Messages in this thread |  | | | Subject | Re: [PATCH] remove throttle_vm_writeout() | | From | Peter Zijlstra <> | | Date | Fri, 05 Oct 2007 10:22:00 +0200 | |
On Thu, 2007-10-04 at 17:48 -0700, Andrew Morton wrote:
> On Fri, 05 Oct 2007 02:12:30 +0200 Miklos Szeredi <miklos@szeredi.hu> wrote:
>
> > > > > > I don't think I understand that. Sure, it _shouldn't_ be a problem. But it
> > > _is_. That's what we're trying to fix, isn't it?> > > > The problem, I believe is in the memory allocation code, not in fuse.
>
> fuse is trying to do something which page reclaim was not designed for.
> Stuff broke.
>
> > In the example, memory allocation may be blocking indefinitely,
> > because we have 4MB under writeback, even though 28MB can still be
> > made available. And that _should_ be fixable.
>
> Well yes. But we need to work out how, without re-breaking the thing which
> throttle_vm_writeout() fixed.
I'm thinking the really_congested thing will also fix this. By only
allowing a limited amount of extra writeback.
> > > > So the only thing the kernel should be careful about, is not to block
> > > > on an allocation if not strictly necessary.> > > > > > > > Actually a trivial fix for this problem could be to just tweak the
> > > > thresholds, so to make the above scenario impossible. Although I'm
> > > > still not convinced, this patch is perfect, because the dirty
> > > > threshold can actually change in time...> > > > > > > > Index: linux/mm/page-writeback.c> > > > ===================================================================
> > > > --- linux.orig/mm/page-writeback.c 2007-10-05 00:31:01.000000000 +0200
> > > > +++ linux/mm/page-writeback.c 2007-10-05 00:50:11.000000000 +0200
> > > > @@ -515,6 +515,12 @@ void throttle_vm_writeout(gfp_t gfp_mask
> > > > for ( ; ; ) {
> > > > get_dirty_limits(&background_thresh, &dirty_thresh, NULL, NULL);
> > > >
> > > > + /*
> > > > + * Make sure the theshold is over the hard limit of
> > > > + * dirty_thresh + ratelimit_pages * nr_cpus
> > > > + */
> > > > + dirty_thresh += ratelimit_pages * num_online_cpus();
> > > > +
> > > > /*
> > > > * Boost the allowable dirty threshold a bit for page
> > > > * allocators so they don't get DoS'ed by heavy writers> > > > > > I can probably kind of guess what you're trying to do here. But if
> > > ratelimit_pages * num_online_cpus() exceeds the size of the offending zone
> > > then things might go bad.> > > > I think the admin can do quite a bit of other damage, by setting
> > dirty_ratio too high.> > > > Maybe this writeback throttling should just have a fixed limit of 80%
> > ZONE_NORMAL, and limit dirty_ratio to something like 50%.> > Bear in mind that the same problem will occur for the 16MB ZONE_DMA, and
> we cannot limit the system-wide dirty-memory threshold to 12MB.
> > iow, throttle_vm_writeout() needs to become zone-aware. Then it only
> throttles when, say, 80% of ZONE_FOO is under writeback.
As it stand 110% of dirty limit can already be larger than say zone_dma
(and likely is), so that is not a new bug - and I don't think its the
thing Miklos runs into.
The problem Miklos is seeing (and I, just in a different form), is that
throttle_vm_writeout() gets stuck because balance_dirty_pages() gets
called once every ratelimit_pages (per cpu). So we can have nr_cpus *
ratelimit_pages extra.....
/me thinks
ok I confused myself.
by calling balance_dirty_pages() once every ratelimit_pages (per cpu)
allows for nr_cpus() * ratelimit_pages extra _dirty_ pages. But
balance_dirty_pages() will make it:
nr_dirty + nr_unstable + nr_writeback < thresh
So even if it writes out all of the dirty pages, we still have:
nr_unstable + nr_writeback < thresh
So at any one time nr_writeback should not exceed thresh. But it does!?
So how do we end up with more writeback pages than that? should we teach
pdflush about these limits as well?
[unhandled content-type:application/pgp-signature] |  |