* Re: [PATCH v3] mm: slub: only warn once when allocating slab obj extensions fails
2025-05-23 16:52 [PATCH v3] mm: slub: only warn once when allocating slab obj extensions fails Usama Arif
@ 2025-05-23 17:08 ` Shakeel Butt
2025-05-23 17:15 ` SeongJae Park
` (3 subsequent siblings)
4 siblings, 0 replies; 7+ messages in thread
From: Shakeel Butt @ 2025-05-23 17:08 UTC (permalink / raw)
To: Usama Arif
Cc: Andrew Morton, surenb, hannes, vlad.wing, linux-mm,
kent.overstreet, cl, rientjes, vbabka, roman.gushchin, harry.yoo,
linux-kernel, kernel-team
On Fri, May 23, 2025 at 05:52:40PM +0100, Usama Arif wrote:
> In memory bound systems, a large number of warnings for failing this
> allocation repeatedly may mask any real issues in the system
> during memory pressure being reported in dmesg. Change this to
> warning only once.
>
> Signed-off-by: Usama Arif <usamaarif642@gmail.com>
> Reported-by: Vlad Poenaru <vlad.wing@gmail.com>
> Closes: https://lore.kernel.org/all/17fab2d6-5a74-4573-bcc3-b75951508f0a@gmail.com/
Reviewed-by: Shakeel Butt <shakeel.butt@linux.dev>
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH v3] mm: slub: only warn once when allocating slab obj extensions fails
2025-05-23 16:52 [PATCH v3] mm: slub: only warn once when allocating slab obj extensions fails Usama Arif
2025-05-23 17:08 ` Shakeel Butt
@ 2025-05-23 17:15 ` SeongJae Park
2025-05-26 9:33 ` Vlastimil Babka
` (2 subsequent siblings)
4 siblings, 0 replies; 7+ messages in thread
From: SeongJae Park @ 2025-05-23 17:15 UTC (permalink / raw)
To: Usama Arif
Cc: SeongJae Park, Andrew Morton, surenb, hannes, shakeel.butt,
vlad.wing, linux-mm, kent.overstreet, cl, rientjes, vbabka,
roman.gushchin, harry.yoo, linux-kernel, kernel-team
On Fri, 23 May 2025 17:52:40 +0100 Usama Arif <usamaarif642@gmail.com> wrote:
> In memory bound systems, a large number of warnings for failing this
> allocation repeatedly may mask any real issues in the system
> during memory pressure being reported in dmesg. Change this to
> warning only once.
>
> Signed-off-by: Usama Arif <usamaarif642@gmail.com>
> Reported-by: Vlad Poenaru <vlad.wing@gmail.com>
> Closes: https://lore.kernel.org/all/17fab2d6-5a74-4573-bcc3-b75951508f0a@gmail.com/
Reviewed-by: SeongJae Park <sj@kernel.org>
Thanks,
SJ
[...]
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH v3] mm: slub: only warn once when allocating slab obj extensions fails
2025-05-23 16:52 [PATCH v3] mm: slub: only warn once when allocating slab obj extensions fails Usama Arif
2025-05-23 17:08 ` Shakeel Butt
2025-05-23 17:15 ` SeongJae Park
@ 2025-05-26 9:33 ` Vlastimil Babka
2025-05-26 10:05 ` Harry Yoo
2025-06-02 21:57 ` Roman Gushchin
4 siblings, 0 replies; 7+ messages in thread
From: Vlastimil Babka @ 2025-05-26 9:33 UTC (permalink / raw)
To: Usama Arif, Andrew Morton, surenb
Cc: hannes, shakeel.butt, vlad.wing, linux-mm, kent.overstreet, cl,
rientjes, roman.gushchin, harry.yoo, linux-kernel, kernel-team
On 5/23/25 18:52, Usama Arif wrote:
> In memory bound systems, a large number of warnings for failing this
> allocation repeatedly may mask any real issues in the system
> during memory pressure being reported in dmesg. Change this to
> warning only once.
>
> Signed-off-by: Usama Arif <usamaarif642@gmail.com>
> Reported-by: Vlad Poenaru <vlad.wing@gmail.com>
> Closes: https://lore.kernel.org/all/17fab2d6-5a74-4573-bcc3-b75951508f0a@gmail.com/
> ---
> v2 -> v3:
> - Put warning back, but only warn once with pr_warn_once.
> v1 -> v2:
> - remove the warning completely. We will have a way in the
> future to indicate that the mem alloc profile is inaccurate.
> ---
> mm/slub.c | 8 +++++---
> 1 file changed, 5 insertions(+), 3 deletions(-)
>
> diff --git a/mm/slub.c b/mm/slub.c
> index dc9e729e1d26..36d7c43a6f2a 100644
> --- a/mm/slub.c
> +++ b/mm/slub.c
> @@ -2102,10 +2102,12 @@ prepare_slab_obj_exts_hook(struct kmem_cache *s, gfp_t flags, void *p)
>
> slab = virt_to_slab(p);
> if (!slab_obj_exts(slab) &&
> - WARN(alloc_slab_obj_exts(slab, s, flags, false),
> - "%s, %s: Failed to create slab extension vector!\n",
> - __func__, s->name))
> + alloc_slab_obj_exts(slab, s, flags, false)) {
> + pr_warn_once("%s, %s: Failed to create slab extension vector!\n",
> + __func__, s->name);
> return NULL;
> + }
> +
I've removed the extra line locally.
Added to slab/for-next, thanks.
>
> return slab_obj_exts(slab) + obj_to_index(s, slab, p);
> }
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH v3] mm: slub: only warn once when allocating slab obj extensions fails
2025-05-23 16:52 [PATCH v3] mm: slub: only warn once when allocating slab obj extensions fails Usama Arif
` (2 preceding siblings ...)
2025-05-26 9:33 ` Vlastimil Babka
@ 2025-05-26 10:05 ` Harry Yoo
2025-06-02 20:14 ` Suren Baghdasaryan
2025-06-02 21:57 ` Roman Gushchin
4 siblings, 1 reply; 7+ messages in thread
From: Harry Yoo @ 2025-05-26 10:05 UTC (permalink / raw)
To: Usama Arif
Cc: Andrew Morton, surenb, hannes, shakeel.butt, vlad.wing, linux-mm,
kent.overstreet, cl, rientjes, vbabka, roman.gushchin,
linux-kernel, kernel-team
On Fri, May 23, 2025 at 05:52:40PM +0100, Usama Arif wrote:
> In memory bound systems, a large number of warnings for failing this
> allocation repeatedly may mask any real issues in the system
> during memory pressure being reported in dmesg. Change this to
> warning only once.
>
> Signed-off-by: Usama Arif <usamaarif642@gmail.com>
> Reported-by: Vlad Poenaru <vlad.wing@gmail.com>
> Closes: https://urldefense.com/v3/__https://lore.kernel.org/all/17fab2d6-5a74-4573-bcc3-b75951508f0a@gmail.com/__;!!ACWV5N9M2RV99hQ!Nv5xX_X4cBQR_HUDO6qfamxI0GEW1eFRCpPj29lANRashIUJDrEWAjU7vlys76FH0voShW07CjpDkzSST46Irg$
> ---
Acked-by: Harry Yoo <harry.yoo@oracle.com>
> v2 -> v3:
> - Put warning back, but only warn once with pr_warn_once.
> v1 -> v2:
> - remove the warning completely. We will have a way in the
> future to indicate that the mem alloc profile is inaccurate.
> ---
> mm/slub.c | 8 +++++---
> 1 file changed, 5 insertions(+), 3 deletions(-)
>
> diff --git a/mm/slub.c b/mm/slub.c
> index dc9e729e1d26..36d7c43a6f2a 100644
> --- a/mm/slub.c
> +++ b/mm/slub.c
> @@ -2102,10 +2102,12 @@ prepare_slab_obj_exts_hook(struct kmem_cache *s, gfp_t flags, void *p)
>
> slab = virt_to_slab(p);
> if (!slab_obj_exts(slab) &&
> - WARN(alloc_slab_obj_exts(slab, s, flags, false),
> - "%s, %s: Failed to create slab extension vector!\n",
> - __func__, s->name))
> + alloc_slab_obj_exts(slab, s, flags, false)) {
> + pr_warn_once("%s, %s: Failed to create slab extension vector!\n",
> + __func__, s->name);
> return NULL;
> + }
> +
>
> return slab_obj_exts(slab) + obj_to_index(s, slab, p);
> }
> --
> 2.47.1
>
--
Cheers,
Harry / Hyeonggon
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH v3] mm: slub: only warn once when allocating slab obj extensions fails
2025-05-26 10:05 ` Harry Yoo
@ 2025-06-02 20:14 ` Suren Baghdasaryan
0 siblings, 0 replies; 7+ messages in thread
From: Suren Baghdasaryan @ 2025-06-02 20:14 UTC (permalink / raw)
To: Harry Yoo
Cc: Usama Arif, Andrew Morton, hannes, shakeel.butt, vlad.wing,
linux-mm, kent.overstreet, cl, rientjes, vbabka, roman.gushchin,
linux-kernel, kernel-team
On Mon, May 26, 2025 at 3:05 AM Harry Yoo <harry.yoo@oracle.com> wrote:
>
> On Fri, May 23, 2025 at 05:52:40PM +0100, Usama Arif wrote:
> > In memory bound systems, a large number of warnings for failing this
> > allocation repeatedly may mask any real issues in the system
> > during memory pressure being reported in dmesg. Change this to
> > warning only once.
> >
> > Signed-off-by: Usama Arif <usamaarif642@gmail.com>
> > Reported-by: Vlad Poenaru <vlad.wing@gmail.com>
> > Closes: https://urldefense.com/v3/__https://lore.kernel.org/all/17fab2d6-5a74-4573-bcc3-b75951508f0a@gmail.com/__;!!ACWV5N9M2RV99hQ!Nv5xX_X4cBQR_HUDO6qfamxI0GEW1eFRCpPj29lANRashIUJDrEWAjU7vlys76FH0voShW07CjpDkzSST46Irg$
> > ---
>
> Acked-by: Harry Yoo <harry.yoo@oracle.com>
Acked-by: Suren Baghdasaryan <surenb@google.com>
>
> > v2 -> v3:
> > - Put warning back, but only warn once with pr_warn_once.
> > v1 -> v2:
> > - remove the warning completely. We will have a way in the
> > future to indicate that the mem alloc profile is inaccurate.
> > ---
> > mm/slub.c | 8 +++++---
> > 1 file changed, 5 insertions(+), 3 deletions(-)
> >
> > diff --git a/mm/slub.c b/mm/slub.c
> > index dc9e729e1d26..36d7c43a6f2a 100644
> > --- a/mm/slub.c
> > +++ b/mm/slub.c
> > @@ -2102,10 +2102,12 @@ prepare_slab_obj_exts_hook(struct kmem_cache *s, gfp_t flags, void *p)
> >
> > slab = virt_to_slab(p);
> > if (!slab_obj_exts(slab) &&
> > - WARN(alloc_slab_obj_exts(slab, s, flags, false),
> > - "%s, %s: Failed to create slab extension vector!\n",
> > - __func__, s->name))
> > + alloc_slab_obj_exts(slab, s, flags, false)) {
> > + pr_warn_once("%s, %s: Failed to create slab extension vector!\n",
> > + __func__, s->name);
> > return NULL;
> > + }
> > +
> >
> > return slab_obj_exts(slab) + obj_to_index(s, slab, p);
> > }
> > --
> > 2.47.1
> >
>
> --
> Cheers,
> Harry / Hyeonggon
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [PATCH v3] mm: slub: only warn once when allocating slab obj extensions fails
2025-05-23 16:52 [PATCH v3] mm: slub: only warn once when allocating slab obj extensions fails Usama Arif
` (3 preceding siblings ...)
2025-05-26 10:05 ` Harry Yoo
@ 2025-06-02 21:57 ` Roman Gushchin
4 siblings, 0 replies; 7+ messages in thread
From: Roman Gushchin @ 2025-06-02 21:57 UTC (permalink / raw)
To: Usama Arif
Cc: Andrew Morton, surenb, hannes, shakeel.butt, vlad.wing, linux-mm,
kent.overstreet, cl, rientjes, vbabka, harry.yoo, linux-kernel,
kernel-team
Usama Arif <usamaarif642@gmail.com> writes:
> In memory bound systems, a large number of warnings for failing this
> allocation repeatedly may mask any real issues in the system
> during memory pressure being reported in dmesg. Change this to
> warning only once.
>
> Signed-off-by: Usama Arif <usamaarif642@gmail.com>
> Reported-by: Vlad Poenaru <vlad.wing@gmail.com>
> Closes: https://lore.kernel.org/all/17fab2d6-5a74-4573-bcc3-b75951508f0a@gmail.com/
Acked-by: Roman Gushchin <roman.gushchin@linux.dev>
Thanks!
^ permalink raw reply [flat|nested] 7+ messages in thread