Re: [PATCH v2] memcg: oom: fix totalpages calculation for memory.swappiness==0

2012-11-08 Thread Michal Hocko
On Wed 07-11-12 14:53:40, Andrew Morton wrote: > On Wed, 7 Nov 2012 23:46:40 +0100 > Michal Hocko wrote: > > > > Realistically, is anyone likely to hurt from this? > > > > The primary motivation for the fix was a real report by a customer. > > Describe it please and I'll copy it to the

Re: [PATCH v2] memcg: oom: fix totalpages calculation for memory.swappiness==0

2012-11-08 Thread Michal Hocko
On Wed 07-11-12 14:53:40, Andrew Morton wrote: On Wed, 7 Nov 2012 23:46:40 +0100 Michal Hocko mho...@suse.cz wrote: Realistically, is anyone likely to hurt from this? The primary motivation for the fix was a real report by a customer. Describe it please and I'll copy it to the

Re: [PATCH v2] memcg: oom: fix totalpages calculation for memory.swappiness==0

2012-11-07 Thread Andrew Morton
On Wed, 7 Nov 2012 23:46:40 +0100 Michal Hocko wrote: > > Realistically, is anyone likely to hurt from this? > > The primary motivation for the fix was a real report by a customer. Describe it please and I'll copy it to the changelog. So that Greg can understand why we sent this at him and so

Re: [PATCH v2] memcg: oom: fix totalpages calculation for memory.swappiness==0

2012-11-07 Thread Michal Hocko
On Wed 07-11-12 14:10:25, Andrew Morton wrote: > On Tue, 16 Oct 2012 00:04:08 +0200 > Michal Hocko wrote: > > > As Kosaki correctly pointed out, the glogal reclaim doesn't have this > > issue because we _do_ swap on swappinnes==0 so the swap space has > > to be considered. So the v2 is just acks

Re: [PATCH v2] memcg: oom: fix totalpages calculation for memory.swappiness==0

2012-11-07 Thread Andrew Morton
On Tue, 16 Oct 2012 00:04:08 +0200 Michal Hocko wrote: > As Kosaki correctly pointed out, the glogal reclaim doesn't have this > issue because we _do_ swap on swappinnes==0 so the swap space has > to be considered. So the v2 is just acks + changelog fix. > > Changes since v1 > - drop a note

Re: [PATCH v2] memcg: oom: fix totalpages calculation for memory.swappiness==0

2012-11-07 Thread Andrew Morton
On Tue, 16 Oct 2012 00:04:08 +0200 Michal Hocko mho...@suse.cz wrote: As Kosaki correctly pointed out, the glogal reclaim doesn't have this issue because we _do_ swap on swappinnes==0 so the swap space has to be considered. So the v2 is just acks + changelog fix. Changes since v1 - drop a

Re: [PATCH v2] memcg: oom: fix totalpages calculation for memory.swappiness==0

2012-11-07 Thread Michal Hocko
On Wed 07-11-12 14:10:25, Andrew Morton wrote: On Tue, 16 Oct 2012 00:04:08 +0200 Michal Hocko mho...@suse.cz wrote: As Kosaki correctly pointed out, the glogal reclaim doesn't have this issue because we _do_ swap on swappinnes==0 so the swap space has to be considered. So the v2 is just

Re: [PATCH v2] memcg: oom: fix totalpages calculation for memory.swappiness==0

2012-11-07 Thread Andrew Morton
On Wed, 7 Nov 2012 23:46:40 +0100 Michal Hocko mho...@suse.cz wrote: Realistically, is anyone likely to hurt from this? The primary motivation for the fix was a real report by a customer. Describe it please and I'll copy it to the changelog. So that Greg can understand why we sent this at

[PATCH v2] memcg: oom: fix totalpages calculation for memory.swappiness==0

2012-10-15 Thread Michal Hocko
As Kosaki correctly pointed out, the glogal reclaim doesn't have this issue because we _do_ swap on swappinnes==0 so the swap space has to be considered. So the v2 is just acks + changelog fix. Changes since v1 - drop a note about global swappiness affected as well from the changelog - stable

[PATCH v2] memcg: oom: fix totalpages calculation for memory.swappiness==0

2012-10-15 Thread Michal Hocko
As Kosaki correctly pointed out, the glogal reclaim doesn't have this issue because we _do_ swap on swappinnes==0 so the swap space has to be considered. So the v2 is just acks + changelog fix. Changes since v1 - drop a note about global swappiness affected as well from the changelog - stable