#asahi-dev · 2026-11-08 22:41
floklijannau: https://github.com/AsahiLinux/linux/commit/fd637905b0d62de21b94ce58524849a45ab742dd breaks other kernel drivers (Tyr for mali). The nixos kernel build configures most things as modules, so breaks for asahi-7.3.3-1. We could configure it out, just wanted to ask first if it's a mistake #asahi-dev · 2026-11-08 22:07
myriahttps://pastebin.com/RHiKKuti haven't tested it myself but supposedly fixes black ttys on m3#asahi-dev · 2026-11-08 22:04
myriajannau: did you pick up the led format patch michael posted for m3 dcp? otherwise ttys will stop working as it'll enable dcp#asahi-dev · 2026-11-08 21:06
JamesCalligeros[m]Go for it#asahi-dev · 2026-11-08 20:59
svendon't think so. hv is still missing m5 pro and a18 pro support but that's not blocking a release#asahi-dev · 2026-11-08 20:58
jannauanything preventing tagging m1n1-2.1.0?#asahi-dev · 2026-11-08 20:33
svenit just has bonus fun like that co-processor now and changed the registers around I think#asahi-dev · 2026-11-08 20:33
svenso I *think* most of the sequencing for the m4pro atcphy stays the same#asahi-dev · 2026-11-08 20:25
jannau995#asahi-dev · 2026-11-08 20:25
jannau`git rev-list --no-merges --count v7.3..asahi/asahi-wip '^v7.3.3'`#asahi-dev · 2026-11-08 20:21
yukayay#asahi-dev · 2026-11-08 20:04
svenm4 pro atcphy looks quite different I think but the good news is that XNU is still very chatty about register names :D#asahi-dev · 2026-11-08 19:50
s0ullight[m]Anyone who can take a gpu trace from macos for me?#asahi-dev · 2026-11-08 16:01
jannauit annoyed me enough write a filter to delete the mails and file an issue#asahi-dev · 2026-11-08 15:54
s0ullight[m]I am hoping to do some draft pr/mr's at some point#asahi-dev · 2026-11-08 15:53
chaos_princessif you are just messing around and not expecting it to go anywhere - do whatever#asahi-dev · 2026-11-08 15:53
chaos_princessfor getting something "real"#asahi-dev · 2026-11-08 15:53
s0ullight[m]For local testing?#asahi-dev · 2026-11-08 15:52
chaos_princessplease, please ask someone for help.#asahi-dev · 2026-11-08 15:52
chaos_princessand if you do things wrong, you deadlock the kernel#asahi-dev · 2026-11-08 15:52
chaos_princesss0ullight[m]: drm fences are extremely subtle (aka poorly specified)#asahi-dev · 2026-11-08 15:51
chaos_princessnot sure how making it more chatty and spammy helps, but it is from that sentiment#asahi-dev · 2026-11-08 15:51
chaos_princessthere was actually a point raised that some people were angry at the bot as it always came bearing bad news#asahi-dev · 2026-11-08 15:51
s0ullight[m]Erm, no. I've mostly been looking at where we loose performance in certain games. Ended up being the virtio implicit sync stuff. Tried some things alyssa didn't like very much but did improve fps by a lot. Ended up stumbling upon lina's comment about fence passing. I got something working locally.#asahi-dev · 2026-11-08 15:50
jannauwho thought that sending "Sashiko has reviewed this patch and found no issues." was good idea?#asahi-dev · 2026-11-08 15:48
chaos_princessplease tell me you are collaborating with at least someone from fdo#asahi-dev · 2026-11-08 15:48
s0ullight[m]I am#asahi-dev · 2026-11-08 15:47
chaos_princessis anyone even working on that#asahi-dev · 2026-11-08 15:40
s0ullight[m]I want to complete figuring out fence passing first. At some point should upstream the hk fixes I have as well#asahi-dev · 2026-11-08 15:38
chaos_princessway back when, marcan also found some more knobs you can poke in the memory controller, but wasn't able to measure anything relevant. if you are curious, it is in his old streams#asahi-dev · 2026-11-08 15:34
s0ullight[m]I'm glad the bandwidth benchmark results with and without pmp ended up relevant.#asahi-dev · 2026-11-08 15:24
chaos_princessno, there are entries for dram banks that you need to turn on on the max, but not on a pro#asahi-dev · 2026-11-08 15:18
s0ullight[m][chaos_princess](https://matrix.to/#/@_oftc_chaos_princess:matrix.org) a bit of a hail mary but could there be a similar entry for the cpu side?#asahi-dev · 2026-11-08 14:47
s0ullight[m]There's no crash log#asahi-dev · 2026-11-08 14:41
chaos_princesscan you get any info of the crash#asahi-dev · 2026-11-08 14:39
s0ullight[m]Didn't help#asahi-dev · 2026-11-08 14:30
chaos_princessso you'd have the chain sgx -> pmp domain -> pmgr thingy#asahi-dev · 2026-11-08 14:29
chaos_princessthere should be examples already#asahi-dev · 2026-11-08 14:29
chaos_princesscan you rewrite the device tree so it is hooked into a gfx power domain, not always-on#asahi-dev · 2026-11-08 14:28
s0ullight[m]Rebooted when I tried to bring back from sleep#asahi-dev · 2026-11-08 14:25
chaos_princessi guess pmp can also clock the fabrics in the other direction#asahi-dev · 2026-11-08 14:24
chaos_princessah#asahi-dev · 2026-11-08 14:24
s0ullight[m]93 was with pmp disabled. It then halved with pmp without entry#asahi-dev · 2026-11-08 14:23
chaos_princessis 93 w/ pmp disabled or is it w/ pmp enabled, but w/o entry?#asahi-dev · 2026-11-08 14:22
chaos_princessgo do a sleep cycle, tell me if your machine survives that#asahi-dev · 2026-11-08 14:21
chaos_princesslol#asahi-dev · 2026-11-08 14:21
s0ullight[m]93#asahi-dev · 2026-11-08 14:21
chaos_princesspmp-less linux?#asahi-dev · 2026-11-08 14:21
s0ullight[m]Metal does 212#asahi-dev · 2026-11-08 14:21
chaos_princesshow much is it supposed to be?