[PATCH 0/8] gpio: omap: cleanup: get rid of system GPIO <-> GPIO offset converseations

[PATCH 0/8] gpio: omap: cleanup: get rid of system GPIO <-> GPIO offset converseations

[grygorii.strashko]

From: Grygorii Strashko <grygorii.strashko@linaro.org>

Now in TI OMAP GPIO driver there are a lot of places where
System GPIO number calculated and then converted to GPIO offset.
What is worse is that in many place such conversation performed twice
or even three times. But actually, we don't need to do that at all, because
- gpiolib always passes GPIO offset to GPIO controller
- OMAP GPIO driver converted to use IRQ domain, so
  struct irq_data->hwirq contains GPIO offset

Hence, it is safe to convert all GPIO OMAP functions to use GPIO
offset instead of system GPIO numbers. Also, this allows to remove
unneeded conversations routines
 #define GPIO_INDEX(bank, gpio)
 #define GPIO_BIT(bank, gpio)
 int omap_irq_to_gpio()

Tested on dra7-evm.

Last two patches have to be tested on OMAP1:
-  gpio: omap: get rid of omap_irq_to_gpio()
-  gpio: omap: get rid of GPIO_INDEX() macro

Based on top of Linux 4.0-rc4 plus patch
'[PATCH 1/2] gpio: omap: irq_shutdown: remove unnecessary call of gpiochip_unlock_as_irq'
http://www.spinics.net/lists/linux-omap/msg116482.html

Grygorii Strashko (8):
  gpio: omap: convert omap_gpio_is_input() to use gpio offset
  gpio: omap: simplify omap_set_gpio_dataout_x()
  gpio: omap: convert debounce functions switch to use gpio offset
  gpio: omap: drop 'gpio' param from omap_gpio_init_irq()
  gpio: omap: convert gpio irq functions to use GPIO offset
  gpio: omap: get rid of GPIO_BIT() macro
  gpio: omap: get rid of omap_irq_to_gpio()
  gpio: omap: get rid of GPIO_INDEX() macro

 drivers/gpio/gpio-omap.c | 130 ++++++++++++++++++++---------------------------
 1 file changed, 55 insertions(+), 75 deletions(-)

-- 
1.9.1

[PATCH 1/8] gpio: omap: convert omap_gpio_is_input() to use gpio offset

[grygorii.strashko]

From: Grygorii Strashko <grygorii.strashko@linaro.org>

Convert omap_gpio_is_input() to use GPIO offset instead of mask and,
in such way, make code simpler and remove few lines of code.

Signed-off-by: Grygorii Strashko <grygorii.strashko@linaro.org>
---
 drivers/gpio/gpio-omap.c | 10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)

diff --git a/drivers/gpio/gpio-omap.c b/drivers/gpio/gpio-omap.c
index 2b2fc4b..ce71239 100644
--- a/drivers/gpio/gpio-omap.c
+++ b/drivers/gpio/gpio-omap.c
@@ -472,11 +472,11 @@ static void omap_disable_gpio_module(struct gpio_bank *bank, unsigned offset)
 	}
 }
 
-static int omap_gpio_is_input(struct gpio_bank *bank, int mask)
+static int omap_gpio_is_input(struct gpio_bank *bank, unsigned offset)
 {
 	void __iomem *reg = bank->base + bank->regs->direction;
 
-	return readl_relaxed(reg) & mask;
+	return readl_relaxed(reg) & BIT(offset);
 }
 
 static void omap_gpio_init_irq(struct gpio_bank *bank, unsigned gpio,
@@ -519,7 +519,7 @@ static int omap_gpio_irq_type(struct irq_data *d, unsigned type)
 	offset = GPIO_INDEX(bank, gpio);
 	retval = omap_set_gpio_triggering(bank, offset, type);
 	omap_gpio_init_irq(bank, gpio, offset);
-	if (!omap_gpio_is_input(bank, BIT(offset))) {
+	if (!omap_gpio_is_input(bank, offset)) {
 		spin_unlock_irqrestore(&bank->lock, flags);
 		return -EINVAL;
 	}
@@ -976,12 +976,10 @@ static int omap_gpio_input(struct gpio_chip *chip, unsigned offset)
 static int omap_gpio_get(struct gpio_chip *chip, unsigned offset)
 {
 	struct gpio_bank *bank;
-	u32 mask;
 
 	bank = container_of(chip, struct gpio_bank, chip);
-	mask = (BIT(offset));
 
-	if (omap_gpio_is_input(bank, mask))
+	if (omap_gpio_is_input(bank, offset))
 		return omap_get_gpio_datain(bank, offset);
 	else
 		return omap_get_gpio_dataout(bank, offset);
-- 
1.9.1

Re: [PATCH 1/8] gpio: omap: convert omap_gpio_is_input() to use gpio offset

[santosh.shilimkar]

On 3/19/15 10:25 AM, grygorii.strashko@linaro.org wrote:
> From: Grygorii Strashko <grygorii.strashko@linaro.org>
>
> Convert omap_gpio_is_input() to use GPIO offset instead of mask and,
> in such way, make code simpler and remove few lines of code.
>
> Signed-off-by: Grygorii Strashko <grygorii.strashko@linaro.org>
> ---

Acked-by: Santosh Shilimkar <ssantosh@kernel.org>

Re: [PATCH 1/8] gpio: omap: convert omap_gpio_is_input() to use gpio offset

[Javier Martinez Canillas]

On Fri, Mar 20, 2015 at 5:42 PM, santosh.shilimkar@oracle.com
<santosh.shilimkar@oracle.com> wrote:
> On 3/19/15 10:25 AM, grygorii.strashko@linaro.org wrote:
>>
>> From: Grygorii Strashko <grygorii.strashko@linaro.org>
>>
>> Convert omap_gpio_is_input() to use GPIO offset instead of mask and,
>> in such way, make code simpler and remove few lines of code.
>>
>> Signed-off-by: Grygorii Strashko <grygorii.strashko@linaro.org>
>> ---
>
>
> Acked-by: Santosh Shilimkar <ssantosh@kernel.org>

Acked-by: Javier Martinez Canillas <javier@dowhile0.org>

[PATCH 2/8] gpio: omap: simplify omap_set_gpio_dataout_x()

[grygorii.strashko]

From: Grygorii Strashko <grygorii.strashko@linaro.org>

Both functions omap_set_gpio_dataout_reg() and
omap_set_gpio_dataout_mask() accept GPIO offset
as 'gpio' input parameter, so rename it to 'offset' and
drop usage of GPIO_BIT() macro.

Signed-off-by: Grygorii Strashko <grygorii.strashko@linaro.org>
---
 drivers/gpio/gpio-omap.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/drivers/gpio/gpio-omap.c b/drivers/gpio/gpio-omap.c
index ce71239..7aeee7b 100644
--- a/drivers/gpio/gpio-omap.c
+++ b/drivers/gpio/gpio-omap.c
@@ -75,7 +75,7 @@ struct gpio_bank {
 	int power_mode;
 	bool workaround_enabled;
 
-	void (*set_dataout)(struct gpio_bank *bank, int gpio, int enable);
+	void (*set_dataout)(struct gpio_bank *bank, unsigned gpio, int enable);
 	int (*get_context_loss_count)(struct device *dev);
 
 	struct omap_gpio_reg_offs *regs;
@@ -119,11 +119,11 @@ static void omap_set_gpio_direction(struct gpio_bank *bank, int gpio,
 
 
 /* set data out value using dedicate set/clear register */
-static void omap_set_gpio_dataout_reg(struct gpio_bank *bank, int gpio,
+static void omap_set_gpio_dataout_reg(struct gpio_bank *bank, unsigned offset,
 				      int enable)
 {
 	void __iomem *reg = bank->base;
-	u32 l = GPIO_BIT(bank, gpio);
+	u32 l = BIT(offset);
 
 	if (enable) {
 		reg += bank->regs->set_dataout;
@@ -137,11 +137,11 @@ static void omap_set_gpio_dataout_reg(struct gpio_bank *bank, int gpio,
 }
 
 /* set data out value using mask register */
-static void omap_set_gpio_dataout_mask(struct gpio_bank *bank, int gpio,
+static void omap_set_gpio_dataout_mask(struct gpio_bank *bank, unsigned offset,
 				       int enable)
 {
 	void __iomem *reg = bank->base + bank->regs->dataout;
-	u32 gpio_bit = GPIO_BIT(bank, gpio);
+	u32 gpio_bit = BIT(offset);
 	u32 l;
 
 	l = readl_relaxed(reg);
-- 
1.9.1

Re: [PATCH 2/8] gpio: omap: simplify omap_set_gpio_dataout_x()

[santosh.shilimkar]

On 3/19/15 10:25 AM, grygorii.strashko@linaro.org wrote:
> From: Grygorii Strashko <grygorii.strashko@linaro.org>
>
> Both functions omap_set_gpio_dataout_reg() and
> omap_set_gpio_dataout_mask() accept GPIO offset
> as 'gpio' input parameter, so rename it to 'offset' and
> drop usage of GPIO_BIT() macro.
>
> Signed-off-by: Grygorii Strashko <grygorii.strashko@linaro.org>
> ---
Acked-by: Santosh Shilimkar <ssantosh@kernel.org>

Re: [PATCH 2/8] gpio: omap: simplify omap_set_gpio_dataout_x()

[Javier Martinez Canillas]

On Fri, Mar 20, 2015 at 5:45 PM, santosh.shilimkar@oracle.com
<santosh.shilimkar@oracle.com> wrote:
> On 3/19/15 10:25 AM, grygorii.strashko@linaro.org wrote:
>>
>> From: Grygorii Strashko <grygorii.strashko@linaro.org>
>>
>> Both functions omap_set_gpio_dataout_reg() and
>> omap_set_gpio_dataout_mask() accept GPIO offset
>> as 'gpio' input parameter, so rename it to 'offset' and
>> drop usage of GPIO_BIT() macro.
>>
>> Signed-off-by: Grygorii Strashko <grygorii.strashko@linaro.org>
>> ---
>
> Acked-by: Santosh Shilimkar <ssantosh@kernel.org>

Acked-by: Javier Martinez Canillas <javier@dowhile0.org>

[PATCH 3/8] gpio: omap: convert debounce functions switch to use gpio offset

[grygorii.strashko]

From: Grygorii Strashko <grygorii.strashko@linaro.org>

Convert debounce functions to use GPIO offset instead of system
GPIO numbers. This allows to drop unneeded conversations between
system GPIO <-> GPIO offset which are done in many places and
many times.
It is safe to do now because:
- gpiolib always passes GPIO offset to GPIO controller
- OMAP GPIO driver converted to use IRQ domain

This is preparation step before removing:
 #define GPIO_INDEX(bank, gpio)
 #define GPIO_BIT(bank, gpio)
 int omap_irq_to_gpio()

Signed-off-by: Grygorii Strashko <grygorii.strashko@linaro.org>
---
 drivers/gpio/gpio-omap.c | 14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

diff --git a/drivers/gpio/gpio-omap.c b/drivers/gpio/gpio-omap.c
index 7aeee7b..2df693b 100644
--- a/drivers/gpio/gpio-omap.c
+++ b/drivers/gpio/gpio-omap.c
@@ -208,13 +208,13 @@ static inline void omap_gpio_dbck_disable(struct gpio_bank *bank)
 /**
  * omap2_set_gpio_debounce - low level gpio debounce time
  * @bank: the gpio bank we're acting upon
- * @gpio: the gpio number on this @gpio
+ * @offset: the gpio number on this @bank
  * @debounce: debounce time to use
  *
  * OMAP's debounce time is in 31us steps so we need
  * to convert and round up to the closest unit.
  */
-static void omap2_set_gpio_debounce(struct gpio_bank *bank, unsigned gpio,
+static void omap2_set_gpio_debounce(struct gpio_bank *bank, unsigned offset,
 				    unsigned debounce)
 {
 	void __iomem		*reg;
@@ -231,7 +231,7 @@ static void omap2_set_gpio_debounce(struct gpio_bank *bank, unsigned gpio,
 	else
 		debounce = (debounce / 0x1f) - 1;
 
-	l = GPIO_BIT(bank, gpio);
+	l = BIT(offset);
 
 	clk_prepare_enable(bank->dbck);
 	reg = bank->base + bank->regs->debounce;
@@ -266,16 +266,16 @@ static void omap2_set_gpio_debounce(struct gpio_bank *bank, unsigned gpio,
 /**
  * omap_clear_gpio_debounce - clear debounce settings for a gpio
  * @bank: the gpio bank we're acting upon
- * @gpio: the gpio number on this @gpio
+ * @offset: the gpio number on this @bank
  *
  * If a gpio is using debounce, then clear the debounce enable bit and if
  * this is the only gpio in this bank using debounce, then clear the debounce
  * time too. The debounce clock will also be disabled when calling this function
  * if this is the only gpio in the bank using debounce.
  */
-static void omap_clear_gpio_debounce(struct gpio_bank *bank, unsigned gpio)
+static void omap_clear_gpio_debounce(struct gpio_bank *bank, unsigned offset)
 {
-	u32 gpio_bit = GPIO_BIT(bank, gpio);
+	u32 gpio_bit = BIT(offset);
 
 	if (!bank->dbck_flag)
 		return;
@@ -659,7 +659,7 @@ static void omap_reset_gpio(struct gpio_bank *bank, int gpio)
 	omap_set_gpio_irqenable(bank, gpio, 0);
 	omap_clear_gpio_irqstatus(bank, gpio);
 	omap_set_gpio_triggering(bank, GPIO_INDEX(bank, gpio), IRQ_TYPE_NONE);
-	omap_clear_gpio_debounce(bank, gpio);
+	omap_clear_gpio_debounce(bank, GPIO_INDEX(bank, gpio));
 }
 
 /* Use disable_irq_wake() and enable_irq_wake() functions from drivers */
-- 
1.9.1

Re: [PATCH 3/8] gpio: omap: convert debounce functions switch to use gpio offset

[Javier Martinez Canillas]

On Thu, Mar 19, 2015 at 6:25 PM,  <grygorii.strashko@linaro.org> wrote:
> From: Grygorii Strashko <grygorii.strashko@linaro.org>
>
> Convert debounce functions to use GPIO offset instead of system
> GPIO numbers. This allows to drop unneeded conversations between
> system GPIO <-> GPIO offset which are done in many places and
> many times.
> It is safe to do now because:
> - gpiolib always passes GPIO offset to GPIO controller
> - OMAP GPIO driver converted to use IRQ domain
>
> This is preparation step before removing:
>  #define GPIO_INDEX(bank, gpio)
>  #define GPIO_BIT(bank, gpio)
>  int omap_irq_to_gpio()
>
> Signed-off-by: Grygorii Strashko <grygorii.strashko@linaro.org>
> ---

Acked-by: Javier Martinez Canillas <javier@dowhile0.org>

[PATCH 4/8] gpio: omap: drop 'gpio' param from omap_gpio_init_irq()

[grygorii.strashko]

From: Grygorii Strashko <grygorii.strashko@linaro.org>

The 'gpio' parameter isn't needed any more as it
duplicates 'offset' parameter, so drop it.

Signed-off-by: Grygorii Strashko <grygorii.strashko@linaro.org>
---
 drivers/gpio/gpio-omap.c | 12 +++++-------
 1 file changed, 5 insertions(+), 7 deletions(-)

diff --git a/drivers/gpio/gpio-omap.c b/drivers/gpio/gpio-omap.c
index 2df693b..ff5d54d 100644
--- a/drivers/gpio/gpio-omap.c
+++ b/drivers/gpio/gpio-omap.c
@@ -479,14 +479,13 @@ static int omap_gpio_is_input(struct gpio_bank *bank, unsigned offset)
 	return readl_relaxed(reg) & BIT(offset);
 }
 
-static void omap_gpio_init_irq(struct gpio_bank *bank, unsigned gpio,
-			       unsigned offset)
+static void omap_gpio_init_irq(struct gpio_bank *bank, unsigned offset)
 {
 	if (!LINE_USED(bank->mod_usage, offset)) {
 		omap_enable_gpio_module(bank, offset);
 		omap_set_gpio_direction(bank, offset, 1);
 	}
-	bank->irq_usage |= BIT(GPIO_INDEX(bank, gpio));
+	bank->irq_usage |= BIT(offset);
 }
 
 static int omap_gpio_irq_type(struct irq_data *d, unsigned type)
@@ -518,7 +517,7 @@ static int omap_gpio_irq_type(struct irq_data *d, unsigned type)
 	spin_lock_irqsave(&bank->lock, flags);
 	offset = GPIO_INDEX(bank, gpio);
 	retval = omap_set_gpio_triggering(bank, offset, type);
-	omap_gpio_init_irq(bank, gpio, offset);
+	omap_gpio_init_irq(bank, offset);
 	if (!omap_gpio_is_input(bank, offset)) {
 		spin_unlock_irqrestore(&bank->lock, flags);
 		return -EINVAL;
@@ -803,15 +802,14 @@ exit:
 static unsigned int omap_gpio_irq_startup(struct irq_data *d)
 {
 	struct gpio_bank *bank = omap_irq_data_get_bank(d);
-	unsigned int gpio = omap_irq_to_gpio(bank, d->hwirq);
 	unsigned long flags;
-	unsigned offset = GPIO_INDEX(bank, gpio);
+	unsigned offset = d->hwirq;
 
 	if (!BANK_USED(bank))
 		pm_runtime_get_sync(bank->dev);
 
 	spin_lock_irqsave(&bank->lock, flags);
-	omap_gpio_init_irq(bank, gpio, offset);
+	omap_gpio_init_irq(bank, offset);
 	spin_unlock_irqrestore(&bank->lock, flags);
 	omap_gpio_unmask_irq(d);
 
-- 
1.9.1

Re: [PATCH 4/8] gpio: omap: drop 'gpio' param from omap_gpio_init_irq()

[Javier Martinez Canillas]

On Thu, Mar 19, 2015 at 6:25 PM,  <grygorii.strashko@linaro.org> wrote:
> From: Grygorii Strashko <grygorii.strashko@linaro.org>
>
> The 'gpio' parameter isn't needed any more as it
> duplicates 'offset' parameter, so drop it.
>
> Signed-off-by: Grygorii Strashko <grygorii.strashko@linaro.org>

Acked-by: Javier Martinez Canillas <javier@dowhile0.org>

[PATCH 5/8] gpio: omap: convert gpio irq functions to use GPIO offset

[grygorii.strashko]

From: Grygorii Strashko <grygorii.strashko@linaro.org>

Convert GPIO IRQ functions to use GPIO offset instead of system
GPIO numbers. This allows to drop unneeded conversations between
system GPIO <-> GPIO offset which are done in many places and
many times.
It is safe to do now because:
- gpiolib always passes GPIO offset to GPIO controller
- OMAP GPIO driver converted to use IRQ domain, so
  struct irq_data->hwirq contains GPIO offset

This is preparation step before removing:
 #define GPIO_INDEX(bank, gpio)
 #define GPIO_BIT(bank, gpio)
 int omap_irq_to_gpio()

Signed-off-by: Grygorii Strashko <grygorii.strashko@linaro.org>
---
 drivers/gpio/gpio-omap.c | 64 +++++++++++++++++++++++++-----------------------
 1 file changed, 33 insertions(+), 31 deletions(-)

diff --git a/drivers/gpio/gpio-omap.c b/drivers/gpio/gpio-omap.c
index ff5d54d..b39faa8 100644
--- a/drivers/gpio/gpio-omap.c
+++ b/drivers/gpio/gpio-omap.c
@@ -549,9 +549,10 @@ static void omap_clear_gpio_irqbank(struct gpio_bank *bank, int gpio_mask)
 	readl_relaxed(reg);
 }
 
-static inline void omap_clear_gpio_irqstatus(struct gpio_bank *bank, int gpio)
+static inline void omap_clear_gpio_irqstatus(struct gpio_bank *bank,
+					     unsigned offset)
 {
-	omap_clear_gpio_irqbank(bank, GPIO_BIT(bank, gpio));
+	omap_clear_gpio_irqbank(bank, BIT(offset));
 }
 
 static u32 omap_get_gpio_irqbank_mask(struct gpio_bank *bank)
@@ -612,13 +613,13 @@ static void omap_disable_gpio_irqbank(struct gpio_bank *bank, int gpio_mask)
 	writel_relaxed(l, reg);
 }
 
-static inline void omap_set_gpio_irqenable(struct gpio_bank *bank, int gpio,
-					   int enable)
+static inline void omap_set_gpio_irqenable(struct gpio_bank *bank,
+					   unsigned offset, int enable)
 {
 	if (enable)
-		omap_enable_gpio_irqbank(bank, GPIO_BIT(bank, gpio));
+		omap_enable_gpio_irqbank(bank, BIT(offset));
 	else
-		omap_disable_gpio_irqbank(bank, GPIO_BIT(bank, gpio));
+		omap_disable_gpio_irqbank(bank, BIT(offset));
 }
 
 /*
@@ -629,14 +630,16 @@ static inline void omap_set_gpio_irqenable(struct gpio_bank *bank, int gpio,
  * enabled. When system is suspended, only selected GPIO interrupts need
  * to have wake-up enabled.
  */
-static int omap_set_gpio_wakeup(struct gpio_bank *bank, int gpio, int enable)
+static int omap_set_gpio_wakeup(struct gpio_bank *bank, unsigned offset,
+				int enable)
 {
-	u32 gpio_bit = GPIO_BIT(bank, gpio);
+	u32 gpio_bit = BIT(offset);
 	unsigned long flags;
 
 	if (bank->non_wakeup_gpios & gpio_bit) {
 		dev_err(bank->dev,
-			"Unable to modify wakeup on non-wakeup GPIO%d\n", gpio);
+			"Unable to modify wakeup on non-wakeup GPIO%d\n",
+			offset);
 		return -EINVAL;
 	}
 
@@ -652,22 +655,22 @@ static int omap_set_gpio_wakeup(struct gpio_bank *bank, int gpio, int enable)
 	return 0;
 }
 
-static void omap_reset_gpio(struct gpio_bank *bank, int gpio)
+static void omap_reset_gpio(struct gpio_bank *bank, unsigned offset)
 {
-	omap_set_gpio_direction(bank, GPIO_INDEX(bank, gpio), 1);
-	omap_set_gpio_irqenable(bank, gpio, 0);
-	omap_clear_gpio_irqstatus(bank, gpio);
-	omap_set_gpio_triggering(bank, GPIO_INDEX(bank, gpio), IRQ_TYPE_NONE);
-	omap_clear_gpio_debounce(bank, GPIO_INDEX(bank, gpio));
+	omap_set_gpio_direction(bank, offset, 1);
+	omap_set_gpio_irqenable(bank, offset, 0);
+	omap_clear_gpio_irqstatus(bank, offset);
+	omap_set_gpio_triggering(bank, offset, IRQ_TYPE_NONE);
+	omap_clear_gpio_debounce(bank, offset);
 }
 
 /* Use disable_irq_wake() and enable_irq_wake() functions from drivers */
 static int omap_gpio_wake_enable(struct irq_data *d, unsigned int enable)
 {
 	struct gpio_bank *bank = omap_irq_data_get_bank(d);
-	unsigned int gpio = omap_irq_to_gpio(bank, d->hwirq);
+	unsigned offset = d->hwirq;
 
-	return omap_set_gpio_wakeup(bank, gpio, enable);
+	return omap_set_gpio_wakeup(bank, offset, enable);
 }
 
 static int omap_gpio_request(struct gpio_chip *chip, unsigned offset)
@@ -705,7 +708,7 @@ static void omap_gpio_free(struct gpio_chip *chip, unsigned offset)
 	spin_lock_irqsave(&bank->lock, flags);
 	bank->mod_usage &= ~(BIT(offset));
 	omap_disable_gpio_module(bank, offset);
-	omap_reset_gpio(bank, bank->chip.base + offset);
+	omap_reset_gpio(bank, offset);
 	spin_unlock_irqrestore(&bank->lock, flags);
 
 	/*
@@ -819,14 +822,13 @@ static unsigned int omap_gpio_irq_startup(struct irq_data *d)
 static void omap_gpio_irq_shutdown(struct irq_data *d)
 {
 	struct gpio_bank *bank = omap_irq_data_get_bank(d);
-	unsigned int gpio = omap_irq_to_gpio(bank, d->hwirq);
 	unsigned long flags;
-	unsigned offset = GPIO_INDEX(bank, gpio);
+	unsigned offset = d->hwirq;
 
 	spin_lock_irqsave(&bank->lock, flags);
 	bank->irq_usage &= ~(BIT(offset));
 	omap_disable_gpio_module(bank, offset);
-	omap_reset_gpio(bank, gpio);
+	omap_reset_gpio(bank, offset);
 	spin_unlock_irqrestore(&bank->lock, flags);
 
 	/*
@@ -840,43 +842,43 @@ static void omap_gpio_irq_shutdown(struct irq_data *d)
 static void omap_gpio_ack_irq(struct irq_data *d)
 {
 	struct gpio_bank *bank = omap_irq_data_get_bank(d);
-	unsigned int gpio = omap_irq_to_gpio(bank, d->hwirq);
+	unsigned offset = d->hwirq;
 
-	omap_clear_gpio_irqstatus(bank, gpio);
+	omap_clear_gpio_irqstatus(bank, offset);
 }
 
 static void omap_gpio_mask_irq(struct irq_data *d)
 {
 	struct gpio_bank *bank = omap_irq_data_get_bank(d);
-	unsigned int gpio = omap_irq_to_gpio(bank, d->hwirq);
+	unsigned offset = d->hwirq;
 	unsigned long flags;
 
 	spin_lock_irqsave(&bank->lock, flags);
-	omap_set_gpio_irqenable(bank, gpio, 0);
-	omap_set_gpio_triggering(bank, GPIO_INDEX(bank, gpio), IRQ_TYPE_NONE);
+	omap_set_gpio_irqenable(bank, offset, 0);
+	omap_set_gpio_triggering(bank, offset, IRQ_TYPE_NONE);
 	spin_unlock_irqrestore(&bank->lock, flags);
 }
 
 static void omap_gpio_unmask_irq(struct irq_data *d)
 {
 	struct gpio_bank *bank = omap_irq_data_get_bank(d);
-	unsigned int gpio = omap_irq_to_gpio(bank, d->hwirq);
+	unsigned offset = d->hwirq;
 	unsigned int irq_mask = GPIO_BIT(bank, gpio);
 	u32 trigger = irqd_get_trigger_type(d);
 	unsigned long flags;
 
 	spin_lock_irqsave(&bank->lock, flags);
 	if (trigger)
-		omap_set_gpio_triggering(bank, GPIO_INDEX(bank, gpio), trigger);
+		omap_set_gpio_triggering(bank, offset, trigger);
 
 	/* For level-triggered GPIOs, the clearing must be done after
 	 * the HW source is cleared, thus after the handler has run */
 	if (bank->level_mask & irq_mask) {
-		omap_set_gpio_irqenable(bank, gpio, 0);
-		omap_clear_gpio_irqstatus(bank, gpio);
+		omap_set_gpio_irqenable(bank, offset, 0);
+		omap_clear_gpio_irqstatus(bank, offset);
 	}
 
-	omap_set_gpio_irqenable(bank, gpio, 1);
+	omap_set_gpio_irqenable(bank, offset, 1);
 	spin_unlock_irqrestore(&bank->lock, flags);
 }
 
-- 
1.9.1

Re: [PATCH 5/8] gpio: omap: convert gpio irq functions to use GPIO offset

[Tony Lindgren]

* grygorii.strashko@linaro.org <grygorii.strashko@linaro.org> [150319 10:26]:
> From: Grygorii Strashko <grygorii.strashko@linaro.org>
> 
> Convert GPIO IRQ functions to use GPIO offset instead of system
> GPIO numbers. This allows to drop unneeded conversations between
> system GPIO <-> GPIO offset which are done in many places and
> many times.
> It is safe to do now because:
> - gpiolib always passes GPIO offset to GPIO controller
> - OMAP GPIO driver converted to use IRQ domain, so
>   struct irq_data->hwirq contains GPIO offset
> 
> This is preparation step before removing:
>  #define GPIO_INDEX(bank, gpio)
>  #define GPIO_BIT(bank, gpio)
>  int omap_irq_to_gpio()
... 
  
>  static void omap_gpio_unmask_irq(struct irq_data *d)
>  {
>  	struct gpio_bank *bank = omap_irq_data_get_bank(d);
> -	unsigned int gpio = omap_irq_to_gpio(bank, d->hwirq);
> +	unsigned offset = d->hwirq;
>  	unsigned int irq_mask = GPIO_BIT(bank, gpio);
>  	u32 trigger = irqd_get_trigger_type(d);
>  	unsigned long flags;

This series up to this patch produces a build error that
would break git bisect:

drivers/gpio/gpio-omap.c: In function ‘omap_gpio_unmask_irq’:
drivers/gpio/gpio-omap.c:866:37: error: ‘gpio’ undeclared (first use in this function)
  unsigned int irq_mask = GPIO_BIT(bank, gpio);

Applying the following patch makes things build again.

Regards,

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

Re: [PATCH 5/8] gpio: omap: convert gpio irq functions to use GPIO offset

[Grygorii.Strashko@linaro.org]

On 03/20/2015 01:03 AM, Tony Lindgren wrote:
> * grygorii.strashko@linaro.org <grygorii.strashko@linaro.org> [150319 10:26]:
>> From: Grygorii Strashko <grygorii.strashko@linaro.org>
>>
>> Convert GPIO IRQ functions to use GPIO offset instead of system
>> GPIO numbers. This allows to drop unneeded conversations between
>> system GPIO <-> GPIO offset which are done in many places and
>> many times.
>> It is safe to do now because:
>> - gpiolib always passes GPIO offset to GPIO controller
>> - OMAP GPIO driver converted to use IRQ domain, so
>>    struct irq_data->hwirq contains GPIO offset
>>
>> This is preparation step before removing:
>>   #define GPIO_INDEX(bank, gpio)
>>   #define GPIO_BIT(bank, gpio)
>>   int omap_irq_to_gpio()
> ...
>
>>   static void omap_gpio_unmask_irq(struct irq_data *d)
>>   {
>>   	struct gpio_bank *bank = omap_irq_data_get_bank(d);
>> -	unsigned int gpio = omap_irq_to_gpio(bank, d->hwirq);
>> +	unsigned offset = d->hwirq;
>>   	unsigned int irq_mask = GPIO_BIT(bank, gpio);
>>   	u32 trigger = irqd_get_trigger_type(d);
>>   	unsigned long flags;
>
> This series up to this patch produces a build error that
> would break git bisect:
>
> drivers/gpio/gpio-omap.c: In function ‘omap_gpio_unmask_irq’:
> drivers/gpio/gpio-omap.c:866:37: error: ‘gpio’ undeclared (first use in this function)
>    unsigned int irq_mask = GPIO_BIT(bank, gpio);

Oh. Thanks for catching that :) - splitting/rebasing issue.

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

Re: [PATCH 5/8] gpio: omap: convert gpio irq functions to use GPIO offset

[santosh.shilimkar]

On 3/19/15 10:25 AM, grygorii.strashko@linaro.org wrote:
> From: Grygorii Strashko <grygorii.strashko@linaro.org>
>
> Convert GPIO IRQ functions to use GPIO offset instead of system
> GPIO numbers. This allows to drop unneeded conversations between
> system GPIO <-> GPIO offset which are done in many places and
> many times.
> It is safe to do now because:
> - gpiolib always passes GPIO offset to GPIO controller
> - OMAP GPIO driver converted to use IRQ domain, so
>    struct irq_data->hwirq contains GPIO offset
>
> This is preparation step before removing:
>   #define GPIO_INDEX(bank, gpio)
>   #define GPIO_BIT(bank, gpio)
>   int omap_irq_to_gpio()
>
> Signed-off-by: Grygorii Strashko <grygorii.strashko@linaro.org>
> ---

Once you take care of the bisect/build issue, feel free to add
Acked-by: Santosh Shilimkar <ssantosh@kernel.org>

Re: [PATCH 5/8] gpio: omap: convert gpio irq functions to use GPIO offset

[Javier Martinez Canillas]

On Fri, Mar 20, 2015 at 5:49 PM, santosh.shilimkar@oracle.com
<santosh.shilimkar@oracle.com> wrote:
> On 3/19/15 10:25 AM, grygorii.strashko@linaro.org wrote:
>>
>> From: Grygorii Strashko <grygorii.strashko@linaro.org>
>>
>> Convert GPIO IRQ functions to use GPIO offset instead of system
>> GPIO numbers. This allows to drop unneeded conversations between
>> system GPIO <-> GPIO offset which are done in many places and
>> many times.
>> It is safe to do now because:
>> - gpiolib always passes GPIO offset to GPIO controller
>> - OMAP GPIO driver converted to use IRQ domain, so
>>    struct irq_data->hwirq contains GPIO offset
>>
>> This is preparation step before removing:
>>   #define GPIO_INDEX(bank, gpio)
>>   #define GPIO_BIT(bank, gpio)
>>   int omap_irq_to_gpio()
>>
>> Signed-off-by: Grygorii Strashko <grygorii.strashko@linaro.org>
>> ---
>
>
> Once you take care of the bisect/build issue, feel free to add
> Acked-by: Santosh Shilimkar <ssantosh@kernel.org>

Same for me, after the rebase/split fixup:

Acked-by: Javier Martinez Canillas <javier@dowhile0.org>

[PATCH 6/8] gpio: omap: get rid of GPIO_BIT() macro

[grygorii.strashko]

From: Grygorii Strashko <grygorii.strashko@linaro.org>

Now OMAP GPIO driver prepared for GPIO_BIT() macro removing.
Do it ;)

Signed-off-by: Grygorii Strashko <grygorii.strashko@linaro.org>
---
 drivers/gpio/gpio-omap.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/gpio/gpio-omap.c b/drivers/gpio/gpio-omap.c
index b39faa8..dd7934a 100644
--- a/drivers/gpio/gpio-omap.c
+++ b/drivers/gpio/gpio-omap.c
@@ -82,7 +82,6 @@ struct gpio_bank {
 };
 
 #define GPIO_INDEX(bank, gpio) (gpio % bank->width)
-#define GPIO_BIT(bank, gpio) (BIT(GPIO_INDEX(bank, gpio)))
 #define GPIO_MOD_CTRL_BIT	BIT(0)
 
 #define BANK_USED(bank) (bank->mod_usage || bank->irq_usage)
@@ -863,7 +862,6 @@ static void omap_gpio_unmask_irq(struct irq_data *d)
 {
 	struct gpio_bank *bank = omap_irq_data_get_bank(d);
 	unsigned offset = d->hwirq;
-	unsigned int irq_mask = GPIO_BIT(bank, gpio);
 	u32 trigger = irqd_get_trigger_type(d);
 	unsigned long flags;
 
@@ -873,7 +871,7 @@ static void omap_gpio_unmask_irq(struct irq_data *d)
 
 	/* For level-triggered GPIOs, the clearing must be done after
 	 * the HW source is cleared, thus after the handler has run */
-	if (bank->level_mask & irq_mask) {
+	if (bank->level_mask & BIT(offset)) {
 		omap_set_gpio_irqenable(bank, offset, 0);
 		omap_clear_gpio_irqstatus(bank, offset);
 	}
-- 
1.9.1

Re: [PATCH 6/8] gpio: omap: get rid of GPIO_BIT() macro

[santosh.shilimkar]

On 3/19/15 10:25 AM, grygorii.strashko@linaro.org wrote:
> From: Grygorii Strashko <grygorii.strashko@linaro.org>
>
> Now OMAP GPIO driver prepared for GPIO_BIT() macro removing.
> Do it ;)
>
> Signed-off-by: Grygorii Strashko <grygorii.strashko@linaro.org>
> ---

Thanks for the cleanup Grygorii.

Acked-by: Santosh Shilimkar <ssantosh@kernel.org>

Re: [PATCH 6/8] gpio: omap: get rid of GPIO_BIT() macro

[Javier Martinez Canillas]

On Fri, Mar 20, 2015 at 5:50 PM, santosh.shilimkar@oracle.com
<santosh.shilimkar@oracle.com> wrote:
> On 3/19/15 10:25 AM, grygorii.strashko@linaro.org wrote:
>>
>> From: Grygorii Strashko <grygorii.strashko@linaro.org>
>>
>> Now OMAP GPIO driver prepared for GPIO_BIT() macro removing.
>> Do it ;)
>>
>> Signed-off-by: Grygorii Strashko <grygorii.strashko@linaro.org>
>> ---
>
>
> Thanks for the cleanup Grygorii.
>
> Acked-by: Santosh Shilimkar <ssantosh@kernel.org>

Indeed!

Acked-by: Javier Martinez Canillas <javier@dowhile0.org>

[RFT OMAP1 PATCH 7/8] gpio: omap: get rid of omap_irq_to_gpio()

[grygorii.strashko]

From: Grygorii Strashko <grygorii.strashko@linaro.org>

Now OMAP GPIO driver prepared for omap_irq_to_gpio() removing.
Do it ;)

Signed-off-by: Grygorii Strashko <grygorii.strashko@linaro.org>
---
 drivers/gpio/gpio-omap.c | 17 +++++------------
 1 file changed, 5 insertions(+), 12 deletions(-)

diff --git a/drivers/gpio/gpio-omap.c b/drivers/gpio/gpio-omap.c
index dd7934a..615e950 100644
--- a/drivers/gpio/gpio-omap.c
+++ b/drivers/gpio/gpio-omap.c
@@ -89,11 +89,6 @@ struct gpio_bank {
 
 static void omap_gpio_unmask_irq(struct irq_data *d);
 
-static int omap_irq_to_gpio(struct gpio_bank *bank, unsigned int gpio_irq)
-{
-	return bank->chip.base + gpio_irq;
-}
-
 static inline struct gpio_bank *omap_irq_data_get_bank(struct irq_data *d)
 {
 	struct gpio_chip *chip = irq_data_get_irq_chip_data(d);
@@ -490,22 +485,21 @@ static void omap_gpio_init_irq(struct gpio_bank *bank, unsigned offset)
 static int omap_gpio_irq_type(struct irq_data *d, unsigned type)
 {
 	struct gpio_bank *bank = omap_irq_data_get_bank(d);
-	unsigned gpio = 0;
 	int retval;
 	unsigned long flags;
-	unsigned offset;
+	unsigned offset = d->hwirq;
 
 	if (!BANK_USED(bank))
 		pm_runtime_get_sync(bank->dev);
 
 #ifdef CONFIG_ARCH_OMAP1
-	if (d->irq > IH_MPUIO_BASE)
+	if (d->irq > IH_MPUIO_BASE) {
+		unsigned gpio = 0;
 		gpio = OMAP_MPUIO(d->irq - IH_MPUIO_BASE);
+		offset = GPIO_INDEX(bank, gpio);
+	}
 #endif
 
-	if (!gpio)
-		gpio = omap_irq_to_gpio(bank, d->hwirq);
-
 	if (type & ~IRQ_TYPE_SENSE_MASK)
 		return -EINVAL;
 
@@ -514,7 +508,6 @@ static int omap_gpio_irq_type(struct irq_data *d, unsigned type)
 		return -EINVAL;
 
 	spin_lock_irqsave(&bank->lock, flags);
-	offset = GPIO_INDEX(bank, gpio);
 	retval = omap_set_gpio_triggering(bank, offset, type);
 	omap_gpio_init_irq(bank, offset);
 	if (!omap_gpio_is_input(bank, offset)) {
-- 
1.9.1

Re: [RFT OMAP1 PATCH 7/8] gpio: omap: get rid of omap_irq_to_gpio()

[Javier Martinez Canillas]

Hello Grygorii,

On Thu, Mar 19, 2015 at 6:25 PM,  <grygorii.strashko@linaro.org> wrote:
> From: Grygorii Strashko <grygorii.strashko@linaro.org>
>
> Now OMAP GPIO driver prepared for omap_irq_to_gpio() removing.
> Do it ;)
>
> Signed-off-by: Grygorii Strashko <grygorii.strashko@linaro.org>

I don't have access to OMAP1 boards to test but Tony did it o an
osk5912 and Aaro can confirm that it doesn't cause regressions on
other OMAP1 boards.

Thanks a lot for the cleanup.

Acked-by: Javier Martinez Canillas <javier@dowhile0.org>

Best regards,
Javier

Re: [RFT OMAP1 PATCH 7/8] gpio: omap: get rid of omap_irq_to_gpio()

[Grygorii.Strashko@linaro.org]

On 03/20/2015 08:56 PM, Javier Martinez Canillas wrote:
> Hello Grygorii,
> 
> On Thu, Mar 19, 2015 at 6:25 PM,  <grygorii.strashko@linaro.org> wrote:
>> From: Grygorii Strashko <grygorii.strashko@linaro.org>
>>
>> Now OMAP GPIO driver prepared for omap_irq_to_gpio() removing.
>> Do it ;)
>>
>> Signed-off-by: Grygorii Strashko <grygorii.strashko@linaro.org>
> 
> I don't have access to OMAP1 boards to test but Tony did it o an
> osk5912 and Aaro can confirm that it doesn't cause regressions on
> other OMAP1 boards.

Thanks. But could you clarify one point to me please? 
"Aaro can confirm that it doesn't cause regressions on
other OMAP1 boards" - Does It mean I can add his Tested-by?
Or I still have to wait for his direct reply? 
Just to avoid any confusions ;)

> Acked-by: Javier Martinez Canillas <javier@dowhile0.org>
> 


-- 
regards,
-grygorii

Re: [RFT OMAP1 PATCH 7/8] gpio: omap: get rid of omap_irq_to_gpio()

[Javier Martinez Canillas]

Hello Grygorii,

On Fri, Mar 20, 2015 at 9:39 PM, Grygorii.Strashko@linaro.org
<grygorii.strashko@linaro.org> wrote:
> On 03/20/2015 08:56 PM, Javier Martinez Canillas wrote:
>> Hello Grygorii,
>>
>> On Thu, Mar 19, 2015 at 6:25 PM,  <grygorii.strashko@linaro.org> wrote:
>>> From: Grygorii Strashko <grygorii.strashko@linaro.org>
>>>
>>> Now OMAP GPIO driver prepared for omap_irq_to_gpio() removing.
>>> Do it ;)
>>>
>>> Signed-off-by: Grygorii Strashko <grygorii.strashko@linaro.org>
>>
>> I don't have access to OMAP1 boards to test but Tony did it o an
>> osk5912 and Aaro can confirm that it doesn't cause regressions on
>> other OMAP1 boards.
>
> Thanks. But could you clarify one point to me please?
> "Aaro can confirm that it doesn't cause regressions on
> other OMAP1 boards" - Does It mean I can add his Tested-by?
> Or I still have to wait for his direct reply?
> Just to avoid any confusions ;)
>

I meant that as far as I know Aaro has access to more OMAP1 systems so
he should be able to test that no regressions are added for those. But
I don't know if he would be able to test and I think your cleanups are
good so that's why I acked the patches.

Sorry for the confusion.

Best regards,
Javier

[RFT OMAP1 PATCH 8/8] gpio: omap: get rid of GPIO_INDEX() macro

[grygorii.strashko]

From: Grygorii Strashko <grygorii.strashko@linaro.org>

Now OMAP GPIO driver prepared for GPIO_INDEX() macro removing.
Do It ;)

Signed-off-by: Grygorii Strashko <grygorii.strashko@linaro.org>
---
 drivers/gpio/gpio-omap.c | 9 ---------
 1 file changed, 9 deletions(-)

diff --git a/drivers/gpio/gpio-omap.c b/drivers/gpio/gpio-omap.c
index 615e950..cd1d5bf 100644
--- a/drivers/gpio/gpio-omap.c
+++ b/drivers/gpio/gpio-omap.c
@@ -81,7 +81,6 @@ struct gpio_bank {
 	struct omap_gpio_reg_offs *regs;
 };
 
-#define GPIO_INDEX(bank, gpio) (gpio % bank->width)
 #define GPIO_MOD_CTRL_BIT	BIT(0)
 
 #define BANK_USED(bank) (bank->mod_usage || bank->irq_usage)
@@ -492,14 +491,6 @@ static int omap_gpio_irq_type(struct irq_data *d, unsigned type)
 	if (!BANK_USED(bank))
 		pm_runtime_get_sync(bank->dev);
 
-#ifdef CONFIG_ARCH_OMAP1
-	if (d->irq > IH_MPUIO_BASE) {
-		unsigned gpio = 0;
-		gpio = OMAP_MPUIO(d->irq - IH_MPUIO_BASE);
-		offset = GPIO_INDEX(bank, gpio);
-	}
-#endif
-
 	if (type & ~IRQ_TYPE_SENSE_MASK)
 		return -EINVAL;
 
-- 
1.9.1

Re: [RFT OMAP1 PATCH 8/8] gpio: omap: get rid of GPIO_INDEX() macro

[Javier Martinez Canillas]

On Thu, Mar 19, 2015 at 6:25 PM,  <grygorii.strashko@linaro.org> wrote:
> From: Grygorii Strashko <grygorii.strashko@linaro.org>
>
> Now OMAP GPIO driver prepared for GPIO_INDEX() macro removing.
> Do It ;)
>
> Signed-off-by: Grygorii Strashko <grygorii.strashko@linaro.org>

Acked-by: Javier Martinez Canillas <javier@dowhile0.org>

Re: [PATCH 0/8] gpio: omap: cleanup: get rid of system GPIO <-> GPIO offset converseations

[Tony Lindgren]

* grygorii.strashko@linaro.org <grygorii.strashko@linaro.org> [150319 10:26]:
> From: Grygorii Strashko <grygorii.strashko@linaro.org>
> 
> Now in TI OMAP GPIO driver there are a lot of places where
> System GPIO number calculated and then converted to GPIO offset.
> What is worse is that in many place such conversation performed twice
> or even three times. But actually, we don't need to do that at all, because
> - gpiolib always passes GPIO offset to GPIO controller
> - OMAP GPIO driver converted to use IRQ domain, so
>   struct irq_data->hwirq contains GPIO offset
> 
> Hence, it is safe to convert all GPIO OMAP functions to use GPIO
> offset instead of system GPIO numbers. Also, this allows to remove
> unneeded conversations routines
>  #define GPIO_INDEX(bank, gpio)
>  #define GPIO_BIT(bank, gpio)
>  int omap_irq_to_gpio()

Right on! This is excellent news. I've tested this set on top of -rc4
plus the patch mentioned below, and I'm not seeing regressions on
the platforms I tested with. Wake up to smsc911x ping with off-idle
still works on omap3.

I only briefly tested on omap1 (osk5912), so I'like to wait for
Aaro's ack before this gets merged.

I found one build error, other than that, for the whole series
please feel free to add:

Tested-by: Tony Lindgren <tony@atomide.com>
 
> Tested on dra7-evm.
> 
> Last two patches have to be tested on OMAP1:
> -  gpio: omap: get rid of omap_irq_to_gpio()
> -  gpio: omap: get rid of GPIO_INDEX() macro
> 
> Based on top of Linux 4.0-rc4 plus patch
> '[PATCH 1/2] gpio: omap: irq_shutdown: remove unnecessary call of gpiochip_unlock_as_irq'
> http://www.spinics.net/lists/linux-omap/msg116482.html
> 
> Grygorii Strashko (8):
>   gpio: omap: convert omap_gpio_is_input() to use gpio offset
>   gpio: omap: simplify omap_set_gpio_dataout_x()
>   gpio: omap: convert debounce functions switch to use gpio offset
>   gpio: omap: drop 'gpio' param from omap_gpio_init_irq()
>   gpio: omap: convert gpio irq functions to use GPIO offset
>   gpio: omap: get rid of GPIO_BIT() macro
>   gpio: omap: get rid of omap_irq_to_gpio()
>   gpio: omap: get rid of GPIO_INDEX() macro
> 
>  drivers/gpio/gpio-omap.c | 130 ++++++++++++++++++++---------------------------
>  1 file changed, 55 insertions(+), 75 deletions(-)
> 
> -- 
> 1.9.1
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-omap" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 

Re: [PATCH 0/8] gpio: omap: cleanup: get rid of system GPIO <-> GPIO offset converseations

[Grygorii.Strashko@linaro.org]

On 03/20/2015 01:00 AM, Tony Lindgren wrote:
> * grygorii.strashko@linaro.org <grygorii.strashko@linaro.org> [150319 10:26]:
>> From: Grygorii Strashko <grygorii.strashko@linaro.org>
>>
>> Now in TI OMAP GPIO driver there are a lot of places where
>> System GPIO number calculated and then converted to GPIO offset.
>> What is worse is that in many place such conversation performed twice
>> or even three times. But actually, we don't need to do that at all, because
>> - gpiolib always passes GPIO offset to GPIO controller
>> - OMAP GPIO driver converted to use IRQ domain, so
>>    struct irq_data->hwirq contains GPIO offset
>>
>> Hence, it is safe to convert all GPIO OMAP functions to use GPIO
>> offset instead of system GPIO numbers. Also, this allows to remove
>> unneeded conversations routines
>>   #define GPIO_INDEX(bank, gpio)
>>   #define GPIO_BIT(bank, gpio)
>>   int omap_irq_to_gpio()
> 
> Right on! This is excellent news. I've tested this set on top of -rc4
> plus the patch mentioned below, and I'm not seeing regressions on
> the platforms I tested with. Wake up to smsc911x ping with off-idle
> still works on omap3.
> 
> I only briefly tested on omap1 (osk5912), so I'like to wait for
> Aaro's ack before this gets merged.
> 
> I found one build error, other than that, for the whole series
> please feel free to add:
> 
> Tested-by: Tony Lindgren <tony@atomide.com>

Thanks Tony. 
Yep. I'll wait for news from Aaro then will re-send if ok.

-- 
regards,
-grygorii

Re: [PATCH 0/8] gpio: omap: cleanup: get rid of system GPIO <-> GPIO offset converseations

[Aaro Koskinen]

Hi,

On Fri, Mar 20, 2015 at 05:11:23PM +0200, Grygorii.Strashko@linaro.org wrote:
> On 03/20/2015 01:00 AM, Tony Lindgren wrote:
> > * grygorii.strashko@linaro.org <grygorii.strashko@linaro.org> [150319 10:26]:
> >> From: Grygorii Strashko <grygorii.strashko@linaro.org>
> >> Now in TI OMAP GPIO driver there are a lot of places where
> >> System GPIO number calculated and then converted to GPIO offset.
> >> What is worse is that in many place such conversation performed twice
> >> or even three times. But actually, we don't need to do that at all, because
> >> - gpiolib always passes GPIO offset to GPIO controller
> >> - OMAP GPIO driver converted to use IRQ domain, so
> >>    struct irq_data->hwirq contains GPIO offset
> >>
> >> Hence, it is safe to convert all GPIO OMAP functions to use GPIO
> >> offset instead of system GPIO numbers. Also, this allows to remove
> >> unneeded conversations routines
> >>   #define GPIO_INDEX(bank, gpio)
> >>   #define GPIO_BIT(bank, gpio)
> >>   int omap_irq_to_gpio()
> > 
> > Right on! This is excellent news. I've tested this set on top of -rc4
> > plus the patch mentioned below, and I'm not seeing regressions on
> > the platforms I tested with. Wake up to smsc911x ping with off-idle
> > still works on omap3.
> > 
> > I only briefly tested on omap1 (osk5912), so I'like to wait for
> > Aaro's ack before this gets merged.
> > 
> > I found one build error, other than that, for the whole series
> > please feel free to add:
> > 
> > Tested-by: Tony Lindgren <tony@atomide.com>
> 
> Thanks Tony. 
> Yep. I'll wait for news from Aaro then will re-send if ok.

Works fine on 770 and E3.

Tested-by: Aaro Koskinen <aaro.koskinen@iki.fi>

A.