#asahi-dev · 2026-09-08 19:58
waltmckso I'm fairly sure that the 1.5x thing is just wrong at least for j413. However I have no idea for other devices#asahi-dev · 2026-09-08 19:58
waltmck(this is comparing two different reconstructions of the same voice recording)#asahi-dev · 2026-09-08 19:57
waltmckI just tested this changing the gains to be the same for all microphones, and it reduces background noise by rought 0.8dB without changing speech volume#asahi-dev · 2026-09-08 19:50
WindDogStudio[m]* I bricked it after trying to manually patch the already installed kernel, must have messed something up as it would get past grub, but not any further, and I would be sent to the linux recovery where all I could do is retry since admin/sudo access is locked. This happened around 5 minutes before I sent my first message earlier today. If anything, it was user error#asahi-dev · 2026-09-08 19:48
waltmckI took a 30 speech sample raw recording from all 3 mics, and they only differed by ~0.1 decibels. If this 1.5 constant were right (and mic 3 is less sensitive), we would expect to see a difference of 3.5 dB#asahi-dev · 2026-09-08 19:48
WindDogStudio[m]jannau: I bricked it after trying to manually patch the already installed kernel, must have messed something up as it wouldnt get past grub, and I would be sent to the linux recovery where all I could do is retry since admin/sudo access is locked. This happened around 5 minutes before I sent my first message earlier today. If anything, it was user error#asahi-dev · 2026-09-08 19:47
waltmckmic 3 gets a 1.5x amplitude boost relative to the other mics before the beamformer ever sees the output#asahi-dev · 2026-09-08 19:47
WindDogStudio[m]jannau: Sorry for the late response, just compiled the asahi-wip kernel. Audio works now! Here is asahi-diagnose after the wip kernel was installed https://winddogstudio.com/asahi-diagnose-20260908-124256.txt#asahi-dev · 2026-09-08 19:46
waltmckYeah that's true, but isn't that what the beamformer is supposed to do?#asahi-dev · 2026-09-08 19:46
chaos_princessthey are in a triangle pattern#asahi-dev · 2026-09-08 19:42
waltmcki.e. https://github.com/AsahiLinux/asahi-audio/blob/687d8cd0f6dc4b12fc0d9c3aebe360656de2882e/firs/j413/mic.json#L27#asahi-dev · 2026-09-08 19:42
waltmckWhy does mic 3 on j413 have 3/2 the gain of the other 2 mics?#asahi-dev · 2026-09-08 19:14
jannaumostly just being busy with the gpu driver for v7.2 and intending to merge it only for v7.2 based kernels#asahi-dev · 2026-09-08 19:12
sofusbtw is there a reason the pmp support for m3 is not merged yet?#asahi-dev · 2026-09-08 18:41
jannauwould be nice if you could re-test before I tag a new kernel#asahi-dev · 2026-09-08 18:37
jannauhow did you brick your install? i.e. how/when does it fail?#asahi-dev · 2026-09-08 18:37
jannauWindDogStudio[m]: fixed fix pushed to asahi-wip#asahi-dev · 2026-09-08 18:19
WindDogStudio[m]Alright, thank you#asahi-dev · 2026-09-08 18:16
jannauI commented on the other two issues in the PR#asahi-dev · 2026-09-08 18:14
WindDogStudio[m]jannau: True, I did have to change the "woofers" to the correct values for it to compile. Forgot to mention that, sorry!#asahi-dev · 2026-09-08 18:11
jannauWindDogStudio[m]: the change from PR 606 wasn't entirely correct and it didn't even build for me (as merged)#asahi-dev · 2026-09-08 17:56
myriathank you sven! I figured as much, but just wanted to check#asahi-dev · 2026-09-08 17:52
svenmyria: ^--#asahi-dev · 2026-09-08 17:52
svenhttps://lore.kernel.org/asahi/dbf17fa6-1af6-467b-8b3d-dca8476dc785@marcan.st/#asahi-dev · 2026-09-08 17:46
myriaspecifically there's some plists on macos describing perf events and what counters they have affinity on for all the generations and I wanted to check before I started reading them. For RE considerations#asahi-dev · 2026-09-08 17:46
myriaIs reading plist files from macos and transforming them into a kernel data structure fine?#asahi-dev · 2026-09-08 17:27
WindDogStudio[m]Hey, I'm the guy who first posted the j615 audio issue on reddit. I tried installing pull request 606, and it doesnt seem to work out of the box. All six speakers are discovered, but speakersafetyd repeated states “negative power / bad ivsense data”. One other thing I found is that there seems to also be a false 150°C temperature report. I would post asahi-diagnose, but I seem to have bricked the asahi install#asahi-dev · 2026-09-08 14:59
sofusmaybe? At the very least the addresses and bw-scratch need another offset#asahi-dev · 2026-09-08 12:41
fl0_id_right. you think it might work (with some changes) for t6031? like if it needs testing. otherwise I'll leave it for now#asahi-dev · 2026-09-08 12:34
sofusyeah, but only for t6030 and the internal display#asahi-dev · 2026-09-08 12:18
fl0_id_sofus I can probably try it later or tomorro. it just adds dcp right?#asahi-dev · 2026-09-08 11:35
sofushow are the disp_reserved_regions derived? or can i assume the other socs also work with t602x#asahi-dev · 2026-09-08 11:32
sofuswhops, should have asked. That works#asahi-dev · 2026-09-08 11:26
chaos_princessi've added an attempt to decode it there#asahi-dev · 2026-09-08 11:26
chaos_princesssofus: https://github.com/asahilinux/m1n1/blob/main/proxyclient/tools/pmp_assist.py#asahi-dev · 2026-09-08 11:24
sofusand in m1n1 using disp_reserved_regions_t602x worked#asahi-dev · 2026-09-08 11:23
sofusthe only thing i had to trace was apple,bw-scratch, the rest is t602x copy paste#asahi-dev · 2026-09-08 11:23
sofusits very hacky, but if anyone wants to use it as a base it can be found here: https://github.com/sofus13/linux/commit/2622d4feae95a3de615a55d1f9d2b43424ec1aa6#asahi-dev · 2026-09-08 11:22
chadmedfixed#asahi-dev · 2026-09-08 11:21
JamesCalligeros[m]Yes#asahi-dev · 2026-09-08 11:18
tryhard[m]chadmed: did u mean speaker_left_woof2 and not speaker_left_woo2f?#asahi-dev · 2026-09-08 10:56
chadmedalso re speakers https://github.com/AsahiLinux/linux/pull/606#asahi-dev · 2026-09-08 10:56
chadmedi hope theres not some weirdness we have to make set_digital_out_mode take longer to account for...#asahi-dev · 2026-09-08 10:54
sofusthanks#asahi-dev · 2026-09-08 10:54
sofusseems to work otherwise#asahi-dev · 2026-09-08 10:54
sofusnot really sure why the first swap fails, https://pastebin.com/n70dUvGG#asahi-dev · 2026-09-08 10:46
chadmedeyyyy#asahi-dev · 2026-09-08 10:46
chaos_princesslooks like it#asahi-dev · 2026-09-08 10:46
chaos_princessaudio-speaker-left-woofer-2 0x000009C40000003A audio-speaker-right-woofer-2 0x000009C40000003D#asahi-dev · 2026-09-08 10:45
chadmedi suspect left woofer 2 is going to be at 3a and right woofer 2 at 3d