From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932712Ab3DFHE4 (ORCPT ); Sat, 6 Apr 2013 03:04:56 -0400 Received: from mail-ea0-f178.google.com ([209.85.215.178]:65286 "EHLO mail-ea0-f178.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932572Ab3DFHEz (ORCPT ); Sat, 6 Apr 2013 03:04:55 -0400 Date: Sat, 6 Apr 2013 09:04:50 +0200 From: Michal Hocko To: KOSAKI Motohiro Cc: Naoya Horiguchi , linux-mm@kvack.org, Andrew Morton , Mel Gorman , Hugh Dickins , KOSAKI Motohiro , Andi Kleen , Hillf Danton , linux-kernel@vger.kernel.org Subject: Re: [PATCH 07/10] mbind: add hugepage migration code to mbind() Message-ID: <20130406070450.GC4501@dhcp22.suse.cz> References: <1363983835-20184-1-git-send-email-n-horiguchi@ah.jp.nec.com> <1363983835-20184-8-git-send-email-n-horiguchi@ah.jp.nec.com> <20130325134926.GZ2154@dhcp22.suse.cz> <515F4ECB.9050105@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <515F4ECB.9050105@gmail.com> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Fri 05-04-13 18:23:07, KOSAKI Motohiro wrote: > >> - if (!new_hpage) > >> + /* > >> + * Getting a new hugepage with alloc_huge_page() (which can happen > >> + * when migration is caused by mbind()) can return ERR_PTR value, > >> + * so we need take care of the case here. > >> + */ > >> + if (!new_hpage || IS_ERR_VALUE(new_hpage)) > >> return -ENOMEM; > > > > Please no. get_new_page returns NULL or a page. You are hooking a wrong > > callback here. The error value doesn't make any sense here. IMO you > > should just wrap alloc_huge_page by something that returns NULL or page. > > I suggest just opposite way. new_vma_page() always return ENOMEM, ENOSPC etc instad > of NULL. and caller propegate it to userland. > I guess userland want to distingush why mbind was failed. Sure, and I wasn't suggesting to change alloc_huge_page. I was just pointing out that new_page_t used to return page or NULL and hugetlb unmap_and_move shouldn't be any different in that direction so using alloc_huge_page is not a good fit here. > Anyway, If new_vma_page() have a change to return both NULL and > -ENOMEM. That's a bug. -- Michal Hocko SUSE Labs From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from psmtp.com (na3sys010amx165.postini.com [74.125.245.165]) by kanga.kvack.org (Postfix) with SMTP id 22BDF6B0137 for ; Sat, 6 Apr 2013 03:04:55 -0400 (EDT) Received: by mail-ea0-f170.google.com with SMTP id a15so1636273eae.29 for ; Sat, 06 Apr 2013 00:04:53 -0700 (PDT) Date: Sat, 6 Apr 2013 09:04:50 +0200 From: Michal Hocko Subject: Re: [PATCH 07/10] mbind: add hugepage migration code to mbind() Message-ID: <20130406070450.GC4501@dhcp22.suse.cz> References: <1363983835-20184-1-git-send-email-n-horiguchi@ah.jp.nec.com> <1363983835-20184-8-git-send-email-n-horiguchi@ah.jp.nec.com> <20130325134926.GZ2154@dhcp22.suse.cz> <515F4ECB.9050105@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <515F4ECB.9050105@gmail.com> Sender: owner-linux-mm@kvack.org List-ID: To: KOSAKI Motohiro Cc: Naoya Horiguchi , linux-mm@kvack.org, Andrew Morton , Mel Gorman , Hugh Dickins , KOSAKI Motohiro , Andi Kleen , Hillf Danton , linux-kernel@vger.kernel.org On Fri 05-04-13 18:23:07, KOSAKI Motohiro wrote: > >> - if (!new_hpage) > >> + /* > >> + * Getting a new hugepage with alloc_huge_page() (which can happen > >> + * when migration is caused by mbind()) can return ERR_PTR value, > >> + * so we need take care of the case here. > >> + */ > >> + if (!new_hpage || IS_ERR_VALUE(new_hpage)) > >> return -ENOMEM; > > > > Please no. get_new_page returns NULL or a page. You are hooking a wrong > > callback here. The error value doesn't make any sense here. IMO you > > should just wrap alloc_huge_page by something that returns NULL or page. > > I suggest just opposite way. new_vma_page() always return ENOMEM, ENOSPC etc instad > of NULL. and caller propegate it to userland. > I guess userland want to distingush why mbind was failed. Sure, and I wasn't suggesting to change alloc_huge_page. I was just pointing out that new_page_t used to return page or NULL and hugetlb unmap_and_move shouldn't be any different in that direction so using alloc_huge_page is not a good fit here. > Anyway, If new_vma_page() have a change to return both NULL and > -ENOMEM. That's a bug. -- Michal Hocko SUSE Labs -- 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