X-Authentication-Warning: delorie.com: mailnull set sender to djgpp-bounces using -f
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
Newsgroups: comp.os.msdos.djgpp
Subject: Re: Memory accessing performance
Date: Fri, 21 Dec 2001 20:05:36 +0200
Lines: 16
Message-ID: <3C2379F0.EEA1B21C@is.elta.co.il>
References: <001f01c18995$b8043360$1400a8c0 AT alex> <3C22E937 DOT 2E7D1336 AT is DOT elta DOT co DOT il> <001701c18a34$23c6ad00$1400a8c0 AT alex>
NNTP-Posting-Host: 192.116.55.139
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Trace: fu-berlin.de 1008958122 19230896 192.116.55.139 (16 [61365])
X-Mailer: Mozilla 4.7 [en] (Win98; I)
X-Accept-Language: en
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Reply-To: djgpp AT delorie DOT com

Alex Oleynikov wrote:
> 
> BTW, in conjunction with that I have another question. I did some profiling
> of the application and analyzed the generated output. The profiler's
> sampling rate was 0.0555 seconds, but as I said there are certain events in
> the program that happen at much higher rate (e.g. polling digital I/O and
> controlling the gates). Is there a way of increasing the profiler's sampling
> rate?

You cannot do that without hacking the sources of the library functions
involved in profiling, mainly the mcount.c module.

In general, I don't think you should bother: if the program runs for
reasonably long periods of time, like many seconds or minutes, the profile
you get gives you a very good idea about where are the ``hot spots'' in your
program.