From alsa-devel-owner@alsa.jcu.cz  Thu Jan 14 16:53:34 1999
Received: from renoir.op.net (root@renoir.op.net [209.152.193.4])
	by marvin.jcu.cz (8.9.1a/8.9.1) with ESMTP id QAA19978
	for <alsa-devel@alsa.jcu.cz>; Thu, 14 Jan 1999 16:52:57 +0100
Received: from someip.ppp.op.net (d-bm2-00.ppp.op.net [209.152.194.32]) by renoir.op.net (o1/$Revision: 1.18 $) with ESMTP id JAA01271; Thu, 14 Jan 1999 09:41:59 -0500 (EST)
Message-Id: <199901141441.JAA01271@renoir.op.net>
To: alsa-devel@alsa.jcu.cz
cc: Thomas Hudson <thudson@cygnus.com>,
        linux-audio-dev@ginette.musique.umontreal.ca
Subject: Re: [linux-audio-dev] Re: Linux scheduler issues. 
In-reply-to: Your message of "Thu, 14 Jan 1999 12:13:51 +0100."
             <369DD16F.12360903@ife.ee.ethz.ch> 
Date: Thu, 14 Jan 1999 09:43:04 -0500
From: Paul Barton-Davis <pbd@op.net>
Reply-To: alsa-devel@alsa.jcu.cz
Sender: alsa-devel-owner@alsa.jcu.cz
Precedence: list

In message <369DD16F.12360903@ife.ee.ethz.ch>you write:
>Paul Barton-Davis wrote:
>
>> if you are prepared to do pentium-specific stuff, you can also use the
>> rdtsc instruction, and get resolution to about 60-70 cycles.
>
>That's what gettimeofday does when there's a TSC available,
>otherwise it uses the timer register, which counts with approx.
>1MHz. At least on x86 PC's, I don't know what it does on
>other archs.

True. But gettimeofday() costs about 450 cycles, and a gcc rdtsc asm
macro costs about 66. When I've got 50usec to do a bunch of floating
point computation, and I need to know the time, I know which one I'm
going to use :)

--p

