#
✕ nick:chadmed
#asahi-dev · 2026-06-23 12:36
chadmedi always find these posts at bedtime, it's rather annoying :P
#asahi-dev · 2026-06-21 13:09
chadmed"the addressing is identical except on this one machine where it has to be shifted right by 4 bits" is the kind of thing the driver should not have to care about given we control the firmware
#asahi-dev · 2026-06-21 13:08
chadmedimo the firmware should abstract away a few more of these stupid little differences between socs otherwise the driver is going to become painful to maintai
#asahi-dev · 2026-06-21 13:04
chadmedi will test on t8103 and t6000 tomorrow
#asahi-dev · 2026-06-21 13:03
chadmedfeel free to test https://github.com/chadmed/linux/tree/avd-test with https://github.com/chadmed/avd-fw/tree/cleanup if youd like
#asahi-dev · 2026-06-21 12:59
chadmedthats the one we were working on all evening
#asahi-dev · 2026-06-21 11:52
chadmedso do i post the ones that didnt get applied as a v2 or resend
#asahi-dev · 2026-06-21 11:41
chadmedhttps://lore.kernel.org/asahi/20250518-mca-fixes-v1-0-ee1015a695f6@gmail.com/ did these commits just never actually get applied lmfao
#asahi-dev · 2026-06-21 11:28
chadmedi think for 7.2 we will squash some of these commits, theyre a bit silly
#asahi-dev · 2026-06-21 11:26
chadmedjannau: please see new-audio and audio-dts. i dropped the graph.dot commits out and fixed macaudio to use the dt prop that ended up being merged (not what was there earlier)
#asahi-dev · 2026-06-21 10:53
chadmedmy t6000 machine is not in production
#asahi-dev · 2026-06-21 10:51
chadmedif you havent gotten to it by tomorrow evening my time ill do it
#asahi-dev · 2026-06-21 10:45
chadmedwhat else needs testing? t6021 worked so i assume t602{0,2} do too
#asahi-dev · 2026-06-21 10:45
chadmedill test t6000 tomorrow evening
#asahi-dev · 2026-06-21 10:39
chadmednever mind then
#asahi-dev · 2026-06-21 10:39
chadmedoh wait no thats not fully implemented yet is it
#asahi-dev · 2026-06-21 10:39
chadmedanyway new problem with hevc just dropped but it's just a null pointer deref so hopefully an easy fix
#asahi-dev · 2026-06-21 10:38
chadmedyeah...
#asahi-dev · 2026-06-21 10:37
chadmedi wonder if this is just not actually rev4. which machines do you have?
#asahi-dev · 2026-06-21 10:36
chadmeddma mask can stay as 41 btw, so it's pmuch just this and the push_rvra/push_address changes
#asahi-dev · 2026-06-21 10:35
chadmedsofus: we did it reddit
#asahi-dev · 2026-06-21 10:32
chadmedone sec
#asahi-dev · 2026-06-21 07:09
chadmedone sec
#asahi-dev · 2026-06-21 07:09
chadmedyeah ive been using v4l2request-copy with mpv
#asahi-dev · 2026-06-21 07:08
chadmedyeah sure, they were PMD not set for 0xc000000 and another address that looked like a buffer. fixed by saving/restoring dart in the runtime pm functions
#asahi-dev · 2026-06-21 05:10
chadmednow we're not getting the vpdone interrupt though, and ffmpeg wont see the thing as a v4l2m2m device
#asahi-dev · 2026-06-21 05:10
chadmedfixed the iova faults by saving and restoring dart state at runtime pm. im still not convinced this is the way forward.
#asahi-dev · 2026-06-21 02:56
chadmedfixed booting on cleanup branch too
#asahi-dev · 2026-06-21 01:37
chadmednvm it boots now, we get iova faults on every frame though :/
#asahi-dev · 2026-06-21 01:27
chadmedsofus: t8112 avd does not boot with either the original or cleaned up fw
#asahi-dev · 2026-06-20 09:21
chadmedyuka: i dont see why not
#asahi-dev · 2026-06-20 07:15
chadmedno problem :)
#asahi-dev · 2026-06-20 06:39
chadmeduse the cleanup branch, ive split the changes out into manageable commits so you can pick and choose what you want
#asahi-dev · 2026-06-20 03:10
chadmedalso please add a licence and assign copyright as you see fit (preferably MIT so that the BSDs can use it, and preferably The Asahi Linux Contributors but entirely up to you what you choose to do)
#asahi-dev · 2026-06-20 03:09
chadmedwhen you get a chance can you please test and make sure i didnt break anything :P
#asahi-dev · 2026-06-20 03:06
chadmedsofus: https://github.com/chadmed/avd-fw/ did some housekeeping
#asahi-dev · 2026-06-19 11:25
chadmedadds ARM to LLVM_TARGETS :p
#asahi-dev · 2026-06-19 11:09
chadmeddo you mind if i fork? i have some suggestions
#asahi-dev · 2026-06-18 11:36
chadmedyuka: https://github.com/AsahiLinux/docs/pull/269
#asahi-dev · 2026-06-17 08:43
chadmeditll get added to dvb at the very least
#asahi-dev · 2026-06-17 08:43
chadmedunless the streamers really put their feet down i suspect itll still end up a "standard" anyway
#asahi-dev · 2026-06-17 08:42
chadmedobv only a problem for avc/hevc/vvc (if it ever happens)
#asahi-dev · 2026-06-17 08:42
chadmedianal but my understanding is that apple has paid whatever licence fees as the "implementor" being that the implementation of the codec is in hardware
#asahi-dev · 2026-06-17 08:40
chadmedah
#asahi-dev · 2026-06-17 08:40
chadmedinsane
#asahi-dev · 2026-06-17 08:40
chadmedno videotoolbox?
#asahi-dev · 2026-06-17 08:39
chadmedwasnt av1 around unofficially before that? or am i thinking of vp9
#asahi-dev · 2026-06-17 08:33
chadmedvideo decoder
#asahi-dev · 2026-06-17 08:33
chadmedavd
#asahi-dev · 2026-06-17 08:02
chadmedyou can probably make that a bit less ambiguous by defining some of those things we know about

plus ›