Re: [PATCH v3 4/6] cpufreq: powernv: Call throttle_check() on receiving OCC_THROTTLE

From: Preeti U Murthy
Date: Thu May 07 2015 - 23:46:57 EST


On 05/08/2015 02:29 AM, Rafael J. Wysocki wrote:
> On Thursday, May 07, 2015 05:49:22 PM Preeti U Murthy wrote:
>> On 05/05/2015 02:11 PM, Preeti U Murthy wrote:
>>> On 05/05/2015 12:03 PM, Shilpasri G Bhat wrote:
>>>> Hi Preeti,
>>>>
>>>> On 05/05/2015 09:30 AM, Preeti U Murthy wrote:
>>>>> Hi Shilpa,
>>>>>
>>>>> On 05/04/2015 02:24 PM, Shilpasri G Bhat wrote:
>>>>>> Re-evaluate the chip's throttled state on recieving OCC_THROTTLE
>>>>>> notification by executing *throttle_check() on any one of the cpu on
>>>>>> the chip. This is a sanity check to verify if we were indeed
>>>>>> throttled/unthrottled after receiving OCC_THROTTLE notification.
>>>>>>
>>>>>> We cannot call *throttle_check() directly from the notification
>>>>>> handler because we could be handling chip1's notification in chip2. So
>>>>>> initiate an smp_call to execute *throttle_check(). We are irq-disabled
>>>>>> in the notification handler, so use a worker thread to smp_call
>>>>>> throttle_check() on any of the cpu in the chipmask.
>>>>>
>>>>> I see that the first patch takes care of reporting *per-chip* throttling
>>>>> for pmax capping condition. But where are we taking care of reporting
>>>>> "pstate set to safe" and "freq control disabled" scenarios per-chip ?
>>>>>
>>>>
>>>> IMO let us not have "psafe" and "freq control disabled" states managed per-chip.
>>>> Because when the above two conditions occur it is likely to happen across all
>>>> chips during an OCC reset cycle. So I am setting 'throttled' to false on
>>>> OCC_ACTIVE and re-verifying if it actually is the case by invoking
>>>> *throttle_check().
>>>
>>> Alright like I pointed in the previous reply, a comment to indicate that
>>> psafe and freq control disabled conditions will fail when occ is
>>> inactive and that all chips face the consequence of this will help.
>>
>> From your explanation on the thread of the first patch of this series,
>> this will not be required.
>>
>> So,
>> Reviewed-by: Preeti U Murthy <preeti@xxxxxxxxxxxxxxxxxx>
>
> OK, so is the whole series reviewed now?

Yes the whole series has been reviewed.

Regards
Preeti U Murthy


>
>

--
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/