#asahi · 2023-03-07 12:35
chadmedideally id like there to be a module parameter to suppress this behaviour because i want to run without the daemon myself (to dogfood userspace dsp safety and easily run tests on new devices without patching the kernel) #asahi · 2023-03-07 12:34
chadmedyeah exactly, the driver should go "huh havent heard from userspace in a while it probably died" and put the codecs into safe mode#asahi · 2023-03-07 12:31
chadmedthis process should be transparent to userspace entirely as the safe mode register does not smush the DVC register or any other setting we expose to userspace #asahi · 2023-03-07 12:30
chadmedthe driver should kick the codecs into safe mode upon any condition where it cannot be satisfied that userspace is guaranteeing safety#asahi · 2023-03-07 12:29
chadmedthe daemon needs to be able to stop all that and just keep kicking the driver#asahi · 2023-03-07 12:29
chadmedwaste of cycles #asahi · 2023-03-07 12:29
chadmedand also that would be wasteful as the daemon would be constantly doing operations on a zeroed buffer #asahi · 2023-03-07 12:28
chadmedwe cant just keep a stream to the pcm open all the time since we need to be able to change sample rate etc#asahi · 2023-03-07 12:28
chadmedalso re speaker safety we need something that the daemon can kinda poll while its not streaming since the idea is for the codec to go into safety mode (a special hardware register mode) when the deamon doesnt kick the driver#asahi · 2023-03-07 12:27
chadmedmort_: err sorry meant to ping you not myself :P #asahi · 2023-03-07 12:26
chadmedchadmed: 0.61 works fine, 0.64 hard deprecated a flag the kernel folks still use so thats a no no#asahi · 2023-03-07 11:49
chadmedrust tooling can be super annoying, especially since kernel rust hasnt really settled yet#asahi · 2023-03-07 11:49
chadmedwhich will install bindgen into your cargo env #asahi · 2023-03-07 11:49
chadmedfor the latest released tag you should be using rust 1.62 and the bindgen-cli crate#asahi · 2023-03-07 11:45
chadmedthe uapi is not stable and the kernel tags usually only work with one or two mesa commits before it breaks again#asahi · 2023-03-07 11:45
chadmeddid you install the corresponding mesa package too? #asahi · 2023-03-07 11:44
chadmedtheyre actually different things afaict #asahi · 2023-03-07 11:43
chadmedtry using bindgen-cli not cbindgen #asahi · 2023-03-07 10:31
chadmedits easier for us to add stuff to our devicetrees that we maintain than it is to get totally new bindings and mechanisms upstreamed #asahi · 2023-03-07 10:30
chadmedjust "go to this code if the idle state is wfi"#asahi · 2023-03-07 10:30
chadmedthat way you dont need to introduce any _new_ checks per se #asahi · 2023-03-07 10:30
chadmedthen have the driver probe based on that#asahi · 2023-03-07 10:29
chadmedif idle-states has binding documentation you could propose just adding "wfi" as a valid value or somethign #asahi · 2023-03-07 10:10
chadmedright right so your plan is to just hack in a path where it keeps the core in WFI if theres no psci interface available? #asahi · 2023-03-07 10:04
chadmedlike isnt that the whole problem#asahi · 2023-03-07 10:03
chadmedcpuidle-arm relies on calling into firmware to handle psci doesnt it?#asahi · 2023-03-06 09:40
chadmedyou could install gentoo :P#asahi · 2023-03-06 08:00
chadmedbut no ones maintained the patches since 5.19 #asahi · 2023-03-06 08:00
chadmedit involves some horrific patches and bypassing some DARTs but it is possible#asahi · 2023-03-06 07:59
chadmed:D#asahi · 2023-03-06 07:58
chadmed(i already tried( #asahi · 2023-03-06 07:58
chadmedRetr0id: dont bother its LOAD sections are 0x1000 aligned#asahi · 2023-03-04 23:57
chadmedbattery life should improve significantly once we merge energy-aware scheduling stuff but there are some issues with interactivity in some edge cases so we want to find some ways to resolve those before doing so#asahi · 2023-03-04 14:55
chadmednah vrr doesnt work at all yet, it was just "please set the refresh rate to 120hz", which dcp said it had done but in actuality hadnt#asahi · 2023-03-04 14:53
chadmedpromotion is just marketing wank for VRR but getting 120hz to work at all is predicated on getting it to work#asahi · 2023-03-04 14:39
chadmedlike if theyve had their chance to respond, do they lose the right to sook when patches break their stuff? #asahi · 2023-03-04 14:39
chadmedis there anyone upstream against just shutting them out if they want to be like that?#asahi · 2023-03-04 14:38
chadmed20% means 5% and only on a day where youve got no work orders to do anything else (these dont exist)#asahi · 2023-03-04 14:37
chadmedhonestly more than i expected from broadcom but yeah :/#asahi · 2023-03-04 14:36
chadmeddo they really just pay one person to look at the mailing list every few weeks so they can pretend they give a shit?#asahi · 2023-03-04 14:34
chadmedthey have their own baseband/modem already #asahi · 2023-03-04 14:32
chadmedi sure hope so, i dont want to sell my shiny new mini to get another laptop i dont need :P#asahi · 2023-03-04 01:25
chadmedmaybe page size mismatches depending on what exactly youre doing, but thats not endemic to apple silicon #asahi · 2023-03-04 01:25
chadmedZeroine: at this stage no#asahi · 2023-03-04 00:21
chadmedits bad enough that it takes me almost an hour to download from scratch, which is pitiful considering i have an easily saturated 100mbps downlink#asahi · 2023-03-04 00:20
chadmedthe overhead is insane #asahi · 2023-03-04 00:20
chadmedfwiw if its download time and space mojang are worried about youd think theyd just bundle up minecraft releases in a compressed archive, the mc launcher for some reason still downloads every single file with its own GET#asahi · 2023-03-04 00:19
chadmedbut apart from that yeah it should work #asahi · 2023-03-04 00:19
chadmedbluetail9: minecraft ships with some graphics libraries that have GLES support stripped out for some reason so you need to replace them with the upstream versions and do some patching #asahi · 2023-03-03 22:53
chadmedi believe someone actually did get minecraft working at some point with patched java libraries!