Re: gettimeofday order of magnitude slower with pmtimer, which is default

[OGAWA Hirofumi]

kernel@xxxxxxxxxxx writes:

> Quoting OGAWA Hirofumi <hirofumi@xxxxxxxxxxxxxxxxxx>:
>
> What I mean is that I've seen profiles of the worst (from Andi) showing up to 5%
> cpu time on some workloads! That's a heck of a lot slower than when it's
> latched.

I see. Thanks.

>> +	if (unlikely(pmtmr_need_workaround)) {
>
> I would not put this in an unlikely because on the machines where
> pmtmr_need_workaround is true this will always be true.

Yes. However, if machines uses buggy chip, I guessed TSC/PIT would be
more proper as time source. But probably you are right, timer_pit.c
seems more slow usually (it uses many I/O port).

I'll remove unlikely(), and also will remove "Use other timer source"
from warning.

BTW, this patch is still quick hack.

At least, we would need to check the ICH4 which says in comment.
However, I couldn't find the PM-Timer Errata in ICH4 spec update.

Do you/anyone know about a ICH4 error?
-- 
OGAWA Hirofumi <hirofumi@xxxxxxxxxxxxxxxxxx>
-
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/