Re: [PATCH 4/4] cpufreq: arm_big_little: add SCPI interface driver

[Sudeep Holla <sudeep.holla@arm.com>]

On 01/05/15 14:19, Jon Medhurst (Tixy) wrote:
> On Mon, 2015-04-27 at 12:40 +0100, Sudeep Holla wrote:
>> On some ARM based systems, a separate Cortex-M based System Control
>> Processor(SCP) provides the overall power, clock, reset and system
>> control including CPU DVFS. SCPI Message Protocol is used to
>> communicate with the SCPI.
>>
>> This patch adds a interface driver for adding OPPs and registering
>> the arm_big_little cpufreq driver for such systems.
>>
>> Signed-off-by: Sudeep Holla <sudeep.holla@arm.com>
>> Cc: Viresh Kumar <viresh.kumar@linaro.org>
>> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net>
>> Cc: linux-pm@vger.kernel.org
>> ---
>>   drivers/cpufreq/Kconfig.arm    |   9 ++++
>>   drivers/cpufreq/Makefile       |   1 +
>>   drivers/cpufreq/scpi-cpufreq.c | 103 +++++++++++++++++++++++++++++++++++++++++
>>   3 files changed, 113 insertions(+)
>>   create mode 100644 drivers/cpufreq/scpi-cpufreq.c
>>
>> diff --git a/drivers/cpufreq/Kconfig.arm b/drivers/cpufreq/Kconfig.arm
>> index 4f3dbc8cf729..9e678bf1687c 100644
>> --- a/drivers/cpufreq/Kconfig.arm
>> +++ b/drivers/cpufreq/Kconfig.arm
>> @@ -24,6 +24,15 @@ config ARM_VEXPRESS_SPC_CPUFREQ
>>             This add the CPUfreq driver support for Versatile Express
>>   	  big.LITTLE platforms using SPC for power management.
>>
>> +config ARM_SCPI_CPUFREQ
>> +        tristate "SCPI based CPUfreq driver"
>> +	depends on ARM_BIG_LITTLE_CPUFREQ && ARM_SCPI_PROTOCOL
>
> And ARM_BIG_LITTLE_CPUFREQ depends on CONFIG_ARM, so we can't build this
> for arm64, which is the only platform (Juno) we have to run these
> patches on.

Yes I know

> Unless you're prepared for a horrid hack...
> https://github.com/ARM-software/linux/commit/b9ceaa0cbd7c57d57ee7e69146cc627697570f6e

I need to check if it can be improved before I post on the list and
hence dropped it in the first version of this patch-set.

> or a pair of less horrid ones...
> http://git.linaro.org/kernel/linux-linaro-tracking.git/commit/4b5dd8ff98613b7e90c8f3214522a00ab6900fe9

No not this one, this was already discussed long back I believe.
It doesn't make sense to add bl-switcher for ARM64 IMO.

> http://git.linaro.org/kernel/linux-linaro-tracking.git/commit/8641dbfe253f6a9061742ca11f769fc1d82c2aaa
>

Yes the Kconfig changes looks better than what I have now.

> Any reason why the above two aren't suitable for mainline Linux? The
> second was actually committed then reverted because it broke arm64
> builds, but the other patch fixes that.
>
Yes IMO since arm-big-little also supports multi-cluster cpufreq(yes I
know it can be merged into cpufreq-dt once we have OPPv2 bindings, but
until then), it should handle the config where CONFIG_BL_SWITCHER=n and
that's what I have attempted in my patch above.