From: Stephen Tweedie <sct@redhat.com>
This patch improves ext3's ability to deal with corruption on-disk. If we
try to delete a metadata block twice, we confuse ext3's internal revoke
error-checking, resulting in a BUG(). But this can occur in practice due
to a corrupt indirect block, so we should attempt to fail gracefully.
Downgrade the assert failure to a JH_EXPECT_BH failure, and return EIO when
it occurs.
This is easily reproduced with a sample ext3 fs image containing an inode
which references the same indirect block more than once. Deleting that
inode will BUG() an unfixed kernel with:
Assertion failure in journal_revoke() at fs/jbd/revoke.c:379:
"!buffer_revoked(bh)"
With the fix, ext3 recovers gracefully.
Signed-off-by: Stephen Tweedie <sct@redhat.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
---
25-akpm/fs/jbd/revoke.c | 7 ++++++-
1 files changed, 6 insertions(+), 1 deletion(-)
diff -puN fs/jbd/revoke.c~ext3-handle-attempted-double-delete-of-metadata fs/jbd/revoke.c
--- 25/fs/jbd/revoke.c~ext3-handle-attempted-double-delete-of-metadata Fri Nov 19 16:18:54 2004
+++ 25-akpm/fs/jbd/revoke.c Fri Nov 19 16:18:54 2004
@@ -376,7 +376,12 @@ int journal_revoke(handle_t *handle, uns
first having the revoke cancelled: it's illegal to free a
block twice without allocating it in between! */
if (bh) {
- J_ASSERT_BH(bh, !buffer_revoked(bh));
+ if (!J_EXPECT_BH(bh, !buffer_revoked(bh),
+ "inconsistent data on disk")) {
+ if (!bh_in)
+ brelse(bh);
+ return -EIO;
+ }
set_buffer_revoked(bh);
set_buffer_revokevalid(bh);
if (bh_in) {
_