From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S262207AbVFURgf (ORCPT ); Tue, 21 Jun 2005 13:36:35 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S262190AbVFURgf (ORCPT ); Tue, 21 Jun 2005 13:36:35 -0400 Received: from viper.oldcity.dca.net ([216.158.38.4]:60836 "HELO viper.oldcity.dca.net") by vger.kernel.org with SMTP id S262208AbVFURgK (ORCPT ); Tue, 21 Jun 2005 13:36:10 -0400 Subject: Re: IBM HDAPS Someone interested? From: Lee Revell To: Vojtech Pavlik Cc: Adam Goode , Pavel Machek , Alejandro Bonilla , linux-kernel@vger.kernel.org, linux-thinkpad@linux-thinkpad.org In-Reply-To: <20050621164519.GA11601@ucw.cz> References: <20050620155720.GA22535@ucw.cz> <005401c575b3_5f5bba90_600cc60a@amer.sykes.com> <20050620163456.GA24111@ucw.cz> <20050620165703.GB477@openzaurus.ucw.cz> <20050620204533.GA9520@ucw.cz> <1119303016.5194.24.camel@lynx.auton.cs.cmu.edu> <1119368259.19357.18.camel@mindpipe> <20050621164519.GA11601@ucw.cz> Content-Type: text/plain Date: Tue, 21 Jun 2005 13:36:03 -0400 Message-Id: <1119375363.4569.2.camel@mindpipe> Mime-Version: 1.0 X-Mailer: Evolution 2.3.1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Tue, 2005-06-21 at 18:45 +0200, Vojtech Pavlik wrote: > On Tue, Jun 21, 2005 at 11:37:38AM -0400, Lee Revell wrote: > > Ugh, if userspace can't meet a 300ms RT constraint, that's a pretty > > shitty OS you have there. > > It's not that you do one measurement in the 300ms. You need to do at least > 100, and some computations, too. > > > This should certainly be done in userspace on Linux. > > So it's a 3ms RT constraint, which is not as easy. > Heh, we do it all the time with JACK. But, I think I was wrong anyway, you'll have to do it in the kernel because you would need PREEMPT to meet that with any degree of certainty. Lee