Re: [RFC PATCH] ARM: clocksource: make ARM_GLOBAL_TIMER selectable
From: Grygorii Strashko
Date: Tue Mar 08 2016 - 05:57:53 EST
On 02/26/2016 10:27 PM, SÃren Brinkmann wrote:
> On Fri, 2016-02-26 at 15:03:19 +0200, Grygorii Strashko wrote:
>> On 02/05/2016 01:39 AM, SÃren Brinkmann wrote:
>>> On Thu, 2016-02-04 at 15:14:47 -0800, Moritz Fischer wrote:
>>>> Hi Soeren,
>>>> On Thu, Feb 4, 2016 at 2:41 PM, SÃren Brinkmann
>>>> <soren.brinkmann@xxxxxxxxxx> wrote:
>>>>> But with this change the 'if !CPU_FREQ' becomes obsolete.
>>>> I'm confused, could you explain that statement? You don't want people
>>>> accidentally running with GT when CPU_FREQ is on, right?
>>> Correct. But with this Kconfig rework you can just deselect it in
>>> Kconfig. The generic HAVE_GT could always be selected.
>> Don't know whom should i ask - but what will be the final conclusion here?
>> Can it be merged?
> I think we don't break anything either way. Would just be some
> additional clean up to get rid of that mentioned constraint (which
> doesn't really work well anyway in the multi-arch kernel). So, no real
> objections to merging it from my side.
I'll re-send it after 4.6-rc.
But What I'm not fully understand is how to get it merged taking into account that
it touches few maches & clocksource :(