Subject: Re: gigabit ethernet cards
To: None <diro@nixsys.bz>
From: Greg A. Woods <woods@weird.com>
List: port-alpha
Date: 12/05/2006 18:29:00
At Sun, 3 Dec 2006 22:44:11 -0500,
diro@nixsys.bz wrote:
> 
> What models have you tested with NetBSD/alpha?

We did a fair bit of testing of various cards before we settled on the
dual-port Intel one shown below.  Some of the saga was posted to
port-alpha.

Note this is with an ES40 and we're running a derivative of 1.6.2_STABLE.

We can apparently only use one port successfully at a time, but that
might be a bug in either the card or the driver.  Search for the details
in the archives.

wm0 at pci1 dev 3 function 0: Intel i82546GB 1000BASE-X Ethernet, rev. 3
wm0: interrupting at dec 6600 irq 32
wm0: Ethernet address 00:04:23:a8:79:28
wm0: 1000baseSX, 1000baseSX-FDX, auto
wm1 at pci1 dev 3 function 1: Intel i82546GB 1000BASE-X Ethernet, rev. 3
wm1: interrupting at dec 6600 irq 33
wm1: Ethernet address 00:04:23:a8:79:28
wm1: 1000baseSX, 1000baseSX-FDX, auto


We are running it with all its checksum offloading capabilities enabled:

wm0: flags=8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
        capabilities=7<IP4CSUM,TCP4CSUM,UDP4CSUM>
        enabled=7<IP4CSUM,TCP4CSUM,UDP4CSUM>
        address: 00:04:23:a8:79:28
        media: Ethernet autoselect (1000baseSX full-duplex)
        status: active



The following card doesn't work, though if it could be made to work then
there was some idea that it might have a slightly better profile under
high small-packet loads but I can't say for sure as I couldn't test it.

Broadcom BCM5703X 10/100/1000 Ethernet (ethernet network, revision 0x02) at pci1 dev 6 function 0 not configured

-- 
						Greg A. Woods

H:+1 416 218-0098 W:+1 416 489-5852 x122 VE3TCP RoboHack <woods@robohack.ca>
Planix, Inc. <woods@planix.com>       Secrets of the Weird <woods@weird.com>