Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756092AbbGaUyZ (ORCPT ); Fri, 31 Jul 2015 16:54:25 -0400 Received: from mail-pd0-f182.google.com ([209.85.192.182]:35374 "EHLO mail-pd0-f182.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754676AbbGaUyV (ORCPT ); Fri, 31 Jul 2015 16:54:21 -0400 Date: Fri, 31 Jul 2015 13:54:18 -0700 (PDT) From: David Rientjes X-X-Sender: rientjes@chino.kir.corp.google.com To: Naoya Horiguchi cc: Andrew Morton , Andi Kleen , Dean Nelson , Tony Luck , "Kirill A. Shutemov" , Hugh Dickins , "linux-mm@kvack.org" , "linux-kernel@vger.kernel.org" , Naoya Horiguchi Subject: Re: [PATCH v2 1/5] mm/memory-failure: unlock_page before put_page In-Reply-To: <1438325105-10059-2-git-send-email-n-horiguchi@ah.jp.nec.com> Message-ID: References: <1438325105-10059-1-git-send-email-n-horiguchi@ah.jp.nec.com> <1438325105-10059-2-git-send-email-n-horiguchi@ah.jp.nec.com> User-Agent: Alpine 2.10 (DEB 1266 2009-07-14) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1338 Lines: 36 On Fri, 31 Jul 2015, Naoya Horiguchi wrote: > In "just unpoisoned" path, we do put_page and then unlock_page, which is a > wrong order and causes "freeing locked page" bug. So let's fix it. > > Signed-off-by: Naoya Horiguchi > --- > mm/memory-failure.c | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) > > diff --git v4.2-rc4.orig/mm/memory-failure.c v4.2-rc4/mm/memory-failure.c > index c53543d89282..04d677048af7 100644 > --- v4.2-rc4.orig/mm/memory-failure.c > +++ v4.2-rc4/mm/memory-failure.c > @@ -1209,9 +1209,9 @@ int memory_failure(unsigned long pfn, int trapno, int flags) > if (!PageHWPoison(p)) { > printk(KERN_ERR "MCE %#lx: just unpoisoned\n", pfn); > atomic_long_sub(nr_pages, &num_poisoned_pages); > + unlock_page(hpage); > put_page(hpage); > - res = 0; > - goto out; > + return 0; > } > if (hwpoison_filter(p)) { > if (TestClearPageHWPoison(p)) Looks like you could do the unlock_page() before either the printk or atomic_long_sub(), but probably not important. Acked-by: David Rientjes -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/