Bug 48551 - e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
Summary: e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
Status: RESOLVED INVALID
Alias: None
Product: Networking
Classification: Unclassified
Component: IPV4 (show other bugs)
Hardware: x86-64 Linux
: P1 normal
Assignee: Stephen Hemminger
URL:
Keywords:
Depends on: 7450
Blocks:
  Show dependency tree
 
Reported: 2012-10-07 08:10 UTC by Dmitry Shilyaev
Modified: 2012-10-08 14:58 UTC (History)
1 user (show)

See Also:
Kernel Version: 2.6.18-308.13.1.el5
Subsystem:
Regression: No
Bisected commit-id:

Attachments
Server configuration (121.13 KB, text/plain)
2012-10-07 08:10 UTC, Dmitry Shilyaev 		Details
Add an attachment (proposed patch, testcase, etc.)

Description Dmitry Shilyaev 2012-10-07 08:10:56 UTC 
Created attachment 82561 [details]
Server configuration

EEPROM fix script, mentioned in the BUG 7450 is not fit our configuration. We have Intel NIC 82576 series
https://bugzilla.kernel.org/show_bug.cgi?id=7450


We have the following errors in message log

Oct  4 20:11:16 carroll kernel: igb 0000:01:00.0: Detected Tx Unit Hang
Oct  4 20:11:16 carroll kernel:   Tx Queue             <0>
Oct  4 20:11:16 carroll kernel:   TDH                  <b2>
Oct  4 20:11:16 carroll kernel:   TDT                  <17>
Oct  4 20:11:16 carroll kernel:   next_to_use          <17>
Oct  4 20:11:16 carroll kernel:   next_to_clean        <b1>
Oct  4 20:11:16 carroll kernel: buffer_info[next_to_clean]
Oct  4 20:11:16 carroll kernel:   time_stamp           <11ac16024>
Oct  4 20:11:16 carroll kernel:   next_to_watch        <b2>
Oct  4 20:11:16 carroll kernel:   jiffies              <11ac16428>
Oct  4 20:11:16 carroll kernel:   desc.status          <17a8000>
Oct  4 20:11:18 carroll kernel: igb 0000:01:00.0: Detected Tx Unit Hang
Oct  4 20:11:18 carroll kernel:   Tx Queue             <0>
Oct  4 20:11:18 carroll kernel:   TDH                  <25>
Oct  4 20:11:18 carroll kernel:   TDT                  <80>
Oct  4 20:11:18 carroll kernel:   next_to_use          <80>
Oct  4 20:11:18 carroll kernel:   next_to_clean        <24>
Oct  4 20:11:18 carroll kernel: buffer_info[next_to_clean]
Oct  4 20:11:18 carroll kernel:   time_stamp           <11ac16550>
Oct  4 20:11:18 carroll kernel:   next_to_watch        <25>
Oct  4 20:11:18 carroll kernel:   jiffies              <11ac16bf8>
Oct  4 20:11:18 carroll kernel:   desc.status          <1788000>
Oct  4 20:11:28 carroll last message repeated 3 times
Oct  4 20:11:54 carroll last message repeated 3 times
Oct  4 20:11:57 carroll last message repeated 3 times
Oct  4 20:13:00 carroll kernel: igb 0000:01:00.0: Detected Tx Unit Hang
Oct  4 20:13:00 carroll kernel:   Tx Queue             <0>
Oct  4 20:13:00 carroll kernel:   TDH                  <96>
Oct  4 20:13:00 carroll kernel:   TDT                  <2b>
Oct  4 20:13:00 carroll kernel:   next_to_use          <2b>
Oct  4 20:13:00 carroll kernel:   next_to_clean        <96>
Oct  4 20:13:00 carroll kernel: buffer_info[next_to_clean]
Oct  4 20:13:00 carroll kernel:   time_stamp           <11ac2f241>
Oct  4 20:13:00 carroll kernel:   next_to_watch        <97>
Oct  4 20:13:00 carroll kernel:   jiffies              <11ac2fa68>
Oct  4 20:13:00 carroll kernel:   desc.status          <1780000>


Here is our configuration:

# ethtool -i eth0
driver: igb
version: 3.0.6-k2-2
firmware-version: 1.4-3
bus-info: 0000:01:00.0

# uname -r
2.6.18-308.13.1.el5

Our configuration in detail is attached to this bug
Comment 1 Alan 2012-10-08 14:58:20 UTC 
RHEL bugs should be reported to the vendor. The kernel is both too old and too different for upstream to care about
Note You need to log in before you can comment on or make changes to this bug.