Re: [PATCH RFC 1/4] mm: throttle MADV_FREE

2015-02-26 Thread Shaohua Li
On Thu, Feb 26, 2015 at 09:42:06AM +0900, Minchan Kim wrote: > Hello, > > On Wed, Feb 25, 2015 at 10:37:48AM -0800, Shaohua Li wrote: > > On Wed, Feb 25, 2015 at 04:11:18PM +0900, Minchan Kim wrote: > > > On Wed, Feb 25, 2015 at 09:08:09AM +0900, Minchan Kim wrote: > > > > Hi Michal, > > > > > >

Re: [PATCH RFC 1/4] mm: throttle MADV_FREE

2015-02-25 Thread Minchan Kim
Hello, On Wed, Feb 25, 2015 at 10:37:48AM -0800, Shaohua Li wrote: > On Wed, Feb 25, 2015 at 04:11:18PM +0900, Minchan Kim wrote: > > On Wed, Feb 25, 2015 at 09:08:09AM +0900, Minchan Kim wrote: > > > Hi Michal, > > > > > > On Tue, Feb 24, 2015 at 04:43:18PM +0100, Michal Hocko wrote: > > > > On

Re: [PATCH RFC 1/4] mm: throttle MADV_FREE

2015-02-25 Thread Shaohua Li
On Wed, Feb 25, 2015 at 04:11:18PM +0900, Minchan Kim wrote: > On Wed, Feb 25, 2015 at 09:08:09AM +0900, Minchan Kim wrote: > > Hi Michal, > > > > On Tue, Feb 24, 2015 at 04:43:18PM +0100, Michal Hocko wrote: > > > On Tue 24-02-15 17:18:14, Minchan Kim wrote: > > > > Recently, Shaohua reported tha

Re: [PATCH RFC 1/4] mm: throttle MADV_FREE

2015-02-25 Thread Michal Hocko
On Wed 25-02-15 16:11:18, Minchan Kim wrote: [...] > diff --git a/mm/madvise.c b/mm/madvise.c > index 6d0fcb8921c2..d41ae76d3e54 100644 > --- a/mm/madvise.c > +++ b/mm/madvise.c > @@ -274,7 +274,9 @@ static int madvise_free_pte_range(pmd_t *pmd, unsigned > long addr, > spinlock_t *ptl; >

Re: [PATCH RFC 1/4] mm: throttle MADV_FREE

2015-02-25 Thread Michal Hocko
On Tue 24-02-15 14:54:01, Shaohua Li wrote: > On Tue, Feb 24, 2015 at 04:43:18PM +0100, Michal Hocko wrote: > > On Tue 24-02-15 17:18:14, Minchan Kim wrote: > > > Recently, Shaohua reported that MADV_FREE is much slower than > > > MADV_DONTNEED in his MADV_FREE bomb test. The reason is many of > >

Re: [PATCH RFC 1/4] mm: throttle MADV_FREE

2015-02-24 Thread Minchan Kim
On Wed, Feb 25, 2015 at 09:08:09AM +0900, Minchan Kim wrote: > Hi Michal, > > On Tue, Feb 24, 2015 at 04:43:18PM +0100, Michal Hocko wrote: > > On Tue 24-02-15 17:18:14, Minchan Kim wrote: > > > Recently, Shaohua reported that MADV_FREE is much slower than > > > MADV_DONTNEED in his MADV_FREE bomb

Re: [PATCH RFC 1/4] mm: throttle MADV_FREE

2015-02-24 Thread Minchan Kim
Hi Michal, On Tue, Feb 24, 2015 at 04:43:18PM +0100, Michal Hocko wrote: > On Tue 24-02-15 17:18:14, Minchan Kim wrote: > > Recently, Shaohua reported that MADV_FREE is much slower than > > MADV_DONTNEED in his MADV_FREE bomb test. The reason is many of > > applications went to stall with direct r

Re: [PATCH RFC 1/4] mm: throttle MADV_FREE

2015-02-24 Thread Shaohua Li
On Tue, Feb 24, 2015 at 04:43:18PM +0100, Michal Hocko wrote: > On Tue 24-02-15 17:18:14, Minchan Kim wrote: > > Recently, Shaohua reported that MADV_FREE is much slower than > > MADV_DONTNEED in his MADV_FREE bomb test. The reason is many of > > applications went to stall with direct reclaim since

Re: [PATCH RFC 1/4] mm: throttle MADV_FREE

2015-02-24 Thread Michal Hocko
On Tue 24-02-15 17:18:14, Minchan Kim wrote: > Recently, Shaohua reported that MADV_FREE is much slower than > MADV_DONTNEED in his MADV_FREE bomb test. The reason is many of > applications went to stall with direct reclaim since kswapd's > reclaim speed isn't fast than applications's allocation sp

[PATCH RFC 1/4] mm: throttle MADV_FREE

2015-02-24 Thread Minchan Kim
Recently, Shaohua reported that MADV_FREE is much slower than MADV_DONTNEED in his MADV_FREE bomb test. The reason is many of applications went to stall with direct reclaim since kswapd's reclaim speed isn't fast than applications's allocation speed so that it causes lots of stall and lock contenti