#asahi-gpu · 2022-12-27 15:42
alyssaand does not affect overhead on the CPU to actually submit work (right now the GPU is idle a ton of the time -- GPU speed ups are then subject to Amdahl's law)#asahi-gpu · 2022-12-27 15:42
alyssabut does not affect the latency nor throughput of system memory (which we often stall on)#asahi-gpu · 2022-12-27 15:41
alyssalina: I would not expect 78% faster for 720 -> 1287, because that's the clock only for actual GPU operation#asahi-gpu · 2022-12-27 11:47
bluetail8oh. Thats good#asahi-gpu · 2022-12-27 10:56
chadmedaiui we cannot overclock this hardware as the pstates are controlled by the platform and we just ask the firmware nicely to switch states#asahi-gpu · 2022-12-27 10:48
bluetail815% is a ton#asahi-gpu · 2022-12-27 10:48
bluetail8is it effectively an overclock or is it within apples legit p state design?#asahi-gpu · 2022-12-27 10:47
bluetail8alyssa will it affect longevity negatively?#asahi-gpu · 2022-12-27 05:04
linahowever, 1 (which is the base on laptops) is 396MHz, so if the laptops are stuck there that's quite a bit worse...#asahi-gpu · 2022-12-27 05:03
linaalyssa: 3 is 720MHz and 6 is 1287 MHz, so I would expect 78% faster if it scales with clock#asahi-gpu · 2022-12-27 04:32
alyssatoday on broken nir_lower_blend: per-RT logic ops?#asahi-gpu · 2022-12-27 03:00
alyssanot 5x faster but that's still substantial#asahi-gpu · 2022-12-27 02:59
alyssat-rex is >15% faster#asahi-gpu · 2022-12-27 02:59
alyssalina: took a cue from the "wtf pstates?" discussion and hacked my DTS to use 0x6 as the new base#asahi-gpu · 2022-12-26 15:30
jndoes git diff -b help?#asahi-gpu · 2022-12-26 15:26
alyssabut it is entirely too boxing day for me to care to review more than I already have#asahi-gpu · 2022-12-26 15:26
alyssathe diff isn't horrible since the indentation isn't changing (3 spaces before and after)#asahi-gpu · 2022-12-26 15:26
alyssalina: !20434 waiting for review :D#asahi-gpu · 2022-12-26 14:52
alyssaokay, that's better :p#asahi-gpu · 2022-12-26 14:52
alyssa 66 files changed, 2742 insertions(+), 2768 deletions(-)#asahi-gpu · 2022-12-26 14:47
alyssabecause this is supposed to be basically the existing style#asahi-gpu · 2022-12-26 14:47
alyssaTBD what it's unhappy about#asahi-gpu · 2022-12-26 14:46
alyssayeah#asahi-gpu · 2022-12-26 14:46
jna clang-format run?#asahi-gpu · 2022-12-26 14:44
alyssaYikes#asahi-gpu · 2022-12-26 14:44
alyssa 66 files changed, 22916 insertions(+), 23168 deletions(-)#asahi-gpu · 2022-12-26 06:56
Ry_DarcyAdditional information to my comment above - just reading the news, no additional load (youtube, videos etc.) on the browser/GPU.#asahi-gpu · 2022-12-26 06:45
corionUhm. Sorry. Paste has apparently overtaken my right click on trackpad. :)#asahi-gpu · 2022-12-26 06:45
corion follows https://paste.debian.net/1265302/ Perhaps this has already been #asahi-gpu · 2022-12-26 06:44
Ry_DarcyMorning all, running Debian Testing ( 6.1.0-asahi) on an M1 Mini using FF 108. The browser has crashed twice this morning. An excerpt of the ring buffer follows https://paste.debian.net/1265302/ Perhaps this has already been reported, no idea. #asahi-gpu · 2022-12-26 06:11
corionlina: thanks!#asahi-gpu · 2022-12-26 03:43
linaI think the pstate machinery isn't running at all...#asahi-gpu · 2022-12-26 03:42
linaI'll have to look into that...#asahi-gpu · 2022-12-26 03:42
linaBut why? The python experiments did boost all the way up...#asahi-gpu · 2022-12-26 03:41
linaApparently we're stuck at pstate 3... that explains the worse than expected performance...#asahi-gpu · 2022-12-26 03:41
linaAnd speaking of that...#asahi-gpu · 2022-12-26 03:41
linacorion: It's not really implemented "properly" yet, but you can try `echo 0x40000 > /sys/module/asahi/parameters/debug_flags` and look at dmesg. It will be spammy though...#asahi-gpu · 2022-12-25 20:41
corionNo idea if that's even possible with m1 on linux. I doubt it is, but figured I'd ask. :)#asahi-gpu · 2022-12-25 20:41
corion... or like nvidia-smi.#asahi-gpu · 2022-12-25 20:39
corionHow could one view the gpu load? Sort of like a radeontop.#asahi-gpu · 2022-12-24 04:29
linaalyssa: Do it!!! ^^#asahi-gpu · 2022-12-23 21:22
alyssa(I don't think that's controversial ..)#asahi-gpu · 2022-12-23 21:22
alyssaif we do clang-format asahi, I would want agxv to be covered by the same set of rules#asahi-gpu · 2022-12-23 21:22
alyssa(when should that happen? Unsure tbh.)#asahi-gpu · 2022-12-23 21:21
alyssaella-0: I don't expect this to affect you much before you're merged upstream#asahi-gpu · 2022-12-23 21:18
alyssaella-0: ^^#asahi-gpu · 2022-12-23 21:18
alyssaoh and maybe there are changes to core on the agxv branch#asahi-gpu · 2022-12-23 21:18
alyssabut it should be able to reformat asahi/mesa (and agx/next and whatever other branches we have hanging around... I guess the compute kernel one?)#asahi-gpu · 2022-12-23 21:18
alyssaI wrote a script to rebase MRs.. I proved it correct but haven't tested yet#asahi-gpu · 2022-12-23 21:04
alyssayou down for clang-format for asahi too?