Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1321061 > unrolled thread
| Started by | Michal Hocko <mhocko@kernel.org> |
|---|---|
| First post | 2016-01-28 22:20 +0100 |
| Last post | 2016-01-29 22:40 +0100 |
| Articles | 6 — 4 participants |
Back to article view | Back to linux.kernel
This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by
below is the oldest one visible, not the original post.
[PATCH 5/3] mm, vmscan: make zone_reclaimable_pages more precise Michal Hocko <mhocko@kernel.org> - 2016-01-28 22:20 +0100
Re: [PATCH 5/3] mm, vmscan: make zone_reclaimable_pages more precise David Rientjes <rientjes@google.com> - 2016-01-29 00:30 +0100
Re: [PATCH 5/3] mm, vmscan: make zone_reclaimable_pages more precise "Hillf Danton" <hillf.zj@alibaba-inc.com> - 2016-01-29 04:50 +0100
Re: [PATCH 5/3] mm, vmscan: make zone_reclaimable_pages more precise Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> - 2016-01-29 11:40 +0100
Re: [PATCH 5/3] mm, vmscan: make zone_reclaimable_pages more precise Michal Hocko <mhocko@kernel.org> - 2016-01-29 16:20 +0100
Re: [PATCH 5/3] mm, vmscan: make zone_reclaimable_pages more precise Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> - 2016-01-29 22:40 +0100
| From | Michal Hocko <mhocko@kernel.org> |
|---|---|
| Date | 2016-01-28 22:20 +0100 |
| Subject | [PATCH 5/3] mm, vmscan: make zone_reclaimable_pages more precise |
| Message-ID | <qW28W-R2-3@gated-at.bofh.it> |
From: Michal Hocko <mhocko@suse.com>
zone_reclaimable_pages is used in should_reclaim_retry which uses it to
calculate the target for the watermark check. This means that precise
numbers are important for the correct decision. zone_reclaimable_pages
uses zone_page_state which can contain stale data with per-cpu diffs
not synced yet (the last vmstat_update might have run 1s in the past).
Use zone_page_state_snapshot in zone_reclaimable_pages instead. None
of the current callers is in a hot path where getting the precise value
(which involves per-cpu iteration) would cause an unreasonable overhead.
Suggested-by: David Rientjes <rientjes@google.com>
Signed-off-by: Michal Hocko <mhocko@suse.com>
---
mm/vmscan.c | 14 +++++++-------
1 file changed, 7 insertions(+), 7 deletions(-)
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 489212252cd6..9145e3f89eab 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -196,21 +196,21 @@ unsigned long zone_reclaimable_pages(struct zone *zone)
{
unsigned long nr;
- nr = zone_page_state(zone, NR_ACTIVE_FILE) +
- zone_page_state(zone, NR_INACTIVE_FILE) +
- zone_page_state(zone, NR_ISOLATED_FILE);
+ nr = zone_page_state_snapshot(zone, NR_ACTIVE_FILE) +
+ zone_page_state_snapshot(zone, NR_INACTIVE_FILE) +
+ zone_page_state_snapshot(zone, NR_ISOLATED_FILE);
if (get_nr_swap_pages() > 0)
- nr += zone_page_state(zone, NR_ACTIVE_ANON) +
- zone_page_state(zone, NR_INACTIVE_ANON) +
- zone_page_state(zone, NR_ISOLATED_ANON);
+ nr += zone_page_state_snapshot(zone, NR_ACTIVE_ANON) +
+ zone_page_state_snapshot(zone, NR_INACTIVE_ANON) +
+ zone_page_state_snapshot(zone, NR_ISOLATED_ANON);
return nr;
}
bool zone_reclaimable(struct zone *zone)
{
- return zone_page_state(zone, NR_PAGES_SCANNED) <
+ return zone_page_state_snapshot(zone, NR_PAGES_SCANNED) <
zone_reclaimable_pages(zone) * 6;
}
--
2.7.0.rc3
[toc] | [next] | [standalone]
| From | David Rientjes <rientjes@google.com> |
|---|---|
| Date | 2016-01-29 00:30 +0100 |
| Subject | Re: [PATCH 5/3] mm, vmscan: make zone_reclaimable_pages more precise |
| Message-ID | <qW4aJ-2gf-1@gated-at.bofh.it> |
| In reply to | #1321061 |
On Thu, 28 Jan 2016, Michal Hocko wrote: > From: Michal Hocko <mhocko@suse.com> > > zone_reclaimable_pages is used in should_reclaim_retry which uses it to > calculate the target for the watermark check. This means that precise > numbers are important for the correct decision. zone_reclaimable_pages > uses zone_page_state which can contain stale data with per-cpu diffs > not synced yet (the last vmstat_update might have run 1s in the past). > > Use zone_page_state_snapshot in zone_reclaimable_pages instead. None > of the current callers is in a hot path where getting the precise value > (which involves per-cpu iteration) would cause an unreasonable overhead. > > Suggested-by: David Rientjes <rientjes@google.com> > Signed-off-by: Michal Hocko <mhocko@suse.com> Acked-by: David Rientjes <rientjes@google.com>
[toc] | [prev] | [next] | [standalone]
| From | "Hillf Danton" <hillf.zj@alibaba-inc.com> |
|---|---|
| Date | 2016-01-29 04:50 +0100 |
| Message-ID | <qW8el-5ca-3@gated-at.bofh.it> |
| In reply to | #1321061 |
>
> From: Michal Hocko <mhocko@suse.com>
>
> zone_reclaimable_pages is used in should_reclaim_retry which uses it to
> calculate the target for the watermark check. This means that precise
> numbers are important for the correct decision. zone_reclaimable_pages
> uses zone_page_state which can contain stale data with per-cpu diffs
> not synced yet (the last vmstat_update might have run 1s in the past).
>
> Use zone_page_state_snapshot in zone_reclaimable_pages instead. None
> of the current callers is in a hot path where getting the precise value
> (which involves per-cpu iteration) would cause an unreasonable overhead.
>
> Suggested-by: David Rientjes <rientjes@google.com>
> Signed-off-by: Michal Hocko <mhocko@suse.com>
> ---
Acked-by: Hillf Danton <hillf.zj@alibaba-inc.com>
> mm/vmscan.c | 14 +++++++-------
> 1 file changed, 7 insertions(+), 7 deletions(-)
>
> diff --git a/mm/vmscan.c b/mm/vmscan.c
> index 489212252cd6..9145e3f89eab 100644
> --- a/mm/vmscan.c
> +++ b/mm/vmscan.c
> @@ -196,21 +196,21 @@ unsigned long zone_reclaimable_pages(struct zone *zone)
> {
> unsigned long nr;
>
> - nr = zone_page_state(zone, NR_ACTIVE_FILE) +
> - zone_page_state(zone, NR_INACTIVE_FILE) +
> - zone_page_state(zone, NR_ISOLATED_FILE);
> + nr = zone_page_state_snapshot(zone, NR_ACTIVE_FILE) +
> + zone_page_state_snapshot(zone, NR_INACTIVE_FILE) +
> + zone_page_state_snapshot(zone, NR_ISOLATED_FILE);
>
> if (get_nr_swap_pages() > 0)
> - nr += zone_page_state(zone, NR_ACTIVE_ANON) +
> - zone_page_state(zone, NR_INACTIVE_ANON) +
> - zone_page_state(zone, NR_ISOLATED_ANON);
> + nr += zone_page_state_snapshot(zone, NR_ACTIVE_ANON) +
> + zone_page_state_snapshot(zone, NR_INACTIVE_ANON) +
> + zone_page_state_snapshot(zone, NR_ISOLATED_ANON);
>
> return nr;
> }
>
> bool zone_reclaimable(struct zone *zone)
> {
> - return zone_page_state(zone, NR_PAGES_SCANNED) <
> + return zone_page_state_snapshot(zone, NR_PAGES_SCANNED) <
> zone_reclaimable_pages(zone) * 6;
> }
>
> --
> 2.7.0.rc3
[toc] | [prev] | [next] | [standalone]
| From | Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> |
|---|---|
| Date | 2016-01-29 11:40 +0100 |
| Message-ID | <qWeD8-1qJ-29@gated-at.bofh.it> |
| In reply to | #1321061 |
Michal Hocko wrote: > From: Michal Hocko <mhocko@suse.com> > > zone_reclaimable_pages is used in should_reclaim_retry which uses it to > calculate the target for the watermark check. This means that precise > numbers are important for the correct decision. zone_reclaimable_pages > uses zone_page_state which can contain stale data with per-cpu diffs > not synced yet (the last vmstat_update might have run 1s in the past). > > Use zone_page_state_snapshot in zone_reclaimable_pages instead. None > of the current callers is in a hot path where getting the precise value > (which involves per-cpu iteration) would cause an unreasonable overhead. > > Suggested-by: David Rientjes <rientjes@google.com> > Signed-off-by: Michal Hocko <mhocko@suse.com> > --- > mm/vmscan.c | 14 +++++++------- > 1 file changed, 7 insertions(+), 7 deletions(-) > I didn't know http://lkml.kernel.org/r/20151021130323.GC8805@dhcp22.suse.cz was forgotten. Anyway, Acked-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
[toc] | [prev] | [next] | [standalone]
| From | Michal Hocko <mhocko@kernel.org> |
|---|---|
| Date | 2016-01-29 16:20 +0100 |
| Message-ID | <qWj05-517-19@gated-at.bofh.it> |
| In reply to | #1321641 |
On Fri 29-01-16 19:35:18, Tetsuo Handa wrote: > Michal Hocko wrote: > > From: Michal Hocko <mhocko@suse.com> > > > > zone_reclaimable_pages is used in should_reclaim_retry which uses it to > > calculate the target for the watermark check. This means that precise > > numbers are important for the correct decision. zone_reclaimable_pages > > uses zone_page_state which can contain stale data with per-cpu diffs > > not synced yet (the last vmstat_update might have run 1s in the past). > > > > Use zone_page_state_snapshot in zone_reclaimable_pages instead. None > > of the current callers is in a hot path where getting the precise value > > (which involves per-cpu iteration) would cause an unreasonable overhead. > > > > Suggested-by: David Rientjes <rientjes@google.com> > > Signed-off-by: Michal Hocko <mhocko@suse.com> > > --- > > mm/vmscan.c | 14 +++++++------- > > 1 file changed, 7 insertions(+), 7 deletions(-) > > > > I didn't know http://lkml.kernel.org/r/20151021130323.GC8805@dhcp22.suse.cz > was forgotten. Anyway, OK, that explains why this sounded so familiar. Sorry I comepletely forgot about it. > Acked-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> Can I change it to your Signed-off-by? -- Michal Hocko SUSE Labs
[toc] | [prev] | [next] | [standalone]
| From | Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> |
|---|---|
| Date | 2016-01-29 22:40 +0100 |
| Message-ID | <qWoVQ-MS-7@gated-at.bofh.it> |
| In reply to | #1321823 |
Michal Hocko wrote: > On Fri 29-01-16 19:35:18, Tetsuo Handa wrote: > > Michal Hocko wrote: > > > From: Michal Hocko <mhocko@suse.com> > > > > > > zone_reclaimable_pages is used in should_reclaim_retry which uses it to > > > calculate the target for the watermark check. This means that precise > > > numbers are important for the correct decision. zone_reclaimable_pages > > > uses zone_page_state which can contain stale data with per-cpu diffs > > > not synced yet (the last vmstat_update might have run 1s in the past). > > > > > > Use zone_page_state_snapshot in zone_reclaimable_pages instead. None > > > of the current callers is in a hot path where getting the precise value > > > (which involves per-cpu iteration) would cause an unreasonable overhead. > > > > > > Suggested-by: David Rientjes <rientjes@google.com> > > > Signed-off-by: Michal Hocko <mhocko@suse.com> > > > --- > > > mm/vmscan.c | 14 +++++++------- > > > 1 file changed, 7 insertions(+), 7 deletions(-) > > > > > > > I didn't know http://lkml.kernel.org/r/20151021130323.GC8805@dhcp22.suse.cz > > was forgotten. Anyway, > > OK, that explains why this sounded so familiar. Sorry I comepletely > forgot about it. > > > Acked-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> > > Can I change it to your Signed-off-by? No problem. > > -- > Michal Hocko > SUSE Labs >
[toc] | [prev] | [standalone]
Back to top | Article view | linux.kernel
csiph-web