Hello Hans, On Fri, Mar 15, 2024 at 04:01:07PM +0100, Hans de Goede wrote: > > > > If they only claimed that the HBA supported PMP, the Linux device > > driver would try to enumerate the devices behind the PMP according > > to the standard. > > > > See AHCI 1.3.1, section 9.2 Port Multiplier Enumeration. > > Or > > SATA-IO - Port Multiplier 1.0, 7.4.2 Device Enumeration. > > > > The PMP standard also describes how you read the device and vendor > > ID of the PMP. > > > > > > Right now, they AMedia? seem to have their own home-made PMP implementation. > > Ah, ok so I think you did see my reply? Checking if we can ignore > the builtin PMP support and uses Linux PMP support instead is > indeed an option. Yes, I saw your reply. I wouldn't call it "Linux PMP support", I would simply call it "PMP", since how you enumerate a PMP is decribed in AHCI and SATA-IO specs. The "builtin PMP support" is not PMP, since it doesn't follow the specs. I would rather call it "non-standard vendor extension" or "pseudo-PMP". > I think we should still merge the revert / dropping of the quirk > while we figure this out though, because not finding people's disks > anymore is a clear regression. I was hoping that someone could try the patch I sent out on Thursday, which sets the PMP supported bit in the CAP register, as that could potentially solve both problems here. I guess if no one has tested it by Monday or Tuesday, we will have no choice but to revert. Kind regards, Niklas
