Return-Path: Delivered-To: new-httpd-archive@hyperreal.org Received: (qmail 16340 invoked by uid 6000); 12 Jan 2000 17:04:09 -0000 Received: (qmail 16330 invoked from network); 12 Jan 2000 17:04:08 -0000 Received: from fwns1d.raleigh.ibm.com (HELO fwns1.raleigh.ibm.com) (204.146.167.235) by taz.hyperreal.org with SMTP; 12 Jan 2000 17:04:08 -0000 Received: from rtpmail03.raleigh.ibm.com (rtpmail03.raleigh.ibm.com [9.37.172.47]) by fwns1.raleigh.ibm.com (8.9.0/8.9.0/RTP-FW-1.2) with ESMTP id MAA28684 for ; Wed, 12 Jan 2000 12:03:59 -0500 Received: from hapkido.raleigh.ibm.com (rbb@hapkido.raleigh.ibm.com [9.37.65.84]) by rtpmail03.raleigh.ibm.com (8.8.5/8.8.5/RTP-ral-1.1) with ESMTP id MAA26816 for ; Wed, 12 Jan 2000 12:03:49 -0500 Date: Wed, 12 Jan 2000 12:01:06 -0500 (EST) From: Ryan Bloom To: new-httpd@apache.org Subject: Re: 1.3.10 and AIX 4.2 In-Reply-To: <20000112174802.B7387@schumann.cx> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: new-httpd-owner@apache.org Precedence: bulk Reply-To: new-httpd@apache.org > > compiling with SHARED_CORE? No garauntees that will work, but maybe gcc > > needs doing that should assign 1 to using_shlib which should cause the > > LD_LIBS stuff I mentioned earlier to happen. > > Could you rephrase the second part of your last sentence? I think I need some time off right now. :-) I'm sorry. All I was suggesting was that gcc on AIX might need SHARED_CORE, which would force using_shlib to be set appropriately. If they aren't using dynamic modules, this is probably a bad idea. I am concerned because I do not know of any bugs reporting this issue, and I don't know when this was last modified. I do know that I have had Apache building on AIX 4.2.X a while ago without dynamic modules, but using xlc. I have never tried to use gcc on AIX. What also bothers me is that if -ld is always needed, is xlc just putting it on there automatically? > > It looks like we need to force use_shlib=1 on this platform, > because the code in os/unix/os-aix-dso.c will always be > pulled in, thus requiring linking against libld. I think I would rather no include the os-aix-dso.c code unless we are either enabling mod_so or compiling with SHARED_CORE. This seems to me to be more intuitive. If the server isn't compiled for dynamic module support, then we shouldn't be including functions to support loading libraries, IMHO. Ryan _______________________________________________________________________ Ryan Bloom rbb@raleigh.ibm.com 4205 S Miami Blvd RTP, NC 27709 Come to the first official Apache Software Foundation Conference!