Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1760150Ab2FARWV (ORCPT ); Fri, 1 Jun 2012 13:22:21 -0400 Received: from mail-qc0-f174.google.com ([209.85.216.174]:37332 "EHLO mail-qc0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1760122Ab2FARWS (ORCPT ); Fri, 1 Jun 2012 13:22:18 -0400 Message-ID: <4FC8FA47.70001@gmail.com> Date: Fri, 01 Jun 2012 13:22:15 -0400 From: KOSAKI Motohiro User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.7; rv:12.0) Gecko/20120428 Thunderbird/12.0.1 MIME-Version: 1.0 To: Nathan Zimmer CC: KOSAKI Motohiro , Andrew Morton , hughd@google.com, npiggin@gmail.com, cl@linux.com, lee.schermerhorn@hp.com, linux-kernel@vger.kernel.org, linux-mm@kvack.org, stable@vger.kernel.org, riel@redhat.com Subject: Re: [PATCH v2] tmpfs not interleaving properly References: <20120531143916.GA16162@gulag1.americas.sgi.com> <4FC7CFEB.5040009@gmail.com> <20120531132515.6af60152.akpm@linux-foundation.org> <4FC7D629.3090801@gmail.com> <20120601142437.GA13739@gulag1.americas.sgi.com> In-Reply-To: <20120601142437.GA13739@gulag1.americas.sgi.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1969 Lines: 50 (6/1/12 10:24 AM), Nathan Zimmer wrote: > On Thu, May 31, 2012 at 04:35:53PM -0400, KOSAKI Motohiro wrote: >> (5/31/12 4:25 PM), Andrew Morton wrote: >>> On Thu, 31 May 2012 16:09:15 -0400 >>> KOSAKI Motohiro wrote: >>> >>>>> --- a/mm/shmem.c >>>>> +++ b/mm/shmem.c >>>>> @@ -929,7 +929,7 @@ static struct page *shmem_alloc_page(gfp_t gfp, >>>>> /* >>>>> * alloc_page_vma() will drop the shared policy reference >>>>> */ >>>>> - return alloc_page_vma(gfp,&pvma, 0); >>>>> + return alloc_page_vma(gfp,&pvma, info->node_offset<< PAGE_SHIFT ); >>>> >>>> 3rd argument of alloc_page_vma() is an address. This is type error. >>> >>> Well, it's an unsigned long... >>> >>> But yes, it is conceptually wrong and *looks* weird. I think we can >>> address that by overcoming our peculair aversion to documenting our >>> code, sigh. This? >> >> Sorry, no. >> >> addr agrument of alloc_pages_vma() have two meanings. >> >> 1) interleave node seed >> 2) look-up key of shmem policy >> >> I think this patch break (2). shmem_get_policy(pol, addr) assume caller honor to >> pass correct address. > > But the pseudo vma we generated in shmem_alloc_page the vm_ops are set to NULL. > So get_vma_policy will return the policy provided by the pseudo vma and not reach > the shmem_get_policy. yes, and it is bug source. we may need to change soon. I guess the right way is to make vm_ops->interleave and interleave_nid uses it if povided. btw, I don't think node_random() is good idea. it is random(pid + jiffies + cycle). current->cpuset_mem_spread_rotor is per-thread value. but you now need per-inode interleave offset. maybe, just inode addition is enough. Why do you need randomness? -- 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/