On Fri, 2018-07-06 at 10:11 -0700, Paul E. McKenney wrote:
> > The preempt state is alread a bit complicated and shadowed in the
> > preempt_count (on some architectures) adding additional bits to it like
> > this is just asking for trouble.
> 
> How about a separate need_resched_rcu() that includes the extra cache
> miss?  Or open-coding the rcu_urgent_qs_requested()?

Peter said "touch two cachelines". He didn't say it was a cache miss.

Given that every single cond_resched() call touches the same cache
line, and every single rcu_all_qs() and similar will also touch it,
it's fairly much guaranteed *not* to be a miss...

... which is why I didn't really understand why he cared.