linux-mips
[Top] [All Lists]

Re: CONFIG_PRINTK_TIME and initial value for jiffies?

To: Clem Taylor <clem.taylor@gmail.com>
Subject: Re: CONFIG_PRINTK_TIME and initial value for jiffies?
From: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Date: Tue, 16 May 2006 01:11:55 +0400
Cc: Linux-MIPS <linux-mips@linux-mips.org>
In-reply-to: <ecb4efd10605151341l33f491f1ueca8a0ce609c989d@mail.gmail.com>
Organization: MontaVista Software Inc.
Original-recipient: rfc822;linux-mips@linux-mips.org
References: <ecb4efd10605151341l33f491f1ueca8a0ce609c989d@mail.gmail.com>
Sender: linux-mips-bounce@linux-mips.org
User-agent: Mozilla/5.0 (X11; U; Linux i686; rv:1.7.2) Gecko/20040803
Hello.

Clem Taylor wrote:
I just switched to 2.6.16.16 from 2.6.14 on a Au1550. I enabled
CONFIG_PRINTK_TIME, and for some reason jiffies doesn't start out near
zero like it does on x86. The first printk() always seems to have a
time of 4284667.296000.

jiffies_64 and wall_jiffies gets initialized to INITIAL_JIFFIES, but
I'm not sure where jiffies is initialized. INITIAL_JIFFIES is -300*HZ
(with some weird casting)

The first line on the 2.6.16.16 Au1550 box is:
[4294667.296000] Linux version 2.6.16.16 (ctaylor@gort) (gcc version
4.1.0) #4 Mon May 15 13:06:37 EDT 2006

   Well, this number looks a lot like -300 secs printed as unsigned --
given the fact that jiffies are unsigned, there's no surprise...
My first guess is that TSC code used on x86 corrected jiffies_64 before they got used by sched_clock(). Probably a bogus guess because TSC should have been used by sched_clock() itsefl then...

MBR, Sergei

<Prev in Thread] Current Thread [Next in Thread>