From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-vk0-f70.google.com (mail-vk0-f70.google.com [209.85.213.70]) by kanga.kvack.org (Postfix) with ESMTP id 01C286B0260 for ; Sat, 17 Sep 2016 12:05:22 -0400 (EDT) Received: by mail-vk0-f70.google.com with SMTP id b133so87960049vka.0 for ; Sat, 17 Sep 2016 09:05:21 -0700 (PDT) Received: from mx1.redhat.com (mx1.redhat.com. [209.132.183.28]) by mx.google.com with ESMTPS id q8si7346710ybb.144.2016.09.17.09.05.17 for (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Sat, 17 Sep 2016 09:05:18 -0700 (PDT) From: Andrea Arcangeli Subject: [PATCH 0/1] mm: vma_merge: fix vm_page_prot SMP race condition against rmap_walk v2 Date: Sat, 17 Sep 2016 18:05:14 +0200 Message-Id: <1474128315-22726-1-git-send-email-aarcange@redhat.com> In-Reply-To: <20160916205441.GB4743@redhat.com> References: <20160916205441.GB4743@redhat.com> Sender: owner-linux-mm@kvack.org List-ID: To: Andrew Morton Cc: linux-mm@kvack.org, Rik van Riel , Hugh Dickins , Mel Gorman , Jan Vorlicek , Aditya Mandaleeka This version 2 supersedes the 2/2 patch I previously sent in this thread, and restricts the fix to case 8 as all other vma_merge cases were already correct as Hugh pointed out, and the testcase of course reproduced only case 8. I verified that this fixes the race condition with the testcase as good as the v1 patch did. The code is the same I sent yesterday inline in the reply, but this further improves a few comments and it's a more proper submit. Andrea Arcangeli (1): mm: vma_merge: fix vm_page_prot SMP race condition against rmap_walk mm/mmap.c | 40 +++++++++++++++++++++++++++++++++++++--- 1 file changed, 37 insertions(+), 3 deletions(-) -- 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