Re: [PATCH, RFC] use hrtimer in sched_clock

[Thomas Gleixner <tglx@linutronix.de>]

[-- Attachment #1: Type: TEXT/PLAIN, Size: 977 bytes --]

On Fri, 11 Jul 2008, Uwe Kleine-König wrote:
> This should make it unnecessary to overwrite sched_clock for a higher
> precision.
> With this patch I get sub-jiffie timing with CONFIG_PRINTK_TIME=y.
> 
> Signed-off-by: Uwe Kleine-König <Uwe.Kleine-Koenig@digi.com>
> Cc: Thomas Gleixner <tglx@linutronix.de>
> Cc: Ingo Molnar <mingo@elte.hu>
> ---
> Hello,
> 
> I tested the patch on arch-arm/mach-ns9xxx and it seem seams to work.
> But I admit I didn't test it deeply and I didn't measure if there is any
> overhead.

There is lots of overhead and your approach is simply wrong. 

>  unsigned long long __attribute__((weak)) sched_clock(void)

The correct way to solve it is to override sched_clock() with a high
resolution implementation for your hardware platform. sched_clock is a
weak function to provide a default implementation based on jiffies.

If you do a grep -r sched_clock arch/arm you'll find a couple of
examples how to override sched_clock().

Thanks,

	tglx