Date:	Thu, 10 May 2001 15:18:31 +0300 (WET)
From:	Andris Pavenis <pavenis AT lanet DOT lv>
To:	Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
Cc:	Laurynas Biveinis <lauras AT softhome DOT net>, djgpp AT delorie DOT com
Subject:	Re: DJGPP is based on ...
In-Reply-To:	<Pine.SUN.3.91.1010510122204.4017C-100000@is>
Message-ID:	<Pine.A41.4.05.10105101505140.65374-100000@ieva06.lanet.lv>
MIME-Version:	1.0
Reply-To:	djgpp AT delorie DOT com
Errors-To:	nobody AT delorie DOT com
X-Mailing-List:	djgpp AT delorie DOT com
X-Unsubscribes-To:	listserv AT delorie DOT com

On Thu, 10 May 2001, Eli Zaretskii wrote:

> > We could (can?) add DJGPP there, if we had enough manpower.
> 
> AFAIK, the DJGPP port _is_ tested (the snapshots are built quite
> regularly, as I glean from Andris's messages), so obviously we already
> have enough manpower.

I think It's not so simple. Yes I'm rather often building gcc-3.0 CVS
snapshots for DJGPP. I already have CVS sources as script that updates
them, bootstraps gcc-3.0 for Linux, runs testsuite and sends results is
running automatically at night. So it remains to look whether I have
number of failures small enough and to run some additional scripts to
build a Linux-to-DJGPP cross-compiler and after that native compiler for
DJGPP (result is creating binary archives). It could be possible to make
them running automatically, but it should require to do some additional
work.

About testing:

I'm not running GCC testsuite for DJGPP (I simply can't do that), so only
real testing is with applications and I'm afraid it not enough to add
DJGPP to secondary platforms as it's difficult to see whether we have some
regressions we have to fix

Andris

- Raw text -