Return-Path: Delivered-To: new-httpd-archive@hyperreal.org Received: (qmail 13406 invoked by uid 6000); 8 Jan 1999 20:46:46 -0000 Received: (qmail 13392 invoked from network); 8 Jan 1999 20:46:44 -0000 Received: from i.meepzor.com (HELO Mail.MeepZor.Com) (root@204.146.167.214) by taz.hyperreal.org with SMTP; 8 Jan 1999 20:46:44 -0000 Received: from Golux.Com (ss02.nc.us.ibm.com [32.97.136.232]) by Mail.MeepZor.Com (8.8.5/8.8.5) with ESMTP id KAA20621; Fri, 8 Jan 1999 10:38:52 -0500 Message-ID: <36966EB1.78E80BCE@Golux.Com> Date: Fri, 08 Jan 1999 15:46:41 -0500 From: Rodent of Unusual Size Organization: The Apache Group X-Mailer: Mozilla 4.06 [en] (WinNT; U) MIME-Version: 1.0 To: new-httpd@apache.org Subject: Re: cvs commit: apache-1.3/src/include ap_mmn.h http_config.h References: <19990108122836.A13869@engelschall.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: new-httpd-owner@apache.org Precedence: bulk Reply-To: new-httpd@apache.org Ralf S. Engelschall wrote: > > But as I said: Putting it at offset 24 is not better for binary compatibility > AFAIK. So I still vote +1 for putting it back to offset 0. And I'm still -1 and vetoing it. With the cookie at offset 0, the core will try to find the version & module name at offsets 4 and 16 respectively. Modules built under 1.3.2 and 1.3.3 will have them at offsets 0 and 12. So 1.3.4+ would not be able to say "mod_foo is too old" for modules compiled for two public releases because it would be looking for the name in the wrong place. It would say "module (null) is too old". That's bogus and stupid when it doesn't need to happen. I'm done with this topic. -- #ken P-)} Ken Coar Apache Group member "Apache Server for Dummies"