#asahi-dev · 2022-11-04 19:35
sventhen there's atcphy which controls the actual physical layer#asahi-dev · 2022-11-04 19:35
svendcpext controls parts of it. mainly the actual display stream and all the link training stuff#asahi-dev · 2022-11-04 19:35
svenit's complicated ;)#asahi-dev · 2022-11-04 19:34
svennot really#asahi-dev · 2022-11-04 19:33
svenbut that's entirely unrelated to the displayport HPD#asahi-dev · 2022-11-04 19:33
svenxhci has a bunch of rings #asahi-dev · 2022-11-04 19:32
svenuh... define usb3 controller#asahi-dev · 2022-11-04 19:17
svenno idea how it gets the HPD updates #asahi-dev · 2022-11-04 19:17
svenweird, I don’t see anything special when tracing macOS :/#asahi-dev · 2022-11-04 16:13
sven*on the typec #asahi-dev · 2022-11-04 16:13
svenit switches pretty reliably between usb3/DP-only/usb3+DP now#asahi-dev · 2022-11-04 16:13
svenbut at least I think I found the missing delay that broke insertion/removal in the type c side #asahi-dev · 2022-11-04 16:10
svenThere’s just no interrupt when that changes happens #asahi-dev · 2022-11-04 16:10
svenapparently there’s some PD messages that is exchanged which changes this “DP SID Status”#asahi-dev · 2022-11-04 16:09
sventps6598x notices, it just doesn’t allow me to trigger an interrupt :(#asahi-dev · 2022-11-04 16:09
svenyeah, I think I accidentally trigger two hotplug events when shutting down but only if tipd shuts down before DCP #asahi-dev · 2022-11-04 15:59
svencould be#asahi-dev · 2022-11-04 15:59
svenoh, you meant for the crash?#asahi-dev · 2022-11-04 15:56
sven*another#asahi-dev · 2022-11-04 15:56
sventhere's probably yet enother out of band mechanism#asahi-dev · 2022-11-04 15:56
svenI've enabled all IRQs in the tps chip and nothing triggered when I unplugged the HDMI cable#asahi-dev · 2022-11-04 15:51
svenjannau: https://f.svpe.de/32f2cafe3d6a663ee824841f8bdd4ee9f1a2dafdc234abd2ffceccaf215cedbf_dcpext-panic.txt no idea how I hit that once. looks like you can also look forward to dcpext fun though!#asahi-dev · 2022-11-04 15:43
sventhe hpd in "Data Status" is incorrect and can get stuck but there is an interrupt triggered when it changes :(#asahi-dev · 2022-11-04 15:43
sven*sigh*, the HPD in "DP SID Status" is correct but no interrupt is triggered when that changes#asahi-dev · 2022-11-04 15:32
svenmacos somehow detects that disconnect though#asahi-dev · 2022-11-04 15:32
svenhrm, looks like macos has another way that's not the tps6598x chip to figure out HPD. my usb3+dp hub doesn't trigger any irq in the tps chip when I disconnect the hdmi cable#asahi-dev · 2022-11-03 18:08
svenbut if I sleep after each reg write in atcphy it works again#asahi-dev · 2022-11-03 18:08
svenno idea, it’s just that removing/inserting the type c side breaks everything right now #asahi-dev · 2022-11-03 17:47
svenurgh.. looks like there's still a delay missing *somewhere* :(#asahi-dev · 2022-11-03 17:40
svenuh oh#asahi-dev · 2022-11-03 17:40
svenapple-dcp 271c00000.dcp: RTKit: syslog message: DPTXController.cpp:2788: AppleDCPDPTXController::handleInterrupt(): FIFO error interrupt COMMON_INT_STA_3=0x00000010#asahi-dev · 2022-11-03 14:35
sven(except the thunderbolt dock ofc :-()#asahi-dev · 2022-11-03 14:35
sventhis still works surprisingly well though with all of the displays and adapters i have#asahi-dev · 2022-11-03 14:24
svenand dcpext gets a bit confused if I just skip that#asahi-dev · 2022-11-03 14:24
svenmy "hpd no longer asserted" code just doesn't work correctly#asahi-dev · 2022-11-03 14:24
svenit's one of those dptx epic calls which essentially trigger HPD#asahi-dev · 2022-11-03 14:23
svenI already do that#asahi-dev · 2022-11-03 14:15
svenlol, if i just remove the dcp disconnect handling dcp gets very confused when I remove/insert the type-c plug and just connect again, tries to set the link rate 2-3 times but finally works again :D#asahi-dev · 2022-11-03 14:02
svenhah, so inserting/removing the HDMI side of the type-c->hdmi adapter works. for some reason inserting/removing the type-c side breaks something though and dcpext ends up complaining about "device busy timeout"#asahi-dev · 2022-11-03 13:05
svenhttps://lore.kernel.org/lkml/cover.1667473008.git.viresh.kumar@linaro.org/#asahi-dev · 2022-11-03 12:59
svenbut well... mailbox *sigh*#asahi-dev · 2022-11-03 12:59
svenyeah, writing something for x and then realizing it doesn't work for y happens all the time#asahi-dev · 2022-11-03 12:57
sveneverything else i've seen usually is just "design decisions taken years/decades ago that aren't valid anymore and haven't been cleaned up yet" #asahi-dev · 2022-11-03 12:57
sventhe only truly bad thing i've seen is mailbox *sigh*#asahi-dev · 2022-11-03 11:22
svenhotplug works exactly once, probably because I don’t know how to shut the previous connection down yet :D#asahi-dev · 2022-11-03 11:21
svenokay, oob hpd forwarding to drm works when I represent the connection using the of graph bindings #asahi-dev · 2022-11-02 19:11
svenyup, 3+DP works #asahi-dev · 2022-11-02 19:07
svenI hardcoded the mode to 3+DP before I implemented the typec stuff for the tps6598x thing today and at least the DP part worked #asahi-dev · 2022-11-02 19:06
svenand I’m pretty sure that will just work