Re: [PATCH V3 2/6] dt-bindings: PCI: tegra: Add PCIe slot supplies regulator entries

[Thierry Reding <thierry.reding@gmail.com>]

[-- Attachment #1: Type: text/plain, Size: 3136 bytes --]

On Wed, Aug 28, 2019 at 10:58:46PM +0530, Vidya Sagar wrote:
> Add optional bindings "vpcie3v3-supply" and "vpcie12v-supply" to describe
> regulators of a PCIe slot's supplies 3.3V and 12V provided the platform
> is designed to have regulator controlled slot supplies.
> 
> Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
> ---
> V3:
> * None
> 
> V2:
> * None
> 
>  .../devicetree/bindings/pci/nvidia,tegra194-pcie.txt      | 8 ++++++++
>  1 file changed, 8 insertions(+)
> 
> diff --git a/Documentation/devicetree/bindings/pci/nvidia,tegra194-pcie.txt b/Documentation/devicetree/bindings/pci/nvidia,tegra194-pcie.txt
> index 0ac1b867ac24..b739f92da58e 100644
> --- a/Documentation/devicetree/bindings/pci/nvidia,tegra194-pcie.txt
> +++ b/Documentation/devicetree/bindings/pci/nvidia,tegra194-pcie.txt
> @@ -104,6 +104,12 @@ Optional properties:
>     specified in microseconds
>  - nvidia,aspm-l0s-entrance-latency-us: ASPM L0s entrance latency to be
>     specified in microseconds
> +- vpcie3v3-supply: A phandle to the regulator node that supplies 3.3V to the slot
> +  if the platform has one such slot. (Ex:- x16 slot owned by C5 controller
> +  in p2972-0000 platform).
> +- vpcie12v-supply: A phandle to the regulator node that supplies 12V to the slot
> +  if the platform has one such slot. (Ex:- x16 slot owned by C5 controller
> +  in p2972-0000 platform).

There's an ongoing discussion regarding the use of optional power
supplies and I'm wondering if we're not abusing this here. Why exactly
are these regulators optional?

The distinction is somewhat subtle, but the other way to look at
modelling this in DT is that the supplies are in fact required, but may
be connected to an always-on regulator with a fixed voltage. Or in some
cases they may also be shorted to ground. In both cases the PCI
controller, or rather the slot that the controller connects to, actually
"requires" the supplies, it's just that we can get away without
describing them because they can't be controlled anyway.

Looking at the PCI connector pinout for PCI Express, I do see a bunch of
+3.3 V and +12 V pins. To me that indicates that the 3.3 V and 12 V
supplies are indeed required for PCI slots. I'm not sure about devices
that are directly connected to the PCI controller, though. I'll need to
go look at some schematics to get a better understanding of these.

Bottom line: I'm wondering if we shouldn't really make these supplies
mandatory and in case where we don't care either just leave them away
(the regulator framework will supply a dummy regulator in that case) or
hook them up to a fixed regulator if that matches the hardware design.

Any thoughts?

Thierry

>  
>  Examples:
>  =========
> @@ -156,6 +162,8 @@ Tegra194:
>  			  0xc2000000 0x18 0x00000000 0x18 0x00000000 0x4 0x00000000>;  /* prefetchable memory (16GB) */
>  
>  		vddio-pex-ctl-supply = <&vdd_1v8ao>;
> +		vpcie3v3-supply = <&vdd_3v3_pcie>;
> +		vpcie12v-supply = <&vdd_12v_pcie>;
>  
>  		phys = <&p2u_hsio_2>, <&p2u_hsio_3>, <&p2u_hsio_4>,
>  		       <&p2u_hsio_5>;
> -- 
> 2.17.1
> 

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]