tech-net archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: CRC errors with gem(4)




To: NetBSD tech-net mailing list <tech-net%netbsd.org@localhost>

Subject: Re: CRC errors with gem(4)

From: Julian Coleman <jdc%coris.org.uk@localhost>

Date: Sat, 5 Jan 2008 20:47:22 +0000


Hi,

> If I understand you correctly, the implication here is that the
> bytes are being transmitted corrupt.

Yes.

> At what packet rate (pps) do you start to see problems?

Sending 10 16k UDP frames with ttcp shows up the problem, so this at most
110pps.

> Is hardware checksum enabled?
> If so, does disabling it improve matters?

No.  No.  The card doesn't really support UDP checksums, so I've disabled it.

> Are there any comments/workarounds in opensolaris code
> for a problem that resembles this?

Unfortunately not.

One thing I tried was to increase the size of the TX descriptor ring.  This
made the problem disappear for 10 frames, but it's still there at 100 frames.

Thanks,

J

-- 
  My other computer also runs NetBSD    /        Sailing at Newbiggin
        http://www.netbsd.org/        /   http://www.newbigginsailingclub.org/



References:

CRC errors with gem(4)
From: Julian Coleman

Re: CRC errors with gem(4)
From: Julian Coleman

Re: CRC errors with gem(4)
From: Darren Reed




Prev by Date: please test: ethernet media-handling patch

Next by Date: LW-IGMPv3 implementation

Previous by Thread: Re: CRC errors with gem(4)

Next by Thread: What is the state of TCP offload support in the kernel?

Indexes:

reverse Date

reverse Thread

Old Index



Home | Main Index | Thread Index | Old Index