cvs commit: src/sys/kern sched_ule.c
Jeff Roberson
jroberson at chesapeake.net
Tue Oct 2 14:31:33 PDT 2007
On Tue, 2 Oct 2007, Peter Grehan wrote:
> Hi Jeff,
>
>> If you can test ULE on UP arm/powerpc I'd appreciate it.
>
> Works fine on ppc after some minor tweaks. I've attached the diff for those
> interested. A couple of questions:
>
> - Can ULE work without preemption ? When it wasn't enabled, the system
> wouldn't switch out of the idle loop. There's a comment in sched_idletd():
It always must use preemption with idletd. You can disable it in all
other cases by omitting the PREEMPTION option...
Ah I see the bug now. I only tested by adjusting
kern.sched.preempt_thresh. I didn't notice that the code in
critical_exit() is conditional. I'll fix that so it should work without
PREEMPTION.
>
> /* ULE relies on preemption for idle interruption. */
>
> I'm asking because PREEMPTION has never been switched on in PPC, and it
> might be nice to enable ULE without having to determine if bugs are in
> PREEMPTION or ULE.
Once I fix critical_exit() it should only preempt idle which is unlikely
to cause bugs, except perhaps in your cpu_idle implementation.
>
> - ppc's cpu_throw() was implemented in C. I modified it to pass the
> thread-lock mutex from the old thread, so the context switch routine would
> simply write it back. Is that an OK thing to do ?
Yes, that's fine.
Thanks!
Jeff
>
> later,
>
> Peter.
>
More information about the freebsd-ppc
mailing list