#asahi-dev · 2026-07-30 11:22
mildsunriseyeah, sorry, it's been so long that I didn't even realize we had a glue driver now, and was still looking for the patches in core.c/drd.c#asahi-dev · 2026-07-30 11:20
mildsunriseahhh it's been moved, txh#asahi-dev · 2026-07-30 11:19
mildsunrisequestion, why are we no longer resetting the dwc3 between role changes? i no longer see sven's patch in asahi{,-wip}#asahi-dev · 2026-06-05 21:09
mildsunriseyuka: yes, the spmi controller driver is fair game, but hpm/tps9598x you should stay clear... the docs (https://asahilinux.org/docs/hw/peripherals/ace3/) should be enough to implement the spmi transport#asahi-dev · 2026-02-10 11:16
mildsunriseoki, good to know :) spmi has interrupt-controller + interrupt in the DT, right?#asahi-dev · 2026-02-10 10:14
mildsunriseyou might wanna grab the patches from the scratchpad branch, since i rebased them on top of asahi#asahi-dev · 2026-02-10 10:14
mildsunrisenoopwafel: hey sry i've been struggling to keep up with irc lately. i was planning on preparing those spmi patches for lkml submission but i seem to be too depressed to get myself to do that, so feel free to! did you find any issues with them or is it just style issues and the like?#asahi-dev · 2026-02-03 14:56
mildsunrisefl0_id: i think i just disabled it for now and didn't look more into it, sorry :(#asahi-dev · 2026-02-03 14:27
mildsunrisedon't look, cherrypick etc#asahi-dev · 2026-02-03 14:26
mildsunrisenoopwafel: sure! please don't look at the patches that begin with [TAINTED] though, since those were written with decomp knowledge#asahi-dev · 2026-02-03 14:23
mildsunrisedevice role has a few issues, most concerning of which is that the dwc3 apparently shits its pants when you talk on an interrupt endpoint#asahi-dev · 2026-02-03 14:19
mildsunrisehost role*#asahi-dev · 2026-02-03 14:19
mildsunrise can confirm that usb2 works in m3 (with the mentioned spmi patches and tipd regmap) but only doesn't role #asahi-dev · 2026-01-27 10:01
mildsunriseyes, it's not a complicated regmap to implement... the interrupts also go through spmi but that will work out of the box with the new driver#asahi-dev · 2025-12-09 09:59
mildsunriseis there any reason we're matching by name and not compatible?#asahi-dev · 2025-12-09 09:58
mildsunrisee.g. https://github.com/AsahiLinux/m1n1/commit/2f03e076af9533daf58809578569c4adee9fc5e4#asahi-dev · 2025-12-09 09:57
mildsunrisejannau: regarding https://github.com/AsahiLinux/m1n1/pull/518/commits/1cbcddbae14b9cf5bc036c92cd424f76cd526ca6 as far as i can see, the layout of the pmu nodes in m3+ is unchanged so I was going to suggest re-using the parsing rules we already have for them (we just need to change to match by compatible since the names have gotten too messy)#asahi-dev · 2025-12-09 09:55
mildsunriseintegralpilot: nice progress!#asahi-dev · 2025-11-30 20:39
mildsunriseah oki, it's consistent then#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 20:29
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 20:26
mildsunrisehey jannau sven, looks like there's a missing "0x" here? https://github.com/AsahiLinux/m1n1/blob/0b23457587c550b26cb3f2795333c1087c4fe5d7/src/usb.c#L289#asahi-dev · 2025-09-24 21:51
mildsunriseoki, thanks anyway#asahi-dev · 2025-09-24 21:30
mildsunrisem1n1 wouldn't hit that bc it doesn't use interrupt endpoints, but most gadgets in linux should#asahi-dev · 2025-09-24 21:28
mildsunrise(by freeze i mean that it stops processing any further transfers)#asahi-dev · 2025-09-24 21:27
mildsunriseunrelated but another issue i'm seeing in m3 is that the dwc3 seems to freeze whenever I attempt an interrupt transfer in device role. i'm also wondering if that happens in m1/m2 or it's an m3 original#asahi-dev · 2025-09-24 21:23
mildsunrisewow, that's good to know#asahi-dev · 2025-09-24 21:22
mildsunriseyou mean the usb-c controller firmware? isn't that a signed blob?#asahi-dev · 2025-09-24 21:21
mildsunrisesad#asahi-dev · 2025-09-24 21:20
mildsunrisethere is no known way to force it, even if done before plugging the cable?#asahi-dev · 2025-09-24 21:18
mildsunriseah, okay, good to know#asahi-dev · 2025-09-24 21:17
mildsunrisein the logs I can see discussion about forcing a particular role, but i understood that to be at the dwc3/atc level#asahi-dev · 2025-09-24 21:17
mildsunrisedoes someone know if telling the USB-C controller to renegotiate a particular data role (e.g. echo device > data_role) works on M1/M2?#asahi-dev · 2025-02-14 15:14
mildsunriseoh okay, got it :) will keep in mind#asahi-dev · 2025-02-14 13:41
mildsunrisenickchan: huh, i see... does vhe affect clock freq?#asahi-dev · 2025-02-13 19:29
mildsunrise(also avoid pmgr, as it's the other thing i did binary RE on, but i don't think i made any patches to that)#asahi-dev · 2025-02-13 19:28
mildsunriseon my fork please avoid HPM / tps6598x. those were developed with binary RE, they are for personal consumption and should not be upstreamed. i published docs instead https://github.com/AsahiLinux/docs/wiki/HW:ACE3