* [PATCH] mm/mempolicy: count MPOL_WEIGHTED_INTERLEAVE to "interleave_hit"
@ 2024-12-27 9:57 Honggyu Kim
2024-12-27 15:37 ` Gregory Price
2024-12-30 1:22 ` Hyeonggon Yoo
0 siblings, 2 replies; 4+ messages in thread
From: Honggyu Kim @ 2024-12-27 9:57 UTC (permalink / raw)
To: Gregory Price
Cc: linux-mm, linux-kernel, kernel_team, Honggyu Kim, Andi Kleen
The fa3bea4e1f82 introduced MPOL_WEIGHTED_INTERLEAVE but it missed
adding its counter to "interleave_hit" of numastat, which is located at
/sys/devices/system/node/nodeN/ directory.
It'd be better to add weighted interleving counter info to the existing
"interleave_hit" instead of introducing a new counter
"weighted_interleave_hit".
Fixes: fa3bea4e1f82 ("mm/mempolicy: introduce MPOL_WEIGHTED_INTERLEAVE for weighted interleaving")
Cc: Gregory Price <gourry@gourry.net>
Cc: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Honggyu Kim <honggyu.kim@sk.com>
---
mm/mempolicy.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index 04f35659717a..162407fbf2bc 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -2268,7 +2268,8 @@ struct page *alloc_pages_mpol_noprof(gfp_t gfp, unsigned int order,
page = __alloc_pages_noprof(gfp, order, nid, nodemask);
- if (unlikely(pol->mode == MPOL_INTERLEAVE) && page) {
+ if (unlikely(pol->mode == MPOL_INTERLEAVE ||
+ pol->mode == MPOL_WEIGHTED_INTERLEAVE) && page) {
/* skip NUMA_INTERLEAVE_HIT update if numa stats is disabled */
if (static_branch_likely(&vm_numa_stat_key) &&
page_to_nid(page) == nid) {
base-commit: 4bbf9020becbfd8fc2c3da790855b7042fad455b
--
2.34.1
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: [PATCH] mm/mempolicy: count MPOL_WEIGHTED_INTERLEAVE to "interleave_hit"
2024-12-27 9:57 [PATCH] mm/mempolicy: count MPOL_WEIGHTED_INTERLEAVE to "interleave_hit" Honggyu Kim
@ 2024-12-27 15:37 ` Gregory Price
2024-12-30 1:22 ` Hyeonggon Yoo
1 sibling, 0 replies; 4+ messages in thread
From: Gregory Price @ 2024-12-27 15:37 UTC (permalink / raw)
To: Honggyu Kim
Cc: Gregory Price, linux-mm, linux-kernel, kernel_team, Andi Kleen
On Fri, Dec 27, 2024 at 06:57:37PM +0900, Honggyu Kim wrote:
> The fa3bea4e1f82 introduced MPOL_WEIGHTED_INTERLEAVE but it missed
> adding its counter to "interleave_hit" of numastat, which is located at
> /sys/devices/system/node/nodeN/ directory.
>
> It'd be better to add weighted interleving counter info to the existing
> "interleave_hit" instead of introducing a new counter
> "weighted_interleave_hit".
>
> Fixes: fa3bea4e1f82 ("mm/mempolicy: introduce MPOL_WEIGHTED_INTERLEAVE for weighted interleaving")
> Cc: Gregory Price <gourry@gourry.net>
> Cc: Andi Kleen <ak@linux.intel.com>
> Signed-off-by: Honggyu Kim <honggyu.kim@sk.com>
> ---
> mm/mempolicy.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
Reviewed-by: Gregory Price <gourry@gourry.net>
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: [PATCH] mm/mempolicy: count MPOL_WEIGHTED_INTERLEAVE to "interleave_hit"
2024-12-27 9:57 [PATCH] mm/mempolicy: count MPOL_WEIGHTED_INTERLEAVE to "interleave_hit" Honggyu Kim
2024-12-27 15:37 ` Gregory Price
@ 2024-12-30 1:22 ` Hyeonggon Yoo
2024-12-31 3:08 ` Honggyu Kim
1 sibling, 1 reply; 4+ messages in thread
From: Hyeonggon Yoo @ 2024-12-30 1:22 UTC (permalink / raw)
To: Honggyu Kim, Gregory Price
Cc: kernel_team, 42.hyeyoo, linux-mm, linux-kernel, Andi Kleen,
Andrew Morton, Yunjeong Mun
On 2024-12-27 6:57 AM, Honggyu Kim wrote:
> The fa3bea4e1f82 introduced MPOL_WEIGHTED_INTERLEAVE but it missed
> adding its counter to "interleave_hit" of numastat, which is located at
> /sys/devices/system/node/nodeN/ directory.
>
> It'd be better to add weighted interleving counter info to the existing
> "interleave_hit" instead of introducing a new counter
> "weighted_interleave_hit".
>
> Fixes: fa3bea4e1f82 ("mm/mempolicy: introduce MPOL_WEIGHTED_INTERLEAVE for weighted interleaving")
> Cc: Gregory Price <gourry@gourry.net>
> Cc: Andi Kleen <ak@linux.intel.com>
> Signed-off-by: Honggyu Kim <honggyu.kim@sk.com>
> ---
> mm/mempolicy.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/mm/mempolicy.c b/mm/mempolicy.c
> index 04f35659717a..162407fbf2bc 100644
> --- a/mm/mempolicy.c
> +++ b/mm/mempolicy.c
> @@ -2268,7 +2268,8 @@ struct page *alloc_pages_mpol_noprof(gfp_t gfp, unsigned int order,
>
> page = __alloc_pages_noprof(gfp, order, nid, nodemask);
>
> - if (unlikely(pol->mode == MPOL_INTERLEAVE) && page) {
> + if (unlikely(pol->mode == MPOL_INTERLEAVE ||
> + pol->mode == MPOL_WEIGHTED_INTERLEAVE) && page) {
> /* skip NUMA_INTERLEAVE_HIT update if numa stats is disabled */
> if (static_branch_likely(&vm_numa_stat_key) &&
> page_to_nid(page) == nid) {
>
> base-commit: 4bbf9020becbfd8fc2c3da790855b7042fad455b
[+Cc Andrew]
Looks good to me,
Reviewed-by: Hyeonggon Yoo <hyeonggon.yoo@sk.com>
We discovered that interleave_hit is not correctly accounted
while testing MPOL_WEIGHTED_INTERLEAVE on libnuma and this patch
fixes the issue.
Tested-by: Yunjeong Mun <yunjeong.mun@sk.com>
For reference:
https://github.com/numactl/numactl/pull/238#issuecomment-2563532910
--
Hyeonggon
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: [PATCH] mm/mempolicy: count MPOL_WEIGHTED_INTERLEAVE to "interleave_hit"
2024-12-30 1:22 ` Hyeonggon Yoo
@ 2024-12-31 3:08 ` Honggyu Kim
0 siblings, 0 replies; 4+ messages in thread
From: Honggyu Kim @ 2024-12-31 3:08 UTC (permalink / raw)
To: Andrew Morton
Cc: kernel_team, 42.hyeyoo, linux-mm, linux-kernel, Hyeonggon Yoo,
Yunjeong Mun, Gregory Price, Andi Kleen
On Mon, 30 Dec 2024 10:22:24 +0900 Hyeonggon Yoo <hyeonggon.yoo@sk.com> wrote:
>
>
> On 2024-12-27 6:57 AM, Honggyu Kim wrote:
> > The fa3bea4e1f82 introduced MPOL_WEIGHTED_INTERLEAVE but it missed
> > adding its counter to "interleave_hit" of numastat, which is located at
> > /sys/devices/system/node/nodeN/ directory.
> >
> > It'd be better to add weighted interleving counter info to the existing
> > "interleave_hit" instead of introducing a new counter
> > "weighted_interleave_hit".
> >
> > Fixes: fa3bea4e1f82 ("mm/mempolicy: introduce MPOL_WEIGHTED_INTERLEAVE for weighted interleaving")
> > Cc: Gregory Price <gourry@gourry.net>
> > Cc: Andi Kleen <ak@linux.intel.com>
> > Signed-off-by: Honggyu Kim <honggyu.kim@sk.com>
> > ---
> > mm/mempolicy.c | 3 ++-
> > 1 file changed, 2 insertions(+), 1 deletion(-)
> >
> > diff --git a/mm/mempolicy.c b/mm/mempolicy.c
> > index 04f35659717a..162407fbf2bc 100644
> > --- a/mm/mempolicy.c
> > +++ b/mm/mempolicy.c
> > @@ -2268,7 +2268,8 @@ struct page *alloc_pages_mpol_noprof(gfp_t gfp, unsigned int order,
> >
> > page = __alloc_pages_noprof(gfp, order, nid, nodemask);
> >
> > - if (unlikely(pol->mode == MPOL_INTERLEAVE) && page) {
> > + if (unlikely(pol->mode == MPOL_INTERLEAVE ||
> > + pol->mode == MPOL_WEIGHTED_INTERLEAVE) && page) {
> > /* skip NUMA_INTERLEAVE_HIT update if numa stats is disabled */
> > if (static_branch_likely(&vm_numa_stat_key) &&
> > page_to_nid(page) == nid) {
> >
> > base-commit: 4bbf9020becbfd8fc2c3da790855b7042fad455b
>
> [+Cc Andrew]
>
> Looks good to me,
> Reviewed-by: Hyeonggon Yoo <hyeonggon.yoo@sk.com>
>
> We discovered that interleave_hit is not correctly accounted
> while testing MPOL_WEIGHTED_INTERLEAVE on libnuma and this patch
> fixes the issue.
>
> Tested-by: Yunjeong Mun <yunjeong.mun@sk.com>
>
> For reference:
> https://github.com/numactl/numactl/pull/238#issuecomment-2563532910
Hi Andrew,
I saw this patch has been merged into mm-unstable only with Gregory's
Reviewed-by tag.
https://lore.kernel.org/mm-commits/20241231002913.D5122C4CED0@smtp.kernel.org
But this patch is contributed by Hyeonggon and Yunjeong together for
their review and testing so please add their tags above in the commit
message together.
Thanks,
Honggyu
> --
> Hyeonggon
>
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2024-12-31 3:08 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2024-12-27 9:57 [PATCH] mm/mempolicy: count MPOL_WEIGHTED_INTERLEAVE to "interleave_hit" Honggyu Kim
2024-12-27 15:37 ` Gregory Price
2024-12-30 1:22 ` Hyeonggon Yoo
2024-12-31 3:08 ` Honggyu Kim
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox