#asahi-re · 2022-06-03 17:12
marcanthe uncore one, yeah, I guess that needs its own test/compatible check#asahi-re · 2022-06-03 17:11
marcanthen it won't try to use them for <8103#asahi-re · 2022-06-03 17:11
marcanthe SYS_IMP_APL_IPI_SR_EL1 thing is a bug at this point, the static_branch_likely(&use_fast_ipi) should be moved outside that if so it doesn't even try to touch that register if Fast IPIs are not in use#asahi-re · 2022-06-03 17:09
marcannot the FIQ side#asahi-re · 2022-06-03 17:09
marcansure, but the AIC side itself works fine?#asahi-re · 2022-06-03 17:07
marcandoes the AIC driver work without changes?#asahi-re · 2022-06-03 17:07
marcanthough I think that should boot in device mode for DFU to work, if the AP doesn't tell it to do anything special#asahi-re · 2022-06-03 17:06
marcanon those devices#asahi-re · 2022-06-03 17:06
marcanthere's also the lightning controller involved, Tristar or whatever, right?#asahi-re · 2022-06-03 17:04
marcannot like that#asahi-re · 2022-06-03 17:04
marcanit isn't#asahi-re · 2022-06-03 17:04
marcanright now it sounds like you have two hosts trying to talk to each other and falsely detecting each other as a full/high speed device for whatever reason#asahi-re · 2022-06-03 17:04
marcanyou need to put it in gadget mode#asahi-re · 2022-06-03 17:04
marcanthat sounds like Linux is treating it as a *host* port, not a *device* port#asahi-re · 2022-06-03 17:01
marcanlinux has a dwc2 driver#asahi-re · 2022-06-03 17:01
marcanI mean in m1n1#asahi-re · 2022-06-03 17:01
marcandifferent hardware then, you'll have to adapt the driver or write your own#asahi-re · 2022-06-03 17:01
marcanM1 uses dwc3#asahi-re · 2022-06-03 17:01
marcandon't those devices use dwc2?#asahi-re · 2022-06-03 16:59
marcanif you end up having to implement the FTL in software, I wish you luck...#asahi-re · 2022-06-03 16:58
marcanthe NAND controller#asahi-re · 2022-06-03 16:58
marcanno ANS taking care of that I guess?#asahi-re · 2022-06-03 16:58
marcanthat sounds painful...#asahi-re · 2022-06-03 16:57
marcanAdryzzOLEDEdition[m]: they just need their hypervisor to convince iOS to boot, not to actually emulate the hardware as it actually functions#asahi-re · 2022-06-03 16:56
marcanhence why I avoid trying to look at their code; it has caused several people much confusion#asahi-re · 2022-06-03 16:55
marcancorellium do a lot of stuff that isn't relevant to anything or outright wrong#asahi-re · 2022-06-03 16:54
marcanand I initially didn't understand what the second bit set did (then I realized it had to do with enabling system-level stuff for the core, e.g. IRQs don't work without it)#asahi-re · 2022-06-03 16:53
marcanIIRC they had the cluster/core stuff wrong#asahi-re · 2022-06-03 16:53
marcanthey might compute it differently but it has to hit the same register to work#asahi-re · 2022-06-03 16:52
marcanI think it did come from corellium though#asahi-re · 2022-06-03 16:32
marcannergzd723: I forget how we figured out the offset, maybe that came from corellium? Either way nowadays we'd be using the hypervisor to find things like that, but I imagine that won't work on A7/A8 era devices.#asahi-re · 2022-05-19 16:02
marcanhttps://oftc.irclog.whitequark.org/asahi/2021-11-16#30381586 <- this is how you use them from Linux userspace (via the kernel driver)#asahi-re · 2022-05-19 15:58
marcanthe Apple implementation does have a userspace access toggle, but I don't think our kernel driver currently supports it and either way it's completely incompatible with the ARM version, so you need custom userspace too at that point#asahi-re · 2022-04-17 08:08
marcanobviously I can't stop people from going off and doing whatever they want outside the project#asahi-re · 2022-04-17 08:07
marcanthe point is we're trying to keep everything developed as part of the Asahi Linux project legally in the clear#asahi-re · 2022-04-17 08:07
marcanI mean you can share it too, that's your own choice; that policy is about the project as a whole#asahi-re · 2022-04-17 08:05
marcanthen we're good :)#asahi-re · 2022-04-17 08:05
marcansame thing; keep in mind that if you're reverse engineering the kext then you can't contribute to the Linux driver a priori#asahi-re · 2022-04-17 08:03
marcanI meant the facetimehd driver#asahi-re · 2022-04-17 07:02
marcancan someone tell them to stick around next time they show up? Or otherwise check logs.#asahi-re · 2022-04-17 07:02
marcan20:41:29 < marcan> this falls under the section "Referencing other open source code" and in particular the aution about Apple-specific stuff#asahi-re · 2022-04-17 07:02
marcan20:41:06 < marcan> millenialhacker: reminder to read https://asahilinux.org/copyright/#asahi-re · 2022-04-17 07:02
marcanI never got an ack on#asahi-re · 2022-04-17 07:02
marcanoh, they're gone...#asahi-re · 2022-04-17 07:01
marcanyou don't care about how the camera works, you only care about what interface ISP exposes for it over the mailbox#asahi-re · 2022-04-17 07:01
marcanmillenialhacker: everything should be controlled via ISP over the RTKit interface. RTKit fundamentally uses 64-bit messages and shared memory for everything.#asahi-re · 2022-04-17 07:00
marcanabsolutely no reason to believe the SEP has anything to do with the camera on these platforms until proven otherwise#asahi-re · 2022-04-17 07:00
marcanthere is no control line from the SEP on M1 Pro/Max and a good chance the one GPIO from the SEP on M1 doesn't do anything#asahi-re · 2022-04-17 07:00
marcanrobinp: that "security switch" is just the thing that turns off the mic/camera when the lid is closed#asahi-re · 2022-04-16 11:42
marcanwe don't know how that driver was developed, so unless we can verify that, we have to treat it similarly as we would Apple code