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 X-Spam-Level: X-Spam-Status: No, score=-15.3 required=3.0 tests=BAYES_00, HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_CR_TRAILER,INCLUDES_PATCH, MAILING_LIST_MULTI,NICE_REPLY_A,SPF_HELO_NONE,SPF_PASS,USER_AGENT_SANE_1 autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 2E590C43461 for ; Thu, 13 May 2021 11:37:43 +0000 (UTC) Received: from kanga.kvack.org (kanga.kvack.org [205.233.56.17]) by mail.kernel.org (Postfix) with ESMTP id D1A946143D for ; Thu, 13 May 2021 11:37:42 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org D1A946143D Authentication-Results: mail.kernel.org; dmarc=none (p=none dis=none) header.from=suse.cz Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=owner-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix) id 44EBD6B0036; Thu, 13 May 2021 07:37:42 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id 3FF216B006E; Thu, 13 May 2021 07:37:42 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 2A0456B0070; Thu, 13 May 2021 07:37:42 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from forelay.hostedemail.com (smtprelay0098.hostedemail.com [216.40.44.98]) by kanga.kvack.org (Postfix) with ESMTP id EF8A56B0036 for ; Thu, 13 May 2021 07:37:41 -0400 (EDT) Received: from smtpin25.hostedemail.com (10.5.19.251.rfc1918.com [10.5.19.251]) by forelay01.hostedemail.com (Postfix) with ESMTP id 7CE4218065075 for ; Thu, 13 May 2021 11:37:41 +0000 (UTC) X-FDA: 78136007922.25.8D9BD89 Received: from mx2.suse.de (mx2.suse.de [195.135.220.15]) by imf06.hostedemail.com (Postfix) with ESMTP id DCD05C000C57 for ; Thu, 13 May 2021 11:37:40 +0000 (UTC) X-Virus-Scanned: by amavisd-new at test-mx.suse.de Received: from relay2.suse.de (unknown [195.135.221.27]) by mx2.suse.de (Postfix) with ESMTP id 70C69B021; Thu, 13 May 2021 11:37:39 +0000 (UTC) To: Marco Elver Cc: Hyeonggon Yoo <42.hyeyoo@gmail.com>, Andrew Morton , iamjoonsoo.kim@lge.com, rientjes@google.com, penberg@kernel.org, cl@linux.com, linux-mm@kvack.org, linux-kernel@vger.kernel.org References: <20210511173448.GA54466@hyeyoo> <20210512195227.245000695c9014242e9a00e5@linux-foundation.org> <20210513031220.GA133011@hyeyoo> <20210512204024.401ff3de38649d7d0f5a45e8@linux-foundation.org> <20210513062809.GA319973@hyeyoo> From: Vlastimil Babka Subject: Re: [PATCH v3] mm, slub: change run-time assertion in kmalloc_index() to compile-time Message-ID: <90c1b2ae-a5c1-07d1-b056-5b3699b6cc77@suse.cz> Date: Thu, 13 May 2021 13:37:38 +0200 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101 Thunderbird/78.10.1 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=utf-8 Content-Language: en-US X-Rspamd-Queue-Id: DCD05C000C57 Authentication-Results: imf06.hostedemail.com; dkim=none; spf=pass (imf06.hostedemail.com: domain of vbabka@suse.cz designates 195.135.220.15 as permitted sender) smtp.mailfrom=vbabka@suse.cz; dmarc=none X-Rspamd-Server: rspam03 X-Stat-Signature: mp49ixakj5epmqa9bsrdrm8sgsrhqess Received-SPF: none (suse.cz>: No applicable sender policy available) receiver=imf06; identity=mailfrom; envelope-from=""; helo=mx2.suse.de; client-ip=195.135.220.15 X-HE-DKIM-Result: none/none X-HE-Tag: 1620905860-336124 Content-Transfer-Encoding: quoted-printable 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: On 5/13/21 12:31 PM, Marco Elver wrote: > On Thu, May 13, 2021 at 10:51AM +0200, Vlastimil Babka wrote: >> On 5/13/21 8:28 AM, Hyeonggon Yoo wrote: >> > On Wed, May 12, 2021 at 08:40:24PM -0700, Andrew Morton wrote: >> >> On Thu, 13 May 2021 12:12:20 +0900 Hyeonggon Yoo <42.hyeyoo@gmail.c= om> wrote: >> >> > On Wed, May 12, 2021 at 07:52:27PM -0700, Andrew Morton wrote: >> >> > > This explodes in mysterious ways. The patch as I have it is ap= pended, >> >> > > for reference. >> >> > >=20 >> >> > > gcc-10.3.0 allmodconfig. >> >> > >=20 >> >> > > This patch suppresses the error: >> >>=20 >> >> Ah, yes, of course, your patch changes kmalloc_index() to require t= hat >> >> it always is called with a constant `size'. kfence_test doesn't do >> >> that. >> >>=20 >> >> kfence is being a bit naughty here - the other kmalloc_index() call= ers >> >> only comple up the call after verifying that `size' is a compile-ti= me >> >> constant. >>=20 >> Agreed. >=20 > It's just a test, and performance doesn't matter for it. Sure. But what if there appear more users where it will matter. Those wou= ld get better performance out of kmalloc_slab(). > The thing is this function lives in , isn't prefixed with > __ or anything like that, so it really does look like a public function= . >=20 >> >> Would something like this work? >>=20 >> I'd prefer if we kept kmalloc_index() for constant sizes only. The bro= ken build >> then warns anyone using it the wrong way that they shouldn't. >=20 > Agreed. Andrew's size_is_constant would do that. Also see my suggestion > below to keep the same interface. >=20 >> Besides, it really >> shouldn't be used outside of slab. >=20 > It's an allocator test. If we want to facilitate testing, it must be > allowed to verify or set up test cases that test boundary conditions > based on internal state. >=20 > In the case of kfence_test it wants:=C2=A0 the cache's alignment to cre= ate > accesses that fall on alignment boundaries; and to verify obj_to_index(= ) > and objs_per_slab_page() are set up correctly. OK. > I think the requirements are: >=20 > 1. Make the interface hard to abuse. Adding the BUILD_BUG_ON does that. Yes. > 2. Facilitate testing. Right. >> But if kfence test really needs this, we could perhaps extract the ind= ex >> determining part out of kmalloc_slab(). >=20 > That would duplicate kmalloc_index()? I don't see the need, let's keep > things simple. They are already "duplicated". But one is tailored for constant sizes, th= e other for variable sizes. >> Hmm or I guess the kfence tests could just use kmalloc_slab() directly= ? >=20 > kmalloc_slab() is internal to slab and should not be exported. So should be kmalloc_index(). However it needs to have the full implement= ation in a header accessible to all kmalloc() users to work, so it's there, vis= ible to anyone. > It'd > require exporting because the tests can be built as modules. That's true. > kmalloc_index() works perfectly fine, and the test really doesn't care > about performance of kmalloc_index(). :-) OK then. > See my suggestion below that builds on Andrew's size_is_constant but > would retain the old interface and support testing. I can accept that, but please also modify/expand the newly added comment.= Now it's *normally* evaluated in compile-time. And there should be warning th= at anyone calling it with size_is_constant =3D=3D false should do that only = in context where performance (and code bloat, most likely too) doesn't matter, such = as unit test. Thanks, Vlastimil > Thanks, > -- Marco >=20 > ------ >8 ------ >=20 > From: Marco Elver > Subject: [PATCH] kfence: test: fix for "mm, slub: change run-time asser= tion in > kmalloc_index() to compile-time" >=20 > Signed-off-by: Marco Elver > --- > include/linux/slab.h | 9 +++++++-- > mm/kfence/kfence_test.c | 5 +++-- > 2 files changed, 10 insertions(+), 4 deletions(-) >=20 > diff --git a/include/linux/slab.h b/include/linux/slab.h > index 27d142564557..7a10bdc4b7a9 100644 > --- a/include/linux/slab.h > +++ b/include/linux/slab.h > @@ -350,7 +350,8 @@ static __always_inline enum kmalloc_cache_type kmal= loc_type(gfp_t flags) > * Note: there's no need to optimize kmalloc_index because it's evalua= ted > * in compile-time. > */ > -static __always_inline unsigned int kmalloc_index(size_t size) > +static __always_inline unsigned int __kmalloc_index(size_t size, > + bool size_is_constant) > { > if (!size) > return 0; > @@ -386,11 +387,15 @@ static __always_inline unsigned int kmalloc_index= (size_t size) > if (size <=3D 16 * 1024 * 1024) return 24; > if (size <=3D 32 * 1024 * 1024) return 25; > =20 > - BUILD_BUG_ON_MSG(1, "unexpected size in kmalloc_index()"); > + if (size_is_constant) > + BUILD_BUG_ON_MSG(1, "unexpected size in kmalloc_index()"); > + else > + BUG(); > =20 > /* Will never be reached. Needed because the compiler may complain */ > return -1; > } > +#define kmalloc_index(s) __kmalloc_index(s, true) > #endif /* !CONFIG_SLOB */ > =20 > void *__kmalloc(size_t size, gfp_t flags) __assume_kmalloc_alignment _= _malloc; > diff --git a/mm/kfence/kfence_test.c b/mm/kfence/kfence_test.c > index 4acf4251ee04..7f24b9bcb2ec 100644 > --- a/mm/kfence/kfence_test.c > +++ b/mm/kfence/kfence_test.c > @@ -197,7 +197,7 @@ static void test_cache_destroy(void) > =20 > static inline size_t kmalloc_cache_alignment(size_t size) > { > - return kmalloc_caches[kmalloc_type(GFP_KERNEL)][kmalloc_index(size)]-= >align; > + return kmalloc_caches[kmalloc_type(GFP_KERNEL)][__kmalloc_index(size,= false)]->align; > } > =20 > /* Must always inline to match stack trace against caller. */ > @@ -267,7 +267,8 @@ static void *test_alloc(struct kunit *test, size_t = size, gfp_t gfp, enum allocat > =20 > if (is_kfence_address(alloc)) { > struct page *page =3D virt_to_head_page(alloc); > - struct kmem_cache *s =3D test_cache ?: kmalloc_caches[kmalloc_type(= GFP_KERNEL)][kmalloc_index(size)]; > + struct kmem_cache *s =3D test_cache ?: > + kmalloc_caches[kmalloc_type(GFP_KERNEL)][__kmalloc_index(size, fa= lse)]; > =20 > /* > * Verify that various helpers return the right values >=20