[PATCH v3 1/5] dt-bindings: PCI: starfive,jh7110-pcie: Add enable-gpios property
Hal Feng
hal.feng at starfivetech.com
Thu Nov 20 20:23:35 PST 2025
> On 12.11.25 21:54, Manivannan Sadhasivam wrote:
> On Thu, Nov 20, 2025 at 04:29:42PM +0800, Hal Feng wrote:
> > Add enable-gpios property for controlling the PCI bus device power.
> > This property had been supported in the driver but not added in the
> > dt-bindings.
> >
> > Acked-by: Conor Dooley <conor.dooley at microchip.com>
> > Fixes: 22fe32239770 ("dt-bindings: PCI: Add StarFive JH7110 PCIe
> > controller")
> > Signed-off-by: Hal Feng <hal.feng at starfivetech.com>
> > ---
> > .../devicetree/bindings/pci/starfive,jh7110-pcie.yaml | 4 ++++
> > 1 file changed, 4 insertions(+)
> >
> > diff --git
> > a/Documentation/devicetree/bindings/pci/starfive,jh7110-pcie.yaml
> > b/Documentation/devicetree/bindings/pci/starfive,jh7110-pcie.yaml
> > index 33c80626e8ec..1e36f92ec852 100644
> > --- a/Documentation/devicetree/bindings/pci/starfive,jh7110-pcie.yaml
> > +++ b/Documentation/devicetree/bindings/pci/starfive,jh7110-pcie.yaml
> > @@ -59,6 +59,10 @@ properties:
> > description:
> > The phandle to System Register Controller syscon node.
> >
> > + enable-gpios:
> > + description: GPIO used to enable the PCI bus device power
>
> This feels wrong to me. Is this GPIO associated with the PCIe controller? I bet
> this is just controlling some regulator that powers the VDD of the PCIe
> device/slot. If so, this should be added as a part of the regulator node and
> referenced in the PCIe node using the existing -supply properties.
This GPIO just controls the power of PCIe devices, not PCIe controller.
I think there may be no design adding power control GPIOs for the PCIe controller,
because usually we don't need to control the PCIe controller power but the PCIe
device power.
I find a similar "pwren-gpios" in
Documentation/devicetree/bindings/pci/sifive,fu740-pcie.yaml .
It uses the GPIO to control the power of PCIe devices too.
Could I continue to do so? Thanks.
Best regards,
Hal
More information about the linux-riscv
mailing list