Re: [PATCH v5 2/4] arm64: dts: Add the device tree entry for the APM X-Gene PCIe MSI node

[Duc Dang <dhdang@apm.com>]

On Tue, Apr 21, 2015 at 8:19 AM, Marc Zyngier <marc.zyngier@arm.com> wrote:
> On 21/04/15 05:04, Duc Dang wrote:
>> There is single MSI block in X-Gene v1 SOC which serves all 5 PCIe ports.
>>
>> Signed-off-by: Duc Dang <dhdang@apm.com>
>> Signed-off-by: Tanmay Inamdar <tinamdar@apm.com>
>> ---
>>  arch/arm64/boot/dts/apm/apm-storm.dtsi | 27 +++++++++++++++++++++++++++
>>  1 file changed, 27 insertions(+)
>>
>> diff --git a/arch/arm64/boot/dts/apm/apm-storm.dtsi b/arch/arm64/boot/dts/apm/apm-storm.dtsi
>> index f1ad9c2..4b719c9 100644
>> --- a/arch/arm64/boot/dts/apm/apm-storm.dtsi
>> +++ b/arch/arm64/boot/dts/apm/apm-storm.dtsi
>> @@ -354,6 +354,28 @@
>>                       };
>>               };
>>
>> +             msi: msi@79000000 {
>> +                     compatible = "apm,xgene1-msi";
>> +                     msi-controller;
>> +                     reg = <0x00 0x79000000 0x0 0x900000>;
>
> I've been repeatedly puzzled by the size of this region. In patch 1, you
> say:
>
> + * X-Gene v1 has 16 groups of MSI termination registers MSInIRx, where
> + * n is group number (0..F), x is index of registers in each group (0..7)
> + * The registers layout is like following:
> + * MSI0IR0                     base_addr
> + * MSI0IR1                     base_addr +  0x10000
> + * ...                         ...
> + * MSI0IR6                     base_addr +  0x60000
> + * MSI0IR7                     base_addr +  0x70000
> + * MSI1IR0                     base_addr +  0x80000
> + * MSI1IR1                     base_addr +  0x90000
> + * ...                         ...
> + * MSI1IR7                     base_addr +  0xF0000
> + * MSI2IR0                     base_addr + 0x100000
> + * ...                         ...
> + * MSIFIR0                     base_addr + 0x780000
> + * MSIFIR1                     base_addr + 0x790000
> + * ...                         ...
> + * MSIFIR7                     base_addr + 0x7F0000
>
> which implies that the size of the region is 0x800000. Or is there
> something hidden in the last 16 64k pages?

The registers listed in the first patch are termination registers.
>From offset 0x800000 to 0x8f0000, the MSI controller provides status
registers for software to check if there is pending MSI interrupt in
each MSI group (in the code we read MSI_INTx registers starting from
0x800000 to check if there is an MSI interrupt pending).
>
> Thanks,
>
>         M.
> --
> Jazz is not dead. It just smells funny...

Regards,
Duc Dang.