Re: [PATCH v2] mm: vmscan: do not throttle based on pfmemalloc reserves if node has no reclaimable pages

2015-05-08 Thread Nishanth Aravamudan
On 08.05.2015 [15:47:26 -0700], Andrew Morton wrote: > On Wed, 06 May 2015 11:28:12 +0200 Vlastimil Babka wrote: > > > On 05/06/2015 12:09 AM, Nishanth Aravamudan wrote: > > > On 03.04.2015 [10:45:56 -0700], Nishanth Aravamudan wrote: > > >>> What I find somewhat worrying though is that we could

Re: [PATCH v2] mm: vmscan: do not throttle based on pfmemalloc reserves if node has no reclaimable pages

2015-05-08 Thread Andrew Morton
On Wed, 06 May 2015 11:28:12 +0200 Vlastimil Babka wrote: > On 05/06/2015 12:09 AM, Nishanth Aravamudan wrote: > > On 03.04.2015 [10:45:56 -0700], Nishanth Aravamudan wrote: > >>> What I find somewhat worrying though is that we could potentially > >>> break the pfmemalloc_watermark_ok() test in

Re: [PATCH v2] mm: vmscan: do not throttle based on pfmemalloc reserves if node has no reclaimable pages

2015-05-08 Thread Andrew Morton
On Wed, 06 May 2015 11:28:12 +0200 Vlastimil Babka vba...@suse.cz wrote: On 05/06/2015 12:09 AM, Nishanth Aravamudan wrote: On 03.04.2015 [10:45:56 -0700], Nishanth Aravamudan wrote: What I find somewhat worrying though is that we could potentially break the pfmemalloc_watermark_ok() test

Re: [PATCH v2] mm: vmscan: do not throttle based on pfmemalloc reserves if node has no reclaimable pages

2015-05-08 Thread Nishanth Aravamudan
On 08.05.2015 [15:47:26 -0700], Andrew Morton wrote: On Wed, 06 May 2015 11:28:12 +0200 Vlastimil Babka vba...@suse.cz wrote: On 05/06/2015 12:09 AM, Nishanth Aravamudan wrote: On 03.04.2015 [10:45:56 -0700], Nishanth Aravamudan wrote: What I find somewhat worrying though is that we

Re: [PATCH v2] mm: vmscan: do not throttle based on pfmemalloc reserves if node has no reclaimable pages

2015-05-06 Thread Vlastimil Babka
On 05/06/2015 12:09 AM, Nishanth Aravamudan wrote: On 03.04.2015 [10:45:56 -0700], Nishanth Aravamudan wrote: What I find somewhat worrying though is that we could potentially break the pfmemalloc_watermark_ok() test in situations where zone_reclaimable_pages(zone) == 0 is a transient situation

Re: [PATCH v2] mm: vmscan: do not throttle based on pfmemalloc reserves if node has no reclaimable pages

2015-05-06 Thread Vlastimil Babka
On 05/06/2015 12:09 AM, Nishanth Aravamudan wrote: On 03.04.2015 [10:45:56 -0700], Nishanth Aravamudan wrote: What I find somewhat worrying though is that we could potentially break the pfmemalloc_watermark_ok() test in situations where zone_reclaimable_pages(zone) == 0 is a transient situation

Re: [PATCH v2] mm: vmscan: do not throttle based on pfmemalloc reserves if node has no reclaimable pages

2015-05-05 Thread Nishanth Aravamudan
On 03.04.2015 [10:45:56 -0700], Nishanth Aravamudan wrote: > On 03.04.2015 [09:57:35 +0200], Vlastimil Babka wrote: > > On 03/31/2015 11:48 AM, Michal Hocko wrote: > > >On Fri 27-03-15 15:23:50, Nishanth Aravamudan wrote: > > >>On 27.03.2015 [13:17:59 -0700], Dave Hansen wrote: > > >>>On

Re: [PATCH v2] mm: vmscan: do not throttle based on pfmemalloc reserves if node has no reclaimable pages

2015-05-05 Thread Nishanth Aravamudan
On 03.04.2015 [10:45:56 -0700], Nishanth Aravamudan wrote: On 03.04.2015 [09:57:35 +0200], Vlastimil Babka wrote: On 03/31/2015 11:48 AM, Michal Hocko wrote: On Fri 27-03-15 15:23:50, Nishanth Aravamudan wrote: On 27.03.2015 [13:17:59 -0700], Dave Hansen wrote: On 03/27/2015 12:28 PM,

Re: [PATCH v2] mm: vmscan: do not throttle based on pfmemalloc reserves if node has no reclaimable pages

2015-04-03 Thread Nishanth Aravamudan
On 03.04.2015 [20:24:45 +0200], Michal Hocko wrote: > On Fri 03-04-15 10:43:57, Nishanth Aravamudan wrote: > > On 31.03.2015 [11:48:29 +0200], Michal Hocko wrote: > [...] > > > I would expect kswapd would be looping endlessly because the zone > > > wouldn't be balanced obviously. But I would be

Re: [PATCH v2] mm: vmscan: do not throttle based on pfmemalloc reserves if node has no reclaimable pages

2015-04-03 Thread Michal Hocko
On Fri 03-04-15 10:43:57, Nishanth Aravamudan wrote: > On 31.03.2015 [11:48:29 +0200], Michal Hocko wrote: [...] > > I would expect kswapd would be looping endlessly because the zone > > wouldn't be balanced obviously. But I would be wrong... because > > pgdat_balanced is doing this: > >

Re: [PATCH v2] mm: vmscan: do not throttle based on pfmemalloc reserves if node has no reclaimable pages

2015-04-03 Thread Nishanth Aravamudan
On 03.04.2015 [09:57:35 +0200], Vlastimil Babka wrote: > On 03/31/2015 11:48 AM, Michal Hocko wrote: > >On Fri 27-03-15 15:23:50, Nishanth Aravamudan wrote: > >>On 27.03.2015 [13:17:59 -0700], Dave Hansen wrote: > >>>On 03/27/2015 12:28 PM, Nishanth Aravamudan wrote: > @@ -2585,7 +2585,7 @@

Re: [PATCH v2] mm: vmscan: do not throttle based on pfmemalloc reserves if node has no reclaimable pages

2015-04-03 Thread Nishanth Aravamudan
On 31.03.2015 [11:48:29 +0200], Michal Hocko wrote: > On Fri 27-03-15 15:23:50, Nishanth Aravamudan wrote: > > On 27.03.2015 [13:17:59 -0700], Dave Hansen wrote: > > > On 03/27/2015 12:28 PM, Nishanth Aravamudan wrote: > > > > @@ -2585,7 +2585,7 @@ static bool pfmemalloc_watermark_ok(pg_data_t >

Re: [PATCH v2] mm: vmscan: do not throttle based on pfmemalloc reserves if node has no reclaimable pages

2015-04-03 Thread Vlastimil Babka
On 03/31/2015 11:48 AM, Michal Hocko wrote: On Fri 27-03-15 15:23:50, Nishanth Aravamudan wrote: On 27.03.2015 [13:17:59 -0700], Dave Hansen wrote: On 03/27/2015 12:28 PM, Nishanth Aravamudan wrote: @@ -2585,7 +2585,7 @@ static bool pfmemalloc_watermark_ok(pg_data_t *pgdat) for (i =

Re: [PATCH v2] mm: vmscan: do not throttle based on pfmemalloc reserves if node has no reclaimable pages

2015-04-03 Thread Vlastimil Babka
On 03/31/2015 11:48 AM, Michal Hocko wrote: On Fri 27-03-15 15:23:50, Nishanth Aravamudan wrote: On 27.03.2015 [13:17:59 -0700], Dave Hansen wrote: On 03/27/2015 12:28 PM, Nishanth Aravamudan wrote: @@ -2585,7 +2585,7 @@ static bool pfmemalloc_watermark_ok(pg_data_t *pgdat) for (i =

Re: [PATCH v2] mm: vmscan: do not throttle based on pfmemalloc reserves if node has no reclaimable pages

2015-04-03 Thread Nishanth Aravamudan
On 31.03.2015 [11:48:29 +0200], Michal Hocko wrote: On Fri 27-03-15 15:23:50, Nishanth Aravamudan wrote: On 27.03.2015 [13:17:59 -0700], Dave Hansen wrote: On 03/27/2015 12:28 PM, Nishanth Aravamudan wrote: @@ -2585,7 +2585,7 @@ static bool pfmemalloc_watermark_ok(pg_data_t *pgdat)

Re: [PATCH v2] mm: vmscan: do not throttle based on pfmemalloc reserves if node has no reclaimable pages

2015-04-03 Thread Nishanth Aravamudan
On 03.04.2015 [20:24:45 +0200], Michal Hocko wrote: On Fri 03-04-15 10:43:57, Nishanth Aravamudan wrote: On 31.03.2015 [11:48:29 +0200], Michal Hocko wrote: [...] I would expect kswapd would be looping endlessly because the zone wouldn't be balanced obviously. But I would be wrong...

Re: [PATCH v2] mm: vmscan: do not throttle based on pfmemalloc reserves if node has no reclaimable pages

2015-04-03 Thread Michal Hocko
On Fri 03-04-15 10:43:57, Nishanth Aravamudan wrote: On 31.03.2015 [11:48:29 +0200], Michal Hocko wrote: [...] I would expect kswapd would be looping endlessly because the zone wouldn't be balanced obviously. But I would be wrong... because pgdat_balanced is doing this: /*

Re: [PATCH v2] mm: vmscan: do not throttle based on pfmemalloc reserves if node has no reclaimable pages

2015-04-03 Thread Nishanth Aravamudan
On 03.04.2015 [09:57:35 +0200], Vlastimil Babka wrote: On 03/31/2015 11:48 AM, Michal Hocko wrote: On Fri 27-03-15 15:23:50, Nishanth Aravamudan wrote: On 27.03.2015 [13:17:59 -0700], Dave Hansen wrote: On 03/27/2015 12:28 PM, Nishanth Aravamudan wrote: @@ -2585,7 +2585,7 @@ static bool

Re: [PATCH v2] mm: vmscan: do not throttle based on pfmemalloc reserves if node has no reclaimable pages

2015-03-31 Thread Michal Hocko
On Fri 27-03-15 15:23:50, Nishanth Aravamudan wrote: > On 27.03.2015 [13:17:59 -0700], Dave Hansen wrote: > > On 03/27/2015 12:28 PM, Nishanth Aravamudan wrote: > > > @@ -2585,7 +2585,7 @@ static bool pfmemalloc_watermark_ok(pg_data_t > > > *pgdat) > > > > > > for (i = 0; i <=

Re: [PATCH v2] mm: vmscan: do not throttle based on pfmemalloc reserves if node has no reclaimable pages

2015-03-31 Thread Michal Hocko
On Fri 27-03-15 15:23:50, Nishanth Aravamudan wrote: On 27.03.2015 [13:17:59 -0700], Dave Hansen wrote: On 03/27/2015 12:28 PM, Nishanth Aravamudan wrote: @@ -2585,7 +2585,7 @@ static bool pfmemalloc_watermark_ok(pg_data_t *pgdat) for (i = 0; i = ZONE_NORMAL; i++) {

[PATCH v2] mm: vmscan: do not throttle based on pfmemalloc reserves if node has no reclaimable pages

2015-03-27 Thread Nishanth Aravamudan
On 27.03.2015 [13:17:59 -0700], Dave Hansen wrote: > On 03/27/2015 12:28 PM, Nishanth Aravamudan wrote: > > @@ -2585,7 +2585,7 @@ static bool pfmemalloc_watermark_ok(pg_data_t *pgdat) > > > > for (i = 0; i <= ZONE_NORMAL; i++) { > > zone = >node_zones[i]; > > -

[PATCH v2] mm: vmscan: do not throttle based on pfmemalloc reserves if node has no reclaimable pages

2015-03-27 Thread Nishanth Aravamudan
On 27.03.2015 [13:17:59 -0700], Dave Hansen wrote: On 03/27/2015 12:28 PM, Nishanth Aravamudan wrote: @@ -2585,7 +2585,7 @@ static bool pfmemalloc_watermark_ok(pg_data_t *pgdat) for (i = 0; i = ZONE_NORMAL; i++) { zone = pgdat-node_zones[i]; - if