#asahi-dev · 2023-05-30 13:10
marcandoing another build...#asahi-dev · 2023-05-30 13:08
marcancould be#asahi-dev · 2023-05-30 13:06
marcanlet's do that#asahi-dev · 2023-05-30 13:06
marcanbut yeah maybe I should do it the other way around#asahi-dev · 2023-05-30 13:06
marcankettenis: I only chose the start value because the Linux UAPI only says the driver is allowed to round the stop value :D#asahi-dev · 2023-05-30 12:56
marcanah, probably because the way SMC works it doesn't really do this properly#asahi-dev · 2023-05-30 12:56
marcannope... that didn't work. argh.#asahi-dev · 2023-05-30 12:40
marcanrunning another build now#asahi-dev · 2023-05-30 12:39
marcanI moved the GPIOs to the parent OF node which I think will make the of fwnode stuff defer the probe for us#asahi-dev · 2023-05-30 12:39
marcanit's a dependency problem with SMC (because the way we do this we can't defer at that point)#asahi-dev · 2023-05-30 12:37
marcanit actually worked for me on j413 but it complained about the GPIOs anyway#asahi-dev · 2023-05-30 12:37
marcanjannau: that was the other thing.#asahi-dev · 2023-05-30 12:36
marcantagging -2 kernel with the PMU thing and another thing that I found is broken#asahi-dev · 2023-05-30 12:35
marcanthe UAPI spec says the driver "may round the value to a supported value" so I'm considering this an extreme case of rounding :p#asahi-dev · 2023-05-30 12:34
marcanso I arbitrarily chose the start threshold to drive the actual effect#asahi-dev · 2023-05-30 12:34
marcanit's binary, it's either 75/80 or 100/100#asahi-dev · 2023-05-30 12:07
marcannew package set pushed to asahi-dev#asahi-dev · 2023-05-30 11:46
marcanjannau: ah yes, will be in -2#asahi-dev · 2023-05-30 11:45
marcanmps: the one I just tagged#asahi-dev · 2023-05-30 11:34
marcantagged asahi-6.3-1#asahi-dev · 2023-05-30 11:25
marcanI think we can have a tag soon#asahi-dev · 2023-05-30 11:25
marcanpushed the new charge control stuff#asahi-dev · 2023-05-29 15:17
marcansven: nice :D#asahi-dev · 2023-05-28 13:16
marcansome packet thing#asahi-dev · 2023-05-28 13:16
marcanit's higher layer, yeah#asahi-dev · 2023-05-28 11:45
marcanyeah#asahi-dev · 2023-05-28 11:40
marcanis tb+altmode a thing? like 2 lanes for TB 2 for DP?#asahi-dev · 2023-05-28 11:40
marcanyeah, could be#asahi-dev · 2023-05-28 11:32
marcanah not on the MBPs at least, but then I wonder how that works over a single ATC...#asahi-dev · 2023-05-28 11:31
marcanjannau: strings suggest the peering stuff is for atc, HDMI should be on lpdp no?#asahi-dev · 2023-05-28 11:25
marcan13.2 actually, but yeah that also makes sense#asahi-dev · 2023-05-28 10:21
marcanwild guess though, it could also just be a tiling thing (but I thought we decided that was done with a single dispext)#asahi-dev · 2023-05-28 10:20
marcanit would make sense that dispext is in charge of the DP protocol so MST is implemented with a side channel with another dispext, while the dual pipe thing is for Thunderbolt only which uses discrete DP tunneled SST streams for multimonitor use cases.#asahi-dev · 2023-05-28 10:19
marcanthat wasn't there before... DisplayPort MST support ganging up two dispexts (at a layer above the crossbar)?#asahi-dev · 2023-05-28 10:19
marcansven, jannau: from latest DCP firmware ^^#asahi-dev · 2023-05-28 10:19
marcanGot peer ID %d#asahi-dev · 2023-05-28 10:19
marcandispext%d: peer pipe dispext%d, max pixel width %d, max pixel rate %d#asahi-dev · 2023-05-28 10:19
marcanpeer-pipe-cfg#asahi-dev · 2023-05-27 05:03
marcanthe HV also intercepts power domain controls so macOS doesn't go trying to shut down the ones the HV uses#asahi-dev · 2023-05-27 05:02
marcanm1n1 also has code to forward this to the Linux DT for the same reason#asahi-dev · 2023-05-27 05:01
marcaneach port has an independent controller, DART, etc. so we can just steal one#asahi-dev · 2023-05-27 05:01
marcannicolas17: we remove specifically the one port m1n1 uses#asahi-dev · 2023-05-27 03:08
marcanpovik: awesome :D#asahi-dev · 2023-05-23 00:47
marcanalyssa: that's lina's 4K stuff#asahi-dev · 2023-05-22 12:44
marcanso the KDE settings are mildly confusing but accurate for what libinput does today#asahi-dev · 2023-05-22 12:44
marcandoesn't look like it AFAICT#asahi-dev · 2023-05-22 12:43
marcan_jannau_: I mean I don't know whether swapping right/middle for clickpads is exposed#asahi-dev · 2023-05-22 12:33
marcannot sure if libinput exposes that or not#asahi-dev · 2023-05-22 12:33
marcanfull click is fixed at 2=right 3=middle#asahi-dev · 2023-05-22 12:33
marcanand from what I can tell that setting affects *only* tap to click