Re: early printk timings way off

[Tim Schmielau]

On Fri, 16 Sep 2005, Tim Bird wrote:

> UPDATE:
> [Based on Tim Schmielau's analysis, maybe it's not the raw TSC
> value, but an unititialized jiffy value coming back from sched_clock().
> In this case, the value is worthless until after time_init().

Well, it's not uninitialized, but initialized to a very high value. So 
the differences between the large values still contain useful information.

> This may be why you're seeing a jump in the first "real" value
> returned.

Yep.

> Previously on x86, the pre-time_init() value was useful (wrong as an absolute
> number, but right in relatives values.)

This should still be the case.


So, one jump (probably the first) happens when time_init sets use_tsc.
Do we understand the other jump as well?

Tim
-
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/