lkml.org 
[lkml]   [1998]   [Oct]   [12]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
SubjectRe: TCP bad checksum with 2.1.125
Date
On Mon, 12 Oct 1998, Mike A. Harris wrote:

> I just got the following by typing "dmesg".
>
> TCPv4 bad checksum from 209.133.83.18:0050 to
> 206.248.78.137:0431, len=632/632/652
> TCPv4 bad checksum from 199.183.24.237:0050 to
> 206.248.78.137:0445, len=1480/1480/1500
>
>
> I don't know if this is a bug, or just an informative message, so
> I leave it to you all to decide.
>
> I'm using 2.1.125 compiled for uniprocessor on an AMD-K6-200.
>
> If any other info is needed, let me know.

I'm having the same problem, and I think it must be a bug because I
saw my first corrupted ftp transfer ever.

The problem only occurs when connecting to a customers "new" server,
a Digital 3000R/SMP. The NIC is A DE434/5:

eth0: DE434/5 at 0x7400 (PCI bus 0, device 10), h/w address 08:00:2b:e6:a5:14,
and requires IRQ15 (provided by PCI BIOS).
de4x5.c:V0.542 1998/9/15 davies@maniac.ultranet.com

I have made some tests with 2 local machines which both have 3c509 NIC's.

local 3000R
2.1.124 <-> 2.1.125 TCPv4 bad checksum
2.1.125 <-> 2.1.125-SMP TCPv4 bad checksum
2.1.125 <-> 2.1.125-UP TCPv4 bad checksum
2.0.33 <-> 2.1.125-SMP OK
2.0.33 <-> 2.1.125-UP OK

local1 local2
2.1.125 <-> 2.1.125 OK

I suspect it has something to do with changes made to
linux/drivers/net/de4x5.c in patch-2.1.125.gz?

But why does the errors not show up when connecting from 2.0.33?

Can you see the same pattern?


Peter L. Hansen



-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 13:44    [W:0.047 / U:0.132 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site