Re: [patch] cifs: check offset in decode_ntlmssp_challenge()

[Dan Carpenter <dan.carpenter@oracle.com>]

[-- Attachment #1: Type: text/plain, Size: 1175 bytes --]

On Tue, Jan 31, 2012 at 05:49:37AM -0500, Jeff Layton wrote:
> On Tue, 31 Jan 2012 11:52:01 +0300
> Dan Carpenter <dan.carpenter@oracle.com> wrote:
> 
> > We should check that we're not copying memory from beyond the end of the
> > blob.
> > 
> > Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
> > 
> > diff --git a/fs/cifs/sess.c b/fs/cifs/sess.c
> > index d85efad..eb76741 100644
> > --- a/fs/cifs/sess.c
> > +++ b/fs/cifs/sess.c
> > @@ -395,6 +395,10 @@ static int decode_ntlmssp_challenge(char *bcc_ptr, int blob_len,
> >  	ses->ntlmssp->server_flags = le32_to_cpu(pblob->NegotiateFlags);
> >  	tioffset = le32_to_cpu(pblob->TargetInfoArray.BufferOffset);
> >  	tilen = le16_to_cpu(pblob->TargetInfoArray.Length);
> > +	if (tioffset > blob_len || tioffset + tilen > blob_len) {
> > +		cERROR(1, "tioffset + tilen too high %u + %u", tioffset, tilen);
> > +		return -EINVAL;
> > +	}
> 
> Good catch.
> 
> Do we really need a || here though? Would it not be sufficient to check
> if tioffset + tilen > blob_len? Or are you concerned about that value
> wrapping?

Yep.  I was concerned about value wrapping.

regards,
dan carpenter


[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 836 bytes --]