[PATCH 00/03] staging: Emma Mobile USB driver and KZM9D board code

[PATCH 00/03] staging: Emma Mobile USB driver and KZM9D board code

[Magnus Damm]

Emma Mobile USB driver and KZM9D board code

[PATCH 01/03] staging: emxx_udc: Add Emma Mobile USB Gadget driver
[PATCH 02/03] staging: emxx_udc: I/O memory and IRQ resource support
[PATCH 03/03] staging: board: kzm9d: Board staging support for emxx_udc

This patch series contains an old USB Gadget driver for Emma Mobile
that gets slightly adjusted to make use of the platform device interface
which in turn is used to add USB Gadget support to the KZM9D board.

Two separate staging components are included in this series:
1) the emxx_udc driver - from out-of-tree Android 2.6.35.7
2) board staging support for KZM9D - platform device for DT-only KZM9D

The two components above will be used to continously improve the driver
and board integration code until the driver can be moved out of staging
and/or DT bindings are available so the board staging platform device code
can be replaced with a DT node.

Olof, patch 3/3 contains some board staging code for KZM9D. As it is
today the EMEV2 SoC is DT-only and we neither have defconfig nor Kconfig
definitions for the KZM9D board (emev2-kzm9d.dts). Instead of again adding
board code to the ARM subarchitecture I hereby propose that we use staging to
integrate non-DT drivers while they are improved to get DT support.

Greg, please let me know your opinions. I'd be happy to rework the code.

Signed-off-by: Magnus Damm <damm+renesas@opensource.se>
---

 Written against renesas-devel-v3.15-rc5-20140521

 drivers/staging/Kconfig             |    4 
 drivers/staging/Makefile            |    2 
 drivers/staging/board/Kconfig       |    7 
 drivers/staging/board/Makefile      |    1 
 drivers/staging/board/board.h       |   17 
 drivers/staging/board/kzm9d.c       |   18 
 drivers/staging/emxx_udc/Kconfig    |   10 
 drivers/staging/emxx_udc/Makefile   |    1 
 drivers/staging/emxx_udc/emxx_udc.c | 3616 ++++++++++++++++++++++++++++++++++-
 drivers/staging/emxx_udc/emxx_udc.h |  671 ++++++
 10 files changed, 4299 insertions(+), 48 deletions(-)

[PATCH 02/03] staging: emxx_udc: I/O memory and IRQ resource support

[Magnus Damm]

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

Adjust the emxx_udc driver to make use of the standard
driver model to pass I/O memory and IRQ as resources
instead of hard coding those things in the driver.

Needs more work - the VBUS signal is yet not handled.

Signed-off-by: Magnus Damm <damm+renesas@opensource.se>
---

 drivers/staging/emxx_udc/emxx_udc.c |   57 +++++++++++------------------------
 drivers/staging/emxx_udc/emxx_udc.h |    9 -----
 2 files changed, 18 insertions(+), 48 deletions(-)

--- 0002/drivers/staging/emxx_udc/emxx_udc.c
+++ work/drivers/staging/emxx_udc/emxx_udc.c	2014-05-22 16:20:02.000000000 +0900
@@ -3356,37 +3356,38 @@ static int nbu2ss_drv_probe(struct platf
 {
 	int	status = -ENODEV;
 	struct nbu2ss_udc	*udc;
+	struct resource *r;
+	int irq;
+	void __iomem *mmio_base;
 
 	udc = &udc_controller;
 	memset(udc, 0, sizeof(struct nbu2ss_udc));
 
 	platform_set_drvdata(pdev, udc);
 
-	/* IO Memory Region */
-	if (!request_mem_region(USB_BASE_ADDRESS, USB_BASE_SIZE
-				, driver_name)) {
+	/* require I/O memory and IRQ to be provided as resources */
+	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	mmio_base = devm_request_and_ioremap(&pdev->dev, r);
+	if (IS_ERR(mmio_base)) {
+		dev_err(&pdev->dev, "failed to map I/O memory\n");
+		return PTR_ERR(mmio_base);
+	}
 
-		ERR("request_mem_region failed\n");
-		return -EBUSY;
+	irq = platform_get_irq(pdev, 0);
+	if (irq < 0) {
+		dev_err(&pdev->dev, "failed to get IRQ\n");
+		return irq;
 	}
+	status = devm_request_irq(&pdev->dev, irq, _nbu2ss_udc_irq,
+				  0, driver_name, udc);
 
 	/* IO Memory */
-	udc->p_regs = (PT_FC_REGS)ioremap(USB_BASE_ADDRESS, USB_BASE_SIZE);
-	if (!udc->p_regs) {
-		ERR("request_io_mem failed\n");
-		goto cleanup3;
-	}
+	udc->p_regs = (PT_FC_REGS)mmio_base;
 
 	/* USB Function Controller Interrupt */
-	status = request_irq(USB_UDC_IRQ_1,
-				_nbu2ss_udc_irq,
-				IRQF_DISABLED,
-				driver_name,
-				udc);
-
 	if (status != 0) {
 		ERR("request_irq(USB_UDC_IRQ_1) failed\n");
-		goto cleanup2;
+		goto cleanup1;
 	}
 
 	/* Driver Initialization */
@@ -3412,18 +3413,6 @@ static int nbu2ss_drv_probe(struct platf
 	return status;
 
 cleanup1:
-	/* Interrupt Handler - Release */
-	free_irq(USB_UDC_IRQ_1, udc);
-
-cleanup2:
-	/* IO Memory - Release */
-	if (udc->p_regs)
-		iounmap(udc->p_regs);
-
-cleanup3:
-	/* IO Memory Region - Release */
-	release_mem_region(USB_BASE_ADDRESS, USB_BASE_SIZE);
-
 	return status;
 }
 
@@ -3456,18 +3445,8 @@ static int __exit nbu2ss_drv_remove(stru
 	}
 
 	/* Interrupt Handler - Release */
-	free_irq(USB_UDC_IRQ_1, udc);
-
-	/* Interrupt Handler - Release */
 	free_irq(INT_VBUS, udc);
 
-	/* IO Memory - Release */
-	if (udc->p_regs)
-		iounmap(udc->p_regs);
-
-	/* IO Memory Region - Release */
-	release_mem_region(USB_BASE_ADDRESS, USB_BASE_SIZE);
-
 	return 0;
 }
 
--- 0002/drivers/staging/emxx_udc/emxx_udc.h
+++ work/drivers/staging/emxx_udc/emxx_udc.h	2014-05-22 16:20:02.000000000 +0900
@@ -47,20 +47,11 @@
 
 
 /*------------ Board dependence(Resource) */
-#define USB_BASE_ADDRESS	EMXX_USBS1_BASE
-#define USB_BASE_SIZE		0x2000
-
-#define USB_UDC_IRQ_0		INT_USBF0
-#define USB_UDC_IRQ_1		INT_USBF1
 #define	VBUS_VALUE		GPIO_VBUS
 
 /* below hacked up for staging integration */
 #define GPIO_VBUS 0 /* GPIO_P153 on KZM9D */
 #define INT_VBUS 0 /* IRQ for GPIO_P153 */
-#define INT_USBF0 158
-#define INT_USBF1 159
-#define EMXX_USBS0_BASE 0xe2700000
-#define EMXX_USBS1_BASE 0xe2800000
 
 /*------------ Board dependence(Wait) */
 

[PATCH 03/03] staging: board: kzm9d: Board staging support for emxx_udc

[Magnus Damm]

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

Add staging board support for the KZM9D board and add
an emxx_udc platform device to allow in-tree continous 
development of the driver on the KZM9D board.

When DT bindings are ready for the emxx_udc driver then
the platform device in the KZM9D staging board code can
easily be removed. Until then we use platform devices
to continously improve the driver and integration code.

Signed-off-by: Magnus Damm <damm+renesas@opensource.se>
---

 drivers/staging/Kconfig        |    2 ++
 drivers/staging/Makefile       |    1 +
 drivers/staging/board/Kconfig  |    7 +++++++
 drivers/staging/board/Makefile |    1 +
 drivers/staging/board/board.h  |   17 +++++++++++++++++
 drivers/staging/board/kzm9d.c  |   18 ++++++++++++++++++
 6 files changed, 46 insertions(+)

--- 0002/drivers/staging/Kconfig
+++ work/drivers/staging/Kconfig	2014-05-22 15:31:57.000000000 +0900
@@ -112,6 +112,8 @@ source "drivers/staging/media/Kconfig"
 
 source "drivers/staging/android/Kconfig"
 
+source "drivers/staging/board/Kconfig"
+
 source "drivers/staging/ozwpan/Kconfig"
 
 source "drivers/staging/gdm72xx/Kconfig"
--- 0002/drivers/staging/Makefile
+++ work/drivers/staging/Makefile	2014-05-22 15:31:57.000000000 +0900
@@ -49,6 +49,7 @@ obj-$(CONFIG_TOUCHSCREEN_CLEARPAD_TM1217
 obj-$(CONFIG_TOUCHSCREEN_SYNAPTICS_I2C_RMI4)	+= ste_rmi4/
 obj-$(CONFIG_MFD_NVEC)		+= nvec/
 obj-$(CONFIG_ANDROID)		+= android/
+obj-$(CONFIG_STAGING_BOARD)	+= board/
 obj-$(CONFIG_USB_WPAN_HCD)	+= ozwpan/
 obj-$(CONFIG_WIMAX_GDM72XX)	+= gdm72xx/
 obj-$(CONFIG_LTE_GDM724X)	+= gdm724x/
--- /dev/null
+++ work/drivers/staging/board/Kconfig	2014-05-22 15:31:58.000000000 +0900
@@ -0,0 +1,7 @@
+config STAGING_BOARD
+	boolean "Staging Board Support"
+	help
+	  Select to enable per-board staging support code.
+
+	  If in doubt, say N here.
+
--- /dev/null
+++ work/drivers/staging/board/Makefile	2014-05-22 15:31:58.000000000 +0900
@@ -0,0 +1 @@
+obj-$(CONFIG_ARCH_SHMOBILE_MULTI)	+= kzm9d.o
--- /dev/null
+++ work/drivers/staging/board/board.h	2014-05-22 15:31:58.000000000 +0900
@@ -0,0 +1,17 @@
+#ifndef __BOARD_H__
+#define __BOARD_H__
+#include <linux/init.h>
+#include <linux/of.h>
+
+#define board_staging(str, fn)			\
+static int __init runtime_board_check(void)	\
+{						\
+	if (of_machine_is_compatible(str))	\
+		fn();				\
+						\
+	return 0;				\
+}						\
+						\
+late_initcall(runtime_board_check)
+
+#endif /* __BOARD_H__ */
--- /dev/null
+++ work/drivers/staging/board/kzm9d.c	2014-05-22 16:19:00.000000000 +0900
@@ -0,0 +1,18 @@
+/* Staging board support for KZM9D. Enable not-yet-DT-capable devices here. */
+
+#include <linux/kernel.h>
+#include <linux/platform_device.h>
+#include "board.h"
+
+static const struct resource usbs1_resources[] __initconst = {
+	DEFINE_RES_MEM(0xe2800000, 0x2000),
+	DEFINE_RES_IRQ(159),
+};
+
+static void __init kzm9d_init(void)
+{
+	platform_device_register_simple("emxx_udc", -1, usbs1_resources,
+					ARRAY_SIZE(usbs1_resources));
+}
+
+board_staging("renesas,kzm9d", kzm9d_init);

Re: [PATCH 00/03] staging: Emma Mobile USB driver and KZM9D board code

[Dan Carpenter]

Generally we take things as is into staging so there is no need for a
re-work if you don't want to.  But we do need a TODO file.  Read the
TODO files for other staging drivers.

regards,
dan carpenter

Re: [PATCH 03/03] staging: board: kzm9d: Board staging support for emxx_udc

[Geert Uytterhoeven]

Hi Magnus,

On Thu, May 22, 2014 at 10:54 AM, Magnus Damm <magnus.damm@gmail.com> wrote:
> Add staging board support for the KZM9D board and add
> an emxx_udc platform device to allow in-tree continous
> development of the driver on the KZM9D board.
>
> When DT bindings are ready for the emxx_udc driver then
> the platform device in the KZM9D staging board code can
> easily be removed. Until then we use platform devices
> to continously improve the driver and integration code.

Nice!

>  drivers/staging/Kconfig        |    2 ++
>  drivers/staging/Makefile       |    1 +
>  drivers/staging/board/Kconfig  |    7 +++++++
>  drivers/staging/board/Makefile |    1 +
>  drivers/staging/board/board.h  |   17 +++++++++++++++++
>  drivers/staging/board/kzm9d.c  |   18 ++++++++++++++++++
>  6 files changed, 46 insertions(+)

As you're adding infrastructure and a first user, I think this should
be split in two patches:
  - Introduction of drivers/staging/board infrastructure,
  - Addition of kzm9d board support.

> --- /dev/null
> +++ work/drivers/staging/board/Kconfig  2014-05-22 15:31:58.000000000 +0900
> @@ -0,0 +1,7 @@
> +config STAGING_BOARD
> +       boolean "Staging Board Support"
> +       help
> +         Select to enable per-board staging support code.
> +
> +         If in doubt, say N here.
> +

No board-specific Kconfig symbols, so the actual individual board
support itself is
always compiled-in (but see below)?

> --- /dev/null
> +++ work/drivers/staging/board/Makefile 2014-05-22 15:31:58.000000000 +0900
> @@ -0,0 +1 @@
> +obj-$(CONFIG_ARCH_SHMOBILE_MULTI)      += kzm9d.o

Ah, it's included in all shmobile multi-platform kernels.

What about testing for CONFIG_ARCH_EMEV2 instead of
CONFIG_ARCH_SHMOBILE_MULTI?
Or adding a KZM9D-specific Kconfig symbol to .../board/Kconfig?

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 00/03] staging: Emma Mobile USB driver and KZM9D board code

[Geert Uytterhoeven]

Hi Magnus,

Thanks for your series! This looks like a good solution for interim board
support.

On Thu, May 22, 2014 at 10:53 AM, Magnus Damm <magnus.damm@gmail.com> wrote:
> Olof, patch 3/3 contains some board staging code for KZM9D. As it is
> today the EMEV2 SoC is DT-only and we neither have defconfig nor Kconfig
> definitions for the KZM9D board (emev2-kzm9d.dts).

Although there's no KZM9D-specific defconfig anymore, support for it is
included in shmobile_defconfig, right?

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 03/03] staging: board: kzm9d: Board staging support for emxx_udc

[Magnus Damm]

Hi Geert,

On Thu, May 22, 2014 at 6:19 PM, Geert Uytterhoeven
<geert@linux-m68k.org> wrote:
> Hi Magnus,
>
> On Thu, May 22, 2014 at 10:54 AM, Magnus Damm <magnus.damm@gmail.com> wrote:
>> Add staging board support for the KZM9D board and add
>> an emxx_udc platform device to allow in-tree continous
>> development of the driver on the KZM9D board.
>>
>> When DT bindings are ready for the emxx_udc driver then
>> the platform device in the KZM9D staging board code can
>> easily be removed. Until then we use platform devices
>> to continously improve the driver and integration code.
>
> Nice!

Thanks, I hope it can turn into something useful!

>>  drivers/staging/Kconfig        |    2 ++
>>  drivers/staging/Makefile       |    1 +
>>  drivers/staging/board/Kconfig  |    7 +++++++
>>  drivers/staging/board/Makefile |    1 +
>>  drivers/staging/board/board.h  |   17 +++++++++++++++++
>>  drivers/staging/board/kzm9d.c  |   18 ++++++++++++++++++
>>  6 files changed, 46 insertions(+)
>
> As you're adding infrastructure and a first user, I think this should
> be split in two patches:
>   - Introduction of drivers/staging/board infrastructure,
>   - Addition of kzm9d board support.

I agree!

>> --- /dev/null
>> +++ work/drivers/staging/board/Kconfig  2014-05-22 15:31:58.000000000 +0900
>> @@ -0,0 +1,7 @@
>> +config STAGING_BOARD
>> +       boolean "Staging Board Support"
>> +       help
>> +         Select to enable per-board staging support code.
>> +
>> +         If in doubt, say N here.
>> +
>
> No board-specific Kconfig symbols, so the actual individual board
> support itself is
> always compiled-in (but see below)?

Yes, the user has a single switch to flip at this point.

>> --- /dev/null
>> +++ work/drivers/staging/board/Makefile 2014-05-22 15:31:58.000000000 +0900
>> @@ -0,0 +1 @@
>> +obj-$(CONFIG_ARCH_SHMOBILE_MULTI)      += kzm9d.o
>
> Ah, it's included in all shmobile multi-platform kernels.

Correct!

> What about testing for CONFIG_ARCH_EMEV2 instead of
> CONFIG_ARCH_SHMOBILE_MULTI?

That is quite fine with me.

> Or adding a KZM9D-specific Kconfig symbol to .../board/Kconfig?

Yeah, we could do that but my gut feeling is that it is better and
easier to just keep it simple. I'm happy to adjust in any direction
here though, so whatever makes people happy...!

Thanks for your feedback!

/ magnus

Re: [PATCH 00/03] staging: Emma Mobile USB driver and KZM9D board code

[Magnus Damm]

Hi Dan,

On Thu, May 22, 2014 at 6:17 PM, Dan Carpenter <dan.carpenter@oracle.com> wrote:
> Generally we take things as is into staging so there is no need for a
> re-work if you don't want to.  But we do need a TODO file.  Read the
> TODO files for other staging drivers.

Thanks for pointing that out, my apologies for being random and
skipping that! I will add a TODO file to v2!

Cheers,

/ magnus

On Thu, May 22, 2014 at 6:17 PM, Dan Carpenter <dan.carpenter@oracle.com> wrote:
> Generally we take things as is into staging so there is no need for a
> re-work if you don't want to.  But we do need a TODO file.  Read the
> TODO files for other staging drivers.
>
> regards,
> dan carpenter
>

Re: [PATCH 00/03] staging: Emma Mobile USB driver and KZM9D board code

[Magnus Damm]

Hi Geert,

On Thu, May 22, 2014 at 6:21 PM, Geert Uytterhoeven
<geert@linux-m68k.org> wrote:
> Hi Magnus,
>
> Thanks for your series! This looks like a good solution for interim board
> support.

Thanks!

> On Thu, May 22, 2014 at 10:53 AM, Magnus Damm <magnus.damm@gmail.com> wrote:
>> Olof, patch 3/3 contains some board staging code for KZM9D. As it is
>> today the EMEV2 SoC is DT-only and we neither have defconfig nor Kconfig
>> definitions for the KZM9D board (emev2-kzm9d.dts).
>
> Although there's no KZM9D-specific defconfig anymore, support for it is
> included in shmobile_defconfig, right?

Correct, I forgot to mention that.

Thanks,

/ magnus

Re: [PATCH 00/03] staging: Emma Mobile USB driver and KZM9D board code

[Greg KH]

On Thu, May 22, 2014 at 06:32:03PM +0900, Magnus Damm wrote:
> Hi Dan,
> 
> On Thu, May 22, 2014 at 6:17 PM, Dan Carpenter <dan.carpenter@oracle.com> wrote:
> > Generally we take things as is into staging so there is no need for a
> > re-work if you don't want to.  But we do need a TODO file.  Read the
> > TODO files for other staging drivers.
> 
> Thanks for pointing that out, my apologies for being random and
> skipping that! I will add a TODO file to v2!

Other than the missing TODO file, the idea looks good, I don't have any
objections to taking this.

thanks,

greg k-h