Hi,
CCing lkml on this,
* Yannick Brosseau (yannick.bross...@gmail.com) wrote:
> Hi all,
>
> We discovered a performance regression in recent kernels with LTTng
> related to the use of fadvise DONTNEED.
> A call to this syscall is present in the LTTng consumer.
>
> The following kernel commit c
On Mon, 17 Jun 2013 10:13:57 -0400 Mathieu Desnoyers
wrote:
> Hi,
>
> CCing lkml on this,
>
> * Yannick Brosseau (yannick.bross...@gmail.com) wrote:
> > Hi all,
> >
> > We discovered a performance regression in recent kernels with LTTng
> > related to the use of fadvise DONTNEED.
> > A call t
On Mon, Jun 17, 2013 at 02:24:59PM -0700, Andrew Morton wrote:
> On Mon, 17 Jun 2013 10:13:57 -0400 Mathieu Desnoyers
> wrote:
>
> > Hi,
> >
> > CCing lkml on this,
> >
> > * Yannick Brosseau (yannick.bross...@gmail.com) wrote:
> > > Hi all,
> > >
> > > We discovered a performance regression
On Tue, Jun 18, 2013 at 10:29:26AM +0100, Mel Gorman wrote:
> On Mon, Jun 17, 2013 at 02:24:59PM -0700, Andrew Morton wrote:
> > On Mon, 17 Jun 2013 10:13:57 -0400 Mathieu Desnoyers
> > wrote:
> >
> > > Hi,
> > >
> > > CCing lkml on this,
> > >
> > > * Yannick Brosseau (yannick.bross...@gmail.
* Mel Gorman (mgor...@suse.de) wrote:
> On Tue, Jun 18, 2013 at 10:29:26AM +0100, Mel Gorman wrote:
> > On Mon, Jun 17, 2013 at 02:24:59PM -0700, Andrew Morton wrote:
> > > On Mon, 17 Jun 2013 10:13:57 -0400 Mathieu Desnoyers
> > > wrote:
> > >
> > > > Hi,
> > > >
> > > > CCing lkml on this,
>
* Rob van der Heij (rvdh...@gmail.com) wrote:
> Wouldn't you batch the calls to drop the pages from cache rather than drop
> one packet at a time?
By default for kernel tracing, lttng's trace packets are 1MB, so I
consider the call to fadvise to be already batched by applying it to 1MB
packets rat
On Thu, Jun 20, 2013 at 08:20:16AM -0400, Mathieu Desnoyers wrote:
> * Rob van der Heij (rvdh...@gmail.com) wrote:
> > Wouldn't you batch the calls to drop the pages from cache rather than drop
> > one packet at a time?
>
> By default for kernel tracing, lttng's trace packets are 1MB, so I
> consi
* Dave Chinner (da...@fromorbit.com) wrote:
> On Thu, Jun 20, 2013 at 08:20:16AM -0400, Mathieu Desnoyers wrote:
> > * Rob van der Heij (rvdh...@gmail.com) wrote:
> > > Wouldn't you batch the calls to drop the pages from cache rather than drop
> > > one packet at a time?
> >
> > By default for ker
* Mathieu Desnoyers (mathieu.desnoy...@efficios.com) wrote:
> * Dave Chinner (da...@fromorbit.com) wrote:
> > On Thu, Jun 20, 2013 at 08:20:16AM -0400, Mathieu Desnoyers wrote:
> > > * Rob van der Heij (rvdh...@gmail.com) wrote:
> > > > Wouldn't you batch the calls to drop the pages from cache rath
On Tue, Jul 02, 2013 at 08:55:14PM -0400, Mathieu Desnoyers wrote:
> * Mathieu Desnoyers (mathieu.desnoy...@efficios.com) wrote:
> > * Dave Chinner (da...@fromorbit.com) wrote:
> > > On Thu, Jun 20, 2013 at 08:20:16AM -0400, Mathieu Desnoyers wrote:
> > > > * Rob van der Heij (rvdh...@gmail.com) wr
Mel Gorman writes:
>> > I just tried replacing my sync_file_range()+fadvise() calls and instead
>> > pass the O_DIRECT flag to open(). Unfortunately, I must be doing
>> > something very wrong, because I get only 1/3rd of the throughput, and
>> > the page cache fills up. Any idea why ?
>>
>> Sinc
On 2013-07-03 04:47, Mel Gorman wrote:
>> Given it is just a hint, we should be allowed to perform page
>> > deactivation lazily. Is there any fundamental reason to wait for worker
>> > threads on each CPU to complete their lru drain before returning from
>> > fadvise() to user-space ?
>> >
> Only
On Wed, Jul 03, 2013 at 10:53:08AM -0400, Jeff Moyer wrote:
> Mel Gorman writes:
>
> >> > I just tried replacing my sync_file_range()+fadvise() calls and instead
> >> > pass the O_DIRECT flag to open(). Unfortunately, I must be doing
> >> > something very wrong, because I get only 1/3rd of the th
* Dave Chinner (da...@fromorbit.com) wrote:
> On Wed, Jul 03, 2013 at 10:53:08AM -0400, Jeff Moyer wrote:
> > Mel Gorman writes:
> >
> > >> > I just tried replacing my sync_file_range()+fadvise() calls and instead
> > >> > pass the O_DIRECT flag to open(). Unfortunately, I must be doing
> > >> >
On Wed, Jul 03, 2013 at 08:31:03PM -0400, Mathieu Desnoyers wrote:
> * Dave Chinner (da...@fromorbit.com) wrote:
> > On Wed, Jul 03, 2013 at 10:53:08AM -0400, Jeff Moyer wrote:
> > > Mel Gorman writes:
> > >
> > > >> > I just tried replacing my sync_file_range()+fadvise() calls and
> > > >> > in
* Dave Chinner (da...@fromorbit.com) wrote:
> On Wed, Jul 03, 2013 at 08:31:03PM -0400, Mathieu Desnoyers wrote:
> > * Dave Chinner (da...@fromorbit.com) wrote:
> > > On Wed, Jul 03, 2013 at 10:53:08AM -0400, Jeff Moyer wrote:
> > > > Mel Gorman writes:
> > > >
> > > > >> > I just tried replacing
On Wed, Jul 03, 2013 at 02:47:58PM -0400, Yannick Brosseau wrote:
> On 2013-07-03 04:47, Mel Gorman wrote:
> >> Given it is just a hint, we should be allowed to perform page
> >> > deactivation lazily. Is there any fundamental reason to wait for worker
> >> > threads on each CPU to complete their l
On Mon, 17 Jun 2013 17:39:36 -0400 Rapha__l Beamonte
wrote:
> 2013/6/17 Andrew Morton
>
> > That change wasn't terribly efficient - if there are any unpopulated
> > pages in the range (which is quite likely), fadvise() will now always
> > call invalidate_mapping_pages() a second time.
> >
> >
* Andrew Morton (a...@linux-foundation.org) wrote:
> On Mon, 17 Jun 2013 17:39:36 -0400 Rapha__l Beamonte
> wrote:
>
> > 2013/6/17 Andrew Morton
> >
> > > That change wasn't terribly efficient - if there are any unpopulated
> > > pages in the range (which is quite likely), fadvise() will now a
On Mon, 17 Jun 2013 22:15:28 -0400 Mathieu Desnoyers
wrote:
> * Andrew Morton (a...@linux-foundation.org) wrote:
> > On Mon, 17 Jun 2013 17:39:36 -0400 Rapha__l Beamonte
> > wrote:
> >
> > > 2013/6/17 Andrew Morton
> > >
> > > > That change wasn't terribly efficient - if there are any unpop
20 matches
Mail list logo