summaryrefslogtreecommitdiffstats
path: root/fs
diff options
context:
space:
mode:
authorLinus Torvalds <torvalds@linux-foundation.org>2012-05-28 18:54:45 +0200
committerLinus Torvalds <torvalds@linux-foundation.org>2012-05-28 18:54:45 +0200
commit90324cc1b11a211e37eabd8cb863e1a1561d6b1d (patch)
treec8b79c6850420a114ca6660c1b44fc486b1ba86d /fs
parentMerge branch 'next' of git://git.monstr.eu/linux-2.6-microblaze (diff)
parentwriteback: Avoid iput() from flusher thread (diff)
downloadlinux-90324cc1b11a211e37eabd8cb863e1a1561d6b1d.tar.xz
linux-90324cc1b11a211e37eabd8cb863e1a1561d6b1d.zip
Merge tag 'writeback' of git://git.kernel.org/pub/scm/linux/kernel/git/wfg/linux
Pull writeback tree from Wu Fengguang: "Mainly from Jan Kara to avoid iput() in the flusher threads." * tag 'writeback' of git://git.kernel.org/pub/scm/linux/kernel/git/wfg/linux: writeback: Avoid iput() from flusher thread vfs: Rename end_writeback() to clear_inode() vfs: Move waiting for inode writeback from end_writeback() to evict_inode() writeback: Refactor writeback_single_inode() writeback: Remove wb->list_lock from writeback_single_inode() writeback: Separate inode requeueing after writeback writeback: Move I_DIRTY_PAGES handling writeback: Move requeueing when I_SYNC set to writeback_sb_inodes() writeback: Move clearing of I_SYNC into inode_sync_complete() writeback: initialize global_dirty_limit fs: remove 8 bytes of padding from struct writeback_control on 64 bit builds mm: page-writeback.c: local functions should not be exposed globally
Diffstat (limited to 'fs')
-rw-r--r--fs/9p/vfs_inode.c2
-rw-r--r--fs/affs/inode.c2
-rw-r--r--fs/afs/inode.c2
-rw-r--r--fs/autofs4/inode.c2
-rw-r--r--fs/bfs/inode.c2
-rw-r--r--fs/binfmt_misc.c2
-rw-r--r--fs/block_dev.c2
-rw-r--r--fs/btrfs/inode.c2
-rw-r--r--fs/cifs/cifsfs.c2
-rw-r--r--fs/coda/inode.c2
-rw-r--r--fs/ecryptfs/super.c2
-rw-r--r--fs/exofs/inode.c4
-rw-r--r--fs/ext2/inode.c2
-rw-r--r--fs/ext3/inode.c6
-rw-r--r--fs/ext4/super.c2
-rw-r--r--fs/fat/inode.c2
-rw-r--r--fs/freevxfs/vxfs_inode.c2
-rw-r--r--fs/fs-writeback.c336
-rw-r--r--fs/fuse/inode.c2
-rw-r--r--fs/gfs2/super.c2
-rw-r--r--fs/hfs/inode.c2
-rw-r--r--fs/hfsplus/super.c2
-rw-r--r--fs/hostfs/hostfs_kern.c2
-rw-r--r--fs/hpfs/inode.c2
-rw-r--r--fs/hppfs/hppfs.c2
-rw-r--r--fs/hugetlbfs/inode.c2
-rw-r--r--fs/inode.c15
-rw-r--r--fs/jffs2/fs.c2
-rw-r--r--fs/jfs/inode.c2
-rw-r--r--fs/logfs/readwrite.c2
-rw-r--r--fs/minix/inode.c2
-rw-r--r--fs/ncpfs/inode.c2
-rw-r--r--fs/nfs/inode.c4
-rw-r--r--fs/nilfs2/inode.c4
-rw-r--r--fs/ntfs/inode.c2
-rw-r--r--fs/ocfs2/dlmfs/dlmfs.c2
-rw-r--r--fs/ocfs2/inode.c2
-rw-r--r--fs/omfs/inode.c2
-rw-r--r--fs/proc/inode.c2
-rw-r--r--fs/pstore/inode.c2
-rw-r--r--fs/reiserfs/inode.c4
-rw-r--r--fs/sysfs/inode.c2
-rw-r--r--fs/sysv/inode.c2
-rw-r--r--fs/ubifs/super.c2
-rw-r--r--fs/udf/inode.c2
-rw-r--r--fs/ufs/inode.c2
-rw-r--r--fs/xfs/xfs_super.c2
47 files changed, 267 insertions, 186 deletions
diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c
index 014c8dd62962..57ccb7537dae 100644
--- a/fs/9p/vfs_inode.c
+++ b/fs/9p/vfs_inode.c
@@ -448,7 +448,7 @@ void v9fs_evict_inode(struct inode *inode)
struct v9fs_inode *v9inode = V9FS_I(inode);
truncate_inode_pages(inode->i_mapping, 0);
- end_writeback(inode);
+ clear_inode(inode);
filemap_fdatawrite(inode->i_mapping);
#ifdef CONFIG_9P_FSCACHE
diff --git a/fs/affs/inode.c b/fs/affs/inode.c
index 88a4b0b50058..8bc4a59f4e7e 100644
--- a/fs/affs/inode.c
+++ b/fs/affs/inode.c
@@ -264,7 +264,7 @@ affs_evict_inode(struct inode *inode)
}
invalidate_inode_buffers(inode);
- end_writeback(inode);
+ clear_inode(inode);
affs_free_prealloc(inode);
cache_page = (unsigned long)AFFS_I(inode)->i_lc;
if (cache_page) {
diff --git a/fs/afs/inode.c b/fs/afs/inode.c
index d890ae3b2ce6..95cffd38239f 100644
--- a/fs/afs/inode.c
+++ b/fs/afs/inode.c
@@ -423,7 +423,7 @@ void afs_evict_inode(struct inode *inode)
ASSERTCMP(inode->i_ino, ==, vnode->fid.vnode);
truncate_inode_pages(&inode->i_data, 0);
- end_writeback(inode);
+ clear_inode(inode);
afs_give_up_callback(vnode);
diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c
index 6e488ebe7784..8a4fed8ead30 100644
--- a/fs/autofs4/inode.c
+++ b/fs/autofs4/inode.c
@@ -100,7 +100,7 @@ static int autofs4_show_options(struct seq_file *m, struct dentry *root)
static void autofs4_evict_inode(struct inode *inode)
{
- end_writeback(inode);
+ clear_inode(inode);
kfree(inode->i_private);
}
diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c
index e23dc7c8b884..9870417c26e7 100644
--- a/fs/bfs/inode.c
+++ b/fs/bfs/inode.c
@@ -174,7 +174,7 @@ static void bfs_evict_inode(struct inode *inode)
truncate_inode_pages(&inode->i_data, 0);
invalidate_inode_buffers(inode);
- end_writeback(inode);
+ clear_inode(inode);
if (inode->i_nlink)
return;
diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c
index 613aa0618235..790b3cddca67 100644
--- a/fs/binfmt_misc.c
+++ b/fs/binfmt_misc.c
@@ -505,7 +505,7 @@ static struct inode *bm_get_inode(struct super_block *sb, int mode)
static void bm_evict_inode(struct inode *inode)
{
- end_writeback(inode);
+ clear_inode(inode);
kfree(inode->i_private);
}
diff --git a/fs/block_dev.c b/fs/block_dev.c
index ba11c30f302d..c2bbe1fb1326 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -487,7 +487,7 @@ static void bdev_evict_inode(struct inode *inode)
struct list_head *p;
truncate_inode_pages(&inode->i_data, 0);
invalidate_inode_buffers(inode); /* is it needed here? */
- end_writeback(inode);
+ clear_inode(inode);
spin_lock(&bdev_lock);
while ( (p = bdev->bd_inodes.next) != &bdev->bd_inodes ) {
__bd_forget(list_entry(p, struct inode, i_devices));
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 61b16c641ce0..ceb7b9c9edcc 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -3756,7 +3756,7 @@ void btrfs_evict_inode(struct inode *inode)
btrfs_end_transaction(trans, root);
btrfs_btree_balance_dirty(root, nr);
no_delete:
- end_writeback(inode);
+ clear_inode(inode);
return;
}
diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index 541ef81f6ae8..0a0fa0250e99 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -272,7 +272,7 @@ static void
cifs_evict_inode(struct inode *inode)
{
truncate_inode_pages(&inode->i_data, 0);
- end_writeback(inode);
+ clear_inode(inode);
cifs_fscache_release_inode_cookie(inode);
}
diff --git a/fs/coda/inode.c b/fs/coda/inode.c
index 2870597b5c9d..f1813120d753 100644
--- a/fs/coda/inode.c
+++ b/fs/coda/inode.c
@@ -244,7 +244,7 @@ static void coda_put_super(struct super_block *sb)
static void coda_evict_inode(struct inode *inode)
{
truncate_inode_pages(&inode->i_data, 0);
- end_writeback(inode);
+ clear_inode(inode);
coda_cache_clear_inode(inode);
}
diff --git a/fs/ecryptfs/super.c b/fs/ecryptfs/super.c
index 2dd946b636d2..e879cf8ff0b1 100644
--- a/fs/ecryptfs/super.c
+++ b/fs/ecryptfs/super.c
@@ -133,7 +133,7 @@ static int ecryptfs_statfs(struct dentry *dentry, struct kstatfs *buf)
static void ecryptfs_evict_inode(struct inode *inode)
{
truncate_inode_pages(&inode->i_data, 0);
- end_writeback(inode);
+ clear_inode(inode);
iput(ecryptfs_inode_to_lower(inode));
}
diff --git a/fs/exofs/inode.c b/fs/exofs/inode.c
index ea5e1f97806a..5badb0c039de 100644
--- a/fs/exofs/inode.c
+++ b/fs/exofs/inode.c
@@ -1473,7 +1473,7 @@ void exofs_evict_inode(struct inode *inode)
goto no_delete;
inode->i_size = 0;
- end_writeback(inode);
+ clear_inode(inode);
/* if we are deleting an obj that hasn't been created yet, wait.
* This also makes sure that create_done cannot be called with an
@@ -1503,5 +1503,5 @@ void exofs_evict_inode(struct inode *inode)
return;
no_delete:
- end_writeback(inode);
+ clear_inode(inode);
}
diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c
index f9fa95f8443d..264d315f6c47 100644
--- a/fs/ext2/inode.c
+++ b/fs/ext2/inode.c
@@ -90,7 +90,7 @@ void ext2_evict_inode(struct inode * inode)
}
invalidate_inode_buffers(inode);
- end_writeback(inode);
+ clear_inode(inode);
ext2_discard_reservation(inode);
rsv = EXT2_I(inode)->i_block_alloc_info;
diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c
index a09790a412b1..9a4a5c48b1c9 100644
--- a/fs/ext3/inode.c
+++ b/fs/ext3/inode.c
@@ -272,18 +272,18 @@ void ext3_evict_inode (struct inode *inode)
if (ext3_mark_inode_dirty(handle, inode)) {
/* If that failed, just dquot_drop() and be done with that */
dquot_drop(inode);
- end_writeback(inode);
+ clear_inode(inode);
} else {
ext3_xattr_delete_inode(handle, inode);
dquot_free_inode(inode);
dquot_drop(inode);
- end_writeback(inode);
+ clear_inode(inode);
ext3_free_inode(handle, inode);
}
ext3_journal_stop(handle);
return;
no_delete:
- end_writeback(inode);
+ clear_inode(inode);
dquot_drop(inode);
}
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index 1867a98e0c49..35b5954489ee 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -1007,7 +1007,7 @@ static void destroy_inodecache(void)
void ext4_clear_inode(struct inode *inode)
{
invalidate_inode_buffers(inode);
- end_writeback(inode);
+ clear_inode(inode);
dquot_drop(inode);
ext4_discard_preallocations(inode);
if (EXT4_I(inode)->jinode) {
diff --git a/fs/fat/inode.c b/fs/fat/inode.c
index 21687e31acc0..b3d290c1b513 100644
--- a/fs/fat/inode.c
+++ b/fs/fat/inode.c
@@ -454,7 +454,7 @@ static void fat_evict_inode(struct inode *inode)
fat_truncate_blocks(inode, 0);
}
invalidate_inode_buffers(inode);
- end_writeback(inode);
+ clear_inode(inode);
fat_cache_inval_inode(inode);
fat_detach(inode);
}
diff --git a/fs/freevxfs/vxfs_inode.c b/fs/freevxfs/vxfs_inode.c
index cf9ef918a2a9..ef67c95f12d4 100644
--- a/fs/freevxfs/vxfs_inode.c
+++ b/fs/freevxfs/vxfs_inode.c
@@ -355,6 +355,6 @@ void
vxfs_evict_inode(struct inode *ip)
{
truncate_inode_pages(&ip->i_data, 0);
- end_writeback(ip);
+ clear_inode(ip);
call_rcu(&ip->i_rcu, vxfs_i_callback);
}
diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
index 539f36cf3e4a..8d2fb8c88cf3 100644
--- a/fs/fs-writeback.c
+++ b/fs/fs-writeback.c
@@ -231,11 +231,8 @@ static void requeue_io(struct inode *inode, struct bdi_writeback *wb)
static void inode_sync_complete(struct inode *inode)
{
- /*
- * Prevent speculative execution through
- * spin_unlock(&wb->list_lock);
- */
-
+ inode->i_state &= ~I_SYNC;
+ /* Waiters must see I_SYNC cleared before being woken up */
smp_mb();
wake_up_bit(&inode->i_state, __I_SYNC);
}
@@ -329,10 +326,12 @@ static int write_inode(struct inode *inode, struct writeback_control *wbc)
}
/*
- * Wait for writeback on an inode to complete.
+ * Wait for writeback on an inode to complete. Called with i_lock held.
+ * Caller must make sure inode cannot go away when we drop i_lock.
*/
-static void inode_wait_for_writeback(struct inode *inode,
- struct bdi_writeback *wb)
+static void __inode_wait_for_writeback(struct inode *inode)
+ __releases(inode->i_lock)
+ __acquires(inode->i_lock)
{
DEFINE_WAIT_BIT(wq, &inode->i_state, __I_SYNC);
wait_queue_head_t *wqh;
@@ -340,70 +339,119 @@ static void inode_wait_for_writeback(struct inode *inode,
wqh = bit_waitqueue(&inode->i_state, __I_SYNC);
while (inode->i_state & I_SYNC) {
spin_unlock(&inode->i_lock);
- spin_unlock(&wb->list_lock);
__wait_on_bit(wqh, &wq, inode_wait, TASK_UNINTERRUPTIBLE);
- spin_lock(&wb->list_lock);
spin_lock(&inode->i_lock);
}
}
/*
- * Write out an inode's dirty pages. Called under wb->list_lock and
- * inode->i_lock. Either the caller has an active reference on the inode or
- * the inode has I_WILL_FREE set.
- *
- * If `wait' is set, wait on the writeout.
- *
- * The whole writeout design is quite complex and fragile. We want to avoid
- * starvation of particular inodes when others are being redirtied, prevent
- * livelocks, etc.
+ * Wait for writeback on an inode to complete. Caller must have inode pinned.
*/
-static int
-writeback_single_inode(struct inode *inode, struct bdi_writeback *wb,
- struct writeback_control *wbc)
+void inode_wait_for_writeback(struct inode *inode)
{
- struct address_space *mapping = inode->i_mapping;
- long nr_to_write = wbc->nr_to_write;
- unsigned dirty;
- int ret;
+ spin_lock(&inode->i_lock);
+ __inode_wait_for_writeback(inode);
+ spin_unlock(&inode->i_lock);
+}
- assert_spin_locked(&wb->list_lock);
- assert_spin_locked(&inode->i_lock);
+/*
+ * Sleep until I_SYNC is cleared. This function must be called with i_lock
+ * held and drops it. It is aimed for callers not holding any inode reference
+ * so once i_lock is dropped, inode can go away.
+ */
+static void inode_sleep_on_writeback(struct inode *inode)
+ __releases(inode->i_lock)
+{
+ DEFINE_WAIT(wait);
+ wait_queue_head_t *wqh = bit_waitqueue(&inode->i_state, __I_SYNC);
+ int sleep;
- if (!atomic_read(&inode->i_count))
- WARN_ON(!(inode->i_state & (I_WILL_FREE|I_FREEING)));
- else
- WARN_ON(inode->i_state & I_WILL_FREE);
+ prepare_to_wait(wqh, &wait, TASK_UNINTERRUPTIBLE);
+ sleep = inode->i_state & I_SYNC;
+ spin_unlock(&inode->i_lock);
+ if (sleep)
+ schedule();
+ finish_wait(wqh, &wait);
+}
- if (inode->i_state & I_SYNC) {
+/*
+ * Find proper writeback list for the inode depending on its current state and
+ * possibly also change of its state while we were doing writeback. Here we
+ * handle things such as livelock prevention or fairness of writeback among
+ * inodes. This function can be called only by flusher thread - noone else
+ * processes all inodes in writeback lists and requeueing inodes behind flusher
+ * thread's back can have unexpected consequences.
+ */
+static void requeue_inode(struct inode *inode, struct bdi_writeback *wb,
+ struct writeback_control *wbc)
+{
+ if (inode->i_state & I_FREEING)
+ return;
+
+ /*
+ * Sync livelock prevention. Each inode is tagged and synced in one
+ * shot. If still dirty, it will be redirty_tail()'ed below. Update
+ * the dirty time to prevent enqueue and sync it again.
+ */
+ if ((inode->i_state & I_DIRTY) &&
+ (wbc->sync_mode == WB_SYNC_ALL || wbc->tagged_writepages))
+ inode->dirtied_when = jiffies;
+
+ if (wbc->pages_skipped) {
/*
- * If this inode is locked for writeback and we are not doing
- * writeback-for-data-integrity, move it to b_more_io so that
- * writeback can proceed with the other inodes on s_io.
- *
- * We'll have another go at writing back this inode when we
- * completed a full scan of b_io.
+ * writeback is not making progress due to locked
+ * buffers. Skip this inode for now.
*/
- if (wbc->sync_mode != WB_SYNC_ALL) {
+ redirty_tail(inode, wb);
+ return;
+ }
+
+ if (mapping_tagged(inode->i_mapping, PAGECACHE_TAG_DIRTY)) {
+ /*
+ * We didn't write back all the pages. nfs_writepages()
+ * sometimes bales out without doing anything.
+ */
+ if (wbc->nr_to_write <= 0) {
+ /* Slice used up. Queue for next turn. */
requeue_io(inode, wb);
- trace_writeback_single_inode_requeue(inode, wbc,
- nr_to_write);
- return 0;
+ } else {
+ /*
+ * Writeback blocked by something other than
+ * congestion. Delay the inode for some time to
+ * avoid spinning on the CPU (100% iowait)
+ * retrying writeback of the dirty page/inode
+ * that cannot be performed immediately.
+ */
+ redirty_tail(inode, wb);
}
-
+ } else if (inode->i_state & I_DIRTY) {
/*
- * It's a data-integrity sync. We must wait.
+ * Filesystems can dirty the inode during writeback operations,
+ * such as delayed allocation during submission or metadata
+ * updates after data IO completion.
*/
- inode_wait_for_writeback(inode, wb);
+ redirty_tail(inode, wb);
+ } else {
+ /* The inode is clean. Remove from writeback lists. */
+ list_del_init(&inode->i_wb_list);
}
+}
- BUG_ON(inode->i_state & I_SYNC);
+/*
+ * Write out an inode and its dirty pages. Do not update the writeback list
+ * linkage. That is left to the caller. The caller is also responsible for
+ * setting I_SYNC flag and calling inode_sync_complete() to clear it.
+ */
+static int
+__writeback_single_inode(struct inode *inode, struct bdi_writeback *wb,
+ struct writeback_control *wbc)
+{
+ struct address_space *mapping = inode->i_mapping;
+ long nr_to_write = wbc->nr_to_write;
+ unsigned dirty;
+ int ret;
- /* Set I_SYNC, reset I_DIRTY_PAGES */
- inode->i_state |= I_SYNC;
- inode->i_state &= ~I_DIRTY_PAGES;
- spin_unlock(&inode->i_lock);
- spin_unlock(&wb->list_lock);
+ WARN_ON(!(inode->i_state & I_SYNC));
ret = do_writepages(mapping, wbc);
@@ -424,6 +472,9 @@ writeback_single_inode(struct inode *inode, struct bdi_writeback *wb,
* write_inode()
*/
spin_lock(&inode->i_lock);
+ /* Clear I_DIRTY_PAGES if we've written out all dirty pages */
+ if (!mapping_tagged(mapping, PAGECACHE_TAG_DIRTY))
+ inode->i_state &= ~I_DIRTY_PAGES;
dirty = inode->i_state & I_DIRTY;
inode->i_state &= ~(I_DIRTY_SYNC | I_DIRTY_DATASYNC);
spin_unlock(&inode->i_lock);
@@ -433,60 +484,67 @@ writeback_single_inode(struct inode *inode, struct bdi_writeback *wb,
if (ret == 0)
ret = err;
}
+ trace_writeback_single_inode(inode, wbc, nr_to_write);
+ return ret;
+}
+
+/*
+ * Write out an inode's dirty pages. Either the caller has an active reference
+ * on the inode or the inode has I_WILL_FREE set.
+ *
+ * This function is designed to be called for writing back one inode which
+ * we go e.g. from filesystem. Flusher thread uses __writeback_single_inode()
+ * and does more profound writeback list handling in writeback_sb_inodes().
+ */
+static int
+writeback_single_inode(struct inode *inode, struct bdi_writeback *wb,
+ struct writeback_control *wbc)
+{
+ int ret = 0;
- spin_lock(&wb->list_lock);
spin_lock(&inode->i_lock);
- inode->i_state &= ~I_SYNC;
- if (!(inode->i_state & I_FREEING)) {
+ if (!atomic_read(&inode->i_count))
+ WARN_ON(!(inode->i_state & (I_WILL_FREE|I_FREEING)));
+ else
+ WARN_ON(inode->i_state & I_WILL_FREE);
+
+ if (inode->i_state & I_SYNC) {
+ if (wbc->sync_mode != WB_SYNC_ALL)
+ goto out;
/*
- * Sync livelock prevention. Each inode is tagged and synced in
- * one shot. If still dirty, it will be redirty_tail()'ed below.
- * Update the dirty time to prevent enqueue and sync it again.
+ * It's a data-integrity sync. We must wait. Since callers hold
+ * inode reference or inode has I_WILL_FREE set, it cannot go
+ * away under us.
*/
- if ((inode->i_state & I_DIRTY) &&
- (wbc->sync_mode == WB_SYNC_ALL || wbc->tagged_writepages))
- inode->dirtied_when = jiffies;
-
- if (mapping_tagged(mapping, PAGECACHE_TAG_DIRTY)) {
- /*
- * We didn't write back all the pages. nfs_writepages()
- * sometimes bales out without doing anything.
- */
- inode->i_state |= I_DIRTY_PAGES;
- if (wbc->nr_to_write <= 0) {
- /*
- * slice used up: queue for next turn
- */
- requeue_io(inode, wb);
- } else {
- /*
- * Writeback blocked by something other than
- * congestion. Delay the inode for some time to
- * avoid spinning on the CPU (100% iowait)
- * retrying writeback of the dirty page/inode
- * that cannot be performed immediately.
- */
- redirty_tail(inode, wb);
- }
- } else if (inode->i_state & I_DIRTY) {
- /*
- * Filesystems can dirty the inode during writeback
- * operations, such as delayed allocation during
- * submission or metadata updates after data IO
- * completion.
- */
- redirty_tail(inode, wb);
- } else {
- /*
- * The inode is clean. At this point we either have
- * a reference to the inode or it's on it's way out.
- * No need to add it back to the LRU.
- */
- list_del_init(&inode->i_wb_list);
- }
+ __inode_wait_for_writeback(inode);
}
+ WARN_ON(inode->i_state & I_SYNC);
+ /*
+ * Skip inode if it is clean. We don't want to mess with writeback
+ * lists in this function since flusher thread may be doing for example
+ * sync in parallel and if we move the inode, it could get skipped. So
+ * here we make sure inode is on some writeback list and leave it there
+ * unless we have completely cleaned the inode.
+ */
+ if (!(inode->i_state & I_DIRTY))
+ goto out;
+ inode->i_state |= I_SYNC;
+ spin_unlock(&inode->i_lock);
+
+ ret = __writeback_single_inode(inode, wb, wbc);
+
+ spin_lock(&wb->list_lock);
+ spin_lock(&inode->i_lock);
+ /*
+ * If inode is clean, remove it from writeback lists. Otherwise don't
+ * touch it. See comment above for explanation.
+ */
+ if (!(inode->i_state & I_DIRTY))
+ list_del_init(&inode->i_wb_list);
+ spin_unlock(&wb->list_lock);
inode_sync_complete(inode);
- trace_writeback_single_inode(inode, wbc, nr_to_write);
+out:
+ spin_unlock(&inode->i_lock);
return ret;
}
@@ -580,29 +638,57 @@ static long writeback_sb_inodes(struct super_block *sb,
redirty_tail(inode, wb);
continue;
}
- __iget(inode);
+ if ((inode->i_state & I_SYNC) && wbc.sync_mode != WB_SYNC_ALL) {
+ /*
+ * If this inode is locked for writeback and we are not
+ * doing writeback-for-data-integrity, move it to
+ * b_more_io so that writeback can proceed with the
+ * other inodes on s_io.
+ *
+ * We'll have another go at writing back this inode
+ * when we completed a full scan of b_io.
+ */
+ spin_unlock(&inode->i_lock);
+ requeue_io(inode, wb);
+ trace_writeback_sb_inodes_requeue(inode);
+ continue;
+ }
+ spin_unlock(&wb->list_lock);
+
+ /*
+ * We already requeued the inode if it had I_SYNC set and we
+ * are doing WB_SYNC_NONE writeback. So this catches only the
+ * WB_SYNC_ALL case.
+ */
+ if (inode->i_state & I_SYNC) {
+ /* Wait for I_SYNC. This function drops i_lock... */
+ inode_sleep_on_writeback(inode);
+ /* Inode may be gone, start again */
+ continue;
+ }
+ inode->i_state |= I_SYNC;
+ spin_unlock(&inode->i_lock);
+
write_chunk = writeback_chunk_size(wb->bdi, work);
wbc.nr_to_write = write_chunk;
wbc.pages_skipped = 0;
- writeback_single_inode(inode, wb, &wbc);
+ /*
+ * We use I_SYNC to pin the inode in memory. While it is set
+ * evict_inode() will wait so the inode cannot be freed.
+ */
+ __writeback_single_inode(inode, wb, &wbc);
work->nr_pages -= write_chunk - wbc.nr_to_write;
wrote += write_chunk - wbc.nr_to_write;
+ spin_lock(&wb->list_lock);
+ spin_lock(&inode->i_lock);
if (!(inode->i_state & I_DIRTY))
wrote++;
- if (wbc.pages_skipped) {
- /*
- * writeback is not making progress due to locked
- * buffers. Skip this inode for now.
- */
- redirty_tail(inode, wb);
- }
+ requeue_inode(inode, wb, &wbc);
+ inode_sync_complete(inode);
spin_unlock(&inode->i_lock);
- spin_unlock(&wb->list_lock);
- iput(inode);
- cond_resched();
- spin_lock(&wb->list_lock);
+ cond_resched_lock(&wb->list_lock);
/*
* bail out to wb_writeback() often enough to check
* background threshold and other termination conditions.
@@ -796,8 +882,10 @@ static long wb_writeback(struct bdi_writeback *wb,
trace_writeback_wait(wb->bdi, work);
inode = wb_inode(wb->b_more_io.prev);
spin_lock(&inode->i_lock);
- inode_wait_for_writeback(inode, wb);
- spin_unlock(&inode->i_lock);
+ spin_unlock(&wb->list_lock);
+ /* This function drops i_lock... */
+ inode_sleep_on_writeback(inode);
+ spin_lock(&wb->list_lock);
}
}
spin_unlock(&wb->list_lock);
@@ -1331,7 +1419,6 @@ EXPORT_SYMBOL(sync_inodes_sb);
int write_inode_now(struct inode *inode, int sync)
{
struct bdi_writeback *wb = &inode_to_bdi(inode)->wb;
- int ret;
struct writeback_control wbc = {
.nr_to_write = LONG_MAX,
.sync_mode = sync ? WB_SYNC_ALL : WB_SYNC_NONE,
@@ -1343,12 +1430,7 @@ int write_inode_now(struct inode *inode, int sync)
wbc.nr_to_write = 0;
might_sleep();
- spin_lock(&wb->list_lock);
- spin_lock(&inode->i_lock);
- ret = writeback_single_inode(inode, wb, &wbc);
- spin_unlock(&inode->i_lock);
- spin_unlock(&wb->list_lock);
- return ret;
+ return writeback_single_inode(inode, wb, &wbc);
}
EXPORT_SYMBOL(write_inode_now);
@@ -1365,15 +1447,7 @@ EXPORT_SYMBOL(write_inode_now);
*/
int sync_inode(struct inode *inode, struct writeback_control *wbc)
{
- struct bdi_writeback *wb = &inode_to_bdi(inode)->wb;
- int ret;
-
- spin_lock(&wb->list_lock);
- spin_lock(&inode->i_lock);
- ret = writeback_single_inode(inode, wb, wbc);
- spin_unlock(&inode->i_lock);
- spin_unlock(&wb->list_lock);
- return ret;
+ return writeback_single_inode(inode, &inode_to_bdi(inode)->wb, wbc);
}
EXPORT_SYMBOL(sync_inode);
diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c
index 26783eb2b1fc..56f6dcf30768 100644
--- a/fs/fuse/inode.c
+++ b/fs/fuse/inode.c
@@ -122,7 +122,7 @@ static void fuse_destroy_inode(struct inode *inode)
static void fuse_evict_inode(struct inode *inode)
{
truncate_inode_pages(&inode->i_data, 0);
- end_writeback(inode);
+ clear_inode(inode);
if (inode->i_sb->s_flags & MS_ACTIVE) {
struct fuse_conn *fc = get_fuse_conn(inode);
struct fuse_inode *fi = get_fuse_inode(inode);
diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c
index 6172fa77ad59..713e621c240b 100644
--- a/fs/gfs2/super.c
+++ b/fs/gfs2/super.c
@@ -1554,7 +1554,7 @@ out_unlock:
out:
/* Case 3 starts here */
truncate_inode_pages(&inode->i_data, 0);
- end_writeback(inode);
+ clear_inode(inode);
gfs2_dir_hash_inval(ip);
ip->i_gl->gl_object = NULL;
flush_delayed_work_sync(&ip->i_gl->gl_work);
diff --git a/fs/hfs/inode.c b/fs/hfs/inode.c
index 737dbeb64320..761ec06354b4 100644
--- a/fs/hfs/inode.c
+++ b/fs/hfs/inode.c
@@ -532,7 +532,7 @@ out:
void hfs_evict_inode(struct inode *inode)
{
truncate_inode_pages(&inode->i_data, 0);
- end_writeback(inode);
+ clear_inode(inode);
if (HFS_IS_RSRC(inode) && HFS_I(inode)->rsrc_inode) {
HFS_I(HFS_I(inode)->rsrc_inode)->rsrc_inode = NULL;
iput(HFS_I(inode)->rsrc_inode);
diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c
index ceb1c281eefb..a9bca4b8768b 100644
--- a/fs/hfsplus/super.c
+++ b/fs/hfsplus/super.c
@@ -154,7 +154,7 @@ static void hfsplus_evict_inode(struct inode *inode)
{
dprint(DBG_INODE, "hfsplus_evict_inode: %lu\n", inode->i_ino);
truncate_inode_pages(&inode->i_data, 0);
- end_writeback(inode);
+ clear_inode(inode);
if (HFSPLUS_IS_RSRC(inode)) {
HFSPLUS_I(HFSPLUS_I(inode)->rsrc_inode)->rsrc_inode = NULL;
iput(HFSPLUS_I(inode)->rsrc_inode);
diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c
index 07c516bfea76..2afa5bbccf9b 100644
--- a/fs/hostfs/hostfs_kern.c
+++ b/fs/hostfs/hostfs_kern.c
@@ -240,7 +240,7 @@ static struct inode *hostfs_alloc_inode(struct super_block *sb)
static void hostfs_evict_inode(struct inode *inode)
{
truncate_inode_pages(&inode->i_data, 0);
- end_writeback(inode);
+ clear_inode(inode);
if (HOSTFS_I(inode)->fd != -1) {
close_file(&HOSTFS_I(inode)->fd);
HOSTFS_I(inode)->fd = -1;
diff --git a/fs/hpfs/inode.c b/fs/hpfs/inode.c
index 3b2cec29972b..b43066cbdc6a 100644
--- a/fs/hpfs/inode.c
+++ b/fs/hpfs/inode.c
@@ -299,7 +299,7 @@ void hpfs_write_if_changed(struct inode *inode)
void hpfs_evict_inode(struct inode *inode)
{
truncate_inode_pages(&inode->i_data, 0);
- end_writeback(inode);
+ clear_inode(inode);
if (!inode->i_nlink) {
hpfs_lock(inode->i_sb);
hpfs_remove_fnode(inode->i_sb, inode->i_ino);
diff --git a/fs/hppfs/hppfs.c b/fs/hppfs/hppfs.c
index a80e45a690ac..d4f93b52cec5 100644
--- a/fs/hppfs/hppfs.c
+++ b/fs/hppfs/hppfs.c
@@ -614,7 +614,7 @@ static struct inode *hppfs_alloc_inode(struct super_block *sb)
void hppfs_evict_inode(struct inode *ino)
{
- end_writeback(ino);
+ clear_inode(ino);
dput(HPPFS_I(ino)->proc_dentry);
mntput(ino->i_sb->s_fs_info);
}
diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
index 001ef01d2fe2..cc9281b6c628 100644
--- a/fs/hugetlbfs/inode.c
+++ b/fs/hugetlbfs/inode.c
@@ -393,7 +393,7 @@ static void truncate_hugepages(struct inode *inode, loff_t lstart)
static void hugetlbfs_evict_inode(struct inode *inode)
{
truncate_hugepages(inode, 0);
- end_writeback(inode);
+ clear_inode(inode);
}
static inline void
diff --git a/fs/inode.c b/fs/inode.c
index da93f7d160d4..6bc8761cc333 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -486,7 +486,7 @@ void __remove_inode_hash(struct inode *inode)
}
EXPORT_SYMBOL(__remove_inode_hash);
-void end_writeback(struct inode *inode)
+void clear_inode(struct inode *inode)
{
might_sleep();
/*
@@ -500,11 +500,10 @@ void end_writeback(struct inode *inode)
BUG_ON(!list_empty(&inode->i_data.private_list));
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);
+EXPORT_SYMBOL(clear_inode);
/*
* Free the inode passed in, removing it from the lists it is still connected
@@ -531,12 +530,20 @@ static void evict(struct inode *inode)
inode_sb_list_del(inode);
+ /*
+ * Wait for flusher thread to be done with the inode so that filesystem
+ * does not start destroying it while writeback is still running. Since
+ * the inode has I_FREEING set, flusher thread won't start new work on
+ * the inode. We just have to wait for running writeback to finish.
+ */
+ inode_wait_for_writeback(inode);
+
if (op->evict_inode) {
op->evict_inode(inode);
} else {
if (inode->i_data.nrpages)
truncate_inode_pages(&inode->i_data, 0);
- end_writeback(inode);
+ clear_inode(inode);
}
if (S_ISBLK(inode->i_mode) && inode->i_bdev)
bd_forget(inode);
diff --git a/fs/jffs2/fs.c b/fs/jffs2/fs.c
index bb6f993ebca9..3d3092eda811 100644
--- a/fs/jffs2/fs.c
+++ b/fs/jffs2/fs.c
@@ -240,7 +240,7 @@ void jffs2_evict_inode (struct inode *inode)
jffs2_dbg(1, "%s(): ino #%lu mode %o\n",
__func__, inode->i_ino, inode->i_mode);
truncate_inode_pages(&inode->i_data, 0);
- end_writeback(inode);
+ clear_inode(inode);
jffs2_do_clear_inode(c, f);
}
diff --git a/fs/jfs/inode.c b/fs/jfs/inode.c
index 77b69b27f825..4692bf3ca8cb 100644
--- a/fs/jfs/inode.c
+++ b/fs/jfs/inode.c
@@ -169,7 +169,7 @@ void jfs_evict_inode(struct inode *inode)
} else {
truncate_inode_pages(&inode->i_data, 0);
}
- end_writeback(inode);
+ clear_inode(inode);
dquot_drop(inode);
}
diff --git a/fs/logfs/readwrite.c b/fs/logfs/readwrite.c
index e3ab5e5a904c..f1cb512c5019 100644
--- a/fs/logfs/readwrite.c
+++ b/fs/logfs/readwrite.c
@@ -2175,7 +2175,7 @@ void logfs_evict_inode(struct inode *inode)
}
}
truncate_inode_pages(&inode->i_data, 0);
- end_writeback(inode);
+ clear_inode(inode);
/* Cheaper version of write_inode. All changes are concealed in
* aliases, which are moved back. No write to the medium happens.
diff --git a/fs/minix/inode.c b/fs/minix/inode.c
index fcb05d2c6b5f..2a503ad020d5 100644
--- a/fs/minix/inode.c
+++ b/fs/minix/inode.c
@@ -32,7 +32,7 @@ static void minix_evict_inode(struct inode *inode)
minix_truncate(inode);
}
invalidate_inode_buffers(inode);
- end_writeback(inode);
+ clear_inode(inode);
if (!inode->i_nlink)
minix_free_inode(inode);
}
diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c
index 87484fb8d177..333df07ae3bd 100644
--- a/fs/ncpfs/inode.c
+++ b/fs/ncpfs/inode.c
@@ -292,7 +292,7 @@ static void
ncp_evict_inode(struct inode *inode)
{
truncate_inode_pages(&inode->i_data, 0);
- end_writeback(inode);
+ clear_inode(inode);
if (S_ISDIR(inode->i_mode)) {
DDPRINTK("ncp_evict_inode: put directory %ld\n", inode->i_ino);
diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
index e8bbfa5b3500..c6073139b402 100644
--- a/fs/nfs/inode.c
+++ b/fs/nfs/inode.c
@@ -121,7 +121,7 @@ static void nfs_clear_inode(struct inode *inode)
void nfs_evict_inode(struct inode *inode)
{
truncate_inode_pages(&inode->i_data, 0);
- end_writeback(inode);
+ clear_inode(inode);
nfs_clear_inode(inode);
}
@@ -1500,7 +1500,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr)
void nfs4_evict_inode(struct inode *inode)
{
truncate_inode_pages(&inode->i_data, 0);
- end_writeback(inode);
+ clear_inode(inode);
pnfs_return_layout(inode);
pnfs_destroy_layout(NFS_I(inode));
/* If we are holding a delegation, return it! */
diff --git a/fs/nilfs2/inode.c b/fs/nilfs2/inode.c
index 8f7b95ac1f7e..7cc64465ec26 100644
--- a/fs/nilfs2/inode.c
+++ b/fs/nilfs2/inode.c
@@ -734,7 +734,7 @@ void nilfs_evict_inode(struct inode *inode)
if (inode->i_nlink || !ii->i_root || unlikely(is_bad_inode(inode))) {
if (inode->i_data.nrpages)
truncate_inode_pages(&inode->i_data, 0);
- end_writeback(inode);
+ clear_inode(inode);
nilfs_clear_inode(inode);
return;
}
@@ -746,7 +746,7 @@ void nilfs_evict_inode(struct inode *inode)
/* TODO: some of the following operations may fail. */
nilfs_truncate_bmap(ii, 0);
nilfs_mark_inode_dirty(inode);
- end_writeback(inode);
+ clear_inode(inode);
ret = nilfs_ifile_delete_inode(ii->i_root->ifile, inode->i_ino);
if (!ret)
diff --git a/fs/ntfs/inode.c b/fs/ntfs/inode.c
index 2eaa66652944..c6dbd3db6ca8 100644
--- a/fs/ntfs/inode.c
+++ b/fs/ntfs/inode.c
@@ -2258,7 +2258,7 @@ void ntfs_evict_big_inode(struct inode *vi)
ntfs_inode *ni = NTFS_I(vi);
truncate_inode_pages(&vi->i_data, 0);
- end_writeback(vi);
+ clear_inode(vi);
#ifdef NTFS_RW
if (NInoDirty(ni)) {
diff --git a/fs/ocfs2/dlmfs/dlmfs.c b/fs/ocfs2/dlmfs/dlmfs.c
index 3b5825ef3193..e31d6ae013ab 100644
--- a/fs/ocfs2/dlmfs/dlmfs.c
+++ b/fs/ocfs2/dlmfs/dlmfs.c
@@ -367,7 +367,7 @@ static void dlmfs_evict_inode(struct inode *inode)
int status;
struct dlmfs_inode_private *ip;
- end_writeback(inode);
+ clear_inode(inode);
mlog(0, "inode %lu\n", inode->i_ino);
diff --git a/fs/ocfs2/inode.c b/fs/ocfs2/inode.c
index 17454a904d7b..735514ca400f 100644
--- a/fs/ocfs2/inode.c
+++ b/fs/ocfs2/inode.c
@@ -1069,7 +1069,7 @@ static void ocfs2_clear_inode(struct inode *inode)
int status;
struct ocfs2_inode_info *oi = OCFS2_I(inode);
- end_writeback(inode);
+ clear_inode(inode);
trace_ocfs2_clear_inode((unsigned long long)oi->ip_blkno,
inode->i_nlink);
diff --git a/fs/omfs/inode.c b/fs/omfs/inode.c
index dbc842222589..e6213b3725d1 100644
--- a/fs/omfs/inode.c
+++ b/fs/omfs/inode.c
@@ -184,7 +184,7 @@ int omfs_sync_inode(struct inode *inode)
static void omfs_evict_inode(struct inode *inode)
{
truncate_inode_pages(&inode->i_data, 0);
- end_writeback(inode);
+ clear_inode(inode);
if (inode->i_nlink)
return;
diff --git a/fs/proc/inode.c b/fs/proc/inode.c
index 554ecc54799f..7ac817b64a71 100644
--- a/fs/proc/inode.c
+++ b/fs/proc/inode.c
@@ -33,7 +33,7 @@ static void proc_evict_inode(struct inode *inode)
const struct proc_ns_operations *ns_ops;
truncate_inode_pages(&inode->i_data, 0);
- end_writeback(inode);
+ clear_inode(inode);
/* Stop tracking associated processes */
put_pid(PROC_I(inode)->pid);
diff --git a/fs/pstore/inode.c b/fs/pstore/inode.c
index 19507889bb7f..aeb19e68e086 100644
--- a/fs/pstore/inode.c
+++ b/fs/pstore/inode.c
@@ -85,7 +85,7 @@ static void pstore_evict_inode(struct inode *inode)
struct pstore_private *p = inode->i_private;
unsigned long flags;
- end_writeback(inode);
+ clear_inode(inode);
if (p) {
spin_lock_irqsave(&allpstore_lock, flags);
list_del(&p->list);
diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c
index 494c315c7417..59d06871a850 100644
--- a/fs/reiserfs/inode.c
+++ b/fs/reiserfs/inode.c
@@ -76,14 +76,14 @@ void reiserfs_evict_inode(struct inode *inode)
;
}
out:
- end_writeback(inode); /* note this must go after the journal_end to prevent deadlock */
+ clear_inode(inode); /* note this must go after the journal_end to prevent deadlock */
dquot_drop(inode);
inode->i_blocks = 0;
reiserfs_write_unlock_once(inode->i_sb, depth);
return;
no_delete:
- end_writeback(inode);
+ clear_inode(inode);
dquot_drop(inode);
}
diff --git a/fs/sysfs/inode.c b/fs/sysfs/inode.c
index 907c2b3af758..0ce3ccf7f401 100644
--- a/fs/sysfs/inode.c
+++ b/fs/sysfs/inode.c
@@ -310,7 +310,7 @@ void sysfs_evict_inode(struct inode *inode)
struct sysfs_dirent *sd = inode->i_private;
truncate_inode_pages(&inode->i_data, 0);
- end_writeback(inode);
+ clear_inode(inode);
sysfs_put(sd);
}
diff --git a/fs/sysv/inode.c b/fs/sysv/inode.c
index 3da5ce25faf0..08d0b2568cd3 100644
--- a/fs/sysv/inode.c
+++ b/fs/sysv/inode.c
@@ -316,7 +316,7 @@ static void sysv_evict_inode(struct inode *inode)
sysv_truncate(inode);
}
invalidate_inode_buffers(inode);
- end_writeback(inode);
+ clear_inode(inode);
if (!inode->i_nlink)
sysv_free_inode(inode);
}
diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c
index 001acccac0d6..5862dd9d2784 100644
--- a/fs/ubifs/super.c
+++ b/fs/ubifs/super.c
@@ -378,7 +378,7 @@ out:
smp_wmb();
}
done:
- end_writeback(inode);
+ clear_inode(inode);
}
static void ubifs_dirty_inode(struct inode *inode, int flags)
diff --git a/fs/udf/inode.c b/fs/udf/inode.c
index 7d7528008359..873e1bab9c4c 100644
--- a/fs/udf/inode.c
+++ b/fs/udf/inode.c
@@ -80,7 +80,7 @@ void udf_evict_inode(struct inode *inode)
} else
truncate_inode_pages(&inode->i_data, 0);
invalidate_inode_buffers(inode);
- end_writeback(inode);
+ clear_inode(inode);
if (iinfo->i_alloc_type != ICBTAG_FLAG_AD_IN_ICB &&
inode->i_size != iinfo->i_lenExtents) {
udf_warn(inode->i_sb, "Inode %lu (mode %o) has inode size %llu different from extent length %llu. Filesystem need not be standards compliant.\n",
diff --git a/fs/ufs/inode.c b/fs/ufs/inode.c
index 7cdd3953d67e..dd7c89d8a1c1 100644
--- a/fs/ufs/inode.c
+++ b/fs/ufs/inode.c
@@ -895,7 +895,7 @@ void ufs_evict_inode(struct inode * inode)
}
invalidate_inode_buffers(inode);
- end_writeback(inode);
+ clear_inode(inode);
if (want_delete) {
lock_ufs(inode->i_sb);
diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c
index 2fcfd5b0b046..0d9de41a7151 100644
--- a/fs/xfs/xfs_super.c
+++ b/fs/xfs/xfs_super.c
@@ -932,7 +932,7 @@ xfs_fs_evict_inode(
trace_xfs_evict_inode(ip);
truncate_inode_pages(&inode->i_data, 0);
- end_writeback(inode);
+ clear_inode(inode);
XFS_STATS_INC(vn_rele);
XFS_STATS_INC(vn_remove);
XFS_STATS_DEC(vn_active);