* Query on PGREFILL, PGSCAN and PGSTEAL
@ 2020-05-06 14:34 Shakeel Butt
2020-05-06 15:14 ` Roman Gushchin
0 siblings, 1 reply; 3+ messages in thread
From: Shakeel Butt @ 2020-05-06 14:34 UTC (permalink / raw)
To: Roman Gushchin; +Cc: Michal Hocko, Johannes Weiner, Andrew Morton, Linux MM
Hi Roman,
The commit 2262185c5b28 ("mm: per-cgroup memory reclaim stats") added
the memcg reclaim stats. However for the global PGSCAN and PGSTEAL
counters it restricts them to be updated only for global reclaim while
the PGREFILL global counter is updated for both global and cgroup
reclaim. Is that intentional?
thanks,
Shakeel
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: Query on PGREFILL, PGSCAN and PGSTEAL
2020-05-06 14:34 Query on PGREFILL, PGSCAN and PGSTEAL Shakeel Butt
@ 2020-05-06 15:14 ` Roman Gushchin
2020-05-06 15:32 ` Shakeel Butt
0 siblings, 1 reply; 3+ messages in thread
From: Roman Gushchin @ 2020-05-06 15:14 UTC (permalink / raw)
To: Shakeel Butt; +Cc: Michal Hocko, Johannes Weiner, Andrew Morton, Linux MM
On Wed, May 06, 2020 at 07:34:16AM -0700, Shakeel Butt wrote:
> Hi Roman,
>
> The commit 2262185c5b28 ("mm: per-cgroup memory reclaim stats") added
> the memcg reclaim stats. However for the global PGSCAN and PGSTEAL
> counters it restricts them to be updated only for global reclaim while
> the PGREFILL global counter is updated for both global and cgroup
> reclaim. Is that intentional?
Hello, Shakeel!
It looks like it was so long before 2262185c5b28. And I don't know any
particular reason for it. I guess it was unintentional.
Thanks!
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: Query on PGREFILL, PGSCAN and PGSTEAL
2020-05-06 15:14 ` Roman Gushchin
@ 2020-05-06 15:32 ` Shakeel Butt
0 siblings, 0 replies; 3+ messages in thread
From: Shakeel Butt @ 2020-05-06 15:32 UTC (permalink / raw)
To: Roman Gushchin; +Cc: Michal Hocko, Johannes Weiner, Andrew Morton, Linux MM
On Wed, May 6, 2020 at 8:15 AM Roman Gushchin <guro@fb.com> wrote:
>
> On Wed, May 06, 2020 at 07:34:16AM -0700, Shakeel Butt wrote:
> > Hi Roman,
> >
> > The commit 2262185c5b28 ("mm: per-cgroup memory reclaim stats") added
> > the memcg reclaim stats. However for the global PGSCAN and PGSTEAL
> > counters it restricts them to be updated only for global reclaim while
> > the PGREFILL global counter is updated for both global and cgroup
> > reclaim. Is that intentional?
>
> Hello, Shakeel!
>
> It looks like it was so long before 2262185c5b28. And I don't know any
> particular reason for it. I guess it was unintentional.
>
Indeed it was done in 904249aa6801 ("mm: fix up the vmscan stat in
vmstat"). I think we should make them consistent. I will see if these
global reclaim checks should be removed.
thanks,
Shakeel
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2020-05-06 15:32 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-05-06 14:34 Query on PGREFILL, PGSCAN and PGSTEAL Shakeel Butt
2020-05-06 15:14 ` Roman Gushchin
2020-05-06 15:32 ` Shakeel Butt
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox