#
✕ nick:eiln
#asahi-dev · 2023-09-10 16:05
eilnand I verified t6000 dsid before
#asahi-dev · 2023-09-10 16:03
eilnI'm guessing the bootargs are at 0xe12f80
#asahi-dev · 2023-09-10 16:02
eilncan you trace (open photobooth app) and hexdump heap_top - heap_top + 0x4000?
#asahi-dev · 2023-09-10 16:01
eilnsorry I'm reinstalling macos rn I can't test the tracer
#asahi-dev · 2023-09-10 15:58
eilnit's definitely the bootargs, sigh
#asahi-dev · 2023-09-10 15:51
eilnor the cmd_iova offset is wrong
#asahi-dev · 2023-09-10 15:49
eilnhttps://pastebin.com/jmif91Ah
#asahi-dev · 2023-09-10 15:48
eilnit might be the wall of spmi registers
#asahi-dev · 2023-09-10 15:47
eilnwe didn't send those yet. we're stuck on cmd_print_enable
#asahi-dev · 2023-09-10 15:43
eilnhttps://gist.github.com/ah-/34d440894e30980f4b110d0e758b9e0f
#asahi-dev · 2023-09-10 15:42
eilnt6000? it's not mine
#asahi-dev · 2023-09-10 15:41
eilnt6000
#asahi-dev · 2023-09-10 15:41
eilnISPASC: ASSERT: ./h10isp/common/misc/H13/CSystemConfiguratorConnectivityH13Jade.cpp, 472: 0
#asahi-dev · 2023-09-10 15:41
eilnISPASC: ConnectivityTableCreate: optical card rev 0x0
#asahi-dev · 2023-09-10 15:41
eilnmarcan: ah- timed out here (enable APPLE_ISP_DEBUG to print fw logs)
#asahi-dev · 2023-09-10 15:18
eilnand it boots all the way?
#asahi-dev · 2023-09-10 15:18
eilnit's fine, I figured out a hack during the ANE days
#asahi-dev · 2023-09-10 15:18
eilnthe "tunables" masks unused regs tho, 0x64/0x68/0x6c. interesting
#asahi-dev · 2023-09-10 15:13
eilnwhy would they do this? isn't this easier?
#asahi-dev · 2023-09-10 15:12
eilndart-tunables-instance-* is in the dt, and is required for the hack
#asahi-dev · 2023-09-10 14:53
eilnmy understanding is that they are not real darts. and initing them as separate iommus might erase the special tunables
#asahi-dev · 2023-09-10 14:51
eilnwhere is it not mirrored?
#asahi-dev · 2023-09-10 14:33
eilnISP/ANE/AVE just needs TTBR and TLB invalidation mirrored
#asahi-dev · 2023-09-10 14:19
eilnnothing hardware yet, opcode 0x004 is PRINT_ENABLE. only 0xe00000 vs t8103 0x1800000 is suspicious though.
#asahi-dev · 2023-09-10 14:15
eilnare we sure of the heap_top size? you can run 'log show' in macos to get that string
#asahi-dev · 2023-09-10 14:13
eilnit didnt for me. I just rebased on asahi-wip. can you print apple_dart_setup_resv_locked to see if it's getting mapped?
#asahi-dev · 2023-09-10 14:09
eilnwithout it dart_resv can't find it. I had this issue earlier
#asahi-dev · 2023-09-10 14:08
eilnI pushed to isp-dapf with yours cherry-picked
#asahi-dev · 2023-09-10 14:05
eilnthat means the asc hasn't even booted, likely an err in ctrr setup. I patched src/isp.c to add the heap to the phandle so of_iommu can find it. is that there? ah-'s t6000 booted before
#asahi-dev · 2023-09-10 11:26
eilnsounds good, because I messed up my signature lol
#asahi-dev · 2023-09-10 11:22
eilnwait my key expired
#asahi-dev · 2023-09-10 11:11
eilnhow far do your other machines go before halting? :P
#asahi-dev · 2023-09-10 11:11
eilnsorry I wasn't clear. it doesn't have to alloc bottom-up anymore. but it'd be helpful (for me) to leave as is until we confirm the firmware boot mess across all variants. I also haven't checked vm_size for all dts. (rightfully) moving to iova.c is trivial, and I'll handle it right after
#asahi-dev · 2023-09-10 11:10
eilnmarcan: I thought unk_size was firmware size too, but 0x180000 worked on 13.5 (0x100000). turns out unk_size only has to be greater than the current size, which the 0x180000 unknowingly covered.
#asahi-dev · 2023-09-10 11:10
eilnrebased and pushed the just-works heap mapping
#asahi-dev · 2023-09-07 18:04
eilnthe power domains, the one required read the revision version register, not being in the dt, is such a nice touch
#asahi-dev · 2023-09-07 17:34
eilnquickly inverting the iovas in m1n1 does not break it. frankly, it was pasted from the ane code, and also the familiar iovas make debugging a lot easier. I will check on the bottom-up allocator though
#asahi-dev · 2023-09-07 17:34
eilnwith the heap gone (thank you), the only iova arithmetic is in firmware boot stage 2, but that's all contained within the ipc surface so it shouldn't matter.
#asahi-dev · 2023-09-07 17:34
eilnas for the ANE, DMA to/from the upper iovas (I think it was piodma_size <= iova <= vm_size) is noticeably slower, at least 10x, and their kext actually does tricks to avoid using this range. it seems loading/unloading an entire transformers model is cheaper
#asahi-dev · 2023-09-07 17:34
eilnthe drm mm allocator is because iova.c can't allocate bottom up. there was a patch some time ago but it wasn't reviewed. some stuff about zero iovas
#asahi-dev · 2023-09-07 17:33
eilnANE/AVE also use greek mythology code names, particularly the Hyperion/Crius family tree
#asahi-dev · 2023-09-07 17:33
eilnalso I can finally boot the ANE firmware with my cursed ISP knowledge. not sure what to do with this
#asahi-dev · 2023-09-07 17:33
eilnISP is not the only one with a static non-iboot carveout. see ANE and AVE ;)
#asahi-dev · 2023-09-07 17:33
eilnshower thought. piodma could mean peripheral IO DMA device
#asahi-dev · 2023-09-07 15:11
eilnyeah, that 13.5 issue I had some time ago was this
#asahi-dev · 2023-09-07 15:08
eilnbut the sizes are different per OS version too. that's not in the kext
#asahi-dev · 2023-09-07 15:03
eilnoh that. sigh, they have a firmware for each fucking device I dont know why they couldn't have took care of that shit in there. t8103 is probably all the same, but it might not be
#asahi-dev · 2023-09-07 15:00
eilnthe hard-coded values are per-sensor, which are unrelated to the board type (e.g. m1 pro & m2 air)
#asahi-dev · 2023-09-07 14:59
eilnmarcan: following ANE I think "revision" means board type. it makes sense at that point in probe too. it's not in adt but there might be a register.
#asahi-dev · 2023-09-07 13:37
eilnleft another comment which might be useful

plus ›