[2566] in Kerberos_V5_Development
Re: Prototype hell
daemon@ATHENA.MIT.EDU (Ken Raeburn)
Mon Oct 13 03:47:42 1997
To: Tom Yu <tlyu@MIT.EDU>
Cc: Assar Westerlund <assar@sics.se>, "Theodore Y. Ts'o" <tytso@MIT.EDU>,
Ken Hornstein <kenh@cmf.nrl.navy.mil>, krbdev@MIT.EDU
From: Ken Raeburn <raeburn@cygnus.com>
Date: 13 Oct 1997 03:44:31 -0400
In-Reply-To: Tom Yu's message of "Mon, 13 Oct 1997 02:09:10 -0400"
Tom Yu <tlyu@MIT.EDU> writes:
> Hm. It could be that the Ferguson code assumes that certain things
> are exactly 32 bits wide for the purpose of lazy mod-2^32 behavior. I
> don't know enough about DES implementations to tell if this is the
> case for certain though.
Try looking for arithmetic right shifts done without proper masking.
I'd be suspicious of any uses of casts or memcpy in the des code too.
Setting krb5_int32 (etc) to "long" (or whatever) and building on an
Alpha may be a good way to start tracking down some of these problems.
