From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-wg0-f49.google.com (mail-wg0-f49.google.com [74.125.82.49]) by kanga.kvack.org (Postfix) with ESMTP id 48124900016 for ; Fri, 14 Nov 2014 08:33:17 -0500 (EST) Received: by mail-wg0-f49.google.com with SMTP id x13so19430574wgg.8 for ; Fri, 14 Nov 2014 05:33:16 -0800 (PST) Received: from mx2.suse.de (cantor2.suse.de. [195.135.220.15]) by mx.google.com with ESMTPS id q9si3689513wij.61.2014.11.14.05.33.16 for (version=TLSv1 cipher=ECDHE-RSA-RC4-SHA bits=128/128); Fri, 14 Nov 2014 05:33:16 -0800 (PST) From: Mel Gorman Subject: [PATCH 7/7] mm: numa: Add paranoid check around pte_protnone_numa Date: Fri, 14 Nov 2014 13:33:06 +0000 Message-Id: <1415971986-16143-8-git-send-email-mgorman@suse.de> In-Reply-To: <1415971986-16143-1-git-send-email-mgorman@suse.de> References: <1415971986-16143-1-git-send-email-mgorman@suse.de> Sender: owner-linux-mm@kvack.org List-ID: To: Linux Kernel Cc: Linux-MM , Aneesh Kumar , Hugh Dickins , Dave Jones , Rik van Riel , Ingo Molnar , Kirill Shutemov , Sasha Levin , Linus Torvalds , Mel Gorman pte_protnone_numa is only safe to use after VMA checks for PROT_NONE are complete. Treating a real PROT_NONE PTE as a NUMA hinting fault is going to result in strangeness so add a check for it. BUG_ON looks like overkill but if this is hit then it's a serious bug that could result in corruption so do not even try recovering. It would have been more comprehensive to check VMA flags in pte_protnone_numa but it would have made the API ugly just for a debugging check. Signed-off-by: Mel Gorman --- mm/huge_memory.c | 3 +++ mm/memory.c | 3 +++ 2 files changed, 6 insertions(+) diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 4f02010..ae3f3e0 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -1274,6 +1274,9 @@ int do_huge_pmd_numa_page(struct mm_struct *mm, struct vm_area_struct *vma, bool migrated = false; int flags = 0; + /* A PROT_NONE fault should not end up here */ + BUG_ON(!(vma->vm_flags & (VM_READ | VM_EXEC | VM_WRITE))); + ptl = pmd_lock(mm, pmdp); if (unlikely(!pmd_same(pmd, *pmdp))) goto out_unlock; diff --git a/mm/memory.c b/mm/memory.c index 360c3e3..5d45026 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -3116,6 +3116,9 @@ static int do_numa_page(struct mm_struct *mm, struct vm_area_struct *vma, bool migrated = false; int flags = 0; + /* A PROT_NONE fault should not end up here */ + BUG_ON(!(vma->vm_flags & (VM_READ | VM_EXEC | VM_WRITE))); + /* * The "pte" at this point cannot be used safely without * validation through pte_unmap_same(). It's of NUMA type but -- 1.8.4.5 -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org