public inbox for linux-xfs@vger.kernel.org
 help / color / mirror / Atom feed
From: Dave Chinner <david@fromorbit.com>
To: xfs@oss.sgi.com
Subject: [PATCH 15/21] xfs: shortform directory offsets change for dir3 format
Date: Tue, 12 Mar 2013 23:30:48 +1100	[thread overview]
Message-ID: <1363091454-8852-16-git-send-email-david@fromorbit.com> (raw)
In-Reply-To: <1363091454-8852-1-git-send-email-david@fromorbit.com>

From: Dave Chinner <dchinner@redhat.com>

Because the header size for the CRC enabled directory blocks is
larger, the offset of the first entry into a directory block is
different to the dir2 format. The shortform directory stores the
dirent's offset so that it doesn't change when moving from shortform
to block form and back again, and hence it needs to take into
account the different header sizes to maintain the correct offsets.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
---
 fs/xfs/xfs_dir2_format.h |   25 ++++++++++++++-----------
 fs/xfs/xfs_dir2_sf.c     |   10 +++++-----
 2 files changed, 19 insertions(+), 16 deletions(-)

diff --git a/fs/xfs/xfs_dir2_format.h b/fs/xfs/xfs_dir2_format.h
index 8bb4efe..8d33f03 100644
--- a/fs/xfs/xfs_dir2_format.h
+++ b/fs/xfs/xfs_dir2_format.h
@@ -228,16 +228,6 @@ xfs_dir2_sf_nextentry(struct xfs_dir2_sf_hdr *hdr,
 	xfs_dir2_byte_to_db(mp, XFS_DIR2_DATA_OFFSET)
 
 /*
- * Offsets of . and .. in data space (always block 0)
- */
-#define	XFS_DIR2_DATA_DOT_OFFSET	\
-	((xfs_dir2_data_aoff_t)sizeof(struct xfs_dir2_data_hdr))
-#define	XFS_DIR2_DATA_DOTDOT_OFFSET	\
-	(XFS_DIR2_DATA_DOT_OFFSET + xfs_dir2_data_entsize(1))
-#define	XFS_DIR2_DATA_FIRST_OFFSET		\
-	(XFS_DIR2_DATA_DOTDOT_OFFSET + xfs_dir2_data_entsize(2))
-
-/*
  * Describe a free area in the data block.
  *
  * The freespace will be formatted as a xfs_dir2_data_unused_t.
@@ -378,7 +368,20 @@ xfs_dir3_data_unused_p(struct xfs_dir2_data_hdr *hdr)
 
 /*
  * Offsets of . and .. in data space (always block 0)
- */
+ *
+ * The macros are used for shortform directories as they have no headers to read
+ * the magic number out of. Shortform directories need to know the size of the
+ * data block header because the sfe embeds the block offset of the entry into
+ * it so that it doesn't change when format conversion occurs. Bad Things Happen
+ * if we don't follow this rule.
+ */
+#define	XFS_DIR3_DATA_DOT_OFFSET(mp)	\
+	xfs_dir3_data_hdr_size(xfs_sb_version_hascrc(&(mp)->m_sb))
+#define	XFS_DIR3_DATA_DOTDOT_OFFSET(mp)	\
+	(XFS_DIR3_DATA_DOT_OFFSET(mp) + xfs_dir2_data_entsize(1))
+#define	XFS_DIR3_DATA_FIRST_OFFSET(mp)		\
+	(XFS_DIR3_DATA_DOTDOT_OFFSET(mp) + xfs_dir2_data_entsize(2))
+
 static inline xfs_dir2_data_aoff_t
 xfs_dir3_data_dot_offset(struct xfs_dir2_data_hdr *hdr)
 {
diff --git a/fs/xfs/xfs_dir2_sf.c b/fs/xfs/xfs_dir2_sf.c
index b975622..6157424 100644
--- a/fs/xfs/xfs_dir2_sf.c
+++ b/fs/xfs/xfs_dir2_sf.c
@@ -535,7 +535,7 @@ xfs_dir2_sf_addname_hard(
 	 * to insert the new entry.
 	 * If it's going to end up at the end then oldsfep will point there.
 	 */
-	for (offset = XFS_DIR2_DATA_FIRST_OFFSET,
+	for (offset = XFS_DIR3_DATA_FIRST_OFFSET(dp->i_mount),
 	      oldsfep = xfs_dir2_sf_firstentry(oldsfp),
 	      add_datasize = xfs_dir2_data_entsize(args->namelen),
 	      eof = (char *)oldsfep == &buf[old_isize];
@@ -617,7 +617,7 @@ xfs_dir2_sf_addname_pick(
 
 	sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
 	size = xfs_dir2_data_entsize(args->namelen);
-	offset = XFS_DIR2_DATA_FIRST_OFFSET;
+	offset = XFS_DIR3_DATA_FIRST_OFFSET(mp);
 	sfep = xfs_dir2_sf_firstentry(sfp);
 	holefit = 0;
 	/*
@@ -688,7 +688,7 @@ xfs_dir2_sf_check(
 	dp = args->dp;
 
 	sfp = (xfs_dir2_sf_hdr_t *)dp->i_df.if_u1.if_data;
-	offset = XFS_DIR2_DATA_FIRST_OFFSET;
+	offset = XFS_DIR3_DATA_FIRST_OFFSET(dp->i_mount);
 	ino = xfs_dir2_sf_get_parent_ino(sfp);
 	i8count = ino > XFS_DIR2_MAX_SHORT_INUM;
 
@@ -812,9 +812,9 @@ xfs_dir2_sf_getdents(
 	 * mp->m_dirdatablk.
 	 */
 	dot_offset = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk,
-					     XFS_DIR2_DATA_DOT_OFFSET);
+					     XFS_DIR3_DATA_DOT_OFFSET(mp));
 	dotdot_offset = xfs_dir2_db_off_to_dataptr(mp, mp->m_dirdatablk,
-						XFS_DIR2_DATA_DOTDOT_OFFSET);
+						XFS_DIR3_DATA_DOTDOT_OFFSET(mp));
 
 	/*
 	 * Put . entry unless we're starting past it.
-- 
1.7.10.4

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

  parent reply	other threads:[~2013-03-12 12:31 UTC|newest]

Thread overview: 51+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-03-12 12:30 [PATCH 00/21] xfs: metadata CRCs, third version Dave Chinner
2013-03-12 12:30 ` [PATCH 01/21] xfs: ensure we capture IO errors correctly Dave Chinner
2013-03-12 12:30 ` [PATCH 02/21] xfs: increase hexdump output in xfs_corruption_error Dave Chinner
2013-03-14 21:18   ` Ben Myers
2013-03-15  1:13     ` Dave Chinner
2013-03-12 12:30 ` [PATCH 03/21] xfs: take inode version into account in XFS_LITINO Dave Chinner
2013-03-12 12:30 ` [PATCH 04/21] xfs: add support for large btree blocks Dave Chinner
2013-03-12 12:30 ` [PATCH 05/21] xfs: add CRC checks to the AGF Dave Chinner
2013-03-12 12:30 ` [PATCH 06/21] xfs: add CRC checks to the AGFL Dave Chinner
2013-03-12 12:30 ` [PATCH 07/21] xfs: add CRC checks to the AGI Dave Chinner
2013-03-12 12:30 ` [PATCH 08/21] xfs: add CRC checks for quota blocks Dave Chinner
2013-03-12 12:30 ` [PATCH 09/21] xfs: add version 3 inode format with CRCs Dave Chinner
2013-03-14 16:03   ` Ben Myers
2013-03-14 19:01     ` Ben Myers
2013-03-15  1:11     ` Dave Chinner
2013-03-26 22:56       ` Dave Chinner
2013-03-27  0:53         ` Ben Myers
2013-03-27  1:48           ` Dave Chinner
2013-04-02 22:44             ` Ben Myers
2013-04-03  4:08               ` Dave Chinner
2013-04-02 22:49   ` Ben Myers
2013-03-12 12:30 ` [PATCH 10/21] xfs: add CRC checks to remote symlinks Dave Chinner
2013-03-20 21:14   ` Ben Myers
2013-03-21  1:22     ` Dave Chinner
2013-03-21 14:59       ` Ben Myers
2013-03-20 22:03   ` Ben Myers
2013-03-21  1:32     ` Dave Chinner
2013-03-12 12:30 ` [PATCH 11/21] xfs: add CRC checks to block format directory blocks Dave Chinner
2013-03-26 18:39   ` Ben Myers
2013-03-26 21:40     ` Dave Chinner
2013-03-12 12:30 ` [PATCH 12/21] xfs: add CRC checking to dir2 free blocks Dave Chinner
2013-03-28 23:40   ` Ben Myers
2013-03-29  3:13     ` Dave Chinner
2013-03-12 12:30 ` [PATCH 13/21] xfs: add CRC checking to dir2 data blocks Dave Chinner
2013-04-03 22:13   ` Ben Myers
2013-03-12 12:30 ` [PATCH 14/21] xfs: add CRC checking to dir2 leaf blocks Dave Chinner
2013-03-12 12:30 ` Dave Chinner [this message]
2013-03-12 12:30 ` [PATCH 16/21] xfs: add CRCs to dir2/da node blocks Dave Chinner
2013-03-12 12:30 ` [PATCH 17/21] xfs: add CRCs to attr leaf blocks Dave Chinner
2013-03-12 12:30 ` [PATCH 18/21] xfs: split remote attribute code out Dave Chinner
2013-03-12 12:30 ` [PATCH 19/21] xfs: add CRC protection to remote attributes Dave Chinner
2013-03-12 12:30 ` [PATCH 20/21] xfs: add buffer types to directory and attribute buffers Dave Chinner
2013-03-12 12:30 ` [PATCH 21/21] xfs: add CRC checks to the superblock Dave Chinner
2013-03-26 20:58   ` Chandra Seetharaman
2013-03-27  1:06     ` Dave Chinner
2013-03-27 23:07       ` Chandra Seetharaman
2013-03-28  1:36         ` Dave Chinner
2013-03-12 12:43 ` [PATCH 22/21] xfs: Fix magic number assert in xfs_dir3_leaf_log_bests Dave Chinner
2013-03-13  0:29 ` [PATCH 23/21] xfs: fix endian issues reported by sparse Dave Chinner
2013-03-13  1:34 ` [PATCH 24/21] xfs: buffer type overruns blf_flags field Dave Chinner
2013-03-14 21:41 ` [PATCH 00/21] xfs: metadata CRCs, third version Ben Myers

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=1363091454-8852-16-git-send-email-david@fromorbit.com \
    --to=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