Re: [PATCH RESEND] thermal: power_allocator: trace the real requested power

[Punit Agrawal <punit.agrawal@arm.com>]

Hi Eduardo, Rui,

Javi Merino <javi.merino@arm.com> writes:

> On Fri, Jul 03, 2015 at 10:24:33AM +0100, Javi Merino wrote:
>> The power allocator governor uses ftrace to output a bunch of internal
>> data for debugging and tuning.  Currently, the requested power it
>> outputs is the "weighted" requested power, that is, what each cooling
>> device has requested multiplied by the cooling device weight.  It is
>> more useful to trace the real request, without any weight being
>> applied.
>> 
>> This commit only affects the data traced, there is no functional change.
>> 
>> Cc: Zhang Rui <rui.zhang@intel.com>
>> Cc: Eduardo Valentin <edubezval@gmail.com>
>> Signed-off-by: Javi Merino <javi.merino@arm.com>
>> ---
>>  drivers/thermal/power_allocator.c | 26 ++++++++++++++++----------
>>  1 file changed, 16 insertions(+), 10 deletions(-)
>> 
>> diff --git a/drivers/thermal/power_allocator.c b/drivers/thermal/power_allocator.c
>> index 4672250b329f..63a448f9d93b 100644
>> --- a/drivers/thermal/power_allocator.c
>> +++ b/drivers/thermal/power_allocator.c
>> @@ -229,7 +229,8 @@ static int allocate_power(struct thermal_zone_device *tz,
>>  	struct thermal_instance *instance;
>>  	struct power_allocator_params *params = tz->governor_data;
>>  	u32 *req_power, *max_power, *granted_power, *extra_actor_power;
>> -	u32 total_req_power, max_allocatable_power;
>> +	u32 *weighted_req_power;
>> +	u32 total_req_power, max_allocatable_power, total_weighted_req_power;
>>  	u32 total_granted_power, power_range;
>>  	int i, num_actors, total_weight, ret = 0;
>>  	int trip_max_desired_temperature = params->trip_max_desired_temperature;
>> @@ -247,16 +248,17 @@ static int allocate_power(struct thermal_zone_device *tz,
>>  	}
>>  
>>  	/*
>> -	 * We need to allocate three arrays of the same size:
>> -	 * req_power, max_power and granted_power.  They are going to
>> -	 * be needed until this function returns.  Allocate them all
>> -	 * in one go to simplify the allocation and deallocation
>> -	 * logic.
>> +	 * We need to allocate five arrays of the same size:
>> +	 * req_power, max_power, granted_power, extra_actor_power and
>> +	 * weighted_req_power.  They are going to be needed until this
>> +	 * function returns.  Allocate them all in one go to simplify
>> +	 * the allocation and deallocation logic.
>>  	 */
>>  	BUILD_BUG_ON(sizeof(*req_power) != sizeof(*max_power));
>>  	BUILD_BUG_ON(sizeof(*req_power) != sizeof(*granted_power));
>>  	BUILD_BUG_ON(sizeof(*req_power) != sizeof(*extra_actor_power));
>> -	req_power = devm_kcalloc(&tz->device, num_actors * 4,
>> +	BUILD_BUG_ON(sizeof(*req_power) != sizeof(*weighted_req_power));
>> +	req_power = devm_kcalloc(&tz->device, num_actors * 5,
>>  				 sizeof(*req_power), GFP_KERNEL);
>>  	if (!req_power) {
>>  		ret = -ENOMEM;
>> @@ -266,8 +268,10 @@ static int allocate_power(struct thermal_zone_device *tz,
>>  	max_power = &req_power[num_actors];
>>  	granted_power = &req_power[2 * num_actors];
>>  	extra_actor_power = &req_power[3 * num_actors];
>> +	weighted_req_power = &req_power[4 * num_actors];
>>  
>>  	i = 0;
>> +	total_weighted_req_power = 0;
>>  	total_req_power = 0;
>>  	max_allocatable_power = 0;
>>  
>> @@ -289,13 +293,14 @@ static int allocate_power(struct thermal_zone_device *tz,
>>  		else
>>  			weight = instance->weight;
>>  
>> -		req_power[i] = frac_to_int(weight * req_power[i]);
>> +		weighted_req_power[i] = frac_to_int(weight * req_power[i]);
>>  
>>  		if (power_actor_get_max_power(cdev, tz, &max_power[i]))
>>  			continue;
>>  
>>  		total_req_power += req_power[i];
>>  		max_allocatable_power += max_power[i];
>> +		total_weighted_req_power += weighted_req_power[i];
>>  
>>  		i++;
>>  	}
>> @@ -303,8 +308,9 @@ static int allocate_power(struct thermal_zone_device *tz,
>>  	power_range = pid_controller(tz, current_temp, control_temp,
>>  				     max_allocatable_power);
>>  
>> -	divvy_up_power(req_power, max_power, num_actors, total_req_power,
>> -		       power_range, granted_power, extra_actor_power);
>> +	divvy_up_power(weighted_req_power, max_power, num_actors,
>> +		       total_weighted_req_power, power_range, granted_power,
>> +		       extra_actor_power);
>>  
>>  	total_granted_power = 0;
>>  	i = 0;
>
> Gentle ping

This is a simple change that should be merged now to prevent user
visible behaviour change once 4.2 is released.

Seeing that it's been on the list for three weeks without any
complaints, could this be picked up for 4.2?

Thanks.
> --
> To unsubscribe from this list: send the line "unsubscribe linux-pm" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html