Re: [PATCH 1/4] rcu: Speed up calling of RCU tasks callbacks

From: Steven Rostedt <rostedt@goodmis.org>
To: Joel Fernandes <joel@joelfernandes.org>
Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>,
	Joel Fernandes <joelaf@google.com>,
	linux-kernel@vger.kernel.org,
	Peter Zilstra <peterz@infradead.org>,
	Ingo Molnar <mingo@redhat.com>, Boqun Feng <boqun.feng@gmail.com>,
	byungchul.park@lge.com, kernel-team@android.com,
	Josh Triplett <josh@joshtriplett.org>,
	Lai Jiangshan <jiangshanlai@gmail.com>,
	Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Subject: Re: [PATCH 1/4] rcu: Speed up calling of RCU tasks callbacks
Date: Wed, 23 May 2018 21:35:53 -0400	[thread overview]
Message-ID: <20180523213553.7a0ffd6d@gandalf.local.home> (raw)
In-Reply-To: <20180524005119.GA170821@joelaf.mtv.corp.google.com>

On Wed, 23 May 2018 17:51:19 -0700
Joel Fernandes <joel@joelfernandes.org> wrote:

> Shouldn't this assignment be done outside the loop? I believe the variable
> will be initialized on each iteration.
> 
> A program like this doesn't terminate:
> 
> #include<stdio.h>
> 
> int main() {
> 	for (;;) {
> 		int i = 10;
> 		if (!(i--))
> 			break;
> 	}
> 
> 	return 0;
> }

Hey, it compiled, booted and tested. SHIP IT! ;-)

But yeah, a little egg on my face for that one. I blame it for the
first code I wrote right off of coming back from vacation.

> 
> Otherwise looks good to me, I would initialize fract to 10 so its consistent
> with "HZ/10" in other parts of the code but I'm ok with either number.

I was thinking about that, but for some reason I thought 15. Not sure
why. I'm fine with dropping it down to 10. I'll send out a more proper
patch tomorrow.

Thanks!

-- Steve