#asahi-dev · 2021-10-03 13:07
marcanit's all kinda weird#asahi-dev · 2021-10-03 13:07
marcanyeah, I guess it gets presented as a "fake" nvram to the firmware?#asahi-dev · 2021-10-03 13:06
marcanyou just need to set up the right pointer#asahi-dev · 2021-10-03 13:06
marcanm1n1 already does that (thanks to the patch by kettenis)#asahi-dev · 2021-10-03 13:06
marcanthere's the whole alias dance to set the mac#asahi-dev · 2021-10-03 13:05
marcanit's a child of the pcie bus node#asahi-dev · 2021-10-03 13:05
marcansee how the ethernet mac gets set#asahi-dev · 2021-10-03 13:05
marcanbased on pcie port#asahi-dev · 2021-10-03 13:05
marcanyou add it#asahi-dev · 2021-10-03 13:04
marcan:')#asahi-dev · 2021-10-03 13:03
marcanI'm not saying I don't like their code out of NIH syndrome, you know :p#asahi-dev · 2021-10-02 02:26
marcanj_ey: ASoC is right#asahi-dev · 2021-09-24 12:50
marcanor just refuse to run really, or even just auto-chainload#asahi-dev · 2021-09-24 12:49
marcanalso I really need to add a "get version" thing to m1n1 and a big fat blinking red text warning to the HV on mismatch#asahi-dev · 2021-09-24 12:49
marcanit was a dev kernel assert that I'd fixed for UP but borked on SMP by accident, and never tested on dev since then#asahi-dev · 2021-09-24 12:49
marcansven: I fixed the hv thing#asahi-dev · 2021-09-24 01:44
marcannevermidn me#asahi-dev · 2021-09-24 01:44
marcanno, I am doing that,wtf#asahi-dev · 2021-09-24 01:43
marcanohh wait#asahi-dev · 2021-09-24 00:58
marcanI believe so#asahi-dev · 2021-09-24 00:52
marcanso it's not a regression, it's something broken there in SMP support#asahi-dev · 2021-09-24 00:52
marcanalso it's breaking when spinning up the extra CPUs#asahi-dev · 2021-09-24 00:52
marcanand with linux#asahi-dev · 2021-09-24 00:52
marcanworks for me on newer versions though :/#asahi-dev · 2021-09-24 00:51
marcanthat's like a stack overflow or something#asahi-dev · 2021-09-24 00:51
marcanurgh#asahi-dev · 2021-09-24 00:49
marcancan you also p.hv_translate(hv.ctx.sp[1], True) #asahi-dev · 2021-09-24 00:47
marcaneither that or something's really bugged in the HV#asahi-dev · 2021-09-24 00:46
marcanthat looks like the stack pointer went bananas and ended up in IO space#asahi-dev · 2021-09-24 00:45
marcan11.4 no third digit?#asahi-dev · 2021-09-24 00:45
marcanwhat are you running again, exactly?#asahi-dev · 2021-09-24 00:45
marcanalso uh that's weird#asahi-dev · 2021-09-24 00:44
marcanbacktrace?#asahi-dev · 2021-09-24 00:43
marcanbut I'm not sure#asahi-dev · 2021-09-24 00:43
marcanso the API is more like "driver front side", not a cut down the middle#asahi-dev · 2021-09-24 00:43
marcanand I think in the case of those they moved entire drivers outright#asahi-dev · 2021-09-24 00:43
marcanat least I don't think it has the multichannel asynchronous callback stuff#asahi-dev · 2021-09-24 00:43
marcanI hope#asahi-dev · 2021-09-24 00:43
marcan(but at least I think it's the same for all the *other *eps)#asahi-dev · 2021-09-24 00:42
marcanit's a completely different RPC to the main dcpep#asahi-dev · 2021-09-24 00:42
marcanI worked out the ring buffer format for the other endpoints (see the tracer), but I never looked at the actual messages beyond a cursory glance#asahi-dev · 2021-09-24 00:42
marcannot everything has or will have an EDID#asahi-dev · 2021-09-24 00:42
marcanbut also someone should fix GNOME#asahi-dev · 2021-09-24 00:41
marcanas in i2c over one of the other endpoints#asahi-dev · 2021-09-24 00:41
marcanalyssa: the i2c stuff should AIUI#asahi-dev · 2021-09-23 06:35
marcanis in UTC+9#asahi-dev · 2021-09-21 13:07
marcanyeah...#asahi-dev · 2021-09-21 11:27
marcansuggestions welcome :)#asahi-dev · 2021-09-21 11:27
marcanit was inspired by the clock thing but it doesn't really fit#asahi-dev · 2021-09-21 11:27
marcankettenis: yeah, I'm not sure what to call that