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 0BA4AC6FD18 for ; Wed, 29 Mar 2023 00:13:16 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id 5BD316B0078; Tue, 28 Mar 2023 20:13:15 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id 56D896B007B; Tue, 28 Mar 2023 20:13:15 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 40E436B007D; Tue, 28 Mar 2023 20:13:15 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (smtprelay0016.hostedemail.com [216.40.44.16]) by kanga.kvack.org (Postfix) with ESMTP id 340D56B0078 for ; Tue, 28 Mar 2023 20:13:15 -0400 (EDT) Received: from smtpin03.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay09.hostedemail.com (Postfix) with ESMTP id EB92480B5D for ; Wed, 29 Mar 2023 00:13:14 +0000 (UTC) X-FDA: 80620011108.03.FC57262 Received: from mail-yb1-f172.google.com (mail-yb1-f172.google.com [209.85.219.172]) by imf26.hostedemail.com (Postfix) with ESMTP id 32A0D14000A for ; Wed, 29 Mar 2023 00:13:12 +0000 (UTC) Authentication-Results: imf26.hostedemail.com; dkim=pass header.d=google.com header.s=20210112 header.b=RAXol0bR; dmarc=pass (policy=reject) header.from=google.com; spf=pass (imf26.hostedemail.com: domain of jiaqiyan@google.com designates 209.85.219.172 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=1680048793; 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=AJ2OZZkUNXmjvJGJI6YqUVg4cGbpkB8QFhZoxIQY8PU=; b=CS/qBJv2ppxUiQJFv8hUQtfqq3ODIPUrlNFBipKQpmLjZNI176vim8yVml+sDet77JUL6A x1jcRX8eMdUWW1M60rnh+5TBRIXZ/EyLtMmEgzZU567V0GVvWH/iqFE7CkrsWPQEFdheC+ nN4o4S+by3s7Kev/uqRAqiUMgwTwngI= ARC-Authentication-Results: i=1; imf26.hostedemail.com; dkim=pass header.d=google.com header.s=20210112 header.b=RAXol0bR; dmarc=pass (policy=reject) header.from=google.com; spf=pass (imf26.hostedemail.com: domain of jiaqiyan@google.com designates 209.85.219.172 as permitted sender) smtp.mailfrom=jiaqiyan@google.com ARC-Seal: i=1; s=arc-20220608; d=hostedemail.com; t=1680048793; a=rsa-sha256; cv=none; b=c+b/QnlppBRfSsaKvcWhEycpY6XUMeVN8IZZiN6aurfPKWVULfYBK06+0r+BP8JYhZ9kS2 kB3I/qK5+8Dm9sET4zqxQAmiiLfN0nwQfMGNzmbbo1STUOS3UdyeF45lMtFw5IeKmnXB4g M6tfZtwRtwQ1q5p3RjKU7mkcLKbrm+Q= Received: by mail-yb1-f172.google.com with SMTP id k17so17345044ybm.11 for ; Tue, 28 Mar 2023 17:13:12 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20210112; t=1680048792; 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=AJ2OZZkUNXmjvJGJI6YqUVg4cGbpkB8QFhZoxIQY8PU=; b=RAXol0bRrVlfJoE67aV84mQVbKI8iOodXBUyAnKECkxjY4i3kxxJEmsLbOEx2IrdwZ jP+Bi/7E9Qds18zsociP7e/QO3iHacbVmVMutwOM7jF7/vIQ31ez8VBzAUO/rpIXD9Vf 9AUbQTm1YnHxnwLx1KTqOLsOfglln2UFn4b6iXZARqO07jXd3YOfFxDBNUqhRzmSiG8e OnZ+IDZTX02cLb7YCm8Csrqts5HDHhXO5QqTC1z3OFyIo9Qsn02YKYnQl39pkE6PwzAP IYN8jygbuSC/TClapgkKuJO+wWw58h/GHqDQMRKKMZzS8i8J2Z/pvCnaHbLoZnqANHzx ouNg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; t=1680048792; 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=AJ2OZZkUNXmjvJGJI6YqUVg4cGbpkB8QFhZoxIQY8PU=; b=yXFNOiVsfsC/lXoVXcA0HP/VrqUQyo25acqRHDpql178aPKBUqrC+QDP/fEph9qXXf l3Ly7nEIv+hMWFlLKCWl0JKFy67MAeYGXdR5XQLefOwT7ethQ8ip5/1WA/dsRlwYrmAC OxZgtf7tAdcIvFbHlB3U15wf3i1DznZ+1SliKw3iIv1IUVOPBV2NoEqxa8Wd9zyx3Pph eEN00J0tztjn4fX/dn1WT6M+ReqSejPAOP9JY3+EnrAELs3onyBdhZscy2U4H/LsOglD GXHjwNWsfJsQZOcGNkH6YiqXW9x/TUHGObMJ59fkFovqtp2Xtx/yEQQAI07tN7RfsxAD xXZw== X-Gm-Message-State: AAQBX9eAmwCHd1XN9n3cFJ02Hxr9RmOd5ShqDN8FDOQeSOMDr8nOiQ/5 QgMWN62efAZR+0pkE7Cc251wjnObpRyd1NDL1m/5kg== X-Google-Smtp-Source: AKy350ZhDNqeCa73b0qsfQaIYageTj72YrfQzaRBneBjPEsjHxd87592FwTKCc3cKNJ5P3L9ZLJhvrtuRNBMoB6WYr4= X-Received: by 2002:a25:d88a:0:b0:ac9:cb97:bd0e with SMTP id p132-20020a25d88a000000b00ac9cb97bd0emr8651280ybg.5.1680048792153; Tue, 28 Mar 2023 17:13:12 -0700 (PDT) MIME-Version: 1.0 References: <20230327211548.462509-1-jiaqiyan@google.com> <20230327211548.462509-2-jiaqiyan@google.com> In-Reply-To: From: Jiaqi Yan Date: Tue, 28 Mar 2023 17:13:01 -0700 Message-ID: Subject: Re: [PATCH v11 1/3] mm/khugepaged: recover from poisoned anonymous memory To: Yang Shi Cc: kirill.shutemov@linux.intel.com, kirill@shutemov.name, tongtiangen@huawei.com, tony.luck@intel.com, naoya.horiguchi@nec.com, linmiaohe@huawei.com, linux-mm@kvack.org, akpm@linux-foundation.org, osalvador@suse.de, wangkefeng.wang@huawei.com, stevensd@chromium.org, hughd@google.com Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Rspamd-Queue-Id: 32A0D14000A X-Rspamd-Server: rspam09 X-Rspam-User: X-Stat-Signature: 3x1gapknju1bnkabpo6jpbfy9kbgf6rj X-HE-Tag: 1680048792-6966 X-HE-Meta: U2FsdGVkX180VIPJ72KOlw/2V0d4ZT63jzA4f7+rwLUGEqHNEQaglVf+OxcrQtT5GP9soQY92ocE+Qfgew7E8xsdtBncbDUOaKYTuOfGIG7kAoCmUqpcOtmIvfXuOAVorwB4f1PrD84flii0YSI8aL/2RlV0ySYwrNXYAI1htVl07re6Ek/8stD9gCJ9VultbaJ4fOB/t5irqQDG3IXHcpC46UjuHvxkAUHFcTdf0KngO0i+g7LY379fVhaTcu2vu4HDGwe6DN+q+pORGwBFQNvMxC49f0onQlLZ0QD1+iWZOLJzq7ba27KJaCQaKyHuIUpTBPEF3npM0A+xm/oOz0nGlcxZvmZkh63WwYaddfW5g5/RAlCCczjaDIonIPs+jg22yJ2hXM3OpebvOMW/oW+RlbhIHyOWQk2NCXFK8MWSFgxDqDe1/VjIkTHn9DLMUKUbtETxUAkQVBXp335REn7xRNYzafeQxnmWET6re2XuJpqy2KEDsjsaacw5SNKuKLzp60h/5MQ+WTaMhIzpnDrJ0yKu1FiNni/wQMA3jxiw6yWYhlMeQEF0yiuM4/izlid7KnU+9pRzY04dmXST78T9LjS+6jy1RNH1Bpf61NsF68V6qcpcOGlSMOaM0GqbTFIJQ3mESBzez2b+GYUvUJUiAGKPeWcRMYvygSwK4qK1OMZD+Q0Z8637OcCa/Pa1WrsuAyxZrfH/t4CYML3/mP/wovfk6yzXZIupf82qTbPNX+aTI4EPdFZXES1fv8di3oTxE5MPcO1vwvQIoN4AsMw42A2Ir6Fh5CJ8Fl1uCkHEqczinn42okNoHy0Lf5xtqF0VQPBm/ij/YcbB3cwdi7dkUyFaVeWmNOvhXVajfxYqVWooHCf6T+2bSlukzuwO3m8uTCPzcUwKs1vXnfQ9WK/Dezd0dCWJO2382uAMOp98g1JsILuJxvcENtV5WNkIUAFRUmgxOL0e7scAgco MF7603Mt /IriCLD5LQEyzc0ML8HSY874c+VXHgjvRd0lRnXiAfYeZ9H7eSGJ3s2NQuGRqDdgZAIOz+RlhEDNCq8p2OolIirgibGMbyIfZovMVu64Dwea0WW56JyVQ0BPoBSyK6+cAKTmVlo06NNDHFHzqNEWXC89we0PD2N70gvKkFADgW1S34EoOu/3zykOFs7GWI7aV+gRejBuEAA2njhzX/iIzirz8P+//tzDYImT1o4v9KGW4iKPZJkOYga/g/FLGwE3YtIOxW5YSeyG7/Y29XGZYjj5/ZnOAOYLGAvLoaGdhTlNNiAU= 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 Tue, Mar 28, 2023 at 8:59=E2=80=AFAM Yang Shi wrot= e: > > On Mon, Mar 27, 2023 at 2:15=E2=80=AFPM Jiaqi Yan w= rote: > > > > Make __collapse_huge_page_copy return whether copying anonymous pages > > succeeded, and make collapse_huge_page handle the return status. > > > > Break existing PTE scan loop into two for-loops. The first loop copies > > source pages into target huge page, and can fail gracefully when runnin= g > > into memory errors in source pages. If copying all pages succeeds, the > > second loop releases and clears up these normal pages. Otherwise, the > > second loop rolls back the page table and page states by: > > - re-establishing the original PTEs-to-PMD connection. > > - releasing source pages back to their LRU list. > > > > Tested manually: > > 0. Enable khugepaged on system under test. > > 1. Start a two-thread application. Each thread allocates a chunk of > > non-huge anonymous memory buffer. > > 2. Pick 4 random buffer locations (2 in each thread) and inject > > uncorrectable memory errors at corresponding physical addresses. > > 3. Signal both threads to make their memory buffer collapsible, i.e. > > calling madvise(MADV_HUGEPAGE). > > 4. Wait and check kernel log: khugepaged is able to recover from poison= ed > > pages and skips collapsing them. > > 5. Signal both threads to inspect their buffer contents and make sure n= o > > data corruption. > > > > Signed-off-by: Jiaqi Yan > > Reviewed-by: Yang Shi > > Just a nit below: > > > --- > > include/trace/events/huge_memory.h | 3 +- > > mm/khugepaged.c | 114 +++++++++++++++++++++++++---- > > 2 files changed, 103 insertions(+), 14 deletions(-) > > > > diff --git a/include/trace/events/huge_memory.h b/include/trace/events/= huge_memory.h > > index 3e6fb05852f9a..46cce509957ba 100644 > > --- a/include/trace/events/huge_memory.h > > +++ b/include/trace/events/huge_memory.h > > @@ -36,7 +36,8 @@ > > EM( SCAN_ALLOC_HUGE_PAGE_FAIL, "alloc_huge_page_failed") = \ > > EM( SCAN_CGROUP_CHARGE_FAIL, "ccgroup_charge_failed") = \ > > EM( SCAN_TRUNCATED, "truncated") = \ > > - EMe(SCAN_PAGE_HAS_PRIVATE, "page_has_private") = \ > > + EM( SCAN_PAGE_HAS_PRIVATE, "page_has_private") = \ > > + EMe(SCAN_COPY_MC, "copy_poisoned_page") = \ > > > > #undef EM > > #undef EMe > > diff --git a/mm/khugepaged.c b/mm/khugepaged.c > > index bee7fd7db380a..bef68286345c8 100644 > > --- a/mm/khugepaged.c > > +++ b/mm/khugepaged.c > > @@ -55,6 +55,7 @@ enum scan_result { > > SCAN_CGROUP_CHARGE_FAIL, > > SCAN_TRUNCATED, > > SCAN_PAGE_HAS_PRIVATE, > > + SCAN_COPY_MC, > > }; > > > > #define CREATE_TRACE_POINTS > > @@ -681,20 +682,22 @@ static int __collapse_huge_page_isolate(struct vm= _area_struct *vma, > > return result; > > } > > > > -static void __collapse_huge_page_copy(pte_t *pte, struct page *page, > > - struct vm_area_struct *vma, > > - unsigned long address, > > - spinlock_t *ptl, > > - struct list_head *compound_pageli= st) > > +static void __collapse_huge_page_copy_succeeded(pte_t *pte, > > + pmd_t *pmd, > > + struct vm_area_struct *= vma, > > + unsigned long address, > > + spinlock_t *ptl, > > + struct list_head *compo= und_pagelist) > > { > > - struct page *src_page, *tmp; > > + struct page *src_page; > > + struct page *tmp; > > pte_t *_pte; > > - for (_pte =3D pte; _pte < pte + HPAGE_PMD_NR; > > - _pte++, page++, address +=3D PAGE_SIZE)= { > > - pte_t pteval =3D *_pte; > > + pte_t pteval; > > > > + for (_pte =3D pte; _pte < pte + HPAGE_PMD_NR; > > + _pte++, address +=3D PAGE_SIZE) { > > + pteval =3D *_pte; > > if (pte_none(pteval) || is_zero_pfn(pte_pfn(pteval))) { > > - clear_user_highpage(page, address); > > add_mm_counter(vma->vm_mm, MM_ANONPAGES, 1); > > if (is_zero_pfn(pte_pfn(pteval))) { > > /* > > @@ -706,7 +709,6 @@ static void __collapse_huge_page_copy(pte_t *pte, s= truct page *page, > > } > > } else { > > src_page =3D pte_page(pteval); > > - copy_user_highpage(page, src_page, address, vma= ); > > if (!PageCompound(src_page)) > > release_pte_page(src_page); > > /* > > @@ -733,6 +735,88 @@ static void __collapse_huge_page_copy(pte_t *pte, = struct page *page, > > } > > } > > > > +static void __collapse_huge_page_copy_failed(pte_t *pte, > > + pmd_t *pmd, > > + pmd_t orig_pmd, > > + struct vm_area_struct *vma= , > > + unsigned long address, > > It looks like "address" is not used at all. It could be removed. Thanks for catching this! pmd in __collapse_huge_page_copy_succeeded can also be dropped. > > > + struct list_head *compound= _pagelist) > > +{ > > + spinlock_t *pmd_ptl; > > + > > + /* > > + * Re-establish the PMD to point to the original page table > > + * entry. Restoring PMD needs to be done prior to releasing > > + * pages. Since pages are still isolated and locked here, > > + * acquiring anon_vma_lock_write is unnecessary. > > + */ > > + pmd_ptl =3D pmd_lock(vma->vm_mm, pmd); > > + pmd_populate(vma->vm_mm, pmd, pmd_pgtable(orig_pmd)); > > + spin_unlock(pmd_ptl); > > + /* > > + * Release both raw and compound pages isolated > > + * in __collapse_huge_page_isolate. > > + */ > > + release_pte_pages(pte, pte + HPAGE_PMD_NR, compound_pagelist); > > +} > > + > > +/* > > + * __collapse_huge_page_copy - attempts to copy memory contents from r= aw > > + * pages to a hugepage. Cleans up the raw pages if copying succeeds; > > + * otherwise restores the original page table and releases isolated ra= w pages. > > + * Returns SCAN_SUCCEED if copying succeeds, otherwise returns SCAN_CO= PY_MC. > > + * > > + * @pte: starting of the PTEs to copy from > > + * @page: the new hugepage to copy contents to > > + * @pmd: pointer to the new hugepage's PMD > > + * @orig_pmd: the original raw pages' PMD > > + * @vma: the original raw pages' virtual memory area > > + * @address: starting address to copy > > + * @pte_ptl: lock on raw pages' PTEs > > + * @compound_pagelist: list that stores compound pages > > + */ > > +static int __collapse_huge_page_copy(pte_t *pte, > > + struct page *page, > > + pmd_t *pmd, > > + pmd_t orig_pmd, > > + struct vm_area_struct *vma, > > + unsigned long address, > > + spinlock_t *pte_ptl, > > + struct list_head *compound_pagelis= t) > > +{ > > + struct page *src_page; > > + pte_t *_pte; > > + pte_t pteval; > > + unsigned long _address; > > + int result =3D SCAN_SUCCEED; > > + > > + /* > > + * Copying pages' contents is subject to memory poison at any i= teration. > > + */ > > + for (_pte =3D pte, _address =3D address; _pte < pte + HPAGE_PMD= _NR; > > + _pte++, page++, _address +=3D PAGE_SIZE) { > > + pteval =3D *_pte; > > + if (pte_none(pteval) || is_zero_pfn(pte_pfn(pteval))) { > > + clear_user_highpage(page, _address); > > + continue; > > + } > > + src_page =3D pte_page(pteval); > > + if (copy_mc_user_highpage(page, src_page, _address, vma= ) > 0) { > > + result =3D SCAN_COPY_MC; > > + break; > > + } > > + } > > + > > + if (likely(result =3D=3D SCAN_SUCCEED)) > > + __collapse_huge_page_copy_succeeded(pte, pmd, vma, addr= ess, > > + pte_ptl, compound_p= agelist); > > + else > > + __collapse_huge_page_copy_failed(pte, pmd, orig_pmd, vm= a, > > + address, compound_page= list); > > + > > + return result; > > +} > > + > > static void khugepaged_alloc_sleep(void) > > { > > DEFINE_WAIT(wait); > > @@ -1106,9 +1190,13 @@ static int collapse_huge_page(struct mm_struct *= mm, unsigned long address, > > */ > > anon_vma_unlock_write(vma->anon_vma); > > > > - __collapse_huge_page_copy(pte, hpage, vma, address, pte_ptl, > > - &compound_pagelist); > > + result =3D __collapse_huge_page_copy(pte, hpage, pmd, _pmd, > > + vma, address, pte_ptl, > > + &compound_pagelist); > > pte_unmap(pte); > > + if (unlikely(result !=3D SCAN_SUCCEED)) > > + goto out_up_write; > > + > > /* > > * spin_lock() below is not the equivalent of smp_wmb(), but > > * the smp_wmb() inside __SetPageUptodate() can be reused to > > -- > > 2.40.0.348.gf938b09366-goog > >