Hi Eddie,
2014-12-11 13:47 GMT+01:00 Eddie Huang <eddie.huang@xxxxxxxxxxxx>:
> On Wed, 2014-12-10 at 15:50 +0100, Matthias Brugger wrote:
>> 2014-12-10 15:27 GMT+01:00 Yingjoe Chen <yingjoe.chen@xxxxxxxxxxxx>:
>> >
>> > Hi,
>> >
>> > On Wed, 2014-12-10 at 18:50 +0800, Eddie Huang wrote:
>> > <...>
>> >> diff --git a/arch/arm64/boot/dts/mt8173-evb.dts b/arch/arm64/boot/dts/mt8173-evb.dts
>> >> new file mode 100644
>> >> index 0000000..adf26dd
>> >> --- /dev/null
>> >> +++ b/arch/arm64/boot/dts/mt8173-evb.dts
>> > <...>
>> >> + timer {
>> >> + compatible = "arm,armv8-timer";
>> >> + interrupt-parent = <&gic>;
>> >> + interrupts = <1 13 0x8>,
>> >> + <1 14 0x8>,
>> >> + <1 11 0x8>,
>> >> + <1 10 0x8>;
>> >> + clock-frequency = <13000000>;
>> >
>> > I believe our firmware doesn't need this line. Please remove it.
>>
>> The point here would be to know if you need to enable a special timer
>> from the mtk-timer block to get the arch timer working.
>> In any case, you will need some sort of timer. This dts does not
>> describe the mtk-timer (may in the mt8173 it does not exist) but
>> defines the clocks clk26m and clk32k. So if you don't use the
>> mtk-timer, please remove the clocks as there isn't a block using them.
>>
>
> MT8173 has two timer set: CPUGPT and APBGPT, and use CPUGPT to enable
> arch_timer. Previous series only have APBGPT. MT8173 still need enable
> CPUGPT to get arch timer working, we put this in loader, and transparent
> to kernel. So I will remove clk26m and clk32k in next version.
Ok, so if this is done in the bootloader, there shouldn't be a problem
then. Perfect.
Just one more comment on the dts nodes. Can you use the makros defined
for the interrupt type
e.g "GIC_PPI 14 IRQ_TYPE_LEVEL_LOW" instead of "1 14 0x8". This makes
the dts easier to read.
Thanks,
Matthias
>
--
motzblog.wordpress.com
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
