From: "Harry Yoo (Oracle)" <harry@kernel•org>
To: Thomas Gleixner <tglx@kernel•org>
Cc: LKML <linux-kernel@vger•kernel.org>,
Vlastimil Babka <vbabka@kernel•org>,
linux-mm@kvack•org, Arnd Bergmann <arnd@arndb•de>,
x86@kernel•org, Lu Baolu <baolu.lu@linux•intel.com>,
iommu@lists•linux.dev,
Michael Grzeschik <m.grzeschik@pengutronix•de>,
netdev@vger•kernel.org, linux-wireless@vger•kernel.org,
Herbert Xu <herbert@gondor•apana.org.au>,
linux-crypto@vger•kernel.org,
David Woodhouse <dwmw2@infradead•org>,
Bernie Thompson <bernie@plugable•com>,
linux-fbdev@vger•kernel.org, Theodore Tso <tytso@mit•edu>,
linux-ext4@vger•kernel.org,
Andrew Morton <akpm@linux-foundation•org>,
Uladzislau Rezki <urezki@gmail•com>,
Marco Elver <elver@google•com>,
Dmitry Vyukov <dvyukov@google•com>,
kasan-dev@googlegroups•com,
Andrey Ryabinin <ryabinin.a.a@gmail•com>,
Thomas Sailer <t.sailer@alumni•ethz.ch>,
linux-hams@vger•kernel.org,
"Jason A. Donenfeld" <Jason@zx2c4•com>,
Richard Henderson <richard.henderson@linaro•org>,
linux-alpha@vger•kernel.org, Russell King <linux@armlinux•org.uk>,
linux-arm-kernel@lists•infradead.org,
Catalin Marinas <catalin.marinas@arm•com>,
Huacai Chen <chenhuacai@kernel•org>,
loongarch@lists•linux.dev,
Geert Uytterhoeven <geert@linux-m68k•org>,
linux-m68k@lists•linux-m68k.org,
Dinh Nguyen <dinguyen@kernel•org>,
Jonas Bonn <jonas@southpole•se>,
linux-openrisc@vger•kernel.org, Helge Deller <deller@gmx•de>,
linux-parisc@vger•kernel.org,
Michael Ellerman <mpe@ellerman•id.au>,
linuxppc-dev@lists•ozlabs.org, Paul Walmsley <pjw@kernel•org>,
linux-riscv@lists•infradead.org,
Heiko Carstens <hca@linux•ibm.com>,
linux-s390@vger•kernel.org,
"David S. Miller" <davem@davemloft•net>,
sparclinux@vger•kernel.org, Hao Li <hao.li@linux•dev>,
Christoph Lameter <cl@gentwo•org>,
David Rientjes <rientjes@google•com>,
Roman Gushchin <roman.gushchin@linux•dev>,
Shengming Hu <hu.shengming@zte•com.cn>
Subject: Re: [patch 14/38] slub: Use prandom instead of get_cycles()
Date: Mon, 13 Apr 2026 18:07:01 +0900 [thread overview]
Message-ID: <adyyNeVTkXQlnh_2@hyeyoo> (raw)
In-Reply-To: <20260410120318.525653921@kernel.org>
[Resending after fixing broken email headers]
On Fri, Apr 10, 2026 at 02:19:37PM +0200, Thomas Gleixner wrote:
> The decision whether to scan remote nodes is based on a 'random' number
> retrieved via get_cycles(). get_cycles() is about to be removed.
>
> There is already prandom state in the code, so use that instead.
>
> Signed-off-by: Thomas Gleixner <tglx@kernel•org>
> Cc: Vlastimil Babka <vbabka@kernel•org>
> Cc: linux-mm@kvack•org
> ---
Acked-by: Harry Yoo (Oracle) <harry@kernel•org>
Is this for this merge window?
This may conflict with upcoming changes on freelist shuffling [1]
(not queued for slab/for-next yet though), but it should be easy to
resolve.
[Cc'ing Shengming and SLAB ALLOCATOR folks]
[1] https://lore.kernel.org/linux-mm/20260409204352095kKWVYKtZImN59ybO6iRNj@zte.com.cn
--
Cheers,
Harry / Hyeonggon
> mm/slub.c | 37 +++++++++++++++++++++++--------------
> 1 file changed, 23 insertions(+), 14 deletions(-)
>
> --- a/mm/slub.c
> +++ b/mm/slub.c
> @@ -3302,6 +3302,25 @@ static inline struct slab *alloc_slab_pa
> return slab;
> }
>
> +#if defined(CONFIG_SLAB_FREELIST_RANDOM) || defined(CONFIG_NUMA)
> +static DEFINE_PER_CPU(struct rnd_state, slab_rnd_state);
> +
> +static unsigned int slab_get_prandom_state(unsigned int limit)
> +{
> + struct rnd_state *state;
> + unsigned int res;
> +
> + /*
> + * An interrupt or NMI handler might interrupt and change
> + * the state in the middle, but that's safe.
> + */
> + state = &get_cpu_var(slab_rnd_state);
> + res = prandom_u32_state(state) % limit;
> + put_cpu_var(slab_rnd_state);
> + return res;
> +}
> +#endif
> +
> #ifdef CONFIG_SLAB_FREELIST_RANDOM
> /* Pre-initialize the random sequence cache */
> static int init_cache_random_seq(struct kmem_cache *s)
> @@ -3365,8 +3384,6 @@ static void *next_freelist_entry(struct
> return (char *)start + idx;
> }
>
> -static DEFINE_PER_CPU(struct rnd_state, slab_rnd_state);
> -
> /* Shuffle the single linked freelist based on a random pre-computed sequence */
> static bool shuffle_freelist(struct kmem_cache *s, struct slab *slab,
> bool allow_spin)
> @@ -3383,15 +3400,7 @@ static bool shuffle_freelist(struct kmem
> if (allow_spin) {
> pos = get_random_u32_below(freelist_count);
> } else {
> - struct rnd_state *state;
> -
> - /*
> - * An interrupt or NMI handler might interrupt and change
> - * the state in the middle, but that's safe.
> - */
> - state = &get_cpu_var(slab_rnd_state);
> - pos = prandom_u32_state(state) % freelist_count;
> - put_cpu_var(slab_rnd_state);
> + pos = slab_get_prandom_state(freelist_count);
> }
>
> page_limit = slab->objects * s->size;
> @@ -3882,7 +3891,7 @@ static void *get_from_any_partial(struct
> * with available objects.
> */
> if (!s->remote_node_defrag_ratio ||
> - get_cycles() % 1024 > s->remote_node_defrag_ratio)
> + slab_get_prandom_state(1024) > s->remote_node_defrag_ratio)
> return NULL;
>
> do {
> @@ -7102,7 +7111,7 @@ static unsigned int
>
> /* see get_from_any_partial() for the defrag ratio description */
> if (!s->remote_node_defrag_ratio ||
> - get_cycles() % 1024 > s->remote_node_defrag_ratio)
> + slab_get_prandom_state(1024) > s->remote_node_defrag_ratio)
> return 0;
>
> do {
> @@ -8421,7 +8430,7 @@ void __init kmem_cache_init_late(void)
> flushwq = alloc_workqueue("slub_flushwq", WQ_MEM_RECLAIM | WQ_PERCPU,
> 0);
> WARN_ON(!flushwq);
> -#ifdef CONFIG_SLAB_FREELIST_RANDOM
> +#if defined(CONFIG_SLAB_FREELIST_RANDOM) || defined(CONFIG_NUMA)
> prandom_init_once(&slab_rnd_state);
> #endif
> }
>
>
next prev parent reply other threads:[~2026-04-13 9:07 UTC|newest]
Thread overview: 73+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-04-10 12:18 [patch 00/38] treewide: Cleanup LATCH, CLOCK_TICK_RATE and get_cycles() [ab]use Thomas Gleixner
2026-04-10 12:18 ` [patch 01/38] percpu: Sanitize __percpu_qual include hell Thomas Gleixner
2026-04-10 12:18 ` [patch 02/38] x86: Cleanup include recursion hell Thomas Gleixner
2026-04-10 20:55 ` [patch V1.1 " Thomas Gleixner
2026-04-10 12:18 ` [patch 03/38] x86/apm: Remove last LATCH usage Thomas Gleixner
2026-04-10 12:18 ` [patch 04/38] x86: Use PIT_TICK_RATE instead of CLOCK_TICK_RATE Thomas Gleixner
2026-04-10 12:18 ` [patch 05/38] treewide: Remove CLOCK_TICK_RATE Thomas Gleixner
2026-04-15 6:40 ` Christophe Leroy (CS GROUP)
2026-04-16 11:22 ` Geert Uytterhoeven
2026-04-10 12:18 ` [patch 06/38] calibrate: Rework delay timer calibration Thomas Gleixner
2026-04-10 12:19 ` [patch 07/38] treewide: Consolidate cycles_t Thomas Gleixner
2026-04-13 9:15 ` Ojaswin Mujoo
2026-04-15 6:43 ` Christophe Leroy (CS GROUP)
2026-04-16 19:32 ` Thomas Gleixner
2026-04-16 11:22 ` Geert Uytterhoeven
2026-04-10 12:19 ` [patch 08/38] x86/tsc: Use rdtsc() instead of get_cycles() Thomas Gleixner
2026-04-10 12:19 ` [patch 09/38] iommu/vt-d: Use sched_clock() " Thomas Gleixner
2026-04-10 13:45 ` Baolu Lu
2026-04-10 15:14 ` Thomas Gleixner
2026-04-10 12:19 ` [patch 10/38] arcnet: Remove function timing code Thomas Gleixner
2026-04-13 15:29 ` David Woodhouse
2026-04-10 12:19 ` [patch 11/38] misc: sgi-gru: Remove get_cycles() [ab]use Thomas Gleixner
2026-04-10 20:56 ` [patch V1.1 " Thomas Gleixner
2026-04-10 12:19 ` [patch 12/38] wifi: wil6210: Replace get_cyles() usage Thomas Gleixner
2026-04-10 12:19 ` [patch 13/38] crypto: tcrypt: Replace get_cycles() with ktime_get() Thomas Gleixner
2026-04-10 12:19 ` [patch 14/38] slub: Use prandom instead of get_cycles() Thomas Gleixner
2026-04-13 9:00 ` Vlastimil Babka (SUSE)
2026-04-13 9:07 ` Harry Yoo (Oracle) [this message]
2026-04-13 13:02 ` hu.shengming
2026-04-13 13:45 ` Vlastimil Babka (SUSE)
2026-04-10 12:19 ` [patch 15/38] ptp: ptp_vmclock: Replace get_cycles() usage Thomas Gleixner
2026-04-13 15:33 ` David Woodhouse
2026-04-13 19:30 ` Arnd Bergmann
2026-04-10 12:19 ` [patch 16/38] fbdev: udlfb: Replace get_cycles() with ktime_get() Thomas Gleixner
2026-04-10 12:19 ` [patch 17/38] ext4: Replace get_cycles() usage " Thomas Gleixner
2026-04-13 14:46 ` Arnd Bergmann
2026-04-10 12:19 ` [patch 18/38] lib/tests: Replace get_cycles() " Thomas Gleixner
2026-04-16 10:24 ` Geert Uytterhoeven
2026-04-10 12:20 ` [patch 19/38] kcsan: Replace get_cycles() usage Thomas Gleixner
2026-04-10 13:39 ` Marco Elver
2026-04-10 12:20 ` [patch 20/38] kasan: sw_tags: Replace get_cycles() by random_get_entropy() Thomas Gleixner
2026-04-10 12:20 ` [patch 21/38] hamradio: baycom_epp: Remove BAYCOM_DEBUG Thomas Gleixner
2026-04-10 12:20 ` [patch 22/38] random: Provide CONFIG_ARCH_HAS_RANDOM_ENTROPY Thomas Gleixner
2026-04-10 12:20 ` [patch 23/38] alpha: Select ARCH_HAS_RANDOM_ENTROPY Thomas Gleixner
2026-04-12 13:22 ` Magnus Lindholm
2026-04-10 12:20 ` [patch 24/38] ARM: " Thomas Gleixner
2026-04-10 12:20 ` [patch 25/38] arm64: " Thomas Gleixner
2026-04-10 12:20 ` [patch 26/38] loongarch: " Thomas Gleixner
2026-04-10 12:20 ` [patch 27/38] m68k: " Thomas Gleixner
2026-04-10 15:31 ` Daniel Palmer
2026-04-16 11:22 ` Geert Uytterhoeven
2026-04-10 12:20 ` [patch 28/38] mips: " Thomas Gleixner
2026-04-13 5:47 ` Maciej W. Rozycki
2026-04-10 12:20 ` [patch 29/38] nios2: " Thomas Gleixner
2026-04-10 12:20 ` [patch 30/38] openrisc: " Thomas Gleixner
2026-04-12 8:56 ` Stafford Horne
2026-04-10 12:21 ` [patch 31/38] parisc: " Thomas Gleixner
2026-04-14 12:41 ` Helge Deller
2026-04-10 12:21 ` [patch 32/38] powerpc/spufs: Use mftb() directly Thomas Gleixner
2026-04-13 14:43 ` Arnd Bergmann
2026-04-15 6:38 ` Christophe Leroy (CS GROUP)
2026-04-21 6:48 ` Mukesh Kumar Chaurasiya
2026-04-10 12:21 ` [patch 33/38] powerpc: Select ARCH_HAS_RANDOM_ENTROPY Thomas Gleixner
2026-04-15 6:47 ` Christophe Leroy (CS GROUP)
2026-04-21 11:22 ` Mukesh Kumar Chaurasiya
2026-04-10 12:21 ` [patch 34/38] riscv: " Thomas Gleixner
2026-04-10 12:21 ` [patch 35/38] s390: " Thomas Gleixner
2026-04-16 13:42 ` Heiko Carstens
2026-04-16 19:29 ` Thomas Gleixner
2026-04-10 12:21 ` [patch 36/38] sparc: Select ARCH_HAS_RANDOM_ENTROPY for SPARC64 Thomas Gleixner
2026-04-10 12:21 ` [patch 37/38] x86: Select ARCH_HAS_RANDOM_ENTROPY Thomas Gleixner
2026-04-10 12:21 ` [patch 38/38] treewide: Remove asm/timex.h includes from generic code Thomas Gleixner
2026-04-13 14:45 ` Arnd Bergmann
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=adyyNeVTkXQlnh_2@hyeyoo \
--to=harry@kernel$(echo .)org \
--cc=Jason@zx2c4$(echo .)com \
--cc=akpm@linux-foundation$(echo .)org \
--cc=arnd@arndb$(echo .)de \
--cc=baolu.lu@linux$(echo .)intel.com \
--cc=bernie@plugable$(echo .)com \
--cc=catalin.marinas@arm$(echo .)com \
--cc=chenhuacai@kernel$(echo .)org \
--cc=cl@gentwo$(echo .)org \
--cc=davem@davemloft$(echo .)net \
--cc=deller@gmx$(echo .)de \
--cc=dinguyen@kernel$(echo .)org \
--cc=dvyukov@google$(echo .)com \
--cc=dwmw2@infradead$(echo .)org \
--cc=elver@google$(echo .)com \
--cc=geert@linux-m68k$(echo .)org \
--cc=hao.li@linux$(echo .)dev \
--cc=hca@linux$(echo .)ibm.com \
--cc=herbert@gondor$(echo .)apana.org.au \
--cc=hu.shengming@zte$(echo .)com.cn \
--cc=iommu@lists$(echo .)linux.dev \
--cc=jonas@southpole$(echo .)se \
--cc=kasan-dev@googlegroups$(echo .)com \
--cc=linux-alpha@vger$(echo .)kernel.org \
--cc=linux-arm-kernel@lists$(echo .)infradead.org \
--cc=linux-crypto@vger$(echo .)kernel.org \
--cc=linux-ext4@vger$(echo .)kernel.org \
--cc=linux-fbdev@vger$(echo .)kernel.org \
--cc=linux-hams@vger$(echo .)kernel.org \
--cc=linux-kernel@vger$(echo .)kernel.org \
--cc=linux-m68k@lists$(echo .)linux-m68k.org \
--cc=linux-mm@kvack$(echo .)org \
--cc=linux-openrisc@vger$(echo .)kernel.org \
--cc=linux-parisc@vger$(echo .)kernel.org \
--cc=linux-riscv@lists$(echo .)infradead.org \
--cc=linux-s390@vger$(echo .)kernel.org \
--cc=linux-wireless@vger$(echo .)kernel.org \
--cc=linux@armlinux$(echo .)org.uk \
--cc=linuxppc-dev@lists$(echo .)ozlabs.org \
--cc=loongarch@lists$(echo .)linux.dev \
--cc=m.grzeschik@pengutronix$(echo .)de \
--cc=mpe@ellerman$(echo .)id.au \
--cc=netdev@vger$(echo .)kernel.org \
--cc=pjw@kernel$(echo .)org \
--cc=richard.henderson@linaro$(echo .)org \
--cc=rientjes@google$(echo .)com \
--cc=roman.gushchin@linux$(echo .)dev \
--cc=ryabinin.a.a@gmail$(echo .)com \
--cc=sparclinux@vger$(echo .)kernel.org \
--cc=t.sailer@alumni$(echo .)ethz.ch \
--cc=tglx@kernel$(echo .)org \
--cc=tytso@mit$(echo .)edu \
--cc=urezki@gmail$(echo .)com \
--cc=vbabka@kernel$(echo .)org \
--cc=x86@kernel$(echo .)org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
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