* [PATCH] mm: vmscan: fix memcg-aware shrinkers not called on global reclaim
@ 2016-08-01 13:03 ` Vladimir Davydov
0 siblings, 0 replies; 14+ messages in thread
From: Vladimir Davydov @ 2016-08-01 13:03 UTC (permalink / raw)
To: Andrew Morton
Cc: Mel Gorman, Hillf Danton, Johannes Weiner, Vlastimil Babka,
Michal Hocko, Joonsoo Kim, Minchan Kim, Rik van Riel, linux-mm,
linux-kernel
We must call shrink_slab() for each memory cgroup on both global and
memcg reclaim in shrink_node_memcg(). Commit d71df22b55099 accidentally
changed that so that now shrink_slab() is only called with memcg != NULL
on memcg reclaim. As a result, memcg-aware shrinkers (including
dentry/inode) are never invoked on global reclaim. Fix that.
Fixes: d71df22b55099 ("mm, vmscan: begin reclaiming pages on a per-node basis")
Signed-off-by: Vladimir Davydov <vdavydov@virtuozzo.com>
---
mm/vmscan.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 650d26832569..374d95d04178 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -2561,7 +2561,7 @@ static bool shrink_node(pg_data_t *pgdat, struct scan_control *sc)
shrink_node_memcg(pgdat, memcg, sc, &lru_pages);
node_lru_pages += lru_pages;
- if (!global_reclaim(sc))
+ if (memcg)
shrink_slab(sc->gfp_mask, pgdat->node_id,
memcg, sc->nr_scanned - scanned,
lru_pages);
--
2.1.4
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply related [flat|nested] 14+ messages in thread* Re: [PATCH] mm: vmscan: fix memcg-aware shrinkers not called on global reclaim
2016-08-01 13:03 ` Vladimir Davydov
@ 2016-08-01 13:18 ` Michal Hocko
-1 siblings, 0 replies; 14+ messages in thread
From: Michal Hocko @ 2016-08-01 13:18 UTC (permalink / raw)
To: Vladimir Davydov
Cc: Andrew Morton, Mel Gorman, Hillf Danton, Johannes Weiner,
Vlastimil Babka, Joonsoo Kim, Minchan Kim, Rik van Riel,
linux-mm, linux-kernel
On Mon 01-08-16 16:03:10, Vladimir Davydov wrote:
> We must call shrink_slab() for each memory cgroup on both global and
> memcg reclaim in shrink_node_memcg(). Commit d71df22b55099 accidentally
> changed that so that now shrink_slab() is only called with memcg != NULL
> on memcg reclaim. As a result, memcg-aware shrinkers (including
> dentry/inode) are never invoked on global reclaim. Fix that.
>
> Fixes: d71df22b55099 ("mm, vmscan: begin reclaiming pages on a per-node basis")
I guess you meant b2e18757f2c9. I do not see d71df22b55099 anywhere.
> Signed-off-by: Vladimir Davydov <vdavydov@virtuozzo.com>
The fix looks ok to me otherwise
Acked-by: Michal Hocko <mhocko@suse.com>
> ---
> mm/vmscan.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/mm/vmscan.c b/mm/vmscan.c
> index 650d26832569..374d95d04178 100644
> --- a/mm/vmscan.c
> +++ b/mm/vmscan.c
> @@ -2561,7 +2561,7 @@ static bool shrink_node(pg_data_t *pgdat, struct scan_control *sc)
> shrink_node_memcg(pgdat, memcg, sc, &lru_pages);
> node_lru_pages += lru_pages;
>
> - if (!global_reclaim(sc))
> + if (memcg)
> shrink_slab(sc->gfp_mask, pgdat->node_id,
> memcg, sc->nr_scanned - scanned,
> lru_pages);
> --
> 2.1.4
>
> --
> To unsubscribe, send a message with 'unsubscribe linux-mm' in
> the body to majordomo@kvack.org. For more info on Linux MM,
> see: http://www.linux-mm.org/ .
> Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
--
Michal Hocko
SUSE Labs
^ permalink raw reply [flat|nested] 14+ messages in thread* Re: [PATCH] mm: vmscan: fix memcg-aware shrinkers not called on global reclaim
@ 2016-08-01 13:18 ` Michal Hocko
0 siblings, 0 replies; 14+ messages in thread
From: Michal Hocko @ 2016-08-01 13:18 UTC (permalink / raw)
To: Vladimir Davydov
Cc: Andrew Morton, Mel Gorman, Hillf Danton, Johannes Weiner,
Vlastimil Babka, Joonsoo Kim, Minchan Kim, Rik van Riel,
linux-mm, linux-kernel
On Mon 01-08-16 16:03:10, Vladimir Davydov wrote:
> We must call shrink_slab() for each memory cgroup on both global and
> memcg reclaim in shrink_node_memcg(). Commit d71df22b55099 accidentally
> changed that so that now shrink_slab() is only called with memcg != NULL
> on memcg reclaim. As a result, memcg-aware shrinkers (including
> dentry/inode) are never invoked on global reclaim. Fix that.
>
> Fixes: d71df22b55099 ("mm, vmscan: begin reclaiming pages on a per-node basis")
I guess you meant b2e18757f2c9. I do not see d71df22b55099 anywhere.
> Signed-off-by: Vladimir Davydov <vdavydov@virtuozzo.com>
The fix looks ok to me otherwise
Acked-by: Michal Hocko <mhocko@suse.com>
> ---
> mm/vmscan.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/mm/vmscan.c b/mm/vmscan.c
> index 650d26832569..374d95d04178 100644
> --- a/mm/vmscan.c
> +++ b/mm/vmscan.c
> @@ -2561,7 +2561,7 @@ static bool shrink_node(pg_data_t *pgdat, struct scan_control *sc)
> shrink_node_memcg(pgdat, memcg, sc, &lru_pages);
> node_lru_pages += lru_pages;
>
> - if (!global_reclaim(sc))
> + if (memcg)
> shrink_slab(sc->gfp_mask, pgdat->node_id,
> memcg, sc->nr_scanned - scanned,
> lru_pages);
> --
> 2.1.4
>
> --
> To unsubscribe, send a message with 'unsubscribe linux-mm' in
> the body to majordomo@kvack.org. For more info on Linux MM,
> see: http://www.linux-mm.org/ .
> Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
--
Michal Hocko
SUSE Labs
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 14+ messages in thread* Re: [PATCH] mm: vmscan: fix memcg-aware shrinkers not called on global reclaim
2016-08-01 13:18 ` Michal Hocko
@ 2016-08-01 13:21 ` Vladimir Davydov
-1 siblings, 0 replies; 14+ messages in thread
From: Vladimir Davydov @ 2016-08-01 13:21 UTC (permalink / raw)
To: Michal Hocko
Cc: Andrew Morton, Mel Gorman, Hillf Danton, Johannes Weiner,
Vlastimil Babka, Joonsoo Kim, Minchan Kim, Rik van Riel,
linux-mm, linux-kernel
On Mon, Aug 01, 2016 at 03:18:40PM +0200, Michal Hocko wrote:
> On Mon 01-08-16 16:03:10, Vladimir Davydov wrote:
> > We must call shrink_slab() for each memory cgroup on both global and
> > memcg reclaim in shrink_node_memcg(). Commit d71df22b55099 accidentally
> > changed that so that now shrink_slab() is only called with memcg != NULL
> > on memcg reclaim. As a result, memcg-aware shrinkers (including
> > dentry/inode) are never invoked on global reclaim. Fix that.
> >
> > Fixes: d71df22b55099 ("mm, vmscan: begin reclaiming pages on a per-node basis")
>
> I guess you meant b2e18757f2c9. I do not see d71df22b55099 anywhere.
I'm basing on top of v4.7-mmotm-2016-07-28-16-33 and there it's
d71df22b55099.
>
> > Signed-off-by: Vladimir Davydov <vdavydov@virtuozzo.com>
>
> The fix looks ok to me otherwise
>
> Acked-by: Michal Hocko <mhocko@suse.com>
Thanks!
^ permalink raw reply [flat|nested] 14+ messages in thread* Re: [PATCH] mm: vmscan: fix memcg-aware shrinkers not called on global reclaim
@ 2016-08-01 13:21 ` Vladimir Davydov
0 siblings, 0 replies; 14+ messages in thread
From: Vladimir Davydov @ 2016-08-01 13:21 UTC (permalink / raw)
To: Michal Hocko
Cc: Andrew Morton, Mel Gorman, Hillf Danton, Johannes Weiner,
Vlastimil Babka, Joonsoo Kim, Minchan Kim, Rik van Riel,
linux-mm, linux-kernel
On Mon, Aug 01, 2016 at 03:18:40PM +0200, Michal Hocko wrote:
> On Mon 01-08-16 16:03:10, Vladimir Davydov wrote:
> > We must call shrink_slab() for each memory cgroup on both global and
> > memcg reclaim in shrink_node_memcg(). Commit d71df22b55099 accidentally
> > changed that so that now shrink_slab() is only called with memcg != NULL
> > on memcg reclaim. As a result, memcg-aware shrinkers (including
> > dentry/inode) are never invoked on global reclaim. Fix that.
> >
> > Fixes: d71df22b55099 ("mm, vmscan: begin reclaiming pages on a per-node basis")
>
> I guess you meant b2e18757f2c9. I do not see d71df22b55099 anywhere.
I'm basing on top of v4.7-mmotm-2016-07-28-16-33 and there it's
d71df22b55099.
>
> > Signed-off-by: Vladimir Davydov <vdavydov@virtuozzo.com>
>
> The fix looks ok to me otherwise
>
> Acked-by: Michal Hocko <mhocko@suse.com>
Thanks!
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 14+ messages in thread* Re: [PATCH] mm: vmscan: fix memcg-aware shrinkers not called on global reclaim
2016-08-01 13:21 ` Vladimir Davydov
@ 2016-08-01 13:27 ` Michal Hocko
-1 siblings, 0 replies; 14+ messages in thread
From: Michal Hocko @ 2016-08-01 13:27 UTC (permalink / raw)
To: Vladimir Davydov
Cc: Andrew Morton, Mel Gorman, Hillf Danton, Johannes Weiner,
Vlastimil Babka, Joonsoo Kim, Minchan Kim, Rik van Riel,
linux-mm, linux-kernel
On Mon 01-08-16 16:21:45, Vladimir Davydov wrote:
> On Mon, Aug 01, 2016 at 03:18:40PM +0200, Michal Hocko wrote:
> > On Mon 01-08-16 16:03:10, Vladimir Davydov wrote:
> > > We must call shrink_slab() for each memory cgroup on both global and
> > > memcg reclaim in shrink_node_memcg(). Commit d71df22b55099 accidentally
> > > changed that so that now shrink_slab() is only called with memcg != NULL
> > > on memcg reclaim. As a result, memcg-aware shrinkers (including
> > > dentry/inode) are never invoked on global reclaim. Fix that.
> > >
> > > Fixes: d71df22b55099 ("mm, vmscan: begin reclaiming pages on a per-node basis")
> >
> > I guess you meant b2e18757f2c9. I do not see d71df22b55099 anywhere.
>
> I'm basing on top of v4.7-mmotm-2016-07-28-16-33 and there it's
> d71df22b55099.
But this sha is unstable. THe patch you are referring to is already
sitting in the Linus tree so please use his (stable sha instead).
--
Michal Hocko
SUSE Labs
^ permalink raw reply [flat|nested] 14+ messages in thread* Re: [PATCH] mm: vmscan: fix memcg-aware shrinkers not called on global reclaim
@ 2016-08-01 13:27 ` Michal Hocko
0 siblings, 0 replies; 14+ messages in thread
From: Michal Hocko @ 2016-08-01 13:27 UTC (permalink / raw)
To: Vladimir Davydov
Cc: Andrew Morton, Mel Gorman, Hillf Danton, Johannes Weiner,
Vlastimil Babka, Joonsoo Kim, Minchan Kim, Rik van Riel,
linux-mm, linux-kernel
On Mon 01-08-16 16:21:45, Vladimir Davydov wrote:
> On Mon, Aug 01, 2016 at 03:18:40PM +0200, Michal Hocko wrote:
> > On Mon 01-08-16 16:03:10, Vladimir Davydov wrote:
> > > We must call shrink_slab() for each memory cgroup on both global and
> > > memcg reclaim in shrink_node_memcg(). Commit d71df22b55099 accidentally
> > > changed that so that now shrink_slab() is only called with memcg != NULL
> > > on memcg reclaim. As a result, memcg-aware shrinkers (including
> > > dentry/inode) are never invoked on global reclaim. Fix that.
> > >
> > > Fixes: d71df22b55099 ("mm, vmscan: begin reclaiming pages on a per-node basis")
> >
> > I guess you meant b2e18757f2c9. I do not see d71df22b55099 anywhere.
>
> I'm basing on top of v4.7-mmotm-2016-07-28-16-33 and there it's
> d71df22b55099.
But this sha is unstable. THe patch you are referring to is already
sitting in the Linus tree so please use his (stable sha instead).
--
Michal Hocko
SUSE Labs
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH] mm: vmscan: fix memcg-aware shrinkers not called on global reclaim
2016-08-01 13:03 ` Vladimir Davydov
@ 2016-08-01 15:15 ` Johannes Weiner
-1 siblings, 0 replies; 14+ messages in thread
From: Johannes Weiner @ 2016-08-01 15:15 UTC (permalink / raw)
To: Vladimir Davydov
Cc: Andrew Morton, Mel Gorman, Hillf Danton, Vlastimil Babka,
Michal Hocko, Joonsoo Kim, Minchan Kim, Rik van Riel, linux-mm,
linux-kernel
On Mon, Aug 01, 2016 at 04:03:10PM +0300, Vladimir Davydov wrote:
> We must call shrink_slab() for each memory cgroup on both global and
> memcg reclaim in shrink_node_memcg(). Commit d71df22b55099 accidentally
> changed that so that now shrink_slab() is only called with memcg != NULL
> on memcg reclaim. As a result, memcg-aware shrinkers (including
> dentry/inode) are never invoked on global reclaim. Fix that.
>
> Fixes: d71df22b55099 ("mm, vmscan: begin reclaiming pages on a per-node basis")
> Signed-off-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Ouch, I missed that in the review. The change looked so obviously
correct, but the shrink_slab() interface is a little deceiving. It
would be great if shrink_slab() could handle root_mem_cgroup/NULL and
then we'd always call it from inside the loop. But AFAICS we need the
global call to get the cumulative scanned/lru_pages ratio from all of
the memcgs reclaimed... grr. Oh well.
This fix looks correct to me, anyway.
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
^ permalink raw reply [flat|nested] 14+ messages in thread* Re: [PATCH] mm: vmscan: fix memcg-aware shrinkers not called on global reclaim
@ 2016-08-01 15:15 ` Johannes Weiner
0 siblings, 0 replies; 14+ messages in thread
From: Johannes Weiner @ 2016-08-01 15:15 UTC (permalink / raw)
To: Vladimir Davydov
Cc: Andrew Morton, Mel Gorman, Hillf Danton, Vlastimil Babka,
Michal Hocko, Joonsoo Kim, Minchan Kim, Rik van Riel, linux-mm,
linux-kernel
On Mon, Aug 01, 2016 at 04:03:10PM +0300, Vladimir Davydov wrote:
> We must call shrink_slab() for each memory cgroup on both global and
> memcg reclaim in shrink_node_memcg(). Commit d71df22b55099 accidentally
> changed that so that now shrink_slab() is only called with memcg != NULL
> on memcg reclaim. As a result, memcg-aware shrinkers (including
> dentry/inode) are never invoked on global reclaim. Fix that.
>
> Fixes: d71df22b55099 ("mm, vmscan: begin reclaiming pages on a per-node basis")
> Signed-off-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Ouch, I missed that in the review. The change looked so obviously
correct, but the shrink_slab() interface is a little deceiving. It
would be great if shrink_slab() could handle root_mem_cgroup/NULL and
then we'd always call it from inside the loop. But AFAICS we need the
global call to get the cumulative scanned/lru_pages ratio from all of
the memcgs reclaimed... grr. Oh well.
This fix looks correct to me, anyway.
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH] mm: vmscan: fix memcg-aware shrinkers not called on global reclaim
2016-08-01 13:03 ` Vladimir Davydov
@ 2016-08-02 0:11 ` Minchan Kim
-1 siblings, 0 replies; 14+ messages in thread
From: Minchan Kim @ 2016-08-02 0:11 UTC (permalink / raw)
To: Vladimir Davydov
Cc: Andrew Morton, Mel Gorman, Hillf Danton, Johannes Weiner,
Vlastimil Babka, Michal Hocko, Joonsoo Kim, Rik van Riel,
linux-mm, linux-kernel
On Mon, Aug 01, 2016 at 04:03:10PM +0300, Vladimir Davydov wrote:
> We must call shrink_slab() for each memory cgroup on both global and
> memcg reclaim in shrink_node_memcg(). Commit d71df22b55099 accidentally
> changed that so that now shrink_slab() is only called with memcg != NULL
> on memcg reclaim. As a result, memcg-aware shrinkers (including
> dentry/inode) are never invoked on global reclaim. Fix that.
>
> Fixes: d71df22b55099 ("mm, vmscan: begin reclaiming pages on a per-node basis")
> Signed-off-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Acked-by: Minchan Kim <minchan@kernel.org>
^ permalink raw reply [flat|nested] 14+ messages in thread* Re: [PATCH] mm: vmscan: fix memcg-aware shrinkers not called on global reclaim
@ 2016-08-02 0:11 ` Minchan Kim
0 siblings, 0 replies; 14+ messages in thread
From: Minchan Kim @ 2016-08-02 0:11 UTC (permalink / raw)
To: Vladimir Davydov
Cc: Andrew Morton, Mel Gorman, Hillf Danton, Johannes Weiner,
Vlastimil Babka, Michal Hocko, Joonsoo Kim, Rik van Riel,
linux-mm, linux-kernel
On Mon, Aug 01, 2016 at 04:03:10PM +0300, Vladimir Davydov wrote:
> We must call shrink_slab() for each memory cgroup on both global and
> memcg reclaim in shrink_node_memcg(). Commit d71df22b55099 accidentally
> changed that so that now shrink_slab() is only called with memcg != NULL
> on memcg reclaim. As a result, memcg-aware shrinkers (including
> dentry/inode) are never invoked on global reclaim. Fix that.
>
> Fixes: d71df22b55099 ("mm, vmscan: begin reclaiming pages on a per-node basis")
> Signed-off-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Acked-by: Minchan Kim <minchan@kernel.org>
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH] mm: vmscan: fix memcg-aware shrinkers not called on global reclaim
2016-08-01 13:03 ` Vladimir Davydov
@ 2016-08-02 9:32 ` Mel Gorman
-1 siblings, 0 replies; 14+ messages in thread
From: Mel Gorman @ 2016-08-02 9:32 UTC (permalink / raw)
To: Vladimir Davydov
Cc: Andrew Morton, Hillf Danton, Johannes Weiner, Vlastimil Babka,
Michal Hocko, Joonsoo Kim, Minchan Kim, Rik van Riel, linux-mm,
linux-kernel
On Mon, Aug 01, 2016 at 04:03:10PM +0300, Vladimir Davydov wrote:
> We must call shrink_slab() for each memory cgroup on both global and
> memcg reclaim in shrink_node_memcg(). Commit d71df22b55099 accidentally
> changed that so that now shrink_slab() is only called with memcg != NULL
> on memcg reclaim. As a result, memcg-aware shrinkers (including
> dentry/inode) are never invoked on global reclaim. Fix that.
>
> Fixes: d71df22b55099 ("mm, vmscan: begin reclaiming pages on a per-node basis")
> Signed-off-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Acked-by: Mel Gorman <mgorman@techsingularity.net>
--
Mel Gorman
SUSE Labs
^ permalink raw reply [flat|nested] 14+ messages in thread* Re: [PATCH] mm: vmscan: fix memcg-aware shrinkers not called on global reclaim
@ 2016-08-02 9:32 ` Mel Gorman
0 siblings, 0 replies; 14+ messages in thread
From: Mel Gorman @ 2016-08-02 9:32 UTC (permalink / raw)
To: Vladimir Davydov
Cc: Andrew Morton, Hillf Danton, Johannes Weiner, Vlastimil Babka,
Michal Hocko, Joonsoo Kim, Minchan Kim, Rik van Riel, linux-mm,
linux-kernel
On Mon, Aug 01, 2016 at 04:03:10PM +0300, Vladimir Davydov wrote:
> We must call shrink_slab() for each memory cgroup on both global and
> memcg reclaim in shrink_node_memcg(). Commit d71df22b55099 accidentally
> changed that so that now shrink_slab() is only called with memcg != NULL
> on memcg reclaim. As a result, memcg-aware shrinkers (including
> dentry/inode) are never invoked on global reclaim. Fix that.
>
> Fixes: d71df22b55099 ("mm, vmscan: begin reclaiming pages on a per-node basis")
> Signed-off-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Acked-by: Mel Gorman <mgorman@techsingularity.net>
--
Mel Gorman
SUSE Labs
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
^ permalink raw reply [flat|nested] 14+ messages in thread