 *nfs4_decode_dirent(struct xdr_stream *xdr, struct nfs_entry *entry, out_overflow: print_overflow_msg(__func__, xdr); - return ERR_PTR(-EIO); + return ERR_PTR(-EAGAIN); } /* diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index 137b549e63d..b68536cc904 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -115,7 +115,7 @@ int nfs_set_page_tag_locked(struct nfs_page *req) { if (!nfs_lock_request_dontget(req)) return 0; - if (req->wb_page != NULL) + if (test_bit(PG_MAPPED, &req->wb_flags)) radix_tree_tag_set(&NFS_I(req->wb_context->path.dentry->d_inode)->nfs_page_tree, req->wb_index, NFS_PAGE_TAG_LOCKED); return 1; } @@ -125,7 +125,7 @@ int nfs_set_page_tag_locked(struct nfs_page *req) */ void nfs_clear_page_tag_locked(struct nfs_page *req) { - if (req->wb_page != NULL) { + if (test_bit(PG_MAPPED, &req->wb_flags)) { struct inode *inode = req->wb_context->path.dentry->d_inode; struct nfs_inode *nfsi = NFS_I(inode); diff --git a/fs/nfs/read.c b/fs/nfs/read.c index e4b62c6f5a6..aedcaa7f291 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -152,7 +152,6 @@ static void nfs_readpage_release(struct nfs_page *req) (long long)NFS_FILEID(req->wb_context->path.dentry->d_inode), req->wb_bytes, (long long)req_offset(req)); - nfs_clear_request(req); nfs_release_request(req); } diff --git a/fs/nfs/super.c b/fs/nfs/super.c index 0a42e8f4adc..4100630c9a5 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -39,7 +39,6 @@ #include <linux/nfs_mount.h> #include <linux/nfs4_mount.h> #include <linux/lockd/bind.h> -#include <linux/smp_lock.h> #include <linux/seq_file.h> #include <linux/mount.h> #include <linux/mnt_namespace.h> @@ -67,6 +66,12 @@ #define NFSDBG_FACILITY NFSDBG_VFS +#ifdef CONFIG_NFS_V3 +#define NFS_DEFAULT_VERSION 3 +#else +#define NFS_DEFAULT_VERSION 2 +#endif + enum { /* Mount options that take no arguments */ Opt_soft, Opt_hard, @@ -1064,12 +1069,10 @@ static int nfs_parse_mount_options(char *raw, mnt->flags |= NFS_MOUNT_VER3; mnt->version = 3; break; -#ifdef CONFIG_NFS_V4 case Opt_v4: mnt->flags &= ~NFS_MOUNT_VER3; mnt->version = 4; break; -#endif case Opt_udp: mnt->flags &= ~NFS_MOUNT_TCP; mnt->nfs_server.protocol = XPRT_TRANSPORT_UDP; @@ -1281,12 +1284,10 @@ static int nfs_parse_mount_options(char *raw, mnt->flags |= NFS_MOUNT_VER3; mnt->version = 3; break; -#ifdef CONFIG_NFS_V4 case NFS4_VERSION: mnt->flags &= ~NFS_MOUNT_VER3; mnt->version = 4; break; -#endif default: goto out_invalid_value; } @@ -2277,7 +2278,7 @@ static int nfs_get_sb(struct file_system_type *fs_type, }; int error = -ENOMEM; - data = nfs_alloc_parsed_mount_data(3); + data = nfs_alloc_parsed_mount_data(NFS_DEFAULT_VERSION); mntfh = nfs_alloc_fhandle(); if (data == NULL || mntfh == NULL) goto out_free_fh; diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c index 7bdec853140..8fe9eb47a97 100644 --- a/fs/nfs/unlink.c +++ b/fs/nfs/unlink.c @@ -496,7 +496,7 @@ nfs_sillyrename(struct inode *dir, struct dentry *dentry) dfprintk(VFS, "NFS: silly-rename(%s/%s, ct=%d)\n", dentry->d_parent->d_name.name, dentry->d_name.name, - atomic_read(&dentry->d_count)); + dentry->d_count); nfs_inc_stats(dir, NFSIOS_SILLYRENAME); /* diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 4c14c17a527..10d648ea128 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -390,6 +390,7 @@ static int nfs_inode_add_request(struct inode *inode, struct nfs_page *req) if (nfs_have_delegation(inode, FMODE_WRITE)) nfsi->change_attr++; } + set_bit(PG_MAPPED, &req->wb_flags); SetPagePrivate(req->wb_page); set_page_private(req->wb_page, (unsigned long)req); nfsi->npages++; @@ -415,6 +416,7 @@ static void nfs_inode_remove_request(struct nfs_page *req) spin_lock(&inode->i_lock); set_page_private(req->wb_page, 0); ClearPagePrivate(req->wb_page); + clear_bit(PG_MAPPED, &req->wb_flags); radix_tree_delete(&nfsi->nfs_page_tree, req->wb_index); nfsi->npages--; if (!nfsi->npages) { @@ -422,7 +424,6 @@ static void nfs_inode_remove_request(struct nfs_page *req) iput(inode); } else spin_unlock(&inode->i_lock); - nfs_clear_request(req); nfs_release_request(req); } diff --git a/fs/nfsd/nfs3xdr.c b/fs/nfsd/nfs3xdr.c index 2a533a0af2a..7e84a852cda 100644 --- a/fs/nfsd/nfs3xdr.c +++ b/fs/nfsd/nfs3xdr.c @@ -260,9 +260,11 @@ void fill_post_wcc(struct svc_fh *fhp) err = vfs_getattr(fhp->fh_export->ex_path.mnt, fhp->fh_dentry, &fhp->fh_post_attr); fhp->fh_post_change = fhp->fh_dentry->d_inode->i_version; - if (err) + if (err) { fhp->fh_post_saved = 0; - else + /* Grab the ctime anyway - set_change_info might use it */ + fhp->fh_post_attr.ctime = fhp->fh_dentry->d_inode->i_ctime; + } else fhp->fh_post_saved = 1; } diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index f1e5ec6b510..116cab970e0 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -673,16 +673,17 @@ static void nfsd4_hash_conn(struct nfsd4_conn *conn, struct nfsd4_session *ses) spin_unlock(&clp->cl_lock); } -static void nfsd4_register_conn(struct nfsd4_conn *conn) +static int nfsd4_register_conn(struct nfsd4_conn *conn) { conn->cn_xpt_user.callback = nfsd4_conn_lost; - register_xpt_user(conn->cn_xprt, &conn->cn_xpt_user); + return register_xpt_user(conn->cn_xprt, &conn->cn_xpt_user); } static __be32 nfsd4_new_conn(struct svc_rqst *rqstp, struct nfsd4_session *ses) { struct nfsd4_conn *conn; u32 flags = NFS4_CDFC4_FORE; + int ret; if (ses->se_flags & SESSION4_BACK_CHAN) flags |= NFS4_CDFC4_BACK; @@ -690,7 +691,10 @@ static __be32 nfsd4_new_conn(struct svc_rqst *rqstp, struct nfsd4_session *ses) if (!conn) return nfserr_jukebox; nfsd4_hash_conn(conn, ses); - nfsd4_register_conn(conn); + ret = nfsd4_register_conn(conn); + if (ret) + /* oops; xprt is already down: */ + nfsd4_conn_lost(&conn->cn_xpt_user); return nfs_ok; } @@ -1644,6 +1648,7 @@ static void nfsd4_sequence_check_conn(struct nfsd4_conn *new, struct nfsd4_sessi { struct nfs4_client *clp = ses->se_client; struct nfsd4_conn *c; + int ret; spin_lock(&clp->cl_lock); c = __nfsd4_find_conn(new->cn_xprt, ses); @@ -1654,7 +1659,10 @@ static void nfsd4_sequence_check_conn(struct nfsd4_conn *new, struct nfsd4_sessi } __nfsd4_hash_conn(new, ses); spin_unlock(&clp->cl_lock); - nfsd4_register_conn(new); + ret = nfsd4_register_conn(new); + if (ret) + /* oops; xprt is already down: */ + nfsd4_conn_lost(&new->cn_xpt_user); return; } @@ -2254,7 +2262,7 @@ nfs4_file_downgrade(struct nfs4_file *fp, unsigned int share_access) * Spawn a thread to perform a recall on the delegation represented * by the lease (file_lock) * - * Called from break_lease() with lock_kernel() held. + * Called from break_lease() with lock_flocks() held. * Note: we assume break_lease will only call this *once* for any given * lease. */ @@ -2278,7 +2286,7 @@ void nfsd_break_deleg_cb(struct file_lock *fl) list_add_tail(&dp->dl_recall_lru, &del_recall_lru); spin_unlock(&recall_lock); - /* only place dl_time is set. protected by lock_kernel*/ + /* only place dl_time is set. protected by lock_flocks*/ dp->dl_time = get_seconds(); /* @@ -2295,7 +2303,7 @@ void nfsd_break_deleg_cb(struct file_lock *fl) /* * The file_lock is being reapd. * - * Called by locks_free_lock() with lock_kernel() held. + * Called by locks_free_lock() with lock_flocks() held. */ static void nfsd_release_deleg_cb(struct file_lock *fl) @@ -2310,7 +2318,7 @@ void nfsd_release_deleg_cb(struct file_lock *fl) } /* - * Called from setlease() with lock_kernel() held + * Called from setlease() with lock_flocks() held */ static int nfsd_same_client_deleg_cb(struct file_lock *onlist, struct file_lock *try) diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 184938fcff0..3a359023c9f 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -1756,8 +1756,7 @@ nfsd_rename(struct svc_rqst *rqstp, struct svc_fh *ffhp, char *fname, int flen, goto out_dput_new; if (svc_msnfs(ffhp) && - ((atomic_read(&odentry->d_count) > 1) - || (atomic_read(&ndentry->d_count) > 1))) { + ((odentry->d_count > 1) || (ndentry->d_count > 1))) { host_err = -EPERM; goto out_dput_new; } @@ -1843,7 +1842,7 @@ nfsd_unlink(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, if (type != S_IFDIR) { /* It's UNLINK */ #ifdef MSNFS if ((fhp->fh_export->ex_flags & NFSEXP_MSNFS) && - (atomic_read(&rdentry->d_count) > 1)) { + (rdentry->d_count > 1)) { host_err = -EPERM; } else #endif diff --git a/fs/nfsd/xdr4.h b/fs/nfsd/xdr4.h index 4d476ff08ae..60fce3dc5cb 100644 --- a/fs/nfsd/xdr4.h +++ b/fs/nfsd/xdr4.h @@ -484,18 +484,17 @@ static inline bool nfsd4_not_cached(struct nfsd4_compoundres *resp) static inline void set_change_info(struct nfsd4_change_info *cinfo, struct svc_fh *fhp) { - BUG_ON(!fhp->fh_pre_saved || !fhp->fh_post_saved); - cinfo->atomic = 1; + BUG_ON(!fhp->fh_pre_saved); + cinfo->atomic = fhp->fh_post_saved; cinfo->change_supported = IS_I_VERSION(fhp->fh_dentry->d_inode); - if (cinfo->change_supported) { - cinfo->before_change = fhp->fh_pre_change; - cinfo->after_change = fhp->fh_post_change; - } else { - cinfo->before_ctime_sec = fhp->fh_pre_ctime.tv_sec; - cinfo->before_ctime_nsec = fhp->fh_pre_ctime.tv_nsec; - cinfo->after_ctime_sec = fhp->fh_post_attr.ctime.tv_sec; - cinfo->after_ctime_nsec = fhp->fh_post_attr.ctime.tv_nsec; - } + + cinfo->before_change = fhp->fh_pre_change; + cinfo->after_change = fhp->fh_post_change; + cinfo->before_ctime_sec = fhp->fh_pre_ctime.tv_sec; + cinfo->before_ctime_nsec = fhp->fh_pre_ctime.tv_nsec; + cinfo->after_ctime_sec = fhp->fh_post_attr.ctime.tv_sec; + cinfo->after_ctime_nsec = fhp->fh_post_attr.ctime.tv_nsec; + } int nfs4svc_encode_voidres(struct svc_rqst *, __be32 *, void *); diff --git a/fs/nilfs2/dat.c b/fs/nilfs2/dat.c index 49c844dab33..59e5fe742f7 100644 --- a/fs/nilfs2/dat.c +++ b/fs/nilfs2/dat.c @@ -335,7 +335,7 @@ int nilfs_dat_move(struct inode *dat, __u64 vblocknr, sector_t blocknr) * the device at this point. * * To prevent nilfs_dat_translate() from returning the - * uncommited block number, this makes a copy of the entry + * uncommitted block number, this makes a copy of the entry * buffer and redirects nilfs_dat_translate() to the copy. */ if (!buffer_nilfs_redirected(entry_bh)) { diff --git a/fs/nilfs2/gcinode.c b/fs/nilfs2/gcinode.c index 33ad25ddd5c..caf9a6a3fb5 100644 --- a/fs/nilfs2/gcinode.c +++ b/fs/nilfs2/gcinode.c @@ -176,7 +176,6 @@ int nilfs_gccache_wait_and_mark_dirty(struct buffer_head *bh) int nilfs_init_gcinode(struct inode *inode) { struct nilfs_inode_info *ii = NILFS_I(inode); - struct the_nilfs *nilfs = NILFS_SB(inode->i_sb)->s_nilfs; inode->i_mode = S_IFREG; mapping_set_gfp_mask(inode->i_mapping, GFP_NOFS); @@ -186,14 +185,6 @@ int nilfs_init_gcinode(struct inode *inode) ii->i_flags = 0; nilfs_bmap_init_gc(ii->i_bmap); - /* - * Add the inode to GC inode list. Garbage Collection - * is serialized and no two processes manipulate the - * list simultaneously. - */ - igrab(inode); - list_add(&NILFS_I(inode)->i_dirty, &nilfs->ns_gc_inodes); - return 0; } diff --git a/fs/nilfs2/inode.c b/fs/nilfs2/inode.c index 71d4bc8464e..77b48c8fab1 100644 --- a/fs/nilfs2/inode.c +++ b/fs/nilfs2/inode.c @@ -785,15 +785,19 @@ out_err: return err; } -int nilfs_permission(struct inode *inode, int mask) +int nilfs_permission(struct inode *inode, int mask, unsigned int flags) { - struct nilfs_root *root = NILFS_I(inode)->i_root; + struct nilfs_root *root; + + if (flags & IPERM_FLAG_RCU) + return -ECHILD; + root = NILFS_I(inode)->i_root; if ((mask & MAY_WRITE) && root && root->cno != NILFS_CPTREE_CURRENT_CNO) return -EROFS; /* snapshot is not writable */ - return generic_permission(inode, mask, NULL); + return generic_permission(inode, mask, flags, NULL); } int nilfs_load_inode_block(struct nilfs_sb_info *sbi, struct inode *inode, diff --git a/fs/nilfs2/ioctl.c b/fs/nilfs2/ioctl.c index 3e90f86d5bf..b185e937a33 100644 --- a/fs/nilfs2/ioctl.c +++ b/fs/nilfs2/ioctl.c @@ -337,6 +337,7 @@ static int nilfs_ioctl_move_blocks(struct super_block *sb, struct nilfs_argv *argv, void *buf) { size_t nmembs = argv->v_nmembs; + struct the_nilfs *nilfs = NILFS_SB(sb)->s_nilfs; struct inode *inode; struct nilfs_vdesc *vdesc; struct buffer_head *bh, *n; @@ -349,10 +350,21 @@ static int nilfs_ioctl_move_blocks(struct super_block *sb, ino = vdesc->vd_ino; cno = vdesc->vd_cno; inode = nilfs_iget_for_gc(sb, ino, cno); - if (unlikely(inode == NULL)) { - ret = -ENOMEM; + if (IS_ERR(inode)) { + ret = PTR_ERR(inode); goto failed; } + if (list_empty(&NILFS_I(inode)->i_dirty)) { + /* + * Add the inode to GC inode list. Garbage Collection + * is serialized and no two processes manipulate the + * list simultaneously. + */ + igrab(inode); + list_add(&NILFS_I(inode)->i_dirty, + &nilfs->ns_gc_inodes); + } + do { ret = nilfs_ioctl_move_inode_block(inode, vdesc, &buffers); diff --git a/fs/nilfs2/nilfs.h b/fs/nilfs2/nilfs.h index f7560da5a56..0ca98823db5 100644 --- a/fs/nilfs2/nilfs.h +++ b/fs/nilfs2/nilfs.h @@ -256,7 +256,7 @@ extern void nilfs_update_inode(struct inode *, struct buffer_head *); extern void nilfs_truncate(struct inode *); extern void nilfs_evict_inode(struct inode *); extern int nilfs_setattr(struct dentry *, struct iattr *); -int nilfs_permission(struct inode *inode, int mask); +int nilfs_permission(struct inode *inode, int mask, unsigned int flags); extern int nilfs_load_inode_block(struct nilfs_sb_info *, struct inode *, struct buffer_head **); extern int nilfs_inode_dirty(struct inode *); diff --git a/fs/nilfs2/super.c b/fs/nilfs2/super.c index f804d41ec9d..e2dcc9c733f 100644 --- a/fs/nilfs2/super.c +++ b/fs/nilfs2/super.c @@ -162,10 +162,13 @@ struct inode *nilfs_alloc_inode(struct super_block *sb) return &ii->vfs_inode; } -void nilfs_destroy_inode(struct inode *inode) +static void nilfs_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); struct nilfs_mdt_info *mdi = NILFS_MDT(inode); + INIT_LIST_HEAD(&inode->i_dentry); + if (mdi) { kfree(mdi->mi_bgl); /* kfree(NULL) is safe */ kfree(mdi); @@ -173,6 +176,11 @@ void nilfs_destroy_inode(struct inode *inode) kmem_cache_free(nilfs_inode_cachep, NILFS_I(inode)); } +void nilfs_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, nilfs_i_callback); +} + static int nilfs_sync_super(struct nilfs_sb_info *sbi, int flag) { struct the_nilfs *nilfs = sbi->s_nilfs; @@ -838,7 +846,7 @@ static int nilfs_attach_snapshot(struct super_block *s, __u64 cno, static int nilfs_tree_was_touched(struct dentry *root_dentry) { - return atomic_read(&root_dentry->d_count) > 1; + return root_dentry->d_count > 1; } /** diff --git a/fs/notify/fanotify/fanotify.c b/fs/notify/fanotify/fanotify.c index b04f88eed09..f35794b97e8 100644 --- a/fs/notify/fanotify/fanotify.c +++ b/fs/notify/fanotify/fanotify.c @@ -92,7 +92,11 @@ static int fanotify_get_response_from_access(struct fsnotify_group *group, pr_debug("%s: group=%p event=%p\n", __func__, group, event); - wait_event(group->fanotify_data.access_waitq, event->response); + wait_event(group->fanotify_data.access_waitq, event->response || + atomic_read(&group->fanotify_data.bypass_perm)); + + if (!event->response) /* bypass_perm set */ + return 0; /* userspace responded, convert to something usable */ spin_lock(&event->lock); diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c index 063224812b7..8b61220cffc 100644 --- a/fs/notify/fanotify/fanotify_user.c +++ b/fs/notify/fanotify/fanotify_user.c @@ -106,20 +106,29 @@ static int create_fd(struct fsnotify_group *group, struct fsnotify_event *event) return client_fd; } -static ssize_t fill_event_metadata(struct fsnotify_group *group, +static int fill_event_metadata(struct fsnotify_group *group, struct fanotify_event_metadata *metadata, struct fsnotify_event *event) { + int ret = 0; + pr_debug("%s: group=%p metadata=%p event=%p\n", __func__, group, metadata, event); metadata->event_len = FAN_EVENT_METADATA_LEN; + metadata->metadata_len = FAN_EVENT_METADATA_LEN; metadata->vers = FANOTIFY_METADATA_VERSION; metadata->mask = event->mask & FAN_ALL_OUTGOING_EVENTS; metadata->pid = pid_vnr(event->tgid); - metadata->fd = create_fd(group, event); + if (unlikely(event->mask & FAN_Q_OVERFLOW)) + metadata->fd = FAN_NOFD; + else { + metadata->fd = create_fd(group, event); + if (metadata->fd < 0) + ret = metadata->fd; + } - return metadata->fd; + return ret; } #ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS @@ -200,7 +209,7 @@ static int prepare_for_access_response(struct fsnotify_group *group, mutex_lock(&group->fanotify_data.access_mutex); - if (group->fanotify_data.bypass_perm) { + if (atomic_read(&group->fanotify_data.bypass_perm)) { mutex_unlock(&group->fanotify_data.access_mutex); kmem_cache_free(fanotify_response_event_cache, re); event->response = FAN_ALLOW; @@ -257,24 +266,34 @@ static ssize_t copy_event_to_user(struct fsnotify_group *group, pr_debug("%s: group=%p event=%p\n", __func__, group, event); - fd = fill_event_metadata(group, &fanotify_event_metadata, event); - if (fd < 0) - return fd; + ret = fill_event_metadata(group, &fanotify_event_metadata, event); + if (ret < 0) + goto out; + fd = fanotify_event_metadata.fd; ret = prepare_for_access_response(group, event, fd); if (ret) goto out_close_fd; ret = -EFAULT; - if (copy_to_user(buf, &fanotify_event_metadata, FAN_EVENT_METADATA_LEN)) + if (copy_to_user(buf, &fanotify_event_metadata, + fanotify_event_metadata.event_len)) goto out_kill_access_response; - return FAN_EVENT_METADATA_LEN; + return fanotify_event_metadata.event_len; out_kill_access_response: remove_access_response(group, event, fd); out_close_fd: - sys_close(fd); + if (fd != FAN_NOFD) + sys_close(fd); +out: +#ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS + if (event->mask & FAN_ALL_PERM_EVENTS) { + event->response = FAN_DENY; + wake_up(&group->fanotify_data.access_waitq); + } +#endif return ret; } @@ -382,7 +401,7 @@ static int fanotify_release(struct inode *ignored, struct file *file) mutex_lock(&group->fanotify_data.access_mutex); - group->fanotify_data.bypass_perm = true; + atomic_inc(&group->fanotify_data.bypass_perm); list_for_each_entry_safe(re, lre, &group->fanotify_data.access_list, list) { pr_debug("%s: found group=%p re=%p event=%p\n", __func__, group, @@ -586,11 +605,10 @@ static int fanotify_add_vfsmount_mark(struct fsnotify_group *group, { struct fsnotify_mark *fsn_mark; __u32 added; + int ret = 0; fsn_mark = fsnotify_find_vfsmount_mark(group, mnt); if (!fsn_mark) { - int ret; - if (atomic_read(&group->num_marks) > group->fanotify_data.max_marks) return -ENOSPC; @@ -600,17 +618,16 @@ static int fanotify_add_vfsmount_mark(struct fsnotify_group *group, fsnotify_init_mark(fsn_mark, fanotify_free_mark); ret = fsnotify_add_mark(fsn_mark, group, NULL, mnt, 0); - if (ret) { - fanotify_free_mark(fsn_mark); - return ret; - } + if (ret) + goto err; } added = fanotify_mark_add_to_mask(fsn_mark, mask, flags); - fsnotify_put_mark(fsn_mark); + if (added & ~mnt->mnt_fsnotify_mask) fsnotify_recalc_vfsmount_mask(mnt); - - return 0; +err: + fsnotify_put_mark(fsn_mark); + return ret; } static int fanotify_add_inode_mark(struct fsnotify_group *group, @@ -619,6 +636,7 @@ static int fanotify_add_inode_mark(struct fsnotify_group *group, { struct fsnotify_mark *fsn_mark; __u32 added; + int ret = 0; pr_debug("%s: group=%p inode=%p\n", __func__, group, inode); @@ -634,8 +652,6 @@ static int fanotify_add_inode_mark(struct fsnotify_group *group, fsn_mark = fsnotify_find_inode_mark(group, inode); if (!fsn_mark) { - int ret; - if (atomic_read(&group->num_marks) > group->fanotify_data.max_marks) return -ENOSPC; @@ -645,16 +661,16 @@ static int fanotify_add_inode_mark(struct fsnotify_group *group, fsnotify_init_mark(fsn_mark, fanotify_free_mark); ret = fsnotify_add_mark(fsn_mark, group, inode, NULL, 0); - if (ret) { - fanotify_free_mark(fsn_mark); - return ret; - } + if (ret) + goto err; } added = fanotify_mark_add_to_mask(fsn_mark, mask, flags); - fsnotify_put_mark(fsn_mark); + if (added & ~inode->i_fsnotify_mask) fsnotify_recalc_inode_mask(inode); - return 0; +err: + fsnotify_put_mark(fsn_mark); + return ret; } /* fanotify syscalls */ @@ -687,8 +703,10 @@ SYSCALL_DEFINE2(fanotify_init, unsigned int, flags, unsigned int, event_f_flags) /* fsnotify_alloc_group takes a ref. Dropped in fanotify_release */ group = fsnotify_alloc_group(&fanotify_fsnotify_ops); - if (IS_ERR(group)) + if (IS_ERR(group)) { + free_uid(user); return PTR_ERR(group); + } group->fanotify_data.user = user; atomic_inc(&user->fanotify_listeners); @@ -698,6 +716,7 @@ SYSCALL_DEFINE2(fanotify_init, unsigned int, flags, unsigned int, event_f_flags) mutex_init(&group->fanotify_data.access_mutex); init_waitqueue_head(&group->fanotify_data.access_waitq); INIT_LIST_HEAD(&group->fanotify_data.access_list); + atomic_set(&group->fanotify_data.bypass_perm, 0); #endif switch (flags & FAN_ALL_CLASS_BITS) { case FAN_CLASS_NOTIF: @@ -764,8 +783,10 @@ SYSCALL_DEFINE(fanotify_mark)(int fanotify_fd, unsigned int flags, if (flags & ~FAN_ALL_MARK_FLAGS) return -EINVAL; switch (flags & (FAN_MARK_ADD | FAN_MARK_REMOVE | FAN_MARK_FLUSH)) { - case FAN_MARK_ADD: + case FAN_MARK_ADD: /* fallthrough */ case FAN_MARK_REMOVE: + if (!mask) + return -EINVAL; case FAN_MARK_FLUSH: break; default: diff --git a/fs/notify/fsnotify.c b/fs/notify/fsnotify.c index 20dc218707c..79b47cbb5cd 100644 --- a/fs/notify/fsnotify.c +++ b/fs/notify/fsnotify.c @@ -59,7 +59,7 @@ void __fsnotify_update_child_dentry_flags(struct inode *inode) /* determine if the children should tell inode about their events */ watched = fsnotify_inode_watches_children(inode); - spin_lock(&dcache_lock); + spin_lock(&inode->i_lock); /* run all of the dentries associated with this inode. Since this is a * directory, there damn well better only be one item on this list */ list_for_each_entry(alias, &inode->i_dentry, d_alias) { @@ -68,19 +68,21 @@ void __fsnotify_update_child_dentry_flags(struct inode *inode) /* run all of the children of the original inode and fix their * d_flags to indicate parental interest (their parent is the * original inode) */ + spin_lock(&alias->d_lock); list_for_each_entry(child, &alias->d_subdirs, d_u.d_child) { if (!child->d_inode) continue; - spin_lock(&child->d_lock); + spin_lock_nested(&child->d_lock, DENTRY_D_LOCK_NESTED); if (watched) child->d_flags |= DCACHE_FSNOTIFY_PARENT_WATCHED; else child->d_flags &= ~DCACHE_FSNOTIFY_PARENT_WATCHED; spin_unlock(&child->d_lock); } + spin_unlock(&alias->d_lock); } - spin_unlock(&dcache_lock); + spin_unlock(&inode->i_lock); } /* Notify this dentry's parent about a child's events. */ diff --git a/fs/notify/inotify/inotify_user.c b/fs/notify/inotify/inotify_user.c index 444c305a468..4cd5d5d78f9 100644 --- a/fs/notify/inotify/inotify_user.c +++ b/fs/notify/inotify/inotify_user.c @@ -752,6 +752,7 @@ SYSCALL_DEFINE1(inotify_init1, int, flags) if (ret >= 0) return ret; + fsnotify_put_group(group); atomic_dec(&user->inotify_devs); out_free_uid: free_uid(user); diff --git a/fs/ntfs/inode.c b/fs/ntfs/inode.c index 93622b175fc..a627ed82c0a 100644 --- a/fs/ntfs/inode.c +++ b/fs/ntfs/inode.c @@ -332,6 +332,13 @@ struct inode *ntfs_alloc_big_inode(struct super_block *sb) return NULL; } +static void ntfs_i_callback(struct rcu_head *head) +{ + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); + kmem_cache_free(ntfs_big_inode_cache, NTFS_I(inode)); +} + void ntfs_destroy_big_inode(struct inode *inode) { ntfs_inode *ni = NTFS_I(inode); @@ -340,7 +347,7 @@ void ntfs_destroy_big_inode(struct inode *inode) BUG_ON(ni->page); if (!atomic_dec_and_test(&ni->count)) BUG(); - kmem_cache_free(ntfs_big_inode_cache, NTFS_I(inode)); + call_rcu(&inode->i_rcu, ntfs_i_callback); } static inline ntfs_inode *ntfs_alloc_extent_inode(void) diff --git a/fs/ocfs2/acl.c b/fs/ocfs2/acl.c index 391915093fe..704f6b1742f 100644 --- a/fs/ocfs2/acl.c +++ b/fs/ocfs2/acl.c @@ -291,13 +291,17 @@ static int ocfs2_set_acl(handle_t *handle, return ret; } -int ocfs2_check_acl(struct inode *inode, int mask) +int ocfs2_check_acl(struct inode *inode, int mask, unsigned int flags) { - struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); + struct ocfs2_super *osb; struct buffer_head *di_bh = NULL; struct posix_acl *acl; int ret = -EAGAIN; + if (flags & IPERM_FLAG_RCU) + return -ECHILD; + + osb = OCFS2_SB(inode->i_sb); if (!(osb->s_mount_opt & OCFS2_MOUNT_POSIX_ACL)) return ret; diff --git a/fs/ocfs2/acl.h b/fs/ocfs2/acl.h index 5c5d31f0585..4fe7c9cf4bf 100644 --- a/fs/ocfs2/acl.h +++ b/fs/ocfs2/acl.h @@ -26,7 +26,7 @@ struct ocfs2_acl_entry { __le32 e_id; }; -extern int ocfs2_check_acl(struct inode *, int); +extern int ocfs2_check_acl(struct inode *, int, unsigned int); extern int ocfs2_acl_chmod(struct inode *); extern int ocfs2_init_acl(handle_t *, struct inode *, struct inode *, struct buffer_head *, struct buffer_head *, diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c index f1e962cb3b7..0d7c5540ad6 100644 --- a/fs/ocfs2/aops.c +++ b/fs/ocfs2/aops.c @@ -573,11 +573,14 @@ static void ocfs2_dio_end_io(struct kiocb *iocb, /* this io's submitter should not have unlocked this before we could */ BUG_ON(!ocfs2_iocb_is_rw_locked(iocb)); + if (ocfs2_iocb_is_sem_locked(iocb)) { + up_read(&inode->i_alloc_sem); + ocfs2_iocb_clear_sem_locked(iocb); + } + ocfs2_iocb_clear_rw_locked(iocb); level = ocfs2_iocb_rw_locked_level(iocb); - if (!level) - up_read(&inode->i_alloc_sem); ocfs2_rw_unlock(inode, level); if (is_async) diff --git a/fs/ocfs2/aops.h b/fs/ocfs2/aops.h index 76bfdfda691..eceb456037c 100644 --- a/fs/ocfs2/aops.h +++ b/fs/ocfs2/aops.h @@ -68,8 +68,27 @@ static inline void ocfs2_iocb_set_rw_locked(struct kiocb *iocb, int level) else clear_bit(1, (unsigned long *)&iocb->private); } + +/* + * Using a named enum representing lock types in terms of #N bit stored in + * iocb->private, which is going to be used for communication bewteen + * ocfs2_dio_end_io() and ocfs2_file_aio_write/read(). + */ +enum ocfs2_iocb_lock_bits { + OCFS2_IOCB_RW_LOCK = 0, + OCFS2_IOCB_RW_LOCK_LEVEL, + OCFS2_IOCB_SEM, + OCFS2_IOCB_NUM_LOCKS +}; + #define ocfs2_iocb_clear_rw_locked(iocb) \ - clear_bit(0, (unsigned long *)&iocb->private) + clear_bit(OCFS2_IOCB_RW_LOCK, (unsigned long *)&iocb->private) #define ocfs2_iocb_rw_locked_level(iocb) \ - test_bit(1, (unsigned long *)&iocb->private) + test_bit(OCFS2_IOCB_RW_LOCK_LEVEL, (unsigned long *)&iocb->private) +#define ocfs2_iocb_set_sem_locked(iocb) \ + set_bit(OCFS2_IOCB_SEM, (unsigned long *)&iocb->private) +#define ocfs2_iocb_clear_sem_locked(iocb) \ + clear_bit(OCFS2_IOCB_SEM, (unsigned long *)&iocb->private) +#define ocfs2_iocb_is_sem_locked(iocb) \ + test_bit(OCFS2_IOCB_SEM, (unsigned long *)&iocb->private) #endif /* OCFS2_FILE_H */ diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c index 52c7557f3e2..9f26ac9be2a 100644 --- a/fs/ocfs2/cluster/heartbeat.c +++ b/fs/ocfs2/cluster/heartbeat.c @@ -1964,8 +1964,10 @@ static struct config_item *o2hb_heartbeat_group_make_item(struct config_group *g if (reg == NULL) return ERR_PTR(-ENOMEM); - if (strlen(name) > O2HB_MAX_REGION_NAME_LEN) - return ERR_PTR(-ENAMETOOLONG); + if (strlen(name) > O2HB_MAX_REGION_NAME_LEN) { + ret = -ENAMETOOLONG; + goto free; + } spin_lock(&o2hb_live_lock); reg->hr_region_num = 0; @@ -1974,7 +1976,8 @@ static struct config_item *o2hb_heartbeat_group_make_item(struct config_group *g O2NM_MAX_REGIONS); if (reg->hr_region_num >= O2NM_MAX_REGIONS) { spin_unlock(&o2hb_live_lock); - return ERR_PTR(-EFBIG); + ret = -EFBIG; + goto free; } set_bit(reg->hr_region_num, o2hb_region_bitmap); } @@ -1986,10 +1989,13 @@ static struct config_item *o2hb_heartbeat_group_make_item(struct config_group *g ret = o2hb_debug_region_init(reg, o2hb_debug_dir); if (ret) { config_item_put(®->hr_item); - return ERR_PTR(ret); + goto free; } return ®->hr_item; +free: + kfree(reg); + return ERR_PTR(ret); } static void o2hb_heartbeat_group_drop_item(struct config_group *group, diff --git a/fs/ocfs2/cluster/masklog.c b/fs/ocfs2/cluster/masklog.c index c7fba396392..6c61771469a 100644 --- a/fs/ocfs2/cluster/masklog.c +++ b/fs/ocfs2/cluster/masklog.c @@ -113,10 +113,11 @@ static struct mlog_attribute mlog_attrs[MLOG_MAX_BITS] = { define_mask(QUOTA), define_mask(REFCOUNT), define_mask(BASTS), + define_mask(RESERVATIONS), + define_mask(CLUSTER), define_mask(ERROR), define_mask(NOTICE), define_mask(KTHREAD), - define_mask(RESERVATIONS), }; static struct attribute *mlog_attr_ptrs[MLOG_MAX_BITS] = {NULL, }; diff --git a/fs/ocfs2/cluster/masklog.h b/fs/ocfs2/cluster/masklog.h index ea2ed9f56c9..34d6544357d 100644 --- a/fs/ocfs2/cluster/masklog.h +++ b/fs/ocfs2/cluster/masklog.h @@ -81,7 +81,7 @@ #include <linux/sched.h> /* bits that are frequently given and infrequently matched in the low word */ -/* NOTE: If you add a flag, you need to also update mlog.c! */ +/* NOTE: If you add a flag, you need to also update masklog.c! */ #define ML_ENTRY 0x0000000000000001ULL /* func call entry */ #define ML_EXIT 0x0000000000000002ULL /* func call exit */ #define ML_TCP 0x0000000000000004ULL /* net cluster/tcp.c */ @@ -114,13 +114,14 @@ #define ML_XATTR 0x0000000020000000ULL /* ocfs2 extended attributes */ #define ML_QUOTA 0x0000000040000000ULL /* ocfs2 quota operations */ #define ML_REFCOUNT 0x0000000080000000ULL /* refcount tree operations */ -#define ML_BASTS 0x0000001000000000ULL /* dlmglue asts and basts */ +#define ML_BASTS 0x0000000100000000ULL /* dlmglue asts and basts */ +#define ML_RESERVATIONS 0x0000000200000000ULL /* ocfs2 alloc reservations */ +#define ML_CLUSTER 0x0000000400000000ULL /* cluster stack */ + /* bits that are infrequently given and frequently matched in the high word */ -#define ML_ERROR 0x0000000100000000ULL /* sent to KERN_ERR */ -#define ML_NOTICE 0x0000000200000000ULL /* setn to KERN_NOTICE */ -#define ML_KTHREAD 0x0000000400000000ULL /* kernel thread activity */ -#define ML_RESERVATIONS 0x0000000800000000ULL /* ocfs2 alloc reservations */ -#define ML_CLUSTER 0x0000001000000000ULL /* cluster stack */ +#define ML_ERROR 0x1000000000000000ULL /* sent to KERN_ERR */ +#define ML_NOTICE 0x2000000000000000ULL /* setn to KERN_NOTICE */ +#define ML_KTHREAD 0x4000000000000000ULL /* kernel thread activity */ #define MLOG_INITIAL_AND_MASK (ML_ERROR|ML_NOTICE) #define MLOG_INITIAL_NOT_MASK (ML_ENTRY|ML_EXIT) diff --git a/fs/ocfs2/dcache.c b/fs/ocfs2/dcache.c index edaded48e7e..6d80ecc7834 100644 --- a/fs/ocfs2/dcache.c +++ b/fs/ocfs2/dcache.c @@ -52,9 +52,15 @@ void ocfs2_dentry_attach_gen(struct dentry *dentry) static int ocfs2_dentry_revalidate(struct dentry *dentry, struct nameidata *nd) { - struct inode *inode = dentry->d_inode; + struct inode *inode; int ret = 0; /* if all else fails, just return false */ - struct ocfs2_super *osb = OCFS2_SB(dentry->d_sb); + struct ocfs2_super *osb; + + if (nd->flags & LOOKUP_RCU) + return -ECHILD; + + inode = dentry->d_inode; + osb = OCFS2_SB(dentry->d_sb); mlog_entry("(0x%p, '%.*s')\n", dentry, dentry->d_name.len, dentry->d_name.name); @@ -169,23 +175,25 @@ struct dentry *ocfs2_find_local_alias(struct inode *inode, struct list_head *p; struct dentry *dentry = NULL; - spin_lock(&dcache_lock); - + spin_lock(&inode->i_lock); list_for_each(p, &inode->i_dentry) { dentry = list_entry(p, struct dentry, d_alias); + spin_lock(&dentry->d_lock); if (ocfs2_match_dentry(dentry, parent_blkno, skip_unhashed)) { mlog(0, "dentry found: %.*s\n", dentry->d_name.len, dentry->d_name.name); - dget_locked(dentry); + dget_dlock(dentry); + spin_unlock(&dentry->d_lock); break; } + spin_unlock(&dentry->d_lock); dentry = NULL; } - spin_unlock(&dcache_lock); + spin_unlock(&inode->i_lock); return dentry; } @@ -476,7 +484,6 @@ static void ocfs2_dentry_iput(struct dentry *dentry, struct inode *inode) out: iput(inode); - ocfs2_dentry_attach_gen(dentry); } /* diff --git a/fs/ocfs2/dir.c b/fs/ocfs2/dir.c index c49f6de0e7a..d417b3f9b0c 100644 --- a/fs/ocfs2/dir.c +++ b/fs/ocfs2/dir.c @@ -2461,8 +2461,10 @@ static int ocfs2_dx_dir_attach_index(struct ocfs2_super *osb, di->i_dx_root = cpu_to_le64(dr_blkno); + spin_lock(&OCFS2_I(dir)->ip_lock); OCFS2_I(dir)->ip_dyn_features |= OCFS2_INDEXED_DIR_FL; di->i_dyn_features = cpu_to_le16(OCFS2_I(dir)->ip_dyn_features); + spin_unlock(&OCFS2_I(dir)->ip_lock); ocfs2_journal_dirty(handle, di_bh); @@ -4466,8 +4468,10 @@ static int ocfs2_dx_dir_remove_index(struct inode *dir, goto out_commit; } + spin_lock(&OCFS2_I(dir)->ip_lock); OCFS2_I(dir)->ip_dyn_features &= ~OCFS2_INDEXED_DIR_FL; di->i_dyn_features = cpu_to_le16(OCFS2_I(dir)->ip_dyn_features); + spin_unlock(&OCFS2_I(dir)->ip_lock); di->i_dx_root = cpu_to_le64(0ULL); ocfs2_journal_dirty(handle, di_bh); diff --git a/fs/ocfs2/dlm/dlmdomain.c b/fs/ocfs2/dlm/dlmdomain.c index 58a93b95373..cc2aaa96cfe 100644 --- a/fs/ocfs2/dlm/dlmdomain.c +++ b/fs/ocfs2/dlm/dlmdomain.c @@ -959,7 +959,7 @@ static int dlm_match_regions(struct dlm_ctxt *dlm, r += O2HB_MAX_REGION_NAME_LEN; } - local = kmalloc(sizeof(qr->qr_regions), GFP_KERNEL); + local = kmalloc(sizeof(qr->qr_regions), GFP_ATOMIC); if (!local) { status = -ENOMEM; goto bail; diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c index f564b0e5f80..59f0f6bdfc6 100644 --- a/fs/ocfs2/dlm/dlmmaster.c +++ b/fs/ocfs2/dlm/dlmmaster.c @@ -2346,7 +2346,8 @@ static void dlm_deref_lockres_worker(struct dlm_work_item *item, void *data) */ static int dlm_is_lockres_migrateable(struct dlm_ctxt *dlm, struct dlm_lock_resource *res, - int *numlocks) + int *numlocks, + int *hasrefs) { int ret; int i; @@ -2356,6 +2357,9 @@ static int dlm_is_lockres_migrateable(struct dlm_ctxt *dlm, assert_spin_locked(&res->spinlock); + *numlocks = 0; + *hasrefs = 0; + ret = -EINVAL; if (res->owner == DLM_LOCK_RES_OWNER_UNKNOWN) { mlog(0, "cannot migrate lockres with unknown owner!\n"); @@ -2386,7 +2390,13 @@ static int dlm_is_lockres_migrateable(struct dlm_ctxt *dlm, } *numlocks = count; - mlog(0, "migrateable lockres having %d locks\n", *numlocks); + + count = find_next_bit(res->refmap, O2NM_MAX_NODES, 0); + if (count < O2NM_MAX_NODES) + *hasrefs = 1; + + mlog(0, "%s: res %.*s, Migrateable, locks %d, refs %d\n", dlm->name, + res->lockname.len, res->lockname.name, *numlocks, *hasrefs); leave: return ret; @@ -2408,7 +2418,7 @@ static int dlm_migrate_lockres(struct dlm_ctxt *dlm, const char *name; unsigned int namelen; int mle_added = 0; - int numlocks; + int numlocks, hasrefs; int wake = 0; if (!dlm_grab(dlm)) @@ -2417,13 +2427,13 @@ static int dlm_migrate_lockres(struct dlm_ctxt *dlm, name = res->lockname.name; namelen = res->lockname.len; - mlog(0, "migrating %.*s to %u\n", namelen, name, target); + mlog(0, "%s: Migrating %.*s to %u\n", dlm->name, namelen, name, target); /* * ensure this lockres is a proper candidate for migration */ spin_lock(&res->spinlock); - ret = dlm_is_lockres_migrateable(dlm, res, &numlocks); + ret = dlm_is_lockres_migrateable(dlm, res, &numlocks, &hasrefs); if (ret < 0) { spin_unlock(&res->spinlock); goto leave; @@ -2431,10 +2441,8 @@ static int dlm_migrate_lockres(struct dlm_ctxt *dlm, spin_unlock(&res->spinlock); /* no work to do */ - if (numlocks == 0) { - mlog(0, "no locks were found on this lockres! done!\n"); + if (numlocks == 0 && !hasrefs) goto leave; - } /* * preallocate up front @@ -2459,14 +2467,14 @@ static int dlm_migrate_lockres(struct dlm_ctxt *dlm, * find a node to migrate the lockres to */ - mlog(0, "picking a migration node\n"); spin_lock(&dlm->spinlock); /* pick a new node */ if (!test_bit(target, dlm->domain_map) || target >= O2NM_MAX_NODES) { target = dlm_pick_migration_target(dlm, res); } - mlog(0, "node %u chosen for migration\n", target); + mlog(0, "%s: res %.*s, Node %u chosen for migration\n", dlm->name, + namelen, name, target); if (target >= O2NM_MAX_NODES || !test_bit(target, dlm->domain_map)) { @@ -2667,7 +2675,7 @@ int dlm_empty_lockres(struct dlm_ctxt *dlm, struct dlm_lock_resource *res) { int ret; int lock_dropped = 0; - int numlocks; + int numlocks, hasrefs; spin_lock(&res->spinlock); if (res->owner != dlm->node_num) { @@ -2681,8 +2689,8 @@ int dlm_empty_lockres(struct dlm_ctxt *dlm, struct dlm_lock_resource *res) } /* No need to migrate a lockres having no locks */ - ret = dlm_is_lockres_migrateable(dlm, res, &numlocks); - if (ret >= 0 && numlocks == 0) { + ret = dlm_is_lockres_migrateable(dlm, res, &numlocks, &hasrefs); + if (ret >= 0 && numlocks == 0 && !hasrefs) { spin_unlock(&res->spinlock); goto leave; } @@ -2915,6 +2923,12 @@ static u8 dlm_pick_migration_target(struct dlm_ctxt *dlm, } queue++; } + + nodenum = find_next_bit(res->refmap, O2NM_MAX_NODES, 0); + if (nodenum < O2NM_MAX_NODES) { + spin_unlock(&res->spinlock); + return nodenum; + } spin_unlock(&res->spinlock); mlog(0, "have not found a suitable target yet! checking domain map\n"); diff --git a/fs/ocfs2/dlmfs/dlmfs.c b/fs/ocfs2/dlmfs/dlmfs.c index b2df490a19e..8c5c0eddc36 100644 --- a/fs/ocfs2/dlmfs/dlmfs.c +++ b/fs/ocfs2/dlmfs/dlmfs.c @@ -351,11 +351,18 @@ static struct inode *dlmfs_alloc_inode(struct super_block *sb) return &ip->ip_vfs_inode; } -static void dlmfs_destroy_inode(struct inode *inode) +static void dlmfs_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(dlmfs_inode_cache, DLMFS_I(inode)); } +static void dlmfs_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, dlmfs_i_callback); +} + static void dlmfs_evict_inode(struct inode *inode) { int status; diff --git a/fs/ocfs2/export.c b/fs/ocfs2/export.c index 19ad145d2af..6adafa57606 100644 --- a/fs/ocfs2/export.c +++ b/fs/ocfs2/export.c @@ -138,7 +138,7 @@ check_gen: result = d_obtain_alias(inode); if (!IS_ERR(result)) - result->d_op = &ocfs2_dentry_ops; + d_set_d_op(result, &ocfs2_dentry_ops); else mlog_errno(PTR_ERR(result)); @@ -176,7 +176,7 @@ static struct dentry *ocfs2_get_parent(struct dentry *child) parent = d_obtain_alias(ocfs2_iget(OCFS2_SB(dir->i_sb), blkno, 0, 0)); if (!IS_ERR(parent)) - parent->d_op = &ocfs2_dentry_ops; + d_set_d_op(parent, &ocfs2_dentry_ops); bail_unlock: ocfs2_inode_unlock(dir, 0); diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index 77b4c04a280..bdadbae0909 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -1307,10 +1307,13 @@ bail: return err; } -int ocfs2_permission(struct inode *inode, int mask) +int ocfs2_permission(struct inode *inode, int mask, unsigned int flags) { int ret; + if (flags & IPERM_FLAG_RCU) + return -ECHILD; + mlog_entry_void(); ret = ocfs2_inode_lock(inode, NULL, 0); @@ -1320,7 +1323,7 @@ int ocfs2_permission(struct inode *inode, int mask) goto out; } - ret = generic_permission(inode, mask, ocfs2_check_acl); + ret = generic_permission(inode, mask, flags, ocfs2_check_acl); ocfs2_inode_unlock(inode, 0); out: @@ -2241,11 +2244,15 @@ static ssize_t ocfs2_file_aio_write(struct kiocb *iocb, mutex_lock(&inode->i_mutex); + ocfs2_iocb_clear_sem_locked(iocb); + relock: /* to match setattr's i_mutex -> i_alloc_sem -> rw_lock ordering */ if (direct_io) { down_read(&inode->i_alloc_sem); have_alloc_sem = 1; + /* communicate with ocfs2_dio_end_io */ + ocfs2_iocb_set_sem_locked(iocb); } /* @@ -2382,8 +2389,10 @@ out: ocfs2_rw_unlock(inode, rw_level); out_sems: - if (have_alloc_sem) + if (have_alloc_sem) { up_read(&inode->i_alloc_sem); + ocfs2_iocb_clear_sem_locked(iocb); + } mutex_unlock(&inode->i_mutex); @@ -2527,6 +2536,8 @@ static ssize_t ocfs2_file_aio_read(struct kiocb *iocb, goto bail; } + ocfs2_iocb_clear_sem_locked(iocb); + /* * buffered reads protect themselves in ->readpage(). O_DIRECT reads * need locks to protect pending reads from racing with truncate. @@ -2534,6 +2545,7 @@ static ssize_t ocfs2_file_aio_read(struct kiocb *iocb, if (filp->f_flags & O_DIRECT) { down_read(&inode->i_alloc_sem); have_alloc_sem = 1; + ocfs2_iocb_set_sem_locked(iocb); ret = ocfs2_rw_lock(inode, 0); if (ret < 0) { @@ -2575,8 +2587,10 @@ static ssize_t ocfs2_file_aio_read(struct kiocb *iocb, } bail: - if (have_alloc_sem) + if (have_alloc_sem) { up_read(&inode->i_alloc_sem); + ocfs2_iocb_clear_sem_locked(iocb); + } if (rw_level != -1) ocfs2_rw_unlock(inode, rw_level); mlog_exit(ret); diff --git a/fs/ocfs2/file.h b/fs/ocfs2/file.h index 97bf761c9e7..f5afbbef670 100644 --- a/fs/ocfs2/file.h +++ b/fs/ocfs2/file.h @@ -61,7 +61,7 @@ int ocfs2_zero_extend(struct inode *inode, struct buffer_head *di_bh, int ocfs2_setattr(struct dentry *dentry, struct iattr *attr); int ocfs2_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat); -int ocfs2_permission(struct inode *inode, int mask); +int ocfs2_permission(struct inode *inode, int mask, unsigned int flags); int ocfs2_should_update_atime(struct inode *inode, struct vfsmount *vfsmnt); diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c index ff5744e1e36..d14cad6e2e4 100644 --- a/fs/ocfs2/namei.c +++ b/fs/ocfs2/namei.c @@ -147,7 +147,7 @@ static struct dentry *ocfs2_lookup(struct inode *dir, struct dentry *dentry, spin_unlock(&oi->ip_lock); bail_add: - dentry->d_op = &ocfs2_dentry_ops; + d_set_d_op(dentry, &ocfs2_dentry_ops); ret = d_splice_alias(inode, dentry); if (inode) { @@ -415,7 +415,7 @@ static int ocfs2_mknod(struct inode *dir, mlog_errno(status); goto leave; } - dentry->d_op = &ocfs2_dentry_ops; + d_set_d_op(dentry, &ocfs2_dentry_ops); status = ocfs2_add_entry(handle, dentry, inode, OCFS2_I(inode)->ip_blkno, parent_fe_bh, @@ -743,7 +743,7 @@ static int ocfs2_link(struct dentry *old_dentry, } ihold(inode); - dentry->d_op = &ocfs2_dentry_ops; + d_set_d_op(dentry, &ocfs2_dentry_ops); d_instantiate(dentry, inode); out_commit: @@ -1794,7 +1794,7 @@ static int ocfs2_symlink(struct inode *dir, mlog_errno(status); goto bail; } - dentry->d_op = &ocfs2_dentry_ops; + d_set_d_op(dentry, &ocfs2_dentry_ops); status = ocfs2_add_entry(handle, dentry, inode, le64_to_cpu(fe->i_blkno), parent_fe_bh, @@ -2459,7 +2459,7 @@ int ocfs2_mv_orphaned_inode_to_new(struct inode *dir, goto out_commit; } - dentry->d_op = &ocfs2_dentry_ops; + d_set_d_op(dentry, &ocfs2_dentry_ops); d_instantiate(dentry, inode); status = 0; out_commit: diff --git a/fs/ocfs2/ocfs2.h b/fs/ocfs2/ocfs2.h index d8408217e3b..70dd3b1798f 100644 --- a/fs/ocfs2/ocfs2.h +++ b/fs/ocfs2/ocfs2.h @@ -159,7 +159,9 @@ struct ocfs2_lock_res { char l_name[OCFS2_LOCK_ID_MAX_LEN]; unsigned int l_ro_holders; unsigned int l_ex_holders; - unsigned char l_level; + signed char l_level; + signed char l_requested; + signed char l_blocking; /* Data packed - type enum ocfs2_lock_type */ unsigned char l_type; @@ -169,8 +171,6 @@ struct ocfs2_lock_res { unsigned char l_action; /* Data packed - enum type ocfs2_unlock_action */ unsigned char l_unlock_action; - unsigned char l_requested; - unsigned char l_blocking; unsigned int l_pending_gen; spinlock_t l_lock; diff --git a/fs/ocfs2/ocfs2_fs.h b/fs/ocfs2/ocfs2_fs.h index c2e4f8222e2..bf2e7764920 100644 --- a/fs/ocfs2/ocfs2_fs.h +++ b/fs/ocfs2/ocfs2_fs.h @@ -350,7 +350,7 @@ enum { #define OCFS2_LAST_LOCAL_SYSTEM_INODE LOCAL_GROUP_QUOTA_SYSTEM_INODE NUM_SYSTEM_INODES }; -#define NUM_GLOBAL_SYSTEM_INODES OCFS2_LAST_GLOBAL_SYSTEM_INODE +#define NUM_GLOBAL_SYSTEM_INODES OCFS2_FIRST_LOCAL_SYSTEM_INODE #define NUM_LOCAL_SYSTEM_INODES \ (NUM_SYSTEM_INODES - OCFS2_FIRST_LOCAL_SYSTEM_INODE) diff --git a/fs/ocfs2/stack_user.c b/fs/ocfs2/stack_user.c index 252e7c82f92..a5ebe421195 100644 --- a/fs/ocfs2/stack_user.c +++ b/fs/ocfs2/stack_user.c @@ -190,7 +190,7 @@ static struct ocfs2_live_connection *ocfs2_connection_find(const char *name) return c; } - return c; + return NULL; } /* diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index f02c0ef3157..17ff46fa8a1 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -41,7 +41,6 @@ #include <linux/mount.h> #include <linux/seq_file.h> #include <linux/quotaops.h> -#include <linux/smp_lock.h> #define MLOG_MASK_PREFIX ML_SUPER #include <cluster/masklog.h> @@ -570,11 +569,18 @@ static struct inode *ocfs2_alloc_inode(struct super_block *sb) return &oi->vfs_inode; } -static void ocfs2_destroy_inode(struct inode *inode) +static void ocfs2_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(ocfs2_inode_cachep, OCFS2_I(inode)); } +static void ocfs2_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, ocfs2_i_callback); +} + static unsigned long long ocfs2_max_file_offset(unsigned int bbits, unsigned int cbits) { diff --git a/fs/openpromfs/inode.c b/fs/openpromfs/inode.c index ddb1f41376e..a2a5bff774e 100644 --- a/fs/openpromfs/inode.c +++ b/fs/openpromfs/inode.c @@ -343,11 +343,18 @@ static struct inode *openprom_alloc_inode(struct super_block *sb) return &oi->vfs_inode; } -static void openprom_destroy_inode(struct inode *inode) +static void openprom_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(op_inode_cachep, OP_I(inode)); } +static void openprom_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, openprom_i_callback); +} + static struct inode *openprom_iget(struct super_block *sb, ino_t ino) { struct inode *inode; @@ -418,7 +425,7 @@ out_no_root: static struct dentry *openprom_mount(struct file_system_type *fs_type, int flags, const char *dev_name, void *data) { - return mount_single(fs_type, flags, data, openprom_fill_super) + return mount_single(fs_type, flags, data, openprom_fill_super); } static struct file_system_type openprom_fs_type = { diff --git a/fs/pipe.c b/fs/pipe.c index a8012a95572..68f1f8e4e23 100644 --- a/fs/pipe.c +++ b/fs/pipe.c @@ -999,12 +999,12 @@ struct file *create_write_pipe(int flags) goto err; err = -ENOMEM; - path.dentry = d_alloc(pipe_mnt->mnt_sb->s_root, &name); + path.dentry = d_alloc_pseudo(pipe_mnt->mnt_sb, &name); if (!path.dentry) goto err_inode; path.mnt = mntget(pipe_mnt); - path.dentry->d_op = &pipefs_dentry_operations; + d_set_d_op(path.dentry, &pipefs_dentry_operations); d_instantiate(path.dentry, inode); err = -ENFILE; @@ -1199,12 +1199,24 @@ int pipe_proc_fn(struct ctl_table *table, int write, void __user *buf, return ret; } +/* + * After the inode slimming patch, i_pipe/i_bdev/i_cdev share the same + * location, so checking ->i_pipe is not enough to verify that this is a + * pipe. + */ +struct pipe_inode_info *get_pipe_info(struct file *file) +{ + struct inode *i = file->f_path.dentry->d_inode; + + return S_ISFIFO(i->i_mode) ? i->i_pipe : NULL; +} + long pipe_fcntl(struct file *file, unsigned int cmd, unsigned long arg) { struct pipe_inode_info *pipe; long ret; - pipe = file->f_path.dentry->d_inode->i_pipe; + pipe = get_pipe_info(file); if (!pipe) return -EBADF; @@ -1241,6 +1253,10 @@ out: return ret; } +static const struct super_operations pipefs_ops = { + .destroy_inode = free_inode_nonrcu, +}; + /* * pipefs should _never_ be mounted by userland - too much of security hassle, * no real gain from having the whole whorehouse mounted. So we don't need @@ -1250,7 +1266,7 @@ out: static struct dentry *pipefs_mount(struct file_system_type *fs_type, int flags, const char *dev_name, void *data) { - return mount_pseudo(fs_type, "pipe:", NULL, PIPEFS_MAGIC); + return mount_pseudo(fs_type, "pipe:", &pipefs_ops, PIPEFS_MAGIC); } static struct file_system_type pipe_fs_type = { @@ -1276,7 +1292,7 @@ static int __init init_pipe_fs(void) static void __exit exit_pipe_fs(void) { unregister_filesystem(&pipe_fs_type); - mntput(pipe_mnt); + mntput_long(pipe_mnt); } fs_initcall(init_pipe_fs); diff --git a/fs/pnode.c b/fs/pnode.c index 8066b8dd748..d42514e3238 100644 --- a/fs/pnode.c +++ b/fs/pnode.c @@ -288,7 +288,7 @@ out: */ static inline int do_refcount_check(struct vfsmount *mnt, int count) { - int mycount = atomic_read(&mnt->mnt_count) - mnt->mnt_ghosts; + int mycount = mnt_get_count(mnt) - mnt->mnt_ghosts; return (mycount > count); } @@ -300,7 +300,7 @@ static inline int do_refcount_check(struct vfsmount *mnt, int count) * Check if any of these mounts that **do not have submounts** * have more references than 'refcnt'. If so return busy. * - * vfsmount lock must be held for read or write + * vfsmount lock must be held for write */ int propagate_mount_busy(struct vfsmount *mnt, int refcnt) { diff --git a/fs/proc/Makefile b/fs/proc/Makefile index 2758e2afc51..288a49e098b 100644 --- a/fs/proc/Makefile +++ b/fs/proc/Makefile @@ -15,6 +15,7 @@ proc-y += devices.o proc-y += interrupts.o proc-y += loadavg.o proc-y += meminfo.o +proc-y += proc_console.o proc-y += stat.o proc-y += uptime.o proc-y += version.o diff --git a/fs/proc/base.c b/fs/proc/base.c index f3d02ca461e..b20962c71a5 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -1407,6 +1407,82 @@ static const struct file_operations proc_pid_sched_operations = { #endif +#ifdef CONFIG_SCHED_AUTOGROUP +/* + * Print out autogroup related information: + */ +static int sched_autogroup_show(struct seq_file *m, void *v) +{ + struct inode *inode = m->private; + struct task_struct *p; + + p = get_proc_task(inode); + if (!p) + return -ESRCH; + proc_sched_autogroup_show_task(p, m); + + put_task_struct(p); + + return 0; +} + +static ssize_t +sched_autogroup_write(struct file *file, const char __user *buf, + size_t count, loff_t *offset) +{ + struct inode *inode = file->f_path.dentry->d_inode; + struct task_struct *p; + char buffer[PROC_NUMBUF]; + long nice; + int err; + + memset(buffer, 0, sizeof(buffer)); + if (count > sizeof(buffer) - 1) + count = sizeof(buffer) - 1; + if (copy_from_user(buffer, buf, count)) + return -EFAULT; + + err = strict_strtol(strstrip(buffer), 0, &nice); + if (err) + return -EINVAL; + + p = get_proc_task(inode); + if (!p) + return -ESRCH; + + err = nice; + err = proc_sched_autogroup_set_nice(p, &err); + if (err) + count = err; + + put_task_struct(p); + + return count; +} + +static int sched_autogroup_open(struct inode *inode, struct file *filp) +{ + int ret; + + ret = single_open(filp, sched_autogroup_show, NULL); + if (!ret) { + struct seq_file *m = filp->private_data; + + m->private = inode; + } + return ret; +} + +static const struct file_operations proc_pid_sched_autogroup_operations = { + .open = sched_autogroup_open, + .read = seq_read, + .write = sched_autogroup_write, + .llseek = seq_lseek, + .release = single_release, +}; + +#endif /* CONFIG_SCHED_AUTOGROUP */ + static ssize_t comm_write(struct file *file, const char __user *buf, size_t count, loff_t *offset) { @@ -1574,7 +1650,7 @@ static int do_proc_readlink(struct path *path, char __user *buffer, int buflen) if (!tmp) return -ENOMEM; - pathname = d_path_with_unreachable(path, tmp, PAGE_SIZE); + pathname = d_path(path, tmp, PAGE_SIZE); len = PTR_ERR(pathname); if (IS_ERR(pathname)) goto out; @@ -1719,10 +1795,16 @@ static int pid_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat */ static int pid_revalidate(struct dentry *dentry, struct nameidata *nd) { - struct inode *inode = dentry->d_inode; - struct task_struct *task = get_proc_task(inode); + struct inode *inode; + struct task_struct *task; const struct cred *cred; + if (nd && nd->flags & LOOKUP_RCU) + return -ECHILD; + + inode = dentry->d_inode; + task = get_proc_task(inode); + if (task) { if ((inode->i_mode == (S_IFDIR|S_IRUGO|S_IXUGO)) || task_dumpable(task)) { @@ -1744,7 +1826,7 @@ static int pid_revalidate(struct dentry *dentry, struct nameidata *nd) return 0; } -static int pid_delete_dentry(struct dentry * dentry) +static int pid_delete_dentry(const struct dentry * dentry) { /* Is the task we represent dead? * If so, then don't put the dentry on the lru list, @@ -1888,12 +1970,19 @@ static int proc_fd_link(struct inode *inode, struct path *path) static int tid_fd_revalidate(struct dentry *dentry, struct nameidata *nd) { - struct inode *inode = dentry->d_inode; - struct task_struct *task = get_proc_task(inode); - int fd = proc_fd(inode); + struct inode *inode; + struct task_struct *task; + int fd; struct files_struct *files; const struct cred *cred; + if (nd && nd->flags & LOOKUP_RCU) + return -ECHILD; + + inode = dentry->d_inode; + task = get_proc_task(inode); + fd = proc_fd(inode); + if (task) { files = get_files_struct(task); if (files) { @@ -1969,7 +2058,7 @@ static struct dentry *proc_fd_instantiate(struct inode *dir, inode->i_op = &proc_pid_link_inode_operations; inode->i_size = 64; ei->op.proc_get_link = proc_fd_link; - dentry->d_op = &tid_fd_dentry_operations; + d_set_d_op(dentry, &tid_fd_dentry_operations); d_add(dentry, inode); /* Close the race of the process dying before we return the dentry */ if (tid_fd_revalidate(dentry, NULL)) @@ -2101,11 +2190,13 @@ static const struct file_operations proc_fd_operations = { * /proc/pid/fd needs a special permission handler so that a process can still * access /proc/self/fd after it has executed a setuid(). */ -static int proc_fd_permission(struct inode *inode, int mask) +static int proc_fd_permission(struct inode *inode, int mask, unsigned int flags) { int rv; - rv = generic_permission(inode, mask, NULL); + if (flags & IPERM_FLAG_RCU) + return -ECHILD; + rv = generic_permission(inode, mask, flags, NULL); if (rv == 0) return 0; if (task_pid(current) == proc_pid(inode)) @@ -2137,7 +2228,7 @@ static struct dentry *proc_fdinfo_instantiate(struct inode *dir, ei->fd = fd; inode->i_mode = S_IFREG | S_IRUSR; inode->i_fop = &proc_fdinfo_file_operations; - dentry->d_op = &tid_fd_dentry_operations; + d_set_d_op(dentry, &tid_fd_dentry_operations); d_add(dentry, inode); /* Close the race of the process dying before we return the dentry */ if (tid_fd_revalidate(dentry, NULL)) @@ -2196,7 +2287,7 @@ static struct dentry *proc_pident_instantiate(struct inode *dir, if (p->fop) inode->i_fop = p->fop; ei->op = p->op; - dentry->d_op = &pid_dentry_operations; + d_set_d_op(dentry, &pid_dentry_operations); d_add(dentry, inode); /* Close the race of the process dying before we return the dentry */ if (pid_revalidate(dentry, NULL)) @@ -2563,8 +2654,14 @@ static const struct pid_entry proc_base_stuff[] = { */ static int proc_base_revalidate(struct dentry *dentry, struct nameidata *nd) { - struct inode *inode = dentry->d_inode; - struct task_struct *task = get_proc_task(inode); + struct inode *inode; + struct task_struct *task; + + if (nd->flags & LOOKUP_RCU) + return -ECHILD; + + inode = dentry->d_inode; + task = get_proc_task(inode); if (task) { put_task_struct(task); return 1; @@ -2615,7 +2712,7 @@ static struct dentry *proc_base_instantiate(struct inode *dir, if (p->fop) inode->i_fop = p->fop; ei->op = p->op; - dentry->d_op = &proc_base_dentry_operations; + d_set_d_op(dentry, &proc_base_dentry_operations); d_add(dentry, inode); error = NULL; out: @@ -2733,6 +2830,9 @@ static const struct pid_entry tgid_base_stuff[] = { #ifdef CONFIG_SCHED_DEBUG REG("sched", S_IRUGO|S_IWUSR, proc_pid_sched_operations), #endif +#ifdef CONFIG_SCHED_AUTOGROUP + REG("autogroup", S_IRUGO|S_IWUSR, proc_pid_sched_autogroup_operations), +#endif REG("comm", S_IRUGO|S_IWUSR, proc_pid_set_comm_operations), #ifdef CONFIG_HAVE_ARCH_TRACEHOOK INF("syscall", S_IRUSR, proc_pid_syscall), @@ -2926,7 +3026,7 @@ static struct dentry *proc_pid_instantiate(struct inode *dir, inode->i_nlink = 2 + pid_entry_count_dirs(tgid_base_stuff, ARRAY_SIZE(tgid_base_stuff)); - dentry->d_op = &pid_dentry_operations; + d_set_d_op(dentry, &pid_dentry_operations); d_add(dentry, inode); /* Close the race of the process dying before we return the dentry */ @@ -3169,7 +3269,7 @@ static struct dentry *proc_task_instantiate(struct inode *dir, inode->i_nlink = 2 + pid_entry_count_dirs(tid_base_stuff, ARRAY_SIZE(tid_base_stuff)); - dentry->d_op = &pid_dentry_operations; + d_set_d_op(dentry, &pid_dentry_operations); d_add(dentry, inode); /* Close the race of the process dying before we return the dentry */ diff --git a/fs/proc/generic.c b/fs/proc/generic.c index dd29f033766..f766be29d2c 100644 --- a/fs/proc/generic.c +++ b/fs/proc/generic.c @@ -400,7 +400,7 @@ static const struct inode_operations proc_link_inode_operations = { * smarter: we could keep a "volatile" flag in the * inode to indicate which ones to keep. */ -static int proc_delete_dentry(struct dentry * dentry) +static int proc_delete_dentry(const struct dentry * dentry) { return 1; } @@ -439,7 +439,7 @@ struct dentry *proc_lookup_de(struct proc_dir_entry *de, struct inode *dir, out_unlock: if (inode) { - dentry->d_op = &proc_dentry_operations; + d_set_d_op(dentry, &proc_dentry_operations); d_add(dentry, inode); return NULL; } diff --git a/fs/proc/inode.c b/fs/proc/inode.c index 9c2b5f48487..6bcb926b101 100644 --- a/fs/proc/inode.c +++ b/fs/proc/inode.c @@ -16,7 +16,6 @@ #include <linux/limits.h> #include <linux/init.h> #include <linux/module.h> -#include <linux/smp_lock.h> #include <linux/sysctl.h> #include <linux/slab.h> @@ -66,11 +65,18 @@ static struct inode *proc_alloc_inode(struct super_block *sb) return inode; } -static void proc_destroy_inode(struct inode *inode) +static void proc_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(proc_inode_cachep, PROC_I(inode)); } +static void proc_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, proc_i_callback); +} + static void init_once(void *foo) { struct proc_inode *ei = (struct proc_inode *) foo; diff --git a/fs/proc/proc_console.c b/fs/proc/proc_console.c new file mode 100644 index 00000000000..8a707609f52 --- /dev/null +++ b/fs/proc/proc_console.c @@ -0,0 +1,114 @@ +/* + * Copyright (c) 2010 Werner Fink, Jiri Slaby + * + * Licensed under GPLv2 + */ + +#include <linux/console.h> +#include <linux/kernel.h> +#include <linux/proc_fs.h> +#include <linux/seq_file.h> +#include <linux/tty_driver.h> + +/* + * This is handler for /proc/consoles + */ +static int show_console_dev(struct seq_file *m, void *v) +{ + static const struct { + short flag; + char name; + } con_flags[] = { + { CON_ENABLED, 'E' }, + { CON_CONSDEV, 'C' }, + { CON_BOOT, 'B' }, + { CON_PRINTBUFFER, 'p' }, + { CON_BRL, 'b' }, + { CON_ANYTIME, 'a' }, + }; + char flags[ARRAY_SIZE(con_flags) + 1]; + struct console *con = v; + unsigned int a; + int len; + dev_t dev = 0; + + if (con->device) { + const struct tty_driver *driver; + int index; + driver = con->device(con, &index); + if (driver) { + dev = MKDEV(driver->major, driver->minor_start); + dev += index; + } + } + + for (a = 0; a < ARRAY_SIZE(con_flags); a++) + flags[a] = (con->flags & con_flags[a].flag) ? + con_flags[a].name : ' '; + flags[a] = 0; + + seq_printf(m, "%s%d%n", con->name, con->index, &len); + len = 21 - len; + if (len < 1) + len = 1; + seq_printf(m, "%*c%c%c%c (%s)", len, ' ', con->read ? 'R' : '-', + con->write ? 'W' : '-', con->unblank ? 'U' : '-', + flags); + if (dev) + seq_printf(m, " %4d:%d", MAJOR(dev), MINOR(dev)); + + seq_printf(m, "\n"); + + return 0; +} + +static void *c_start(struct seq_file *m, loff_t *pos) +{ + struct console *con; + loff_t off = 0; + + acquire_console_sem(); + for_each_console(con) + if (off++ == *pos) + break; + + return con; +} + +static void *c_next(struct seq_file *m, void *v, loff_t *pos) +{ + struct console *con = v; + ++*pos; + return con->next; +} + +static void c_stop(struct seq_file *m, void *v) +{ + release_console_sem(); +} + +static const struct seq_operations consoles_op = { + .start = c_start, + .next = c_next, + .stop = c_stop, + .show = show_console_dev +}; + +static int consoles_open(struct inode *inode, struct file *file) +{ + return seq_open(file, &consoles_op); +} + +static const struct file_operations proc_consoles_operations = { + .open = consoles_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; + +static int register_proc_consoles(void) +{ + proc_create("consoles", 0, NULL, &proc_consoles_operations); + return 0; +} +module_init(register_proc_consoles); diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c index b652cb00906..09a1f92a34e 100644 --- a/fs/proc/proc_sysctl.c +++ b/fs/proc/proc_sysctl.c @@ -5,6 +5,7 @@ #include <linux/sysctl.h> #include <linux/proc_fs.h> #include <linux/security.h> +#include <linux/namei.h> #include "internal.h" static const struct dentry_operations proc_sys_dentry_operations; @@ -120,7 +121,7 @@ static struct dentry *proc_sys_lookup(struct inode *dir, struct dentry *dentry, goto out; err = NULL; - dentry->d_op = &proc_sys_dentry_operations; + d_set_d_op(dentry, &proc_sys_dentry_operations); d_add(dentry, inode); out: @@ -201,7 +202,7 @@ static int proc_sys_fill_cache(struct file *filp, void *dirent, dput(child); return -ENOMEM; } else { - child->d_op = &proc_sys_dentry_operations; + d_set_d_op(child, &proc_sys_dentry_operations); d_add(child, inode); } } else { @@ -294,7 +295,7 @@ out: return ret; } -static int proc_sys_permission(struct inode *inode, int mask) +static int proc_sys_permission(struct inode *inode, int mask,unsigned int flags) { /* * sysctl entries that are not writeable, @@ -304,6 +305,9 @@ static int proc_sys_permission(struct inode *inode, int mask) struct ctl_table *table; int error; + if (flags & IPERM_FLAG_RCU) + return -ECHILD; + /* Executable files are not allowed under /proc/sys/ */ if ((mask & MAY_EXEC) && S_ISREG(inode->i_mode)) return -EACCES; @@ -389,23 +393,30 @@ static const struct inode_operations proc_sys_dir_operations = { static int proc_sys_revalidate(struct dentry *dentry, struct nameidata *nd) { + if (nd->flags & LOOKUP_RCU) + return -ECHILD; return !PROC_I(dentry->d_inode)->sysctl->unregistering; } -static int proc_sys_delete(struct dentry *dentry) +static int proc_sys_delete(const struct dentry *dentry) { return !!PROC_I(dentry->d_inode)->sysctl->unregistering; } -static int proc_sys_compare(struct dentry *dir, struct qstr *qstr, - struct qstr *name) +static int proc_sys_compare(const struct dentry *parent, + const struct inode *pinode, + const struct dentry *dentry, const struct inode *inode, + unsigned int len, const char *str, const struct qstr *name) { - struct dentry *dentry = container_of(qstr, struct dentry, d_name); - if (qstr->len != name->len) + /* Although proc doesn't have negative dentries, rcu-walk means + * that inode here can be NULL */ + if (!inode) + return 0; + if (name->len != len) return 1; - if (memcmp(qstr->name, name->name, name->len)) + if (memcmp(name->name, str, len)) return 1; - return !sysctl_is_seen(PROC_I(dentry->d_inode)->sysctl); + return !sysctl_is_seen(PROC_I(inode)->sysctl); } static const struct dentry_operations proc_sys_dentry_operations = { diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index da6b01d70f0..c126c83b9a4 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -706,6 +706,7 @@ static int pagemap_hugetlb_range(pte_t *pte, unsigned long hmask, * skip over unmapped regions. */ #define PAGEMAP_WALK_SIZE (PMD_SIZE) +#define PAGEMAP_WALK_MASK (PMD_MASK) static ssize_t pagemap_read(struct file *file, char __user *buf, size_t count, loff_t *ppos) { @@ -776,7 +777,7 @@ static ssize_t pagemap_read(struct file *file, char __user *buf, unsigned long end; pm.pos = 0; - end = start_vaddr + PAGEMAP_WALK_SIZE; + end = (start_vaddr + PAGEMAP_WALK_SIZE) & PAGEMAP_WALK_MASK; /* overflow ? */ if (end < start_vaddr || end > end_vaddr) end = end_vaddr; diff --git a/fs/proc/vmcore.c b/fs/proc/vmcore.c index 2367fb3f70b..74802bc5ded 100644 --- a/fs/proc/vmcore.c +++ b/fs/proc/vmcore.c @@ -499,7 +499,7 @@ static int __init parse_crash_elf64_headers(void) /* Do some basic Verification. */ if (memcmp(ehdr.e_ident, ELFMAG, SELFMAG) != 0 || (ehdr.e_type != ET_CORE) || - !vmcore_elf_check_arch(&ehdr) || + !vmcore_elf64_check_arch(&ehdr) || ehdr.e_ident[EI_CLASS] != ELFCLASS64 || ehdr.e_ident[EI_VERSION] != EV_CURRENT || ehdr.e_version != EV_CURRENT || diff --git a/fs/qnx4/inode.c b/fs/qnx4/inode.c index fcada42f1aa..e63b4171d58 100644 --- a/fs/qnx4/inode.c +++ b/fs/qnx4/inode.c @@ -425,11 +425,18 @@ static struct inode *qnx4_alloc_inode(struct super_block *sb) return &ei->vfs_inode; } -static void qnx4_destroy_inode(struct inode *inode) +static void qnx4_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(qnx4_inode_cachep, qnx4_i(inode)); } +static void qnx4_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, qnx4_i_callback); +} + static void init_once(void *foo) { struct qnx4_inode_info *ei = (struct qnx4_inode_info *) foo; diff --git a/fs/read_write.c b/fs/read_write.c index 431a0ed610c..5d431bacbea 100644 --- a/fs/read_write.c +++ b/fs/read_write.c @@ -9,7 +9,6 @@ #include <linux/fcntl.h> #include <linux/file.h> #include <linux/uio.h> -#include <linux/smp_lock.h> #include <linux/fsnotify.h> #include <linux/security.h> #include <linux/module.h> diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c index 41656d40dc5..0bae036831e 100644 --- a/fs/reiserfs/inode.c +++ b/fs/reiserfs/inode.c @@ -8,7 +8,6 @@ #include <linux/reiserfs_acl.h> #include <linux/reiserfs_xattr.h> #include <linux/exportfs.h> -#include <linux/smp_lock.h> #include <linux/pagemap.h> #include <linux/highmem.h> #include <linux/slab.h> diff --git a/fs/reiserfs/ioctl.c b/fs/reiserfs/ioctl.c index adf22b485ce..79265fdc317 100644 --- a/fs/reiserfs/ioctl.c +++ b/fs/reiserfs/ioctl.c @@ -9,7 +9,6 @@ #include <linux/time.h> #include <asm/uaccess.h> #include <linux/pagemap.h> -#include <linux/smp_lock.h> #include <linux/compat.h> /* @@ -184,12 +183,11 @@ int reiserfs_unpack(struct inode *inode, struct file *filp) return 0; } - /* we need to make sure nobody is changing the file size beneath - ** us - */ - reiserfs_mutex_lock_safe(&inode->i_mutex, inode->i_sb); depth = reiserfs_write_lock_once(inode->i_sb); + /* we need to make sure nobody is changing the file size beneath us */ + reiserfs_mutex_lock_safe(&inode->i_mutex, inode->i_sb); + write_from = inode->i_size & (blocksize - 1); /* if we are on a block boundary, we are already unpacked. */ if (write_from == 0) { diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index 076c8b19468..d31bce1a9f9 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -43,7 +43,6 @@ #include <linux/fcntl.h> #include <linux/stat.h> #include <linux/string.h> -#include <linux/smp_lock.h> #include <linux/buffer_head.h> #include <linux/workqueue.h> #include <linux/writeback.h> diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index 3bf7a6457f4..2575682a9ea 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -28,7 +28,6 @@ #include <linux/mount.h> #include <linux/namei.h> #include <linux/crc32.h> -#include <linux/smp_lock.h> struct file_system_type reiserfs_fs_type; @@ -530,11 +529,18 @@ static struct inode *reiserfs_alloc_inode(struct super_block *sb) return &ei->vfs_inode; } -static void reiserfs_destroy_inode(struct inode *inode) +static void reiserfs_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(reiserfs_inode_cachep, REISERFS_I(inode)); } +static void reiserfs_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, reiserfs_i_callback); +} + static void init_once(void *foo) { struct reiserfs_inode_info *ei = (struct reiserfs_inode_info *)foo; diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c index 5d04a7828e7..3cfb2e93364 100644 --- a/fs/reiserfs/xattr.c +++ b/fs/reiserfs/xattr.c @@ -870,11 +870,14 @@ out: return err; } -static int reiserfs_check_acl(struct inode *inode, int mask) +static int reiserfs_check_acl(struct inode *inode, int mask, unsigned int flags) { struct posix_acl *acl; int error = -EAGAIN; /* do regular unix permission checks by default */ + if (flags & IPERM_FLAG_RCU) + return -ECHILD; + acl = reiserfs_get_acl(inode, ACL_TYPE_ACCESS); if (acl) { @@ -951,8 +954,10 @@ static int xattr_mount_check(struct super_block *s) return 0; } -int reiserfs_permission(struct inode *inode, int mask) +int reiserfs_permission(struct inode *inode, int mask, unsigned int flags) { + if (flags & IPERM_FLAG_RCU) + return -ECHILD; /* * We don't do permission checks on the internal objects. * Permissions are determined by the "owning" object. @@ -965,13 +970,16 @@ int reiserfs_permission(struct inode *inode, int mask) * Stat data v1 doesn't support ACLs. */ if (get_inode_sd_version(inode) != STAT_DATA_V1) - return generic_permission(inode, mask, reiserfs_check_acl); + return generic_permission(inode, mask, flags, + reiserfs_check_acl); #endif - return generic_permission(inode, mask, NULL); + return generic_permission(inode, mask, flags, NULL); } static int xattr_hide_revalidate(struct dentry *dentry, struct nameidata *nd) { + if (nd->flags & LOOKUP_RCU) + return -ECHILD; return -EPERM; } @@ -990,7 +998,7 @@ int reiserfs_lookup_privroot(struct super_block *s) strlen(PRIVROOT_NAME)); if (!IS_ERR(dentry)) { REISERFS_SB(s)->priv_root = dentry; - dentry->d_op = &xattr_lookup_poison_ops; + d_set_d_op(dentry, &xattr_lookup_poison_ops); if (dentry->d_inode) dentry->d_inode->i_flags |= S_PRIVATE; } else diff --git a/fs/reiserfs/xattr_acl.c b/fs/reiserfs/xattr_acl.c index 536d697a8a2..90d2fcb67a3 100644 --- a/fs/reiserfs/xattr_acl.c +++ b/fs/reiserfs/xattr_acl.c @@ -472,7 +472,9 @@ int reiserfs_acl_chmod(struct inode *inode) struct reiserfs_transaction_handle th; size_t size = reiserfs_xattr_nblocks(inode, reiserfs_acl_size(clone->a_count)); - reiserfs_write_lock(inode->i_sb); + int depth; + + depth = reiserfs_write_lock_once(inode->i_sb); error = journal_begin(&th, inode->i_sb, size * 2); if (!error) { int error2; @@ -482,7 +484,7 @@ int reiserfs_acl_chmod(struct inode *inode) if (error2) error = error2; } - reiserfs_write_unlock(inode->i_sb); + reiserfs_write_unlock_once(inode->i_sb, depth); } posix_acl_release(clone); return error; diff --git a/fs/romfs/super.c b/fs/romfs/super.c index 6647f90e55c..2305e3121cb 100644 --- a/fs/romfs/super.c +++ b/fs/romfs/super.c @@ -400,11 +400,18 @@ static struct inode *romfs_alloc_inode(struct super_block *sb) /* * return a spent inode to the slab cache */ -static void romfs_destroy_inode(struct inode *inode) +static void romfs_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(romfs_inode_cachep, ROMFS_I(inode)); } +static void romfs_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, romfs_i_callback); +} + /* * get filesystem statistics */ diff --git a/fs/splice.c b/fs/splice.c index 8f1dfaecc8f..ce2f02579e3 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -1311,18 +1311,6 @@ long do_splice_direct(struct file *in, loff_t *ppos, struct file *out, static int splice_pipe_to_pipe(struct pipe_inode_info *ipipe, struct pipe_inode_info *opipe, size_t len, unsigned int flags); -/* - * After the inode slimming patch, i_pipe/i_bdev/i_cdev share the same - * location, so checking ->i_pipe is not enough to verify that this is a - * pipe. - */ -static inline struct pipe_inode_info *pipe_info(struct inode *inode) -{ - if (S_ISFIFO(inode->i_mode)) - return inode->i_pipe; - - return NULL; -} /* * Determine where to splice to/from. @@ -1336,8 +1324,8 @@ static long do_splice(struct file *in, loff_t __user *off_in, loff_t offset, *off; long ret; - ipipe = pipe_info(in->f_path.dentry->d_inode); - opipe = pipe_info(out->f_path.dentry->d_inode); + ipipe = get_pipe_info(in); + opipe = get_pipe_info(out); if (ipipe && opipe) { if (off_in || off_out) @@ -1555,7 +1543,7 @@ static long vmsplice_to_user(struct file *file, const struct iovec __user *iov, int error; long ret; - pipe = pipe_info(file->f_path.dentry->d_inode); + pipe = get_pipe_info(file); if (!pipe) return -EBADF; @@ -1642,7 +1630,7 @@ static long vmsplice_to_pipe(struct file *file, const struct iovec __user *iov, }; long ret; - pipe = pipe_info(file->f_path.dentry->d_inode); + pipe = get_pipe_info(file); if (!pipe) return -EBADF; @@ -2022,8 +2010,8 @@ static int link_pipe(struct pipe_inode_info *ipipe, static long do_tee(struct file *in, struct file *out, size_t len, unsigned int flags) { - struct pipe_inode_info *ipipe = pipe_info(in->f_path.dentry->d_inode); - struct pipe_inode_info *opipe = pipe_info(out->f_path.dentry->d_inode); + struct pipe_inode_info *ipipe = get_pipe_info(in); + struct pipe_inode_info *opipe = get_pipe_info(out); int ret = -EINVAL; /* diff --git a/fs/squashfs/super.c b/fs/squashfs/super.c index 24de30ba34c..20700b9f2b4 100644 --- a/fs/squashfs/super.c +++ b/fs/squashfs/super.c @@ -440,11 +440,18 @@ static struct inode *squashfs_alloc_inode(struct super_block *sb) } -static void squashfs_destroy_inode(struct inode *inode) +static void squashfs_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(squashfs_inode_cachep, squashfs_i(inode)); } +static void squashfs_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, squashfs_i_callback); +} + static struct file_system_type squashfs_fs_type = { .owner = THIS_MODULE, diff --git a/fs/super.c b/fs/super.c index ca696155cd9..823e061faa8 100644 --- a/fs/super.c +++ b/fs/super.c @@ -30,6 +30,7 @@ #include <linux/idr.h> #include <linux/mutex.h> #include <linux/backing-dev.h> +#include <linux/rculist_bl.h> #include "internal.h" @@ -71,7 +72,7 @@ static struct super_block *alloc_super(struct file_system_type *type) INIT_LIST_HEAD(&s->s_files); #endif INIT_LIST_HEAD(&s->s_instances); - INIT_HLIST_HEAD(&s->s_anon); + INIT_HLIST_BL_HEAD(&s->s_anon); INIT_LIST_HEAD(&s->s_inodes); INIT_LIST_HEAD(&s->s_dentry_lru); init_rwsem(&s->s_umount); @@ -1139,7 +1140,7 @@ static struct vfsmount *fs_set_subtype(struct vfsmount *mnt, const char *fstype) return mnt; err: - mntput(mnt); + mntput_long(mnt); return ERR_PTR(err); } diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c index 7e54bac8c4b..ea9120a830d 100644 --- a/fs/sysfs/dir.c +++ b/fs/sysfs/dir.c @@ -231,7 +231,7 @@ void release_sysfs_dirent(struct sysfs_dirent * sd) goto repeat; } -static int sysfs_dentry_delete(struct dentry *dentry) +static int sysfs_dentry_delete(const struct dentry *dentry) { struct sysfs_dirent *sd = dentry->d_fsdata; return !!(sd->s_flags & SYSFS_FLAG_REMOVED); @@ -239,9 +239,13 @@ static int sysfs_dentry_delete(struct dentry *dentry) static int sysfs_dentry_revalidate(struct dentry *dentry, struct nameidata *nd) { - struct sysfs_dirent *sd = dentry->d_fsdata; + struct sysfs_dirent *sd; int is_dir; + if (nd->flags & LOOKUP_RCU) + return -ECHILD; + + sd = dentry->d_fsdata; mutex_lock(&sysfs_mutex); /* The sysfs dirent has been deleted */ @@ -701,7 +705,7 @@ static struct dentry * sysfs_lookup(struct inode *dir, struct dentry *dentry, /* instantiate and hash dentry */ ret = d_find_alias(inode); if (!ret) { - dentry->d_op = &sysfs_dentry_ops; + d_set_d_op(dentry, &sysfs_dentry_ops); dentry->d_fsdata = sysfs_get(sd); d_add(dentry, inode); } else { diff --git a/fs/sysfs/inode.c b/fs/sysfs/inode.c index cffb1fd8ba3..30ac2734558 100644 --- a/fs/sysfs/inode.c +++ b/fs/sysfs/inode.c @@ -348,13 +348,18 @@ int sysfs_hash_and_remove(struct sysfs_dirent *dir_sd, const void *ns, const cha return -ENOENT; } -int sysfs_permission(struct inode *inode, int mask) +int sysfs_permission(struct inode *inode, int mask, unsigned int flags) { - struct sysfs_dirent *sd = inode->i_private; + struct sysfs_dirent *sd; + + if (flags & IPERM_FLAG_RCU) + return -ECHILD; + + sd = inode->i_private; mutex_lock(&sysfs_mutex); sysfs_refresh_inode(sd, inode); mutex_unlock(&sysfs_mutex); - return generic_permission(inode, mask, NULL); + return generic_permission(inode, mask, flags, NULL); } diff --git a/fs/sysfs/sysfs.h b/fs/sysfs/sysfs.h index d9be60a2e95..ffaaa816bfb 100644 --- a/fs/sysfs/sysfs.h +++ b/fs/sysfs/sysfs.h @@ -200,7 +200,7 @@ static inline void __sysfs_put(struct sysfs_dirent *sd) struct inode *sysfs_get_inode(struct super_block *sb, struct sysfs_dirent *sd); void sysfs_evict_inode(struct inode *inode); int sysfs_sd_setattr(struct sysfs_dirent *sd, struct iattr *iattr); -int sysfs_permission(struct inode *inode, int mask); +int sysfs_permission(struct inode *inode, int mask, unsigned int flags); int sysfs_setattr(struct dentry *dentry, struct iattr *iattr); int sysfs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat); int sysfs_setxattr(struct dentry *dentry, const char *name, const void *value, diff --git a/fs/sysv/inode.c b/fs/sysv/inode.c index de44d067b9e..0630eb969a2 100644 --- a/fs/sysv/inode.c +++ b/fs/sysv/inode.c @@ -333,11 +333,18 @@ static struct inode *sysv_alloc_inode(struct super_block *sb) return &si->vfs_inode; } -static void sysv_destroy_inode(struct inode *inode) +static void sysv_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(sysv_inode_cachep, SYSV_I(inode)); } +static void sysv_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, sysv_i_callback); +} + static void init_once(void *p) { struct sysv_inode_info *si = (struct sysv_inode_info *)p; diff --git a/fs/sysv/namei.c b/fs/sysv/namei.c index 11e7f7d11cd..b5e68da2db3 100644 --- a/fs/sysv/namei.c +++ b/fs/sysv/namei.c @@ -27,7 +27,8 @@ static int add_nondir(struct dentry *dentry, struct inode *inode) return err; } -static int sysv_hash(struct dentry *dentry, struct qstr *qstr) +static int sysv_hash(const struct dentry *dentry, const struct inode *inode, + struct qstr *qstr) { /* Truncate the name in place, avoids having to define a compare function. */ @@ -47,7 +48,7 @@ static struct dentry *sysv_lookup(struct inode * dir, struct dentry * dentry, st struct inode * inode = NULL; ino_t ino; - dentry->d_op = dir->i_sb->s_root->d_op; + d_set_d_op(dentry, dir->i_sb->s_root->d_op); if (dentry->d_name.len > SYSV_NAMELEN) return ERR_PTR(-ENAMETOOLONG); ino = sysv_inode_by_name(dentry); diff --git a/fs/sysv/super.c b/fs/sysv/super.c index 3d9c62be0c1..76712aefc4a 100644 --- a/fs/sysv/super.c +++ b/fs/sysv/super.c @@ -346,7 +346,7 @@ static int complete_read_super(struct super_block *sb, int silent, int size) if (sbi->s_forced_ro) sb->s_flags |= MS_RDONLY; if (sbi->s_truncate) - sb->s_root->d_op = &sysv_dentry_operations; + d_set_d_op(sb->s_root, &sysv_dentry_operations); return 1; } diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c index 91fac54c70e..6e11c2975dc 100644 --- a/fs/ubifs/super.c +++ b/fs/ubifs/super.c @@ -272,12 +272,20 @@ static struct inode *ubifs_alloc_inode(struct super_block *sb) return &ui->vfs_inode; }; +static void ubifs_i_callback(struct rcu_head *head) +{ + struct inode *inode = container_of(head, struct inode, i_rcu); + struct ubifs_inode *ui = ubifs_inode(inode); + INIT_LIST_HEAD(&inode->i_dentry); + kmem_cache_free(ubifs_inode_slab, ui); +} + static void ubifs_destroy_inode(struct inode *inode) { struct ubifs_inode *ui = ubifs_inode(inode); kfree(ui->data); - kmem_cache_free(ubifs_inode_slab, inode); + call_rcu(&inode->i_rcu, ubifs_i_callback); } /* diff --git a/fs/udf/super.c b/fs/udf/super.c index 4a5c7c61836..b539d53320f 100644 --- a/fs/udf/super.c +++ b/fs/udf/super.c @@ -139,11 +139,18 @@ static struct inode *udf_alloc_inode(struct super_block *sb) return &ei->vfs_inode; } -static void udf_destroy_inode(struct inode *inode) +static void udf_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(udf_inode_cachep, UDF_I(inode)); } +static void udf_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, udf_i_callback); +} + static void init_once(void *foo) { struct udf_inode_info *ei = (struct udf_inode_info *)foo; diff --git a/fs/ufs/super.c b/fs/ufs/super.c index 2c47daed56d..2c61ac5d4e4 100644 --- a/fs/ufs/super.c +++ b/fs/ufs/super.c @@ -1412,11 +1412,18 @@ static struct inode *ufs_alloc_inode(struct super_block *sb) return &ei->vfs_inode; } -static void ufs_destroy_inode(struct inode *inode) +static void ufs_i_callback(struct rcu_head *head) { + struct inode *inode = container_of(head, struct inode, i_rcu); + INIT_LIST_HEAD(&inode->i_dentry); kmem_cache_free(ufs_inode_cachep, UFS_I(inode)); } +static void ufs_destroy_inode(struct inode *inode) +{ + call_rcu(&inode->i_rcu, ufs_i_callback); +} + static void init_once(void *foo) { struct ufs_inode_info *ei = (struct ufs_inode_info *) foo; diff --git a/fs/xfs/linux-2.6/xfs_acl.c b/fs/xfs/linux-2.6/xfs_acl.c index b2771862fd3..39f4f809bb6 100644 --- a/fs/xfs/linux-2.6/xfs_acl.c +++ b/fs/xfs/linux-2.6/xfs_acl.c @@ -219,12 +219,13 @@ xfs_set_acl(struct inode *inode, int type, struct posix_acl *acl) } int -xfs_check_acl(struct inode *inode, int mask) +xfs_check_acl(struct inode *inode, int mask, unsigned int flags) { - struct xfs_inode *ip = XFS_I(inode); + struct xfs_inode *ip; struct posix_acl *acl; int error = -EAGAIN; + ip = XFS_I(inode); trace_xfs_check_acl(ip); /* @@ -234,6 +235,12 @@ xfs_check_acl(struct inode *inode, int mask) if (!XFS_IFORK_Q(ip)) return -EAGAIN; + if (flags & IPERM_FLAG_RCU) { + if (!negative_cached_acl(inode, ACL_TYPE_ACCESS)) + return -ECHILD; + return -EAGAIN; + } + acl = xfs_get_acl(inode, ACL_TYPE_ACCESS); if (IS_ERR(acl)) return PTR_ERR(acl); diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c index c9af48fffcd..691f61223ed 100644 --- a/fs/xfs/linux-2.6/xfs_aops.c +++ b/fs/xfs/linux-2.6/xfs_aops.c @@ -934,7 +934,6 @@ xfs_aops_discard_page( struct xfs_inode *ip = XFS_I(inode); struct buffer_head *bh, *head; loff_t offset = page_offset(page); - ssize_t len = 1 << inode->i_blkbits; if (!xfs_is_delayed_page(page, IO_DELAY)) goto out_invalidate; @@ -949,58 +948,14 @@ xfs_aops_discard_page( xfs_ilock(ip, XFS_ILOCK_EXCL); bh = head = page_buffers(page); do { - int done; - xfs_fileoff_t offset_fsb; - xfs_bmbt_irec_t imap; - int nimaps = 1; int error; - xfs_fsblock_t firstblock; - xfs_bmap_free_t flist; + xfs_fileoff_t start_fsb; if (!buffer_delay(bh)) goto next_buffer; - offset_fsb = XFS_B_TO_FSBT(ip->i_mount, offset); - - /* - * Map the range first and check that it is a delalloc extent - * before trying to unmap the range. Otherwise we will be - * trying to remove a real extent (which requires a - * transaction) or a hole, which is probably a bad idea... - */ - error = xfs_bmapi(NULL, ip, offset_fsb, 1, - XFS_BMAPI_ENTIRE, NULL, 0, &imap, - &nimaps, NULL); - - if (error) { - /* something screwed, just bail */ - if (!XFS_FORCED_SHUTDOWN(ip->i_mount)) { - xfs_fs_cmn_err(CE_ALERT, ip->i_mount, - "page discard failed delalloc mapping lookup."); - } - break; - } - if (!nimaps) { - /* nothing there */ - goto next_buffer; - } - if (imap.br_startblock != DELAYSTARTBLOCK) { - /* been converted, ignore */ - goto next_buffer; - } - WARN_ON(imap.br_blockcount == 0); - - /* - * Note: while we initialise the firstblock/flist pair, they - * should never be used because blocks should never be - * allocated or freed for a delalloc extent and hence we need - * don't cancel or finish them after the xfs_bunmapi() call. - */ - xfs_bmap_init(&flist, &firstblock); - error = xfs_bunmapi(NULL, ip, offset_fsb, 1, 0, 1, &firstblock, - &flist, &done); - - ASSERT(!flist.xbf_count && !flist.xbf_first); + start_fsb = XFS_B_TO_FSBT(ip->i_mount, offset); + error = xfs_bmap_punch_delalloc_range(ip, start_fsb, 1); if (error) { /* something screwed, just bail */ if (!XFS_FORCED_SHUTDOWN(ip->i_mount)) { @@ -1010,7 +965,7 @@ xfs_aops_discard_page( break; } next_buffer: - offset += len; + offset += 1 << inode->i_blkbits; } while ((bh = bh->b_this_page) != head); @@ -1111,11 +1066,12 @@ xfs_vm_writepage( uptodate = 0; /* - * A hole may still be marked uptodate because discard_buffer - * leaves the flag set. + * set_page_dirty dirties all buffers in a page, independent + * of their state. The dirty state however is entirely + * meaningless for holes (!mapped && uptodate), so skip + * buffers covering holes here. */ if (!buffer_mapped(bh) && buffer_uptodate(bh)) { - ASSERT(!buffer_dirty(bh)); imap_valid = 0; continue; } @@ -1504,11 +1460,42 @@ xfs_vm_write_failed( struct inode *inode = mapping->host; if (to > inode->i_size) { - struct iattr ia = { - .ia_valid = ATTR_SIZE | ATTR_FORCE, - .ia_size = inode->i_size, - }; - xfs_setattr(XFS_I(inode), &ia, XFS_ATTR_NOLOCK); + /* + * punch out the delalloc blocks we have already allocated. We + * don't call xfs_setattr() to do this as we may be in the + * middle of a multi-iovec write and so the vfs inode->i_size + * will not match the xfs ip->i_size and so it will zero too + * much. Hence we jus truncate the page cache to zero what is + * necessary and punch the delalloc blocks directly. + */ + struct xfs_inode *ip = XFS_I(inode); + xfs_fileoff_t start_fsb; + xfs_fileoff_t end_fsb; + int error; + + truncate_pagecache(inode, to, inode->i_size); + + /* + * Check if there are any blocks that are outside of i_size + * that need to be trimmed back. + */ + start_fsb = XFS_B_TO_FSB(ip->i_mount, inode->i_size) + 1; + end_fsb = XFS_B_TO_FSB(ip->i_mount, to); + if (end_fsb <= start_fsb) + return; + + xfs_ilock(ip, XFS_ILOCK_EXCL); + error = xfs_bmap_punch_delalloc_range(ip, start_fsb, + end_fsb - start_fsb); + if (error) { + /* something screwed, just bail */ + if (!XFS_FORCED_SHUTDOWN(ip->i_mount)) { + xfs_fs_cmn_err(CE_ALERT, ip->i_mount, + "xfs_vm_write_failed: unable to clean up ino %lld", + ip->i_ino); + } + } + xfs_iunlock(ip, XFS_ILOCK_EXCL); } } diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c index 63fd2c07cb5..4c5deb6e9e3 100644 --- a/fs/xfs/linux-2.6/xfs_buf.c +++ b/fs/xfs/linux-2.6/xfs_buf.c @@ -488,29 +488,16 @@ found: spin_unlock(&pag->pag_buf_lock); xfs_perag_put(pag); - /* Attempt to get the semaphore without sleeping, - * if this does not work then we need to drop the - * spinlock and do a hard attempt on the semaphore. - */ - if (down_trylock(&bp->b_sema)) { + if (xfs_buf_cond_lock(bp)) { + /* failed, so wait for the lock if requested. */ if (!(flags & XBF_TRYLOCK)) { - /* wait for buffer ownership */ xfs_buf_lock(bp); XFS_STATS_INC(xb_get_locked_waited); } else { - /* We asked for a trylock and failed, no need - * to look at file offset and length here, we - * know that this buffer at least overlaps our - * buffer and is locked, therefore our buffer - * either does not exist, or is this buffer. - */ xfs_buf_rele(bp); XFS_STATS_INC(xb_busy_locked); return NULL; } - } else { - /* trylock worked */ - XB_SET_OWNER(bp); } if (bp->b_flags & XBF_STALE) { @@ -876,10 +863,18 @@ xfs_buf_rele( */ /* - * Locks a buffer object, if it is not already locked. - * Note that this in no way locks the underlying pages, so it is only - * useful for synchronizing concurrent use of buffer objects, not for - * synchronizing independent access to the underlying pages. + * Locks a buffer object, if it is not already locked. Note that this in + * no way locks the underlying pages, so it is only useful for + * synchronizing concurrent use of buffer objects, not for synchronizing + * independent access to the underlying pages. + * + * If we come across a stale, pinned, locked buffer, we know that we are + * being asked to lock a buffer that has been reallocated. Because it is + * pinned, we know that the log has not been pushed to disk and hence it + * will still be locked. Rather than continuing to have trylock attempts + * fail until someone else pushes the log, push it ourselves before + * returning. This means that the xfsaild will not get stuck trying + * to push on stale inode buffers. */ int xfs_buf_cond_lock( @@ -890,6 +885,8 @@ xfs_buf_cond_lock( locked = down_trylock(&bp->b_sema) == 0; if (locked) XB_SET_OWNER(bp); + else if (atomic_read(&bp->b_pin_count) && (bp->b_flags & XBF_STALE)) + xfs_log_force(bp->b_target->bt_mount, 0); trace_xfs_buf_cond_lock(bp, _RET_IP_); return locked ? 0 : -EBUSY; @@ -1781,7 +1778,6 @@ xfs_buf_delwri_split( INIT_LIST_HEAD(list); spin_lock(dwlk); list_for_each_entry_safe(bp, n, dwq, b_list) { - trace_xfs_buf_delwri_split(bp, _RET_IP_); ASSERT(bp->b_flags & XBF_DELWRI); if (!XFS_BUF_ISPINNED(bp) && !xfs_buf_cond_lock(bp)) { @@ -1795,6 +1791,7 @@ xfs_buf_delwri_split( _XBF_RUN_QUEUES); bp->b_flags |= XBF_WRITE; list_move_tail(&bp->b_list, list); + trace_xfs_buf_delwri_split(bp, _RET_IP_); } else skipped++; } diff --git a/fs/xfs/linux-2.6/xfs_ioctl.c b/fs/xfs/linux-2.6/xfs_ioctl.c index 2ea238f6d38..ad442d9e392 100644 --- a/fs/xfs/linux-2.6/xfs_ioctl.c +++ b/fs/xfs/linux-2.6/xfs_ioctl.c @@ -416,7 +416,7 @@ xfs_attrlist_by_handle( if (IS_ERR(dentry)) return PTR_ERR(dentry); - kbuf = kmalloc(al_hreq.buflen, GFP_KERNEL); + kbuf = kzalloc(al_hreq.buflen, GFP_KERNEL); if (!kbuf) goto out_dput; diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c index 96107efc0c6..94d5fd6a297 100644 --- a/fs/xfs/linux-2.6/xfs_iops.c +++ b/fs/xfs/linux-2.6/xfs_iops.c @@ -762,7 +762,8 @@ xfs_setup_inode( inode->i_state = I_NEW; inode_sb_list_add(inode); - insert_inode_hash(inode); + /* make the inode look hashed for the writeback code */ + hlist_add_fake(&inode->i_hash); inode->i_mode = ip->i_d.di_mode; inode->i_nlink = ip->i_d.di_nlink; diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c index 9f3a78fe6ae..064f964d4f3 100644 --- a/fs/xfs/linux-2.6/xfs_super.c +++ b/fs/xfs/linux-2.6/xfs_super.c @@ -353,9 +353,6 @@ xfs_parseargs( mp->m_qflags &= ~XFS_OQUOTA_ENFD; } else if (!strcmp(this_char, MNTOPT_DELAYLOG)) { mp->m_flags |= XFS_MOUNT_DELAYLOG; - cmn_err(CE_WARN, - "Enabling EXPERIMENTAL delayed logging feature " - "- use at your own risk.\n"); } else if (!strcmp(this_char, MNTOPT_NODELAYLOG)) { mp->m_flags &= ~XFS_MOUNT_DELAYLOG; } else if (!strcmp(this_char, "ihashsize")) { diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c index 37d33254981..afb0d7cfad1 100644 --- a/fs/xfs/linux-2.6/xfs_sync.c +++ b/fs/xfs/linux-2.6/xfs_sync.c @@ -853,6 +853,7 @@ restart: if (trylock) { if (!mutex_trylock(&pag->pag_ici_reclaim_lock)) { skipped++; + xfs_perag_put(pag); continue; } first_index = pag->pag_ici_reclaim_cursor; diff --git a/fs/xfs/xfs_acl.h b/fs/xfs/xfs_acl.h index 0135e2a669d..11dd72070cb 100644 --- a/fs/xfs/xfs_acl.h +++ b/fs/xfs/xfs_acl.h @@ -42,7 +42,7 @@ struct xfs_acl { #define SGI_ACL_DEFAULT_SIZE (sizeof(SGI_ACL_DEFAULT)-1) #ifdef CONFIG_XFS_POSIX_ACL -extern int xfs_check_acl(struct inode *inode, int mask); +extern int xfs_check_acl(struct inode *inode, int mask, unsigned int flags); extern struct posix_acl *xfs_get_acl(struct inode *inode, int type); extern int xfs_inherit_acl(struct inode *inode, struct posix_acl *default_acl); extern int xfs_acl_chmod(struct inode *inode); diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c index 8abd12e32e1..4111cd3966c 100644 --- a/fs/xfs/xfs_bmap.c +++ b/fs/xfs/xfs_bmap.c @@ -5471,8 +5471,13 @@ xfs_getbmap( if (error) goto out_unlock_iolock; } - - ASSERT(ip->i_delayed_blks == 0); + /* + * even after flushing the inode, there can still be delalloc + * blocks on the inode beyond EOF due to speculative + * preallocation. These are not removed until the release + * function is called or the inode is inactivated. Hence we + * cannot assert here that ip->i_delayed_blks == 0. + */ } lock = xfs_ilock_map_shared(ip); @@ -6070,3 +6075,79 @@ xfs_bmap_disk_count_leaves( *count += xfs_bmbt_disk_get_blockcount(frp); } } + +/* + * dead simple method of punching delalyed allocation blocks from a range in + * the inode. Walks a block at a time so will be slow, but is only executed in + * rare error cases so the overhead is not critical. This will alays punch out + * both the start and end blocks, even if the ranges only partially overlap + * them, so it is up to the caller to ensure that partial blocks are not + * passed in. + */ +int +xfs_bmap_punch_delalloc_range( + struct xfs_inode *ip, + xfs_fileoff_t start_fsb, + xfs_fileoff_t length) +{ + xfs_fileoff_t remaining = length; + int error = 0; + + ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); + + do { + int done; + xfs_bmbt_irec_t imap; + int nimaps = 1; + xfs_fsblock_t firstblock; + xfs_bmap_free_t flist; + + /* + * Map the range first and check that it is a delalloc extent + * before trying to unmap the range. Otherwise we will be + * trying to remove a real extent (which requires a + * transaction) or a hole, which is probably a bad idea... + */ + error = xfs_bmapi(NULL, ip, start_fsb, 1, + XFS_BMAPI_ENTIRE, NULL, 0, &imap, + &nimaps, NULL); + + if (error) { + /* something screwed, just bail */ + if (!XFS_FORCED_SHUTDOWN(ip->i_mount)) { + xfs_fs_cmn_err(CE_ALERT, ip->i_mount, + "Failed delalloc mapping lookup ino %lld fsb %lld.", + ip->i_ino, start_fsb); + } + break; + } + if (!nimaps) { + /* nothing there */ + goto next_block; + } + if (imap.br_startblock != DELAYSTARTBLOCK) { + /* been converted, ignore */ + goto next_block; + } + WARN_ON(imap.br_blockcount == 0); + + /* + * Note: while we initialise the firstblock/flist pair, they + * should never be used because blocks should never be + * allocated or freed for a delalloc extent and hence we need + * don't cancel or finish them after the xfs_bunmapi() call. + */ + xfs_bmap_init(&flist, &firstblock); + error = xfs_bunmapi(NULL, ip, start_fsb, 1, 0, 1, &firstblock, + &flist, &done); + if (error) + break; + + ASSERT(!flist.xbf_count && !flist.xbf_first); +next_block: + start_fsb++; + remaining--; + } while(remaining > 0); + + return error; +} diff --git a/fs/xfs/xfs_bmap.h b/fs/xfs/xfs_bmap.h index 71ec9b6ecdf..3651191daea 100644 --- a/fs/xfs/xfs_bmap.h +++ b/fs/xfs/xfs_bmap.h @@ -394,6 +394,11 @@ xfs_bmap_count_blocks( int whichfork, int *count); +int +xfs_bmap_punch_delalloc_range( + struct xfs_inode *ip, + xfs_fileoff_t start_fsb, + xfs_fileoff_t length); #endif /* __KERNEL__ */ #endif /* __XFS_BMAP_H__ */ diff --git a/fs/xfs/xfs_dfrag.c b/fs/xfs/xfs_dfrag.c index 3b9582c60a2..e60490bc00a 100644 --- a/fs/xfs/xfs_dfrag.c +++ b/fs/xfs/xfs_dfrag.c @@ -377,6 +377,19 @@ xfs_swap_extents( ip->i_d.di_format = tip->i_d.di_format; tip->i_d.di_format = tmp; + /* + * The extents in the source inode could still contain speculative + * preallocation beyond EOF (e.g. the file is open but not modified + * while defrag is in progress). In that case, we need to copy over the + * number of delalloc blocks the data fork in the source inode is + * tracking beyond EOF so that when the fork is truncated away when the + * temporary inode is unlinked we don't underrun the i_delayed_blks + * counter on that inode. + */ + ASSERT(tip->i_delayed_blks == 0); + tip->i_delayed_blks = ip->i_delayed_blks; + ip->i_delayed_blks = 0; + ilf_fields = XFS_ILOG_CORE; switch(ip->i_d.di_format) { diff --git a/fs/xfs/xfs_error.c b/fs/xfs/xfs_error.c index ed999026766..c78cc6a3d87 100644 --- a/fs/xfs/xfs_error.c +++ b/fs/xfs/xfs_error.c @@ -58,6 +58,7 @@ xfs_error_trap(int e) int xfs_etest[XFS_NUM_INJECT_ERROR]; int64_t xfs_etest_fsid[XFS_NUM_INJECT_ERROR]; char * xfs_etest_fsname[XFS_NUM_INJECT_ERROR]; +int xfs_error_test_active; int xfs_error_test(int error_tag, int *fsidp, char *expression, @@ -108,6 +109,7 @@ xfs_errortag_add(int error_tag, xfs_mount_t *mp) len = strlen(mp->m_fsname); xfs_etest_fsname[i] = kmem_alloc(len + 1, KM_SLEEP); strcpy(xfs_etest_fsname[i], mp->m_fsname); + xfs_error_test_active++; return 0; } } @@ -137,6 +139,7 @@ xfs_errortag_clearall(xfs_mount_t *mp, int loud) xfs_etest_fsid[i] = 0LL; kmem_free(xfs_etest_fsname[i]); xfs_etest_fsname[i] = NULL; + xfs_error_test_active--; } } diff --git a/fs/xfs/xfs_error.h b/fs/xfs/xfs_error.h index c2c1a072bb8..f338847f80b 100644 --- a/fs/xfs/xfs_error.h +++ b/fs/xfs/xfs_error.h @@ -127,13 +127,14 @@ extern void xfs_corruption_error(const char *tag, int level, #define XFS_RANDOM_BMAPIFORMAT XFS_RANDOM_DEFAULT #ifdef DEBUG +extern int xfs_error_test_active; extern int xfs_error_test(int, int *, char *, int, char *, unsigned long); #define XFS_NUM_INJECT_ERROR 10 #define XFS_TEST_ERROR(expr, mp, tag, rf) \ - ((expr) || \ + ((expr) || (xfs_error_test_active && \ xfs_error_test((tag), (mp)->m_fixedfsid, "expr", __LINE__, __FILE__, \ - (rf))) + (rf)))) extern int xfs_errortag_add(int error_tag, xfs_mount_t *mp); extern int xfs_errortag_clearall(xfs_mount_t *mp, int loud); diff --git a/fs/xfs/xfs_filestream.c b/fs/xfs/xfs_filestream.c index 9b715dce569..9124425b7f2 100644 --- a/fs/xfs/xfs_filestream.c +++ b/fs/xfs/xfs_filestream.c @@ -744,9 +744,15 @@ xfs_filestream_new_ag( * If the file's parent directory is known, take its iolock in exclusive * mode to prevent two sibling files from racing each other to migrate * themselves and their parent to different AGs. + * + * Note that we lock the parent directory iolock inside the child + * iolock here. That's fine as we never hold both parent and child + * iolock in any other place. This is different from the ilock, + * which requires locking of the child after the parent for namespace + * operations. */ if (pip) - xfs_ilock(pip, XFS_IOLOCK_EXCL); + xfs_ilock(pip, XFS_IOLOCK_EXCL | XFS_IOLOCK_PARENT); /* * A new AG needs to be found for the file. If the file's parent diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c index 0cdd26932d8..d7de5a3f786 100644 --- a/fs/xfs/xfs_iget.c +++ b/fs/xfs/xfs_iget.c @@ -91,6 +91,17 @@ xfs_inode_alloc( return ip; } +STATIC void +xfs_inode_free_callback( + struct rcu_head *head) +{ + struct inode *inode = container_of(head, struct inode, i_rcu); + struct xfs_inode *ip = XFS_I(inode); + + INIT_LIST_HEAD(&inode->i_dentry); + kmem_zone_free(xfs_inode_zone, ip); +} + void xfs_inode_free( struct xfs_inode *ip) @@ -134,7 +145,7 @@ xfs_inode_free( ASSERT(!spin_is_locked(&ip->i_flags_lock)); ASSERT(completion_done(&ip->i_flush)); - kmem_zone_free(xfs_inode_zone, ip); + call_rcu(&ip->i_vnode.i_rcu, xfs_inode_free_callback); } /* diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c index c7ac020705d..7c8d30c453c 100644 --- a/fs/xfs/xfs_inode_item.c +++ b/fs/xfs/xfs_inode_item.c @@ -657,18 +657,37 @@ xfs_inode_item_unlock( } /* - * This is called to find out where the oldest active copy of the - * inode log item in the on disk log resides now that the last log - * write of it completed at the given lsn. Since we always re-log - * all dirty data in an inode, the latest copy in the on disk log - * is the only one that matters. Therefore, simply return the - * given lsn. + * This is called to find out where the oldest active copy of the inode log + * item in the on disk log resides now that the last log write of it completed + * at the given lsn. Since we always re-log all dirty data in an inode, the + * latest copy in the on disk log is the only one that matters. Therefore, + * simply return the given lsn. + * + * If the inode has been marked stale because the cluster is being freed, we + * don't want to (re-)insert this inode into the AIL. There is a race condition + * where the cluster buffer may be unpinned before the inode is inserted into + * the AIL during transaction committed processing. If the buffer is unpinned + * before the inode item has been committed and inserted, then it is possible + * for the buffer to be written and IO completions before the inode is inserted + * into the AIL. In that case, we'd be inserting a clean, stale inode into the + * AIL which will never get removed. It will, however, get reclaimed which + * triggers an assert in xfs_inode_free() complaining about freein an inode + * still in the AIL. + * + * To avoid this, return a lower LSN than the one passed in so that the + * transaction committed code will not move the inode forward in the AIL but + * will still unpin it properly. */ STATIC xfs_lsn_t xfs_inode_item_committed( struct xfs_log_item *lip, xfs_lsn_t lsn) { + struct xfs_inode_log_item *iip = INODE_ITEM(lip); + struct xfs_inode *ip = iip->ili_inode; + + if (xfs_iflags_test(ip, XFS_ISTALE)) + return lsn - 1; return lsn; } diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index b1498ab5a39..19e9dfa1c25 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -275,6 +275,7 @@ xfs_free_perag( pag = radix_tree_delete(&mp->m_perag_tree, agno); spin_unlock(&mp->m_perag_lock); ASSERT(pag); + ASSERT(atomic_read(&pag->pag_ref) == 0); call_rcu(&pag->rcu_head, __xfs_free_perag); } } diff --git a/fs/xfs/xfs_quota.h b/fs/xfs/xfs_quota.h index e0e64b113bd..9bb6eda4cd2 100644 --- a/fs/xfs/xfs_quota.h +++ b/fs/xfs/xfs_quota.h @@ -346,8 +346,17 @@ xfs_qm_vop_dqalloc(struct xfs_inode *ip, uid_t uid, gid_t gid, prid_t prid, #define xfs_trans_mod_dquot_byino(tp, ip, fields, delta) #define xfs_trans_apply_dquot_deltas(tp) #define xfs_trans_unreserve_and_mod_dquots(tp) -#define xfs_trans_reserve_quota_nblks(tp, ip, nblks, ninos, flags) (0) -#define xfs_trans_reserve_quota_bydquots(tp, mp, u, g, nb, ni, fl) (0) +static inline int xfs_trans_reserve_quota_nblks(struct xfs_trans *tp, + struct xfs_inode *ip, long nblks, long ninos, uint flags) +{ + return 0; +} +static inline int xfs_trans_reserve_quota_bydquots(struct xfs_trans *tp, + struct xfs_mount *mp, struct xfs_dquot *udqp, + struct xfs_dquot *gdqp, long nblks, long nions, uint flags) +{ + return 0; +} #define xfs_qm_vop_create_dqattach(tp, ip, u, g) #define xfs_qm_vop_rename_dqattach(it) (0) #define xfs_qm_vop_chown(tp, ip, old, new) (NULL) @@ -357,11 +366,14 @@ xfs_qm_vop_dqalloc(struct xfs_inode *ip, uid_t uid, gid_t gid, prid_t prid, #define xfs_qm_dqdetach(ip) #define xfs_qm_dqrele(d) #define xfs_qm_statvfs(ip, s) -#define xfs_qm_sync(mp, fl) (0) +static inline int xfs_qm_sync(struct xfs_mount *mp, int flags) +{ + return 0; +} #define xfs_qm_newmount(mp, a, b) (0) #define xfs_qm_mount_quotas(mp) #define xfs_qm_unmount(mp) -#define xfs_qm_unmount_quotas(mp) (0) +#define xfs_qm_unmount_quotas(mp) #endif /* CONFIG_XFS_QUOTA */ #define xfs_trans_unreserve_quota_nblks(tp, ip, nblks, ninos, flags) \ diff --git a/fs/xfs/xfs_rename.c b/fs/xfs/xfs_rename.c index d2af0a8381a..77a59891734 100644 --- a/fs/xfs/xfs_rename.c +++ b/fs/xfs/xfs_rename.c @@ -297,6 +297,7 @@ xfs_rename( * it and some incremental backup programs won't work without it. */ xfs_trans_ichgtime(tp, src_ip, XFS_ICHGTIME_CHG); + xfs_trans_log_inode(tp, src_ip, XFS_ILOG_CORE); /* * Adjust the link count on src_dp. This is necessary when |