From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from aserp2120.oracle.com ([141.146.126.78]:33000 "EHLO aserp2120.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751658AbeCVRto (ORCPT ); Thu, 22 Mar 2018 13:49:44 -0400 Date: Thu, 22 Mar 2018 10:49:38 -0700 From: "Darrick J. Wong" Subject: Re: [PATCH 11/9] xfs: flag inode corruption if parent ptr doesn't get us a real inode Message-ID: <20180322174938.GY4818@magnolia> References: <152107377037.19571.8618901963505842632.stgit@magnolia> <20180321032158.GR1757@magnolia> <20180322143435.GG16617@bfoster.bfoster> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20180322143435.GG16617@bfoster.bfoster> Sender: linux-xfs-owner@vger.kernel.org List-ID: List-Id: xfs To: Brian Foster Cc: linux-xfs@vger.kernel.org On Thu, Mar 22, 2018 at 10:34:35AM -0400, Brian Foster wrote: > On Tue, Mar 20, 2018 at 08:21:58PM -0700, Darrick J. Wong wrote: > > From: Darrick J. Wong > > > > If a directory's parent inode pointer doesn't point to an inode, flag > > the directory as corrupt. Enable IGET_UNTRUSTED here so that we > > can confirm with the inobt that the inode is present and allocated. > > > > Signed-off-by: Darrick J. Wong > > --- > > fs/xfs/scrub/parent.c | 9 +++++++-- > > 1 file changed, 7 insertions(+), 2 deletions(-) > > > > diff --git a/fs/xfs/scrub/parent.c b/fs/xfs/scrub/parent.c > > index 0d38514..d1547ca 100644 > > --- a/fs/xfs/scrub/parent.c > > +++ b/fs/xfs/scrub/parent.c > > @@ -167,9 +167,14 @@ xfs_scrub_parent_validate( > > * if the parent pointer erroneously points to a file, we > > * can't use DONTCACHE here because DONTCACHE inodes can trigger > > * immediate inactive cleanup of the inode. > > + * > > + * If _iget returns -EINVAL then the parent inode number is garbage > > + * and the directory is corrupt. > > */ > > - error = xfs_iget(mp, sc->tp, dnum, 0, 0, &dp); > > - if (!xfs_scrub_fblock_xref_process_error(sc, XFS_DATA_FORK, 0, &error)) > > + error = xfs_iget(mp, sc->tp, dnum, XFS_IGET_UNTRUSTED, 0, &dp); > > + if (error == -EINVAL) > > + error = -EFSCORRUPTED; > > + if (!xfs_scrub_fblock_process_error(sc, XFS_DATA_FORK, 0, &error)) > > Ok, but it looks like we're also changing the error semantics here > (xref_process_error() -> process_error()). Could you comment on that in > the commit log as well? Ok, though it occurs to me that iget returning EFSCORRUPTED is still an xref error, not a corruption error, so this really ought to be: error = xfs_iget(mp, sc->tp, dnum, XFS_IGET_UNTRUSTED, 0, &dp); if (error == -EINVAL) { error = -EFSCORRUPTED; xfs_scrub_fblock_process_error(sc, XFS_DATA_FORK, 0, &error); goto out; } if (!xfs_scrub_fblock_xref_process_error(sc, XFS_DATA_FORK, 0, &error)) goto out; So I'll fix the code + commit message and send out a new patch. Thanks for reviewing! --D > With that, looks fine: > > Reviewed-by: Brian Foster > > > goto out; > > if (dp == sc->ip || !S_ISDIR(VFS_I(dp)->i_mode)) { > > xfs_scrub_fblock_set_corrupt(sc, XFS_DATA_FORK, 0); > > -- > > To unsubscribe from this list: send the line "unsubscribe linux-xfs" in > > the body of a message to majordomo@vger.kernel.org > > More majordomo info at http://vger.kernel.org/majordomo-info.html > -- > To unsubscribe from this list: send the line "unsubscribe linux-xfs" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html