Re: [PATCH] powerpc/mm: Fix RECLAIM_DISTANCE

2017-02-07 Thread Gavin Shan
On Tue, Jan 31, 2017 at 04:01:05PM +1100, Gavin Shan wrote: >On Tue, Jan 31, 2017 at 08:11:39AM +1100, Michael Ellerman wrote: >>Gavin Shan writes: >> >>I'd like to see some test results from multi-node systems. >> >>I'd also like to understand what has changed since we

Re: [PATCH] powerpc/mm: Fix RECLAIM_DISTANCE

2017-01-30 Thread Gavin Shan
On Tue, Jan 31, 2017 at 04:01:05PM +1100, Gavin Shan wrote: >On Tue, Jan 31, 2017 at 08:11:39AM +1100, Michael Ellerman wrote: >>Gavin Shan writes: >> >>I'd like to see some test results from multi-node systems. >> >>I'd also like to understand what has changed since we

Re: [PATCH] powerpc/mm: Fix RECLAIM_DISTANCE

2017-01-30 Thread Gavin Shan
On Tue, Jan 31, 2017 at 03:58:16PM +1100, Anton Blanchard wrote: >Hi, > > >> Anton, I think the behaviour looks good. Actually, it's not very >> relevant to the issue addressed by the patch. I will reply to >> Michael's reply about the reason. There are two nodes in your system >> and the memory

Re: [PATCH] powerpc/mm: Fix RECLAIM_DISTANCE

2017-01-30 Thread Gavin Shan
On Tue, Jan 31, 2017 at 08:11:39AM +1100, Michael Ellerman wrote: >Gavin Shan writes: > >I'd like to see some test results from multi-node systems. > >I'd also like to understand what has changed since we changed >RECLAIM_DISTANCE in the first place, ie. why did it used

Re: [PATCH] powerpc/mm: Fix RECLAIM_DISTANCE

2017-01-30 Thread Anton Blanchard
Hi, > Anton, I think the behaviour looks good. Actually, it's not very > relevant to the issue addressed by the patch. I will reply to > Michael's reply about the reason. There are two nodes in your system > and the memory is expected to be allocated from node-0. If node-0 > doesn't have enough

Re: [PATCH] powerpc/mm: Fix RECLAIM_DISTANCE

2017-01-30 Thread Gavin Shan
On Mon, Jan 30, 2017 at 12:02:40PM +1100, Anton Blanchard wrote: >Hi, > >> Anton suggested that NUMA distances in powerpc mattered and hurted >> performance without this setting. We need to validate to see if this >> is still true. A simple way to start would be benchmarking > >The original issue

Re: [PATCH] powerpc/mm: Fix RECLAIM_DISTANCE

2017-01-30 Thread Michael Ellerman
Gavin Shan writes: > On Mon, Jan 30, 2017 at 12:02:40PM +1100, Anton Blanchard wrote: >>> Anton suggested that NUMA distances in powerpc mattered and hurted >>> performance without this setting. We need to validate to see if this >>> is still true. A simple way to

Re: [PATCH] powerpc/mm: Fix RECLAIM_DISTANCE

2017-01-29 Thread Gavin Shan
On Mon, Jan 30, 2017 at 12:02:40PM +1100, Anton Blanchard wrote: >> Anton suggested that NUMA distances in powerpc mattered and hurted >> performance without this setting. We need to validate to see if this >> is still true. A simple way to start would be benchmarking > >The original issue was

Re: [PATCH] powerpc/mm: Fix RECLAIM_DISTANCE

2017-01-29 Thread Anton Blanchard
Hi, > Anton suggested that NUMA distances in powerpc mattered and hurted > performance without this setting. We need to validate to see if this > is still true. A simple way to start would be benchmarking The original issue was that we never reclaimed local clean pagecache. I just tried all

Re: [PATCH] powerpc/mm: Fix RECLAIM_DISTANCE

2017-01-27 Thread Balbir Singh
On Wed, Jan 25, 2017 at 03:58:22PM +1100, Gavin Shan wrote: > On Wed, Jan 25, 2017 at 09:27:44AM +0530, Balbir Singh wrote: > >On Tue, Jan 24, 2017 at 10:32:28AM +1100, Gavin Shan wrote: > >> When @node_reclaim_mode ("/proc/sys/vm/zone_reclaim_mode") is enabled, > >> the nodes in the specified

Re: [PATCH] powerpc/mm: Fix RECLAIM_DISTANCE

2017-01-24 Thread Gavin Shan
On Wed, Jan 25, 2017 at 09:27:44AM +0530, Balbir Singh wrote: >On Tue, Jan 24, 2017 at 10:32:28AM +1100, Gavin Shan wrote: >> When @node_reclaim_mode ("/proc/sys/vm/zone_reclaim_mode") is enabled, >> the nodes in the specified distance (< RECLAIM_DISTANCE) to the preferred >> one will be checked

Re: [PATCH] powerpc/mm: Fix RECLAIM_DISTANCE

2017-01-24 Thread Balbir Singh
On Tue, Jan 24, 2017 at 10:32:28AM +1100, Gavin Shan wrote: > When @node_reclaim_mode ("/proc/sys/vm/zone_reclaim_mode") is enabled, > the nodes in the specified distance (< RECLAIM_DISTANCE) to the preferred > one will be checked for page direct reclaim in the fast path, as below > function call

[PATCH] powerpc/mm: Fix RECLAIM_DISTANCE

2017-01-23 Thread Gavin Shan
When @node_reclaim_mode ("/proc/sys/vm/zone_reclaim_mode") is enabled, the nodes in the specified distance (< RECLAIM_DISTANCE) to the preferred one will be checked for page direct reclaim in the fast path, as below function call chain indicates. Currently, RECLAIM_DISTANCE is set to 10, equal to