Re: + mm-balance_dirty_pages-reduce-calls-to-global_page_state-to-reduce-c ache-references.patch added to -mm tree

[Richard Kennedy <richard@rsk.demon.co.uk>]

On Thu, 2009-09-03 at 10:22 +0800, Wu Fengguang wrote:
> On Wed, Sep 02, 2009 at 09:53:31PM +0800, Richard Kennedy wrote:
> > On Wed, 2009-09-02 at 12:45 +0200, Peter Zijlstra wrote:
> > > On Wed, 2009-09-02 at 17:57 +0800, Wu Fengguang wrote:
> > > > On Wed, Sep 02, 2009 at 04:31:40PM +0800, Peter Zijlstra wrote:
> > > > > On Sat, 2009-08-22 at 20:11 +0200, Peter Zijlstra wrote:
> > > > > > > > +           /* always throttle if over threshold */
> > > > > > > > +           if (nr_reclaimable + nr_writeback < dirty_thresh) {
> > > > > > > 
> > > > > > > That 'if' is a big behavior change. It effectively blocks every one
> > > > > > > and canceled Peter's proportional throttling work: the less a process
> > > > > > > dirtied, the less it should be throttled.
> > > > > > 
> > > > > > Hmm, I think you're right, I had not considered that, thanks for
> > > > > > catching that.
> > > > > 
> > > > > So in retrospect I think I might have been wrong here.
> > > > > 
> > > > > The per task thing causes the bdi limit to be lower than the bdi limit
> > > > > based on writeback speed alone. That is, the more a task dirties, the
> > > > > lower the bdi limit is as seen for that task.
> > > > 
> > > > Right. If I understand it right, there will be a safety margin of about
> > > > (1/8) * dirty_limit for 1 heavy dirtier case, and that gap scales down
> > > > when there are more concurrent heavy dirtiers.
> > > 
> > > Right, with say 4 heavy writers the gap will be 1/4-th of 1/8-th, which
> > > is 1/32-nd.
> > > 
> > > With the side node that I think 1/8 is too much on large memory systems,
> > > and I have posted a sqrt patch numerous times, but I don't think we've
> > > ever found out if that helps or not...
> > > 
> > > > In principle, the ceiling will be a bit higher for a light dirtier to
> > > > make it easy to pass in the presence of more heavy dirtiers.
> > > 
> > > Right.
> > > 
> > > > > So if we get a task that generates tons of dirty pages (dd) then it
> > > > > won't ever actually hit the full dirty limit, even if its the only task
> > > > > on the system, and this outer if() will always be true.
> > > > 
> > > > Right, we have the safety margin :)
> > > > 
> > > > > Only when we actually saturate the full dirty limit will we fall through
> > > > > and throttle, but that is ok -- we want to enforce the full limit.
> > > > > 
> > > > > In short, a very aggressive dirtier will have a bdi limit lower than the
> > > > > total limit (at all times) leaving a little room at the top for the
> > > > > occasional dirtier to make quick progress.
> > > > > 
> > > > > Wu, does that cover the scenario you had in mind?
> > > > 
> > > > Yes thanks! Please correct me if wrong:
> > > > - the lower-ceiling-for-heavier-dirtier algorithm in task_dirty_limit()
> > > >   is elegant enough to prevent heavy dirtier to block light ones
> > > 
> > > ack
> > > 
> > > > - the test (nr_reclaimable + nr_writeback < dirty_thresh) is not
> > > >   relevant in normal, but can be kept for safety in the form of
> > > > 
> > > >           if (bdi_nr_reclaimable + bdi_nr_writeback < bdi_thresh &&
> > > >               nr_reclaimable + nr_writeback < dirty_thresh)
> > > >                   break;
> > > 
> > > ack
> > > 
> > > > - clip_bdi_dirty_limit() could be removed: we have been secured by the
> > > >   above test
> > > 
> > > ack.
> > 
> > 
> > I've noticed that there's a difference in the handling of the
> > dirty_exceeded flag, because this change no longer clips the bdi_thresh
> > then the flag may get cleared more quickly here :-  
> > 
> > 	if (bdi_nr_reclaimable + bdi_nr_writeback < bdi_thresh &&
> > 	    bdi->dirty_exceeded)
> > 		bdi->dirty_exceeded = 0;
> > 
> > So it then could call balance_dirty_pages a lot less often.
> 
> I guess in normal situations, clip_bdi_dirty_limit() is simply a
> no-op, or just lowers bdi_thresh slightly (otherwise could a bug).
> So it could be removed without causing much side effects, including
> the influence on dirty_exceeded.
> 
> > I've got an updated version of this patch that moves the clip_bdi logic
> > up into balance_dirty_pages that should be closer to the existing
> > behavior & tests so far look good. I can post it for comments if you're
> > interested ?
> 
> So I suggested just remove clip_bdi_dirty_limit(). To be sure, could
> run with the following patch and check if big numbers are showed.
> 
> Thanks,
> Fengguang
> ---
Yes, writing to one disk there's no difference, but what about writing
to multiple disks?

Can't we get into the situation where
	nr_reclaimable + nr_writeback >= dirty_threshold
and a bdi is
	bdi_nr_reclaimable + bdi_nr_writeback < bdi_thresh

With the old code that clips the bdi, the dirty_exceeded flag will stay
set, so balance_dirty_pages_ratelimited will check every 8 new dirty
pages.

And with the new code dirty_exceeded get cleared & ratelimited drops
back to checking every 32 new pages.

I'm not sure what difference this will have in practice, but I don't
have a RAID array to test it so I'm just guessing.

regards
Richard