RHEL Engineering is moving the tracking of its product development work on RHEL 6 through RHEL 9 to Red Hat Jira (issues.redhat.com). If you're a Red Hat customer, please continue to file support cases via the Red Hat customer portal. If you're not, please head to the "RHEL project" in Red Hat Jira and file new tickets here. Individual Bugzilla bugs in the statuses "NEW", "ASSIGNED", and "POST" are being migrated throughout September 2023. Bugs of Red Hat partners with an assigned Engineering Partner Manager (EPM) are migrated in late September as per pre-agreed dates. Bugs against components "kernel", "kernel-rt", and "kpatch" are only migrated if still in "NEW" or "ASSIGNED". If you cannot log in to RH Jira, please consult article #7032570. That failing, please send an e-mail to the RH Jira admins at rh-issues@redhat.com to troubleshoot your issue as a user management inquiry. The email creates a ServiceNow ticket with Red Hat. Individual Bugzilla bugs that are migrated will be moved to status "CLOSED", resolution "MIGRATED", and set with "MigratedToJIRA" in "Keywords". The link to the successor Jira issue will be found under "Links", have a little "two-footprint" icon next to it, and direct you to the "RHEL project" in Red Hat Jira (issue links are of type "https://issues.redhat.com/browse/RHEL-XXXX", where "X" is a digit). This same link will be available in a blue banner at the top of the page informing you that that bug has been migrated.
Bug 463183 - [LTC 6.0 FEAT] 201700:Support Intel 10Gb Ethernet (Knox) with MSI-X interrupt enabled.
Summary: [LTC 6.0 FEAT] 201700:Support Intel 10Gb Ethernet (Knox) with MSI-X interrupt...
Keywords:
Status: CLOSED CURRENTRELEASE
Alias: None
Product: Red Hat Enterprise Linux 6
Classification: Red Hat
Component: kernel
Version: 6.0
Hardware: ppc64
OS: All
high
high
Target Milestone: alpha
: 6.0
Assignee: Ameet Paranjape
QA Contact: Martin Jenner
URL:
Whiteboard:
Depends On:
Blocks: 356741
TreeView+ depends on / blocked
 
Reported: 2008-09-22 15:10 UTC by IBM Bug Proxy
Modified: 2010-10-18 14:26 UTC (History)
7 users (show)

Fixed In Version:
Doc Type: Enhancement
Doc Text:
Clone Of:
Environment:
Last Closed: 2009-11-12 23:15:46 UTC
Target Upstream Version:
Embargoed:

Attachments (Terms of Use)

Description IBM Bug Proxy 2008-09-22 15:10:57 UTC 
=Comment: #0=================================================
Emily J. Ratliff <emilyr.com> - 2008-09-16 18:22 EDT
1. Feature Overview:
Feature Id:	[201700]
a. Name of Feature:	Support Intel 10Gb Ethernet (Knox) with MSI-X interrupt enabled.
b. Feature Description
Update the ixgbe driver in order to enable the MSI-X interrupt support for the Intel-based 10Gb
PCI-E adapter on p6 servers.

2. Feature Details:
Sponsor:	PPC
Architectures:
ppc64

Arch Specificity: Purely Common Code
Affects Kernel Modules: Yes
Delivery Mechanism: Direct from community
Category:	Device Drivers and IO
Request Type:	Driver - Update Version
d. Upstream Acceptance:	In Progress
Sponsor Priority	1
f. Severity: High
IBM Confidential:	no
Code Contribution:	3rd party code
g. Component Version Target:	TBD

3. Business Case
The MSI-X interrupt provides better performance and throughput out of the high bandwidth (10Gb)
networking device.

4. Primary contact at Red Hat: 
John Jarvis
jjarvis

5. Primary contacts at Partner:
Project Management Contact:
Stephanie Glass, sglass.com, 512-838-9284

Technical contact(s):
Daisy Chang, daisyc.com
Paul Larson, plars.com

IBM Manager:
Wendel Voigt, wvoigt.com
Comment 1 Bill Nottingham 2008-10-01 20:38:32 UTC 
What kernel version is this currently targeted for release in?
Comment 2 IBM Bug Proxy 2009-03-03 16:02:27 UTC 
The ixgbe driver in the kernel already has msi-x, and Breno tested this with linux-next that should go in (with kernel msi-x support patches) to 2.6.30 and found no problems.
Comment 3 Bill Nottingham 2009-03-03 22:13:16 UTC 
OK, marking as MODIFIED.

The feature requested has already been accepted into the upstream code base
planned for the next major release of Red Hat Enterprise Linux.

When the next milestone release of Red Hat Enterprise Linux 6 is available,
please verify that the feature requested is present and functioning as
desired.
Comment 4 IBM Bug Proxy 2009-03-04 12:41:19 UTC 
Moving feature to Accepted state since it is in RH cvs tree.
Comment 5 Kevin W Monroe 2009-11-12 23:15:46 UTC 
Closing as support is available in current release.
Comment 6 IBM Bug Proxy 2010-05-07 19:01:58 UTC 
------- Comment From leitao.ibm.com 2010-05-07 14:59 EDT-------
I just had chance to verify this feature, and MSI-X is not enabled on ixgbe.
It seems that only MSI is running on ixgbe.

Capabilities: [50] MSI: Enable+ Count=1/1 Maskable- 64bit+
Capabilities: [60] MSI-X: Enable- Count=18 Masked-

Adding Andr?? as the owner fo this bug, since Paul left IBM.
Comment 7 IBM Bug Proxy 2010-05-10 23:01:14 UTC 
------- Comment From adetsch.com 2010-05-10 18:51 EDT-------
This patch made MSI-X stop to work in ixgbe + powerpc.
http://git.kernel.org/?p=linux/kernel/git/davem/net-2.6.git;a=commitdiff;h=342bde1b70c79bfc8509b017b3987f3c7541ff8e

After this change, we request 3 msi-x vectors, but pci_enable_msix returns only 2, which is less than MIN_MSIX_COUNT (== 3), and leads to a MSI fallback.

Trying vectors = 4,  pci_enable_msix returns the requested number, 4, and MSI-X works fine.

In summary: the commit seems to expose a problem with the pci_enable_msix logic in powerpc.

I'll keep working on that.
Comment 8 IBM Bug Proxy 2010-05-10 23:50:39 UTC 
------- Comment From adetsch.com 2010-05-10 19:40 EDT-------
Enabling debug on rtas msi code, this is the point where we fail:
rtas_msi: ibm,change_msi(func=4,num=3), got 2 rc = 2
rtas_msi: rtas_change_msi() failed

"num=3" corresponds to the msi-x vector size. "got 2" means that the rtas call returned only 2 irqs and not 3.
Comment 9 IBM Bug Proxy 2010-05-21 17:30:54 UTC 
------- Comment From adetsch.com 2010-05-21 13:20 EDT-------
Feature verified.
There was a recent "regression" in the driver, that made MSI-X not be enabled on machine with 2 or less cpu cores. And this problem is present in the version in RHEL6.
Not a driver regression actually, but a driver change that exposed a firmware limitation. This is still yet to be handled, and will be the subject of another bug.
Note You need to log in before you can comment on or make changes to this bug.