[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 #2 from Martin Birgmeier <d8zNeCFG at aon.at> ---
Another scenario
================

Just for experimenting, I issued the following commands:

ifconfig em0 -rxcsum .............. result: connectivity to other machines lost
for existing connections (TCP and UDP), new connections can be established
ifconfig em0 rxcsum ............... result: same
ifconfig em0 down ................. result: same
ifocnfig em0 up ................... result: same

In addition, these actions resulted in the following message being repeated on
the console every few seconds:

TX(0) desc avail = 1024, pidx = 0

I had to reboot the machine to get it working normally again.

[0]# grep em0 /var/log/messages
...
Jan 17 21:29:18 mizar kernel: em0: link state changed to DOWN
Jan 17 21:29:20 mizar kernel: em0: TX(0) desc avail = 1024, pidx = 0
Jan 17 21:30:02 mizar kernel: em0: TX(0) desc avail = 1024, pidx = 0
Jan 17 21:36:14 mizar kernel: em0: TX(0) desc avail = 1024, pidx = 0
Jan 17 21:36:22 mizar kernel: em0: TX(0) desc avail = 1024, pidx = 0
Jan 17 21:36:25 mizar kernel: em0: TX(0) desc avail = 1024, pidx = 0
Jan 17 21:36:28 mizar kernel: em0: TX(0) desc avail = 1024, pidx = 0
Jan 17 21:36:31 mizar kernel: em0: TX(0) desc avail = 1024, pidx = 0
Jan 17 21:37:59 mizar kernel: Syncing disks, vnodes remaining... 11 12 em0:
TX(0) desc avail = 1024, pidx = 0
Jan 17 21:37:59 mizar kernel: 8 8 em0: TX(0) desc avail = 1024, pidx = 0
Jan 17 21:37:59 mizar kernel: 8 8 8 em0: TX(0) desc avail = 1024, pidx = 0
Jan 17 21:37:59 mizar kernel: 8 2 2 em0: TX(0) desc avail = 1024, pidx = 0
Jan 17 21:37:59 mizar kernel: 1 0 0 em0: TX(0) desc avail = 1024, pidx = 0
Jan 17 21:37:59 mizar kernel: 0 0 0 em0: TX(0) desc avail = 1024, pidx = 0
Jan 17 21:37:59 mizar kernel: Waiting (max 60 seconds) for system thread
`bufspacedaemon-0' to stop... em0: TX(0) desc avail = 1024, pidx = 0
Jan 17 21:37:59 mizar kernel: Waiting (max 60 seconds) for system thread
`bufspacedaemon-6' to stop... em0: TX(0) desc avail = 1024, pidx = 0
Jan 17 21:37:59 mizar kernel: em0: <Intel(R) PRO/1000 Network Connection> port
0x6040-0x605f mem 0xd5300000-0xd531ffff,0xd532b000-0xd532bfff irq 20 at device
25.0 on pci0
Jan 17 21:37:59 mizar kernel: em0: attach_pre capping queues at 1
Jan 17 21:37:59 mizar kernel: em0: using 1024 tx descriptors and 1024 rx
descriptors
Jan 17 21:37:59 mizar kernel: em0: msix_init qsets capped at 1
Jan 17 21:37:59 mizar kernel: em0: Unable to map MSIX table 
Jan 17 21:37:59 mizar kernel: em0: Using an MSI interrupt
Jan 17 21:37:59 mizar kernel: em0: allocated for 1 tx_queues
Jan 17 21:37:59 mizar kernel: em0: allocated for 1 rx_queues
Jan 17 21:37:59 mizar kernel: em0: Ethernet address: f0:de:f1:98:86:a9
Jan 17 21:37:59 mizar kernel: em0: link state changed to UP
[0]#

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