Re: [PATCH] linux-2.5.66_monotonic-clock_A3

john stultz (johnstul@us.ibm.com)
31 Mar 2003 10:39:09 -0800


On Fri, 2003-03-28 at 23:57, Andi Kleen wrote:
> john stultz <johnstul@us.ibm.com> writes:
>
> > + do {
> > + seq = read_seqbegin(&xtime_lock);
> > + ret = timer->monotonic_clock();
> > + } while (read_seqretry(&xtime_lock, seq));
>
> Why does it need to check xtime lock ? xtime should be independent
> of the monotonic time because it can be changed.

Ok, fair enough. I was using the xtime lock to protect monotonic_base in
(updated in mark_offset()), but since that isn't obvious I should be
more explicit and use a different lock.

> Also doing seqlocks around hardware register reads is quite nasty,
> because a hardware register read can be hundreds of cycles and you're
> very likely to get retries. If you really need a seqlock I would
> move it into the low level function and do it after the hardware access.
> But as far as I can see it can be just removed.

I should be able to use a regular rwlock w/o troubles. I change that and
resubmit.

thanks for the feedback!
-john

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/