On 04.01.2021 18:39, Lukas Wunner wrote: > On Thu, Dec 31, 2020 at 10:38:12AM +0100, Heiner Kallweit wrote: >> On 31.12.2020 05:07, Lukas Wunner wrote: >>> FWIW, if platform_pci_power_manageable() returns true, it can probably >>> be assumed that allowing runtime PM by default is okay. So as a first >>> step, you may want to call that instead of adding a new callback. >> >> I don't think that's sufficient. Most likely all the broken old systems >> return true for platform_pci_power_manageable(). > > platform_pci_power_manageable() is not a global flag, but rather > a per-device flag whether the platform is capable of power-managing > that device. E.g. for the ACPI platform, it indicates that objects > such as _PS0 or _PS3 are present in the device's namespace. > > My point is that if the platform can power-manage a device, > then it ought to be safe to enable runtime PM by default for it. > Not sure about that. Just that the BIOS claims it can power-manage the device, doesn't rule out that it's broken and fails to do so.
> If you insist on a "big hammer" approach by turning on runtime PM > by default for everything, you risk regressions. You can avoid > that by going for a smart approach which enables runtime PM in > cases when it's safe. > > Thanks, > > Lukas > Heiner
