Re: commit 0bf1457f0cfca7b mm: vmscan: do not swap anon pages just because free+file is low causes heavy performance regression on paging

Hi Rik, On Tue, Apr 22, 2014 at 10:40:17AM -0400, Rik van Riel wrote: On 04/22/2014 07:57 AM, Christian Borntraeger wrote: On 22/04/14 12:55, Christian Borntraeger wrote: While preparing/testing some KVM on s390 patches for the next merge window (target is kvm/next which is based on

commit 0bf1457f0cfca7b mm: vmscan: do not swap anon pages just because free+file is low causes heavy performance regression on paging

While preparing/testing some KVM on s390 patches for the next merge window (target is kvm/next which is based on 3.15-rc1) I faced a very severe performance hickup on guest paging (all anonymous memory). All memory bound guests are in D state now and the system is barely unusable. Reverting

Re: commit 0bf1457f0cfca7b mm: vmscan: do not swap anon pages just because free+file is low causes heavy performance regression on paging

On 22/04/14 12:55, Christian Borntraeger wrote: While preparing/testing some KVM on s390 patches for the next merge window (target is kvm/next which is based on 3.15-rc1) I faced a very severe performance hickup on guest paging (all anonymous memory). All memory bound guests are in D state

Re: commit 0bf1457f0cfca7b mm: vmscan: do not swap anon pages just because free+file is low causes heavy performance regression on paging

On Tue, Apr 22, 2014 at 10:40:17AM -0400, Rik van Riel wrote: On 04/22/2014 07:57 AM, Christian Borntraeger wrote: On 22/04/14 12:55, Christian Borntraeger wrote: While preparing/testing some KVM on s390 patches for the next merge window (target is kvm/next which is based on 3.15-rc1) I

Re: commit 0bf1457f0cfca7b mm: vmscan: do not swap anon pages just because free+file is low causes heavy performance regression on paging

Hi Christian, On Tue, Apr 22, 2014 at 12:55:37PM +0200, Christian Borntraeger wrote: While preparing/testing some KVM on s390 patches for the next merge window (target is kvm/next which is based on 3.15-rc1) I faced a very severe performance hickup on guest paging (all anonymous memory).

Re: commit 0bf1457f0cfca7b mm: vmscan: do not swap anon pages just because free+file is low causes heavy performance regression on paging

On 04/22/2014 07:57 AM, Christian Borntraeger wrote: On 22/04/14 12:55, Christian Borntraeger wrote: While preparing/testing some KVM on s390 patches for the next merge window (target is kvm/next which is based on 3.15-rc1) I faced a very severe performance hickup on guest paging (all

Re: commit 0bf1457f0cfca7b mm: vmscan: do not swap anon pages just because free+file is low causes heavy performance regression on paging

On 22/04/14 17:06, Johannes Weiner wrote: Hi Christian, On Tue, Apr 22, 2014 at 12:55:37PM +0200, Christian Borntraeger wrote: While preparing/testing some KVM on s390 patches for the next merge window (target is kvm/next which is based on 3.15-rc1) I faced a very severe performance

Re: commit 0bf1457f0cfca7b mm: vmscan: do not swap anon pages just because free+file is low causes heavy performance regression on paging

On Tue, Apr 22, 2014 at 11:06:56AM -0400, Johannes Weiner wrote: Hi Christian, On Tue, Apr 22, 2014 at 12:55:37PM +0200, Christian Borntraeger wrote: While preparing/testing some KVM on s390 patches for the next merge window (target is kvm/next which is based on 3.15-rc1) I faced a very