[PATCH 0/2] arm64: dts: qcom: x1e80100-yoga: add wifi calibration variant

[PATCH 0/2] arm64: dts: qcom: x1e80100-yoga: add wifi calibration variant

[Patrick Wildt]

This series adds the missing calibration variant devicetree property
which is needed to load the calibration data and use the ath12k wifi
on the Lenovo Yoga Slim 7x.

Patrick Wildt (2):
  dt-bindings: net: wireless: add ath12k pcie bindings
  arm64: dts: qcom: x1e80100-yoga: add wifi calibration variant

 .../net/wireless/qcom,ath12k-pci.yaml         | 59 +++++++++++++++++++
 .../dts/qcom/x1e80100-lenovo-yoga-slim7x.dts  |  9 +++
 arch/arm64/boot/dts/qcom/x1e80100.dtsi        | 10 ++++
 3 files changed, 78 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/net/wireless/qcom,ath12k-pci.yaml

-- 
2.45.2

[PATCH 1/2] dt-bindings: net: wireless: add ath12k pcie bindings

[Patrick Wildt]

Add devicetree bindings for Qualcomm ath12k PCIe devices such as WCN7850
for which the calibration data variant may need to be described.

Signed-off-by: Patrick Wildt <patrick@blueri.se>
---
 .../net/wireless/qcom,ath12k-pci.yaml         | 59 +++++++++++++++++++
 1 file changed, 59 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/net/wireless/qcom,ath12k-pci.yaml

diff --git a/Documentation/devicetree/bindings/net/wireless/qcom,ath12k-pci.yaml b/Documentation/devicetree/bindings/net/wireless/qcom,ath12k-pci.yaml
new file mode 100644
index 000000000000..8f18868ee726
--- /dev/null
+++ b/Documentation/devicetree/bindings/net/wireless/qcom,ath12k-pci.yaml
@@ -0,0 +1,59 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+# Copyright (c) 2024 Linaro Limited
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/net/wireless/qcom,ath12k-pci.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: Qualcomm Technologies ath12k wireless devices (PCIe)
+
+maintainers:
+  - Kalle Valo <kvalo@kernel.org>
+  - Jeff Johnson <jjohnson@kernel.org>
+
+description: |
+  Qualcomm Technologies IEEE 802.11ax PCIe devices
+
+properties:
+  compatible:
+    enum:
+      - pci17cb,1107  # WCN7850
+
+  reg:
+    maxItems: 1
+
+  qcom,ath12k-calibration-variant:
+    $ref: /schemas/types.yaml#/definitions/string
+    description: |
+      string to uniquely identify variant of the calibration data for designs
+      with colliding bus and device ids
+
+required:
+  - compatible
+  - reg
+
+additionalProperties: false
+
+examples:
+  - |
+    pcie {
+        #address-cells = <3>;
+        #size-cells = <2>;
+
+        pcie@0 {
+            device_type = "pci";
+            reg = <0x0 0x0 0x0 0x0 0x0>;
+            #address-cells = <3>;
+            #size-cells = <2>;
+            ranges;
+
+            bus-range = <0x01 0xff>;
+
+            wifi@0 {
+                compatible = "pci17cb,1107";
+                reg = <0x10000 0x0 0x0 0x0 0x0>;
+
+                qcom,ath12k-calibration-variant = "LES790";
+            };
+        };
+    };
-- 
2.45.2

[PATCH 2/2] arm64: dts: qcom: x1e80100-yoga: add wifi calibration variant

[Patrick Wildt]

Describe the bus topology for PCIe domain 4 and add the ath12k
calibration variant so that the board file (calibration data) can be
loaded.

Signed-off-by: Patrick Wildt <patrick@blueri.se>
---
 .../boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts      |  9 +++++++++
 arch/arm64/boot/dts/qcom/x1e80100.dtsi                 | 10 ++++++++++
 2 files changed, 19 insertions(+)

diff --git a/arch/arm64/boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts b/arch/arm64/boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts
index fbff558f5b07..f569f0fbd1fc 100644
--- a/arch/arm64/boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts
+++ b/arch/arm64/boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts
@@ -635,6 +635,15 @@ &pcie4_phy {
 	status = "okay";
 };
 
+&pcie4_port0 {
+	wifi@0 {
+		compatible = "pci17cb,1107";
+		reg = <0x10000 0x0 0x0 0x0 0x0>;
+
+		qcom,ath12k-calibration-variant = "LES790";
+	};
+};
+
 &pcie6a {
 	perst-gpios = <&tlmm 152 GPIO_ACTIVE_LOW>;
 	wake-gpios = <&tlmm 154 GPIO_ACTIVE_LOW>;
diff --git a/arch/arm64/boot/dts/qcom/x1e80100.dtsi b/arch/arm64/boot/dts/qcom/x1e80100.dtsi
index 7bca5fcd7d52..70eeacd4f9ad 100644
--- a/arch/arm64/boot/dts/qcom/x1e80100.dtsi
+++ b/arch/arm64/boot/dts/qcom/x1e80100.dtsi
@@ -3085,6 +3085,16 @@ &mc_virt SLAVE_EBI1 QCOM_ICC_TAG_ALWAYS>,
 			phy-names = "pciephy";
 
 			status = "disabled";
+
+			pcie4_port0: pcie@0 {
+				device_type = "pci";
+				reg = <0x0 0x0 0x0 0x0 0x0>;
+				bus-range = <0x01 0xff>;
+
+				#address-cells = <3>;
+				#size-cells = <2>;
+				ranges;
+			};
 		};
 
 		pcie4_phy: phy@1c0e000 {
-- 
2.45.2

Re: [PATCH 1/2] dt-bindings: net: wireless: add ath12k pcie bindings

[Andrew Lunn]

On Mon, Jul 15, 2024 at 09:39:51PM +0200, Patrick Wildt wrote:
> Add devicetree bindings for Qualcomm ath12k PCIe devices such as WCN7850
> for which the calibration data variant may need to be described.

Hi Patrick

General, the device tree binding and the needed changes to the driver
to implement the binding are in the same patchset. I don't see
anything implementing qcom,ath12k-calibration-variant here? Does the
driver already support this, and you are just fixing up missing
documentation?

	Andrew

Re: [PATCH 1/2] dt-bindings: net: wireless: add ath12k pcie bindings

[Patrick Wildt]

Am Mon, Jul 15, 2024 at 10:54:18PM +0200 schrieb Andrew Lunn:
> On Mon, Jul 15, 2024 at 09:39:51PM +0200, Patrick Wildt wrote:
> > Add devicetree bindings for Qualcomm ath12k PCIe devices such as WCN7850
> > for which the calibration data variant may need to be described.
> 
> Hi Patrick
> 
> General, the device tree binding and the needed changes to the driver
> to implement the binding are in the same patchset. I don't see
> anything implementing qcom,ath12k-calibration-variant here? Does the
> driver already support this, and you are just fixing up missing
> documentation?
> 
> 	Andrew

Hi there,

technically I could try and make that change, but I don't actually run
this driver or Linux at all; this change is for running OpenBSD on that
machine with a correctly defined device tree.

The realities of Linux being the de facto upstream for device tree
bindings force me to submit changes here so they end up being usable
for other operating systems as well.

I would assume that eventually someone that runs Linux will adjust the
ath12k driver as well, because this kind of binding has been used for
both ath10k and ath11k and this is just a copy with a name change for
ath12k.

Cheers,
Patrick

Re: [PATCH 2/2] arm64: dts: qcom: x1e80100-yoga: add wifi calibration variant

[Dmitry Baryshkov]

On Mon, Jul 15, 2024 at 09:40:41PM GMT, Patrick Wildt wrote:
> Describe the bus topology for PCIe domain 4 and add the ath12k
> calibration variant so that the board file (calibration data) can be
> loaded.
> 
> Signed-off-by: Patrick Wildt <patrick@blueri.se>
> ---
>  .../boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts      |  9 +++++++++
>  arch/arm64/boot/dts/qcom/x1e80100.dtsi                 | 10 ++++++++++
>  2 files changed, 19 insertions(+)
> 
> diff --git a/arch/arm64/boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts b/arch/arm64/boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts
> index fbff558f5b07..f569f0fbd1fc 100644
> --- a/arch/arm64/boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts
> +++ b/arch/arm64/boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts
> @@ -635,6 +635,15 @@ &pcie4_phy {
>  	status = "okay";
>  };
>  
> +&pcie4_port0 {
> +	wifi@0 {
> +		compatible = "pci17cb,1107";
> +		reg = <0x10000 0x0 0x0 0x0 0x0>;
> +
> +		qcom,ath12k-calibration-variant = "LES790";

It doesn't look like it follows the rest of the calibration variants.

Something like "Lenovo_Y7x" or "Lenovo_Yoga7x" sounds more logical.

> +	};
> +};
> +
>  &pcie6a {
>  	perst-gpios = <&tlmm 152 GPIO_ACTIVE_LOW>;
>  	wake-gpios = <&tlmm 154 GPIO_ACTIVE_LOW>;
> diff --git a/arch/arm64/boot/dts/qcom/x1e80100.dtsi b/arch/arm64/boot/dts/qcom/x1e80100.dtsi
> index 7bca5fcd7d52..70eeacd4f9ad 100644
> --- a/arch/arm64/boot/dts/qcom/x1e80100.dtsi
> +++ b/arch/arm64/boot/dts/qcom/x1e80100.dtsi
> @@ -3085,6 +3085,16 @@ &mc_virt SLAVE_EBI1 QCOM_ICC_TAG_ALWAYS>,
>  			phy-names = "pciephy";
>  
>  			status = "disabled";
> +
> +			pcie4_port0: pcie@0 {
> +				device_type = "pci";
> +				reg = <0x0 0x0 0x0 0x0 0x0>;
> +				bus-range = <0x01 0xff>;
> +
> +				#address-cells = <3>;
> +				#size-cells = <2>;
> +				ranges;
> +			};
>  		};
>  
>  		pcie4_phy: phy@1c0e000 {
> -- 
> 2.45.2
> 

-- 
With best wishes
Dmitry

Re: [PATCH 2/2] arm64: dts: qcom: x1e80100-yoga: add wifi calibration variant

[Patrick Wildt]

Am Tue, Jul 16, 2024 at 12:51:53AM +0300 schrieb Dmitry Baryshkov:
> On Mon, Jul 15, 2024 at 09:40:41PM GMT, Patrick Wildt wrote:
> > Describe the bus topology for PCIe domain 4 and add the ath12k
> > calibration variant so that the board file (calibration data) can be
> > loaded.
> > 
> > Signed-off-by: Patrick Wildt <patrick@blueri.se>
> > ---
> >  .../boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts      |  9 +++++++++
> >  arch/arm64/boot/dts/qcom/x1e80100.dtsi                 | 10 ++++++++++
> >  2 files changed, 19 insertions(+)
> > 
> > diff --git a/arch/arm64/boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts b/arch/arm64/boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts
> > index fbff558f5b07..f569f0fbd1fc 100644
> > --- a/arch/arm64/boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts
> > +++ b/arch/arm64/boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts
> > @@ -635,6 +635,15 @@ &pcie4_phy {
> >  	status = "okay";
> >  };
> >  
> > +&pcie4_port0 {
> > +	wifi@0 {
> > +		compatible = "pci17cb,1107";
> > +		reg = <0x10000 0x0 0x0 0x0 0x0>;
> > +
> > +		qcom,ath12k-calibration-variant = "LES790";
> 
> It doesn't look like it follows the rest of the calibration variants.
> 
> Something like "Lenovo_Y7x" or "Lenovo_Yoga7x" sounds more logical.

This is what's both in the DSDT

  Device (WLN)
  {
    [...]
    Name (BDFE, "BDF_LES790")

and kvalo's board-2.bin for this machine:

  $ strings board-2.bin | grep LES
  bus=pci,vendor=17cb,device=1107,subsystem-vendor=17aa,subsystem-device=e0e9,qmi-chip-id=2,qmi-board-id=255,variant=LES790

I don't think we can hand-pick these strings, they come from whoever
decided upon them and fed them into ACPI tables and QC's binaries.

Re: [PATCH 1/2] dt-bindings: net: wireless: add ath12k pcie bindings

[Konrad Dybcio]

On 15.07.2024 11:10 PM, Patrick Wildt wrote:
> Am Mon, Jul 15, 2024 at 10:54:18PM +0200 schrieb Andrew Lunn:
>> On Mon, Jul 15, 2024 at 09:39:51PM +0200, Patrick Wildt wrote:
>>> Add devicetree bindings for Qualcomm ath12k PCIe devices such as WCN7850
>>> for which the calibration data variant may need to be described.
>>
>> Hi Patrick
>>
>> General, the device tree binding and the needed changes to the driver
>> to implement the binding are in the same patchset. I don't see
>> anything implementing qcom,ath12k-calibration-variant here? Does the
>> driver already support this, and you are just fixing up missing
>> documentation?
>>
>> 	Andrew
> 
> Hi there,
> 
> technically I could try and make that change, but I don't actually run
> this driver or Linux at all; this change is for running OpenBSD on that
> machine with a correctly defined device tree.
> 
> The realities of Linux being the de facto upstream for device tree
> bindings force me to submit changes here so they end up being usable
> for other operating systems as well.
> 
> I would assume that eventually someone that runs Linux will adjust the
> ath12k driver as well, because this kind of binding has been used for
> both ath10k and ath11k and this is just a copy with a name change for
> ath12k.

Paraphrasing a certain company.. Linux <3 OpenBSD!

Since as you mentioned Linux is the de facto upstream for dt-bindings,
many people identify those two.. I think a solution that would keep
everyone reasonably happy is dropping a line about your usage in the
cover letter

Konrad

Re: [PATCH 2/2] arm64: dts: qcom: x1e80100-yoga: add wifi calibration variant

[Konrad Dybcio]

On 16.07.2024 12:01 AM, Patrick Wildt wrote:
> Am Tue, Jul 16, 2024 at 12:51:53AM +0300 schrieb Dmitry Baryshkov:
>> On Mon, Jul 15, 2024 at 09:40:41PM GMT, Patrick Wildt wrote:
>>> Describe the bus topology for PCIe domain 4 and add the ath12k
>>> calibration variant so that the board file (calibration data) can be
>>> loaded.
>>>
>>> Signed-off-by: Patrick Wildt <patrick@blueri.se>
>>> ---
>>>  .../boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts      |  9 +++++++++
>>>  arch/arm64/boot/dts/qcom/x1e80100.dtsi                 | 10 ++++++++++
>>>  2 files changed, 19 insertions(+)
>>>
>>> diff --git a/arch/arm64/boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts b/arch/arm64/boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts
>>> index fbff558f5b07..f569f0fbd1fc 100644
>>> --- a/arch/arm64/boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts
>>> +++ b/arch/arm64/boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts
>>> @@ -635,6 +635,15 @@ &pcie4_phy {
>>>  	status = "okay";
>>>  };
>>>  
>>> +&pcie4_port0 {
>>> +	wifi@0 {
>>> +		compatible = "pci17cb,1107";
>>> +		reg = <0x10000 0x0 0x0 0x0 0x0>;
>>> +
>>> +		qcom,ath12k-calibration-variant = "LES790";
>>
>> It doesn't look like it follows the rest of the calibration variants.
>>
>> Something like "Lenovo_Y7x" or "Lenovo_Yoga7x" sounds more logical.
> 
> This is what's both in the DSDT
> 
>   Device (WLN)
>   {
>     [...]
>     Name (BDFE, "BDF_LES790")
> 
> and kvalo's board-2.bin for this machine:
> 
>   $ strings board-2.bin | grep LES
>   bus=pci,vendor=17cb,device=1107,subsystem-vendor=17aa,subsystem-device=e0e9,qmi-chip-id=2,qmi-board-id=255,variant=LES790
> 
> I don't think we can hand-pick these strings, they come from whoever
> decided upon them and fed them into ACPI tables and QC's binaries.

Huh, great nod from qca!
Looks like there's even a bit more hw supported in there..

Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>

Konrad

Re: [PATCH 2/2] arm64: dts: qcom: x1e80100-yoga: add wifi calibration variant

[Konrad Dybcio]

On 16.07.2024 1:00 AM, Konrad Dybcio wrote:
> On 16.07.2024 12:01 AM, Patrick Wildt wrote:
>> Am Tue, Jul 16, 2024 at 12:51:53AM +0300 schrieb Dmitry Baryshkov:
>>> On Mon, Jul 15, 2024 at 09:40:41PM GMT, Patrick Wildt wrote:
>>>> Describe the bus topology for PCIe domain 4 and add the ath12k
>>>> calibration variant so that the board file (calibration data) can be
>>>> loaded.
>>>>
>>>> Signed-off-by: Patrick Wildt <patrick@blueri.se>
>>>> ---
>>>>  .../boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts      |  9 +++++++++
>>>>  arch/arm64/boot/dts/qcom/x1e80100.dtsi                 | 10 ++++++++++
>>>>  2 files changed, 19 insertions(+)
>>>>
>>>> diff --git a/arch/arm64/boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts b/arch/arm64/boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts
>>>> index fbff558f5b07..f569f0fbd1fc 100644
>>>> --- a/arch/arm64/boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts
>>>> +++ b/arch/arm64/boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts
>>>> @@ -635,6 +635,15 @@ &pcie4_phy {
>>>>  	status = "okay";
>>>>  };
>>>>  
>>>> +&pcie4_port0 {
>>>> +	wifi@0 {
>>>> +		compatible = "pci17cb,1107";
>>>> +		reg = <0x10000 0x0 0x0 0x0 0x0>;
>>>> +
>>>> +		qcom,ath12k-calibration-variant = "LES790";
>>>
>>> It doesn't look like it follows the rest of the calibration variants.
>>>
>>> Something like "Lenovo_Y7x" or "Lenovo_Yoga7x" sounds more logical.
>>
>> This is what's both in the DSDT
>>
>>   Device (WLN)
>>   {
>>     [...]
>>     Name (BDFE, "BDF_LES790")
>>
>> and kvalo's board-2.bin for this machine:
>>
>>   $ strings board-2.bin | grep LES
>>   bus=pci,vendor=17cb,device=1107,subsystem-vendor=17aa,subsystem-device=e0e9,qmi-chip-id=2,qmi-board-id=255,variant=LES790
>>
>> I don't think we can hand-pick these strings, they come from whoever
>> decided upon them and fed them into ACPI tables and QC's binaries.
> 
> Huh, great nod from qca!
> Looks like there's even a bit more hw supported in there..
> 
> Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>

Ref: https://git.codelinaro.org/clo/ath-firmware/ath12k-firmware/-/blob/main/WCN7850/hw2.0/board-2.bin

Konrad

Re: [PATCH 1/2] dt-bindings: net: wireless: add ath12k pcie bindings

[Krzysztof Kozlowski]

On 15/07/2024 21:39, Patrick Wildt wrote:
> Add devicetree bindings for Qualcomm ath12k PCIe devices such as WCN7850
> for which the calibration data variant may need to be described.
> 
> Signed-off-by: Patrick Wildt <patrick@blueri.se>
> ---
>  .../net/wireless/qcom,ath12k-pci.yaml         | 59 +++++++++++++++++++
>  1 file changed, 59 insertions(+)
>  create mode 100644 Documentation/devicetree/bindings/net/wireless/qcom,ath12k-pci.yaml
> 
> diff --git a/Documentation/devicetree/bindings/net/wireless/qcom,ath12k-pci.yaml b/Documentation/devicetree/bindings/net/wireless/qcom,ath12k-pci.yaml
> new file mode 100644
> index 000000000000..8f18868ee726
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/net/wireless/qcom,ath12k-pci.yaml
> @@ -0,0 +1,59 @@
> +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
> +# Copyright (c) 2024 Linaro Limited
> +%YAML 1.2
> +---
> +$id: http://devicetree.org/schemas/net/wireless/qcom,ath12k-pci.yaml#
> +$schema: http://devicetree.org/meta-schemas/core.yaml#
> +
> +title: Qualcomm Technologies ath12k wireless devices (PCIe)
> +
> +maintainers:
> +  - Kalle Valo <kvalo@kernel.org>
> +  - Jeff Johnson <jjohnson@kernel.org>
> +
> +description: |
> +  Qualcomm Technologies IEEE 802.11ax PCIe devices
> +
> +properties:
> +  compatible:
> +    enum:
> +      - pci17cb,1107  # WCN7850

That's a duplicate of:
https://lore.kernel.org/all/20240605122106.23818-1-brgl@bgdev.pl/

AFAIK, this was already merged. I don't get why after ~1 month it is not
in the linux-next.

Best regards,
Krzysztof

Re: [PATCH 2/2] arm64: dts: qcom: x1e80100-yoga: add wifi calibration variant

[Manivannan Sadhasivam]

On Mon, Jul 15, 2024 at 09:40:41PM +0200, Patrick Wildt wrote:
> Describe the bus topology for PCIe domain 4 and add the ath12k
> calibration variant so that the board file (calibration data) can be
> loaded.
> 
> Signed-off-by: Patrick Wildt <patrick@blueri.se>

Reviewed-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>

- Mani

> ---
>  .../boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts      |  9 +++++++++
>  arch/arm64/boot/dts/qcom/x1e80100.dtsi                 | 10 ++++++++++
>  2 files changed, 19 insertions(+)
> 
> diff --git a/arch/arm64/boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts b/arch/arm64/boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts
> index fbff558f5b07..f569f0fbd1fc 100644
> --- a/arch/arm64/boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts
> +++ b/arch/arm64/boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts
> @@ -635,6 +635,15 @@ &pcie4_phy {
>  	status = "okay";
>  };
>  
> +&pcie4_port0 {
> +	wifi@0 {
> +		compatible = "pci17cb,1107";
> +		reg = <0x10000 0x0 0x0 0x0 0x0>;
> +
> +		qcom,ath12k-calibration-variant = "LES790";
> +	};
> +};
> +
>  &pcie6a {
>  	perst-gpios = <&tlmm 152 GPIO_ACTIVE_LOW>;
>  	wake-gpios = <&tlmm 154 GPIO_ACTIVE_LOW>;
> diff --git a/arch/arm64/boot/dts/qcom/x1e80100.dtsi b/arch/arm64/boot/dts/qcom/x1e80100.dtsi
> index 7bca5fcd7d52..70eeacd4f9ad 100644
> --- a/arch/arm64/boot/dts/qcom/x1e80100.dtsi
> +++ b/arch/arm64/boot/dts/qcom/x1e80100.dtsi
> @@ -3085,6 +3085,16 @@ &mc_virt SLAVE_EBI1 QCOM_ICC_TAG_ALWAYS>,
>  			phy-names = "pciephy";
>  
>  			status = "disabled";
> +
> +			pcie4_port0: pcie@0 {
> +				device_type = "pci";
> +				reg = <0x0 0x0 0x0 0x0 0x0>;
> +				bus-range = <0x01 0xff>;
> +
> +				#address-cells = <3>;
> +				#size-cells = <2>;
> +				ranges;
> +			};
>  		};
>  
>  		pcie4_phy: phy@1c0e000 {
> -- 
> 2.45.2
> 
> 

-- 
மணிவண்ணன் சதாசிவம்

Re: [PATCH 2/2] arm64: dts: qcom: x1e80100-yoga: add wifi calibration variant

[Dmitry Baryshkov]

On Tue, Jul 16, 2024 at 12:01:28AM GMT, Patrick Wildt wrote:
> Am Tue, Jul 16, 2024 at 12:51:53AM +0300 schrieb Dmitry Baryshkov:
> > On Mon, Jul 15, 2024 at 09:40:41PM GMT, Patrick Wildt wrote:
> > > Describe the bus topology for PCIe domain 4 and add the ath12k
> > > calibration variant so that the board file (calibration data) can be
> > > loaded.
> > > 
> > > Signed-off-by: Patrick Wildt <patrick@blueri.se>
> > > ---
> > >  .../boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts      |  9 +++++++++
> > >  arch/arm64/boot/dts/qcom/x1e80100.dtsi                 | 10 ++++++++++
> > >  2 files changed, 19 insertions(+)
> > > 
> > > diff --git a/arch/arm64/boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts b/arch/arm64/boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts
> > > index fbff558f5b07..f569f0fbd1fc 100644
> > > --- a/arch/arm64/boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts
> > > +++ b/arch/arm64/boot/dts/qcom/x1e80100-lenovo-yoga-slim7x.dts
> > > @@ -635,6 +635,15 @@ &pcie4_phy {
> > >  	status = "okay";
> > >  };
> > >  
> > > +&pcie4_port0 {
> > > +	wifi@0 {
> > > +		compatible = "pci17cb,1107";
> > > +		reg = <0x10000 0x0 0x0 0x0 0x0>;
> > > +
> > > +		qcom,ath12k-calibration-variant = "LES790";
> > 
> > It doesn't look like it follows the rest of the calibration variants.
> > 
> > Something like "Lenovo_Y7x" or "Lenovo_Yoga7x" sounds more logical.
> 
> This is what's both in the DSDT
> 
>   Device (WLN)
>   {
>     [...]
>     Name (BDFE, "BDF_LES790")
> 
> and kvalo's board-2.bin for this machine:
> 
>   $ strings board-2.bin | grep LES
>   bus=pci,vendor=17cb,device=1107,subsystem-vendor=17aa,subsystem-device=e0e9,qmi-chip-id=2,qmi-board-id=255,variant=LES790
> 
> I don't think we can hand-pick these strings, they come from whoever
> decided upon them and fed them into ACPI tables and QC's binaries.

Ack, if Kalle has already selected this string, we can't argue.


Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>


-- 
With best wishes
Dmitry

Re: [PATCH 0/2] arm64: dts: qcom: x1e80100-yoga: add wifi calibration variant

[Jeff Johnson]

On 7/15/2024 12:38 PM, Patrick Wildt wrote:
> This series adds the missing calibration variant devicetree property
> which is needed to load the calibration data and use the ath12k wifi
> on the Lenovo Yoga Slim 7x.
> 
> Patrick Wildt (2):
>   dt-bindings: net: wireless: add ath12k pcie bindings
>   arm64: dts: qcom: x1e80100-yoga: add wifi calibration variant
> 
>  .../net/wireless/qcom,ath12k-pci.yaml         | 59 +++++++++++++++++++
>  .../dts/qcom/x1e80100-lenovo-yoga-slim7x.dts  |  9 +++
>  arch/arm64/boot/dts/qcom/x1e80100.dtsi        | 10 ++++
>  3 files changed, 78 insertions(+)
>  create mode 100644 Documentation/devicetree/bindings/net/wireless/qcom,ath12k-pci.yaml
> 

+ath12k mailing list.

Qualcomm expects, on x86 and Qualcomm-based ARM devices, that this information
come from ACPI.

That support is currently under review:
https://lore.kernel.org/all/20240717111023.78798-1-quic_lingbok@quicinc.com/

/jeff

Re: [PATCH 0/2] arm64: dts: qcom: x1e80100-yoga: add wifi calibration variant

[Rob Clark]

On Thu, Jul 18, 2024 at 7:40 AM Jeff Johnson <quic_jjohnson@quicinc.com> wrote:
>
> On 7/15/2024 12:38 PM, Patrick Wildt wrote:
> > This series adds the missing calibration variant devicetree property
> > which is needed to load the calibration data and use the ath12k wifi
> > on the Lenovo Yoga Slim 7x.
> >
> > Patrick Wildt (2):
> >   dt-bindings: net: wireless: add ath12k pcie bindings
> >   arm64: dts: qcom: x1e80100-yoga: add wifi calibration variant
> >
> >  .../net/wireless/qcom,ath12k-pci.yaml         | 59 +++++++++++++++++++
> >  .../dts/qcom/x1e80100-lenovo-yoga-slim7x.dts  |  9 +++
> >  arch/arm64/boot/dts/qcom/x1e80100.dtsi        | 10 ++++
> >  3 files changed, 78 insertions(+)
> >  create mode 100644 Documentation/devicetree/bindings/net/wireless/qcom,ath12k-pci.yaml
> >
>
> +ath12k mailing list.
>
> Qualcomm expects, on x86 and Qualcomm-based ARM devices, that this information
> come from ACPI.

That is fine for acpi boot.  But it never prevented adding needed
information in dt.

BR,
-R

> That support is currently under review:
> https://lore.kernel.org/all/20240717111023.78798-1-quic_lingbok@quicinc.com/
>
> /jeff
>

Re: [PATCH 0/2] arm64: dts: qcom: x1e80100-yoga: add wifi calibration variant

[Dmitry Baryshkov]

On Thu, Jul 18, 2024 at 07:40:13AM GMT, Jeff Johnson wrote:
> On 7/15/2024 12:38 PM, Patrick Wildt wrote:
> > This series adds the missing calibration variant devicetree property
> > which is needed to load the calibration data and use the ath12k wifi
> > on the Lenovo Yoga Slim 7x.
> > 
> > Patrick Wildt (2):
> >   dt-bindings: net: wireless: add ath12k pcie bindings
> >   arm64: dts: qcom: x1e80100-yoga: add wifi calibration variant
> > 
> >  .../net/wireless/qcom,ath12k-pci.yaml         | 59 +++++++++++++++++++
> >  .../dts/qcom/x1e80100-lenovo-yoga-slim7x.dts  |  9 +++
> >  arch/arm64/boot/dts/qcom/x1e80100.dtsi        | 10 ++++
> >  3 files changed, 78 insertions(+)
> >  create mode 100644 Documentation/devicetree/bindings/net/wireless/qcom,ath12k-pci.yaml
> > 
> 
> +ath12k mailing list.
> 
> Qualcomm expects, on x86 and Qualcomm-based ARM devices, that this information
> come from ACPI.

Unfortunately we can not use ACPI on Qualcomm-based ARM devices. They
all are manually converted to DT. 

> 
> That support is currently under review:
> https://lore.kernel.org/all/20240717111023.78798-1-quic_lingbok@quicinc.com/
> 
> /jeff

-- 
With best wishes
Dmitry

Re: (subset) [PATCH 0/2] arm64: dts: qcom: x1e80100-yoga: add wifi calibration variant

[Bjorn Andersson]

On Mon, 15 Jul 2024 21:38:11 +0200, Patrick Wildt wrote:
> This series adds the missing calibration variant devicetree property
> which is needed to load the calibration data and use the ath12k wifi
> on the Lenovo Yoga Slim 7x.
> 
> Patrick Wildt (2):
>   dt-bindings: net: wireless: add ath12k pcie bindings
>   arm64: dts: qcom: x1e80100-yoga: add wifi calibration variant
> 
> [...]

Applied, thanks!

[2/2] arm64: dts: qcom: x1e80100-yoga: add wifi calibration variant
      commit: 8bc7cb73df8644423758c79d4504d501c8ef3854

Best regards,
-- 
Bjorn Andersson <andersson@kernel.org>

Re: [PATCH 2/2] arm64: dts: qcom: x1e80100-yoga: add wifi calibration variant

[Bjorn Helgaas]

On Mon, Jul 15, 2024 at 09:40:41PM +0200, Patrick Wildt wrote:
> Describe the bus topology for PCIe domain 4 and add the ath12k
> calibration variant so that the board file (calibration data) can be
> loaded.

> +++ b/arch/arm64/boot/dts/qcom/x1e80100.dtsi
> @@ -3085,6 +3085,16 @@ &mc_virt SLAVE_EBI1 QCOM_ICC_TAG_ALWAYS>,
>  			phy-names = "pciephy";
>  
>  			status = "disabled";
> +
> +			pcie4_port0: pcie@0 {
> +				device_type = "pci";
> +				reg = <0x0 0x0 0x0 0x0 0x0>;
> +				bus-range = <0x01 0xff>;

Hi Patrick, what's the purpose of this bus-range?  IIUC this describes
a Root Port, where we can read and configure the secondary/subordinate
bus numbers from the RP config space, so it seems like we don't need
to describe them here.

> +				#address-cells = <3>;
> +				#size-cells = <2>;
> +				ranges;
> +			};
>  		};
>  
>  		pcie4_phy: phy@1c0e000 {
> -- 
> 2.45.2
>