On 11 May 2025 20:59:12 CEST, Mikko Juhani Korhonen <mjkorhon@xxxxxxxxx> wrote: >su 11.5.2025 klo 21.53 Niklas Cassel (cassel@xxxxxxxxxx) kirjoitti: >> >linux-6.14.5/.config:CONFIG_SATA_MOBILE_LPM_POLICY=3 (good) >> >linux-6.8.12/.config:CONFIG_SATA_MOBILE_LPM_POLICY=3 (bad) > >Oops got this mixed up, my bad! I meant: >linux-6.8.12/.config:CONFIG_SATA_MOBILE_LPM_POLICY=3 (god) >linux-6.14.5/.config:CONFIG_SATA_MOBILE_LPM_POLICY=3 (bad) Ok, I see were things went wrong now. Damien asked for the good output for 6.8, rather than the good output for 6.14.5, when you plugged the drive in to a port that was working. Basically, we wanted to compare the good 6.14.5 output with the bad 6.14.5 output, especially the lpm-pol output. The 6.8 output is not interesting at all IMO, because that is before we enabled LPM for all controllers that claim to support it. Sorry for the confusion. If you get a different lpm-pol print, when booting the exact same 6.14.5 kernel binary, but just having the SATA drive plugged in to different ports, and you have hotplug disabled for all ports in BIOS, suggests a BIOS bug to me. Tomorrow, I will send a patch that will print, per port, if BIOS has marked the port as external/ hotplug capable or not. (Because right now, when we see lpm-pol 0, we don't know if it is because of BIOS settings or Kconfigs). Kind regards, Niklas
