From Bugzilla Helper:
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.5b) Gecko/20030827
Description of problem:
Using ethtool, I can not get the broadcom card to set to full duplex. The
command i'm using (which works for me on the other machines) is "ethtool -s eth0
duplex full speed 100 autoneg off"
Version-Release number of selected component (if applicable):
Steps to Reproduce:
2. ethtool -s eth0 duplex full speed 100 autoneg off
3. nothing happens
Actual Results: duplex stayed the same
Expected Results: duplex switch to full
Linux ln3012 2.4.21-3.ELsmp #1 SMP Fri Sep 19 14:06:12 EDT 2003 i686 i686 i386
description: "Broadcom Tigon3 ethernet driver"
author: "David S. Miller (firstname.lastname@example.org) and Jeff Garzik
parm: tg3_debug int, description "Tigon3 bitmapped debugging message
I rebooted it a few more times with no luck, tried differnt options on ethtool,
ethtool -s eth0 autoneg off
wait a second
ethtool -s eth0 duplex full
and it came up. I'm not sure if it came up because I did it like that, or if I
just got lucky. I have it up and running so the developers can continue to test
on it, but I still think its broken.
On that last comment, it booted to the older test kernel, so it seems that the
older one works, but the new kernel doesn't. I just rebooted to the newest test
kernel, and just turning autoneg off, causes me to loose link.
what is tg3 version?
show us # ethtool -i eth0 and # ethtool eth0
I know that Jeff G. doesn't like that people turn off autoneg. Maybe with this
driver release it cannot be disabled
I've also tried ethtool without using autoneg off, but it won't set the card to
ethtool -i eth0
Settings for eth0:
Supported ports: [ MII ]
Supported link modes: 10baseT/Half 10baseT/Full
Supports auto-negotiation: Yes
Advertised link modes: 10baseT/Half 10baseT/Full
Advertised auto-negotiation: Yes
Port: Twisted Pair
Supports Wake-on: g
Current message level: 0x000000ff (255)
Link detected: yes
Amit from Dell has been able to reproduce this bug. It needs to be
fixed in the next errata.
Issue Tracker 29166 opened as sev 2.
Jeff points out that this could have happened because of a change in
ethtool as well as tg3 -- can we check latest ethtool from
http://sf.net/projects/gkernel/ as well as earlier ethtool, perhaps
tested ethtool from RHL9 (1.6-5) and the latest from
http://sf.net/projects/gkernel/ (1.8) -> neither worked
- looks like this is a tg3 issue
just dropped a RHL9 kernel (2.4.20-20.9) on this system and ethtool
starting working properly ...
Created attachment 95742 [details]
Fix for ETHTOOL_SSET bug in tg3 driver.
The attached patch should fix the bug, problems were added to
ETHTOOL_SSET handling during the conversion of the tg3 driver
over to ethtool_ops.
FROM ISSUE TRACKER....
Event posted 11-06-2003 03:30pm by Bhutani with duration of 0.00
First things first - Patch works great!
I am assuming that this fix will be in QU1. I know Red Hat was working
on a "DKMS like" device driver update model. Are there any updates on
that? In the absence of such a mechanism and considering that this is
a critical enough bug to be a high volume call generator, Dell will
have to update this driver in the factory using DKMS. Event posted
11-11-2003 11:13am by ltroan with duration of 0.30 I've appended your
last paragraph to Issue Tracker 28574 and requested comments on our
support for DKMS.
Bhutani assigned to issue for Dell-Engineering.
Status set to: Fix Pending
Event posted 11-18-2003 12:51pm by Bhutani with duration of 0.00
Attaching the DKMS version of tg3 v2.2 driver that will be installed
with Taroon in Factory and DSA. FYI, you will need the latest version
of DKMS from www.lerhaupt.com/dkms before you can install this RPM.
File uploaded: tg3-2.2-1dkms.noarch.rpm
Event posted 11-26-2003 03:26pm by Bhutani with duration of 0.00
Updated version of the tg3 DKMS package that will be Factory installed
for Taroon on all platforms attached. As before you will need the
latest version of DKMS from www.lerhaupt.com/dkms before you can
install this RPM.
File uploaded: tg3-2.2-3dkms.noarch.rpm
It seems that (1) this bug actually is resolved, as marked in
RHSA-2004-017, and that (2) bug #108115 is a duplicate of this bug. Am
I wrong or am I right?
Fixed in RHEL 3 U1.
An errata has been issued which should help the problem
described in this bug report. This report is therefore being
closed with a resolution of ERRATA. For more information
on the solution and/or where to find the updated files,
please follow the link below. You may reopen this bug report
if the solution does not work for you.
*** Bug 108115 has been marked as a duplicate of this bug. ***