From: Dave Chinner <david@fromorbit.com>
To: Christoph Hellwig <hch@infradead.org>
Cc: xfs@oss.sgi.com
Subject: Re: [PATCH 03/19] xfs: uncached buffer reads need to return an error
Date: Fri, 12 Oct 2012 13:28:29 +1100 [thread overview]
Message-ID: <20121012022829.GL2739@dastard> (raw)
In-Reply-To: <20121011221137.GJ2739@dastard>
On Fri, Oct 12, 2012 at 09:11:37AM +1100, Dave Chinner wrote:
> On Thu, Oct 11, 2012 at 05:38:02PM -0400, Christoph Hellwig wrote:
> > > index 917e121..dee14eb 100644
> > > --- a/fs/xfs/xfs_fsops.c
> > > +++ b/fs/xfs/xfs_fsops.c
> > > @@ -149,6 +149,11 @@ xfs_growfs_data_private(
> > > XFS_FSS_TO_BB(mp, 1), 0, NULL);
> > > if (!bp)
> > > return EIO;
> > > + if (bp->b_error) {
> > > + int error = bp->b_error;
> > > + xfs_buf_relse(bp);
> > > + return error;
> > > + }
> > > xfs_buf_relse(bp);
> >
> > > + if (bp->b_error) {
> > > + error = bp->b_error;
> > > + if (loud)
> > > + xfs_warn(mp, "SB validate failed");
> > > + goto release_buf;
> > > + }
> >
> > > + if (bp->b_error) {
> > > + error = bp->b_error;
> > > + xfs_buf_relse(bp);
> > > + return error;
> > > + }
> >
> > > + if (!bp || bp->b_error) {
> > > xfs_warn(mp, "realtime device size check failed");
> > > + if (bp)
> > > + xfs_buf_relse(bp);
> > > return EIO;
> > > }
> > > xfs_buf_relse(bp);
> >
> > It seems like all these callers would be a lot cleaner if we'd just
> > return the error as the return value, and a buffer as an indirect
> > pointer if and only if the read succeeded.
>
> The number of callers is relatively small, and the knock-on effect
> through the subsequent patches isn't that bad, so changing the
> interface is probably the right thing to do here. I'll rework this
> patch appropriately.
Then again, maybe I won't. I just remembered the main reason for
returning a buffer with an error set on it. That is that there are
circumstances in which we might want to attempt repair of a buffer
that failed verification, and we cannot do that without returning
the buffer.
So even if we return the buffer as an indirect pointer, we'd still
need to return it when particular types of errors are detected on
the buffer (i.e. EFSCORRUPTED from the verifier). Hence the error
handling in xfs_buf_read_uncached() would become more complex and
difficult to get right (e.g. different EIO vs EFSCORRUPTED vs ...
handling) and it wouldn't simplify the error handling at the call
site, either.
Cheers,
Dave.
--
Dave Chinner
david@fromorbit.com
_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs
next prev parent reply other threads:[~2012-10-12 2:27 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-10-09 3:50 [PATCH 00/19] xfs: buffer read verifier infrastructure Dave Chinner
2012-10-09 3:50 ` [PATCH 01/19] xfs: growfs: don't read garbage for new secondary superblocks Dave Chinner
2012-10-11 21:34 ` Christoph Hellwig
2012-10-09 3:50 ` [PATCH 02/19] xfs: make buffer read verication an IO completion function Dave Chinner
2012-10-11 21:36 ` Christoph Hellwig
2012-10-09 3:50 ` [PATCH 03/19] xfs: uncached buffer reads need to return an error Dave Chinner
2012-10-11 21:38 ` Christoph Hellwig
2012-10-11 22:11 ` Dave Chinner
2012-10-12 2:28 ` Dave Chinner [this message]
2012-10-09 3:50 ` [PATCH 04/19] xfs: verify superblocks as they are read from disk Dave Chinner
2012-10-11 21:41 ` Christoph Hellwig
2012-10-11 22:28 ` Dave Chinner
2012-10-09 3:50 ` [PATCH 05/19] xfs: verify AGF blocks " Dave Chinner
2012-10-11 21:42 ` Christoph Hellwig
2012-10-09 3:50 ` [PATCH 06/19] xfs: verify AGI " Dave Chinner
2012-10-11 21:43 ` Christoph Hellwig
2012-10-09 3:50 ` [PATCH 07/19] xfs: verify AGFL " Dave Chinner
2012-10-11 21:44 ` Christoph Hellwig
2012-10-11 21:52 ` Dave Chinner
2012-10-09 3:50 ` [PATCH 08/19] xfs: verify inode buffers " Dave Chinner
2012-10-11 21:45 ` Christoph Hellwig
2012-10-11 21:55 ` Dave Chinner
2012-10-09 3:51 ` [PATCH 09/19] xfs: verify btree blocks " Dave Chinner
2012-10-09 3:51 ` [PATCH 10/19] xfs: verify dquot " Dave Chinner
2012-10-11 21:48 ` Christoph Hellwig
2012-10-11 22:08 ` Dave Chinner
2012-10-09 3:51 ` [PATCH 11/19] xfs: add verifier callback to directorry read code Dave Chinner
2012-10-11 21:48 ` Christoph Hellwig
2012-10-09 3:51 ` [PATCH 12/19] xfs: factor dir2 block read operations Dave Chinner
2012-10-09 3:51 ` [PATCH 13/19] xfs: verify dir2 block format buffers Dave Chinner
2012-10-09 3:51 ` [PATCH 14/19] xfs: factor dir2 free block reading Dave Chinner
2012-10-09 3:51 ` [PATCH 15/19] xfs: factor out dir2 data " Dave Chinner
2012-10-09 3:51 ` [PATCH 16/19] xfs: factor dir2 leaf read Dave Chinner
2012-10-09 3:51 ` [PATCH 17/19] xfs: factor and verify attr leaf reads Dave Chinner
2012-10-09 3:51 ` [PATCH 18/19] xfs: add xfs_da_node verification Dave Chinner
2012-10-09 3:51 ` [PATCH 19/19] xfs: Add verifiers to dir2 data readahead Dave Chinner
2012-10-11 12:09 ` [PATCH 00/19] xfs: buffer read verifier infrastructure Mark Tinguely
2012-10-11 21:42 ` Dave Chinner
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=20121012022829.GL2739@dastard \
--to=david@fromorbit.com \
--cc=hch@infradead.org \
--cc=xfs@oss.sgi.com \
/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