em(4) interface hangs under 8.0-RELEASE

[Jeremy Chadwick]

On Sat, Mar 06, 2010 at 12:08:22PM -0800, Nick Rogers wrote:
> Yes, this was the first em(4) problem I ran into when upgrading from
> 7.2-RELEASE to 8.0-RELEASE. Yourself and others on another thread eventually
> recommended turning off TSO and what not. I never had a chance to thoroughly
> test this solution on this particular hardware because we had already
> switched to a different set of interfaces (on-motherboard bge(4)). We also
> had that ALTQ problem popup on em which I'm sure you remember, which
> prevented me from going back to the em interfaces for a while.

I've re-read your ALTQ post (Subject "em(4) + ALTQ broken") and I
noticed you didn't provide any details regarding *how* you're using ALTQ
on your systems (specifically, no pf.conf directives provided).

We use pf and ALTQ on all our RELENG_7 and RELENG_8 systems, exclusively
using em(4), without any problems.  I should note that we only utilise
the ALTQ pieces of pf.conf on RELENG_7, but ALTQ is included in our
RELENG_8 systems' kernels.  The ALTQ feature we use is "bandwidth" for
rate-limiting certain IPs bound to em(4) interfaces.  We've used this
successfully at both 100mbit and 1000mbit interface rates.

I can provide specific details of the systems (pciconf -lvc from them,
including OS release + etc.) if you'd like to compare.

I'm almost certain the NIC model you use differs from ours, including
the fact that our NICs are PCIe-bound, and do use MSI + TSO + all forms
of checksum offloading.

Why I care: upgrading our RELENG_7 machine which uses ALTQ directives is
on my to-do list, and if this feature is somehow broken under RELENG_8,
I need to know in advance so I can use ipfw + dummynet instead.

-- 
| Jeremy Chadwick                                   jdc at parodius.com |
| Parodius Networking                       http://www.parodius.com/ |
| UNIX Systems Administrator                  Mountain View, CA, USA |
| Making life hard for others since 1977.              PGP: 4BD6C0CB |