From patchwork Thu Oct 5 16:57:10 2017 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Will Deacon X-Patchwork-Id: 114952 Delivered-To: patch@linaro.org Received: by 10.80.163.170 with SMTP id s39csp430734edb; Thu, 5 Oct 2017 09:57:12 -0700 (PDT) X-Google-Smtp-Source: AOwi7QAmzKmYPRk2c10A9y5t+vBafy2aNWjpRatlBH7xcd1CjiXl5oJCvgQldMbfxIQqmqAldTpx X-Received: by 10.98.31.14 with SMTP id f14mr6367386pff.235.1507222632655; Thu, 05 Oct 2017 09:57:12 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1507222632; cv=none; d=google.com; s=arc-20160816; b=BEG3RmlrBafypiMULgQcLBODMwBNMwXb5uBrwWbFwQ2i/8GnhVH+sibybVeC0viEct qrum14tJNDYhMv7Ptj5JvIgqlao1gQfYC3nY3xsDY/uqIj2Nw97aJyx3PK2qNHHxhUiL EU6oCtg2lqGhrT4OkMUNS9FrI4KGCz4WkQBSNWdMkNfXQYGJaOf+cEZuEc/2gBe+mXM0 FGgKNQNvJ+CSdtCBpaSuCj9QBxNFTyUCDLgPDwljVd9nq7eQYXgjPj5jn5p9EZqZ89Y9 IagFzHWWFgzt6zlRSnxNvnfyu6ZmcA2uV2UOlok+PI0aAcegAMtiGMzeKnqOunwIP/tO BYhw== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=list-id:precedence:sender:message-id:date:subject:cc:to:from :arc-authentication-results; bh=3R/miDMxmM/Xr34Sjtl43DXN/jiZzUsz8rem5pfZ3aQ=; b=01gg3ESSCGrxsh0+ILeE3imqILeisi/jEcmc1kAXi8FyzE3sxo7bSY1cTkPtkNUJX6 U5bBjBXaKkOpFXjucMlZ3rWVipD49GReBVYL5OPzf0Y6XItl6ezygxKpOHEGMACugIqv R2nEv2rft+CgwtpZ2NYM5VL95ic2OivsDytIiZ3mhWYk+bVImyq0SgAsG0k6RbUdzK3o 2un9TSRWW4qKs3K0We97oJDBBcCAQPLzJZEHY4DqC5o9CqPHe4L3QkDyIJnRP/voyWaF etWY8jHXq/4BvSHCENRmL8k8W9pWG7HFu37QaY31eDcUofKJM3p29aoGK1dcvUdEdi0v oYpQ== ARC-Authentication-Results: i=1; mx.google.com; spf=pass (google.com: best guess record for domain of stable-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) smtp.mailfrom=stable-owner@vger.kernel.org Return-Path: Received: from vger.kernel.org (vger.kernel.org. [209.132.180.67]) by mx.google.com with ESMTP id 71si2492341pfs.5.2017.10.05.09.57.12; Thu, 05 Oct 2017 09:57:12 -0700 (PDT) Received-SPF: pass (google.com: best guess record for domain of stable-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67; Authentication-Results: mx.google.com; spf=pass (google.com: best guess record for domain of stable-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) smtp.mailfrom=stable-owner@vger.kernel.org Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751380AbdJEQ5L (ORCPT + 8 others); Thu, 5 Oct 2017 12:57:11 -0400 Received: from foss.arm.com ([217.140.101.70]:49162 "EHLO foss.arm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751335AbdJEQ5K (ORCPT ); Thu, 5 Oct 2017 12:57:10 -0400 Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.72.51.249]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id 4D0A11596; Thu, 5 Oct 2017 09:57:10 -0700 (PDT) Received: from edgewater-inn.cambridge.arm.com (usa-sjc-imap-foss1.foss.arm.com [10.72.51.249]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPA id 1EEE03F53D; Thu, 5 Oct 2017 09:57:10 -0700 (PDT) Received: by edgewater-inn.cambridge.arm.com (Postfix, from userid 1000) id 7D0391AE2F43; Thu, 5 Oct 2017 17:57:11 +0100 (BST) From: Will Deacon To: linux-mm@kvack.org Cc: linux-kernel@vger.kernel.org, akpm@linux-foundation.org, Will Deacon , "Kirill A. Shutemov" , "Paul E. McKenney" , Peter Zijlstra , Subject: [PATCH] mm: page_vma_mapped: Ensure pmd is loaded with READ_ONCE outside of lock Date: Thu, 5 Oct 2017 17:57:10 +0100 Message-Id: <1507222630-5839-1-git-send-email-will.deacon@arm.com> X-Mailer: git-send-email 2.1.4 Sender: stable-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: stable@vger.kernel.org Loading the pmd without holding the pmd_lock exposes us to races with concurrent updaters of the page tables but, worse still, it also allows the compiler to cache the pmd value in a register and reuse it later on, even if we've performed a READ_ONCE in between and seen a more recent value. In the case of page_vma_mapped_walk, this leads to the following crash when the pmd loaded for the initial pmd_trans_huge check is all zeroes and a subsequent valid table entry is loaded by check_pmd. We then proceed into map_pte, but the compiler re-uses the zero entry inside pte_offset_map, resulting in a junk pointer being installed in pvmw->pte: [ 254.032812] PC is at check_pte+0x20/0x170 [ 254.032948] LR is at page_vma_mapped_walk+0x2e0/0x540 [...] [ 254.036114] Process doio (pid: 2463, stack limit = 0xffff00000f2e8000) [ 254.036361] Call trace: [ 254.038977] [] check_pte+0x20/0x170 [ 254.039137] [] page_vma_mapped_walk+0x2e0/0x540 [ 254.039332] [] page_mkclean_one+0xac/0x278 [ 254.039489] [] rmap_walk_file+0xf0/0x238 [ 254.039642] [] rmap_walk+0x64/0xa0 [ 254.039784] [] page_mkclean+0x90/0xa8 [ 254.040029] [] clear_page_dirty_for_io+0x84/0x2a8 [ 254.040311] [] mpage_submit_page+0x34/0x98 [ 254.040518] [] mpage_process_page_bufs+0x164/0x170 [ 254.040743] [] mpage_prepare_extent_to_map+0x134/0x2b8 [ 254.040969] [] ext4_writepages+0x484/0xe30 [ 254.041175] [] do_writepages+0x44/0xe8 [ 254.041372] [] __filemap_fdatawrite_range+0xbc/0x110 [ 254.041568] [] file_write_and_wait_range+0x48/0xd8 [ 254.041739] [] ext4_sync_file+0x80/0x4b8 [ 254.041907] [] vfs_fsync_range+0x64/0xc0 [ 254.042106] [] SyS_msync+0x194/0x1e8 This patch fixes the problem by ensuring that READ_ONCE is used before the initial checks on the pmd, and this value is subsequently used when checking whether or not the pmd is present. pmd_check is removed and the pmd_present check is inlined directly. Cc: "Kirill A. Shutemov" Cc: "Paul E. McKenney" Cc: Peter Zijlstra Cc: Fixes: f27176cfc363 ("mm: convert page_mkclean_one() to use page_vma_mapped_walk()") Tested-by: Yury Norov Tested-by: Richard Ruigrok Signed-off-by: Will Deacon --- mm/page_vma_mapped.c | 25 ++++++++++--------------- 1 file changed, 10 insertions(+), 15 deletions(-) -- 2.1.4 diff --git a/mm/page_vma_mapped.c b/mm/page_vma_mapped.c index 6a03946469a9..6b85f5464246 100644 --- a/mm/page_vma_mapped.c +++ b/mm/page_vma_mapped.c @@ -6,17 +6,6 @@ #include "internal.h" -static inline bool check_pmd(struct page_vma_mapped_walk *pvmw) -{ - pmd_t pmde; - /* - * Make sure we don't re-load pmd between present and !trans_huge check. - * We need a consistent view. - */ - pmde = READ_ONCE(*pvmw->pmd); - return pmd_present(pmde) && !pmd_trans_huge(pmde); -} - static inline bool not_found(struct page_vma_mapped_walk *pvmw) { page_vma_mapped_walk_done(pvmw); @@ -116,6 +105,7 @@ bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw) pgd_t *pgd; p4d_t *p4d; pud_t *pud; + pmd_t pmde; /* The only possible pmd mapping has been handled on last iteration */ if (pvmw->pmd && !pvmw->pte) @@ -148,7 +138,13 @@ bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw) if (!pud_present(*pud)) return false; pvmw->pmd = pmd_offset(pud, pvmw->address); - if (pmd_trans_huge(*pvmw->pmd) || is_pmd_migration_entry(*pvmw->pmd)) { + /* + * Make sure the pmd value isn't cached in a register by the + * compiler and used as a stale value after we've observed a + * subsequent update. + */ + pmde = READ_ONCE(*pvmw->pmd); + if (pmd_trans_huge(pmde) || is_pmd_migration_entry(pmde)) { pvmw->ptl = pmd_lock(mm, pvmw->pmd); if (likely(pmd_trans_huge(*pvmw->pmd))) { if (pvmw->flags & PVMW_MIGRATION) @@ -175,9 +171,8 @@ bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw) spin_unlock(pvmw->ptl); pvmw->ptl = NULL; } - } else { - if (!check_pmd(pvmw)) - return false; + } else if (!pmd_present(pmde)) { + return false; } if (!map_pte(pvmw)) goto next_pte;