On Tue, 22 Oct 2013, Prasad Koya wrote: > Hi > > I have a USB subsystem question. If there is a specific mailing list for > USB queries I could post it there. The specific mailing list for USB queries is <linux-usb@xxxxxxxxxxxxxxx>. It's listed in the kernel's MAINTAINERS file. > I'm seeing this problem with both 2.6.38 > and 3.4 kernels. Have you tried anything more up to date, such as 3.11? > I have an embedded system that boots off flash. If the main kernel crashes, Why does your kernel crash? That should be the first thing you want to fix. > we go into another kernel to save dmesg from previous kernel. Inside > crash_kexec'ed kernel, we wait for same flash that main kernel booted off > to be mounted before we copy out the dmesg from old kernel. At times, I see > this error inside crash_kexec'ed kernel: > > usb 1-3: khubd timed out on ep0in len=0/64 > > and the USB flash fails to come up with ehci driver. CONFIG_PM is not > defined. > > Here are some more debugs with return codes around this. I > tried USB_PORT_FEAT_SUSPEND before USB_PORT_FEAT_RESET and I get EPIPE. > > [ 6.628118] usb 1-3: khubd timed out on ep0in len=0/64 status -2 > [ 6.628124] usb_start_wait_urb 78: retval -110 status -2 > [ 6.628128] usb_internal_control_msg retv -110 len 0 > [ 6.628132] usb_control_msg pipe 80000080 reqtype 80 ret -110 > [ 6.628135] hub_port_init: iter 0 maxpktsize 0 r -110 > [ 6.628140] usb 1-3: usb_start_wait_urb timeout 5000 > [ 7.793311] usb_start_wait_urb 78: retval -71 status -71 > [ 7.793316] usb_internal_control_msg retv -71 len 0 > [ 7.793319] usb_control_msg pipe 80000080 reqtype 80 ret -71 > [ 7.793323] hub_port_init: iter 1 maxpktsize 0 r -71 > [ 7.793328] usb 1-3: usb_start_wait_urb timeout 5000 > [ 7.801435] usb_start_wait_urb 78: retval -71 status -71 > [ 7.801439] usb_internal_control_msg retv -71 len 0 > [ 7.801443] usb_control_msg pipe 80000080 reqtype 80 ret -71 > [ 7.801446] hub_port_init: iter 2 maxpktsize 0 r -71 > [ 7.801452] usb usb1: usb_start_wait_urb timeout 1000 > [ 7.801459] usb_start_wait_urb 78: retval -32 status -32 > [ 7.801461] usb_internal_control_msg retv -32 len 0 > [ 7.801464] usb_control_msg pipe 80000100 reqtype 23 ret -32 > [ 7.801468] hub 1-0:1.0: suspend port 3 (err = -32) It looks like the firmware in the USB flash device has crashed. That could explain the original kernel crash. Or maybe the kernel crashed while the flash device was in the middle of executing a command. Lots of USB storage devices don't reset properly when that happens. > Appreciate any help/pointers. The kernel log from the time of the original crash might provide some clues. 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
