Re: [RFC v1 2/4] PCI: Add support for drivers to decide bridge D3 policy

[Mario Limonciello <mario.limonciello@amd.com>]

On 10/14/2023 05:53, Lukas Wunner wrote:
> On Mon, Oct 09, 2023 at 05:56:51PM -0500, Mario Limonciello wrote:
>> commit 9d26d3a8f1b0 ("PCI: Put PCIe ports into D3 during suspend")
>> changed pci_bridge_d3_possible() so that any vendor's PCIe ports
>> from modern machines (>=2015) are allowed to be put into D3.
>>
>> This policy change has worked for most machines, but the behavior
>> is improved with `pcie_port_pm=off` on some others.
>>
>> Add support for drivers to register a callback that they can optionally
>> use to decide the policy on supported machines.
> 
> I would assume that drivers can decide the policy already today through
> pci_d3cold_enable() / pci_d3cold_disable().
> 
> Why is this not sufficient and what's the benefit of the more complex
> approach proposed here?
> 

This approach allows multiple drivers to participate.  Realistically 
however I don't know "many" drivers will participate in the first place.

 From our earlier conversations on the quirk thread I think your 
proposal will work as well for this, I'll try it.  Thanks.