| Message ID | 20170612212412.22719-2-enric.balletbo@collabora.com (mailing list archive) |
|---|---|
| State | Not Applicable, archived |
| Headers | show |
* Enric Balletbo i Serra <enric.balletbo@collabora.com> [170612 14:28]: > The interrupt for power button is static data that comes from the > datasheet, there is no reason to need to define this value on every > board so seams reasonable put this information into the common tps65217 > file. I think there's a problem with these patches where we no longer know if the interrupts are wired up on a board. For example, if the USB PHY VBUS interrupt is not connected, the phy driver needs poll the cable state. And on some PMICs some pins can be muxed to GPIO mode, don't remember if that can be done with tps65217. Regards, Tony
Hi Tony, 2017-06-13 9:35 GMT+02:00 Tony Lindgren <tony@atomide.com>: > * Enric Balletbo i Serra <enric.balletbo@collabora.com> [170612 14:28]: >> The interrupt for power button is static data that comes from the >> datasheet, there is no reason to need to define this value on every >> board so seams reasonable put this information into the common tps65217 >> file. > > I think there's a problem with these patches where we no longer know > if the interrupts are wired up on a board. For example, if the USB PHY > VBUS interrupt is not connected, the phy driver needs poll the > cable state. > Hmm, the tps65217 doesn't have a USB PHY like others tps, so not sure how could affect this patch. Did you see this issue in any board that uses tps65217 or any am335x based board which commonly uses this pmic? > And on some PMICs some pins can be muxed to GPIO mode, don't > remember if that can be done with tps65217. > i don't think this is the case of the boards that use tps65217 as the pins can't be muxed to GPIO mode. > Regards, > > Tony Regards, Enric
* Enric Balletbo Serra <eballetbo@gmail.com> [170613 02:12]: > Hi Tony, > > 2017-06-13 9:35 GMT+02:00 Tony Lindgren <tony@atomide.com>: > > * Enric Balletbo i Serra <enric.balletbo@collabora.com> [170612 14:28]: > >> The interrupt for power button is static data that comes from the > >> datasheet, there is no reason to need to define this value on every > >> board so seams reasonable put this information into the common tps65217 > >> file. > > > > I think there's a problem with these patches where we no longer know > > if the interrupts are wired up on a board. For example, if the USB PHY > > VBUS interrupt is not connected, the phy driver needs poll the > > cable state. > > > > Hmm, the tps65217 doesn't have a USB PHY like others tps, so not sure > how could affect this patch. Did you see this issue in any board that > uses tps65217 or any am335x based board which commonly uses this pmic? Oh sorry I misread your patch. I thought you're removing the interrupts from dts, but you're just moving the dts entries to the common dtsi file :) So seems just fine with me. Regards, Tony
* Tony Lindgren <tony@atomide.com> [170613 22:00]: > * Enric Balletbo Serra <eballetbo@gmail.com> [170613 02:12]: > > Hi Tony, > > > > 2017-06-13 9:35 GMT+02:00 Tony Lindgren <tony@atomide.com>: > > > * Enric Balletbo i Serra <enric.balletbo@collabora.com> [170612 14:28]: > > >> The interrupt for power button is static data that comes from the > > >> datasheet, there is no reason to need to define this value on every > > >> board so seams reasonable put this information into the common tps65217 > > >> file. > > > > > > I think there's a problem with these patches where we no longer know > > > if the interrupts are wired up on a board. For example, if the USB PHY > > > VBUS interrupt is not connected, the phy driver needs poll the > > > cable state. > > > > > > > Hmm, the tps65217 doesn't have a USB PHY like others tps, so not sure > > how could affect this patch. Did you see this issue in any board that > > uses tps65217 or any am335x based board which commonly uses this pmic? > > Oh sorry I misread your patch. I thought you're removing > the interrupts from dts, but you're just moving the dts entries > to the common dtsi file :) So seems just fine with me. Applying patches 1 and 2 of this series into omap-for-v4.14/dt thanks. Regards, Tony
diff --git a/arch/arm/boot/dts/am335x-bone-common.dtsi b/arch/arm/boot/dts/am335x-bone-common.dtsi index bfa4258..89daaeb 100644 --- a/arch/arm/boot/dts/am335x-bone-common.dtsi +++ b/arch/arm/boot/dts/am335x-bone-common.dtsi @@ -323,7 +323,6 @@ }; pwrbutton { - interrupts = <2>; status = "okay"; }; diff --git a/arch/arm/boot/dts/am335x-chiliboard.dts b/arch/arm/boot/dts/am335x-chiliboard.dts index 5c25f3a..59431b2 100644 --- a/arch/arm/boot/dts/am335x-chiliboard.dts +++ b/arch/arm/boot/dts/am335x-chiliboard.dts @@ -195,7 +195,6 @@ }; pwrbutton { - interrupts = <2>; status = "okay"; }; }; diff --git a/arch/arm/boot/dts/tps65217.dtsi b/arch/arm/boot/dts/tps65217.dtsi index 1973159..399baaa 100644 --- a/arch/arm/boot/dts/tps65217.dtsi +++ b/arch/arm/boot/dts/tps65217.dtsi @@ -25,6 +25,7 @@ pwrbutton { compatible = "ti,tps65217-pwrbutton"; + interrupts = <2>; status = "disabled"; };
The interrupt for power button is static data that comes from the datasheet, there is no reason to need to define this value on every board so seams reasonable put this information into the common tps65217 file. Signed-off-by: Enric Balletbo i Serra <enric.balletbo@collabora.com> --- arch/arm/boot/dts/am335x-bone-common.dtsi | 1 - arch/arm/boot/dts/am335x-chiliboard.dts | 1 - arch/arm/boot/dts/tps65217.dtsi | 1 + 3 files changed, 1 insertion(+), 2 deletions(-)