#
…

2022-11-20

00:59capta1nt0ad has joined #asahi-dev
01:16jlco has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
02:00Dcow has joined #asahi-dev
02:19riker77_ has joined #asahi-dev
02:24riker77 has quit [Ping timeout: 480 seconds]
02:24riker77_ is now known as riker77
02:26tanty has quit [Remote host closed the connection]
02:27tanty has joined #asahi-dev
02:40capta1nt0ad has quit [Quit: Konversation terminated!]
02:50user982492 has joined #asahi-dev
03:14psykose_ has quit [Remote host closed the connection]
03:14psykose has joined #asahi-dev
03:15lamlam has joined #asahi-dev
03:16lamlami believe i had a brainfart yesterday
03:16lamlamWHO LIKES CONVOLUTIONS
03:16lamlamAND MATMULS
03:17lamlamTHE ANE WORKS
03:17lamlamIT WORKS
03:18lamlamTESTS ARE NCHW 2D CONV
03:19lamlamall it really is a multiply-add circuit, so it can be configured to do matmuls/elemwise too :)
03:19lamlamfrom apple themselves:
03:19lamlam"Neural processor circuit is a *configurable circuit* that performs neural network operations on the input data based at least on kernel data"
03:19lamlamkernel (image kernel) just being input2
03:20lamlamtldr, does MADDs with input1 with input2, with LOTS of config options
03:20lamlamill send the poc soon
03:24lamlamcaveats, the *entire* 0x4000 - 0x20000 range is configs so lots of work to do
03:24tpw_rulessounds pretty convoluted
03:24lamlambut ive got the hardware queue/ push / trigger interface down
03:38marcanI think I'm going to push the new installer (just the installer) ahead of the actual image updates. I'm tired of people running into the same old diskutil failures and posting everything except the actual useful info on github :p
03:48capta1nt0ad has joined #asahi-dev
04:02marcanpushed installer v0.5 to prod
04:06tpw_rulesi guess it would be hard until this current dev push is finished, but next time you update the main asahi non-dev repo please tag the PKGBUILD repo or set up a separate branch or whatever
04:10lamlam has quit [Quit: Page closed]
05:08kevans91 has joined #asahi-dev
05:46jnrubyrubyruby has joined #asahi-dev
06:00jnrubyrubyruby has quit [Quit: Page closed]
06:37psykose has quit [Ping timeout: 480 seconds]
06:48marcantpw_rules: that's the plan
06:49marcanI'll make a separate branch that tracks the commit at which asahi-dev gets synced to asahi
06:54capta1nt0ad has quit [Remote host closed the connection]
07:31user982492 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
08:07weitcis has joined #asahi-dev
08:32SSJ_GZ has joined #asahi-dev
09:00capta1nt0ad has joined #asahi-dev
09:12capta1nt0ad has quit [Remote host closed the connection]
09:14chadmedcan we just port coreaudio to linux seriously...
09:15povikwhat now?
09:15chadmedi just want AudioUnits :P
09:16capta1nt0ad has joined #asahi-dev
09:16chadmedtheyre using an AU to split off bass to a virtual channel to apply independent dynamics filtering so that the cones dont hit the chassis on big bass hits
09:17chadmedtheres no sane way to copy this with IRs and LADSPA plugins inside pipewire, so the choices are nonexistent bass at low volumes or popping noises at high volumes
09:18chadmedsame goes for equal loudness, you cant link a sink's volume control to a ladspa plugin port so the equal loudness plugin doesnt actually do anything when inserted into the chain
09:21chadmedthe actual sound is good now though, i figured out how to deal with those horrible honking mids
09:22weitcis has quit [Remote host closed the connection]
09:23chadmedand most _good_ sources sound fine at low volume because theyve been mastered properly, but web content is going to be a bit of a wild west
09:25kode54the AU in question would be using a crossover strategy
09:25kode54then you apply your dynamics filtering to the output of the crossover
09:25chadmedyeah i know how it works, apples dsp graph is just stored on disk in plain text so we can see what they do
09:26kode54but there's no useful crossover filters for Linux?
09:26kode54I know someone made some unfortunately closed source speaker crossover plugin for foobar2000, of all things
09:26chadmedthere are, but the way pipewire loads that stuff in means we cant actually control them after theyre loaded, so equal loudness is never going to work
09:27chadmedand in terms of doing the dynamics on bass and sub bass, a filter chain can only link a port exactly once
09:27kode54I thought it was down to placing a dynamics compressor on the output
09:28chadmedso we cant, for example, split off the bass to do DRC on it, then link that back into the convolver with the non-compressed signal, then sum those and send off to the correct cone
09:28chadmedhonestly its not the end of the world if we have to live with less bass than macos
09:29chadmedmacos has too much
09:29kode54you can't do convolver per speaker?
09:29kode54crossover splitter on the correct ranges, DRC on the outputs, convolve, send to speakers
09:29chadmedthats what im doing now, but pipewire wont let you then sum two inputs into the convolver (i.e. the bass and everything else)
09:30kode54you need merged signal for the convolver
09:30chadmedyeah theres no "subwoofer" per se, the woofer drivers are dual opposed
09:31kode54I didn't think there was a subwoofer
09:31kode54I thought there was a six-way crossover going on
09:31chadmedso theyre both full range and need 20-1200, but we only want to apply dynamics to 20-400ish
09:31kode54well, three way per channel
09:31chadmedwe cant do that
09:31kode54have you looked into a multi-band compressor then?
09:31kode54or does Creative own the patent on that with their Crystalizer bs
09:32chadmednah lsp has a multiband compressor
09:32kode54then it needs to be a compressor that only applies to specific band that's configurable
09:32chadmedi might play around with that, the challenge there being configuring it via pipewire's text files
09:33kode54maybe figure out a setting in some LADSPA config app, and then see if you can recreate the settings as a pipewire settings file
09:33kode54former to see if it will work first
09:33chadmedyeah im using carla, from which i can export each plugins config
09:34chadmedits just figuring out what pipewire calls the plugins controls etc
09:34kode54ah, that's a pain
09:34kode54meanwhile
09:34kode54I'm thinking, with the way dev is progressing
09:34kode54it's on the right track to replace my macOS setup by the time Apple decides if they ever want to drop support for the M1 series
09:35kode54no idea how many years that will take
09:35kode54maybe I might jump ship sooner if it turns out the Linux hardware acceleration works much better for games than macOS native games
09:35kode54that's offtopic though
09:35kode54I wish you luck with your speaker configuration woes
09:36kode54ideally a multi-band compressor will let you pick a band to apply the dynamics compression to, and merge that with the unprocessed signal ranges
09:36kode54all without having to use complex graph setup
09:37kode54I'm hoping for that that the plugin will just have controls to create N bands and set their frequency ranges, possibly rolloff behavior
09:37kode54and possibly dynamics behavior to apply
09:37kode54I'd have to look at the filter in carla to know for sure
09:37chadmedthe lsp one has 8 bands hard coded but theyre all bypass-by-default so i can just configure one from pipewire
09:37chadmedthe only thing now is that we introduce a dependency on lsp-plugins
09:38chadmedand ive been trying to avoid that
09:40kode54ah, yeah
09:40kode54maybe instead you can use some coding wizardry to generate ideal filters for the crossover
09:41kode54and bundle the ideal filter with a dynamics compressor that recombines the output
09:41kode54ie. make an ideal crossover and package it separately
09:41chadmedi am the last person in the universe who should be allowed near dsp code :P
09:41kode54but then it's single purpose
09:42chadmedthats also kinda hacky and part of the motivation for this is to demonstrate A Better Way(tm) to handle these sorts of devices on bog-standard linux
09:42kode54so I guess you may be forced to make systems depend on that lsp-plugins package
09:43chadmedyeah, which is fine tbh if it means people can use their speakers without crushing the cones against the chassis
10:11thepigguy has joined #asahi-dev
10:44chadmedok if someones game to test ive pushed to asahi-audio
10:44chadmedit works quite well actually
10:46chadmeddownside is the cpu use...
10:57bluetail has quit [Ping timeout: 480 seconds]
10:58bluetail has joined #asahi-dev
11:01chadmed has quit [Ping timeout: 480 seconds]
11:44thepigguy has quit [Quit: thepigguy]
11:50Dcow has quit []
11:53jlco has joined #asahi-dev
11:54___nick___ has joined #asahi-dev
11:55___nick___ has quit []
11:57___nick___ has joined #asahi-dev
13:16Dcow has joined #asahi-dev
13:24Dcow has quit [Quit: Konversation terminated!]
13:30Dcow has joined #asahi-dev
13:39Dcowis there a special way to prepare initramfs on arch asahi linux? apparently mkinitcpio don't like me (stuck on booting)
14:01Dcow has quit [Remote host closed the connection]
14:14kettenis_ is now known as kettenis
14:25gladiac has joined #asahi-dev
14:26ah- has joined #asahi-dev
14:27gladiac has quit []
14:28gladiac has joined #asahi-dev
14:29alyssais ASan supposed to work with 16K pages...?
14:31j`eynot yet
14:31j`eysome patches on lkml, lemme get the link
14:31alyssaugh okay
14:31alyssathat explains that then
14:32alyssaj`ey: userspace libasan I mean, not KASAN
14:32j`eysorry, I dont know then
14:37bisko has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
14:37bisko has joined #asahi-dev
14:48bisko has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
15:01bisko has joined #asahi-dev
15:10Dcow has joined #asahi-dev
15:11Dcowis it possible to boot installed m1n1 (in efi partition) and get logs from kernel boot with proxymode?
15:12bisko has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
15:19bisko has joined #asahi-dev
15:34simon_ has joined #asahi-dev
15:43Dcow has quit [Remote host closed the connection]
15:46simon_ is now known as NewNickname
15:46NewNickname is now known as siontx
15:50Dcow has joined #asahi-dev
15:56kettenisso that was fun; apparently the SMC throws a fit if you send a command to the SPI keyboard/touchpad and don't consume the reply quickly enough
16:24gladiac is now known as Guest2033
16:24gladiac has joined #asahi-dev
16:30Guest2033 has quit [Ping timeout: 480 seconds]
16:36Dcow has quit [Remote host closed the connection]
16:42Dcow has joined #asahi-dev
16:53Dcow has quit [Remote host closed the connection]
17:01siontx has quit [Remote host closed the connection]
17:12siontx has joined #asahi-dev
17:20siontx has quit [Ping timeout: 480 seconds]
17:51jannaukettenis: weird
17:52jannaudcp works on M2!
17:54jannau3 changed lines in drivers/gpu/drm/apple and all three changes will make sven happy
17:54svenwhy will they make me happy? :D
17:54jannaudma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(36));
17:55jannauso the dma_set_mask_and_coherent() calls are not the default anymore
17:55svenah, i remember "complaining" about that ;)
17:57jannauonly dart, devicetree and m1n1 changes
17:58jannauany idea how the aperture start and end could be expressed in the devicetree?
18:00jannauI have currently "apple,vm-start" and "apple,vm-size" copied from the ADT
18:03jannausince I had to change dart anyway I've switched to a shadow L1 table for the locked darts
18:03svenhrm, so you actually need to enforce vm-start and vm-size?
18:05jannaunot sure, in practice probably not
18:05siontx has joined #asahi-dev
18:05svenI don't think there's any documented property for that. iirc there is one iommu that has a custom property ("xxxx,dma-window" maybe)
18:05svenbut if it's not required i'd just skip it
18:06svenquite a few darts have those vm-start/vm-size properties and so far we just ignore all of them
18:06jannaubut I would expect problems if dcp gets gets addresses below vm-start
18:07svencan't you mark that region as reserved if it actually causes problems?
18:07jannauI think none of them has a non-zero vm-start
18:07jannauI still need something to set a 36-bit aperture
18:07svendidn't the pcie ones have that as well?
18:08svendo all DARTs on the M2 support 36 bit? or is there a register where you can get the aperture from?
18:10svenif so i'd just set the domain aperture to that for the M2 darts
18:11jannauwe have DART_T8110_PARAMS3_VA_WIDTH
18:11jannauso, yes we can just use that
18:12svenluckily the iova allocator starts handing out allocations at the very end of its aperture so it'll be very obvious if this breaks anything else :D
18:12TramtristM2 work progressing 👀 🎉
18:13siontx has quit [Ping timeout: 480 seconds]
18:20jannaudcp and usb still work
18:21jannaushouldn't a larger aperture cause only problems for devices that set a larger dma mask
18:21svenbut dwc3 will do that
18:23svenbut essentially only devices that support >32 bit dma should do that and those should all work fine if the iommu also supports that I think
18:41bisko has quit [Read error: Connection reset by peer]
18:42bisko has joined #asahi-dev
18:45gladiac has quit [Quit: k thx bye]
19:01siontx has joined #asahi-dev
19:08jannaunow with all firmware in place everything still seems to work without vm-base/vm-size
19:09tanty has quit []
19:10jannauand using 36-bit VA width
19:10tanty has joined #asahi-dev
19:21lkron has joined #asahi-dev
20:00lkron has quit [Quit: Lost terminal]
20:16jlco has quit [Ping timeout: 480 seconds]
20:16jlco has joined #asahi-dev
20:37jannausigh, https://github.com/AsahiLinux/linux/issues/69 "... we can do ..." filling feature requests is not material participation in development
20:38svenlooks like I can close it :D
20:44kettenisbtw, after the upgrade to macOS 12.6.1, nvme advertises an rtkit endpoint numbered 10
20:44kettenishas that one seen before? and to we have a clue what it is?
20:45svenI remember seeing that somewhere else as well
20:45svenbut I don’t remember what it was
20:51jannauit was added in the ventura (beta) firmware as well
20:54jluthra has quit [Remote host closed the connection]
20:54jluthra has joined #asahi-dev
20:55ashi has joined #asahi-dev
20:58erik3424 has quit [Quit: erik3424]
20:58ashi has quit []
21:04ketteniswouldn't surprise me if 13.0.1 and 12.6.1 use the same nvme firmware
21:07___nick___ has quit [Ping timeout: 480 seconds]
21:11alyssajannau: nice
21:18amarioguy2 has joined #asahi-dev
21:18amarioguy2 has quit [Remote host closed the connection]
21:18bisko has quit [Read error: Connection reset by peer]
21:19bisko has joined #asahi-dev
21:31user982492 has joined #asahi-dev
21:34siontx has quit []
21:44Dcow has joined #asahi-dev
21:45Dcowany hint how to fix: m1n1.proxy.UartTimeout: Expected 1 bytes, got 0 bytes
21:46alyssatry again
21:46Dcowtried like 10 times, zero succes
21:46alyssatry unplugging all usb devices from the proxyclient machine, except for the cable to the m1
21:46alyssatry switching which usb port on the proxyclient machine is used
21:47alyssa(I get that issue unless I use the right port ....)
21:47alyssa(and have nothing plugged to the left port ...)
21:48Dcowwhat is right port for j316?
21:48Dcowconnection to the proxy is detected, is it mean that port is right one?
21:49alyssa has left #asahi-dev [#asahi-dev]
21:49Dcowtype-c to type-c cable is ok to use?
21:49svenplease move this to #asahi
21:50Dcowsure, sorry
22:38user982492 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
22:41tanty has quit []
22:46tanty has joined #asahi-dev
22:51Dcow__ has joined #asahi-dev
22:53Dcow has quit [Quit: dcow]
22:56SSJ_GZ has quit [Ping timeout: 480 seconds]
22:56elvishjerricco has quit [Read error: No route to host]
22:57elvishjerricco has joined #asahi-dev
23:00yrlf has quit [Quit: The Lounge - https://thelounge.chat]
23:00lamlam has joined #asahi-dev
23:02lamlamwas about to commit ane patch rn and github soft-banned me?? "Your account has been flagged. Because of that, your profile is hidden from the public"
23:09lamlam has quit [Quit: Page closed]
23:28povikso you get private repos for free now? :p
23:28povikthey used to charge for that not so long ago!
23:29povikon a more serious note, i am curious what will turn out to be the grounds for it
23:29poviki assume you haven't done anything outright illegal there
23:32weitcis has joined #asahi-dev
23:36weitcis_ has joined #asahi-dev
23:36weitcis has quit [Read error: Connection reset by peer]
23:46chadmed has joined #asahi-dev
23:53weitcis_ has quit [Ping timeout: 480 seconds]
…