Re: [RFC v5 0/3] mm: make swapin readahead to gain more thp performance

2016-03-03 Thread Ebru Akagunduz
On Fri, Feb 26, 2016 at 09:51:56AM -0500, Rik van Riel wrote: > On Thu, 2016-02-25 at 22:17 -0800, Hugh Dickins wrote: > > On Fri, 26 Feb 2016, Ebru Akagunduz wrote: > > > in Thu, Feb 25, 2016 at 05:35:50PM -0500, Rik van Riel wrote: > >  > > > > Am I forgetting anything obvious? > > > >  > > > > I

Re: [RFC v5 0/3] mm: make swapin readahead to gain more thp performance

2016-02-26 Thread Rik van Riel
On Thu, 2016-02-25 at 22:17 -0800, Hugh Dickins wrote: > On Fri, 26 Feb 2016, Ebru Akagunduz wrote: > > in Thu, Feb 25, 2016 at 05:35:50PM -0500, Rik van Riel wrote: >  > > > Am I forgetting anything obvious? > > >  > > > Is this too aggressive? > > >  > > > Not aggressive enough? > > >  > > > Coul

Re: [RFC v5 0/3] mm: make swapin readahead to gain more thp performance

2016-02-25 Thread Hugh Dickins
On Fri, 26 Feb 2016, Ebru Akagunduz wrote: > in Thu, Feb 25, 2016 at 05:35:50PM -0500, Rik van Riel wrote: > > On Wed, 2016-02-24 at 23:36 -0800, Hugh Dickins wrote: > > >  > > > Doesn't this imply that __collapse_huge_page_swapin() will initiate > > > all > > > the necessary swapins for a THP, the

Re: [RFC v5 0/3] mm: make swapin readahead to gain more thp performance

2016-02-25 Thread Ebru Akagunduz
in Thu, Feb 25, 2016 at 05:35:50PM -0500, Rik van Riel wrote: > On Wed, 2016-02-24 at 23:36 -0800, Hugh Dickins wrote: > >  > > Doesn't this imply that __collapse_huge_page_swapin() will initiate > > all > > the necessary swapins for a THP, then (given the > > FAULT_FLAG_ALLOW_RETRY) > > not wait f

Re: [RFC v5 0/3] mm: make swapin readahead to gain more thp performance

2016-02-25 Thread Ebru Akagunduz
On Wed, Feb 24, 2016 at 11:36:30PM -0800, Hugh Dickins wrote: > On Mon, 14 Sep 2015, Andrew Morton wrote: > > On Mon, 14 Sep 2015 22:31:42 +0300 Ebru Akagunduz > > wrote: > > > > > This patch series makes swapin readahead up to a > > > certain number to gain more thp performance and adds > > > t

Re: [RFC v5 0/3] mm: make swapin readahead to gain more thp performance

2016-02-25 Thread Rik van Riel
On Wed, 2016-02-24 at 23:36 -0800, Hugh Dickins wrote: >  > Doesn't this imply that __collapse_huge_page_swapin() will initiate > all > the necessary swapins for a THP, then (given the > FAULT_FLAG_ALLOW_RETRY) > not wait for them to complete, so khugepaged will give up on that > extent > and move

Re: [RFC v5 0/3] mm: make swapin readahead to gain more thp performance

2016-02-24 Thread Hugh Dickins
On Mon, 14 Sep 2015, Andrew Morton wrote: > On Mon, 14 Sep 2015 22:31:42 +0300 Ebru Akagunduz > wrote: > > > This patch series makes swapin readahead up to a > > certain number to gain more thp performance and adds > > tracepoint for khugepaged_scan_pmd, collapse_huge_page, > > __collapse_huge_p

Re: [RFC v5 0/3] mm: make swapin readahead to gain more thp performance

2015-09-14 Thread Andrew Morton
On Mon, 14 Sep 2015 22:31:42 +0300 Ebru Akagunduz wrote: > This patch series makes swapin readahead up to a > certain number to gain more thp performance and adds > tracepoint for khugepaged_scan_pmd, collapse_huge_page, > __collapse_huge_page_isolate. I'll merge this series for testing. Hopef

[RFC v5 0/3] mm: make swapin readahead to gain more thp performance

2015-09-14 Thread Ebru Akagunduz
This patch series makes swapin readahead up to a certain number to gain more thp performance and adds tracepoint for khugepaged_scan_pmd, collapse_huge_page, __collapse_huge_page_isolate. This patch series was written to deal with programs that access most, but not all, of their memory after they