On Monday, November 26, 2012 07:57:30 AM Mark Langsdorf wrote:

On 11/24/2012 04:05 AM, Rafael J. Wysocki wrote:

quoted

On Saturday, November 17, 2012 03:50:48 PM Borislav Petkov wrote:

quoted

On Tue, Nov 13, 2012 at 02:13:38PM -0500, Mark Langsdorf wrote:

quoted

Although cpufreq_driver has a flag field, no part of cpufreq_driver
is directly passed to the cpufreq_stat code. Only cpufreq_policy
is. It's cleaner to do passes of the while loop than to copy the
cpufreq_driver.flag field into cpufreq_policy and then store it again
in cpufreq_stats.

That maybe so but this newly added loop which is only Calxeda-relevant
is called in cpufreq_stat_notifier_trans, which is the frequency change
notifier call, AFAICT.

Drivers only go through the loop if they can't find an exact frequency.
So every driver that isn't Calxeda shouldn't see the issue.

quoted

quoted

So you probably need to find a slick way of detecting calxeda hw
somewhere along the init path of cpufreq_stats_init and set a
hw-specific flag instead of adding that cost to each driver.

Mark, I suppose you'd like me to take this series for v3.8, but the above
comment from Boris has to be addressed for that.

I think I'd rather drop this particular patch and not have cpufreq_stat
support for Highbank. Redesigning it to meet Boris' requirements is
going to take more time than I currently have available.

Would it be acceptable to drop this patch and fix the issues with
patches 4 and 6 to get the series in?

Yes, it would, but please resubmit ASAP.

Thanks,
Rafael


-- 
I speak only for myself.
Rafael J. Wysocki, Intel Open Source Technology Center.