From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id F16FAD4A603 for ; Fri, 16 Jan 2026 05:45:41 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id 10D306B0088; Fri, 16 Jan 2026 00:45:41 -0500 (EST) Received: by kanga.kvack.org (Postfix, from userid 40) id 0BBA26B0089; Fri, 16 Jan 2026 00:45:41 -0500 (EST) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id ED5216B008A; Fri, 16 Jan 2026 00:45:40 -0500 (EST) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (smtprelay0015.hostedemail.com [216.40.44.15]) by kanga.kvack.org (Postfix) with ESMTP id DDB1C6B0088 for ; Fri, 16 Jan 2026 00:45:40 -0500 (EST) Received: from smtpin11.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay08.hostedemail.com (Postfix) with ESMTP id ECEEA140562 for ; Fri, 16 Jan 2026 05:45:39 +0000 (UTC) X-FDA: 84336739998.11.38F093D Received: from mail-qt1-f181.google.com (mail-qt1-f181.google.com [209.85.160.181]) by imf05.hostedemail.com (Postfix) with ESMTP id 17F34100005 for ; Fri, 16 Jan 2026 05:45:37 +0000 (UTC) Authentication-Results: imf05.hostedemail.com; dkim=pass header.d=google.com header.s=20230601 header.b=grarBrjT; arc=pass ("google.com:s=arc-20240605:i=1"); spf=pass (imf05.hostedemail.com: domain of surenb@google.com designates 209.85.160.181 as permitted sender) smtp.mailfrom=surenb@google.com; dmarc=pass (policy=reject) header.from=google.com ARC-Seal: i=2; s=arc-20220608; d=hostedemail.com; t=1768542338; a=rsa-sha256; cv=pass; b=OjVatJEUgZossTLTsOFkUKL2LKGOp4/mFmCf9DmHpQIrMU8imokO6mrNANnftPjtEvr43B 5swQpxnMm+nwIbCDUXbttH9tYSfcV6VIyzcCpbX0vFYjfTtqmnU9/EuwxGTvU7UugAuE/g 5DvwD44n8i28qgk16VyLBs1lqakRCD4= ARC-Authentication-Results: i=2; imf05.hostedemail.com; dkim=pass header.d=google.com header.s=20230601 header.b=grarBrjT; arc=pass ("google.com:s=arc-20240605:i=1"); spf=pass (imf05.hostedemail.com: domain of surenb@google.com designates 209.85.160.181 as permitted sender) smtp.mailfrom=surenb@google.com; dmarc=pass (policy=reject) header.from=google.com ARC-Message-Signature: i=2; a=rsa-sha256; c=relaxed/relaxed; d=hostedemail.com; s=arc-20220608; t=1768542338; h=from:from:sender:reply-to:subject:subject:date:date: message-id:message-id:to:to:cc:cc:mime-version:mime-version: content-type:content-type: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references:dkim-signature; bh=oD4lKTyPruUjM+2ZGAjFWRV4Lq8q1KPCB1ITiVRP8FE=; b=fjfiatshf17TheodwenVi20xiDlAMnuEdNBDRhhjndGp4H7OrHIw+Spy1rv/+rYlyyjSmD Xko4mP0zrX32l3tESQH5nuMxcC784o65IUGq8rOiHhmfdDUB8w/w0nM8AyYxQHgOyDfv/Y D/J1e2Wri9KalJDxGXtk7MaBJ8Xai58= Received: by mail-qt1-f181.google.com with SMTP id d75a77b69052e-5014acad6f2so227881cf.1 for ; Thu, 15 Jan 2026 21:45:37 -0800 (PST) ARC-Seal: i=1; a=rsa-sha256; t=1768542337; cv=none; d=google.com; s=arc-20240605; b=M94N3yhF3xeEto2z5b202b0KqC1fMGJw5Z7YMX8OSd6KPsXEr/+FaVRabtnP/H9sWW sKJScr2mvbSCthNIL99kfSTt1rUFrELD/Vxh+Ot9jITDk0oT1FHGmYczT7L7ify+/8zu V7s6SUN4eu5WY7QgEaamsp4jHHAfZ4mp6/lGQkwvN2oEF19qCbDjNA8TuZ0Ccmv05MKM KCRE39YbCHjOmDFsPKq42UJ4BaryngXKlb8uT+/u5A3TvVGOJ4TbuRb8cyKSc8EiSg7/ F4GlDPcuZQo4qx488w0BCgwkyBET71tgv8iLtl8bupo3+c2ggH4XhhJsuYXGFp1H6svh LK9A== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20240605; h=content-transfer-encoding:cc:to:subject:message-id:date:from :in-reply-to:references:mime-version:dkim-signature; bh=oD4lKTyPruUjM+2ZGAjFWRV4Lq8q1KPCB1ITiVRP8FE=; fh=qQYi905PnJBNgpyxLc8gl7/GUgIvqnzmGMSDgiPjiig=; b=cLIZlBfGa97ec1yL1mQHKPLmOg1FfTpAlK/gsWgYoeaB+H6P4PEYfZKbvkCj/xma7K 5/BkRxHssgv1+OZSAqR8/GfTvIzDB6TZH0ww0Umv1oMVLmUIyKZdHMmrUYfItQqJNYI8 kGpNpl6s63W40SeAW6bNr2kBYYhtSoPlWh7eJ4uXygT9a3qqOX4U8B8G91OlDzBJ/SEe Ra2GC2b77heyDY0uNfU111SHXeq9ruw9mW/WodZRtsANN8ZP0HVg6YAzcdi8U9YZ+17Q 4zXmakSUgtPXZsIYmofDLMig/hAchAnKxsg3EuEVyT/YoB632qK5XsnQZTkKFDUp8GXM GnOQ==; darn=kvack.org ARC-Authentication-Results: i=1; mx.google.com; arc=none DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1768542337; x=1769147137; darn=kvack.org; h=content-transfer-encoding:cc:to:subject:message-id:date:from :in-reply-to:references:mime-version:from:to:cc:subject:date :message-id:reply-to; bh=oD4lKTyPruUjM+2ZGAjFWRV4Lq8q1KPCB1ITiVRP8FE=; b=grarBrjTV2aCDWc7abR1EytN+gECdSIjMjzFG/OytCmsrRaVG6CimNVTWGN7eGywR4 pQPIyB+Jkkc3+j7ADopgY/uC1aJ+SRzqBZAy89F6dE7Hz2biPQBKJhsfoXQ76ofwPnH4 S36fVWB1VswONwIQ9AA08Ve4qVrJIQPk4JYgMaqD3DlmYtF1kttflD1y0vOJMm48WZZ7 bUicNUl2wc51hOhHrX10dAqA+V/EcSnBiD3RmzS3mIt/UwH6YCDZFAQNzfy9+ieoApb/ 8bFoe8ZIN3hod86SgqsCbJNcm0ExZWgW3nYHFYW+dZDNJ4rnmxJJKMkFe3pTyulR8zAg vHug== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1768542337; x=1769147137; h=content-transfer-encoding:cc:to:subject:message-id:date:from :in-reply-to:references:mime-version:x-gm-gg:x-gm-message-state:from :to:cc:subject:date:message-id:reply-to; bh=oD4lKTyPruUjM+2ZGAjFWRV4Lq8q1KPCB1ITiVRP8FE=; b=OeScUx+RuxazF851WEanVO7lQqSzRsFKlFrBIl9aXznS9DGSd9kml4CweyfELZRTmb kkZxG+EYHzmNFwkIGQ0GaeyRZdbgJXqNtVoTnb6jSa3Me7WkePxGDAk405MRX3bE2oDx aFcQiFqid8RkNoNUjXoyOBuIbXPRejPIG894CZiurcIACRWioFm+v0igV0Wu7CEyl0BK isnLxK5jbWh5u7l5NAgzl2L/qj6HfuBx/Nw/dG2j4Yvu8w/EJbrX1DB/xI+j8nlhtYEu TEIDLyi90MrGGOHn9KiJoFvmO62n9dngKhpQ59N1f5KmEYn+z7dkYdBBkmqrG/c7Ol6A Ee5A== X-Forwarded-Encrypted: i=1; AJvYcCX1sW0ReRhNw4UhMr2vfWBdhWMjJcC9t8C1UINFFMyCtsf41ED3Hl9wNCW72rRvY39EnOBNdvQvcw==@kvack.org X-Gm-Message-State: AOJu0YyUoxqe3CcyVNxVXD+zR/s9zEaQtrdDedH+bLspAONLj5+4vUb2 UXM+zPtIBqUeDu2WtPp4Di/IsXpJgggFPmOxlDuOuvdxjs0SadeO8vvURB2XQtT0b+grsGlHYHq /bZ2VKSsiEcBXPdtRD07SaZL+xCXC6nt2KpjLk863 X-Gm-Gg: AY/fxX5ga7/NatRIdF2CaOK7mYvgTHTfCit4kDOJDujkO/aW5lR7uuTlBwiqLutdh8Y tEJ1XcSmdVCiULVDyX6icdywpkiglYZUIp94L0K012JeQVZwi2/cGwN0p46r2ipUnbswMIW7Whd 4L6HyCfLnBv/sXm6TFO9JoIAYDH9qDMI3KChi3sAJ3Bzixyn6wWYooKissnJEiHBoMLDIscxkq5 iMNkqBZKNzjd8KJsdf3QqPfBrUWuH38k6ZKOM0ZhHnZU9DHyalKPUTkWxtyc1bLWWFZymu50dj0 +oPPumEatloJcQUmoIQrTbfLLARlX6J8Ig== X-Received: by 2002:ac8:5a93:0:b0:4ed:8103:8c37 with SMTP id d75a77b69052e-502a23ba0bemr6815851cf.12.1768542336851; Thu, 15 Jan 2026 21:45:36 -0800 (PST) MIME-Version: 1.0 References: <20260112-sheaves-for-all-v2-0-98225cfb50cf@suse.cz> <20260112-sheaves-for-all-v2-4-98225cfb50cf@suse.cz> In-Reply-To: <20260112-sheaves-for-all-v2-4-98225cfb50cf@suse.cz> From: Suren Baghdasaryan Date: Fri, 16 Jan 2026 05:45:26 +0000 X-Gm-Features: AZwV_Qj313U65TQKJGcZKomalyp4G4s3S9dwtw3N-deAJ8pCUFL8HFCSglYguDc Message-ID: Subject: Re: [PATCH RFC v2 04/20] slab: add sheaves to most caches To: Vlastimil Babka Cc: Harry Yoo , Petr Tesarik , Christoph Lameter , David Rientjes , Roman Gushchin , Hao Li , Andrew Morton , Uladzislau Rezki , "Liam R. Howlett" , Sebastian Andrzej Siewior , Alexei Starovoitov , linux-mm@kvack.org, linux-kernel@vger.kernel.org, linux-rt-devel@lists.linux.dev, bpf@vger.kernel.org, kasan-dev@googlegroups.com Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Rspamd-Server: rspam03 X-Rspamd-Queue-Id: 17F34100005 X-Stat-Signature: bsue5akj7rddidk6uq7f79jackuit7u4 X-Rspam-User: X-HE-Tag: 1768542337-135717 X-HE-Meta: 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 X-Bogosity: Ham, tests=bogofilter, spamicity=0.000000, version=1.2.4 Sender: owner-linux-mm@kvack.org Precedence: bulk X-Loop: owner-majordomo@kvack.org List-ID: List-Subscribe: List-Unsubscribe: On Mon, Jan 12, 2026 at 3:17=E2=80=AFPM Vlastimil Babka wr= ote: > > In the first step to replace cpu (partial) slabs with sheaves, enable > sheaves for almost all caches. Treat args->sheaf_capacity as a minimum, > and calculate sheaf capacity with a formula that roughly follows the > formula for number of objects in cpu partial slabs in set_cpu_partial(). > > This should achieve roughly similar contention on the barn spin lock as > there's currently for node list_lock without sheaves, to make > benchmarking results comparable. It can be further tuned later. > > Don't enable sheaves for bootstrap caches as that wouldn't work. In > order to recognize them by SLAB_NO_OBJ_EXT, make sure the flag exists > even for !CONFIG_SLAB_OBJ_EXT. > > This limitation will be lifted for kmalloc caches after the necessary > bootstrapping changes. > > Signed-off-by: Vlastimil Babka One nit but otherwise LGTM. Reviewed-by: Suren Baghdasaryan > --- > include/linux/slab.h | 6 ------ > mm/slub.c | 51 ++++++++++++++++++++++++++++++++++++++++++++++= +---- > 2 files changed, 47 insertions(+), 10 deletions(-) > > diff --git a/include/linux/slab.h b/include/linux/slab.h > index 2482992248dc..2682ee57ec90 100644 > --- a/include/linux/slab.h > +++ b/include/linux/slab.h > @@ -57,9 +57,7 @@ enum _slab_flag_bits { > #endif > _SLAB_OBJECT_POISON, > _SLAB_CMPXCHG_DOUBLE, > -#ifdef CONFIG_SLAB_OBJ_EXT > _SLAB_NO_OBJ_EXT, > -#endif > _SLAB_FLAGS_LAST_BIT > }; > > @@ -238,11 +236,7 @@ enum _slab_flag_bits { > #define SLAB_TEMPORARY SLAB_RECLAIM_ACCOUNT /* Objects are sh= ort-lived */ > > /* Slab created using create_boot_cache */ > -#ifdef CONFIG_SLAB_OBJ_EXT > #define SLAB_NO_OBJ_EXT __SLAB_FLAG_BIT(_SLAB_NO_OBJ_EXT) > -#else > -#define SLAB_NO_OBJ_EXT __SLAB_FLAG_UNUSED > -#endif > > /* > * ZERO_SIZE_PTR will be returned for zero sized kmalloc requests. > diff --git a/mm/slub.c b/mm/slub.c > index 8ffeb3ab3228..6e05e3cc5c49 100644 > --- a/mm/slub.c > +++ b/mm/slub.c > @@ -7857,6 +7857,48 @@ static void set_cpu_partial(struct kmem_cache *s) > #endif > } > > +static unsigned int calculate_sheaf_capacity(struct kmem_cache *s, > + struct kmem_cache_args *args= ) > + > +{ > + unsigned int capacity; > + size_t size; > + > + > + if (IS_ENABLED(CONFIG_SLUB_TINY) || s->flags & SLAB_DEBUG_FLAGS) > + return 0; > + > + /* bootstrap caches can't have sheaves for now */ > + if (s->flags & SLAB_NO_OBJ_EXT) > + return 0; > + > + /* > + * For now we use roughly similar formula (divided by two as ther= e are > + * two percpu sheaves) as what was used for percpu partial slabs,= which > + * should result in similar lock contention (barn or list_lock) > + */ > + if (s->size >=3D PAGE_SIZE) > + capacity =3D 4; > + else if (s->size >=3D 1024) > + capacity =3D 12; > + else if (s->size >=3D 256) > + capacity =3D 26; > + else > + capacity =3D 60; > + > + /* Increment capacity to make sheaf exactly a kmalloc size bucket= */ > + size =3D struct_size_t(struct slab_sheaf, objects, capacity); > + size =3D kmalloc_size_roundup(size); > + capacity =3D (size - struct_size_t(struct slab_sheaf, objects, 0)= ) / sizeof(void *); > + > + /* > + * Respect an explicit request for capacity that's typically moti= vated by > + * expected maximum size of kmem_cache_prefill_sheaf() to not end= up > + * using low-performance oversize sheaves > + */ > + return max(capacity, args->sheaf_capacity); > +} > + > /* > * calculate_sizes() determines the order and the distribution of data w= ithin > * a slab object. > @@ -7991,6 +8033,10 @@ static int calculate_sizes(struct kmem_cache_args = *args, struct kmem_cache *s) > if (s->flags & SLAB_RECLAIM_ACCOUNT) > s->allocflags |=3D __GFP_RECLAIMABLE; > > + /* kmalloc caches need extra care to support sheaves */ > + if (!is_kmalloc_cache(s)) nit: All the checks for the cases when sheaves should not be used (like SLAB_DEBUG_FLAGS and SLAB_NO_OBJ_EXT) are done inside calculate_sheaf_capacity(). Only this is_kmalloc_cache() one is here. It would be nice to have all of them in the same place but maybe you have a reason for keeping it here? > + s->sheaf_capacity =3D calculate_sheaf_capacity(s, args); > + > /* > * Determine the number of objects per slab > */ > @@ -8595,15 +8641,12 @@ int do_kmem_cache_create(struct kmem_cache *s, co= nst char *name, > > set_cpu_partial(s); > > - if (args->sheaf_capacity && !IS_ENABLED(CONFIG_SLUB_TINY) > - && !(s->flags & SLAB_DEBUG_FLAGS)= ) { > + if (s->sheaf_capacity) { > s->cpu_sheaves =3D alloc_percpu(struct slub_percpu_sheave= s); > if (!s->cpu_sheaves) { > err =3D -ENOMEM; > goto out; > } > - // TODO: increase capacity to grow slab_sheaf up to next = kmalloc size? > - s->sheaf_capacity =3D args->sheaf_capacity; > } > > #ifdef CONFIG_NUMA > > -- > 2.52.0 >