From owner-freebsd-emulation  Mon Jun 23 19:44:19 1997
Return-Path: <owner-emulation>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.5/8.8.5) id TAA16230
          for emulation-outgoing; Mon, 23 Jun 1997 19:44:19 -0700 (PDT)
Received: from kithrup.com (kithrup.com [205.179.156.40])
          by hub.freebsd.org (8.8.5/8.8.5) with SMTP id TAA16225
          for <emulation@FreeBSD.ORG>; Mon, 23 Jun 1997 19:44:17 -0700 (PDT)
Received: (from sef@localhost) by kithrup.com (8.6.8/8.6.6) id TAA07504; Mon, 23 Jun 1997 19:44:12 -0700
Date: Mon, 23 Jun 1997 19:44:12 -0700
From: Sean Eric Fagan <sef@Kithrup.COM>
Message-Id: <199706240244.TAA07504@kithrup.com>
To: emulation@FreeBSD.ORG, msagre@cactus.fi.uba.ar
Subject: Re: ORACLE problem with ibcs2_getgroups
Sender: owner-emulation@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

I have no idea what this code is trying to do.  (The original code, that
is.)  Well, okay, I think I understand it now.

However... according to the SCO system I just checked (which, admittedly, is
not SysVr3.2, which is what the original iBCS2 stuff was based on):  if
getgroups() is called with a first parameter of 0, it is supposed to return
the number of supplemental groups, although it won't set any.  Your patch
does not allow that.  (For that matter, I don't think the old one did,
either ;).)  Instead, the code should check for SCARG(uap, gidsetsize) being
0, and, if so, it should set SCARG(&sa, gidsetsize) to NGROUPS_MAX *
sizeof(gid_t).

Without looking at it more, I think that Oracle is checking to see how much
space it should allocate for the supplementary groups, but passing in a size
of 0 is causing our getgroups() routine to return an error.

But I haven't looked at it too closely yet, so I could be wrong.

Sean.