public inbox for linux-xfs@vger.kernel.org
 help / color / mirror / Atom feed
From: "Darrick J. Wong" <darrick.wong@oracle.com>
To: Christoph Hellwig <hch@infradead.org>
Cc: sandeen@sandeen.net, linux-xfs@vger.kernel.org
Subject: Re: [PATCH 01/16] xfs_repair: fix missing dir buffer corruption checks
Date: Mon, 11 May 2020 10:36:03 -0700	[thread overview]
Message-ID: <20200511173603.GB6714@magnolia> (raw)
In-Reply-To: <20200511164421.GA6714@magnolia>

On Mon, May 11, 2020 at 09:44:21AM -0700, Darrick J. Wong wrote:
> On Sat, May 09, 2020 at 10:08:50AM -0700, Christoph Hellwig wrote:
> > On Sat, May 09, 2020 at 09:29:58AM -0700, Darrick J. Wong wrote:
> > > @@ -140,11 +140,24 @@ _("can't read %s block %u for inode %" PRIu64 "\n"),
> > >  		if (whichfork == XFS_DATA_FORK &&
> > >  		    (nodehdr.magic == XFS_DIR2_LEAFN_MAGIC ||
> > >  		    nodehdr.magic == XFS_DIR3_LEAFN_MAGIC)) {
> > > +			int bad = 0;
> > > +
> > >  			if (i != -1) {
> > >  				do_warn(
> > >  _("found non-root LEAFN node in inode %" PRIu64 " bno = %u\n"),
> > >  					da_cursor->ino, bno);
> > > +				bad++;
> > >  			}
> > > +
> > > +			/* corrupt leafn node; rebuild the dir. */
> > > +			if (!bad &&
> > > +			    (bp->b_error == -EFSBADCRC ||
> > > +			     bp->b_error == -EFSCORRUPTED)) {
> > > +				do_warn(
> > > +_("corrupt %s LEAFN block %u for inode %" PRIu64 "\n"),
> > > +					FORKNAME(whichfork), bno, da_cursor->ino);
> > > +			}
> > > +
> > 
> > So this doesn't really change any return value, but just the error
> > message.  But looking at this code I wonder why we don't check
> > b_error first thing after reading the buffer, as checking the magic
> > for a corrupt buffer seems a little pointless.
> 
> <shrug> In the first hunk I was merely following what we did for DA_NODE
> blocks (check magic, then check for corruption errors) but I guess I
> could just pull that up in the file.  I'll have a look and see what
> happens if I do that.

...and on re-examination, the other da_read_buf callers in repair/dir2.c
ought to look for EFSCORRUPTED (they already look for EFSBADCRC) and
complain about that.  Seeing as the directory salvage is done separately
in phase6.c, I guess there's no harm in jumping out early in phases 3/4.

--D

> --D

  reply	other threads:[~2020-05-11 17:36 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-05-09 16:29 [PATCH 00/16] xfs_repair: catch things that xfs_check misses Darrick J. Wong
2020-05-09 16:29 ` [PATCH 01/16] xfs_repair: fix missing dir buffer corruption checks Darrick J. Wong
2020-05-09 17:08   ` Christoph Hellwig
2020-05-11 16:44     ` Darrick J. Wong
2020-05-11 17:36       ` Darrick J. Wong [this message]
2020-05-12 17:29   ` [PATCH v2 " Darrick J. Wong
2020-05-13  6:22     ` Christoph Hellwig
2020-05-13 15:35       ` Darrick J. Wong
2020-05-09 16:30 ` [PATCH 02/16] xfs_repair: warn when we would have rebuilt a directory Darrick J. Wong
2020-05-09 17:09   ` Christoph Hellwig
2020-05-09 16:30 ` [PATCH 03/16] xfs_repair: check for AG btree records that would wrap around Darrick J. Wong
2020-05-09 17:09   ` Christoph Hellwig
2020-05-09 16:30 ` [PATCH 04/16] xfs_repair: fix bnobt and refcountbt record order checks Darrick J. Wong
2020-05-09 17:10   ` Christoph Hellwig
2020-05-09 16:30 ` [PATCH 05/16] xfs_repair: check for out-of-order inobt records Darrick J. Wong
2020-05-09 17:11   ` Christoph Hellwig
2020-05-09 16:30 ` [PATCH 06/16] xfs_repair: fix rmapbt record order check Darrick J. Wong
2020-05-10  7:33   ` Christoph Hellwig
2020-05-09 16:30 ` [PATCH 07/16] xfs_repair: tag inobt vs finobt errors properly Darrick J. Wong
2020-05-09 17:14   ` Christoph Hellwig
2020-05-09 16:30 ` [PATCH 08/16] xfs_repair: complain about bad interior btree pointers Darrick J. Wong
2020-05-09 17:15   ` Christoph Hellwig
2020-05-09 16:30 ` [PATCH 09/16] xfs_repair: convert to libxfs_verify_agbno Darrick J. Wong
2020-05-09 17:18   ` Christoph Hellwig
2020-05-11 16:22     ` Darrick J. Wong
2020-05-12  8:07       ` Christoph Hellwig
2020-05-09 16:31 ` [PATCH 10/16] xfs_repair: refactor verify_dfsbno_range Darrick J. Wong
2020-05-10  7:24   ` Christoph Hellwig
2020-05-09 16:31 ` [PATCH 11/16] xfs_repair: remove verify_dfsbno Darrick J. Wong
2020-05-10  7:24   ` Christoph Hellwig
2020-05-09 16:31 ` [PATCH 12/16] xfs_repair: remove verify_aginum Darrick J. Wong
2020-05-10  7:25   ` Christoph Hellwig
2020-05-09 16:31 ` [PATCH 13/16] xfs_repair: mark entire free space btree record as free1 Darrick J. Wong
2020-05-10  7:26   ` Christoph Hellwig
2020-05-09 16:31 ` [PATCH 14/16] xfs_repair: complain about free space only seen by one btree Darrick J. Wong
2020-05-10  7:26   ` Christoph Hellwig
2020-05-09 16:31 ` [PATCH 15/16] xfs_repair: complain about extents in unknown state Darrick J. Wong
2020-05-10  7:27   ` Christoph Hellwig
2020-05-09 16:31 ` [PATCH 16/16] xfs_repair: complain about any nonzero inprogress value, not just 1 Darrick J. Wong
2020-05-10  7:27   ` Christoph Hellwig

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20200511173603.GB6714@magnolia \
    --to=darrick.wong@oracle.com \
    --cc=hch@infradead.org \
    --cc=linux-xfs@vger.kernel.org \
    --cc=sandeen@sandeen.net \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox