Re: [PATCH v2] PCI: pciehp: Fix hotplug on Catlow Lake with unreliable PME status

[Kuppuswamy Sathyanarayanan]

Hi Lukas,

On 2/17/2026 10:22 AM, Lukas Wunner wrote:
> On Tue, Feb 17, 2026 at 09:01:25AM -0800, Kuppuswamy Sathyanarayanan wrote:
>> On 2/14/2026 7:11 AM, Lukas Wunner wrote:
>>> I've just realized that pcie_disable_interrupt() isn't called from
>>> pciehp_suspend() if pme_is_native() is true.  Should disabling
>                                            ^^^^
> Sorry, I meant "if pme_is_native() is *false*".
> My brain was apparently half asleep when I wrote this.
> 
>>> runtime PM cause a power regression, an alternative solution may be
>>> to make pcie_disable_interrupt() conditional on a new pme_is_broken()
>>> which checks for affected Catlow Lake PCH Root Ports.
>>>
>>> The pm_runtime_disable() approach is slightly preferred because
>>> it keeps pciehp code clean.
>>
>> I think pcie_disable_interrupt() is called from pciehp_suspend() when
>> pme_is_native() is true. Looking at the code:
>>
>> static void pciehp_disable_interrupt(struct pcie_device *dev)
>> {
>>         /*
>>          * Disable hotplug interrupt so that it does not trigger
>>          * immediately when the downstream link goes down.
>>          */
>>         if (pme_is_native(dev->port))
>>                 pcie_disable_interrupt(get_service_data(dev));
>> }
> 
> What I had in mind is something like:
> 
> -	if (pme_is_native(dev))
> +	if (pme_is_native(dev) && !pme_is_broken(dev))
> 		pcie_disable_interrupt(get_service_data(dev));
> 
> Again, the pm_runtime_disable() (or pm_runtime_get_sync()) approach
> is slightly preferred because it keeps pciehp code clean and confines
> the issue to a quirk that only needs to be compiled in on x86.

Got it. Thanks for clarifying.

> 
> Thanks,
> 
> Lukas
> 

-- 
Sathyanarayanan Kuppuswamy
Linux Kernel Developer