public inbox for linux-xfs@vger.kernel.org
 help / color / mirror / Atom feed
* [PATCH] Check block magic number so we scan only valid blocks.
@ 2010-07-07 15:09 Peter Watkins
  2010-07-07 23:52 ` Christoph Hellwig
  0 siblings, 1 reply; 3+ messages in thread
From: Peter Watkins @ 2010-07-07 15:09 UTC (permalink / raw)
  To: xfs; +Cc: Peter Watkins

Hello,

Occasionally I've hit a SEGV while querying free space in xfs_db on a mounted
file system. In scanfunc_bno, block->bb_numrecs has crazy values. And bb_magic 
is not XFS_ABTB_MAGIC.

Does a check like this, similiar to other places, make sense?
Should scanfunc_cnt make a similiar check?

Signed-off-by: Peter Watkins <treestem@gmail.com>
---
 db/freesp.c |    3 +++
 1 files changed, 3 insertions(+), 0 deletions(-)

diff --git a/db/freesp.c b/db/freesp.c
index e1902c6..7825628 100644
--- a/db/freesp.c
+++ b/db/freesp.c
@@ -286,6 +286,9 @@ scanfunc_bno(
 	xfs_alloc_ptr_t		*pp;
 	xfs_alloc_rec_t		*rp;
 
+	if (be32_to_cpu(block->bb_magic) != XFS_ABTB_MAGIC)
+		return;
+
 	if (level == 0) {
 		rp = XFS_ALLOC_REC_ADDR(mp, block, 1);
 		for (i = 0; i < be16_to_cpu(block->bb_numrecs); i++)
-- 
1.6.0.4

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

^ permalink raw reply related	[flat|nested] 3+ messages in thread

* Re: [PATCH] Check block magic number so we scan only valid blocks.
  2010-07-07 15:09 [PATCH] Check block magic number so we scan only valid blocks Peter Watkins
@ 2010-07-07 23:52 ` Christoph Hellwig
  2010-07-08 19:20   ` Peter Watkins
  0 siblings, 1 reply; 3+ messages in thread
From: Christoph Hellwig @ 2010-07-07 23:52 UTC (permalink / raw)
  To: Peter Watkins; +Cc: xfs

On Wed, Jul 07, 2010 at 11:09:11AM -0400, Peter Watkins wrote:
> Hello,
> 
> Occasionally I've hit a SEGV while querying free space in xfs_db on a mounted
> file system. In scanfunc_bno, block->bb_numrecs has crazy values. And bb_magic 
> is not XFS_ABTB_MAGIC.
> 
> Does a check like this, similiar to other places, make sense?

Yes, it's reasonable.  As long as you'd don't expect directly poking at
a life partition giving correct results..


> Should scanfunc_cnt make a similiar check?

Yes.

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

^ permalink raw reply	[flat|nested] 3+ messages in thread

* [PATCH] Check block magic number so we scan only valid blocks.
  2010-07-07 23:52 ` Christoph Hellwig
@ 2010-07-08 19:20   ` Peter Watkins
  0 siblings, 0 replies; 3+ messages in thread
From: Peter Watkins @ 2010-07-08 19:20 UTC (permalink / raw)
  To: xfs; +Cc: Peter Watkins

OK, thanks for taking a look.

Here's the complete patch in case anyone is interested.

 -Peter

Signed-off-by: Peter Watkins <treestem@gmail.com>
---
 db/freesp.c |    6 ++++++
 1 files changed, 6 insertions(+), 0 deletions(-)

diff --git a/db/freesp.c b/db/freesp.c
index e1902c6..c4dabad 100644
--- a/db/freesp.c
+++ b/db/freesp.c
@@ -286,6 +286,9 @@ scanfunc_bno(
 	xfs_alloc_ptr_t		*pp;
 	xfs_alloc_rec_t		*rp;
 
+	if (be32_to_cpu(block->bb_magic) != XFS_ABTB_MAGIC)
+		return;
+
 	if (level == 0) {
 		rp = XFS_ALLOC_REC_ADDR(mp, block, 1);
 		for (i = 0; i < be16_to_cpu(block->bb_numrecs); i++)
@@ -310,6 +313,9 @@ scanfunc_cnt(
 	xfs_alloc_ptr_t		*pp;
 	xfs_alloc_rec_t		*rp;
 
+	if (be32_to_cpu(block->bb_magic) != XFS_ABTC_MAGIC)
+		return;
+
 	if (level == 0) {
 		rp = XFS_ALLOC_REC_ADDR(mp, block, 1);
 		for (i = 0; i < be16_to_cpu(block->bb_numrecs); i++)
-- 
1.6.0.4

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

^ permalink raw reply related	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2010-07-08 19:18 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2010-07-07 15:09 [PATCH] Check block magic number so we scan only valid blocks Peter Watkins
2010-07-07 23:52 ` Christoph Hellwig
2010-07-08 19:20   ` Peter Watkins

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox