[PATCH v2 0/2] PCI: qcom: Binding fix

[Manivannan Sadhasivam <manivannan.sadhasivam@oss.qualcomm.com>]

Hi,

This series fixes the binding issue around the PERST# and PHY properties.
The binding issue was reported in [1], while discussing a DTS fix [2].

The binding fix provided in this series is not sufficient enough to spot all the
shenanigans, especially keeping one property in Controller node and another in
Root Port node. But this series does catch the DTS issue fixed by [2]:

arch/arm64/boot/dts/qcom/sm8750-mtp.dtb: pcie@1c00000 (qcom,pcie-sm8750): 'oneOf' conditional failed, one must be fixed:
	'phys' is a required property
	False schema does not allow [[148, 102, 1]]
	False schema does not allow [[148, 104, 0]]
	from schema $id: http://devicetree.org/schemas/pci/qcom,pcie-sm8550.yaml#
arch/arm64/boot/dts/qcom/sm8750-mtp.dtb: pcie@1c00000 (qcom,pcie-sm8750): pcie@0: 'oneOf' conditional failed, one must be fixed:
	'reset-gpios' is a required property
	'wake-gpios' is a required property
	False schema does not allow [[34]]
	from schema $id: http://devicetree.org/schemas/pci/qcom,pcie-sm8550.yaml#
arch/arm64/boot/dts/qcom/sm8750-mtp.dtb: pcie@1c00000 (qcom,pcie-sm8750): Unevaluated properties are not allowed ('#address-cells', '#interrupt-cells', '#size-cells', 'bus-range', 'device_type', 'dma-coherent', 'interconnect-names', 'interconnects', 'interrupt-map', 'interrupt-map-mask', 'iommu-map', 'linux,pci-domain', 'msi-map', 'msi-map-mask', 'num-lanes', 'operating-points-v2', 'opp-table', 'pcie@0', 'perst-gpios', 'power-domains', 'ranges', 'wake-gpios' were unexpected)
	from schema $id: http://devicetree.org/schemas/pci/qcom,pcie-sm8550.yaml#

Thanks to Dmitry for suggesting the binding fix!

[1] https://lore.kernel.org/linux-pci/8f2e0631-6c59-4298-b36e-060708970ced@oss.qualcomm.com
[2] https://lore.kernel.org/all/20251008-sm8750-v1-1-daeadfcae980@oss.qualcomm.com

Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@oss.qualcomm.com>
---
Changes in v2:
- Dropped the binding patch that got applied.
- Dropped mandating the WAKE# property in binding as not all platforms are
  supplying it and also fixed the binding example reported by the bot.
- Reworked the driver patch to just make 'phy' property as optional since all
  platforms specify PERST#.
- Link to v1: https://lore.kernel.org/r/20251010-pci-binding-v1-0-947c004b5699@oss.qualcomm.com

---
Manivannan Sadhasivam (2):
      dt-bindings: PCI: qcom: Enforce check for PHY, PERST# properties
      PCI: qcom: Treat PHY as optional for the new binding

 .../devicetree/bindings/pci/qcom,pcie-common.yaml        | 16 ++++++++++++++++
 .../devicetree/bindings/pci/qcom,pcie-sc8180x.yaml       |  3 +++
 drivers/pci/controller/dwc/pcie-qcom.c                   |  2 +-
 3 files changed, 20 insertions(+), 1 deletion(-)
---
base-commit: 5472d60c129f75282d94ae5ad072ee6dfb7c7246
change-id: 20251010-pci-binding-7d4d7799c6ed

Best regards,
-- 
Manivannan Sadhasivam <manivannan.sadhasivam@oss.qualcomm.com>