From: Christoph Hellwig <hch@infradead.org>
To: Dave Chinner <david@fromorbit.com>
Cc: xfs@oss.sgi.com
Subject: Re: [PATCH 8/9] xfs: introduce xfs_buf_submit[_wait]
Date: Fri, 15 Aug 2014 21:55:31 -0700 [thread overview]
Message-ID: <20140816045531.GA7876@infradead.org> (raw)
In-Reply-To: <20140815233743.GX26465@dastard>
On Sat, Aug 16, 2014 at 09:37:43AM +1000, Dave Chinner wrote:
> xfs_zero_remaining_bytes is uses an uncached buffer, so we're not
> using the buffer cache at all for the blocks being zeroed. That is
> why it does the flag twiddling dance it does. However, consolidation
> all the different block zeroing functions we have is an exercise for
> a different day....
Well, we're using buffers. Anyway, below is what I think it should
look like when using buffers. Although I wonder how either the old
or new variant pass the verifier check in _xfs_buf_ioapply for v5
filesystems given that we don't pass any ops in.
diff --git a/fs/xfs/xfs_bmap_util.c b/fs/xfs/xfs_bmap_util.c
index 2f1e30d..c495dce 100644
--- a/fs/xfs/xfs_bmap_util.c
+++ b/fs/xfs/xfs_bmap_util.c
@@ -1122,14 +1122,6 @@ xfs_zero_remaining_bytes(
if (endoff > XFS_ISIZE(ip))
endoff = XFS_ISIZE(ip);
- bp = xfs_buf_get_uncached(XFS_IS_REALTIME_INODE(ip) ?
- mp->m_rtdev_targp : mp->m_ddev_targp,
- BTOBB(mp->m_sb.sb_blocksize), 0);
- if (!bp)
- return -ENOMEM;
-
- xfs_buf_unlock(bp);
-
for (offset = startoff; offset <= endoff; offset = lastoffset + 1) {
uint lock_mode;
@@ -1152,42 +1144,26 @@ xfs_zero_remaining_bytes(
ASSERT(imap.br_startblock != DELAYSTARTBLOCK);
if (imap.br_state == XFS_EXT_UNWRITTEN)
continue;
- XFS_BUF_UNDONE(bp);
- XFS_BUF_UNWRITE(bp);
- XFS_BUF_READ(bp);
- XFS_BUF_SET_ADDR(bp, xfs_fsb_to_db(ip, imap.br_startblock));
- if (XFS_FORCED_SHUTDOWN(mp)) {
- error = -EIO;
- break;
- }
- xfs_buf_iorequest(bp);
- error = xfs_buf_iowait(bp);
- if (error) {
- xfs_buf_ioerror_alert(bp,
- "xfs_zero_remaining_bytes(read)");
- break;
+ bp = xfs_buf_read_uncached(XFS_IS_REALTIME_INODE(ip) ?
+ mp->m_rtdev_targp : mp->m_ddev_targp,
+ xfs_fsb_to_db(ip, imap.br_startblock),
+ BTOBB(mp->m_sb.sb_blocksize),
+ 0, NULL);
+ if (!bp)
+ return -ENOMEM;
+ if (bp->b_error) {
+ error = bp->b_error;
+ xfs_buf_relse(bp);
+ return error;
}
+
memset(bp->b_addr +
(offset - XFS_FSB_TO_B(mp, imap.br_startoff)),
0, lastoffset - offset + 1);
- XFS_BUF_UNDONE(bp);
- XFS_BUF_UNREAD(bp);
- XFS_BUF_WRITE(bp);
- if (XFS_FORCED_SHUTDOWN(mp)) {
- error = -EIO;
- break;
- }
- xfs_buf_iorequest(bp);
- error = xfs_buf_iowait(bp);
- if (error) {
- xfs_buf_ioerror_alert(bp,
- "xfs_zero_remaining_bytes(write)");
- break;
- }
+ xfs_bwrite(bp);
}
- xfs_buf_free(bp);
return error;
}
_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs
next prev parent reply other threads:[~2014-08-16 4:55 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-08-15 6:38 [RFC PATCH 0/9] xfs: clean up xfs_buf io interfaces Dave Chinner
2014-08-15 6:38 ` [PATCH 1/9] xfs: synchronous buffer IO needs a reference Dave Chinner
2014-08-15 13:18 ` Brian Foster
2014-08-15 23:17 ` Dave Chinner
2014-08-18 14:15 ` Brian Foster
2014-08-29 0:18 ` Christoph Hellwig
2014-08-15 6:39 ` [PATCH 2/9] xfs: xfs_buf_ioend and xfs_buf_iodone_work duplicate functionality Dave Chinner
2014-08-15 13:18 ` Brian Foster
2014-08-15 23:21 ` Dave Chinner
2014-08-18 14:15 ` Brian Foster
2014-08-29 0:22 ` Christoph Hellwig
2014-08-29 0:55 ` Dave Chinner
2014-08-15 6:39 ` [PATCH 3/9] xfs: rework xfs_buf_bio_endio error handling Dave Chinner
2014-08-15 13:18 ` Brian Foster
2014-08-15 23:25 ` Dave Chinner
2014-08-29 0:23 ` Christoph Hellwig
2014-08-15 6:39 ` [PATCH 4/9] xfs: kill xfs_bdstrat_cb Dave Chinner
2014-08-29 0:24 ` Christoph Hellwig
2014-08-15 6:39 ` [PATCH 5/9] xfs: xfs_bioerror can die Dave Chinner
2014-08-15 14:35 ` Brian Foster
2014-08-15 23:27 ` Dave Chinner
2014-08-29 0:28 ` Christoph Hellwig
2014-08-29 1:05 ` Dave Chinner
2014-08-15 6:39 ` [PATCH 6/9] xfs: kill xfs_bioerror_relse Dave Chinner
2014-08-29 0:32 ` Christoph Hellwig
2014-08-29 1:12 ` Dave Chinner
2014-08-29 18:26 ` Christoph Hellwig
2014-08-30 0:05 ` Dave Chinner
2014-08-15 6:39 ` [PATCH 7/9] xfs: clean up xfs_trans_buf_read_map Dave Chinner
2014-08-15 6:39 ` [PATCH 8/9] xfs: introduce xfs_buf_submit[_wait] Dave Chinner
2014-08-15 13:10 ` Christoph Hellwig
2014-08-15 23:37 ` Dave Chinner
2014-08-16 4:55 ` Christoph Hellwig [this message]
2014-08-15 14:35 ` Brian Foster
2014-08-15 23:39 ` Dave Chinner
2014-08-18 14:16 ` Brian Foster
2014-08-15 16:13 ` Brian Foster
2014-08-15 23:58 ` Dave Chinner
2014-08-18 14:26 ` Brian Foster
2014-08-15 6:39 ` [PATCH 9/9] xfs: check xfs_buf_read_uncached returns correctly Dave Chinner
2014-08-15 12:56 ` Christoph Hellwig
2014-08-15 23:58 ` Dave Chinner
2014-08-29 0:37 ` 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=20140816045531.GA7876@infradead.org \
--to=hch@infradead.org \
--cc=david@fromorbit.com \
--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