Bug 218943
| Summary: | No SuperSpeedPlus on AM4/5 Hubs | ||
|---|---|---|---|
| Product: | Drivers | Reporter: | Gerard (gerarderloper) |
| Component: | USB | Assignee: | Default virtual assignee for Drivers/USB (drivers_usb) |
| Status: | NEW --- | ||
| Severity: | normal | CC: | accounts+kernel, gerarderloper, larsm17, mathias.nyman |
| Priority: | P3 | ||
| Hardware: | AMD | ||
| OS: | Linux | ||
| Kernel Version: | 6.8+ | Subsystem: | |
| Regression: | No | Bisected commit-id: | |
| Attachments: |
lsusb verbose list of devices and hubs
inxi system information fresh dmesg output debug bus speed USBView (patch) Win10USBCapture |
||
Created attachment 306433 [details]
inxi system information
(In reply to Jarrard from comment #1) > Created attachment 306433 [details] > inxi system information Device-3: AVerMedia Live Gamer Ultra 2.1 driver: snd-usb-audio,uvcvideo type: USB rev: 3.2 speed: 5 Gb/s lanes: 1 mode: 3.2 gen-1x1 bus-ID: 4-4:4 chip-ID: 07ca:2553 class-ID: 0102 serial: <filter> As can be seen the speed is WRONG at 5 Gb/s. Something is really going wrong here. These USB3.2 Gen 2 ports are 10Gbps as sold by mfr and confirmed under Linux. Something is going wrong when configuring my USB ports. (In reply to Jarrard from comment #2) > (In reply to Jarrard from comment #1) > > Created attachment 306433 [details] > > inxi system information > > Device-3: AVerMedia Live Gamer Ultra 2.1 driver: snd-usb-audio,uvcvideo > type: USB rev: 3.2 speed: 5 Gb/s lanes: 1 mode: 3.2 gen-1x1 bus-ID: 4-4:4 > chip-ID: 07ca:2553 class-ID: 0102 serial: <filter> > > As can be seen the speed is WRONG at 5 Gb/s. Something is really going > wrong here. These USB3.2 Gen 2 ports are 10Gbps as sold by mfr and confirmed > under Linux. Something is going wrong when configuring my USB ports. CONFIRMED UNDER WINDOWS!!!! Please reboot and post `dmesg`. Can you check the xhci roothub port speed of connected devices: mount -t debugfs none /sys/kernel/debug/ cat /sys/kernel/debug/usb/xhci/<address of device>/ports/port*/portsc If host uses default speed IDs then PortSpeed 4 is 5Gbps PortSpeed 5 is 10Gbps (Gen2x1) PortSpeed 6 is 10Gbps (2x5Gbps, Gen 1x2) example: cat /sys/kernel/debug/usb/xhci/0000:00:0d.0/ports/port*/portsc 0x0a0002a0 Powered Not-connected Disabled Link:RxDetect PortSpeed:0 Change: Wake: WCE WOE 0x000002a0 Powered Not-connected Disabled Link:RxDetect PortSpeed:0 Change: Wake: 0x000002a0 Powered Not-connected Disabled Link:RxDetect PortSpeed:0 Change: Wake: 0x00001603 Powered Connected Enabled Link:U0 PortSpeed:5 Change: Wake: 0x000002a0 Powered Not-connected Disabled Link:RxDetect PortSpeed:0 Change: Wake: Created attachment 306440 [details]
fresh dmesg output
(In reply to Artem S. Tashkinov from comment #4) > Please reboot and post `dmesg`. Added Created attachment 306441 [details]
debug bus speed
(In reply to Mathias Nyman from comment #5) > Can you check the xhci roothub port speed of connected devices: added as debug bus speed. None of them go above speed 4. I can also inspect the device in USBView app and it confirms this. Device is working under SuperSpeedPlus for windows. If I boot to Linux (dual boot) it sets to SuperSpeed(NO PLUS) and then reboot (not ac off) to Windows it is stuck in 5Gbps mode. Only a full AC off will reset the device. Linux is forcing the device into 5Gbps mode somehow. USBView Pastebin from Win10 pastebin.com/HDJm5wDg Just to reiterated again. I see this behaviour on a AM4 machine and a AM5 machine! Two completely different computers with identical issue, linux forcing ports to 5Gbps when 10Gbps device is connected. Your device was initially detected and trainer as a 10Gpps SSP device in Linux There were some issues reading descriptors from the device, and giving it an address, so port was power cycled. After this the device was enumerated as a SS 5Gbps device. There are probably some small differences in how Windows and Linux do device enumeration that cause this. [ 363.960026] usb 4-4: new SuperSpeed Plus Gen 2x1 USB device number 4 using xhci_hcd [ 369.344592] usb 4-4: unable to get BOS descriptor or descriptor too short [ 369.370574] usb 4-4: unable to read config index 0 descriptor/start: -71 [ 369.370586] usb 4-4: can't read configurations, error -71 ... [ 370.325161] usb 4-4: Device not responding to setup address. [ 370.537171] usb 4-4: Device not responding to setup address. [ 370.744627] usb 4-4: device not accepting address 5, error -71 [ 370.753259] usb usb4-port4: attempt power cycle [ 371.861790] usb 4-4: new SuperSpeed USB device number 6 using xhci_hcd [ 372.067721] usb 4-4: LPM exit latency is zeroed, disabling LPM. [ 372.157551] usb 4-4: New USB device found, idVendor=07ca, idProduct=2553, bcdDevice= 1.00 [ 372.157560] usb 4-4: New USB device strings: Mfr=6, Product=7, SerialNumber=3 [ 372.157564] usb 4-4: Product: Live Gamer Ultra 2.1 Logs with dynamic debug enabled for usb core and xhci host could possibly show more info <boot> mount -t debugfs none /sys/kernel/debug echo 'module xhci_hcd =p' >/sys/kernel/debug/dynamic_debug/control echo 'module usbcore =p' >/sys/kernel/debug/dynamic_debug/control <connect device> <send dmesg> (In reply to Mathias Nyman from comment #12) > Your device was initially detected and trainer as a 10Gpps SSP device in > Linux > > There were some issues reading descriptors from the device, and giving it an > address, so port was power cycled. > After this the device was enumerated as a SS 5Gbps device. > > There are probably some small differences in how Windows and Linux do device > enumeration that cause this. > > [ 363.960026] usb 4-4: new SuperSpeed Plus Gen 2x1 USB device number 4 > using xhci_hcd > [ 369.344592] usb 4-4: unable to get BOS descriptor or descriptor too short > [ 369.370574] usb 4-4: unable to read config index 0 descriptor/start: -71 > [ 369.370586] usb 4-4: can't read configurations, error -71 > ... > [ 370.325161] usb 4-4: Device not responding to setup address. > [ 370.537171] usb 4-4: Device not responding to setup address. > [ 370.744627] usb 4-4: device not accepting address 5, error -71 > [ 370.753259] usb usb4-port4: attempt power cycle > [ 371.861790] usb 4-4: new SuperSpeed USB device number 6 using xhci_hcd > [ 372.067721] usb 4-4: LPM exit latency is zeroed, disabling LPM. > [ 372.157551] usb 4-4: New USB device found, idVendor=07ca, idProduct=2553, > bcdDevice= 1.00 > [ 372.157560] usb 4-4: New USB device strings: Mfr=6, Product=7, > SerialNumber=3 > [ 372.157564] usb 4-4: Product: Live Gamer Ultra 2.1 > > > Logs with dynamic debug enabled for usb core and xhci host could possibly > show more info > > <boot> > mount -t debugfs none /sys/kernel/debug > echo 'module xhci_hcd =p' >/sys/kernel/debug/dynamic_debug/control > echo 'module usbcore =p' >/sys/kernel/debug/dynamic_debug/control > <connect device> > <send dmesg> Ok Cheers I'll check this out and post results if I find anything. And yeah the USB devices under Linux do get power cycled from time to time as they can freeze up occasionally when used in OBS.
> <boot>
> mount -t debugfs none /sys/kernel/debug
> echo 'module xhci_hcd =p' >/sys/kernel/debug/dynamic_debug/control
> echo 'module usbcore =p' >/sys/kernel/debug/dynamic_debug/control
> <connect device>
> <send dmesg>
These commands are not allowed under Kubuntu which is the distro I'm using atm (did complete ssd wipe).
[ 668.628381] usb 3-4: new high-speed USB device number 4 using xhci_hcd [ 668.755970] usb 3-4: New USB device found, idVendor=1a40, idProduct=0101, bcdDevice= 1.11 [ 668.755978] usb 3-4: New USB device strings: Mfr=0, Product=1, SerialNumber=0 [ 668.755983] usb 3-4: Product: USB 2.0 Hub [ 668.818075] hub 3-4:1.0: USB hub found [ 668.818344] hub 3-4:1.0: 4 ports detected [ 669.113942] usb 4-4: new SuperSpeed USB device number 5 using xhci_hcd [ 674.233892] usb 4-4: unable to get BOS descriptor or descriptor too short [ 674.258279] usb 4-4: unable to read config index 0 descriptor/start: -71 [ 674.258288] usb 4-4: can't read configurations, error -71 [ 674.409839] usb 3-4.3: new full-speed USB device number 5 using xhci_hcd [ 674.628846] usb 3-4.3: New USB device found, idVendor=07ca, idProduct=d553, bcdDevice= 1.01 [ 674.628853] usb 3-4.3: New USB device strings: Mfr=1, Product=2, SerialNumber=3 [ 674.628858] usb 3-4.3: Product: Live Gamer Ultra 2.1-RGB [ 674.628862] usb 3-4.3: Manufacturer: AVerMedia Computer Inc. [ 674.628866] usb 3-4.3: SerialNumber: 5204140800332 [ 674.683236] hid-generic 0003:07CA:D553.000B: hiddev5,hidraw10: USB HID v1.11 Device [AVerMedia Computer Inc. Live Gamer Ultra 2.1-RGB] on usb-0000:30:00.3-4.3/input1 [ 675.205949] usb 4-4: Device not responding to setup address. [ 675.418073] usb 4-4: Device not responding to setup address. [ 675.625936] usb 4-4: device not accepting address 6, error -71 [ 675.634016] usb usb4-port4: attempt power cycle [ 676.742634] usb 4-4: new SuperSpeed USB device number 7 using xhci_hcd [ 676.950771] usb 4-4: LPM exit latency is zeroed, disabling LPM. [ 677.048948] usb 4-4: New USB device found, idVendor=07ca, idProduct=2553, bcdDevice= 1.00 [ 677.048957] usb 4-4: New USB device strings: Mfr=6, Product=7, SerialNumber=3 [ 677.048962] usb 4-4: Product: Live Gamer Ultra 2.1 [ 677.048966] usb 4-4: Manufacturer: AVerMedia [ 677.048970] usb 4-4: SerialNumber: 5204140800332 [ 677.147762] uvcvideo 4-4:1.1: Unknown video format 30313050-0000-0010-8000-00aa00389b71 [ 677.147779] usb 4-4: Found UVC 1.00 device Live Gamer Ultra 2.1 (07ca:2553) [ 677.547565] usb 4-4: 3:2: failed to get current value for ch 0 (-22) Nothing about superspeedplus sorry to day. I don't understand the descriptor and can't debug further as dynamic debug is not allowed under this distro atm. I(In reply to Jarrard from comment #15) > [ 668.628381] usb 3-4: new high-speed USB device number 4 using xhci_hcd > [ 668.755970] usb 3-4: New USB device found, idVendor=1a40, idProduct=0101, > bcdDevice= 1.11 > [ 668.755978] usb 3-4: New USB device strings: Mfr=0, Product=1, > SerialNumber=0 > [ 668.755983] usb 3-4: Product: USB 2.0 Hub > [ 668.818075] hub 3-4:1.0: USB hub found > [ 668.818344] hub 3-4:1.0: 4 ports detected > [ 669.113942] usb 4-4: new SuperSpeed USB device number 5 using xhci_hcd > [ 674.233892] usb 4-4: unable to get BOS descriptor or descriptor too short > [ 674.258279] usb 4-4: unable to read config index 0 descriptor/start: -71 > [ 674.258288] usb 4-4: can't read configurations, error -71 > [ 674.409839] usb 3-4.3: new full-speed USB device number 5 using xhci_hcd > [ 674.628846] usb 3-4.3: New USB device found, idVendor=07ca, > idProduct=d553, bcdDevice= 1.01 > [ 674.628853] usb 3-4.3: New USB device strings: Mfr=1, Product=2, > SerialNumber=3 > [ 674.628858] usb 3-4.3: Product: Live Gamer Ultra 2.1-RGB > [ 674.628862] usb 3-4.3: Manufacturer: AVerMedia Computer Inc. > [ 674.628866] usb 3-4.3: SerialNumber: 5204140800332 > [ 674.683236] hid-generic 0003:07CA:D553.000B: hiddev5,hidraw10: USB HID > v1.11 Device [AVerMedia Computer Inc. Live Gamer Ultra 2.1-RGB] on > usb-0000:30:00.3-4.3/input1 > [ 675.205949] usb 4-4: Device not responding to setup address. > [ 675.418073] usb 4-4: Device not responding to setup address. > [ 675.625936] usb 4-4: device not accepting address 6, error -71 > [ 675.634016] usb usb4-port4: attempt power cycle > [ 676.742634] usb 4-4: new SuperSpeed USB device number 7 using xhci_hcd > [ 676.950771] usb 4-4: LPM exit latency is zeroed, disabling LPM. > [ 677.048948] usb 4-4: New USB device found, idVendor=07ca, idProduct=2553, > bcdDevice= 1.00 > [ 677.048957] usb 4-4: New USB device strings: Mfr=6, Product=7, > SerialNumber=3 > [ 677.048962] usb 4-4: Product: Live Gamer Ultra 2.1 > [ 677.048966] usb 4-4: Manufacturer: AVerMedia > [ 677.048970] usb 4-4: SerialNumber: 5204140800332 > [ 677.147762] uvcvideo 4-4:1.1: Unknown video format > 30313050-0000-0010-8000-00aa00389b71 > [ 677.147779] usb 4-4: Found UVC 1.00 device Live Gamer Ultra 2.1 > (07ca:2553) > [ 677.547565] usb 4-4: 3:2: failed to get current value for ch 0 (-22) > > > > > > Nothing about superspeedplus sorry to day. I don't understand the > descriptor and can't debug further as dynamic debug is not allowed under > this distro atm. Interesting enough dynamic debug is compiled in with this release of kernel for kubuntu but it spits a permission denied with any of those cat commands. I don't really understand the issue. Pretty sure I was able to do them under Arch so maybe I should go back to that instead of using kubuntu. For some reason I find Arch simpler, lol Ok seems I'm a bit of a idiot and had secure boot enabled. Now the commands work. Got a whole bunch of stuff. [ 619.877598] xhci_hcd 0000:30:00.3: Get port status 4-4 read: 0xe2a1203, return 0x310203 [ 619.877612] xhci_hcd 0000:30:00.3: xhci_get_isoc_frame_id: index 0, reg 0x64b start_frame_id 0xca, end_frame_id 0x448, start_frame 0xd5 [ 619.877627] xhci_hcd 0000:30:00.3: clear port4 reset change, portsc: 0xe0a1203 [ 619.877641] xhci_hcd 0000:30:00.3: clear port4 warm(BH) reset change, portsc: 0xe021203 [ 619.877654] xhci_hcd 0000:30:00.3: clear port4 link state change, portsc: 0xe021203 [ 619.877667] xhci_hcd 0000:30:00.3: clear port4 connect change, portsc: 0xe001203 [ 619.877679] xhci_hcd 0000:30:00.3: Get port status 4-4 read: 0xe001203, return 0x203 ... [ 619.929597] xhci_hcd 0000:30:00.3: Set root hub portnum to 8 [ 619.929602] xhci_hcd 0000:30:00.3: Set fake root hub portnum to 4 [ 619.929606] xhci_hcd 0000:30:00.3: udev->tt = 0000000000000000 [ 619.929610] xhci_hcd 0000:30:00.3: udev->ttport = 0x0 [ 619.929616] xhci_hcd 0000:30:00.3: // Ding dong! ... [ 619.937709] usb 4-4: Device not responding to setup address. [ 619.937754] xhci_hcd 0000:30:00.3: // Ding dong! [ 619.937775] xhci_hcd 0000:30:00.3: // Ding dong! [ 619.937811] xhci_hcd 0000:30:00.3: Slot 7 output ctx = 0x0x00000000fffc2000 (dma) [ 619.937817] xhci_hcd 0000:30:00.3: Slot 7 input ctx = 0x0x00000000fff42000 (dma) [ 619.937823] xhci_hcd 0000:30:00.3: Set slot id 7 dcbaa entry 00000000dcd62c49 to 0xfffc2000 [ 619.938588] xhci_hcd 0000:30:00.3: xhci_get_isoc_frame_id: index 0, reg 0x833 start_frame_id 0x107, end_frame_id 0x485, start_frame 0x112 [ 619.938720] xhci_hcd 0000:30:00.3: Port change event, 4-4, id 8, portsc: 0xe0202a0 [ 619.938726] xhci_hcd 0000:30:00.3: handle_port_status: starting usb4 port polling. ... [ 620.141591] xhci_hcd 0000:30:00.3: Set root hub portnum to 8 [ 620.141596] xhci_hcd 0000:30:00.3: Set fake root hub portnum to 4 [ 620.141600] xhci_hcd 0000:30:00.3: udev->tt = 0000000000000000 [ 620.141604] xhci_hcd 0000:30:00.3: udev->ttport = 0x0 [ 620.141610] xhci_hcd 0000:30:00.3: // Ding dong! ... [ 620.149708] usb 4-4: Device not responding to setup address. [ 620.149752] xhci_hcd 0000:30:00.3: // Ding dong! [ 620.149774] xhci_hcd 0000:30:00.3: // Ding dong! [ 620.149810] xhci_hcd 0000:30:00.3: Slot 7 output ctx = 0x0x00000000fffc2000 (dma) [ 620.149816] xhci_hcd 0000:30:00.3: Slot 7 input ctx = 0x0x00000000fff42000 (dma) [ 620.149822] xhci_hcd 0000:30:00.3: Set slot id 7 dcbaa entry 00000000dcd62c49 to 0xfffc2000 ... [ 620.357571] usb 4-4: device not accepting address 19, error -71 [ 620.357587] xhci_hcd 0000:30:00.3: xhci_get_isoc_frame_id: index 0, reg 0x154b start_frame_id 0x2aa, end_frame_id 0x628, start_frame 0x2b5 ... [ 620.365636] xhci_hcd 0000:30:00.3: // Ding dong! [ 620.365655] xhci_hcd 0000:30:00.3: Bad real port. [ 620.365662] usb usb4-port4: attempt power cycle [ 620.365668] xhci_hcd 0000:30:00.3: set port power 4-4 OFF, portsc: 0xe021263 ... [ 620.573567] xhci_hcd 0000:30:00.3: set port power 4-4 ON, portsc: 0xe020080 ... Here is a more truncated dmesg log, ran it a few times and this seems to be the sum of the log when plugging the device in. Removed most the xhci_get_isoc_frame_id: index 0........ stuff Keep in mind this USB capture card has around 3 devices on it from what I can see. Video, Audio, RGB Lighting. [ 1560.044886] xhci_hcd 0000:30:00.3: udev->tt = 0000000000000000 [ 1560.044890] xhci_hcd 0000:30:00.3: udev->ttport = 0x0 [ 1560.044895] xhci_hcd 0000:30:00.3: // Ding dong! [ 1560.052981] usb 4-4: Device not responding to setup address. [ 1560.053033] xhci_hcd 0000:30:00.3: // Ding dong! [ 1560.053055] xhci_hcd 0000:30:00.3: // Ding dong! [ 1560.053092] xhci_hcd 0000:30:00.3: Slot 7 output ctx = 0x0x00000000fff26000 (dma) [ 1560.053098] xhci_hcd 0000:30:00.3: Slot 7 input ctx = 0x0x00000000fff17000 (dma) [ 1560.053105] xhci_hcd 0000:30:00.3: Set slot id 7 dcbaa entry 00000000dcd62c49 to 0xfff26000 [ 1560.053130] xhci_hcd 0000:30:00.3: xhci_get_isoc_frame_id: index 0, reg 0xac3 start_frame_id 0x159, end_frame_id 0x4d7, start_frame 0x164 [ 1560.054017] xhci_hcd 0000:30:00.3: Port change event, 4-4, id 8, portsc: 0xe0202a0 [ 1560.054025] xhci_hcd 0000:30:00.3: handle_port_status: starting usb4 port polling. [ 1560.256853] xhci_hcd 0000:30:00.3: Set root hub portnum to 8 [ 1560.256857] xhci_hcd 0000:30:00.3: Set fake root hub portnum to 4 [ 1560.256861] xhci_hcd 0000:30:00.3: udev->tt = 0000000000000000 [ 1560.256865] xhci_hcd 0000:30:00.3: udev->ttport = 0x0 [ 1560.256871] xhci_hcd 0000:30:00.3: // Ding dong! [ 1560.264980] usb 4-4: Device not responding to setup address. [ 1560.265022] xhci_hcd 0000:30:00.3: // Ding dong! [ 1560.265045] xhci_hcd 0000:30:00.3: // Ding dong! [ 1560.265079] xhci_hcd 0000:30:00.3: Slot 7 output ctx = 0x0x00000000fff26000 (dma) [ 1560.265085] xhci_hcd 0000:30:00.3: Slot 7 input ctx = 0x0x00000000fff17000 (dma) [ 1560.265091] xhci_hcd 0000:30:00.3: Set slot id 7 dcbaa entry 00000000dcd62c49 to 0xfff26000 [ 1560.473834] usb 4-4: device not accepting address 25, error -71 [ 1560.473846] xhci_hcd 0000:30:00.3: Set port 4-4 link state, portsc: 0xe021203, write 0xe011261 [ 1560.481893] xhci_hcd 0000:30:00.3: // Ding dong! [ 1560.481911] xhci_hcd 0000:30:00.3: Bad real port. [ 1560.481918] usb usb4-port4: attempt power cycle [ 1560.481924] xhci_hcd 0000:30:00.3: set port power 4-4 OFF, portsc: 0xe021263 [ 1560.688829] xhci_hcd 0000:30:00.3: set port power 4-4 ON, portsc: 0xe020080 [ 1560.792813] xhci_hcd 0000:30:00.3: // Ding dong! [ 1560.792835] xhci_hcd 0000:30:00.3: Slot 7 output ctx = 0x0x00000000fff26000 (dma) [ 1560.792841] xhci_hcd 0000:30:00.3: Slot 7 input ctx = 0x0x00000000fff17000 (dma) [ 1560.792849] xhci_hcd 0000:30:00.3: Set slot id 7 dcbaa entry 00000000dcd62c49 to 0xfff26000 [ 1560.792888] xhci_hcd 0000:30:00.3: Get port status 4-4 read: 0xe0202a0, return 0x102a0 [ 1560.792912] xhci_hcd 0000:30:00.3: set port reset, actual port 4-4 status = 0xe0202b0 [ 1560.854820] xhci_hcd 0000:30:00.3: Get port status 4-4 read: 0xe2a02a0, return 0x3102a0 [ 1560.854839] usb usb4-port4: not reset yet, waiting 60ms [ 1560.916802] xhci_hcd 0000:30:00.3: Get port status 4-4 read: 0xe2a02a0, return 0x3102a0 [ 1560.916831] usb usb4-port4: not reset yet, waiting 200ms [ 1561.120792] xhci_hcd 0000:30:00.3: Get port status 4-4 read: 0xe2a02a0, return 0x3102a0 [ 1561.120840] usb usb4-port4: not reset yet, waiting 200ms [ 1561.328770] xhci_hcd 0000:30:00.3: Get port status 4-4 read: 0xe2a02a0, return 0x3102a0 [ 1561.328790] usb usb4-port4: not reset yet, waiting 200ms [ 1561.536760] xhci_hcd 0000:30:00.3: Get port status 4-4 read: 0xe2a1203, return 0x310203 [ 1561.536780] xhci_hcd 0000:30:00.3: clear port4 reset change, portsc: 0xe0a1203 [ 1561.536794] xhci_hcd 0000:30:00.3: clear port4 warm(BH) reset change, portsc: 0xe021203 [ 1561.536810] xhci_hcd 0000:30:00.3: clear port4 link state change, portsc: 0xe021203 [ 1561.536824] xhci_hcd 0000:30:00.3: clear port4 connect change, portsc: 0xe001203 [ 1561.536837] xhci_hcd 0000:30:00.3: Get port status 4-4 read: 0xe001203, return 0x203 [ 1561.588756] xhci_hcd 0000:30:00.3: Set root hub portnum to 8 [ 1561.588762] xhci_hcd 0000:30:00.3: Set fake root hub portnum to 4 [ 1561.588766] xhci_hcd 0000:30:00.3: udev->tt = 0000000000000000 [ 1561.588770] xhci_hcd 0000:30:00.3: udev->ttport = 0x0 [ 1561.588776] xhci_hcd 0000:30:00.3: // Ding dong! [ 1561.589112] xhci_hcd 0000:30:00.3: xhci_get_isoc_frame_id: index 0, reg 0x3ac3 start_frame_id 0x759, end_frame_id 0x2d7, start_frame 0x764 [ 1561.589376] xhci_hcd 0000:30:00.3: Successful setup address command [ 1561.589383] xhci_hcd 0000:30:00.3: Op regs DCBAA ptr = 0x000000fffff000 [ 1561.589387] xhci_hcd 0000:30:00.3: Slot ID 7 dcbaa entry @00000000dcd62c49 = 0x000000fff26000 [ 1561.589393] xhci_hcd 0000:30:00.3: Output Context DMA address = 0xfff26000 [ 1561.589397] xhci_hcd 0000:30:00.3: Internal device address = 7 [ 1561.589404] usb 4-4: new SuperSpeed USB device number 26 using xhci_hcd [ 1561.784721] xhci_hcd 0000:30:00.3: xhci_hub_status_data: stopping usb4 port polling [ 1561.811955] usb 4-4: LPM exit latency is zeroed, disabling LPM. [ 1561.857239] usb 4-4: skipped 1 descriptor after configuration [ 1561.857246] usb 4-4: skipped 5 descriptors after interface [ 1561.857253] usb 4-4: skipped 2 descriptors after endpoint [ 1561.857259] usb 4-4: skipped 42 descriptors after interface [ 1561.857265] usb 4-4: skipped 2 descriptors after endpoint [ 1561.857270] usb 4-4: skipped 5 descriptors after interface [ 1561.857275] usb 4-4: skipped 2 descriptors after interface [ 1561.857281] usb 4-4: skipped 2 descriptors after endpoint [ 1561.870967] xhci_hcd 0000:30:00.3: Waiting for status stage event [ 1561.871029] usb 4-4: default language 0x0409 [ 1561.910158] usb 4-4: udev 26, busnum 4, minor = 409 [ 1561.910165] usb 4-4: New USB device found, idVendor=07ca, idProduct=2553, bcdDevice= 1.00 [ 1561.910172] usb 4-4: New USB device strings: Mfr=6, Product=7, SerialNumber=3 [ 1561.910177] usb 4-4: Product: Live Gamer Ultra 2.1 [ 1561.910183] usb 4-4: Manufacturer: AVerMedia [ 1561.910188] usb 4-4: SerialNumber: 5204140800332 [ 1561.910426] usb 4-4: usb_probe_device [ 1561.910432] usb 4-4: configuration #1 chosen from 1 choice [ 1561.910446] xhci_hcd 0000:30:00.3: add ep 0x85, slot id 7, new drop flags = 0x0, new add flags = 0x800 [ 1561.910456] xhci_hcd 0000:30:00.3: add ep 0x82, slot id 7, new drop flags = 0x0, new add flags = 0x820 [ 1561.910463] xhci_hcd 0000:30:00.3: xhci_check_bandwidth called for udev 0000000005f01f8b [ 1561.910472] xhci_hcd 0000:30:00.3: // Ding dong! [ 1561.911116] xhci_hcd 0000:30:00.3: xhci_get_isoc_frame_id: index 0, reg 0x4d3 [ 1561.916735] xhci_hcd 0000:30:00.3: Successful Endpoint Configure command [ 1561.916772] xhci_hcd 0000:30:00.3: // Ding dong! [ 1561.916784] xhci_hcd 0000:30:00.3: Stopped on No-op or Link TRB for slot 7 ep 10 [ 1561.916795] xhci_hcd 0000:30:00.3: // Ding dong! [ 1561.932735] xhci_hcd 0000:30:00.3: // Ding dong! [ 1561.932764] xhci_hcd 0000:30:00.3: Stopped on No-op or Link TRB for slot 7 ep 4 [ 1561.932772] xhci_hcd 0000:30:00.3: // Ding dong! Plugged in a USB3 10Gbps HUB (red ports). Under windows usbview shows it as USB3.2 SuperSpeedPlus 10Gbps and the USB devices connected agree. UNDER LINUX it is a 5Gbps hub and no SuperSpeedPlus. I can't believe this sort of massive USB port bug has existed this long under Linux, does nobody test AM4/AM5 usb port speed compatibility? Anyway, still no clue on how to fix this, way above my pay grade. On Wed, Jun 12, 2024, bugzilla-daemon@kernel.org wrote: > > https://urldefense.com/v3/__https://bugzilla.kernel.org/show_bug.cgi?id=218943__;!!A4F2R9G_pg!cCuzUsUD-2ePSsdCBgCODNgftuH0LhiYyBq6--AbqKmW5wPBuGEQCnSgOWhEOf4rPi-n6bd02HAx-VvD0vN56GomiUx1GIRM$ > > --- Comment #19 from Jarrard (jarrard@proton.me) --- > Plugged in a USB3 10Gbps HUB (red ports). Under windows usbview shows it as > USB3.2 SuperSpeedPlus 10Gbps and the USB devices connected agree. > > UNDER LINUX it is a 5Gbps hub and no SuperSpeedPlus. I can't believe this > sort > of massive USB port bug has existed this long under Linux, does nobody test > AM4/AM5 usb port speed compatibility? > > Anyway, still no clue on how to fix this, way above my pay grade. > Are you using AsMedia host? Then I suspect it's a similar issue that was reported before, and the issue is likely in the host controller reporting the wrong port ID and not the driver issue. It's a different AsMedia host version, but you can checkout this [*] thread for more info. I suspect the test patch I provided there should resolve your issue. It's been a while, you may need to tweak that change to apply for your latest kernel. [*] https://lore.kernel.org/all/6908aa69-469b-8f92-8e19-60685f524f9c@synopsys.com/ BR, Thinh (In reply to Thinh.Nguyen from comment #20) > On Wed, Jun 12, 2024, bugzilla-daemon@kernel.org wrote: > > > > > https://urldefense.com/v3/__https://bugzilla.kernel.org/show_bug.cgi?id=218943__;!!A4F2R9G_pg!cCuzUsUD-2ePSsdCBgCODNgftuH0LhiYyBq6--AbqKmW5wPBuGEQCnSgOWhEOf4rPi-n6bd02HAx-VvD0vN56GomiUx1GIRM$ > > > > --- Comment #19 from Jarrard (jarrard@proton.me) --- > > Plugged in a USB3 10Gbps HUB (red ports). Under windows usbview shows it > as > > USB3.2 SuperSpeedPlus 10Gbps and the USB devices connected agree. > > > > UNDER LINUX it is a 5Gbps hub and no SuperSpeedPlus. I can't believe this > > sort > > of massive USB port bug has existed this long under Linux, does nobody test > > AM4/AM5 usb port speed compatibility? > > > > Anyway, still no clue on how to fix this, way above my pay grade. > > > > Are you using AsMedia host? Then I suspect it's a similar issue that was > reported before, and the issue is likely in the host controller > reporting the wrong port ID and not the driver issue. > > It's a different AsMedia host version, but you can checkout this [*] > thread for more info. I suspect the test patch I provided there should > resolve your issue. It's been a while, you may need to tweak that change > to apply for your latest kernel. > > [*] > https://lore.kernel.org/all/6908aa69-469b-8f92-8e19-60685f524f9c@synopsys. > com/ > > BR, > Thinh Thanks I'll check that out sometime. I don't normally compile my own kernels so will need to set it up for doing so under this Kubuntu install. I applied the patch the LQX under Arch and I think it fixed the USB3.2 hub to now report 10Gbps but Live Gamer Ultra 2.1 device is still hand shaking at 5Gbps sadly. Might just need to go back to Windows OBS for this one. I don't understand how this can happen after watching YT videos of other people getting 10Gbps under Linux with this device. Very annoying issue. Maybe one day, must be a issue with combination of AM4/AM5 with Via-USB controllers or something. Created attachment 306461 [details]
USBView (patch)
Here is a image showing what I see in USB View app.
When the HUB is connected it does show as 10Gbps but AsMedia device is still 5Gbps. HOWEVER here we clearly see the VIA hub is 5Gbps which is confusing to me.
I don't understand what is going on but this seems like a pretty big problem. Guessing INTEL CPU/MOBO users don't have this issue.
This is with XANMOD kernel with USB Patch applied as suggested.
Perhaps there is a modification to the patch that can be done to fix this specific asmedia device? I do not know much about programming so can't really figure it out, its a small miracle I even got the patch applied successfully (according to logs).
Sent a ticket away to AverMedia about this issue directing them here. I hope they can provide a Firmware fix for this AM4/5 issue. That is assuming its NOT a kernel bug as I understand it. For reference here is one video where he talks about the 5Gbps port connection speed issue under Linux with a Intel system. It seems this may also be the issue for Intel systems so perhaps I was wrong in assuming it was just a AMD issue. https://youtu.be/imvTVL7oS2w?t=589 AverMedia said they won't be fixing this for Linux. So until a exception flag can be compiled into these USB speed checks for this device. Its stuck at 5Gbps. Wish I was a Kernel code-master, probably fixable within a day for myself. Thanks for help none the less. The OBS server will remain on Win10. (I have dual boot in case miracle happens and this issue is resolved) (In reply to Jarrard from comment #23) > When the HUB is connected it does show as 10Gbps but AsMedia device is still > 5Gbps. HOWEVER here we clearly see the VIA hub is 5Gbps which is confusing > to me. I can't see any AsMedia device, maybe you mean the AVerMedia device? What do you mean with the hub shows as 10Gbps? I can only see one hub, the ViaLabs VL813 and it is clearly a 5Gbps hub, it is identifying itself as USB3.0 (55Gbps) and the mfgr specs also says that it is a 5Gbps hub. (In reply to Lars Melin from comment #27) > (In reply to Jarrard from comment #23) > > I can't see any AsMedia device, maybe you mean the AVerMedia device? > What do you mean with the hub shows as 10Gbps? > > I can only see one hub, the ViaLabs VL813 and it is clearly a 5Gbps hub, it > is identifying itself as USB3.0 (55Gbps) and the mfgr specs also says that > it is a 5Gbps hub. Yes that was a typo on my part, its AverMedia I have 10Gbps hub connect to same port now and it shows as 10Gbps. The VIA Hub at 5Gbps is wrong and possibly created by the AverMedia device which is also wrong. But it does not appear when I use a physical 10Gbps hub however the AverMedia device is still wrong. (In reply to Jarrard from comment #28) > I have 10Gbps hub connect to same port now and it shows as 10Gbps. > The VIA Hub at 5Gbps is wrong and possibly created by the AverMedia device > which is also wrong. But it does not appear when I use a physical 10Gbps hub > however the AverMedia device is still wrong. The detection of a hub can not be wrong, what you see is descriptors read from a hub so it is present and probably built into the AverMedia device which from what your usbdevice screenshoot shows also have an RGB light controller as a separate device. The only reason for having a hub is to be able to connect more devices to a bus so you should at least find out where it comes from. Does the hub appear if you disconnect the AverMedia device? Does it appear if connect another USB device to the same computer port? I guess not, I guess it is inside the AVerMedia device. Created attachment 306466 [details]
Win10USBCapture
Yeah it disappears.
Have a look at the device and hub here under Windows, no VIA Hub, must be a Linux thing.
(In reply to Jarrard from comment #30) > Created attachment 306466 [details] > Win10USBCapture > > Yeah it disappears. > > Have a look at the device and hub here under Windows, no VIA Hub, must be a > Linux thing. Forgot to mention this is with it connected to the 10Gbps HUB in the same port the capture card was directly connected too. I just decided to use this hub as it expands the 10Gbps ports I have. It does appear to disappear the VIA hub under Linux which must be made by the capture card IF a core hub device isn't found? Either way it does not change the AverMedia capture connection speed, still stuck at 5Gbps for Linux. To conclude this bug report I have moved the OBS server to windows as a temp solution but would like to return it to Linux eventually. However for that to happen I think 10Gbps would need to be manually flagged on this vendor+device ID with some patch. I have tried the suggested PATCH set already and it allows the HUB to connect at 10Gbps but the device itself remains at 5Gbps so further patching would be need. I do not know how to do this. If anyone is able to help with a manual kernel patch to test other then the one in the comments here then I'm happy to test. Thankyou. On Fri, Jun 21, 2024, bugzilla-daemon@kernel.org wrote: > > https://urldefense.com/v3/__https://bugzilla.kernel.org/show_bug.cgi?id=218943__;!!A4F2R9G_pg!fODPpvu8QMI1SH2NW38mJgNxScddtxlr7-FRbFln8qt4ISm6Ib7P6OXrtb1YRRyf5zD7koRGYGLNbyWW0Ts_xUuGkHNHpFhB$ > > --- Comment #32 from Jarrard (jarrard@proton.me) --- > To conclude this bug report I have moved the OBS server to windows as a temp > solution but would like to return it to Linux eventually. > > However for that to happen I think 10Gbps would need to be manually flagged > on > this vendor+device ID with some patch. > > I have tried the suggested PATCH set already and it allows the HUB to connect > at 10Gbps but the device itself remains at 5Gbps so further patching would be > need. I do not know how to do this. So it's not AsMedia host? Regardless, if the patch allows the host to see SuperSpeed Plus hub, then it confirms the host controller reported the wrong port speed ID in its root port. The patch gets the speed from device notification instead. The device notification sublink speed should be propagated for devices under the hub. The fact that you still see 5Gbps means there are other problems either due to the host controller or the hub. > > If anyone is able to help with a manual kernel patch to test other then the > one > in the comments here then I'm happy to test. > > Thankyou. > It'd be better for the vendor to provide the support and patches to resolve this since they have the knowledge and the tools to debug their hardware. I can only provide clues for whoever needs to proceed further. BR, Thinh (In reply to Thinh.Nguyen from comment #33) > On Fri, Jun 21, 2024, bugzilla-daemon@kernel.org wrote: > > > > > https://urldefense.com/v3/__https://bugzilla.kernel.org/show_bug.cgi?id=218943__;!!A4F2R9G_pg!fODPpvu8QMI1SH2NW38mJgNxScddtxlr7-FRbFln8qt4ISm6Ib7P6OXrtb1YRRyf5zD7koRGYGLNbyWW0Ts_xUuGkHNHpFhB$ > > > > It'd be better for the vendor to provide the support and patches to > resolve this since they have the knowledge and the tools to debug their > hardware. > > I can only provide clues for whoever needs to proceed further. > > BR, > Thinh I cannot modify my posts so where I say AsMedia I actually meant AverMedia. Also I talked to the vendor and they do not support linux and that was the end of it. |
Created attachment 306432 [details] lsusb verbose list of devices and hubs Having major problem getting my USB capture card and any usb device that is rated at 10Gbps working under Linux at 10Gbps. They all default to 5Gbps. Things done: Checked systems on Windows, 10Gbps confirmed Checked devices on Windows "" Checked multiple 10Gbps+ cables under Windows, all works. Checked Linux kernel to ensure superspeedplus Linux system won't use superspeedplus on devices and via usb hubs, all are forced to 5Gbps. I have tested on Arch and Ubuntu with multiple cables and systems. Device in question is USB Live Gamer 2.1 Capture Card from AsMedia. It works flawlessly under Windows at 10Gbps! I've added a lsusb list and as can be showned the root hub supports superspeedplus but all devices are 5000M however the very last one is AsMedia device should should be 10000M I've tried many ports, I can't figure out what is wrong. There is a youtube video showing this device works at 10Gbps under Linux but this system can't. If I can't fix this I will need to abandon Linux and move back to Windows for the streaming due to this bug.