Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On 10/10/2017 05:39 PM, Dave Hansen wrote: > On 10/10/2017 07:57 AM, Michal Hocko wrote: >>> But, let's be honest, this leaves us with an option that nobody is ever >>> going to turn on. IOW, nobody except a very small portion of our users >>> will ever see any benefit from this. >> But aren't

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On 10/10/2017 05:39 PM, Dave Hansen wrote: > On 10/10/2017 07:57 AM, Michal Hocko wrote: >>> But, let's be honest, this leaves us with an option that nobody is ever >>> going to turn on. IOW, nobody except a very small portion of our users >>> will ever see any benefit from this. >> But aren't

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On Tue, 10 Oct 2017 19:51:43 +0200 Michal Hocko wrote: > On Tue 10-10-17 08:39:47, Dave Hansen wrote: > > On 10/10/2017 07:57 AM, Michal Hocko wrote: > > >> But, let's be honest, this leaves us with an option that nobody is ever > > >> going to turn on. IOW, nobody except a

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On Tue, 10 Oct 2017 19:51:43 +0200 Michal Hocko wrote: > On Tue 10-10-17 08:39:47, Dave Hansen wrote: > > On 10/10/2017 07:57 AM, Michal Hocko wrote: > > >> But, let's be honest, this leaves us with an option that nobody is ever > > >> going to turn on. IOW, nobody except a very small portion

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On Tue 10-10-17 08:39:47, Dave Hansen wrote: > On 10/10/2017 07:57 AM, Michal Hocko wrote: > >> But, let's be honest, this leaves us with an option that nobody is ever > >> going to turn on. IOW, nobody except a very small portion of our users > >> will ever see any benefit from this. > > But

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On Tue 10-10-17 08:39:47, Dave Hansen wrote: > On 10/10/2017 07:57 AM, Michal Hocko wrote: > >> But, let's be honest, this leaves us with an option that nobody is ever > >> going to turn on. IOW, nobody except a very small portion of our users > >> will ever see any benefit from this. > > But

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On 10/10/2017 07:57 AM, Michal Hocko wrote: >> But, let's be honest, this leaves us with an option that nobody is ever >> going to turn on. IOW, nobody except a very small portion of our users >> will ever see any benefit from this. > But aren't those small groups who would like to squeeze every

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On 10/10/2017 07:57 AM, Michal Hocko wrote: >> But, let's be honest, this leaves us with an option that nobody is ever >> going to turn on. IOW, nobody except a very small portion of our users >> will ever see any benefit from this. > But aren't those small groups who would like to squeeze every

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On Tue, 10 Oct 2017, Michal Hocko wrote: > > But, let's be honest, this leaves us with an option that nobody is ever > > going to turn on. IOW, nobody except a very small portion of our users > > will ever see any benefit from this. > > But aren't those small groups who would like to squeeze

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On Tue, 10 Oct 2017, Michal Hocko wrote: > > But, let's be honest, this leaves us with an option that nobody is ever > > going to turn on. IOW, nobody except a very small portion of our users > > will ever see any benefit from this. > > But aren't those small groups who would like to squeeze

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On Tue 10-10-17 07:53:50, Dave Hansen wrote: > On 10/10/2017 07:31 AM, Michal Hocko wrote: > > On Tue 10-10-17 07:29:31, Dave Hansen wrote: > >> On 10/09/2017 10:49 PM, Michal Hocko wrote: > >>> Anyway I still stand by my position that this sounds over-engineered and > >>> a simple 0/1 resp.

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On Tue 10-10-17 07:53:50, Dave Hansen wrote: > On 10/10/2017 07:31 AM, Michal Hocko wrote: > > On Tue 10-10-17 07:29:31, Dave Hansen wrote: > >> On 10/09/2017 10:49 PM, Michal Hocko wrote: > >>> Anyway I still stand by my position that this sounds over-engineered and > >>> a simple 0/1 resp.

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On 10/10/2017 07:31 AM, Michal Hocko wrote: > On Tue 10-10-17 07:29:31, Dave Hansen wrote: >> On 10/09/2017 10:49 PM, Michal Hocko wrote: >>> Anyway I still stand by my position that this sounds over-engineered and >>> a simple 0/1 resp. on/off interface would be both simpler and safer. If >>>

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On 10/10/2017 07:31 AM, Michal Hocko wrote: > On Tue 10-10-17 07:29:31, Dave Hansen wrote: >> On 10/09/2017 10:49 PM, Michal Hocko wrote: >>> Anyway I still stand by my position that this sounds over-engineered and >>> a simple 0/1 resp. on/off interface would be both simpler and safer. If >>>

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On Tue 10-10-17 07:29:31, Dave Hansen wrote: > On 10/09/2017 10:49 PM, Michal Hocko wrote: > > On Mon 09-10-17 09:55:49, Michal Hocko wrote: > >> I haven't checked closely but what happens (or should happen) when you > >> do a partial read? Should you get an inconsistent results? Or is this > >>

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On Tue 10-10-17 07:29:31, Dave Hansen wrote: > On 10/09/2017 10:49 PM, Michal Hocko wrote: > > On Mon 09-10-17 09:55:49, Michal Hocko wrote: > >> I haven't checked closely but what happens (or should happen) when you > >> do a partial read? Should you get an inconsistent results? Or is this > >>

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On 10/09/2017 10:49 PM, Michal Hocko wrote: > On Mon 09-10-17 09:55:49, Michal Hocko wrote: >> I haven't checked closely but what happens (or should happen) when you >> do a partial read? Should you get an inconsistent results? Or is this >> impossible? > Well, after thinking about it little bit

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On 10/09/2017 10:49 PM, Michal Hocko wrote: > On Mon 09-10-17 09:55:49, Michal Hocko wrote: >> I haven't checked closely but what happens (or should happen) when you >> do a partial read? Should you get an inconsistent results? Or is this >> impossible? > Well, after thinking about it little bit

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On 2017年10月10日 13:49, Michal Hocko wrote: > On Mon 09-10-17 09:55:49, Michal Hocko wrote: >> I haven't checked closely but what happens (or should happen) when you >> do a partial read? Should you get an inconsistent results? Or is this >> impossible? > > Well, after thinking about it little

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On 2017年10月10日 13:49, Michal Hocko wrote: > On Mon 09-10-17 09:55:49, Michal Hocko wrote: >> I haven't checked closely but what happens (or should happen) when you >> do a partial read? Should you get an inconsistent results? Or is this >> impossible? > > Well, after thinking about it little

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On Mon 09-10-17 09:55:49, Michal Hocko wrote: > I haven't checked closely but what happens (or should happen) when you > do a partial read? Should you get an inconsistent results? Or is this > impossible? Well, after thinking about it little bit more, partial reads are always inconsistent so this

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On Mon 09-10-17 09:55:49, Michal Hocko wrote: > I haven't checked closely but what happens (or should happen) when you > do a partial read? Should you get an inconsistent results? Or is this > impossible? Well, after thinking about it little bit more, partial reads are always inconsistent so this

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On Mon 09-10-17 14:34:11, kemi wrote: > > > On 2017年10月03日 17:23, Michal Hocko wrote: > > On Thu 28-09-17 14:11:41, Kemi Wang wrote: > >> This is the second step which introduces a tunable interface that allow > >> numa stats configurable for optimizing zone_statistics(), as suggested by > >>

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On Mon 09-10-17 14:34:11, kemi wrote: > > > On 2017年10月03日 17:23, Michal Hocko wrote: > > On Thu 28-09-17 14:11:41, Kemi Wang wrote: > >> This is the second step which introduces a tunable interface that allow > >> numa stats configurable for optimizing zone_statistics(), as suggested by > >>

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On 2017年10月03日 17:23, Michal Hocko wrote: > On Thu 28-09-17 14:11:41, Kemi Wang wrote: >> This is the second step which introduces a tunable interface that allow >> numa stats configurable for optimizing zone_statistics(), as suggested by >> Dave Hansen and Ying Huang. >> >>

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On 2017年10月03日 17:23, Michal Hocko wrote: > On Thu 28-09-17 14:11:41, Kemi Wang wrote: >> This is the second step which introduces a tunable interface that allow >> numa stats configurable for optimizing zone_statistics(), as suggested by >> Dave Hansen and Ying Huang. >> >>

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On 2017年09月29日 15:03, Vlastimil Babka wrote: > On 09/28/2017 08:11 AM, Kemi Wang wrote: >> This is the second step which introduces a tunable interface that allow >> numa stats configurable for optimizing zone_statistics(), as suggested by >> Dave Hansen and Ying Huang. >> >>

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On 2017年09月29日 15:03, Vlastimil Babka wrote: > On 09/28/2017 08:11 AM, Kemi Wang wrote: >> This is the second step which introduces a tunable interface that allow >> numa stats configurable for optimizing zone_statistics(), as suggested by >> Dave Hansen and Ying Huang. >> >>

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On Thu 28-09-17 14:11:41, Kemi Wang wrote: > This is the second step which introduces a tunable interface that allow > numa stats configurable for optimizing zone_statistics(), as suggested by > Dave Hansen and Ying Huang. > >

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On Thu 28-09-17 14:11:41, Kemi Wang wrote: > This is the second step which introduces a tunable interface that allow > numa stats configurable for optimizing zone_statistics(), as suggested by > Dave Hansen and Ying Huang. > >

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

[+CC linux-api] On 09/28/2017 08:11 AM, Kemi Wang wrote: > This is the second step which introduces a tunable interface that allow > numa stats configurable for optimizing zone_statistics(), as suggested by > Dave Hansen and Ying Huang. > >

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

[+CC linux-api] On 09/28/2017 08:11 AM, Kemi Wang wrote: > This is the second step which introduces a tunable interface that allow > numa stats configurable for optimizing zone_statistics(), as suggested by > Dave Hansen and Ying Huang. > >

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On 09/28/2017 11:29 PM, Andrew Morton wrote: > On Thu, 28 Sep 2017 14:11:41 +0800 Kemi Wang wrote: > >> This is the second step which introduces a tunable interface that allow >> numa stats configurable for optimizing zone_statistics(), as suggested by >> Dave Hansen and

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On 09/28/2017 11:29 PM, Andrew Morton wrote: > On Thu, 28 Sep 2017 14:11:41 +0800 Kemi Wang wrote: > >> This is the second step which introduces a tunable interface that allow >> numa stats configurable for optimizing zone_statistics(), as suggested by >> Dave Hansen and Ying Huang. > > Looks

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On 09/28/2017 08:11 AM, Kemi Wang wrote: > This is the second step which introduces a tunable interface that allow > numa stats configurable for optimizing zone_statistics(), as suggested by > Dave Hansen and Ying Huang. > >

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On 09/28/2017 08:11 AM, Kemi Wang wrote: > This is the second step which introduces a tunable interface that allow > numa stats configurable for optimizing zone_statistics(), as suggested by > Dave Hansen and Ying Huang. > >

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On 2017年09月29日 05:29, Andrew Morton wrote: > On Thu, 28 Sep 2017 14:11:41 +0800 Kemi Wang wrote: > >> This is the second step which introduces a tunable interface that allow >> numa stats configurable for optimizing zone_statistics(), as suggested by >> Dave Hansen and

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On 2017年09月29日 05:29, Andrew Morton wrote: > On Thu, 28 Sep 2017 14:11:41 +0800 Kemi Wang wrote: > >> This is the second step which introduces a tunable interface that allow >> numa stats configurable for optimizing zone_statistics(), as suggested by >> Dave Hansen and Ying Huang. > > Looks

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On Thu, 28 Sep 2017 14:11:41 +0800 Kemi Wang wrote: > This is the second step which introduces a tunable interface that allow > numa stats configurable for optimizing zone_statistics(), as suggested by > Dave Hansen and Ying Huang. Looks OK I guess. I fiddled with it a

Re: [PATCH v3] mm, sysctl: make NUMA stats configurable

On Thu, 28 Sep 2017 14:11:41 +0800 Kemi Wang wrote: > This is the second step which introduces a tunable interface that allow > numa stats configurable for optimizing zone_statistics(), as suggested by > Dave Hansen and Ying Huang. Looks OK I guess. I fiddled with it a lot. Please consider:

[PATCH v3] mm, sysctl: make NUMA stats configurable

This is the second step which introduces a tunable interface that allow numa stats configurable for optimizing zone_statistics(), as suggested by Dave Hansen and Ying Huang. = When page allocation performance becomes a

[PATCH v3] mm, sysctl: make NUMA stats configurable

This is the second step which introduces a tunable interface that allow numa stats configurable for optimizing zone_statistics(), as suggested by Dave Hansen and Ying Huang. = When page allocation performance becomes a