#
✕ nick:phire
#asahi-gpu · 2022-02-10 14:19
phireat the structure pointed to by Submit3D
#asahi-gpu · 2022-02-10 14:17
phireThis might be another set of ringbuffers
#asahi-gpu · 2022-02-09 15:11
phireAlright, I've found a set of circular buffers
#asahi-gpu · 2022-02-09 13:22
phires/lind/lid/
#asahi-gpu · 2022-02-09 13:22
phirethis apparently works: ssh in, then close the lind
#asahi-gpu · 2022-02-09 13:19
phireand setting zero brightness doesn't prevent rendering
#asahi-gpu · 2022-02-09 13:18
phiresetting the "prevent sleep when display off" option didn't stop it from sleeping
#asahi-gpu · 2022-02-09 13:08
phirein macos settings?
#asahi-gpu · 2022-02-09 13:03
phiredid stop most of the kicks. I think it still drew the white apple logo
#asahi-gpu · 2022-02-09 13:01
phirewell... m1n1 crashes if I boot when the lid closed
#asahi-gpu · 2022-02-09 12:55
phirehmm, lid closed might work
#asahi-gpu · 2022-02-09 12:53
phireyes... I agree it is odd that only that example would cause them
#asahi-gpu · 2022-02-09 12:51
phirewhat I mean is that I've only seen in while running the metal spinning cube example, not during login or on the desktop
#asahi-gpu · 2022-02-09 11:46
phiremonitoring doesn't show any updates to this structure
#asahi-gpu · 2022-02-09 11:45
phirethat's an interesting panic
#asahi-gpu · 2022-02-09 11:44
phireGFX PANIC - Host-mapped FW allocations are disabled, but FW only supports enabled (init_data_addr=0xffffffa0000b0000)! - agx_background(2)
#asahi-gpu · 2022-02-09 11:44
phireyeah, I haven't even tried to look for the per-context stuff, because I assume there needs to be a global queue to set the context
#asahi-gpu · 2022-02-09 11:40
phireif there is an address to a shared memory queue, it should be there
#asahi-gpu · 2022-02-09 11:39
phireI'm current mapping out this "init data" struct
#asahi-gpu · 2022-02-09 11:32
phireyeah, the general idea
#asahi-gpu · 2022-02-09 11:30
phirethough the DCP ASC looks nothing kike the GFX ASC. I doubt they are related
#asahi-gpu · 2022-02-09 11:30
phireand way more sane than my other theories
#asahi-gpu · 2022-02-09 11:30
phireyeah, sounds right
#asahi-gpu · 2022-02-09 10:16
phire GFX PANIC - Unable to grab init data from host - agx_background(2)
#asahi-gpu · 2022-02-09 10:15
phirewell, unless you do something stupid like senting a null pointer on it's init data
#asahi-gpu · 2022-02-09 10:13
phireEP 0x20 doesn't seem to crash no matter what you send it
#asahi-gpu · 2022-02-09 10:13
phireep 0x21 crashes when you send it message types it doesn't exist, but doesn't seem to do any checking on the args to the 0x83 message
#asahi-gpu · 2022-02-09 10:11
phireforcing crashes hasn't told me that much
#asahi-gpu · 2022-02-09 08:58
phireRTKit Task List: rtk_ep_work, power, agx_background, agx_recovery, agx_interrupt, agx_power, agx_sampler
#asahi-gpu · 2022-02-09 08:57
phireoh... thank you crashlog and your very creative endpoint naming scheme of "user01" and "user02"
#asahi-gpu · 2022-02-09 08:48
phireyes, it's very privliaged
#asahi-gpu · 2022-02-09 08:36
phirethanks. Prehaps I should be trying harder to crash it
#asahi-gpu · 2022-02-09 04:59
phirehttps://github.com/phire/m1n1/blob/gpu-RE/phire-gpu-infodump.md
#asahi-gpu · 2022-02-09 02:31
phireyou still interested at looking into gpu stuff today?
#asahi-gpu · 2022-02-09 02:30
phireright, I'm writing up an infodump for you marcan
#asahi-gpu · 2022-02-08 07:24
phireI guess it makes total sense for gfx-asc to have a direct communication channel for those
#asahi-gpu · 2022-02-08 06:46
phireMMC is memory controller?
#asahi-gpu · 2022-02-08 06:46
phireit has the PMGR and MCC devices mapped in
#asahi-gpu · 2022-02-08 06:45
phireah, there is an offset
#asahi-gpu · 2022-02-08 06:39
phireand the ADT doesnt' seem to know about any of these ranges
#asahi-gpu · 2022-02-08 06:36
phireand a few extra
#asahi-gpu · 2022-02-08 06:36
phirethen kernel adds a bunch more IO ranges, including the sgx ranges
#asahi-gpu · 2022-02-08 06:35
phireok, at boot, gfx-asc only has it's own IO ranges mapped into it's pagetable
#asahi-gpu · 2022-02-08 02:40
phireI don't have enought time to infodump right now
#asahi-gpu · 2022-02-08 02:39
phirehow about tomorrow if I haven't made any progress?
#asahi-gpu · 2022-02-08 02:38
phireI'm not 100% stuck yet, but maybe
#asahi-gpu · 2022-02-08 02:27
phire(that's probally not a memory region)
#asahi-gpu · 2022-02-08 02:25
phire(except for ttbat-phys-addr-base, which I don't understand)
#asahi-gpu · 2022-02-08 02:25
phirebut I have traced the entire IO range around the sgx registers, and every single memory region mentioned in the sgx node
#asahi-gpu · 2022-02-08 02:24
phireI haven't tired tracing *everything* yet. I suspect would need a good handle on when a work submission might be happen before that would be viable

plus ›