** Description changed:
Virtual Machines with SR-IOV VFs from an Intel E810-XXV [8086:159b] get
no traffic flow and produce error messages in both the host and guest
during network configuration.
Environment: Ubuntu OpenStack Focal-Ussuri with OVN
Host Kernel: v5.15.0-41-generic
** Description changed:
Virtual Machines with SR-IOV VFs from an Intel E810-XXV [8086:159b] get
no traffic flow and produce error messages in both the host and guest
during network configuration.
Environment: Ubuntu OpenStack Focal-Ussuri with OVN
Host Kernel: v5.15.0-41-generic
For users of i40e, seems this depends on the following host patch where the
max_mtu from the host PF that is now used was not set on i40e (but was for ice).
https://github.com/torvalds/linux/commit/372539def2824c43b6afe2403045b140f65c5acc
It's also queued for stable. We may want to pull this at
The patch has now made it upstream into the v6.0-rc7 release:
https://github.com/torvalds/linux/commit/399c98c4dc50b7eb7e9f24da7ffdda6f025676ef
It has also been pushed to the stable-queue for the upstream 5.4 and 5.15
kernels:
https://www.spinics.net/lists/stable-commits/msg266383.html
Been working with Trent and Intel, and below [1] is the patch for the
MTU fix
My test kernels are available in PPA [2] for both 5.4 and 5.15
[1]
I have created a test kernel with the patches listed in comment #4 and
comment #6. It does not include the mtu fix from comment #5
https://people.canonical.com/~mreed/misc/lp_1983656_e810_sriov/
--
You received this bug notification because you are a member of Kernel
Packages, which is
I may have missed this one... I meant to add it if it's not already
picked up in a different commit:
ice: Fix not stopping Tx queues for VFs
https://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next.git/commit/drivers/net/ethernet/intel/ice?id=b385cca47363316c6d9a74ae9db407bbc281f815
--
I was unable to locate the patch from comment #5 in linux-next or
linus's tree as of yet. This patch may require a separate PR.
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs.launchpad.net/bugs/1983656
The following patches are already applied to the jammy master-next kernel
(Ubuntu-5.15.0-48.54-2-gd250073d5172)
5951a2b9812d v5.16-rc6 iavf: Fix VLAN feature flags after VFR
e6ba5273d4ed v5.16-rc6 ice: Fix race conditions between virtchnl handling and
VF ndo ops
b385cca47363 v5.16-rc6 ice: Fix
additional patch to pick up that is supposed to help address the issue
with MTUs
ice: Fix interface being down after reset with link-down-on-close flag on
Update from upstream:
Please check out the other patches below that were sent together with the one
patch I listed in my previous reply. Cherry-picking patches can be challenging
with ice driver from later kernels as new features and bug fixes are sent
continuously in each kernel and the
Updaet: We found that this bug was only triggering when you try to
change the MTU to anything other than 1500. With MTU=1500 the VMs work
fine even with the current 5.15 kernel.
With MTU=1550 or MTU=9000 it fails even with the upstream 5.19 kernel
and these patches. So seems we need to report and
Also, this will be resolved in Kinetic via 5.19 and Jammy 22.04.2 via
HWE.
** Tags added: servcert-472
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs.launchpad.net/bugs/1983656
Title:
SR-IOV VFs no
Hi Trent,
What is the expectation here? To update 5.15? Or also 5.4?
** Changed in: linux (Ubuntu)
Status: Confirmed => Incomplete
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
14 matches
Mail list logo