From: Dave Chinner <david@fromorbit.com>
To: Christoph Hellwig <hch@infradead.org>
Cc: linux-xfs@vger.kernel.org, chandanbabu@kernel.org
Subject: Re: [PATCH 3/4] xfs: handle allocation failure in xfs_dquot_disk_alloc()
Date: Thu, 4 Apr 2024 08:49:13 +1100 [thread overview]
Message-ID: <Zg3O2d8GozDujblD@dread.disaster.area> (raw)
In-Reply-To: <Zg1ieA0NA9Bd_i3P@infradead.org>
On Wed, Apr 03, 2024 at 07:06:48AM -0700, Christoph Hellwig wrote:
> On Wed, Apr 03, 2024 at 08:38:18AM +1100, Dave Chinner wrote:
> > @@ -356,6 +356,23 @@ xfs_dquot_disk_alloc(
> > if (error)
> > goto err_cancel;
> >
> > + if (nmaps == 0) {
> > + /*
> > + * Unexpected ENOSPC - the transaction reservation should have
> > + * guaranteed that this allocation will succeed. We don't know
> > + * why this happened, so just back out gracefully.
>
> So looking at this code, xfs_dquot_disk_alloc allocates it's own
> transaction, and does so without a space reservation.
It does have a valid space reservation:
error = xfs_trans_alloc(mp, &M_RES(mp)->tr_qm_dqalloc,
XFS_QM_DQALLOC_SPACE_RES(mp), 0, 0, &tp);
The space reservation is XFS_QM_DQALLOC_SPACE_RES(mp):
#define XFS_QM_DQALLOC_SPACE_RES(mp) \
(XFS_EXTENTADD_SPACE_RES(mp, XFS_DATA_FORK) + \
XFS_DQUOT_CLUSTER_SIZE_FSB)
which is correct for allocating a single XFS_DQUOT_CLUSTER_SIZE_FSB
sized extent.
> In other words:
> an ENOSPC is entirely expected here in the current form.
I disagree with that assessment. :)
> The code, just
> like many other callers of xfs_bmapi_write, just fails to handle
> the weird 0 return value and zero nmaps convention properly.
Yes, that's exactly what this patch is fixing regardless of the
cause of the failure. It's the right thing to do - error handling by
assumption (i.e. ASSERT()) is simply poor code....
-Dave.
--
Dave Chinner
david@fromorbit.com
next prev parent reply other threads:[~2024-04-03 21:49 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-04-02 21:38 [PATCH 0/4] xfs: fixes for 6.9-rcX Dave Chinner
2024-04-02 21:38 ` [PATCH 1/4] xfs: use kvmalloc for xattr buffers Dave Chinner
2024-04-03 3:43 ` Darrick J. Wong
2024-04-03 4:39 ` Christoph Hellwig
2024-04-03 6:16 ` Dave Chinner
2024-04-03 6:19 ` Christoph Hellwig
2024-04-17 14:35 ` Pankaj Raghav (Samsung)
2024-04-02 21:38 ` [PATCH 2/4] xfs: xfs_alloc_file_space() fails to detect ENOSPC Dave Chinner
2024-04-03 3:46 ` Darrick J. Wong
2024-04-03 4:40 ` Christoph Hellwig
2024-04-03 6:34 ` Dave Chinner
2024-04-03 18:23 ` Christoph Hellwig
2024-04-02 21:38 ` [PATCH 3/4] xfs: handle allocation failure in xfs_dquot_disk_alloc() Dave Chinner
2024-04-03 3:48 ` Darrick J. Wong
2024-04-03 4:41 ` Christoph Hellwig
2024-04-03 4:54 ` Darrick J. Wong
2024-04-03 4:56 ` Christoph Hellwig
2024-04-03 5:04 ` Darrick J. Wong
2024-04-03 6:41 ` Dave Chinner
2024-04-03 14:06 ` Christoph Hellwig
2024-04-03 21:49 ` Dave Chinner [this message]
2024-04-02 21:38 ` [PATCH 4/4] xfs: validate block count for XFS_IOC_SET_RESBLKS Dave Chinner
2024-04-03 3:53 ` Darrick J. Wong
2024-04-03 6:55 ` Dave Chinner
2024-04-03 4:43 ` 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=Zg3O2d8GozDujblD@dread.disaster.area \
--to=david@fromorbit.com \
--cc=chandanbabu@kernel.org \
--cc=hch@infradead.org \
--cc=linux-xfs@vger.kernel.org \
/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;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).