summaryrefslogtreecommitdiff
path: root/fs
diff options
context:
space:
mode:
Diffstat (limited to 'fs')
-rw-r--r--fs/9p/vfs_inode.c41
-rw-r--r--fs/cifs/cifs_dfs_ref.c3
-rw-r--r--fs/configfs/symlink.c4
-rw-r--r--fs/ecryptfs/inode.c24
-rw-r--r--fs/fcntl.c102
-rw-r--r--fs/hppfs/hppfs.c18
-rw-r--r--fs/namei.c23
-rw-r--r--fs/namespace.c14
-rw-r--r--fs/notify/inotify/inotify_fsnotify.c2
-rw-r--r--fs/notify/inotify/inotify_user.c4
-rw-r--r--fs/proc/base.c1
-rw-r--r--fs/ramfs/file-nommu.c26
-rw-r--r--fs/xfs/linux-2.6/xfs_super.c14
-rw-r--r--fs/xfs/linux-2.6/xfs_sync.c183
-rw-r--r--fs/xfs/linux-2.6/xfs_sync.h2
-rw-r--r--fs/xfs/quota/xfs_qm_syscalls.c2
-rw-r--r--fs/xfs/xfs_dfrag.c106
-rw-r--r--fs/xfs/xfs_iget.c1
-rw-r--r--fs/xfs/xfs_inode.c21
-rw-r--r--fs/xfs/xfs_rtalloc.c2
20 files changed, 322 insertions, 271 deletions
diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c
index 18f74ec4dce9..9d03d1ebca6f 100644
--- a/fs/9p/vfs_inode.c
+++ b/fs/9p/vfs_inode.c
@@ -1001,44 +1001,6 @@ done:
}
/**
- * v9fs_vfs_readlink - read a symlink's location
- * @dentry: dentry for symlink
- * @buffer: buffer to load symlink location into
- * @buflen: length of buffer
- *
- */
-
-static int v9fs_vfs_readlink(struct dentry *dentry, char __user * buffer,
- int buflen)
-{
- int retval;
- int ret;
- char *link = __getname();
-
- if (unlikely(!link))
- return -ENOMEM;
-
- if (buflen > PATH_MAX)
- buflen = PATH_MAX;
-
- P9_DPRINTK(P9_DEBUG_VFS, " dentry: %s (%p)\n", dentry->d_name.name,
- dentry);
-
- retval = v9fs_readlink(dentry, link, buflen);
-
- if (retval > 0) {
- if ((ret = copy_to_user(buffer, link, retval)) != 0) {
- P9_DPRINTK(P9_DEBUG_ERROR,
- "problem copying to user: %d\n", ret);
- retval = ret;
- }
- }
-
- __putname(link);
- return retval;
-}
-
-/**
* v9fs_vfs_follow_link - follow a symlink path
* @dentry: dentry for symlink
* @nd: nameidata
@@ -1230,7 +1192,6 @@ static const struct inode_operations v9fs_dir_inode_operations_ext = {
.rmdir = v9fs_vfs_rmdir,
.mknod = v9fs_vfs_mknod,
.rename = v9fs_vfs_rename,
- .readlink = v9fs_vfs_readlink,
.getattr = v9fs_vfs_getattr,
.setattr = v9fs_vfs_setattr,
};
@@ -1253,7 +1214,7 @@ static const struct inode_operations v9fs_file_inode_operations = {
};
static const struct inode_operations v9fs_symlink_inode_operations = {
- .readlink = v9fs_vfs_readlink,
+ .readlink = generic_readlink,
.follow_link = v9fs_vfs_follow_link,
.put_link = v9fs_vfs_put_link,
.getattr = v9fs_vfs_getattr,
diff --git a/fs/cifs/cifs_dfs_ref.c b/fs/cifs/cifs_dfs_ref.c
index fea9e898c4ba..b44ce0a0711c 100644
--- a/fs/cifs/cifs_dfs_ref.c
+++ b/fs/cifs/cifs_dfs_ref.c
@@ -269,7 +269,7 @@ static int add_mount_helper(struct vfsmount *newmnt, struct nameidata *nd,
int err;
mntget(newmnt);
- err = do_add_mount(newmnt, &nd->path, nd->path.mnt->mnt_flags, mntlist);
+ err = do_add_mount(newmnt, &nd->path, nd->path.mnt->mnt_flags | MNT_SHRINKABLE, mntlist);
switch (err) {
case 0:
path_put(&nd->path);
@@ -371,7 +371,6 @@ cifs_dfs_follow_mountpoint(struct dentry *dentry, struct nameidata *nd)
if (IS_ERR(mnt))
goto out_err;
- nd->path.mnt->mnt_flags |= MNT_SHRINKABLE;
rc = add_mount_helper(mnt, nd, &cifs_dfs_automount_list);
out:
diff --git a/fs/configfs/symlink.c b/fs/configfs/symlink.c
index c8afa6b1d91d..32a5f46b1157 100644
--- a/fs/configfs/symlink.c
+++ b/fs/configfs/symlink.c
@@ -121,8 +121,10 @@ static int get_target(const char *symname, struct path *path,
ret = -ENOENT;
path_put(path);
}
- } else
+ } else {
ret = -EPERM;
+ path_put(path);
+ }
}
return ret;
diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c
index 429ca0b3ba08..7f8545032930 100644
--- a/fs/ecryptfs/inode.c
+++ b/fs/ecryptfs/inode.c
@@ -715,31 +715,31 @@ static void *ecryptfs_follow_link(struct dentry *dentry, struct nameidata *nd)
/* Released in ecryptfs_put_link(); only release here on error */
buf = kmalloc(len, GFP_KERNEL);
if (!buf) {
- rc = -ENOMEM;
+ buf = ERR_PTR(-ENOMEM);
goto out;
}
old_fs = get_fs();
set_fs(get_ds());
rc = dentry->d_inode->i_op->readlink(dentry, (char __user *)buf, len);
set_fs(old_fs);
- if (rc < 0)
- goto out_free;
- else
+ if (rc < 0) {
+ kfree(buf);
+ buf = ERR_PTR(rc);
+ } else
buf[rc] = '\0';
- rc = 0;
- nd_set_link(nd, buf);
- goto out;
-out_free:
- kfree(buf);
out:
- return ERR_PTR(rc);
+ nd_set_link(nd, buf);
+ return NULL;
}
static void
ecryptfs_put_link(struct dentry *dentry, struct nameidata *nd, void *ptr)
{
- /* Free the char* */
- kfree(nd_get_link(nd));
+ char *buf = nd_get_link(nd);
+ if (!IS_ERR(buf)) {
+ /* Free the char* */
+ kfree(buf);
+ }
}
/**
diff --git a/fs/fcntl.c b/fs/fcntl.c
index 2cf93ec40a67..97e01dc0d95f 100644
--- a/fs/fcntl.c
+++ b/fs/fcntl.c
@@ -618,60 +618,90 @@ static DEFINE_RWLOCK(fasync_lock);
static struct kmem_cache *fasync_cache __read_mostly;
/*
- * fasync_helper() is used by almost all character device drivers
- * to set up the fasync queue. It returns negative on error, 0 if it did
- * no changes and positive if it added/deleted the entry.
+ * Remove a fasync entry. If successfully removed, return
+ * positive and clear the FASYNC flag. If no entry exists,
+ * do nothing and return 0.
+ *
+ * NOTE! It is very important that the FASYNC flag always
+ * match the state "is the filp on a fasync list".
+ *
+ * We always take the 'filp->f_lock', in since fasync_lock
+ * needs to be irq-safe.
*/
-int fasync_helper(int fd, struct file * filp, int on, struct fasync_struct **fapp)
+static int fasync_remove_entry(struct file *filp, struct fasync_struct **fapp)
{
struct fasync_struct *fa, **fp;
- struct fasync_struct *new = NULL;
int result = 0;
- if (on) {
- new = kmem_cache_alloc(fasync_cache, GFP_KERNEL);
- if (!new)
- return -ENOMEM;
+ spin_lock(&filp->f_lock);
+ write_lock_irq(&fasync_lock);
+ for (fp = fapp; (fa = *fp) != NULL; fp = &fa->fa_next) {
+ if (fa->fa_file != filp)
+ continue;
+ *fp = fa->fa_next;
+ kmem_cache_free(fasync_cache, fa);
+ filp->f_flags &= ~FASYNC;
+ result = 1;
+ break;
}
+ write_unlock_irq(&fasync_lock);
+ spin_unlock(&filp->f_lock);
+ return result;
+}
+
+/*
+ * Add a fasync entry. Return negative on error, positive if
+ * added, and zero if did nothing but change an existing one.
+ *
+ * NOTE! It is very important that the FASYNC flag always
+ * match the state "is the filp on a fasync list".
+ */
+static int fasync_add_entry(int fd, struct file *filp, struct fasync_struct **fapp)
+{
+ struct fasync_struct *new, *fa, **fp;
+ int result = 0;
+
+ new = kmem_cache_alloc(fasync_cache, GFP_KERNEL);
+ if (!new)
+ return -ENOMEM;
- /*
- * We need to take f_lock first since it's not an IRQ-safe
- * lock.
- */
spin_lock(&filp->f_lock);
write_lock_irq(&fasync_lock);
for (fp = fapp; (fa = *fp) != NULL; fp = &fa->fa_next) {
- if (fa->fa_file == filp) {
- if(on) {
- fa->fa_fd = fd;
- kmem_cache_free(fasync_cache, new);
- } else {
- *fp = fa->fa_next;
- kmem_cache_free(fasync_cache, fa);
- result = 1;
- }
- goto out;
- }
+ if (fa->fa_file != filp)
+ continue;
+ fa->fa_fd = fd;
+ kmem_cache_free(fasync_cache, new);
+ goto out;
}
- if (on) {
- new->magic = FASYNC_MAGIC;
- new->fa_file = filp;
- new->fa_fd = fd;
- new->fa_next = *fapp;
- *fapp = new;
- result = 1;
- }
+ new->magic = FASYNC_MAGIC;
+ new->fa_file = filp;
+ new->fa_fd = fd;
+ new->fa_next = *fapp;
+ *fapp = new;
+ result = 1;
+ filp->f_flags |= FASYNC;
+
out:
- if (on)
- filp->f_flags |= FASYNC;
- else
- filp->f_flags &= ~FASYNC;
write_unlock_irq(&fasync_lock);
spin_unlock(&filp->f_lock);
return result;
}
+/*
+ * fasync_helper() is used by almost all character device drivers
+ * to set up the fasync queue, and for regular files by the file
+ * lease code. It returns negative on error, 0 if it did no changes
+ * and positive if it added/deleted the entry.
+ */
+int fasync_helper(int fd, struct file * filp, int on, struct fasync_struct **fapp)
+{
+ if (!on)
+ return fasync_remove_entry(filp, fapp);
+ return fasync_add_entry(fd, filp, fapp);
+}
+
EXPORT_SYMBOL(fasync_helper);
void __kill_fasync(struct fasync_struct *fa, int sig, int band)
diff --git a/fs/hppfs/hppfs.c b/fs/hppfs/hppfs.c
index a5089a6dd67a..7239efc690d8 100644
--- a/fs/hppfs/hppfs.c
+++ b/fs/hppfs/hppfs.c
@@ -646,22 +646,27 @@ static const struct super_operations hppfs_sbops = {
static int hppfs_readlink(struct dentry *dentry, char __user *buffer,
int buflen)
{
- struct dentry *proc_dentry;
-
- proc_dentry = HPPFS_I(dentry->d_inode)->proc_dentry;
+ struct dentry *proc_dentry = HPPFS_I(dentry->d_inode)->proc_dentry;
return proc_dentry->d_inode->i_op->readlink(proc_dentry, buffer,
buflen);
}
static void *hppfs_follow_link(struct dentry *dentry, struct nameidata *nd)
{
- struct dentry *proc_dentry;
-
- proc_dentry = HPPFS_I(dentry->d_inode)->proc_dentry;
+ struct dentry *proc_dentry = HPPFS_I(dentry->d_inode)->proc_dentry;
return proc_dentry->d_inode->i_op->follow_link(proc_dentry, nd);
}
+static void hppfs_put_link(struct dentry *dentry, struct nameidata *nd,
+ void *cookie)
+{
+ struct dentry *proc_dentry = HPPFS_I(dentry->d_inode)->proc_dentry;
+
+ if (proc_dentry->d_inode->i_op->put_link)
+ proc_dentry->d_inode->i_op->put_link(proc_dentry, nd, cookie);
+}
+
static const struct inode_operations hppfs_dir_iops = {
.lookup = hppfs_lookup,
};
@@ -669,6 +674,7 @@ static const struct inode_operations hppfs_dir_iops = {
static const struct inode_operations hppfs_link_iops = {
.readlink = hppfs_readlink,
.follow_link = hppfs_follow_link,
+ .put_link = hppfs_put_link,
};
static struct inode *get_inode(struct super_block *sb, struct dentry *dentry)
diff --git a/fs/namei.c b/fs/namei.c
index b55440baf7ab..94a5e60779f9 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -561,6 +561,7 @@ static __always_inline int __do_follow_link(struct path *path, struct nameidata
dget(dentry);
}
mntget(path->mnt);
+ nd->last_type = LAST_BIND;
cookie = dentry->d_inode->i_op->follow_link(dentry, nd);
error = PTR_ERR(cookie);
if (!IS_ERR(cookie)) {
@@ -1603,11 +1604,12 @@ struct file *do_filp_open(int dfd, const char *pathname,
struct file *filp;
struct nameidata nd;
int error;
- struct path path, save;
+ struct path path;
struct dentry *dir;
int count = 0;
int will_truncate;
int flag = open_to_namei_flags(open_flag);
+ int force_reval = 0;
/*
* O_SYNC is implemented as __O_SYNC|O_DSYNC. As many places only
@@ -1619,7 +1621,7 @@ struct file *do_filp_open(int dfd, const char *pathname,
open_flag |= O_DSYNC;
if (!acc_mode)
- acc_mode = MAY_OPEN | ACC_MODE(flag);
+ acc_mode = MAY_OPEN | ACC_MODE(open_flag);
/* O_TRUNC implies we need access checks for write permissions */
if (flag & O_TRUNC)
@@ -1659,9 +1661,12 @@ struct file *do_filp_open(int dfd, const char *pathname,
/*
* Create - we need to know the parent.
*/
+reval:
error = path_init(dfd, pathname, LOOKUP_PARENT, &nd);
if (error)
return ERR_PTR(error);
+ if (force_reval)
+ nd.flags |= LOOKUP_REVAL;
error = path_walk(pathname, &nd);
if (error) {
if (nd.root.mnt)
@@ -1853,17 +1858,7 @@ do_link:
error = security_inode_follow_link(path.dentry, &nd);
if (error)
goto exit_dput;
- save = nd.path;
- path_get(&save);
error = __do_follow_link(&path, &nd);
- if (error == -ESTALE) {
- /* nd.path had been dropped */
- nd.path = save;
- path_get(&nd.path);
- nd.flags |= LOOKUP_REVAL;
- error = __do_follow_link(&path, &nd);
- }
- path_put(&save);
path_put(&path);
if (error) {
/* Does someone understand code flow here? Or it is only
@@ -1873,6 +1868,10 @@ do_link:
release_open_intent(&nd);
if (nd.root.mnt)
path_put(&nd.root);
+ if (error == -ESTALE && !force_reval) {
+ force_reval = 1;
+ goto reval;
+ }
return ERR_PTR(error);
}
nd.flags &= ~LOOKUP_PARENT;
diff --git a/fs/namespace.c b/fs/namespace.c
index 7d70d63ceb29..c768f733c8d6 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -965,10 +965,12 @@ EXPORT_SYMBOL(may_umount_tree);
int may_umount(struct vfsmount *mnt)
{
int ret = 1;
+ down_read(&namespace_sem);
spin_lock(&vfsmount_lock);
if (propagate_mount_busy(mnt, 2))
ret = 0;
spin_unlock(&vfsmount_lock);
+ up_read(&namespace_sem);
return ret;
}
@@ -1352,12 +1354,12 @@ static int attach_recursive_mnt(struct vfsmount *source_mnt,
if (err)
goto out_cleanup_ids;
+ spin_lock(&vfsmount_lock);
+
if (IS_MNT_SHARED(dest_mnt)) {
for (p = source_mnt; p; p = next_mnt(p, source_mnt))
set_mnt_shared(p);
}
-
- spin_lock(&vfsmount_lock);
if (parent_path) {
detach_mnt(source_mnt, parent_path);
attach_mnt(source_mnt, path);
@@ -1534,8 +1536,12 @@ static int do_remount(struct path *path, int flags, int mnt_flags,
err = change_mount_flags(path->mnt, flags);
else
err = do_remount_sb(sb, flags, data, 0);
- if (!err)
+ if (!err) {
+ spin_lock(&vfsmount_lock);
+ mnt_flags |= path->mnt->mnt_flags & MNT_PNODE_MASK;
path->mnt->mnt_flags = mnt_flags;
+ spin_unlock(&vfsmount_lock);
+ }
up_write(&sb->s_umount);
if (!err) {
security_sb_post_remount(path->mnt, flags, data);
@@ -1665,6 +1671,8 @@ int do_add_mount(struct vfsmount *newmnt, struct path *path,
{
int err;
+ mnt_flags &= ~(MNT_SHARED | MNT_WRITE_HOLD);
+
down_write(&namespace_sem);
/* Something was mounted here while we slept */
while (d_mountpoint(path->dentry) &&
diff --git a/fs/notify/inotify/inotify_fsnotify.c b/fs/notify/inotify/inotify_fsnotify.c
index c9ee67b442e1..1afb0a10229f 100644
--- a/fs/notify/inotify/inotify_fsnotify.c
+++ b/fs/notify/inotify/inotify_fsnotify.c
@@ -121,7 +121,7 @@ static int idr_callback(int id, void *p, void *data)
if (warned)
return 0;
- warned = false;
+ warned = true;
entry = p;
ientry = container_of(entry, struct inotify_inode_mark_entry, fsn_entry);
diff --git a/fs/notify/inotify/inotify_user.c b/fs/notify/inotify/inotify_user.c
index 8271cf05c957..a94e8bd8eb1f 100644
--- a/fs/notify/inotify/inotify_user.c
+++ b/fs/notify/inotify/inotify_user.c
@@ -552,7 +552,7 @@ retry:
spin_lock(&group->inotify_data.idr_lock);
ret = idr_get_new_above(&group->inotify_data.idr, &tmp_ientry->fsn_entry,
- group->inotify_data.last_wd,
+ group->inotify_data.last_wd+1,
&tmp_ientry->wd);
spin_unlock(&group->inotify_data.idr_lock);
if (ret) {
@@ -632,7 +632,7 @@ static struct fsnotify_group *inotify_new_group(struct user_struct *user, unsign
spin_lock_init(&group->inotify_data.idr_lock);
idr_init(&group->inotify_data.idr);
- group->inotify_data.last_wd = 1;
+ group->inotify_data.last_wd = 0;
group->inotify_data.user = user;
group->inotify_data.fa = NULL;
diff --git a/fs/proc/base.c b/fs/proc/base.c
index 18d5cc62d8ed..e42bbd843ed1 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -1419,7 +1419,6 @@ static void *proc_pid_follow_link(struct dentry *dentry, struct nameidata *nd)
goto out;
error = PROC_I(inode)->op.proc_get_link(inode, &nd->path);
- nd->last_type = LAST_BIND;
out:
return ERR_PTR(error);
}
diff --git a/fs/ramfs/file-nommu.c b/fs/ramfs/file-nommu.c
index 2efc57173fd7..1739a4aba25f 100644
--- a/fs/ramfs/file-nommu.c
+++ b/fs/ramfs/file-nommu.c
@@ -123,30 +123,6 @@ add_error:
/*****************************************************************************/
/*
- * check that file shrinkage doesn't leave any VMAs dangling in midair
- */
-static int ramfs_nommu_check_mappings(struct inode *inode,
- size_t newsize, size_t size)
-{
- struct vm_area_struct *vma;
- struct prio_tree_iter iter;
-
- /* search for VMAs that fall within the dead zone */
- vma_prio_tree_foreach(vma, &iter, &inode->i_mapping->i_mmap,
- newsize >> PAGE_SHIFT,
- (size + PAGE_SIZE - 1) >> PAGE_SHIFT
- ) {
- /* found one - only interested if it's shared out of the page
- * cache */
- if (vma->vm_flags & VM_SHARED)
- return -ETXTBSY; /* not quite true, but near enough */
- }
-
- return 0;
-}
-
-/*****************************************************************************/
-/*
*
*/
static int ramfs_nommu_resize(struct inode *inode, loff_t newsize, loff_t size)
@@ -164,7 +140,7 @@ static int ramfs_nommu_resize(struct inode *inode, loff_t newsize, loff_t size)
/* check that a decrease in size doesn't cut off any shared mappings */
if (newsize < size) {
- ret = ramfs_nommu_check_mappings(inode, newsize, size);
+ ret = nommu_shrink_inode_mappings(inode, size, newsize);
if (ret < 0)
return ret;
}
diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c
index 09783cc444ac..77414db10dc2 100644
--- a/fs/xfs/linux-2.6/xfs_super.c
+++ b/fs/xfs/linux-2.6/xfs_super.c
@@ -954,16 +954,14 @@ xfs_fs_destroy_inode(
ASSERT_ALWAYS(!xfs_iflags_test(ip, XFS_IRECLAIM));
/*
- * If we have nothing to flush with this inode then complete the
- * teardown now, otherwise delay the flush operation.
+ * We always use background reclaim here because even if the
+ * inode is clean, it still may be under IO and hence we have
+ * to take the flush lock. The background reclaim path handles
+ * this more efficiently than we can here, so simply let background
+ * reclaim tear down all inodes.
*/
- if (!xfs_inode_clean(ip)) {
- xfs_inode_set_reclaim_tag(ip);
- return;
- }
-
out_reclaim:
- xfs_ireclaim(ip);
+ xfs_inode_set_reclaim_tag(ip);
}
/*
diff --git a/fs/xfs/linux-2.6/xfs_sync.c b/fs/xfs/linux-2.6/xfs_sync.c
index 6fed97a8cd3e..1f5e4bb5e970 100644
--- a/fs/xfs/linux-2.6/xfs_sync.c
+++ b/fs/xfs/linux-2.6/xfs_sync.c
@@ -65,7 +65,6 @@ xfs_inode_ag_lookup(
* as the tree is sparse and a gang lookup walks to find
* the number of objects requested.
*/
- read_lock(&pag->pag_ici_lock);
if (tag == XFS_ICI_NO_TAG) {
nr_found = radix_tree_gang_lookup(&pag->pag_ici_root,
(void **)&ip, *first_index, 1);
@@ -74,7 +73,7 @@ xfs_inode_ag_lookup(
(void **)&ip, *first_index, 1, tag);
}
if (!nr_found)
- goto unlock;
+ return NULL;
/*
* Update the index for the next lookup. Catch overflows
@@ -84,13 +83,8 @@ xfs_inode_ag_lookup(
*/
*first_index = XFS_INO_TO_AGINO(mp, ip->i_ino + 1);
if (*first_index < XFS_INO_TO_AGINO(mp, ip->i_ino))
- goto unlock;
-
+ return NULL;
return ip;
-
-unlock:
- read_unlock(&pag->pag_ici_lock);
- return NULL;
}
STATIC int
@@ -100,7 +94,8 @@ xfs_inode_ag_walk(
int (*execute)(struct xfs_inode *ip,
struct xfs_perag *pag, int flags),
int flags,
- int tag)
+ int tag,
+ int exclusive)
{
struct xfs_perag *pag = &mp->m_perag[ag];
uint32_t first_index;
@@ -114,10 +109,20 @@ restart:
int error = 0;
xfs_inode_t *ip;
+ if (exclusive)
+ write_lock(&pag->pag_ici_lock);
+ else
+ read_lock(&pag->pag_ici_lock);
ip = xfs_inode_ag_lookup(mp, pag, &first_index, tag);
- if (!ip)
+ if (!ip) {
+ if (exclusive)
+ write_unlock(&pag->pag_ici_lock);
+ else
+ read_unlock(&pag->pag_ici_lock);
break;
+ }
+ /* execute releases pag->pag_ici_lock */
error = execute(ip, pag, flags);
if (error == EAGAIN) {
skipped++;
@@ -125,9 +130,8 @@ restart:
}
if (error)
last_error = error;
- /*
- * bail out if the filesystem is corrupted.
- */
+
+ /* bail out if the filesystem is corrupted. */
if (error == EFSCORRUPTED)
break;
@@ -148,7 +152,8 @@ xfs_inode_ag_iterator(
int (*execute)(struct xfs_inode *ip,
struct xfs_perag *pag, int flags),
int flags,
- int tag)
+ int tag,
+ int exclusive)
{
int error = 0;
int last_error = 0;
@@ -157,7 +162,8 @@ xfs_inode_ag_iterator(
for (ag = 0; ag < mp->m_sb.sb_agcount; ag++) {
if (!mp->m_perag[ag].pag_ici_init)
continue;
- error = xfs_inode_ag_walk(mp, ag, execute, flags, tag);
+ error = xfs_inode_ag_walk(mp, ag, execute, flags, tag,
+ exclusive);
if (error) {
last_error = error;
if (error == EFSCORRUPTED)
@@ -174,30 +180,31 @@ xfs_sync_inode_valid(
struct xfs_perag *pag)
{
struct inode *inode = VFS_I(ip);
+ int error = EFSCORRUPTED;
/* nothing to sync during shutdown */
- if (XFS_FORCED_SHUTDOWN(ip->i_mount)) {
- read_unlock(&pag->pag_ici_lock);
- return EFSCORRUPTED;
- }
+ if (XFS_FORCED_SHUTDOWN(ip->i_mount))
+ goto out_unlock;
- /*
- * If we can't get a reference on the inode, it must be in reclaim.
- * Leave it for the reclaim code to flush. Also avoid inodes that
- * haven't been fully initialised.
- */
- if (!igrab(inode)) {
- read_unlock(&pag->pag_ici_lock);
- return ENOENT;
- }
- read_unlock(&pag->pag_ici_lock);
+ /* avoid new or reclaimable inodes. Leave for reclaim code to flush */
+ error = ENOENT;
+ if (xfs_iflags_test(ip, XFS_INEW | XFS_IRECLAIMABLE | XFS_IRECLAIM))
+ goto out_unlock;
- if (is_bad_inode(inode) || xfs_iflags_test(ip, XFS_INEW)) {
+ /* If we can't grab the inode, it must on it's way to reclaim. */
+ if (!igrab(inode))
+ goto out_unlock;
+
+ if (is_bad_inode(inode)) {
IRELE(ip);
- return ENOENT;
+ goto out_unlock;
}
- return 0;
+ /* inode is valid */
+ error = 0;
+out_unlock:
+ read_unlock(&pag->pag_ici_lock);
+ return error;
}
STATIC int
@@ -282,7 +289,7 @@ xfs_sync_data(
ASSERT((flags & ~(SYNC_TRYLOCK|SYNC_WAIT)) == 0);
error = xfs_inode_ag_iterator(mp, xfs_sync_inode_data, flags,
- XFS_ICI_NO_TAG);
+ XFS_ICI_NO_TAG, 0);
if (error)
return XFS_ERROR(error);
@@ -304,7 +311,7 @@ xfs_sync_attr(
ASSERT((flags & ~SYNC_WAIT) == 0);
return xfs_inode_ag_iterator(mp, xfs_sync_inode_attr, flags,
- XFS_ICI_NO_TAG);
+ XFS_ICI_NO_TAG, 0);
}
STATIC int
@@ -664,60 +671,6 @@ xfs_syncd_stop(
kthread_stop(mp->m_sync_task);
}
-STATIC int
-xfs_reclaim_inode(
- xfs_inode_t *ip,
- int sync_mode)
-{
- xfs_perag_t *pag = xfs_get_perag(ip->i_mount, ip->i_ino);
-
- /* The hash lock here protects a thread in xfs_iget_core from
- * racing with us on linking the inode back with a vnode.
- * Once we have the XFS_IRECLAIM flag set it will not touch
- * us.
- */
- write_lock(&pag->pag_ici_lock);
- spin_lock(&ip->i_flags_lock);
- if (__xfs_iflags_test(ip, XFS_IRECLAIM) ||
- !__xfs_iflags_test(ip, XFS_IRECLAIMABLE)) {
- spin_unlock(&ip->i_flags_lock);
- write_unlock(&pag->pag_ici_lock);
- return -EAGAIN;
- }
- __xfs_iflags_set(ip, XFS_IRECLAIM);
- spin_unlock(&ip->i_flags_lock);
- write_unlock(&pag->pag_ici_lock);
- xfs_put_perag(ip->i_mount, pag);
-
- /*
- * If the inode is still dirty, then flush it out. If the inode
- * is not in the AIL, then it will be OK to flush it delwri as
- * long as xfs_iflush() does not keep any references to the inode.
- * We leave that decision up to xfs_iflush() since it has the
- * knowledge of whether it's OK to simply do a delwri flush of
- * the inode or whether we need to wait until the inode is
- * pulled from the AIL.
- * We get the flush lock regardless, though, just to make sure
- * we don't free it while it is being flushed.
- */
- xfs_ilock(ip, XFS_ILOCK_EXCL);
- xfs_iflock(ip);
-
- /*
- * In the case of a forced shutdown we rely on xfs_iflush() to
- * wait for the inode to be unpinned before returning an error.
- */
- if (!is_bad_inode(VFS_I(ip)) && xfs_iflush(ip, sync_mode) == 0) {
- /* synchronize with xfs_iflush_done */
- xfs_iflock(ip);
- xfs_ifunlock(ip);
- }
-
- xfs_iunlock(ip, XFS_ILOCK_EXCL);
- xfs_ireclaim(ip);
- return 0;
-}
-
void
__xfs_inode_set_reclaim_tag(
struct xfs_perag *pag,
@@ -760,19 +713,55 @@ __xfs_inode_clear_reclaim_tag(
}
STATIC int
-xfs_reclaim_inode_now(
+xfs_reclaim_inode(
struct xfs_inode *ip,
struct xfs_perag *pag,
- int flags)
+ int sync_mode)
{
- /* ignore if already under reclaim */
- if (xfs_iflags_test(ip, XFS_IRECLAIM)) {
- read_unlock(&pag->pag_ici_lock);
+ /*
+ * The radix tree lock here protects a thread in xfs_iget from racing
+ * with us starting reclaim on the inode. Once we have the
+ * XFS_IRECLAIM flag set it will not touch us.
+ */
+ spin_lock(&ip->i_flags_lock);
+ ASSERT_ALWAYS(__xfs_iflags_test(ip, XFS_IRECLAIMABLE));
+ if (__xfs_iflags_test(ip, XFS_IRECLAIM)) {
+ /* ignore as it is already under reclaim */
+ spin_unlock(&ip->i_flags_lock);
+ write_unlock(&pag->pag_ici_lock);
return 0;
}
- read_unlock(&pag->pag_ici_lock);
+ __xfs_iflags_set(ip, XFS_IRECLAIM);
+ spin_unlock(&ip->i_flags_lock);
+ write_unlock(&pag->pag_ici_lock);
- return xfs_reclaim_inode(ip, flags);
+ /*
+ * If the inode is still dirty, then flush it out. If the inode
+ * is not in the AIL, then it will be OK to flush it delwri as
+ * long as xfs_iflush() does not keep any references to the inode.
+ * We leave that decision up to xfs_iflush() since it has the
+ * knowledge of whether it's OK to simply do a delwri flush of
+ * the inode or whether we need to wait until the inode is
+ * pulled from the AIL.
+ * We get the flush lock regardless, though, just to make sure
+ * we don't free it while it is being flushed.
+ */
+ xfs_ilock(ip, XFS_ILOCK_EXCL);
+ xfs_iflock(ip);
+
+ /*
+ * In the case of a forced shutdown we rely on xfs_iflush() to
+ * wait for the inode to be unpinned before returning an error.
+ */
+ if (!is_bad_inode(VFS_I(ip)) && xfs_iflush(ip, sync_mode) == 0) {
+ /* synchronize with xfs_iflush_done */
+ xfs_iflock(ip);
+ xfs_ifunlock(ip);
+ }
+
+ xfs_iunlock(ip, XFS_ILOCK_EXCL);
+ xfs_ireclaim(ip);
+ return 0;
}
int
@@ -780,6 +769,6 @@ xfs_reclaim_inodes(
xfs_mount_t *mp,
int mode)
{
- return xfs_inode_ag_iterator(mp, xfs_reclaim_inode_now, mode,
- XFS_ICI_RECLAIM_TAG);
+ return xfs_inode_ag_iterator(mp, xfs_reclaim_inode, mode,
+ XFS_ICI_RECLAIM_TAG, 1);
}
diff --git a/fs/xfs/linux-2.6/xfs_sync.h b/fs/xfs/linux-2.6/xfs_sync.h
index a500b4d91835..ea932b43335d 100644
--- a/fs/xfs/linux-2.6/xfs_sync.h
+++ b/fs/xfs/linux-2.6/xfs_sync.h
@@ -54,6 +54,6 @@ void __xfs_inode_clear_reclaim_tag(struct xfs_mount *mp, struct xfs_perag *pag,
int xfs_sync_inode_valid(struct xfs_inode *ip, struct xfs_perag *pag);
int xfs_inode_ag_iterator(struct xfs_mount *mp,
int (*execute)(struct xfs_inode *ip, struct xfs_perag *pag, int flags),
- int flags, int tag);
+ int flags, int tag, int write_lock);
#endif
diff --git a/fs/xfs/quota/xfs_qm_syscalls.c b/fs/xfs/quota/xfs_qm_syscalls.c
index 71af76fe8a23..873e07e29074 100644
--- a/fs/xfs/quota/xfs_qm_syscalls.c
+++ b/fs/xfs/quota/xfs_qm_syscalls.c
@@ -891,7 +891,7 @@ xfs_qm_dqrele_all_inodes(
uint flags)
{
ASSERT(mp->m_quotainfo);
- xfs_inode_ag_iterator(mp, xfs_dqrele_inode, flags, XFS_ICI_NO_TAG);
+ xfs_inode_ag_iterator(mp, xfs_dqrele_inode, flags, XFS_ICI_NO_TAG, 0);
}
/*------------------------------------------------------------------------*/
diff --git a/fs/xfs/xfs_dfrag.c b/fs/xfs/xfs_dfrag.c
index d1483a4f71b8..84ca1cf16a1e 100644
--- a/fs/xfs/xfs_dfrag.c
+++ b/fs/xfs/xfs_dfrag.c
@@ -114,10 +114,82 @@ xfs_swapext(
return error;
}
+/*
+ * We need to check that the format of the data fork in the temporary inode is
+ * valid for the target inode before doing the swap. This is not a problem with
+ * attr1 because of the fixed fork offset, but attr2 has a dynamically sized
+ * data fork depending on the space the attribute fork is taking so we can get
+ * invalid formats on the target inode.
+ *
+ * E.g. target has space for 7 extents in extent format, temp inode only has
+ * space for 6. If we defragment down to 7 extents, then the tmp format is a
+ * btree, but when swapped it needs to be in extent format. Hence we can't just
+ * blindly swap data forks on attr2 filesystems.
+ *
+ * Note that we check the swap in both directions so that we don't end up with
+ * a corrupt temporary inode, either.
+ *
+ * Note that fixing the way xfs_fsr sets up the attribute fork in the source
+ * inode will prevent this situation from occurring, so all we do here is
+ * reject and log the attempt. basically we are putting the responsibility on
+ * userspace to get this right.
+ */
+static int
+xfs_swap_extents_check_format(
+ xfs_inode_t *ip, /* target inode */
+ xfs_inode_t *tip) /* tmp inode */
+{
+
+ /* Should never get a local format */
+ if (ip->i_d.di_format == XFS_DINODE_FMT_LOCAL ||
+ tip->i_d.di_format == XFS_DINODE_FMT_LOCAL)
+ return EINVAL;
+
+ /*
+ * if the target inode has less extents that then temporary inode then
+ * why did userspace call us?
+ */
+ if (ip->i_d.di_nextents < tip->i_d.di_nextents)
+ return EINVAL;
+
+ /*
+ * if the target inode is in extent form and the temp inode is in btree
+ * form then we will end up with the target inode in the wrong format
+ * as we already know there are less extents in the temp inode.
+ */
+ if (ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS &&
+ tip->i_d.di_format == XFS_DINODE_FMT_BTREE)
+ return EINVAL;
+
+ /* Check temp in extent form to max in target */
+ if (tip->i_d.di_format == XFS_DINODE_FMT_EXTENTS &&
+ XFS_IFORK_NEXTENTS(tip, XFS_DATA_FORK) > ip->i_df.if_ext_max)
+ return EINVAL;
+
+ /* Check target in extent form to max in temp */
+ if (ip->i_d.di_format == XFS_DINODE_FMT_EXTENTS &&
+ XFS_IFORK_NEXTENTS(ip, XFS_DATA_FORK) > tip->i_df.if_ext_max)
+ return EINVAL;
+
+ /* Check root block of temp in btree form to max in target */
+ if (tip->i_d.di_format == XFS_DINODE_FMT_BTREE &&
+ XFS_IFORK_BOFF(ip) &&
+ tip->i_df.if_broot_bytes > XFS_IFORK_BOFF(ip))
+ return EINVAL;
+
+ /* Check root block of target in btree form to max in temp */
+ if (ip->i_d.di_format == XFS_DINODE_FMT_BTREE &&
+ XFS_IFORK_BOFF(tip) &&
+ ip->i_df.if_broot_bytes > XFS_IFORK_BOFF(tip))
+ return EINVAL;
+
+ return 0;
+}
+
int
xfs_swap_extents(
- xfs_inode_t *ip,
- xfs_inode_t *tip,
+ xfs_inode_t *ip, /* target inode */
+ xfs_inode_t *tip, /* tmp inode */
xfs_swapext_t *sxp)
{
xfs_mount_t *mp;
@@ -161,13 +233,6 @@ xfs_swap_extents(
goto out_unlock;
}
- /* Should never get a local format */
- if (ip->i_d.di_format == XFS_DINODE_FMT_LOCAL ||
- tip->i_d.di_format == XFS_DINODE_FMT_LOCAL) {
- error = XFS_ERROR(EINVAL);
- goto out_unlock;
- }
-
if (VN_CACHED(VFS_I(tip)) != 0) {
error = xfs_flushinval_pages(tip, 0, -1,
FI_REMAPF_LOCKED);
@@ -189,13 +254,12 @@ xfs_swap_extents(
goto out_unlock;
}
- /*
- * If the target has extended attributes, the tmp file
- * must also in order to ensure the correct data fork
- * format.
- */
- if ( XFS_IFORK_Q(ip) != XFS_IFORK_Q(tip) ) {
- error = XFS_ERROR(EINVAL);
+ /* check inode formats now that data is flushed */
+ error = xfs_swap_extents_check_format(ip, tip);
+ if (error) {
+ xfs_fs_cmn_err(CE_NOTE, mp,
+ "%s: inode 0x%llx format is incompatible for exchanging.",
+ __FILE__, ip->i_ino);
goto out_unlock;
}
@@ -276,6 +340,16 @@ xfs_swap_extents(
*tifp = *tempifp; /* struct copy */
/*
+ * Fix the in-memory data fork values that are dependent on the fork
+ * offset in the inode. We can't assume they remain the same as attr2
+ * has dynamic fork offsets.
+ */
+ ifp->if_ext_max = XFS_IFORK_SIZE(ip, XFS_DATA_FORK) /
+ (uint)sizeof(xfs_bmbt_rec_t);
+ tifp->if_ext_max = XFS_IFORK_SIZE(tip, XFS_DATA_FORK) /
+ (uint)sizeof(xfs_bmbt_rec_t);
+
+ /*
* Fix the on-disk inode values
*/
tmp = (__uint64_t)ip->i_d.di_nblocks;
diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c
index fa402a6bbbcf..155e798f30a1 100644
--- a/fs/xfs/xfs_iget.c
+++ b/fs/xfs/xfs_iget.c
@@ -73,7 +73,6 @@ xfs_inode_alloc(
ASSERT(atomic_read(&ip->i_pincount) == 0);
ASSERT(!spin_is_locked(&ip->i_flags_lock));
ASSERT(completion_done(&ip->i_flush));
- ASSERT(!rwsem_is_locked(&ip->i_iolock.mr_lock));
mrlock_init(&ip->i_iolock, MRLOCK_BARRIER, "xfsio", ip->i_ino);
diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index 391d36b0e68c..ef77fd88c8e3 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -2842,13 +2842,9 @@ xfs_iflush(
/*
* If the inode isn't dirty, then just release the inode flush lock and
- * do nothing. Treat stale inodes the same; we cannot rely on the
- * backing buffer remaining stale in cache for the remaining life of
- * the stale inode and so xfs_itobp() below may give us a buffer that
- * no longer contains inodes below. Doing this stale check here also
- * avoids forcing the log on pinned, stale inodes.
+ * do nothing.
*/
- if (xfs_inode_clean(ip) || xfs_iflags_test(ip, XFS_ISTALE)) {
+ if (xfs_inode_clean(ip)) {
xfs_ifunlock(ip);
return 0;
}
@@ -2872,6 +2868,19 @@ xfs_iflush(
xfs_iunpin_wait(ip);
/*
+ * For stale inodes we cannot rely on the backing buffer remaining
+ * stale in cache for the remaining life of the stale inode and so
+ * xfs_itobp() below may give us a buffer that no longer contains
+ * inodes below. We have to check this after ensuring the inode is
+ * unpinned so that it is safe to reclaim the stale inode after the
+ * flush call.
+ */
+ if (xfs_iflags_test(ip, XFS_ISTALE)) {
+ xfs_ifunlock(ip);
+ return 0;
+ }
+
+ /*
* This may have been unpinned because the filesystem is shutting
* down forcibly. If that's the case we must not write this inode
* to disk, because the log record didn't make it to disk!
diff --git a/fs/xfs/xfs_rtalloc.c b/fs/xfs/xfs_rtalloc.c
index 9e15a1185362..6be05f756d59 100644
--- a/fs/xfs/xfs_rtalloc.c
+++ b/fs/xfs/xfs_rtalloc.c
@@ -1517,6 +1517,8 @@ xfs_rtfree_range(
*/
error = xfs_rtfind_forw(mp, tp, end, mp->m_sb.sb_rextents - 1,
&postblock);
+ if (error)
+ return error;
/*
* If there are blocks not being freed at the front of the
* old extent, add summary data for them to be allocated.
generated by cgit v1.2.3 (git 2.0.4) at 2025-06-04 04:27:49 +0000