GPIO chip select support in McSPI

GPIO chip select support in McSPI

[Ben Gamari]

I've included the OMAP list so that I can hopefully get some feedback
from folks more familiar with this code.

Background:

I've been working on adding support for GPIO chip select lines to the
McSPI driver. Grant has been working with me to try getting the
in-kernel interface right and we have finally converged on a solution
whereby a table of GPIO lines will be passed to the McSPI driver through
platform_device.device.platform_data. Unfortunately, as explained below,
there is no clear path to support this in the current McSPI
initialization code.


On Thu, 03 Mar 2011 14:42:07 -0700, Grant Likely <grant.likely@secretlab.ca> wrote:
> The spi_master platform device is registered somewhere in the
> inappropriate support code. You need to arrange to make sure the
> correct pdata is attached to it when it gets registered. There
> /should/ be a mechanism for doing so.
> 
> All spi devices already specify a cs number anyway. What you can do is
> use the cs number as the lookup index into the gpio table, and
> remember to reserve a cs# for the 'native' cs line.
> 
I've done as you suggested and added support to the McSPI driver by
passing a GPIO table through platform_data. A patch will be coming
shortly.

Unfortunately, I'm really not sure how to restructure the OMAP code to
pass this information along. Currently the McSPI devices are registered
in arch/arm/mach-omap2/devices.c (omap_init_mcspi). In order to make the
GPIO CS support configurable by board code, we need some way to change
the (currently static) platform_devices prior to registration. Does
anyone have any suggestions on how this code could be refactored to
allow for this with minimal code duplication? Obviously we could just
move the platform_devices into the board files but this seems like a lot
of code duplication to support functionality that few boards will use.

Thoughts?

- Ben

[PATCH] mcspi: Add support for GPIO chip select lines

[Ben Gamari]

Many applications require more chip select lines than the board or
processor allow. Introduce a mechanism to allow use of GPIO pins as
chip select lines with the McSPI controller. To use this functionality,
one simply provides a table mapping CS line numbers to GPIO numbers in
omap2_mcspi_platform_config.cs_gpios.

Signed-off-by: Ben Gamari <bgamari.foss@gmail.com>
---
 arch/arm/plat-omap/include/plat/mcspi.h |    3 ++-
 drivers/spi/omap2_mcspi.c               |   15 ++++++++++-----
 2 files changed, 12 insertions(+), 6 deletions(-)

diff --git a/arch/arm/plat-omap/include/plat/mcspi.h b/arch/arm/plat-omap/include/plat/mcspi.h
index 1254e49..cac1d84 100644
--- a/arch/arm/plat-omap/include/plat/mcspi.h
+++ b/arch/arm/plat-omap/include/plat/mcspi.h
@@ -2,7 +2,8 @@
 #define _OMAP2_MCSPI_H
 
 struct omap2_mcspi_platform_config {
-	unsigned short	num_cs;
+	unsigned short num_cs;
+	int *cs_gpios;
 };
 
 struct omap2_mcspi_device_config {
diff --git a/drivers/spi/omap2_mcspi.c b/drivers/spi/omap2_mcspi.c
index abb1ffb..59cbed4 100644
--- a/drivers/spi/omap2_mcspi.c
+++ b/drivers/spi/omap2_mcspi.c
@@ -35,6 +35,7 @@
 #include <linux/slab.h>
 
 #include <linux/spi/spi.h>
+#include <linux/gpio.h>
 
 #include <plat/dma.h>
 #include <plat/clock.h>
@@ -235,11 +236,15 @@ static void omap2_mcspi_set_enable(const struct spi_device *spi, int enable)
 
 static void omap2_mcspi_force_cs(struct spi_device *spi, int cs_active)
 {
-	u32 l;
-
-	l = mcspi_cached_chconf0(spi);
-	MOD_REG_BIT(l, OMAP2_MCSPI_CHCONF_FORCE, cs_active);
-	mcspi_write_chconf0(spi, l);
+	struct omap2_mcspi_platform_config *pconfig = spi->master->dev.platform_data;
+	if (pconfig->cs_gpios) {
+		int gpio = pconfig->cs_gpios[spi->chip_select];
+		gpio_set_value(gpio, cs_active);
+	} else {
+		u32 l = mcspi_cached_chconf0(spi);
+		MOD_REG_BIT(l, OMAP2_MCSPI_CHCONF_FORCE, cs_active);
+		mcspi_write_chconf0(spi, l);
+	}
 }
 
 static void omap2_mcspi_set_master_mode(struct spi_master *master)
-- 
1.7.0.4

Re: [PATCH] mcspi: Add support for GPIO chip select lines

[Grant Likely]

On Sun, Mar 13, 2011 at 03:05:19PM -0400, Ben Gamari wrote:
> Many applications require more chip select lines than the board or
> processor allow. Introduce a mechanism to allow use of GPIO pins as
> chip select lines with the McSPI controller. To use this functionality,
> one simply provides a table mapping CS line numbers to GPIO numbers in
> omap2_mcspi_platform_config.cs_gpios.
> 
> Signed-off-by: Ben Gamari <bgamari.foss@gmail.com>
> ---
>  arch/arm/plat-omap/include/plat/mcspi.h |    3 ++-
>  drivers/spi/omap2_mcspi.c               |   15 ++++++++++-----
>  2 files changed, 12 insertions(+), 6 deletions(-)
> 
> diff --git a/arch/arm/plat-omap/include/plat/mcspi.h b/arch/arm/plat-omap/include/plat/mcspi.h
> index 1254e49..cac1d84 100644
> --- a/arch/arm/plat-omap/include/plat/mcspi.h
> +++ b/arch/arm/plat-omap/include/plat/mcspi.h
> @@ -2,7 +2,8 @@
>  #define _OMAP2_MCSPI_H
>  
>  struct omap2_mcspi_platform_config {
> -	unsigned short	num_cs;
> +	unsigned short num_cs;
> +	int *cs_gpios;
>  };
>  
>  struct omap2_mcspi_device_config {
> diff --git a/drivers/spi/omap2_mcspi.c b/drivers/spi/omap2_mcspi.c
> index abb1ffb..59cbed4 100644
> --- a/drivers/spi/omap2_mcspi.c
> +++ b/drivers/spi/omap2_mcspi.c
> @@ -35,6 +35,7 @@
>  #include <linux/slab.h>
>  
>  #include <linux/spi/spi.h>
> +#include <linux/gpio.h>
>  
>  #include <plat/dma.h>
>  #include <plat/clock.h>
> @@ -235,11 +236,15 @@ static void omap2_mcspi_set_enable(const struct spi_device *spi, int enable)
>  
>  static void omap2_mcspi_force_cs(struct spi_device *spi, int cs_active)
>  {
> -	u32 l;
> -
> -	l = mcspi_cached_chconf0(spi);
> -	MOD_REG_BIT(l, OMAP2_MCSPI_CHCONF_FORCE, cs_active);
> -	mcspi_write_chconf0(spi, l);
> +	struct omap2_mcspi_platform_config *pconfig = spi->master->dev.platform_data;
> +	if (pconfig->cs_gpios) {
> +		int gpio = pconfig->cs_gpios[spi->chip_select];
> +		gpio_set_value(gpio, cs_active);
> +	} else {
> +		u32 l = mcspi_cached_chconf0(spi);
> +		MOD_REG_BIT(l, OMAP2_MCSPI_CHCONF_FORCE, cs_active);
> +		mcspi_write_chconf0(spi, l);
> +	}

What if the board wanted to use both the native SPI ss line as well as
one or more GPIOs?  You probably want to reserve cs0 for the native
gpio line.

Otherwise this patch looks good to me.

g.

Re: [PATCH] mcspi: Add support for GPIO chip select lines

[Ben Gamari]

On Mon, 14 Mar 2011 13:27:18 -0600, Grant Likely <grant.likely@secretlab.ca> wrote:
> What if the board wanted to use both the native SPI ss line as well as
> one or more GPIOs?  You probably want to reserve cs0 for the native
> gpio line.
> 
Hmm, I had thought about this and assumed it would be easiest to punt on
this, requiring the user to use the native line as a GPIO. This of
course assumes that all of the CS lines also have pinmux configurations
as GPIO pins. Is this not a good assumption? 

> Otherwise this patch looks good to me.
> 
Thanks!

- Ben

Re: [PATCH] mcspi: Add support for GPIO chip select lines

[Grant Likely]

On Mon, Mar 14, 2011 at 10:06:40PM -0400, Ben Gamari wrote:
> On Mon, 14 Mar 2011 13:27:18 -0600, Grant Likely <grant.likely@secretlab.ca> wrote:
> > What if the board wanted to use both the native SPI ss line as well as
> > one or more GPIOs?  You probably want to reserve cs0 for the native
> > gpio line.
> > 
> Hmm, I had thought about this and assumed it would be easiest to punt on
> this, requiring the user to use the native line as a GPIO. This of
> course assumes that all of the CS lines also have pinmux configurations
> as GPIO pins. Is this not a good assumption? 

As a general principle I would say no since it would mean adding a 2nd
GPIO device will potentially break the first if the infrastructure
isn't in place to handle the first line as a gpio.

g.

Re: GPIO chip select support in McSPI

[Grant Likely]

On Sun, Mar 13, 2011 at 03:04:11PM -0400, Ben Gamari wrote:
> I've included the OMAP list so that I can hopefully get some feedback
> from folks more familiar with this code.
> 
> Background:
> 
> I've been working on adding support for GPIO chip select lines to the
> McSPI driver. Grant has been working with me to try getting the
> in-kernel interface right and we have finally converged on a solution
> whereby a table of GPIO lines will be passed to the McSPI driver through
> platform_device.device.platform_data. Unfortunately, as explained below,
> there is no clear path to support this in the current McSPI
> initialization code.
> 
> 
> On Thu, 03 Mar 2011 14:42:07 -0700, Grant Likely <grant.likely@secretlab.ca> wrote:
> > The spi_master platform device is registered somewhere in the
> > inappropriate support code. You need to arrange to make sure the
> > correct pdata is attached to it when it gets registered. There
> > /should/ be a mechanism for doing so.
> > 
> > All spi devices already specify a cs number anyway. What you can do is
> > use the cs number as the lookup index into the gpio table, and
> > remember to reserve a cs# for the 'native' cs line.
> > 
> I've done as you suggested and added support to the McSPI driver by
> passing a GPIO table through platform_data. A patch will be coming
> shortly.
> 
> Unfortunately, I'm really not sure how to restructure the OMAP code to
> pass this information along. Currently the McSPI devices are registered
> in arch/arm/mach-omap2/devices.c (omap_init_mcspi). In order to make the
> GPIO CS support configurable by board code, we need some way to change
> the (currently static) platform_devices prior to registration. Does
> anyone have any suggestions on how this code could be refactored to
> allow for this with minimal code duplication? Obviously we could just
> move the platform_devices into the board files but this seems like a lot
> of code duplication to support functionality that few boards will use.

I see two solutions:
1) platform code registers a bus notifier so that it gets called back
before the device gets bound to a driver.  Then it can augment the
platform data.
2) add an api to arch/arm/mach-omap2/devices.c for providing a cs
table before the device gets registered (must be called before
arch_initcall() time).

g.

Re: GPIO chip select support in McSPI

[Ben Gamari]

On Mon, 14 Mar 2011 13:25:36 -0600, Grant Likely <grant.likely@secretlab.ca> wrote:
> I see two solutions:
> 1) platform code registers a bus notifier so that it gets called back
> before the device gets bound to a driver.  Then it can augment the
> platform data.
> 
This sounds like it might be a bit involved and I'm not terribly
familiar with these facets of the driver model. This would probably be
the more flexible approach but I think I'll stick with the simpler
option.

> 2) add an api to arch/arm/mach-omap2/devices.c for providing a cs
> table before the device gets registered (must be called before
> arch_initcall() time).
> 
This is along the lines of what I was thinking. Any thoughts on how to
get function called before arch_initcall()?

- Ben

Re: GPIO chip select support in McSPI

[Grant Likely]

On Mon, Mar 14, 2011 at 10:22:31PM -0400, Ben Gamari wrote:
> On Mon, 14 Mar 2011 13:25:36 -0600, Grant Likely <grant.likely@secretlab.ca> wrote:
> > I see two solutions:
> > 1) platform code registers a bus notifier so that it gets called back
> > before the device gets bound to a driver.  Then it can augment the
> > platform data.
> > 
> This sounds like it might be a bit involved and I'm not terribly
> familiar with these facets of the driver model. This would probably be
> the more flexible approach but I think I'll stick with the simpler
> option.
> 
> > 2) add an api to arch/arm/mach-omap2/devices.c for providing a cs
> > table before the device gets registered (must be called before
> > arch_initcall() time).
> > 
> This is along the lines of what I was thinking. Any thoughts on how to
> get function called before arch_initcall()?

You can call it from .init_machine()

g.

> 
> - Ben

Re: McSPI questions pertaining to GPIO chip select support

[Raju Sana]

Hi Ben,

Could you please post your patch as I am also working the similar design.
I need to modify the existing McSPI driver to support 3 codec chips.

Thanks.
Venkat Raju.


On Thu, Jan 28, 2010 at 9:40 AM, Ben Gamari <bgamari.foss-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:

> Hey all!
>
> Recently I've been thinking about adding support to the McSPI driver for
> using GPIO pins as chip selects. As a starting point, I've browsed the
> driver source trying to identify what changes would be necessary to add
> this support. It seems like the rough idea is,
>
>  a) add a cs_gpio field to the device struct (omap2_mcspi)
>  b) modify omap2_mcspi_force_cs() to use cs_gpio instead of the
>     CHxCONF[FORCE] bit if cs_gpio is valid
>
> Unfortunately, I'm having a bit of difficulty seeing how the existing
> configuration works. In master mode, it seems that the controller is
> configured in single forced channel mode (MCSPI_MODULECTRL_SINGLE).  As
> far as I can tell, this means that the driver is responsible for
> managing the chip selects through the MCSPI_CHxCONF[FORCE] bits. It
> seems that this is ideal as little will need to be changed to
> incorporate GPIO CS support.
>
> Nevertheless, the timing of the various operations isn't making sense to
> me. As far as I understand it, the callgraph is as follows,
>
>
> omap2_mcspi_work():
>  loop over messages
>    omap2_mcspi_set_enable(1)
>    loop over transfers:
>      configure bus parameters
>      if (!cs_active) cs = 1
>      if (dma):
>        omap2_mcspi_txrx_dma()
>      else:
>        omap2_mcspi_txrx_pio()
>      if (cs_change) cs = 0
>    end loop
>
>    if (cs_active) cs = 0
>    omap2_mcspi_set_enable(0)
>  end loop
>
>
> It seems that omap2_mcspi_set_enable() is called to enable the channel
> (MCSPI_CHCTRLx[EN] is set) before force_cs() is called. If I'm
> interpreting the hardware documentation correctly, the enable bit (among
> other things) starts the SPI clock, meaning that the chip will begin
> clocking out zeros to the device before CS is brought high.
> Furthermore, even after CS is brought high, it seems there could be a
> fair amount of time before the chip begins actually clocking out valid
> data.
>
> As I am writing this, it now occurs to me that perhaps the chip doesn't
> start SPI_CLK until it has data to send. It seems that if this were the
> case my concerns would be resolved. However, it is not at all clear that
> this is the case from the documentation. When exactly does the
> controller start the bus clock? Is this stated explicitly anywhere in
> the documentation?
>
> Any input you could offer would be greatly appreciated.
> Thanks for your time.
>
> - Ben
>
>
> P.S. The more and more I think about it, it seems like my hypothesis is
> the only logical behavior. It would be great if someone could confirm my
> suspicions, however. It really seems like this should be in the
> documentation either way.
>
> P.P.S. I apologize if this is something someone familiar with SPI should
> just know. My knowledge is a little spotty on this sort of thing.
> --
> To unsubscribe from this list: send the line "unsubscribe linux-omap" in
> the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>
------------------------------------------------------------------------------
Special Offer -- Download ArcSight Logger for FREE!
Finally, a world-class log management solution at an even better 
price-free! And you'll get a free "Love Thy Logs" t-shirt when you
download Logger. Secure your free ArcSight Logger TODAY!
http://p.sf.net/sfu/arcsisghtdev2dev

GPIO chip select support

[Ben Gamari]

Thanks for the poke. I've been meaning to send this set out again anyways.

This is the patch I am currently using (on top of v3.0). Conceptually it's
quite similar to the first patches I sent out but some implementation details
were slightly reworked to account for the generalization of the McSPI driver to
other silicon.

I would say that the patch is close to mergable save the fact that there is
currently no clean way to set up GPIO CS pins from a board file. I'll send the
horrible hack I currently use shortly. If anyone has any ideas on how this
might be done cleanly please let me know. It's sad that the mere lack of a
configuration interface is the reason this isn't upstream.

Cheers,

- Ben

[PATCH] mcspi: Add support for GPIO chip select lines

[Ben Gamari]

Many applications require more chip select lines than the board or
processor allow. Introduce a mechanism to allow use of GPIO pins as
chip select lines with the McSPI controller. To use this
tionality,
one simply provides a table mapping CS line numbers to GPIO numbers

omap2_mcspi_platform_config.cs_gpios.

Signed-off-by: Ben Gamari <bgamari.foss@gmail.com>
---
 arch/arm/plat-omap/include/plat/mcspi.h |    1 +
 drivers/spi/omap2_mcspi.c               |   19 +++++++++++++++++--
 2 files changed, 18 insertions(+), 2 deletions(-)

diff --git a/arch/arm/plat-omap/include/plat/mcspi.h b/arch/arm/plat-omap/include/plat/mcspi.h
index 3d51b18..c5670f2 100644
--- a/arch/arm/plat-omap/include/plat/mcspi.h
+++ b/arch/arm/plat-omap/include/plat/mcspi.h
@@ -10,6 +10,7 @@
 struct omap2_mcspi_platform_config {
 	unsigned short	num_cs;
 	unsigned int regs_offset;
+	int *cs_gpios;
 };
 
 struct omap2_mcspi_dev_attr {
diff --git a/drivers/spi/omap2_mcspi.c b/drivers/spi/omap2_mcspi.c
index f41c906..20dec5d 100644
--- a/drivers/spi/omap2_mcspi.c
+++ b/drivers/spi/omap2_mcspi.c
@@ -36,6 +36,7 @@
 #include <linux/pm_runtime.h>
 
 #include <linux/spi/spi.h>
+#include <linux/gpio.h>
 
 #include <plat/dma.h>
 #include <plat/clock.h>
@@ -121,6 +122,7 @@ struct omap2_mcspi {
 	/* SPI1 has 4 channels, while SPI2 has 2 */
 	struct omap2_mcspi_dma	*dma_channels;
 	struct  device		*dev;
+        int                     *cs_gpios;
 };
 
 struct omap2_mcspi_cs {
@@ -227,7 +229,11 @@ static void omap2_mcspi_set_enable(const struct spi_device *spi, int enable)
 static void omap2_mcspi_force_cs(struct spi_device *spi, int cs_active)
 {
 	u32 l;
+	struct omap2_mcspi* mcspi = spi_master_get_devdata(spi->master);
+	if (mcspi->cs_gpios)
+		gpio_set_value(mcspi->cs_gpios[spi->chip_select], cs_active);
 
+	// TXS times out unless we force the CHCONF reg as well
 	l = mcspi_cached_chconf0(spi);
 	MOD_REG_BIT(l, OMAP2_MCSPI_CHCONF_FORCE, cs_active);
 	mcspi_write_chconf0(spi, l);
@@ -1088,6 +1094,8 @@ static int __init omap2_mcspi_probe(struct platform_device *pdev)
 	struct omap2_mcspi	*mcspi;
 	struct resource		*r;
 	int			status = 0, i;
+	struct omap2_mcspi_platform_config* pconfig = pdev->dev.platform_data;
+	int                     num_dma;
 
 	master = spi_alloc_master(&pdev->dev, sizeof *mcspi);
 	if (master == NULL) {
@@ -1110,6 +1118,13 @@ static int __init omap2_mcspi_probe(struct platform_device *pdev)
 
 	mcspi = spi_master_get_devdata(master);
 	mcspi->master = master;
+	if (pconfig && pconfig->cs_gpios) {
+		mcspi->cs_gpios = pconfig->cs_gpios;
+		num_dma = 1;
+	} else {
+		mcspi->cs_gpios = NULL;
+		num_dma = master->num_chipselect;
+	}
 
 	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	if (r == NULL) {
@@ -1139,14 +1154,14 @@ static int __init omap2_mcspi_probe(struct platform_device *pdev)
 	INIT_LIST_HEAD(&mcspi->msg_queue);
 	INIT_LIST_HEAD(&omap2_mcspi_ctx[master->bus_num - 1].cs);
 
-	mcspi->dma_channels = kcalloc(master->num_chipselect,
+	mcspi->dma_channels = kcalloc(num_dma,
 			sizeof(struct omap2_mcspi_dma),
 			GFP_KERNEL);
 
 	if (mcspi->dma_channels == NULL)
 		goto err2;
 
-	for (i = 0; i < master->num_chipselect; i++) {
+	for (i = 0; i < num_dma; i++) {
 		char dma_ch_name[14];
 		struct resource *dma_res;
 
-- 
1.7.4.1

Re: McSPI questions pertaining to GPIO chip select support

[jassi brar]

On Thu, Jan 28, 2010 at 1:10 PM, Ben Gamari <bgamari.foss-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
> Hey all!
>
> Recently I've been thinking about adding support to the McSPI driver for
> using GPIO pins as chip selects. As a starting point, I've browsed the
> driver source trying to identify what changes would be necessary to add
> this support. It seems like the rough idea is,
You may loot at drivers/spi/spi_s3c64xx.c in Grant Likely's tree.

It doesn't put limit on the number of CS and not even on whether
the CS mechanism is simple GPIO toggling(though the callback
type is defined to match gpio_set_value).
For platform side of it, you need to look in to Ben Dooks' tree.

hth

------------------------------------------------------------------------------
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com

[PATCH] mcspi: Add support for GPIO chip select lines

[Ben Gamari]

This mechanism is in large part stolen from the s3c64xx-spi module. To
use this functionality, one simply must define a set_level function to
set the CS state and a omap2_mcspi_csinfo struct for each chip select in
the board file.

Each spi_board_info.controller_data should then be set
to point to the appropriate csinfo struct. This will cause the driver to
call the csinfo->set_level function instead of toggling the McSPI chip
select lines.

Signed-off-by: Ben Gamari <bgamari.foss@gmail.com>
---
 arch/arm/plat-omap/include/plat/mcspi.h |   14 ++++++++++++++
 drivers/spi/omap2_mcspi.c               |   14 +++++++++-----
 2 files changed, 23 insertions(+), 5 deletions(-)

diff --git a/arch/arm/plat-omap/include/plat/mcspi.h b/arch/arm/plat-omap/include/plat/mcspi.h
index 1254e49..ab84b8d 100644
--- a/arch/arm/plat-omap/include/plat/mcspi.h
+++ b/arch/arm/plat-omap/include/plat/mcspi.h
@@ -1,6 +1,20 @@
 #ifndef _OMAP2_MCSPI_H
 #define _OMAP2_MCSPI_H
 
+/**
+ * struct omap2_mcspi_csinfo - Chip Select description
+ * @line: Custom 'identity' of the CS line
+ * @set_level: Function to set the state of a given CS line
+ *
+ * This is to allow use of GPIO lines as CS lines. Allocate and initialize one
+ * in the machine init code and make spi_board_info.controller_data point to
+ * it.
+ */
+struct omap2_mcspi_csinfo {
+        unsigned line;
+        void (*set_level)(unsigned line_id, int lvl);
+};
+
 struct omap2_mcspi_platform_config {
 	unsigned short	num_cs;
 };
diff --git a/drivers/spi/omap2_mcspi.c b/drivers/spi/omap2_mcspi.c
index 2a651e6..92ccbd6 100644
--- a/drivers/spi/omap2_mcspi.c
+++ b/drivers/spi/omap2_mcspi.c
@@ -35,6 +35,7 @@
 #include <linux/slab.h>
 
 #include <linux/spi/spi.h>
+#include <linux/gpio.h>
 
 #include <plat/dma.h>
 #include <plat/clock.h>
@@ -235,11 +236,14 @@ static void omap2_mcspi_set_enable(const struct spi_device *spi, int enable)
 
 static void omap2_mcspi_force_cs(struct spi_device *spi, int cs_active)
 {
-	u32 l;
-
-	l = mcspi_cached_chconf0(spi);
-	MOD_REG_BIT(l, OMAP2_MCSPI_CHCONF_FORCE, cs_active);
-	mcspi_write_chconf0(spi, l);
+	if (spi->controller_data) {
+		struct omap2_mcspi_csinfo *csinfo = spi->controller_data;
+		(*csinfo->set_level)(csinfo->line, cs_active);
+	} else {
+		u32 l = mcspi_cached_chconf0(spi);
+		MOD_REG_BIT(l, OMAP2_MCSPI_CHCONF_FORCE, cs_active);
+		mcspi_write_chconf0(spi, l);
+	}
 }
 
 static void omap2_mcspi_set_master_mode(struct spi_master *master)
-- 
1.7.1

Re: [PATCH] mcspi: Add support for GPIO chip select lines

[Tony Lindgren]

* Ben Gamari <bgamari.foss@gmail.com> [101221 09:56]:
> This mechanism is in large part stolen from the s3c64xx-spi module. To
> use this functionality, one simply must define a set_level function to
> set the CS state and a omap2_mcspi_csinfo struct for each chip select in
> the board file.
> 
> Each spi_board_info.controller_data should then be set
> to point to the appropriate csinfo struct. This will cause the driver to
> call the csinfo->set_level function instead of toggling the McSPI chip
> select lines.
> 
> Signed-off-by: Ben Gamari <bgamari.foss@gmail.com>

This one should go via the SPI list:

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

> ---
>  arch/arm/plat-omap/include/plat/mcspi.h |   14 ++++++++++++++
>  drivers/spi/omap2_mcspi.c               |   14 +++++++++-----
>  2 files changed, 23 insertions(+), 5 deletions(-)
> 
> diff --git a/arch/arm/plat-omap/include/plat/mcspi.h b/arch/arm/plat-omap/include/plat/mcspi.h
> index 1254e49..ab84b8d 100644
> --- a/arch/arm/plat-omap/include/plat/mcspi.h
> +++ b/arch/arm/plat-omap/include/plat/mcspi.h
> @@ -1,6 +1,20 @@
>  #ifndef _OMAP2_MCSPI_H
>  #define _OMAP2_MCSPI_H
>  
> +/**
> + * struct omap2_mcspi_csinfo - Chip Select description
> + * @line: Custom 'identity' of the CS line
> + * @set_level: Function to set the state of a given CS line
> + *
> + * This is to allow use of GPIO lines as CS lines. Allocate and initialize one
> + * in the machine init code and make spi_board_info.controller_data point to
> + * it.
> + */
> +struct omap2_mcspi_csinfo {
> +        unsigned line;
> +        void (*set_level)(unsigned line_id, int lvl);
> +};
> +
>  struct omap2_mcspi_platform_config {
>  	unsigned short	num_cs;
>  };
> diff --git a/drivers/spi/omap2_mcspi.c b/drivers/spi/omap2_mcspi.c
> index 2a651e6..92ccbd6 100644
> --- a/drivers/spi/omap2_mcspi.c
> +++ b/drivers/spi/omap2_mcspi.c
> @@ -35,6 +35,7 @@
>  #include <linux/slab.h>
>  
>  #include <linux/spi/spi.h>
> +#include <linux/gpio.h>
>  
>  #include <plat/dma.h>
>  #include <plat/clock.h>
> @@ -235,11 +236,14 @@ static void omap2_mcspi_set_enable(const struct spi_device *spi, int enable)
>  
>  static void omap2_mcspi_force_cs(struct spi_device *spi, int cs_active)
>  {
> -	u32 l;
> -
> -	l = mcspi_cached_chconf0(spi);
> -	MOD_REG_BIT(l, OMAP2_MCSPI_CHCONF_FORCE, cs_active);
> -	mcspi_write_chconf0(spi, l);
> +	if (spi->controller_data) {
> +		struct omap2_mcspi_csinfo *csinfo = spi->controller_data;
> +		(*csinfo->set_level)(csinfo->line, cs_active);
> +	} else {
> +		u32 l = mcspi_cached_chconf0(spi);
> +		MOD_REG_BIT(l, OMAP2_MCSPI_CHCONF_FORCE, cs_active);
> +		mcspi_write_chconf0(spi, l);
> +	}
>  }
>  
>  static void omap2_mcspi_set_master_mode(struct spi_master *master)
> -- 
> 1.7.1
> 
> --
> 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: [PATCH] mcspi: Add support for GPIO chip select lines

[Grant Likely]

On Tue, Dec 21, 2010 at 10:56 AM, Ben Gamari <bgamari.foss@gmail.com> wrote:
> This mechanism is in large part stolen from the s3c64xx-spi module. To
> use this functionality, one simply must define a set_level function to
> set the CS state and a omap2_mcspi_csinfo struct for each chip select in
> the board file.
>
> Each spi_board_info.controller_data should then be set
> to point to the appropriate csinfo struct. This will cause the driver to
> call the csinfo->set_level function instead of toggling the McSPI chip
> select lines.
>
> Signed-off-by: Ben Gamari <bgamari.foss@gmail.com>

I'd rather see the spi driver modified to use the gpio api directly.
The drivers are already tending in that direction and it doesn't
require machine specific set_level functions to be defined.

g.

> ---
>  arch/arm/plat-omap/include/plat/mcspi.h |   14 ++++++++++++++
>  drivers/spi/omap2_mcspi.c               |   14 +++++++++-----
>  2 files changed, 23 insertions(+), 5 deletions(-)
>
> diff --git a/arch/arm/plat-omap/include/plat/mcspi.h b/arch/arm/plat-omap/include/plat/mcspi.h
> index 1254e49..ab84b8d 100644
> --- a/arch/arm/plat-omap/include/plat/mcspi.h
> +++ b/arch/arm/plat-omap/include/plat/mcspi.h
> @@ -1,6 +1,20 @@
>  #ifndef _OMAP2_MCSPI_H
>  #define _OMAP2_MCSPI_H
>
> +/**
> + * struct omap2_mcspi_csinfo - Chip Select description
> + * @line: Custom 'identity' of the CS line
> + * @set_level: Function to set the state of a given CS line
> + *
> + * This is to allow use of GPIO lines as CS lines. Allocate and initialize one
> + * in the machine init code and make spi_board_info.controller_data point to
> + * it.
> + */
> +struct omap2_mcspi_csinfo {
> +        unsigned line;
> +        void (*set_level)(unsigned line_id, int lvl);
> +};
> +
>  struct omap2_mcspi_platform_config {
>        unsigned short  num_cs;
>  };
> diff --git a/drivers/spi/omap2_mcspi.c b/drivers/spi/omap2_mcspi.c
> index 2a651e6..92ccbd6 100644
> --- a/drivers/spi/omap2_mcspi.c
> +++ b/drivers/spi/omap2_mcspi.c
> @@ -35,6 +35,7 @@
>  #include <linux/slab.h>
>
>  #include <linux/spi/spi.h>
> +#include <linux/gpio.h>
>
>  #include <plat/dma.h>
>  #include <plat/clock.h>
> @@ -235,11 +236,14 @@ static void omap2_mcspi_set_enable(const struct spi_device *spi, int enable)
>
>  static void omap2_mcspi_force_cs(struct spi_device *spi, int cs_active)
>  {
> -       u32 l;
> -
> -       l = mcspi_cached_chconf0(spi);
> -       MOD_REG_BIT(l, OMAP2_MCSPI_CHCONF_FORCE, cs_active);
> -       mcspi_write_chconf0(spi, l);
> +       if (spi->controller_data) {
> +               struct omap2_mcspi_csinfo *csinfo = spi->controller_data;
> +               (*csinfo->set_level)(csinfo->line, cs_active);
> +       } else {
> +               u32 l = mcspi_cached_chconf0(spi);
> +               MOD_REG_BIT(l, OMAP2_MCSPI_CHCONF_FORCE, cs_active);
> +               mcspi_write_chconf0(spi, l);
> +       }
>  }
>
>  static void omap2_mcspi_set_master_mode(struct spi_master *master)
> --
> 1.7.1
>
>



-- 
Grant Likely, B.Sc., P.Eng.
Secret Lab Technologies Ltd.
--
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: [PATCH] mcspi: Add support for GPIO chip select lines

[Ben Gamari]

On Thu, 23 Dec 2010 14:38:57 -0700, Grant Likely <grant.likely@secretlab.ca> wrote:
> On Tue, Dec 21, 2010 at 10:56 AM, Ben Gamari <bgamari.foss@gmail.com> wrote:
> > This mechanism is in large part stolen from the s3c64xx-spi module. To
> > use this functionality, one simply must define a set_level function to
> > set the CS state and a omap2_mcspi_csinfo struct for each chip select in
> > the board file.
> >
> > Each spi_board_info.controller_data should then be set
> > to point to the appropriate csinfo struct. This will cause the driver to
> > call the csinfo->set_level function instead of toggling the McSPI chip
> > select lines.
> >
> > Signed-off-by: Ben Gamari <bgamari.foss@gmail.com>
> 
> I'd rather see the spi driver modified to use the gpio api directly.
> The drivers are already tending in that direction and it doesn't
> require machine specific set_level functions to be defined.
> 
The reason I left this up to the board is it's easy to foresee cases
where you want a non-trivial mapping between logical CS numbers and CS
pin states. In my case, I using a 2-to-4 multiplexer as the source of
chip select.

- Ben

Re: [PATCH] mcspi: Add support for GPIO chip select lines

[Grant Likely]

On Thu, Dec 23, 2010 at 4:09 PM, Ben Gamari <bgamari.foss@gmail.com> wrote:
> On Thu, 23 Dec 2010 14:38:57 -0700, Grant Likely <grant.likely@secretlab.ca> wrote:
>> On Tue, Dec 21, 2010 at 10:56 AM, Ben Gamari <bgamari.foss@gmail.com> wrote:
>> > This mechanism is in large part stolen from the s3c64xx-spi module. To
>> > use this functionality, one simply must define a set_level function to
>> > set the CS state and a omap2_mcspi_csinfo struct for each chip select in
>> > the board file.
>> >
>> > Each spi_board_info.controller_data should then be set
>> > to point to the appropriate csinfo struct. This will cause the driver to
>> > call the csinfo->set_level function instead of toggling the McSPI chip
>> > select lines.
>> >
>> > Signed-off-by: Ben Gamari <bgamari.foss@gmail.com>
>>
>> I'd rather see the spi driver modified to use the gpio api directly.
>> The drivers are already tending in that direction and it doesn't
>> require machine specific set_level functions to be defined.
>>
> The reason I left this up to the board is it's easy to foresee cases
> where you want a non-trivial mapping between logical CS numbers and CS
> pin states. In my case, I using a 2-to-4 multiplexer as the source of
> chip select.

Hi Ben,

I understand and appreciate the motivation.  However in practice, the
gpio api is sufficient for pretty much any use case, even when the
backing gpio controller driver ends up driving some oddball device
with different constraints.  The big downside to using a callback is
that it forces all users to do the extra work of implementing the
callback.  With the gpio api, only the oddball cases have to do extra
code (to adapt the custom device to the gpio api).

g.

Re: [PATCH] mcspi: Add support for GPIO chip select lines

[Ben Gamari]

On Thu, 23 Dec 2010 17:37:12 -0700, Grant Likely <grant.likely@secretlab.ca> wrote:
> On Thu, Dec 23, 2010 at 4:09 PM, Ben Gamari <bgamari.foss@gmail.com> wrote:
> > The reason I left this up to the board is it's easy to foresee cases
> > where you want a non-trivial mapping between logical CS numbers and CS
> > pin states. In my case, I using a 2-to-4 multiplexer as the source of
> > chip select.
> 
> Hi Ben,
> 
> I understand and appreciate the motivation.  However in practice, the
> gpio api is sufficient for pretty much any use case, even when the
> backing gpio controller driver ends up driving some oddball device
> with different constraints.  The big downside to using a callback is
> that it forces all users to do the extra work of implementing the
> callback.  With the gpio api, only the oddball cases have to do extra
> code (to adapt the custom device to the gpio api).

I understand your concerns, but I'm not sure how to satisfy them without
crippling the design's ability to accomodate my use-case. I can't pass a
GPIO line per spi_board_info since in my case of a multiplexed CS
configuration a single pin's state does not uniquely determine the
desired CS. The only other option I can think of is that we somehow
provide a list of GPIOs for each bus and map the CS numbers to
permutations of GPIO states. Unfortunately, I don't know of any suitable
structure to put this GPIO list in. Perhaps I'm missing something obvious?

- Ben

Re: [PATCH] mcspi: Add support for GPIO chip select lines

[Grant Likely]

On Thu, Dec 23, 2010 at 09:27:20PM -0500, Ben Gamari wrote:
> On Thu, 23 Dec 2010 17:37:12 -0700, Grant Likely <grant.likely@secretlab.ca> wrote:
> > On Thu, Dec 23, 2010 at 4:09 PM, Ben Gamari <bgamari.foss@gmail.com> wrote:
> > > The reason I left this up to the board is it's easy to foresee cases
> > > where you want a non-trivial mapping between logical CS numbers and CS
> > > pin states. In my case, I using a 2-to-4 multiplexer as the source of
> > > chip select.
> > 
> > Hi Ben,
> > 
> > I understand and appreciate the motivation.  However in practice, the
> > gpio api is sufficient for pretty much any use case, even when the
> > backing gpio controller driver ends up driving some oddball device
> > with different constraints.  The big downside to using a callback is
> > that it forces all users to do the extra work of implementing the
> > callback.  With the gpio api, only the oddball cases have to do extra
> > code (to adapt the custom device to the gpio api).
> 
> I understand your concerns, but I'm not sure how to satisfy them without
> crippling the design's ability to accomodate my use-case. I can't pass a
> GPIO line per spi_board_info since in my case of a multiplexed CS
> configuration a single pin's state does not uniquely determine the
> desired CS. The only other option I can think of is that we somehow
> provide a list of GPIOs for each bus and map the CS numbers to
> permutations of GPIO states. Unfortunately, I don't know of any suitable
> structure to put this GPIO list in. Perhaps I'm missing something obvious?

Close, but not quite.  Assign one gpio number to each cs state, and
write a gpio controller driver that maps the linux-gpio number to the
physical gpio state permutation.  The mapping from gpio# to ss# is
1:1, but the driver behind the gpio# can do whatever you need it to
do.

g.

Re: [PATCH] mcspi: Add support for GPIO chip select lines

[Ben Gamari]

On Thu, 23 Dec 2010 20:28:19 -0700, Grant Likely <grant.likely@secretlab.ca> wrote:
> On Thu, Dec 23, 2010 at 09:27:20PM -0500, Ben Gamari wrote:
> > I understand your concerns, but I'm not sure how to satisfy them without
> > crippling the design's ability to accomodate my use-case. I can't pass a
> > GPIO line per spi_board_info since in my case of a multiplexed CS
> > configuration a single pin's state does not uniquely determine the
> > desired CS. The only other option I can think of is that we somehow
> > provide a list of GPIOs for each bus and map the CS numbers to
> > permutations of GPIO states. Unfortunately, I don't know of any suitable
> > structure to put this GPIO list in. Perhaps I'm missing something obvious?
> 
> Close, but not quite.  Assign one gpio number to each cs state, and
> write a gpio controller driver that maps the linux-gpio number to the
> physical gpio state permutation.  The mapping from gpio# to ss# is
> 1:1, but the driver behind the gpio# can do whatever you need it to
> do.
> 
I see. I'm still not convinced that this is the route to take,
however. It seems like this virtual gpio interface is not only pretty
clunky (simple board file glue turns into an entire gpio chip driver),
it seems like this is a very inaccurate and not very useful way to
expose a multiplexed CS configuration; e.g. what is this chip driver to
do if the user tries to set two CS pins at once? Is there precedent for
using the GPIO subsystem in this way?

Cheers,

- Ben

Re: [PATCH] mcspi: Add support for GPIO chip select lines

[Grant Likely]

On Fri, Dec 24, 2010 at 01:05:56AM -0500, Ben Gamari wrote:
> On Thu, 23 Dec 2010 20:28:19 -0700, Grant Likely <grant.likely@secretlab.ca> wrote:
> > On Thu, Dec 23, 2010 at 09:27:20PM -0500, Ben Gamari wrote:
> > > I understand your concerns, but I'm not sure how to satisfy them without
> > > crippling the design's ability to accomodate my use-case. I can't pass a
> > > GPIO line per spi_board_info since in my case of a multiplexed CS
> > > configuration a single pin's state does not uniquely determine the
> > > desired CS. The only other option I can think of is that we somehow
> > > provide a list of GPIOs for each bus and map the CS numbers to
> > > permutations of GPIO states. Unfortunately, I don't know of any suitable
> > > structure to put this GPIO list in. Perhaps I'm missing something obvious?
> > 
> > Close, but not quite.  Assign one gpio number to each cs state, and
> > write a gpio controller driver that maps the linux-gpio number to the
> > physical gpio state permutation.  The mapping from gpio# to ss# is
> > 1:1, but the driver behind the gpio# can do whatever you need it to
> > do.
> > 
> I see. I'm still not convinced that this is the route to take,
> however. It seems like this virtual gpio interface is not only pretty
> clunky (simple board file glue turns into an entire gpio chip driver),
> it seems like this is a very inaccurate and not very useful way to
> expose a multiplexed CS configuration; e.g. what is this chip driver to
> do if the user tries to set two CS pins at once?

(Sorry for the really laggy reply)

I don't see it being that big a deal.  gpio drivers are pretty
lightweight and it is fine to have domain-specific limitations on how
gpios for a particular "gpio controller" behave.  In your example, if
the hardware doesn't support enabling 2 CS pins at once, then you make
a choice, either setting one when the other is already set simply does
not work; or it could reset the other state.  Either choice would be
fine.  The SPI driver must do the right thing regardless and deselect
the previous CS line before enabling a new one.

The other alternative would be to implement a new SPI chipselect
common infrastructure, but IMO, that would just end up looking like a
duplication of the gpio infrastructure.

Regardless, my point still stands, platform callbacks for what amounts
to gpio manipulations doesn't make a whole lot of sense.

> Is there precedent for
> using the GPIO subsystem in this way?

Yes, in drivers/spi look at mpc52xx_spi.c, davinci_spi.c, ath79_spi.c,
atmel_spi.c, and many more.  Grep for gpio in drivers/spi.

g.

Re: [PATCH] mcspi: Add support for GPIO chip select lines

[Ben Gamari]

On Sat, 12 Feb 2011 01:33:24 -0700, Grant Likely <grant.likely@secretlab.ca> wrote:
> (Sorry for the really laggy reply)
> 
No worries, I was just starting to think about this again myself.

> I don't see it being that big a deal.  gpio drivers are pretty
> lightweight and it is fine to have domain-specific limitations on how
> gpios for a particular "gpio controller" behave.  In your example, if
> the hardware doesn't support enabling 2 CS pins at once, then you make
> a choice, either setting one when the other is already set simply does
> not work; or it could reset the other state.  Either choice would be
> fine.  The SPI driver must do the right thing regardless and deselect
> the previous CS line before enabling a new one.
> 
Having now finally having had a chance to implement a gpio_chip driver,
I agree. It's quite simple and seems to fill the need pretty well

> The other alternative would be to implement a new SPI chipselect
> common infrastructure, but IMO, that would just end up looking like a
> duplication of the gpio infrastructure.
> 
I think you are probably right. GPIO seems to be a good fit.

Patch coming shortly.  I use spi_board_info.controller_data to pass the
gpio number to the controller which doesn't seem like the best way
forward, but I wasn't sure how else to proceed. Also, since the gpio
numbers are dynamically determined spi_board_info.controller_data must
be filled in at runtime, which is also a bit of a bummer (although not
easily solved as far as I can tell). Let me know what you think.

 - Ben