Re: [RFC 13/13] mm, compaction: fix and improve watermark handling

2016-05-18 Thread Mel Gorman
On Wed, May 18, 2016 at 04:27:53PM +0200, Michal Hocko wrote: > > > > - __compaction_suitable() then checks the low watermark plus a (2 << > > > > order) gap > > > > to decide if there's enough free memory to perform compaction. This > > > > check > > > > > > And this was a real head

Re: [RFC 13/13] mm, compaction: fix and improve watermark handling

2016-05-18 Thread Mel Gorman
On Wed, May 18, 2016 at 04:27:53PM +0200, Michal Hocko wrote: > > > > - __compaction_suitable() then checks the low watermark plus a (2 << > > > > order) gap > > > > to decide if there's enough free memory to perform compaction. This > > > > check > > > > > > And this was a real head

Re: [RFC 13/13] mm, compaction: fix and improve watermark handling

2016-05-18 Thread Michal Hocko
On Wed 18-05-16 14:50:04, Mel Gorman wrote: > On Mon, May 16, 2016 at 11:25:05AM +0200, Michal Hocko wrote: > > On Tue 10-05-16 09:36:03, Vlastimil Babka wrote: > > > Compaction has been using watermark checks when deciding whether it was > > > successful, and whether compaction is at all

Re: [RFC 13/13] mm, compaction: fix and improve watermark handling

2016-05-18 Thread Michal Hocko
On Wed 18-05-16 14:50:04, Mel Gorman wrote: > On Mon, May 16, 2016 at 11:25:05AM +0200, Michal Hocko wrote: > > On Tue 10-05-16 09:36:03, Vlastimil Babka wrote: > > > Compaction has been using watermark checks when deciding whether it was > > > successful, and whether compaction is at all

Re: [RFC 13/13] mm, compaction: fix and improve watermark handling

2016-05-18 Thread Mel Gorman
On Mon, May 16, 2016 at 11:25:05AM +0200, Michal Hocko wrote: > On Tue 10-05-16 09:36:03, Vlastimil Babka wrote: > > Compaction has been using watermark checks when deciding whether it was > > successful, and whether compaction is at all suitable. There are few > > problems > > with these checks.

Re: [RFC 13/13] mm, compaction: fix and improve watermark handling

2016-05-18 Thread Mel Gorman
On Mon, May 16, 2016 at 11:25:05AM +0200, Michal Hocko wrote: > On Tue 10-05-16 09:36:03, Vlastimil Babka wrote: > > Compaction has been using watermark checks when deciding whether it was > > successful, and whether compaction is at all suitable. There are few > > problems > > with these checks.

Re: [RFC 13/13] mm, compaction: fix and improve watermark handling

2016-05-16 Thread Michal Hocko
On Mon 16-05-16 11:50:22, Vlastimil Babka wrote: > On 05/16/2016 11:25 AM, Michal Hocko wrote: > > On Tue 10-05-16 09:36:03, Vlastimil Babka wrote: > > > Compaction has been using watermark checks when deciding whether it was > > > successful, and whether compaction is at all suitable. There are

Re: [RFC 13/13] mm, compaction: fix and improve watermark handling

2016-05-16 Thread Michal Hocko
On Mon 16-05-16 11:50:22, Vlastimil Babka wrote: > On 05/16/2016 11:25 AM, Michal Hocko wrote: > > On Tue 10-05-16 09:36:03, Vlastimil Babka wrote: > > > Compaction has been using watermark checks when deciding whether it was > > > successful, and whether compaction is at all suitable. There are

Re: [RFC 13/13] mm, compaction: fix and improve watermark handling

2016-05-16 Thread Vlastimil Babka
On 05/16/2016 11:25 AM, Michal Hocko wrote: On Tue 10-05-16 09:36:03, Vlastimil Babka wrote: Compaction has been using watermark checks when deciding whether it was successful, and whether compaction is at all suitable. There are few problems with these checks. - __compact_finished() uses low

Re: [RFC 13/13] mm, compaction: fix and improve watermark handling

2016-05-16 Thread Vlastimil Babka
On 05/16/2016 11:25 AM, Michal Hocko wrote: On Tue 10-05-16 09:36:03, Vlastimil Babka wrote: Compaction has been using watermark checks when deciding whether it was successful, and whether compaction is at all suitable. There are few problems with these checks. - __compact_finished() uses low

Re: [RFC 13/13] mm, compaction: fix and improve watermark handling

2016-05-16 Thread Michal Hocko
On Tue 10-05-16 09:36:03, Vlastimil Babka wrote: > Compaction has been using watermark checks when deciding whether it was > successful, and whether compaction is at all suitable. There are few problems > with these checks. > > - __compact_finished() uses low watermark in a check that has to pass

Re: [RFC 13/13] mm, compaction: fix and improve watermark handling

2016-05-16 Thread Michal Hocko
On Tue 10-05-16 09:36:03, Vlastimil Babka wrote: > Compaction has been using watermark checks when deciding whether it was > successful, and whether compaction is at all suitable. There are few problems > with these checks. > > - __compact_finished() uses low watermark in a check that has to pass

[RFC 13/13] mm, compaction: fix and improve watermark handling

2016-05-10 Thread Vlastimil Babka
Compaction has been using watermark checks when deciding whether it was successful, and whether compaction is at all suitable. There are few problems with these checks. - __compact_finished() uses low watermark in a check that has to pass if the direct compaction is to finish and allocation

[RFC 13/13] mm, compaction: fix and improve watermark handling

2016-05-10 Thread Vlastimil Babka
Compaction has been using watermark checks when deciding whether it was successful, and whether compaction is at all suitable. There are few problems with these checks. - __compact_finished() uses low watermark in a check that has to pass if the direct compaction is to finish and allocation