[PATCH] ARM: shmobile: salvage lost CONFIG_REGULATOR

[PATCH] ARM: shmobile: salvage lost CONFIG_REGULATOR

[Kuninori Morimoto]

From: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>

eb4513482d1c4882e3633528ccc48d77581b7bb1
(ARM: shmobile: Enable kzm9g board in multiplatform defconfig)
dropped CONFIG_REGULATOR from shmobile_defconfig.
So, we lost CONFIG_REGULATOR_GPIO or others REGULATORs
Let's salvage these

This patch is using "make savedefconfig"

Reported-by: Magnus Damm <damm+renesas@opensource.se>
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
---
Geert, Magnus

Can you check this patch ?

 arch/arm/configs/shmobile_defconfig |    4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/arch/arm/configs/shmobile_defconfig b/arch/arm/configs/shmobile_defconfig
index 6512c41..46c6623 100644
--- a/arch/arm/configs/shmobile_defconfig
+++ b/arch/arm/configs/shmobile_defconfig
@@ -18,7 +18,6 @@ CONFIG_ARCH_R8A7779=y
 CONFIG_ARCH_R8A7790=y
 CONFIG_ARCH_R8A7791=y
 CONFIG_ARCH_R8A7794=y
-CONFIG_ARCH_SH73A0=y
 CONFIG_MACH_MARZEN=y
 CONFIG_CPU_BPREDICT_DISABLE=y
 CONFIG_PL310_ERRATA_588369=y
@@ -55,7 +54,6 @@ CONFIG_IP_PNP_DHCP=y
 CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
 CONFIG_DEVTMPFS=y
 CONFIG_DEVTMPFS_MOUNT=y
-CONFIG_SIMPLE_PM_BUS=y
 CONFIG_MTD=y
 CONFIG_MTD_BLOCK=y
 CONFIG_MTD_M25P80=y
@@ -118,6 +116,7 @@ CONFIG_WATCHDOG=y
 CONFIG_DA9063_WATCHDOG=y
 CONFIG_MFD_AS3711=y
 CONFIG_MFD_DA9063=y
+CONFIG_REGULATOR=y
 CONFIG_REGULATOR_AS3711=y
 CONFIG_REGULATOR_DA9210=y
 CONFIG_REGULATOR_GPIO=y
@@ -170,7 +169,6 @@ CONFIG_RTC_DRV_RS5C372=y
 CONFIG_RTC_DRV_S35390A=y
 CONFIG_DMADEVICES=y
 CONFIG_SH_DMAE=y
-CONFIG_RCAR_DMAC=y
 # CONFIG_IOMMU_SUPPORT is not set
 CONFIG_IIO=y
 CONFIG_AK8975=y
-- 
1.7.9.5

Re: [PATCH] ARM: shmobile: salvage lost CONFIG_REGULATOR

[Geert Uytterhoeven]

Hi Morimoto-san,

On Fri, Jan 23, 2015 at 8:55 AM, Kuninori Morimoto
<kuninori.morimoto.gx@renesas.com> wrote:
> From: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
>
> eb4513482d1c4882e3633528ccc48d77581b7bb1
> (ARM: shmobile: Enable kzm9g board in multiplatform defconfig)
> dropped CONFIG_REGULATOR from shmobile_defconfig.
> So, we lost CONFIG_REGULATOR_GPIO or others REGULATORs
> Let's salvage these

On which version did you see this problem?
I still have CONFIG_REGULATOR_GPIO=y in the generated .config.

My first guess was that there's a select in -next (and thus in renesas-drivers,
where I generated the patch), that's not yet in renesas-devel (selects are evil,
and cause havoc when combined with savedefconfig).

However, I can't reproduce it on
renesas-devel-20150119-v3.19-rc5 (which still has kzm9g multiplatform),
renesas-drivers-2015-01-19-v3.19-rc5 (which still has kzm9g multiplatform),
renesas-devel-20150123-v3.19-rc5 (which lost kzm9g multiplatform)

Note that you remove:
  - CONFIG_ARCH_SH73A0 and CONFIG_SIMPLE_PM_BUS
    (which indicates your local repo no longer has kzm9g multiplatform)
  - CONFIG_RCAR_DMAC (which is in -next and renesas-drivers,
    but not in renesas-devel).

> This patch is using "make savedefconfig"

Hence that removes everything which is not (yet) in your tree, while I think
we do want to enable features that will only appear in -next, as Simon has
to send pull requests for the defconfigs before the features arrive in Linus'
tree.

> Reported-by: Magnus Damm <damm+renesas@opensource.se>
> Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
> ---
> Geert, Magnus
>
> Can you check this patch ?
>
>  arch/arm/configs/shmobile_defconfig |    4 +---
>  1 file changed, 1 insertion(+), 3 deletions(-)
>
> diff --git a/arch/arm/configs/shmobile_defconfig b/arch/arm/configs/shmobile_defconfig
> index 6512c41..46c6623 100644
> --- a/arch/arm/configs/shmobile_defconfig
> +++ b/arch/arm/configs/shmobile_defconfig
> @@ -18,7 +18,6 @@ CONFIG_ARCH_R8A7779=y
>  CONFIG_ARCH_R8A7790=y
>  CONFIG_ARCH_R8A7791=y
>  CONFIG_ARCH_R8A7794=y
> -CONFIG_ARCH_SH73A0=y
>  CONFIG_MACH_MARZEN=y
>  CONFIG_CPU_BPREDICT_DISABLE=y
>  CONFIG_PL310_ERRATA_588369=y
> @@ -55,7 +54,6 @@ CONFIG_IP_PNP_DHCP=y
>  CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
>  CONFIG_DEVTMPFS=y
>  CONFIG_DEVTMPFS_MOUNT=y
> -CONFIG_SIMPLE_PM_BUS=y
>  CONFIG_MTD=y
>  CONFIG_MTD_BLOCK=y
>  CONFIG_MTD_M25P80=y
> @@ -118,6 +116,7 @@ CONFIG_WATCHDOG=y
>  CONFIG_DA9063_WATCHDOG=y
>  CONFIG_MFD_AS3711=y
>  CONFIG_MFD_DA9063=y
> +CONFIG_REGULATOR=y
>  CONFIG_REGULATOR_AS3711=y
>  CONFIG_REGULATOR_DA9210=y
>  CONFIG_REGULATOR_GPIO=y
> @@ -170,7 +169,6 @@ CONFIG_RTC_DRV_RS5C372=y
>  CONFIG_RTC_DRV_S35390A=y
>  CONFIG_DMADEVICES=y
>  CONFIG_SH_DMAE=y
> -CONFIG_RCAR_DMAC=y
>  # CONFIG_IOMMU_SUPPORT is not set
>  CONFIG_IIO=y
>  CONFIG_AK8975=y

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] ARM: shmobile: salvage lost CONFIG_REGULATOR

[Kuninori Morimoto]

Hi Geert

Thank you for your help

> > eb4513482d1c4882e3633528ccc48d77581b7bb1
> > (ARM: shmobile: Enable kzm9g board in multiplatform defconfig)
> > dropped CONFIG_REGULATOR from shmobile_defconfig.
> > So, we lost CONFIG_REGULATOR_GPIO or others REGULATORs
> > Let's salvage these
> 
> On which version did you see this problem?
> I still have CONFIG_REGULATOR_GPIO=y in the generated .config.
> 
> My first guess was that there's a select in -next (and thus in renesas-drivers,
> where I generated the patch), that's not yet in renesas-devel (selects are evil,
> and cause havoc when combined with savedefconfig).
> 
> However, I can't reproduce it on
> renesas-devel-20150119-v3.19-rc5 (which still has kzm9g multiplatform),
> renesas-drivers-2015-01-19-v3.19-rc5 (which still has kzm9g multiplatform),
> renesas-devel-20150123-v3.19-rc5 (which lost kzm9g multiplatform)

Hmm...

 renesas-devel-20150122-v3.19-rc5 : I (and Magnus) used, and it was broken
 renesas-devel-20150123-v3.19-rc5 : it doens't have issue

So, this patch was not needed.
Thank you for pointing it.

Best regards
---
Kuninori Morimoto

Re: [PATCH] ARM: shmobile: salvage lost CONFIG_REGULATOR

[Simon Horman]

Hi Geert, Hi Morimoto-san,

I believe this problem is present in renesas-next-20150122-v3.19-rc1,
but not in renesas-devel-20150123-v3.19-rc5 due to the presence there of
the following in the latter:

From Magnus Damm <damm+renesas@opensource.se>

ARM: shmobile: sh73a0: Add Multiplatform support

Enable sh73a0 Multiplatform support for the generic sh73a0
machine vector. No board support is enabled, and the board
code for KZM9G DT Reference is left by itself.

Signed-off-by: Magnus Damm <damm+renesas@opensource.se>
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>

diff --git a/arch/arm/mach-shmobile/Kconfig b/arch/arm/mach-shmobile/Kconfig
index 2f36c85..8ee2f28 100644
--- a/arch/arm/mach-shmobile/Kconfig
+++ b/arch/arm/mach-shmobile/Kconfig
@@ -78,6 +78,11 @@ config ARCH_R8A7794
 	bool "R-Car E2 (R8A77940)"
 	select ARCH_RCAR_GEN2
 
+config ARCH_SH73A0
+	bool "SH-Mobile AG5 (R8A73A00)"
+	select ARCH_RMOBILE
+	select RENESAS_INTC_IRQPIN
+
 comment "Renesas ARM SoCs Board Type"
 
 config MACH_MARZEN
[patch truncated]


It may have been better if the defconfig problem that Morimoto-san
mentions below had not gone into next for v3.20 while the change above
ended up being deferred for v3.21. But that is how things have
played out. And the result seems to be that Morimoto-san's patch
will resolve a regression that has been queued up for v3.20.

With the above in mind I plan to queue up the change below
as a fix for v3.20.


On Fri, Jan 23, 2015 at 10:34:41AM +0100, Geert Uytterhoeven wrote:
> Hi Morimoto-san,
> 
> On Fri, Jan 23, 2015 at 8:55 AM, Kuninori Morimoto
> <kuninori.morimoto.gx@renesas.com> wrote:
> > From: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
> >
> > eb4513482d1c4882e3633528ccc48d77581b7bb1
> > (ARM: shmobile: Enable kzm9g board in multiplatform defconfig)
> > dropped CONFIG_REGULATOR from shmobile_defconfig.
> > So, we lost CONFIG_REGULATOR_GPIO or others REGULATORs
> > Let's salvage these
> 
> On which version did you see this problem?
> I still have CONFIG_REGULATOR_GPIO=y in the generated .config.
> 
> My first guess was that there's a select in -next (and thus in renesas-drivers,
> where I generated the patch), that's not yet in renesas-devel (selects are evil,
> and cause havoc when combined with savedefconfig).
> 
> However, I can't reproduce it on
> renesas-devel-20150119-v3.19-rc5 (which still has kzm9g multiplatform),
> renesas-drivers-2015-01-19-v3.19-rc5 (which still has kzm9g multiplatform),
> renesas-devel-20150123-v3.19-rc5 (which lost kzm9g multiplatform)
> 
> Note that you remove:
>   - CONFIG_ARCH_SH73A0 and CONFIG_SIMPLE_PM_BUS
>     (which indicates your local repo no longer has kzm9g multiplatform)
>   - CONFIG_RCAR_DMAC (which is in -next and renesas-drivers,
>     but not in renesas-devel).
> 
> > This patch is using "make savedefconfig"
> 
> Hence that removes everything which is not (yet) in your tree, while I think
> we do want to enable features that will only appear in -next, as Simon has
> to send pull requests for the defconfigs before the features arrive in Linus'
> tree.
> 
> > Reported-by: Magnus Damm <damm+renesas@opensource.se>
> > Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
> > ---
> > Geert, Magnus
> >
> > Can you check this patch ?
> >
> >  arch/arm/configs/shmobile_defconfig |    4 +---
> >  1 file changed, 1 insertion(+), 3 deletions(-)
> >
> > diff --git a/arch/arm/configs/shmobile_defconfig b/arch/arm/configs/shmobile_defconfig
> > index 6512c41..46c6623 100644
> > --- a/arch/arm/configs/shmobile_defconfig
> > +++ b/arch/arm/configs/shmobile_defconfig
> > @@ -18,7 +18,6 @@ CONFIG_ARCH_R8A7779=y
> >  CONFIG_ARCH_R8A7790=y
> >  CONFIG_ARCH_R8A7791=y
> >  CONFIG_ARCH_R8A7794=y
> > -CONFIG_ARCH_SH73A0=y
> >  CONFIG_MACH_MARZEN=y
> >  CONFIG_CPU_BPREDICT_DISABLE=y
> >  CONFIG_PL310_ERRATA_588369=y
> > @@ -55,7 +54,6 @@ CONFIG_IP_PNP_DHCP=y
> >  CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
> >  CONFIG_DEVTMPFS=y
> >  CONFIG_DEVTMPFS_MOUNT=y
> > -CONFIG_SIMPLE_PM_BUS=y
> >  CONFIG_MTD=y
> >  CONFIG_MTD_BLOCK=y
> >  CONFIG_MTD_M25P80=y
> > @@ -118,6 +116,7 @@ CONFIG_WATCHDOG=y
> >  CONFIG_DA9063_WATCHDOG=y
> >  CONFIG_MFD_AS3711=y
> >  CONFIG_MFD_DA9063=y
> > +CONFIG_REGULATOR=y
> >  CONFIG_REGULATOR_AS3711=y
> >  CONFIG_REGULATOR_DA9210=y
> >  CONFIG_REGULATOR_GPIO=y
> > @@ -170,7 +169,6 @@ CONFIG_RTC_DRV_RS5C372=y
> >  CONFIG_RTC_DRV_S35390A=y
> >  CONFIG_DMADEVICES=y
> >  CONFIG_SH_DMAE=y
> > -CONFIG_RCAR_DMAC=y
> >  # CONFIG_IOMMU_SUPPORT is not set
> >  CONFIG_IIO=y
> >  CONFIG_AK8975=y
> 
> 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] ARM: shmobile: salvage lost CONFIG_REGULATOR

[Simon Horman]

On Mon, Jan 26, 2015 at 10:24:04AM +0900, Simon Horman wrote:
> Hi Geert, Hi Morimoto-san,
> 
> I believe this problem is present in renesas-next-20150122-v3.19-rc1,
> but not in renesas-devel-20150123-v3.19-rc5 due to the presence there of
> the following in the latter:
> 
> >From Magnus Damm <damm+renesas@opensource.se>
> 
> ARM: shmobile: sh73a0: Add Multiplatform support
> 
> Enable sh73a0 Multiplatform support for the generic sh73a0
> machine vector. No board support is enabled, and the board
> code for KZM9G DT Reference is left by itself.
> 
> Signed-off-by: Magnus Damm <damm+renesas@opensource.se>
> Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
> Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
> 
> diff --git a/arch/arm/mach-shmobile/Kconfig b/arch/arm/mach-shmobile/Kconfig
> index 2f36c85..8ee2f28 100644
> --- a/arch/arm/mach-shmobile/Kconfig
> +++ b/arch/arm/mach-shmobile/Kconfig
> @@ -78,6 +78,11 @@ config ARCH_R8A7794
>  	bool "R-Car E2 (R8A77940)"
>  	select ARCH_RCAR_GEN2
>  
> +config ARCH_SH73A0
> +	bool "SH-Mobile AG5 (R8A73A00)"
> +	select ARCH_RMOBILE
> +	select RENESAS_INTC_IRQPIN
> +
>  comment "Renesas ARM SoCs Board Type"
>  
>  config MACH_MARZEN
> [patch truncated]
> 
> 
> It may have been better if the defconfig problem that Morimoto-san
> mentions below had not gone into next for v3.20 while the change above
> ended up being deferred for v3.21. But that is how things have
> played out. And the result seems to be that Morimoto-san's patch
> will resolve a regression that has been queued up for v3.20.
> 
> With the above in mind I plan to queue up the change below
> as a fix for v3.20.

As I would like to treat this as a fix I have made a more minimal
version of Morimoto-san's patch. I have also  updated the changelog to
supply a bit more information.

I will post the result as "[PATCH] ARM: shmobile: Select CONFIG_REGULATOR
in defconfig once again".

Re: [PATCH] ARM: shmobile: salvage lost CONFIG_REGULATOR

[Kuninori Morimoto]

Hi Simon

> I believe this problem is present in renesas-next-20150122-v3.19-rc1,
> but not in renesas-devel-20150123-v3.19-rc5 due to the presence there of
> the following in the latter:
(snip)
> It may have been better if the defconfig problem that Morimoto-san
> mentions below had not gone into next for v3.20 while the change above
> ended up being deferred for v3.21. But that is how things have
> played out. And the result seems to be that Morimoto-san's patch
> will resolve a regression that has been queued up for v3.20.
> 
> With the above in mind I plan to queue up the change below
> as a fix for v3.20.

Thank you for your help.
My patch is including below, because "make savedefconfig" did it

-CONFIG_ARCH_SH73A0=y
-CONFIG_SIMPLE_PM_BUS=y
-CONFIG_RCAR_DMAC=y

But, if these are based on merged branch (= Geert mentioned to it),
I think I should send v2 (which includes REGURALOR only),
otherwise, it will create other issues.

My patch was based on renesas-devel-20150122-v3.19-rc5.
but, can you show me best branch/tag for this issue ?

Best regards
---
Kuninori Morimoto

Re: [PATCH] ARM: shmobile: salvage lost CONFIG_REGULATOR

[Simon Horman]

On Mon, Jan 26, 2015 at 02:28:26AM +0000, Kuninori Morimoto wrote:
> 
> Hi Simon
> 
> > I believe this problem is present in renesas-next-20150122-v3.19-rc1,
> > but not in renesas-devel-20150123-v3.19-rc5 due to the presence there of
> > the following in the latter:
> (snip)
> > It may have been better if the defconfig problem that Morimoto-san
> > mentions below had not gone into next for v3.20 while the change above
> > ended up being deferred for v3.21. But that is how things have
> > played out. And the result seems to be that Morimoto-san's patch
> > will resolve a regression that has been queued up for v3.20.
> > 
> > With the above in mind I plan to queue up the change below
> > as a fix for v3.20.
> 
> Thank you for your help.
> My patch is including below, because "make savedefconfig" did it
> 
> -CONFIG_ARCH_SH73A0=y
> -CONFIG_SIMPLE_PM_BUS=y
> -CONFIG_RCAR_DMAC=y
> 
> But, if these are based on merged branch (= Geert mentioned to it),
> I think I should send v2 (which includes REGURALOR only),
> otherwise, it will create other issues.
> 
> My patch was based on renesas-devel-20150122-v3.19-rc5.
> but, can you show me best branch/tag for this issue ?

renesas-next-20150122-v3.19-rc1

The config items above would probably also be removed by savedefconfig
in renesas-next-20150122-v3.19-rc1 too. However, I think a minimal
approach is best for a fix: just removing REGURALOR as you suggest.

As you (now) know I have sent a patch to do that, which
I suppose crossed in-flight with your email.