Mailing-List:	contact cygwin-help AT sourceware DOT cygnus DOT com; run by ezmlm
List-Subscribe:	<mailto:cygwin-subscribe AT sources DOT redhat DOT com>
List-Archive:	<http://sources.redhat.com/ml/cygwin/>
List-Post:	<mailto:cygwin AT sources DOT redhat DOT com>
List-Help:	<mailto:cygwin-help AT sources DOT redhat DOT com>, <http://sources.redhat.com/ml/#faqs>
Sender:	cygwin-owner AT sources DOT redhat DOT com
Delivered-To:	mailing list cygwin AT sources DOT redhat DOT com
Message-Id:	<200008011346.AA09859@mlx.com>
Mime-Version:	1.0 (NeXT Mail 3.3 v148.2.1)
From:	MarketLogix <mlx AT mlx DOT com>
Date:	Tue, 1 Aug 2000 06:46:43 -0700
To:	cygwin AT sourceware DOT cygnus DOT com
Subject:	Re: Upgrading from b20.1 to 1.1.x - now my static linking fails !
Reply-To:	mlx AT san DOT rr DOT com
References:	<200007311649 DOT AA08997 AT mlx DOT com>
	<200007311800 DOT OAA23243 AT envy DOT delorie DOT com>
	<200007311832 DOT AA09069 AT mlx DOT com>
	<200007311852 DOT OAA26794 AT envy DOT delorie DOT com>
	<200007312201 DOT AA09210 AT mlx DOT com> <3985FA17 DOT 7E725ED9 AT ece DOT gatech DOT edu>
	<200007312219 DOT AA09223 AT mlx DOT com>

O.K.

Looks like I finally got by this.

The culprit looks to be gcc-2.95.2-2 (release 2)

One last time, here's the situation:

Assume that there exists a module called MY.c that contains an
eclectic mixture of C & ObjC source. I then attempt to create
both a relocatable DLL and a static library around this module.

The DLL to be used by non-cygwin Apps, the .a to be used by cygwin Apps.
To further clarify: the static (.a) library here is NOT an import
library for the DLL, its a separate & wholly individual FULL STATIC
binary representation of the same source module MY.c ...

And here's the final scorecard:

1st scenario:

gcc-2.95.2-2 + latest/binutils  = good MY.dll BUT X.exe -lMY.a link fails .

2nd scenario:

gcc-2.95.2-2 + release/binutils = bad  MY.dll BUT  X.exe -lMY.a links O.K. .

3rd scenario: (USABLE ONE)

gcc-2.95.2   + latest/binutils  = good MY.dll AND  X.exe -lMY.a links O.K.

This seems somewhat of an aberration because it only seems to happen
with 1 of my half a dozen or so libary modules. But I'm fairly certain
that since gcc-2.95.2-2 is the future, we haven't seen the last of it.

When I get more time, I will try to drill down deeper to unearth the
exact cause, but I'd bet even money that my skillset in this area is
such that I could easily burn a week - so for now I need to get back
to "direct-paying" work.

later,

bisk


--
Want to unsubscribe from this list?
Send a message to cygwin-unsubscribe AT sourceware DOT cygnus DOT com

- Raw text -