#
✕ nick:modwizcode
#asahi-dev · 2021-01-31 04:56
modwizcodeI still haven't watched the chicken bits stream, from the registers I've had to poke, what's the L2 stuff pertain to?
#asahi-dev · 2021-01-31 04:54
modwizcodeEverytime you write to more HID registers I get to stub them out in my qemu branch :p
#asahi-dev · 2021-01-31 04:54
modwizcodeAnyway how'd you figure this one out?]
#asahi-dev · 2021-01-31 04:54
modwizcodeI'm guess it's a good gate to wait to enable anything that might be buggy until there's an OS capable of dealing with it
#asahi-dev · 2021-01-31 04:53
modwizcodeYeah but I mean like the whole caching system is disabled with MMU off
#asahi-dev · 2021-01-31 04:53
modwizcodeAlso I wonder if some of that speculation is disabled unless the MMU is on, like caching.
#asahi-dev · 2021-01-31 04:52
modwizcodeI feel like that's because most other CPUs aren't quite as speculative and the intel and amd ones have a microcode to make it behave more probably
#asahi-dev · 2021-01-30 17:30
modwizcodeGood news is I should be able to get a simple FB up and running today. Then maybe add the AIC
#asahi-dev · 2021-01-30 17:30
modwizcodeUnforutnately there's no easy clean way to get qemu to define a function as the "final" reset handler, so stuff like initializing the HCR_EL2 reg before jumping in and pointing it to the boot args is only certain if you have the cpu itself execute it as a sort of bootloader blob. You can hook reset but if you use any built-in loader code it will hook reset and break :/
#asahi-dev · 2021-01-30 17:25
modwizcodeAnd i cleaned up the code handling boot args and the stub device tree a bit
#asahi-dev · 2021-01-30 17:25
modwizcodeIncidientally I have improved the qemu code a bit. I now load things at 0x10 instead of 0x0 to not confuse code.
#asahi-dev · 2021-01-30 17:24
modwizcodeShould go fix it right now that I have hot air to use.
#asahi-dev · 2021-01-30 17:24
modwizcodeI cannot remember the exact size (I went to go look but it's unmarked) but it was small enough that I accidentally knocked other things off the board and almost broke it completely
#asahi-dev · 2021-01-30 17:22
modwizcodeI should probably get something like that.
#asahi-dev · 2021-01-30 17:21
modwizcodeThe other day I bought like 10-20 smd tiny little ones to fix a eval board. I have no use for more than 1
#asahi-dev · 2021-01-30 17:20
modwizcodemarcan: Then I'd have to buy something to store them in?
#asahi-dev · 2021-01-30 17:03
modwizcodeshould be
#asahi-dev · 2021-01-29 23:55
modwizcodekettenis: What about SCTLR_EL1?
#asahi-dev · 2021-01-29 23:53
modwizcodeYeah I was like why aren't interrupts firing (in qemu I have the uart IRQ connected directly to the core's IRQ line lol)
#asahi-dev · 2021-01-29 23:44
modwizcodethanks
#asahi-dev · 2021-01-29 23:33
modwizcodemarcan (or anyone else): Could you get me the contents of hcr_el2 upon entry into mini from iBoot? I'm curious about what it configures since it affects how el2 takes interrupts.
#asahi-dev · 2021-01-29 18:22
modwizcodenight
#asahi-dev · 2021-01-29 18:22
modwizcodeheh smp funnnn
#asahi-dev · 2021-01-29 18:17
modwizcodeI should try throwing linux at this thing. I've gotta make another patch to point the framebuffer into some unused ram though.
#asahi-dev · 2021-01-29 18:17
modwizcodenice
#asahi-dev · 2021-01-29 18:16
modwizcodeoh you added the libfdt stuff too
#asahi-dev · 2021-01-29 18:15
modwizcodety :)
#asahi-dev · 2021-01-29 18:14
modwizcodeI just made a tiny patch that adds a call to adt_check_header in adt_first_property_offset and that fixed it (but that's obviously not a good patch)
#asahi-dev · 2021-01-29 18:12
modwizcodeWell it's sort of used in the higher level code I think
#asahi-dev · 2021-01-29 18:12
modwizcodeThere's all the definitions for the checking code, but they're not used at all :p
#asahi-dev · 2021-01-29 18:12
modwizcodeYeah :)
#asahi-dev · 2021-01-29 18:11
modwizcodeso it will just keep looking forever
#asahi-dev · 2021-01-29 18:10
modwizcodebut nothing ever checks the condition to report 0
#asahi-dev · 2021-01-29 18:10
modwizcodeIt just keeps going unless the property offset isn't 0.
#asahi-dev · 2021-01-29 18:10
modwizcodeSo line 81 in adt.c
#asahi-dev · 2021-01-29 18:10
modwizcodeHang on
#asahi-dev · 2021-01-29 18:10
modwizcodeYou should never have a 0 size tree except for the nonsense I'm doing
#asahi-dev · 2021-01-29 18:09
modwizcodeI don't think you have to pass size
#asahi-dev · 2021-01-29 18:09
modwizcodeThat's fine but it has functions to check validity of device tree nodes, it should use them
#asahi-dev · 2021-01-29 18:08
modwizcodeBut even if you give it one (a stub) it ignores if there's no properties/children and doesn't give up on looking for nodes
#asahi-dev · 2021-01-29 18:08
modwizcodeYeah
#asahi-dev · 2021-01-29 18:08
modwizcodeu32 devtree_size;
#asahi-dev · 2021-01-29 18:08
modwizcodeSo if you say null devtree with 0 size. It'll still try to find the device-type and target
#asahi-dev · 2021-01-29 18:07
modwizcodeIt ignores the size arg too
#asahi-dev · 2021-01-29 18:07
modwizcodeThat's why m1n1 needs to check the args
#asahi-dev · 2021-01-29 18:07
modwizcodeYeah
#asahi-dev · 2021-01-29 18:07
modwizcodethat's the other thing
#asahi-dev · 2021-01-29 18:07
modwizcodeSo yeah
#asahi-dev · 2021-01-29 18:07
modwizcode nope literally 0
#asahi-dev · 2021-01-29 18:07
modwizcodeBecause I could and I don't have real hardware docs in front of me at the moment to make saner choices

plus ›