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 E25C7C87FCB for ; Thu, 29 Aug 2024 23:42:05 +0000 (UTC) Received: by kanga.kvack.org (Postfix) id 64DB16B0085; Thu, 29 Aug 2024 19:42:05 -0400 (EDT) Received: by kanga.kvack.org (Postfix, from userid 40) id 5FDAA6B008C; Thu, 29 Aug 2024 19:42:05 -0400 (EDT) X-Delivered-To: int-list-linux-mm@kvack.org Received: by kanga.kvack.org (Postfix, from userid 63042) id 4A9176B0092; Thu, 29 Aug 2024 19:42:05 -0400 (EDT) X-Delivered-To: linux-mm@kvack.org Received: from relay.hostedemail.com (smtprelay0011.hostedemail.com [216.40.44.11]) by kanga.kvack.org (Postfix) with ESMTP id 2B4296B0085 for ; Thu, 29 Aug 2024 19:42:05 -0400 (EDT) Received: from smtpin12.hostedemail.com (a10.router.float.18 [10.200.18.1]) by unirelay10.hostedemail.com (Postfix) with ESMTP id A1A23C0A16 for ; Thu, 29 Aug 2024 23:42:04 +0000 (UTC) X-FDA: 82506908568.12.4C7729E Received: from bombadil.infradead.org (bombadil.infradead.org [198.137.202.133]) by imf22.hostedemail.com (Postfix) with ESMTP id 9E691C0007 for ; Thu, 29 Aug 2024 23:42:01 +0000 (UTC) Authentication-Results: imf22.hostedemail.com; dkim=pass header.d=infradead.org header.s=bombadil.20210309 header.b=Gr8cjZsD; dmarc=fail reason="No valid SPF, DKIM not aligned (relaxed)" header.from=kernel.org (policy=quarantine); spf=none (imf22.hostedemail.com: domain of mcgrof@infradead.org has no SPF policy when checking 198.137.202.133) smtp.mailfrom=mcgrof@infradead.org ARC-Seal: i=1; s=arc-20220608; d=hostedemail.com; t=1724974849; a=rsa-sha256; cv=none; b=YuAeobkmbyWbxB4d0C+Z6URHQgTJAupX3fF/xyMNVKfEaRz/+5Cr1MEjbZlZ0SB7v5NuhV FfVt0tHl1CoaQkHacKC4qqFQ/TvpHb3uAbS+0vAe379r3bYrEFQQybGWHbrCiThP3wE/W5 aOPkLF2XAfnj+XMZI6i09fGQbzK1OfY= ARC-Authentication-Results: i=1; imf22.hostedemail.com; dkim=pass header.d=infradead.org header.s=bombadil.20210309 header.b=Gr8cjZsD; dmarc=fail reason="No valid SPF, DKIM not aligned (relaxed)" header.from=kernel.org (policy=quarantine); spf=none (imf22.hostedemail.com: domain of mcgrof@infradead.org has no SPF policy when checking 198.137.202.133) smtp.mailfrom=mcgrof@infradead.org ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=hostedemail.com; s=arc-20220608; t=1724974849; h=from:from:sender: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=ybJSaS8XOS7QsdfbQRICGW0IbfDKuXL+/iCoPAdx+KY=; b=inu5MPZzAUicFIbmV6HNdR9/pi/K6G4EmLYExBSxGOIVGx903GXXUtz4NWa6zs7iyO3JZE SmGbKF8uyNpkrQeoWDyEwbrJ28lSYtpwYzC1Ije/YTq4kBxy8S+mrcxbBfMDl3j/nRk5TN XUbYIsf0mQD922OQ7JdD9ZNWAMl3EUQ= DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=infradead.org; s=bombadil.20210309; h=Sender:In-Reply-To: Content-Transfer-Encoding:Content-Type:MIME-Version:References:Message-ID: Subject:Cc:To:From:Date:Reply-To:Content-ID:Content-Description; bh=ybJSaS8XOS7QsdfbQRICGW0IbfDKuXL+/iCoPAdx+KY=; b=Gr8cjZsDERIV8FeBImsAYONIfE Ik/7Ze1hHVdNPBXXLv1TvWeGIsiBY8jb6g9XsyHEPq3qXcBbP5rtGFjMhM3ZSALp3ErHqE5ATFH6b Y4QXmq0oioCRmpW+BnqhpSLuM2NoAy31dhv2n3U2qf2MnL9Ntc4bI+5SEFcuCKkVP8SlG3QWadWqc xVvDL467Sq/1Q0XioN6DrbmFpDWo131QwtfPG0Qlwxkiqm18oQx2Fu+h/yVPjpCR2mX6bxNE5LfkF TgZA9go07M6fgmvh0rQuOwJfoCSxgkbDbbCyJJza3RHBIMvnqaaFQXeJBjdspOLhejlRp/tW0o75q GcK9Qdqg==; Received: from mcgrof by bombadil.infradead.org with local (Exim 4.97.1 #2 (Red Hat Linux)) id 1sjolw-000000043rD-23DS; Thu, 29 Aug 2024 23:41:48 +0000 Date: Thu, 29 Aug 2024 16:41:48 -0700 From: Luis Chamberlain To: Zi Yan Cc: Matthew Wilcox , Sven Schnelle , "Pankaj Raghav (Samsung)" , brauner@kernel.org, akpm@linux-foundation.org, chandan.babu@oracle.com, linux-fsdevel@vger.kernel.org, djwong@kernel.org, hare@suse.de, gost.dev@samsung.com, linux-xfs@vger.kernel.org, hch@lst.de, david@fromorbit.com, yang@os.amperecomputing.com, linux-kernel@vger.kernel.org, linux-mm@kvack.org, john.g.garry@oracle.com, cl@os.amperecomputing.com, p.raghav@samsung.com, ryan.roberts@arm.com, David Howells , linux-s390@vger.kernel.org Subject: Re: [PATCH v13 04/10] mm: split a folio in minimum folio order chunks Message-ID: References: <20240822135018.1931258-1-kernel@pankajraghav.com> <20240822135018.1931258-5-kernel@pankajraghav.com> <221FAE59-097C-4D31-A500-B09EDB07C285@nvidia.com> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline Content-Transfer-Encoding: quoted-printable In-Reply-To: <221FAE59-097C-4D31-A500-B09EDB07C285@nvidia.com> X-Rspamd-Server: rspam12 X-Rspamd-Queue-Id: 9E691C0007 X-Stat-Signature: wq1ue9yua3dh9witb3uqkmueda8p17ik X-Rspamd-Pre-Result: action=add header; module=dmarc; Action set by DMARC X-Rspam-User: X-Rspam: Yes X-HE-Tag: 1724974921-857631 X-HE-Meta: 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 IomeyGK+ hQdNbsejAV66Az4GkxD30ZhSG5mRGdzZbkWAtTqs2lGRVCvc0mH03Zy697CrJiK/WPE5VxIvvEx7kW8ja0Me9xoCKugh0gLRcK524/hJOZq/qfy3hWoW9l294U0L3h+0DfL0dAFt0iJgoiPMrUgfDT6dXHX0T4liwBQTxWW0Ox+kQdZDDadCaRZTdwYDpmQ1hkID7gG+ypBXTWZ25UeW25j+nsCOYZvHZOIbqK5jTIthHQc+o2NKfHhIjy6uPAUf8nnETpHSu78ZftXsYLtOIo8iNTvSxx4/KMLYmLxz1jRA3JvsqtLJMZIesZOLLou6UX/C0vmoemg0/XvEJxgF/piV7q2g+efIM0lP8ph+SjVbMzcLxhYga0fbiSYvwbl5HxWyF+OkSlxtiWvw= 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 Thu, Aug 29, 2024 at 06:12:26PM -0400, Zi Yan wrote: > The issue is that the change to split_huge_page() makes split_huge_page_t= o_list_to_order() > unlocks the wrong subpage. split_huge_page() used to pass the =E2=80=9Cpa= ge=E2=80=9D pointer > to split_huge_page_to_list_to_order(), which keeps that =E2=80=9Cpage=E2= =80=9D still locked. > But this patch changes the =E2=80=9Cpage=E2=80=9D passed into split_huge_= page_to_list_to_order() > always to the head page. >=20 > This fixes the crash on my x86 VM, but it can be improved: >=20 > diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h > index 7c50aeed0522..eff5d2fb5d4e 100644 > --- a/include/linux/huge_mm.h > +++ b/include/linux/huge_mm.h > @@ -320,10 +320,7 @@ bool can_split_folio(struct folio *folio, int *pextr= a_pins); > int split_huge_page_to_list_to_order(struct page *page, struct list_head= *list, > unsigned int new_order); > int split_folio_to_list(struct folio *folio, struct list_head *list); > -static inline int split_huge_page(struct page *page) > -{ > - return split_folio(page_folio(page)); > -} > +int split_huge_page(struct page *page); > void deferred_split_folio(struct folio *folio); >=20 > void __split_huge_pmd(struct vm_area_struct *vma, pmd_t *pmd, > diff --git a/mm/huge_memory.c b/mm/huge_memory.c > index c29af9451d92..4d723dab4336 100644 > --- a/mm/huge_memory.c > +++ b/mm/huge_memory.c > @@ -3297,6 +3297,25 @@ int split_huge_page_to_list_to_order(struct page *= page, struct list_head *list, > return ret; > } >=20 > +int split_huge_page(struct page *page) > +{ > + unsigned int min_order =3D 0; > + struct folio *folio =3D page_folio(page); > + > + if (folio_test_anon(folio)) > + goto out; > + > + if (!folio->mapping) { > + if (folio_test_pmd_mappable(folio)) > + count_vm_event(THP_SPLIT_PAGE_FAILED); > + return -EBUSY; > + } > + > + min_order =3D mapping_min_folio_order(folio->mapping); > +out: > + return split_huge_page_to_list_to_order(page, NULL, min_order); > +} > + > int split_folio_to_list(struct folio *folio, struct list_head *list) > { > unsigned int min_order =3D 0; Confirmed, and also although you suggest it can be improved, I thought that we could do that by sharing more code and putting things in the headers, the below also fixes this but tries to share more code, but I think it is perhaps less easier to understand than your patch. So I think your patch is cleaner and easier as a fix. diff --git a/include/linux/huge_mm.h b/include/linux/huge_mm.h index c275aa9cc105..99cd9c7bf55b 100644 --- a/include/linux/huge_mm.h +++ b/include/linux/huge_mm.h @@ -97,6 +97,7 @@ extern struct kobj_attribute thpsize_shmem_enabled_attr; (!!thp_vma_allowable_orders(vma, vm_flags, tva_flags, BIT(order))) =20 #define split_folio(f) split_folio_to_list(f, NULL) +#define split_folio_to_list(f, list) split_page_folio_to_list(&f->page, f,= list) =20 #ifdef CONFIG_PGTABLE_HAS_HUGE_LEAVES #define HPAGE_PMD_SHIFT PMD_SHIFT @@ -331,10 +332,11 @@ unsigned long thp_get_unmapped_area_vmflags(struct fi= le *filp, unsigned long add bool can_split_folio(struct folio *folio, int caller_pins, int *pextra_pin= s); int split_huge_page_to_list_to_order(struct page *page, struct list_head *= list, unsigned int new_order); -int split_folio_to_list(struct folio *folio, struct list_head *list); +int split_page_folio_to_list(struct page *page, struct folio *folio, + struct list_head *list); static inline int split_huge_page(struct page *page) { - return split_folio(page_folio(page)); + return split_page_folio_to_list(page, page_folio(page), NULL); } void deferred_split_folio(struct folio *folio); =20 @@ -511,7 +513,9 @@ static inline int split_huge_page(struct page *page) return 0; } =20 -static inline int split_folio_to_list(struct folio *folio, struct list_hea= d *list) +static inline int split_page_folio_to_list(struct page *page, + struct folio *folio, + struct list_head *list) { return 0; } diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 169f1a71c95d..b115bfe63b52 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -3529,7 +3529,8 @@ int split_huge_page_to_list_to_order(struct page *pag= e, struct list_head *list, return ret; } =20 -int split_folio_to_list(struct folio *folio, struct list_head *list) +int split_page_folio_to_list(struct page *page, struct folio *folio, + struct list_head *list) { unsigned int min_order =3D 0; =20 @@ -3544,8 +3545,7 @@ int split_folio_to_list(struct folio *folio, struct l= ist_head *list) =20 min_order =3D mapping_min_folio_order(folio->mapping); out: - return split_huge_page_to_list_to_order(&folio->page, list, - min_order); + return split_huge_page_to_list_to_order(page, list, min_order); } =20 void __folio_undo_large_rmappable(struct folio *folio)