Re: [PATCH] rcu: Remove the unnecessary separate function, rcu_preempt_do_callback()

[Byungchul Park]

On 2/27/2018 8:35 AM, Byungchul Park wrote:
> On 2/27/2018 3:22 AM, Paul E. McKenney wrote:
> > On Mon, Feb 26, 2018 at 12:15:14PM -0500, Steven Rostedt wrote:
> > > On Mon, 26 Feb 2018 14:11:36 +0900
> > > Byungchul Park <byungchul.park@xxxxxxx> wrote:
> > >
> > > > rcu_preemptp_do_callback() was introduced in commit 09223371dea(rcu:
> > > > Use softirq to address performance regression), where it had to be
> > > > distinguished between in the case CONFIG_TREE_PREEMPT_RCU is set and
> > > > it's not.
> > > >
> > > > Now that the code was cleaned up so that rcu_preemt_do_callback() is
> > > > only called in rcu_kthread_do_work() in the same file, tree_plugin.h,
> > > > we don't have to keep the separate function anymore. Remove it for a
> > > > better readability.
> > >
> > > Looks good to me (looks like commit f8b7fc6b51 "rcu: use softirq
> > > instead of kthreads except when RCU_BOOST=y" cleaned up the ifdefs and
> > > removed the requirement).
> > >
> > > Reviewed-by: Steven Rostedt (VMware) <rostedt@xxxxxxxxxxx>
> >
> > Thank you both! I have queued a slightly modified patch for testing
> > and further review. Please see below and let me know if I messed
> > something up.
> >
> > ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ Thanx, Paul
> >
> > ------------------------------------------------------------------------
> >
> > commit b8a3012ddba397d4a18d9fd4a00432f8c2626bd6
> > Author: Byungchul Park <byungchul.park@xxxxxxx>
> > Date:ÂÂ Mon Feb 26 14:11:36 2018 +0900
> >
> > ÂÂÂÂ rcu: Inline rcu_preempt_do_callback() into its sole caller
> > ÂÂÂÂ The rcu_preempt_do_callbacks() function was introduced in commit
> > ÂÂÂÂ 09223371dea(rcu: Use softirq to address performance regression), 
> > where it
> > ÂÂÂÂ was necessary to handle kernel builds both containing and not 
> > containing
> >  RCU-preempt. Since then, various changes (most notably f8b7fc6b51
> > ÂÂÂÂ ("rcu: use softirq instead of kthreads except when RCU_BOOST=y")) 
> > have
> > ÂÂÂÂ resulted in this function being invoked only from 
> > rcu_kthread_do_work(),
> > ÂÂÂÂ which is present only in kernels containing RCU-preempt, which in 
> > turn
> > ÂÂÂÂ means that the rcu_preempt_do_callbacks() function is no longer 
> > needed.
> > ÂÂÂÂ This commit therefore inlines rcu_preempt_do_callbacks() into its
> > ÂÂÂÂ sole remaining caller and also removes the rcu_state_p and 
> > rcu_data_p
> > ÂÂÂÂ indirection for added clarity.
> > ÂÂÂÂ Signed-off-by: Byungchul Park <byungchul.park@xxxxxxx>
> > ÂÂÂÂ Reviewed-by: Steven Rostedt (VMware) <rostedt@xxxxxxxxxxx>
> > ÂÂÂÂ [ paulmck: Remove the rcu_state_p and rcu_data_p indirection. ]
> >
> > diff --git a/kernel/rcu/tree.h b/kernel/rcu/tree.h
> > index dc6f2319fc21..9dd0ea77faed 100644
> > --- a/kernel/rcu/tree.h
> > +++ b/kernel/rcu/tree.h
> > @@ -449,7 +449,6 @@ static void rcu_preempt_boost_start_gp(struct 
> > rcu_node *rnp);
> > Â static void invoke_rcu_callbacks_kthread(void);
> > Â static bool rcu_is_callbacks_kthread(void);
> > Â #ifdef CONFIG_RCU_BOOST
> > -static void rcu_preempt_do_callbacks(void);
> > Â static int rcu_spawn_one_boost_kthread(struct rcu_state *rsp,
> > ÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂÂ struct rcu_node *rnp);
> > Â #endif /* #ifdef CONFIG_RCU_BOOST */
> > diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h
> > index 26d7a31e81cb..b0d7f9ba6bf2 100644
> > --- a/kernel/rcu/tree_plugin.h
> > +++ b/kernel/rcu/tree_plugin.h
> > @@ -686,15 +686,6 @@ static void rcu_preempt_check_callbacks(void)
> > ÂÂÂÂÂÂÂÂÂ t->rcu_read_unlock_special.b.need_qs = true;
> > Â }
> > -#ifdef CONFIG_RCU_BOOST
> > -
> > -static void rcu_preempt_do_callbacks(void)
> > -{
> > -ÂÂÂ rcu_do_batch(rcu_state_p, this_cpu_ptr(rcu_data_p));
> > -}
> > -
> > -#endif /* #ifdef CONFIG_RCU_BOOST */
> > -
> > Â /**
> > ÂÂ * call_rcu() - Queue an RCU callback for invocation after a grace 
> > period.
> > ÂÂ * @head: structure to be used for queueing the RCU updates.
> > @@ -1170,7 +1161,7 @@ static void rcu_kthread_do_work(void)
> > Â {
> > ÂÂÂÂÂ rcu_do_batch(&rcu_sched_state, this_cpu_ptr(&rcu_sched_data));
> > ÂÂÂÂÂ rcu_do_batch(&rcu_bh_state, this_cpu_ptr(&rcu_bh_data));
> > -ÂÂÂ rcu_preempt_do_callbacks();
> > +ÂÂÂ rcu_do_batch(&rcu_preempt_state, this_cpu_ptr(&rcu_preempt_data));
>
> OMG. Sorry for the mistake and thank you very much for fixing it.
>
> I will be more careful.

Ah. Logically no difference between mine and your fixed one.

But anyway yours looks much better! Thank you~ :)

--
Thanks,
Byungchul