2023-10-04 14:57:48

by Duje Mihanović

[permalink] [raw]
Subject: [PATCH RFC v5 0/6] ARM: pxa: GPIO descriptor conversions

Hello,

Small series to convert some of the board files in the mach-pxa directory
to use the new GPIO descriptor interface.

Most notably, the am200epd, am300epd and Spitz matrix keypad among
others are not converted in this series.

Signed-off-by: Duje Mihanović <[email protected]>
---
Changes in v5:
- Address maintainer comments:
- Rename "reset generator" GPIO to "reset"
- Rename ads7846_wait_for_sync() to ads7846_wait_for_sync_gpio()
- Properly bail out when requesting USB host GPIO fails
- Use dev_err_probe() when requesting touchscreen sync GPIO fails
- Use static gpio_desc for gumstix bluetooth reset
- Pulse gumstix bluetooth reset line correctly (assert, then deassert)
- Fix style issue in ads7846_wait_for_sync_gpio()
- Update trailers
- Link to v4: https://lore.kernel.org/r/[email protected]

Changes in v4:
- Address maintainer comments:
- Move wait_for_sync() from spitz.c to driver
- Register LED platform device before getting its gpiod-s
- Add Linus' Reviewed-by
- Link to v3: https://lore.kernel.org/r/[email protected]

Changes in v3:
- Address maintainer comments:
- Use GPIO_LOOKUP_IDX for LEDs
- Drop unnecessary NULL assignments
- Don't give up on *all* SPI devices if hsync cannot be set up
- Add Linus' Acked-by
- Link to v2: https://lore.kernel.org/r/[email protected]

Changes in v2:
- Address maintainer comments:
- Change mentions of function to function()
- Drop cast in OHCI driver dev_warn() call
- Use %pe in OHCI and reset drivers
- Use GPIO _optional() API in OHCI driver
- Drop unnecessary not-null check in OHCI driver
- Use pr_err() instead of printk() in reset driver
- Rebase on v6.6-rc3
- Link to v1: https://lore.kernel.org/r/[email protected]

---
Duje Mihanović (6):
ARM: pxa: Convert Spitz OHCI to GPIO descriptors
ARM: pxa: Convert Spitz LEDs to GPIO descriptors
ARM: pxa: Convert Spitz CF power control to GPIO descriptors
ARM: pxa: Convert reset driver to GPIO descriptors
ARM: pxa: Convert gumstix Bluetooth to GPIO descriptors
input: ads7846: Move wait_for_sync() logic to driver

arch/arm/mach-pxa/gumstix.c | 22 ++++++------
arch/arm/mach-pxa/reset.c | 39 +++++++--------------
arch/arm/mach-pxa/reset.h | 3 +-
arch/arm/mach-pxa/spitz.c | 69 +++++++++++++++++++++++++------------
drivers/input/touchscreen/ads7846.c | 22 ++++++++----
drivers/usb/host/ohci-pxa27x.c | 7 ++++
include/linux/spi/ads7846.h | 1 -
7 files changed, 94 insertions(+), 69 deletions(-)
---
base-commit: 6465e260f48790807eef06b583b38ca9789b6072
change-id: 20230807-pxa-gpio-3ce25d574814

Best regards,
--
Duje Mihanović <[email protected]>



2023-10-04 14:58:05

by Duje Mihanović

[permalink] [raw]
Subject: [PATCH RFC v5 6/6] input: ads7846: Move wait_for_sync() logic to driver

If this code is left in the board file, the sync GPIO would have to be
separated into another lookup table during conversion to the GPIO
descriptor API (which is also done in this patch).

The only user of this code (Sharp Spitz) is also converted in this
patch.

Suggested-by: Linus Walleij <[email protected]>
Reviewed-by: Linus Walleij <[email protected]>
Acked-by: Mark Brown <[email protected]>
Signed-off-by: Duje Mihanović <[email protected]>
---
arch/arm/mach-pxa/spitz.c | 12 ++----------
drivers/input/touchscreen/ads7846.c | 22 +++++++++++++++-------
include/linux/spi/ads7846.h | 1 -
3 files changed, 17 insertions(+), 18 deletions(-)

diff --git a/arch/arm/mach-pxa/spitz.c b/arch/arm/mach-pxa/spitz.c
index 26ec29c9cd1b..e8bc0249b503 100644
--- a/arch/arm/mach-pxa/spitz.c
+++ b/arch/arm/mach-pxa/spitz.c
@@ -520,22 +520,12 @@ static inline void spitz_leds_init(void) {}
* SSP Devices
******************************************************************************/
#if defined(CONFIG_SPI_PXA2XX) || defined(CONFIG_SPI_PXA2XX_MODULE)
-static void spitz_ads7846_wait_for_hsync(void)
-{
- while (gpio_get_value(SPITZ_GPIO_HSYNC))
- cpu_relax();
-
- while (!gpio_get_value(SPITZ_GPIO_HSYNC))
- cpu_relax();
-}
-
static struct ads7846_platform_data spitz_ads7846_info = {
.model = 7846,
.vref_delay_usecs = 100,
.x_plate_ohms = 419,
.y_plate_ohms = 486,
.pressure_max = 1024,
- .wait_for_sync = spitz_ads7846_wait_for_hsync,
};

static struct gpiod_lookup_table spitz_ads7846_gpio_table = {
@@ -543,6 +533,8 @@ static struct gpiod_lookup_table spitz_ads7846_gpio_table = {
.table = {
GPIO_LOOKUP("gpio-pxa", SPITZ_GPIO_TP_INT,
"pendown", GPIO_ACTIVE_LOW),
+ GPIO_LOOKUP("gpio-pxa", SPITZ_GPIO_HSYNC,
+ "sync", GPIO_ACTIVE_LOW),
{ }
},
};
diff --git a/drivers/input/touchscreen/ads7846.c b/drivers/input/touchscreen/ads7846.c
index faea40dd66d0..139b0f3735d0 100644
--- a/drivers/input/touchscreen/ads7846.c
+++ b/drivers/input/touchscreen/ads7846.c
@@ -138,8 +138,7 @@ struct ads7846 {
void *filter_data;
int (*get_pendown_state)(void);
struct gpio_desc *gpio_pendown;
-
- void (*wait_for_sync)(void);
+ struct gpio_desc *sync;
};

enum ads7846_filter {
@@ -636,9 +635,15 @@ static const struct attribute_group ads784x_attr_group = {
};

/*--------------------------------------------------------------------------*/
-
-static void null_wait_for_sync(void)
+static void ads7846_wait_for_sync_gpio(struct ads7846 *ts)
{
+ if (!ts->sync)
+ return;
+ while (!gpiod_get_value(ts->sync))
+ cpu_relax();
+
+ while (gpiod_get_value(ts->sync))
+ cpu_relax();
}

static int ads7846_debounce_filter(void *ads, int data_idx, int *val)
@@ -803,7 +808,7 @@ static void ads7846_read_state(struct ads7846 *ts)
packet->last_cmd_idx = 0;

while (true) {
- ts->wait_for_sync();
+ ads7846_wait_for_sync_gpio(ts);

m = &ts->msg[msg_idx];
error = spi_sync(ts->spi, m);
@@ -1261,8 +1266,6 @@ static int ads7846_probe(struct spi_device *spi)
ts->penirq_recheck_delay_usecs =
pdata->penirq_recheck_delay_usecs;

- ts->wait_for_sync = pdata->wait_for_sync ? : null_wait_for_sync;
-
snprintf(ts->phys, sizeof(ts->phys), "%s/input0", dev_name(dev));
snprintf(ts->name, sizeof(ts->name), "ADS%d Touchscreen", ts->model);

@@ -1361,6 +1364,11 @@ static int ads7846_probe(struct spi_device *spi)
if (err)
return err;

+ ts->sync = devm_gpiod_get_optional(dev, "sync", GPIOD_IN);
+ if (IS_ERR(ts->sync))
+ return dev_err_probe(dev, PTR_ERR(ts->sync),
+ "Failed to get sync GPIO");
+
err = input_register_device(input_dev);
if (err)
return err;
diff --git a/include/linux/spi/ads7846.h b/include/linux/spi/ads7846.h
index a04c1c34c344..fa7c4f119023 100644
--- a/include/linux/spi/ads7846.h
+++ b/include/linux/spi/ads7846.h
@@ -38,7 +38,6 @@ struct ads7846_platform_data {
int gpio_pendown_debounce; /* platform specific debounce time for
* the gpio_pendown */
int (*get_pendown_state)(void);
- void (*wait_for_sync)(void);
bool wakeup;
unsigned long irq_flags;
};

--
2.42.0


2023-10-05 14:32:42

by Andy Shevchenko

[permalink] [raw]
Subject: Re: [PATCH RFC v5 0/6] ARM: pxa: GPIO descriptor conversions

On Wed, Oct 04, 2023 at 04:56:24PM +0200, Duje Mihanović wrote:
> Hello,
>
> Small series to convert some of the board files in the mach-pxa directory
> to use the new GPIO descriptor interface.
>
> Most notably, the am200epd, am300epd and Spitz matrix keypad among
> others are not converted in this series.

Why is it still RFC?
I believe it's already good enough to be considered as a real material.
OTOH "RFT" might make sense. I'm not sure there are any users on the
planet Earth that have this Sharp device up and running with newest
kernels.


--
With Best Regards,
Andy Shevchenko


2023-10-05 16:12:39

by Duje Mihanović

[permalink] [raw]
Subject: Re: [PATCH RFC v5 0/6] ARM: pxa: GPIO descriptor conversions

On Thursday, October 5, 2023 11:14:05 AM CEST Andy Shevchenko wrote:
> On Wed, Oct 04, 2023 at 04:56:24PM +0200, Duje Mihanović wrote:
> > Hello,
> >
> > Small series to convert some of the board files in the mach-pxa directory
> > to use the new GPIO descriptor interface.
> >
> > Most notably, the am200epd, am300epd and Spitz matrix keypad among
> > others are not converted in this series.
>
> Why is it still RFC?
> I believe it's already good enough to be considered as a real material.
> OTOH "RFT" might make sense. I'm not sure there are any users on the
> planet Earth that have this Sharp device up and running with newest
> kernels.

Will add RFT in an eventual v6.

Regards,
Duje




2023-10-06 07:12:29

by Bartosz Golaszewski

[permalink] [raw]
Subject: Re: [PATCH RFC v5 6/6] input: ads7846: Move wait_for_sync() logic to driver

On Wed, Oct 4, 2023 at 4:56 PM Duje Mihanović <[email protected]> wrote:
>
> If this code is left in the board file, the sync GPIO would have to be
> separated into another lookup table during conversion to the GPIO
> descriptor API (which is also done in this patch).
>
> The only user of this code (Sharp Spitz) is also converted in this
> patch.
>
> Suggested-by: Linus Walleij <[email protected]>
> Reviewed-by: Linus Walleij <[email protected]>
> Acked-by: Mark Brown <[email protected]>
> Signed-off-by: Duje Mihanović <[email protected]>
> ---

Reviewed-by: Bartosz Golaszewski <[email protected]>