Message-Id: <200006091910.WAA26625@alpha.netvision.net.il>
Date: Fri, 09 Jun 2000 22:09:28 +0200
X-Mailer: Emacs 20.6 (via feedmail 8.1.emacs20_6 I) and Blat ver 1.8.5b
From: "Eli Zaretskii" <eliz AT is DOT elta DOT co DOT il>
To: djgpp AT delorie DOT com
CC: krogg AT gtcom DOT net, djgpp AT delorie DOT com
In-reply-to: <Pine.LNX.4.21.0006090804270.1733-100000@roadrunner.grendel.net>
	(message from Kalum Somaratna aka Grendel on Fri, 9 Jun 2000 08:05:01
	+0600 (LKT))
Subject: Re: HELP...Gprof tore something up!
References:  <Pine DOT LNX DOT 4 DOT 21 DOT 0006090804270 DOT 1733-100000 AT roadrunner DOT grendel DOT net>
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
Precedence: bulk

> Date: Fri, 9 Jun 2000 08:05:01 +0600 (LKT)
> From: Kalum Somaratna aka Grendel <kalum AT lintux DOT cx>
>
> The -pg causes profiling code to be included in the executable which
> results in a noticible performance hit..so please remove it...

That's not really true: all the profiling code does is to hook the
timer tick interrupt and run a 5-line C function when the timer
ticks.  For 18 times a second, this is nowhere near a performance
hit.

One other thing that profiling does is to call another function each
time the application calls one of its functions (not the library
functions).  I don't think this should mean a performance hit, either,
unless the application has lots of small functions which are called a
lot.