https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=7680
--- Comment #9 from Chris Elston <celston@xxxxxxxxxxx> 2012-08-30 05:04:10 PDT ---
Thanks for the review, Jakub.
(In reply to comment #8)
> Quick review:
>
> (In reply to comment #3)
> > Created attachment 9030 [details]
> > Split sha1_hmac for incremental use
>
> For consistency: s/sha1_hmac_starts/sha1_hmac_init/g ?
This is internally consistent within sha1.c, which already has:
sha1_starts, sha1_update and sha1_finish
so I went with:
sha1_hmac_starts, sha1_hmac_update and sha1_hmac_finish.
Whereas md5.c has:
md5_init, md5_append and md5_finish
and therefore I went with:
md5_hmac_init, md5_hmac_append and md5_hmac_finish
So the choice is which pattern we'd like to be consistent with :)
> (In reply to comment #5)
> > Created attachment 9032 [details]
> > Add L2TPv3 control message authentication check
>
> se_alloc() never returns NULL.
I didn't know that, thanks. I'll amend accordingly.
> Also I don't see much sense for gotos in this patch, could you remove it?
ack.
--
Configure bugmail: https://bugs.wireshark.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching all bug changes.