From: Alexander Nyberg <alexn@dsv.su.se>
There's a missing failure handling check here that would possibly lead to a
null dereference later on, I'm not sure about the correct return value
however. I haven't tried it as I'm not sure how to trigger the case ;)
Found by the Coverity tool.
Signed-off-by: Alexander Nyberg <alexn@dsv.su.se>
Signed-off-by: Andrew Morton <akpm@osdl.org>
---
25-akpm/fs/ext3/ialloc.c | 5 ++++-
1 files changed, 4 insertions(+), 1 deletion(-)
diff -puN fs/ext3/ialloc.c~ext3_new_inode-failure-handling-missing-check fs/ext3/ialloc.c
--- 25/fs/ext3/ialloc.c~ext3_new_inode-failure-handling-missing-check 2005-02-28 14:50:24.000000000 -0800
+++ 25-akpm/fs/ext3/ialloc.c 2005-02-28 14:50:24.000000000 -0800
@@ -460,9 +460,12 @@ struct inode *ext3_new_inode(handle_t *h
goto out;
for (i = 0; i < sbi->s_groups_count; i++) {
+ err = -EIO;
+
gdp = ext3_get_group_desc(sb, group, &bh2);
+ if (!gdp)
+ goto fail;
- err = -EIO;
brelse(bitmap_bh);
bitmap_bh = read_inode_bitmap(sb, group);
if (!bitmap_bh)
_