Bug 229433 - need to update forcedeth driver for MCP55 platforms
Summary: need to update forcedeth driver for MCP55 platforms
Status: CLOSED DUPLICATE of bug 197797
Alias: None
Product: Red Hat Enterprise Linux 4
Classification: Red Hat
Component: kernel   
(Show other bugs)
Version: 4.4
Hardware: x86_64 Linux
Target Milestone: ---
: ---
Assignee: Andy Gospodarek
QA Contact: Brian Brock
Depends On:
TreeView+ depends on / blocked
Reported: 2007-02-21 00:57 UTC by Hans-Helmar Althaus
Modified: 2014-06-29 22:58 UTC (History)
6 users (show)

Fixed In Version:
Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of:
Last Closed: 2007-03-14 18:40:18 UTC
Type: ---
Regression: ---
Mount Type: ---
Documentation: ---
Verified Versions:
Category: ---
oVirt Team: ---
RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: ---

Attachments (Terms of Use)

Description Hans-Helmar Althaus 2007-02-21 00:57:00 UTC
Description of problem:

the forcedeth driver you are using in rhel4 fails to
setup link on most gigabit switches

Version-Release number of selected component (if applicable):

2.6.9-42 with forcedeth 0.41

How reproducible:

Connect for example a hp xw9400 on a hp procurve switch
Actual results:

no traffic

Expected results:

normal communication

Additional info:

the link light will go on, but the
RX counter remains on zero

the current driver aviable from nvidia
or plain vamilla kernel works.

Comment 1 Andy Gospodarek 2007-03-12 21:37:15 UTC
I have an xw9300 and tested it today without any problems.  Can you send output
from 'ethtool ethX' for that interface?

A sysreport might be nice as well, but we can start with the ethtool output if
you like.

Comment 2 Hans-Helmar Althaus 2007-03-13 07:51:02 UTC
I am sorry i have installed the driver provided from nvidia because we need this
machine working. With this driver or attached on a 100 MBit ethernet i did never
see any problems for now. The problem for me is that i have to hack the
redhat.initrd for pxe-installation of this workstation (i have to unpack the
initrd and to replace forcedeth module). The problem is known to nvidia and hp
(xw9400 and maybe also fujtsu-siemens v840) - please ask anybody there if you
dont trust me, i can provide you a contact person if you like.

The output of ethtool (eth0 is a nvidia interface but the driver is the on from
nvidia webside, forcedeth 0.59 !! ):

[root@fsc-amd-2 ~]# ethtool eth0
Settings for eth0:
        Supported ports: [ MII ]
        Supported link modes:   10baseT/Half 10baseT/Full 
                                100baseT/Half 100baseT/Full 
        Supports auto-negotiation: Yes
        Advertised link modes:  10baseT/Half 10baseT/Full 
                                100baseT/Half 100baseT/Full 
        Advertised auto-negotiation: Yes
        Speed: 1000Mb/s
        Duplex: Full
        Port: MII
        PHYAD: 0
        Transceiver: externel
        Auto-negotiation: on
        Supports Wake-on: g
        Wake-on: d
        Link detected: yes

Comment 3 Andy Gospodarek 2007-03-13 13:05:56 UTC
Thanks for the feedback.  Please send me your contact person at HP, so I can get
more detail about which switches have this problem.

I have not heard complaints about the xw9400 and forcedeth on 4.4 and I've not
personally experienced and problems with my xw9400 and an unmanaged D-Link
gigabit switch, so I would like to understand what devices won't link up with
this hardware.

Comment 4 Andy Gospodarek 2007-03-14 18:40:18 UTC
Good news.  I played around some more and noticed that eth0 does have problems
linking up at gigabit speeds and confirmed that this is fixed in our latest
development kernels (though I did find another problem).  

Closing this as a duplicate of BZ 197797.

*** This bug has been marked as a duplicate of 197797 ***

Note You need to log in before you can comment on or make changes to this bug.