Bug 108213

Summary: LTC5169 - ethtool -S not working with the e1000 driver
Product: Red Hat Enterprise Linux 3 Reporter: Olof Johansson <olof>
Component: kernelAssignee: Julie DeWandel <jdewand>
Status: CLOSED CURRENTRELEASE QA Contact: Brian Brock <bbrock>
Severity: medium Docs Contact:
Priority: medium    
Version: 3.0CC: msw, petrides, riel
Target Milestone: ---   
Target Release: ---   
Hardware: powerpc   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2004-01-19 19:04:35 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
Patch for ethtool -S <netdev> against RHEL 3 released product.
none
A second way to solve this problem none

Description Olof Johansson 2003-10-28 17:12:17 UTC
From Bugzilla Helper:
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.3.1) Gecko/20030425

Description of problem:
p630
1 Goliad adapter

Software Environment:
RHEL3

Steps to Reproduce:
1.run ethtool -S eth1

Actual Results:
it will return:
Operation not supported.

Expected Results:
Should give all the Intel Gigabit adapter statistis. The e1000 driver supports
it, the problem is withShould give all the Intel Gigabit adapter statistis. 

Ethtool needs to get updated to support "-S" for the e1000 driver.

The problem in SLES8 was that both ETHTOOL_GSTRINGS and ETHTOOL_GSTATS were not
mapped by arch/ppc64/kernel/ioctl32.c::ethtool_ioctl()

They need to check these on RedHat.

------- Additional Comment #4 From Olof Johansson 2003-10-28 12:11 -------

This has been fixed in ameslab BK, so RedHat can pick it up from there. I'll
open up a mirror bug.



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


How reproducible:
Always

Steps to Reproduce:
See above  

Additional info:

Comment 1 Julie DeWandel 2003-10-29 20:15:55 UTC
I will attach a proposed patch to the RHEL 3 sources to correct this problem. If
you guys (IBM) get a chance to test it, it would be appreciated. I have verified
it here.

Comment 2 Julie DeWandel 2003-10-29 20:19:40 UTC
Created attachment 95588 [details]
Patch for ethtool -S <netdev> against RHEL 3 released product.

The problem was only a problem for ethtool compiled as a 32-bit application.
Worked fine as a 64-bit app.

Comment 3 Olof Johansson 2003-10-31 02:23:46 UTC
------- Additional Comments From olof.com  2003-30-10 21:21 -------
Rick, can you test the patch RH provided? They have modified the ioctl layers,
so the BK patch doesn't apply to their tree.

Comment 4 Olof Johansson 2003-10-31 02:40:09 UTC
------- Additional Comments From olof.com  2003-30-10 21:21 -------
Rick, can you test the patch RH provided? They have modified the ioctl layers,
so the BK patch doesn't apply to their tree.

Comment 5 Olof Johansson 2003-10-31 19:53:17 UTC
ricardoz.com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |FIXEDAWAITINGTEST
         Resolution|                            |FIX_BY_DISTRO




------- Additional Comments From ricardoz.com  2003-31-10 13:07 -------
I just tested the patch and it works now. Defect has been fixed.

Comment 6 Olof Johansson 2003-10-31 19:53:43 UTC
------- Additional Comments From ricardoz.com  2003-31-10 13:09 -------
When would we expect this patch to be in the RHEL3 distro. ?

Comment 7 Julie DeWandel 2003-10-31 20:06:12 UTC
Thank you for testing this patch. A suggestion was made from a kernel team
member for an improved version which I'd like to explore. My plan is to try the
second version and post it to here if all goes well. Either way, I expect this
issue to be resolved within the coming week.

But to answer your question, this problem will be fixed in Update 1 for RHEL 3.

Comment 8 Julie DeWandel 2003-10-31 22:01:19 UTC
Created attachment 95637 [details]
A second way to solve this problem

This patch creates a much simpler ethtool_ioctl interface for 32-bit apps. All
ioctl commands are supported this way (both now and with future drivers). This
code borrow heavily from the 2.5 code. A patch review by IBM would be
appreciated as would testing of this patch. It has been tested here and appears
to work, but futher verification is always nice.

Comment 9 Olof Johansson 2003-10-31 22:42:55 UTC
------- Additional Comments From olof.com  2003-31-10 17:44 -------
Julie, appropriate review method would be to post on linuxppc64-dev. Thanks.

Comment 10 Julie DeWandel 2003-11-06 14:45:55 UTC
The second patch was submitted on 05-OCT-03 and should appear in Update 1.

Comment 11 Julie DeWandel 2004-01-19 19:04:35 UTC
Closing on our side after verifying with IBM.