RE: 3.6.10: Intel: ixgbe 0000:01:00.0 eth4: Detected Tx Unit Hang

[Tantilov, Emil S]

>-----Original Message-----
>From: linux-kernel-owner@xxxxxxxxxxxxxxx [mailto:linux-kernel-
>owner@xxxxxxxxxxxxxxx] On Behalf Of Justin Piszcz
>Sent: Saturday, December 15, 2012 7:49 AM
>To: linux-kernel@xxxxxxxxxxxxxxx
>Subject: 3.6.10: Intel: ixgbe 0000:01:00.0 eth4: Detected Tx Unit Hang
>
>Hello,
>
>Kernel 3.6.10, first time I have seen this that I can remember (on 10GbE)
>anyway, is this a known issue with 3.6.10?
>
>When the link went down is when I rebooted/etc the remote host attached on
>the other end.
>I've not changed anything physically with the hardware and have been on
>3.6.0-3.6.9 and noticed this when I moved to 3.6.10.
>
>[10270.229200] ixgbe 0000:01:00.0 eth4: NIC Link is Down
>[10276.124937] ixgbe 0000:01:00.0 eth4: NIC Link is Up 10 Gbps, Flow
>Control: RX/TX
>[24529.430997] ixgbe 0000:01:00.0 eth4: Detected Tx Unit Hang
>[24529.430997]   Tx Queue             <10>
>[24529.430997]   TDH, TDT             <4e>, <51>
>[24529.430997]   next_to_use          <51>
>[24529.430997]   next_to_clean        <4e>
>[24529.430997] tx_buffer_info[next_to_clean]
>[24529.430997]   time_stamp           <10172668f>
>[24529.430997]   jiffies              <101726ea4>
>[24529.431011] ixgbe 0000:01:00.0 eth4: tx hang 1 detected on queue 10,
>resetting adapter
>[24529.431028] ixgbe 0000:01:00.0 eth4: Reset adapter
>
>Thoughts?

I don't believe we have seen Tx hangs in validation. If you could narrow down the conditions that lead to the Tx hang that would help a lot. Also the output of ethtool -S eth4 after the Tx hang occurs can be useful to get an idea of the load on the interface.

Thanks,
Emil


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/