Originally I sent a not-working patch to the mailing list [1], turns out the patch is far from correct. Bjørn Mork suggests that we can cover the USB3 pair diff pins with tape to do some experiment, but the vendor told me that there's no such pin. According to the attached log, looks like xhci driver failed to get status from port after resume, eventually make the cdc_mbim probe failed: [ 10.038428] xhci_hcd 0000:00:14.0: get port status, actual port 1 status = 0x202c0 [ 10.038429] xhci_hcd 0000:00:14.0: Get port status returned 0x102c0 [ 10.038463] usb usb2-port2: status 0001.02c0 after resume, -19 [ 10.038465] usb 2-2: can't resume, status -19 [ 10.038465] usb usb2-port2: logical disconnect [ 10.038472] xhci_hcd 0000:00:14.0: Cannot set link state. [ 10.038528] usb usb2-port2: cannot disable (err = -32) [ 10.038575] cdc_mbim 2-2:1.12: usb_probe_interface [ 10.038577] cdc_mbim 2-2:1.12: usb_probe_interface - got id [ 10.038579] cdc_mbim: probe of 2-2:1.12 failed with error -22 [1] https://patchwork.ozlabs.org/patch/691177/
Attachment:
dmesg
Description: Binary data
