[PATCH v2 1/1] PINCTRL: Warn if direct IRQ GPIO set to output

[PATCH v2 1/1] PINCTRL: Warn if direct IRQ GPIO set to output

[eric.ernst]

From: Eric Ernst <eric.ernst@linux.intel.com>

For Baytrail, you should never set a GPIO set to direct_irq
to output mode.  When direct_irq_en is set for a GPIO, it is
tied directly to an APIC internally, and making the pad output
does not make any sense. Assert a WARN() in the event this happens.

Signed-off-by: Eric Ernst <eric.ernst@linux.intel.com>
---
 drivers/pinctrl/pinctrl-baytrail.c |   19 +++++++++++++++----
 1 file changed, 15 insertions(+), 4 deletions(-)

diff --git a/drivers/pinctrl/pinctrl-baytrail.c b/drivers/pinctrl/pinctrl-baytrail.c
index e59983423991..fde3767df254 100644
--- a/drivers/pinctrl/pinctrl-baytrail.c
+++ b/drivers/pinctrl/pinctrl-baytrail.c
@@ -47,6 +47,7 @@
 #define BYT_TRIG_POS		BIT(25)
 #define BYT_TRIG_LVL		BIT(24)
 #define BYT_PIN_MUX		0x07
+#define BYT_DIRECTIRQ		BIT(27)
 
 /* BYT_VAL_REG register bits */
 #define BYT_INPUT_EN		BIT(2)  /* 0: input enabled (active low)*/
@@ -256,19 +257,29 @@ static int byt_gpio_direction_output(struct gpio_chip *chip,
 				     unsigned gpio, int value)
 {
 	struct byt_gpio *vg = to_byt_gpio(chip);
-	void __iomem *reg = byt_gpio_reg(chip, gpio, BYT_VAL_REG);
+	void __iomem *conf_reg = byt_gpio_reg(chip, gpio, BYT_CONF0_REG);
+	void __iomem *value_reg = byt_gpio_reg(chip, gpio, BYT_VAL_REG);
 	unsigned long flags;
 	u32 reg_val;
 
 	spin_lock_irqsave(&vg->lock, flags);
 
-	reg_val = readl(reg) | BYT_DIR_MASK;
+	/*
+	 * Before making any direction modifications, do a check if gpio
+	 * is set for direct IRQ.  On baytrail, setting GPIO to output does
+	 * not make sense, so let's at least warn the caller before they shoot
+	 * themselves in the foot.
+	 */
+	WARN((readl(conf_reg) & BYT_DIRECTIRQ),
+		"Potential Error: Setting GPIO with direct_irq_en to output");
+
+	reg_val = readl(value_reg) | BYT_DIR_MASK;
 	reg_val &= ~BYT_OUTPUT_EN;
 
 	if (value)
-		writel(reg_val | BYT_LEVEL, reg);
+		writel(reg_val | BYT_LEVEL, value_reg);
 	else
-		writel(reg_val & ~BYT_LEVEL, reg);
+		writel(reg_val & ~BYT_LEVEL, value_reg);
 
 	spin_unlock_irqrestore(&vg->lock, flags);
 
-- 
1.7.9.5

Re: [PATCH v2 1/1] PINCTRL: Warn if direct IRQ GPIO set to output

[Mika Westerberg]

On Mon, Jun 02, 2014 at 02:12:07PM -0700, eric.ernst@linux.intel.com wrote:
> From: Eric Ernst <eric.ernst@linux.intel.com>
> 
> For Baytrail, you should never set a GPIO set to direct_irq
> to output mode.  When direct_irq_en is set for a GPIO, it is
> tied directly to an APIC internally, and making the pad output
> does not make any sense. Assert a WARN() in the event this happens.

Subject should probably be:

pinctrl: baytrail: Warn if direct IRQ GPIO is set to output

> Signed-off-by: Eric Ernst <eric.ernst@linux.intel.com>
> ---
>  drivers/pinctrl/pinctrl-baytrail.c |   19 +++++++++++++++----
>  1 file changed, 15 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/pinctrl/pinctrl-baytrail.c b/drivers/pinctrl/pinctrl-baytrail.c
> index e59983423991..fde3767df254 100644
> --- a/drivers/pinctrl/pinctrl-baytrail.c
> +++ b/drivers/pinctrl/pinctrl-baytrail.c
> @@ -47,6 +47,7 @@
>  #define BYT_TRIG_POS		BIT(25)
>  #define BYT_TRIG_LVL		BIT(24)
>  #define BYT_PIN_MUX		0x07
> +#define BYT_DIRECTIRQ		BIT(27)

Please move this definition to be first, like:

/* BYT_CONF0_REG register bits */
#define BYT_DIRECT_IRQ_EN	BIT(27)
#define BYT_TRIG_NEG		BIT(26)
#define BYT_TRIG_POS		BIT(25)

and I would call it BYT_DIRECT_IRQ_EN since it's name in datasheet is
"direct_irq_en".

>  
>  /* BYT_VAL_REG register bits */
>  #define BYT_INPUT_EN		BIT(2)  /* 0: input enabled (active low)*/
> @@ -256,19 +257,29 @@ static int byt_gpio_direction_output(struct gpio_chip *chip,
>  				     unsigned gpio, int value)
>  {
>  	struct byt_gpio *vg = to_byt_gpio(chip);
> -	void __iomem *reg = byt_gpio_reg(chip, gpio, BYT_VAL_REG);
> +	void __iomem *conf_reg = byt_gpio_reg(chip, gpio, BYT_CONF0_REG);
> +	void __iomem *value_reg = byt_gpio_reg(chip, gpio, BYT_VAL_REG);

Not sure if it is necessary to rename reg -> value_reg. It just makes
the patch bigger than it has to be since you also need to rename stuff
below.

Otherwise looks good.

>  	unsigned long flags;
>  	u32 reg_val;
>  
>  	spin_lock_irqsave(&vg->lock, flags);
>  
> -	reg_val = readl(reg) | BYT_DIR_MASK;
> +	/*
> +	 * Before making any direction modifications, do a check if gpio
> +	 * is set for direct IRQ.  On baytrail, setting GPIO to output does
> +	 * not make sense, so let's at least warn the caller before they shoot
> +	 * themselves in the foot.
> +	 */
> +	WARN((readl(conf_reg) & BYT_DIRECTIRQ),
> +		"Potential Error: Setting GPIO with direct_irq_en to output");
> +
> +	reg_val = readl(value_reg) | BYT_DIR_MASK;
>  	reg_val &= ~BYT_OUTPUT_EN;
>  
>  	if (value)
> -		writel(reg_val | BYT_LEVEL, reg);
> +		writel(reg_val | BYT_LEVEL, value_reg);
>  	else
> -		writel(reg_val & ~BYT_LEVEL, reg);
> +		writel(reg_val & ~BYT_LEVEL, value_reg);
>  
>  	spin_unlock_irqrestore(&vg->lock, flags);
>  
> -- 
> 1.7.9.5

Re: [PATCH v2 1/1] PINCTRL: Warn if direct IRQ GPIO set to output

[Mika Westerberg]

On Mon, Jun 02, 2014 at 02:12:07PM -0700, eric.ernst@linux.intel.com wrote:
> +	WARN((readl(conf_reg) & BYT_DIRECTIRQ),
> +		"Potential Error: Setting GPIO with direct_irq_en to output");

Also you don't need parentheses here.

[PATCH v3 1/1] pinctrl: baytrail: Warn if direct IRQ GPIO set to output

[eric.ernst]

From: Eric Ernst <eric.ernst@linux.intel.com>

For Baytrail, you should never set a GPIO set to direct_irq
to output mode.  When direct_irq_en is set for a GPIO, it is
tied directly to an APIC internally, and making the pad output
does not make any sense. Assert a WARN() in the event this happens.

Signed-off-by: Eric Ernst <eric.ernst@linux.intel.com>
---
 drivers/pinctrl/pinctrl-baytrail.c |   11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/drivers/pinctrl/pinctrl-baytrail.c b/drivers/pinctrl/pinctrl-baytrail.c
index e59983423991..7f0a2bac7c82 100644
--- a/drivers/pinctrl/pinctrl-baytrail.c
+++ b/drivers/pinctrl/pinctrl-baytrail.c
@@ -43,6 +43,7 @@
 #define BYT_INT_STAT_REG	0x800
 
 /* BYT_CONF0_REG register bits */
+#define BYT_DIRECT_IRQ_EN	BIT(27)
 #define BYT_TRIG_NEG		BIT(26)
 #define BYT_TRIG_POS		BIT(25)
 #define BYT_TRIG_LVL		BIT(24)
@@ -256,12 +257,22 @@ static int byt_gpio_direction_output(struct gpio_chip *chip,
 				     unsigned gpio, int value)
 {
 	struct byt_gpio *vg = to_byt_gpio(chip);
+	void __iomem *conf_reg = byt_gpio_reg(chip, gpio, BYT_CONF0_REG);
 	void __iomem *reg = byt_gpio_reg(chip, gpio, BYT_VAL_REG);
 	unsigned long flags;
 	u32 reg_val;
 
 	spin_lock_irqsave(&vg->lock, flags);
 
+	/*
+	 * Before making any direction modifications, do a check if gpio
+	 * is set for direct IRQ.  On baytrail, setting GPIO to output does
+	 * not make sense, so let's at least warn the caller before they shoot
+	 * themselves in the foot.
+	 */
+	WARN(readl(conf_reg) & BYT_DIRECT_IRQ_EN,
+		"Potential Error: Setting GPIO with direct_irq_en to output");
+
 	reg_val = readl(reg) | BYT_DIR_MASK;
 	reg_val &= ~BYT_OUTPUT_EN;
 
-- 
1.7.9.5

Re: [PATCH v3 1/1] pinctrl: baytrail: Warn if direct IRQ GPIO set to output

[Mika Westerberg]

On Tue, Jun 03, 2014 at 03:25:16PM -0700, eric.ernst@linux.intel.com wrote:
> From: Eric Ernst <eric.ernst@linux.intel.com>
> 
> For Baytrail, you should never set a GPIO set to direct_irq
> to output mode.  When direct_irq_en is set for a GPIO, it is
> tied directly to an APIC internally, and making the pad output
> does not make any sense. Assert a WARN() in the event this happens.
> 
> Signed-off-by: Eric Ernst <eric.ernst@linux.intel.com>

Looks good now thanks,

Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com>

Re: [PATCH v3 1/1] pinctrl: baytrail: Warn if direct IRQ GPIO set to output

[Linus Walleij]

On Wed, Jun 4, 2014 at 12:25 AM,  <eric.ernst@linux.intel.com> wrote:

> From: Eric Ernst <eric.ernst@linux.intel.com>
>
> For Baytrail, you should never set a GPIO set to direct_irq
> to output mode.  When direct_irq_en is set for a GPIO, it is
> tied directly to an APIC internally, and making the pad output
> does not make any sense. Assert a WARN() in the event this happens.
>
> Signed-off-by: Eric Ernst <eric.ernst@linux.intel.com>

This does not apply to my devel branch or Torvald's HEAD.
Please rebase and repost, and include Mika's ACK.

Yours,
Linus Walleij

[PATCH v4 1/1] pinctrl: baytrail: Warn if direct IRQ GPIO set to output

[eric.ernst]

From: Eric Ernst <eric.ernst@linux.intel.com>

For Baytrail, you should never set a GPIO set to direct_irq
to output mode.  When direct_irq_en is set for a GPIO, it is
tied directly to an APIC internally, and making the pad output
does not make any sense. Assert a WARN() in the event this happens.

Signed-off-by: Eric Ernst <eric.ernst@linux.intel.com>
Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com>
---
 drivers/pinctrl/pinctrl-baytrail.c |   10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/pinctrl/pinctrl-baytrail.c b/drivers/pinctrl/pinctrl-baytrail.c
index 975572e2f260..c34add934216 100644
--- a/drivers/pinctrl/pinctrl-baytrail.c
+++ b/drivers/pinctrl/pinctrl-baytrail.c
@@ -303,12 +303,22 @@ static int byt_gpio_direction_output(struct gpio_chip *chip,
 				     unsigned gpio, int value)
 {
 	struct byt_gpio *vg = to_byt_gpio(chip);
+	void __iomem *conf_reg = byt_gpio_reg(chip, gpio, BYT_CONF0_REG);
 	void __iomem *reg = byt_gpio_reg(chip, gpio, BYT_VAL_REG);
 	unsigned long flags;
 	u32 reg_val;
 
 	spin_lock_irqsave(&vg->lock, flags);
 
+	/*
+	 * Before making any direction modifications, do a check if gpio
+	 * is set for direct IRQ.  On baytrail, setting GPIO to output does
+	 * not make sense, so let's at least warn the caller before they shoot
+	 * themselves in the foot.
+	 */
+	WARN(readl(conf_reg) & BYT_DIRECT_IRQ_EN,
+		"Potential Error: Setting GPIO with direct_irq_en to output");
+
 	reg_val = readl(reg) | BYT_DIR_MASK;
 	reg_val &= ~BYT_OUTPUT_EN;
 
-- 
1.7.9.5

Re: [PATCH v4 1/1] pinctrl: baytrail: Warn if direct IRQ GPIO set to output

[Linus Walleij]

On Thu, Jun 12, 2014 at 8:06 PM,  <eric.ernst@linux.intel.com> wrote:

> From: Eric Ernst <eric.ernst@linux.intel.com>
>
> For Baytrail, you should never set a GPIO set to direct_irq
> to output mode.  When direct_irq_en is set for a GPIO, it is
> tied directly to an APIC internally, and making the pad output
> does not make any sense. Assert a WARN() in the event this happens.
>
> Signed-off-by: Eric Ernst <eric.ernst@linux.intel.com>
> Acked-by: Mika Westerberg <mika.westerberg@linux.intel.com>

Patch applied.

Sorry for the delay.

Yours,
Linus Walleij

Re: [PATCH v4 1/1] pinctrl: baytrail: Warn if direct IRQ GPIO set to output

[Linus Walleij]

On Thu, Jun 12, 2014 at 8:06 PM,  <eric.ernst@linux.intel.com> wrote:

> +       WARN(readl(conf_reg) & BYT_DIRECT_IRQ_EN,
> +               "Potential Error: Setting GPIO with direct_irq_en to output");

Hm in the v4 version you somehow managed to drop the definition
of BYT_DIRECT_IRQ_EN...

I patched it back in. Let's hope it compiles now.

Yours,
Linus Walleij

Re: [PATCH v4 1/1] pinctrl: baytrail: Warn if direct IRQ GPIO set to output

[eric ernst]

Mea culpa - thanks Linus.  I saw a note from kbuild test
robot indicating an issue, but given your response, I am
assuming it was from your first build attempt.  Thanks.

On 14-07-07 02:14 PM, Linus Walleij wrote:
> On Thu, Jun 12, 2014 at 8:06 PM,  <eric.ernst@linux.intel.com> wrote:
>
>> +       WARN(readl(conf_reg) & BYT_DIRECT_IRQ_EN,
>> +               "Potential Error: Setting GPIO with direct_irq_en to output");
> Hm in the v4 version you somehow managed to drop the definition
> of BYT_DIRECT_IRQ_EN...
>
> I patched it back in. Let's hope it compiles now.
>
> Yours,
> Linus Walleij