Re: [PATCH v4 0/8] ARM: davinci: complete the conversion to using the reset framework

From: Bartosz Golaszewski
Date: Tue Mar 27 2018 - 05:30:02 EST


2018-03-27 8:09 GMT+02:00 Sekhar Nori <nsekhar@xxxxxx>:
> Hi Bart,
>
> On Friday 23 March 2018 10:46 PM, Bartosz Golaszewski wrote:
>> 2018-03-23 18:08 GMT+01:00 Stephen Boyd <sboyd@xxxxxxxxxx>:
>>> Quoting Bartosz Golaszewski (2018-03-23 09:55:47)
>>>> 2018-03-23 17:49 GMT+01:00 Stephen Boyd <sboyd@xxxxxxxxxx>:
>>>>> Quoting Bartosz Golaszewski (2018-03-23 06:04:47)
>>>>>> From: Bartosz Golaszewski <bgolaszewski@xxxxxxxxxxxx>
>>>>>>
>>>>>> This series converts the only user of the handcoded, mach-specific reset
>>>>>> routines in the davinci platform to using the reset framework.
>>>>>>
>>>>>> Patch 1 modifies the way lookup entries are registered with the reset
>>>>>> framework.
>>>>>>
>>>>>> Patches 2-4 add necessary lookups/DT-properties.
>>>>>>
>>>>>> Patches 5-7 convert the davinci-rproc driver to the reset framework.
>>>>>>
>>>>>> Patch 8 removes now dead code.
>>>>>>
>>>>>> Philipp: it turned out that it's indeed better to use the reset
>>>>>> controller's device name for the entry lookup.
>>>>>>
>>>>>> Tested both in DT and legacy modes by booting the examples from
>>>>>> ti-ipc-rtos recipe in meta-ti.
>>>>>>
>>>>>> This series applies on top of David Lechner's common-clk-v9 branch[1]
>>>>>> with Philipp Zabel's reset/next branch[2] pulled in.
>>>>>>
>>>>>> It can be found in my github tree as well[3].
>>>>>>
>>>>>
>>>>> What's the merge strategy for the rest of the patches? I should apply
>>>>> the clk ones after the next -rc1?
>>>>
>>>> Or maybe Philipp can provide us with an immutable branch with the reset patches?
>>>>
>>>> The you could apply the driver patches and let Sekhar take all the
>>>> platform code?
>>>>
>>>
>>> Ok that could work too.
>>
>> Ohad, Bjorn can you ack the remoteproc patches? Are you OK with
>> Stephen taking them through the clock tree? Otherwise it would get
>> complicated since they depend on the first clk patch and the last clk
>> patch depends on them.
>
> I will not be queuing the DTS patches for v4.17. They depend on clock
> framework DTS patches which itself I plan to queue for v4.18. Given
> this, I think the best bet is:
>
> Ohad/Bjorn queue remoteproc patches already acked by Suman and having no
> dependencies. That is, 5/8 and 6/8.
>
> If Philipp can provide an immutable branch with reset changes, it will
> be great and Stephen can queue 4/8. If not, its best to resend that
> patch to Stephen once v4.17-rc1 is out.
>
> The remaining patches need to wait till v4.18 (or even v4.19 if
> dependencies don't pan out well).
>
> Thanks,
> Sekhar

Very good, thanks.

I've just sent out a v3 with three non-reset remoteproc patches that
can already be applied by Suman.

Thanks,
Bartosz