#asahi-dev · 2025-12-02 15:50
nickchanintegralpilot: i think that's a guest trap ISS, how about you check AFSR1_EL2?#asahi-dev · 2025-12-02 15:41
nickchanhm okay ISS is 0x20#asahi-dev · 2025-12-02 15:39
nickchanintegralpilot: all apple ISA extension exceptions uses EC 0x3f#asahi-dev · 2025-12-01 03:55
chadmed__so having only one codec on each half of the bus pull it down while the other half has an active codec is the correct approach#asahi-dev · 2025-12-01 03:54
chadmed__we cant have left woofer 1 pull the bus down while left tweeter is trying to transmit#asahi-dev · 2025-12-01 03:54
chadmed__using the NOT of the tx mask probably wont work since then there will be bus conflicts on each side#asahi-dev · 2025-11-30 23:44
SpencerJust installed Asahi Debian! Is there any good resources that document the SEP besides the Asahi Docs?#asahi-dev · 2025-11-30 20:39
mildsunriseah oki, it's consistent then#asahi-dev · 2025-11-30 20:36
sven*on m1#asahi-dev · 2025-11-30 20:36
svenyou only need to switch the power state for host mode even in m1 iirc#asahi-dev · 2025-11-30 20:35
mildsunrise(well they are in state 7 at boot, which seems to force it to device role)#asahi-dev · 2025-11-30 20:32
mildsunrisejannau: maybe in M1/M2 the hpms are in powered state already? in M3 they certainly are not#asahi-dev · 2025-11-30 20:30
mildsunriseahhh, yeah fair enough#asahi-dev · 2025-11-30 20:30
jannau`name[3] - '0';`#asahi-dev · 2025-11-30 20:30
chaos_princessuse character literal instead of 0x30, that way i don't have to take extra time to understand what that code was supposed to do :P#asahi-dev · 2025-11-30 20:30
jannaudoesn't seem to cause issue although I think I've seen i2c power error messages#asahi-dev · 2025-11-30 20:29
mildsunrisechaos_princess: i'm afraid i don't quite follow ^^'#asahi-dev · 2025-11-30 20:28
mildsunriseAFAIU this means no hpm will survive the "idx >= USB_IODEV_COUNT" check, so we aren't powering up the hpms, right? i'm surprised nobody complained though#asahi-dev · 2025-11-30 20:28
chaos_princessi'd even argue for replacing with '0'#asahi-dev · 2025-11-30 20:26
mildsunrisehey jannau sven, looks like there's a missing "0x" here? https://github.com/AsahiLinux/m1n1/blob/0b23457587c550b26cb3f2795333c1087c4fe5d7/src/usb.c#L289#asahi-dev · 2025-11-30 13:31
jannaubut I'll have to fight with life time for the rtkit buffer anyway#asahi-dev · 2025-11-30 13:31
jannaurtkit probably should have used `struct iosys_map`#asahi-dev · 2025-11-30 13:16
chadmed_i havent tested making the idle slot config the NOT of the tx mask but that might also work if the soc is ORing the lines anyway. the zeroes will do nothing and it will slightly simplify our implementation#asahi-dev · 2025-11-30 12:31
chadmedyoull see why its named like that im sure :P #asahi-dev · 2025-11-30 12:30
sveni already like the branch name :>#asahi-dev · 2025-11-30 12:30
chadmedill clean it up for submission during the week perhaps#asahi-dev · 2025-11-30 12:30
chadmedhttps://github.com/chadmed/linux/tree/tdm-idle-clusterfuck should work on j314 and j316 _only_ if anyone wants to test/check #asahi-dev · 2025-11-30 12:23
sven\o/#asahi-dev · 2025-11-30 12:22
chadmedand is generic and will work with any combination of machine driver and codec so long as the codec driver supports the op#asahi-dev · 2025-11-30 12:21
chadmedthe important thing is it works :P#asahi-dev · 2025-11-30 12:21
chadmedwell, at least i hope its saner#asahi-dev · 2025-11-30 12:21
chadmedi finally found a saner way to handle the idle tdm slot config#asahi-dev · 2025-11-29 17:54
jannautoo late to avoid these 2 issues. complex macros like try_init! are always a pain to debug#asahi-dev · 2025-11-29 17:46
chaos_princess"too late" was unclear to me, you meant we should have merged it before 6.18?#asahi-dev · 2025-11-29 17:30
jannau1. one of the Globals structs is "packed" resulting in reference to potentially unaligned fields#asahi-dev · 2025-11-29 17:29
jannau1. a variable with the same name as a field is used twice, changing its type from u32 to &u32#asahi-dev · 2025-11-29 17:28
jannauwe're too late with moving agx init to m1n1. https://lore.kernel.org/rust-for-linux/20250905140047.3325945-1-lossin@kernel.org/ in 6.18 breaks in asahi in 2 different ways#asahi-dev · 2025-11-29 12:17
integralpilotalso does anyone know what ESR_EL1 = 0xfe000020, it's implementation-defined so means some kind of apple-specific exception#asahi-dev · 2025-11-29 12:13
integralpilotSP_EL1 is a higher-half address that isn't mapped by the hypervisor to any physical address#asahi-dev · 2025-11-29 12:10
integralpilothmmm after some further digging i'm a bit more confused#asahi-dev · 2025-11-29 11:37
integralpilotcurrently looking into fleh_dispatch64_common more closely rn. fleh = first level exception handler, iirc?#asahi-dev · 2025-11-29 11:36
integralpilotit was near the LR so i suspected it was involved, but looks like some form of lookup table that the code that causes the issue gets data from#asahi-dev · 2025-11-29 11:33
integralpiloti think the "sel instruction" is actually in a data section after disassembling further sorry, so it might not be an instruction at all#asahi-dev · 2025-11-29 11:31
mazcould well be. I wouldn't be surprised if they had started with some embryonic SME hacking way before M4.#asahi-dev · 2025-11-29 11:30
jannaucould it be related to apple's vendor AMX halfway transitioning to SME?#asahi-dev · 2025-11-29 11:29
mazdo you have a SMSTART before that?#asahi-dev · 2025-11-29 11:29
integralpilotand this is the kernel variant for an M3 chipset#asahi-dev · 2025-11-29 11:28
integralpilotthe instruction is sel z0.d, p8, z8.d, z21.d#asahi-dev · 2025-11-29 11:27
mazwell, the registers are.#asahi-dev · 2025-11-29 11:27
mazthey are the same instructions.