#asahi-dev · 2021-11-11 09:36
marcandid we actually break i2c? I thought we tested this#asahi-dev · 2021-11-11 09:36
marcanmaz: is this the X1000?#asahi-dev · 2021-11-11 02:06
marcanpovik: it says 2 hours but then ends up taking 30-40 minutes#asahi-dev · 2021-11-11 02:06
marcanfor the hypervisor you want a proper secondary macos install#asahi-dev · 2021-11-10 18:50
marcanwe don't *want* to emulate anything macOS does layers above that (other than at extremely general publicly documented design levels *if and only if* they make sense)#asahi-dev · 2021-11-10 18:49
marcanthen we can build whatever we want on top#asahi-dev · 2021-11-10 18:49
marcanyeah, the important bit is figuring out the hardware#asahi-dev · 2021-11-10 18:46
marcanmuch safer to work out the binaries by differential analysis and such#asahi-dev · 2021-11-10 18:46
marcanbe particularly wary of looking at the userspace driver, as that's where you get tons of logic; we do have a hard ban on looking at the userspace GPU driver for this reason.#asahi-dev · 2021-11-10 18:45
marcanthose are two fairly disconnected aspects/drivers#asahi-dev · 2021-11-10 18:45
marcanbut nobody has looked at the ANS interface#asahi-dev · 2021-11-10 18:45
marcangeohot and some others looked at some stuff#asahi-dev · 2021-11-10 18:45
marcanyou're familiar with the existing userspace side ANE reversing?#asahi-dev · 2021-11-10 18:45
marcanyup, that's the general plan with anything#asahi-dev · 2021-11-10 18:44
marcan(also, for register names that show up in strings I prefer to rewrite them too)#asahi-dev · 2021-11-10 18:44
marcanI hope so, but you know...#asahi-dev · 2021-11-10 18:43
marcanah yeah, strings OK in my book, as you're largely just seeing log messages and such; just don't go around copying function names or something for no reason#asahi-dev · 2021-11-10 18:41
marcanyeah, especially if you want to get the big picture (which is where you run into copyright issues too)#asahi-dev · 2021-11-10 18:40
marcansquags_: I think ANS is largely fronted by an ASC, so you should be able to use the ASC tracing functionality in the hypervisor (and just write an ANS module as you learn more about it)#asahi-dev · 2021-11-10 18:39
marcanconversely, I worked out all of AIC/AICv2 without any disasm (though I might peek at the AICv2 driver a bit later only to confirm how the capability registers are interpreted; again a very specific thing that yields non-copyrightable information about the hardware)#asahi-dev · 2021-11-10 18:37
marcanso I ask that people don't disasm unless I am familiar with how they work, because I've been burned by that before#asahi-dev · 2021-11-10 18:36
marcane.g. I did some kext disasm the past two days because I was looking for very specific bits of information that I knew were not copyrightable (and would've taken me a significant detour to get with the hypervisor)#asahi-dev · 2021-11-10 18:35
marcandisassembling isn't entirely banned outright, but I can't blindly trust that people will know how to do that without getting into trouble#asahi-dev · 2021-11-10 18:34
marcansquags_: basically yes. you should probably use the hypervisor to work out the hardware interface.#asahi-dev · 2021-11-10 17:47
marcanthe moved all the interesting bits to the KDK :/#asahi-dev · 2021-11-10 17:47
marcannot like there's anything terribly interesting there these days...#asahi-dev · 2021-11-10 17:09
marcanif what macos does is what the linux code says... then I wouldn't be surprised if macos is wrong :D#asahi-dev · 2021-11-10 17:09
marcan(i.e. that other bit turns on)#asahi-dev · 2021-11-10 17:08
marcanactually I say "it's what macos does" but rather, it's what I see in the mmiotrace#asahi-dev · 2021-11-10 17:08
marcanalso I have no idea if this does anything useful, but it's another thing macos does that stood out a bit: https://github.com/AsahiLinux/linux/commit/3a623fc4d1bb996e9412f19d322703e98679077c#asahi-dev · 2021-11-10 17:08
marcanI *think* that is correct, at least it's what I see in the mmiotrace#asahi-dev · 2021-11-10 17:07
marcanmaz: fwiw: https://github.com/AsahiLinux/linux/commit/05bed0e4f3a921581ad236d7659707d7193cbe2c#asahi-dev · 2021-11-10 16:20
marcan(force pushed because it was the latest commit anyway)#asahi-dev · 2021-11-10 16:20
marcanfirefox317: whoops, no, fixed#asahi-dev · 2021-11-10 14:02
marcanno idea why#asahi-dev · 2021-11-10 14:02
marcankettenis: I think "sync" just didn't work in busybox for some reason#asahi-dev · 2021-11-10 07:40
marcanthough I do think I found and fixed a bug#asahi-dev · 2021-11-10 07:39
marcanmaz: good news, the linux pci side doesn't actually need any changes for m1pro after all#asahi-dev · 2021-11-10 00:29
marcankettenis: really? this is a genesys chip... did they just use a designware SD controller inside?#asahi-dev · 2021-11-09 12:05
marcanbut usually there are at least some controls#asahi-dev · 2021-11-09 12:05
marcanreading levels/status at least#asahi-dev · 2021-11-09 11:56
marcanbattery management#asahi-dev · 2021-11-09 11:55
marcanwell, there are some obvious ones like GPIO, BMIC#asahi-dev · 2021-11-09 11:49
marcanit's not much different from e.g. some of the power management schemes used on other SoCs#asahi-dev · 2021-11-09 11:49
marcanI mean a sub-node#asahi-dev · 2021-11-09 11:49
marcansure, but why not have a node?#asahi-dev · 2021-11-09 11:45
marcanand enable/disable things depending on the device config#asahi-dev · 2021-11-09 11:44
marcanthat would allow us to have different compatibles for different features#asahi-dev · 2021-11-09 11:44
marcanthen SMC just does the scan#asahi-dev · 2021-11-09 11:44
marcancould just be a bunch of sub-nodes of SMC, not declared as a bus