arm64: dts: ls1028a: add interrupt to Root Complex Event Collector
diff mbox series

Message ID 20210209005259.29725-1-michael@walle.cc
State Accepted
Commit dfee46f17b4a5e84a8a0a4eb77b083370105862e
Headers show
Series
  • arm64: dts: ls1028a: add interrupt to Root Complex Event Collector
Related show

Commit Message

Michael Walle Feb. 9, 2021, 12:52 a.m. UTC
The legacy interrupt INT_A is hardwired to the event collector. RCEC is
bascially supported starting with v5.11. Having a correct interrupt, will
make RCEC at least probe correctly.

There are still issues with how RCEC is implemented in the RCiEP on the
LS1028A. RCEC will report an error, but it cannot find the correct
subdevice.

Signed-off-by: Michael Walle <michael@walle.cc>
---
 arch/arm64/boot/dts/freescale/fsl-ls1028a.dtsi | 6 ++++++
 1 file changed, 6 insertions(+)

Comments

Vladimir Oltean Feb. 10, 2021, 3:40 p.m. UTC | #1
On Tue, Feb 09, 2021 at 01:52:59AM +0100, Michael Walle wrote:
> The legacy interrupt INT_A is hardwired to the event collector. RCEC is
> bascially supported starting with v5.11. Having a correct interrupt, will
> make RCEC at least probe correctly.
> 
> There are still issues with how RCEC is implemented in the RCiEP on the
> LS1028A. RCEC will report an error, but it cannot find the correct
> subdevice.
> 
> Signed-off-by: Michael Walle <michael@walle.cc>
> ---

Reviewed-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Tested-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Michael Walle March 3, 2021, 8:47 p.m. UTC | #2
Hi,

Am 2021-02-09 01:52, schrieb Michael Walle:
> The legacy interrupt INT_A is hardwired to the event collector. RCEC is
> bascially supported starting with v5.11. Having a correct interrupt, 
> will
> make RCEC at least probe correctly.
> 
> There are still issues with how RCEC is implemented in the RCiEP on the
> LS1028A. RCEC will report an error, but it cannot find the correct
> subdevice.
> 
> Signed-off-by: Michael Walle <michael@walle.cc>

Any news on this?

-michael
Shawn Guo March 4, 2021, 11:37 a.m. UTC | #3
On Tue, Feb 09, 2021 at 01:52:59AM +0100, Michael Walle wrote:
> The legacy interrupt INT_A is hardwired to the event collector. RCEC is
> bascially supported starting with v5.11. Having a correct interrupt, will
> make RCEC at least probe correctly.
> 
> There are still issues with how RCEC is implemented in the RCiEP on the
> LS1028A. RCEC will report an error, but it cannot find the correct
> subdevice.
> 
> Signed-off-by: Michael Walle <michael@walle.cc>
> ---
>  arch/arm64/boot/dts/freescale/fsl-ls1028a.dtsi | 6 ++++++
>  1 file changed, 6 insertions(+)
> 
> diff --git a/arch/arm64/boot/dts/freescale/fsl-ls1028a.dtsi b/arch/arm64/boot/dts/freescale/fsl-ls1028a.dtsi
> index 262fbad8f0ec..c1f2f402ad53 100644
> --- a/arch/arm64/boot/dts/freescale/fsl-ls1028a.dtsi
> +++ b/arch/arm64/boot/dts/freescale/fsl-ls1028a.dtsi
> @@ -1114,6 +1114,12 @@
>  					full-duplex;
>  				};
>  			};
> +
> +			rcec@1f,0 {

Just curious how unit-address comes to '1f,0'?

Shawn

> +				reg = <0x00f800 0 0 0 0>;
> +				/* IEP INT_A */
> +				interrupts = <GIC_SPI 94 IRQ_TYPE_LEVEL_HIGH>;
> +			};
>  		};
>  
>  		rcpm: power-controller@1e34040 {
> -- 
> 2.20.1
>
Michael Walle March 4, 2021, 11:59 a.m. UTC | #4
Am 2021-03-04 12:37, schrieb Shawn Guo:
> On Tue, Feb 09, 2021 at 01:52:59AM +0100, Michael Walle wrote:
>> The legacy interrupt INT_A is hardwired to the event collector. RCEC 
>> is
>> bascially supported starting with v5.11. Having a correct interrupt, 
>> will
>> make RCEC at least probe correctly.
>> 
>> There are still issues with how RCEC is implemented in the RCiEP on 
>> the
>> LS1028A. RCEC will report an error, but it cannot find the correct
>> subdevice.
>> 
>> Signed-off-by: Michael Walle <michael@walle.cc>
>> ---
>>  arch/arm64/boot/dts/freescale/fsl-ls1028a.dtsi | 6 ++++++
>>  1 file changed, 6 insertions(+)
>> 
>> diff --git a/arch/arm64/boot/dts/freescale/fsl-ls1028a.dtsi 
>> b/arch/arm64/boot/dts/freescale/fsl-ls1028a.dtsi
>> index 262fbad8f0ec..c1f2f402ad53 100644
>> --- a/arch/arm64/boot/dts/freescale/fsl-ls1028a.dtsi
>> +++ b/arch/arm64/boot/dts/freescale/fsl-ls1028a.dtsi
>> @@ -1114,6 +1114,12 @@
>>  					full-duplex;
>>  				};
>>  			};
>> +
>> +			rcec@1f,0 {
> 
> Just curious how unit-address comes to '1f,0'?

You mean that it's pci dev 1f func 0?
0x00f800 >> 11 == 0x1f

See also:
https://elixir.bootlin.com/linux/v5.12-rc1/source/scripts/dtc/checks.c#L916

-michael

> 
> Shawn
> 
>> +				reg = <0x00f800 0 0 0 0>;
>> +				/* IEP INT_A */
>> +				interrupts = <GIC_SPI 94 IRQ_TYPE_LEVEL_HIGH>;
>> +			};
>>  		};
>> 
>>  		rcpm: power-controller@1e34040 {
>> --
>> 2.20.1
>>
Shawn Guo March 5, 2021, 12:35 a.m. UTC | #5
On Tue, Feb 09, 2021 at 01:52:59AM +0100, Michael Walle wrote:
> The legacy interrupt INT_A is hardwired to the event collector. RCEC is
> bascially supported starting with v5.11. Having a correct interrupt, will
> make RCEC at least probe correctly.
> 
> There are still issues with how RCEC is implemented in the RCiEP on the
> LS1028A. RCEC will report an error, but it cannot find the correct
> subdevice.
> 
> Signed-off-by: Michael Walle <michael@walle.cc>

Applied, thanks.

Patch
diff mbox series

diff --git a/arch/arm64/boot/dts/freescale/fsl-ls1028a.dtsi b/arch/arm64/boot/dts/freescale/fsl-ls1028a.dtsi
index 262fbad8f0ec..c1f2f402ad53 100644
--- a/arch/arm64/boot/dts/freescale/fsl-ls1028a.dtsi
+++ b/arch/arm64/boot/dts/freescale/fsl-ls1028a.dtsi
@@ -1114,6 +1114,12 @@ 
 					full-duplex;
 				};
 			};
+
+			rcec@1f,0 {
+				reg = <0x00f800 0 0 0 0>;
+				/* IEP INT_A */
+				interrupts = <GIC_SPI 94 IRQ_TYPE_LEVEL_HIGH>;
+			};
 		};
 
 		rcpm: power-controller@1e34040 {