#asahi-dev · 2022-11-01 19:51
sveni'll probably have to disassemble xnu and find its iodelays :(#asahi-dev · 2022-11-01 19:51
svenwell, every single test requires me too boot ~20 times to be sure it works#asahi-dev · 2022-11-01 19:49
sventhe mmio reads/writes look exactly the same as XNU's. it's a delay somewhere that I can't easily see in the mmio trace#asahi-dev · 2022-11-01 19:49
sventhat too#asahi-dev · 2022-11-01 19:48
svenwell, unless I add a usleep(10); after every single register poke :p#asahi-dev · 2022-11-01 19:48
svenand it's not reliable yet#asahi-dev · 2022-11-01 19:47
svenit's only supported in my local tree for now#asahi-dev · 2022-11-01 19:46
svendisplay port via type-c ports#asahi-dev · 2022-11-01 19:15
svenbah.. I just can't find the last udelay that's missing to make altmode work every single time :/#asahi-dev · 2022-11-01 16:05
svenhuh, that sounds broken. I thought there’s a reset on timeout #asahi-dev · 2022-11-01 16:00
svenwhich is also very annoying to test :(#asahi-dev · 2022-11-01 16:00
svenI’m hunting down yet another udelay that’s missing somewhere i atcphy.c. every now and then altmode doesn’t come up #asahi-dev · 2022-11-01 15:59
svenlooks like everything’s broken today!#asahi-dev · 2022-11-01 15:52
svenamarioguy can just fix it in v4 I guess #asahi-dev · 2022-11-01 15:51
svenirq isn’t even upstream yet #asahi-dev · 2022-11-01 15:51
svenoh, but then there’s no follow up check if XEN was actually set#asahi-dev · 2022-11-01 15:48
sventhe irq path rereads status, doesn’t it?#asahi-dev · 2022-11-01 15:42
svencause it worked so far#asahi-dev · 2022-11-01 15:41
sventhose boards had like two chips on the i2c bus#asahi-dev · 2022-11-01 15:41
svenin the pasemi thing? #asahi-dev · 2022-11-01 15:24
svenguess there was never a follow up#asahi-dev · 2022-11-01 15:24
svenhttps://lore.kernel.org/lkml/20220409130013.1474412-1-trix@redhat.com/#asahi-dev · 2022-11-01 15:23
sveni guess that one never got merged then :/#asahi-dev · 2022-11-01 15:23
svenmarcan: huh, even on the latest rc? I vaguely remember a patch that fixed that issue#asahi-dev · 2022-11-01 15:23
svenjannau: not totally sure yet, but afaict mst creates a new connector on the fly #asahi-dev · 2022-11-01 15:04
svenand wouldn't DP MST use two connectors anyway?#asahi-dev · 2022-11-01 15:02
sveni.e. there's no OOB event required#asahi-dev · 2022-11-01 15:02
svenDP MST over thunderbolt/usb-c on intel machine happens by magic afaict#asahi-dev · 2022-11-01 15:02
svennot really :(#asahi-dev · 2022-11-01 15:00
sven(drm_connector_oob_hotplug_event)#asahi-dev · 2022-11-01 14:59
svenhrm, what happens to the oob hotplug notification if you have a single type c port connected to multiple drm connectors?#asahi-dev · 2022-11-01 14:58
sven(and I suspect that dpin0/1 will become another phy)#asahi-dev · 2022-11-01 14:58
svenso in my head I still have a drm connector connected to a single phy and a crossbar mux. in that case we'd need two drm connectors for the dcpextX,0/dcpextX,1 case since both muxes have to be configured#asahi-dev · 2022-11-01 14:54
svenhrm, or maybe fake the tiled displays into a single connector#asahi-dev · 2022-11-01 14:54
svenwe'll have to be able to describe tiled displays which will use dpin0/1 at the same time#asahi-dev · 2022-11-01 14:53
svenI think we'll also need three connectors (dpphy, dpin0/1) for each type-c port#asahi-dev · 2022-11-01 14:52
svenbut maybe having it as a dcpext subnode makes more sense#asahi-dev · 2022-11-01 14:52
svenbecause that thing has to decide the muxing#asahi-dev · 2022-11-01 14:52
svenoh, i was thinking about having the connector subnode on the display-subsystem#asahi-dev · 2022-11-01 14:51
sveneach type-c port also has its own display crossbar mux but we can just ignore that for now#asahi-dev · 2022-11-01 14:51
svenso each type-c port has its own atcphy (and probably will get another phy for dpin0/1)#asahi-dev · 2022-11-01 14:50
svenhmm?#asahi-dev · 2022-11-01 14:49
svenit’ll just get another graph connection to the drm connector #asahi-dev · 2022-11-01 14:48
svenusb will remain more or less like it’s in that old atc wip branch #asahi-dev · 2022-11-01 14:48
svendrm#asahi-dev · 2022-11-01 14:38
svenalso, can I volunteer you to make this work on t600x with more than one dcpext once I get it in a reasonable shape on t8013? :>#asahi-dev · 2022-11-01 14:36
sventhe first part is probably easy if the connectors are just sub nodes. we could just give them a phys=<atcphyX> property #asahi-dev · 2022-11-01 14:35
svenand a way to encode that connectors on die 0 can only be routed to dcpext on die 0#asahi-dev · 2022-11-01 14:34
svenjannau: we’ll also need some way to encode which connector maps to which atcphy