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 4CD51C433FE for ; Thu, 10 Mar 2022 11:46:40 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id 781B78D0002; Thu, 10 Mar 2022 06:46:39 -0500 (EST) Received: by kanga.kvack.org (Postfix, from userid 40) id 72F3F8D0001; Thu, 10 Mar 2022 06:46:39 -0500 (EST) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 61DA78D0002; Thu, 10 Mar 2022 06:46:39 -0500 (EST) X-Delivered-To: linux-mm@kvack.org Received: from forelay.hostedemail.com (smtprelay0071.hostedemail.com [216.40.44.71]) by kanga.kvack.org (Postfix) with ESMTP id 53F688D0001 for ; Thu, 10 Mar 2022 06:46:39 -0500 (EST) Received: from smtpin19.hostedemail.com (10.5.19.251.rfc1918.com [10.5.19.251]) by forelay01.hostedemail.com (Postfix) with ESMTP id EDD09181C9650 for ; Thu, 10 Mar 2022 11:46:38 +0000 (UTC) X-FDA: 79228299276.19.5A0311F Received: from szxga01-in.huawei.com (szxga01-in.huawei.com [45.249.212.187]) by imf19.hostedemail.com (Postfix) with ESMTP id 79FBB1A001A for ; Thu, 10 Mar 2022 11:46:37 +0000 (UTC) Received: from canpemm500002.china.huawei.com (unknown [172.30.72.56]) by szxga01-in.huawei.com (SkyGuard) with ESMTP id 4KDnGH0lmszcb1m; Thu, 10 Mar 2022 19:41:43 +0800 (CST) Received: from [10.174.177.76] (10.174.177.76) by canpemm500002.china.huawei.com (7.192.104.244) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_128_GCM_SHA256) id 15.1.2308.21; Thu, 10 Mar 2022 19:46:33 +0800 Subject: Re: [PATCH 4/4] mm/memory-failure.c: fix potential VM_BUG_ON_PAGE in split_huge_page_to_list To: Yang Shi CC: =?UTF-8?B?SE9SSUdVQ0hJIE5BT1lBKOWggOWPoyDnm7TkuZ8p?= , "akpm@linux-foundation.org" , "linux-mm@kvack.org" , "linux-kernel@vger.kernel.org" References: <20220228140245.24552-1-linmiaohe@huawei.com> <20220228140245.24552-5-linmiaohe@huawei.com> <20220304082804.GC3778609@hori.linux.bs1.fc.nec.co.jp> <2311bee4-cc11-93fc-6992-6c327a150e3d@huawei.com> From: Miaohe Lin Message-ID: Date: Thu, 10 Mar 2022 19:46:32 +0800 User-Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:78.0) Gecko/20100101 Thunderbird/78.6.0 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset="utf-8" Content-Language: en-US X-Originating-IP: [10.174.177.76] X-ClientProxiedBy: dggems701-chm.china.huawei.com (10.3.19.178) To canpemm500002.china.huawei.com (7.192.104.244) X-CFilter-Loop: Reflected X-Rspam-User: X-Rspamd-Server: rspam02 X-Rspamd-Queue-Id: 79FBB1A001A X-Stat-Signature: 4csfiz5mcfzo3ypaswdpd9muxqz7dikb Authentication-Results: imf19.hostedemail.com; dkim=none; dmarc=pass (policy=quarantine) header.from=huawei.com; spf=pass (imf19.hostedemail.com: domain of linmiaohe@huawei.com designates 45.249.212.187 as permitted sender) smtp.mailfrom=linmiaohe@huawei.com X-HE-Tag: 1646912797-513270 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 2022/3/9 2:47, Yang Shi wrote: > On Tue, Mar 8, 2022 at 4:36 AM Miaohe Lin wrote: >> >> On 2022/3/8 3:53, Yang Shi wrote: >>> On Sun, Mar 6, 2022 at 11:07 PM Miaohe Lin wro= te: >>>> >>>> On 2022/3/4 16:28, HORIGUCHI NAOYA(=E5=A0=80=E5=8F=A3 =E7=9B=B4=E4=B9= =9F) wrote: >>>>> On Mon, Feb 28, 2022 at 10:02:45PM +0800, Miaohe Lin wrote: >>>>>> The huge zero page could reach here and if we ever try to split it= , the >>>>>> VM_BUG_ON_PAGE will be triggered in split_huge_page_to_list(). Als= o the >>>>>> non-lru compound movable pages could be taken for transhuge pages.= Skip >>>>>> these pages by checking PageLRU because huge zero page isn't lru p= age as >>>>>> non-lru compound movable pages. >>>>> >>>>> It seems that memory_failure() also fails at get_any_page() with "h= wpoison: >>>>> unhandlable page" message. >>>>> >>>>> [16478.203474] page:00000000b6acdbd1 refcount:1 mapcount:0 mappin= g:0000000000000000 index:0x0 pfn:0x1810b4 >>>>> [16478.206612] flags: 0x57ffffc0801000(reserved|hwpoison|node=3D1= |zone=3D2|lastcpupid=3D0x1fffff) >>>>> [16478.209411] raw: 0057ffffc0801000 fffff11bc6042d08 fffff11bc60= 42d08 0000000000000000 >>>>> [16478.211921] raw: 0000000000000000 0000000000000000 00000001fff= fffff 0000000000000000 >>>>> [16478.214473] page dumped because: hwpoison: unhandlable page >>>>> [16478.216386] Memory failure: 0x1810b4: recovery action for unkn= own page: Ignored >>>>> >>>>> We can't handle errors on huge (or normal) zero page, so the curren= t >>>> >>>> Sorry for confusing commit log again. I should have a coffee before = I make this patch. >>>> Huge or normal zero page will fail at get_any_page because they're n= either HWPoisonHandlable >>>> nor PageHuge. >>>> >>>>> behavior seems to me more suitable than "unsplit thp". >>>>> >>>>> Or if you have some producer to reach the following path with huge = zero >>>>> page, could you share it? >>>>> >>>> >>>> What I mean is that non-lru movable compound page can reach here une= xpected because __PageMovable(page) >>>> is handleable now. So get_any_page could succeed to grab the page re= fcnt. And since it's compound page, >>>> it will go through the split_huge_page_to_list because PageTransHuge= checks PageHead(page) which can also >>>> be true for compound page. But this type of pages is unexpected for = split_huge_page_to_list. >>> >>> Can we really handle non-LRU movable pages in memory failure >>> (uncorrectable errors)? Typically they are balloon, zsmalloc, etc. >>> Assuming we run into a base (4K) non-LRU movable page, we could reach >>> as far as identify_page_state(), it should not fall into any category >>> except me_unknown. So it seems we could just simply make it >>> unhandlable. >> >> There is the comment from memory_failure: >> /* >> * We ignore non-LRU pages for good reasons. >> * - PG_locked is only well defined for LRU pages and a few ot= hers >> * - to avoid races with __SetPageLocked() >> * - to avoid races with __SetPageSlab*() (and more non-atomic= ops) >> * The check (unnecessarily) ignores LRU pages being isolated = and >> * walked by the page reclaim code, however that's not a big l= oss. >> */ >> >> So we could not handle non-LRU movable pages. >> >> What do you mean is something like below? >> >> diff --git a/mm/memory-failure.c b/mm/memory-failure.c >> index 5444a8ef4867..d80dbe0f20b6 100644 >> --- a/mm/memory-failure.c >> +++ b/mm/memory-failure.c >> @@ -1784,6 +1784,13 @@ int memory_failure(unsigned long pfn, int flags= ) >> } >> } >> >> + if (__PageMovable(hpage)) { >> + put_page(p); >> + action_result(pfn, MF_MSG_MOVALBE_PAGE, MF_IGNORED); >> + res =3D -EBUSY; >> + goto unlock_mutex; >> + } >> + >> if (PageTransHuge(hpage)) { >> /* >> * The flag must be set after the refcount is bumped >> >> >> i.e. Simply make non-LRU movable pages unhandlable ? >=20 I think about the below code more carefully and I found that this will ma= ke hwpoison_filter can't handle the non-LRU movable pages now. Because non-L= RU movable pages return early now and thus can't reach the hwpoison_filter. = This results in a inconsistent behavior with previous one. So I think the orig= in fixup of this patch is more suitable. What do you think? Thanks. > I'd prefer this personally. Something like the below (compile test only= ): >=20 > diff --git a/mm/memory-failure.c b/mm/memory-failure.c > index 5444a8ef4867..789e40909ade 100644 > --- a/mm/memory-failure.c > +++ b/mm/memory-failure.c > @@ -1176,12 +1176,18 @@ void ClearPageHWPoisonTakenOff(struct page *pag= e) > * does not return true for hugetlb or device memory pages, so it's as= sumed > * to be called only in the context where we never have such pages. > */ > -static inline bool HWPoisonHandlable(struct page *page) > +static inline bool HWPoisonHandlable(struct page *page, unsigned long = flags) > { > - return PageLRU(page) || __PageMovable(page) || is_free_buddy_page(pag= e); > + bool movable =3D false; > + > + /* Soft offline could mirgate non-LRU movable pages */ > + if ((flags & MF_SOFT_OFFLINE) && __PageMovable(page)) > + movable =3D true; > + > + return movable || PageLRU(page) || is_free_buddy_page(page); > } >=20 > -static int __get_hwpoison_page(struct page *page) > +static int __get_hwpoison_page(struct page *page, unsigned long flags) > { > struct page *head =3D compound_head(page); > int ret =3D 0; > @@ -1196,7 +1202,7 @@ static int __get_hwpoison_page(struct page *page) > * for any unsupported type of page in order to reduce the risk of > * unexpected races caused by taking a page refcount. > */ > - if (!HWPoisonHandlable(head)) > + if (!HWPoisonHandlable(head, flags)) > return -EBUSY; >=20 > if (get_page_unless_zero(head)) { > @@ -1221,7 +1227,7 @@ static int get_any_page(struct page *p, unsigned > long flags) >=20 > try_again: > if (!count_increased) { > - ret =3D __get_hwpoison_page(p); > + ret =3D __get_hwpoison_page(p, flags); > if (!ret) { > if (page_count(p)) { > /* We raced with an allocation, retry. */ > @@ -1249,7 +1255,7 @@ static int get_any_page(struct page *p, unsigned > long flags) > } > } >=20 > - if (PageHuge(p) || HWPoisonHandlable(p)) { > + if (PageHuge(p) || HWPoisonHandlable(p, flags)) { > ret =3D 1; > } else { > /* >=20 >> >>> >>> But it should be handlable for soft-offline since it could be migrate= d. >>> >> >> Yes, non-LRU movable pages can be simply migrated. >> >> Many thanks. >> >>> >>>> Does this make sense for you? Thanks Naoya. >>>> >>>>> Thanks, >>>>> Naoya Horiguchi >>>>> >>>>>> >>>>>> Signed-off-by: Miaohe Lin >>>>>> --- >>>>>> mm/memory-failure.c | 14 ++++++++++++++ >>>>>> 1 file changed, 14 insertions(+) >>>>>> >>>>>> diff --git a/mm/memory-failure.c b/mm/memory-failure.c >>>>>> index 23bfd809dc8c..ac6492e36978 100644 >>>>>> --- a/mm/memory-failure.c >>>>>> +++ b/mm/memory-failure.c >>>>>> @@ -1792,6 +1792,20 @@ int memory_failure(unsigned long pfn, int f= lags) >>>>>> } >>>>>> >>>>>> if (PageTransHuge(hpage)) { >>>>>> + /* >>>>>> + * The non-lru compound movable pages could be taken = for >>>>>> + * transhuge pages. Also huge zero page could reach h= ere >>>>>> + * and if we ever try to split it, the VM_BUG_ON_PAGE= will >>>>>> + * be triggered in split_huge_page_to_list(). Skip th= ese >>>>>> + * pages by checking PageLRU because huge zero page i= sn't >>>>>> + * lru page as non-lru compound movable pages. >>>>>> + */ >>>>>> + if (!PageLRU(hpage)) { >>>>>> + put_page(p); >>>>>> + action_result(pfn, MF_MSG_UNSPLIT_THP, MF_IGN= ORED); >>>>>> + res =3D -EBUSY; >>>>>> + goto unlock_mutex; >>>>>> + } >>>>>> /* >>>>>> * The flag must be set after the refcount is bumped >>>>>> * otherwise it may race with THP split. >>>>>> -- >>>>>> 2.23.0 >>>> >>>> >>> . >>> >> > . >=20