On 7/28/19 12:03 PM, Marc Zyngier wrote: > On Sat, 27 Jul 2019 18:53:15 +0100, > Martin Blumenstingl wrote: >> >> EBU provides an interrupt line for the PCI_INTA interrupt. Route >> easy50712's PCI interrupt to EBU so the interrupt line is configured >> correctly (using IRQ_TYPE_LEVEL_LOW, this was previously hardcoded in >> the PCI driver) and ACKed properly. >> >> Signed-off-by: Martin Blumenstingl Acked-by: Hauke Mehrtens >> --- >> arch/mips/boot/dts/lantiq/easy50712.dts | 4 +++- >> 1 file changed, 3 insertions(+), 1 deletion(-) >> >> diff --git a/arch/mips/boot/dts/lantiq/easy50712.dts b/arch/mips/boot/dts/lantiq/easy50712.dts >> index 1ce20b7d05cb..33c26b93cfc9 100644 >> --- a/arch/mips/boot/dts/lantiq/easy50712.dts >> +++ b/arch/mips/boot/dts/lantiq/easy50712.dts >> @@ -1,6 +1,8 @@ >> // SPDX-License-Identifier: GPL-2.0 >> /dts-v1/; >> >> +#include >> + >> /include/ "danube.dtsi" >> >> / { >> @@ -105,7 +107,7 @@ >> lantiq,bus-clock = <33333333>; >> interrupt-map-mask = <0xf800 0x0 0x0 0x7>; >> interrupt-map = < >> - 0x7000 0 0 1 &icu0 29 1 // slot 14, irq 29 >> + 0x7000 0 0 1 &ebu0 0 IRQ_TYPE_LEVEL_LOW // slot 14 >> >; >> gpios-reset = <&gpio 21 0>; >> req-mask = <0x1>; /* GNT1 */ >> -- >> 2.22.0 >> > > Are you OK with breaking compatibility between kernel and DT? It > usually isn't very nice for users... I am fine with such changes. I am not aware of any board using this SoC which ships the kernel and the device tree file as different binaries, it is always either attached or patched into the kernel and never in the boot loader. Hauke