From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Message-ID: <4FC69035.3000509@gmail.com> Date: Wed, 30 May 2012 17:25:09 -0400 From: KOSAKI Motohiro MIME-Version: 1.0 To: Ben Hutchings CC: KOSAKI Motohiro , Andi Kleen , Christoph Lameter , Linus Torvalds , linux-kernel@vger.kernel.org, linux-mm@kvack.org, Andrew Morton , Dave Jones , Mel Gorman , stable@vger.kernel.org, hughd@google.com, sivanich@sgi.com Subject: Re: [PATCH 0/6] mempolicy memory corruption fixlet References: <1338368529-21784-1-git-send-email-kosaki.motohiro@gmail.com> <20120530184638.GU27374@one.firstfloor.org> <20120530193234.GV27374@one.firstfloor.org> <20120530195244.GX27374@one.firstfloor.org> <20120530212235.GB20051@decadent.org.uk> In-Reply-To: <20120530212235.GB20051@decadent.org.uk> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: (5/30/12 5:22 PM), Ben Hutchings wrote: > On Wed, May 30, 2012 at 04:00:55PM -0400, KOSAKI Motohiro wrote: >> On Wed, May 30, 2012 at 3:52 PM, Andi Kleen wrote: >>> On Wed, May 30, 2012 at 02:42:42PM -0500, Christoph Lameter wrote: >>>> On Wed, 30 May 2012, Andi Kleen wrote: >>>> >>>>> On Wed, May 30, 2012 at 01:50:02PM -0500, Christoph Lameter wrote: >>>>>> On Wed, 30 May 2012, Andi Kleen wrote: >>>>>> >>>>>>> I always regretted that cpusets were no done with custom node lists. >>>>>>> That would have been much cleaner and also likely faster than what we have. >>>>>> >>>>>> Could shared memory policies ignore cpuset constraints? >>>>> >>>>> Only if noone uses cpusets as a "security" mechanism, just for a "soft policy" >>>>> Even with soft policy you could well break someone's setup. >>>> >>>> Well at least lets exempt shared memory from memory migration and memory >>>> policy updates. That seems to be causing many of these issues. >>> >>> Migration on the page level is needed for the memory error handling. >>> >>> Updates: you mean not allowing to set the policy when there are already >>> multiple mappers? I could see that causing some unexpected behaviour. Presumably >>> a standard database will only set it at the beginning, but I don't know >>> if that would work for all users. >> >> We don't need to kill migration core. We only need to kill that mbind(2) updates >> vma->policy of shmem. > [...] > > So should I (and Greg) drop 'mm: mempolicy: Let vma_merge and > vma_split handle vma->vm_policy linkages' from the pending stable > releases? Or is that OK as an interim fix until these changes go > into mainline? Please drop. It screw up mbind(2).