Alright, I have a little more detail, but it's still largely a puzzle. My current machine state (expected to remain this way until the next reboot), is that any time I plug a USB device into a port under the xHCI controller in the TB18DC dock, the xHCI controller dies a horrible death.
This is 100% reproducible right now. Unbinding the xHCI driver and rebinding works, as does something firmer to the entire thunderbolt PCI chain and rescanning. Doing so results in every USB device plugged in at the time of the rebind/rescan working perfectly, until such time as I next plug a USB device in. That device can be on a port on the dock, or on a port on any hub connected to a port on the dock. I can remove the thunderbolt PCI devices software wise, unplug the dock from the laptop, unplug the dock from power, then reconnect the dock to power, plug it back into the laptop, and once everything comes back up I am again in a temporarily working state... Until I plug a USB device into the system. At which point it fails again, in exactly the same way. Given that a reboot will solve the problem for some time, this is starting to make me think that either A, there is something very wonky going on in Dell SMBIOS land, or B, there is something wonky going on somewhere in the kernel space. It being an issue with the dock itself seems difficult to support, given that the dock is getting a full power cycle and it's not changing anything, and yet a reboot of the laptop without a full power cycle of the dock seems to resolve the issue for a little while. Here are some more kernel messages of the death, this time with more debugging enabled: Feb 08 20:03:46 chimera kernel: xhci_hcd 0000:0c:00.0: // Ding dong! Feb 08 20:03:46 chimera kernel: xhci_hcd 0000:0c:00.0: Slot 3 output ctx = 0xaa5a13000 (dma) Feb 08 20:03:46 chimera kernel: xhci_hcd 0000:0c:00.0: Slot 3 input ctx = 0xacd98c000 (dma) Feb 08 20:03:46 chimera kernel: xhci_hcd 0000:0c:00.0: Set slot id 3 dcbaa entry 000000006f1f9598 to 0xaa5a13000 Feb 08 20:03:46 chimera kernel: usb 3-1.4: new full-speed USB device number 7 using xhci_hcd Feb 08 20:03:46 chimera kernel: xhci_hcd 0000:0c:00.0: Set root hub portnum to 3 Feb 08 20:03:46 chimera kernel: xhci_hcd 0000:0c:00.0: Set fake root hub portnum to 1 Feb 08 20:03:46 chimera kernel: xhci_hcd 0000:0c:00.0: udev->tt = 000000008fd0cd08 Feb 08 20:03:46 chimera kernel: xhci_hcd 0000:0c:00.0: udev->ttport = 0x4 Feb 08 20:03:46 chimera kernel: xhci_hcd 0000:0c:00.0: // Ding dong! Feb 08 20:03:46 chimera kernel: xhci_hcd 0000:0c:00.0: Successful setup context command Feb 08 20:03:46 chimera kernel: xhci_hcd 0000:0c:00.0: Op regs DCBAA ptr = 0x00000acbcc6000 Feb 08 20:03:46 chimera kernel: xhci_hcd 0000:0c:00.0: Slot ID 3 dcbaa entry @000000006f1f9598 = 0x00000aa5a13000 Feb 08 20:03:46 chimera kernel: xhci_hcd 0000:0c:00.0: Output Context DMA address = 0xaa5a13000 Feb 08 20:03:46 chimera kernel: xhci_hcd 0000:0c:00.0: Internal device address = 0 Feb 08 20:03:46 chimera kernel: xhci_hcd 0000:0c:00.0: Waiting for status stage event Feb 08 20:03:47 chimera kernel: xhci_hcd 0000:0c:00.0: Cancel URB 00000000b99abcb2, dev 1, ep 0x0, starting at offset 0xaa6cbddc0 Feb 08 20:03:47 chimera kernel: xhci_hcd 0000:0c:00.0: // Ding dong! Feb 08 20:03:52 chimera kernel: xhci_hcd 0000:0c:00.0: xHCI host not responding to stop endpoint command. Feb 08 20:03:52 chimera kernel: xhci_hcd 0000:0c:00.0: // Halt the HC Feb 08 20:03:52 chimera kernel: xhci_hcd 0000:0c:00.0: Host halt failed, -110 Feb 08 20:03:52 chimera kernel: xhci_hcd 0000:0c:00.0: xHCI host controller not responding, assume dead Feb 08 20:03:52 chimera kernel: xhci_hcd 0000:0c:00.0: Killing URBs for slot ID 1, ep index 0 Feb 08 20:03:52 chimera kernel: xhci_hcd 0000:0c:00.0: Killing URBs for slot ID 1, ep index 2 Feb 08 20:03:52 chimera kernel: xhci_hcd 0000:0c:00.0: Killing URBs for slot ID 2, ep index 0 Feb 08 20:03:52 chimera kernel: xhci_hcd 0000:0c:00.0: Killing URBs for slot ID 2, ep index 2 Feb 08 20:03:52 chimera kernel: xhci_hcd 0000:0c:00.0: Killing URBs for slot ID 3, ep index 0 Feb 08 20:03:52 chimera kernel: xhci_hcd 0000:0c:00.0: Killing URBs for slot ID 4, ep index 0 Feb 08 20:03:52 chimera kernel: xhci_hcd 0000:0c:00.0: Killing URBs for slot ID 4, ep index 2 Feb 08 20:03:52 chimera kernel: xhci_hcd 0000:0c:00.0: Killing URBs for slot ID 4, ep index 3 Feb 08 20:03:52 chimera kernel: r8152 4-1.2:1.0 eth0: Tx status -108 Feb 08 20:03:52 chimera kernel: r8152 4-1.2:1.0 eth0: Tx status -108 Feb 08 20:03:52 chimera kernel: r8152 4-1.2:1.0 eth0: Tx status -108 Feb 08 20:03:52 chimera kernel: r8152 4-1.2:1.0 eth0: Tx status -108 Feb 08 20:03:52 chimera kernel: xhci_hcd 0000:0c:00.0: Killing URBs for slot ID 4, ep index 6 Feb 08 20:03:52 chimera kernel: xhci_hcd 0000:0c:00.0: Killing URBs for slot ID 5, ep index 0 Feb 08 20:03:52 chimera kernel: xhci_hcd 0000:0c:00.0: Killing URBs for slot ID 5, ep index 2 Feb 08 20:03:52 chimera kernel: xhci_hcd 0000:0c:00.0: Killing URBs for slot ID 5, ep index 10 Feb 08 20:03:52 chimera kernel: xhci_hcd 0000:0c:00.0: Killing URBs for slot ID 5, ep index 15 Feb 08 20:03:52 chimera kernel: xhci_hcd 0000:0c:00.0: Killing URBs for slot ID 6, ep index 0 Feb 08 20:03:52 chimera kernel: xhci_hcd 0000:0c:00.0: Killing URBs for slot ID 6, ep index 2 Feb 08 20:03:52 chimera kernel: xhci_hcd 0000:0c:00.0: Killing URBs for slot ID 7, ep index 0 Feb 08 20:03:52 chimera kernel: xhci_hcd 0000:0c:00.0: Killing URBs for slot ID 7, ep index 5 Feb 08 20:03:52 chimera kernel: xhci_hcd 0000:0c:00.0: Killing URBs for slot ID 7, ep index 8 Feb 08 20:03:52 chimera kernel: xhci_hcd 0000:0c:00.0: HC died; cleaning up Feb 08 20:03:52 chimera kernel: xhci_hcd 0000:0c:00.0: xHCI host controller is dead. Of note, sometimes when I kick this off I get CPU soft lockup watchdog notices, I need to do some more digging, but this _seems_ to indicate either a lock being held deep inside some stuff, or that this really is a SMBIOS issue where the SMBIOS is going out to lunch for a little bit. If this does prove to be a SMM/SMBIOS issue, then there will likely be nothing at all that we can do to fix it except for bugging Dell. -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1766076 Title: USB over thunderbolt turns off every once in a while To manage notifications about this bug go to: https://bugs.launchpad.net/dell-sputnik/+bug/1766076/+subscriptions -- ubuntu-bugs mailing list [email protected] https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
