[RFC][PATCH] Add support for hook switch on ams-delta

[RFC][PATCH] Add support for hook switch on ams-delta

[Janusz Krzysztofik]

[-- Attachment #1: Type: text/plain, Size: 302 bytes --]

Hi,

This trivial patch adds gpio-keys compatible platform device definition to 
ams-delta board, that supports hook switch found on this videophone. It is 
derived from similiar definitions found in other boards code. The patch is 
based on linux-2.6.30-rc5. Any comments are welcome.

Cheers,
Janusz

[-- Attachment #2: ams-delta-hook-switch.patch --]
[-- Type: text/x-diff, Size: 1934 bytes --]

diff -Npru a/arch/arm/mach-omap1/board-ams-delta.c b/arch/arm/mach-omap1/board-ams-delta.c
--- a/arch/arm/mach-omap1/board-ams-delta.c	2009-05-17 17:58:18.000000000 +0200
+++ b/arch/arm/mach-omap1/board-ams-delta.c	2009-05-17 16:22:59.000000000 +0200
@@ -15,6 +15,7 @@
 #include <linux/kernel.h>
 #include <linux/init.h>
 #include <linux/input.h>
+#include <linux/gpio_keys.h>
 #include <linux/platform_device.h>
 
 #include <mach/hardware.h>
@@ -213,10 +214,35 @@ static struct platform_device ams_delta_
 	.id	= -1
 };
 
+static struct gpio_keys_button ams_delta_gpio_keys_buttons[] = {
+	[0] = {
+		.desc		= "hook_switch",
+		.type		= EV_SW,		/* or EV_KEY ? */
+		.code		= SW_HEADPHONE_INSERT,	/* or a new one ? */
+		.active_low	= 1,
+		.gpio		= AMS_DELTA_GPIO_PIN_HOOK_SWITCH,
+		.debounce_interval = 10,
+	},
+};
+
+static struct gpio_keys_platform_data ams_delta_gpio_keys = {
+	.buttons		= ams_delta_gpio_keys_buttons,
+	.nbuttons		= ARRAY_SIZE(ams_delta_gpio_keys_buttons),
+};
+
+static struct platform_device ams_delta_gpio_keys_device = {
+	.name			= "gpio-keys",
+	.id			= -1,
+	.dev			= {
+		.platform_data	= &ams_delta_gpio_keys,
+	},
+};
+
 static struct platform_device *ams_delta_devices[] __initdata = {
 	&ams_delta_kp_device,
 	&ams_delta_lcd_device,
 	&ams_delta_led_device,
+	&ams_delta_gpio_keys_device,
 };
 
 static void __init ams_delta_init(void)
@@ -233,6 +259,13 @@ static void __init ams_delta_init_irq(vo
 
 	omap_usb_init(&ams_delta_usb_config);
 	platform_add_devices(ams_delta_devices, ARRAY_SIZE(ams_delta_devices));
+
+	omap_cfg_reg(P20_1610_GPIO4); /* is this required? */
+
+	/* The hook switch state could be exposed over sysfs with gpio_export().
+	 * This should be done after the gpio-keys driver calls gpio_request(),
+	 * but I don't know how to do this from outside of the driver. */
+	/* gpio_export(AMS_DELTA_GPIO_PIN_HOOK_SWITCH, 0); */
 }
 
 static void __init ams_delta_map_io(void)

Re: [RFC][PATCH] Add support for hook switch on ams-delta

[Jonathan McDowell]

On Tue, May 26, 2009 at 08:57:20PM +0200, Janusz Krzysztofik wrote:
> This trivial patch adds gpio-keys compatible platform device definition to 
> ams-delta board, that supports hook switch found on this videophone. It is 
> derived from similiar definitions found in other boards code. The patch is 
> based on linux-2.6.30-rc5. Any comments are welcome.

I'm obviously too late as I've seen the "Applied" mail, but some
comments:

* I don't think SW_HEADPHONE_INSERT is appropriate. input guys, is it
  not reasonable to have SW_PHONE_HOOK or similar? I can't find anything
  else in tree I know to be a VOIP phone that has the concept of a hook
  switch except perhaps the TuxScreen and it doesn't have anything set
  up that I could find. I think you're correct that EV_SW is
  appropriate; it may remain in the off-hook state for some time and
  AFAICT the gpio-key driver would produce a repeating key press if you
  used EV_KEY.

* We assume the bootloader does the appropriate GPIO pin setup for us,
  so I don't think your omap_cfg_reg is required but it doesn't hurt in
  the unlikely event we ever replace the Amstrad PBL.

* The commented out code to include the GPIO status in sysfs shouldn't
  be included. Does the input layer not provide a way to obtain the
  state of the switch?

> diff -Npru a/arch/arm/mach-omap1/board-ams-delta.c b/arch/arm/mach-omap1/board-ams-delta.c
> --- a/arch/arm/mach-omap1/board-ams-delta.c	2009-05-17 17:58:18.000000000 +0200
> +++ b/arch/arm/mach-omap1/board-ams-delta.c	2009-05-17 16:22:59.000000000 +0200
> @@ -15,6 +15,7 @@
>  #include <linux/kernel.h>
>  #include <linux/init.h>
>  #include <linux/input.h>
> +#include <linux/gpio_keys.h>
>  #include <linux/platform_device.h>
>  
>  #include <mach/hardware.h>
> @@ -213,10 +214,35 @@ static struct platform_device ams_delta_
>  	.id	= -1
>  };
>  
> +static struct gpio_keys_button ams_delta_gpio_keys_buttons[] = {
> +	[0] = {
> +		.desc		= "hook_switch",
> +		.type		= EV_SW,		/* or EV_KEY ? */
> +		.code		= SW_HEADPHONE_INSERT,	/* or a new one ? */
> +		.active_low	= 1,
> +		.gpio		= AMS_DELTA_GPIO_PIN_HOOK_SWITCH,
> +		.debounce_interval = 10,
> +	},
> +};
> +
> +static struct gpio_keys_platform_data ams_delta_gpio_keys = {
> +	.buttons		= ams_delta_gpio_keys_buttons,
> +	.nbuttons		= ARRAY_SIZE(ams_delta_gpio_keys_buttons),
> +};
> +
> +static struct platform_device ams_delta_gpio_keys_device = {
> +	.name			= "gpio-keys",
> +	.id			= -1,
> +	.dev			= {
> +		.platform_data	= &ams_delta_gpio_keys,
> +	},
> +};
> +
>  static struct platform_device *ams_delta_devices[] __initdata = {
>  	&ams_delta_kp_device,
>  	&ams_delta_lcd_device,
>  	&ams_delta_led_device,
> +	&ams_delta_gpio_keys_device,
>  };
>  
>  static void __init ams_delta_init(void)
> @@ -233,6 +259,13 @@ static void __init ams_delta_init_irq(vo
>  
>  	omap_usb_init(&ams_delta_usb_config);
>  	platform_add_devices(ams_delta_devices, ARRAY_SIZE(ams_delta_devices));
> +
> +	omap_cfg_reg(P20_1610_GPIO4); /* is this required? */
> +
> +	/* The hook switch state could be exposed over sysfs with gpio_export().
> +	 * This should be done after the gpio-keys driver calls gpio_request(),
> +	 * but I don't know how to do this from outside of the driver. */
> +	/* gpio_export(AMS_DELTA_GPIO_PIN_HOOK_SWITCH, 0); */
>  }
>  
>  static void __init ams_delta_map_io(void)


J.

-- 
Revd. Jonathan McDowell, ULC | I don't sleep, I dream.

Re: [RFC][PATCH] Add support for hook switch on ams-delta

[Janusz Krzysztofik]

Hi Jonathan,

Wednesday 03 June 2009 00:04:33 Jonathan McDowell napisał(a):
> I'm obviously too late as I've seen the "Applied" mail,

No problem, I'm glad to hear from you.

> but some 
> comments:
>
> * I don't think SW_HEADPHONE_INSERT is appropriate. input guys, is it 
>   not reasonable to have SW_PHONE_HOOK or similar?

I do share you point of view. However, I didn't want to start a discussion if 
there is a need for another symbol or not before the patch got any 
acceptance.

> * We assume the bootloader does the appropriate GPIO pin setup for us,
>   so I don't think your omap_cfg_reg is required but it doesn't hurt in
>   the unlikely event we ever replace the Amstrad PBL.

OK, let it stay there. Do you see a need for replaceing it with a new 
ams_delta_hook_switch_init() function call that just calls omap_cfg_reg()?

> * The commented out code to include the GPIO status in sysfs shouldn't
>   be included.

Yes, I put it there to get a feedback.

>   Does the input layer not provide a way to obtain the 
>   state of the switch?

Yes, it does, with EVIOCGSW ioctl()[1]. I personally don't like this way of 
getting the switch status and would rather see it available over sysfs. 
However, input guys may have their own preferences and gpio-keys driver 
belongs to them.

Thanks,
Janusz

[1] http://www.spinics.net/lists/linux-input/msg03482.html

P.S. I include my Signed-off-by, as Tony have requested, to avoid the code 
without one floating around.

Signed-off-by: Janusz Krzysztofik <jkrzyszt@tis.icnet.pl>

> > diff -Npru a/arch/arm/mach-omap1/board-ams-delta.c
> > b/arch/arm/mach-omap1/board-ams-delta.c ---
> > a/arch/arm/mach-omap1/board-ams-delta.c	2009-05-17 17:58:18.000000000
> > +0200 +++ b/arch/arm/mach-omap1/board-ams-delta.c	2009-05-17
> > 16:22:59.000000000 +0200 @@ -15,6 +15,7 @@
> >  #include <linux/kernel.h>
> >  #include <linux/init.h>
> >  #include <linux/input.h>
> > +#include <linux/gpio_keys.h>
> >  #include <linux/platform_device.h>
> >
> >  #include <mach/hardware.h>
> > @@ -213,10 +214,35 @@ static struct platform_device ams_delta_
> >  	.id	= -1
> >  };
> >
> > +static struct gpio_keys_button ams_delta_gpio_keys_buttons[] = {
> > +	[0] = {
> > +		.desc		= "hook_switch",
> > +		.type		= EV_SW,		/* or EV_KEY ? */
> > +		.code		= SW_HEADPHONE_INSERT,	/* or a new one ? */
> > +		.active_low	= 1,
> > +		.gpio		= AMS_DELTA_GPIO_PIN_HOOK_SWITCH,
> > +		.debounce_interval = 10,
> > +	},
> > +};
> > +
> > +static struct gpio_keys_platform_data ams_delta_gpio_keys = {
> > +	.buttons		= ams_delta_gpio_keys_buttons,
> > +	.nbuttons		= ARRAY_SIZE(ams_delta_gpio_keys_buttons),
> > +};
> > +
> > +static struct platform_device ams_delta_gpio_keys_device = {
> > +	.name			= "gpio-keys",
> > +	.id			= -1,
> > +	.dev			= {
> > +		.platform_data	= &ams_delta_gpio_keys,
> > +	},
> > +};
> > +
> >  static struct platform_device *ams_delta_devices[] __initdata = {
> >  	&ams_delta_kp_device,
> >  	&ams_delta_lcd_device,
> >  	&ams_delta_led_device,
> > +	&ams_delta_gpio_keys_device,
> >  };
> >
> >  static void __init ams_delta_init(void)
> > @@ -233,6 +259,13 @@ static void __init ams_delta_init_irq(vo
> >
> >  	omap_usb_init(&ams_delta_usb_config);
> >  	platform_add_devices(ams_delta_devices, ARRAY_SIZE(ams_delta_devices));
> > +
> > +	omap_cfg_reg(P20_1610_GPIO4); /* is this required? */
> > +
> > +	/* The hook switch state could be exposed over sysfs with
> > gpio_export(). +	 * This should be done after the gpio-keys driver calls
> > gpio_request(), +	 * but I don't know how to do this from outside of the
> > driver. */ +	/* gpio_export(AMS_DELTA_GPIO_PIN_HOOK_SWITCH, 0); */
> >  }
> >
> >  static void __init ams_delta_map_io(void)
>
> J.


--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [RFC][PATCH] Add support for hook switch on ams-delta

[Jonathan McDowell]

On Wed, Jun 03, 2009 at 11:03:20AM +0200, Janusz Krzysztofik wrote:
> Wednesday 03 June 2009 00:04:33 Jonathan McDowell napisał(a):
> > I'm obviously too late as I've seen the "Applied" mail,
> 
> No problem, I'm glad to hear from you.
> 
> > but some 
> > comments:
> >
> > * I don't think SW_HEADPHONE_INSERT is appropriate. input guys, is it 
> >   not reasonable to have SW_PHONE_HOOK or similar?
> 
> I do share you point of view. However, I didn't want to start a discussion if 
> there is a need for another symbol or not before the patch got any 
> acceptance.
> 
> > * We assume the bootloader does the appropriate GPIO pin setup for us,
> >   so I don't think your omap_cfg_reg is required but it doesn't hurt in
> >   the unlikely event we ever replace the Amstrad PBL.
> 
> OK, let it stay there. Do you see a need for replaceing it with a new 
> ams_delta_hook_switch_init() function call that just calls omap_cfg_reg()?

I don't see a need for this; it's always present and not a lot of code
to have in the init function as it stands.

> > * The commented out code to include the GPIO status in sysfs shouldn't
> >   be included.
> 
> Yes, I put it there to get a feedback.
> 
> >   Does the input layer not provide a way to obtain the 
> >   state of the switch?
> 
> Yes, it does, with EVIOCGSW ioctl()[1]. I personally don't like this way of 
> getting the switch status and would rather see it available over sysfs. 
> However, input guys may have their own preferences and gpio-keys driver 
> belongs to them.
 
I think that's a discussion to have with the input guys rather than
putting a hack in the platform file then.

So really the only issue with the patch that remains is if it justifies
adding a new SW_PHONE_HOOK switch type?

J.

-- 
] http://www.earth.li/~noodles/ []  Purranoia: The fear that your cat  [
]  PGP/GPG Key @ the.earth.li   []         is up to something.         [
] via keyserver, web or email.  []                                     [
] RSA: 4096/2DA8B985            []                                     [
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [RFC][PATCH] Add support for hook switch on ams-delta

[Janusz Krzysztofik]

Hi,

Thursday 04 June 2009 20:18:33 Jonathan McDowell napisał(a):
> On Wed, Jun 03, 2009 at 11:03:20AM +0200, Janusz Krzysztofik wrote:
> > Wednesday 03 June 2009 00:04:33 Jonathan McDowell napisał(a):
> >
> > > * We assume the bootloader does the appropriate GPIO pin setup for us,
> > >   so I don't think your omap_cfg_reg is required but it doesn't hurt in
> > >   the unlikely event we ever replace the Amstrad PBL.
> >
> > OK, let it stay there. Do you see a need for replaceing it with a new
> > ams_delta_hook_switch_init() function call that just calls
> > omap_cfg_reg()?
>
> I don't see a need for this; it's always present and not a lot of code
> to have in the init function as it stands.

Fine, I'll only add a comment explainig the purpose of the call then.

> > >   Does the input layer not provide a way to obtain the
> > >   state of the switch?
> >
> > Yes, it does, with EVIOCGSW ioctl()[1]. I personally don't like this way
> > of getting the switch status and would rather see it available over
> > sysfs. However, input guys may have their own preferences and gpio-keys
> > driver belongs to them.
>
> I think that's a discussion to have with the input guys rather than
> putting a hack in the platform file then.

Sure. I have a patch for gpio-keys.c ready to submit, let's see how far we can 
get with the idea of exporting a gpio-keys driven switch state over gpiolib 
sysfs.

> So really the only issue with the patch that remains is if it justifies
> adding a new SW_PHONE_HOOK switch type?

I've just submitted a patch that adds new symbol definition to 
include/linux/input.h. Do I have to wait for acknowledgement before I 
resubmit my modified patch that depends on that, or can I submit it now?

Thanks,
Janusz
--
To unsubscribe from this list: send the line "unsubscribe linux-input" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [RFC][PATCH] Add support for hook switch on ams-delta

[Mark Brown]

On Tue, Jun 02, 2009 at 11:04:33PM +0100, Jonathan McDowell wrote:

> * The commented out code to include the GPIO status in sysfs shouldn't
>   be included. Does the input layer not provide a way to obtain the
>   state of the switch?

It does, but only via an ioctl().

Re: [RFC][PATCH] Add support for hook switch on ams-delta

[Mark Brown]

On Mon, Mar 29, 2010 at 02:46:48PM +0100, Mark Brown wrote:
> On Tue, Jun 02, 2009 at 11:04:33PM +0100, Jonathan McDowell wrote:
> 
> > * The commented out code to include the GPIO status in sysfs shouldn't
> >   be included. Does the input layer not provide a way to obtain the
> >   state of the switch?
> 
> It does, but only via an ioctl().

Sorry, clearly I'm managing to get some extremely old mail dripping
through here :(