#asahi-dev · 2024-05-13 23:36
chadmed_yeah, which represents silence. we dont want the dsp graph to do anything with a bad sample at all#asahi-dev · 2024-05-13 23:35
chadmed_so if clients try to send NaN, infinites, etc we just ignore them instead of passing them through the dsp chain #asahi-dev · 2024-05-13 23:34
chadmed_non-finite protection#asahi-dev · 2024-05-13 23:29
chadmed_are we happy to just merge those speaker commits now? the lsp-plugins patch has been in releases for months now, and we have NFP as well as proven-good last-ditch limiters on the back end of the dsp graph anyway #asahi-dev · 2024-05-13 22:40
chadmed_many such cases#asahi-dev · 2024-05-13 22:39
chadmed_<sven> it’s surprising usb3 worked _at all_ the way I implemented it#asahi-dev · 2024-05-12 22:51
chadmed_theres efforts to sort that out in upstream r4l aiui #asahi-dev · 2024-05-12 22:51
chadmed_yeah, all of the rust bindings :p #asahi-dev · 2024-05-12 22:40
chadmed_ill try again tonight then. thats promising though#asahi-dev · 2024-05-12 22:37
chadmed_anyway jannau no if i patch out the circular dependencies the kernel just fails to build. something in there assumes acpi is configured or some such#asahi-dev · 2024-05-12 22:37
chadmed_its just over wifi on the j415#asahi-dev · 2024-05-12 22:37
chadmed_my internet is rock solid and when im connected via the studio its fine#asahi-dev · 2024-05-12 22:36
chadmed_its some brcmfmac thing #asahi-dev · 2024-05-12 22:36
chadmed_oh god is my laptop at home having a fit again #asahi-dev · 2024-05-12 22:18
chadmed_idk if that will work of course but its an idea#asahi-dev · 2024-05-12 22:18
chadmed_now that we've fixed the crashing we can probably find ways around the other silly behaviour. e.g. make the primary plane dest rectangle x+64, y+64 then offset the fb placement on that surface so that the cursor plane technically does not clip the edges of the primary surface#asahi-dev · 2024-05-12 22:16
chadmed_again this makes sense given DCP's backstory in the apple sip extended universe#asahi-dev · 2024-05-12 22:16
chadmed_leio: in our case blending = compositing. we can only have two surfaces#asahi-dev · 2024-05-12 00:32
chadmed_im not sure if it's going to be better to put in the work to add like min_x min_y specifications to planes generically or just make the dcp driver reject a swap if the source fb is too small. will compositors respect that? might just test it later#asahi-dev · 2024-05-12 00:22
chadmed_unfortunately this doesnt really explain why it cant do audio bitstream formats. this has been a critical shortcoming of the apple tv for a very long time#asahi-dev · 2024-05-12 00:22
chadmed_this is an issue for dcp as it does not support swap surfaces smaller than 32x32, likely owing to its original use case in apple tvs and idevices. just like how no one needs 4K RAM, who needs a 32x32 framebuffer on a television#asahi-dev · 2024-05-12 00:20
chadmed_but that's kind of irrelevant for us. for optimal video playback etc we need an overlay plane and compositors need to know about it. the issue here is that drm has moved to "universal" planes because most modern hardware doesnt have special snowflake planes _just_ for cursors#asahi-dev · 2024-05-12 00:19
chadmed_nicolas17: i have a suspicion that it doesnt just based on the cursor's behaviour, but i havent traced it or anything#asahi-dev · 2024-05-07 10:47
chadmed_using that also gets us friendly identifiers for free once displayid is wired up #asahi-dev · 2024-05-07 10:46
chadmed_seems possible to rework the firmware string parsing helpers to not take snd_interval/snd_mask for the sieve and just look for whatever the codec has requested. if it cant find it just throw an error back.#asahi-dev · 2024-05-07 10:44
chadmed_i still think this could and should be reworked as an asoc driver. the only thing that doesnt really make sense to me is the avep element parsing. aiui DCPAVAudioService reports the sink's supported capabilities and the driver is putting constraints on the pcm at runtime based on the connected sink#asahi-dev · 2024-05-07 10:41
chadmed_interestingly the samples were dumped _after_ forcing a hard shutdown and my tv played them as the screen went black... dunno how that works#asahi-dev · 2024-05-07 10:41
chadmed_jannau: https://0x0.st/X88Q.txt sio explosion when attempting to use iec958 passthrough in vlc#asahi-dev · 2024-05-06 07:44
chadmed_*sigh* i should boot the 14" pro up more often#asahi-dev · 2024-05-06 07:43
chadmed_>Emerging (44 of 328)#asahi-dev · 2024-05-06 07:00
chadmed_every sink _should_ gracefully handle it#asahi-dev · 2024-05-06 07:00
chadmed_aiui it's up to the sink to parse and decode whatever bitstream the source sends it, which is why linux can get away with iec958 being a passthrough catch-all "digital" mode#asahi-dev · 2024-05-06 06:58
chadmed_(adding SNDRV_PCM_FMTBIT_IEC958_SUBFRAME_LE to supported formats and sending it)#asahi-dev · 2024-05-06 06:57
chadmed_im about to do something like that :p#asahi-dev · 2024-05-06 06:52
chadmed_if it's just pcm frames we might not even need to do anything extra other than that#asahi-dev · 2024-05-06 06:51
chadmed_so we need to tell alsa that it supports the iec958 pcm formats and figure out how to make dcp do that (assuming it supports it [which it should]) #asahi-dev · 2024-05-06 06:49
chadmed_otherwise apple are bit-banging dolby, dts, etc #asahi-dev · 2024-05-06 06:49
chadmed_it should support at least passing through the bitstream to whatever is on the other end of the hdmi out#asahi-dev · 2024-04-29 07:03
chadmed_yeah ok the resize stutter is almost completely gone, 58fps while resizing glxgears and nothing noticeable at all with native wayland windows#asahi-dev · 2024-04-29 07:02
chadmed_jannau: ahhhh that stutter no sorry disregard my report then, i was testing the page half-speed page flips#asahi-dev · 2024-04-28 10:56
chadmed_what the heck#asahi-dev · 2024-04-28 10:56
chadmed_oh and 60 #asahi-dev · 2024-04-28 10:55
chadmed_and 120 too seems like#asahi-dev · 2024-04-28 10:55
chadmed_jannau: still borked at 144 and 165 hz#asahi-dev · 2024-04-28 10:26
chadmed_ill pick and test on the studio#asahi-dev · 2024-04-28 10:24
chadmed_is that included in the new tag?#asahi-dev · 2024-04-28 10:21
chadmed_thanks :)#asahi-dev · 2024-04-28 09:53
chadmed_huh... rust just caused an oops...#asahi-dev · 2024-04-23 22:20
chadmed_jannau: thats the same person who kept bugging me about the audio "firmware" #asahi-dev · 2024-04-14 23:14
chadmed_is there something fundamentally broken about power domains in the kernel? we seem to have these sorts of issues quite frequently with various hw blocks...