Re: capget() overflows buffers.

["Andrew G. Morgan" <morgan@kernel.org>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Chris Wright wrote:
| Hmm, it would be kind of nice to have a formalized way get the size,
| perhaps it would help with KaiGai's request for caps printed out.
| Something that tells us either the number of u32s, or the max bit
| supported?

Serge has already provided one  with the call,

~   sys_prctl(PR_CAPBSET_READ, x);

returns -EINVAL if (x > max-supported-capability).

(Ref: 3b7391de67da515c91f48aa371de77cb6cc5c07e)

|> | All looks good.  I think we need to issue some warnings, because
|> | at least Fedora 9 and openSUSE 11 are/will be 2.6.25 based.
|>
|> Do any of the above answers help? (FWIW I attached the patch to the
|> redhat bug.)
|
| Yes, thanks.  But I still think we need to print a warning (unfortunately
| we can't distinguish libcap from non-libcap app), because apps that
| aren't using libcap should really be updated (either pull new update
| from vendor or recompiled by end user).

Just to be clear, you are not referring to a warning that the
application is stuck in a 32-bit capability world, because we already
have one of those: warn_legacy_capability_use(). You are referring to a
warning that might indicate a problem with code like that given in your
example - in which case I'll respond to that part of the thread...

Cheers

Andrew

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.6 (GNU/Linux)

iD8DBQFIN5xI+bHCR3gb8jsRAsz6AKDHOeOO8953r2bRJ3RXZaRdBnlGUwCdG3oX
CvWy/iQVmfVdpeRIWLa7N/w=
=R7Bq
-----END PGP SIGNATURE-----