Date: Wed, 12 May 1999 00:56:30 +0100 (BST)
From: Daniel Barker <sokal AT holyrood DOT ed DOT ac DOT uk>
To: djgpp AT delorie DOT com
Subject: Re: Failed compilation of GCC 2.8.1
In-Reply-To: <199905111429.QAA30060@acp3bf.physik.rwth-aachen.de>
Message-ID: <Pine.GSO.3.95.990512004941.16213C-100000@holyrood.ed.ac.uk>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Reply-To: djgpp AT delorie DOT com
X-Mailing-List: djgpp AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk

Thank you very much for your reply. 

> I've played around with all three types of counting statistics in gcc
> (-fprofile-arcs, -ax and -a style), recently. None of them work
> out-of-the box in gcc-2.8.1 for DJGPP. I have patches to make '-a'
> work nicely, including using the output counts in a 'gprof --line'
> analysis. 

This sounds interesting. All I want is to record source line coverage, for
dynamic testing and the development of test suites. My program
(http://www.icmb.ed.ac.uk/sokal.html) is ported widely both with and
without my knowledge, and I think a test suite with high coverage would be
a big reassurance for myself and other users. 

Obviously, the tool to do this should be reliable ... are you hesitant to
post the patches because there's still debugging to do, or just because
you have not documented them yet? If simply the latter, I would be
interested in giving them a try. Note, though, I am fairly new to
compiling DJGPP packages and am having quite serious trouble just
compiling the standard GCC 2.8.1 source distribution under Windows 98 (see
also my post to the news group tonight).

Thanks again.


Daniel Barker.