Mailing-List: contact cygwin-developers-help AT sourceware DOT cygnus DOT com; run by ezmlm
Sender: cygwin-developers-owner AT sourceware DOT cygnus DOT com
Delivered-To: mailing list cygwin-developers AT sourceware DOT cygnus DOT com
From: Chris Faylor <cgf AT cygnus DOT com>
Date: Mon, 26 Apr 1999 11:31:57 -0400
To: fortinj AT ibm DOT net
Cc: "cygwin-developers AT sourceware DOT cygnus DOT com" <cygwin-developers AT sourceware DOT cygnus DOT com>
Subject: Re: [fortinj AT ibm DOT net: New snapshots]
Message-ID: <19990426113157.B1019@cygnus.com>
References: <Pine DOT SUN DOT 3 DOT 93 DOT 990425231240 DOT 10028G-100000 AT modi DOT xraylith DOT wisc DOT edu> <37247C45 DOT CC82329D AT ibm DOT net>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-Mailer: Mutt 0.95.3i
In-Reply-To: <37247C45.CC82329D@ibm.net>; from fortinj@ibm.net on Mon, Apr 26, 1999 at 10:46:29AM -0400

The original problem that you reported was due to the fact that DJ's machine
was producing incorrect binaries.  That should be corrected except that now
we're seeing other problems in building the cygwin utilities.  When that's
resolved a new snapshot should appear.

In the meantime if you haven't updated egcs to a snapshot you should still be
able to build a copy of cygwin.

cgf

On Mon, Apr 26, 1999 at 10:46:29AM -0400, fortinj AT ibm DOT net wrote:
>All,
>	I'm assuming this patch requires a bebuild of egcs. Is this the 1.1.2
>version available (source, that is) at Mumit's site?
>
>Thanks,
>John Fortin
>fortinj AT ibm DOT net
>
>Mumit Khan wrote:
>> 
>> On Sat, 24 Apr 1999, DJ Delorie wrote:
>> 
>> >
>> > > Could you please forward the patch to me as well? This is a side effect
>> > > of the recent stack alignment patch John Wehle.
>> >
>> > Take out or uncomment this line in gcc/config/i386.h:
>> >
>> > #define PREFERRED_STACK_BOUNDARY 128
>> >
>> > That will disable it for every function, not just stdcalls.  The right
>> > solution would be to disable it just for stdcalls.
>> >
>> 
>> I'm currently bootstrapping using Richard Henderson's patch (actually 2
>> patches), and so far it looks it's doing the right thing.