#asahi-dev · 2023-04-16 12:05
marcan12.4 you mean?#asahi-dev · 2023-04-16 11:49
marcanOTOH it will need to be added to u-boot for MTP...#asahi-dev · 2023-04-16 11:49
marcanI'm just going to remove it from m1n1 and we'll see if anything breaks :p#asahi-dev · 2023-04-16 11:47
marcanbut u-boot doesn't implement it...?#asahi-dev · 2023-04-16 11:46
marcanhowever, I still see no messages in my old DCP logs#asahi-dev · 2023-04-16 11:45
marcanha.#asahi-dev · 2023-04-16 11:43
marcanit's not implented in u-boot though#asahi-dev · 2023-04-16 11:41
marcanso I have no idea where the logic came from#asahi-dev · 2023-04-16 11:41
marcanhowever it sends no messages#asahi-dev · 2023-04-16 11:41
marcannvme does have the endpoint#asahi-dev · 2023-04-16 11:40
marcanI don't see oslog in my most recent dcp log#asahi-dev · 2023-04-16 11:33
marcandoes anyone remember what we added oslog for? the current behavior seems patently wrong as far as MTP goes, but I have no idea why it's there at all#asahi-dev · 2023-04-16 11:21
marcanfixing that#asahi-dev · 2023-04-16 11:21
marcanjannau: also the mtp protocol init message parsing was wrong#asahi-dev · 2023-04-16 11:07
marcanwe need to actually implement that properly#asahi-dev · 2023-04-16 11:01
marcanjannau: it's oslog#asahi-dev · 2023-04-16 10:42
marcanI'll figure it out#asahi-dev · 2023-04-16 10:42
marcanjannau: yeah it's broken :/#asahi-dev · 2023-04-16 10:38
marcanI made it cover everything, looks like it's 1MiB#asahi-dev · 2023-04-16 10:34
marcanwait sec, sram first#asahi-dev · 2023-04-16 10:33
marcanjannau: you're right, it's broken :/#asahi-dev · 2023-04-16 09:57
marcanbut it's definitely broken#asahi-dev · 2023-04-16 09:57
marcanI suspect the clock shutdown might take a while or there is some other timing nuance that makes it work for my usual kernel builds, and also it works for me the first time on my repro kernel#asahi-dev · 2023-04-16 09:55
marcanpovik: https://github.com/AsahiLinux/linux/commit/93065f7023fbe738d733b4c5a96f87c653301c81 (in that branch because lazy)#asahi-dev · 2023-04-16 09:24
marcanand it's just broken, it pulls the clock before switching off so it never had a chance#asahi-dev · 2023-04-16 09:24
marcankettenis: it's a logic issue#asahi-dev · 2023-04-16 09:24
marcanpovik: we are disabling the MCA clocks before muting the codec#asahi-dev · 2023-04-16 08:58
marcangoing to try to fix it#asahi-dev · 2023-04-16 08:58
marcanand it is related to clocking and clock switching#asahi-dev · 2023-04-16 08:57
marcanpovik: reproduced the cs42l83 issue that altdistros are hitting. it clearly has to do something with kernel config, timing, or something like that.#asahi-dev · 2023-04-16 08:19
marcanfixed#asahi-dev · 2023-04-16 08:19
marcangood catch#asahi-dev · 2023-04-16 08:17
marcan*ADT+PMGR#asahi-dev · 2023-04-16 08:17
marcanjannau: confirmed on j493 ADT#asahi-dev · 2023-04-16 08:14
marcannor uses 200m, everything else uses 24#asahi-dev · 2023-04-16 08:14
marcanjannau: it should be 24m, that's cargo cult copypasta from the nor spi#asahi-dev · 2023-04-16 08:12
marcansame with the distro scripts that do updates#asahi-dev · 2023-04-16 08:11
marcankettenis: yup#asahi-dev · 2023-04-16 07:55
marcanpushed the SPI DTs to t602x/bringup#asahi-dev · 2023-04-16 07:40
marcansetting that one way will work with gpio, the other way with spi#asahi-dev · 2023-04-16 07:40
marcandid you set the pinmux properly?#asahi-dev · 2023-04-16 07:40
marcanhm#asahi-dev · 2023-04-16 07:40
marcanwhy?#asahi-dev · 2023-04-16 07:39
marcanI guess I might as well do spi2 and spi4#asahi-dev · 2023-04-16 07:39
marcananyway I confirmed the spi1 pins on t602x#asahi-dev · 2023-04-16 07:39
marcanonly CS since we do use that one differently from Apple (I think they use direct GPIO for that instead of the native CS?)#asahi-dev · 2023-04-16 07:39
marcanI think we don't actually need to set the extra pinmuxes#asahi-dev · 2023-04-16 07:17
marcanI think if you guess wrong it will still work but might mess something else up...#asahi-dev · 2023-04-16 07:13
marcanI need to do that for the NOR SPI on t602x...#asahi-dev · 2023-04-16 07:13
marcanmonitoring GPIO registers with regmon or gpiola while messing with SPI