Messages in this thread | | | Date | Tue, 1 Jul 2014 16:55:23 +0200 | From | Oleg Nesterov <> | Subject | Re: [PATCH v4 0/2] rcu: uninline rcu_lock_acquire() and rcu_lock_release() |
| |
On 06/30, Paul E. McKenney wrote: > > On Mon, Jun 30, 2014 at 06:18:37PM +0200, Oleg Nesterov wrote: > > Hello, > > > > May be correct this time ;) Based on paulmck/linux-rcu.git rcu/next. > > > > 2/2 is new and hopefully trivial. But! the numbers look suspiciously > > good, I do not understand where does the difference come from... > > Probably from rcu_dereference_raw() and rcu_dereference_check(..., 1). ;-)
Yes, sure, this was the motivation for the patch. But I didn't expect the 50k difference ;)
OK, I understand now. I forgot that every list_for_each_rcu/list_entry_rcu has rcu_dereference_raw().
> Queued and kicked off testing, both mine and (indirectly) Fengguang's.
Thanks!
Oleg.
| |