Re: [PATCH v2 1/2] PCI/ASPM: Override the ASPM and Clock PM states set by BIOS for devicetree platforms

[Jon Hunter <jonathanh@nvidia.com>]

On 16/02/2026 14:18, Manivannan Sadhasivam wrote:
> On Mon, Feb 16, 2026 at 02:03:41PM +0000, Jon Hunter wrote:
>> Hi Mani, Krishna,
>>
>> On 23/01/2026 13:56, Manivannan Sadhasivam wrote:
>>
>> ...
>>
>>> So to properly fix it, we need the controller drivers to perform below checks
>>> for all devices under the Root bus(ses) before initiating D3Cold:
>>>
>>> 1. Check if the device state is D3Hot. If it is not D3Hot, then the device is
>>> expected to stay in the current D-state by the client driver, so D3Cold should
>>> not be initiated.
>>>
>>> 2. Check if the device is wakeup capable. If it is, then check if it can support
>>> wakeup from D3Cold (with WAKE#).
>>>
>>> Only if both conditions are satisfied for all the devices under the Root busses,
>>> then the host controller driver should initiate D3Cold sequence.
>>>
>>> Krishna is going to post a patch that performs the above checks for
>>> pcie-designware-host driver. But since the above checks are platform agnostic,
>>> we should introduce a helper and resuse it across other controllers as well.
>>
>>
>> Do you have a rough idea of when you will be posting patches for this?
>>
> 
> Krishna posted the series a couple of weeks before but forgot to CC you:
> https://lore.kernel.org/linux-pci/20260128-d3cold-v1-0-dd8f3f0ce824@oss.qualcomm.com/
> 
> You are expected to use the helper pci_host_common_can_enter_d3cold() in the
> suspend path.

Thanks! I will take a look.

Jon

-- 
nvpublic