postscriptum on RT_PREEMPT on ARM

Am 03.12.2012 um 05:22 schrieb Michael Haberler:

>  The main difference is that Xenomai ports are pegged to very few kernel 
> versions as starting point - realistically 2, maybe three for the 
> adventurous; whereas RT_PREEMPT has been available for many major kernel 
> versions so far and will likely be the first RT option to be in Linux 
> mainline; Xenomai might follow later when their 'Xenomai 3' strategy pans 
> out. 

I had assumed 'building an RT_PREEMPT kernel is a tad easier or not necessary 
on the PC, so it will be easier on an ARM too' - until I tried, that is.

Well nonsense, wrong analogy.

A key element for an RT kernel of any kind is high-resolution timer support 
being in place in the kernel for results to make any sense. That can be assumed 
granted on PC/x86, but it cannot with the average ARM board. There will be 
_some_ timer support, but not necessarily a hires one.

It is likely in place if somebody ported Xenomai to that board to start with, 
because that's clearly the more popular option in this league. And it might 
still mean the RT_PREEMPT kernel needs those patches carried over from whatever 
source manually. 

That, of course, makes the 'easier build' assumption moot, it can very well be 
harder, which raises the question why do it in the first place if one can have 
better latency with Xenomai to start with with less work.

So much for the value of re-applying past experiences;)

-m


------------------------------------------------------------------------------
LogMeIn Rescue: Anywhere, Anytime Remote support for IT. Free Trial
Remotely access PCs and mobile devices and provide instant support
Improve your efficiency, and focus on delivering more value-add services
Discover what IT Professionals Know. Rescue delivers
http://p.sf.net/sfu/logmein_12329d2d
_______________________________________________
Emc-developers mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/emc-developers

Reply via email to