#asahi · 2021-05-27 20:09
svenlol#asahi · 2021-05-27 14:23
svenor maybe it's not MMIO but some shared memory region for one of the subprocessors?#asahi · 2021-05-27 14:23
svenfor MMIO reads/writes?#asahi · 2021-05-27 14:18
svensimd in pmgr :D#asahi · 2021-05-27 12:51
svenmarcan: uh, does it always run on the same core?#asahi · 2021-05-25 17:33
svenwhich didn't exist on any x86 macs :)#asahi · 2021-05-25 17:33
sveni thought their last update still only mentions APRR#asahi · 2021-05-25 17:33
svendoes apple even mention GXF anywhere?#asahi · 2021-05-25 13:52
svenyeah, that sounds nasty#asahi · 2021-05-25 13:49
svennice#asahi · 2021-05-25 13:49
sven\o/#asahi · 2021-05-25 13:40
svenI'm still surprised the phy bringup is also enough for host mode#asahi · 2021-05-25 13:39
svenfair enough :D#asahi · 2021-05-25 13:38
svenyou probably just need usb_phy_bringup or however it's called if you want to use dwc3 in uboot/openbsd#asahi · 2021-05-25 13:36
svenglad you figured it out ;)#asahi · 2021-05-25 13:35
svenlol#asahi · 2021-05-25 13:00
svenyes, debug_putc in GL2 isn't working apparently#asahi · 2021-05-25 12:55
svenGL1 is xnu#asahi · 2021-05-25 12:55
svenGL2 *is* the hypervisor#asahi · 2021-05-25 12:54
sventhere's no need to inject anything into xnu's pagetables#asahi · 2021-05-25 12:54
svenuh, it's a hv. he just uses two stage translations#asahi · 2021-05-25 12:50
svenwith some mystery bits I don't know about#asahi · 2021-05-25 12:50
svenhm, yes#asahi · 2021-05-25 12:49
svenso it faults from GL1 to GL2 and then just hangs?#asahi · 2021-05-25 12:48
svenweird#asahi · 2021-05-25 12:47
svenit shouldn't make a difference but...#asahi · 2021-05-25 12:47
svenmarcan: did you try in the middle of the hv after xnu already enabled gxf in el1?#asahi · 2021-05-25 12:43
sventhe SP was strange. iirc macOS always sets it up from TPIDR_GL1/2#asahi · 2021-05-25 12:42
sven:D :D#asahi · 2021-05-25 12:41
svencan you read the SP with that coresigh stuff?#asahi · 2021-05-25 12:40
svenoh wait.. the SP in GXF is weird fwiw#asahi · 2021-05-25 12:40
sveni.e. the exception vector is not marked as executable in the SPRR register or the stack isn't writable?#asahi · 2021-05-25 12:39
svenhuh.. maybe some weird permission remap thing?#asahi · 2021-05-25 12:39
svenpipcet[m]: what makes you think that?#asahi · 2021-05-25 12:38
svenfrom what I can tell it's always using the same pagetables#asahi · 2021-05-25 12:38
svenor, well, I haven't found anything like that#asahi · 2021-05-25 12:38
svenI don't think so#asahi · 2021-05-25 12:38
svenTTBAR?#asahi · 2021-05-25 12:33
svenlike, it just stop working?#asahi · 2021-05-25 12:33
svenwow :D#asahi · 2021-05-25 11:08
svenbut who knows :-)#asahi · 2021-05-25 11:08
svenI hope that doesn't matter though since it shouldn't try to speak to the m1n1 port once it has been removed from the ADT#asahi · 2021-05-25 11:07
sventrue#asahi · 2021-05-25 10:48
sventhough at least in theory they both also have a separate phy mmio space. but who knows what kinda of fun surprises are lurking in there#asahi · 2021-05-25 10:48
svenyeah, it also wouldn't surprise me if some phy stuff is actually shared between the two ports and breaks#asahi · 2021-05-25 09:19
svenis there any other reason you have in mind for sharing that USB port?#asahi · 2021-05-25 09:18
svenit's also not really required. it's possible to mux the Mac OS serial output over m1n1's usb driver together with the MMIO trace#asahi · 2021-05-25 09:18
svenyou'd have two drivers competing over a single HW block#asahi · 2021-05-25 09:17
svenhow would that work?#asahi · 2021-05-25 09:13
svenwhy do you want to know that?