writeback: introduce .tagged_writepages for the WB_SYNC_NONE sync stage

sync(2) is performed in two stages: the WB_SYNC_NONE sync and the
WB_SYNC_ALL sync. Identify the first stage with .tagged_writepages and
do livelock prevention for it, too.

Jan's commit f446daaea9 ("mm: implement writeback livelock avoidance
using page tagging") is a partial fix in that it only fixed the
WB_SYNC_ALL phase livelock.

Although ext4 is tested to no longer livelock with commit f446daaea9,
it may due to some "redirty_tail() after pages_skipped" effect which
is by no means a guarantee for _all_ the file systems.

Note that writeback_inodes_sb() is called by not only sync(), they are
treated the same because the other callers also need livelock prevention.

Impact:  It changes the order in which pages/inodes are synced to disk.
Now in the WB_SYNC_NONE stage, it won't proceed to write the next inode
until finished with the current inode.

Acked-by: Jan Kara <jack@suse.cz>
CC: Dave Chinner <david@fromorbit.com>
Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>

diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index a5763e3..8558b6c 100644 (file)
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -2741,7 +2741,7 @@ static int write_cache_pages_da(struct address_space *mapping,
        index = wbc->range_start >> PAGE_CACHE_SHIFT;
        end = wbc->range_end >> PAGE_CACHE_SHIFT;
 
-       if (wbc->sync_mode == WB_SYNC_ALL)
+       if (wbc->sync_mode == WB_SYNC_ALL || wbc->tagged_writepages)
                tag = PAGECACHE_TAG_TOWRITE;
        else
                tag = PAGECACHE_TAG_DIRTY;
@@ -2973,7 +2973,7 @@ static int ext4_da_writepages(struct address_space *mapping,
        }
 
 retry:
-       if (wbc->sync_mode == WB_SYNC_ALL)
+       if (wbc->sync_mode == WB_SYNC_ALL || wbc->tagged_writepages)
                tag_pages_for_writeback(mapping, index, end);
 
        while (!ret && wbc->nr_to_write > 0) {

diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
index 0f015a0..5ed2ce9 100644 (file)
--- a/fs/fs-writeback.c
+++ b/fs/fs-writeback.c
@@ -36,6 +36,7 @@ struct wb_writeback_work {
        long nr_pages;
        struct super_block *sb;
        enum writeback_sync_modes sync_mode;
+       unsigned int tagged_writepages:1;
        unsigned int for_kupdate:1;
        unsigned int range_cyclic:1;
        unsigned int for_background:1;
@@ -650,6 +651,7 @@ static long wb_writeback(struct bdi_writeback *wb,
 {
        struct writeback_control wbc = {
                .sync_mode              = work->sync_mode,
+               .tagged_writepages      = work->tagged_writepages,
                .older_than_this        = NULL,
                .for_kupdate            = work->for_kupdate,
                .for_background         = work->for_background,
@@ -657,7 +659,7 @@ static long wb_writeback(struct bdi_writeback *wb,
        };
        unsigned long oldest_jif;
        long wrote = 0;
-       long write_chunk;
+       long write_chunk = MAX_WRITEBACK_PAGES;
        struct inode *inode;
 
        if (wbc.for_kupdate) {
@@ -683,9 +685,7 @@ static long wb_writeback(struct bdi_writeback *wb,
         *                   (quickly) tag currently dirty pages
         *                   (maybe slowly) sync all tagged pages
         */
-       if (wbc.sync_mode == WB_SYNC_NONE)
-               write_chunk = MAX_WRITEBACK_PAGES;
-       else
+       if (wbc.sync_mode == WB_SYNC_ALL || wbc.tagged_writepages)
                write_chunk = LONG_MAX;
 
        wbc.wb_start = jiffies; /* livelock avoidance */
@@ -1188,10 +1188,11 @@ void writeback_inodes_sb_nr(struct super_block *sb, unsigned long nr)
 {
        DECLARE_COMPLETION_ONSTACK(done);
        struct wb_writeback_work work = {
-               .sb             = sb,
-               .sync_mode      = WB_SYNC_NONE,
-               .done           = &done,
-               .nr_pages       = nr,
+               .sb                     = sb,
+               .sync_mode              = WB_SYNC_NONE,
+               .tagged_writepages      = 1,
+               .done                   = &done,
+               .nr_pages               = nr,
        };
 
        WARN_ON(!rwsem_is_locked(&sb->s_umount));

diff --git a/include/linux/writeback.h b/include/linux/writeback.h
index 17e7ccc..3f6542c 100644 (file)
--- a/include/linux/writeback.h
+++ b/include/linux/writeback.h
@@ -47,6 +47,7 @@ struct writeback_control {
        unsigned encountered_congestion:1; /* An output: a queue is full */
        unsigned for_kupdate:1;         /* A kupdate writeback */
        unsigned for_background:1;      /* A background writeback */
+       unsigned tagged_writepages:1;   /* tag-and-write to avoid livelock */
        unsigned for_reclaim:1;         /* Invoked from the page allocator */
        unsigned range_cyclic:1;        /* range_start is cyclic */
        unsigned more_io:1;             /* more io to be dispatched */

diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index 31f6988..955fe35 100644 (file)
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -892,12 +892,12 @@ int write_cache_pages(struct address_space *mapping,
                        range_whole = 1;
                cycled = 1; /* ignore range_cyclic tests */
        }
-       if (wbc->sync_mode == WB_SYNC_ALL)
+       if (wbc->sync_mode == WB_SYNC_ALL || wbc->tagged_writepages)
                tag = PAGECACHE_TAG_TOWRITE;
        else
                tag = PAGECACHE_TAG_DIRTY;
 retry:
-       if (wbc->sync_mode == WB_SYNC_ALL)
+       if (wbc->sync_mode == WB_SYNC_ALL || wbc->tagged_writepages)
                tag_pages_for_writeback(mapping, index, end);
        done_index = index;
        while (!done && (index <= end)) {