[PATCH 1/1] drivers: bus: Move the OMAP interconnect driver to drivers/bus/

[PATCH 1/1] drivers: bus: Move the OMAP interconnect driver to drivers/bus/

[Santosh Shilimkar]

OMAP interconnect drivers are used for the interconnect error handling.
Since they are bus driver, lets move it to newly created drivers/bus.

Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Tony Lindgren <tony@atomide.com>
Tested-by: Lokesh Vutla <lokeshvutla@ti.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
---
Patch just moves OMAP interconnect drivers as is to the newly created
driver/bus/* directory. Patch is generated against "arm-soc/drivers/ocp2scp"
tree and test on all OMAP boards.

 arch/arm/mach-omap2/Kconfig                        |    2 ++
 arch/arm/mach-omap2/Makefile                       |    5 -----
 drivers/bus/Kconfig                                |    6 ++++++
 drivers/bus/Makefile                               |    3 +++
 {arch/arm/mach-omap2 => drivers/bus}/omap_l3_noc.c |    0
 {arch/arm/mach-omap2 => drivers/bus}/omap_l3_noc.h |    0
 {arch/arm/mach-omap2 => drivers/bus}/omap_l3_smx.c |    0
 {arch/arm/mach-omap2 => drivers/bus}/omap_l3_smx.h |    0
 8 files changed, 11 insertions(+), 5 deletions(-)
 rename {arch/arm/mach-omap2 => drivers/bus}/omap_l3_noc.c (100%)
 rename {arch/arm/mach-omap2 => drivers/bus}/omap_l3_noc.h (100%)
 rename {arch/arm/mach-omap2 => drivers/bus}/omap_l3_smx.c (100%)
 rename {arch/arm/mach-omap2 => drivers/bus}/omap_l3_smx.h (100%)

diff --git a/arch/arm/mach-omap2/Kconfig b/arch/arm/mach-omap2/Kconfig
index dd2db02..7d3c8ab 100644
--- a/arch/arm/mach-omap2/Kconfig
+++ b/arch/arm/mach-omap2/Kconfig
@@ -44,6 +44,7 @@ config ARCH_OMAP3
 	select ARM_CPU_SUSPEND if PM
 	select MULTI_IRQ_HANDLER
 	select SOC_HAS_OMAP2_SDRC
+	select OMAP_INTERCONNECT
 
 config ARCH_OMAP4
 	bool "TI OMAP4"
@@ -63,6 +64,7 @@ config ARCH_OMAP4
 	select USB_ARCH_HAS_EHCI if USB_SUPPORT
 	select ARM_CPU_SUSPEND if PM
 	select ARCH_NEEDS_CPU_IDLE_COUPLED
+	select OMAP_INTERCONNECT
 
 config SOC_OMAP5
 	bool "TI OMAP5"
diff --git a/arch/arm/mach-omap2/Makefile b/arch/arm/mach-omap2/Makefile
index f6a24b3..7fed980 100644
--- a/arch/arm/mach-omap2/Makefile
+++ b/arch/arm/mach-omap2/Makefile
@@ -199,11 +199,6 @@ obj-$(CONFIG_ARCH_OMAP4)		+= omap_hwmod_44xx_data.o
 # EMU peripherals
 obj-$(CONFIG_OMAP3_EMU)			+= emu.o
 
-# L3 interconnect
-obj-$(CONFIG_ARCH_OMAP3)		+= omap_l3_smx.o
-obj-$(CONFIG_ARCH_OMAP4)		+= omap_l3_noc.o
-obj-$(CONFIG_SOC_OMAP5)			+= omap_l3_noc.o
-
 obj-$(CONFIG_OMAP_MBOX_FWK)		+= mailbox_mach.o
 mailbox_mach-objs			:= mailbox.o
 
diff --git a/drivers/bus/Kconfig b/drivers/bus/Kconfig
index 6270415..bbec35d 100644
--- a/drivers/bus/Kconfig
+++ b/drivers/bus/Kconfig
@@ -12,4 +12,10 @@ config OMAP_OCP2SCP
 	  OCP2SCP and in OMAP5, both USB PHY and SATA PHY is connected via
 	  OCP2SCP.
 
+config OMAP_INTERCONNECT
+	tristate "OMAP INTERCONNECT DRIVER"
+	depends on ARCH_OMAP2PLUS
+
+	help
+	  Driver to enable OMAP interconnect error handling driver.
 endmenu
diff --git a/drivers/bus/Makefile b/drivers/bus/Makefile
index 0ec50bc..45d997c 100644
--- a/drivers/bus/Makefile
+++ b/drivers/bus/Makefile
@@ -3,3 +3,6 @@
 #
 
 obj-$(CONFIG_OMAP_OCP2SCP)	+= omap-ocp2scp.o
+
+# Interconnect bus driver for OMAP SoCs.
+obj-$(CONFIG_OMAP_INTERCONNECT)	+= omap_l3_smx.o omap_l3_noc.o
diff --git a/arch/arm/mach-omap2/omap_l3_noc.c b/drivers/bus/omap_l3_noc.c
similarity index 100%
rename from arch/arm/mach-omap2/omap_l3_noc.c
rename to drivers/bus/omap_l3_noc.c
diff --git a/arch/arm/mach-omap2/omap_l3_noc.h b/drivers/bus/omap_l3_noc.h
similarity index 100%
rename from arch/arm/mach-omap2/omap_l3_noc.h
rename to drivers/bus/omap_l3_noc.h
diff --git a/arch/arm/mach-omap2/omap_l3_smx.c b/drivers/bus/omap_l3_smx.c
similarity index 100%
rename from arch/arm/mach-omap2/omap_l3_smx.c
rename to drivers/bus/omap_l3_smx.c
diff --git a/arch/arm/mach-omap2/omap_l3_smx.h b/drivers/bus/omap_l3_smx.h
similarity index 100%
rename from arch/arm/mach-omap2/omap_l3_smx.h
rename to drivers/bus/omap_l3_smx.h
-- 
1.7.9.5

Re: [PATCH 1/1] drivers: bus: Move the OMAP interconnect driver to drivers/bus/

[Tony Lindgren]

* Santosh Shilimkar <santosh.shilimkar@ti.com> [120914 02:21]:
> OMAP interconnect drivers are used for the interconnect error handling.
> Since they are bus driver, lets move it to newly created drivers/bus.
> 
> Cc: Arnd Bergmann <arnd@arndb.de>
> Cc: Tony Lindgren <tony@atomide.com>
> Tested-by: Lokesh Vutla <lokeshvutla@ti.com>
> Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
> ---
> Patch just moves OMAP interconnect drivers as is to the newly created
> driver/bus/* directory. Patch is generated against "arm-soc/drivers/ocp2scp"
> tree and test on all OMAP boards.

Great, looks like this should not conflict with other
omap patches queued, so Arnd should probably take this into
the bus branch:

Acked-by: Tony Lindgren <tony@atomide.com>

Re: [PATCH 1/1] drivers: bus: Move the OMAP interconnect driver to drivers/bus/

[Arnd Bergmann]

On Monday 17 September 2012, Tony Lindgren wrote:
> * Santosh Shilimkar <santosh.shilimkar@ti.com> [120914 02:21]:
> > OMAP interconnect drivers are used for the interconnect error handling.
> > Since they are bus driver, lets move it to newly created drivers/bus.
> > 
> > Cc: Arnd Bergmann <arnd@arndb.de>
> > Cc: Tony Lindgren <tony@atomide.com>
> > Tested-by: Lokesh Vutla <lokeshvutla@ti.com>
> > Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
> > ---
> > Patch just moves OMAP interconnect drivers as is to the newly created
> > driver/bus/* directory. Patch is generated against "arm-soc/drivers/ocp2scp"
> > tree and test on all OMAP boards.
> 
> Great, looks like this should not conflict with other
> omap patches queued, so Arnd should probably take this into
> the bus branch:
> 
> Acked-by: Tony Lindgren <tony@atomide.com>

Ok, added to the branch. In the future please Cc both Olof and me when submitting
patches for inclusion.

Olof: I figured it would be easier if I just add this one than relaying it to
you.

	Arnd

Re: [PATCH 1/1] drivers: bus: Move the OMAP interconnect driver to drivers/bus/

[Olof Johansson]

On Wed, Sep 19, 2012 at 7:58 AM, Arnd Bergmann <arnd@arndb.de> wrote:
> On Monday 17 September 2012, Tony Lindgren wrote:
>> * Santosh Shilimkar <santosh.shilimkar@ti.com> [120914 02:21]:
>> > OMAP interconnect drivers are used for the interconnect error handling.
>> > Since they are bus driver, lets move it to newly created drivers/bus.
>> >
>> > Cc: Arnd Bergmann <arnd@arndb.de>
>> > Cc: Tony Lindgren <tony@atomide.com>
>> > Tested-by: Lokesh Vutla <lokeshvutla@ti.com>
>> > Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
>> > ---
>> > Patch just moves OMAP interconnect drivers as is to the newly created
>> > driver/bus/* directory. Patch is generated against "arm-soc/drivers/ocp2scp"
>> > tree and test on all OMAP boards.
>>
>> Great, looks like this should not conflict with other
>> omap patches queued, so Arnd should probably take this into
>> the bus branch:
>>
>> Acked-by: Tony Lindgren <tony@atomide.com>
>
> Ok, added to the branch. In the future please Cc both Olof and me when submitting
> patches for inclusion.
>
> Olof: I figured it would be easier if I just add this one than relaying it to
> you.

Sounds good, thanks for the heads up.


-Olof

Re: [PATCH 1/1] drivers: bus: Move the OMAP interconnect driver to drivers/bus/

[Shilimkar, Santosh]

On Wed, Sep 19, 2012 at 8:28 PM, Arnd Bergmann <arnd@arndb.de> wrote:
> On Monday 17 September 2012, Tony Lindgren wrote:
>> * Santosh Shilimkar <santosh.shilimkar@ti.com> [120914 02:21]:
>> > OMAP interconnect drivers are used for the interconnect error handling.
>> > Since they are bus driver, lets move it to newly created drivers/bus.
>> >
>> > Cc: Arnd Bergmann <arnd@arndb.de>
>> > Cc: Tony Lindgren <tony@atomide.com>
>> > Tested-by: Lokesh Vutla <lokeshvutla@ti.com>
>> > Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
>> > ---
>> > Patch just moves OMAP interconnect drivers as is to the newly created
>> > driver/bus/* directory. Patch is generated against "arm-soc/drivers/ocp2scp"
>> > tree and test on all OMAP boards.
>>
>> Great, looks like this should not conflict with other
>> omap patches queued, so Arnd should probably take this into
>> the bus branch:
>>
>> Acked-by: Tony Lindgren <tony@atomide.com>
>
> Ok, added to the branch. In the future please Cc both Olof and me when submitting
> patches for inclusion.
>
Thanks Arnd. Sorry I missed to Cc Olof on the patch. Will take care of
this in future.

Regards
santosh

Re: [PATCH 1/1] drivers: bus: Move the OMAP interconnect driver to drivers/bus/

[Arnd Bergmann]

On Monday 17 September 2012, Tony Lindgren wrote:
> * Santosh Shilimkar <santosh.shilimkar@ti.com> [120914 02:21]:
> > OMAP interconnect drivers are used for the interconnect error handling.
> > Since they are bus driver, lets move it to newly created drivers/bus.
> > 
> > Cc: Arnd Bergmann <arnd@arndb.de>
> > Cc: Tony Lindgren <tony@atomide.com>
> > Tested-by: Lokesh Vutla <lokeshvutla@ti.com>
> > Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
> > ---
> > Patch just moves OMAP interconnect drivers as is to the newly created
> > driver/bus/* directory. Patch is generated against "arm-soc/drivers/ocp2scp"
> > tree and test on all OMAP boards.
> 
> Great, looks like this should not conflict with other
> omap patches queued, so Arnd should probably take this into
> the bus branch:
> 
> Acked-by: Tony Lindgren <tony@atomide.com>

It turns out that the patch actually did conflict and we now have a broken
omap2plus_defconfig. The patch below seems to fix it, but please verify
that this makes sense.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>

diff --git a/drivers/bus/omap_l3_noc.c b/drivers/bus/omap_l3_noc.c
index f447e02..ab911a3 100644
--- a/drivers/bus/omap_l3_noc.c
+++ b/drivers/bus/omap_l3_noc.c
@@ -28,7 +28,6 @@
 #include <linux/kernel.h>
 #include <linux/slab.h>
 
-#include "soc.h"
 #include "omap_l3_noc.h"
 
 /*
@@ -191,7 +190,7 @@ static int __devinit omap4_l3_probe(struct platform_device *pdev)
 			IRQF_DISABLED, "l3-dbg-irq", l3);
 	if (ret) {
 		pr_crit("L3: request_irq failed to register for 0x%x\n",
-						9 + OMAP44XX_IRQ_GIC_START);
+						l3->debug_irq);
 		goto err3;
 	}
 
@@ -201,7 +200,7 @@ static int __devinit omap4_l3_probe(struct platform_device *pdev)
 			IRQF_DISABLED, "l3-app-irq", l3);
 	if (ret) {
 		pr_crit("L3: request_irq failed to register for 0x%x\n",
-						10 + OMAP44XX_IRQ_GIC_START);
+						l3->app_irq);
 		goto err4;
 	}
 

Re: [PATCH 1/1] drivers: bus: Move the OMAP interconnect driver to drivers/bus/

[Tony Lindgren]

* Arnd Bergmann <arnd@arndb.de> [120919 11:59]:
> On Monday 17 September 2012, Tony Lindgren wrote:
> > * Santosh Shilimkar <santosh.shilimkar@ti.com> [120914 02:21]:
> > > OMAP interconnect drivers are used for the interconnect error handling.
> > > Since they are bus driver, lets move it to newly created drivers/bus.
> > > 
> > > Cc: Arnd Bergmann <arnd@arndb.de>
> > > Cc: Tony Lindgren <tony@atomide.com>
> > > Tested-by: Lokesh Vutla <lokeshvutla@ti.com>
> > > Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
> > > ---
> > > Patch just moves OMAP interconnect drivers as is to the newly created
> > > driver/bus/* directory. Patch is generated against "arm-soc/drivers/ocp2scp"
> > > tree and test on all OMAP boards.
> > 
> > Great, looks like this should not conflict with other
> > omap patches queued, so Arnd should probably take this into
> > the bus branch:
> > 
> > Acked-by: Tony Lindgren <tony@atomide.com>
> 
> It turns out that the patch actually did conflict and we now have a broken
> omap2plus_defconfig. The patch below seems to fix it, but please verify
> that this makes sense.

Heh a conflict caused by hard coded irqs for the error message :)
Looks like a good fix that removes some unncecessary dependencies:

Acked-by: Tony Lindgren <tony@atomide.com>
 
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> 
> diff --git a/drivers/bus/omap_l3_noc.c b/drivers/bus/omap_l3_noc.c
> index f447e02..ab911a3 100644
> --- a/drivers/bus/omap_l3_noc.c
> +++ b/drivers/bus/omap_l3_noc.c
> @@ -28,7 +28,6 @@
>  #include <linux/kernel.h>
>  #include <linux/slab.h>
>  
> -#include "soc.h"
>  #include "omap_l3_noc.h"
>  
>  /*
> @@ -191,7 +190,7 @@ static int __devinit omap4_l3_probe(struct platform_device *pdev)
>  			IRQF_DISABLED, "l3-dbg-irq", l3);
>  	if (ret) {
>  		pr_crit("L3: request_irq failed to register for 0x%x\n",
> -						9 + OMAP44XX_IRQ_GIC_START);
> +						l3->debug_irq);
>  		goto err3;
>  	}
>  
> @@ -201,7 +200,7 @@ static int __devinit omap4_l3_probe(struct platform_device *pdev)
>  			IRQF_DISABLED, "l3-app-irq", l3);
>  	if (ret) {
>  		pr_crit("L3: request_irq failed to register for 0x%x\n",
> -						10 + OMAP44XX_IRQ_GIC_START);
> +						l3->app_irq);
>  		goto err4;
>  	}
>  

Re: [PATCH 1/1] drivers: bus: Move the OMAP interconnect driver to drivers/bus/

[Shilimkar, Santosh]

On Thu, Sep 20, 2012 at 12:27 AM, Arnd Bergmann <arnd@arndb.de> wrote:
> On Monday 17 September 2012, Tony Lindgren wrote:
>> * Santosh Shilimkar <santosh.shilimkar@ti.com> [120914 02:21]:
>> > OMAP interconnect drivers are used for the interconnect error handling.
>> > Since they are bus driver, lets move it to newly created drivers/bus.
>> >
>> > Cc: Arnd Bergmann <arnd@arndb.de>
>> > Cc: Tony Lindgren <tony@atomide.com>
>> > Tested-by: Lokesh Vutla <lokeshvutla@ti.com>
>> > Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
>> > ---
>> > Patch just moves OMAP interconnect drivers as is to the newly created
>> > driver/bus/* directory. Patch is generated against "arm-soc/drivers/ocp2scp"
>> > tree and test on all OMAP boards.
>>
>> Great, looks like this should not conflict with other
>> omap patches queued, so Arnd should probably take this into
>> the bus branch:
>>
>> Acked-by: Tony Lindgren <tony@atomide.com>
>
> It turns out that the patch actually did conflict and we now have a broken
> omap2plus_defconfig. The patch below seems to fix it, but please verify
> that this makes sense.
>
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
>
Looks correct.

Acked-by: Santosh Shilimkar <santosh.shilimkar@ti.com>

Re: [PATCH 1/1] drivers: bus: Move the OMAP interconnect driver to drivers/bus/

[Olof Johansson]

On Wed, Sep 19, 2012 at 10:56 PM, Shilimkar, Santosh
<santosh.shilimkar@ti.com> wrote:
> On Thu, Sep 20, 2012 at 12:27 AM, Arnd Bergmann <arnd@arndb.de> wrote:
>> On Monday 17 September 2012, Tony Lindgren wrote:
>>> * Santosh Shilimkar <santosh.shilimkar@ti.com> [120914 02:21]:
>>> > OMAP interconnect drivers are used for the interconnect error handling.
>>> > Since they are bus driver, lets move it to newly created drivers/bus.
>>> >
>>> > Cc: Arnd Bergmann <arnd@arndb.de>
>>> > Cc: Tony Lindgren <tony@atomide.com>
>>> > Tested-by: Lokesh Vutla <lokeshvutla@ti.com>
>>> > Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
>>> > ---
>>> > Patch just moves OMAP interconnect drivers as is to the newly created
>>> > driver/bus/* directory. Patch is generated against "arm-soc/drivers/ocp2scp"
>>> > tree and test on all OMAP boards.
>>>
>>> Great, looks like this should not conflict with other
>>> omap patches queued, so Arnd should probably take this into
>>> the bus branch:
>>>
>>> Acked-by: Tony Lindgren <tony@atomide.com>
>>
>> It turns out that the patch actually did conflict and we now have a broken
>> omap2plus_defconfig. The patch below seems to fix it, but please verify
>> that this makes sense.
>>
>> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
>>
> Looks correct.
>
> Acked-by: Santosh Shilimkar <santosh.shilimkar@ti.com>

I ended up applying something very similar to this and carried the
ACKs on that, but I broke it up in two pieces; one for the
drivers/ocp2scp branch to fix the irq numbers, and another to just
for-next to deal with the soc.h include file (since that should
probably have been fixed at branch merge time).

Pushed out to for-next as well.


-Olof

Re: [PATCH 1/1] drivers: bus: Move the OMAP interconnect driver to drivers/bus/

[Tony Lindgren]

* Olof Johansson <olof@lixom.net> [120920 16:23]:
> On Wed, Sep 19, 2012 at 10:56 PM, Shilimkar, Santosh
> <santosh.shilimkar@ti.com> wrote:
> > On Thu, Sep 20, 2012 at 12:27 AM, Arnd Bergmann <arnd@arndb.de> wrote:
> >> On Monday 17 September 2012, Tony Lindgren wrote:
> >>> * Santosh Shilimkar <santosh.shilimkar@ti.com> [120914 02:21]:
> >>> > OMAP interconnect drivers are used for the interconnect error handling.
> >>> > Since they are bus driver, lets move it to newly created drivers/bus.
> >>> >
> >>> > Cc: Arnd Bergmann <arnd@arndb.de>
> >>> > Cc: Tony Lindgren <tony@atomide.com>
> >>> > Tested-by: Lokesh Vutla <lokeshvutla@ti.com>
> >>> > Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
> >>> > ---
> >>> > Patch just moves OMAP interconnect drivers as is to the newly created
> >>> > driver/bus/* directory. Patch is generated against "arm-soc/drivers/ocp2scp"
> >>> > tree and test on all OMAP boards.
> >>>
> >>> Great, looks like this should not conflict with other
> >>> omap patches queued, so Arnd should probably take this into
> >>> the bus branch:
> >>>
> >>> Acked-by: Tony Lindgren <tony@atomide.com>
> >>
> >> It turns out that the patch actually did conflict and we now have a broken
> >> omap2plus_defconfig. The patch below seems to fix it, but please verify
> >> that this makes sense.
> >>
> >> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> >>
> > Looks correct.
> >
> > Acked-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
> 
> I ended up applying something very similar to this and carried the
> ACKs on that, but I broke it up in two pieces; one for the
> drivers/ocp2scp branch to fix the irq numbers, and another to just
> for-next to deal with the soc.h include file (since that should
> probably have been fixed at branch merge time).
> 
> Pushed out to for-next as well.

OK thanks!

Tony