From: Hyeonggon Yoo <42.hyeyoo@gmail.com>
To: Vlastimil Babka <vbabka@suse.cz>
Cc: Christoph Lameter <cl@linux.com>,
David Rientjes <rientjes@google.com>,
Joonsoo Kim <iamjoonsoo.kim@lge.com>,
Pekka Enberg <penberg@kernel.org>,
Roman Gushchin <roman.gushchin@linux.dev>,
Andrew Morton <akpm@linux-foundation.org>,
Linus Torvalds <torvalds@linux-foundation.org>,
Matthew Wilcox <willy@infradead.org>,
patches@lists.linux.dev, linux-mm@kvack.org,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH 11/12] mm, slub: don't aggressively inline with CONFIG_SLUB_TINY
Date: Mon, 28 Nov 2022 22:19:19 +0900 [thread overview]
Message-ID: <Y4S1V9MNglgserVG@hyeyoo> (raw)
In-Reply-To: <20221121171202.22080-12-vbabka@suse.cz>
On Mon, Nov 21, 2022 at 06:12:01PM +0100, Vlastimil Babka wrote:
> SLUB fastpaths use __always_inline to avoid function calls. With
> CONFIG_SLUB_TINY we would rather save the memory. Add a
> __fastpath_inline macro that's __always_inline normally but empty with
> CONFIG_SLUB_TINY.
>
> bloat-o-meter results on x86_64 mm/slub.o:
>
> add/remove: 3/1 grow/shrink: 1/8 up/down: 865/-1784 (-919)
> Function old new delta
> kmem_cache_free 20 281 +261
> slab_alloc_node.isra - 245 +245
> slab_free.constprop.isra - 231 +231
> __kmem_cache_alloc_lru.isra - 128 +128
> __kmem_cache_release 88 83 -5
> __kmem_cache_create 1446 1436 -10
> __kmem_cache_free 271 142 -129
> kmem_cache_alloc_node 330 127 -203
> kmem_cache_free_bulk.part 826 613 -213
> __kmem_cache_alloc_node 230 10 -220
> kmem_cache_alloc_lru 325 12 -313
> kmem_cache_alloc 325 10 -315
> kmem_cache_free.part 376 - -376
> Total: Before=26103, After=25184, chg -3.52%
>
> Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
> ---
> mm/slub.c | 14 ++++++++++----
> 1 file changed, 10 insertions(+), 4 deletions(-)
>
> diff --git a/mm/slub.c b/mm/slub.c
> index 7f1cd702c3b4..d54466e76503 100644
> --- a/mm/slub.c
> +++ b/mm/slub.c
> @@ -187,6 +187,12 @@ do { \
> #define USE_LOCKLESS_FAST_PATH() (false)
> #endif
>
> +#ifndef CONFIG_SLUB_TINY
> +#define __fastpath_inline __always_inline
> +#else
> +#define __fastpath_inline
> +#endif
> +
> #ifdef CONFIG_SLUB_DEBUG
> #ifdef CONFIG_SLUB_DEBUG_ON
> DEFINE_STATIC_KEY_TRUE(slub_debug_enabled);
> @@ -3386,7 +3392,7 @@ static __always_inline void maybe_wipe_obj_freeptr(struct kmem_cache *s,
> *
> * Otherwise we can simply pick the next object from the lockless free list.
> */
> -static __always_inline void *slab_alloc_node(struct kmem_cache *s, struct list_lru *lru,
> +static __fastpath_inline void *slab_alloc_node(struct kmem_cache *s, struct list_lru *lru,
> gfp_t gfpflags, int node, unsigned long addr, size_t orig_size)
> {
> void *object;
> @@ -3412,13 +3418,13 @@ static __always_inline void *slab_alloc_node(struct kmem_cache *s, struct list_l
> return object;
> }
>
> -static __always_inline void *slab_alloc(struct kmem_cache *s, struct list_lru *lru,
> +static __fastpath_inline void *slab_alloc(struct kmem_cache *s, struct list_lru *lru,
> gfp_t gfpflags, unsigned long addr, size_t orig_size)
> {
> return slab_alloc_node(s, lru, gfpflags, NUMA_NO_NODE, addr, orig_size);
> }
>
> -static __always_inline
> +static __fastpath_inline
> void *__kmem_cache_alloc_lru(struct kmem_cache *s, struct list_lru *lru,
> gfp_t gfpflags)
> {
> @@ -3733,7 +3739,7 @@ static void do_slab_free(struct kmem_cache *s,
> }
> #endif /* CONFIG_SLUB_TINY */
>
> -static __always_inline void slab_free(struct kmem_cache *s, struct slab *slab,
> +static __fastpath_inline void slab_free(struct kmem_cache *s, struct slab *slab,
> void *head, void *tail, void **p, int cnt,
> unsigned long addr)
> {
> --
> 2.38.1
Acked-by: Hyeonggon Yoo <42.hyeyoo@gmail.com>
--
Thanks,
Hyeonggon
next prev parent reply other threads:[~2022-11-28 13:19 UTC|newest]
Thread overview: 60+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-11-21 17:11 [PATCH 00/12] Introduce CONFIG_SLUB_TINY and deprecate SLOB Vlastimil Babka
2022-11-21 17:11 ` [PATCH 01/12] mm, slab: ignore hardened usercopy parameters when disabled Vlastimil Babka
2022-11-21 21:35 ` Kees Cook
2022-11-23 14:23 ` Vlastimil Babka
2022-11-24 11:16 ` Hyeonggon Yoo
2022-11-24 11:26 ` Vlastimil Babka
2022-11-24 12:33 ` Hyeonggon Yoo
2022-11-21 17:11 ` [PATCH 02/12] mm, slub: add CONFIG_SLUB_TINY Vlastimil Babka
2022-11-24 1:08 ` Roman Gushchin
2022-11-24 11:33 ` Hyeonggon Yoo
2022-11-25 7:55 ` Vlastimil Babka
2022-11-21 17:11 ` [PATCH 03/12] mm, slub: disable SYSFS support with CONFIG_SLUB_TINY Vlastimil Babka
2022-11-24 1:12 ` Roman Gushchin
2022-11-24 9:00 ` Vlastimil Babka
2022-11-21 17:11 ` [PATCH 04/12] mm, slub: retain no free slabs on partial list " Vlastimil Babka
2022-11-24 1:12 ` Roman Gushchin
2022-11-24 11:38 ` Hyeonggon Yoo
2022-11-21 17:11 ` [PATCH 05/12] mm, slub: lower the default slub_max_order " Vlastimil Babka
2022-11-24 1:16 ` Roman Gushchin
2022-11-24 11:40 ` Hyeonggon Yoo
2022-11-21 17:11 ` [PATCH 06/12] mm, slub: don't create kmalloc-rcl caches " Vlastimil Babka
2022-11-23 13:53 ` Vlastimil Babka
2022-11-24 12:06 ` Hyeonggon Yoo
2022-11-24 12:12 ` Vlastimil Babka
2022-11-24 12:55 ` Hyeonggon Yoo
2022-11-24 13:23 ` Hyeonggon Yoo
2022-11-24 14:25 ` Hyeonggon Yoo
2022-11-21 17:11 ` [PATCH 07/12] mm, slab: ignore SLAB_RECLAIM_ACCOUNT " Vlastimil Babka
2022-11-24 1:20 ` Roman Gushchin
2022-11-24 9:09 ` Vlastimil Babka
2022-11-24 9:21 ` Christoph Lameter
2022-11-27 23:11 ` Vlastimil Babka
2022-11-21 17:11 ` [PATCH 08/12] mm, slub: refactor free debug processing Vlastimil Babka
2022-11-27 10:18 ` Hyeonggon Yoo
2022-11-21 17:11 ` [PATCH 09/12] mm, slub: split out allocations from pre/post hooks Vlastimil Babka
2022-11-27 10:54 ` Hyeonggon Yoo
2022-11-27 23:01 ` Vlastimil Babka
2022-11-28 13:06 ` Hyeonggon Yoo
2022-11-21 17:12 ` [PATCH 10/12] mm, slub: remove percpu slabs with CONFIG_SLUB_TINY Vlastimil Babka
2022-11-27 11:05 ` Hyeonggon Yoo
2022-12-12 10:54 ` Vlastimil Babka
2022-12-12 13:11 ` Dennis Zhou
2022-12-13 3:04 ` Baoquan He
2022-12-13 14:02 ` Hyeonggon Yoo
2022-12-18 10:16 ` Hyeonggon Yoo
2022-11-21 17:12 ` [PATCH 11/12] mm, slub: don't aggressively inline " Vlastimil Babka
2022-11-28 13:19 ` Hyeonggon Yoo [this message]
2022-11-21 17:12 ` [PATCH 12/12] mm, slob: rename CONFIG_SLOB to CONFIG_SLOB_DEPRECATED Vlastimil Babka
2022-11-21 18:41 ` Aaro Koskinen
2022-11-21 19:42 ` Vlastimil Babka
2022-11-22 6:47 ` Damien Le Moal
2022-11-22 16:08 ` Arnd Bergmann
2022-11-24 1:21 ` Roman Gushchin
2022-12-02 17:59 ` Palmer Dabbelt
2022-12-05 12:25 ` Damien Le Moal
2022-12-13 13:41 ` Hyeonggon Yoo
2022-11-22 16:33 ` [PATCH 00/12] Introduce CONFIG_SLUB_TINY and deprecate SLOB Arnd Bergmann
2022-11-22 16:59 ` Vlastimil Babka
2022-11-22 17:15 ` Arnd Bergmann
2022-11-24 20:30 ` Mike Rapoport
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=Y4S1V9MNglgserVG@hyeyoo \
--to=42.hyeyoo@gmail.com \
--cc=akpm@linux-foundation.org \
--cc=cl@linux.com \
--cc=iamjoonsoo.kim@lge.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=patches@lists.linux.dev \
--cc=penberg@kernel.org \
--cc=rientjes@google.com \
--cc=roman.gushchin@linux.dev \
--cc=torvalds@linux-foundation.org \
--cc=vbabka@suse.cz \
--cc=willy@infradead.org \
/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;
as well as URLs for NNTP newsgroup(s).