On Wed, Nov 29, 2017 at 03:32:08PM +0900, Joonsoo Kim wrote:
> On Thu, Nov 23, 2017 at 02:08:43PM +, Mel Gorman wrote:
>
> > 3. Another reason a linear scanner was used was because we wanted to
> >clear entire pageblocks we were migrating from and pack the target
> >pageblocks as much
On 11/29/2017 12:34 AM, Andrew Morton wrote:
> On Wed, 22 Nov 2017 15:52:55 +0100 Vlastimil Babka wrote:
>
>>
>> Thanks a lot, that's very encouraging!
>
> Yup.
>
> Should we proceed with this patch for now, or wait for something better
> to come along?
I'm working on the refined version, so w
On Wed, Nov 22, 2017 at 03:52:55PM +0100, Vlastimil Babka wrote:
> On 11/22/2017 03:33 PM, Johannes Weiner wrote:
> > From: Vlastimil Babka
> >
> > The goal of direct compaction is to quickly make a high-order page available
> > for the pending allocation. The free page scanner can add significan
On Thu, Nov 23, 2017 at 02:08:43PM +, Mel Gorman wrote:
> 3. Another reason a linear scanner was used was because we wanted to
>clear entire pageblocks we were migrating from and pack the target
>pageblocks as much as possible. This was to reduce the amount of
>migration required o
On Wed, 22 Nov 2017 15:52:55 +0100 Vlastimil Babka wrote:
> On 11/22/2017 03:33 PM, Johannes Weiner wrote:
> > From: Vlastimil Babka
> >
> > The goal of direct compaction is to quickly make a high-order page available
> > for the pending allocation. The free page scanner can add significant
>
On Fri, Nov 24, 2017 at 02:49:34PM +0100, Vlastimil Babka wrote:
> On 11/24/2017 11:57 AM, Mel Gorman wrote:
> > On Thu, Nov 23, 2017 at 10:15:17PM +0100, Vlastimil Babka wrote:
> >> Hmm this really reads like about the migration scanner. That one is
> >> unchanged by this patch, there is still a l
On 11/24/2017 11:57 AM, Mel Gorman wrote:
> On Thu, Nov 23, 2017 at 10:15:17PM +0100, Vlastimil Babka wrote:
>> Hmm this really reads like about the migration scanner. That one is
>> unchanged by this patch, there is still a linear scanner. In fact, it
>> gets better, because now it can see the who
On Thu, Nov 23, 2017 at 10:15:17PM +0100, Vlastimil Babka wrote:
> On 11/23/2017 03:08 PM, Mel Gorman wrote:
> >
> > 1. This indirectly uses __rmqueue to allocate a MIGRATE_MOVABLE page but
> >that is allowed to fallback to other pageblocks and potentially even
> >steal them. I think it's
On 11/23/2017 03:08 PM, Mel Gorman wrote:
>
> 1. This indirectly uses __rmqueue to allocate a MIGRATE_MOVABLE page but
>that is allowed to fallback to other pageblocks and potentially even
>steal them. I think it's very bad that an attempt to defragment can
>itself indirectly cause mor
On Wed, Nov 22, 2017 at 09:33:21AM -0500, Johannes Weiner wrote:
> From: Vlastimil Babka
>
> The goal of direct compaction is to quickly make a high-order page available
> for the pending allocation. The free page scanner can add significant latency
> when searching for migration targets, althoug
On 11/22/2017 03:33 PM, Johannes Weiner wrote:
> From: Vlastimil Babka
>
> The goal of direct compaction is to quickly make a high-order page available
> for the pending allocation. The free page scanner can add significant latency
> when searching for migration targets, although to succeed the c
From: Vlastimil Babka
The goal of direct compaction is to quickly make a high-order page available
for the pending allocation. The free page scanner can add significant latency
when searching for migration targets, although to succeed the compaction, the
only important limit on the target free pa
12 matches
Mail list logo
