From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753926AbaIBMbx (ORCPT ); Tue, 2 Sep 2014 08:31:53 -0400 Received: from mx1.redhat.com ([209.132.183.28]:32507 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750954AbaIBMbw (ORCPT ); Tue, 2 Sep 2014 08:31:52 -0400 Date: Tue, 2 Sep 2014 08:31:34 -0400 From: Rafael Aquini To: Konstantin Khlebnikov Cc: Andrew Morton , Konstantin Khlebnikov , Linux Kernel Mailing List , linux-mm@kvack.org, Andrey Ryabinin , Sasha Levin Subject: Re: [PATCH v2 2/6] mm/balloon_compaction: keep ballooned pages away from normal migration path Message-ID: <20140902123133.GC14419@t510.redhat.com> References: <20140830163834.29066.98205.stgit@zurg> <20140830164113.29066.25879.stgit@zurg> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20140830164113.29066.25879.stgit@zurg> User-Agent: Mutt/1.5.23 (2014-03-12) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Sat, Aug 30, 2014 at 08:41:13PM +0400, Konstantin Khlebnikov wrote: > From: Konstantin Khlebnikov > > Proper testing shows yet another problem in balloon migration: it works only > once for each page. balloon_page_movable() check page flags and page_count. > In __unmap_and_move page is locked, reference counter is elevated, so > balloon_page_movable() _always_ fails here. As result in __unmap_and_move() > migration goes to the normal migration path. > > Balloon ->migratepage() is so special, it returns MIGRATEPAGE_BALLOON_SUCCESS > instead of MIGRATEPAGE_SUCCESS. After that in move_to_new_page() successfully > migrated page got NULL into its mapping pointer and loses connectivity with > balloon and ability for further migration. > > It's safe to use __is_movable_balloon_page here: page is isolated and pinned. > > Signed-off-by: Konstantin Khlebnikov > Cc: stable # v3.8 > --- > include/linux/balloon_compaction.h | 5 +++++ > mm/migrate.c | 2 +- > 2 files changed, 6 insertions(+), 1 deletion(-) > > diff --git a/include/linux/balloon_compaction.h b/include/linux/balloon_compaction.h > index 53d482e..284fc1d 100644 > --- a/include/linux/balloon_compaction.h > +++ b/include/linux/balloon_compaction.h > @@ -258,6 +258,11 @@ static inline void balloon_page_delete(struct page *page) > list_del(&page->lru); > } > > +static inline bool __is_movable_balloon_page(struct page *page) > +{ > + return false; > +} > + > static inline bool balloon_page_movable(struct page *page) > { > return false; > diff --git a/mm/migrate.c b/mm/migrate.c > index 905b1aa..57c94f9 100644 > --- a/mm/migrate.c > +++ b/mm/migrate.c > @@ -873,7 +873,7 @@ static int __unmap_and_move(struct page *page, struct page *newpage, > } > } > > - if (unlikely(balloon_page_movable(page))) { > + if (unlikely(__is_movable_balloon_page(page))) { > /* > * A ballooned page does not need any special attention from > * physical to virtual reverse mapping procedures. > Acked-by: Rafael Aquini From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-vc0-f173.google.com (mail-vc0-f173.google.com [209.85.220.173]) by kanga.kvack.org (Postfix) with ESMTP id D1DDC6B0036 for ; Tue, 2 Sep 2014 08:31:47 -0400 (EDT) Received: by mail-vc0-f173.google.com with SMTP id im17so6868680vcb.32 for ; Tue, 02 Sep 2014 05:31:47 -0700 (PDT) Received: from mx1.redhat.com (mx1.redhat.com. [209.132.183.28]) by mx.google.com with ESMTPS id j7si2175526vcy.50.2014.09.02.05.31.45 for (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Tue, 02 Sep 2014 05:31:46 -0700 (PDT) Date: Tue, 2 Sep 2014 08:31:34 -0400 From: Rafael Aquini Subject: Re: [PATCH v2 2/6] mm/balloon_compaction: keep ballooned pages away from normal migration path Message-ID: <20140902123133.GC14419@t510.redhat.com> References: <20140830163834.29066.98205.stgit@zurg> <20140830164113.29066.25879.stgit@zurg> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20140830164113.29066.25879.stgit@zurg> Sender: owner-linux-mm@kvack.org List-ID: To: Konstantin Khlebnikov Cc: Andrew Morton , Konstantin Khlebnikov , Linux Kernel Mailing List , linux-mm@kvack.org, Andrey Ryabinin , Sasha Levin On Sat, Aug 30, 2014 at 08:41:13PM +0400, Konstantin Khlebnikov wrote: > From: Konstantin Khlebnikov > > Proper testing shows yet another problem in balloon migration: it works only > once for each page. balloon_page_movable() check page flags and page_count. > In __unmap_and_move page is locked, reference counter is elevated, so > balloon_page_movable() _always_ fails here. As result in __unmap_and_move() > migration goes to the normal migration path. > > Balloon ->migratepage() is so special, it returns MIGRATEPAGE_BALLOON_SUCCESS > instead of MIGRATEPAGE_SUCCESS. After that in move_to_new_page() successfully > migrated page got NULL into its mapping pointer and loses connectivity with > balloon and ability for further migration. > > It's safe to use __is_movable_balloon_page here: page is isolated and pinned. > > Signed-off-by: Konstantin Khlebnikov > Cc: stable # v3.8 > --- > include/linux/balloon_compaction.h | 5 +++++ > mm/migrate.c | 2 +- > 2 files changed, 6 insertions(+), 1 deletion(-) > > diff --git a/include/linux/balloon_compaction.h b/include/linux/balloon_compaction.h > index 53d482e..284fc1d 100644 > --- a/include/linux/balloon_compaction.h > +++ b/include/linux/balloon_compaction.h > @@ -258,6 +258,11 @@ static inline void balloon_page_delete(struct page *page) > list_del(&page->lru); > } > > +static inline bool __is_movable_balloon_page(struct page *page) > +{ > + return false; > +} > + > static inline bool balloon_page_movable(struct page *page) > { > return false; > diff --git a/mm/migrate.c b/mm/migrate.c > index 905b1aa..57c94f9 100644 > --- a/mm/migrate.c > +++ b/mm/migrate.c > @@ -873,7 +873,7 @@ static int __unmap_and_move(struct page *page, struct page *newpage, > } > } > > - if (unlikely(balloon_page_movable(page))) { > + if (unlikely(__is_movable_balloon_page(page))) { > /* > * A ballooned page does not need any special attention from > * physical to virtual reverse mapping procedures. > Acked-by: Rafael Aquini -- 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