Re: HEADS UP: call for nve(4) users to test a patch

Andrew Gallatin wrote:
> 
> Maxime Henrion writes:
>  > This patch just reduces the size of the TX ring by one.  Many NIC chips
>  > in existence today have such bugs and require similar fixes, so I'm not
> 
> I'm sorry, but it does not work for me on my Nforce4 based
> motherboard running FreeBSD/amd64:
> 
> nve0: <NVIDIA nForce MCP9 Networking Adapter> port 0xb400-0xb407 mem 0xfebf9000-0xfebf9fff irq 22 at device 10.0 on pci0
> nve0: Ethernet address 00:01:29:f5:6b:91
> miibus1: <MII bus> on nve0
> ciphy0: <Cicada CS8201 10/100/1000TX PHY> on miibus1
> ciphy0:  10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, 1000baseT, 1000baseT-FDX, auto
> nve0: Ethernet address: 00:01:29:f5:6b:91
> nve0: [GIANT-LOCKED]
> nve0: link state changed to DOWN
> nve0: link state changed to UP
> nve0: device timeout (2)

Ok, I should have said that this patch will only help if you get a
"device timeout (64)" message, so it won't help here unfortunately.

> I'm totally unable to send or receive any traffic under FreeBSD with this
> nic.  It doesn't randomly timeout, it never manages to transmit (or receive)
> anything.
> 
> I can provide more details about this board upon request..
> 
> 
>  > really surprised.  It also seems Linux's forcedeth driver does such a
>  > thing, but it's hard to tell because it uses an entirely different API
>  > than us.
> 
> The Solaris driver (from http://homepage2.nifty.com/mrym3/taiyodo/eng) uses the
> same interface as we do, and uses the same version of the nvidia libs.  It
> has been bulletproof for me so far on this same machine when it is booted
> into Solaris.

I'll look into this ASAP.

Thanks,
Maxime