From patchwork Thu Feb 17 03:00:38 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: luofei X-Patchwork-Id: 12749248 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 A5C8CC433EF for ; Thu, 17 Feb 2022 03:01:31 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id B9C196B0078; Wed, 16 Feb 2022 22:01:30 -0500 (EST) Received: by kanga.kvack.org (Postfix, from userid 40) id B4A106B0075; Wed, 16 Feb 2022 22:01:30 -0500 (EST) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id A12076B0078; Wed, 16 Feb 2022 22:01:30 -0500 (EST) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (relay.hostedemail.com [64.99.140.27]) by kanga.kvack.org (Postfix) with ESMTP id 929F06B0074 for ; Wed, 16 Feb 2022 22:01:30 -0500 (EST) Received: from smtpin02.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay12.hostedemail.com (Postfix) with ESMTP id 66A681201DC for ; Thu, 17 Feb 2022 03:01:30 +0000 (UTC) X-FDA: 79150771140.02.EAA7C84 Received: from spam.unicloud.com (mx.gosinoic.com [220.194.70.58]) by imf27.hostedemail.com (Postfix) with ESMTP id 2ACE940006 for ; Thu, 17 Feb 2022 03:01:28 +0000 (UTC) Received: from eage.unicloud.com ([220.194.70.35]) by spam.unicloud.com with ESMTP id 21H30oTu032234; Thu, 17 Feb 2022 11:00:50 +0800 (GMT-8) (envelope-from luofei@unicloud.com) Received: from localhost.localdomain (10.10.1.7) by zgys-ex-mb09.Unicloud.com (10.10.0.24) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_CBC_SHA256_P256) id 15.1.2375.17; Thu, 17 Feb 2022 11:00:49 +0800 From: luofei To: , , , , , , , CC: , , , , luofei Subject: [PATCH v2] hw/poison: Add in-use hugepage filter judgement and avoid filter page impact on mce handler Date: Wed, 16 Feb 2022 22:00:38 -0500 Message-ID: <20220217030038.1552124-1-luofei@unicloud.com> X-Mailer: git-send-email 2.27.0 MIME-Version: 1.0 X-Originating-IP: [10.10.1.7] X-ClientProxiedBy: zgys-ex-mb11.Unicloud.com (10.10.0.28) To zgys-ex-mb09.Unicloud.com (10.10.0.24) X-DNSRBL: X-MAIL: spam.unicloud.com 21H30oTu032234 X-Stat-Signature: d56ykb6gthpuo5jrbxjb78mobuiojc9w X-Rspamd-Server: rspam02 X-Rspamd-Queue-Id: 2ACE940006 Authentication-Results: imf27.hostedemail.com; dkim=none; dmarc=none; spf=pass (imf27.hostedemail.com: domain of luofei@unicloud.com designates 220.194.70.58 as permitted sender) smtp.mailfrom=luofei@unicloud.com X-Rspam-User: X-HE-Tag: 1645066888-498240 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: After successfully obtaining the reference count of the huge page, it is still necessary to call hwpoison_filter() to make a filter judgement, otherwise the filter hugepage will be unmaped and the related process may be killed. Also when the huge page meets the filter conditions, it should not be regarded as successful memory_failure() processing for mce handler, but should return a value to inform the caller, otherwise the caller regards the error page has been identified and isolated, which may lead to calling set_mce_nospec() to change page attribute, etc. Signed-off-by: luofei --- arch/x86/kernel/cpu/mce/core.c | 22 +++++++++++----------- include/linux/mm.h | 1 + mm/memory-failure.c | 25 +++++++++++++++++++++++-- 3 files changed, 35 insertions(+), 13 deletions(-) diff --git a/arch/x86/kernel/cpu/mce/core.c b/arch/x86/kernel/cpu/mce/core.c index 5818b837fd4d..c2b99c60225f 100644 --- a/arch/x86/kernel/cpu/mce/core.c +++ b/arch/x86/kernel/cpu/mce/core.c @@ -612,7 +612,7 @@ static int uc_decode_notifier(struct notifier_block *nb, unsigned long val, return NOTIFY_DONE; pfn = mce->addr >> PAGE_SHIFT; - if (!memory_failure(pfn, 0)) { + if (!memory_failure(pfn, MF_MCE_HANDLE)) { set_mce_nospec(pfn, whole_page(mce)); mce->kflags |= MCE_HANDLED_UC; } @@ -1286,7 +1286,7 @@ static void kill_me_now(struct callback_head *ch) static void kill_me_maybe(struct callback_head *cb) { struct task_struct *p = container_of(cb, struct task_struct, mce_kill_me); - int flags = MF_ACTION_REQUIRED; + int flags = MF_ACTION_REQUIRED | MF_MCE_HANDLE; int ret; p->mce_count = 0; @@ -1300,14 +1300,14 @@ static void kill_me_maybe(struct callback_head *cb) set_mce_nospec(p->mce_addr >> PAGE_SHIFT, p->mce_whole_page); sync_core(); return; - } - - /* - * -EHWPOISON from memory_failure() means that it already sent SIGBUS - * to the current process with the proper error info, so no need to - * send SIGBUS here again. - */ - if (ret == -EHWPOISON) + } else if (ret == -EHWPOISON || ret == 1) + /* + * -EHWPOISON from memory_failure() means that it already sent SIGBUS + * to the current process with the proper error info, so no need to + * send SIGBUS here again. + * + * 1 means it's a filter page, no need to deal with. + */ return; pr_err("Memory error not recovered"); @@ -1320,7 +1320,7 @@ static void kill_me_never(struct callback_head *cb) p->mce_count = 0; pr_err("Kernel accessed poison in user space at %llx\n", p->mce_addr); - if (!memory_failure(p->mce_addr >> PAGE_SHIFT, 0)) + if (!memory_failure(p->mce_addr >> PAGE_SHIFT, MF_MCE_HANDLE)) set_mce_nospec(p->mce_addr >> PAGE_SHIFT, p->mce_whole_page); } diff --git a/include/linux/mm.h b/include/linux/mm.h index 213cc569b192..f4703f948e9a 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -3188,6 +3188,7 @@ enum mf_flags { MF_MUST_KILL = 1 << 2, MF_SOFT_OFFLINE = 1 << 3, MF_UNPOISON = 1 << 4, + MF_MCE_HANDLE = 1 << 5, }; extern int memory_failure(unsigned long pfn, int flags); extern void memory_failure_queue(unsigned long pfn, int flags); diff --git a/mm/memory-failure.c b/mm/memory-failure.c index 97a9ed8f87a9..1a0bd91a685b 100644 --- a/mm/memory-failure.c +++ b/mm/memory-failure.c @@ -1526,7 +1526,10 @@ static int memory_failure_hugetlb(unsigned long pfn, int flags) if (TestClearPageHWPoison(head)) num_poisoned_pages_dec(); unlock_page(head); - return 0; + if (flags & MF_MCE_HANDLE) + return 1; + else + return 0; } unlock_page(head); res = MF_FAILED; @@ -1545,6 +1548,17 @@ static int memory_failure_hugetlb(unsigned long pfn, int flags) lock_page(head); page_flags = head->flags; + if (hwpoison_filter(p)) { + if (TestClearPageHWPoison(head)) + num_poisoned_pages_dec(); + put_page(p); + if (flags & MF_MCE_HANDLE) + res = 1; + else + res = 0; + goto out; + } + /* * TODO: hwpoison for pud-sized hugetlb doesn't work right now, so * simply disable it. In order to make it work properly, we need @@ -1613,7 +1627,10 @@ static int memory_failure_dev_pagemap(unsigned long pfn, int flags, goto out; if (hwpoison_filter(page)) { - rc = 0; + if (flags & MF_MCE_HANDLE) + rc = 1; + else + rc = 0; goto unlock; } @@ -1837,6 +1854,10 @@ int memory_failure(unsigned long pfn, int flags) num_poisoned_pages_dec(); unlock_page(p); put_page(p); + if (flags & MF_MCE_HANDLE) + res = 1; + else + res = 0; goto unlock_mutex; }