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 81E41C87FD4 for ; Thu, 29 Aug 2024 19:21:58 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id C18EF6B00A5; Thu, 29 Aug 2024 15:21:57 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id BC7A76B00A8; Thu, 29 Aug 2024 15:21:57 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id A1CA66B00A9; Thu, 29 Aug 2024 15:21:57 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (smtprelay0012.hostedemail.com [216.40.44.12]) by kanga.kvack.org (Postfix) with ESMTP id 7F59B6B00A5 for ; Thu, 29 Aug 2024 15:21:57 -0400 (EDT) Received: from smtpin16.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay07.hostedemail.com (Postfix) with ESMTP id 24831160F00 for ; Thu, 29 Aug 2024 19:21:57 +0000 (UTC) X-FDA: 82506253074.16.6004900 Received: from mail-lf1-f47.google.com (mail-lf1-f47.google.com [209.85.167.47]) by imf04.hostedemail.com (Postfix) with ESMTP id 44CA840021 for ; Thu, 29 Aug 2024 19:21:54 +0000 (UTC) Authentication-Results: imf04.hostedemail.com; dkim=pass header.d=google.com header.s=20230601 header.b=ptM5wC6a; dmarc=pass (policy=reject) header.from=google.com; spf=pass (imf04.hostedemail.com: domain of jiaqiyan@google.com designates 209.85.167.47 as permitted sender) smtp.mailfrom=jiaqiyan@google.com ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=hostedemail.com; s=arc-20220608; t=1724959215; 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=XtYmS9ziQ+XBhfl3YdNZ87YkAlH4hGP3vnWtNfxEJKQ=; b=XgODHQA/wlxCKSkBa/0XKwfMunjaUutG2K3pzEXXkCBN3vcgu7Ra4BRppk80rYWT5lbJjs EHOhWzbrAJ+dLHHuFRN6Q7h4JPvTsfW5bVVRJplMWv6Vkb8A2BEp2PJ3ZZSgGecrDHBxBL ENrVqMNi7D+miL4wy+sk55+IYLoMTe4= ARC-Seal: i=1; s=arc-20220608; d=hostedemail.com; t=1724959215; a=rsa-sha256; cv=none; b=ntLZQRIN1MUQpxz2ktieP4ctS1XSCsALywSL4Efynm8Y8HW2QZ/MEI6SuxL+EZjvfsE52K /5g7mx/ZyL/eSXMp2vxba2V7S/R6wo/DxgR+H5NE8oLuM5RHoZDiaUGNhZnG1TfAwtq9Of PNUMBP4Vs2yPzR+7+26zTH0PKCIVTOs= ARC-Authentication-Results: i=1; imf04.hostedemail.com; dkim=pass header.d=google.com header.s=20230601 header.b=ptM5wC6a; dmarc=pass (policy=reject) header.from=google.com; spf=pass (imf04.hostedemail.com: domain of jiaqiyan@google.com designates 209.85.167.47 as permitted sender) smtp.mailfrom=jiaqiyan@google.com Received: by mail-lf1-f47.google.com with SMTP id 2adb3069b0e04-5334b61b6c2so69e87.1 for ; Thu, 29 Aug 2024 12:21:54 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20230601; t=1724959313; x=1725564113; 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=XtYmS9ziQ+XBhfl3YdNZ87YkAlH4hGP3vnWtNfxEJKQ=; b=ptM5wC6ahbcaY8t2YRKBMlBPY9fsGLZRnhG+S2IQdTmJyzGtEneaAGsE3NQuT2SaDw H9jth2QPGhZ/QXTWhlbs3bsQ96U8nN+QTEUulgGsXLC+S0CsJdH5vQQXUc9QAvWDlozY TzYr6yw5u161jUesHeZgbtui8Fr3hksR+alHWzzYtMOP5KFUAxoH++/R8p8LfCo1Z8su yX/RnvCFQO1gQW5sURnX6aZR0GNJqzB4jony/MjmYiJH406OwyreQuUhAYjnz03h2ybx YyhQwuhu8PIe1Ywq99bIA36nnPUio6iOrLuUNtIeZfRLs9L4isNNJdHBUH9NIIXyl9pl 6qqQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1724959313; x=1725564113; h=content-transfer-encoding:cc:to:subject:message-id:date:from :in-reply-to:references:mime-version:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=XtYmS9ziQ+XBhfl3YdNZ87YkAlH4hGP3vnWtNfxEJKQ=; b=WAtXGHO4LONqDN3OVWGUTEFFrABHIjlEyKWHnc+B61V3od2HpYsi3SvsME2lesjbhT fi4Mcx7VrWsqjseaLZJ5sba1ErmVDEV8B0OJIZYQ3Z9oP0xEd7/wIrBVAh9MHuZbXn2v RnwGXChkES67gDGltK+gkACMXuVcVCNrhlGdTfTaBy7I1Fbit09kOgNmPnbgeEEj7YKd CS023xMjhT35umAtWms+ae0hLiUky5+KkS+2uw6rxyjvY7VVw7bRuJAt7clf0jN2NpxZ n1kpAKEbj4/zGDwPPRtLgo+TJAIrJ/rQ3OaMJWqZ7op+1CBYv10qyG9/GyW0VWwC1dK7 SEaQ== X-Forwarded-Encrypted: i=1; AJvYcCXQQZiJWSkQ4FYXoKTqRQaMy13Y0df2QAivguqIVdcMqJzwrfrBOB8YaEaW4PG7uhRsQkPEKI0jVA==@kvack.org X-Gm-Message-State: AOJu0YyGViRkmhzN7VtHklj57wWkgJshNODkUJqce0KFtV7Lw2CuPeQ5 tYxdWO+TUTh1JSXnQQtrkt3k/VnbCwFcNpWVNFjXcAs2n15oVrksXvi0JfmxxA2enY4AWS/oJw1 uPvAFz2y3lA8uFp3wkl7icFmrVqO/Lq5IPIzn X-Google-Smtp-Source: AGHT+IGPGsmRJGrjlfmPy0Q890/rIIpu+Epc0j8/i49ZLk24mfPdNhJtw1ZZE/u7EAIS8O2vn2KBYv7g9erpJjM6yjo= X-Received: by 2002:a05:6512:3a8e:b0:533:4620:ebe6 with SMTP id 2adb3069b0e04-535455b9261mr19389e87.4.1724959313018; Thu, 29 Aug 2024 12:21:53 -0700 (PDT) MIME-Version: 1.0 References: <20240826204353.2228736-1-peterx@redhat.com> <20240828142422.GU3773488@nvidia.com> <20240828234958.GE3773488@nvidia.com> In-Reply-To: <20240828234958.GE3773488@nvidia.com> From: Jiaqi Yan Date: Thu, 29 Aug 2024 12:21:39 -0700 Message-ID: Subject: Re: [PATCH v2 00/19] mm: Support huge pfnmaps To: Jason Gunthorpe Cc: Peter Xu , linux-kernel@vger.kernel.org, linux-mm@kvack.org, Gavin Shan , Catalin Marinas , x86@kernel.org, Ingo Molnar , Andrew Morton , Paolo Bonzini , Dave Hansen , Thomas Gleixner , Alistair Popple , kvm@vger.kernel.org, linux-arm-kernel@lists.infradead.org, Sean Christopherson , Oscar Salvador , Borislav Petkov , Zi Yan , Axel Rasmussen , David Hildenbrand , Yan Zhao , Will Deacon , Kefeng Wang , Alex Williamson , ankita@nvidia.com Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Rspamd-Server: rspam07 X-Rspamd-Queue-Id: 44CA840021 X-Stat-Signature: fs7dozjf84mg583dxipu5gm9ejs4qjh4 X-Rspam-User: X-HE-Tag: 1724959314-926446 X-HE-Meta: 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 Kxpbr0yL 2qHzwelcIkDiKrIFvXqES2tZMRNau89gXe4ROZpwZcT0fSMgd3hJhRFH7YZSn2HdD9ZQPz8TOmAV6VpbZcP7fjr4lOqzA+n5CfW3Gjq2Iv0faMyZRizQYf0OlTGCQN/9zVBu+H/qQ8mtZRjkqNwKMDyVzoHf1ET8qV8SQPZGztJskK27ziYjWlyB5iW1kzEdCWhDLL7Zr5CeVkY6dmSRbtLlvYCpOWxFhDHa/GnNzPc6Dm32eaXTpOXdUJQ== 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 Wed, Aug 28, 2024 at 4:50=E2=80=AFPM Jason Gunthorpe wr= ote: > > On Wed, Aug 28, 2024 at 09:10:34AM -0700, Jiaqi Yan wrote: > > On Wed, Aug 28, 2024 at 7:24=E2=80=AFAM Jason Gunthorpe wrote: > > > > > > On Tue, Aug 27, 2024 at 05:42:21PM -0700, Jiaqi Yan wrote: > > > > > > > Instead of removing the whole pud, can driver or memory_failure do > > > > something similar to non-struct-page-version of split_huge_page? So > > > > driver doesn't need to re-fault good pages back? > > > > > > It would be far nicer if we didn't have to poke a hole in a 1G mappin= g > > > just for memory failure reporting. > > > > If I follow this, which of the following sounds better? 1. remove pud > > and rely on the driver to re-fault PFNs that it knows are not poisoned > > (what Peter suggested), or 2. keep the pud and allow access to both > > good and bad PFNs. > > In practice I think people will need 2, as breaking up a 1G mapping > just because a few bits are bad will destroy the VM performance. > Totally agreed. > For this the expectation would be for the VM to co-operate and not > keep causing memory failures, or perhaps for the platform to spare in > good memory somehow. Yes, whether a VM gets into a memory-error-consumption loop maliciously or accidentally, a reasonable VMM should have means to detect and break it. > > > Or provide some knob (configured by ?) so that kernel + driver can > > switch between the two? > > This is also sounding reasonable, especially if we need some > alternative protocol to signal userspace about the failed memory > besides fault and SIGBUS. To clarify, what on my mind is a knob say named "sysctl_enable_hard_offline", configured by userspace. To apply to Ankit's memory_failure_pfn patch[*]: static int memory_failure_pfn(unsigned long pfn, int flags) { struct interval_tree_node *node; int res =3D MF_FAILED; LIST_HEAD(tokill); mutex_lock(&pfn_space_lock); for (node =3D interval_tree_iter_first(&pfn_space_itree, pfn, pfn); node= ; node =3D interval_tree_iter_next(node, pfn, pfn)) { struct pfn_address_space *pfn_space =3D container_of(node, struct pfn_address_space, node); if (pfn_space->ops) pfn_space->ops->failure(pfn_space, pfn); collect_procs_pgoff(NULL, pfn_space->mapping, pfn, &tokill); if (sysctl_enable_hard_offline) unmap_mapping_range(pfn_space->mapping, pfn << PAGE_SHIFT, PAGE_SIZE, 0); res =3D MF_RECOVERED; } mutex_unlock(&pfn_space_lock); if (res =3D=3D MF_FAILED) return action_result(pfn, MF_MSG_PFN_MAP, res); flags |=3D MF_ACTION_REQUIRED | MF_MUST_KILL; kill_procs(&tokill, true, false, pfn, flags); return action_result(pfn, MF_MSG_PFN_MAP, MF_RECOVERED); } I think we still want to attempt to SIGBUS userspace, regardless of doing unmap_mapping_range or not. [*] https://lore.kernel.org/lkml/20231123003513.24292-2-ankita@nvidia.com/#= t > > Jason