From: "Liam R. Howlett" <Liam.Howlett@oracle.com>
To: Vlastimil Babka <vbabka@suse.cz>
Cc: Harry Yoo <harry.yoo@oracle.com>,
Petr Tesarik <ptesarik@suse.com>,
Christoph Lameter <cl@gentwo.org>,
David Rientjes <rientjes@google.com>,
Roman Gushchin <roman.gushchin@linux.dev>,
Hao Li <hao.li@linux.dev>,
Andrew Morton <akpm@linux-foundation.org>,
Uladzislau Rezki <urezki@gmail.com>,
Suren Baghdasaryan <surenb@google.com>,
Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
Alexei Starovoitov <ast@kernel.org>,
linux-mm@kvack.org, linux-kernel@vger.kernel.org,
linux-rt-devel@lists.linux.dev, bpf@vger.kernel.org,
kasan-dev@googlegroups.com
Subject: Re: [PATCH v4 09/22] slab: handle kmalloc sheaves bootstrap
Date: Tue, 27 Jan 2026 13:30:52 -0500 [thread overview]
Message-ID: <tm7yjp4phbf24quv5vdjw3juhusvzk7dyassrtrejqyhbieie7@ml5okfvozh6j> (raw)
In-Reply-To: <20260123-sheaves-for-all-v4-9-041323d506f7@suse.cz>
* Vlastimil Babka <vbabka@suse.cz> [260123 01:53]:
> Enable sheaves for kmalloc caches. For other types than KMALLOC_NORMAL,
> we can simply allow them in calculate_sizes() as they are created later
> than KMALLOC_NORMAL caches and can allocate sheaves and barns from
> those.
>
> For KMALLOC_NORMAL caches we perform additional step after first
> creating them without sheaves. Then bootstrap_cache_sheaves() simply
> allocates and initializes barns and sheaves and finally sets
> s->sheaf_capacity to make them actually used.
>
> Afterwards the only caches left without sheaves (unless SLUB_TINY or
> debugging is enabled) are kmem_cache and kmem_cache_node. These are only
> used when creating or destroying other kmem_caches. Thus they are not
> performance critical and we can simply leave it that way.
>
> Reviewed-by: Harry Yoo <harry.yoo@oracle.com>
> Reviewed-by: Hao Li <hao.li@linux.dev>
> Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Reviewed-by: Liam R. Howlett <Liam.Howlett@oracle.com>
> ---
> mm/slub.c | 88 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++---
> 1 file changed, 84 insertions(+), 4 deletions(-)
>
> diff --git a/mm/slub.c b/mm/slub.c
> index 4ca6bd944854..22acc249f9c0 100644
> --- a/mm/slub.c
> +++ b/mm/slub.c
> @@ -2593,7 +2593,8 @@ static void *setup_object(struct kmem_cache *s, void *object)
> return object;
> }
>
> -static struct slab_sheaf *alloc_empty_sheaf(struct kmem_cache *s, gfp_t gfp)
> +static struct slab_sheaf *__alloc_empty_sheaf(struct kmem_cache *s, gfp_t gfp,
> + unsigned int capacity)
> {
> struct slab_sheaf *sheaf;
> size_t sheaf_size;
> @@ -2611,7 +2612,7 @@ static struct slab_sheaf *alloc_empty_sheaf(struct kmem_cache *s, gfp_t gfp)
> if (s->flags & SLAB_KMALLOC)
> gfp |= __GFP_NO_OBJ_EXT;
>
> - sheaf_size = struct_size(sheaf, objects, s->sheaf_capacity);
> + sheaf_size = struct_size(sheaf, objects, capacity);
> sheaf = kzalloc(sheaf_size, gfp);
>
> if (unlikely(!sheaf))
> @@ -2624,6 +2625,12 @@ static struct slab_sheaf *alloc_empty_sheaf(struct kmem_cache *s, gfp_t gfp)
> return sheaf;
> }
>
> +static inline struct slab_sheaf *alloc_empty_sheaf(struct kmem_cache *s,
> + gfp_t gfp)
> +{
> + return __alloc_empty_sheaf(s, gfp, s->sheaf_capacity);
> +}
> +
> static void free_empty_sheaf(struct kmem_cache *s, struct slab_sheaf *sheaf)
> {
> kfree(sheaf);
> @@ -8144,8 +8151,11 @@ static int calculate_sizes(struct kmem_cache_args *args, struct kmem_cache *s)
> if (s->flags & SLAB_RECLAIM_ACCOUNT)
> s->allocflags |= __GFP_RECLAIMABLE;
>
> - /* kmalloc caches need extra care to support sheaves */
> - if (!is_kmalloc_cache(s))
> + /*
> + * For KMALLOC_NORMAL caches we enable sheaves later by
> + * bootstrap_kmalloc_sheaves() to avoid recursion
> + */
> + if (!is_kmalloc_normal(s))
> s->sheaf_capacity = calculate_sheaf_capacity(s, args);
>
> /*
> @@ -8640,6 +8650,74 @@ static struct kmem_cache * __init bootstrap(struct kmem_cache *static_cache)
> return s;
> }
>
> +/*
> + * Finish the sheaves initialization done normally by init_percpu_sheaves() and
> + * init_kmem_cache_nodes(). For normal kmalloc caches we have to bootstrap it
> + * since sheaves and barns are allocated by kmalloc.
> + */
> +static void __init bootstrap_cache_sheaves(struct kmem_cache *s)
> +{
> + struct kmem_cache_args empty_args = {};
> + unsigned int capacity;
> + bool failed = false;
> + int node, cpu;
> +
> + capacity = calculate_sheaf_capacity(s, &empty_args);
> +
> + /* capacity can be 0 due to debugging or SLUB_TINY */
> + if (!capacity)
> + return;
> +
> + for_each_node_mask(node, slab_nodes) {
> + struct node_barn *barn;
> +
> + barn = kmalloc_node(sizeof(*barn), GFP_KERNEL, node);
> +
> + if (!barn) {
> + failed = true;
> + goto out;
> + }
> +
> + barn_init(barn);
> + get_node(s, node)->barn = barn;
> + }
> +
> + for_each_possible_cpu(cpu) {
> + struct slub_percpu_sheaves *pcs;
> +
> + pcs = per_cpu_ptr(s->cpu_sheaves, cpu);
> +
> + pcs->main = __alloc_empty_sheaf(s, GFP_KERNEL, capacity);
> +
> + if (!pcs->main) {
> + failed = true;
> + break;
> + }
> + }
> +
> +out:
> + /*
> + * It's still early in boot so treat this like same as a failure to
> + * create the kmalloc cache in the first place
> + */
> + if (failed)
> + panic("Out of memory when creating kmem_cache %s\n", s->name);
> +
> + s->sheaf_capacity = capacity;
> +}
> +
> +static void __init bootstrap_kmalloc_sheaves(void)
> +{
> + enum kmalloc_cache_type type;
> +
> + for (type = KMALLOC_NORMAL; type <= KMALLOC_RANDOM_END; type++) {
> + for (int idx = 0; idx < KMALLOC_SHIFT_HIGH + 1; idx++) {
> + if (kmalloc_caches[type][idx])
> + bootstrap_cache_sheaves(kmalloc_caches[type][idx]);
> + }
> + }
> +}
> +
> void __init kmem_cache_init(void)
> {
> static __initdata struct kmem_cache boot_kmem_cache,
> @@ -8683,6 +8761,8 @@ void __init kmem_cache_init(void)
> setup_kmalloc_cache_index_table();
> create_kmalloc_caches();
>
> + bootstrap_kmalloc_sheaves();
> +
> /* Setup random freelists for each cache */
> init_freelist_randomization();
>
>
> --
> 2.52.0
>
next prev parent reply other threads:[~2026-01-27 18:31 UTC|newest]
Thread overview: 69+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-01-23 6:52 [PATCH v4 00/22] slab: replace cpu (partial) slabs with sheaves Vlastimil Babka
2026-01-23 6:52 ` [PATCH v4 01/22] mm/slab: add rcu_barrier() to kvfree_rcu_barrier_on_cache() Vlastimil Babka
2026-01-27 16:08 ` Liam R. Howlett
2026-01-23 6:52 ` [PATCH v4 02/22] mm/slab: fix false lockdep warning in __kfree_rcu_sheaf() Vlastimil Babka
2026-01-23 12:03 ` Sebastian Andrzej Siewior
2026-01-24 10:58 ` Harry Yoo
2026-01-23 6:52 ` [PATCH v4 03/22] slab: add SLAB_CONSISTENCY_CHECKS to SLAB_NEVER_MERGE Vlastimil Babka
2026-01-23 6:52 ` [PATCH v4 04/22] mm/slab: move and refactor __kmem_cache_alias() Vlastimil Babka
2026-01-27 16:17 ` Liam R. Howlett
2026-01-27 16:59 ` Vlastimil Babka
2026-01-23 6:52 ` [PATCH v4 05/22] mm/slab: make caches with sheaves mergeable Vlastimil Babka
2026-01-27 16:23 ` Liam R. Howlett
2026-01-23 6:52 ` [PATCH v4 06/22] slab: add sheaves to most caches Vlastimil Babka
2026-01-26 6:36 ` Hao Li
2026-01-26 8:39 ` Vlastimil Babka
2026-01-26 13:59 ` Breno Leitao
2026-01-27 16:34 ` Liam R. Howlett
2026-01-27 17:01 ` Vlastimil Babka
2026-01-29 7:24 ` Zhao Liu
2026-01-29 8:21 ` Vlastimil Babka
2026-01-30 7:15 ` Zhao Liu
2026-02-04 18:01 ` Vlastimil Babka
2026-01-23 6:52 ` [PATCH v4 07/22] slab: introduce percpu sheaves bootstrap Vlastimil Babka
2026-01-26 6:13 ` Hao Li
2026-01-26 8:42 ` Vlastimil Babka
2026-01-27 17:31 ` Liam R. Howlett
2026-01-23 6:52 ` [PATCH v4 08/22] slab: make percpu sheaves compatible with kmalloc_nolock()/kfree_nolock() Vlastimil Babka
2026-01-23 18:05 ` Alexei Starovoitov
2026-01-27 17:36 ` Liam R. Howlett
2026-01-29 8:25 ` Vlastimil Babka
2026-01-23 6:52 ` [PATCH v4 09/22] slab: handle kmalloc sheaves bootstrap Vlastimil Babka
2026-01-27 18:30 ` Liam R. Howlett [this message]
2026-01-23 6:52 ` [PATCH v4 10/22] slab: add optimized sheaf refill from partial list Vlastimil Babka
2026-01-26 7:12 ` Hao Li
2026-01-29 7:43 ` Harry Yoo
2026-01-29 8:29 ` Vlastimil Babka
2026-01-27 20:05 ` Liam R. Howlett
2026-01-29 8:01 ` Harry Yoo
2026-01-23 6:52 ` [PATCH v4 11/22] slab: remove cpu (partial) slabs usage from allocation paths Vlastimil Babka
2026-01-23 18:17 ` Alexei Starovoitov
2026-01-23 6:52 ` [PATCH v4 12/22] slab: remove SLUB_CPU_PARTIAL Vlastimil Babka
2026-01-23 6:52 ` [PATCH v4 13/22] slab: remove the do_slab_free() fastpath Vlastimil Babka
2026-01-23 18:15 ` Alexei Starovoitov
2026-01-23 6:52 ` [PATCH v4 14/22] slab: remove defer_deactivate_slab() Vlastimil Babka
2026-01-23 17:31 ` Alexei Starovoitov
2026-01-23 6:52 ` [PATCH v4 15/22] slab: simplify kmalloc_nolock() Vlastimil Babka
2026-01-23 6:52 ` [PATCH v4 16/22] slab: remove struct kmem_cache_cpu Vlastimil Babka
2026-01-23 6:52 ` [PATCH v4 17/22] slab: remove unused PREEMPT_RT specific macros Vlastimil Babka
2026-01-23 6:52 ` [PATCH v4 18/22] slab: refill sheaves from all nodes Vlastimil Babka
2026-01-27 14:28 ` Mateusz Guzik
2026-01-27 22:04 ` Vlastimil Babka
2026-01-29 9:16 ` Harry Yoo
2026-01-23 6:52 ` [PATCH v4 19/22] slab: update overview comments Vlastimil Babka
2026-01-23 6:52 ` [PATCH v4 20/22] slab: remove frozen slab checks from __slab_free() Vlastimil Babka
2026-01-29 7:16 ` Harry Yoo
2026-01-23 6:52 ` [PATCH v4 21/22] mm/slub: remove DEACTIVATE_TO_* stat items Vlastimil Babka
2026-01-29 7:21 ` Harry Yoo
2026-01-23 6:53 ` [PATCH v4 22/22] mm/slub: cleanup and repurpose some " Vlastimil Babka
2026-01-29 7:40 ` Harry Yoo
2026-01-29 15:18 ` [PATCH v4 00/22] slab: replace cpu (partial) slabs with sheaves Hao Li
2026-01-29 15:28 ` Vlastimil Babka
2026-01-29 16:06 ` Hao Li
2026-01-29 16:44 ` Liam R. Howlett
2026-01-30 4:38 ` Hao Li
2026-01-30 4:50 ` Hao Li
2026-01-30 6:17 ` Hao Li
2026-02-04 18:02 ` Vlastimil Babka
2026-02-04 18:24 ` Christoph Lameter (Ampere)
2026-02-06 16:44 ` Vlastimil Babka
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=tm7yjp4phbf24quv5vdjw3juhusvzk7dyassrtrejqyhbieie7@ml5okfvozh6j \
--to=liam.howlett@oracle.com \
--cc=akpm@linux-foundation.org \
--cc=ast@kernel.org \
--cc=bigeasy@linutronix.de \
--cc=bpf@vger.kernel.org \
--cc=cl@gentwo.org \
--cc=hao.li@linux.dev \
--cc=harry.yoo@oracle.com \
--cc=kasan-dev@googlegroups.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linux-rt-devel@lists.linux.dev \
--cc=ptesarik@suse.com \
--cc=rientjes@google.com \
--cc=roman.gushchin@linux.dev \
--cc=surenb@google.com \
--cc=urezki@gmail.com \
--cc=vbabka@suse.cz \
/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