* [PATCH 1/4 v4] of_spi: add generic binding support to specify cs gpio @ 2012-03-07 12:23 Jean-Christophe PLAGNIOL-VILLARD [not found] ` <1331122989-14451-1-git-send-email-plagnioj-sclMFOaUSTBWk0Htik3J/w@public.gmane.org> 0 siblings, 1 reply; 4+ messages in thread From: Jean-Christophe PLAGNIOL-VILLARD @ 2012-03-07 12:23 UTC (permalink / raw) To: linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r Cc: spi-devel-general-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f, devicetree-discuss-uLR06cmDAlY/bJ5BZ2RsiQ This will allow to use gpio for chip select with no modification in the driver binding When use the cs-gpios, the gpio number will be passed via the cs_gpio field and the number of chip select will automatically increased. Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj-sclMFOaUSTBWk0Htik3J/w@public.gmane.org> Cc: devicetree-discuss-uLR06cmDAlY/bJ5BZ2RsiQ@public.gmane.org Cc: spi-devel-general-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org --- v4: use cs_gpio to pass the gpio number Best Regards, J. Documentation/devicetree/bindings/spi/spi-bus.txt | 6 ++ drivers/spi/spi.c | 57 +++++++++++++++++++- include/linux/spi/spi.h | 6 ++ 3 files changed, 66 insertions(+), 3 deletions(-) diff --git a/Documentation/devicetree/bindings/spi/spi-bus.txt b/Documentation/devicetree/bindings/spi/spi-bus.txt index e782add..c253379 100644 --- a/Documentation/devicetree/bindings/spi/spi-bus.txt +++ b/Documentation/devicetree/bindings/spi/spi-bus.txt @@ -12,6 +12,7 @@ The SPI master node requires the following properties: - #size-cells - should be zero. - compatible - name of SPI bus controller following generic names recommended practice. +- cs-gpios - (optional) gpios chip select. No other properties are required in the SPI bus node. It is assumed that a driver for an SPI bus device will understand that it is an SPI bus. However, the binding does not attempt to define the specific method for @@ -21,6 +22,8 @@ assumption that board specific platform code will be used to manage chip selects. Individual drivers can define additional properties to support describing the chip select layout. +If cs-gpios is used the number of chip select will automatically increased. + SPI slave nodes must be children of the SPI master node and can contain the following properties. - reg - (required) chip select address of device. @@ -34,6 +37,9 @@ contain the following properties. - spi-cs-high - (optional) Empty property indicating device requires chip select active high +If a gpio chipselect is used for the SPI slave the gpio number will be passed +via the controller_data + SPI example for an MPC5200 SPI bus: spi@f00 { #address-cells = <1>; diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c index b2ccdea..c1d0955 100644 --- a/drivers/spi/spi.c +++ b/drivers/spi/spi.c @@ -28,6 +28,7 @@ #include <linux/mod_devicetable.h> #include <linux/spi/spi.h> #include <linux/of_spi.h> +#include <linux/of_gpio.h> #include <linux/pm_runtime.h> #include <linux/export.h> @@ -322,6 +323,7 @@ struct spi_device *spi_alloc_device(struct spi_master *master) spi->dev.parent = &master->dev; spi->dev.bus = &spi_bus_type; spi->dev.release = spidev_release; + spi->cs_gpio = -EINVAL; device_initialize(&spi->dev); return spi; } @@ -339,15 +341,16 @@ EXPORT_SYMBOL_GPL(spi_alloc_device); int spi_add_device(struct spi_device *spi) { static DEFINE_MUTEX(spi_add_lock); - struct device *dev = spi->master->dev.parent; + struct spi_master *master = spi->master; + struct device *dev = master->dev.parent; struct device *d; int status; /* Chipselects are numbered 0..max; validate. */ - if (spi->chip_select >= spi->master->num_chipselect) { + if (spi->chip_select >= master->num_chipselect) { dev_err(dev, "cs%d >= max %d\n", spi->chip_select, - spi->master->num_chipselect); + master->num_chipselect); return -EINVAL; } @@ -371,6 +374,13 @@ int spi_add_device(struct spi_device *spi) goto done; } + if (master->num_gpio_cs && + spi->chip_select >= master->first_gpio_cs) { + int num = spi->chip_select - master->first_gpio_cs; + + spi->cs_gpio = master->cs_gpios[num]; + } + /* Drivers may modify this initial i/o setup, but will * normally rely on the device being setup. Devices * using SPI_CS_HIGH can't coexist well otherwise... @@ -561,6 +571,43 @@ struct spi_master *spi_alloc_master(struct device *dev, unsigned size) } EXPORT_SYMBOL_GPL(spi_alloc_master); +#ifdef CONFIG_OF +static int of_spi_register_master(struct spi_master *master) +{ + int nb, i; + int *cs; + struct device_node *np = master->dev.of_node; + + if (!np) + return 0; + + nb = of_gpio_named_count(np, "cs-gpios"); + + if (nb < 1) + return 0; + + cs = devm_kzalloc(&master->dev, sizeof(int) * nb, GFP_KERNEL); + master->cs_gpios = cs; + + if (!master->cs_gpios) + return -ENOMEM; + + master->first_gpio_cs = master->num_chipselect; + master->num_chipselect += nb; + master->num_gpio_cs = nb; + + for (i = 0; i < nb; i++) + cs[i] = of_get_named_gpio(np, "cs-gpios", i); + + return 0; +} +#else +static int of_spi_register_master(struct spi_master *master) +{ + return 0; +} +#endif + /** * spi_register_master - register SPI master controller * @master: initialized master, originally from spi_alloc_master() @@ -592,6 +639,10 @@ int spi_register_master(struct spi_master *master) if (!dev) return -ENODEV; + status = of_spi_register_master(master); + if (status) + return status; + /* even if it's just one always-selected device, there must * be at least one chipselect */ diff --git a/include/linux/spi/spi.h b/include/linux/spi/spi.h index 176fce9..7dcf031 100644 --- a/include/linux/spi/spi.h +++ b/include/linux/spi/spi.h @@ -89,6 +89,7 @@ struct spi_device { void *controller_state; void *controller_data; char modalias[SPI_NAME_SIZE]; + int cs_gpio; /* chip select gpio */ /* * likely need more hooks for more protocol options affecting how @@ -318,6 +319,11 @@ struct spi_master { /* called on release() to free memory provided by spi_master */ void (*cleanup)(struct spi_device *spi); + + /* gpio chip select */ + int *cs_gpios; + int first_gpio_cs; + int num_gpio_cs; }; static inline void *spi_master_get_devdata(struct spi_master *master) -- 1.7.7 ^ permalink raw reply related [flat|nested] 4+ messages in thread
[parent not found: <1331122989-14451-1-git-send-email-plagnioj-sclMFOaUSTBWk0Htik3J/w@public.gmane.org>]
* [PATCH 2/4 v4] spi/atmel: add DT support [not found] ` <1331122989-14451-1-git-send-email-plagnioj-sclMFOaUSTBWk0Htik3J/w@public.gmane.org> @ 2012-03-07 12:23 ` Jean-Christophe PLAGNIOL-VILLARD 2012-03-09 15:14 ` [PATCH 1/4 v4] of_spi: add generic binding support to specify cs gpio Grant Likely 1 sibling, 0 replies; 4+ messages in thread From: Jean-Christophe PLAGNIOL-VILLARD @ 2012-03-07 12:23 UTC (permalink / raw) To: linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r Cc: spi-devel-general-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f, devicetree-discuss-uLR06cmDAlY/bJ5BZ2RsiQ, Jean-Christophe PLAGNIOL-VILLARD The atmel_spi use only gpio for chip select. Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj-sclMFOaUSTBWk0Htik3J/w@public.gmane.org> Cc: devicetree-discuss-uLR06cmDAlY/bJ5BZ2RsiQ@public.gmane.org Cc: spi-devel-general-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org --- .../devicetree/bindings/spi/spi_atmel.txt | 6 +++++ drivers/spi/spi-atmel.c | 21 ++++++++++++++++--- 2 files changed, 23 insertions(+), 4 deletions(-) create mode 100644 Documentation/devicetree/bindings/spi/spi_atmel.txt diff --git a/Documentation/devicetree/bindings/spi/spi_atmel.txt b/Documentation/devicetree/bindings/spi/spi_atmel.txt new file mode 100644 index 0000000..7ec3d8d --- /dev/null +++ b/Documentation/devicetree/bindings/spi/spi_atmel.txt @@ -0,0 +1,6 @@ +Atmel SPI device + +Required properties: +- compatible : should be "atmel,at91rm9200-spi". +- reg: Address and length of the register set for the device +- interrupts: Should contain macb interrupt diff --git a/drivers/spi/spi-atmel.c b/drivers/spi/spi-atmel.c index 16d6a83..efbc5f8 100644 --- a/drivers/spi/spi-atmel.c +++ b/drivers/spi/spi-atmel.c @@ -19,6 +19,7 @@ #include <linux/interrupt.h> #include <linux/spi/spi.h> #include <linux/slab.h> +#include <linux/of.h> #include <asm/io.h> #include <mach/board.h> @@ -695,7 +696,7 @@ static int atmel_spi_setup(struct spi_device *spi) u32 scbr, csr; unsigned int bits = spi->bits_per_word; unsigned long bus_hz; - unsigned int npcs_pin; + int npcs_pin; int ret; as = spi_master_get_devdata(spi->master); @@ -767,7 +768,9 @@ static int atmel_spi_setup(struct spi_device *spi) csr |= SPI_BF(DLYBCT, 0); /* chipselect must have been muxed as GPIO (e.g. in board setup) */ - npcs_pin = (unsigned int)spi->controller_data; + if (!gpio_is_valid(spi->cs_gpio)) + spi->cs_gpio = (int)spi->controller_data; + npcs_pin = spi->cs_gpio; asd = spi->controller_state; if (!asd) { asd = kzalloc(sizeof(struct atmel_spi_device), GFP_KERNEL); @@ -887,7 +890,7 @@ static void atmel_spi_cleanup(struct spi_device *spi) { struct atmel_spi *as = spi_master_get_devdata(spi->master); struct atmel_spi_device *asd = spi->controller_state; - unsigned gpio = (unsigned) spi->controller_data; + unsigned gpio = spi->cs_gpio; unsigned long flags; if (!asd) @@ -938,7 +941,8 @@ static int __devinit atmel_spi_probe(struct platform_device *pdev) master->mode_bits = SPI_CPOL | SPI_CPHA | SPI_CS_HIGH; master->bus_num = pdev->id; - master->num_chipselect = 4; + master->dev.of_node = pdev->dev.of_node; + master->num_chipselect = master->dev.of_node ? 0 : 4; master->setup = atmel_spi_setup; master->transfer = atmel_spi_transfer; master->cleanup = atmel_spi_cleanup; @@ -1064,11 +1068,20 @@ static int atmel_spi_resume(struct platform_device *pdev) #define atmel_spi_resume NULL #endif +#if defined(CONFIG_OF) +static const struct of_device_id atmel_spi_dt_ids[] = { + { .compatible = "atmel,at91rm9200-spi" }, + { /* sentinel */ } +}; + +MODULE_DEVICE_TABLE(of, atmel_spi_dt_ids); +#endif static struct platform_driver atmel_spi_driver = { .driver = { .name = "atmel_spi", .owner = THIS_MODULE, + .of_match_table = of_match_ptr(atmel_spi_dt_ids), }, .suspend = atmel_spi_suspend, .resume = atmel_spi_resume, -- 1.7.7 ------------------------------------------------------------------------------ Virtualization & Cloud Management Using Capacity Planning Cloud computing makes use of virtualization - but cloud computing also focuses on allowing computing to be delivered as a service. http://www.accelacomm.com/jaw/sfnl/114/51521223/ ^ permalink raw reply related [flat|nested] 4+ messages in thread
* Re: [PATCH 1/4 v4] of_spi: add generic binding support to specify cs gpio [not found] ` <1331122989-14451-1-git-send-email-plagnioj-sclMFOaUSTBWk0Htik3J/w@public.gmane.org> 2012-03-07 12:23 ` [PATCH 2/4 v4] spi/atmel: add DT support Jean-Christophe PLAGNIOL-VILLARD @ 2012-03-09 15:14 ` Grant Likely 2012-03-09 18:02 ` Jean-Christophe PLAGNIOL-VILLARD 1 sibling, 1 reply; 4+ messages in thread From: Grant Likely @ 2012-03-09 15:14 UTC (permalink / raw) To: Jean-Christophe PLAGNIOL-VILLARD, linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r Cc: spi-devel-general-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f, devicetree-discuss-uLR06cmDAlY/bJ5BZ2RsiQ On Wed, 7 Mar 2012 13:23:06 +0100, Jean-Christophe PLAGNIOL-VILLARD <plagnioj-sclMFOaUSTBWk0Htik3J/w@public.gmane.org> wrote: > This will allow to use gpio for chip select with no modification in the > driver binding > > When use the cs-gpios, the gpio number will be passed via the cs_gpio field > and the number of chip select will automatically increased. > > Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj-sclMFOaUSTBWk0Htik3J/w@public.gmane.org> > Cc: devicetree-discuss-uLR06cmDAlY/bJ5BZ2RsiQ@public.gmane.org > Cc: spi-devel-general-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org > --- > v4: > use cs_gpio to pass the gpio number > > Best Regards, > J. > Documentation/devicetree/bindings/spi/spi-bus.txt | 6 ++ > drivers/spi/spi.c | 57 +++++++++++++++++++- > include/linux/spi/spi.h | 6 ++ > 3 files changed, 66 insertions(+), 3 deletions(-) > > diff --git a/Documentation/devicetree/bindings/spi/spi-bus.txt b/Documentation/devicetree/bindings/spi/spi-bus.txt > index e782add..c253379 100644 > --- a/Documentation/devicetree/bindings/spi/spi-bus.txt > +++ b/Documentation/devicetree/bindings/spi/spi-bus.txt > @@ -12,6 +12,7 @@ The SPI master node requires the following properties: > - #size-cells - should be zero. > - compatible - name of SPI bus controller following generic names > recommended practice. > +- cs-gpios - (optional) gpios chip select. > No other properties are required in the SPI bus node. It is assumed > that a driver for an SPI bus device will understand that it is an SPI bus. > However, the binding does not attempt to define the specific method for > @@ -21,6 +22,8 @@ assumption that board specific platform code will be used to manage > chip selects. Individual drivers can define additional properties to > support describing the chip select layout. > > +If cs-gpios is used the number of chip select will automatically increased. > + > SPI slave nodes must be children of the SPI master node and can > contain the following properties. > - reg - (required) chip select address of device. > @@ -34,6 +37,9 @@ contain the following properties. > - spi-cs-high - (optional) Empty property indicating device requires > chip select active high > > +If a gpio chipselect is used for the SPI slave the gpio number will be passed > +via the controller_data > + > SPI example for an MPC5200 SPI bus: > spi@f00 { > #address-cells = <1>; > diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c > index b2ccdea..c1d0955 100644 > --- a/drivers/spi/spi.c > +++ b/drivers/spi/spi.c > @@ -28,6 +28,7 @@ > #include <linux/mod_devicetable.h> > #include <linux/spi/spi.h> > #include <linux/of_spi.h> > +#include <linux/of_gpio.h> > #include <linux/pm_runtime.h> > #include <linux/export.h> > > @@ -322,6 +323,7 @@ struct spi_device *spi_alloc_device(struct spi_master *master) > spi->dev.parent = &master->dev; > spi->dev.bus = &spi_bus_type; > spi->dev.release = spidev_release; > + spi->cs_gpio = -EINVAL; > device_initialize(&spi->dev); > return spi; > } > @@ -339,15 +341,16 @@ EXPORT_SYMBOL_GPL(spi_alloc_device); > int spi_add_device(struct spi_device *spi) > { > static DEFINE_MUTEX(spi_add_lock); > - struct device *dev = spi->master->dev.parent; > + struct spi_master *master = spi->master; > + struct device *dev = master->dev.parent; > struct device *d; > int status; > > /* Chipselects are numbered 0..max; validate. */ > - if (spi->chip_select >= spi->master->num_chipselect) { > + if (spi->chip_select >= master->num_chipselect) { > dev_err(dev, "cs%d >= max %d\n", > spi->chip_select, > - spi->master->num_chipselect); > + master->num_chipselect); > return -EINVAL; > } > > @@ -371,6 +374,13 @@ int spi_add_device(struct spi_device *spi) > goto done; > } > > + if (master->num_gpio_cs && > + spi->chip_select >= master->first_gpio_cs) { > + int num = spi->chip_select - master->first_gpio_cs; > + > + spi->cs_gpio = master->cs_gpios[num]; spi->cs_gpio isn't used anywhere. > + } > + > /* Drivers may modify this initial i/o setup, but will > * normally rely on the device being setup. Devices > * using SPI_CS_HIGH can't coexist well otherwise... > @@ -561,6 +571,43 @@ struct spi_master *spi_alloc_master(struct device *dev, unsigned size) > } > EXPORT_SYMBOL_GPL(spi_alloc_master); > > +#ifdef CONFIG_OF > +static int of_spi_register_master(struct spi_master *master) > +{ > + int nb, i; > + int *cs; > + struct device_node *np = master->dev.of_node; > + > + if (!np) > + return 0; > + > + nb = of_gpio_named_count(np, "cs-gpios"); > + > + if (nb < 1) > + return 0; > + > + cs = devm_kzalloc(&master->dev, sizeof(int) * nb, GFP_KERNEL); > + master->cs_gpios = cs; > + > + if (!master->cs_gpios) > + return -ENOMEM; > + > + master->first_gpio_cs = master->num_chipselect; > + master->num_chipselect += nb; > + master->num_gpio_cs = nb; I think this can be simplified. Rather than trying to optimize by having a set of integrated cs lines followed by a gpio set, I think it would be better to have a single cs_gpios table that covers the whole range of num_chipselect; even if they overlap natural cs lines. That will make for an easier to use binding since users won't need to remember how many natural cs lines are on the device when filling the cs-gpios property. It will also make the code simpler because first_gpio_cs and num_gpio_cs can both be eliminated. It also means the core can have exact knowledge about which gpio is associated with which cs number. > + > + for (i = 0; i < nb; i++) > + cs[i] = of_get_named_gpio(np, "cs-gpios", i); > + > + return 0; > +} > +#else > +static int of_spi_register_master(struct spi_master *master) > +{ > + return 0; > +} > +#endif > + > /** > * spi_register_master - register SPI master controller > * @master: initialized master, originally from spi_alloc_master() > @@ -592,6 +639,10 @@ int spi_register_master(struct spi_master *master) > if (!dev) > return -ENODEV; > > + status = of_spi_register_master(master); > + if (status) > + return status; > + > /* even if it's just one always-selected device, there must > * be at least one chipselect > */ > diff --git a/include/linux/spi/spi.h b/include/linux/spi/spi.h > index 176fce9..7dcf031 100644 > --- a/include/linux/spi/spi.h > +++ b/include/linux/spi/spi.h > @@ -89,6 +89,7 @@ struct spi_device { > void *controller_state; > void *controller_data; > char modalias[SPI_NAME_SIZE]; > + int cs_gpio; /* chip select gpio */ > > /* > * likely need more hooks for more protocol options affecting how > @@ -318,6 +319,11 @@ struct spi_master { > > /* called on release() to free memory provided by spi_master */ > void (*cleanup)(struct spi_device *spi); > + > + /* gpio chip select */ > + int *cs_gpios; > + int first_gpio_cs; > + int num_gpio_cs; > }; > > static inline void *spi_master_get_devdata(struct spi_master *master) > -- > 1.7.7 > > _______________________________________________ > devicetree-discuss mailing list > devicetree-discuss-uLR06cmDAlY/bJ5BZ2RsiQ@public.gmane.org > https://lists.ozlabs.org/listinfo/devicetree-discuss -- Grant Likely, B.Sc, P.Eng. Secret Lab Technologies,Ltd. ------------------------------------------------------------------------------ Virtualization & Cloud Management Using Capacity Planning Cloud computing makes use of virtualization - but cloud computing also focuses on allowing computing to be delivered as a service. http://www.accelacomm.com/jaw/sfnl/114/51521223/ ^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH 1/4 v4] of_spi: add generic binding support to specify cs gpio 2012-03-09 15:14 ` [PATCH 1/4 v4] of_spi: add generic binding support to specify cs gpio Grant Likely @ 2012-03-09 18:02 ` Jean-Christophe PLAGNIOL-VILLARD 0 siblings, 0 replies; 4+ messages in thread From: Jean-Christophe PLAGNIOL-VILLARD @ 2012-03-09 18:02 UTC (permalink / raw) To: Grant Likely Cc: spi-devel-general-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f, devicetree-discuss-uLR06cmDAlY/bJ5BZ2RsiQ, linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r On 08:14 Fri 09 Mar , Grant Likely wrote: > On Wed, 7 Mar 2012 13:23:06 +0100, Jean-Christophe PLAGNIOL-VILLARD <plagnioj-sclMFOaUSTBWk0Htik3J/w@public.gmane.org> wrote: > > This will allow to use gpio for chip select with no modification in the > > driver binding > > > > When use the cs-gpios, the gpio number will be passed via the cs_gpio field > > and the number of chip select will automatically increased. > > > > Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj-sclMFOaUSTBWk0Htik3J/w@public.gmane.org> > > Cc: devicetree-discuss-uLR06cmDAlY/bJ5BZ2RsiQ@public.gmane.org > > Cc: spi-devel-general-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org > > --- > > v4: > > use cs_gpio to pass the gpio number > > > > Best Regards, > > J. > > Documentation/devicetree/bindings/spi/spi-bus.txt | 6 ++ > > drivers/spi/spi.c | 57 +++++++++++++++++++- > > include/linux/spi/spi.h | 6 ++ > > 3 files changed, 66 insertions(+), 3 deletions(-) > > > > diff --git a/Documentation/devicetree/bindings/spi/spi-bus.txt b/Documentation/devicetree/bindings/spi/spi-bus.txt > > index e782add..c253379 100644 > > --- a/Documentation/devicetree/bindings/spi/spi-bus.txt > > +++ b/Documentation/devicetree/bindings/spi/spi-bus.txt > > @@ -12,6 +12,7 @@ The SPI master node requires the following properties: > > - #size-cells - should be zero. > > - compatible - name of SPI bus controller following generic names > > recommended practice. > > +- cs-gpios - (optional) gpios chip select. > > No other properties are required in the SPI bus node. It is assumed > > that a driver for an SPI bus device will understand that it is an SPI bus. > > However, the binding does not attempt to define the specific method for > > @@ -21,6 +22,8 @@ assumption that board specific platform code will be used to manage > > chip selects. Individual drivers can define additional properties to > > support describing the chip select layout. > > > > +If cs-gpios is used the number of chip select will automatically increased. > > + > > SPI slave nodes must be children of the SPI master node and can > > contain the following properties. > > - reg - (required) chip select address of device. > > @@ -34,6 +37,9 @@ contain the following properties. > > - spi-cs-high - (optional) Empty property indicating device requires > > chip select active high > > > > +If a gpio chipselect is used for the SPI slave the gpio number will be passed > > +via the controller_data > > + > > SPI example for an MPC5200 SPI bus: > > spi@f00 { > > #address-cells = <1>; > > diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c > > index b2ccdea..c1d0955 100644 > > --- a/drivers/spi/spi.c > > +++ b/drivers/spi/spi.c > > @@ -28,6 +28,7 @@ > > #include <linux/mod_devicetable.h> > > #include <linux/spi/spi.h> > > #include <linux/of_spi.h> > > +#include <linux/of_gpio.h> > > #include <linux/pm_runtime.h> > > #include <linux/export.h> > > > > @@ -322,6 +323,7 @@ struct spi_device *spi_alloc_device(struct spi_master *master) > > spi->dev.parent = &master->dev; > > spi->dev.bus = &spi_bus_type; > > spi->dev.release = spidev_release; > > + spi->cs_gpio = -EINVAL; > > device_initialize(&spi->dev); > > return spi; > > } > > @@ -339,15 +341,16 @@ EXPORT_SYMBOL_GPL(spi_alloc_device); > > int spi_add_device(struct spi_device *spi) > > { > > static DEFINE_MUTEX(spi_add_lock); > > - struct device *dev = spi->master->dev.parent; > > + struct spi_master *master = spi->master; > > + struct device *dev = master->dev.parent; > > struct device *d; > > int status; > > > > /* Chipselects are numbered 0..max; validate. */ > > - if (spi->chip_select >= spi->master->num_chipselect) { > > + if (spi->chip_select >= master->num_chipselect) { > > dev_err(dev, "cs%d >= max %d\n", > > spi->chip_select, > > - spi->master->num_chipselect); > > + master->num_chipselect); > > return -EINVAL; > > } > > > > @@ -371,6 +374,13 @@ int spi_add_device(struct spi_device *spi) > > goto done; > > } > > > > + if (master->num_gpio_cs && > > + spi->chip_select >= master->first_gpio_cs) { > > + int num = spi->chip_select - master->first_gpio_cs; > > + > > + spi->cs_gpio = master->cs_gpios[num]; > > spi->cs_gpio isn't used anywhere. driver as example atmel spi I choose to let the driver to manage it as they may need to manange specific register when using both hw and gpio cs > > > + } > > + > > /* Drivers may modify this initial i/o setup, but will > > * normally rely on the device being setup. Devices > > * using SPI_CS_HIGH can't coexist well otherwise... > > @@ -561,6 +571,43 @@ struct spi_master *spi_alloc_master(struct device *dev, unsigned size) > > } > > EXPORT_SYMBOL_GPL(spi_alloc_master); > > > > +#ifdef CONFIG_OF > > +static int of_spi_register_master(struct spi_master *master) > > +{ > > + int nb, i; > > + int *cs; > > + struct device_node *np = master->dev.of_node; > > + > > + if (!np) > > + return 0; > > + > > + nb = of_gpio_named_count(np, "cs-gpios"); > > + > > + if (nb < 1) > > + return 0; > > + > > + cs = devm_kzalloc(&master->dev, sizeof(int) * nb, GFP_KERNEL); > > + master->cs_gpios = cs; > > + > > + if (!master->cs_gpios) > > + return -ENOMEM; > > + > > + master->first_gpio_cs = master->num_chipselect; > > + master->num_chipselect += nb; > > + master->num_gpio_cs = nb; > > I think this can be simplified. Rather than trying to optimize by having > a set of integrated cs lines followed by a gpio set, I think it would be > better to have a single cs_gpios table that covers the whole range of > num_chipselect; even if they overlap natural cs lines. That will make for > an easier to use binding since users won't need to remember how many natural > cs lines are on the device when filling the cs-gpios property. It will > also make the code simpler because first_gpio_cs and num_gpio_cs can both > be eliminated. It also means the core can have exact knowledge about which > gpio is associated with which cs number. ok why not so I feel the array with -EINVAL for hw cs; Best Regards, J. ^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2012-03-09 18:02 UTC | newest] Thread overview: 4+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2012-03-07 12:23 [PATCH 1/4 v4] of_spi: add generic binding support to specify cs gpio Jean-Christophe PLAGNIOL-VILLARD [not found] ` <1331122989-14451-1-git-send-email-plagnioj-sclMFOaUSTBWk0Htik3J/w@public.gmane.org> 2012-03-07 12:23 ` [PATCH 2/4 v4] spi/atmel: add DT support Jean-Christophe PLAGNIOL-VILLARD 2012-03-09 15:14 ` [PATCH 1/4 v4] of_spi: add generic binding support to specify cs gpio Grant Likely 2012-03-09 18:02 ` Jean-Christophe PLAGNIOL-VILLARD
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).