On Fri, 2007-06-22 at 20:36 -0400, Theodore Tso wrote:
> On Wed, Jun 20, 2007 at 03:03:08PM +0530, Kalpak Shah wrote:
> > In ext2fs_swap_inode_full() only the first (GOOD_OLD_INODE_SIZE +
> > i_extra_isize)bytes are copied into inode. The rest of the inode is
> > not zeroed. So memset the inode to zero if swapfs is enabled.
>
> This was due to the bug where we weren't dealing with the i_extra_size
> correctly, right? ext2fs_swap_inode_full *should* be swapping the
> extra fields and copying it into the inode. If not, that's should be
> the real bug, and adding the memset(inode, 0, bufset) doesn't seem to
> be useful.
>
> Am I missing something?
Hi Ted,
In e2fsck_pass1(), a buffer is allocated for the scratch inode,
inode = (struct ext2_inode *) e2fsck_allocate_memory(ctx, inode_size,
"scratch inode");
Now on big-endian systems, while swapping, ext2fs_swap_inode_full()
swaps only 128+extra_isize bytes and the EAs if they are present. Now if
inode N has EAs, (and this is the inode in the "scratch inode") then
inode N+1 also carries seems to have them since the "scratch inode" was
never zeroed. In ext2fs_swap_inode_full(), this occurs:
if (ext2fs_swab32(*eaf) != EXT2_EXT_ATTR_MAGIC)
return; /* it seems no magic here */
So as I said only the first 128+extra_isize bytes are actually copied
into the *to* inode.
Thanks,
Kalpak.
>
> - Ted
-
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
