[Bug 235031] [em] em0: poor NFS performance, strange behavior

[bugzilla-noreply at freebsd.org]

https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=235031

--- Comment #17 from Martin Birgmeier <d8zNeCFG at aon.at> ---
Because bug #234550 might be related to this one, I tried to issue "ifconfig
-lro em0". This resulted in the interface not working anymore, with the already
known syslog messages:

Jan 20 09:29:18 mizar kernel: em0: link state changed to DOWN
Jan 20 09:29:20 mizar kernel: em0: TX(0) desc avail = 1024, pidx = 0
Jan 20 09:34:36 mizar kernel: em0: TX(0) desc avail = 1024, pidx = 0
Jan 20 09:34:42 mizar kernel: em0: TX(0) desc avail = 1024, pidx = 0
Jan 20 09:34:51 mizar kernel: em0: TX(0) desc avail = 1024, pidx = 0
Jan 20 09:34:54 mizar kernel: em0: TX(0) desc avail = 1024, pidx = 0

None of the following commands (from the shell history) helped:

 4981  2019-01-20 09:29  0:01  ifconfig em0 -lro
 4982  2019-01-20 09:29  0:00  ifconfig em0 up
 4985  2019-01-20 09:30  0:00  ifconfig em0
 4987  2019-01-20 09:30  0:03  ifconfig em0 lro up
 4989  2019-01-20 09:31  0:00  ifconfig em0
 4990  2019-01-20 09:32  0:00  ifconfig em0 down
 4991  2019-01-20 09:32  0:02  ifconfig em0 up
 4999  2019-01-20 09:40  0:00  ifconfig em0

Then I tried adding "-lro" in rc.conf:

ifconfig_em0="-lro inet $hostname"

This resulted in a rather long wait for the interface to be available to some
rc processes. When the machine was finally up, NFS to it did not work at all,
it just hung.

I removed the -lro flag again.

What is the issue that using ifconfig on this interface breaks it?

-- Martin

-- 
You are receiving this mail because:
You are the assignee for the bug.