#asahi-dev · 2021-12-14 12:29
marcan:D#asahi-dev · 2021-12-14 12:29
marcanonly this DART is listed under this IRQ in the ADT#asahi-dev · 2021-12-14 12:28
marcanI expected breakage but not *this*#asahi-dev · 2021-12-14 12:27
marcanI'm trying to get the audio jack to work on t6000... and getting an unhandled IRQ... on the DART fault handler IRQ?!#asahi-dev · 2021-12-14 12:27
marcanah yes :)#asahi-dev · 2021-12-14 12:25
marcanwe can do that if you want#asahi-dev · 2021-12-14 12:25
marcanah, the MAINTAINERS thing? don't think so#asahi-dev · 2021-12-14 12:12
marcan(if it's a new entry it doesn't need to be a separate commit)#asahi-dev · 2021-12-14 12:11
marcanre MAINTAINERS, you can add a new entry and volunteer yourself, or you can shove it under the apple SoC stuff; depends on how much you want to commit ;)#asahi-dev · 2021-12-14 12:11
marcanpovik: MCA should probably pull in ADMAC in the kconfig, otherwise it's pretty pointless#asahi-dev · 2021-12-14 11:23
marcanI imagine if refclk changes (though I get the feeling it never will) they'd change all the multipliers anyway#asahi-dev · 2021-12-14 11:23
marcanif we knew the full clocktree I would absolutely do it that way... but we don't :(#asahi-dev · 2021-12-14 11:22
marcannp#asahi-dev · 2021-12-14 11:22
marcankinda harder to do that with ratios :)#asahi-dev · 2021-12-14 11:22
marcanI think we're better off with fixed frequency clocks since we don't know; that way, if something ends up being dynamic or changing, we can teach m1n1 to propagate frequencies from the ADT#asahi-dev · 2021-12-14 11:21
marcanalso, is nco_inp just a guess on the clocktree based on the frequency?#asahi-dev · 2021-12-14 11:21
marcanyeah#asahi-dev · 2021-12-14 11:14
marcanbut you don't have to worry about that for now#asahi-dev · 2021-12-14 11:14
marcanI guess that will make mca the first driver to actually do that with pmgr#asahi-dev · 2021-12-14 11:14
marcanpovik: https://github.com/povik/linux/commit/f5685ae510d5e9cce34c66666e9b5d335c99e848 FWIW since these are actually separate domains intended to be driven independently, you'd want "proper" PD support eventually (which involves a funny hack the PM code does for you of making up virtual devices)#asahi-dev · 2021-12-14 11:06
marcan:D#asahi-dev · 2021-12-14 11:05
marcansven: maybe send a revision to that soonish to poke things?#asahi-dev · 2021-12-14 10:09
marcanlet me know :)#asahi-dev · 2021-12-14 10:09
marcanI think Jannau can point to twitter, can't remember if povik has one too?#asahi-dev · 2021-12-14 10:03
marcanpovik: the userspace stuff is a whole different story anyway (there's profiles and stuff, we'll probably have to create some on top of figuring out what to do for the speaker crossovers)#asahi-dev · 2021-12-14 10:02
marcankettenis: ah yes, totally forgot about that, will do!#asahi-dev · 2021-12-14 09:54
marcan(table will change based on what I do with the hid and audio stuff today :))#asahi-dev · 2021-12-14 09:54
marcansince it has the fancy table you'll probably want to run it locally; just install hugo and `hugo serve` at the repo root#asahi-dev · 2021-12-14 09:53
marcanWIP report is here if anyone wants to look (still not proofread): https://github.com/AsahiLinux/AsahiLinux.github.io/commit/fbfa2ed93d040b20bd984d1e5d1dff8699820fcb#asahi-dev · 2021-12-14 09:52
marcanawesome, I'll give it a shot after dinner :)#asahi-dev · 2021-12-14 09:44
marcanI also noticed the Touch Bar multitouch is also ostensibly HID over SPI, but possibly a different protocol? unclear#asahi-dev · 2021-12-14 09:43
marcan_jannau_: btw, I do have access to an older Intel MacBook with SPI touchpad, so I can test that when we get there#asahi-dev · 2021-12-14 09:38
marcanI think he means hid-generic works in multitouch *mode* because it still picks up the first few bytes of the reports?#asahi-dev · 2021-12-14 09:28
marcanI basically want linux-asahi to be a testing ground for drivers headed in the right direction, even if they're rough, at least at this point in the game#asahi-dev · 2021-12-14 09:27
marcanpovik: also happy to merge in the audio work if you think that's okay#asahi-dev · 2021-12-14 09:24
marcanjannau: if it works well enough to use I might just merge it into asahi in its current state, might as well :)#asahi-dev · 2021-12-14 09:23
marcan(well, not green, green is for upstreamed drivers, but you get the idea)#asahi-dev · 2021-12-14 09:22
marcanI'm writing the progress report, so if you think the jack should trivially work everywhere I might as well test it and color that entire row green (yes I'm doing a table thing this time)#asahi-dev · 2021-12-14 09:22
marcanpovik: awesome! remind me again, what's the state of the audio stuff? works on the mini, have we tested the headphones jack or speakers on anything else?#asahi-dev · 2021-12-13 12:03
marcan_jannau_: clearly we need to do this: https://hub.marcan.st/t/better_apple_chime.mp3#asahi-dev · 2021-12-13 11:35
marcandefinitely not :p#asahi-dev · 2021-12-13 11:33
marcanotherwise they'd use it for iBoot too#asahi-dev · 2021-12-13 11:33
marcanvery unlikely they'd be doing it in hardware#asahi-dev · 2021-12-13 11:33
marcanpovik: iBoot has pre-filtered boot chime samples with channels for each speaker, so yeah#asahi-dev · 2021-12-13 10:47
marcanthat should all be done in userspace, of course; not sure if all the ALSA/PA profile thing can handle EQ/convolution filters, but if it can't, well, maybe it's time to start that conversation#asahi-dev · 2021-12-13 10:47
marcanmy idea is to grab an impulse response of what macOS does :-)#asahi-dev · 2021-12-13 10:47
marcanpovik: the multispeaker thing needs some kind of filter/crossover thing#asahi-dev · 2021-12-13 08:06
marcanit does a lot of stuff :)#asahi-dev · 2021-12-13 08:06
marcannot surprising#asahi-dev · 2021-12-13 07:42
marcanso that checks out