From: "Vlastimil Babka (SUSE)" <vbabka@kernel.org>
To: Suren Baghdasaryan <surenb@google.com>
Cc: Harry Yoo <harry@kernel.org>, Hao Li <hao.li@linux.dev>,
Christoph Lameter <cl@gentwo.org>,
David Rientjes <rientjes@google.com>,
Roman Gushchin <roman.gushchin@linux.dev>,
Alexei Starovoitov <ast@kernel.org>,
Andrew Morton <akpm@linux-foundation.org>,
Johannes Weiner <hannes@cmpxchg.org>,
Michal Hocko <mhocko@kernel.org>,
Shakeel Butt <shakeel.butt@linux.dev>,
Alexander Potapenko <glider@google.com>,
Marco Elver <elver@google.com>,
Dmitry Vyukov <dvyukov@google.com>,
kasan-dev@googlegroups.com, linux-mm@kvack.org,
linux-kernel@vger.kernel.org, cgroups@vger.kernel.org
Subject: Re: [PATCH v2 09/16] mm/slab: pass alloc_flags through slab_post_alloc_hook() chain
Date: Mon, 15 Jun 2026 13:33:03 +0200 [thread overview]
Message-ID: <e079b443-76aa-4b7f-a91c-e5152bad1af5@kernel.org> (raw)
In-Reply-To: <CAJuCfpF0mcV3TUCNi981YO=uT=5p_7OOY1S6zdgwm5PMMV3w8g@mail.gmail.com>
On 6/15/26 06:35, Suren Baghdasaryan wrote:
> On Wed, Jun 10, 2026 at 8:41 AM Vlastimil Babka (SUSE)
> <vbabka@kernel.org> wrote:
>> @@ -4568,9 +4577,8 @@ struct kmem_cache *slab_pre_alloc_hook(struct kmem_cache *s, gfp_t flags)
>> }
>>
>> static __fastpath_inline
>> -bool slab_post_alloc_hook(struct kmem_cache *s, struct list_lru *lru,
>> - gfp_t flags, size_t size, void **p,
>> - unsigned int orig_size)
>> +bool slab_post_alloc_hook(struct kmem_cache *s, gfp_t flags, size_t size,
>> + void **p, struct slab_alloc_context *ac)
>
> Would if be possible to make this last parameter a ""const struct
> slab_alloc_context*" (here and in other functions accepting it)? I
> think these functions accept it as an input parameter only and are not
> supposed to change it, right? Makes it easy to veriy that
> slab_alloc_context is not changed between consequitive calls reusing
> it, for example inside slab_alloc_node().
Uh, ok, did that. Also changed orig_size to size_t.
>> {
>> bool init = slab_want_init_on_alloc(flags, s);
>> unsigned int zero_size = s->object_size;
>> @@ -4590,7 +4598,7 @@ bool slab_post_alloc_hook(struct kmem_cache *s, struct list_lru *lru,
>> * orig_size if we track it.
>> */
>> if (slub_debug_orig_size(s))
>> - zero_size = orig_size;
>> + zero_size = ac->orig_size;
>>
>> /*
>> * When slab_debug is enabled, avoid memory initialization integrated
>> @@ -4616,14 +4624,14 @@ bool slab_post_alloc_hook(struct kmem_cache *s, struct list_lru *lru,
>> !kasan_has_integrated_init())
>> && !is_kfence_address(p[i]))
>> memset(p[i], 0, zero_size);
>> - if (gfpflags_allow_spinning(flags))
>> + if (alloc_flags_allow_spinning(ac->alloc_flags))
>> kmemleak_alloc_recursive(p[i], s->object_size, 1,
>> s->flags, init_flags);
>> kmsan_slab_alloc(s, p[i], init_flags);
>> - alloc_tagging_slab_alloc_hook(s, p[i], flags);
>> + alloc_tagging_slab_alloc_hook(s, p[i], flags, ac->alloc_flags);
>> }
>>
>> - return memcg_slab_post_alloc_hook(s, lru, flags, size, p);
>> + return memcg_slab_post_alloc_hook(s, flags, size, p, ac);
>> }
>>
>> /*
>> @@ -4918,6 +4926,12 @@ static __fastpath_inline void *slab_alloc_node(struct kmem_cache *s, struct list
>> {
>> const unsigned int alloc_flags = SLAB_ALLOC_DEFAULT;
>> void *object;
>> + struct slab_alloc_context ac = {
>> + .caller_addr = addr,
>> + .orig_size = orig_size,
>> + .alloc_flags = alloc_flags,
>> + .lru = lru,
>> + };
>>
>> s = slab_pre_alloc_hook(s, gfpflags);
>> if (unlikely(!s))
>> @@ -4929,14 +4943,8 @@ static __fastpath_inline void *slab_alloc_node(struct kmem_cache *s, struct list
>>
>> object = alloc_from_pcs(s, gfpflags, alloc_flags, node);
>>
>> - if (unlikely(!object)) {
>> - struct slab_alloc_context ac = {
>> - .caller_addr = addr,
>> - .orig_size = orig_size,
>> - .alloc_flags = alloc_flags,
>> - };
>> + if (!object)
>
> Any reason "unlikely" is removed?
No, fixed, thanks!
>> object = __slab_alloc_node(s, gfpflags, node, &ac);
>> - }
>>
>> maybe_wipe_obj_freeptr(s, object);
>>
next prev parent reply other threads:[~2026-06-15 11:33 UTC|newest]
Thread overview: 74+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-06-10 15:40 [PATCH v2 00/16] mm/slab: introduce alloc_flags and slab_alloc_context Vlastimil Babka (SUSE)
2026-06-10 15:40 ` [PATCH v2 01/16] mm/slab: do not limit zeroing to orig_size when only red zoning is enabled Vlastimil Babka (SUSE)
2026-06-11 4:28 ` Harry Yoo
2026-06-12 3:47 ` Hao Li
2026-06-10 15:40 ` [PATCH v2 02/16] mm/slab: do not init any kfence objects on allocation Vlastimil Babka (SUSE)
2026-06-11 3:19 ` Harry Yoo
2026-06-11 8:34 ` Vlastimil Babka (SUSE)
2026-06-11 14:47 ` Vlastimil Babka (SUSE)
2026-06-11 15:11 ` Harry Yoo
2026-06-11 16:37 ` Vlastimil Babka (SUSE)
2026-06-15 1:28 ` Suren Baghdasaryan
2026-06-15 8:52 ` Vlastimil Babka (SUSE)
2026-06-10 15:40 ` [PATCH v2 03/16] mm/slab: stop inlining __slab_alloc_node() Vlastimil Babka (SUSE)
2026-06-12 3:48 ` Hao Li
2026-06-15 1:33 ` Suren Baghdasaryan
2026-06-10 15:40 ` [PATCH v2 04/16] mm/slab: introduce slab_alloc_context Vlastimil Babka (SUSE)
2026-06-11 4:49 ` Harry Yoo
2026-06-12 3:10 ` Hao Li
2026-06-12 9:51 ` Vlastimil Babka (SUSE)
2026-06-15 1:41 ` Suren Baghdasaryan
2026-06-10 15:40 ` [PATCH v2 05/16] mm/slab: introduce alloc_flags and SLAB_ALLOC_TRYLOCK Vlastimil Babka (SUSE)
2026-06-11 4:57 ` Harry Yoo
2026-06-11 6:40 ` Harry Yoo
2026-06-11 8:51 ` Vlastimil Babka (SUSE)
2026-06-12 3:49 ` Hao Li
2026-06-15 2:00 ` Suren Baghdasaryan
2026-06-15 2:01 ` Suren Baghdasaryan
2026-06-15 2:16 ` Alexei Starovoitov
2026-06-15 9:02 ` Vlastimil Babka (SUSE)
2026-06-10 15:40 ` [PATCH v2 06/16] mm/slab: add alloc_flags to slab_alloc_context Vlastimil Babka (SUSE)
2026-06-11 5:06 ` Harry Yoo
2026-06-12 3:50 ` Hao Li
2026-06-15 2:20 ` Suren Baghdasaryan
2026-06-10 15:40 ` [PATCH v2 07/16] mm/slab: replace struct partial_context with slab_alloc_context Vlastimil Babka (SUSE)
2026-06-11 6:05 ` Harry Yoo
2026-06-15 2:36 ` Suren Baghdasaryan
2026-06-15 10:01 ` Vlastimil Babka (SUSE)
2026-06-12 4:04 ` Hao Li
2026-06-12 9:56 ` Vlastimil Babka (SUSE)
2026-06-10 15:40 ` [PATCH v2 08/16] mm/slab: pass alloc_flags to new slab allocation Vlastimil Babka (SUSE)
2026-06-11 7:52 ` Harry Yoo
2026-06-15 10:14 ` Vlastimil Babka (SUSE)
2026-06-12 5:26 ` Hao Li
2026-06-12 9:59 ` Vlastimil Babka (SUSE)
2026-06-15 4:10 ` Suren Baghdasaryan
2026-06-10 15:40 ` [PATCH v2 09/16] mm/slab: pass alloc_flags through slab_post_alloc_hook() chain Vlastimil Babka (SUSE)
2026-06-15 4:35 ` Suren Baghdasaryan
2026-06-15 11:33 ` Vlastimil Babka (SUSE) [this message]
2026-06-10 15:40 ` [PATCH v2 10/16] mm/slab: replace slab_alloc_node() parameters with slab_alloc_context Vlastimil Babka (SUSE)
2026-06-12 5:28 ` Hao Li
2026-06-15 4:39 ` Suren Baghdasaryan
2026-06-10 15:40 ` [PATCH v2 11/16] mm/slab: allow kmem_cache_alloc_bulk() with any gfp flags Vlastimil Babka (SUSE)
2026-06-12 3:21 ` Hao Li
2026-06-12 10:05 ` Vlastimil Babka (SUSE)
2026-06-15 4:48 ` Suren Baghdasaryan
2026-06-10 15:40 ` [PATCH v2 12/16] mm/slab: pass slab_alloc_context to __do_kmalloc_node() Vlastimil Babka (SUSE)
2026-06-12 5:34 ` Hao Li
2026-06-15 4:58 ` Suren Baghdasaryan
2026-06-15 11:08 ` Vlastimil Babka (SUSE)
2026-06-10 15:40 ` [PATCH v2 13/16] mm/slab: allow __GFP_NOMEMALLOC and __GFP_NOWARN for kmalloc_nolock() Vlastimil Babka (SUSE)
2026-06-12 6:57 ` Hao Li
2026-06-15 5:06 ` Suren Baghdasaryan
2026-06-10 15:40 ` [PATCH v2 14/16] mm/slab: introduce kmalloc_flags() Vlastimil Babka (SUSE)
2026-06-12 8:02 ` Hao Li
2026-06-15 5:14 ` Suren Baghdasaryan
2026-06-10 15:40 ` [PATCH v2 15/16] mm/slab: remove __GFP_NO_OBJ_EXT usage from alloc_slab_obj_exts() Vlastimil Babka (SUSE)
2026-06-11 16:28 ` Vlastimil Babka (SUSE)
2026-06-12 6:54 ` Hao Li
2026-06-12 10:17 ` Vlastimil Babka (SUSE)
2026-06-12 11:29 ` Hao Li
2026-06-15 5:38 ` Suren Baghdasaryan
2026-06-15 11:11 ` Vlastimil Babka (SUSE)
2026-06-10 15:40 ` [PATCH v2 16/16] mm/slab: replace __GFP_NO_OBJ_EXT with SLAB_ALLOC_NO_RECURSE for sheaves Vlastimil Babka (SUSE)
2026-06-12 8:16 ` Hao Li
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=e079b443-76aa-4b7f-a91c-e5152bad1af5@kernel.org \
--to=vbabka@kernel.org \
--cc=akpm@linux-foundation.org \
--cc=ast@kernel.org \
--cc=cgroups@vger.kernel.org \
--cc=cl@gentwo.org \
--cc=dvyukov@google.com \
--cc=elver@google.com \
--cc=glider@google.com \
--cc=hannes@cmpxchg.org \
--cc=hao.li@linux.dev \
--cc=harry@kernel.org \
--cc=kasan-dev@googlegroups.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@kernel.org \
--cc=rientjes@google.com \
--cc=roman.gushchin@linux.dev \
--cc=shakeel.butt@linux.dev \
--cc=surenb@google.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox