Re: [RFC] sched: Add a new lockless wake-from-idle implementation
From: Ingo Molnar
Date: Thu Feb 13 2014 - 04:38:16 EST
* Andy Lutomirski <luto@xxxxxxxxxxxxxx> wrote:
> This is a strawman proposal to simplify the idle implementation, eliminate
> a race
>
> Benefits over current code:
> - ttwu_queue_remote doesn't use an IPI unless needed
Cool.
> - The diffstat should speak for itself :)
Neat!
> - Less racy. Spurious IPIs are possible, but only in narrow windows or
> when two wakeups occur in rapid succession.
> - Seems to work (?)
>
> Issues:
> - Am I doing the percpu stuff right?
> - Needs work on non-x86 architectures
Absolutely, and with the least amount of disruption possible, as
people are not very good at testing 'all' of them.
> - The !CONFIG_SMP case needs to be checked
Which also happens to be the default for half of all non-x86 arches.
> - Is "idlepoll" a good name for the new code? It doesn't have *that*
> much to do with the idle state. Maybe cpukick?
'cpukick', hands down.
> If this turns out okay, TIF_NEED_RESCHED could possibly be deleted as well.
Cool ...
Thanks,
Ingo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/