From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jaegeuk Kim Subject: Re: [PATCH] f2fs: fix wrong sum_page pointer in f2fs_gc Date: Sun, 23 Oct 2016 14:40:35 -0700 Message-ID: <20161023214035.GA63620@jaegeuk> References: <20161012232304.6667-1-jaegeuk@kernel.org> <20161013171421.GA6363@jaegeuk> <3ed5607f-0fe6-02cb-8c95-782913e65b5b@kernel.org> <20161014165700.GA67182@jaegeuk> <3a96264b-b187-a5ce-2279-f9edd3748838@kernel.org> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Received: from sog-mx-4.v43.ch3.sourceforge.com ([172.29.43.194] helo=mx.sourceforge.net) by sfs-ml-2.v29.ch3.sourceforge.com with esmtp (Exim 4.76) (envelope-from ) id 1byQVX-0002RV-PA for linux-f2fs-devel@lists.sourceforge.net; Sun, 23 Oct 2016 21:40:43 +0000 Received: from mail.kernel.org ([198.145.29.136]) by sog-mx-4.v43.ch3.sourceforge.com with esmtps (TLSv1:AES256-SHA:256) (Exim 4.76) id 1byQVW-0006EX-Oh for linux-f2fs-devel@lists.sourceforge.net; Sun, 23 Oct 2016 21:40:43 +0000 Content-Disposition: inline In-Reply-To: <3a96264b-b187-a5ce-2279-f9edd3748838@kernel.org> List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: linux-f2fs-devel-bounces@lists.sourceforge.net To: Chao Yu Cc: linux-f2fs-devel@lists.sourceforge.net On Sun, Oct 23, 2016 at 08:26:15PM +0800, Chao Yu wrote: > On 2016/10/15 0:57, Jaegeuk Kim wrote: > > On Fri, Oct 14, 2016 at 09:25:59PM +0800, Chao Yu wrote: > >> On 2016/10/14 1:14, Jaegeuk Kim wrote: > >>> On Thu, Oct 13, 2016 at 06:33:19PM +0800, Chao Yu wrote: > >>>> Hi Jaegeuk, > >>>> > >>>> On 2016/10/13 7:23, Jaegeuk Kim wrote: > >>>>> This patch fixes using a wrong pointer for sum_page in f2fs_gc. > >>>>> > >>>>> Signed-off-by: Jaegeuk Kim > >>>>> --- > >>>>> fs/f2fs/gc.c | 10 +++++----- > >>>>> 1 file changed, 5 insertions(+), 5 deletions(-) > >>>>> > >>>>> diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c > >>>>> index e48142f..9c18917 100644 > >>>>> --- a/fs/f2fs/gc.c > >>>>> +++ b/fs/f2fs/gc.c > >>>>> @@ -854,16 +854,16 @@ static int do_garbage_collect(struct f2fs_sb_info *sbi, > >>>>> > >>>>> for (segno = start_segno; segno < end_segno; segno++) { > >>>>> > >>>>> - if (get_valid_blocks(sbi, segno, 1) == 0 || > >>>>> - unlikely(f2fs_cp_error(sbi))) > >>>>> - goto next; > >>>>> - > >>>>> /* find segment summary of victim */ > >>>>> sum_page = find_get_page(META_MAPPING(sbi), > >>>>> GET_SUM_BLOCK(sbi, segno)); > >>>>> - f2fs_bug_on(sbi, !PageUptodate(sum_page)); > >>>>> f2fs_put_page(sum_page, 0); > >>>>> > >>>>> + if (get_valid_blocks(sbi, segno, 1) == 0 || > >>>>> + !PageUptodate(sum_page) || > >>>> > >>>> Why uptodate flag of summary page can be cleared? someone truncates it? > >>> > >>> Well, it looks like no problem to remove this, since it will hit > >>> f2fs_cp_error(). I just intended to handle the above f2fs_bug_on here. > >> > >> If summary page becomes non-uptodate, it seems there is a bug in vfs/mm/f2fs, > >> why not keep it there to detect bug? > > > > I found that the above get_sum_page() loop can give a page which is not uptodate > > caused by EIO. Then, we would get the above bug_on, if we move f2fs_cp_error() > > case like this patch. > > > > Thoughts? > > Seems this patch has already been merged into linus' tree... Yup. > Anyway, I can understand why you remove f2fs_bug_on & add PageUptodate here, but > why should we repositon below check? > > if (get_valid_blocks(sbi, segno, 1) == 0 || > unlikely(f2fs_cp_error(sbi))) Given description, this fixes using wrong sum_page pointer by: next: f2fs_put_page(sum_page, 0); Thanks, ------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, SlashDot.org! http://sdm.link/slashdot