#asahi-dev · 2023-12-29 11:42
svenin other news, i was apparently doing the pipehandler switch entirely wrong#asahi-dev · 2023-12-28 21:29
svenis that dp only atcphy still connected to a tipd chip?#asahi-dev · 2023-12-28 20:57
svenok, i'll take a look#asahi-dev · 2023-12-28 20:53
svenhaven't read that far yet. i've only looked at the atcphy code 'cause i'm merging your changes with my local new code#asahi-dev · 2023-12-28 20:51
svenok, guess that means we have no other choice than that apple,mode-fixed-dp property :/#asahi-dev · 2023-12-28 20:45
svenjanneg: just to confirm: atcphy3 is used on e.g. j314 only for dp2hdmi but there are other t600x machines where that same atcphy3 is connected to a regular type c port?#asahi-dev · 2023-12-21 15:57
svenyeah, if you chainload and can see serial over the SBU1/2 lines before that chainload you should be able to see them#asahi-dev · 2023-12-21 15:48
svenwhen the SBU lines aren't muxed to UART RX/TX those chars will just go to nowhere#asahi-dev · 2023-12-21 15:47
svensounds like whatever you use to mux out the serial lines isn't quite ready that early#asahi-dev · 2023-12-20 09:57
svenbroken suspend is because xhci doesn’t wake up the ports, isn’t it?#asahi-dev · 2023-12-20 09:57
svenit’s not used after that function is done #asahi-dev · 2023-12-20 09:57
sven0x14 hardcoded some values for the phy interface when we switch the phy #asahi-dev · 2023-12-20 09:56
svenI don’t think so#asahi-dev · 2023-12-19 17:30
svenyou can also mention that dp altmode and thunderbolt is in progress and should come sometime in 2024#asahi-dev · 2023-12-19 17:20
svenok, i'll take a look at the atcphy changes and merge them with my current code then#asahi-dev · 2023-12-19 17:16
svenjanneg: that hdmi out didn't need change inside atcphy, did it?#asahi-dev · 2023-12-19 17:15
svenmy vacation finally starts tomorrow so i'm planning to clean that up a bit and get more people to test it#asahi-dev · 2023-12-19 17:14
sven(0x14 vs. 14 *sigh*)#asahi-dev · 2023-12-19 17:14
svenusb3 seems to be very stable now, including switching dwc3's PIPE phy over#asahi-dev · 2023-12-19 17:14
svenjanneg: no worries, i completely broke dp altmode with that new tipd code anyway :D#asahi-dev · 2023-12-19 17:12
sventhere's random crap like dumping and reversing what feels like half of that cd321x firmware e.g. that just takes time#asahi-dev · 2023-12-19 17:12
svenwe know how it (and all the other usbc mess like tbt) works and very slowly getting there#asahi-dev · 2023-12-19 17:12
svendp alt mode doesn't work anywhere#asahi-dev · 2023-12-19 17:11
svenas usual, the last 20% take the most of the time#asahi-dev · 2023-12-19 17:11
svenyeah, we're slowly picking away at dp altmode#asahi-dev · 2023-12-18 21:52
svenyup…#asahi-dev · 2023-12-18 21:34
svenalso not smart: writing writel(14,..) instead of writel(0x14,…) and then spending weeks wondering why atcphy is sometimes unstable #asahi-dev · 2023-12-15 08:25
svenah, nice #asahi-dev · 2023-12-15 08:24
svenfor display port the edid is sent over the aux channel which is slow but still a differential pair iirc #asahi-dev · 2023-12-15 08:22
svenpretty sure i could easily hack that to allow fully custom edids. and also pretty sure something like that exists on aliexpress ;)#asahi-dev · 2023-12-15 08:22
sveni have a cheap hdmi splitter that allows to change the edid #asahi-dev · 2023-12-15 08:16
svenapparently there’s „port dfu“ now which might require personalized firmware updates :/#asahi-dev · 2023-12-15 08:16
svenI wonder if the tipd swd is connected to some soc gpio though. but then again, the newer typec controllers are probably more locked down if they are similar to the new iPhone ones #asahi-dev · 2023-12-15 08:15
sven*that‘s#asahi-dev · 2023-12-15 08:15
sventhere’s a single flag that stored on flash which disables most customization we could do afaict #asahi-dev · 2023-12-15 08:07
svenit looks like we also can’t easily fix monitors that negotiate 4 lane DP when they could do 2 lane + usb 3 instead :(#asahi-dev · 2023-12-05 05:25
svenmy m1n1 is a few weeks old I think #asahi-dev · 2023-12-05 05:25
svenhuh, weird #asahi-dev · 2023-12-04 08:31
svenjanneg: do you have a usb3+dp hub and can do a atc and tipd trace for me? my trace looks a bit strange since it does 4 mode transitions (usb3 -> usb3+dp -> usb3 -> usb3+dp) and i want to confirm this is just this hub being dumb#asahi-dev · 2023-12-02 12:21
svennice :)#asahi-dev · 2023-12-02 12:08
svener.. ignore that &0xf for the second one ofc ;)#asahi-dev · 2023-12-02 12:07
svenyup#asahi-dev · 2023-12-02 12:07
sven>>> ((1<<4)-1)&0xf#asahi-dev · 2023-12-02 12:07
sven>>> (0x442928>>20)&0xf#asahi-dev · 2023-12-02 11:48
svenI think it might get that n from 0x503000a20 (whatever that is)#asahi-dev · 2023-12-02 11:34
svencan you paste the trace where it's 0xf?#asahi-dev · 2023-12-02 11:32
svenok, so LN_TXA_CAL_CTRL is just (1<< LN_TX_CAL_CODE)-1 then