Hello Christian, On Tue, Jan 14, 2025 at 06:51:19PM +0100, Christian Heusel wrote: > > > Looking at the Arch Linux thread, it seems that the person reporting the > > > problem says that even on a working kernel, you only see the device the > > > user presses the open button on the Blu-Ray player. > > > > > > This suggests to me that it sends a hotplug/PhyRdy event when the user > > > presses the open button. > > > > > > In Linux, if we don't detect anything connected to the port during initial > > > boot, and if the port is not marked (by firmware) as either Hotplug Capable, > > > or External, then we mask the PhyRdy interrupt. > > > > > > We could reconsider this, and leave the PhyRdy interrupt enabled even for a > > > port that is not Hotplug Capable or External, if there is no devices detected > > > on the port. Short story: I was thinking about this a lot, and unfortunately I came to the conclusion that we cannot with good conscience write such a workaround. The reason is that we would increase the power usage on all SATA ports that have no devices connected to them at boot, not just for his controller, but for all SATA controller all around the world, just to have a workaround for this non-spec compliant SATA Blu-Ray player. (The reason is that would not be able to offline the port, which will shut down the PHY). I suggest that your user simply does: # echo max_performance > /sys/class/scsi_host/host3/link_power_management_policy or $ sudo sh -c "echo max_performance > /sys/class/scsi_host/host3/link_power_management_policy" Long story: Each port can be marked External or Hotplug capable by firmware, if so, ahci_update_initial_lpm_policy() will override the configured CONFIG_SATA_MOBILE_LPM_POLICY and set it to 0. (This is because hotplug and LPM are mutually exclusive.) In this case, the lpm-pol is printed as 3 for all ports, this means that CONFIG_SATA_MOBILE_LPM_POLICY was set to 3, and that none of the ports were marked as external or hotplug capable by firmware. Note that even though the target LPM policy is printed early in the boot, the actual enablement of LPM is not done now, that is done _after_ all devices have been detected. > [ 0.594488] ata3: SATA max UDMA/133 abar m2048@0xa091a000 port 0xa091a200 irq 126 lpm-pol 3 > [ 0.594491] ata4: SATA max UDMA/133 abar m2048@0xa091a000 port 0xa091a280 irq 126 lpm-pol 3 > [ 0.594494] ata5: SATA max UDMA/133 abar m2048@0xa091a000 port 0xa091a300 irq 126 lpm-pol 3 > [ 0.594497] ata6: SATA max UDMA/133 abar m2048@0xa091a000 port 0xa091a380 irq 126 lpm-pol 3 Note that when the AHCI driver is loaded, it will call ahci_reset_controller(), which does a HBA reset, which will implicitly send a COMRESET to all ports. Thus, when we dump PxSSTS (SStatus), all ports have already been reset once, and we can see that: 0h No device detected and Phy communication not established 1h Device presence detected but Phy communication not established 3h Device presence detected and Phy communication established 4h Phy in offline mode as a result of the interface being disabled or running in a BIST loopback mode > [ 0.594618] ata3: before wakeup PxSSTS 0x133 PxSCTL 0x0 PxCMD 0x4016 PxDEVSLP 0x0 > [ 0.594675] ata5: before wakeup PxSSTS 0x133 PxSCTL 0x0 PxCMD 0x4016 PxDEVSLP 0x0 > [ 0.594717] ata6: before wakeup PxSSTS 0x4 PxSCTL 0x0 PxCMD 0x4016 PxDEVSLP 0x0 > [ 0.595698] ata4: before wakeup PxSSTS 0x0 PxSCTL 0x0 PxCMD 0x4016 PxDEVSLP 0x0 This is stored in the lowest hex value, so: ata3: 0x3 ata5: 0x3 ata6: 0x4 ata4: 0x0 > [ 1.113233] ata4: before reset PxSSTS 0x0 PxSCTL 0x0 PxCMD 0x4016 PxDEVSLP 0x0 > [ 1.113363] ata6: before reset PxSSTS 0x4 PxSCTL 0x0 PxCMD 0x4016 PxDEVSLP 0x0 > [ 1.113481] ata5: before reset PxSSTS 0x133 PxSCTL 0x0 PxCMD 0x4016 PxDEVSLP 0x0 > [ 1.113603] ata3: before reset PxSSTS 0x133 PxSCTL 0x0 PxCMD 0x4016 PxDEVSLP 0x0 After sending a wakeup to force all devices to active mode, there was no change (this was extra test code added by me). > [ 1.926827] ata3: after reset PxSSTS 0x133 PxSCTL 0x300 PxCMD 0x4016 PxDEVSLP 0x0 > [ 1.927224] ata6: after reset PxSSTS 0x4 PxSCTL 0x300 PxCMD 0x4016 PxDEVSLP 0x0 > [ 1.927707] ata4: after reset PxSSTS 0x0 PxSCTL 0x300 PxCMD 0x4016 PxDEVSLP 0x0 > [ 1.928128] ata5: after reset PxSSTS 0x133 PxSCTL 0x300 PxCMD 0x4016 PxDEVSLP 0x0 After reset, we still don't see ata4 moving to 0x3.
