Re: Deadlock possibly caused by too_many_isolated.

[Wu Fengguang]

On Tue, Oct 19, 2010 at 06:41:37AM +0800, Andrew Morton wrote:
> On Tue, 19 Oct 2010 09:31:42 +1100
> Neil Brown <neilb@xxxxxxx> wrote:
> 
> > On Mon, 18 Oct 2010 14:58:59 -0700
> > Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> wrote:
> > 
> > > On Tue, 19 Oct 2010 00:15:04 +0800
> > > Wu Fengguang <fengguang.wu@xxxxxxxxx> wrote:
> > > 
> > > > Neil find that if too_many_isolated() returns true while performing
> > > > direct reclaim we can end up waiting for other threads to complete their
> > > > direct reclaim.  If those threads are allowed to enter the FS or IO to
> > > > free memory, but this thread is not, then it is possible that those
> > > > threads will be waiting on this thread and so we get a circular
> > > > deadlock.
> > > > 
> > > > some task enters direct reclaim with GFP_KERNEL
> > > >   => too_many_isolated() false
> > > >     => vmscan and run into dirty pages
> > > >       => pageout()
> > > >         => take some FS lock
> > > > 	  => fs/block code does GFP_NOIO allocation
> > > > 	    => enter direct reclaim again
> > > > 	      => too_many_isolated() true
> > > > 		=> waiting for others to progress, however the other
> > > > 		   tasks may be circular waiting for the FS lock..
> 
> I'm assuming that the last four "=>"'s here should have been indented
> another stop.

Yup. I'll fix it in next post.

> > > > The fix is to let !__GFP_IO and !__GFP_FS direct reclaims enjoy higher
> > > > priority than normal ones, by honouring them higher throttle threshold.
> > > > 
> > > > Now !GFP_IOFS reclaims won't be waiting for GFP_IOFS reclaims to
> > > > progress. They will be blocked only when there are too many concurrent
> > > > !GFP_IOFS reclaims, however that's very unlikely because the IO-less
> > > > direct reclaims is able to progress much more faster, and they won't
> > > > deadlock each other. The threshold is raised high enough for them, so
> > > > that there can be sufficient parallel progress of !GFP_IOFS reclaims.
> > > 
> > > I'm not sure that this is really a full fix.  Torsten's analysis does
> > > appear to point at the real bug: raid1 has code paths which allocate
> > > more than a single element from a mempool without starting IO against
> > > previous elements.
> > 
> > ... point at "a" real bug.
> > 
> > I think there are two bugs here.
> > The raid1 bug that Torsten mentions is certainly real (and has been around
> > for an embarrassingly long time).
> > The bug that I identified in too_many_isolated is also a real bug and can be
> > triggered without md/raid1 in the mix.
> > So this is not a 'full fix' for every bug in the kernel :-),

> > but it could well be a full fix for this particular bug.

Yeah it aims to be a full fix for one bug.

> Can we just delete the too_many_isolated() logic?  (Crappy comment

If the two cond_resched() calls can be removed from
shrink_page_list(), the major cause of too many pages being
isolated will be gone. However the writeback-waiting logic after
should_reclaim_stall() will also block the direct reclaimer for long
time with pages isolated, which may bite under pathological conditions.

> describes what the code does but not why it does it).

Good point. The comment could be improved as follows.

Thanks,
Fengguang

---
Subject: vmscan: comment too_many_isolated()
From: Wu Fengguang <fengguang.wu@xxxxxxxxx>
Date: Tue Oct 19 09:53:23 CST 2010

Comment "Why it's doing so" rather than "What it does"
as proposed by Andrew Morton.

Signed-off-by: Wu Fengguang <fengguang.wu@xxxxxxxxx>
---
 mm/vmscan.c |    6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

--- linux-next.orig/mm/vmscan.c	2010-10-19 09:29:44.000000000 +0800
+++ linux-next/mm/vmscan.c	2010-10-19 10:21:41.000000000 +0800
@@ -1142,7 +1142,11 @@ int isolate_lru_page(struct page *page)
 }
 
 /*
- * Are there way too many processes in the direct reclaim path already?
+ * A direct reclaimer may isolate SWAP_CLUSTER_MAX pages from the LRU list and
+ * then get resheduled. When there are massive number of tasks doing page
+ * allocation, such sleeping direct reclaimers may keep piling up on each CPU,
+ * the LRU list will go small and be scanned faster than necessary, leading to
+ * unnecessary swapping, thrashing and OOM.
  */
 static int too_many_isolated(struct zone *zone, int file,
 		struct scan_control *sc)
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/