On Mon, Jan 30 2006, Jens Axboe wrote:
> > and, if you want to, please try this patch:
> > 
> > e1000: add 82573 to TSO workaround code
> > 
> > After shipment, it was discovered that the 82571/2 workaround for TSO is
> > needed for the 82573 as well.  This code slightly rearchitects the
> > workaround code to a more clean implementation for multiple adapters.
> 
> I'll give it a go right away. BTW, I tried both with the disable packet
> split option set and unset, made no difference.

No go, same thing. Triggered it after 1 minute of uptime just scp'ing a
file over.

If this is a hardware bug, why am I only seeing it now all of a sudden?
I'll try with e1000 drivers from an older kernel on this box now, just
to be sure. Starting with 2.6.14, although I'm pretty certain that
2.6.15 worked flawlessly as well.

e1000: eth0: e1000_watchdog_task: NIC Link is Up 1000 Mbps Full Duplex
e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
  Tx Queue             <0>
  TDH                  <a3>
  TDT                  <b1>
  next_to_use          <b1>
  next_to_clean        <a1>
buffer_info[next_to_clean]
  time_stamp           <fffd6bb3>
  next_to_watch        <a3>
  jiffies              <fffd7435>
  next_to_watch.status <0>
e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
  Tx Queue             <0>
  TDH                  <a3>
  TDT                  <b1>
  next_to_use          <b1>
  next_to_clean        <a1>
buffer_info[next_to_clean]
  time_stamp           <fffd6bb3>
  next_to_watch        <a3>
  jiffies              <fffd7c05>
  next_to_watch.status <0>
e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
  Tx Queue             <0>
  TDH                  <a3>
  TDT                  <b1>
  next_to_use          <b1>
  next_to_clean        <a1>
buffer_info[next_to_clean]
  time_stamp           <fffd6bb3>
  next_to_watch        <a3>
  jiffies              <fffd83d5>
  next_to_watch.status <0>
e1000: eth0: e1000_clean_tx_irq: Detected Tx Unit Hang
  Tx Queue             <0>
  TDH                  <a3>
  TDT                  <b1>
  next_to_use          <b1>
  next_to_clean        <a1>
buffer_info[next_to_clean]
  time_stamp           <fffd6bb3>
  next_to_watch        <a3>
  jiffies              <fffd8ba5>
  next_to_watch.status <0>
NETDEV WATCHDOG: eth0: transmit timed out
e1000: eth0: e1000_watchdog_task: NIC Link is Up 1000 Mbps Full Duplex

-- 
Jens Axboe

-
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Reply via email to