From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754720AbdLLM5c (ORCPT ); Tue, 12 Dec 2017 07:57:32 -0500 Received: from mail.linuxfoundation.org ([140.211.169.12]:33814 "EHLO mail.linuxfoundation.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754692AbdLLM50 (ORCPT ); Tue, 12 Dec 2017 07:57:26 -0500 From: Greg Kroah-Hartman To: linux-kernel@vger.kernel.org Cc: Greg Kroah-Hartman , stable@vger.kernel.org, "Kirill A. Shutemov" , Andrea Arcangeli , Hillf Danton , Andrew Morton , Linus Torvalds , Jack Wang Subject: [PATCH 4.9 060/148] thp: fix MADV_DONTNEED vs. numa balancing race Date: Tue, 12 Dec 2017 13:44:30 +0100 Message-Id: <20171212124434.597052315@linuxfoundation.org> X-Mailer: git-send-email 2.15.1 In-Reply-To: <20171212124431.207182779@linuxfoundation.org> References: <20171212124431.207182779@linuxfoundation.org> User-Agent: quilt/0.65 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org 4.9-stable review patch. If anyone has any objections, please let me know. ------------------ From: Kirill A. Shutemov commit ced108037c2aa542b3ed8b7afd1576064ad1362a upstream. In case prot_numa, we are under down_read(mmap_sem). It's critical to not clear pmd intermittently to avoid race with MADV_DONTNEED which is also under down_read(mmap_sem): CPU0: CPU1: change_huge_pmd(prot_numa=1) pmdp_huge_get_and_clear_notify() madvise_dontneed() zap_pmd_range() pmd_trans_huge(*pmd) == 0 (without ptl) // skip the pmd set_pmd_at(); // pmd is re-established The race makes MADV_DONTNEED miss the huge pmd and don't clear it which may break userspace. Found by code analysis, never saw triggered. Link: http://lkml.kernel.org/r/20170302151034.27829-3-kirill.shutemov@linux.intel.com Signed-off-by: Kirill A. Shutemov Cc: Andrea Arcangeli Cc: Hillf Danton Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds [jwang: adjust context for 4.9 ] Signed-off-by: Jack Wang Signed-off-by: Greg Kroah-Hartman --- mm/huge_memory.c | 34 +++++++++++++++++++++++++++++++++- 1 file changed, 33 insertions(+), 1 deletion(-) --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -1531,7 +1531,39 @@ int change_huge_pmd(struct vm_area_struc if (prot_numa && pmd_protnone(*pmd)) goto unlock; - entry = pmdp_huge_get_and_clear_notify(mm, addr, pmd); + /* + * In case prot_numa, we are under down_read(mmap_sem). It's critical + * to not clear pmd intermittently to avoid race with MADV_DONTNEED + * which is also under down_read(mmap_sem): + * + * CPU0: CPU1: + * change_huge_pmd(prot_numa=1) + * pmdp_huge_get_and_clear_notify() + * madvise_dontneed() + * zap_pmd_range() + * pmd_trans_huge(*pmd) == 0 (without ptl) + * // skip the pmd + * set_pmd_at(); + * // pmd is re-established + * + * The race makes MADV_DONTNEED miss the huge pmd and don't clear it + * which may break userspace. + * + * pmdp_invalidate() is required to make sure we don't miss + * dirty/young flags set by hardware. + */ + entry = *pmd; + pmdp_invalidate(vma, addr, pmd); + + /* + * Recover dirty/young flags. It relies on pmdp_invalidate to not + * corrupt them. + */ + if (pmd_dirty(*pmd)) + entry = pmd_mkdirty(entry); + if (pmd_young(*pmd)) + entry = pmd_mkyoung(entry); + entry = pmd_modify(entry, newprot); if (preserve_write) entry = pmd_mkwrite(entry);