Re: [PATCH v3 0/7] PCI/pwrctrl: Major rework to integrate pwrctrl devices with controller drivers

[Manivannan Sadhasivam]

On Tue, Dec 30, 2025 at 11:35:55AM +0100, Niklas Cassel wrote:
> On Mon, Dec 29, 2025 at 10:56:51PM +0530, Manivannan Sadhasivam via B4 Relay wrote:
> > Hi,
> > 
> > This series provides a major rework for the PCI power control (pwrctrl)
> > framework to enable the pwrctrl devices to be controlled by the PCI controller
> > drivers.
> > 
> > Problem Statement
> > =================
> > 
> > Currently, the pwrctrl framework faces two major issues:
> > 
> > 1. Missing PERST# integration
> 
> AFAICT, and from reading your reply here:
> 
> https://lore.kernel.org/linux-pci/ibvk4it7th4bi6djoxshjqjh7zusbulzpndac5jtqkqovvgcei@5sycben7pqkk/
> 
>   I suppose maybe you plan to enhance pwrctrl so it can assert/deassert
>   individual PERST# in the hierarchy?
> 
> "No, that plan has been dropped for good. For now, PERST# will be handled
> entirely by the controller drivers. Sharing the PERST# handling with pwrctrl
> proved to be a pain and it looks more clean (after the API introduction) to
> handle PERST# in controller drivers."
> 
> Thus, it seems that even after this series, pwrctrl will be missing PERST#
> integration. Perhaps the cover letter could be rephrased to more clearly
> highlight this.
> 
> Because it seems a bit weird that the first point of the problem statement
> (missing PERST# integration) will still be the case after this series.
> 

Maybe the wording has confused you. I intend to say that the previous
implementation doesn't integrate PERST# properly because controller drivers
deasserted PERST# even before the pwrctrl driver has turned on the power and
both of them worked asynchronously. But with this series they become synchronous
as the controller drivers have the control over the pwrctrl and PERST# and they
can assert/deassert PERST# following the spec timing requirements.

So even though PERST# is not part of the pwrctrl, the integration is now in
place. But I agree that the wording could be improved. I will rephrase it if I
happen to send next version.

- Mani

-- 
மணிவண்ணன் சதாசிவம்