| Message ID | 20170526180609.2699-5-uwe@kleine-koenig.org |
|---|---|
| State | New |
| Headers | show |
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 >
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 >>
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>; }; };
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? Best regards Uwe arch/arm/boot/dts/vf-colibri-eval-v3.dtsi | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)