RE: [PATCH] sched: Folding nohz load accounting more accurate

From: "Doug Smythies" <dsmythies@telus.net>
To: "'Peter Zijlstra'" <peterz@infradead.org>,
	"'Yong Zhang'" <yong.zhang0@gmail.com>
Cc: "'Charles Wang'" <muming.wq@gmail.com>,
	linux-kernel@vger.kernel.org, "'Ingo Molnar'" <mingo@redhat.com>,
	"'Tao Ma'" <tm@tao.ma>, '含黛' <handai.szj@taobao.com>,
	"'Thomas Gleixner'" <tglx@linutronix.de>,
	"Doug Smythies" <dsmythies@telus.net>
Subject: RE: [PATCH] sched: Folding nohz load accounting more accurate
Date: Tue, 19 Jun 2012 08:50:29 -0700	[thread overview]
Message-ID: <000001cd4e33$496f6c30$dc4e4490$@net> (raw)
In-Reply-To: <1340097532.21745.19.camel@twins>

[-- Attachment #1: Type: text/plain, Size: 1051 bytes --]

>> On Tue, 2012-06-19 at 14:08 +0800, Yong Zhang wrote:
>  On 2012.06.19 02:19 -0700, Peter Zijlstra wrote:

>>>  	/*
>>> +	 * If we're still outside the sample window, we're done.
>>>  	 */
>>> +	if (time_before(jiffies, this_rq->calc_load_update))
>>> +		return;

>> 	else if (time_before(jiffies, calc_load_update + 10)
>> 		this_rq->calc_load_update = calc_load_update + LOAD_FREQ;
>> 	else
>> 		this_rq->calc_load_update = calc_load_update;
>> 
>> Otherwise if you woke after the sample window, we loose on sample?
>> And maybe we need local variable to cache calc_load_update.

> Ah indeed, although I'd write it like:

>	this_rq->calc_load_update = calc_load_update;
>	if (time_before(jiffies, this_rq->calc_load_update + 10)
>		this_rq->calc_load_update += LOAD_FREQ;

Note missing end brace:
	if (time_before(jiffies, this_rq->calc_load_update + 10))

My automated 63 hour test has been terminated, the code changed
and the test re-started.

The attached png file is what I had so far, but it will be replaced.
Summary: Looked good, so far.

[-- Attachment #2: n_processes_peter35.png --]
[-- Type: image/png, Size: 34780 bytes --]