* [PATCH 01/04] ARM: shmobile: Only build clocks when COMMON_CLK=n
2013-07-09 5:43 [PATCH 00/04] ARM: shmobile: ARCH_MULTIPLATFORM for EMEV2 and KZM9D Magnus Damm
@ 2013-07-09 5:43 ` Magnus Damm
2013-07-09 7:43 ` Arnd Bergmann
2013-07-09 5:43 ` [PATCH 02/04] ARM: shmobile: Introduce ARCH_SHMOBILE_MULTI Magnus Damm
` (3 subsequent siblings)
4 siblings, 1 reply; 16+ messages in thread
From: Magnus Damm @ 2013-07-09 5:43 UTC (permalink / raw)
To: linux-arm-kernel
From: Magnus Damm <damm@opensource.se>
Move shared clock.c file and per-SoC clock-xxx.c files
to only build when CONFIG_COMMON_CLK != y.
The regular ARCH_SHMOBILE case with legacy SH clock framework
will build just as before with this patch applied, however
the case when COMMON_CLK=y will exclude all the clock files.
Signed-off-by: Magnus Damm <damm@opensource.se>
---
arch/arm/mach-shmobile/Makefile | 30 +++++++++++++++++++++---------
1 file changed, 21 insertions(+), 9 deletions(-)
--- 0002/arch/arm/mach-shmobile/Makefile
+++ work/arch/arm/mach-shmobile/Makefile 2013-07-09 12:21:29.000000000 +0900
@@ -3,17 +3,28 @@
#
# Common objects
-obj-y := timer.o console.o clock.o
+obj-y := timer.o console.o
# CPU objects
-obj-$(CONFIG_ARCH_SH7372) += setup-sh7372.o clock-sh7372.o intc-sh7372.o
-obj-$(CONFIG_ARCH_SH73A0) += setup-sh73a0.o clock-sh73a0.o intc-sh73a0.o
-obj-$(CONFIG_ARCH_R8A73A4) += setup-r8a73a4.o clock-r8a73a4.o
-obj-$(CONFIG_ARCH_R8A7740) += setup-r8a7740.o clock-r8a7740.o intc-r8a7740.o
-obj-$(CONFIG_ARCH_R8A7778) += setup-r8a7778.o clock-r8a7778.o
-obj-$(CONFIG_ARCH_R8A7779) += setup-r8a7779.o clock-r8a7779.o intc-r8a7779.o
-obj-$(CONFIG_ARCH_R8A7790) += setup-r8a7790.o clock-r8a7790.o
-obj-$(CONFIG_ARCH_EMEV2) += setup-emev2.o clock-emev2.o
+obj-$(CONFIG_ARCH_SH7372) += setup-sh7372.o intc-sh7372.o
+obj-$(CONFIG_ARCH_SH73A0) += setup-sh73a0.o intc-sh73a0.o
+obj-$(CONFIG_ARCH_R8A73A4) += setup-r8a73a4.o
+obj-$(CONFIG_ARCH_R8A7740) += setup-r8a7740.o intc-r8a7740.o
+obj-$(CONFIG_ARCH_R8A7778) += setup-r8a7778.o
+obj-$(CONFIG_ARCH_R8A7779) += setup-r8a7779.o intc-r8a7779.o
+obj-$(CONFIG_ARCH_R8A7790) += setup-r8a7790.o
+obj-$(CONFIG_ARCH_EMEV2) += setup-emev2.o
+
+# Clock objects
+clk-y := clock.o
+clk-$(CONFIG_ARCH_SH7372) += clock-sh7372.o
+clk-$(CONFIG_ARCH_SH73A0) += clock-sh73a0.o
+clk-$(CONFIG_ARCH_R8A73A4) += clock-r8a73a4.o
+clk-$(CONFIG_ARCH_R8A7740) += clock-r8a7740.o
+clk-$(CONFIG_ARCH_R8A7778) += clock-r8a7778.o
+clk-$(CONFIG_ARCH_R8A7779) += clock-r8a7779.o
+clk-$(CONFIG_ARCH_R8A7790) += clock-r8a7790.o
+clk-$(CONFIG_ARCH_EMEV2) += clock-emev2.o
# SMP objects
smp-y := platsmp.o headsmp.o
@@ -53,3 +64,4 @@ obj-$(CONFIG_MACH_KZM9G_REFERENCE) += bo
# Framework support
obj-$(CONFIG_SMP) += $(smp-y)
+obj-y += $(clk-y$(CONFIG_COMMON_CLK))
^ permalink raw reply [flat|nested] 16+ messages in thread* [PATCH 01/04] ARM: shmobile: Only build clocks when COMMON_CLK=n
2013-07-09 5:43 ` [PATCH 01/04] ARM: shmobile: Only build clocks when COMMON_CLK=n Magnus Damm
@ 2013-07-09 7:43 ` Arnd Bergmann
2013-07-09 7:45 ` Laurent Pinchart
0 siblings, 1 reply; 16+ messages in thread
From: Arnd Bergmann @ 2013-07-09 7:43 UTC (permalink / raw)
To: linux-arm-kernel
On Tuesday 09 July 2013, Magnus Damm wrote:
> +obj-y += $(clk-y$(CONFIG_COMMON_CLK))
It took me a bit to understand the Makefile trick you are using here.
I assume other readers will have the same problem. How about just
doing
+ifndef CONFIG_COMMON_CLK
obj-$(CONFIG_ARCH_SH7372) += setup-sh7372.o intc-sh7372.o
obj-$(CONFIG_ARCH_SH73A0) += setup-sh73a0.o intc-sh73a0.o
obj-$(CONFIG_ARCH_R8A73A4) += setup-r8a73a4.o
obj-$(CONFIG_ARCH_R8A7740) += setup-r8a7740.o intc-r8a7740.o
obj-$(CONFIG_ARCH_R8A7778) += setup-r8a7778.o
obj-$(CONFIG_ARCH_R8A7779) += setup-r8a7779.o intc-r8a7779.o
obj-$(CONFIG_ARCH_R8A7790) += setup-r8a7790.o
obj-$(CONFIG_ARCH_EMEV2) += setup-emev2.o
+endif
Arnd
^ permalink raw reply [flat|nested] 16+ messages in thread
* [PATCH 01/04] ARM: shmobile: Only build clocks when COMMON_CLK=n
2013-07-09 7:43 ` Arnd Bergmann
@ 2013-07-09 7:45 ` Laurent Pinchart
2013-07-09 9:09 ` Magnus Damm
0 siblings, 1 reply; 16+ messages in thread
From: Laurent Pinchart @ 2013-07-09 7:45 UTC (permalink / raw)
To: linux-arm-kernel
Hi Arnd and Magnus,
On Tuesday 09 July 2013 09:43:07 Arnd Bergmann wrote:
> On Tuesday 09 July 2013, Magnus Damm wrote:
> > +obj-y += $(clk-y$(CONFIG_COMMON_CLK))
>
> It took me a bit to understand the Makefile trick you are using here.
> I assume other readers will have the same problem. How about just
> doing
>
> +ifndef CONFIG_COMMON_CLK
> obj-$(CONFIG_ARCH_SH7372) += setup-sh7372.o intc-sh7372.o
> obj-$(CONFIG_ARCH_SH73A0) += setup-sh73a0.o intc-sh73a0.o
> obj-$(CONFIG_ARCH_R8A73A4) += setup-r8a73a4.o
> obj-$(CONFIG_ARCH_R8A7740) += setup-r8a7740.o intc-r8a7740.o
> obj-$(CONFIG_ARCH_R8A7778) += setup-r8a7778.o
> obj-$(CONFIG_ARCH_R8A7779) += setup-r8a7779.o intc-r8a7779.o
> obj-$(CONFIG_ARCH_R8A7790) += setup-r8a7790.o
> obj-$(CONFIG_ARCH_EMEV2) += setup-emev2.o
> +endif
I was about to mention the same, so I think that would be a good idea.
--
Regards,
Laurent Pinchart
^ permalink raw reply [flat|nested] 16+ messages in thread
* [PATCH 01/04] ARM: shmobile: Only build clocks when COMMON_CLK=n
2013-07-09 7:45 ` Laurent Pinchart
@ 2013-07-09 9:09 ` Magnus Damm
2013-07-09 21:06 ` Arnd Bergmann
0 siblings, 1 reply; 16+ messages in thread
From: Magnus Damm @ 2013-07-09 9:09 UTC (permalink / raw)
To: linux-arm-kernel
Hi Laurent and Arnd,
On Tue, Jul 9, 2013 at 4:45 PM, Laurent Pinchart
<laurent.pinchart@ideasonboard.com> wrote:
> Hi Arnd and Magnus,
>
> On Tuesday 09 July 2013 09:43:07 Arnd Bergmann wrote:
>> On Tuesday 09 July 2013, Magnus Damm wrote:
>> > +obj-y += $(clk-y$(CONFIG_COMMON_CLK))
>>
>> It took me a bit to understand the Makefile trick you are using here.
>> I assume other readers will have the same problem. How about just
>> doing
>>
>> +ifndef CONFIG_COMMON_CLK
>> obj-$(CONFIG_ARCH_SH7372) += setup-sh7372.o intc-sh7372.o
>> obj-$(CONFIG_ARCH_SH73A0) += setup-sh73a0.o intc-sh73a0.o
>> obj-$(CONFIG_ARCH_R8A73A4) += setup-r8a73a4.o
>> obj-$(CONFIG_ARCH_R8A7740) += setup-r8a7740.o intc-r8a7740.o
>> obj-$(CONFIG_ARCH_R8A7778) += setup-r8a7778.o
>> obj-$(CONFIG_ARCH_R8A7779) += setup-r8a7779.o intc-r8a7779.o
>> obj-$(CONFIG_ARCH_R8A7790) += setup-r8a7790.o
>> obj-$(CONFIG_ARCH_EMEV2) += setup-emev2.o
>> +endif
>
> I was about to mention the same, so I think that would be a good idea.
Sure, that may be a more common way to handle this. =)
To be clear, I assume you wanted that ifndef to cover the clock-xxx.c
files instead of setup-xxx.c. So I will do that in next version.
Thanks for your comments!
Cheers,
/ magnus
^ permalink raw reply [flat|nested] 16+ messages in thread
* [PATCH 01/04] ARM: shmobile: Only build clocks when COMMON_CLK=n
2013-07-09 9:09 ` Magnus Damm
@ 2013-07-09 21:06 ` Arnd Bergmann
0 siblings, 0 replies; 16+ messages in thread
From: Arnd Bergmann @ 2013-07-09 21:06 UTC (permalink / raw)
To: linux-arm-kernel
On Tuesday 09 July 2013, Magnus Damm wrote:
> >>
> >> +ifndef CONFIG_COMMON_CLK
> >> obj-$(CONFIG_ARCH_SH7372) += setup-sh7372.o intc-sh7372.o
> >> obj-$(CONFIG_ARCH_SH73A0) += setup-sh73a0.o intc-sh73a0.o
> >> obj-$(CONFIG_ARCH_R8A73A4) += setup-r8a73a4.o
> >> obj-$(CONFIG_ARCH_R8A7740) += setup-r8a7740.o intc-r8a7740.o
> >> obj-$(CONFIG_ARCH_R8A7778) += setup-r8a7778.o
> >> obj-$(CONFIG_ARCH_R8A7779) += setup-r8a7779.o intc-r8a7779.o
> >> obj-$(CONFIG_ARCH_R8A7790) += setup-r8a7790.o
> >> obj-$(CONFIG_ARCH_EMEV2) += setup-emev2.o
> >> +endif
> >
> > I was about to mention the same, so I think that would be a good idea.
>
> Sure, that may be a more common way to handle this. =)
>
> To be clear, I assume you wanted that ifndef to cover the clock-xxx.c
> files instead of setup-xxx.c. So I will do that in next version.
>
Yes, that's right, sorry for screwing up the example.
Arnd
^ permalink raw reply [flat|nested] 16+ messages in thread
* [PATCH 02/04] ARM: shmobile: Introduce ARCH_SHMOBILE_MULTI
2013-07-09 5:43 [PATCH 00/04] ARM: shmobile: ARCH_MULTIPLATFORM for EMEV2 and KZM9D Magnus Damm
2013-07-09 5:43 ` [PATCH 01/04] ARM: shmobile: Only build clocks when COMMON_CLK=n Magnus Damm
@ 2013-07-09 5:43 ` Magnus Damm
2013-07-09 5:43 ` [PATCH 03/04] ARM: shmobile: Add EMEV2 and KZM9D to ARCH_SHMOBILE_MULTI Magnus Damm
` (2 subsequent siblings)
4 siblings, 0 replies; 16+ messages in thread
From: Magnus Damm @ 2013-07-09 5:43 UTC (permalink / raw)
To: linux-arm-kernel
From: Magnus Damm <damm@opensource.se>
Add ARCH_SHMOBILE_MULTI to mach-shmobile that can be used
to enable ARCH_MULTIPLATFORM on selected SoCs and boards.
The headers stay under arch/arm/mach-shmobile/include/mach/
for now, they can and will be migrated independently over time.
Signed-off-by: Magnus Damm <damm@opensource.se>
---
arch/arm/Makefile | 1 +
arch/arm/mach-shmobile/Kconfig | 21 +++++++++++++++++++++
arch/arm/mach-shmobile/Makefile | 2 ++
3 files changed, 24 insertions(+)
--- 0001/arch/arm/Makefile
+++ work/arch/arm/Makefile 2013-07-08 19:17:00.000000000 +0900
@@ -181,6 +181,7 @@ machine-$(CONFIG_ARCH_EXYNOS) += exynos
machine-$(CONFIG_ARCH_SA1100) += sa1100
machine-$(CONFIG_ARCH_SHARK) += shark
machine-$(CONFIG_ARCH_SHMOBILE) += shmobile
+machine-$(CONFIG_ARCH_SHMOBILE_MULTI) += shmobile
machine-$(CONFIG_ARCH_TEGRA) += tegra
machine-$(CONFIG_ARCH_U300) += u300
machine-$(CONFIG_ARCH_U8500) += ux500
--- 0002/arch/arm/mach-shmobile/Kconfig
+++ work/arch/arm/mach-shmobile/Kconfig 2013-07-08 19:19:21.000000000 +0900
@@ -1,3 +1,24 @@
+config ARCH_SHMOBILE_MULTI
+ bool "SH-Mobile Series" if ARCH_MULTI_V7
+ depends on MMU
+ select CPU_V7
+ select GENERIC_CLOCKEVENTS
+ select HAVE_ARM_SCU if SMP
+ select HAVE_ARM_TWD if LOCAL_TIMERS
+ select HAVE_SMP
+ select ARM_GIC
+ select MIGHT_HAVE_CACHE_L2X0
+ select NO_IOPORT
+ select PINCTRL
+ select ARCH_REQUIRE_GPIOLIB
+ select CLKDEV_LOOKUP
+
+if ARCH_SHMOBILE_MULTI
+
+comment "SH-Mobile System Type"
+
+endif
+
if ARCH_SHMOBILE
comment "SH-Mobile System Type"
--- 0003/arch/arm/mach-shmobile/Makefile
+++ work/arch/arm/mach-shmobile/Makefile 2013-07-08 19:18:20.000000000 +0900
@@ -2,6 +2,8 @@
# Makefile for the linux kernel.
#
+ccflags-$(CONFIG_ARCH_MULTIPLATFORM) := -I$(srctree)/arch/arm/mach-shmobile/include
+
# Common objects
obj-y := timer.o console.o
^ permalink raw reply [flat|nested] 16+ messages in thread* [PATCH 03/04] ARM: shmobile: Add EMEV2 and KZM9D to ARCH_SHMOBILE_MULTI
2013-07-09 5:43 [PATCH 00/04] ARM: shmobile: ARCH_MULTIPLATFORM for EMEV2 and KZM9D Magnus Damm
2013-07-09 5:43 ` [PATCH 01/04] ARM: shmobile: Only build clocks when COMMON_CLK=n Magnus Damm
2013-07-09 5:43 ` [PATCH 02/04] ARM: shmobile: Introduce ARCH_SHMOBILE_MULTI Magnus Damm
@ 2013-07-09 5:43 ` Magnus Damm
2013-07-09 7:37 ` Arnd Bergmann
2013-07-09 5:44 ` [PATCH 04/04] ARM: shmobile: Allow ARCH_SHMOBILE_MULTI timer configuration Magnus Damm
2013-07-09 7:45 ` [PATCH 00/04] ARM: shmobile: ARCH_MULTIPLATFORM for EMEV2 and KZM9D Arnd Bergmann
4 siblings, 1 reply; 16+ messages in thread
From: Magnus Damm @ 2013-07-09 5:43 UTC (permalink / raw)
To: linux-arm-kernel
From: Magnus Damm <damm@opensource.se>
Enable build of EMEV2 and KZM9D DT Reference in
case of ARCH_MULTIPLATFORM and ARCH_SHMOBILE_MULTI.
A weak symbol covers the case when the clock-emev2.c
file is not built when COMMON_CLK=y.
Signed-off-by: Magnus Damm <damm@opensource.se>
---
arch/arm/boot/dts/Makefile | 1 +
arch/arm/mach-shmobile/Kconfig | 15 +++++++++++++++
arch/arm/mach-shmobile/setup-emev2.c | 5 +++++
3 files changed, 21 insertions(+)
--- 0002/arch/arm/boot/dts/Makefile
+++ work/arch/arm/boot/dts/Makefile 2013-07-09 13:10:52.000000000 +0900
@@ -171,6 +171,7 @@ dtb-$(CONFIG_ARCH_SHMOBILE) += emev2-kzm
sh73a0-kzm9g-reference.dtb \
r8a73a4-ape6evm.dtb \
sh7372-mackerel.dtb
+dtb-$(CONFIG_ARCH_SHMOBILE_MULTI) += emev2-kzm9d-reference.dtb
dtb-$(CONFIG_ARCH_SOCFPGA) += socfpga_cyclone5.dtb \
socfpga_vt.dtb
dtb-$(CONFIG_ARCH_SPEAR13XX) += spear1310-evb.dtb \
--- 0004/arch/arm/mach-shmobile/Kconfig
+++ work/arch/arm/mach-shmobile/Kconfig 2013-07-09 13:03:45.000000000 +0900
@@ -17,6 +17,21 @@ if ARCH_SHMOBILE_MULTI
comment "SH-Mobile System Type"
+config ARCH_EMEV2
+ bool "Emma Mobile EV2"
+
+comment "SH-Mobile Board Type"
+
+config MACH_KZM9D_REFERENCE
+ bool "KZM9D board - Reference Device Tree Implementation"
+ depends on ARCH_EMEV2
+ select REGULATOR_FIXED_VOLTAGE if REGULATOR
+ ---help---
+ Use reference implementation of KZM9D board support
+ which makes a greater use of device tree at the expense
+ of not supporting a number of devices.
+
+ This is intended to aid developers
endif
if ARCH_SHMOBILE
--- 0001/arch/arm/mach-shmobile/setup-emev2.c
+++ work/arch/arm/mach-shmobile/setup-emev2.c 2013-07-09 13:03:45.000000000 +0900
@@ -190,6 +190,11 @@ void __init emev2_add_standard_devices(v
emev2_register_pmu();
}
+void __init __weak emev2_clock_init(void)
+{
+ /* in case legacy clocks are disabled then do nothing */
+}
+
void __init emev2_init_delay(void)
{
shmobile_setup_delay(533, 1, 3); /* Cortex-A9 @ 533MHz */
^ permalink raw reply [flat|nested] 16+ messages in thread* [PATCH 03/04] ARM: shmobile: Add EMEV2 and KZM9D to ARCH_SHMOBILE_MULTI
2013-07-09 5:43 ` [PATCH 03/04] ARM: shmobile: Add EMEV2 and KZM9D to ARCH_SHMOBILE_MULTI Magnus Damm
@ 2013-07-09 7:37 ` Arnd Bergmann
2013-07-09 9:15 ` Magnus Damm
0 siblings, 1 reply; 16+ messages in thread
From: Arnd Bergmann @ 2013-07-09 7:37 UTC (permalink / raw)
To: linux-arm-kernel
On Tuesday 09 July 2013, Magnus Damm wrote:
> +void __init __weak emev2_clock_init(void)
> +{
> + /* in case legacy clocks are disabled then do nothing */
> +}
Nitpicking:
I don't like the __weak symbols too much, in general I prefer
doing either
if (IS_ENABLED(CONFIG_SHMOBILE_CLOCKS)
emev2_clock_init()
or using an inline alternative in the header file.
Arnd
^ permalink raw reply [flat|nested] 16+ messages in thread* [PATCH 03/04] ARM: shmobile: Add EMEV2 and KZM9D to ARCH_SHMOBILE_MULTI
2013-07-09 7:37 ` Arnd Bergmann
@ 2013-07-09 9:15 ` Magnus Damm
2013-07-09 21:07 ` Arnd Bergmann
0 siblings, 1 reply; 16+ messages in thread
From: Magnus Damm @ 2013-07-09 9:15 UTC (permalink / raw)
To: linux-arm-kernel
Hi Arnd,
On Tue, Jul 9, 2013 at 4:37 PM, Arnd Bergmann <arnd@arndb.de> wrote:
> On Tuesday 09 July 2013, Magnus Damm wrote:
>> +void __init __weak emev2_clock_init(void)
>> +{
>> + /* in case legacy clocks are disabled then do nothing */
>> +}
>
> Nitpicking:
>
> I don't like the __weak symbols too much, in general I prefer
> doing either
>
> if (IS_ENABLED(CONFIG_SHMOBILE_CLOCKS)
> emev2_clock_init()
>
> or using an inline alternative in the header file.
Sure, good idea. I plan on using the following if you don't mind:
if (!IS_ENABLED(CONFIG_COMMON_CLK)
emev2_clock_init()
Cheers,
/ magnus
^ permalink raw reply [flat|nested] 16+ messages in thread
* [PATCH 04/04] ARM: shmobile: Allow ARCH_SHMOBILE_MULTI timer configuration
2013-07-09 5:43 [PATCH 00/04] ARM: shmobile: ARCH_MULTIPLATFORM for EMEV2 and KZM9D Magnus Damm
` (2 preceding siblings ...)
2013-07-09 5:43 ` [PATCH 03/04] ARM: shmobile: Add EMEV2 and KZM9D to ARCH_SHMOBILE_MULTI Magnus Damm
@ 2013-07-09 5:44 ` Magnus Damm
2013-07-09 7:45 ` [PATCH 00/04] ARM: shmobile: ARCH_MULTIPLATFORM for EMEV2 and KZM9D Arnd Bergmann
4 siblings, 0 replies; 16+ messages in thread
From: Magnus Damm @ 2013-07-09 5:44 UTC (permalink / raw)
To: linux-arm-kernel
From: Magnus Damm <damm@opensource.se>
Move the timer configuration bits to allow use in
case of ARCH_SHMOBILE or ARCH_SHMOBILE_MULTI.
The timers all make use of the regular driver model
so they are safe to enable for ARCH_MULTIPLATFORM.
Signed-off-by: Magnus Damm <damm@opensource.se>
---
arch/arm/mach-shmobile/Kconfig | 16 +++++++++++-----
1 file changed, 11 insertions(+), 5 deletions(-)
--- 0005/arch/arm/mach-shmobile/Kconfig
+++ work/arch/arm/mach-shmobile/Kconfig 2013-07-09 12:07:23.000000000 +0900
@@ -30,6 +30,8 @@ config MACH_KZM9D_REFERENCE
of not supporting a number of devices.
This is intended to aid developers
+
+comment "SH-Mobile System Configuration"
endif
if ARCH_SHMOBILE
@@ -243,6 +245,15 @@ config CPU_HAS_INTEVT
bool
default y
+config SH_CLK_CPG
+ bool
+
+source "drivers/sh/Kconfig"
+
+endif
+
+if ARCH_SHMOBILE || ARCH_SHMOBILE_MULTI
+
menu "Timer and clock configuration"
config SHMOBILE_TIMER_HZ
@@ -277,9 +288,4 @@ config EM_TIMER_STI
endmenu
-config SH_CLK_CPG
- bool
-
-source "drivers/sh/Kconfig"
-
endif
^ permalink raw reply [flat|nested] 16+ messages in thread* [PATCH 00/04] ARM: shmobile: ARCH_MULTIPLATFORM for EMEV2 and KZM9D
2013-07-09 5:43 [PATCH 00/04] ARM: shmobile: ARCH_MULTIPLATFORM for EMEV2 and KZM9D Magnus Damm
` (3 preceding siblings ...)
2013-07-09 5:44 ` [PATCH 04/04] ARM: shmobile: Allow ARCH_SHMOBILE_MULTI timer configuration Magnus Damm
@ 2013-07-09 7:45 ` Arnd Bergmann
2013-07-09 9:11 ` Magnus Damm
4 siblings, 1 reply; 16+ messages in thread
From: Arnd Bergmann @ 2013-07-09 7:45 UTC (permalink / raw)
To: linux-arm-kernel
On Tuesday 09 July 2013, Magnus Damm wrote:
> ARM: shmobile: ARCH_MULTIPLATFORM for EMEV2 and KZM9D
>
> [PATCH 01/04] ARM: shmobile: Only build clocks when COMMON_CLK=n
> [PATCH 02/04] ARM: shmobile: Introduce ARCH_SHMOBILE_MULTI
> [PATCH 03/04] ARM: shmobile: Add EMEV2 and KZM9D to ARCH_SHMOBILE_MULTI
> [PATCH 04/04] ARM: shmobile: Allow ARCH_SHMOBILE_MULTI timer configuration
>
> This patch set extends mach-shmobile to allow a subset of SoCs and
> boards to be built for ARCH_MULTIPLATFORM. With this series applied
> the EMEV2 SoC and the KZM9D board can be built in two ways - either
> as good old ARCH_SHMOBILE or new ARCH_SHMOBILE_MULTI. The latter
> requires common clock framework for proper run time operation.
>
> The goal is to over time move all DT reference implementations from
> ARCH_SHMOBILE to ARCH_SHMOBILE_MULTI. Until CCF is available for a
> certain SoC it is to me acceptable for a DT reference board to allow
> build in both ARCH_SHMOBILE and ARCH_SHMOBILE_MULTI. The reason for
> this is to allow incremental development and easy testing during
> development of CCF for a certain SoC. The KZM9D board is now in this
> state while it is waiting for common clocks to become ready.
>
> In the near future a prototype hack will be posted that comments
> out the clock framework bits on EMEV2 and will allow boot of KZM9D
> in the case of ARCH_SHMOBILE_MULTI even though CCF is not yet
> available.
>
> Signed-off-by: Magnus Damm <damm@opensource.se>
Great stuff, very glad you got this far!
I have two small style comments, nothing important.
Arnd
^ permalink raw reply [flat|nested] 16+ messages in thread* [PATCH 00/04] ARM: shmobile: ARCH_MULTIPLATFORM for EMEV2 and KZM9D
2013-07-09 7:45 ` [PATCH 00/04] ARM: shmobile: ARCH_MULTIPLATFORM for EMEV2 and KZM9D Arnd Bergmann
@ 2013-07-09 9:11 ` Magnus Damm
0 siblings, 0 replies; 16+ messages in thread
From: Magnus Damm @ 2013-07-09 9:11 UTC (permalink / raw)
To: linux-arm-kernel
On Tue, Jul 9, 2013 at 4:45 PM, Arnd Bergmann <arnd@arndb.de> wrote:
> On Tuesday 09 July 2013, Magnus Damm wrote:
>> ARM: shmobile: ARCH_MULTIPLATFORM for EMEV2 and KZM9D
>>
>> [PATCH 01/04] ARM: shmobile: Only build clocks when COMMON_CLK=n
>> [PATCH 02/04] ARM: shmobile: Introduce ARCH_SHMOBILE_MULTI
>> [PATCH 03/04] ARM: shmobile: Add EMEV2 and KZM9D to ARCH_SHMOBILE_MULTI
>> [PATCH 04/04] ARM: shmobile: Allow ARCH_SHMOBILE_MULTI timer configuration
>>
>> This patch set extends mach-shmobile to allow a subset of SoCs and
>> boards to be built for ARCH_MULTIPLATFORM. With this series applied
>> the EMEV2 SoC and the KZM9D board can be built in two ways - either
>> as good old ARCH_SHMOBILE or new ARCH_SHMOBILE_MULTI. The latter
>> requires common clock framework for proper run time operation.
>>
>> The goal is to over time move all DT reference implementations from
>> ARCH_SHMOBILE to ARCH_SHMOBILE_MULTI. Until CCF is available for a
>> certain SoC it is to me acceptable for a DT reference board to allow
>> build in both ARCH_SHMOBILE and ARCH_SHMOBILE_MULTI. The reason for
>> this is to allow incremental development and easy testing during
>> development of CCF for a certain SoC. The KZM9D board is now in this
>> state while it is waiting for common clocks to become ready.
>>
>> In the near future a prototype hack will be posted that comments
>> out the clock framework bits on EMEV2 and will allow boot of KZM9D
>> in the case of ARCH_SHMOBILE_MULTI even though CCF is not yet
>> available.
>>
>> Signed-off-by: Magnus Damm <damm@opensource.se>
>
> Great stuff, very glad you got this far!
Thanks, I'm happy to hear that!
> I have two small style comments, nothing important.
Sure, I will fix up and resend. Thanks for your feedback!
/ magnus
^ permalink raw reply [flat|nested] 16+ messages in thread