#asahi-dev · 2025-03-01 19:02
mischa85jannau: thanks. i got 18 ranges defined in the adt. is there any way to know which registers belong to ATC0_USB_AON, ATC0_COMMON, AFI?#asahi-dev · 2025-03-01 18:43
chaos_princessfair :P#asahi-dev · 2025-03-01 18:43
jannauyes, I wasn't motivated#asahi-dev · 2025-03-01 18:41
chaos_princesshttps://github.com/AsahiLinux/linux/commit/90a611c90ed0a60ac6fef340a05f55ff45ae1d23 - this one can just become copy_from_slice, i will fix it later if nobody else gets to it first#asahi-dev · 2025-03-01 18:40
jannaumischa85: see "/arm-io/ranges" in the adt. if a reg value falls into (bus_addr, size) the effective address is parent_addr + reg value #asahi-dev · 2025-03-01 18:32
mischa85i can succesfully set a register at 0x5008300000, which is what adt_get_reg outputs with an idx of 1. how can i find out the raw registers in macos so i can hardcode them in m1n1?#asahi-dev · 2025-03-01 15:25
chaos_princessthanks, will take a look#asahi-dev · 2025-03-01 15:06
jannauchaos_princess: I left the sep/aop fixes as separate fixup commits if you want to look at them. the newer dma allocator abstraction required larger changes. I'll squash them in a few days#asahi-dev · 2025-03-01 10:55
mischa85parse/interpret#asahi-dev · 2025-03-01 10:55
mischa85while playing with the adt parser, there are a couple of bits that *are* in the binary devicetree ripped by m1n1, and not in the ipsw devicetree supplied with apple firmware. now how to parse them...#asahi-dev · 2025-03-01 09:36
mischa85489 devices, 61 regs#asahi-dev · 2025-03-01 09:30
mischa85i'm desperately looking for some sort of value = n -> reg[n]#asahi-dev · 2025-03-01 01:18
chadmedcurious that the compatibles for the cores are still sawtooth/everest#asahi-dev · 2025-03-01 00:13
mischa85but as i was fearing, ps-regs is indeed, missing.#asahi-dev · 2025-03-01 00:12
mischa85but as i was fearing, ps-regs is indeed, missing.#asahi-dev · 2025-02-28 19:32
opticronoof, sorry#asahi-dev · 2025-02-28 19:32
opticron[/quote]#asahi-dev · 2025-02-28 19:32
opticronSTM32F46RE#asahi-dev · 2025-02-28 19:32
opticron[quote="vidhu, post:1, topic:82"]#asahi-dev · 2025-02-28 19:06
mischa85what i should have been doing, though, is calling it unk2_1 as unk2_0 was already being used. derp.#asahi-dev · 2025-02-28 18:22
mischa85i called the new value unk2_0 and it corresponds to what is found as unk2 in the ipsw ripped adt#asahi-dev · 2025-02-28 18:17
mischa85J616cAP ADT: https://gist.github.com/mischa85/d5e84fcb173eada3902620d28f93df6f#asahi-dev · 2025-02-28 17:44
mischa85i needed to modify m1n1.adt, one of the blank spaces in devices now contains an int32ul of unk2#asahi-dev · 2025-02-28 17:42
mischa85now i need to censor this a bit i guess#asahi-dev · 2025-02-28 17:42
mischa85alright i have succesfully ripped a m1n1 devtree of J616cAP, Mac16,5, AppleARM#asahi-dev · 2025-02-28 08:36
sventhey also changed some between m1 and M1 Pro iirc even though the MMIO interface wasn’t changed as far as we could tell #asahi-dev · 2025-02-28 08:35
nickchanand by mostly compatible I mean I can read the rtc counter just fine using m1n1's spmi.py#asahi-dev · 2025-02-28 08:35
svenyeah, you can’t trust the adt compatibles #asahi-dev · 2025-02-28 08:35
nickchanmaybe I should take these compatibles less seriously#asahi-dev · 2025-02-28 08:34
jannaunickchan: probably a different clang-format version#asahi-dev · 2025-02-28 08:33
nickchaneven though it has another adt compatible#asahi-dev · 2025-02-28 08:33
nickchanhuh so it turns out that the SPMI in A11 *is* mostly compatible with the one in M1#asahi-dev · 2025-02-28 08:25
nickchanif i run make format on the m1n1 src on my machines then a difference near the el3 call handling in src/exception.c will always be generated that the github actions don't generate#asahi-dev · 2025-02-27 19:19
mischa85j'ey: no linux boot on t6041 yet. i do have a usb sound driver on my github which i didn't take upstream yet#asahi-dev · 2025-02-27 19:17
mischa85alyssa: stories about burnouts really hit close to home. i've been there. i deeply respect the people who did their magic on all this, and seeing people going over their limits really saddens me.#asahi-dev · 2025-02-27 19:17
sven\o/#asahi-dev · 2025-02-27 19:08
alyssamaz: thanks :)#asahi-dev · 2025-02-27 19:06
mazguess I'll rework the series once I'm on "holiday" in two weeks.#asahi-dev · 2025-02-27 19:06
mazso I happen to have this box keeping my feet warm.#asahi-dev · 2025-02-27 19:05
jannauyes, m2 pro and max (and double max)#asahi-dev · 2025-02-27 19:05
mazbingo.#asahi-dev · 2025-02-27 19:05
maz"compatible = "apple,t6020-pcie";"#asahi-dev · 2025-02-27 18:56
svencompatible is t6020 so that would be M2 Pro#asahi-dev · 2025-02-27 18:55
alyssai *think* base m2 is the m1-compatible pcie but m2 pro+ is the new one? i think?#asahi-dev · 2025-02-27 18:55
alyssawhy do apple socs all need confusing names#asahi-dev · 2025-02-27 18:55
alyssai think#asahi-dev · 2025-02-27 18:55
alyssaoh wait m2 pro indeed#asahi-dev · 2025-02-27 18:53
mazhmm, only M2 pro here.#asahi-dev · 2025-02-27 18:52
sveni first want to get all the M1 stuff upstream. if no one has taken care of pcie by then I guess I'll get a m2 max from that opencollective money and take care of it#asahi-dev · 2025-02-27 18:50
alyssai have the hw but it's my kernel build machine so makes for an awkward DUT