#
✕ nick:marcan
#asahi-dev · 2021-02-09 17:42
marcan300 baud modems, and a lot of "make it twice as fast"
#asahi-dev · 2021-02-09 12:58
marcanthough I haven't tried it
#asahi-dev · 2021-02-09 12:58
marcanyes
#asahi-dev · 2021-02-09 11:22
marcanalso beyond that signal integrity starts to matter a bit more :)
#asahi-dev · 2021-02-09 11:22
marcanit's the fastest it goes on the default clock and sampling settings
#asahi-dev · 2021-02-09 11:14
marcanbgb: if you're using linux.py, your baud rate is 1500000
#asahi-dev · 2021-02-09 10:43
marcanI should add input to ttymode
#asahi-dev · 2021-02-09 10:43
marcanbgb: yeah, just cancel that and use picocom or whatever
#asahi-dev · 2021-02-09 07:29
marcanI don't see why it wouldn't work
#asahi-dev · 2021-02-09 07:28
marcantarzeau: oh, it would have to be ported to the Apple Silicon version of Hypervisor.framework I guess
#asahi-dev · 2021-02-09 03:23
marcanthat accesses the true TTBRx_EL1 register
#asahi-dev · 2021-02-09 03:23
marcaninstead, you need to use TTBRx_EL12
#asahi-dev · 2021-02-09 03:23
marcanso you cannot access TTBRx_EL1 as that name any more
#asahi-dev · 2021-02-09 03:22
marcanwhen E2H==1, TTBRx_EL1 is *redirected* to TTBRx_EL2
#asahi-dev · 2021-02-09 03:22
marcanwhen E2H==0, they are independent. TTBR0_EL2 is used in EL2 and TTBRx_EL1 is used in EL1.
#asahi-dev · 2021-02-09 03:22
marcanthere are two register sets: TTBRx_EL2 and TTBRx_EL1
#asahi-dev · 2021-02-09 03:22
marcanbgb: the ARM ARM is confusing
#asahi-dev · 2021-02-08 16:19
marcannever_released: sounds unlikely though, why hardcode this on if not to simplify the implementation?
#asahi-dev · 2021-02-08 14:16
marcanbgb: just for the record, if you *could* clear E2H you'd have to change the mmu code, as things there work differently in those two modes
#asahi-dev · 2021-02-08 14:16
marcanjix: yeah but you or someone else might still wonder :)
#asahi-dev · 2021-02-08 14:15
marcanbgb: and yes, this is a spec violation
#asahi-dev · 2021-02-08 14:15
marcan(you could do type 2 with E2H==0 but it would suck)
#asahi-dev · 2021-02-08 14:14
marcanbasically E2H==0 means type 1 hypervisor, E2H==1 means type 2 hypervisor
#asahi-dev · 2021-02-08 14:14
marcanit means you can run a "normal" OS as EL2 because it aliases all the EL1 registers to EL2 registers
#asahi-dev · 2021-02-08 14:14
marcanjix: that bit erases EL1, basically
#asahi-dev · 2021-02-08 14:13
marcanbgb: we cannot
#asahi-dev · 2021-02-08 09:55
marcanyes
#asahi-dev · 2021-02-08 07:49
marcanyes, so does the m1n1 base address and certain bits of virt_base
#asahi-dev · 2021-02-08 06:49
marcanI'm actually not even sure if virt_base is based on that address...
#asahi-dev · 2021-02-08 06:46
marcanthe only thing those vmaddrs are used for is internal to the mach-o format, and as a quirk, also as the base for some pointers passed from iBoot, but those are kASLR slid anyway so that base vmaddr isn't correct either, iBoot tells you the one it *actually* used
#asahi-dev · 2021-02-08 06:45
marcaniBoot loads m1n1 to wherever it wants, in fact it's randomized in a 32MB window every boot due to kASLR
#asahi-dev · 2021-02-08 06:45
marcanbgb: vmaddr is not used for anything other than computing addresses
#asahi-dev · 2021-02-08 05:20
marcanand iBoot translates it to wherever the hell it wants in physical address space
#asahi-dev · 2021-02-08 05:19
marcanbasically it just means all addresses in the mach-o format are relative to that number
#asahi-dev · 2021-02-08 05:18
marcanso it's really a completely arbitrary number at this point, with no significance to anything whatsoever at runtime; I just don't know if iBoot has any restrictions on it
#asahi-dev · 2021-02-08 05:17
marcanit's not even the real virtual address of xnu any more, thanks to kASLR
#asahi-dev · 2021-02-08 05:17
marcanit's just a hack because mach-o is a mildly brain-dead format that does not support the concept of physical addresses
#asahi-dev · 2021-02-08 05:17
marcanfor us, anyway
#asahi-dev · 2021-02-08 05:16
marcanthat address means nothing, for what it's worth
#asahi-dev · 2021-02-08 05:16
marcan(back when I was writing that mach-o header thing I started out making it similar to xnu to eliminate errors, since I had to get that much to work blind)
#asahi-dev · 2021-02-08 05:16
marcanbgb: it's the same as xnu, though I suspect "0" would work but I never tried it
#asahi-dev · 2021-02-07 18:20
marcanbastilian: if it says USB-IF *and* SuperSpeed or USB3.0 or 5Gbps then it will work
#asahi-dev · 2021-02-07 17:46
marcanbastilian: the USB-IF certified one will *definitely* be OK as long as they aren't lying; the other one is probably OK unless it's a really horrible one.
#asahi-dev · 2021-02-07 17:45
marcanAIUI USB4 is mostly a superset of TBT3, it's just that the doc I read was a USB doc and thus won't mention TBT3
#asahi-dev · 2021-02-07 15:16
marcanand other things
#asahi-dev · 2021-02-07 15:16
marcanI mean it would break DP otherwise :p
#asahi-dev · 2021-02-07 15:12
marcanonly in USB4 mode does it turn into a digital thing
#asahi-dev · 2021-02-07 15:11
marcangruetzkopf: the spec seems to say SBU should be passive by default
#asahi-dev · 2021-02-07 15:08
marcangruetzkopf: I think the basic (short) apple tb cable is passive
#asahi-dev · 2021-02-07 13:29
marcanI can't think of why a fancy cable would mess with SBU1/2

plus ›