Skip to contentSkip to conversation

fix inode_table test in ext234_check_descriptors

PATCH In review Eric Sandeen 1 participant 2 messages Jul 23, 2007

Subsystems (2)
ext4 file system ext2 file system
lore
Share & tools
lore

Eric Sandeen

author
#0 lore raw

PATCH 1/1 fix inode_table test in ext234_check_descriptors

ext[234]_check_descriptors sanity checks block group descriptor geometry at mount time, testing whether the block bitmap, inode bitmap, and inode table reside wholly within the blockgroup. However, the inode table test is off by one so that if the last block in the inode table resides on the last block of the block group, the test incorrectly fails. This is because it tests the last block as (start + length) rather than (start + length - 1).

This can be seen by trying to mount a filesystem made such as:
mkfs.ext2 -F -b 1024 -m 0 -g 256 -N 3744 fsfile 1024
which yields:
EXT2-fs error (device loop0): ext2_check_descriptors: Inode table for group 0 not in group (block 101)!
EXT2-fs: group descriptors corrupted!

There is a similar bug in e2fsprogs, patch already sent for that.

(I wonder if inside(), outside(), and/or in_range() should someday be used in this and other tests throughout the ext filesystems...)

Signed-off-by Eric Sandeen <[email protected]>
fs/ext2/super.c +1 -1
Index: linux-2.6.22-rc4/fs/ext2/super.c
===================================================================
@@ -579,7 +579,7 @@ static int ext2_check_descriptors (struc
 			return 0;
 		}
 		if (le32_to_cpu(gdp->bg_inode_table) < first_block ||
-		    le32_to_cpu(gdp->bg_inode_table) + sbi->s_itb_per_group >
+		    le32_to_cpu(gdp->bg_inode_table) + sbi->s_itb_per_group - 1 >
 		    last_block)
 		{
 			ext2_error (sb, "ext2_check_descriptors",
fs/ext3/super.c +1 -1
Index: linux-2.6.22-rc4/fs/ext3/super.c
===================================================================
@@ -1211,7 +1211,7 @@ static int ext3_check_descriptors (struc
 			return 0;
 		}
 		if (le32_to_cpu(gdp->bg_inode_table) < first_block ||
-		    le32_to_cpu(gdp->bg_inode_table) + sbi->s_itb_per_group >
+		    le32_to_cpu(gdp->bg_inode_table) + sbi->s_itb_per_group - 1 >
 		    last_block)
 		{
 			ext3_error (sb, "ext3_check_descriptors",
fs/ext4/super.c +1 -1
Index: linux-2.6.22-rc4/fs/ext4/super.c
===================================================================
@@ -1269,7 +1269,7 @@ static int ext4_check_descriptors (struc
 		}
 		inode_table = ext4_inode_table(sb, gdp);
 		if (inode_table < first_block ||
-		    inode_table + sbi->s_itb_per_group > last_block)
+		    inode_table + sbi->s_itb_per_group - 1 > last_block)
 		{
 			ext4_error (sb, "ext4_check_descriptors",
 				    "Inode table for group %d"
author
#1 lore raw
Quoted 8 lines · Eric Sandeen

ext[234]_check_descriptors sanity checks block group descriptor geometry at mount time, testing whether the block bitmap, inode bitmap, and inode table reside wholly within the blockgroup. However, the inode table test is off by one so that if the last block in the inode table resides on the last block of the block group, the test incorrectly fails. This is because it tests the last block as (start + length) rather than (start + length - 1).

Quoted 5 lines

This can be seen by trying to mount a filesystem made such as:
mkfs.ext2 -F -b 1024 -m 0 -g 256 -N 3744 fsfile 1024
which yields:
EXT2-fs error (device loop0): ext2_check_descriptors: Inode table for group 0 not in group (block 101)!
EXT2-fs: group descriptors corrupted!

Hm, sorry, after seeing Miklos' last patches... should I be sending these as 3 different patches for ext2, 3, and 4...?

-Eric (who really doesn't want to make Andrew mad at him) :)