From nobody Sun Dec 28 21:19:44 2025 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 vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 0C022C4167B for ; Mon, 4 Dec 2023 19:34:58 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S232801AbjLDTet (ORCPT ); Mon, 4 Dec 2023 14:34:49 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:45772 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229509AbjLDTes (ORCPT ); Mon, 4 Dec 2023 14:34:48 -0500 Received: from smtp-out1.suse.de (smtp-out1.suse.de [IPv6:2a07:de40:b251:101:10:150:64:1]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id EFD56C1 for ; Mon, 4 Dec 2023 11:34:53 -0800 (PST) Received: from imap1.dmz-prg2.suse.org (imap1.dmz-prg2.suse.org [10.150.64.97]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (4096 bits) server-digest SHA256) (No client certificate requested) by smtp-out1.suse.de (Postfix) with ESMTPS id 8DF80220FD; Mon, 4 Dec 2023 19:34:52 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=suse.cz; s=susede2_rsa; t=1701718492; h=from:from:reply-to: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; bh=EqcJTAjZFB8lQHDKYDjIZchVKot1dBAP9whijzHZRBs=; b=3aoiT3paks0mcXiEeezN/xPw9jv7Di3tra3+7n/QDct0IcBebFNeztkLnzZNR1eA+r+5P7 5HTYlXZjd6w/d/4hPiYwAl8U4a6RDNwPAenVha70NBnhh7hjIhw5Mnrb5mEfP6x/vm7DcI RvtCEN+PAXA5BPRclQuH0Zu5ugFOvzQ= DKIM-Signature: v=1; a=ed25519-sha256; c=relaxed/relaxed; d=suse.cz; s=susede2_ed25519; t=1701718492; h=from:from:reply-to: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; bh=EqcJTAjZFB8lQHDKYDjIZchVKot1dBAP9whijzHZRBs=; b=LJvwh73CRk0ogt5DyHZYk9KKPNPbooLArhcWhOV7Z8vEi3KnNjT7llf/nkqXmRF2A2S/E1 I9N7BuX+zof3jCDw== Received: from imap1.dmz-prg2.suse.org (localhost [127.0.0.1]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (4096 bits) server-digest SHA256) (No client certificate requested) by imap1.dmz-prg2.suse.org (Postfix) with ESMTPS id 70C1D13AC1; Mon, 4 Dec 2023 19:34:52 +0000 (UTC) Received: from dovecot-director2.suse.de ([2a07:de40:b281:106:10:150:64:167]) by imap1.dmz-prg2.suse.org with ESMTPSA id KHI2G9wpbmUPMwAAD6G6ig (envelope-from ); Mon, 04 Dec 2023 19:34:52 +0000 From: Vlastimil Babka Date: Mon, 04 Dec 2023 20:34:40 +0100 Subject: [PATCH 1/4] mm/slub: fix bulk alloc and free stats MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Message-Id: <20231204-slub-cleanup-hooks-v1-1-88b65f7cd9d5@suse.cz> References: <20231204-slub-cleanup-hooks-v1-0-88b65f7cd9d5@suse.cz> In-Reply-To: <20231204-slub-cleanup-hooks-v1-0-88b65f7cd9d5@suse.cz> To: Christoph Lameter , Pekka Enberg , David Rientjes , Joonsoo Kim Cc: Andrew Morton , Roman Gushchin , Hyeonggon Yoo <42.hyeyoo@gmail.com>, Alexander Potapenko , Marco Elver , Dmitry Vyukov , linux-mm@kvack.org, linux-kernel@vger.kernel.org, kasan-dev@googlegroups.com, Vlastimil Babka X-Mailer: b4 0.12.4 Authentication-Results: smtp-out1.suse.de; none X-Spamd-Result: default: False [0.20 / 50.00]; ARC_NA(0.00)[]; RCVD_VIA_SMTP_AUTH(0.00)[]; MID_RHS_MATCH_FROM(0.00)[]; FROM_HAS_DN(0.00)[]; TO_DN_SOME(0.00)[]; FREEMAIL_ENVRCPT(0.00)[gmail.com]; TO_MATCH_ENVRCPT_ALL(0.00)[]; TAGGED_RCPT(0.00)[]; MIME_GOOD(-0.10)[text/plain]; NEURAL_HAM_LONG(-1.00)[-1.000]; BAYES_HAM(-0.00)[15.64%]; RCVD_COUNT_THREE(0.00)[3]; DKIM_SIGNED(0.00)[suse.cz:s=susede2_rsa,suse.cz:s=susede2_ed25519]; NEURAL_HAM_SHORT(-0.20)[-1.000]; RCPT_COUNT_TWELVE(0.00)[14]; DBL_BLOCKED_OPENRESOLVER(0.00)[suse.cz:email]; FUZZY_BLOCKED(0.00)[rspamd.com]; FROM_EQ_ENVFROM(0.00)[]; MIME_TRACE(0.00)[0:+]; FREEMAIL_CC(0.00)[linux-foundation.org,linux.dev,gmail.com,google.com,kvack.org,vger.kernel.org,googlegroups.com,suse.cz]; RCVD_TLS_ALL(0.00)[]; SUSPICIOUS_RECIPS(1.50)[] Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org The SLUB sysfs stats enabled CONFIG_SLUB_STATS have two deficiencies identified wrt bulk alloc/free operations: - Bulk allocations from cpu freelist are not counted. Add the ALLOC_FASTPATH counter there. - Bulk fastpath freeing will count a list of multiple objects with a single FREE_FASTPATH inc. Add a stat_add() variant to count them all. Signed-off-by: Vlastimil Babka Reviewed-by: Chengming Zhou --- mm/slub.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/mm/slub.c b/mm/slub.c index 3f8b95757106..d7b0ca6012e0 100644 --- a/mm/slub.c +++ b/mm/slub.c @@ -396,6 +396,14 @@ static inline void stat(const struct kmem_cache *s, en= um stat_item si) #endif } =20 +static inline +void stat_add(const struct kmem_cache *s, enum stat_item si, int v) +{ +#ifdef CONFIG_SLUB_STATS + raw_cpu_add(s->cpu_slab->stat[si], v); +#endif +} + /* * The slab lists for all objects. */ @@ -4268,7 +4276,7 @@ static __always_inline void do_slab_free(struct kmem_= cache *s, =20 local_unlock(&s->cpu_slab->lock); } - stat(s, FREE_FASTPATH); + stat_add(s, FREE_FASTPATH, cnt); } #else /* CONFIG_SLUB_TINY */ static void do_slab_free(struct kmem_cache *s, @@ -4545,6 +4553,7 @@ static inline int __kmem_cache_alloc_bulk(struct kmem= _cache *s, gfp_t flags, c->freelist =3D get_freepointer(s, object); p[i] =3D object; maybe_wipe_obj_freeptr(s, p[i]); + stat(s, ALLOC_FASTPATH); } c->tid =3D next_tid(c->tid); local_unlock_irqrestore(&s->cpu_slab->lock, irqflags); --=20 2.43.0 From nobody Sun Dec 28 21:19:44 2025 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 vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 29FD6C10DC1 for ; Mon, 4 Dec 2023 19:35:03 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S233277AbjLDTey (ORCPT ); Mon, 4 Dec 2023 14:34:54 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:45800 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S230317AbjLDTes (ORCPT ); Mon, 4 Dec 2023 14:34:48 -0500 Received: from smtp-out2.suse.de (smtp-out2.suse.de [IPv6:2a07:de40:b251:101:10:150:64:2]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 5DDE1E6 for ; Mon, 4 Dec 2023 11:34:54 -0800 (PST) Received: from imap1.dmz-prg2.suse.org (imap1.dmz-prg2.suse.org [10.150.64.97]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (4096 bits) server-digest SHA256) (No client certificate requested) by smtp-out2.suse.de (Postfix) with ESMTPS id A51611FE6E; Mon, 4 Dec 2023 19:34:52 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=suse.cz; s=susede2_rsa; t=1701718492; h=from:from:reply-to: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; bh=4fSahCwhPLa3fIglvL1TRmtTOqYm3xhkbDoawZHdw9A=; b=cYuegURdK6TarNJSSo68uWkNBOrhbmxZDmoXF/uS7znw/iEtrWLpnV32Dfv5L/PQtD6jMV QULfAq7/mAGtPm9Cbv6R2L7cGyXMaX4ps56QutuI/yGpFkZL7SUzA/8QS4L54o79rRjq4f NWPBKnLkVzZDYQTjYHbbwLf2ZFVPFKY= DKIM-Signature: v=1; a=ed25519-sha256; c=relaxed/relaxed; d=suse.cz; s=susede2_ed25519; t=1701718492; h=from:from:reply-to: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; bh=4fSahCwhPLa3fIglvL1TRmtTOqYm3xhkbDoawZHdw9A=; b=z8JNsCGQ0Xr3e3YcoHYBA6JiZnYEPfBf9mIKjtgLx1BoQFMrFN0F1iTKMetguIwoGikLxp evtQCq2pw+erbpCw== Received: from imap1.dmz-prg2.suse.org (localhost [127.0.0.1]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (4096 bits) server-digest SHA256) (No client certificate requested) by imap1.dmz-prg2.suse.org (Postfix) with ESMTPS id 8875313AC2; Mon, 4 Dec 2023 19:34:52 +0000 (UTC) Received: from dovecot-director2.suse.de ([2a07:de40:b281:106:10:150:64:167]) by imap1.dmz-prg2.suse.org with ESMTPSA id OCn7INwpbmUPMwAAD6G6ig (envelope-from ); Mon, 04 Dec 2023 19:34:52 +0000 From: Vlastimil Babka Date: Mon, 04 Dec 2023 20:34:41 +0100 Subject: [PATCH 2/4] mm/slub: introduce __kmem_cache_free_bulk() without free hooks MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Message-Id: <20231204-slub-cleanup-hooks-v1-2-88b65f7cd9d5@suse.cz> References: <20231204-slub-cleanup-hooks-v1-0-88b65f7cd9d5@suse.cz> In-Reply-To: <20231204-slub-cleanup-hooks-v1-0-88b65f7cd9d5@suse.cz> To: Christoph Lameter , Pekka Enberg , David Rientjes , Joonsoo Kim Cc: Andrew Morton , Roman Gushchin , Hyeonggon Yoo <42.hyeyoo@gmail.com>, Alexander Potapenko , Marco Elver , Dmitry Vyukov , linux-mm@kvack.org, linux-kernel@vger.kernel.org, kasan-dev@googlegroups.com, Vlastimil Babka X-Mailer: b4 0.12.4 Authentication-Results: smtp-out2.suse.de; none X-Spamd-Result: default: False [3.33 / 50.00]; ARC_NA(0.00)[]; RCVD_VIA_SMTP_AUTH(0.00)[]; BAYES_SPAM(2.93)[93.30%]; FROM_HAS_DN(0.00)[]; TO_DN_SOME(0.00)[]; TO_MATCH_ENVRCPT_ALL(0.00)[]; FREEMAIL_ENVRCPT(0.00)[gmail.com]; TAGGED_RCPT(0.00)[]; MIME_GOOD(-0.10)[text/plain]; NEURAL_HAM_LONG(-1.00)[-1.000]; R_RATELIMIT(0.00)[to_ip_from(RLtz7ce9b89hw8xzamye9qeynd)]; RCVD_COUNT_THREE(0.00)[3]; DKIM_SIGNED(0.00)[suse.cz:s=susede2_rsa,suse.cz:s=susede2_ed25519]; RCPT_COUNT_TWELVE(0.00)[14]; DBL_BLOCKED_OPENRESOLVER(0.00)[suse.cz:email]; FUZZY_BLOCKED(0.00)[rspamd.com]; FROM_EQ_ENVFROM(0.00)[]; MIME_TRACE(0.00)[0:+]; FREEMAIL_CC(0.00)[linux-foundation.org,linux.dev,gmail.com,google.com,kvack.org,vger.kernel.org,googlegroups.com,suse.cz]; RCVD_TLS_ALL(0.00)[]; MID_RHS_MATCH_FROM(0.00)[]; SUSPICIOUS_RECIPS(1.50)[] Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Currently, when __kmem_cache_alloc_bulk() fails, it frees back the objects that were allocated before the failure, using kmem_cache_free_bulk(). Because kmem_cache_free_bulk() calls the free hooks (KASAN etc.) and those expect objects that were processed by the post alloc hooks, slab_post_alloc_hook() is called before kmem_cache_free_bulk(). This is wasteful, although not a big concern in practice for the rare error path. But in order to efficiently handle percpu array batch refill and free in the near future, we will also need a variant of kmem_cache_free_bulk() that avoids the free hooks. So introduce it now and use it for the failure path. As a consequence, __kmem_cache_alloc_bulk() no longer needs the objcg parameter, remove it. Signed-off-by: Vlastimil Babka Reviewed-by: Chengming Zhou --- mm/slub.c | 33 ++++++++++++++++++++++++++------- 1 file changed, 26 insertions(+), 7 deletions(-) diff --git a/mm/slub.c b/mm/slub.c index d7b0ca6012e0..0742564c4538 100644 --- a/mm/slub.c +++ b/mm/slub.c @@ -4478,6 +4478,27 @@ int build_detached_freelist(struct kmem_cache *s, si= ze_t size, return same; } =20 +/* + * Internal bulk free of objects that were not initialised by the post all= oc + * hooks and thus should not be processed by the free hooks + */ +static void __kmem_cache_free_bulk(struct kmem_cache *s, size_t size, void= **p) +{ + if (!size) + return; + + do { + struct detached_freelist df; + + size =3D build_detached_freelist(s, size, p, &df); + if (!df.slab) + continue; + + do_slab_free(df.s, df.slab, df.freelist, df.tail, df.cnt, + _RET_IP_); + } while (likely(size)); +} + /* Note that interrupts must be enabled when calling this function. */ void kmem_cache_free_bulk(struct kmem_cache *s, size_t size, void **p) { @@ -4499,7 +4520,7 @@ EXPORT_SYMBOL(kmem_cache_free_bulk); =20 #ifndef CONFIG_SLUB_TINY static inline int __kmem_cache_alloc_bulk(struct kmem_cache *s, gfp_t flag= s, - size_t size, void **p, struct obj_cgroup *objcg) + size_t size, void **p) { struct kmem_cache_cpu *c; unsigned long irqflags; @@ -4563,14 +4584,13 @@ static inline int __kmem_cache_alloc_bulk(struct km= em_cache *s, gfp_t flags, =20 error: slub_put_cpu_ptr(s->cpu_slab); - slab_post_alloc_hook(s, objcg, flags, i, p, false, s->object_size); - kmem_cache_free_bulk(s, i, p); + __kmem_cache_free_bulk(s, i, p); return 0; =20 } #else /* CONFIG_SLUB_TINY */ static int __kmem_cache_alloc_bulk(struct kmem_cache *s, gfp_t flags, - size_t size, void **p, struct obj_cgroup *objcg) + size_t size, void **p) { int i; =20 @@ -4593,8 +4613,7 @@ static int __kmem_cache_alloc_bulk(struct kmem_cache = *s, gfp_t flags, return i; =20 error: - slab_post_alloc_hook(s, objcg, flags, i, p, false, s->object_size); - kmem_cache_free_bulk(s, i, p); + __kmem_cache_free_bulk(s, i, p); return 0; } #endif /* CONFIG_SLUB_TINY */ @@ -4614,7 +4633,7 @@ int kmem_cache_alloc_bulk(struct kmem_cache *s, gfp_t= flags, size_t size, if (unlikely(!s)) return 0; =20 - i =3D __kmem_cache_alloc_bulk(s, flags, size, p, objcg); + i =3D __kmem_cache_alloc_bulk(s, flags, size, p); =20 /* * memcg and kmem_cache debug support and memory initialization. --=20 2.43.0 From nobody Sun Dec 28 21:19:44 2025 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 vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 96D2EC4167B for ; Mon, 4 Dec 2023 19:35:10 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S233614AbjLDTfA (ORCPT ); Mon, 4 Dec 2023 14:35:00 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:45814 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S232466AbjLDTet (ORCPT ); Mon, 4 Dec 2023 14:34:49 -0500 Received: from smtp-out2.suse.de (smtp-out2.suse.de [195.135.223.131]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id BCA2CF0 for ; Mon, 4 Dec 2023 11:34:54 -0800 (PST) Received: from imap1.dmz-prg2.suse.org (imap1.dmz-prg2.suse.org [10.150.64.97]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (4096 bits) server-digest SHA256) (No client certificate requested) by smtp-out2.suse.de (Postfix) with ESMTPS id BDC741FE6F; Mon, 4 Dec 2023 19:34:52 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=suse.cz; s=susede2_rsa; t=1701718492; h=from:from:reply-to: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; bh=oMJ38qh2PsFVzYJdgL9aE/PQtVEO5BW0t1lp5LOzGhE=; b=heFvZSEqKXFYYnE56gLumyjwnovfMoThgCUklAnz/7rCIit0pSFbwcimdNm5j3c/yxeUWW gkrHrXu27VHKyotHjOARtOBi8vu9/U7UBOCZ6dklVeHUlRia8ZCmSyeJ+Yffki83sZt/QD oVMt4iNE5pnET1BPfPzYzcZjH7a7qF0= DKIM-Signature: v=1; a=ed25519-sha256; c=relaxed/relaxed; d=suse.cz; s=susede2_ed25519; t=1701718492; h=from:from:reply-to: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; bh=oMJ38qh2PsFVzYJdgL9aE/PQtVEO5BW0t1lp5LOzGhE=; b=AYjmL9saWR49xKlagqu2kHY52YN3HlACqJSDOaY4oQSGsQj6skj3DVaT9RDS9UPQoco+TP i62y5/OcDjMfblCA== Received: from imap1.dmz-prg2.suse.org (localhost [127.0.0.1]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (4096 bits) server-digest SHA256) (No client certificate requested) by imap1.dmz-prg2.suse.org (Postfix) with ESMTPS id A101E13AC1; Mon, 4 Dec 2023 19:34:52 +0000 (UTC) Received: from dovecot-director2.suse.de ([2a07:de40:b281:106:10:150:64:167]) by imap1.dmz-prg2.suse.org with ESMTPSA id 6Gn/JtwpbmUPMwAAD6G6ig (envelope-from ); Mon, 04 Dec 2023 19:34:52 +0000 From: Vlastimil Babka Date: Mon, 04 Dec 2023 20:34:42 +0100 Subject: [PATCH 3/4] mm/slub: handle bulk and single object freeing separately MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Message-Id: <20231204-slub-cleanup-hooks-v1-3-88b65f7cd9d5@suse.cz> References: <20231204-slub-cleanup-hooks-v1-0-88b65f7cd9d5@suse.cz> In-Reply-To: <20231204-slub-cleanup-hooks-v1-0-88b65f7cd9d5@suse.cz> To: Christoph Lameter , Pekka Enberg , David Rientjes , Joonsoo Kim Cc: Andrew Morton , Roman Gushchin , Hyeonggon Yoo <42.hyeyoo@gmail.com>, Alexander Potapenko , Marco Elver , Dmitry Vyukov , linux-mm@kvack.org, linux-kernel@vger.kernel.org, kasan-dev@googlegroups.com, Vlastimil Babka X-Mailer: b4 0.12.4 Authentication-Results: smtp-out2.suse.de; none X-Spamd-Result: default: False [-2.60 / 50.00]; ARC_NA(0.00)[]; RCVD_VIA_SMTP_AUTH(0.00)[]; BAYES_HAM(-3.00)[100.00%]; FROM_HAS_DN(0.00)[]; TO_DN_SOME(0.00)[]; TO_MATCH_ENVRCPT_ALL(0.00)[]; FREEMAIL_ENVRCPT(0.00)[gmail.com]; TAGGED_RCPT(0.00)[]; MIME_GOOD(-0.10)[text/plain]; NEURAL_HAM_LONG(-1.00)[-1.000]; R_RATELIMIT(0.00)[to_ip_from(RLtz7ce9b89hw8xzamye9qeynd)]; RCVD_COUNT_THREE(0.00)[3]; DKIM_SIGNED(0.00)[suse.cz:s=susede2_rsa,suse.cz:s=susede2_ed25519]; RCPT_COUNT_TWELVE(0.00)[14]; DBL_BLOCKED_OPENRESOLVER(0.00)[suse.cz:email]; FUZZY_BLOCKED(0.00)[rspamd.com]; FROM_EQ_ENVFROM(0.00)[]; MIME_TRACE(0.00)[0:+]; FREEMAIL_CC(0.00)[linux-foundation.org,linux.dev,gmail.com,google.com,kvack.org,vger.kernel.org,googlegroups.com,suse.cz]; RCVD_TLS_ALL(0.00)[]; MID_RHS_MATCH_FROM(0.00)[]; SUSPICIOUS_RECIPS(1.50)[] Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Currently we have a single function slab_free() handling both single object freeing and bulk freeing with necessary hooks, the latter case requiring slab_free_freelist_hook(). It should be however better to distinguish the two use cases for the following reasons: - code simpler to follow for the single object case - better code generation - although inlining should eliminate the slab_free_freelist_hook() for single object freeing in case no debugging options are enabled, it seems it's not perfect. When e.g. KASAN is enabled, we're imposing additional unnecessary overhead for single object freeing. - preparation to add percpu array caches in near future Therefore, simplify slab_free() for the single object case by dropping unnecessary parameters and calling only slab_free_hook() instead of slab_free_freelist_hook(). Rename the bulk variant to slab_free_bulk() and adjust callers accordingly. While at it, flip (and document) slab_free_hook() return value so that it returns true when the freeing can proceed, which matches the logic of slab_free_freelist_hook() and is not confusingly the opposite. Additionally we can simplify a bit by changing the tail parameter of do_slab_free() when freeing a single object - instead of NULL we can set it equal to head. bloat-o-meter shows small code reduction with a .config that has KASAN etc disabled: add/remove: 0/0 grow/shrink: 0/4 up/down: 0/-118 (-118) Function old new delta kmem_cache_alloc_bulk 1203 1196 -7 kmem_cache_free 861 835 -26 __kmem_cache_free 741 704 -37 kmem_cache_free_bulk 911 863 -48 Signed-off-by: Vlastimil Babka Reviewed-by: Chengming Zhou --- mm/slub.c | 59 +++++++++++++++++++++++++++++++++++------------------------ 1 file changed, 35 insertions(+), 24 deletions(-) diff --git a/mm/slub.c b/mm/slub.c index 0742564c4538..ed2fa92e914c 100644 --- a/mm/slub.c +++ b/mm/slub.c @@ -2037,9 +2037,12 @@ static inline void memcg_slab_free_hook(struct kmem_= cache *s, struct slab *slab, /* * Hooks for other subsystems that check memory allocations. In a typical * production configuration these hooks all should produce no code at all. + * + * Returns true if freeing of the object can proceed, false if its reuse + * was delayed by KASAN quarantine. */ -static __always_inline bool slab_free_hook(struct kmem_cache *s, - void *x, bool init) +static __always_inline +bool slab_free_hook(struct kmem_cache *s, void *x, bool init) { kmemleak_free_recursive(x, s->flags); kmsan_slab_free(s, x); @@ -2072,7 +2075,7 @@ static __always_inline bool slab_free_hook(struct kme= m_cache *s, s->size - s->inuse - rsize); } /* KASAN might put x into memory quarantine, delaying its reuse. */ - return kasan_slab_free(s, x, init); + return !kasan_slab_free(s, x, init); } =20 static inline bool slab_free_freelist_hook(struct kmem_cache *s, @@ -2082,7 +2085,7 @@ static inline bool slab_free_freelist_hook(struct kme= m_cache *s, =20 void *object; void *next =3D *head; - void *old_tail =3D *tail ? *tail : *head; + void *old_tail =3D *tail; =20 if (is_kfence_address(next)) { slab_free_hook(s, next, false); @@ -2098,8 +2101,8 @@ static inline bool slab_free_freelist_hook(struct kme= m_cache *s, next =3D get_freepointer(s, object); =20 /* If object's reuse doesn't have to be delayed */ - if (likely(!slab_free_hook(s, object, - slab_want_init_on_free(s)))) { + if (likely(slab_free_hook(s, object, + slab_want_init_on_free(s)))) { /* Move object to the new freelist */ set_freepointer(s, object, *head); *head =3D object; @@ -2114,9 +2117,6 @@ static inline bool slab_free_freelist_hook(struct kme= m_cache *s, } } while (object !=3D old_tail); =20 - if (*head =3D=3D *tail) - *tail =3D NULL; - return *head !=3D NULL; } =20 @@ -4227,7 +4227,6 @@ static __always_inline void do_slab_free(struct kmem_= cache *s, struct slab *slab, void *head, void *tail, int cnt, unsigned long addr) { - void *tail_obj =3D tail ? : head; struct kmem_cache_cpu *c; unsigned long tid; void **freelist; @@ -4246,14 +4245,14 @@ static __always_inline void do_slab_free(struct kme= m_cache *s, barrier(); =20 if (unlikely(slab !=3D c->slab)) { - __slab_free(s, slab, head, tail_obj, cnt, addr); + __slab_free(s, slab, head, tail, cnt, addr); return; } =20 if (USE_LOCKLESS_FAST_PATH()) { freelist =3D READ_ONCE(c->freelist); =20 - set_freepointer(s, tail_obj, freelist); + set_freepointer(s, tail, freelist); =20 if (unlikely(!__update_cpu_freelist_fast(s, freelist, head, tid))) { note_cmpxchg_failure("slab_free", s, tid); @@ -4270,7 +4269,7 @@ static __always_inline void do_slab_free(struct kmem_= cache *s, tid =3D c->tid; freelist =3D c->freelist; =20 - set_freepointer(s, tail_obj, freelist); + set_freepointer(s, tail, freelist); c->freelist =3D head; c->tid =3D next_tid(tid); =20 @@ -4283,15 +4282,27 @@ static void do_slab_free(struct kmem_cache *s, struct slab *slab, void *head, void *tail, int cnt, unsigned long addr) { - void *tail_obj =3D tail ? : head; - - __slab_free(s, slab, head, tail_obj, cnt, addr); + __slab_free(s, slab, head, tail, cnt, addr); } #endif /* CONFIG_SLUB_TINY */ =20 -static __fastpath_inline void slab_free(struct kmem_cache *s, struct slab = *slab, - void *head, void *tail, void **p, int cnt, - unsigned long addr) +static __fastpath_inline +void slab_free(struct kmem_cache *s, struct slab *slab, void *object, + unsigned long addr) +{ + bool init; + + memcg_slab_free_hook(s, slab, &object, 1); + + init =3D !is_kfence_address(object) && slab_want_init_on_free(s); + + if (likely(slab_free_hook(s, object, init))) + do_slab_free(s, slab, object, object, 1, addr); +} + +static __fastpath_inline +void slab_free_bulk(struct kmem_cache *s, struct slab *slab, void *head, + void *tail, void **p, int cnt, unsigned long addr) { memcg_slab_free_hook(s, slab, p, cnt); /* @@ -4305,7 +4316,7 @@ static __fastpath_inline void slab_free(struct kmem_c= ache *s, struct slab *slab, #ifdef CONFIG_KASAN_GENERIC void ___cache_free(struct kmem_cache *cache, void *x, unsigned long addr) { - do_slab_free(cache, virt_to_slab(x), x, NULL, 1, addr); + do_slab_free(cache, virt_to_slab(x), x, x, 1, addr); } #endif =20 @@ -4349,7 +4360,7 @@ void kmem_cache_free(struct kmem_cache *s, void *x) if (!s) return; trace_kmem_cache_free(_RET_IP_, x, s); - slab_free(s, virt_to_slab(x), x, NULL, &x, 1, _RET_IP_); + slab_free(s, virt_to_slab(x), x, _RET_IP_); } EXPORT_SYMBOL(kmem_cache_free); =20 @@ -4395,7 +4406,7 @@ void kfree(const void *object) =20 slab =3D folio_slab(folio); s =3D slab->slab_cache; - slab_free(s, slab, x, NULL, &x, 1, _RET_IP_); + slab_free(s, slab, x, _RET_IP_); } EXPORT_SYMBOL(kfree); =20 @@ -4512,8 +4523,8 @@ void kmem_cache_free_bulk(struct kmem_cache *s, size_= t size, void **p) if (!df.slab) continue; =20 - slab_free(df.s, df.slab, df.freelist, df.tail, &p[size], df.cnt, - _RET_IP_); + slab_free_bulk(df.s, df.slab, df.freelist, df.tail, &p[size], + df.cnt, _RET_IP_); } while (likely(size)); } EXPORT_SYMBOL(kmem_cache_free_bulk); --=20 2.43.0 From nobody Sun Dec 28 21:19:44 2025 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 vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 21DBDC10DC1 for ; Mon, 4 Dec 2023 19:35:06 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S233459AbjLDTe5 (ORCPT ); Mon, 4 Dec 2023 14:34:57 -0500 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:45804 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S231779AbjLDTes (ORCPT ); Mon, 4 Dec 2023 14:34:48 -0500 Received: from smtp-out2.suse.de (smtp-out2.suse.de [195.135.223.131]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id BCEB9101 for ; Mon, 4 Dec 2023 11:34:54 -0800 (PST) Received: from imap1.dmz-prg2.suse.org (imap1.dmz-prg2.suse.org [IPv6:2a07:de40:b281:104:10:150:64:97]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (4096 bits) server-digest SHA256) (No client certificate requested) by smtp-out2.suse.de (Postfix) with ESMTPS id D710B1FE70; Mon, 4 Dec 2023 19:34:52 +0000 (UTC) Received: from imap1.dmz-prg2.suse.org (localhost [127.0.0.1]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (4096 bits) server-digest SHA256) (No client certificate requested) by imap1.dmz-prg2.suse.org (Postfix) with ESMTPS id B9C9B139AA; Mon, 4 Dec 2023 19:34:52 +0000 (UTC) Received: from dovecot-director2.suse.de ([2a07:de40:b281:106:10:150:64:167]) by imap1.dmz-prg2.suse.org with ESMTPSA id KLwRLdwpbmUPMwAAD6G6ig (envelope-from ); Mon, 04 Dec 2023 19:34:52 +0000 From: Vlastimil Babka Date: Mon, 04 Dec 2023 20:34:43 +0100 Subject: [PATCH 4/4] mm/slub: free KFENCE objects in slab_free_hook() MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Message-Id: <20231204-slub-cleanup-hooks-v1-4-88b65f7cd9d5@suse.cz> References: <20231204-slub-cleanup-hooks-v1-0-88b65f7cd9d5@suse.cz> In-Reply-To: <20231204-slub-cleanup-hooks-v1-0-88b65f7cd9d5@suse.cz> To: Christoph Lameter , Pekka Enberg , David Rientjes , Joonsoo Kim Cc: Andrew Morton , Roman Gushchin , Hyeonggon Yoo <42.hyeyoo@gmail.com>, Alexander Potapenko , Marco Elver , Dmitry Vyukov , linux-mm@kvack.org, linux-kernel@vger.kernel.org, kasan-dev@googlegroups.com, Vlastimil Babka X-Mailer: b4 0.12.4 X-Spamd-Bar: +++++ Authentication-Results: smtp-out2.suse.de; dkim=none; dmarc=none; spf=softfail (smtp-out2.suse.de: 2a07:de40:b281:104:10:150:64:97 is neither permitted nor denied by domain of vbabka@suse.cz) smtp.mailfrom=vbabka@suse.cz X-Rspamd-Server: rspamd2 X-Spamd-Result: default: False [5.39 / 50.00]; RCVD_VIA_SMTP_AUTH(0.00)[]; SPAMHAUS_XBL(0.00)[2a07:de40:b281:104:10:150:64:97:from]; TO_DN_SOME(0.00)[]; R_SPF_SOFTFAIL(4.60)[~all:c]; R_RATELIMIT(0.00)[to_ip_from(RLhc4kaujr6ihojcnjq7c1jwbi)]; RCVD_COUNT_THREE(0.00)[3]; MX_GOOD(-0.01)[]; FROM_EQ_ENVFROM(0.00)[]; R_DKIM_NA(2.20)[]; MIME_TRACE(0.00)[0:+]; BAYES_HAM(-3.00)[100.00%]; MID_RHS_MATCH_FROM(0.00)[]; ARC_NA(0.00)[]; FROM_HAS_DN(0.00)[]; TO_MATCH_ENVRCPT_ALL(0.00)[]; FREEMAIL_ENVRCPT(0.00)[gmail.com]; TAGGED_RCPT(0.00)[]; MIME_GOOD(-0.10)[text/plain]; DMARC_NA(1.20)[suse.cz]; NEURAL_HAM_LONG(-1.00)[-1.000]; RCPT_COUNT_TWELVE(0.00)[14]; DBL_BLOCKED_OPENRESOLVER(0.00)[suse.cz:email]; FUZZY_BLOCKED(0.00)[rspamd.com]; FREEMAIL_CC(0.00)[linux-foundation.org,linux.dev,gmail.com,google.com,kvack.org,vger.kernel.org,googlegroups.com,suse.cz]; RCVD_TLS_ALL(0.00)[]; SUSPICIOUS_RECIPS(1.50)[] X-Rspamd-Queue-Id: D710B1FE70 Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org When freeing an object that was allocated from KFENCE, we do that in the slowpath __slab_free(), relying on the fact that KFENCE "slab" cannot be the cpu slab, so the fastpath has to fallback to the slowpath. This optimization doesn't help much though, because is_kfence_address() is checked earlier anyway during the free hook processing or detached freelist building. Thus we can simplify the code by making the slab_free_hook() free the KFENCE object immediately, similarly to KASAN quarantine. In slab_free_hook() we can place kfence_free() above init processing, as callers have been making sure to set init to false for KFENCE objects. This simplifies slab_free(). This places it also above kasan_slab_free() which is ok as that skips KFENCE objects anyway. While at it also determine the init value in slab_free_freelist_hook() outside of the loop. This change will also make introducing per cpu array caches easier. Tested-by: Marco Elver Signed-off-by: Vlastimil Babka Reviewed-by: Chengming Zhou --- mm/slub.c | 22 ++++++++++------------ 1 file changed, 10 insertions(+), 12 deletions(-) diff --git a/mm/slub.c b/mm/slub.c index ed2fa92e914c..e38c2b712f6c 100644 --- a/mm/slub.c +++ b/mm/slub.c @@ -2039,7 +2039,7 @@ static inline void memcg_slab_free_hook(struct kmem_c= ache *s, struct slab *slab, * production configuration these hooks all should produce no code at all. * * Returns true if freeing of the object can proceed, false if its reuse - * was delayed by KASAN quarantine. + * was delayed by KASAN quarantine, or it was returned to KFENCE. */ static __always_inline bool slab_free_hook(struct kmem_cache *s, void *x, bool init) @@ -2057,6 +2057,9 @@ bool slab_free_hook(struct kmem_cache *s, void *x, bo= ol init) __kcsan_check_access(x, s->object_size, KCSAN_ACCESS_WRITE | KCSAN_ACCESS_ASSERT); =20 + if (kfence_free(kasan_reset_tag(x))) + return false; + /* * As memory initialization might be integrated into KASAN, * kasan_slab_free and initialization memset's must be @@ -2086,23 +2089,25 @@ static inline bool slab_free_freelist_hook(struct k= mem_cache *s, void *object; void *next =3D *head; void *old_tail =3D *tail; + bool init; =20 if (is_kfence_address(next)) { slab_free_hook(s, next, false); - return true; + return false; } =20 /* Head and tail of the reconstructed freelist */ *head =3D NULL; *tail =3D NULL; =20 + init =3D slab_want_init_on_free(s); + do { object =3D next; next =3D get_freepointer(s, object); =20 /* If object's reuse doesn't have to be delayed */ - if (likely(slab_free_hook(s, object, - slab_want_init_on_free(s)))) { + if (likely(slab_free_hook(s, object, init))) { /* Move object to the new freelist */ set_freepointer(s, object, *head); *head =3D object; @@ -4103,9 +4108,6 @@ static void __slab_free(struct kmem_cache *s, struct = slab *slab, =20 stat(s, FREE_SLOWPATH); =20 - if (kfence_free(head)) - return; - if (IS_ENABLED(CONFIG_SLUB_TINY) || kmem_cache_debug(s)) { free_to_partial_list(s, slab, head, tail, cnt, addr); return; @@ -4290,13 +4292,9 @@ static __fastpath_inline void slab_free(struct kmem_cache *s, struct slab *slab, void *object, unsigned long addr) { - bool init; - memcg_slab_free_hook(s, slab, &object, 1); =20 - init =3D !is_kfence_address(object) && slab_want_init_on_free(s); - - if (likely(slab_free_hook(s, object, init))) + if (likely(slab_free_hook(s, object, slab_want_init_on_free(s)))) do_slab_free(s, slab, object, object, 1, addr); } =20 --=20 2.43.0