Re: [Patch v9 3/8] arm,arm64,drivers:Add infrastructure to store and update instantaneous thermal pressure

[Thara Gopinath]

On 02/13/2020 09:38 AM, Amit Kucheria wrote:
> On Thu, Feb 13, 2020 at 7:35 PM Thara Gopinath
> <thara.gopinath@xxxxxxxxxx> wrote:
>>
>> On 02/13/2020 07:25 AM, Amit Kucheria wrote:
>>> On Wed, Jan 29, 2020 at 4:06 AM Thara Gopinath
>>> <thara.gopinath@xxxxxxxxxx> wrote:
>>>>
>>>> Add architecture specific APIs to update and track thermal pressure on a
>>>> per cpu basis. A per cpu variable thermal_pressure is introduced to keep
>>>> track of instantaneous per cpu thermal pressure. Thermal pressure is the
>>>> delta between maximum capacity and capped capacity due to a thermal event.
>>>
>>> s/capped/decreased to have consistent use throughout the series e.g. in patch 1.
>>>
>>> Though personally, I like "capped capacity"  in which case
>>> s/decreased/capped in patch 1 and elsewhere.
>>
>> I will fix this
>>>
>>>>
>>>> topology_get_thermal_pressure can be hooked into the scheduler specified
>>>> arch_cpu_thermal_capacity to retrieve instantaneous thermal pressure of a
>>>> cpu.
>>>>
>>>> arch_set_thermal_pressure can be used to update the thermal pressure.
>>>>
>>>> Considering topology_get_thermal_pressure reads thermal_pressure and
>>>> arch_set_thermal_pressure writes into thermal_pressure, one can argue for
>>>> some sort of locking mechanism to avoid a stale value.  But considering
>>>> topology_get_thermal_pressure can be called from a system critical path
>>>> like scheduler tick function, a locking mechanism is not ideal. This means
>>>> that it is possible the thermal_pressure value used to calculate average
>>>> thermal pressure for a cpu can be stale for upto 1 tick period.
>>>>
>>>> Signed-off-by: Thara Gopinath <thara.gopinath@xxxxxxxxxx>
>>>> ---
>>>>
>>>> v6->v7:
>>>>         - Changed the input argument in arch_set_thermal_pressure from
>>>>           capped capacity to delta capacity(thermal pressure) as per
>>>>           Ionela's review comments.
>>>>
>>>>  arch/arm/include/asm/topology.h   |  3 +++
>>>>  arch/arm64/include/asm/topology.h |  3 +++
>>>
>>> Any particular reason to enable this for arm/arm64 in this patch
>>> itself? I'd have enabled them in two separate patches after this one.
>>
>> No reason. No reason not to as well as arch_topology is "Arm specific
>> cpu topology file" and changes are one-liners.
> 
> One reason to do this, IMHO, is to keep platform conversions separate
> from the core infrastructure in a series, so the core can get merged
> while platform maintainers can take their time to decide if, when, how
> to merge this.

That makes sense. I will split it out. I did not think of it from a
merging point of view.

> 


-- 
Warm Regards
Thara