[PATCH] ARM: Add ARCH_HAS_L2X0 and make CACHE_L2X0 config depend on that

[PATCH] ARM: Add ARCH_HAS_L2X0 and make CACHE_L2X0 config depend on that

[Olof Johansson]

It normally scales better if each platform that has <foo> selects the
ARCH_HAS_<foo> option, instead of having to add every single one to the
depends line.

Signed-off-by: Olof Johansson <olof@lixom.net>
---
 arch/arm/Kconfig               |    2 ++
 arch/arm/mach-mx3/Kconfig      |    2 ++
 arch/arm/mach-realview/Kconfig |    5 +++++
 arch/arm/mach-vexpress/Kconfig |    1 +
 arch/arm/mm/Kconfig            |    8 +++++---
 arch/arm/plat-omap/Kconfig     |    1 +
 6 files changed, 16 insertions(+), 3 deletions(-)

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 1f254bd..6800b4b 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -748,6 +748,7 @@ config ARCH_U8500
 	select GENERIC_CLOCKEVENTS
 	select COMMON_CLKDEV
 	select ARCH_REQUIRE_GPIOLIB
+	select ARCH_HAS_L2X0
 	help
 	  Support for ST-Ericsson's Ux500 architecture
 
@@ -759,6 +760,7 @@ config ARCH_NOMADIK
 	select COMMON_CLKDEV
 	select GENERIC_CLOCKEVENTS
 	select ARCH_REQUIRE_GPIOLIB
+	select ARCH_HAS_L2X0
 	help
 	  Support for the Nomadik platform by ST-Ericsson
 
diff --git a/arch/arm/mach-mx3/Kconfig b/arch/arm/mach-mx3/Kconfig
index 344753f..a4cd462 100644
--- a/arch/arm/mach-mx3/Kconfig
+++ b/arch/arm/mach-mx3/Kconfig
@@ -3,12 +3,14 @@ if ARCH_MX3
 config ARCH_MX31
 	select ARCH_HAS_RNGA
 	select ARCH_MXC_AUDMUX_V2
+	select ARCH_HAS_L2X0
 	bool
 
 config ARCH_MX35
 	bool
 	select ARCH_MXC_IOMUX_V3
 	select ARCH_MXC_AUDMUX_V2
+	select ARCH_HAS_L2X0
 
 comment "MX3 platforms:"
 
diff --git a/arch/arm/mach-realview/Kconfig b/arch/arm/mach-realview/Kconfig
index ee5e392..24077e8 100644
--- a/arch/arm/mach-realview/Kconfig
+++ b/arch/arm/mach-realview/Kconfig
@@ -11,6 +11,7 @@ config REALVIEW_EB_A9MP
 	bool "Support Multicore Cortex-A9"
 	depends on MACH_REALVIEW_EB
 	select CPU_V7
+	select ARCH_HAS_L2X0
 	help
 	  Enable support for the Cortex-A9MPCore tile on the Realview platform.
 
@@ -18,6 +19,7 @@ config REALVIEW_EB_ARM11MP
 	bool "Support ARM11MPCore tile"
 	depends on MACH_REALVIEW_EB
 	select CPU_V6
+	select ARCH_HAS_L2X0
 	help
 	  Enable support for the ARM11MPCore tile on the Realview platform.
 
@@ -35,6 +37,7 @@ config MACH_REALVIEW_PB11MP
 	select CPU_V6
 	select ARM_GIC
 	select HAVE_PATA_PLATFORM
+	select ARCH_HAS_L2X0
 	help
 	  Include support for the ARM(R) RealView MPCore Platform Baseboard.
 	  PB11MPCore is a platform with an on-board ARM11MPCore and has
@@ -44,6 +47,7 @@ config MACH_REALVIEW_PB1176
 	bool "Support RealView/PB1176 platform"
 	select CPU_V6
 	select ARM_GIC
+	select ARCH_HAS_L2X0
 	help
 	  Include support for the ARM(R) RealView ARM1176 Platform Baseboard.
 
@@ -72,6 +76,7 @@ config MACH_REALVIEW_PBX
 	select HAVE_PATA_PLATFORM
 	select ARCH_SPARSEMEM_ENABLE if CPU_V7 && !REALVIEW_HIGH_PHYS_OFFSET
 	select ZONE_DMA if SPARSEMEM
+	select ARCH_HAS_L2X0
 	help
 	  Include support for the ARM(R) RealView PBX platform.
 
diff --git a/arch/arm/mach-vexpress/Kconfig b/arch/arm/mach-vexpress/Kconfig
index 3f19b66..8cc4281 100644
--- a/arch/arm/mach-vexpress/Kconfig
+++ b/arch/arm/mach-vexpress/Kconfig
@@ -5,5 +5,6 @@ config ARCH_VEXPRESS_CA9X4
 	bool "Versatile Express Cortex-A9x4 tile"
 	select CPU_V7
 	select ARM_GIC
+	select ARCH_HAS_L2X0
 
 endmenu
diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig
index 346ae14..419b957 100644
--- a/arch/arm/mm/Kconfig
+++ b/arch/arm/mm/Kconfig
@@ -759,11 +759,13 @@ config CACHE_FEROCEON_L2_WRITETHROUGH
 	  Say Y here to use the Feroceon L2 cache in writethrough mode.
 	  Unless you specifically require this, say N for writeback mode.
 
+config ARCH_HAS_L2X0
+	bool
+	default n
+
 config CACHE_L2X0
 	bool "Enable the L2x0 outer cache controller"
-	depends on REALVIEW_EB_ARM11MP || MACH_REALVIEW_PB11MP || MACH_REALVIEW_PB1176 || \
-		   REALVIEW_EB_A9MP || ARCH_MX35 || ARCH_MX31 || MACH_REALVIEW_PBX || \
-		   ARCH_NOMADIK || ARCH_OMAP4 || ARCH_U8500 || ARCH_VEXPRESS_CA9X4
+	depends on ARCH_HAS_L2X0
 	default y
 	select OUTER_CACHE
 	select OUTER_CACHE_SYNC
diff --git a/arch/arm/plat-omap/Kconfig b/arch/arm/plat-omap/Kconfig
index 78b49a6..2e42717 100644
--- a/arch/arm/plat-omap/Kconfig
+++ b/arch/arm/plat-omap/Kconfig
@@ -38,6 +38,7 @@ config ARCH_OMAP4
 	depends on ARCH_OMAP2PLUS
 	select CPU_V7
 	select ARM_GIC
+	select ARCH_HAS_L2X0
 
 endchoice
 
-- 
1.5.6.5

[PATCH] ARM: Add ARCH_HAS_L2X0 and make CACHE_L2X0 config depend on that

[Daniel Walker]

On Fri, 2010-06-11 at 11:03 -0500, Olof Johansson wrote:

> diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig
> index 346ae14..419b957 100644
> --- a/arch/arm/mm/Kconfig
> +++ b/arch/arm/mm/Kconfig
> @@ -759,11 +759,13 @@ config CACHE_FEROCEON_L2_WRITETHROUGH
>  	  Say Y here to use the Feroceon L2 cache in writethrough mode.
>  	  Unless you specifically require this, say N for writeback mode.
>  
> +config ARCH_HAS_L2X0
> +	bool
> +	default n
> +

I made a similar change not too long ago, but Russell wanted the define
to be HAVE_CACHE_L2X0 ..

Daniel

[PATCH v2] ARM: Add HAVE_CACHE_L2X0

[Olof Johansson]

It normally scales better if each platform that has <foo> selects the
HAVE_<foo> option, instead of having to add every single one to the
depends line for CONFIG_<foo>.

Signed-off-by: Olof Johansson <olof@lixom.net>

---

On Fri, Jun 11, 2010 at 10:58:07AM -0700, Daniel Walker wrote:

> I made a similar change not too long ago, but Russell wanted the define
> to be HAVE_CACHE_L2X0 ..

Ah, I missed it. I'm not attached to the option name so since I didn't
see you repost, here's my version... :)


-Olof

---

 arch/arm/Kconfig               |    2 ++
 arch/arm/mach-mx3/Kconfig      |    2 ++
 arch/arm/mach-realview/Kconfig |    5 +++++
 arch/arm/mach-vexpress/Kconfig |    1 +
 arch/arm/mm/Kconfig            |    8 +++++---
 arch/arm/plat-omap/Kconfig     |    1 +
 6 files changed, 16 insertions(+), 3 deletions(-)

diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 1f254bd..fee073d 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -748,6 +748,7 @@ config ARCH_U8500
 	select GENERIC_CLOCKEVENTS
 	select COMMON_CLKDEV
 	select ARCH_REQUIRE_GPIOLIB
+	select HAVE_CACHE_L2X0
 	help
 	  Support for ST-Ericsson's Ux500 architecture
 
@@ -759,6 +760,7 @@ config ARCH_NOMADIK
 	select COMMON_CLKDEV
 	select GENERIC_CLOCKEVENTS
 	select ARCH_REQUIRE_GPIOLIB
+	select HAVE_CACHE_L2X0
 	help
 	  Support for the Nomadik platform by ST-Ericsson
 
diff --git a/arch/arm/mach-mx3/Kconfig b/arch/arm/mach-mx3/Kconfig
index 344753f..103cad3 100644
--- a/arch/arm/mach-mx3/Kconfig
+++ b/arch/arm/mach-mx3/Kconfig
@@ -3,12 +3,14 @@ if ARCH_MX3
 config ARCH_MX31
 	select ARCH_HAS_RNGA
 	select ARCH_MXC_AUDMUX_V2
+	select HAVE_CACHE_L2X0
 	bool
 
 config ARCH_MX35
 	bool
 	select ARCH_MXC_IOMUX_V3
 	select ARCH_MXC_AUDMUX_V2
+	select HAVE_CACHE_L2X0
 
 comment "MX3 platforms:"
 
diff --git a/arch/arm/mach-realview/Kconfig b/arch/arm/mach-realview/Kconfig
index ee5e392..f188ea3 100644
--- a/arch/arm/mach-realview/Kconfig
+++ b/arch/arm/mach-realview/Kconfig
@@ -11,6 +11,7 @@ config REALVIEW_EB_A9MP
 	bool "Support Multicore Cortex-A9"
 	depends on MACH_REALVIEW_EB
 	select CPU_V7
+	select HAVE_CACHE_L2X0
 	help
 	  Enable support for the Cortex-A9MPCore tile on the Realview platform.
 
@@ -18,6 +19,7 @@ config REALVIEW_EB_ARM11MP
 	bool "Support ARM11MPCore tile"
 	depends on MACH_REALVIEW_EB
 	select CPU_V6
+	select HAVE_CACHE_L2X0
 	help
 	  Enable support for the ARM11MPCore tile on the Realview platform.
 
@@ -35,6 +37,7 @@ config MACH_REALVIEW_PB11MP
 	select CPU_V6
 	select ARM_GIC
 	select HAVE_PATA_PLATFORM
+	select HAVE_CACHE_L2X0
 	help
 	  Include support for the ARM(R) RealView MPCore Platform Baseboard.
 	  PB11MPCore is a platform with an on-board ARM11MPCore and has
@@ -44,6 +47,7 @@ config MACH_REALVIEW_PB1176
 	bool "Support RealView/PB1176 platform"
 	select CPU_V6
 	select ARM_GIC
+	select HAVE_CACHE_L2X0
 	help
 	  Include support for the ARM(R) RealView ARM1176 Platform Baseboard.
 
@@ -72,6 +76,7 @@ config MACH_REALVIEW_PBX
 	select HAVE_PATA_PLATFORM
 	select ARCH_SPARSEMEM_ENABLE if CPU_V7 && !REALVIEW_HIGH_PHYS_OFFSET
 	select ZONE_DMA if SPARSEMEM
+	select HAVE_CACHE_L2X0
 	help
 	  Include support for the ARM(R) RealView PBX platform.
 
diff --git a/arch/arm/mach-vexpress/Kconfig b/arch/arm/mach-vexpress/Kconfig
index 3f19b66..97d8592 100644
--- a/arch/arm/mach-vexpress/Kconfig
+++ b/arch/arm/mach-vexpress/Kconfig
@@ -5,5 +5,6 @@ config ARCH_VEXPRESS_CA9X4
 	bool "Versatile Express Cortex-A9x4 tile"
 	select CPU_V7
 	select ARM_GIC
+	select HAVE_CACHE_L2X0
 
 endmenu
diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig
index 346ae14..14d25f3 100644
--- a/arch/arm/mm/Kconfig
+++ b/arch/arm/mm/Kconfig
@@ -759,11 +759,13 @@ config CACHE_FEROCEON_L2_WRITETHROUGH
 	  Say Y here to use the Feroceon L2 cache in writethrough mode.
 	  Unless you specifically require this, say N for writeback mode.
 
+config HAVE_CACHE_L2X0
+	bool
+	default n
+
 config CACHE_L2X0
 	bool "Enable the L2x0 outer cache controller"
-	depends on REALVIEW_EB_ARM11MP || MACH_REALVIEW_PB11MP || MACH_REALVIEW_PB1176 || \
-		   REALVIEW_EB_A9MP || ARCH_MX35 || ARCH_MX31 || MACH_REALVIEW_PBX || \
-		   ARCH_NOMADIK || ARCH_OMAP4 || ARCH_U8500 || ARCH_VEXPRESS_CA9X4
+	depends on HAVE_CACHE_L2X0
 	default y
 	select OUTER_CACHE
 	select OUTER_CACHE_SYNC
diff --git a/arch/arm/plat-omap/Kconfig b/arch/arm/plat-omap/Kconfig
index 78b49a6..e7972d9 100644
--- a/arch/arm/plat-omap/Kconfig
+++ b/arch/arm/plat-omap/Kconfig
@@ -38,6 +38,7 @@ config ARCH_OMAP4
 	depends on ARCH_OMAP2PLUS
 	select CPU_V7
 	select ARM_GIC
+	select HAVE_CACHE_L2X0
 
 endchoice
 
-- 
1.5.6.5

[PATCH] ARM: Add ARCH_HAS_L2X0 and make CACHE_L2X0 config depend on that

[Ben Dooks]

On Fri, Jun 11, 2010 at 10:58:07AM -0700, Daniel Walker wrote:
> On Fri, 2010-06-11 at 11:03 -0500, Olof Johansson wrote:
> 
> > diff --git a/arch/arm/mm/Kconfig b/arch/arm/mm/Kconfig
> > index 346ae14..419b957 100644
> > --- a/arch/arm/mm/Kconfig
> > +++ b/arch/arm/mm/Kconfig
> > @@ -759,11 +759,13 @@ config CACHE_FEROCEON_L2_WRITETHROUGH
> >  	  Say Y here to use the Feroceon L2 cache in writethrough mode.
> >  	  Unless you specifically require this, say N for writeback mode.
> >  
> > +config ARCH_HAS_L2X0
> > +	bool
> > +	default n
> > +
> 
> I made a similar change not too long ago, but Russell wanted the define
> to be HAVE_CACHE_L2X0 ..

possibly, I was thinking ARM_L2X0 would also be an option, since if I
remeber correctly it is selecting the driver not the availability of
it.

-- 
Ben

Q:      What's a light-year?
A:      One-third less calories than a regular year.

[PATCH] ARM: Add ARCH_HAS_L2X0 and make CACHE_L2X0 config depend on that

[Olof Johansson]

On Sun, Jun 13, 2010 at 06:16:26AM +0100, Ben Dooks wrote:
> On Fri, Jun 11, 2010 at 10:58:07AM -0700, Daniel Walker wrote:
> >
> > I made a similar change not too long ago, but Russell wanted the define
> > to be HAVE_CACHE_L2X0 ..
> 
> possibly, I was thinking ARM_L2X0 would also be an option, since if I
> remeber correctly it is selecting the driver not the availability of
> it.

Yeah, there are lots of different possible names. :)

Let's just go with the one Russell preferred and not worry too much
about it. There's some presendence to using HAVE_.* already.


-Olof

[PATCH] ARM: Add ARCH_HAS_L2X0 and make CACHE_L2X0 config depend on that

[Eric Miao]

On Mon, Jun 14, 2010 at 10:59 PM, Olof Johansson <olof@lixom.net> wrote:
> On Sun, Jun 13, 2010 at 06:16:26AM +0100, Ben Dooks wrote:
>> On Fri, Jun 11, 2010 at 10:58:07AM -0700, Daniel Walker wrote:
>> >
>> > I made a similar change not too long ago, but Russell wanted the define
>> > to be HAVE_CACHE_L2X0 ..
>>
>> possibly, I was thinking ARM_L2X0 would also be an option, since if I
>> remeber correctly it is selecting the driver not the availability of
>> it.
>
> Yeah, there are lots of different possible names. :)
>
> Let's just go with the one Russell preferred and not worry too much
> about it. There's some presendence to using HAVE_.* already.
>

Sometimes I am thinking that maybe ARCH_HAS_* has some benefits:

HAVE_* for those kernel-wide and general capabilities
ARCH_HAS_* for those ARM machine-class specific capabilities?

So to make things a bit more clear? If ARCH_HAS_* is widely adopted
on architectures other than ARM, maybe we can then switch that name
to HAVE_*?

[PATCH] ARM: Add ARCH_HAS_L2X0 and make CACHE_L2X0 config depend on that

[Olof Johansson]

On Wed, Jun 16, 2010 at 11:43:29AM +0800, Eric Miao wrote:
> On Mon, Jun 14, 2010 at 10:59 PM, Olof Johansson <olof@lixom.net> wrote:
> > On Sun, Jun 13, 2010 at 06:16:26AM +0100, Ben Dooks wrote:
> >> On Fri, Jun 11, 2010 at 10:58:07AM -0700, Daniel Walker wrote:
> >> >
> >> > I made a similar change not too long ago, but Russell wanted the define
> >> > to be HAVE_CACHE_L2X0 ..
> >>
> >> possibly, I was thinking ARM_L2X0 would also be an option, since if I
> >> remeber correctly it is selecting the driver not the availability of
> >> it.
> >
> > Yeah, there are lots of different possible names. :)
> >
> > Let's just go with the one Russell preferred and not worry too much
> > about it. There's some presendence to using HAVE_.* already.
> >
> 
> Sometimes I am thinking that maybe ARCH_HAS_* has some benefits:
> 
> HAVE_* for those kernel-wide and general capabilities
> ARCH_HAS_* for those ARM machine-class specific capabilities?
> 
> So to make things a bit more clear? If ARCH_HAS_* is widely adopted
> on architectures other than ARM, maybe we can then switch that name
> to HAVE_*?

We're nowhere near depleting the HAVE_.* namespace, it's quite sparsely
used at the moment. I don't see a need to overengineer things here.

A bike shed of any color will do, etc.


-Olof

[PATCH] ARM: Add ARCH_HAS_L2X0 and make CACHE_L2X0 config depend on that

[Eric Miao]

On Wed, Jun 16, 2010 at 10:13 PM, Olof Johansson <olof@lixom.net> wrote:
> On Wed, Jun 16, 2010 at 11:43:29AM +0800, Eric Miao wrote:
>> On Mon, Jun 14, 2010 at 10:59 PM, Olof Johansson <olof@lixom.net> wrote:
>> > On Sun, Jun 13, 2010 at 06:16:26AM +0100, Ben Dooks wrote:
>> >> On Fri, Jun 11, 2010 at 10:58:07AM -0700, Daniel Walker wrote:
>> >> >
>> >> > I made a similar change not too long ago, but Russell wanted the define
>> >> > to be HAVE_CACHE_L2X0 ..
>> >>
>> >> possibly, I was thinking ARM_L2X0 would also be an option, since if I
>> >> remeber correctly it is selecting the driver not the availability of
>> >> it.
>> >
>> > Yeah, there are lots of different possible names. :)
>> >
>> > Let's just go with the one Russell preferred and not worry too much
>> > about it. There's some presendence to using HAVE_.* already.
>> >
>>
>> Sometimes I am thinking that maybe ARCH_HAS_* has some benefits:
>>
>> HAVE_* for those kernel-wide and general capabilities
>> ARCH_HAS_* for those ARM machine-class specific capabilities?
>>
>> So to make things a bit more clear? If ARCH_HAS_* is widely adopted
>> on architectures other than ARM, maybe we can then switch that name
>> to HAVE_*?
>
> We're nowhere near depleting the HAVE_.* namespace, it's quite sparsely
> used at the moment.

No I'm not talking about depleting the HAVE_* , just wondering if there is a way
to make things a bit clearer.

> I don't see a need to overengineer things here.
>
> A bike shed of any color will do, etc.
>

Agreed.