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 1358853 - Provide vfio no-iommu through FD channel for NFV non-support
Summary: Provide vfio no-iommu through FD channel for NFV non-support
Keywords:
Status: CLOSED CURRENTRELEASE
Alias: None
Product: Red Hat Enterprise Linux 7
Classification: Red Hat
Component: Driver Update Disk
Version: 7.2
Hardware: Unspecified
OS: Unspecified
unspecified
unspecified
Target Milestone: rc
: ---
Assignee: Petr Oros
QA Contact: Virtualization Bugs
URL:
Whiteboard:
Depends On: 1299662
Blocks:
TreeView+ depends on / blocked
 
Reported: 2016-07-21 15:30 UTC by Alex Williamson
Modified: 2016-10-18 14:01 UTC (History)
11 users (show)

Fixed In Version: kmod-vfio-0.3-3.el7fdb.x86_64.rpm
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
Environment:
Last Closed: 2016-10-18 14:01:17 UTC
Target Upstream Version:
Embargoed:


Attachments (Terms of Use)

Description Alex Williamson 2016-07-21 15:30:44 UTC
Description of problem:

A request has been made to make vfio no-iommu mode kernel modules available to NFV partners on RHEL7.2 through the "FD" channel.  This mode currently exists in RHEL7.3 via bug 1299662.  Importantly, this no-iommu mode is not supported and taints the kernel when used.  The justification for providing it at all is that it's more detectable when it's used due to the kernel tainting and it promotes the vfio device interface which provides additional features, such as MSI/X interrupt support vs the UIO driver.  However, vfio is a userspace driver interface which can only be considered secure when operating in conjunction with the protection of an iommu.  This unsupported no-iommu mode is meant to bridge the gap for physical systems without an iommu as well as virtual machines where the interaction of a virtual iommu with a physical iommu is not sufficiently advanced to provide the necessary protection.

RHEL7.2 already includes vfio support in the base kernel, though of course not with this requested feature.  The driver update would therefore need to replace the kernel delivered modules with these updates.  Somehow these modules should only be approved for customer and partner use for this specific feature and not for general KVM device assignment as the driver update will not be qualified beyond the no-iommu use case.

Version-Release number of selected component (if applicable):
Latest rhel7.2.z kernel

How reproducible:


Steps to Reproduce:
1.
2.
3.

Actual results:


Expected results:


Additional info:

Comment 2 Anita Tragler 2016-07-24 19:42:32 UTC
The goal is to get VFIO no-iommu to NFV vendors and partners so that they can validate DPDK VNF using RHEL for guest. Need to provide VFIO no-iommu kernel modules via Fast Datapath Beta channel till RHEL 7.3  becomes available.

VFIO no-iommu is not supported since it  taints the guest kernel but provides more protection than *UIO* drivers.

RHEL host already supports VFIO with IOMMU providing secure DMA access for DPDK usecases


For DPDK guest, eventually when vIOMMU (worked on upstream) is available then RHEL guests can use VFIO vIOMMU which would provide secure DMA access

Comment 12 Whitney Chadwick 2016-09-14 21:04:38 UTC
set need info flags for me and Anita to +


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