On 21.11.2016 23:52, Simon Arlott wrote:
I have a 5-port USB 2.0 hub attached to a USB 3.0 host by a long length
of Cat 5e cable (via a pair of cheap USB<->Cat5e adapters from eBay)
that only works at full speed and doesn't stay connected all the time.
This wouldn't be a problem except that the kernel can get stuck in a
loop with the "URB transfer length is wrong, xHC issue?" error.
The message "hub_ext_port_status failed (err = -71)" in particular can
occur 70+ times for every instance of this error. There's no guarantee
that it will ever disconnect the device and stop doing this, it has
previously suddenly started generating 30MB/minute of log messages for
30 minutes until I unplugged the device.
Would it be ok to rate limit both of these messages to reduce the impact
of poor connections?
2016-06-17T09:56:09.219+01:00 <warning> kernel: xhci_hcd 0000:00:14.0: URB transfer length is wrong, xHC issue? req. len = 4, act. len = 4294967292
Your broken hardware triggered another interesting issue.
The len value 4294967292 looks like a u32 wrapped around.
What kernelversion was this?
-Mathias
--
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
