在 2012-9-5，下午4:11，Minchan Kim 写道：
> This patch introudes MIGRATE_DISCARD mode in migration.
> It drops *clean cache pages* instead of migration so that
> migration latency could be reduced by avoiding (memcpy + page remapping).
> It's useful for CMA because latency of migration is very
在 2012-9-5，下午4:11，Minchan Kim minc...@kernel.org 写道：
This patch introudes MIGRATE_DISCARD mode in migration.
It drops *clean cache pages* instead of migration so that
migration latency could be reduced by avoiding (memcpy + page remapping).
It's useful for CMA because latency of migration is
On Fri, Sep 07, 2012 at 10:32:03AM +0100, Mel Gorman wrote:
> On Fri, Sep 07, 2012 at 09:21:45AM +0100, Mel Gorman wrote:
> >
> > So other than the mix up of order parameters I think this should work.
> >
>
> But I'd be wrong, isolated page accounting is not fixed up so it will
> eventually
On Fri, Sep 07, 2012 at 09:21:45AM +0100, Mel Gorman wrote:
>
> So other than the mix up of order parameters I think this should work.
>
But I'd be wrong, isolated page accounting is not fixed up so it will
eventually hang on too_many_isolated. It turns out it is necessary
to pass in zone after
On 9/7/12, Minchan Kim wrote:
> Hi Kyungmin,
>
> On Fri, Sep 07, 2012 at 04:31:17PM +0900, Kyungmin Park wrote:
>> On 9/7/12, Kyungmin Park wrote:
>> > Hi Minchan,
>> >
>> > I tested Mel patch again with ClearPageActive(page). but after some
>> > testing, it's stall and can't return from
>> >
On Fri, Sep 07, 2012 at 11:24:34AM +0900, Minchan Kim wrote:
> > +unsigned long reclaim_clean_pages_from_list(struct list_head *page_list)
> > +{
> > + struct scan_control sc = {
> > + .gfp_mask = GFP_KERNEL,
> > + .priority = DEF_PRIORITY,
> > + };
> > + unsigned long
Hi Kyungmin,
On Fri, Sep 07, 2012 at 04:31:17PM +0900, Kyungmin Park wrote:
> On 9/7/12, Kyungmin Park wrote:
> > Hi Minchan,
> >
> > I tested Mel patch again with ClearPageActive(page). but after some
> > testing, it's stall and can't return from
> > reclaim_clean_pages_from_list().
> >
> >
On Fri, Sep 07, 2012 at 11:26:01AM +0900, Minchan Kim wrote:
> On Fri, Sep 07, 2012 at 09:57:12AM +0900, Kyungmin Park wrote:
> > Hi Mel,
> >
> > After apply your patch, It got the below message.
> > Please note that it's not the latest kernel. it's kernel v3.0.31 + CMA
> > + your patch.
> > It
On 9/7/12, Kyungmin Park wrote:
> Hi Minchan,
>
> I tested Mel patch again with ClearPageActive(page). but after some
> testing, it's stall and can't return from
> reclaim_clean_pages_from_list().
>
> Maybe it's related with unmap feature from yours?
> stall is not happened from your codes until
On 9/7/12, Kyungmin Park kmp...@infradead.org wrote:
Hi Minchan,
I tested Mel patch again with ClearPageActive(page). but after some
testing, it's stall and can't return from
reclaim_clean_pages_from_list(cc.migratepages).
Maybe it's related with unmap feature from yours?
stall is not
On Fri, Sep 07, 2012 at 11:26:01AM +0900, Minchan Kim wrote:
On Fri, Sep 07, 2012 at 09:57:12AM +0900, Kyungmin Park wrote:
Hi Mel,
After apply your patch, It got the below message.
Please note that it's not the latest kernel. it's kernel v3.0.31 + CMA
+ your patch.
It seems it
Hi Kyungmin,
On Fri, Sep 07, 2012 at 04:31:17PM +0900, Kyungmin Park wrote:
On 9/7/12, Kyungmin Park kmp...@infradead.org wrote:
Hi Minchan,
I tested Mel patch again with ClearPageActive(page). but after some
testing, it's stall and can't return from
On Fri, Sep 07, 2012 at 11:24:34AM +0900, Minchan Kim wrote:
+unsigned long reclaim_clean_pages_from_list(struct list_head *page_list)
+{
+ struct scan_control sc = {
+ .gfp_mask = GFP_KERNEL,
+ .priority = DEF_PRIORITY,
+ };
+ unsigned long ret;
+
On 9/7/12, Minchan Kim minc...@kernel.org wrote:
Hi Kyungmin,
On Fri, Sep 07, 2012 at 04:31:17PM +0900, Kyungmin Park wrote:
On 9/7/12, Kyungmin Park kmp...@infradead.org wrote:
Hi Minchan,
I tested Mel patch again with ClearPageActive(page). but after some
testing, it's stall and
On Fri, Sep 07, 2012 at 09:21:45AM +0100, Mel Gorman wrote:
So other than the mix up of order parameters I think this should work.
But I'd be wrong, isolated page accounting is not fixed up so it will
eventually hang on too_many_isolated. It turns out it is necessary
to pass in zone after
On Fri, Sep 07, 2012 at 10:32:03AM +0100, Mel Gorman wrote:
On Fri, Sep 07, 2012 at 09:21:45AM +0100, Mel Gorman wrote:
So other than the mix up of order parameters I think this should work.
But I'd be wrong, isolated page accounting is not fixed up so it will
eventually hang on
Hi Minchan,
I tested Mel patch again with ClearPageActive(page). but after some
testing, it's stall and can't return from
reclaim_clean_pages_from_list().
Maybe it's related with unmap feature from yours?
stall is not happened from your codes until now.
I'll test it more and report any issue if
On Fri, Sep 07, 2012 at 09:57:12AM +0900, Kyungmin Park wrote:
> Hi Mel,
>
> After apply your patch, It got the below message.
> Please note that it's not the latest kernel. it's kernel v3.0.31 + CMA
> + your patch.
> It seems it should not be active but it contains active field.
Yeb. At the
On Thu, Sep 06, 2012 at 10:03:25AM +0100, Mel Gorman wrote:
> On Thu, Sep 06, 2012 at 09:29:35AM +0100, Mel Gorman wrote:
> > On Thu, Sep 06, 2012 at 02:31:12PM +0900, Minchan Kim wrote:
> > > Hi Mel,
> > >
> > > On Wed, Sep 05, 2012 at 11:56:11AM +0100, Mel Gorman wrote:
> > > > On Wed, Sep 05,
Hi Mel,
After apply your patch, It got the below message.
Please note that it's not the latest kernel. it's kernel v3.0.31 + CMA
+ your patch.
It seems it should not be active but it contains active field.
Thank you,
Kyungmin Park
[ 79.160394] c1 BUG: Bad page state in process mediaserver
On Thu, Sep 06, 2012 at 09:29:35AM +0100, Mel Gorman wrote:
> On Thu, Sep 06, 2012 at 02:31:12PM +0900, Minchan Kim wrote:
> > Hi Mel,
> >
> > On Wed, Sep 05, 2012 at 11:56:11AM +0100, Mel Gorman wrote:
> > > On Wed, Sep 05, 2012 at 05:11:13PM +0900, Minchan Kim wrote:
> > > > This patch
On Thu, Sep 06, 2012 at 02:31:12PM +0900, Minchan Kim wrote:
> Hi Mel,
>
> On Wed, Sep 05, 2012 at 11:56:11AM +0100, Mel Gorman wrote:
> > On Wed, Sep 05, 2012 at 05:11:13PM +0900, Minchan Kim wrote:
> > > This patch introudes MIGRATE_DISCARD mode in migration.
> > > It drops *clean cache pages*
On Thu, Sep 06, 2012 at 02:31:12PM +0900, Minchan Kim wrote:
Hi Mel,
On Wed, Sep 05, 2012 at 11:56:11AM +0100, Mel Gorman wrote:
On Wed, Sep 05, 2012 at 05:11:13PM +0900, Minchan Kim wrote:
This patch introudes MIGRATE_DISCARD mode in migration.
It drops *clean cache pages* instead of
On Thu, Sep 06, 2012 at 09:29:35AM +0100, Mel Gorman wrote:
On Thu, Sep 06, 2012 at 02:31:12PM +0900, Minchan Kim wrote:
Hi Mel,
On Wed, Sep 05, 2012 at 11:56:11AM +0100, Mel Gorman wrote:
On Wed, Sep 05, 2012 at 05:11:13PM +0900, Minchan Kim wrote:
This patch introudes
Hi Mel,
After apply your patch, It got the below message.
Please note that it's not the latest kernel. it's kernel v3.0.31 + CMA
+ your patch.
It seems it should not be active but it contains active field.
Thank you,
Kyungmin Park
[ 79.160394] c1 BUG: Bad page state in process mediaserver
On Thu, Sep 06, 2012 at 10:03:25AM +0100, Mel Gorman wrote:
On Thu, Sep 06, 2012 at 09:29:35AM +0100, Mel Gorman wrote:
On Thu, Sep 06, 2012 at 02:31:12PM +0900, Minchan Kim wrote:
Hi Mel,
On Wed, Sep 05, 2012 at 11:56:11AM +0100, Mel Gorman wrote:
On Wed, Sep 05, 2012 at
On Fri, Sep 07, 2012 at 09:57:12AM +0900, Kyungmin Park wrote:
Hi Mel,
After apply your patch, It got the below message.
Please note that it's not the latest kernel. it's kernel v3.0.31 + CMA
+ your patch.
It seems it should not be active but it contains active field.
Yeb. At the moment,
Hi Minchan,
I tested Mel patch again with ClearPageActive(page). but after some
testing, it's stall and can't return from
reclaim_clean_pages_from_list(cc.migratepages).
Maybe it's related with unmap feature from yours?
stall is not happened from your codes until now.
I'll test it more and
Hi Mel,
On Wed, Sep 05, 2012 at 11:56:11AM +0100, Mel Gorman wrote:
> On Wed, Sep 05, 2012 at 05:11:13PM +0900, Minchan Kim wrote:
> > This patch introudes MIGRATE_DISCARD mode in migration.
> > It drops *clean cache pages* instead of migration so that
> > migration latency could be reduced by
On Wed, Sep 05 2012, Minchan Kim wrote:
> This patch introudes MIGRATE_DISCARD mode in migration.
> It drops *clean cache pages* instead of migration so that
> migration latency could be reduced by avoiding (memcpy + page remapping).
> It's useful for CMA because latency of migration is very
On Wed, Sep 05, 2012 at 05:11:13PM +0900, Minchan Kim wrote:
> This patch introudes MIGRATE_DISCARD mode in migration.
> It drops *clean cache pages* instead of migration so that
> migration latency could be reduced by avoiding (memcpy + page remapping).
> It's useful for CMA because latency of
Hi Mel,
On Wed, Sep 05, 2012 at 11:56:11AM +0100, Mel Gorman wrote:
On Wed, Sep 05, 2012 at 05:11:13PM +0900, Minchan Kim wrote:
This patch introudes MIGRATE_DISCARD mode in migration.
It drops *clean cache pages* instead of migration so that
migration latency could be reduced by avoiding
On Wed, Sep 05, 2012 at 05:11:13PM +0900, Minchan Kim wrote:
This patch introudes MIGRATE_DISCARD mode in migration.
It drops *clean cache pages* instead of migration so that
migration latency could be reduced by avoiding (memcpy + page remapping).
It's useful for CMA because latency of
On Wed, Sep 05 2012, Minchan Kim wrote:
This patch introudes MIGRATE_DISCARD mode in migration.
It drops *clean cache pages* instead of migration so that
migration latency could be reduced by avoiding (memcpy + page remapping).
It's useful for CMA because latency of migration is very important
34 matches
Mail list logo
