#asahi-dev · 2022-11-17 10:08
svenbut none of that code is triggered right now with usb3-only fwiw#asahi-dev · 2022-11-17 10:08
svenif you look at the DP function you'll also see quite a few hardcoded values#asahi-dev · 2022-11-17 10:08
svenyeah, it's still on my TODO list ;)#asahi-dev · 2022-11-17 10:07
svenso it's not t6k vs t8103 but actually something that has to be done on both#asahi-dev · 2022-11-17 10:07
sveni'm still mostly sure it's just some if (some_magic_register & some_magic_bit) do_what_looks_like_t6k_stuff#asahi-dev · 2022-11-17 10:06
svenyeah#asahi-dev · 2022-11-17 10:06
svenI don't think the code for t6k is actually different, iirc it doesn't overload the dp functions#asahi-dev · 2022-11-17 10:06
sventhe hardcoded values are DP only, yes#asahi-dev · 2022-11-17 10:05
sveni'm pretty sure there are no differences for USB3#asahi-dev · 2022-11-17 10:05
sven(for DP, not for USB3)#asahi-dev · 2022-11-17 10:05
svenI still have some hardcoded values for some registers that actually come from $somewhere#asahi-dev · 2022-11-17 10:05
svenand what was different on t6000?#asahi-dev · 2022-11-17 10:04
svenDP still kinda worked with just the t8103 compatible so it's fine to have that as a fallback#asahi-dev · 2022-11-17 10:01
svensounds good to me#asahi-dev · 2022-11-16 20:47
svenlike half of acio is up but one MMIO range still SErrors *sigh*#asahi-dev · 2022-11-16 20:47
svenLooks like thunderbolt is even more cursed that usb3/dp #asahi-dev · 2022-11-16 18:09
svenok, let's use "apple,tXXXX-atcphy", "apple,t8103-atcphy" then for all currently released SoCs#asahi-dev · 2022-11-16 11:10
sventhe reset bit is in PIPEHANDLER_AON_GEN fwiw#asahi-dev · 2022-11-16 11:02
sventhe hard reset bit for dwc3 is in the MMIO range that also controls which phy (usb3 vs. usb4 tunneling) is hooked up to dwc3#asahi-dev · 2022-11-16 09:38
svenand there's also still thunderbolt but I hope that won't require any more changes to the atcphy node#asahi-dev · 2022-11-16 09:37
svenhttps://github.com/AsahiLinux/linux/blob/asahi-wip/arch/arm64/boot/dts/apple/t8103.dtsi#asahi-dev · 2022-11-16 09:37
svennote that i'm not sure about the atcphy compatible yet, but it'll probably be either "apple,tXXXX-atcphy", "apple,atcphy" or "apple,tXXXX-atcphy", "apple,t8103-atcphy" since it was introduced with t8103#asahi-dev · 2022-11-16 08:34
svenoh.. because of the phy reference in the dt? :/#asahi-dev · 2022-11-15 20:21
svenstill happy that the pipehandler sequence seems to be stable enough to recover even if that lock fails now#asahi-dev · 2022-11-15 20:20
svenguess i'll try hotplugging some usb3 device >50 times tomorrow and see if I can provoke that -110 error#asahi-dev · 2022-11-15 20:19
sventhere's still that weird ACIOPHY_TOP_BIST_READ_CTRL_LN0_PHY_STATUS_RE behavior I see on t8103#asahi-dev · 2022-11-15 20:17
svennah, it just prints the error from above again#asahi-dev · 2022-11-15 20:17
svenoh.. wait#asahi-dev · 2022-11-15 20:17
sventhat really looks like some kind of race#asahi-dev · 2022-11-15 20:17
sven[ 834.513144] xhci-hcd: probe of xhci-hcd.0.auto failed with error -110#asahi-dev · 2022-11-15 20:17
sven[ 834.513132] xhci-hcd xhci-hcd.0.auto: USB bus 1 deregistered#asahi-dev · 2022-11-15 20:13
svenI never saw that on t8103#asahi-dev · 2022-11-15 20:13
svenmaybe t6k actually needs some additional setup to work properly #asahi-dev · 2022-11-15 20:13
svenyeah, just saw that#asahi-dev · 2022-11-15 20:12
svenno worries, I can use grep ;)#asahi-dev · 2022-11-15 20:08
svenstill surprised the same atcphy code works everywhere so far #asahi-dev · 2022-11-15 20:06
svenbecause it claims xhci probe fails just after the unplug #asahi-dev · 2022-11-15 20:06
svenit looks like there might be a race when you unplugged it#asahi-dev · 2022-11-15 20:05
sven“unfortunately” ;)#asahi-dev · 2022-11-15 20:05
svenjannau: ah, whoops, thanks! #asahi-dev · 2022-11-15 20:04
svenand can you reproduce it? if yes it would be interesting to see what’s going on while enabling more debug messages #asahi-dev · 2022-11-15 20:03
svencan you paste the full dmesg?#asahi-dev · 2022-11-15 20:01
svenand I assume you didn’t suspend the mbp or anything?#asahi-dev · 2022-11-15 19:59
svenit looks like xhci timed out waiting for something #asahi-dev · 2022-11-15 19:58
svenis there anything earlier in the dmesg? it looms like xhci already didn’t come up#asahi-dev · 2022-11-15 19:58
svenoh… pipehandler lock failed :/#asahi-dev · 2022-11-15 19:57
svenso it just didn’t show up? Or how did it break?#asahi-dev · 2022-11-15 19:30
svenit’s probably something dumb again #asahi-dev · 2022-11-15 19:29
svenbut I don’t think it does