#
✕ nick:chadmed
#asahi-dev · 2022-11-17 09:57
chadmedah yeah i see, the hotplug thing doesnt work with pro audio. that complicates things on the DSP front significantly tbh
#asahi-dev · 2022-11-17 09:56
chadmedall pw does is complain that theres no valid verbs for the device for some reason
#asahi-dev · 2022-11-17 09:56
chadmedoh yeah i see, what profile do you use the device in? i dont think this matters in pro audio since i cant reproduce it
#asahi-dev · 2022-11-17 09:49
chadmediirc for pw it was some syntax thing in the speaker detection block that caused it to just give up entirely on parsing the profiles. let me see if i still have it somewhere and can test it against 0.3.60
#asahi-dev · 2022-11-17 09:47
chadmedjannau: is this for UCM? i figured out what we needed to change for pipewire a while ago but i cant remember what :/
#asahi-dev · 2022-11-17 03:27
chadmedthey sound significantly better than the previous lot to me but im really sick atm so would like a second opinion from someone who can hear properly :P
#asahi-dev · 2022-11-17 03:26
chadmedhttps://github.com/chadmed/asahi-audio/tree/testing can someone with a j314 or j316 and a brave streak do me a huge favour and test these new IRs?
#asahi-dev · 2022-11-16 02:14
chadmedthe vulkan one should be a pun about magma or lava
#asahi-dev · 2022-11-16 02:12
chadmedpairs well with please-blow-up-my-speakers
#asahi-dev · 2022-11-15 12:22
chadmedthe EM shouldnt register on rc5 anyway since that fix didnt make it in
#asahi-dev · 2022-11-15 12:19
chadmedunless you picked viresh's fix from linux-pm then EAS wouldnt be registering anyway
#asahi-dev · 2022-11-15 12:16
chadmedwithout it dcp is smooth in x11 but sucks in wayland
#asahi-dev · 2022-11-15 12:15
chadmedit reliably destroys software rendering performance on my 14"
#asahi-dev · 2022-11-15 12:15
chadmeddid you merge "working" EAS by any chance?
#asahi-dev · 2022-11-13 08:14
chadmedhowever now we _really_ need pipewire to hide nodes from the rest of userspace :P
#asahi-dev · 2022-11-13 08:14
chadmedadding the equal loudness compensator to the pipewire filter chain helps a decent amount with sound quality at low volumes, as youd expect
#asahi-dev · 2022-11-08 04:14
chadmedohh right ok, i thought you meant viresh's regulator patch
#asahi-dev · 2022-11-08 01:11
chadmedunless its just as i suspected and making stupid scheduling decisions becuase our power values are stupidly inaccurate because theyre mostly noise...
#asahi-dev · 2022-11-08 01:11
chadmedkazuki: not quite sure how this patch could break sched domains other than there being something wrong deeper in the scheduler past the point it gets power info from the EM
#asahi-dev · 2022-11-07 07:24
chadmedkazuki: how does it break? are tasks not being properly spread across cores/clusters?
#asahi-dev · 2022-11-05 02:21
chadmedhard to A/B test with sw rendering too since its always going to wake up pcores to render stuff, but the clusters do seem to stay in lower pstates for longer now
#asahi-dev · 2022-11-05 02:20
chadmedprobably both because our power values are wildly inaccurate and the added latency of swapping the render threads across clusters constantly to minmax for J/inst
#asahi-dev · 2022-11-05 02:18
chadmedi wouldnt enable it until the GPU driver lands in asahi though since it absolutely slaughters sw rast performance
#asahi-dev · 2022-11-05 02:18
chadmedat least EAS will work for us on 6.2 so its still a win
#asahi-dev · 2022-11-05 02:17
chadmedviresh was nice about it but my god i felt like an idiot
#asahi-dev · 2022-11-05 02:15
chadmedin fairness, the function that was _actually_ failing because of it was doing so silently
#asahi-dev · 2022-11-05 02:15
chadmedamarioguy: at least you didnt lead a maintainer on a wild goose chase over an ignored compiler warning :P
#asahi-dev · 2022-11-04 01:38
chadmedi have to put the opp stuff on ice for a few days now though, but i think the smart long term solution is to detangle the opp-microwatt prop from the regulator stuff altogether, since its stated purpose hasnt really got much to do with regulators anyway and tethering it to that stuff was a lazy hack
#asahi-dev · 2022-11-04 01:36
chadmedzzwysm: the DTs are fine, it looks like theres a race between adding a node to opp_list and parsing the next OPP. there doesnt seem to be any other way that branch could be taken
#asahi-dev · 2022-11-03 14:54
chadmedactually disregard idk whats causing this
#asahi-dev · 2022-11-03 14:50
chadmedthere was actually one issue with viresh's patch, but now with that fixed cpufreq tries to add multiple opp tables per cpu which causes the new parsing stuff to error out since an opp table already exists
#asahi-dev · 2022-11-03 14:49
chadmedmarcan: is the cpufreq driver supposed to try and init multiple times per cpu?
#asahi-dev · 2022-11-03 13:05
chadmedah my bad i thought you were talking about viresh's new patch series
#asahi-dev · 2022-11-03 13:02
chadmed5th patch* i should say
#asahi-dev · 2022-11-03 13:02
chadmedi replied to the 5th message in that patch series
#asahi-dev · 2022-11-03 13:01
chadmedyeah he needs to delete the stuff in the commit message about it fixing AS, see my reply to 5/5
#asahi-dev · 2022-11-03 12:59
chadmedmarcan: no, that patch series does not change the behaviour of that code path at all. it just shuffles stuff around
#asahi-dev · 2022-11-03 12:34
chadmeds/DT/opp table/
#asahi-dev · 2022-11-03 12:34
chadmedwhy it was added to all of that is a mystery to me since it was added specifically for platforms that do not express their regulators in the DT...
#asahi-dev · 2022-11-03 12:33
chadmedso next angle of attack is taking opp-microwatt out of the regulator nonsense entirely
#asahi-dev · 2022-11-03 12:32
chadmedlooks like Viresh doesn't want opp_parse_supplies() to return anything valid without a microvolt property since that breaks some consumers of the regulator struct
#asahi-dev · 2022-11-01 01:56
chadmedyeah theres a PMU driver, accessible through the normal sysfs mechanism i think
#asahi-dev · 2022-11-01 01:55
chadmedaiui we dont yet put the memory controller into the higher pstate, so that could still be the case
#asahi-dev · 2022-10-30 10:26
chadmedthis one works
#asahi-dev · 2022-10-30 10:26
chadmedhttps://lore.kernel.org/asahi/20221030101546.29306-1-jcalligeros99@gmail.com/T/#u
#asahi-dev · 2022-10-30 10:25
chadmedlinux-pm and asahi mailing lists
#asahi-dev · 2022-10-30 10:19
chadmedsent
#asahi-dev · 2022-10-30 10:10
chadmedi do have a built, tested and working version here
#asahi-dev · 2022-10-30 10:10
chadmedill send a v2
#asahi-dev · 2022-10-30 10:09
chadmedyeah i sent the wrong version lmao, first time using git send-email

plus ›