[PATCH v9 00/14] Tegra114 clockframework

[Stephen Warren]

On 04/03/2013 12:27 PM, Stephen Warren wrote:
> On 04/03/2013 11:52 AM, Stephen Warren wrote:
>> On 04/03/2013 08:40 AM, Peter De Schrijver wrote:
>>> This is the nineth version of the Tegra114 clockframework. It is based on the
>>> next-20130320-fixed branch of
>>> git://nv-tegra.nvidia.com/user/swarren/linux-2.6.git,
> ... [it depends on]
>>> http://patchwork.ozlabs.org/patch/233415/
>>
>> That is Joseph's "clocksource: tegra: enable arch_timer". Is this
>> actually a dependency, or an optimization, or ...?
>>
>> Rob Herring objected to that patch since it isn't necessary if his
>> series "ARM arch, sp804 and integrator timer CLKSRC_OF support" is applied.
>>
>> If I apply just your (Peter's) CCF series without Joseph's or Rob's
>> patches, will everything still work, or is this a hard dependency to
>> create a working system?
> 
> I manage to get this series applied on top of my local dev branch, based
> on next-20130328 (it was painful). Testing indicates that there is no
> dependency on Joseph's patch mentioned above; I could boot, log in, and
> play audio (with local patches) with /or/ without Joseph's patch applied.
> 
> However, I still find that I need to revert f7db706 ARM: "7674/1: smp:
> Avoid dummy clockevent being preferred over real hardware clock-event",
> or I get a hard system hang right after the root fs is mounted. As such,
> I not sure if it's worthwhile applying this series since it's unlikely
> to actually work once merged into Torvalds' tree.
> 
> Do you have any idea what the problem is here?
> 
> I'll try rebasing my local test branch on top of the latest linux-next
> just in case some other fix solved this already, but I'm not going to
> hold my breath.

I've found the problem. My bootloader wasn't initializing the TSC and
hence I guess it wasn't working properly, even with Joseph's patch to
add a call to the initialization function.

So, if I:

a) Fix my bootloader to enable TSC.

b) Merge Rob's patch series to enable the TSC. This is needed rather
than Joseph's patch since the two conflict, and I assume that Rob's
patch is going into 3.10.

c) Undo my local revert of "ARM: 7674/1: smp: Avoid dummy clockevent
being preferred over real hardware clock-event".

... then everything now works.

So the only remaining issues are:

1) Should the mux clocks still be defined in the binding document?
Prashant didn't seem to think so.

2) I need Mike's ack to apply this series to the Tegra tree.