On 29.05.2020 22:59, Matthew Garrett wrote: > On Fri, May 29, 2020 at 03:21:35PM -0500, Bjorn Helgaas wrote: > >> Yeah, that makes sense. I can't remember the details, but I'm pretty >> sure there's a reason why we ask for the whole set of things. Seems >> like it solved some problem. I think Matthew Garrett might have been >> involved in that. > > This was https://bugzilla.redhat.com/show_bug.cgi?id=638912 - some > firmware misbehaves unless you pass the same set of supported > functionality as Windows does. >
Current situation means that PME is unusable on all systems where pcie_aspm_support_enabled() returns false, what is basically every system except EXPERT mode is enabled and CONFIG_PCIEASPM is set. So we definitely need to do something. One question is whether the system from the 10yr old bug report actually depends on OSC_PCI_EXPRESS_LTR_CONTROL control, or whether some other change in recent years fixed the issue. Not sure whether the system is still available for re-testing. If worst case we have 10yr old systems breaking with a new kernel then we still would have the workaround to enable CONFIG_PCIEASPM on that system.
