#
✕ nick:jannau
#asahi-dev · 2022-04-19 14:20
jannaushould loop every 178.957 seconds
#asahi-dev · 2022-04-19 13:18
jannaumarcan: dcp method annotations are good, I haven't spotted any other inconsistencies. thanks
#asahi-dev · 2022-04-19 13:12
jannaualso bit 1 (uncached for t6000/t8110) is not set: table (0): 00000000 ... 02000000 -> 0000010ffff44000 [01]
#asahi-dev · 2022-04-19 13:11
jannauI tried that and it didn't made a difference
#asahi-dev · 2022-04-19 13:09
jannauthe page table itself is initialized by iboot
#asahi-dev · 2022-04-19 13:06
jannau0x10ffff47500 is within the first dart-disp0 page table
#asahi-dev · 2022-04-19 13:04
jannauAMCC0 PLANE0 UNEXP_RT_HIT_DIR error: INTSTS1 0x00000100 AFERRLOG0/1/2/3/4 0xfff47500/0xea10f/0x23a120e/0x80/0x80041 ADDR 0x10ffff47500 CMD/SIZE 0x14(ncrdincr_ro)/0x7 AID/TID 0x120e/0x23a @AppleT6000PlatformErrorHandler.cpp:2712
#asahi-dev · 2022-04-19 13:02
jannaumarcan: thanks, t6k2 looks good. note that macos guest panics due to the initialized display
#asahi-dev · 2022-04-19 11:08
jannauthe 8 should be promoted to unsigned as 'int' and 'u32' should have the same rank
#asahi-dev · 2022-04-19 07:45
jannaukde system settings seems to get confused about the refresh rates though. 4k is limited to 30 Hz due to HDMI 1.4 but it offers 60, 50 and 24 Hz instead of 30, 25 and 24
#asahi-dev · 2022-04-19 07:42
jannauyes, a cursed dell 4k display and a well behaved FullHD display
#asahi-dev · 2022-04-19 07:41
jannaudisplays
#asahi-dev · 2022-04-19 07:37
jannauat least dpms and switching between 2 different seems now to work reliable
#asahi-dev · 2022-04-19 07:17
jannauthe next observed Call A131 doesn't look obviously wrong judging by context
#asahi-dev · 2022-04-19 07:14
jannaunot a particularly important task but it's impossible to tell from traces what happened to unobserved calls/callbacks
#asahi-dev · 2022-04-19 07:10
jannauthe next two calls set_config_frame_size and program_config_frame_size are consistent with one removed call
#asahi-dev · 2022-04-19 07:08
jannauCALL > 0x0000002000400002 CMD 0x40 A103 0xc 0x8 00000000000000000000aaaa
#asahi-dev · 2022-04-19 07:07
jannausignature still matches but looks more like it should be 'A103 = Call(void, "get_config_frame_size", width=InOutPtr(uint), height=InOutPtr(uint))'
#asahi-dev · 2022-04-19 07:05
jannaumarcan: how much effort is it to recreate KNOWN_MSGS in trace_dcp.py? I think '"A103": "IOMFB::UPPipe2::test_control(IOMFB_TC_Cmd, unsigned int)"' doesn't exists anymore
#asahi-dev · 2022-04-18 16:39
jannaumillenialhacker: parse instance as ascii and look for DART
#asahi-dev · 2022-04-18 15:17
jannauonly hint in the ADT is 'asc-dram-mask' so it might be asc or something more or less compatible
#asahi-dev · 2022-04-18 15:08
jannaumillenialhacker: ascwrap or m3wrap in the compatible in the adt
#asahi-dev · 2022-04-18 11:56
jannauwhy dcp? why? swaps are not cancelled/errored on connector unplug but are completed on replug after setting the mode
#asahi-dev · 2022-04-17 16:49
jannaumillenialhacker: I would have hoped mac os would complain beforehand. iirc it worked with 75GB, could have been macos 11 though
#asahi-dev · 2022-04-17 16:48
jannaus/bale/blame/
#asahi-dev · 2022-04-17 16:47
jannauI mostly bale the display for this
#asahi-dev · 2022-04-17 16:47
jannauthis is with the nice display. I have a cursed Dell 4k display which disconnects immediately after setting the mode (with a hotplug disconnect event)
#asahi-dev · 2022-04-17 16:44
jannauthis is on the mac studio so I assume iboot just loads the firmware
#asahi-dev · 2022-04-17 16:43
jannaudcp doesn't even seem consistant. On the initial poweron I don't get a hotplug event but I get on after turning it off and on
#asahi-dev · 2022-04-17 16:41
jannaudpms is now somewhat working
#asahi-dev · 2022-04-17 16:40
jannaucursed dcp interface is cursed
#asahi-dev · 2022-04-15 14:16
jannaumillenialhacker: no need for a pr, you can just edit it but I think you can't push to the wiki repository
#asahi-dev · 2022-04-15 08:23
jannauyes, that are different endpoints in DCP's rtkit interface
#asahi-dev · 2022-04-15 07:55
jannauif it is using HDMI audio over the dp to hdmi converter
#asahi-dev · 2022-04-15 07:54
jannauI see nothing in the annotated Calls for the dcp endpoint so it's probably in a different endpoint. "av" maybe or "mcdp29xx"
#asahi-dev · 2022-04-15 07:44
jannauI haven't seen any audio related things in DCP yet but I haven't looked
#asahi-dev · 2022-04-15 07:43
jannauyes, SPI works fine without SIO and the large transfers for touch id should not be a problem either
#asahi-dev · 2022-04-15 07:42
jannauI think displayport only carries linear pcm and no encoded audio
#asahi-dev · 2022-04-15 07:36
jannauand s/hdmi audiio/dp audio/ I assume
#asahi-dev · 2022-04-15 07:35
jannautouch id seems to have larger SPI transfers, see #asahi-re
#asahi-dev · 2022-04-15 07:35
jannauapple uses it as "DMA" replacement for "large" (256 byte) SPI transfers for keyboard/touchpad
#asahi-dev · 2022-04-15 05:52
jannauwoot, a gdbserver for m1n1 pull request
#asahi-dev · 2022-04-15 05:49
jannau'smartio_debug=0xff' as boot args is helpful, should report the EP in decimal
#asahi-dev · 2022-04-15 05:47
jannaunot sure the iomem handling of the sio tracer is complete, I did what was necessary to get the keyboard tracer to work
#asahi-dev · 2022-04-15 05:46
jannaucomment the iomon.poll() and an iomon more specific for mesa
#asahi-dev · 2022-04-15 05:41
jannautouch id is be 0x18 / 0x19
#asahi-dev · 2022-04-15 05:40
jannauchadmed_: see how the keyboard is handed. msg.EP refers to the dma-channels property in the ADT
#asahi-dev · 2022-04-14 17:33
jannaudcp and dcpextN can not share the same mappings since they have both their own heap starting at iova 0
#asahi-dev · 2022-04-14 17:31
jannauthanks, I think the lazy approach will break or become annoying once we support dispextN
#asahi-dev · 2022-04-14 17:11
jannausven: is there a specific reason why dcp needs to use the raw iommu api instead of DMA?

plus ›