On 2011-11-21 16:49:07, Linus Torvalds wrote:
> On Mon, Nov 21, 2011 at 3:32 PM, Michael Halcrow <mhalcrow@xxxxxxxxxx> wrote:
> > Characters with ASCII values greater than the size of
> > filename_rev_map[] are valid filename
> > characters. ecryptfs_decode_from_filename() will access kernel memory
> > beyond that array, and ecryptfs_parse_tag_70_packet() will then
> > decrypt those characters.
>
> Ugh. I really don't like the patch.
>
> Why isn't the patch just this one-liner:
>
> diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c
> index 58609bde3b9f..7c50715c05d6 100644
> --- a/fs/ecryptfs/crypto.c
> +++ b/fs/ecryptfs/crypto.c
> @@ -1943,7 +1943,7 @@ static unsigned char *portable_filename_chars
> = ("-.0123456789ABCD"
>
> /* We could either offset on every reverse map or just pad some 0x00's
> * at the front here */
> -static const unsigned char filename_rev_map[] = {
> +static const unsigned char filename_rev_map[256] = {
> 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 7 */
> 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 15 */
> 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 23 */
>
> instead?
>
> Making invalid characters over \x50 be somehow magically different
> from invalid characters elsewhere seems just totally bogus. There are
> lots of characters that aren't valid, and they have the
> filename_rev_map[] value of 0 elsewhere.
>
> So the simpler one-liner is not only simpler, but gives much saner
> semantics, I think - now invalid character '\x05' gets exactly the
> same result as invalid character '\xf5'.
Good point - I'll get this in proper patch form and send a pull request
your way, along with a couple of other fixes.
Tyler
Attachment:
signature.asc
Description: Digital signature
