#asahi-dev · 2023-01-16 15:14
marcanwhy do we care about LEAP even? if we're bypassing things, can't we just get the raw mic data before LEAP?#asahi-dev · 2023-01-16 15:14
marcanthat depends on the specific hardware involved though#asahi-dev · 2023-01-16 15:13
marcane.g. consider what happens if AOP starts talking to other firmwares or doing something special#asahi-dev · 2023-01-16 15:13
marcanpersonally I'm more of the opinion that we should do whatever macOS does, even if it involves firmware, because diverging from that is more likely to come crashing down in bad ways with updates#asahi-dev · 2023-01-16 15:12
marcanpovik: AIUI AOP also reports accelerometer data on at least some models#asahi-dev · 2023-01-15 14:42
marcangetting to the point where I'd like to ship 6.2, do we have any other known regressions?#asahi-dev · 2023-01-15 14:41
marcanit works though, tested my M1 MBA through suspend/resume etc without issue#asahi-dev · 2023-01-15 14:27
marcanalso not sure if the NACKed byte counts for the increment or not, the datasheet is unclear#asahi-dev · 2023-01-15 14:27
marcanI *think* the counting is such that start/repeat-start each count as 1 byte, separate from the address byte, but I'm not 100% sure#asahi-dev · 2023-01-15 14:27
marcanpovik: the remaining errors I see for cs42l83 on startup are I2C NACKs (-6). just pushed a patch so it'll print xfsta, the bottom byte of which is the byte which caused the NACK.#asahi-dev · 2023-01-15 14:13
marcanyes#asahi-dev · 2023-01-15 14:11
marcan(tap-release-tapandhold power, relatively quickly)#asahi-dev · 2023-01-15 14:11
marcantried the fallback rOS boot gesture?#asahi-dev · 2023-01-15 14:11
marcanI'm not even sure that is supposed to happen if you just screw up the SSD#asahi-dev · 2023-01-15 14:11
marcanthat sounds very, very broken...#asahi-dev · 2023-01-15 14:10
marcanno logo even?#asahi-dev · 2023-01-15 14:10
marcanwhat happens on boot now?#asahi-dev · 2023-01-15 14:10
marcanbest guess, the empty partition slots broke it (macOS never does that)#asahi-dev · 2023-01-15 14:10
marcanyou didn't touch the system recovery partition at the end?#asahi-dev · 2023-01-15 14:07
marcanjeffmiw: how did you make space for resizing the partition in the first place? parted doesn't know how to resize APFS...#asahi-dev · 2023-01-15 14:06
marcan(usbmuxd, libimobiledevice, idevicerestore)#asahi-dev · 2023-01-15 14:06
marcanit should work, modulo you need a bunch of git versions#asahi-dev · 2023-01-15 13:57
marcankernels are weird things#asahi-dev · 2023-01-15 13:56
marcancould just be a random timing difference#asahi-dev · 2023-01-15 13:55
marcanyeah it was just the early enable that was breaking it#asahi-dev · 2023-01-15 13:46
marcanbut everything above that is a giant mess of aliasing#asahi-dev · 2023-01-15 13:46
marcansince this happens every 8 input samples, the overall playback rate is ~correct for f << 48kHz/8 = 6kHz -> 0-3kHz, which means audio is recognizably the right pitch#asahi-dev · 2023-01-15 13:45
marcansince the core was running too fast, the elastic buffer keeps underrunning and that creates 0-samples on the output#asahi-dev · 2023-01-15 13:45
marcanas for what happened: the PLL was running unlocked/too fast, by a factor of 1/8 or 1/4 or thereabouts. there's probably an 8-sample elastic buffer between the input sample data (clocked by the input clock) and the core.#asahi-dev · 2023-01-15 13:31
marcancs42l42 doesn't have the same bug, but I can imagine some codepaths where it would go wrong. I'm going to add a defensive patch.#asahi-dev · 2023-01-15 13:28
marcanmps: I fixed it, no more explanations necessary#asahi-dev · 2023-01-15 13:25
marcanI'm guessing something about the sequence of operations changed in 6.2 and that's what broke it#asahi-dev · 2023-01-15 13:24
marcanjannau, chadmed, povik: fix in bits/070-audio and asahi-6.2-rc3-4, doing an arch build now#asahi-dev · 2023-01-15 13:19
marcanyeah I think that's it#asahi-dev · 2023-01-15 13:17
marcanor maybe the problem is just enabling it early, let's see#asahi-dev · 2023-01-15 13:15
marcanwe need to configure the PLL when the refclk is up#asahi-dev · 2023-01-15 13:15
marcanpovik: I don't think configuring the PLL in hw setup, then just turning it on on unmute is legit#asahi-dev · 2023-01-15 13:15
marcanfixed it.#asahi-dev · 2023-01-15 13:02
marcanjannau: 0x20 is the error flag#asahi-dev · 2023-01-15 12:22
marcanthe resulting aliasing seems to be slightly jittery, which suggests a PLL issue yeah#asahi-dev · 2023-01-15 12:22
marcantook a capture, I see every Nth sample ish getting replaced by zero, where N is circa 8 (might be more than one sample, hard to tell in analog)#asahi-dev · 2023-01-15 11:59
marcanpovik, chadmed: listening to the distortion, this sounds exactly like terribad resampling/aliasing#asahi-dev · 2023-01-15 11:32
marcanI'm going to add a bit more logging to i2c after the unjam fix, in case it's useful to debug. That should at least log jams (and other error conditions) before the start of the next transaction, even if unjam succeeded.#asahi-dev · 2023-01-15 11:23
marcanpovik: ^#asahi-dev · 2023-01-15 11:22
marcanthis all points towards clocking issues in general#asahi-dev · 2023-01-15 11:22
marcanIIRC the codecs use the audio clock as their internal clock when active, so the audio clock being screwed up would explain the i2c bus getting jammed#asahi-dev · 2023-01-14 11:17
marcanwhich is what makes it sound thick without the fundamental#asahi-dev · 2023-01-14 11:17
marcanchadmed: right, after saturation#asahi-dev · 2023-01-14 11:16
marcanmps: https://joeysturgistones.com/blogs/learn/the-beginners-guide-to-multi-band-processing#asahi-dev · 2023-01-14 11:15
marcanit's for the "we are alive -> reset" path