Mailing-List: contact cygwin-help AT cygwin DOT com; run by ezmlm
List-Subscribe: <mailto:cygwin-subscribe AT cygwin DOT com>
List-Archive: <http://sourceware.org/ml/cygwin/>
List-Post: <mailto:cygwin AT cygwin DOT com>
List-Help: <mailto:cygwin-help AT cygwin DOT com>, <http://sourceware.org/ml/#faqs>
Sender: cygwin-owner AT cygwin DOT com
Mail-Followup-To: cygwin AT cygwin DOT com
Delivered-To: mailing list cygwin AT cygwin DOT com
Message-ID: <416D50FD.9020704@x-ray.at>
Date: Wed, 13 Oct 2004 17:59:57 +0200
From: Reini Urban <rurban AT x-ray DOT at>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; de-AT; rv:1.8a3) Gecko/20040817
MIME-Version: 1.0
To: cygwin AT cygwin DOT com
Subject: Re: libtool / assembler problem with -DPIC
References: <501586262435 DOT 20041011122240 AT familiehaase DOT de> <416C85FB DOT 5040508 AT cwilson DOT fastmail DOT fm>
In-Reply-To: <416C85FB.5040508@cwilson.fastmail.fm>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
X-IsSubscribed: yes

Charles Wilson schrieb:
> Gerrit P. Haase wrote:
>> With GNU as PIC is not an noop, when -DPIC is used to invoke gas the
>> generated assembly is broken.  I saw this problem with a
>> reautoconfiscated version of GMP.  This may be unusual, but there was
>> libtool used to invoke gas.
>>
>> While -DPIC is a noop for usual compilation, it is harmful when used
>> as gas flag to compile assembly, I suggest to remove it entirely when
>> target is cygwin. Change would be in libtool.m4 line 4971 ff in
>> libtool-1.5.10.
>>
>> Alternative: don't pass flag through when gas is called.
> 
> I don't see anywhere that "-DPIC" is ever invoked on cygwin.  Please 
> send a patch to libtool.m4 that fixes the problem for you.

The problem is when a user (or makefile) adds this -DPIC, which does no 
harm usually. Gerrit thought that it might be clever to strip it on 
cygwin when gas is involved, since this does harm.
Or maybe emit a warning to fix the makefile for cygwin.
(Gerrit: Really -DPIC, not -fPIC?)

> BTW, somebody mentioned libtool CVS branch-2.0 as being too "cutting 
> edge" for cygwin use...FYI, libtool-cvs-branch2.0 passes ALL regression 
> tests which is better than libtool has EVER done on our platform.  I'm 
> thinking of releasing a test version of the next beta...

Ah! Cannot wait for that.
I couldn't find a fix for my -o <noext> => exe problem so far.
And libtool --debug is soo huge.
-- 
Reini Urban
http://xarch.tu-graz.ac.at/home/rurban/

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Problem reports:       http://cygwin.com/problems.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/