Re: [PATCH] [RFC] PCI/PM: Do not RPM suspend devices without drivers

[Lukas Wunner <lukas@wunner.de>]

On Sat, Oct 12, 2024 at 02:48:48AM +0200, Marek Vasut wrote:
> The pci_host_probe() does reallocate BARs for devices which start up with
> uninitialized BAR addresses set to 0 by calling pci_bus_assign_resources(),
> which updates the device config space content.
> 
> At the same time, pci_pm_runtime_suspend() triggers pci_save_state() for
> all devices which do not have drivers assigned to them to store current
> content of their config space registers.
[...]
> Work around the issue by not suspending pci_bus_type devices which do
> not have driver assigned to them, keep those devices active to prevent
> pci_save_state() from being called. Once a proper driver takes over, it
> can RPM manage the device correctly.

It sounds like you may want to acquire a runtime PM reference
or disable runtime PM for the duration of the bus scan (or at
least device scan) rather than the proposed workaround.

Thanks,

Lukas