[PATCH] drivers: sh: Use ARCH_RENESAS

[PATCH] drivers: sh: Use ARCH_RENESAS

[Simon Horman]

Make use of ARCH_RENESAS in place of ARCH_SHMOBILE.

This is part of an ongoing process to migrate from ARCH_SHMOBILE to
ARCH_RENESAS the motivation for which being that RENESAS seems to be a more
appropriate name than SHMOBILE for the majority of Renesas ARM based SoCs.

ARCH_RENESAS should cover all cases where both CONFIG_OF and
ARCH_SHMOBILE are enabled.

Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
---
 drivers/sh/pm_runtime.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

 Based on v4.5-rc6

diff --git a/drivers/sh/pm_runtime.c b/drivers/sh/pm_runtime.c
index a9bac3bf20de..aa2ce227e3eb 100644
--- a/drivers/sh/pm_runtime.c
+++ b/drivers/sh/pm_runtime.c
@@ -34,7 +34,7 @@ static struct pm_clk_notifier_block platform_bus_notifier = {
 
 static int __init sh_pm_runtime_init(void)
 {
-	if (IS_ENABLED(CONFIG_OF) && IS_ENABLED(CONFIG_ARCH_SHMOBILE)) {
+	if (IS_ENABLED(CONFIG_ARCH_RENESAS)) {
 		if (!of_find_compatible_node(NULL, NULL,
 					     "renesas,cpg-mstp-clocks"))
 			return 0;
-- 
2.1.4

Re: [PATCH] drivers: sh: Use ARCH_RENESAS

[Geert Uytterhoeven]

Hi Simon,

On Wed, Mar 2, 2016 at 2:43 AM, Simon Horman <horms+renesas@verge.net.au> wrote:
> Make use of ARCH_RENESAS in place of ARCH_SHMOBILE.
>
> This is part of an ongoing process to migrate from ARCH_SHMOBILE to
> ARCH_RENESAS the motivation for which being that RENESAS seems to be a more
> appropriate name than SHMOBILE for the majority of Renesas ARM based SoCs.
>
> ARCH_RENESAS should cover all cases where both CONFIG_OF and
> ARCH_SHMOBILE are enabled.
>
> Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

Acked-by: Geert Uytterhoeven <geert+renesas@glider.be>

If you intend to drop ARCH_SHMOBILE from arch/arm/mach-shmobile/Kconfig before
dropping the whole "if (...) { ... }" block below" (cfr. "drivers: sh: Stop
using the legacy clock domain on ARM",
http://www.spinics.net/lists/linux-renesas-soc/msg00869.html).

Note that the SH-people may resurrect (a variant of) the block when they start
migrating to DT and CCF.

> ---
>  drivers/sh/pm_runtime.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
>  Based on v4.5-rc6
>
> diff --git a/drivers/sh/pm_runtime.c b/drivers/sh/pm_runtime.c
> index a9bac3bf20de..aa2ce227e3eb 100644
> --- a/drivers/sh/pm_runtime.c
> +++ b/drivers/sh/pm_runtime.c
> @@ -34,7 +34,7 @@ static struct pm_clk_notifier_block platform_bus_notifier = {
>
>  static int __init sh_pm_runtime_init(void)
>  {
> -       if (IS_ENABLED(CONFIG_OF) && IS_ENABLED(CONFIG_ARCH_SHMOBILE)) {
> +       if (IS_ENABLED(CONFIG_ARCH_RENESAS)) {
>                 if (!of_find_compatible_node(NULL, NULL,
>                                              "renesas,cpg-mstp-clocks"))
>                         return 0;

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] drivers: sh: Use ARCH_RENESAS

[Simon Horman]

On Wed, Mar 02, 2016 at 09:25:54AM +0100, Geert Uytterhoeven wrote:
> Hi Simon,
> 
> On Wed, Mar 2, 2016 at 2:43 AM, Simon Horman <horms+renesas@verge.net.au> wrote:
> > Make use of ARCH_RENESAS in place of ARCH_SHMOBILE.
> >
> > This is part of an ongoing process to migrate from ARCH_SHMOBILE to
> > ARCH_RENESAS the motivation for which being that RENESAS seems to be a more
> > appropriate name than SHMOBILE for the majority of Renesas ARM based SoCs.
> >
> > ARCH_RENESAS should cover all cases where both CONFIG_OF and
> > ARCH_SHMOBILE are enabled.
> >
> > Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
> 
> Acked-by: Geert Uytterhoeven <geert+renesas@glider.be>
> 
> If you intend to drop ARCH_SHMOBILE from arch/arm/mach-shmobile/Kconfig before
> dropping the whole "if (...) { ... }" block below" (cfr. "drivers: sh: Stop
> using the legacy clock domain on ARM",
> http://www.spinics.net/lists/linux-renesas-soc/msg00869.html).

At this point that is my intention.

> Note that the SH-people may resurrect (a variant of) the block when they start
> migrating to DT and CCF.

Yes, I considered that too.

> > ---
> >  drivers/sh/pm_runtime.c | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> >
> >  Based on v4.5-rc6
> >
> > diff --git a/drivers/sh/pm_runtime.c b/drivers/sh/pm_runtime.c
> > index a9bac3bf20de..aa2ce227e3eb 100644
> > --- a/drivers/sh/pm_runtime.c
> > +++ b/drivers/sh/pm_runtime.c
> > @@ -34,7 +34,7 @@ static struct pm_clk_notifier_block platform_bus_notifier = {
> >
> >  static int __init sh_pm_runtime_init(void)
> >  {
> > -       if (IS_ENABLED(CONFIG_OF) && IS_ENABLED(CONFIG_ARCH_SHMOBILE)) {
> > +       if (IS_ENABLED(CONFIG_ARCH_RENESAS)) {
> >                 if (!of_find_compatible_node(NULL, NULL,
> >                                              "renesas,cpg-mstp-clocks"))
> >                         return 0;
> 
> 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] drivers: sh: Use ARCH_RENESAS

[Simon Horman]

On Thu, Mar 03, 2016 at 09:07:13AM +0900, Simon Horman wrote:
> On Wed, Mar 02, 2016 at 09:25:54AM +0100, Geert Uytterhoeven wrote:
> > Hi Simon,
> > 
> > On Wed, Mar 2, 2016 at 2:43 AM, Simon Horman <horms+renesas@verge.net.au> wrote:
> > > Make use of ARCH_RENESAS in place of ARCH_SHMOBILE.
> > >
> > > This is part of an ongoing process to migrate from ARCH_SHMOBILE to
> > > ARCH_RENESAS the motivation for which being that RENESAS seems to be a more
> > > appropriate name than SHMOBILE for the majority of Renesas ARM based SoCs.
> > >
> > > ARCH_RENESAS should cover all cases where both CONFIG_OF and
> > > ARCH_SHMOBILE are enabled.
> > >
> > > Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
> > 
> > Acked-by: Geert Uytterhoeven <geert+renesas@glider.be>
> > 
> > If you intend to drop ARCH_SHMOBILE from arch/arm/mach-shmobile/Kconfig before
> > dropping the whole "if (...) { ... }" block below" (cfr. "drivers: sh: Stop
> > using the legacy clock domain on ARM",
> > http://www.spinics.net/lists/linux-renesas-soc/msg00869.html).
> 
> At this point that is my intention.

I have queued this patch up for v4.6.

[snip]

Re: [PATCH] drivers: sh: Use ARCH_RENESAS

[Geert Uytterhoeven]

Hi Simon,

On Thu, Mar 3, 2016 at 1:56 AM, Simon Horman <horms@verge.net.au> wrote:
> On Thu, Mar 03, 2016 at 09:07:13AM +0900, Simon Horman wrote:
>> On Wed, Mar 02, 2016 at 09:25:54AM +0100, Geert Uytterhoeven wrote:
>> > On Wed, Mar 2, 2016 at 2:43 AM, Simon Horman <horms+renesas@verge.net.au> wrote:
>> > > Make use of ARCH_RENESAS in place of ARCH_SHMOBILE.
>> > >
>> > > This is part of an ongoing process to migrate from ARCH_SHMOBILE to
>> > > ARCH_RENESAS the motivation for which being that RENESAS seems to be a more
>> > > appropriate name than SHMOBILE for the majority of Renesas ARM based SoCs.
>> > >
>> > > ARCH_RENESAS should cover all cases where both CONFIG_OF and
>> > > ARCH_SHMOBILE are enabled.
>> > >
>> > > Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
>> >
>> > Acked-by: Geert Uytterhoeven <geert+renesas@glider.be>
>> >
>> > If you intend to drop ARCH_SHMOBILE from arch/arm/mach-shmobile/Kconfig before
>> > dropping the whole "if (...) { ... }" block below" (cfr. "drivers: sh: Stop
>> > using the legacy clock domain on ARM",
>> > http://www.spinics.net/lists/linux-renesas-soc/msg00869.html).
>>
>> At this point that is my intention.
>
> I have queued this patch up for v4.6.

BTW, you also have to update drivers/Makefile.

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] drivers: sh: Use ARCH_RENESAS

[Simon Horman]

On Mon, Mar 07, 2016 at 01:47:36PM +0100, Geert Uytterhoeven wrote:
> Hi Simon,
> 
> On Thu, Mar 3, 2016 at 1:56 AM, Simon Horman <horms@verge.net.au> wrote:
> > On Thu, Mar 03, 2016 at 09:07:13AM +0900, Simon Horman wrote:
> >> On Wed, Mar 02, 2016 at 09:25:54AM +0100, Geert Uytterhoeven wrote:
> >> > On Wed, Mar 2, 2016 at 2:43 AM, Simon Horman <horms+renesas@verge.net.au> wrote:
> >> > > Make use of ARCH_RENESAS in place of ARCH_SHMOBILE.
> >> > >
> >> > > This is part of an ongoing process to migrate from ARCH_SHMOBILE to
> >> > > ARCH_RENESAS the motivation for which being that RENESAS seems to be a more
> >> > > appropriate name than SHMOBILE for the majority of Renesas ARM based SoCs.
> >> > >
> >> > > ARCH_RENESAS should cover all cases where both CONFIG_OF and
> >> > > ARCH_SHMOBILE are enabled.
> >> > >
> >> > > Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
> >> >
> >> > Acked-by: Geert Uytterhoeven <geert+renesas@glider.be>
> >> >
> >> > If you intend to drop ARCH_SHMOBILE from arch/arm/mach-shmobile/Kconfig before
> >> > dropping the whole "if (...) { ... }" block below" (cfr. "drivers: sh: Stop
> >> > using the legacy clock domain on ARM",
> >> > http://www.spinics.net/lists/linux-renesas-soc/msg00869.html).
> >>
> >> At this point that is my intention.
> >
> > I have queued this patch up for v4.6.
> 
> BTW, you also have to update drivers/Makefile.

Ok, that is a bit of a tricky one as we would need (ARCH_SHMOBILE ||
ARCH_RENESAS) and I'm not sure how to express that other than by
introducing a cover-all Kconfig symbol, which is more or less the opposite
direction to where I was hoping to go with the transition to ARCH_RENESAS.

In the mean time Renesas ARM-based SoCs need to keep selecting
ARCH_SHMOBILE, which is something I was hoping to change sooner than later.

I think your patches at the link above neatly resolve this. At least for
sh-drives. Perhaps I need to convince myself there is no fallout from
patches 1 and 2 of that series and go ahead and queue up patches 3 and 4.

Re: [PATCH] drivers: sh: Use ARCH_RENESAS

[Simon Horman]

On Fri, Mar 11, 2016 at 10:54:15AM +0900, Simon Horman wrote:
> On Mon, Mar 07, 2016 at 01:47:36PM +0100, Geert Uytterhoeven wrote:
> > Hi Simon,
> > 
> > On Thu, Mar 3, 2016 at 1:56 AM, Simon Horman <horms@verge.net.au> wrote:
> > > On Thu, Mar 03, 2016 at 09:07:13AM +0900, Simon Horman wrote:
> > >> On Wed, Mar 02, 2016 at 09:25:54AM +0100, Geert Uytterhoeven wrote:
> > >> > On Wed, Mar 2, 2016 at 2:43 AM, Simon Horman <horms+renesas@verge.net.au> wrote:
> > >> > > Make use of ARCH_RENESAS in place of ARCH_SHMOBILE.
> > >> > >
> > >> > > This is part of an ongoing process to migrate from ARCH_SHMOBILE to
> > >> > > ARCH_RENESAS the motivation for which being that RENESAS seems to be a more
> > >> > > appropriate name than SHMOBILE for the majority of Renesas ARM based SoCs.
> > >> > >
> > >> > > ARCH_RENESAS should cover all cases where both CONFIG_OF and
> > >> > > ARCH_SHMOBILE are enabled.
> > >> > >
> > >> > > Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
> > >> >
> > >> > Acked-by: Geert Uytterhoeven <geert+renesas@glider.be>
> > >> >
> > >> > If you intend to drop ARCH_SHMOBILE from arch/arm/mach-shmobile/Kconfig before
> > >> > dropping the whole "if (...) { ... }" block below" (cfr. "drivers: sh: Stop
> > >> > using the legacy clock domain on ARM",
> > >> > http://www.spinics.net/lists/linux-renesas-soc/msg00869.html).
> > >>
> > >> At this point that is my intention.
> > >
> > > I have queued this patch up for v4.6.
> > 
> > BTW, you also have to update drivers/Makefile.
> 
> Ok, that is a bit of a tricky one as we would need (ARCH_SHMOBILE ||
> ARCH_RENESAS) and I'm not sure how to express that other than by
> introducing a cover-all Kconfig symbol, which is more or less the opposite
> direction to where I was hoping to go with the transition to ARCH_RENESAS.
> 
> In the mean time Renesas ARM-based SoCs need to keep selecting
> ARCH_SHMOBILE, which is something I was hoping to change sooner than later.
> 
> I think your patches at the link above neatly resolve this. At least for
> sh-drives. Perhaps I need to convince myself there is no fallout from
> patches 1 and 2 of that series and go ahead and queue up patches 3 and 4.

I have decided to drop this patch for now as being a partial solution
it doesn't seem to buy us much. I am currently thinking in terms
of queuing up your patches to drop the whole if block etc... for v4.8
which would neatly give a few cycles for the fallout from the dependencies
to appear.

Geert, how does that plan sound to you?

Re: [PATCH] drivers: sh: Use ARCH_RENESAS

[Rich Felker]

On Wed, Mar 02, 2016 at 09:25:54AM +0100, Geert Uytterhoeven wrote:
> Hi Simon,
> 
> On Wed, Mar 2, 2016 at 2:43 AM, Simon Horman <horms+renesas@verge.net.au> wrote:
> > Make use of ARCH_RENESAS in place of ARCH_SHMOBILE.
> >
> > This is part of an ongoing process to migrate from ARCH_SHMOBILE to
> > ARCH_RENESAS the motivation for which being that RENESAS seems to be a more
> > appropriate name than SHMOBILE for the majority of Renesas ARM based SoCs.
> >
> > ARCH_RENESAS should cover all cases where both CONFIG_OF and
> > ARCH_SHMOBILE are enabled.
> >
> > Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
> 
> Acked-by: Geert Uytterhoeven <geert+renesas@glider.be>
> 
> If you intend to drop ARCH_SHMOBILE from arch/arm/mach-shmobile/Kconfig before
> dropping the whole "if (...) { ... }" block below" (cfr. "drivers: sh: Stop
> using the legacy clock domain on ARM",
> http://www.spinics.net/lists/linux-renesas-soc/msg00869.html).
> 
> Note that the SH-people may resurrect (a variant of) the block when they start
> migrating to DT and CCF.

I can't say for sure at this point, but I think instead of searching
for a node by compatible I would want to represent the relationship
between the nodes semantically in the device tree. We can address that
when we get there, though. If we're stuck with device-provided device
trees that don't fully represent the relationships we might have to
keep kludges for searching.

Rich

Re: [PATCH] drivers: sh: Use ARCH_RENESAS

[Simon Horman]

On Mon, Mar 28, 2016 at 09:16:51AM -0400, Rich Felker wrote:
> On Wed, Mar 02, 2016 at 09:25:54AM +0100, Geert Uytterhoeven wrote:
> > Hi Simon,
> > 
> > On Wed, Mar 2, 2016 at 2:43 AM, Simon Horman <horms+renesas@verge.net.au> wrote:
> > > Make use of ARCH_RENESAS in place of ARCH_SHMOBILE.
> > >
> > > This is part of an ongoing process to migrate from ARCH_SHMOBILE to
> > > ARCH_RENESAS the motivation for which being that RENESAS seems to be a more
> > > appropriate name than SHMOBILE for the majority of Renesas ARM based SoCs.
> > >
> > > ARCH_RENESAS should cover all cases where both CONFIG_OF and
> > > ARCH_SHMOBILE are enabled.
> > >
> > > Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
> > 
> > Acked-by: Geert Uytterhoeven <geert+renesas@glider.be>
> > 
> > If you intend to drop ARCH_SHMOBILE from arch/arm/mach-shmobile/Kconfig before
> > dropping the whole "if (...) { ... }" block below" (cfr. "drivers: sh: Stop
> > using the legacy clock domain on ARM",
> > http://www.spinics.net/lists/linux-renesas-soc/msg00869.html).
> > 
> > Note that the SH-people may resurrect (a variant of) the block when they start
> > migrating to DT and CCF.
> 
> I can't say for sure at this point, but I think instead of searching
> for a node by compatible I would want to represent the relationship
> between the nodes semantically in the device tree. We can address that
> when we get there, though. If we're stuck with device-provided device
> trees that don't fully represent the relationships we might have to
> keep kludges for searching.

Thanks for the feedback.

At this point I think it would be best to drop the block in question.
It should not be difficult to add it back if and when it is needed again.

Re: [PATCH] drivers: sh: Use ARCH_RENESAS

[Geert Uytterhoeven]

Hi Rich,

On Mon, Mar 28, 2016 at 3:16 PM, Rich Felker <dalias@libc.org> wrote:
> On Wed, Mar 02, 2016 at 09:25:54AM +0100, Geert Uytterhoeven wrote:
>> On Wed, Mar 2, 2016 at 2:43 AM, Simon Horman <horms+renesas@verge.net.au> wrote:
>> > Make use of ARCH_RENESAS in place of ARCH_SHMOBILE.
>> >
>> > This is part of an ongoing process to migrate from ARCH_SHMOBILE to
>> > ARCH_RENESAS the motivation for which being that RENESAS seems to be a more
>> > appropriate name than SHMOBILE for the majority of Renesas ARM based SoCs.
>> >
>> > ARCH_RENESAS should cover all cases where both CONFIG_OF and
>> > ARCH_SHMOBILE are enabled.
>> >
>> > Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
>>
>> Acked-by: Geert Uytterhoeven <geert+renesas@glider.be>
>>
>> If you intend to drop ARCH_SHMOBILE from arch/arm/mach-shmobile/Kconfig before
>> dropping the whole "if (...) { ... }" block below" (cfr. "drivers: sh: Stop
>> using the legacy clock domain on ARM",
>> http://www.spinics.net/lists/linux-renesas-soc/msg00869.html).
>>
>> Note that the SH-people may resurrect (a variant of) the block when they start
>> migrating to DT and CCF.
>
> I can't say for sure at this point, but I think instead of searching
> for a node by compatible I would want to represent the relationship
> between the nodes semantically in the device tree. We can address that
> when we get there, though. If we're stuck with device-provided device
> trees that don't fully represent the relationships we might have to
> keep kludges for searching.

Great. That's indeed the best approach to follow.

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