Re: [PATCH v2 block/for-linus] writeback: sync_inodes_sb() must write out I_DIRTY_TIME inodes and always call wait_sb_inodes()

[Jan Kara <jack@xxxxxxx>]

On Tue 25-08-15 14:11:52, Tejun Heo wrote:
> e79729123f63 ("writeback: don't issue wb_writeback_work if clean")
> updated writeback path to avoid kicking writeback work items if there
> are no inodes to be written out; unfortunately, the avoidance logic
> was too aggressive and broke sync_inodes_sb().
> 
> * sync_inodes_sb() must write out I_DIRTY_TIME inodes but I_DIRTY_TIME
>   inodes dont't contribute to bdi/wb_has_dirty_io() tests and were
>   being skipped over.
> 
> * inodes are taken off wb->b_dirty/io/more_io lists after writeback
>   starts on them.  sync_inodes_sb() skipping wait_sb_inodes() when
>   bdi_has_dirty_io() breaks it by making it return while writebacks
>   are in-flight.
> 
> This patch fixes the breakages by
> 
> * Removing bdi_has_dirty_io() shortcut from bdi_split_work_to_wbs().
>   The callers are already testing the condition.
> 
> * Removing bdi_has_dirty_io() shortcut from sync_inodes_sb() so that
>   it always calls into bdi_split_work_to_wbs() and wait_sb_inodes().
> 
> * Making bdi_split_work_to_wbs() consider the b_dirty_time list for
>   WB_SYNC_ALL writebacks.
> 
> Kudos to Eryu, Dave and Jan for tracking down the issue.
> 
> Signed-off-by: Tejun Heo <tj@xxxxxxxxxx>
> Fixes: e79729123f63 ("writeback: don't issue wb_writeback_work if clean")
> Link: 
> http://lkml.kernel.org/g/20150812101204.GE17933@xxxxxxxxxxxxxxxxxxxxxxxxxx
> Reported-and-bisected-by: Eryu Guan <eguan@xxxxxxxxxx>
> Cc: Dave Chinner <david@xxxxxxxxxxxxx>
> Cc: Jan Kara <jack@xxxxxxxx>
> Cc: Ted Ts'o <tytso@xxxxxxxxxx>
> ---
>  fs/fs-writeback.c |   22 +++++++++++++---------
>  1 file changed, 13 insertions(+), 9 deletions(-)

The patch looks good. You can add:

Reviewed-by: Jan Kara <jack@xxxxxxxx>

                                                                Honza

> 
> --- a/fs/fs-writeback.c
> +++ b/fs/fs-writeback.c
> @@ -844,14 +844,15 @@ static void bdi_split_work_to_wbs(struct
>       struct wb_iter iter;
>  
>       might_sleep();
> -
> -     if (!bdi_has_dirty_io(bdi))
> -             return;
>  restart:
>       rcu_read_lock();
>       bdi_for_each_wb(wb, bdi, &iter, next_blkcg_id) {
> -             if (!wb_has_dirty_io(wb) ||
> -                 (skip_if_busy && writeback_in_progress(wb)))
> +             /* SYNC_ALL writes out I_DIRTY_TIME too */
> +             if (!wb_has_dirty_io(wb) &&
> +                 (base_work->sync_mode == WB_SYNC_NONE ||
> +                  list_empty(&wb->b_dirty_time)))
> +                     continue;
> +             if (skip_if_busy && writeback_in_progress(wb))
>                       continue;
>  
>               base_work->nr_pages = wb_split_bdi_pages(wb, nr_pages);
> @@ -899,8 +900,7 @@ static void bdi_split_work_to_wbs(struct
>  {
>       might_sleep();
>  
> -     if (bdi_has_dirty_io(bdi) &&
> -         (!skip_if_busy || !writeback_in_progress(&bdi->wb))) {
> +     if (!skip_if_busy || !writeback_in_progress(&bdi->wb)) {
>               base_work->auto_free = 0;
>               base_work->single_wait = 0;
>               base_work->single_done = 0;
> @@ -2275,8 +2275,12 @@ void sync_inodes_sb(struct super_block *
>       };
>       struct backing_dev_info *bdi = sb->s_bdi;
>  
> -     /* Nothing to do? */
> -     if (!bdi_has_dirty_io(bdi) || bdi == &noop_backing_dev_info)
> +     /*
> +      * Can't skip on !bdi_has_dirty() because we should wait for !dirty
> +      * inodes under writeback and I_DIRTY_TIME inodes ignored by
> +      * bdi_has_dirty() need to be written out too.
> +      */
> +     if (bdi == &noop_backing_dev_info)
>               return;
>       WARN_ON(!rwsem_is_locked(&sb->s_umount));
>  
> 
-- 
Jan Kara <jack@xxxxxxxx>
SUSE Labs, CR