Skip to contentSkip to conversation

gpio-tqmx86 fixes

PATCH Superseded Matthias Schiffer 4 participants 28 messages May 29, 2024 – May 30, 2024

Subsystems (2)

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore
#0 lore raw

cover gpio-tqmx86 fixes

This is the first series of improvements to the tqmx86 GPIO driver, which fixes some long-standing bugs - in particular, IRQ_TYPE_EDGE_BOTH can never have worked correctly.

Other patches in the series are code cleanup, which is included as it makes the actual fixes much nicer. I have included the same Fixes tag in all commits, as they will need to be cherry-picked together.

A second series with new features (changing GPIO directions, support more GPIOs on SMARC TQMx86 modules) will be submitted when the fixes have been reviewed and merged.

Gregor Herburger (1):
gpio: tqmx86: fix typo in Kconfig label

Matthias Schiffer (7):
gpio: tqmx86: introduce shadow register for GPIO output value
gpio: tqmx86: change tqmx86_gpio_write() order of arguments to match
regmap API
gpio: tqmx86: introduce _tqmx86_gpio_update_bits() helper
gpio: tqmx86: add macros for interrupt configuration
gpio: tqmx86: store IRQ triggers without offsetting index
gpio: tqmx86: store IRQ trigger type and unmask status separately
gpio: tqmx86: fix broken IRQ_TYPE_EDGE_BOTH interrupt type

 drivers/gpio/Kconfig       |   2 +-
 drivers/gpio/gpio-tqmx86.c | 151 ++++++++++++++++++++++++++-----------
 2 files changed, 106 insertions(+), 47 deletions(-)
Signature · 4 lines
TQ-Systems GmbH | Mühlstraße 2, Gut Delling | 82229 Seefeld, Germany
Amtsgericht München, HRB 105018
Geschäftsführer: Detlef Schneider, Rüdiger Stahl, Stefan Schneider
https://www.tq-group.com/
#1 lore raw

PATCH 1/8 gpio: tqmx86: fix typo in Kconfig label

From: Gregor Herburger <[email protected]>

Fix description for GPIO_TQMX86 from QTMX86 to TQMx86.

Fixes b868db94a6a7 ("gpio: tqmx86: Add GPIO from for this IO controller")
---
 drivers/gpio/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpio/Kconfig +1 -1
index 3dbddec070281..1c28a48915bb2 100644
@@ -1576,7 +1576,7 @@ config GPIO_TPS68470
 	  are "output only" GPIOs.
 
 config GPIO_TQMX86
-	tristate "TQ-Systems QTMX86 GPIO"
+	tristate "TQ-Systems TQMx86 GPIO"
 	depends on MFD_TQMX86 || COMPILE_TEST
 	depends on HAS_IOPORT_MAP
 	select GPIOLIB_IRQCHIP
Signature · 4 lines
TQ-Systems GmbH | Mühlstraße 2, Gut Delling | 82229 Seefeld, Germany
Amtsgericht München, HRB 105018
Geschäftsführer: Detlef Schneider, Rüdiger Stahl, Stefan Schneider
https://www.tq-group.com/
#3 lore raw

PATCH 2/8 gpio: tqmx86: introduce shadow register for GPIO output value

In reply to Matthias Schiffer

The TQMx86 GPIO controller uses the same register address for input and output data. Reading the register will always return current inputs rather than the previously set outputs (regardless of the current direction setting). Therefore, using a RMW pattern does not make sense when setting output values. Instead, the previously set output register value needs to be stored as a shadow register.

As there is no reliable way to get the current output values from the hardware, also initialize all channels to 0, to ensure that stored and actual output values match. This should usually not have any effect in practise, as the TQMx86 UEFI sets all outputs to 0 during boot.

Also prepare for extension of the driver to more than 8 GPIOs by using DECLARE_BITMAP.

Fixes b868db94a6a7 ("gpio: tqmx86: Add GPIO from for this IO controller")
---
 drivers/gpio/gpio-tqmx86.c | 18 +++++++++++-------
 1 file changed, 11 insertions(+), 7 deletions(-)
drivers/gpio/gpio-tqmx86.c +11 -7
index 3a28c1f273c39..b7e2dbbdc4ebe 100644
@@ -6,6 +6,7 @@
  *   Vadim V.Vlasov <[email protected]>
  */
 
+#include <linux/bitmap.h>
 #include <linux/bitops.h>
 #include <linux/errno.h>
 #include <linux/gpio/driver.h>
@@ -38,6 +39,7 @@ struct tqmx86_gpio_data {
 	void __iomem		*io_base;
 	int			irq;
 	raw_spinlock_t		spinlock;
+	DECLARE_BITMAP(output, TQMX86_NGPIO);
 	u8			irq_type[TQMX86_NGPI];
 };
 
@@ -64,15 +66,10 @@ static void tqmx86_gpio_set(struct gpio_chip *chip, unsigned int offset,
 {
 	struct tqmx86_gpio_data *gpio = gpiochip_get_data(chip);
 	unsigned long flags;
-	u8 val;
 
 	raw_spin_lock_irqsave(&gpio->spinlock, flags);
-	val = tqmx86_gpio_read(gpio, TQMX86_GPIOD);
-	if (value)
-		val |= BIT(offset);
-	else
-		val &= ~BIT(offset);
-	tqmx86_gpio_write(gpio, val, TQMX86_GPIOD);
+	__assign_bit(offset, gpio->output, value);
+	tqmx86_gpio_write(gpio, bitmap_get_value8(gpio->output, 0), TQMX86_GPIOD);
 	raw_spin_unlock_irqrestore(&gpio->spinlock, flags);
 }
 
@@ -277,6 +274,13 @@ static int tqmx86_gpio_probe(struct platform_device *pdev)
 
 	tqmx86_gpio_write(gpio, (u8)~TQMX86_DIR_INPUT_MASK, TQMX86_GPIODD);
 
+	/*
+	 * Reading the previous output state is not possible with TQMx86 hardware.
+	 * Initialize all outputs to 0 to have a defined state that matches the
+	 * shadow register.
+	 */
+	tqmx86_gpio_write(gpio, 0, TQMX86_GPIOD);
+
 	chip = &gpio->chip;
 	chip->label = "gpio-tqmx86";
 	chip->owner = THIS_MODULE;
Signature · 4 lines
TQ-Systems GmbH | Mühlstraße 2, Gut Delling | 82229 Seefeld, Germany
Amtsgericht München, HRB 105018
Geschäftsführer: Detlef Schneider, Rüdiger Stahl, Stefan Schneider
https://www.tq-group.com/
Reviewed-by on v1 2/8
#4 lore raw

Re: [PATCH 2/8] gpio: tqmx86: introduce shadow register for GPIO output value

Quoted 12 lines · Matthias Schiffer

The TQMx86 GPIO controller uses the same register address for input and output data. Reading the register will always return current inputs rather than the previously set outputs (regardless of the current direction setting). Therefore, using a RMW pattern does not make sense when setting output values. Instead, the previously set output register value needs to be stored as a shadow register.

As there is no reliable way to get the current output values from the hardware, also initialize all channels to 0, to ensure that stored and actual output values match. This should usually not have any effect in practise, as the TQMx86 UEFI sets all outputs to 0 during boot.

Quoted 5 lines

Also prepare for extension of the driver to more than 8 GPIOs by using DECLARE_BITMAP.

Fixes b868db94a6a7 ("gpio: tqmx86: Add GPIO from for this IO controller")

Andrew

#5 lore raw

PATCH 3/8 gpio: tqmx86: change tqmx86_gpio_write() order of arguments to match regmap API

In reply to Matthias Schiffer

Conversion to actually use regmap does not seem useful for this driver, as regmap can't properly represent separate read-only and write-only registers at the same address, but we can at least match the API to make the code clearer.

No functional change intended.

Fixes b868db94a6a7 ("gpio: tqmx86: Add GPIO from for this IO controller")
---
 drivers/gpio/gpio-tqmx86.c | 22 +++++++++++-----------
 1 file changed, 11 insertions(+), 11 deletions(-)
drivers/gpio/gpio-tqmx86.c +11 -11
index b7e2dbbdc4ebe..613ab9ef2e744 100644
@@ -48,8 +48,8 @@ static u8 tqmx86_gpio_read(struct tqmx86_gpio_data *gd, unsigned int reg)
 	return ioread8(gd->io_base + reg);
 }
 
-static void tqmx86_gpio_write(struct tqmx86_gpio_data *gd, u8 val,
-			      unsigned int reg)
+static void tqmx86_gpio_write(struct tqmx86_gpio_data *gd, unsigned int reg,
+			      u8 val)
 {
 	iowrite8(val, gd->io_base + reg);
 }
@@ -69,7 +69,7 @@ static void tqmx86_gpio_set(struct gpio_chip *chip, unsigned int offset,
 
 	raw_spin_lock_irqsave(&gpio->spinlock, flags);
 	__assign_bit(offset, gpio->output, value);
-	tqmx86_gpio_write(gpio, bitmap_get_value8(gpio->output, 0), TQMX86_GPIOD);
+	tqmx86_gpio_write(gpio, TQMX86_GPIOD, bitmap_get_value8(gpio->output, 0));
 	raw_spin_unlock_irqrestore(&gpio->spinlock, flags);
 }
 
@@ -117,7 +117,7 @@ static void tqmx86_gpio_irq_mask(struct irq_data *data)
 	raw_spin_lock_irqsave(&gpio->spinlock, flags);
 	gpiic = tqmx86_gpio_read(gpio, TQMX86_GPIIC);
 	gpiic &= ~mask;
-	tqmx86_gpio_write(gpio, gpiic, TQMX86_GPIIC);
+	tqmx86_gpio_write(gpio, TQMX86_GPIIC, gpiic);
 	raw_spin_unlock_irqrestore(&gpio->spinlock, flags);
 	gpiochip_disable_irq(&gpio->chip, irqd_to_hwirq(data));
 }
@@ -137,7 +137,7 @@ static void tqmx86_gpio_irq_unmask(struct irq_data *data)
 	gpiic = tqmx86_gpio_read(gpio, TQMX86_GPIIC);
 	gpiic &= ~mask;
 	gpiic |= gpio->irq_type[offset] << (offset * TQMX86_GPII_BITS);
-	tqmx86_gpio_write(gpio, gpiic, TQMX86_GPIIC);
+	tqmx86_gpio_write(gpio, TQMX86_GPIIC, gpiic);
 	raw_spin_unlock_irqrestore(&gpio->spinlock, flags);
 }
 
@@ -170,7 +170,7 @@ static int tqmx86_gpio_irq_set_type(struct irq_data *data, unsigned int type)
 	gpiic = tqmx86_gpio_read(gpio, TQMX86_GPIIC);
 	gpiic &= ~((TQMX86_GPII_MASK) << (offset * TQMX86_GPII_BITS));
 	gpiic |= new_type << (offset * TQMX86_GPII_BITS);
-	tqmx86_gpio_write(gpio, gpiic, TQMX86_GPIIC);
+	tqmx86_gpio_write(gpio, TQMX86_GPIIC, gpiic);
 	raw_spin_unlock_irqrestore(&gpio->spinlock, flags);
 
 	return 0;
@@ -188,7 +188,7 @@ static void tqmx86_gpio_irq_handler(struct irq_desc *desc)
 	chained_irq_enter(irq_chip, desc);
 
 	irq_status = tqmx86_gpio_read(gpio, TQMX86_GPIIS);
-	tqmx86_gpio_write(gpio, irq_status, TQMX86_GPIIS);
+	tqmx86_gpio_write(gpio, TQMX86_GPIIS, irq_status);
 
 	irq_bits = irq_status;
 	for_each_set_bit(i, &irq_bits, TQMX86_NGPI)
@@ -272,14 +272,14 @@ static int tqmx86_gpio_probe(struct platform_device *pdev)
 	raw_spin_lock_init(&gpio->spinlock);
 	gpio->io_base = io_base;
 
-	tqmx86_gpio_write(gpio, (u8)~TQMX86_DIR_INPUT_MASK, TQMX86_GPIODD);
+	tqmx86_gpio_write(gpio, TQMX86_GPIODD, (u8)~TQMX86_DIR_INPUT_MASK);
 
 	/*
 	 * Reading the previous output state is not possible with TQMx86 hardware.
 	 * Initialize all outputs to 0 to have a defined state that matches the
 	 * shadow register.
 	 */
-	tqmx86_gpio_write(gpio, 0, TQMX86_GPIOD);
+	tqmx86_gpio_write(gpio, TQMX86_GPIOD, 0);
 
 	chip = &gpio->chip;
 	chip->label = "gpio-tqmx86";
@@ -300,11 +300,11 @@ static int tqmx86_gpio_probe(struct platform_device *pdev)
 		u8 irq_status;
 
 		/* Mask all interrupts */
-		tqmx86_gpio_write(gpio, 0, TQMX86_GPIIC);
+		tqmx86_gpio_write(gpio, TQMX86_GPIIC, 0);
 
 		/* Clear all pending interrupts */
 		irq_status = tqmx86_gpio_read(gpio, TQMX86_GPIIS);
-		tqmx86_gpio_write(gpio, irq_status, TQMX86_GPIIS);
+		tqmx86_gpio_write(gpio, TQMX86_GPIIS, irq_status);
 
 		girq = &chip->irq;
 		gpio_irq_chip_set_chip(girq, &tqmx86_gpio_irq_chip);
Signature · 4 lines
TQ-Systems GmbH | Mühlstraße 2, Gut Delling | 82229 Seefeld, Germany
Amtsgericht München, HRB 105018
Geschäftsführer: Detlef Schneider, Rüdiger Stahl, Stefan Schneider
https://www.tq-group.com/
#6 lore raw

Re: [PATCH 3/8] gpio: tqmx86: change tqmx86_gpio_write() order of arguments to match regmap API

Quoted 9 lines · Matthias Schiffer

Conversion to actually use regmap does not seem useful for this driver, as regmap can't properly represent separate read-only and write-only registers at the same address, but we can at least match the API to make the code clearer.

No functional change intended.

Fixes b868db94a6a7 ("gpio: tqmx86: Add GPIO from for this IO controller")

This is not a fix.

Bart

#7 lore raw

Re: [PATCH 3/8] gpio: tqmx86: change tqmx86_gpio_write() order of arguments to match regmap API

Quoted 8 lines · Bartosz Golaszewski
Quoted 6 lines · Matthias Schiffer

Conversion to actually use regmap does not seem useful for this driver, as regmap can't properly represent separate read-only and write-only registers at the same address, but we can at least match the API to make the code clearer.

Quoted 5 lines
Quoted 3 lines

No functional change intended.

Fixes b868db94a6a7 ("gpio: tqmx86: Add GPIO from for this IO controller")

This is not a fix.

Agreed.

I'm somewhat conflicted by this patch. It is a step towards using regmap, but then says regmap does not make sense. So why make that step?

Changing the order of parameters like this seems like it is will make back porting bug fixes harder, unless all supported versions are changed, which is why fixes make sense. Does the compiler at least issue a warning if the parameters are used the wrong way around?

Overall, i'm leaning towards just dropping it.

Andrew

#8 lore raw

Re: [PATCH 3/8] gpio: tqmx86: change tqmx86_gpio_write() order of arguments to match regmap API

Quoted 22 lines · Andrew Lunn
Quoted 8 lines · Bartosz Golaszewski
Quoted 6 lines · Matthias Schiffer

Conversion to actually use regmap does not seem useful for this driver, as regmap can't properly represent separate read-only and write-only registers at the same address, but we can at least match the API to make the code clearer.

Quoted 5 lines
Quoted 3 lines

No functional change intended.

Fixes b868db94a6a7 ("gpio: tqmx86: Add GPIO from for this IO controller")

This is not a fix.

Agreed.

I'm somewhat conflicted by this patch. It is a step towards using regmap, but then says regmap does not make sense. So why make that step?

Quoted 8 lines

Changing the order of parameters like this seems like it is will make back porting bug fixes harder, unless all supported versions are changed, which is why fixes make sense. Does the compiler at least issue a warning if the parameters are used the wrong way around?

Overall, i'm leaning towards just dropping it.

Andrew

Okay, will drop this patch.

Matthias

#9 lore raw

PATCH 4/8 gpio: tqmx86: introduce _tqmx86_gpio_update_bits() helper

In reply to Matthias Schiffer

Simplify a lot of code in the driver by introducing helpers for the common RMW pattern. No tqmx86_gpio_update_bits() function with builtin locking is added, as it would become redundant with the following fixes, which further consolidate interrupt configuration register setup.

No functional change intended.

Fixes b868db94a6a7 ("gpio: tqmx86: Add GPIO from for this IO controller")
---
 drivers/gpio/gpio-tqmx86.c | 40 ++++++++++++++++++++++----------------
 1 file changed, 23 insertions(+), 17 deletions(-)
drivers/gpio/gpio-tqmx86.c +23 -17
index 613ab9ef2e744..7a851e1730dd1 100644
@@ -54,6 +54,17 @@ static void tqmx86_gpio_write(struct tqmx86_gpio_data *gd, unsigned int reg,
 	iowrite8(val, gd->io_base + reg);
 }
 
+static void _tqmx86_gpio_update_bits(struct tqmx86_gpio_data *gd,
+				     unsigned int reg, u8 mask, u8 val)
+{
+	u8 tmp = tqmx86_gpio_read(gd, reg);
+
+	tmp &= ~mask;
+	tmp |= val & mask;
+
+	tqmx86_gpio_write(gd, reg, tmp);
+}
+
 static int tqmx86_gpio_get(struct gpio_chip *chip, unsigned int offset)
 {
 	struct tqmx86_gpio_data *gpio = gpiochip_get_data(chip);
@@ -110,15 +121,13 @@ static void tqmx86_gpio_irq_mask(struct irq_data *data)
 	struct tqmx86_gpio_data *gpio = gpiochip_get_data(
 		irq_data_get_irq_chip_data(data));
 	unsigned long flags;
-	u8 gpiic, mask;
+	u8 mask;
 
 	mask = TQMX86_GPII_MASK << (offset * TQMX86_GPII_BITS);
-
 	raw_spin_lock_irqsave(&gpio->spinlock, flags);
-	gpiic = tqmx86_gpio_read(gpio, TQMX86_GPIIC);
-	gpiic &= ~mask;
-	tqmx86_gpio_write(gpio, TQMX86_GPIIC, gpiic);
+	_tqmx86_gpio_update_bits(gpio, TQMX86_GPIIC, mask, 0);
 	raw_spin_unlock_irqrestore(&gpio->spinlock, flags);
+
 	gpiochip_disable_irq(&gpio->chip, irqd_to_hwirq(data));
 }
 
@@ -128,16 +137,14 @@ static void tqmx86_gpio_irq_unmask(struct irq_data *data)
 	struct tqmx86_gpio_data *gpio = gpiochip_get_data(
 		irq_data_get_irq_chip_data(data));
 	unsigned long flags;
-	u8 gpiic, mask;
-
-	mask = TQMX86_GPII_MASK << (offset * TQMX86_GPII_BITS);
+	u8 mask, val;
 
 	gpiochip_enable_irq(&gpio->chip, irqd_to_hwirq(data));
+
+	mask = TQMX86_GPII_MASK << (offset * TQMX86_GPII_BITS);
+	val = gpio->irq_type[offset] << (offset * TQMX86_GPII_BITS);
 	raw_spin_lock_irqsave(&gpio->spinlock, flags);
-	gpiic = tqmx86_gpio_read(gpio, TQMX86_GPIIC);
-	gpiic &= ~mask;
-	gpiic |= gpio->irq_type[offset] << (offset * TQMX86_GPII_BITS);
-	tqmx86_gpio_write(gpio, TQMX86_GPIIC, gpiic);
+	_tqmx86_gpio_update_bits(gpio, TQMX86_GPIIC, mask, val);
 	raw_spin_unlock_irqrestore(&gpio->spinlock, flags);
 }
 
@@ -148,7 +155,7 @@ static int tqmx86_gpio_irq_set_type(struct irq_data *data, unsigned int type)
 	unsigned int offset = (data->hwirq - TQMX86_NGPO);
 	unsigned int edge_type = type & IRQF_TRIGGER_MASK;
 	unsigned long flags;
-	u8 new_type, gpiic;
+	u8 new_type, mask, val;
 
 	switch (edge_type) {
 	case IRQ_TYPE_EDGE_RISING:
@@ -166,11 +173,10 @@ static int tqmx86_gpio_irq_set_type(struct irq_data *data, unsigned int type)
 
 	gpio->irq_type[offset] = new_type;
 
+	mask = TQMX86_GPII_MASK << (offset * TQMX86_GPII_BITS);
+	val = new_type << (offset * TQMX86_GPII_BITS);
 	raw_spin_lock_irqsave(&gpio->spinlock, flags);
-	gpiic = tqmx86_gpio_read(gpio, TQMX86_GPIIC);
-	gpiic &= ~((TQMX86_GPII_MASK) << (offset * TQMX86_GPII_BITS));
-	gpiic |= new_type << (offset * TQMX86_GPII_BITS);
-	tqmx86_gpio_write(gpio, TQMX86_GPIIC, gpiic);
+	_tqmx86_gpio_update_bits(gpio, TQMX86_GPIIC, mask, val);
 	raw_spin_unlock_irqrestore(&gpio->spinlock, flags);
 
 	return 0;
Signature · 4 lines
TQ-Systems GmbH | Mühlstraße 2, Gut Delling | 82229 Seefeld, Germany
Amtsgericht München, HRB 105018
Geschäftsführer: Detlef Schneider, Rüdiger Stahl, Stefan Schneider
https://www.tq-group.com/
#10 lore raw

Re: [PATCH 4/8] gpio: tqmx86: introduce _tqmx86_gpio_update_bits() helper

Quoted 18 lines · Matthias Schiffer

Simplify a lot of code in the driver by introducing helpers for the common RMW pattern. No tqmx86_gpio_update_bits() function with builtin locking is added, as it would become redundant with the following fixes, which further consolidate interrupt configuration register setup.

No functional change intended.

Fixes b868db94a6a7 ("gpio: tqmx86: Add GPIO from for this IO controller")
---
 drivers/gpio/gpio-tqmx86.c | 40 ++++++++++++++++++++++----------------
 1 file changed, 23 insertions(+), 17 deletions(-)
drivers/gpio/gpio-tqmx86.c +0 -0
index 613ab9ef2e744..7a851e1730dd1 100644

drivers/gpio/gpio-tqmx86.c static void tqmx86_gpio_write(struct tqmx86_gpio_data *gd, unsigned int reg, · line 58

Quoted 5 lines
drivers/gpio/gpio-tqmx86.c +2 -0
@@ -54,6 +54,17 @@ static void tqmx86_gpio_write(struct tqmx86_gpio_data *gd, unsigned int reg,
 	iowrite8(val, gd->io_base + reg);
 }
 
+static void _tqmx86_gpio_update_bits(struct tqmx86_gpio_data *gd,
+				     unsigned int reg, u8 mask, u8 val)

Why the _ prefix? This is a local function, it is static, so you don't have name space issues. Functions starting with _ are those you should not call without a good reason, there is generally a version without the _ prefix which is the real function to use. So i would drop the _.

This is also not a fix. Please read:

https://www.kernel.org/doc/html/latest/process/stable-kernel-rules.html

and stick to the rules described there.

I don't know how the GPIO tree works, but for netdev, about a week after fixes are merged, they appear in net-next. So you can then build on top of them for development work.

Andrew

#11 lore raw

Re: [PATCH 4/8] gpio: tqmx86: introduce _tqmx86_gpio_update_bits() helper

Quoted 23 lines · Andrew Lunn
Quoted 21 lines · Matthias Schiffer

Simplify a lot of code in the driver by introducing helpers for the common RMW pattern. No tqmx86_gpio_update_bits() function with builtin locking is added, as it would become redundant with the following fixes, which further consolidate interrupt configuration register setup.

No functional change intended.

Fixes b868db94a6a7 ("gpio: tqmx86: Add GPIO from for this IO controller")
---
 drivers/gpio/gpio-tqmx86.c | 40 ++++++++++++++++++++++----------------
 1 file changed, 23 insertions(+), 17 deletions(-)
drivers/gpio/gpio-tqmx86.c +0 -0
index 613ab9ef2e744..7a851e1730dd1 100644
@@ -54,6 +54,17 @@ static void tqmx86_gpio_write(struct tqmx86_gpio_data *gd, unsigned int reg,
 	iowrite8(val, gd->io_base + reg);
 }
 
Quoted 7 lines

drivers/gpio/gpio-tqmx86.c static void tqmx86_gpio_write(struct tqmx86_gpio_data *gd, unsigned int reg, · line 58

Quoted 2 lines
drivers/gpio/gpio-tqmx86.c +2 -0
+static void _tqmx86_gpio_update_bits(struct tqmx86_gpio_data *gd,
+				     unsigned int reg, u8 mask, u8 val)

Why the _ prefix? This is a local function, it is static, so you don't have name space issues. Functions starting with _ are those you should not call without a good reason, there is generally a version without the _ prefix which is the real function to use. So i would drop the _.

My intention was to mark functions that need to be called while holding the spinlock with a _ prefix. Should I just remove the prefix and add a comment instead?

Matthias

Quoted 12 lines

This is also not a fix. Please read:

https://www.kernel.org/doc/html/latest/process/stable-kernel-rules.html

and stick to the rules described there.

I don't know how the GPIO tree works, but for netdev, about a week after fixes are merged, they appear in net-next. So you can then build on top of them for development work.

Andrew

Signature · 4 lines
TQ-Systems GmbH | Mühlstraße 2, Gut Delling | 82229 Seefeld, Germany
Amtsgericht München, HRB 105018
Geschäftsführer: Detlef Schneider, Rüdiger Stahl, Stefan Schneider
https://www.tq-group.com/
#12 lore raw

Re: [PATCH 4/8] gpio: tqmx86: introduce _tqmx86_gpio_update_bits() helper

Quoted 2 lines

My intention was to mark functions that need to be called while holding the spinlock with a _ prefix. Should I just remove the prefix and add a comment instead?

Yes.

You could also add sparse markup of the locks, or add an assert_spin_locked(lock);

Andrew

#13 lore raw

PATCH 5/8 gpio: tqmx86: add macros for interrupt configuration

In reply to Matthias Schiffer

The new macros introduce a distinction between flags used by the driver to represent interrupt configuration (TQMX86_INT_ prefix) and the flags actually written to the hardware (TQMX86_GPII_ prefix). The TQMX86_INT_TRIG_ values are chosen such that they can be converted to register values by a simple shift (in the TQMX86_GPII_CONFIG() macro), at least for the NONE/FALLING/RISING triggers.

No functional change intended.

Fixes b868db94a6a7 ("gpio: tqmx86: Add GPIO from for this IO controller")
---
 drivers/gpio/gpio-tqmx86.c | 28 ++++++++++++++++------------
 1 file changed, 16 insertions(+), 12 deletions(-)
drivers/gpio/gpio-tqmx86.c +16 -12
index 7a851e1730dd1..d6e77f604f4df 100644
@@ -29,10 +29,14 @@
 #define TQMX86_GPIIC	3	/* GPI Interrupt Configuration Register */
 #define TQMX86_GPIIS	4	/* GPI Interrupt Status Register */
 
-#define TQMX86_GPII_FALLING	BIT(0)
-#define TQMX86_GPII_RISING	BIT(1)
-#define TQMX86_GPII_MASK	(BIT(0) | BIT(1))
-#define TQMX86_GPII_BITS	2
+#define TQMX86_INT_TRIG_NONE	0x0
+#define TQMX86_INT_TRIG_FALLING	0x1
+#define TQMX86_INT_TRIG_RISING	0x2
+#define TQMX86_INT_TRIG_BOTH	0x3
+#define TQMX86_INT_TRIG_MASK	0x3
+
+#define TQMX86_GPII_CONFIG(i, v)	((v) << (2 * (i)))
+#define TQMX86_GPII_MASK(i)		TQMX86_GPII_CONFIG(i, TQMX86_INT_TRIG_MASK)
 
 struct tqmx86_gpio_data {
 	struct gpio_chip	chip;
@@ -123,7 +127,7 @@ static void tqmx86_gpio_irq_mask(struct irq_data *data)
 	unsigned long flags;
 	u8 mask;
 
-	mask = TQMX86_GPII_MASK << (offset * TQMX86_GPII_BITS);
+	mask = TQMX86_GPII_MASK(offset);
 	raw_spin_lock_irqsave(&gpio->spinlock, flags);
 	_tqmx86_gpio_update_bits(gpio, TQMX86_GPIIC, mask, 0);
 	raw_spin_unlock_irqrestore(&gpio->spinlock, flags);
@@ -141,8 +145,8 @@ static void tqmx86_gpio_irq_unmask(struct irq_data *data)
 
 	gpiochip_enable_irq(&gpio->chip, irqd_to_hwirq(data));
 
-	mask = TQMX86_GPII_MASK << (offset * TQMX86_GPII_BITS);
-	val = gpio->irq_type[offset] << (offset * TQMX86_GPII_BITS);
+	mask = TQMX86_GPII_MASK(offset);
+	val = TQMX86_GPII_CONFIG(offset, gpio->irq_type[offset]);
 	raw_spin_lock_irqsave(&gpio->spinlock, flags);
 	_tqmx86_gpio_update_bits(gpio, TQMX86_GPIIC, mask, val);
 	raw_spin_unlock_irqrestore(&gpio->spinlock, flags);
@@ -159,13 +163,13 @@ static int tqmx86_gpio_irq_set_type(struct irq_data *data, unsigned int type)
 
 	switch (edge_type) {
 	case IRQ_TYPE_EDGE_RISING:
-		new_type = TQMX86_GPII_RISING;
+		new_type = TQMX86_INT_TRIG_RISING;
 		break;
 	case IRQ_TYPE_EDGE_FALLING:
-		new_type = TQMX86_GPII_FALLING;
+		new_type = TQMX86_INT_TRIG_FALLING;
 		break;
 	case IRQ_TYPE_EDGE_BOTH:
-		new_type = TQMX86_GPII_FALLING | TQMX86_GPII_RISING;
+		new_type = TQMX86_INT_TRIG_BOTH;
 		break;
 	default:
 		return -EINVAL; /* not supported */
@@ -173,8 +177,8 @@ static int tqmx86_gpio_irq_set_type(struct irq_data *data, unsigned int type)
 
 	gpio->irq_type[offset] = new_type;
 
-	mask = TQMX86_GPII_MASK << (offset * TQMX86_GPII_BITS);
-	val = new_type << (offset * TQMX86_GPII_BITS);
+	mask = TQMX86_GPII_MASK(offset);
+	val = TQMX86_GPII_CONFIG(offset, new_type);
 	raw_spin_lock_irqsave(&gpio->spinlock, flags);
 	_tqmx86_gpio_update_bits(gpio, TQMX86_GPIIC, mask, val);
 	raw_spin_unlock_irqrestore(&gpio->spinlock, flags);
Signature · 4 lines
TQ-Systems GmbH | Mühlstraße 2, Gut Delling | 82229 Seefeld, Germany
Amtsgericht München, HRB 105018
Geschäftsführer: Detlef Schneider, Rüdiger Stahl, Stefan Schneider
https://www.tq-group.com/
#14 lore raw

PATCH 6/8 gpio: tqmx86: store IRQ triggers without offsetting index

In reply to Matthias Schiffer

This will allow us to move all offset handling into a single place in the following commits. The additional irq_type indices remain unused, but the tqmx86_gpio_data size increase is insignificant.

No functional change intended.

Fixes b868db94a6a7 ("gpio: tqmx86: Add GPIO from for this IO controller")
---
 drivers/gpio/gpio-tqmx86.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/gpio/gpio-tqmx86.c +3 -3
index d6e77f604f4df..4b37cc3bdd455 100644
@@ -44,7 +44,7 @@ struct tqmx86_gpio_data {
 	int			irq;
 	raw_spinlock_t		spinlock;
 	DECLARE_BITMAP(output, TQMX86_NGPIO);
-	u8			irq_type[TQMX86_NGPI];
+	u8			irq_type[TQMX86_NGPIO];
 };
 
 static u8 tqmx86_gpio_read(struct tqmx86_gpio_data *gd, unsigned int reg)
@@ -146,7 +146,7 @@ static void tqmx86_gpio_irq_unmask(struct irq_data *data)
 	gpiochip_enable_irq(&gpio->chip, irqd_to_hwirq(data));
 
 	mask = TQMX86_GPII_MASK(offset);
-	val = TQMX86_GPII_CONFIG(offset, gpio->irq_type[offset]);
+	val = TQMX86_GPII_CONFIG(offset, gpio->irq_type[data->hwirq]);
 	raw_spin_lock_irqsave(&gpio->spinlock, flags);
 	_tqmx86_gpio_update_bits(gpio, TQMX86_GPIIC, mask, val);
 	raw_spin_unlock_irqrestore(&gpio->spinlock, flags);
@@ -175,7 +175,7 @@ static int tqmx86_gpio_irq_set_type(struct irq_data *data, unsigned int type)
 		return -EINVAL; /* not supported */
 	}
 
-	gpio->irq_type[offset] = new_type;
+	gpio->irq_type[data->hwirq] = new_type;
 
 	mask = TQMX86_GPII_MASK(offset);
 	val = TQMX86_GPII_CONFIG(offset, new_type);
Signature · 4 lines
TQ-Systems GmbH | Mühlstraße 2, Gut Delling | 82229 Seefeld, Germany
Amtsgericht München, HRB 105018
Geschäftsführer: Detlef Schneider, Rüdiger Stahl, Stefan Schneider
https://www.tq-group.com/
#15 lore raw

PATCH 7/8 gpio: tqmx86: store IRQ trigger type and unmask status separately

In reply to Matthias Schiffer

irq_set_type() should not implicitly unmask the IRQ.

All accesses to the interrupt configuration register are moved to a new helper _tqmx86_gpio_irq_config(). We also introduce the new rule that accessing irq_type must happen while locked, which will become significant for fixing EDGE_BOTH handling.

Fixes b868db94a6a7 ("gpio: tqmx86: Add GPIO from for this IO controller")
---
 drivers/gpio/gpio-tqmx86.c | 41 +++++++++++++++++++++++---------------
 1 file changed, 25 insertions(+), 16 deletions(-)
drivers/gpio/gpio-tqmx86.c +25 -16
index 4b37cc3bdd455..c957be3341774 100644
@@ -34,6 +34,7 @@
 #define TQMX86_INT_TRIG_RISING	0x2
 #define TQMX86_INT_TRIG_BOTH	0x3
 #define TQMX86_INT_TRIG_MASK	0x3
+#define TQMX86_INT_UNMASKED	BIT(2)
 
 #define TQMX86_GPII_CONFIG(i, v)	((v) << (2 * (i)))
 #define TQMX86_GPII_MASK(i)		TQMX86_GPII_CONFIG(i, TQMX86_INT_TRIG_MASK)
@@ -42,6 +43,7 @@ struct tqmx86_gpio_data {
 	struct gpio_chip	chip;
 	void __iomem		*io_base;
 	int			irq;
+	/* Lock must be held for accessing output and irq_type fields */
 	raw_spinlock_t		spinlock;
 	DECLARE_BITMAP(output, TQMX86_NGPIO);
 	u8			irq_type[TQMX86_NGPIO];
@@ -119,17 +121,28 @@ static int tqmx86_gpio_get_direction(struct gpio_chip *chip,
 	return GPIO_LINE_DIRECTION_OUT;
 }
 
+static void _tqmx86_gpio_irq_config(struct tqmx86_gpio_data *gpio, int hwirq)
+{
+	unsigned int offset = hwirq - TQMX86_NGPO;
+	u8 type = TQMX86_INT_TRIG_NONE, mask, val;
+
+	if (gpio->irq_type[hwirq] & TQMX86_INT_UNMASKED)
+		type = gpio->irq_type[hwirq] & TQMX86_INT_TRIG_MASK;
+
+	mask = TQMX86_GPII_MASK(offset);
+	val = TQMX86_GPII_CONFIG(offset, type);
+	_tqmx86_gpio_update_bits(gpio, TQMX86_GPIIC, mask, val);
+}
+
 static void tqmx86_gpio_irq_mask(struct irq_data *data)
 {
-	unsigned int offset = (data->hwirq - TQMX86_NGPO);
 	struct tqmx86_gpio_data *gpio = gpiochip_get_data(
 		irq_data_get_irq_chip_data(data));
 	unsigned long flags;
-	u8 mask;
 
-	mask = TQMX86_GPII_MASK(offset);
 	raw_spin_lock_irqsave(&gpio->spinlock, flags);
-	_tqmx86_gpio_update_bits(gpio, TQMX86_GPIIC, mask, 0);
+	gpio->irq_type[data->hwirq] &= ~TQMX86_INT_UNMASKED;
+	_tqmx86_gpio_irq_config(gpio, data->hwirq);
 	raw_spin_unlock_irqrestore(&gpio->spinlock, flags);
 
 	gpiochip_disable_irq(&gpio->chip, irqd_to_hwirq(data));
@@ -137,18 +150,15 @@ static void tqmx86_gpio_irq_mask(struct irq_data *data)
 
 static void tqmx86_gpio_irq_unmask(struct irq_data *data)
 {
-	unsigned int offset = (data->hwirq - TQMX86_NGPO);
 	struct tqmx86_gpio_data *gpio = gpiochip_get_data(
 		irq_data_get_irq_chip_data(data));
 	unsigned long flags;
-	u8 mask, val;
 
 	gpiochip_enable_irq(&gpio->chip, irqd_to_hwirq(data));
 
-	mask = TQMX86_GPII_MASK(offset);
-	val = TQMX86_GPII_CONFIG(offset, gpio->irq_type[data->hwirq]);
 	raw_spin_lock_irqsave(&gpio->spinlock, flags);
-	_tqmx86_gpio_update_bits(gpio, TQMX86_GPIIC, mask, val);
+	gpio->irq_type[data->hwirq] |= TQMX86_INT_UNMASKED;
+	_tqmx86_gpio_irq_config(gpio, data->hwirq);
 	raw_spin_unlock_irqrestore(&gpio->spinlock, flags);
 }
 
@@ -156,10 +166,9 @@ static int tqmx86_gpio_irq_set_type(struct irq_data *data, unsigned int type)
 {
 	struct tqmx86_gpio_data *gpio = gpiochip_get_data(
 		irq_data_get_irq_chip_data(data));
-	unsigned int offset = (data->hwirq - TQMX86_NGPO);
 	unsigned int edge_type = type & IRQF_TRIGGER_MASK;
 	unsigned long flags;
-	u8 new_type, mask, val;
+	u8 new_type;
 
 	switch (edge_type) {
 	case IRQ_TYPE_EDGE_RISING:
@@ -175,12 +184,12 @@ static int tqmx86_gpio_irq_set_type(struct irq_data *data, unsigned int type)
 		return -EINVAL; /* not supported */
 	}
 
-	gpio->irq_type[data->hwirq] = new_type;
-
-	mask = TQMX86_GPII_MASK(offset);
-	val = TQMX86_GPII_CONFIG(offset, new_type);
 	raw_spin_lock_irqsave(&gpio->spinlock, flags);
-	_tqmx86_gpio_update_bits(gpio, TQMX86_GPIIC, mask, val);
+
+	gpio->irq_type[data->hwirq] &= ~TQMX86_INT_TRIG_MASK;
+	gpio->irq_type[data->hwirq] |= new_type;
+	_tqmx86_gpio_irq_config(gpio, data->hwirq);
+
 	raw_spin_unlock_irqrestore(&gpio->spinlock, flags);
 
 	return 0;
Signature · 4 lines
TQ-Systems GmbH | Mühlstraße 2, Gut Delling | 82229 Seefeld, Germany
Amtsgericht München, HRB 105018
Geschäftsführer: Detlef Schneider, Rüdiger Stahl, Stefan Schneider
https://www.tq-group.com/
#16 lore raw

PATCH 8/8 gpio: tqmx86: fix broken IRQ_TYPE_EDGE_BOTH interrupt type

In reply to Matthias Schiffer

The TQMx86 GPIO controller only supports falling and rising edge triggers, but not both. Fix this by implementing a software both-edge mode that toggles the edge type after every interrupt.

Fixes b868db94a6a7 ("gpio: tqmx86: Add GPIO from for this IO controller")
---
 drivers/gpio/gpio-tqmx86.c | 42 +++++++++++++++++++++++++++++++++++---
 1 file changed, 39 insertions(+), 3 deletions(-)
drivers/gpio/gpio-tqmx86.c +39 -3
index c957be3341774..400415676ad5d 100644
@@ -126,9 +126,15 @@ static void _tqmx86_gpio_irq_config(struct tqmx86_gpio_data *gpio, int hwirq)
 	unsigned int offset = hwirq - TQMX86_NGPO;
 	u8 type = TQMX86_INT_TRIG_NONE, mask, val;
 
-	if (gpio->irq_type[hwirq] & TQMX86_INT_UNMASKED)
+	if (gpio->irq_type[hwirq] & TQMX86_INT_UNMASKED) {
 		type = gpio->irq_type[hwirq] & TQMX86_INT_TRIG_MASK;
 
+		if (type == TQMX86_INT_TRIG_BOTH)
+			type = tqmx86_gpio_get(&gpio->chip, hwirq)
+				? TQMX86_INT_TRIG_FALLING
+				: TQMX86_INT_TRIG_RISING;
+	}
+
 	mask = TQMX86_GPII_MASK(offset);
 	val = TQMX86_GPII_CONFIG(offset, type);
 	_tqmx86_gpio_update_bits(gpio, TQMX86_GPIIC, mask, val);
@@ -200,8 +206,8 @@ static void tqmx86_gpio_irq_handler(struct irq_desc *desc)
 	struct gpio_chip *chip = irq_desc_get_handler_data(desc);
 	struct tqmx86_gpio_data *gpio = gpiochip_get_data(chip);
 	struct irq_chip *irq_chip = irq_desc_get_chip(desc);
-	unsigned long irq_bits;
-	int i = 0;
+	unsigned long irq_bits, flags;
+	int i, hwirq;
 	u8 irq_status;
 
 	chained_irq_enter(irq_chip, desc);
@@ -210,6 +216,36 @@ static void tqmx86_gpio_irq_handler(struct irq_desc *desc)
 	tqmx86_gpio_write(gpio, TQMX86_GPIIS, irq_status);
 
 	irq_bits = irq_status;
+
+	raw_spin_lock_irqsave(&gpio->spinlock, flags);
+	for_each_set_bit(i, &irq_bits, TQMX86_NGPI) {
+		hwirq = i + TQMX86_NGPO;
+
+		/*
+		 * Edge-both triggers are implemented by flipping the edge
+		 * trigger after each interrupt, as the controller only supports
+		 * either rising or falling edge triggers, but not both.
+		 *
+		 * Internally, the TQMx86 GPIO controller has separate status
+		 * registers for rising and falling edge interrupts. GPIIC
+		 * configures which bits from which register are visible in the
+		 * interrupt status register GPIIS and defines what triggers the
+		 * parent IRQ line. Writing to GPIIS always clears both rising
+		 * and falling interrupt flags internally, regardless of the
+		 * currently configured trigger.
+		 *
+		 * In consequence, we can cleanly implement the edge-both
+		 * trigger in software by first clearing the interrupt and then
+		 * setting the new trigger based on the current GPIO input in
+		 * _tqmx86_gpio_irq_config() - even if an edge arrives between
+		 * reading the input and setting the trigger, we will have a new
+		 * interrupt pending.
+		 */
+		if ((gpio->irq_type[hwirq] & TQMX86_INT_TRIG_MASK) == TQMX86_INT_TRIG_BOTH)
+			_tqmx86_gpio_irq_config(gpio, hwirq);
+	}
+	raw_spin_unlock_irqrestore(&gpio->spinlock, flags);
+
 	for_each_set_bit(i, &irq_bits, TQMX86_NGPI)
 		generic_handle_domain_irq(gpio->chip.irq.domain,
 					  i + TQMX86_NGPO);
Signature · 4 lines
TQ-Systems GmbH | Mühlstraße 2, Gut Delling | 82229 Seefeld, Germany
Amtsgericht München, HRB 105018
Geschäftsführer: Detlef Schneider, Rüdiger Stahl, Stefan Schneider
https://www.tq-group.com/
#17 lore raw

Re: [PATCH 8/8] gpio: tqmx86: fix broken IRQ_TYPE_EDGE_BOTH interrupt type

Quoted 3 lines · Matthias Schiffer

The TQMx86 GPIO controller only supports falling and rising edge triggers, but not both. Fix this by implementing a software both-edge mode that toggles the edge type after every interrupt.

Do you have a real use case for this, one that will handle lost interrupts because it cannot swap edge quick enough?

I personally would not do this, because it is dangerous, it gives the impression the device can do both, when in fact it cannot reliably.

For me, the correct fix is to return EOPNOTSUPP or EINVAL for BOTH.

Andrew

#18 lore raw

Re: [PATCH 8/8] gpio: tqmx86: fix broken IRQ_TYPE_EDGE_BOTH interrupt type

Quoted 5 lines · Andrew Lunn
Quoted 3 lines · Matthias Schiffer

The TQMx86 GPIO controller only supports falling and rising edge triggers, but not both. Fix this by implementing a software both-edge mode that toggles the edge type after every interrupt.

Quoted 8 lines

Do you have a real use case for this, one that will handle lost interrupts because it cannot swap edge quick enough?

I personally would not do this, because it is dangerous, it gives the impression the device can do both, when in fact it cannot reliably.

For me, the correct fix is to return EOPNOTSUPP or EINVAL for BOTH.

This was the first thing I tried as well, but it seems that supporting IRQ_TYPE_EDGE_BOTH is mandatory for all GPIO drivers (not doing so results in various error messages when attemting to use *any* type of interrupt for the GPIO; I don't remember the exact errors). For this reason, several drivers implement a similar software solution when the hardware doesn't support it.

Many drivers implement this in a fragile way that will easily break when an edge is missed. On the TQMx86 we are lucky, and this software implementation is actually robust and will not stop reporting edges when one is missed. The reason is explained in detail in the long comment added by this patch.

Matthias

#19 lore raw

Re: [PATCH 8/8] gpio: tqmx86: fix broken IRQ_TYPE_EDGE_BOTH interrupt type

In reply to Matthias Schiffer
Quoted 5 lines · Matthias Schiffer
drivers/gpio/gpio-tqmx86.c +0 -0
index c957be3341774..400415676ad5d 100644

drivers/gpio/gpio-tqmx86.c static void _tqmx86_gpio_irq_config(struct tqmx86_gpio_data *gpio, int hwirq) · line 133

Quoted 9 lines
drivers/gpio/gpio-tqmx86.c +3 -1
@@ -126,9 +126,15 @@ static void _tqmx86_gpio_irq_config(struct tqmx86_gpio_data *gpio, int hwirq)
 	unsigned int offset = hwirq - TQMX86_NGPO;
 	u8 type = TQMX86_INT_TRIG_NONE, mask, val;
 
-	if (gpio->irq_type[hwirq] & TQMX86_INT_UNMASKED)
+	if (gpio->irq_type[hwirq] & TQMX86_INT_UNMASKED) {
 		type = gpio->irq_type[hwirq] & TQMX86_INT_TRIG_MASK;
 
+		if (type == TQMX86_INT_TRIG_BOTH)
+			type = tqmx86_gpio_get(&gpio->chip, hwirq)
                                                            ^^^^^

drivers/gpio/gpio-tqmx86.c static void _tqmx86_gpio_irq_config(struct tqmx86_gpio_data *gpio, int hwirq) · line 138

Quoted 5 lines
drivers/gpio/gpio-tqmx86.c +4 -0
+				? TQMX86_INT_TRIG_FALLING
+				: TQMX86_INT_TRIG_RISING;
+	}
+
 	mask = TQMX86_GPII_MASK(offset);
                                ^^^^^^

drivers/gpio/gpio-tqmx86.c static void _tqmx86_gpio_irq_config(struct tqmx86_gpio_data *gpio, int hwirq) · line 139

Quoted 1 line
drivers/gpio/gpio-tqmx86.c +0 -0
 	val = TQMX86_GPII_CONFIG(offset, type);
                                 ^^^^^^

drivers/gpio/gpio-tqmx86.c static void _tqmx86_gpio_irq_config(struct tqmx86_gpio_data *gpio, int hwirq) · line 140

Quoted 1 line
drivers/gpio/gpio-tqmx86.c +0 -0
 	_tqmx86_gpio_update_bits(gpio, TQMX86_GPIIC, mask, val);

The offset stuff wasn't beautiful and I'm glad you are deleting it. My understanding is that a hwirq is 0-3 for output or 4-7 input. An offset is "hwirq % 4"?

There are a bunch of places which are still marked as taking an offset but they all actually take a hwirq. For example, tqmx86_gpio_get() above. The only things which still actually take an offset are the TQMX86_GPII_MASK() and TQMX86_GPII_CONFIG() macros.

Could you:
1) Modify TQMX86_GPII_MASK() and TQMX86_GPII_CONFIG() to take a hwirq?
2) Rename all the "offset" variables to "hwirq"?

regards,
dan carpenter

#20 lore raw

Re: [PATCH 8/8] gpio: tqmx86: fix broken IRQ_TYPE_EDGE_BOTH interrupt type

Quoted 32 lines · Dan Carpenter
Quoted 5 lines · Matthias Schiffer
drivers/gpio/gpio-tqmx86.c +0 -0
index c957be3341774..400415676ad5d 100644

drivers/gpio/gpio-tqmx86.c static void _tqmx86_gpio_irq_config(struct tqmx86_gpio_data *gpio, int hwirq) · line 133

Quoted 9 lines
drivers/gpio/gpio-tqmx86.c +3 -1
@@ -126,9 +126,15 @@ static void _tqmx86_gpio_irq_config(struct tqmx86_gpio_data *gpio, int hwirq)
 	unsigned int offset = hwirq - TQMX86_NGPO;
 	u8 type = TQMX86_INT_TRIG_NONE, mask, val;
 
-	if (gpio->irq_type[hwirq] & TQMX86_INT_UNMASKED)
+	if (gpio->irq_type[hwirq] & TQMX86_INT_UNMASKED) {
 		type = gpio->irq_type[hwirq] & TQMX86_INT_TRIG_MASK;
 
+		if (type == TQMX86_INT_TRIG_BOTH)
+			type = tqmx86_gpio_get(&gpio->chip, hwirq)
                                                            ^^^^^

drivers/gpio/gpio-tqmx86.c static void _tqmx86_gpio_irq_config(struct tqmx86_gpio_data *gpio, int hwirq) · line 138

Quoted 5 lines
drivers/gpio/gpio-tqmx86.c +4 -0
+				? TQMX86_INT_TRIG_FALLING
+				: TQMX86_INT_TRIG_RISING;
+	}
+
 	mask = TQMX86_GPII_MASK(offset);
                                ^^^^^^

drivers/gpio/gpio-tqmx86.c static void _tqmx86_gpio_irq_config(struct tqmx86_gpio_data *gpio, int hwirq) · line 139

Quoted 1 line
drivers/gpio/gpio-tqmx86.c +0 -0
 	val = TQMX86_GPII_CONFIG(offset, type);
                                 ^^^^^^

drivers/gpio/gpio-tqmx86.c static void _tqmx86_gpio_irq_config(struct tqmx86_gpio_data *gpio, int hwirq) · line 140

Quoted 1 line
drivers/gpio/gpio-tqmx86.c +0 -0
 	_tqmx86_gpio_update_bits(gpio, TQMX86_GPIIC, mask, val);

The offset stuff wasn't beautiful and I'm glad you are deleting it. My understanding is that a hwirq is 0-3 for output or 4-7 input. An offset is "hwirq % 4"?

Quoted 8 lines

There are a bunch of places which are still marked as taking an offset but they all actually take a hwirq. For example, tqmx86_gpio_get() above. The only things which still actually take an offset are the TQMX86_GPII_MASK() and TQMX86_GPII_CONFIG() macros.

Could you:
1) Modify TQMX86_GPII_MASK() and TQMX86_GPII_CONFIG() to take a hwirq?
2) Rename all the "offset" variables to "hwirq"?

Unfortunately, the TQMx86 GPIO is a huge mess, and the mapping between GPIO numbers and IRQ numbers depends on the hardware generation/variant. I don't think it is possible to have GPIO numbers and hwirq numbers differ, is it?

Currently, the driver only supports COM Express modules, where IRQs 0-3 correspond to GPIOs 4-7, while GPIOs 0-3 don't have interrupt support. We will soon be mainlining support for our SMARC modules, which have up to 14 GPIOs, and (on some families) IRQ support for all GPIOs (IRQs 0-13 correspond to GPIOs 0-13).

New interrupt config and status registers have been introduced to support more IRQs - up to 4 config registers (2 bits for each IRQ) and 3 status registers (IRQs 0-3 in the first one, 4-11 in the second one, 12-13 in the third one... so this part is a bit more convoluted than just "hwirq % 4")

As the mapping between GPIOs and IRQs will become dynamic with these changes, I'd rather keep TQMX86_GPII_* using IRQ numbers instead of GPIO numbers. We will be introducing helpers for accessing the interrupt registers; the macros deal with individual register bits, and I think they should be agnostic of the mapping to GPIO/hwirq numbers.

Matthias

Quoted 4 lines

regards,
dan carpenter

Signature · 4 lines
TQ-Systems GmbH | Mühlstraße 2, Gut Delling | 82229 Seefeld, Germany
Amtsgericht München, HRB 105018
Geschäftsführer: Detlef Schneider, Rüdiger Stahl, Stefan Schneider
https://www.tq-group.com/
#21 lore raw

Re: [PATCH 8/8] gpio: tqmx86: fix broken IRQ_TYPE_EDGE_BOTH interrupt type

Quoted 42 lines · Matthias Schiffer
Quoted 32 lines · Dan Carpenter
Quoted 5 lines · Matthias Schiffer
drivers/gpio/gpio-tqmx86.c +0 -0
index c957be3341774..400415676ad5d 100644

drivers/gpio/gpio-tqmx86.c static void _tqmx86_gpio_irq_config(struct tqmx86_gpio_data *gpio, int hwirq) · line 133

Quoted 9 lines
drivers/gpio/gpio-tqmx86.c +3 -1
@@ -126,9 +126,15 @@ static void _tqmx86_gpio_irq_config(struct tqmx86_gpio_data *gpio, int hwirq)
 	unsigned int offset = hwirq - TQMX86_NGPO;
 	u8 type = TQMX86_INT_TRIG_NONE, mask, val;
 
-	if (gpio->irq_type[hwirq] & TQMX86_INT_UNMASKED)
+	if (gpio->irq_type[hwirq] & TQMX86_INT_UNMASKED) {
 		type = gpio->irq_type[hwirq] & TQMX86_INT_TRIG_MASK;
 
+		if (type == TQMX86_INT_TRIG_BOTH)
+			type = tqmx86_gpio_get(&gpio->chip, hwirq)
                                                            ^^^^^

drivers/gpio/gpio-tqmx86.c static void _tqmx86_gpio_irq_config(struct tqmx86_gpio_data *gpio, int hwirq) · line 138

Quoted 5 lines
drivers/gpio/gpio-tqmx86.c +4 -0
+				? TQMX86_INT_TRIG_FALLING
+				: TQMX86_INT_TRIG_RISING;
+	}
+
 	mask = TQMX86_GPII_MASK(offset);
                                ^^^^^^

drivers/gpio/gpio-tqmx86.c static void _tqmx86_gpio_irq_config(struct tqmx86_gpio_data *gpio, int hwirq) · line 139

Quoted 1 line
drivers/gpio/gpio-tqmx86.c +0 -0
 	val = TQMX86_GPII_CONFIG(offset, type);
                                 ^^^^^^

drivers/gpio/gpio-tqmx86.c static void _tqmx86_gpio_irq_config(struct tqmx86_gpio_data *gpio, int hwirq) · line 140

Quoted 1 line
drivers/gpio/gpio-tqmx86.c +0 -0
 	_tqmx86_gpio_update_bits(gpio, TQMX86_GPIIC, mask, val);

The offset stuff wasn't beautiful and I'm glad you are deleting it. My understanding is that a hwirq is 0-3 for output or 4-7 input. An offset is "hwirq % 4"?

Quoted 8 lines

There are a bunch of places which are still marked as taking an offset but they all actually take a hwirq. For example, tqmx86_gpio_get() above. The only things which still actually take an offset are the TQMX86_GPII_MASK() and TQMX86_GPII_CONFIG() macros.

Could you:
1) Modify TQMX86_GPII_MASK() and TQMX86_GPII_CONFIG() to take a hwirq?
2) Rename all the "offset" variables to "hwirq"?

Quoted 6 lines

Unfortunately, the TQMx86 GPIO is a huge mess, and the mapping between GPIO numbers and IRQ numbers depends on the hardware generation/variant. I don't think it is possible to have GPIO numbers and hwirq numbers differ, is it?

Currently, the driver only supports COM Express modules, where IRQs 0-3 correspond to GPIOs 4-7, while GPIOs 0-3 don't have interrupt support.

I'm so confused.

So "offset" is the GPIO number and "hwirq" is the IRQ number? If the IRQ numbers are 0-3 then why do we subtract 4 to get the GPIO number in _tqmx86_gpio_irq_config()?

	unsigned int offset = hwirq - TQMX86_NGPO;

And again, it's just weird to call:

		type = tqmx86_gpio_get(&gpio->chip, hwirq);

where we're passing "hwirq" when tqmx86_gpio_get() takes an "offset" as an argument.

regards,
dan carpenter

#22 lore raw

Re: [PATCH 8/8] gpio: tqmx86: fix broken IRQ_TYPE_EDGE_BOTH interrupt type

Quoted 48 lines · Dan Carpenter
Quoted 46 lines · Matthias Schiffer
Quoted 32 lines · Dan Carpenter
Quoted 5 lines · Matthias Schiffer
drivers/gpio/gpio-tqmx86.c +0 -0
index c957be3341774..400415676ad5d 100644

drivers/gpio/gpio-tqmx86.c static void _tqmx86_gpio_irq_config(struct tqmx86_gpio_data *gpio, int hwirq) · line 133

Quoted 9 lines
drivers/gpio/gpio-tqmx86.c +3 -1
@@ -126,9 +126,15 @@ static void _tqmx86_gpio_irq_config(struct tqmx86_gpio_data *gpio, int hwirq)
 	unsigned int offset = hwirq - TQMX86_NGPO;
 	u8 type = TQMX86_INT_TRIG_NONE, mask, val;
 
-	if (gpio->irq_type[hwirq] & TQMX86_INT_UNMASKED)
+	if (gpio->irq_type[hwirq] & TQMX86_INT_UNMASKED) {
 		type = gpio->irq_type[hwirq] & TQMX86_INT_TRIG_MASK;
 
+		if (type == TQMX86_INT_TRIG_BOTH)
+			type = tqmx86_gpio_get(&gpio->chip, hwirq)
                                                            ^^^^^

drivers/gpio/gpio-tqmx86.c static void _tqmx86_gpio_irq_config(struct tqmx86_gpio_data *gpio, int hwirq) · line 138

Quoted 5 lines
drivers/gpio/gpio-tqmx86.c +4 -0
+				? TQMX86_INT_TRIG_FALLING
+				: TQMX86_INT_TRIG_RISING;
+	}
+
 	mask = TQMX86_GPII_MASK(offset);
                                ^^^^^^

drivers/gpio/gpio-tqmx86.c static void _tqmx86_gpio_irq_config(struct tqmx86_gpio_data *gpio, int hwirq) · line 139

Quoted 1 line
drivers/gpio/gpio-tqmx86.c +0 -0
 	val = TQMX86_GPII_CONFIG(offset, type);
                                 ^^^^^^

drivers/gpio/gpio-tqmx86.c static void _tqmx86_gpio_irq_config(struct tqmx86_gpio_data *gpio, int hwirq) · line 140

Quoted 1 line
drivers/gpio/gpio-tqmx86.c +0 -0
 	_tqmx86_gpio_update_bits(gpio, TQMX86_GPIIC, mask, val);

The offset stuff wasn't beautiful and I'm glad you are deleting it. My understanding is that a hwirq is 0-3 for output or 4-7 input. An offset is "hwirq % 4"?

Quoted 8 lines

There are a bunch of places which are still marked as taking an offset but they all actually take a hwirq. For example, tqmx86_gpio_get() above. The only things which still actually take an offset are the TQMX86_GPII_MASK() and TQMX86_GPII_CONFIG() macros.

Could you:
1) Modify TQMX86_GPII_MASK() and TQMX86_GPII_CONFIG() to take a hwirq?
2) Rename all the "offset" variables to "hwirq"?

Unfortunately, the TQMx86 GPIO is a huge mess, and the mapping between GPIO numbers and IRQ numbers depends on the hardware generation/variant. I don't think it is possible to have GPIO numbers and hwirq numbers differ, is it?

Quoted 7 lines
Quoted 2 lines

Currently, the driver only supports COM Express modules, where IRQs 0-3 correspond to GPIOs 4-7, while GPIOs 0-3 don't have interrupt support.

I'm so confused.

So "offset" is the GPIO number and "hwirq" is the IRQ number? If the IRQ numbers are 0-3 then why do we subtract 4 to get the GPIO number in

The current naming in the driver is confusing and I'll fix that in the next round of refactoring patches.

Generally, hwirq == GPIO number (I have not found a way to change this mapping - if there is one, I'd be interested to try if it makes the code less confusing). "offset" currently always refers to some shift in a hardware register. In tqmx86_gpio_get and tqmx86_gpio_set, offset is a GPIO number. In all functions dealing with IRQs, offset is an IRQ number (which is different from the hwirq number).

Matthias

Quoted 14 lines

_tqmx86_gpio_irq_config()?

	unsigned int offset = hwirq - TQMX86_NGPO;

And again, it's just weird to call:

		type = tqmx86_gpio_get(&gpio->chip, hwirq);

where we're passing "hwirq" when tqmx86_gpio_get() takes an "offset" as an argument.

regards,
dan carpenter

Signature · 4 lines
TQ-Systems GmbH | Mühlstraße 2, Gut Delling | 82229 Seefeld, Germany
Amtsgericht München, HRB 105018
Geschäftsführer: Detlef Schneider, Rüdiger Stahl, Stefan Schneider
https://www.tq-group.com/
#23 lore raw

Re: [PATCH 8/8] gpio: tqmx86: fix broken IRQ_TYPE_EDGE_BOTH interrupt type

Quoted 60 lines · Matthias Schiffer
Quoted 48 lines · Dan Carpenter
Quoted 46 lines · Matthias Schiffer
Quoted 32 lines · Dan Carpenter
Quoted 5 lines · Matthias Schiffer
drivers/gpio/gpio-tqmx86.c +0 -0
index c957be3341774..400415676ad5d 100644

drivers/gpio/gpio-tqmx86.c static void _tqmx86_gpio_irq_config(struct tqmx86_gpio_data *gpio, int hwirq) · line 133

Quoted 9 lines
drivers/gpio/gpio-tqmx86.c +3 -1
@@ -126,9 +126,15 @@ static void _tqmx86_gpio_irq_config(struct tqmx86_gpio_data *gpio, int hwirq)
 	unsigned int offset = hwirq - TQMX86_NGPO;
 	u8 type = TQMX86_INT_TRIG_NONE, mask, val;
 
-	if (gpio->irq_type[hwirq] & TQMX86_INT_UNMASKED)
+	if (gpio->irq_type[hwirq] & TQMX86_INT_UNMASKED) {
 		type = gpio->irq_type[hwirq] & TQMX86_INT_TRIG_MASK;
 
+		if (type == TQMX86_INT_TRIG_BOTH)
+			type = tqmx86_gpio_get(&gpio->chip, hwirq)
                                                            ^^^^^

drivers/gpio/gpio-tqmx86.c static void _tqmx86_gpio_irq_config(struct tqmx86_gpio_data *gpio, int hwirq) · line 138

Quoted 5 lines
drivers/gpio/gpio-tqmx86.c +4 -0
+				? TQMX86_INT_TRIG_FALLING
+				: TQMX86_INT_TRIG_RISING;
+	}
+
 	mask = TQMX86_GPII_MASK(offset);
                                ^^^^^^

drivers/gpio/gpio-tqmx86.c static void _tqmx86_gpio_irq_config(struct tqmx86_gpio_data *gpio, int hwirq) · line 139

Quoted 1 line
drivers/gpio/gpio-tqmx86.c +0 -0
 	val = TQMX86_GPII_CONFIG(offset, type);
                                 ^^^^^^

drivers/gpio/gpio-tqmx86.c static void _tqmx86_gpio_irq_config(struct tqmx86_gpio_data *gpio, int hwirq) · line 140

Quoted 1 line
drivers/gpio/gpio-tqmx86.c +0 -0
 	_tqmx86_gpio_update_bits(gpio, TQMX86_GPIIC, mask, val);

The offset stuff wasn't beautiful and I'm glad you are deleting it. My understanding is that a hwirq is 0-3 for output or 4-7 input. An offset is "hwirq % 4"?

Quoted 8 lines

There are a bunch of places which are still marked as taking an offset but they all actually take a hwirq. For example, tqmx86_gpio_get() above. The only things which still actually take an offset are the TQMX86_GPII_MASK() and TQMX86_GPII_CONFIG() macros.

Could you:
1) Modify TQMX86_GPII_MASK() and TQMX86_GPII_CONFIG() to take a hwirq?
2) Rename all the "offset" variables to "hwirq"?

Unfortunately, the TQMx86 GPIO is a huge mess, and the mapping between GPIO numbers and IRQ numbers depends on the hardware generation/variant. I don't think it is possible to have GPIO numbers and hwirq numbers differ, is it?

Quoted 7 lines
Quoted 2 lines

Currently, the driver only supports COM Express modules, where IRQs 0-3 correspond to GPIOs 4-7, while GPIOs 0-3 don't have interrupt support.

I'm so confused.

So "offset" is the GPIO number and "hwirq" is the IRQ number? If the IRQ numbers are 0-3 then why do we subtract 4 to get the GPIO number in

The current naming in the driver is confusing and I'll fix that in the next round of refactoring patches.

Quoted 7 lines

Generally, hwirq == GPIO number (I have not found a way to change this mapping - if there is one, I'd be interested to try if it makes the code less confusing). "offset" currently always refers to some shift in a hardware register. In tqmx86_gpio_get and tqmx86_gpio_set, offset is a GPIO number. In all functions dealing with IRQs, offset is an IRQ number (which is different from the hwirq number).

Matthias

Ah, I just noticed the "to_irq" function in struct gpio_chip, I guess I simply didn't look in the right place before. Will have a closer look when I rebase the refactoring patches.

Matthias

Quoted 17 lines
Quoted 16 lines

_tqmx86_gpio_irq_config()?

	unsigned int offset = hwirq - TQMX86_NGPO;

And again, it's just weird to call:

		type = tqmx86_gpio_get(&gpio->chip, hwirq);

where we're passing "hwirq" when tqmx86_gpio_get() takes an "offset" as an argument.

regards,
dan carpenter

Signature · 4 lines
TQ-Systems GmbH | Mühlstraße 2, Gut Delling | 82229 Seefeld, Germany
Amtsgericht München, HRB 105018
Geschäftsführer: Detlef Schneider, Rüdiger Stahl, Stefan Schneider
https://www.tq-group.com/
#24 lore raw

Re: [PATCH 8/8] gpio: tqmx86: fix broken IRQ_TYPE_EDGE_BOTH interrupt type

In reply to Matthias Schiffer
Quoted 8 lines

Currently, the driver only supports COM Express modules, where IRQs 0-3 correspond to GPIOs 4-7, while GPIOs 0-3 don't have interrupt support. We will soon be mainlining support for our SMARC modules, which have up to 14 GPIOs, and (on some families) IRQ support for all GPIOs (IRQs 0-13 correspond to GPIOs 0-13).

New interrupt config and status registers have been introduced to support more IRQs - up to 4 config registers (2 bits for each IRQ) and 3 status registers (IRQs 0-3 in the first one, 4-11 in the second one, 12-13 in the third one... so this part is a bit more convoluted than just "hwirq % 4")

Depending on how different it is, you could consider a second driver, rather than make this driver more complex.

Andrew

#25 lore raw
In reply to Matthias Schiffer
Quoted 30 lines · Matthias Schiffer

This is the first series of improvements to the tqmx86 GPIO driver, which fixes some long-standing bugs - in particular, IRQ_TYPE_EDGE_BOTH can never have worked correctly.

Other patches in the series are code cleanup, which is included as it makes the actual fixes much nicer. I have included the same Fixes tag in all commits, as they will need to be cherry-picked together.

A second series with new features (changing GPIO directions, support more GPIOs on SMARC TQMx86 modules) will be submitted when the fixes have been reviewed and merged.

Gregor Herburger (1):
gpio: tqmx86: fix typo in Kconfig label

Matthias Schiffer (7):
gpio: tqmx86: introduce shadow register for GPIO output value
gpio: tqmx86: change tqmx86_gpio_write() order of arguments to match
regmap API
gpio: tqmx86: introduce _tqmx86_gpio_update_bits() helper
gpio: tqmx86: add macros for interrupt configuration
gpio: tqmx86: store IRQ triggers without offsetting index
gpio: tqmx86: store IRQ trigger type and unmask status separately
gpio: tqmx86: fix broken IRQ_TYPE_EDGE_BOTH interrupt type

 drivers/gpio/Kconfig       |   2 +-
 drivers/gpio/gpio-tqmx86.c | 151 ++++++++++++++++++++++++++-----------
 2 files changed, 106 insertions(+), 47 deletions(-)
Quoted 6 lines

--
TQ-Systems GmbH | Mühlstraße 2, Gut Delling | 82229 Seefeld, Germany Amtsgericht München, HRB 105018
Geschäftsführer: Detlef Schneider, Rüdiger Stahl, Stefan Schneider
https://www.tq-group.com/

Hi Matthias!

Not all patches in this series are fixes (as in: warrant being sent upstream outside of the merge window). Please split the series into two with the first one containing actual fixes to real bugs and the second for any refactoring and improvements.

Bart

#26 lore raw
Quoted 39 lines · Bartosz Golaszewski
Quoted 30 lines · Matthias Schiffer

This is the first series of improvements to the tqmx86 GPIO driver, which fixes some long-standing bugs - in particular, IRQ_TYPE_EDGE_BOTH can never have worked correctly.

Other patches in the series are code cleanup, which is included as it makes the actual fixes much nicer. I have included the same Fixes tag in all commits, as they will need to be cherry-picked together.

A second series with new features (changing GPIO directions, support more GPIOs on SMARC TQMx86 modules) will be submitted when the fixes have been reviewed and merged.

Gregor Herburger (1):
gpio: tqmx86: fix typo in Kconfig label

Matthias Schiffer (7):
gpio: tqmx86: introduce shadow register for GPIO output value
gpio: tqmx86: change tqmx86_gpio_write() order of arguments to match
regmap API
gpio: tqmx86: introduce _tqmx86_gpio_update_bits() helper
gpio: tqmx86: add macros for interrupt configuration
gpio: tqmx86: store IRQ triggers without offsetting index
gpio: tqmx86: store IRQ trigger type and unmask status separately
gpio: tqmx86: fix broken IRQ_TYPE_EDGE_BOTH interrupt type

 drivers/gpio/Kconfig       |   2 +-
 drivers/gpio/gpio-tqmx86.c | 151 ++++++++++++++++++++++++++-----------
 2 files changed, 106 insertions(+), 47 deletions(-)
Quoted 6 lines

--
TQ-Systems GmbH | Mühlstraße 2, Gut Delling | 82229 Seefeld, Germany Amtsgericht München, HRB 105018
Geschäftsführer: Detlef Schneider, Rüdiger Stahl, Stefan Schneider
https://www.tq-group.com/

Quoted 8 lines

Hi Matthias!

Not all patches in this series are fixes (as in: warrant being sent upstream outside of the merge window). Please split the series into two with the first one containing actual fixes to real bugs and the second for any refactoring and improvements.

Bart

Hi Bartosz,

as explained in the cover letter, I've found that the fixes become much nicer to implement (and to review) if they are based on the refactoring. I can leave out _tqmx86_gpio_update_bits for now, but removing "add macros for interrupt configuration" and "store IRQ triggers without offsetting index" does make the actual fixes "store IRQ trigger type and unmask status separately" and "fix broken IRQ_TYPE_EDGE_BOTH interrupt type" somewhat uglier.

That being said, you're the maintainer, and I will structure this series in any way you prefer. I can remove the mentioned refactoring, even though it makes the fixes less pleasant. Another option would be that I can submit just the refactoring for -next for now, and leave the fixes for a future series. Let me know how you want to proceed.

Thanks,
Matthias

#27 lore raw
Quoted 59 lines · Matthias Schiffer
Quoted 39 lines · Bartosz Golaszewski
Quoted 30 lines · Matthias Schiffer

This is the first series of improvements to the tqmx86 GPIO driver, which fixes some long-standing bugs - in particular, IRQ_TYPE_EDGE_BOTH can never have worked correctly.

Other patches in the series are code cleanup, which is included as it makes the actual fixes much nicer. I have included the same Fixes tag in all commits, as they will need to be cherry-picked together.

A second series with new features (changing GPIO directions, support more GPIOs on SMARC TQMx86 modules) will be submitted when the fixes have been reviewed and merged.

Gregor Herburger (1):
gpio: tqmx86: fix typo in Kconfig label

Matthias Schiffer (7):
gpio: tqmx86: introduce shadow register for GPIO output value
gpio: tqmx86: change tqmx86_gpio_write() order of arguments to match
regmap API
gpio: tqmx86: introduce _tqmx86_gpio_update_bits() helper
gpio: tqmx86: add macros for interrupt configuration
gpio: tqmx86: store IRQ triggers without offsetting index
gpio: tqmx86: store IRQ trigger type and unmask status separately
gpio: tqmx86: fix broken IRQ_TYPE_EDGE_BOTH interrupt type

 drivers/gpio/Kconfig       |   2 +-
 drivers/gpio/gpio-tqmx86.c | 151 ++++++++++++++++++++++++++-----------
 2 files changed, 106 insertions(+), 47 deletions(-)
Quoted 6 lines

--
TQ-Systems GmbH | Mühlstraße 2, Gut Delling | 82229 Seefeld, Germany Amtsgericht München, HRB 105018
Geschäftsführer: Detlef Schneider, Rüdiger Stahl, Stefan Schneider
https://www.tq-group.com/

Quoted 8 lines

Hi Matthias!

Not all patches in this series are fixes (as in: warrant being sent upstream outside of the merge window). Please split the series into two with the first one containing actual fixes to real bugs and the second for any refactoring and improvements.

Bart

Hi Bartosz,

as explained in the cover letter, I've found that the fixes become much nicer to implement (and to review) if they are based on the refactoring. I can leave out _tqmx86_gpio_update_bits for now, but removing "add macros for interrupt configuration" and "store IRQ triggers without offsetting index" does make the actual fixes "store IRQ trigger type and unmask status separately" and "fix broken IRQ_TYPE_EDGE_BOTH interrupt type" somewhat uglier.

Quoted 7 lines

That being said, you're the maintainer, and I will structure this series in any way you prefer. I can remove the mentioned refactoring, even though it makes the fixes less pleasant. Another option would be that I can submit just the refactoring for -next for now, and leave the fixes for a future series. Let me know how you want to proceed.

Thanks,
Matthias

The question is: do you want these fixes to be backported into stable branches? Because if not then it's true that the ordering doesn't matter. But if you do, then it makes more sense to put fixes first, send them to Torvalds, get them backported and then add refactoring changes on top.

Bart