Re: [PATCH] cpuidle/cpuidle-big_little: fix reading cpu id part number

[Juri Lelli]

Hi all,

On 08/08/14 17:47, Lorenzo Pieralisi wrote:
> On Fri, Aug 08, 2014 at 02:21:05PM +0100, Russell King - ARM Linux wrote:
>> On Fri, Aug 08, 2014 at 01:42:37PM +0100, Juri Lelli wrote:
>>> Commit af040ffc9ba1 ("ARM: make it easier to check the CPU part number
>>> correctly") changed ARM_CPU_PART_X masks, and the way they are returned and
>>> checked against. Usage of read_cpuid_part_number() is now deprecated, and
>>> calling places updated accordingly. This actually broke cpuidle-big_little
>>> initialization, as bl_idle_driver_init() performs a check using and hardcoded
>>> mask on cpu_id.
>>>
>>> Update the check to reflect changes on ARM_CPU_PART_X masks. Also,
>>> make the check easier to understand.
>>
>> I don't like this "let's work out our own way to check the CPU type"
>> stuff which people seem to create.  Let's try and do the job better
>> and have some proper interfaces to do this kind of thing.
>>
>> Maybe we should have smp_cpuid_part(cpu) which returns the CPU vendor/
>> part identifier for the particular CPU in an efficient manner.

Thanks a lot for the comment. Does what below address it?

> 
> Ok, basically an inline function that either read and mask the cpuid
> stashed at boot in cpu_data or read the part number straight away on UP.
> 
> Should it belong in asm/smp_plat.h ?
> 

Ok, this is how it could look like.