#asahi-dev · 2023-01-22 14:02
jannaumarcan: do you still want to "stabilize" asahi-wip soon? I would like to get the component based probing and runtime PM merged before 13.2 support#asahi-dev · 2023-01-22 13:59
jannaumarcan: please replace bits/020-t6000-dart with https://github.com/jannau/linux/tree/bits/020-t6000-dart - update to Thierry's v13 which hopefully gets merged for v6.3#asahi-dev · 2023-01-21 17:01
jannauyes, I was asking if it's possible to convince a on-screen keyboard to use the digitizer#asahi-dev · 2023-01-21 16:56
jannauis on-screen keyboard software configurable enough to "display" a single line of function keys?#asahi-dev · 2023-01-21 13:40
jannauif I connect the adapter with dp cable connected there is only a single dp_sid update. It doesn't generate HPD events since dp_altmode_status_update() consumes that for altmode configuration#asahi-dev · 2023-01-21 13:36
jannauit works somewhat with the exception of dwc3 role switch handling and the missed hpd events#asahi-dev · 2023-01-21 13:30
jannaumacos checks it after it gets AttentionReceived irqs#asahi-dev · 2023-01-21 13:25
jannauare you still looking for HPD irqs? apple seems to use DATA STATUS (0x50) bit 14 / 15 even for dp altmode. annotated as TBT HPD in the tps 65987ddh trm#asahi-dev · 2023-01-20 18:32
jannauI didn't noticed until now (or forgot) that hpmBusManager is an i2c device too. macos reads something after interrupts, maybe the interrupt source?#asahi-dev · 2023-01-20 12:28
jannauyeah! so 80ms for the first meter and 60ms for the remaining kilometers#asahi-dev · 2023-01-20 11:10
jannauI don't remember if the port links have an influence on probe order but if they do it might be stable now#asahi-dev · 2023-01-20 11:08
jannauI'm currently using my dcp tree with component based probing, i.e. without deferred probing so it should be up earlier#asahi-dev · 2023-01-20 11:00
jannauno, the problem I see is that dp_altmode_status_update() is only called once. it set dp altmode up then and never genereates a hotplug event#asahi-dev · 2023-01-20 10:47
jannausven: dp-altmode works even when connected on boot, only the hotplug event is missing in that case. altmodes/displayport.c doesn't issue that if HPD is set during configuration#asahi-dev · 2023-01-20 10:43
jannauprobably to late for t8110 device trees#asahi-dev · 2023-01-20 06:15
jannauamarioguy: pmc 0 and 1 are hardcoded to cycles and instructions#asahi-dev · 2023-01-19 22:37
jannaumarcan: please move a0139ccbde88c52 "fixup! iommu: Parse translated reserved regions" from bits/200-dcp to bits/020-t6000-dart, see kernel test robot complaints#asahi-dev · 2023-01-19 21:14
jannauhttps://oftc.irclog.whitequark.org/asahi-dev/2022-12-04#31685475; and followup on the next day#asahi-dev · 2023-01-19 21:10
jannauI think we spoke about that already last year#asahi-dev · 2023-01-19 21:10
jannauthere are aparently some state tracking issues in there. I always get "Firmware doesn't support alternate mode overriding"#asahi-dev · 2023-01-19 21:08
jannauah, I see it now in tps6598x_displayport_enter()#asahi-dev · 2023-01-19 21:07
jannauyes, I have "usb: typec: tipd: Add DisplayPort alternate mode support"#asahi-dev · 2023-01-19 20:52
jannauand nothing changes that during dp-altmode switching#asahi-dev · 2023-01-19 20:50
jannauI guess it's happening since tps6598x_connect() always sets TYPEC_STATE_USB#asahi-dev · 2023-01-19 18:52
jannauin atcphy_mux_set_work() is still accurate with the current dwc3 changes. I'm hitting the WARN_ON(atcphy->dwc3_online) everytime#asahi-dev · 2023-01-19 18:50
jannaunot sure if the top comment of #asahi-dev · 2023-01-19 18:48
jannauconsole only showing after vt switch is fixed within dcp#asahi-dev · 2023-01-19 18:47
jannauit was yankier than I rememebered. I think I had it at one point working relieable with first connecting the usb-c-dp adapter and then connect the displayport cable#asahi-dev · 2023-01-19 17:03
jannauold branch "works" but was yankier than I remembered#asahi-dev · 2023-01-19 16:46
jannauhmm, no it was working back then despite this issue#asahi-dev · 2023-01-19 16:36
jannauI simply don't remember my last working branch#asahi-dev · 2023-01-19 16:20
jannauno, that branch has the same problem#asahi-dev · 2023-01-19 16:15
jannautesting now my hopefully last working branch to be sure diffing is useful#asahi-dev · 2023-01-19 16:12
jannauthe problem doesn't seem dcp related though but somewhere in dwc3/tipd/typec/atcphy#asahi-dev · 2023-01-19 16:11
jannaunot in my current work branch based on asahi-6.2-rc3-5 with runtime pm for dcp#asahi-dev · 2023-01-19 16:04
jannaudwc3 doesn't powers off the phy on dp connect#asahi-dev · 2023-01-19 16:03
jannauno obvious differences in the relevant parts to the branch which was working (I think, would make sense to verify)#asahi-dev · 2023-01-19 16:02
jannautoo much time since I last looked at dp-altmode. problem that looks familiar but I can't remember details#asahi-dev · 2023-01-19 13:12
jannauis there anything left to do for usb4 or is display altmode / tunneling the only thing left after tb PCIe?#asahi-dev · 2023-01-19 10:53
jannaunot a fan of the larger boot picker though but I guess 4 or more OS installs on a system are rare#asahi-dev · 2023-01-19 10:52
jannauloading the recovery os ramdisk for options seems to faster as well#asahi-dev · 2023-01-19 10:51
jannauhuh, 13.2 beta2 to rc update surprisingly was fast, downloaded just 2.5GB, but no endless processing after the restart#asahi-dev · 2023-01-19 07:37
jannaumaybe we need to use a different naming scheme for tags based on asahi-wip#asahi-dev · 2023-01-19 07:35
jannauit was only in asahi-dev, the normal asahi repo is still at asahi-6.1-2#asahi-dev · 2023-01-18 23:46
jannaukettenis_: 10gb is the known aqc113. no identity information for the 1gb one, probably the same as in the m1 mac mini and imac#asahi-dev · 2023-01-18 19:28
jannauaic-iack-offset is not in the template adt#asahi-dev · 2023-01-18 19:25
jannauwlan claims to be bcm4387 compatible#asahi-dev · 2023-01-18 19:22
jannauboth mac minis have ASM3142 as pcie usb controller#asahi-dev · 2023-01-18 19:07
jannauand the dp/hdmi converter is connected to the switch#asahi-dev · 2023-01-18 19:04
jannaum2 mac mini has 'dp-switch-ufp-endpoint = 0' for dcp