Subject: Re: Hard realtime: Directions?
To: Oliver.Korpilla@gmx.de <Oliver.Korpilla@gmx.de>
From: SODA Noriyuki <soda@sra.co.jp>
List: tech-kern
Date: 04/12/2005 16:41:49
>>>>> On Tue, 12 Apr 2005 09:29:12 +0200,
	"Oliver.Korpilla@gmx.de" <Oliver.Korpilla@gmx.de> said:

> What do you think is meant with:

> "We need real-time scheduling support, POSIX real-time extensions, and 
> thread-safe libraries. There is an increasing number of applications 
> related to video, voice, and control that need hard real-time support. 
> We can start by bullet-proofing our thread support and finishing the 
> re-entrancy issues with our libraries, then continue by evaluating 
> real-time scheduling and making subsystems of our kernel able to use 
> multiple processors."

> From: [Also available at 
> http://www.netbsd.org/Foundation/reports/2004.html] Report of the 2004 
> Annual NetBSD Group Meeting

> Does any of this sound reasonable?

Yes, all of the goals are reasonable, except calling it as "hard
realtime".

> It sounds mostly out of reach at the time, don't you think. One
> could improve on the scheduler, or imnplement POSIX real-time stuff
> like "realtime" priorities - but it would just be a facade with not
> much to back it up, don't you think?

Not just a facade I think, because all of those must improve latency.
--
soda