mbox series

[00/11] clk: ti: clkctrl data split based on clkdm boundaries

Message ID 1535728027-24573-1-git-send-email-t-kristo@ti.com (mailing list archive)
Headers show
Series clk: ti: clkctrl data split based on clkdm boundaries | expand

Message

Tero Kristo Aug. 31, 2018, 3:06 p.m. UTC
Hi,

This series splits up the clkctrl data for TI SoCs based on clockdomain
boundaries, instead of the current CM boundaries. For rest of the SoCs,
the CM boundaries are actually the same as clockdomains, only am3/am4/dra7
are mutants.

This is done in preparation for better supporting Tony's upcoming
interconnect driver data introduction for these SoCs.

One thing to note is that we maintain backwards compatibility with this
series, the compat data is introduced temporarily for this purpose. The
compat data can be ditched after couple of release cycles when nobody
needs it anymore.

The series can be applied independently, as the compat data support
makes care that the switch is only done once DT changes are in place.

-Tero

--
Texas Instruments Finland Oy, Porkkalankatu 22, 00180 Helsinki. Y-tunnus/Business ID: 0615521-4. Kotipaikka/Domicile: Helsinki

Comments

Tony Lindgren Sept. 4, 2018, 3:36 p.m. UTC | #1
* Tero Kristo <t-kristo@ti.com> [180831 15:11]:
> Hi,
> 
> This series splits up the clkctrl data for TI SoCs based on clockdomain
> boundaries, instead of the current CM boundaries. For rest of the SoCs,
> the CM boundaries are actually the same as clockdomains, only am3/am4/dra7
> are mutants.
> 
> This is done in preparation for better supporting Tony's upcoming
> interconnect driver data introduction for these SoCs.
> 
> One thing to note is that we maintain backwards compatibility with this
> series, the compat data is introduced temporarily for this purpose. The
> compat data can be ditched after couple of release cycles when nobody
> needs it anymore.
> 
> The series can be applied independently, as the compat data support
> makes care that the switch is only done once DT changes are in place.

When this is ready for merging, can you guys please set up an
immutable branch against v4.19-rc1 with the drivers/clk changes
for me to merge in too?

I've tested this on am335x and am437x so far and it works fine
for me, so for the whole series:

Tested-by: Tony Lindgren <tony@atomide.com>
Tony Lindgren Sept. 26, 2018, 7:13 p.m. UTC | #2
Hi,

* Tony Lindgren <tony@atomide.com> [180904 15:41]:
> * Tero Kristo <t-kristo@ti.com> [180831 15:11]:
> > Hi,
> > 
> > This series splits up the clkctrl data for TI SoCs based on clockdomain
> > boundaries, instead of the current CM boundaries. For rest of the SoCs,
> > the CM boundaries are actually the same as clockdomains, only am3/am4/dra7
> > are mutants.
> > 
> > This is done in preparation for better supporting Tony's upcoming
> > interconnect driver data introduction for these SoCs.
> > 
> > One thing to note is that we maintain backwards compatibility with this
> > series, the compat data is introduced temporarily for this purpose. The
> > compat data can be ditched after couple of release cycles when nobody
> > needs it anymore.
> > 
> > The series can be applied independently, as the compat data support
> > makes care that the switch is only done once DT changes are in place.
> 
> When this is ready for merging, can you guys please set up an
> immutable branch against v4.19-rc1 with the drivers/clk changes
> for me to merge in too?
> 
> I've tested this on am335x and am437x so far and it works fine
> for me, so for the whole series:
> 
> Tested-by: Tony Lindgren <tony@atomide.com>

Any news on getting this into Linux next and merged? The
immutable is still needed too for the dts changes.

Regards,

Tony
Tero Kristo Sept. 27, 2018, 6:31 a.m. UTC | #3
On 26/09/18 22:13, Tony Lindgren wrote:
> Hi,
> 
> * Tony Lindgren <tony@atomide.com> [180904 15:41]:
>> * Tero Kristo <t-kristo@ti.com> [180831 15:11]:
>>> Hi,
>>>
>>> This series splits up the clkctrl data for TI SoCs based on clockdomain
>>> boundaries, instead of the current CM boundaries. For rest of the SoCs,
>>> the CM boundaries are actually the same as clockdomains, only am3/am4/dra7
>>> are mutants.
>>>
>>> This is done in preparation for better supporting Tony's upcoming
>>> interconnect driver data introduction for these SoCs.
>>>
>>> One thing to note is that we maintain backwards compatibility with this
>>> series, the compat data is introduced temporarily for this purpose. The
>>> compat data can be ditched after couple of release cycles when nobody
>>> needs it anymore.
>>>
>>> The series can be applied independently, as the compat data support
>>> makes care that the switch is only done once DT changes are in place.
>>
>> When this is ready for merging, can you guys please set up an
>> immutable branch against v4.19-rc1 with the drivers/clk changes
>> for me to merge in too?
>>
>> I've tested this on am335x and am437x so far and it works fine
>> for me, so for the whole series:
>>
>> Tested-by: Tony Lindgren <tony@atomide.com>
> 
> Any news on getting this into Linux next and merged? The
> immutable is still needed too for the dts changes.
> 

This is currently blocked from my side, as the patch [1] from Johan got 
in clk-next is conflicting with my series. Asked Stephen for either 
dropping that patch and letting me pick it up but no response so far.

[1] https://lkml.org/lkml/2018/8/22/160

-Tero
--
Texas Instruments Finland Oy, Porkkalankatu 22, 00180 Helsinki. Y-tunnus/Business ID: 0615521-4. Kotipaikka/Domicile: Helsinki
Tony Lindgren Sept. 27, 2018, 2:49 p.m. UTC | #4
* Tero Kristo <t-kristo@ti.com> [180927 06:35]:
> On 26/09/18 22:13, Tony Lindgren wrote:
> > Hi,
> > 
> > * Tony Lindgren <tony@atomide.com> [180904 15:41]:
> > > * Tero Kristo <t-kristo@ti.com> [180831 15:11]:
> > > > Hi,
> > > > 
> > > > This series splits up the clkctrl data for TI SoCs based on clockdomain
> > > > boundaries, instead of the current CM boundaries. For rest of the SoCs,
> > > > the CM boundaries are actually the same as clockdomains, only am3/am4/dra7
> > > > are mutants.
> > > > 
> > > > This is done in preparation for better supporting Tony's upcoming
> > > > interconnect driver data introduction for these SoCs.
> > > > 
> > > > One thing to note is that we maintain backwards compatibility with this
> > > > series, the compat data is introduced temporarily for this purpose. The
> > > > compat data can be ditched after couple of release cycles when nobody
> > > > needs it anymore.
> > > > 
> > > > The series can be applied independently, as the compat data support
> > > > makes care that the switch is only done once DT changes are in place.
> > > 
> > > When this is ready for merging, can you guys please set up an
> > > immutable branch against v4.19-rc1 with the drivers/clk changes
> > > for me to merge in too?
> > > 
> > > I've tested this on am335x and am437x so far and it works fine
> > > for me, so for the whole series:
> > > 
> > > Tested-by: Tony Lindgren <tony@atomide.com>
> > 
> > Any news on getting this into Linux next and merged? The
> > immutable is still needed too for the dts changes.
> > 
> 
> This is currently blocked from my side, as the patch [1] from Johan got in
> clk-next is conflicting with my series. Asked Stephen for either dropping
> that patch and letting me pick it up but no response so far.
> 
> [1] https://lkml.org/lkml/2018/8/22/160

Hmm well the idea is to not redo branches. So how about
send Stephen a pull request against Johan's patch commit
00a461cc32ec ("clk: ti: fix OF child-node lookup")?

That is assuming 00a461cc32ec is considered immutable and
OK to use as a base by Stephen? :)

And assuming that your pull request gets merged, and we
all consider that branch immutable too, then I can merge it
in too for the dts changes. It will bring also Rob's
clean-up patch 00a461cc32ec ("clk: ti: fix OF child-node
lookup") too but that's just fine.

Regards,

Tony
Tero Kristo Oct. 3, 2018, 1:17 p.m. UTC | #5
On 27/09/18 17:49, Tony Lindgren wrote:
> * Tero Kristo <t-kristo@ti.com> [180927 06:35]:
>> On 26/09/18 22:13, Tony Lindgren wrote:
>>> Hi,
>>>
>>> * Tony Lindgren <tony@atomide.com> [180904 15:41]:
>>>> * Tero Kristo <t-kristo@ti.com> [180831 15:11]:
>>>>> Hi,
>>>>>
>>>>> This series splits up the clkctrl data for TI SoCs based on clockdomain
>>>>> boundaries, instead of the current CM boundaries. For rest of the SoCs,
>>>>> the CM boundaries are actually the same as clockdomains, only am3/am4/dra7
>>>>> are mutants.
>>>>>
>>>>> This is done in preparation for better supporting Tony's upcoming
>>>>> interconnect driver data introduction for these SoCs.
>>>>>
>>>>> One thing to note is that we maintain backwards compatibility with this
>>>>> series, the compat data is introduced temporarily for this purpose. The
>>>>> compat data can be ditched after couple of release cycles when nobody
>>>>> needs it anymore.
>>>>>
>>>>> The series can be applied independently, as the compat data support
>>>>> makes care that the switch is only done once DT changes are in place.
>>>>
>>>> When this is ready for merging, can you guys please set up an
>>>> immutable branch against v4.19-rc1 with the drivers/clk changes
>>>> for me to merge in too?
>>>>
>>>> I've tested this on am335x and am437x so far and it works fine
>>>> for me, so for the whole series:
>>>>
>>>> Tested-by: Tony Lindgren <tony@atomide.com>
>>>
>>> Any news on getting this into Linux next and merged? The
>>> immutable is still needed too for the dts changes.
>>>
>>
>> This is currently blocked from my side, as the patch [1] from Johan got in
>> clk-next is conflicting with my series. Asked Stephen for either dropping
>> that patch and letting me pick it up but no response so far.
>>
>> [1] https://lkml.org/lkml/2018/8/22/160
> 
> Hmm well the idea is to not redo branches. So how about
> send Stephen a pull request against Johan's patch commit
> 00a461cc32ec ("clk: ti: fix OF child-node lookup")?
> 
> That is assuming 00a461cc32ec is considered immutable and
> OK to use as a base by Stephen? :)
> 
> And assuming that your pull request gets merged, and we
> all consider that branch immutable too, then I can merge it
> in too for the dts changes. It will bring also Rob's
> clean-up patch 00a461cc32ec ("clk: ti: fix OF child-node
> lookup") too but that's just fine.
> 
> Regards,
> 
> Tony
> 

So, finally queued this up towards 4.20, assuming Stephen accepts it.

The immutable branch I have pushed is available here:

https://git.kernel.org/pub/scm/linux/kernel/git/kristo/linux.git/log/?h=clk-ti-clkctrl-sr

It is also tagged in the pull-req I just sent out.

-Tero
--
Texas Instruments Finland Oy, Porkkalankatu 22, 00180 Helsinki. Y-tunnus/Business ID: 0615521-4. Kotipaikka/Domicile: Helsinki