[PATCH 0/5] ARM: shmobile: Enable drivers/sh/pm_runtime.c on multi-platform

[PATCH 0/5] ARM: shmobile: Enable drivers/sh/pm_runtime.c on multi-platform

[Geert Uytterhoeven]

This patchset enables the PM runtime code in drivers/sh/pm_runtime.c when
running a multi-platform ARM kernel including support for shmobile.
Before this code was only enabled for legacy shmobile kernels, leading to
disabled clocks in multiplatform kernels, depending on implicit reset state
or on the bootloader.

  [1/5] drivers: sh: pm_runtime does not need idle callback
  [2/5] sh: intc: Restrict INTC_USERIMASK to SH4A
  [3/5] drivers: sh: compile drivers/sh/pm_runtime.c if ARCH_SHMOBILE_MULTI
  [4/5] ARM: shmobile: Call sh_pm_runtime_init() from platform code on all
        SoCs
  [5/5] drivers: sh: Remove automatic sh_pm_runtime_init() if
        ARCH_SHMOBILE_LEGACY

Patches 1 and 3 were authored by Ben Dooks, with some changes made by me
(for patch 3). Patch 4 was based on a patch by Ben for r8a7790.

After this it should be safe to revert the various commits to "Work around
core clock issues". Note that this is just a port of the existing
sh(mobile)-specifc PM runtime code from legacy to multi-platform, so
expect more work to fix the core issues.

This was tested on Koelsch, both legacy and multi-platform.

Many thanks to Ben Dooks, Laurent Pinchart, and Magnus Damm for the
valuable input, investigations, and discussions.

Thanks for your comments, and test results on other boards/SoCs (esp.
SoCs supporting power domains, like sh7372, r8a7740, and r8a7779)!

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds

Re: [PATCH 0/5] ARM: shmobile: Enable drivers/sh/pm_runtime.c on multi-platform

[Laurent Pinchart]

Hi Geert,

Thank you for the patches.

On Wednesday 19 March 2014 20:06:45 Geert Uytterhoeven wrote:
> This patchset enables the PM runtime code in drivers/sh/pm_runtime.c when
> running a multi-platform ARM kernel including support for shmobile.
> Before this code was only enabled for legacy shmobile kernels, leading to
> disabled clocks in multiplatform kernels, depending on implicit reset state
> or on the bootloader.
> 
>   [1/5] drivers: sh: pm_runtime does not need idle callback
>   [2/5] sh: intc: Restrict INTC_USERIMASK to SH4A
>   [3/5] drivers: sh: compile drivers/sh/pm_runtime.c if ARCH_SHMOBILE_MULTI
>   [4/5] ARM: shmobile: Call sh_pm_runtime_init() from platform code on all
>         SoCs
>   [5/5] drivers: sh: Remove automatic sh_pm_runtime_init() if
>         ARCH_SHMOBILE_LEGACY
> 
> Patches 1 and 3 were authored by Ben Dooks, with some changes made by me
> (for patch 3). Patch 4 was based on a patch by Ben for r8a7790.
> 
> After this it should be safe to revert the various commits to "Work around
> core clock issues". Note that this is just a port of the existing
> sh(mobile)-specifc PM runtime code from legacy to multi-platform, so
> expect more work to fix the core issues.
> 
> This was tested on Koelsch, both legacy and multi-platform.
> 
> Many thanks to Ben Dooks, Laurent Pinchart, and Magnus Damm for the
> valuable input, investigations, and discussions.
> 
> Thanks for your comments, and test results on other boards/SoCs (esp.
> SoCs supporting power domains, like sh7372, r8a7740, and r8a7779)!

For the whole series,

Reviewed-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>

On r8a7779 legacy only (as r8a7779 isn't supported in multiplatform kernels 
yet),

Tested-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>

-- 
Regards,

Laurent Pinchart

Re: [PATCH 0/5] ARM: shmobile: Enable drivers/sh/pm_runtime.c on multi-platform

[Ben Dooks]

On 19/03/14 20:06, Geert Uytterhoeven wrote:
> This patchset enables the PM runtime code in drivers/sh/pm_runtime.c when
> running a multi-platform ARM kernel including support for shmobile.
> Before this code was only enabled for legacy shmobile kernels, leading to
> disabled clocks in multiplatform kernels, depending on implicit reset state
> or on the bootloader.
>
>    [1/5] drivers: sh: pm_runtime does not need idle callback
>    [2/5] sh: intc: Restrict INTC_USERIMASK to SH4A

I thought there was already a of capable intc driver in the
irq drivers directory?

>    [3/5] drivers: sh: compile drivers/sh/pm_runtime.c if ARCH_SHMOBILE_MULTI
>    [4/5] ARM: shmobile: Call sh_pm_runtime_init() from platform code on all
>          SoCs
>    [5/5] drivers: sh: Remove automatic sh_pm_runtime_init() if
>          ARCH_SHMOBILE_LEGACY
>
> Patches 1 and 3 were authored by Ben Dooks, with some changes made by me
> (for patch 3). Patch 4 was based on a patch by Ben for r8a7790.
>
> After this it should be safe to revert the various commits to "Work around
> core clock issues". Note that this is just a port of the existing
> sh(mobile)-specifc PM runtime code from legacy to multi-platform, so
> expect more work to fix the core issues.
>
> This was tested on Koelsch, both legacy and multi-platform.
>
> Many thanks to Ben Dooks, Laurent Pinchart, and Magnus Damm for the
> valuable input, investigations, and discussions.

Thanks, I will probably not get around to testing this until next
week as I am not in the office until Monday now. I think there is
one more thing we need to do, which is to do a device suspend and
resume like the current davinci implementation.




-- 
Ben Dooks				http://www.codethink.co.uk/
Senior Engineer				Codethink - Providing Genius

Re: [PATCH 0/5] ARM: shmobile: Enable drivers/sh/pm_runtime.c on multi-platform

[Geert Uytterhoeven]

Hi Ben,

On Thu, Mar 20, 2014 at 9:56 AM, Ben Dooks <ben.dooks@codethink.co.uk> wrote:
>>    [2/5] sh: intc: Restrict INTC_USERIMASK to SH4A
>
> I thought there was already a of capable intc driver in the
> irq drivers directory?

SH4A doesn't use OF. Or am I missing your point?

> Thanks, I will probably not get around to testing this until next
> week as I am not in the office until Monday now. I think there is
> one more thing we need to do, which is to do a device suspend and
> resume like the current davinci implementation.

Thanks, I'll have a look at that.

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds

Re: [PATCH 0/5] ARM: shmobile: Enable drivers/sh/pm_runtime.c on multi-platform

[Ben Dooks]

On 20/03/14 10:14, Geert Uytterhoeven wrote:
> Hi Ben,
>
> On Thu, Mar 20, 2014 at 9:56 AM, Ben Dooks <ben.dooks@codethink.co.uk> wrote:
>>>     [2/5] sh: intc: Restrict INTC_USERIMASK to SH4A
>>
>> I thought there was already a of capable intc driver in the
>> irq drivers directory?
>
> SH4A doesn't use OF. Or am I missing your point?
>
>> Thanks, I will probably not get around to testing this until next
>> week as I am not in the office until Monday now. I think there is
>> one more thing we need to do, which is to do a device suspend and
>> resume like the current davinci implementation.
>
> Thanks, I'll have a look at that.

I've already done a patch which I am testing for it.


-- 
Ben Dooks				http://www.codethink.co.uk/
Senior Engineer				Codethink - Providing Genius

Re: [PATCH 0/5] ARM: shmobile: Enable drivers/sh/pm_runtime.c on multi-platform

[Laurent Pinchart]

Hi Geert,

On Thursday 20 March 2014 09:56:17 Ben Dooks wrote:
> On 19/03/14 20:06, Geert Uytterhoeven wrote:
> > This patchset enables the PM runtime code in drivers/sh/pm_runtime.c when
> > running a multi-platform ARM kernel including support for shmobile.
> > Before this code was only enabled for legacy shmobile kernels, leading to
> > disabled clocks in multiplatform kernels, depending on implicit reset
> > state or on the bootloader.
> > 
> >    [1/5] drivers: sh: pm_runtime does not need idle callback
> >    [2/5] sh: intc: Restrict INTC_USERIMASK to SH4A
> 
> I thought there was already a of capable intc driver in the
> irq drivers directory?

I think Ben's point is that drivers/sh/intc might be completely unneeded for 
ARM.

> >    [3/5] drivers: sh: compile drivers/sh/pm_runtime.c if
> >    ARCH_SHMOBILE_MULTI
> >    [4/5] ARM: shmobile: Call sh_pm_runtime_init() from platform code on
> >    all
> >    
> >          SoCs
> >    
> >    [5/5] drivers: sh: Remove automatic sh_pm_runtime_init() if
> >    
> >          ARCH_SHMOBILE_LEGACY
> > 
> > Patches 1 and 3 were authored by Ben Dooks, with some changes made by me
> > (for patch 3). Patch 4 was based on a patch by Ben for r8a7790.
> > 
> > After this it should be safe to revert the various commits to "Work around
> > core clock issues". Note that this is just a port of the existing
> > sh(mobile)-specifc PM runtime code from legacy to multi-platform, so
> > expect more work to fix the core issues.
> > 
> > This was tested on Koelsch, both legacy and multi-platform.
> > 
> > Many thanks to Ben Dooks, Laurent Pinchart, and Magnus Damm for the
> > valuable input, investigations, and discussions.
> 
> Thanks, I will probably not get around to testing this until next
> week as I am not in the office until Monday now. I think there is
> one more thing we need to do, which is to do a device suspend and
> resume like the current davinci implementation.

-- 
Regards,

Laurent Pinchart

Re: [PATCH 0/5] ARM: shmobile: Enable drivers/sh/pm_runtime.c on multi-platform

[Geert Uytterhoeven]

Hi Laurent,

On Thu, Mar 20, 2014 at 1:01 PM, Laurent Pinchart
<laurent.pinchart@ideasonboard.com> wrote:
> On Thursday 20 March 2014 09:56:17 Ben Dooks wrote:
>> On 19/03/14 20:06, Geert Uytterhoeven wrote:
>> > This patchset enables the PM runtime code in drivers/sh/pm_runtime.c when
>> > running a multi-platform ARM kernel including support for shmobile.
>> > Before this code was only enabled for legacy shmobile kernels, leading to
>> > disabled clocks in multiplatform kernels, depending on implicit reset
>> > state or on the bootloader.
>> >
>> >    [1/5] drivers: sh: pm_runtime does not need idle callback
>> >    [2/5] sh: intc: Restrict INTC_USERIMASK to SH4A
>>
>> I thought there was already a of capable intc driver in the
>> irq drivers directory?
>
> I think Ben's point is that drivers/sh/intc might be completely unneeded for
> ARM.

http://www.spinics.net/lists/linux-sh/msg30067.html

"Furthermore, the core intc parts are used by sh7372/mackerel and
 sh73a0/kzm9g (and arch/sh) only, so I'm gonna restrict it to
 ARCH_SHMOBILE_LEGACY and SUPERH only."

Despite the names, sh7372 and sh73a0 are ARM.

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds

Re: [PATCH 0/5] ARM: shmobile: Enable drivers/sh/pm_runtime.c on multi-platform

[Geert Uytterhoeven]

On Thu, Mar 20, 2014 at 1:13 PM, Geert Uytterhoeven
<geert@linux-m68k.org> wrote:
>> I think Ben's point is that drivers/sh/intc might be completely unneeded for
>> ARM.
>
> http://www.spinics.net/lists/linux-sh/msg30067.html
>
> "Furthermore, the core intc parts are used by sh7372/mackerel and
>  sh73a0/kzm9g (and arch/sh) only, so I'm gonna restrict it to
>  ARCH_SHMOBILE_LEGACY and SUPERH only."
>
> Despite the names, sh7372 and sh73a0 are ARM.

Hmm, I think the inclusion of drivers/sh/intc could depend on a new
SH_INTC symbol, to be selected by legacy sh, and ARM sh7372/sh73a0?

Gr{oetje,eeting}s,

                        Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds

Re: [PATCH 0/5] ARM: shmobile: Enable drivers/sh/pm_runtime.c on multi-platform

[Laurent Pinchart]

Hi Geert,

On Thursday 20 March 2014 13:13:28 Geert Uytterhoeven wrote:
> On Thu, Mar 20, 2014 at 1:01 PM, Laurent Pinchart wrote:
> > On Thursday 20 March 2014 09:56:17 Ben Dooks wrote:
> >> On 19/03/14 20:06, Geert Uytterhoeven wrote:
> >> > This patchset enables the PM runtime code in drivers/sh/pm_runtime.c
> >> > when running a multi-platform ARM kernel including support for
> >> > shmobile. Before this code was only enabled for legacy shmobile
> >> > kernels, leading to disabled clocks in multiplatform kernels, depending
> >> > on implicit reset state or on the bootloader.
> >> > 
> >> >    [1/5] drivers: sh: pm_runtime does not need idle callback
> >> >    [2/5] sh: intc: Restrict INTC_USERIMASK to SH4A
> >> 
> >> I thought there was already a of capable intc driver in the
> >> irq drivers directory?
> > 
> > I think Ben's point is that drivers/sh/intc might be completely unneeded
> > for ARM.
> 
> http://www.spinics.net/lists/linux-sh/msg30067.html
> 
> "Furthermore, the core intc parts are used by sh7372/mackerel and
>  sh73a0/kzm9g (and arch/sh) only, so I'm gonna restrict it to
>  ARCH_SHMOBILE_LEGACY and SUPERH only."

My bad, sorry. If it wasn't for sh7372 and sh73a0 being old SoCs that's 
probably something we should fix by moving the code to drivers/irqchip.

> Despite the names, sh7372 and sh73a0 are ARM.

-- 
Regards,

Laurent Pinchart

Re: [PATCH 0/5] ARM: shmobile: Enable drivers/sh/pm_runtime.c on multi-platform

[Ben Dooks]

On 20/03/14 13:19, Laurent Pinchart wrote:
> Hi Geert,
>
> On Thursday 20 March 2014 13:13:28 Geert Uytterhoeven wrote:
>> On Thu, Mar 20, 2014 at 1:01 PM, Laurent Pinchart wrote:
>>> On Thursday 20 March 2014 09:56:17 Ben Dooks wrote:
>>>> On 19/03/14 20:06, Geert Uytterhoeven wrote:
>>>>> This patchset enables the PM runtime code in drivers/sh/pm_runtime.c
>>>>> when running a multi-platform ARM kernel including support for
>>>>> shmobile. Before this code was only enabled for legacy shmobile
>>>>> kernels, leading to disabled clocks in multiplatform kernels, depending
>>>>> on implicit reset state or on the bootloader.
>>>>>
>>>>>     [1/5] drivers: sh: pm_runtime does not need idle callback
>>>>>     [2/5] sh: intc: Restrict INTC_USERIMASK to SH4A
>>>>
>>>> I thought there was already a of capable intc driver in the
>>>> irq drivers directory?
>>>
>>> I think Ben's point is that drivers/sh/intc might be completely unneeded
>>> for ARM.
>>
>> http://www.spinics.net/lists/linux-sh/msg30067.html
>>
>> "Furthermore, the core intc parts are used by sh7372/mackerel and
>>   sh73a0/kzm9g (and arch/sh) only, so I'm gonna restrict it to
>>   ARCH_SHMOBILE_LEGACY and SUPERH only."
>
> My bad, sorry. If it wasn't for sh7372 and sh73a0 being old SoCs that's
> probably something we should fix by moving the code to drivers/irqchip.

Something to look at later?

-- 
Ben Dooks				http://www.codethink.co.uk/
Senior Engineer				Codethink - Providing Genius

Re: [PATCH 0/5] ARM: shmobile: Enable drivers/sh/pm_runtime.c on multi-platform

[Kuninori Morimoto]

Hi

I tested these patches with r8a7770 lager on multiplatform DT.
Sound couldn't works without these patches.

Tested-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>

> This patchset enables the PM runtime code in drivers/sh/pm_runtime.c when
> running a multi-platform ARM kernel including support for shmobile.
> Before this code was only enabled for legacy shmobile kernels, leading to
> disabled clocks in multiplatform kernels, depending on implicit reset state
> or on the bootloader.
> 
>   [1/5] drivers: sh: pm_runtime does not need idle callback
>   [2/5] sh: intc: Restrict INTC_USERIMASK to SH4A
>   [3/5] drivers: sh: compile drivers/sh/pm_runtime.c if ARCH_SHMOBILE_MULTI
>   [4/5] ARM: shmobile: Call sh_pm_runtime_init() from platform code on all
>         SoCs
>   [5/5] drivers: sh: Remove automatic sh_pm_runtime_init() if
>         ARCH_SHMOBILE_LEGACY
> 
> Patches 1 and 3 were authored by Ben Dooks, with some changes made by me
> (for patch 3). Patch 4 was based on a patch by Ben for r8a7790.
> 
> After this it should be safe to revert the various commits to "Work around
> core clock issues". Note that this is just a port of the existing
> sh(mobile)-specifc PM runtime code from legacy to multi-platform, so
> expect more work to fix the core issues.
> 
> This was tested on Koelsch, both legacy and multi-platform.
> 
> Many thanks to Ben Dooks, Laurent Pinchart, and Magnus Damm for the
> valuable input, investigations, and discussions.
> 
> Thanks for your comments, and test results on other boards/SoCs (esp.
> SoCs supporting power domains, like sh7372, r8a7740, and r8a7779)!
> 
> Gr{oetje,eeting}s,
> 
> 						Geert
> 
> --
> Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org
> 
> In personal conversations with technical people, I call myself a hacker. But
> when I'm talking to journalists I just say "programmer" or something like that.
> 							    -- Linus Torvalds
> --
> To unsubscribe from this list: send the line "unsubscribe linux-sh" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html