| Message ID | 20170526180609.2699-1-uwe@kleine-koenig.org |
|---|---|
| State | Superseded, archived |
| Headers | show |
> +- interrupt-controller: Marks the device node as a GPIO controller. Interrupt controller, not GPIO controller. > +- #interrupt-cells: Should be two. The first cell is the pin number and the > + second cell is used to specify irq type flags: > + 1 = low-to-high edge triggered. > + 2 = high-to-low edge triggered. > + 4 = active high level-sensitive. > + 8 = active low level-sensitive. Maybe just reference interrupts.txt? Andrew -- To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On 05/26/2017 08:06 PM, Uwe Kleine-König wrote: > According to the binding documentation and the source code the atmel-gpio > controller takes IRQ_TYPE_* as its flags values, not GPIO_ACTIVE_*. > > This patch uses the right variable type which yields the same result > when compiled. Note that this might be wrong and actually > IRQ_TYPE_LEVEL_LOW is intended by the dt author. > > Signed-off-by: Uwe Kleine-König <uwe@kleine-koenig.org> > --- > Hello, > > can somebody with the hardware or it's documentation please check which > flag is the right one? It's correct, I tested the CAN, so: Acked-by: Marek Vasut <marex@denx.de> > Best regards > Uwe > > arch/arm/boot/dts/at91-sama5d4_ma5d4.dtsi | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) > > diff --git a/arch/arm/boot/dts/at91-sama5d4_ma5d4.dtsi b/arch/arm/boot/dts/at91-sama5d4_ma5d4.dtsi > index b5a5a91bc2ef..b813fdfa2842 100644 > --- a/arch/arm/boot/dts/at91-sama5d4_ma5d4.dtsi > +++ b/arch/arm/boot/dts/at91-sama5d4_ma5d4.dtsi > @@ -75,7 +75,7 @@ > reg = <0>; > clocks = <&clk20m>; > interrupt-parent = <&pioE>; > - interrupts = <6 GPIO_ACTIVE_LOW>; > + interrupts = <6 IRQ_TYPE_EDGE_RISING>; > spi-max-frequency = <10000000>; > }; > > @@ -84,7 +84,7 @@ > reg = <1>; > clocks = <&clk20m>; > interrupt-parent = <&pioE>; > - interrupts = <7 GPIO_ACTIVE_LOW>; > + interrupts = <7 IRQ_TYPE_EDGE_RISING>; > spi-max-frequency = <10000000>; > }; > }; >
On 26/05/2017 at 21:17:37 +0200, Marek Vasut wrote: > On 05/26/2017 08:06 PM, Uwe Kleine-König wrote: > > According to the binding documentation and the source code the atmel-gpio > > controller takes IRQ_TYPE_* as its flags values, not GPIO_ACTIVE_*. > > > > This patch uses the right variable type which yields the same result > > when compiled. Note that this might be wrong and actually > > IRQ_TYPE_LEVEL_LOW is intended by the dt author. > > > > Signed-off-by: Uwe Kleine-König <uwe@kleine-koenig.org> > > --- > > Hello, > > > > can somebody with the hardware or it's documentation please check which > > flag is the right one? > > It's correct, I tested the CAN, so: > > Acked-by: Marek Vasut <marex@denx.de> > > > Best regards > > Uwe > > > > arch/arm/boot/dts/at91-sama5d4_ma5d4.dtsi | 4 ++-- > > 1 file changed, 2 insertions(+), 2 deletions(-) > > Applied, thanks.
* Uwe Kleine-König <uwe@kleine-koenig.org> [170526 11:09]: > According to the binding documentation and the source code the omap-gpio > controller takes IRQ_TYPE_* as its flags values, not GPIO_ACTIVE_*. > > This patch uses the right variable type which yields the same result > when compiled. Note that this might be wrong and actually > IRQ_TYPE_LEVEL_LOW is intended by the dt author. > > Signed-off-by: Uwe Kleine-König <uwe@kleine-koenig.org> > --- > Hello, > > can somebody with the hardware or it's documentation please check which > flag is the right one? I'll wait on this one until we have somebody test it. Regards, Tony -- To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Hi Uwe, Tony, On Wed, May 31, 2017 at 7:19 PM, Tony Lindgren <tony@atomide.com> wrote: > * Uwe Kleine-König <uwe@kleine-koenig.org> [170526 11:09]: >> According to the binding documentation and the source code the omap-gpio >> controller takes IRQ_TYPE_* as its flags values, not GPIO_ACTIVE_*. >> >> This patch uses the right variable type which yields the same result >> when compiled. Note that this might be wrong and actually >> IRQ_TYPE_LEVEL_LOW is intended by the dt author. >> >> Signed-off-by: Uwe Kleine-König <uwe@kleine-koenig.org> >> --- >> Hello, >> >> can somebody with the hardware or it's documentation please check which >> flag is the right one? > > I'll wait on this one until we have somebody test it. I'll look at it this week. Yegor -- To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Fri, May 26, 2017 at 08:06:09PM +0200, Uwe Kleine-König wrote: > According to the binding documentation and the source code the > vf610-gpio controller takes IRQ_TYPE_* as its flags values, not > GPIO_ACTIVE_*. > > This patch uses the right variable type which yields the same result > when compiled. Note that this might be wrong and actually > IRQ_TYPE_LEVEL_LOW is intended by the dt author. > > Signed-off-by: Uwe Kleine-König <uwe@kleine-koenig.org> > --- > Hello, > > can somebody with the hardware or it's documentation please check which > flag is the right one? @Stefan, can you help to confirm? Shawn > > Best regards > Uwe > > arch/arm/boot/dts/vf-colibri-eval-v3.dtsi | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/arch/arm/boot/dts/vf-colibri-eval-v3.dtsi b/arch/arm/boot/dts/vf-colibri-eval-v3.dtsi > index 091b738041a0..3e0c84d79c43 100644 > --- a/arch/arm/boot/dts/vf-colibri-eval-v3.dtsi > +++ b/arch/arm/boot/dts/vf-colibri-eval-v3.dtsi > @@ -117,7 +117,7 @@ > clocks = <&clk16m>; > spi-max-frequency = <10000000>; > interrupt-parent = <&gpio1>; > - interrupts = <11 GPIO_ACTIVE_LOW>; > + interrupts = <11 IRQ_TYPE_EDGE_RISING>; > }; > }; > > -- > 2.11.0 > -- To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On 2017-06-03 20:49, Shawn Guo wrote: > On Fri, May 26, 2017 at 08:06:09PM +0200, Uwe Kleine-König wrote: >> According to the binding documentation and the source code the >> vf610-gpio controller takes IRQ_TYPE_* as its flags values, not >> GPIO_ACTIVE_*. >> >> This patch uses the right variable type which yields the same result >> when compiled. Note that this might be wrong and actually >> IRQ_TYPE_LEVEL_LOW is intended by the dt author. >> >> Signed-off-by: Uwe Kleine-König <uwe@kleine-koenig.org> >> --- >> Hello, >> >> can somebody with the hardware or it's documentation please check which >> flag is the right one? > > @Stefan, can you help to confirm? > Thanks for spotting! According to the data sheet it is a low active signal, so I guess IRQ_TYPE_LEVEL_LOW is correct. But the driver actually explicitly requests IRQF_TRIGGER_FALLING, that has been the case since its inception. IMHO LEVEL_LOW should be more rigid since it helps for missed interrupt edges... -- Stefan >> >> Best regards >> Uwe >> >> arch/arm/boot/dts/vf-colibri-eval-v3.dtsi | 2 +- >> 1 file changed, 1 insertion(+), 1 deletion(-) >> >> diff --git a/arch/arm/boot/dts/vf-colibri-eval-v3.dtsi b/arch/arm/boot/dts/vf-colibri-eval-v3.dtsi >> index 091b738041a0..3e0c84d79c43 100644 >> --- a/arch/arm/boot/dts/vf-colibri-eval-v3.dtsi >> +++ b/arch/arm/boot/dts/vf-colibri-eval-v3.dtsi >> @@ -117,7 +117,7 @@ >> clocks = <&clk16m>; >> spi-max-frequency = <10000000>; >> interrupt-parent = <&gpio1>; >> - interrupts = <11 GPIO_ACTIVE_LOW>; >> + interrupts = <11 IRQ_TYPE_EDGE_RISING>; >> }; >> }; >> >> -- >> 2.11.0 >> -- To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
* Yegor Yefremov <yegorslists@googlemail.com> [170606 07:37]: > On Wed, May 31, 2017 at 7:22 PM, Yegor Yefremov > <yegorslists@googlemail.com> wrote: > > Hi Uwe, Tony, > > > > On Wed, May 31, 2017 at 7:19 PM, Tony Lindgren <tony@atomide.com> wrote: > >> * Uwe Kleine-König <uwe@kleine-koenig.org> [170526 11:09]: > >>> According to the binding documentation and the source code the omap-gpio > >>> controller takes IRQ_TYPE_* as its flags values, not GPIO_ACTIVE_*. > >>> > >>> This patch uses the right variable type which yields the same result > >>> when compiled. Note that this might be wrong and actually > >>> IRQ_TYPE_LEVEL_LOW is intended by the dt author. > >>> > >>> Signed-off-by: Uwe Kleine-König <uwe@kleine-koenig.org> > >>> --- > >>> Hello, > >>> > >>> can somebody with the hardware or it's documentation please check which > >>> flag is the right one? > >> > >> I'll wait on this one until we have somebody test it. > > > > I'll look at it this week. > > This is what works for me (at least it doesn't produce "irq 88: nobody > cared (try booting with the "irqpoll" option)"). > > As for tca6416 it is working with all possible settings (from > IRQ_TYPE_EDGE_RISING till IRQ_TYPE_LEVEL_LOW). "cat /proc/interrupts" > always shows Level type: > > 47: 5 44e07000.gpio 20 Level 1-0020 OK, can you please send a proper patch with description and Signed-off-by? That is unless Uwe cares to update his patch. Otherwise add Reported-by for Uwe :) Regards, Tony -- To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/Documentation/devicetree/bindings/gpio/gpio_atmel.txt b/Documentation/devicetree/bindings/gpio/gpio_atmel.txt index 85f8c0d084fa..b53fe3bce270 100644 --- a/Documentation/devicetree/bindings/gpio/gpio_atmel.txt +++ b/Documentation/devicetree/bindings/gpio/gpio_atmel.txt @@ -5,9 +5,16 @@ Required properties: - reg: Should contain GPIO controller registers location and length - interrupts: Should be the port interrupt shared by all the pins. - #gpio-cells: Should be two. The first cell is the pin number and - the second cell is used to specify optional parameters (currently - unused). + the second cell is used to specify optional parameters to declare if the GPIO + is active high or low. See gpio.txt. - gpio-controller: Marks the device node as a GPIO controller. +- interrupt-controller: Marks the device node as a GPIO controller. +- #interrupt-cells: Should be two. The first cell is the pin number and the + second cell is used to specify irq type flags: + 1 = low-to-high edge triggered. + 2 = high-to-low edge triggered. + 4 = active high level-sensitive. + 8 = active low level-sensitive. optional properties: - #gpio-lines: Number of gpio if absent 32. @@ -21,5 +28,7 @@ Example: #gpio-cells = <2>; gpio-controller; #gpio-lines = <19>; + interrupt-controller; + #interrupt-cells = <2>; };
The second cell in a gpio reference is used to pass GPIO_ACTIVE_LOW or GPIO_ACTIVE_HIGH. The gpio device can also be used as irq controller and a reference can contain the IRQ_TYPE_* values in the second cell. Signed-off-by: Uwe Kleine-König <uwe@kleine-koenig.org> --- Documentation/devicetree/bindings/gpio/gpio_atmel.txt | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-)