X-Authentication-Warning: acp3bf.physik.rwth-aachen.de: broeker owned process doing -bs
Date: Tue, 18 Jul 2000 11:50:45 +0200 (MET DST)
From: Hans-Bernhard Broeker <broeker AT physik DOT rwth-aachen DOT de>
X-Sender: broeker AT acp3bf
To: djgpp-workers AT delorie DOT com
Subject: Re: Fw: x86: Not clobbering %edx at function entry when profiling
In-Reply-To: <200007172342.TAA16962@envy.delorie.com>
Message-ID: <Pine.LNX.4.10.10007181130140.3746-100000@acp3bf>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
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
Precedence: bulk

On Mon, 17 Jul 2000, DJ Delorie wrote:

> > > argument in it when -mregparm is used.  Modern implementations of
> > > mcount do not need the value we load into %edx; it's for the ancient
> 
> DJGPP *does* use the value in %dx, for *both* -pg and -p.  It stores a
> pointer through it, as a sort-of LRU cache for the structures it has
> created in its own memory.  This cached pointer is used to speed up
> profiling.

Another little remark from me, here: be sure you take care of the effect
of the of the '-mschedule-prologue'switch that controls generation of the
function entry/exit code blocks either via RTL or via direct assembly
output. The former used to screw up debugging, esp. of very short
functions, as we discussed a while ago. This switch may influence the way
mcount calls are generated, too, so it may well be you have to do this
kind of modification in *two* places, not just one.

Hans-Bernhard Broeker (broeker AT physik DOT rwth-aachen DOT de)
Even if all the snow were burnt, ashes would remain.