On 18.05.2017 17:30, Christoph Gohle wrote:

I have seen several reports around the internet regarding failing io on 
USB-SATA bridges. However, these reports seem to be partially old and/or fixes 
proposed are implemented in my kernel but don’t fix things. Therefore I thought 
I report here again. If this is know/duplicate please apologise.

I am running on ubuntu 16.04 LTS with kernel

$ ~> uname -a
Linux gserv 4.8.0-52-generic #55~16.04.1-Ubuntu SMP Fri Apr 28 14:36:29 UTC 
2017 x86_64 x86_64 x86_64 GNU/Linux

I was also using the 4.4.0-75-generic version of the kernel before, with same 
results. I am having a JMS567 bridge with (currently) three disks attached.

$ ~> lsusb
Bus 002 Device 002: ID 152d:0567 JMicron Technology Corp. / JMicron USA 
Technology Corp. JMS567 SATA 6Gb/s bridge
Bus 002 Device 001: ID 1d6b:0003 Linux Foundation 3.0 root hub

$ ~> ls -l /dev/disk/by-id/
total 0
lrwxrwxrwx 1 root root  9 May 18 00:20 
usb-JMicron_Generic_DISK00_0123456789ABCDEF-0:0 -> ../../sdc
lrwxrwxrwx 1 root root 10 May 18 00:20 
usb-JMicron_Generic_DISK00_0123456789ABCDEF-0:0-part1 -> ../../sdc1
lrwxrwxrwx 1 root root 10 May 18 00:20 
usb-JMicron_Generic_DISK00_0123456789ABCDEF-0:0-part9 -> ../../sdc9
lrwxrwxrwx 1 root root  9 May 18 00:20 
usb-JMicron_Generic_DISK01_0123456789ABCDEF-0:1 -> ../../sdd
lrwxrwxrwx 1 root root 10 May 18 00:20 
usb-JMicron_Generic_DISK01_0123456789ABCDEF-0:1-part1 -> ../../sdd1
lrwxrwxrwx 1 root root  9 May 18 00:20 
usb-JMicron_Generic_DISK02_0123456789ABCDEF-0:2 -> ../../sde
lrwxrwxrwx 1 root root 10 May 18 00:20 
usb-JMicron_Generic_DISK02_0123456789ABCDEF-0:2-part1 -> ../../sde1
lrwxrwxrwx 1 root root 10 May 18 00:20 
usb-JMicron_Generic_DISK02_0123456789ABCDEF-0:2-part9 -> ../../sde9

I am running a ZFS filesystem on those… (actually two pools…). The system seems 
to run fine as long as there is only reading going on or only writing to one 
disk (possibly from somewhere else)….

However once I start a copy operation (here data coming from one of the above 
devices and going to the two others (in a mirror configuration), I get frequent 
io errors from the kernel:

May 17 22:53:13 gserv kernel: [  474.505548] xhci_hcd 0000:00:14.0: ERROR 
Transfer event for disabled endpoint or incorrect stream ring
May 17 22:53:13 gserv kernel: [  474.505670] xhci_hcd 0000:00:14.0: 
@000000026e54c460 00000000 00000000 04000000 02088001

4.11 kernel has xhci traces enabled, could you try to reproduce it with 4.11?
xhci traces can be enabled with:

echo 1 > /sys/kernel/debug/tracing/events/xhci-hcd/enable

If you know how to reliably reproduce this then please enable tracing just 
triggering this. It generates a lot of data.

To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Reply via email to