Re: printk timestamps jumping back and forth in 2.6.25-rc.

[David Miller]

From: "Alexey Zaytsev" <alexey.zaytsev@xxxxxxxxx>
Date: Thu, 28 Feb 2008 17:26:38 +0300

> commit d713f519332e029d43eca8462629314eee1ded86
> Author: Ingo Molnar <mingo@xxxxxxx>
> Date:   Fri Jan 25 21:07:58 2008 +0100
> 
>     sched: fix CONFIG_PRINT_TIME's reliance on sched_clock()
> 
>     Stefano Brivio reported weird printk timestamp behavior during
>     CPU frequency changes:
> 
>       http://bugzilla.kernel.org/show_bug.cgi?id=9475
> 
>     fix CONFIG_PRINT_TIME's reliance on sched_clock() and use cpu_clock()
>     instead.
> 
>     Reported-and-bisected-by: Stefano Brivio <stefano.brivio@xxxxxxxxx>
>     Signed-off-by: Ingo Molnar <mingo@xxxxxxx>
> 
> 
> which sound like "happy debugging, losers" to me. ;)
> 
> Anything else I can do?

FWIW, I've been seeing jumping printk timestamps ever since this
change on SMP sparc64 as well.  Especially on my Niagara boxes.

The timestamp base seem to be dependant upon which cpu prints the
message.  I guess the per-runqueue timestamps this new scheme is using
are all differently based.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/