05:48 <jglathe_x13s> Tried the Ubuntu 24.04 iso of 04/17/24 (need to write it with rufus, disks utility was worse, Still this xhci USB loop at boot, no luck. Needed regulator_ignore_unused and arm64.nopauth to boot (and see what happens for more than 30 secs). Boot with dtb in the ESP works, but gives the same result - so something in the kernel or initrd modules is off.

07:08 <steev> https://www.amazon.com/Lenovo-ThinkPad-Snapdragon-Unlocked-T-Mobile/dp/B0BVWKM463 "renewed" if anyone wants, but only 600 for the 512GB/16GB ram version

07:25 <jglathe_volterra> nice

12:40 <strongtz[m]> travmurav: I've always been using 0=windows

12:41 <strongtz[m]> I guess the linux option just provides a built-in dtb

12:45 <strongtz[m]> also I fail to find an older tcblaunch sent here (if necessary)

12:46 <travmurav[m]> this should have an old windows 10 one: https://oftc.irclog.whitequark.org/aarch64-laptops/2024-01-27#32866556

12:55 <strongtz[m]> uh, it gets stuck just like before

12:56 <travmurav[m]> yeah I guess it's something different then

12:56 <travmurav[m]> but I guess it's unlikely we reach the custom code in el2 so probably dies somewhere in the hyp

12:57 <travmurav[m]> and that thing is very broken on pre-gunyah hyp at least since any error causes a hyp crash

12:57 <travmurav[m]> and I can almost bet they ^C^V it with the same mistakes into the new hyp

12:58 <travmurav[m]> didn't look at it closely though

13:01 <strongtz[m]> let's see which oem is going to use this

13:03 <strongtz[m]> I did try hooking an rtx3060 via M.2->oculink today lmao

13:04 <strongtz[m]> however there seem to be something stuck in uefi, the internal screen won't light up anymore and the os never seemed to boot

13:13 <strongtz[m]> unfortunately I can't get the gpu working lol. There are some weird issues

13:14 <strongtz[m]> `Timeout waiting for GMU OOB set GPU_SET: 0x4`

13:33 <jenneron[m]> steev: thanks for testing the audio patch, i didn't get to it until now

13:33 <jenneron[m]> this should also partially fix flex 5g, but it still fails, so there is some another issue, will look into it as well

18:48 <jenneron[m]> it seems that there is another regression in 6.8

18:48 <jenneron[m]> on yoga c630

18:49 <jenneron[m]> i got speakers attached to soundwire on flex, but it fails with a slim bus error the same as on c630

19:00 <jenneron[m]> amplifiers*

19:04 <jenneron[m]> hmm, probably not slim bus

19:05 <jenneron[m]> this is the error https://dpaste.com/8JKCRWGG2

19:05 <jenneron[m]> the same as in steev's dmesg from c630

19:05 <jenneron[m]> it has to be between 6.7 and 6.8

19:06 <konradybcio> there have been some.. creative patches regarding slim and nullptrs lately

19:06 <steev> i'd check what was backported to 6.7.8

19:07 <steev> iirc, i see it on 6.7.8 but not 6.7.7

19:08 <steev> hm, no

19:08 <steev> 6.7.8 was only an ntfs patch

20:17 <jenneron[m]> i reverted soundwire to pre-6.8 state and the error is gone

20:17 <jenneron[m]> basically to this state https://github.com/torvalds/linux/blob/95b0f3aa71eb09d067d89395fdb7f2dd87feae18/drivers/soundwire/qcom.c

20:17 <jenneron[m]> but i can't get amplifiers in attached state on flex anymore

20:19 <jenneron[m]> jenneron[m]: i'm actually not sure, maybe it's a coincidence

20:19 <jenneron[m]> it would be better to bisect it on sdm845 instead

21:55 <steev> https://lore.kernel.org/all/20240419150140.92527-1-srinivas.kandagatla@linaro.org/