Re: [patch] Real-Time Preemption, -RT-2.6.12-rc1-V0.7.41-07

[Ingo Molnar]

* Steven Rostedt <rostedt@xxxxxxxxxxx> wrote:

> > i think the 'migrate read-count' method is not adequate either, because
> > all callbacks queued within an RCU read section must be called after the
> > lock has been dropped - while with the migration method CPU#1 would be
> > free to process callbacks queued in the RCU read section still active on
> > CPU#2.
> 
> Although you can't disable preemption for the duration of the
> rcu_readlock, what about pinning the process to a CPU while it has the
> lock.  Would this help solve the migration issue?

yes, but that's rather gross. PREEMPT_BKL was opposed by Linus precisely
because it used such a method - once that was fixed, PREEMPT_BKL got
accepted. It also limits the execution flow quite much. I'd rather not
do it if there's any other method.

	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/