Re: regression in page writeback

[Wu Fengguang <fengguang.wu@intel.com>]

On Wed, Sep 23, 2009 at 09:47:26AM +0800, Andrew Morton wrote:
> On Wed, 23 Sep 2009 09:32:36 +0800 Wu Fengguang <fengguang.wu@intel.com> wrote:
> 
> > On Wed, Sep 23, 2009 at 09:28:32AM +0800, Andrew Morton wrote:
> > > On Wed, 23 Sep 2009 09:17:58 +0800 Wu Fengguang <fengguang.wu@intel.com> wrote:
> > > 
> > > > On Wed, Sep 23, 2009 at 08:54:52AM +0800, Andrew Morton wrote:
> > > > > On Wed, 23 Sep 2009 08:22:20 +0800 Wu Fengguang <fengguang.wu@intel.com> wrote:
> > > > > 
> > > > > > Jens' per-bdi writeback has another improvement. In 2.6.31, when
> > > > > > superblocks A and B both have 100000 dirty pages, it will first
> > > > > > exhaust A's 100000 dirty pages before going on to sync B's.
> > > > > 
> > > > > That would only be true if someone broke 2.6.31.  Did they?
> > > > > 
> > > > > SYSCALL_DEFINE0(sync)
> > > > > {
> > > > > 	wakeup_pdflush(0);
> > > > > 	sync_filesystems(0);
> > > > > 	sync_filesystems(1);
> > > > > 	if (unlikely(laptop_mode))
> > > > > 		laptop_sync_completion();
> > > > > 	return 0;
> > > > > }
> > > > > 
> > > > > the sync_filesystems(0) is supposed to non-blockingly start IO against
> > > > > all devices.  It used to do that correctly.  But people mucked with it
> > > > > so perhaps it no longer does.
> > > > 
> > > > I'm referring to writeback_inodes(). Each invocation of which (to sync
> > > > 4MB) will do the same iteration over superblocks A => B => C ... So if
> > > > A has dirty pages, it will always be served first.
> > > > 
> > > > So if wbc->bdi == NULL (which is true for kupdate/background sync), it
> > > > will have to first exhaust A before going on to B and C.
> > > 
> > > But that works OK.  We fill the first device's queue, then it gets
> > > congested and sync_sb_inodes() does nothing and we advance to the next
> > > queue.
> > > 
> > > If a device has more than a queue's worth of dirty data then we'll
> > > probably leave some of that dirty memory un-queued, so there's some
> > > lack of concurrency in that situation.
> > 
> > Yes, exactly if block device is not fast enough.
> 
> Actually, no.

Sorry my "yes" is mainly for the first paragraph. The concurrency
problem exists for both fast/slow devices.

> If there's still outstanding dirty data for any of those queues, both
> wb_kupdate() and background_writeout() will take a teeny sleep and then
> will re-poll the queues.
> 
> Did that logic get broken?

No, but the "teeny sleep" is normally much smaller. When io queue is
not congested, every io completion event will wakeup the congestion
waiters. Also A's event could wake up B's waiters.

__freed_request() always calls blk_clear_queue_congested() if under
congestion threshold which in turn wakes up congestion waiters:

        if (rl->count[sync] < queue_congestion_off_threshold(q))
                blk_clear_queue_congested(q, sync);

Thanks,
Fengguang