X-Authentication-Warning: delorie.com: mail set sender to djgpp-bounces using -f
Date: Sat, 23 Oct 2004 23:39:36 +0200
From: "Eli Zaretskii" <eliz AT gnu DOT org>
Sender: halo1 AT zahav DOT net DOT il
To: djgpp AT delorie DOT com
Message-ID: <01c4b948$Blat.v2.2.2$f034ea00@zahav.net.il>
Content-Transfer-Encoding: 7BIT
Content-Type: text/plain; charset=ISO-8859-1
X-Mailer: emacs 21.3.50 (via feedmail 8 I) and Blat ver 2.2.2
In-reply-to: <clee5f$j5s$1@coranto.ucs.mun.ca> (message from Jason Mills on
	Sat, 23 Oct 2004 17:45:42 -0230)
Subject: Re: Timer inside interrupt handler
References:  <clee5f$j5s$1 AT coranto DOT ucs DOT mun DOT ca>
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

> From: Jason Mills <jmills AT cs DOT mun DOT ca>
> Date: Sat, 23 Oct 2004 17:45:42 -0230
> 
> Does anyone have any suggestions as too what time functions
> can be safely called inside an interrupt handler? I
> need something with a resolution of > 80ms. Basically,
> I need a time stamp of when the IRS is called.

How about reading the BIOS tick counter?  That's a simple memory read
from a fixed address, so it should be safe.  The resolution you get is
54 msec.

OTOH, invoking any library functions that may issue Int 21h or other
real-mode software interrupts is a no-no in an interrupt handler.