Messages in this thread | | | Date | Wed, 21 Jan 2015 10:30:07 +0100 (CET) | From | Thomas Gleixner <> | Subject | Re: [PATCH] ksoftirqd: Enable IRQs and call cond_resched() before poking RCU |
| |
On Tue, 20 Jan 2015, Paul E. McKenney wrote: > On Tue, Jan 20, 2015 at 02:21:51PM +0100, Thomas Gleixner wrote: > > The whole rcu_note_context_switch() in run_ksoftirqd() is silly. > > > > cond_resched() > > __preempt_count_add(PREEMPT_ACTIVE); > > > > __schedule(); > > preempt_disable(); > > rcu_note_context_switch(); > > .... > > > > __preempt_count_sub(PREEMPT_ACTIVE); > > I agree that if should_resched() returns true as assumed above, then there > is no point to invoking rcu_note_context_switch(). However, the case that > this code applies to is when should_resched() returns false, but RCU is > waiting for a quiescent state from the current CPU. In that case, > cond_resched() won't do anything for RCU, and we do need the > rcu_note_context_switch().
So this should be:
if (!cond_resched()) rcu_note_context_switch();
Hmm? > Of course, it would be better to avoid the extra RCU work in the common > case where cond_resched() does inovke the scheduler. And that is the > point of the following patch, which uses cond_resched_rcu_qs(). > However, this use of cond_resched_rcu_qs() doesn't work in older > kernels. So Calvin's patch is for backporting, and the follow-up > patch for future kernels.
I see.
Thanks,
tglx
| |