On 13 December 2022 20:18:55 EET, Eric Chanudet <echanude@xxxxxxxxxx> wrote: >On Tue, Dec 13, 2022 at 04:18:00AM +0300, Dmitry Baryshkov wrote: >> >diff --git a/arch/arm64/boot/dts/qcom/pm8450a.dtsi b/arch/arm64/boot/dts/qcom/pm8450a.dtsi >> >index 34fc72896761..af761dbfbc66 100644 >> >--- a/arch/arm64/boot/dts/qcom/pm8450a.dtsi >> >+++ b/arch/arm64/boot/dts/qcom/pm8450a.dtsi >> >@@ -13,6 +13,14 @@ pm8450a: pmic@0 { >> > #address-cells = <1>; >> > #size-cells = <0>; >> > >> >+ rtc@6000 { >> >+ compatible = "qcom,pm8941-rtc"; >> >+ reg = <0x6000>; >> >+ reg-names = "rtc", "alarm"; >> >+ interrupts = <0x0 0x61 0x1 IRQ_TYPE_NONE>; >> >> >> 0x60? > >Checking downstream it uses 0x61 for the ppid. This is also the case >for all other descriptions of that RTC I could find. On the other hand, >that does not describe the "alarm" register bank at 0x6100. >Should it be added, if anything to match reg-names? Definitely yes. I think otherwise it breaks the idea of reg-names. > >I tried a quick test of the alarm on sa8540p-ride: >$ echo $(date '+%s' -d '+ 10 seconds') > /sys/class/rtc/rtc0/wakealarm >It logged the interrupt: >172: 3 0 0 0 0 0 0 0 pmic_arb 101777441 Edge pm8xxx_rtc_alarm > -- With best wishes Dmitry