From: Christoph Hellwig <hch@infradead.org>
To: Dave Chinner <david@fromorbit.com>
Cc: xfs@oss.sgi.com
Subject: Re: [PATCH 26/30] xfs_db: avoid libxfs buffer lookup warnings
Date: Mon, 4 Nov 2013 01:12:45 -0800 [thread overview]
Message-ID: <20131104091245.GF23564@infradead.org> (raw)
In-Reply-To: <1383107481-28937-27-git-send-email-david@fromorbit.com>
>
> +/*
> + * We are now using libxfs for our IO backend, so we should always try to use
> + * inode cluster buffers rather than filesystem block sized buffers for reading
> + * inodes. This means that we always use the same buffers as libxfs operations
> + * does, and that avoids buffer cache issues caused by overlapping buffers. This
> + * can be seen clearly when trying to read the root inode. Much of this logic is
> + * similar to libxfs_imap().
> + */
> void
> set_cur_inode(
> xfs_ino_t ino)
> @@ -632,6 +640,9 @@ set_cur_inode(
> xfs_agnumber_t agno;
> xfs_dinode_t *dip;
> int offset;
> + int numblks = blkbb;
> + xfs_agblock_t cluster_agbno;
> +
>
> agno = XFS_INO_TO_AGNO(mp, ino);
> agino = XFS_INO_TO_AGINO(mp, ino);
> @@ -644,6 +655,24 @@ set_cur_inode(
> return;
> }
> cur_agno = agno;
> +
> + if (mp->m_inode_cluster_size > mp->m_sb.sb_blocksize &&
> + mp->m_inoalign_mask) {
> + xfs_agblock_t chunk_agbno;
> + xfs_agblock_t offset_agbno;
> + int blks_per_cluster;
> +
> + blks_per_cluster = mp->m_inode_cluster_size >>
> + mp->m_sb.sb_blocklog;
> + offset_agbno = agbno & mp->m_inoalign_mask;
> + chunk_agbno = agbno - offset_agbno;
> + cluster_agbno = chunk_agbno +
> + ((offset_agbno / blks_per_cluster) * blks_per_cluster);
> + offset += ((agbno - cluster_agbno) * mp->m_sb.sb_inopblock);
> + numblks = XFS_FSB_TO_BB(mp, blks_per_cluster);
> + } else
> + cluster_agbno = agbno;
> +
Seems like this should be a separate patch?
> - if (iocur_top->bp)
> + if (iocur_top->bp) {
> libxfs_putbuf(iocur_top->bp);
> + iocur_top->bp = NULL;
> + }
This should probably be folded into the patch that started using buffers
in xfs_db.
> + int icache_flags; /* cache init flags */
> + int bcache_flags; /* cache init flags */
The icache_flags aren't used anywhere. Also I'd really prefer to have
my patch to kill the inode cache istead of adding more changes to it.
> +#ifdef IO_BCOMPARE_CHECK
> + if (!(libxfs_bcache->c_flags & CACHE_MISCOMPARE_PURGE)) {
> + fprintf(stderr,
> + "%lx: Badness in key lookup (length)\n"
> + "bp=(bno 0x%llx, len %u bytes) key=(bno 0x%llx, len %u bytes)\n",
> + pthread_self(),
> + (unsigned long long)bp->b_bn, (int)bp->b_bcount,
> + (unsigned long long)bkey->blkno,
> + BBTOB(bkey->bblen));
> + }
> #endif
What is the point of the IO_BCOMPARE_CHECK ifdef if we unconditionally
define it?
_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs
next prev parent reply other threads:[~2013-11-04 9:12 UTC|newest]
Thread overview: 76+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-10-30 4:30 [PATCH 00/30 V3] xfsprogs: CRC write support for xfs_db Dave Chinner
2013-10-30 4:30 ` [PATCH 01/30] xfsprogs: fix automatic dependency generation Dave Chinner
2013-10-31 14:44 ` Christoph Hellwig
2013-10-30 4:30 ` [PATCH 02/30] xfs: fix some minor sparse warnings Dave Chinner
2013-10-31 14:44 ` Christoph Hellwig
2013-10-30 4:30 ` [PATCH 03/30] xfs: create a shared header file for format-related information Dave Chinner
2013-10-31 14:45 ` Christoph Hellwig
2013-10-30 4:30 ` [PATCH 04/30] xfs: split dquot buffer operations out Dave Chinner
2013-10-31 14:45 ` Christoph Hellwig
2013-10-30 4:30 ` [PATCH 05/30] xfs: decouple inode and bmap btree header files Dave Chinner
2013-10-31 14:45 ` Christoph Hellwig
2013-10-30 4:30 ` [PATCH 06/30] libxfs: unify xfs_btree.c with kernel code Dave Chinner
2013-10-31 14:46 ` Christoph Hellwig
2013-10-30 4:30 ` [PATCH 07/30] libxfs: bmap btree owner swap support Dave Chinner
2013-10-31 14:47 ` Christoph Hellwig
2013-10-30 4:30 ` [PATCH 08/30] libxfs: xfs_rtalloc.c becomes xfs_rtbitmap.c Dave Chinner
2013-10-31 14:48 ` Christoph Hellwig
2013-10-30 4:31 ` [PATCH 09/30] libxfs: bring across inode buffer readahead verifier changes Dave Chinner
2013-10-31 14:48 ` Christoph Hellwig
2013-10-30 4:31 ` [PATCH 10/30] libxfs: Minor cleanup and bug fix sync Dave Chinner
2013-10-31 14:48 ` Christoph Hellwig
2013-10-30 4:31 ` [PATCH 11/30] xfs: remove newlines from strings passed to __xfs_printk Dave Chinner
2013-10-31 14:49 ` Christoph Hellwig
2013-10-30 4:31 ` [PATCH 12/30] xfs: fix the wrong new_size/rnew_size at xfs_iext_realloc_direct() Dave Chinner
2013-10-31 14:49 ` Christoph Hellwig
2013-10-30 4:31 ` [PATCH 13/30] xfs: fix node forward in xfs_node_toosmall Dave Chinner
2013-10-31 14:49 ` Christoph Hellwig
2013-10-30 4:31 ` [PATCH 14/30] xfs: don't emit corruption noise on fs probes Dave Chinner
2013-10-31 14:49 ` Christoph Hellwig
2013-10-30 4:31 ` [PATCH 15/30] db: separate out straight buffer IO from map based IO Dave Chinner
2013-10-31 14:56 ` Christoph Hellwig
2013-10-31 21:50 ` Dave Chinner
2013-11-01 12:51 ` Christoph Hellwig
2013-10-30 4:31 ` [PATCH 16/30] db: rewrite bbmap to use xfs_buf_map Dave Chinner
2013-10-31 15:06 ` Christoph Hellwig
2013-10-30 4:31 ` [PATCH 17/30] db: rewrite IO engine to use libxfs Dave Chinner
2013-10-31 15:10 ` Christoph Hellwig
2013-10-31 21:59 ` Dave Chinner
2013-10-30 4:31 ` [PATCH 18/30] db: introduce verifier support into set_cur Dave Chinner
2013-11-01 13:07 ` Christoph Hellwig
2013-10-30 4:31 ` [PATCH 19/30] db: indicate if the CRC on a buffer is correct or not Dave Chinner
2013-11-01 13:09 ` Christoph Hellwig
2013-10-30 4:31 ` [PATCH 20/30] db: verify and calculate inode CRCs Dave Chinner
2013-11-04 9:05 ` Christoph Hellwig
2013-11-05 0:18 ` Dave Chinner
2013-10-30 4:31 ` [PATCH 21/30] db: verify and calculate dquot CRCs Dave Chinner
2013-11-04 9:06 ` Christoph Hellwig
2013-11-05 0:25 ` Dave Chinner
2013-10-30 4:31 ` [PATCH 22/30] db: add a special directory buffer verifier Dave Chinner
2013-11-04 9:07 ` Christoph Hellwig
2013-10-30 4:31 ` [PATCH 23/30] db: add a special attribute " Dave Chinner
2013-11-04 9:07 ` Christoph Hellwig
2013-10-30 4:31 ` [PATCH 24/30] db: re-enable write support for v5 filesystems Dave Chinner
2013-11-04 9:07 ` Christoph Hellwig
2013-10-30 4:31 ` [PATCH 25/30] libxfs: fix root inode handling inconsistencies Dave Chinner
2013-10-30 10:23 ` Christoph Hellwig
2013-10-30 21:59 ` Dave Chinner
2013-10-31 4:13 ` [PATCH 25/30 V2] " Dave Chinner
2013-10-31 15:00 ` Christoph Hellwig
2013-10-31 22:04 ` Dave Chinner
2013-11-01 13:03 ` Christoph Hellwig
2013-11-04 23:22 ` Dave Chinner
2013-11-05 9:05 ` Christoph Hellwig
2013-10-30 4:31 ` [PATCH 26/30] xfs_db: avoid libxfs buffer lookup warnings Dave Chinner
2013-11-04 9:12 ` Christoph Hellwig [this message]
2013-11-05 0:52 ` Dave Chinner
2013-11-05 9:06 ` Christoph Hellwig
2013-10-30 4:31 ` [PATCH 27/30] libxfs: work around do_div() not handling 32 bit numerators Dave Chinner
2013-11-04 9:16 ` Christoph Hellwig
2013-11-05 1:55 ` Dave Chinner
2013-10-30 4:31 ` [PATCH 28/30] db: enable metadump on CRC filesystems Dave Chinner
2013-11-04 9:18 ` Christoph Hellwig
2013-10-30 4:31 ` [PATCH 29/30] xfs: support larger inode clusters on v5 filesystems Dave Chinner
2013-11-04 9:18 ` Christoph Hellwig
2013-10-30 4:31 ` [PATCH 30/30] xfsprogs: kill experimental warnings for " Dave Chinner
2013-11-04 9:19 ` 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=20131104091245.GF23564@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