On Mon, 27 Jun 2016, Dmitry K wrote: > Hi, first for the delay in the reply. > > I've assumed the device is being reset because the device itself has a > led indicator that goes ON when plugged in. Few second before the > timeout message in dmesg the led indicator goes OFF. It comes back ON > after dmseg says "USB disconnected" and stays ON. So it looks like the > device resets but not functioning after the reset. I dont know for > sure if the device being reset, I've just assumed. Yes, that does sound like some kind of reset. > Here is the new output from dmesg: > [ +27.168417] usb 5-2: new SuperSpeed USB device number 4 using xhci_hcd > [ +0.016962] usb-storage 5-2:1.0: USB Mass Storage device detected > [ +0.003979] scsi host4: usb-storage 5-2:1.0 > [ +1.002877] scsi 4:0:0:0: Direct-Access Generic STORAGE DEVICE > TS26 PQ: 0 ANSI: 6 > [ +0.000639] scsi 4:0:0:1: Direct-Access Generic STORAGE DEVICE > TS26 PQ: 0 ANSI: 6 > [ +0.000582] scsi 4:0:0:2: Direct-Access Generic STORAGE DEVICE > TS26 PQ: 0 ANSI: 6 > [ +0.009352] sd 4:0:0:0: [sdc] Attached SCSI removable disk > [ +0.000705] sd 4:0:0:2: [sde] Attached SCSI removable disk > [ +0.000647] sd 4:0:0:1: [sdd] Attached SCSI removable disk > [Jun27 19:47] xhci_hcd 0000:06:00.0: Timeout while waiting for setup > device command > [ +5.214252] xhci_hcd 0000:06:00.0: Timeout while waiting for setup > device command > [ +0.206615] usb 5-2: device not accepting address 4, error -62 > [ +3.970753] usb usb5-port2: Cannot enable. Maybe the USB cable is bad? > [ +2.400364] usb usb5-port2: Could not disable after 2000 ms > [ +0.770129] usb 5-2: device not accepting address 4, error -22 > [ +3.970735] usb usb5-port2: Cannot enable. Maybe the USB cable is bad? > [ +2.400448] usb usb5-port2: Could not disable after 2000 ms > [ +0.000155] usb 5-2: USB disconnect, device number 4 > > You can see that the new bus is 5 (I'm not sure why bus number jumped, > maybe I've plugged it into another USB port this time). > > Attached is the usbmon for bus 5 from the moment the device is > connected up until few seconds after it was disconnected. It looks like there's a problem in the xHCI hardware or driver. There may also be a problem with the device -- it's hard to tell. The problem starts here: ffff880210ecf180 466827464 S Bi:5:004:1 -115 13 < ffff880210ecf180 497343652 C Bi:5:004:1 -104 13 = 55534243 01010000 60000000 80 ffff88009fd8ea80 497343691 S Ci:5:001:0 s a3 00 0000 0002 0004 4 < ffff88009fd8ea80 497343703 C Ci:5:001:0 0 4 = 03030000 The trace shows a request for a 13-byte bulk-IN transfer, which timed out and was cancelled after 30 seconds. The trace says that actual_len = 13, meaning that all 13 bytes were received, but this is a mistake; the contents of the buffer had not changed since the last time it was used (not shown above). The last line shows that the port's link state changed to Recovery. I don't know exactly what that indicates, but apparently something went wrong with the connection between the device and host controller. Later parts of the trace show the link state getting even more messed up; the port is in reset and the link state is U2, which makes no sense. It eventually straightens out, but it never starts working properly. Alan Stern -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
