#asahi · 2022-04-07 08:02
Jamie[m]1didn't resume successfully (can't remember if kernel panic or serial disconnected), but went away after reboot #asahi · 2022-04-07 07:17
Jamie[m]130 minutes?#asahi · 2022-04-07 06:27
Jamie[m]1actually there's a weird zigzaggyness to it, so not random at all#asahi · 2022-04-07 06:25
Jamie[m]1as if the display controller forgets what brightness each local dimming zone is supposed to be at, and the brightnesses randomly drift#asahi · 2022-04-07 06:25
Jamie[m]1^ I left my machine paused in hv for a long time and it developed this weird starburst pattern#asahi · 2022-04-07 06:24
Jamie[m]1uploaded an image: (1962KiB) < https://matrix.org/_matrix/media/r0/download/memes.nz/eNOefdjBhusItOlweJcXLlZM/20220407_182415.jpg >#asahi · 2022-04-07 05:18
Jamie[m]1yep#asahi · 2022-04-07 05:17
Jamie[m]1and occasionally linux failing to enumerate the ACMs, or taking several seconds to do so#asahi · 2022-04-07 05:15
Jamie[m]1cursed host machine :(#asahi · 2022-04-07 05:15
Jamie[m]1across 3 cables, both A-C and C-C#asahi · 2022-04-07 05:14
Jamie[m]1as well as lots of "expected 0 bytes got 1"#asahi · 2022-04-07 05:14
Jamie[m]1hmm yeah getting opcode mismatches and reply command mismatches with random bits flipped#asahi · 2022-04-07 05:07
Jamie[m]1about 3/15 attempts at booting now have gone through without any issues#asahi · 2022-04-07 05:07
Jamie[m]1ugh nope that wasn't the whole problem, this machine just has Cursed Usb#asahi · 2022-04-07 04:38
Jamie[m]1lol i've also been trying to debug an issue with hypervisor booting: turns out i had the laptop sitting on this broken device, shorting out the usb pins and resetting the host usb controller -_-#asahi · 2022-04-07 04:37
Jamie[m]1uploaded an image: (2032KiB) < https://matrix.org/_matrix/media/r0/download/memes.nz/lUiEYSrwIROPVrZXFwsprdVV/ima_afe797f.jpeg >#asahi · 2022-03-20 12:04
Jamie[m]1👍#asahi · 2022-03-20 12:03
Jamie[m]1do we have keyboard backlight support (on pro/max MBP)?#asahi · 2022-03-18 16:36
Jamie[m]1s/without/with/#asahi · 2022-03-18 16:36
Jamie[m]1that's how you can decrypt your phone without a 4-digit pin without someone easily brute-forcing it in 10000 attempts#asahi · 2022-03-18 16:35
Jamie[m]1Sobek: (speaking in general terms) it allows you to use a weaker password or biometric auth, and the SEP ratelimits password/biometric attempts and releases the encryption key when it likes what it sees#asahi · 2022-03-18 16:29
Jamie[m]1but it's purer in the sense that it can be built on any machine#asahi · 2022-03-18 16:28
Jamie[m]1actually i'm not sure if nix's build sandbox allows local filesystem access for fixed-output derivations#asahi · 2022-03-18 16:27
Jamie[m]1yeah, i'm gonna run wifi.py on this directory afterwards#asahi · 2022-03-18 16:27
Jamie[m]1tpw_rules: definitely a better move lol#asahi · 2022-03-18 16:26
Jamie[m]1i'm thinking of a specific nixos usecase#asahi · 2022-03-18 16:26
Jamie[m]1yeah fair#asahi · 2022-03-18 16:25
Jamie[m]1still, 18MB < 1.8GB :P#asahi · 2022-03-18 16:25
Jamie[m]1didn't notice that :P#asahi · 2022-03-18 16:25
Jamie[m]1marcan: ah neat lol#asahi · 2022-03-18 16:25
Jamie[m]1I had a dumb idea and had to implement it: using http range requests and DEFLATE trickery to download Wifi firmware from Apple's CDN with only 18MB of transfer (out of a 13GB ipsw) http://github.com/JJJollyjim/firmware-abomination#asahi · 2022-03-09 02:38
Jamie[m]1not sure in matrix, but you can view logs here https://oftc.irclog.whitequark.org/asahi/2022-03-09#asahi · 2022-03-08 08:47
Jamie[m]1clrwf0x80: prebuilt images haven't been released yet#asahi · 2022-02-11 11:56
Jamie[m]1oh is it an ACPI thing? right :/#asahi · 2022-02-11 11:56
Jamie[m]1i didn't realise that was a more widespread practise#asahi · 2022-02-11 11:55
Jamie[m]1^ lol, valve's embedded controller for the steam deck uses apple-style 4-letter attributes#asahi · 2022-02-11 11:55
Jamie[m]1uploaded an image: (1094KiB) < https://matrix.org/_matrix/media/r0/download/memes.nz/kZGWQWIoctKodvjLunHSzZFa/Screen%20Shot%202022-02-12%20at%2012.55.14%20AM.png >#asahi · 2022-01-24 08:03
Jamie[m]1ah gotcha#asahi · 2022-01-24 07:39
Jamie[m]1is the issue that the conversion isn't being performed, and the data is misinterpreted as yuv420?#asahi · 2022-01-24 07:37
Jamie[m]1or rearranging bits i mean#asahi · 2022-01-24 07:37
Jamie[m]1jannau: is nv12 <-> yuv420 not a lossless conversion, just rearranging bytes?#asahi · 2022-01-23 15:15
Jamie[m]1again in vp9, minor differences are visible#asahi · 2022-01-23 15:15
Jamie[m]1uploaded an image: (414KiB) < https://matrix.org/_matrix/media/r0/download/memes.nz/LzPTyJzJBhoQZngoJASDRzXT/Screen%20Shot%202022-01-24%20at%204.15.41%20AM.png >#asahi · 2022-01-23 15:12
Jamie[m]1and for vp9, none fail outright, but 158 produce incorrect output, and 1 fails early after only 3 frames#asahi · 2022-01-23 15:08
Jamie[m]1for hevc: 172 fail outright, 96 produce different output, out of a total 383#asahi · 2022-01-23 14:57
Jamie[m]1in that case, is it a bug that ffmpeg is delegating incompatible videos to hardware?#asahi · 2022-01-23 14:57
Jamie[m]1hmm#asahi · 2022-01-23 14:55
Jamie[m]1what about the conformance ones though?#asahi · 2022-01-23 14:55
Jamie[m]1yeah#asahi · 2022-01-23 14:51
Jamie[m]1In the case of all(?) the others that fail to produce frames, it's [h264 @ 0x137e0ebc0] hardware accelerator failed to decode picture