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]) by smtp.lore.kernel.org (Postfix) with ESMTP id 90E1DD33981 for ; Mon, 28 Oct 2024 13:45:07 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id D143D6B0085; Mon, 28 Oct 2024 09:45:06 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id CC4116B0088; Mon, 28 Oct 2024 09:45:06 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id B64D76B0089; Mon, 28 Oct 2024 09:45:06 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (smtprelay0010.hostedemail.com [216.40.44.10]) by kanga.kvack.org (Postfix) with ESMTP id 992376B0085 for ; Mon, 28 Oct 2024 09:45:06 -0400 (EDT) Received: from smtpin25.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay01.hostedemail.com (Postfix) with ESMTP id 38F161C5662 for ; Mon, 28 Oct 2024 13:45:06 +0000 (UTC) X-FDA: 82723131330.25.188D7D2 Received: from smtp-out2.suse.de (smtp-out2.suse.de [195.135.223.131]) by imf15.hostedemail.com (Postfix) with ESMTP id F2895A002F for ; Mon, 28 Oct 2024 13:44:39 +0000 (UTC) Authentication-Results: imf15.hostedemail.com; dkim=pass header.d=suse.de header.s=susede2_rsa header.b="OWkfzVl/"; dkim=pass header.d=suse.de header.s=susede2_ed25519 header.b=XYRcD20A; dkim=pass header.d=suse.de header.s=susede2_rsa header.b="OWkfzVl/"; dkim=pass header.d=suse.de header.s=susede2_ed25519 header.b=XYRcD20A; dmarc=pass (policy=none) header.from=suse.de; spf=pass (imf15.hostedemail.com: domain of matz@suse.de designates 195.135.223.131 as permitted sender) smtp.mailfrom=matz@suse.de ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=hostedemail.com; s=arc-20220608; t=1730122930; 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: in-reply-to:in-reply-to:references:references:dkim-signature; bh=iR/AN4/y0s+RfThiBPrdliRlimEX/iA9QxHItnJi1QM=; b=grgrC74KrYgOVaXZxEGDD/AAOSN8n3WoEAhHAYYtnAYIGbBl8Ah6/Nk4VDgdoXGBA1NB9X s24GaLBA88FQRf0dfo/VlpdjtT1bIw4Va5NUrQz+r/YoospWblLZFYbMObvwHNofOt7+a+ AmoTJ3kIY3NUfQiDohWCIhULXh2A0vs= ARC-Seal: i=1; s=arc-20220608; d=hostedemail.com; t=1730122930; a=rsa-sha256; cv=none; b=SHewezLyxFCKb6eadchV1VjhF82P87hrpdJ1YxjT+geqABxfvTmEzk8AhJYj5ZU/XLLmIh cCULaspoJRsXf0DBqdD6DWtxUxZcb3y4ytX+cnvLATcwwN5AHuZsY7PD7ejVFp61mCOh3O p2IcjZZOv5oGaPehmCrdwkF8J/dUM5s= ARC-Authentication-Results: i=1; imf15.hostedemail.com; dkim=pass header.d=suse.de header.s=susede2_rsa header.b="OWkfzVl/"; dkim=pass header.d=suse.de header.s=susede2_ed25519 header.b=XYRcD20A; dkim=pass header.d=suse.de header.s=susede2_rsa header.b="OWkfzVl/"; dkim=pass header.d=suse.de header.s=susede2_ed25519 header.b=XYRcD20A; dmarc=pass (policy=none) header.from=suse.de; spf=pass (imf15.hostedemail.com: domain of matz@suse.de designates 195.135.223.131 as permitted sender) smtp.mailfrom=matz@suse.de Received: from knuth.suse.de (unknown [IPv6:2a07:de40:a101:3:9249:faff:fe06:959]) by smtp-out2.suse.de (Postfix) with ESMTP id EDE7E1FD93; Mon, 28 Oct 2024 13:45:01 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_rsa; t=1730123102; h=from:from:reply-to:date:date:message-id:message-id:to:to:cc:cc: mime-version:mime-version:content-type:content-type: in-reply-to:in-reply-to:references:references; bh=iR/AN4/y0s+RfThiBPrdliRlimEX/iA9QxHItnJi1QM=; b=OWkfzVl/Fadez7L7yLlro93ZMyoT1WoW3wMDHMn2Tqg7tWdM8iTYQcadZAA70ad/mycRcF sUj8y6BYW2n0sJTI3MxcAQ9Cl1+eXAruZXSwsXQE5GwBruaXz0ANRL3ExEEaIlPXcPCCHq hkBm2yt9yLaVNGO59EZ+SEWn7fGvLhk= DKIM-Signature: v=1; a=ed25519-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_ed25519; t=1730123102; h=from:from:reply-to:date:date:message-id:message-id:to:to:cc:cc: mime-version:mime-version:content-type:content-type: in-reply-to:in-reply-to:references:references; bh=iR/AN4/y0s+RfThiBPrdliRlimEX/iA9QxHItnJi1QM=; b=XYRcD20AAWLrV7Fh6k8oNLsmQkPaw2PMQHhWUvHnUHd8Cs1l4w0drRPH9e8PhOnOQFTNK+ jvzhreVeb2GdJODw== DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_rsa; t=1730123102; h=from:from:reply-to:date:date:message-id:message-id:to:to:cc:cc: mime-version:mime-version:content-type:content-type: in-reply-to:in-reply-to:references:references; bh=iR/AN4/y0s+RfThiBPrdliRlimEX/iA9QxHItnJi1QM=; b=OWkfzVl/Fadez7L7yLlro93ZMyoT1WoW3wMDHMn2Tqg7tWdM8iTYQcadZAA70ad/mycRcF sUj8y6BYW2n0sJTI3MxcAQ9Cl1+eXAruZXSwsXQE5GwBruaXz0ANRL3ExEEaIlPXcPCCHq hkBm2yt9yLaVNGO59EZ+SEWn7fGvLhk= DKIM-Signature: v=1; a=ed25519-sha256; c=relaxed/relaxed; d=suse.de; s=susede2_ed25519; t=1730123102; h=from:from:reply-to:date:date:message-id:message-id:to:to:cc:cc: mime-version:mime-version:content-type:content-type: in-reply-to:in-reply-to:references:references; bh=iR/AN4/y0s+RfThiBPrdliRlimEX/iA9QxHItnJi1QM=; b=XYRcD20AAWLrV7Fh6k8oNLsmQkPaw2PMQHhWUvHnUHd8Cs1l4w0drRPH9e8PhOnOQFTNK+ jvzhreVeb2GdJODw== Received: by knuth.suse.de (Postfix, from userid 10510) id D5EFD53D6F9; Mon, 28 Oct 2024 14:45:01 +0100 (CET) Received: from localhost (localhost [127.0.0.1]) by knuth.suse.de (Postfix) with ESMTP id C2FBC53D6F8; Mon, 28 Oct 2024 14:45:01 +0100 (CET) Date: Mon, 28 Oct 2024 14:45:01 +0100 (CET) From: Michael Matz To: Lorenzo Stoakes cc: Vlastimil Babka , Andrew Morton , "Liam R. Howlett" , Jann Horn , Thorsten Leemhuis , linux-mm@kvack.org, linux-kernel@vger.kernel.org, Petr Tesarik , Gabriel Krisman Bertazi , Matthias Bodenbinder , stable@vger.kernel.org, Rik van Riel , Yang Shi Subject: Re: [PATCH hotfix 6.12] mm, mmap: limit THP aligment of anonymous mappings to PMD-aligned sizes In-Reply-To: <2b89811b-5957-4fad-8979-86744678d296@lucifer.local> Message-ID: References: <2050f0d4-57b0-481d-bab8-05e8d48fed0c@leemhuis.info> <20241024151228.101841-2-vbabka@suse.cz> <2b89811b-5957-4fad-8979-86744678d296@lucifer.local> MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII X-Rspamd-Action: no action X-Spamd-Bar: + X-Rspamd-Server: rspam07 X-Rspamd-Queue-Id: F2895A002F X-Stat-Signature: gwxrc5ustdwdwyfqg8k5rdfa5mpsm1db X-Rspam-User: X-HE-Tag: 1730123079-184172 X-HE-Meta: U2FsdGVkX1/hhDesf972K6R1KT8sFzh68cKdLYZiHQ/gjEkpvcF2RnI5MEHpBS/zTGMklKN3za+Z5nxDC8xIMqyDfEnsIR/lrI7uGv1IfZ3FeNPLhVsuUD5HXGzdkZcVx+hA4iIBGjo5QnN/aHctx4+XYMue1vTJj+J5o8AOze5mfckbyspZtH51G4SUcOKiKwqGzfMAvKbD+5Si73Q2p7j+OsKz2US0rjnpSKNv3ga462qHpMb951S0oBlpUnm1Tc3nnT0zGRZcWGioXAbr4pdtq2OMptZntE14guxqBw0G2cUW8ZEBwEGSciLMXmsyUglnK+b9/f0fNGdRU4r2i/xDb51Mk4uWbUbpGaZ5AzIxl5OIt9wVk9ItAfnWpduyT0pgl0Z8jKkMwKMINcQ4qwW6o0T5UkYCigZn8cy7cLLFNIDRp4klXAy8q5DFEqL4SAwY2Sodg1ycXDN5BA+3rr5V1zpBcYrWQn3TllVz/yrROLwJPfPlpeciLQeTVZ7ITIRXak4i6cQ2oTRx6rUKreIO6kVS/GCEyey94a+UlB/T5+fp8bbf4F9PdRbGmBHqTW2zlHCz1o/c6fJnCivtkyNHEqh+pRRfmTSCGhhz01tALgFE3NYxWwS3UGpOQRS4a51kvjV+Q3wrQUJj8F/5tU2g/5bm73XyYTI9lq+EbC6aIPqsfZWRgkIUaKa0yMiNP0ac7floFkOb3Ny7FjELkCTh9FsTjm4iKrHKuSRnEArUwdBCg812Zz3UY7nolIc6f07Gtzg9E65jZmBwuM1QmStnOY8TwtsKM0dmi80RNjvmOZ4qPFzON5KURhqoHicums/SuB+izru64qLFm/W0EapXYX6uhiEGFWZW1U+D7qncU1U2khNJL8s2dzojeSsXYv9DZFTYQrn+no2MCaGQDuiAIId+FVCyNkcIGjn4hUmktJC+8k65mjeroGupORIaTWJPlE8OxjHU2HwQrHO lpRCjgnu M73nICHhTdlUs8G2orctRMWi0MC4fW+h6XeK5nepPVHV48KMP/y6ik59Oie1HZdWHKOpJw+0DnoMlJBatk+GD99WVvgFb2wOwF8NE8bfHRKzf5rBREtdQV+Z8djYEPpuDxAyR+X6L+MHoli8V9deaP+zBOerzBEWLIV5dlDWq4P+wLp/4GtIwqd0yybMKof6wPu/O07cSwezb0F53rpiykhD16KJTMWvk4qblnkkcI5FlWh4OORX6ifDzST24tbn6K3XmiNxCghk8U5Y= 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: Hello, On Thu, 24 Oct 2024, Lorenzo Stoakes wrote: > > benchmark seems to create many mappings of 4632kB, which would have > > merged to a large THP-backed area before commit efa7df3e3bb5 and now > > they are fragmented to multiple areas each aligned to PMD boundary with > > gaps between. The regression then seems to be caused mainly due to the > > benchmark's memory access pattern suffering from TLB or cache aliasing > > due to the aligned boundaries of the individual areas. > > Any more details on precisely why? Anything we found out and theorized about is in the suse bugreport. I think the best theory is TLB aliasing when the mixing^Whash function in the given hardware uses too few bits, and most of them in the low 21-12 bits of an address. Of course that then still depends on the particular access pattern. cactuBSSN has about 20 memory streams in the hot loops, and the accesses are fairly regular from step to step (plus/minus certain strides in 3D arrays). When their start addresses all differ only in the upper bits, you will hit TLB aliasing from time to time, and when the dimensions/strides are just right it occurs often, the N-way associativity doesn't save you anymore and you will hit it very very hard. It was interesting to see how broad the range of CPUs and vendors was that exhibited the problem (in various degrees of severity, from 50% to 600% slowdown), and how more recent CPUs don't show the symptom anymore. I guess the micro-arch guys eventually convinced P&R management that hashing another bit or two is worthwhile the silicon :-) Ciao, Michael.