[Simh] idle on VMS and Windows

[Jason Stevens]

I've had issues with the idle on 3-7-3 running BSD4.x on windows... I just
simply replaced it's idle code with the code from 3-7-1 and it seems far
more reliable...

RENO doesn't seem to idle at all under 3-7-3....




On Wed, Jun 11, 2008 at 1:00 AM, Marc Chametzky <marc at bluevine.net> wrote:

> Out of curiosity, I looked into why I wasn't getting idle functionality
> to work on my VAX running on Solaris.
>
> My first issue was that the gcc optimizer appears to have been messing
> up the code in sim_idle(). When I removed "-O", the calculations worked
> right, but idle still wasn't working.
>
> If I'm reading the code right, it looks as though it's a limitation of
> SIMH on Solaris. In vax_stddev.c, clk_tps is set to 100 (ticks/sec),
> which means 10 ms per tick. Unfortunately, the granularity of
> nanosleep() (as determined by clock_getres()) is also 10 ms, which means
> that it can't idle because the granularity isn't small enough to wait
> for the next VAX clock tick.
>
> Does this make sense?
>
> --Marc
>  _______________________________________________
> Simh mailing list
> Simh at trailing-edge.com
> http://mailman.trailing-edge.com/mailman/listinfo/simh
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.trailing-edge.com/pipermail/simh/attachments/20080611/2249a28b/attachment-0003.html>