Re: Timekeeping and scheduling
- From: Martin Lucina <martin@xxxxxxxxxx>
- To: rumpkernel-users@xxxxxxxxxxxxx
- Date: Fri, 6 Mar 2015 11:56:24 +0100
pooka@xxxxxx said:
> rump_schedule() schedules a *rump kernel cpu*, not a host thread.
> sched_yield() yields the *host thread*.
>
> The problem is that the code you posted is essentially a busy loop.
> Since the current timekeeping driver depends on the clock interrupt
> running, and since all interrupts in rump kernels are threads, the
> clock interrupt never runs because the *host thread* the interrupt
> uses is not scheduled in a busy loop.
*slaps self with a wet fish*
Right. And the reason the code works on POSIX without sched_yield() is
because the host scheduler (Linux) is preemptive whereas the -bmk and -xen
host schedulers are cooperative.
> Now, there are actually two issues to consider:
>
> 1) accurate timekeeping which has >clockintr resolution
> 2) detecting if interrupts are pending, and yielding the *host
> thread* in rump_un/schedule() if so
or, eventually, 3) replace/enhance the -xen and -bmk platforms with a
preemptive host scheduler.
Other related posts:
