Problem with timecounters and memory model

[Poul-Henning Kamp]

--------
In message <1056382604.51315.1432925656612.JavaMail.zimbra at embedded-brains.de>,
 Sebastian Huber writes:

>> If you only have one timehand, I can't see how it can work without
>> locks ?
>
>In RTEMS we have two build time configurations: uni-processor and SMP.

That makes sense.

Please note that the default value of 10 timehands was a totally random
number I pulled out of thin air.

The number should be chosen on how long time a thread can expect
to be preempted in kernel-mode, and since I suspect RTEMS to give
quite different guarantees than FreeBSD/UNIX, you may want to
think about that.

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk at FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.