Date:	Tue, 18 Sep 2001 11:31:12 +0300 (WET)
From:	Andris Pavenis <pavenis AT lanet DOT lv>
X-Sender:	pavenis AT ieva06
To:	Juan Manuel Guerrero <ST001906 AT HRZ1 DOT HRZ DOT TU-Darmstadt DOT De>
Cc:	djgpp-workers AT delorie DOT com
Subject:	Re: gcc-3.0.1 binaries
In-Reply-To:	<73986676986@HRZ1.hrz.tu-darmstadt.de>
Message-ID:	<Pine.A41.4.05.10109181126100.28200-100000@ieva06>
MIME-Version:	1.0
Reply-To:	djgpp-workers AT delorie DOT com
Errors-To:	nobody AT delorie DOT com
X-Mailing-List:	djgpp-workers AT delorie DOT com
X-Unsubscribes-To:	listserv AT delorie DOT com

On Sat, 15 Sep 2001, Juan Manuel Guerrero wrote:

> On Sat, 15 Sep 2001, Andris Pavenis wrote:
> > My latest gcc-3.0.1 binaries for DJGPP are available at
> > 	http://www.ltn.lv/~pavenis/gcctest.html
> >
> > I built them using DJGPP v2.03 (+Win2K related patches) built 
> > with gcc-2.95.3
> >
> > Known issues with previous builds:
> >	crashes with some Cyrix CPU's unless internal CPU cache is 	disabled. See:
> > 	http://www.delorie.com/djgpp/mail-archives/browse.cgi?p=djgpp-workers/2001/09/03/07:47:59
> > 	I haven't get any feedback with these binaries yet. Binaries of gcc-3.0
> > 	from v2gnu/alphas worked on these 2 systems. Main difference:
> > 	I used CVS version of DJGPP built with gcc-3.0 for binaries of gcc-3.0
> >
> > Unless there will be serious problems I'm going to upload files
> > next week
> I have downloaded this new release (build 2001-09-10) and it does not work for me.
> It is the same failure as before. Meanwhile I have replace the 64MB PS/2 modules
> by a 128MB PC133 module and gcc301 still fails in the same way. It is still impossible
> to bootstrap gcc301 using djdev203 or latest CVS tree. This weekend I will install linux
> and will try to compile gcc301 (linux native). I will try to get an Cyrix CPU to replace
> my own one to see what happens ASAP.

I saw notes that some Cyrix CPU's are incompatible with some particular
motherboards with resulting GCC crashes (Linux were mentioned there, but I
don't think it's a serious difference):
	http://www.bitwizard.nl/sig11/
Disappearing problems when CPU internal cache is disabled also point to
hardware problems

> http://www.bitwizard.nl/sig11/In conclusion, either my cpu is brocken,
then the issue is no issue at all and can be forgotten.
> Or it is a Cyrix specific issue, then it should be mentioned somewhere in the readme files
> and then be forgotten. I assume that there are no much cyrix users around in this days anyway.

If it's a real incompatibility replacing CPU by very similar one may not
help

Andris

- Raw text -