Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 7 Jul 2006 13:43:12 -0400
From:      John Baldwin <jhb@freebsd.org>
To:        freebsd-current@freebsd.org
Cc:        Mike Jakubik <mikej@rogers.com>, stable@freebsd.org
Subject:   Re: Still getting 'calcru: runtime went backwards'
Message-ID:  <200607071343.14205.jhb@freebsd.org>
In-Reply-To: <44ADC08B.8000408@rogers.com>
References:  <20060629193346.GA2548@dragon.NUXI.org> <44AD6756.4070008@rogers.com> <44ADC08B.8000408@rogers.com>

next in thread | previous in thread | raw e-mail | index | archive | help
On Thursday 06 July 2006 22:01, Mike Jakubik wrote:
> I'm getting a ton of them now, and i found a way to reproduce them. 
> Basically i run a compile session in one terminal, say make buildkernel, 
> and run top in another. As soon as i run top, the messages appear, and 
> they seem to be synchronized with the refresh rate of top, 2 messages 
> per refresh. This is on a 6.1-STABLE as of today.

That is partly because when you run top it queries the resource usage of the 
various processes via fill_kinfo_proc().  When you don't run top, no one is 
asking for the resource usage numbers, so the kernel doesn't waste time 
calculating them.

> ---
> calcru: negative runtime of -261273 usec for pid 12 (swi4: clock)
> calcru: negative runtime of -261273 usec for pid 12 (swi4: clock)
> calcru: negative runtime of -259691 usec for pid 12 (swi4: clock)
> ...

In both cases your errors are for a long-running kernel process that's been up 
since boot.  What's the uptime on your box?

-- 
John Baldwin



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200607071343.14205.jhb>