#
✕ nick:sven
#asahi-dev · 2022-11-17 22:07
sventhats looks like it’s doing what it’s supposed to do
#asahi-dev · 2022-11-17 22:06
svenhuh
#asahi-dev · 2022-11-17 21:56
svenusually it should be xhci shutting down
#asahi-dev · 2022-11-17 21:56
svenanything before that warn?
#asahi-dev · 2022-11-17 21:56
sventhat’s weird. that shouldn’t be any different than disconnecting any other device
#asahi-dev · 2022-11-17 21:50
sven*on
#asahi-dev · 2022-11-17 21:50
svenuh, so this was in the host machine?
#asahi-dev · 2022-11-17 21:41
svens/now/not/
#asahi-dev · 2022-11-17 21:40
svenwas this a disconnect when dwc3 was in device mode?
#asahi-dev · 2022-11-17 21:38
sventhat warn_on essentially means dwc3 did now shut down
#asahi-dev · 2022-11-17 18:37
svenit probably even saves me a second or so compared to running vdmtool this way :D
#asahi-dev · 2022-11-17 18:36
svenit really doesn’t like that
#asahi-dev · 2022-11-17 18:36
svenapparently because I messed up and brought down atcphy while acio was still running
#asahi-dev · 2022-11-17 18:36
svennah, cio is nice enough to do the reboot for me
#asahi-dev · 2022-11-17 18:30
svenugh… guess I’m back to rebooting the soc when trying to bring cio up *sigh*
#asahi-dev · 2022-11-17 15:21
svenbut bleeding edge with a disclaimer what happens to usb on suspend sounds fine with me tbh
#asahi-dev · 2022-11-17 15:20
sveni wouldn’t want to ship this to users who expect this to work
#asahi-dev · 2022-11-17 15:19
svensure, but it’s called -edge for a reason :p
#asahi-dev · 2022-11-17 15:16
svenyeah, it doesn’t really break anything. people who use -edge may just have to reconnect usb devices once it comes out of suspend
#asahi-dev · 2022-11-17 15:03
sven:D
#asahi-dev · 2022-11-17 15:00
sveni guess we really need to take a look at what macos does to do this properly :/
#asahi-dev · 2022-11-17 15:00
svenmaybe the resume needs some additional phy interaction?
#asahi-dev · 2022-11-17 14:59
svenugh
#asahi-dev · 2022-11-17 14:36
sventhat doesn't sound like fun
#asahi-dev · 2022-11-17 14:36
sventrue. given how broken hotplug is on these machines anyway that also wouldn't surprise me
#asahi-dev · 2022-11-17 14:28
svenyeah, that's what I expected. dunno why dwc3 sees a disconnect then :/
#asahi-dev · 2022-11-17 14:26
svenmaybe also try trace_atc.py and see what happens on suspend
#asahi-dev · 2022-11-17 14:22
svenand if dwc3 survives it'll also leave atcphy alone
#asahi-dev · 2022-11-17 14:22
svenunless tipd calls that tps6598x_disconnect it'll leave atcphy alone
#asahi-dev · 2022-11-17 14:21
svenhrm, tipd shouldn't issue any mux_set so atcphy should survive as well I think
#asahi-dev · 2022-11-17 14:20
svenso dwc3 survives?
#asahi-dev · 2022-11-17 14:20
svenhrm, probably
#asahi-dev · 2022-11-17 13:31
sven(after case DWC3_GCTL_PRTCAP_OTG:)
#asahi-dev · 2022-11-17 13:31
svenreplace if (PMSG_IS_AUTO(msg)) with if(true) and it might just survive
#asahi-dev · 2022-11-17 13:30
svenah, wait, it'll hit the DWC3_GCTL_PRTCAP_OTG path anyway
#asahi-dev · 2022-11-17 13:28
svenif that block is skipped it also won't call into atcphy so it should just survive as well
#asahi-dev · 2022-11-17 13:28
sveni'd try to just always make that if with !device_may_wakeup(dwc->dev) false in the suspend and resume functions
#asahi-dev · 2022-11-17 13:26
svenit really would not surprise me if that "reset everything on suspend" is to workaround more crap like that
#asahi-dev · 2022-11-17 13:25
svenbut that also breaks some intel platform
#asahi-dev · 2022-11-17 13:25
sveni'd expect set_mode followed by xhci init
#asahi-dev · 2022-11-17 13:25
sventhere are also two separate ways it orders xhci_init and phy_set_mode
#asahi-dev · 2022-11-17 13:24
sventhere's e.g. some intel platform that breaks if they probe extcon at the wrong time (lol)
#asahi-dev · 2022-11-17 13:24
svenjust look for some of the recent dwc3 threads in linux-usb
#asahi-dev · 2022-11-17 13:23
svenI'm willing to bet that is done to work around a quirk on some random hardware
#asahi-dev · 2022-11-17 13:03
svenso if you manage to skip dwc3_core_exit we might be good
#asahi-dev · 2022-11-17 13:02
svencore_exit also calls into atcphy to shut down
#asahi-dev · 2022-11-17 13:02
svenwho knows
#asahi-dev · 2022-11-17 13:02
svenmaybe just setting device_may_wakeup and/or PMSG_IS_AUTO just skips that
#asahi-dev · 2022-11-17 13:01
sven:>
#asahi-dev · 2022-11-17 13:01
sven<sven> i probably won't get around to fixing suspend/resume but you can probably hack it by issuing tps_disconnect in the tipd suspend handler and tps_connect in the resume handler

plus ›