Commit 5f33a080 authored by Shaohua Li's avatar Shaohua Li Committed by Linus Torvalds

mm/vmscan.c: set correct defer count for shrinker

Our system uses significantly more slab memory with memcg enabled with
the latest kernel.  With 3.10 kernel, slab uses 2G memory, while with
4.6 kernel, 6G memory is used.  The shrinker has problem.  Let's see we
have two memcg for one shrinker.  In do_shrink_slab:

1. Check cg1.  nr_deferred = 0, assume total_scan = 700.  batch size
   is 1024, then no memory is freed.  nr_deferred = 700

2. Check cg2.  nr_deferred = 700.  Assume freeable = 20, then
   total_scan = 10 or 40.  Let's assume it's 10.  No memory is freed.
   nr_deferred = 10.

The deferred share of cg1 is lost in this case.  kswapd will free no
memory even run above steps again and again.

The fix makes sure one memcg's deferred share isn't lost.

Link: http://lkml.kernel.org/r/2414be961b5d25892060315fbb56bb19d81d0c07.1476227351.git.shli@fb.comSigned-off-by: default avatarShaohua Li <shli@fb.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Vladimir Davydov <vdavydov@parallels.com>
Cc: <stable@vger.kernel.org>	[4.0+]
Signed-off-by: default avatarAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
parent 3e321587
...@@ -291,6 +291,7 @@ static unsigned long do_shrink_slab(struct shrink_control *shrinkctl, ...@@ -291,6 +291,7 @@ static unsigned long do_shrink_slab(struct shrink_control *shrinkctl,
int nid = shrinkctl->nid; int nid = shrinkctl->nid;
long batch_size = shrinker->batch ? shrinker->batch long batch_size = shrinker->batch ? shrinker->batch
: SHRINK_BATCH; : SHRINK_BATCH;
long scanned = 0, next_deferred;
freeable = shrinker->count_objects(shrinker, shrinkctl); freeable = shrinker->count_objects(shrinker, shrinkctl);
if (freeable == 0) if (freeable == 0)
...@@ -312,7 +313,9 @@ static unsigned long do_shrink_slab(struct shrink_control *shrinkctl, ...@@ -312,7 +313,9 @@ static unsigned long do_shrink_slab(struct shrink_control *shrinkctl,
pr_err("shrink_slab: %pF negative objects to delete nr=%ld\n", pr_err("shrink_slab: %pF negative objects to delete nr=%ld\n",
shrinker->scan_objects, total_scan); shrinker->scan_objects, total_scan);
total_scan = freeable; total_scan = freeable;
} next_deferred = nr;
} else
next_deferred = total_scan;
/* /*
* We need to avoid excessive windup on filesystem shrinkers * We need to avoid excessive windup on filesystem shrinkers
...@@ -369,17 +372,22 @@ static unsigned long do_shrink_slab(struct shrink_control *shrinkctl, ...@@ -369,17 +372,22 @@ static unsigned long do_shrink_slab(struct shrink_control *shrinkctl,
count_vm_events(SLABS_SCANNED, nr_to_scan); count_vm_events(SLABS_SCANNED, nr_to_scan);
total_scan -= nr_to_scan; total_scan -= nr_to_scan;
scanned += nr_to_scan;
cond_resched(); cond_resched();
} }
if (next_deferred >= scanned)
next_deferred -= scanned;
else
next_deferred = 0;
/* /*
* move the unused scan count back into the shrinker in a * move the unused scan count back into the shrinker in a
* manner that handles concurrent updates. If we exhausted the * manner that handles concurrent updates. If we exhausted the
* scan, there is no need to do an update. * scan, there is no need to do an update.
*/ */
if (total_scan > 0) if (next_deferred > 0)
new_nr = atomic_long_add_return(total_scan, new_nr = atomic_long_add_return(next_deferred,
&shrinker->nr_deferred[nid]); &shrinker->nr_deferred[nid]);
else else
new_nr = atomic_long_read(&shrinker->nr_deferred[nid]); new_nr = atomic_long_read(&shrinker->nr_deferred[nid]);
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment