Hi, On 2022-09-09 21:20, Hans de Goede wrote: > Hi, > > On 9/9/22 20:32, Arvid Norlander wrote: >> Hi, >> >> On 2022-09-09 19:42, Hans de Goede wrote: >>> Hi, >>> >>> On 9/9/22 19:29, Arvid Norlander wrote: >>>> Hi, >>>> >>>> Given the changes, do you want me to test this again? If so, on what branch? >>> >>> I have just pushed this new version + all your pending toshiba_acpi patches >>> to my review-hans branch: >>> >>> https://git.kernel.org/pub/scm/linux/kernel/git/pdx86/platform-drivers-x86.git/log/?h=review-hans\ >>> >>> If you can give this branch a quick test and let me know if everything works >>> as expected that would be great. >> >> My changes (battery charge threshold, fan HWMON support, ECO LED work as >> expected. >> >> Display does come on after resume. Unfortunately the computer also >> completely locks up (I can't even switch from X to a VT). SSH is dead. >> When I suspend with a VT active the cursor is no longer blinking on resume. >> So something else is severly broken in this version and I cannot fully test >> this. >> >> I notice your tree is based on rc1. I was previously testing with 5.19 as >> well as rc2 and newer. Is it possible this is a bug from mainline? > > Ah yes, sorry about that. There is a scsi bug which causes any sata > disks to become inaccessible after suspend/resume in rc1. > > If you cherry pick this commit on top of my tree the problem should be gone: > > https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=785538bfdd682c8e962341d585f9b88262a0475e This fixed it. Resume works properly, as does backlight controls after resume. Tested-by: Arvid Norlander <lkml@xxxxxxxxx> (For the relevant toshiba_acpi patches) However I notice that the laptop is hung for about a second after resume. I had not noticed this before. But I went back and checked on 5.19 and it happens there too, so presumably I just ended up being hyper-aware about resume issues after this... > >> Or is >> it something introduced by your tree? I'm wondering it it would be easier >> to first rebase your tree on the latest RC rather than trying to bisect it >> as is. >> >> Or the easiest test is perhaps: Does suspend and resume with this tree work >> on your computer? I remember hearing that there were SATA issues in early >> 6.0-rcs (don't remember the details), the Toshiba uses mSATA. > > Right, see above. > > Regards, > > Hans > Best regards, Arvid Norlander
