From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-ee0-f44.google.com (mail-ee0-f44.google.com [74.125.83.44]) by kanga.kvack.org (Postfix) with ESMTP id 6D1896B003B for ; Tue, 10 Dec 2013 10:51:41 -0500 (EST) Received: by mail-ee0-f44.google.com with SMTP id b57so2322195eek.17 for ; Tue, 10 Dec 2013 07:51:40 -0800 (PST) Received: from mx2.suse.de (cantor2.suse.de. [195.135.220.15]) by mx.google.com with ESMTP id 5si14864214eei.165.2013.12.10.07.51.40 for ; Tue, 10 Dec 2013 07:51:40 -0800 (PST) From: Mel Gorman Subject: [PATCH 05/18] mm: numa: Do not clear PTE for pte_numa update Date: Tue, 10 Dec 2013 15:51:23 +0000 Message-Id: <1386690695-27380-6-git-send-email-mgorman@suse.de> In-Reply-To: <1386690695-27380-1-git-send-email-mgorman@suse.de> References: <1386690695-27380-1-git-send-email-mgorman@suse.de> Sender: owner-linux-mm@kvack.org List-ID: To: Andrew Morton Cc: Alex Thorlton , Rik van Riel , Linux-MM , LKML , Mel Gorman The TLB must be flushed if the PTE is updated but change_pte_range is clearing the PTE while marking PTEs pte_numa without necessarily flushing the TLB if it reinserts the same entry. Without the flush, it's conceivable that two processors have different TLBs for the same virtual address and at the very least it would generate spurious faults. This patch only unmaps the pages in change_pte_range for a full protection change. Cc: stable@vger.kernel.org Signed-off-by: Mel Gorman Reviewed-by: Rik van Riel --- mm/mprotect.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/mm/mprotect.c b/mm/mprotect.c index 2666797..0a07e2d 100644 --- a/mm/mprotect.c +++ b/mm/mprotect.c @@ -52,13 +52,14 @@ static unsigned long change_pte_range(struct vm_area_struct *vma, pmd_t *pmd, pte_t ptent; bool updated = false; - ptent = ptep_modify_prot_start(mm, addr, pte); if (!prot_numa) { + ptent = ptep_modify_prot_start(mm, addr, pte); ptent = pte_modify(ptent, newprot); updated = true; } else { struct page *page; + ptent = *pte; page = vm_normal_page(vma, addr, oldpte); if (page) { if (!pte_numa(oldpte)) { @@ -79,7 +80,10 @@ static unsigned long change_pte_range(struct vm_area_struct *vma, pmd_t *pmd, if (updated) pages++; - ptep_modify_prot_commit(mm, addr, pte, ptent); + + /* Only !prot_numa always clears the pte */ + if (!prot_numa) + ptep_modify_prot_commit(mm, addr, pte, ptent); } else if (IS_ENABLED(CONFIG_MIGRATION) && !pte_file(oldpte)) { swp_entry_t entry = pte_to_swp_entry(oldpte); -- 1.8.4 -- 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