[patch added to 3.12-stable] libxfs: pack the agfl header structure so XFS_AGFL_SIZE is correct

From: "Darrick J. Wong" <redacted>

This patch has been added to the 3.12 stable tree. If you have any
objections, please let us know.

===============

commit 96f859d52bcb1c6ea6f3388d39862bf7143e2f30 upstream.

Because struct xfs_agfl is 36 bytes long and has a 64-bit integer
inside it, gcc will quietly round the structure size up to the nearest
64 bits -- in this case, 40 bytes.  This results in the XFS_AGFL_SIZE
macro returning incorrect results for v5 filesystems on 64-bit
machines (118 items instead of 119).  As a result, a 32-bit xfs_repair
will see garbage in AGFL item 119 and complain.

Therefore, tell gcc not to pad the structure so that the AGFL size
calculation is correct.

Signed-off-by: Darrick J. Wong <redacted>
Reviewed-by: Dave Chinner <redacted>
Signed-off-by: Dave Chinner <david@fromorbit.com>
Signed-off-by: Jiri Slaby <redacted>
---
 fs/xfs/xfs_ag.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/xfs/xfs_ag.h b/fs/xfs/xfs_ag.h
index 1cb740afd674..78ec58fc282e 100644
--- a/fs/xfs/xfs_ag.h
+++ b/fs/xfs/xfs_ag.h

@@ -224,7 +224,7 @@ typedef struct xfs_agfl {
 	__be64		agfl_lsn;
 	__be32		agfl_crc;
 	__be32		agfl_bno[];	/* actually XFS_AGFL_SIZE(mp) */
-} xfs_agfl_t;
+} __attribute__((packed)) xfs_agfl_t;
 
 /*
  * tags for inode radix tree

-- 
2.7.1

From: Dave Chinner <redacted>

This patch has been added to the 3.12 stable tree. If you have any
objections, please let us know.

===============

commit b79f4a1c68bb99152d0785ee4ea3ab4396cdacc6 upstream.

When we do inode readahead in log recovery, we do can do the
readahead before we've replayed the icreate transaction that stamps
the buffer with inode cores. The inode readahead verifier catches
this and marks the buffer as !done to indicate that it doesn't yet
contain valid inodes.

In adding buffer error notification  (i.e. setting b_error = -EIO at
the same time as as we clear the done flag) to such a readahead
verifier failure, we can then get subsequent inode recovery failing
with this error:

XFS (dm-0): metadata I/O error: block 0xa00060 ("xlog_recover_do..(read#2)") error 5 numblks 32

This occurs when readahead completion races with icreate item replay
such as:

	inode readahead
		find buffer
		lock buffer
		submit RA io
	....
	icreate recovery
	    xfs_trans_get_buffer
		find buffer
		lock buffer
		<blocks on RA completion>
	.....
	<ra completion>
		fails verifier
		clear XBF_DONE
		set bp->b_error = -EIO
		release and unlock buffer
	<icreate gains lock>
	icreate initialises buffer
	marks buffer as done
	adds buffer to delayed write queue
	releases buffer

At this point, we have an initialised inode buffer that is up to
date but has an -EIO state registered against it. When we finally
get to recovering an inode in that buffer:

	inode item recovery
	    xfs_trans_read_buffer
		find buffer
		lock buffer
		sees XBF_DONE is set, returns buffer
	    sees bp->b_error is set
		fail log recovery!

Essentially, we need xfs_trans_get_buf_map() to clear the error status of
the buffer when doing a lookup. This function returns uninitialised
buffers, so the buffer returned can not be in an error state and
none of the code that uses this function expects b_error to be set
on return. Indeed, there is an ASSERT(!bp->b_error); in the
transaction case in xfs_trans_get_buf_map() that would have caught
this if log recovery used transactions....

This patch firstly changes the inode readahead failure to set -EIO
on the buffer, and secondly changes xfs_buf_get_map() to never
return a buffer with an error state set so this first change doesn't
cause unexpected log recovery failures.

Signed-off-by: Dave Chinner <redacted>
Reviewed-by: Brian Foster <redacted>
Signed-off-by: Dave Chinner <david@fromorbit.com>
Signed-off-by: Jiri Slaby <redacted>
---
 fs/xfs/xfs_buf.c       |  7 +++++++
 fs/xfs/xfs_inode_buf.c | 12 +++++++-----
 2 files changed, 14 insertions(+), 5 deletions(-)

diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c
index 263470075ea2..c4a4ad0cd33e 100644
--- a/fs/xfs/xfs_buf.c
+++ b/fs/xfs/xfs_buf.c

@@ -596,6 +596,13 @@ found:
 		}
 	}
 
+	/*
+	 * Clear b_error if this is a lookup from a caller that doesn't expect
+	 * valid data to be found in the buffer.
+	 */
+	if (!(flags & XBF_READ))
+		xfs_buf_ioerror(bp, 0);
+
 	XFS_STATS_INC(xb_get);
 	trace_xfs_buf_get(bp, flags, _RET_IP_);
 	return bp;

diff --git a/fs/xfs/xfs_inode_buf.c b/fs/xfs/xfs_inode_buf.c
index 63382d37f565..4b1447b3a9e4 100644
--- a/fs/xfs/xfs_inode_buf.c
+++ b/fs/xfs/xfs_inode_buf.c

@@ -66,11 +66,12 @@ xfs_inobp_check(
  * has not had the inode cores stamped into it. Hence for readahead, the buffer
  * may be potentially invalid.
  *
- * If the readahead buffer is invalid, we don't want to mark it with an error,
- * but we do want to clear the DONE status of the buffer so that a followup read
- * will re-read it from disk. This will ensure that we don't get an unnecessary
- * warnings during log recovery and we don't get unnecssary panics on debug
- * kernels.
+ * If the readahead buffer is invalid, we need to mark it with an error and
+ * clear the DONE status of the buffer so that a followup read will re-read it
+ * from disk. We don't report the error otherwise to avoid warnings during log
+ * recovery and we don't get unnecssary panics on debug kernels. We use EIO here
+ * because all we want to do is say readahead failed; there is no-one to report
+ * the error to, so this will distinguish it from a non-ra verifier failure.
  */
 static void
 xfs_inode_buf_verify(

@@ -98,6 +99,7 @@ xfs_inode_buf_verify(
 						XFS_RANDOM_ITOBP_INOTOBP))) {
 			if (readahead) {
 				bp->b_flags &= ~XBF_DONE;
+				xfs_buf_ioerror(bp, -EIO);
 				return;
 			}
 

-- 
2.7.1