#
✕ nick:marcan
#asahi-dev · 2022-10-01 23:41
marcanjannau: no objections (as long as we keep the current support too)
#asahi-dev · 2022-10-01 23:40
marcanwoke up, and of course Manjaro is doing a Manjaro. Sigh.
#asahi-dev · 2022-10-01 15:27
marcanyou're in .de right? I can ping the sponsor about this, maybe just wait until next week :)
#asahi-dev · 2022-10-01 15:24
marcanhence why I said the T2 thing
#asahi-dev · 2022-10-01 15:24
marcansven: yup
#asahi-dev · 2022-10-01 15:22
marcanI assume so? it'd be dumb for it not to
#asahi-dev · 2022-10-01 15:21
marcanbut the display stuff should be standard I'd hope
#asahi-dev · 2022-10-01 15:20
marcanwell, I wouldn't be surprised if it has some cursed thunderbolt stuff a la T2
#asahi-dev · 2022-10-01 15:20
marcanprobably
#asahi-dev · 2022-10-01 15:18
marcan(otherwise I'll get it, when I have the bandwidth for this)
#asahi-dev · 2022-10-01 15:18
marcanyou're in line for that if you sign up for it ;)
#asahi-dev · 2022-10-01 15:18
marcandunno if I mentioned we have an apple studio display donor
#asahi-dev · 2022-10-01 15:17
marcanheh
#asahi-dev · 2022-10-01 15:17
marcan(3 displays, two have to have the same clock)
#asahi-dev · 2022-10-01 15:17
marcan(would still be very useful, I have an IVB laptop with that exact limitation)
#asahi-dev · 2022-10-01 15:16
marcanlike does it only work with identical displays?
#asahi-dev · 2022-10-01 15:16
marcanwhat I don't know is how it deals with EDIDs and modes
#asahi-dev · 2022-10-01 15:16
marcanbut it would also allow 4 displays on the M1 Pro machines, which is cute
#asahi-dev · 2022-10-01 15:16
marcan(this would require some funky logic in the driver to allocate DCPs discretely first, then start oversubscribing them if the user turns on too many outputs)
#asahi-dev · 2022-10-01 15:15
marcaninternally you'd just tile it in space and have the layers not overlap
#asahi-dev · 2022-10-01 15:15
marcancan we just take one for each output stream and support separate streams with separate FBs that way, at the cost of losing the cursor plane or similar? probably a lot nicer to the compositor than trying to do something horrible with fake-tiled framebuffers
#asahi-dev · 2022-10-01 15:14
marcanplanes
#asahi-dev · 2022-10-01 15:14
marcanper display
#asahi-dev · 2022-10-01 15:14
marcandisplay layers
#asahi-dev · 2022-10-01 15:14
marcanre hack mode support, DCP does two layers right?
#asahi-dev · 2022-10-01 15:12
marcanand if I had to guess a realistic reason for the 4 display limitation, something something real time memory bandwidth limits? though it doesn't make much sense considering how much real bandwidth this thing has. could also just be marketing / keeping both Studio variants equal
#asahi-dev · 2022-10-01 15:11
marcanah, then yeah, it all checks out
#asahi-dev · 2022-10-01 15:10
marcanjust add the die offset to the xbar regs, does it exist?
#asahi-dev · 2022-10-01 15:10
marcanyes
#asahi-dev · 2022-10-01 15:09
marcannot the first time the ADT lies
#asahi-dev · 2022-10-01 15:08
marcanwhich makes sense, because why would you actually want that? one die can drive all of its ATCs
#asahi-dev · 2022-10-01 15:08
marcanso this sounds like there is no inter-die routing
#asahi-dev · 2022-10-01 15:07
marcanmost likely related to their secret VR project
#asahi-dev · 2022-10-01 15:07
marcanthat's one of the ProRes blocks
#asahi-dev · 2022-10-01 15:07
marcansven: 8 (4x DCPEXT on one die) + StreamCodec
#asahi-dev · 2022-10-01 15:06
marcanjannau: they do, I tried that
#asahi-dev · 2022-10-01 02:08
marcaneither way, we do need a DRM connector created upfront for every physical output, even if they aren't mapped to any DCPs until hotplug
#asahi-dev · 2022-10-01 02:07
marcanbut if we can't even get EDID without a DCP...
#asahi-dev · 2022-10-01 02:07
marcankind of sucks since ideally users would be able to use as many monitors as they want as long as they only actually turn on a certain number (this is how crtc limitations can be implemented in DRM)
#asahi-dev · 2022-10-01 02:07
marcanso yeah, maybe we do have to greedily allocate DCPs as soon as anything is plugged in, due to the architecture
#asahi-dev · 2022-10-01 02:06
marcansven: I want to say the stream allocation needs to happen when the DCP DRM driver actually wants to turn on an output, not when there is merely a monitor connected... but then how do you get EDID ahead of time?
#asahi-dev · 2022-09-30 10:56
marcanbut worth noting
#asahi-dev · 2022-09-30 10:56
marcannot sure what level the DCP interface is at of course
#asahi-dev · 2022-09-30 10:56
marcanbut also, povik: HDMI/DP audio is its own pile of cursedness with things like dolby/DTS/multichannel PCM modes, so this might be more DRM subsystem stuff than ASoC subsystem stuff (I think DRM has helpers for some of this, infoframes/etc?)
#asahi-dev · 2022-09-30 10:55
marcanI honestly expected some extra hardware involved, some DMA engine
#asahi-dev · 2022-09-30 10:55
marcanI'm surprised they're pushing audio through the EPIC interface
#asahi-dev · 2022-09-30 02:14
marcanpovik: let me know if I should drop or include the binding patch in the end
#asahi-dev · 2022-09-29 18:17
marcandepends on whether the maintainer hates backing off patches :)
#asahi-dev · 2022-09-29 18:16
marcanAh, yeah, they should probably go together at least. But if one is already in dmaengine, maybe the other one should just go in there too now
#asahi-dev · 2022-09-29 17:44
marcanwe already had that for some other bindings

plus ›