** Description changed:
This ticket is an FR to update iproute2 package from 6.1.0 to latest
upstream version.
This updated is needed to support latest mlx5 driver features.
- Missing support examples (iproute v6.7):
- 1)
- ipsec
+ Missing support examples
+ 1) related feature: ipsec
Public bug reported:
This ticket is an FR to update iproute2 package from 6.1.0 to latest
upstream version.
This updated is needed to support latest mlx5 driver features.
Missing support examples (iproute v6.7):
1)
ipsec
https://lore.kernel.org/netdev/20231003180557.GC51282@unreal/T/
994e80e9
Hi,
We tested the feature with 5.15.0-28 from proposed.
The features is working and SMFS mode is enabled for Connectx-7.
No degradation detected comparing 5.15.0-28 from proposed and 5.15.0-28
compiled without the feature patch set.
** Tags removed: verification-needed-jammy
** Tags added:
Hi Dan,
Those test are ASAP which uses steering as its infrastructure thus this
test set when configured to run with software steering verifies the
functionality.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
I applied same test plan on 5.15.0-25-generic with and without the feature
patchset on CX6-dx.
In both cases the test kernel had the the problematic patch from (bug
https://bugs.launchpad.net/bugs/1967754) reverted.
I got the exact same results as in comment #6 so the patch set does not
Hi,
I run some tests to verify the feature. SMFS was enabled on CX7 setup
# devlink dev param show pci/:08:00.0 name flow_steering_mode
pci/:08:00.0:
name flow_steering_mode type driver-specific
values:
cmode runtime value smfs
but some of the asap test failed (see below),
with cf530217408e ("devlink: Notify users when objects are accessible")
i was able to change steering mode
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1967754
Title:
ubuntu 22.04 jammy, mlx5,
https://lore.kernel.org/netdev/cover.1632565508.git.leo...@nvidia.com/
It seems jammy kernel needs cf530217408e ("devlink: Notify users when objects
are accessible") from v5.16-rc1
(see link) the old devlink_register (before cf530217408e) implementation do
not publish the params so we cannot
** Description changed:
-
- on kernel 5.15.0-23-generic devlink cannot show/change flow steering mode
+ on kernel 5.15.0-23-generic devlink cannot show/change flow steering
+ mode
[test case]
# devlink dev param show pci/:24:00.0 name flow_steering_mode
pci/:24:00.0:
-
Public bug reported:
on kernel 5.15.0-23-generic devlink cannot show/change flow steering mode
[test case]
# devlink dev param show pci/:24:00.0 name flow_steering_mode
pci/:24:00.0:
name flow_steering_mode type driver-specific
values:
(flow steering mode description
Hi Dan,
98576013bf283 net/mlx5: DR, Add missing string for action type SAMPLER
was omitted by mistake, sorry for that.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1966194
Title:
[Jammy, mlx5,
reply to comment #4
on cx7 setup with 5.15.0-23-generic and Ib_virt capability set to 0
we see the correct speed is exposed by ibv_devinfo
so the feature works as expected
# ibv_devinfo -v|grep active_speed
active_speed: 100.0 Gbps (128)
Public bug reported:
[Impact]
Add support for software steering on cx7
[Test Case]
configure software steering on cx7 setup
run asap testing
(reference https://github.com/Mellanox/ovs-tests)
[Regression Potential]
TBD
[Other Info]
Feature patchset:
All patches are cleanly applied on Jamy
** Description changed:
[Impact]
Although focal kernel supports non FDB domains, rdma-core support is missing.
we need the below rdma-core patches to add this support
[Test Case]
- install latest DPDK from github
+ install latest DPDK from github
$ git clone git://dpdk.org/dpdk
$
** Summary changed:
- mlx5: steering, Add support for non FDB domain
+ focal ,mlx5, steering, Add support for non FDB domain
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1963926
Title:
focal
Public bug reported:
[Impact]
Although focal kernel supports non FDB domains, rdma-core support is missing.
we need the below rdma-core patches to add this support
[Test Case]
install latest DPDK from github
$ git clone git://dpdk.org/dpdk
$ cd
$ mason build
$ ninja -C build
- Run testpmd
Public bug reported:
[Impact]
Add support for NDR (next data rate) in IB MAD.
This speed is supported From ConnectX7 onwards.
And in addition user space support for NDR was extended.
[Test Case]
TBD
[Regression Potential]
TBD
[Other Info]
Feature patchset: (kernel v5.17-rc1)
fbdb0ba7051e
** Description changed:
[impact]
Add Support for flow counters offset for bulk counters on kernel side.
Currently Ubuntu 20.04 supports 'flow counters offset for bulk counters' on
user space side only with these 3 rdma-core patches from v27.0
3956cf74 mlx5: Add support for bulk flow
Hi Tim,
It extends devx interface and might affects any user space application
that uses rdma-core (which already has relevant patches), dpdk for
instance.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
Public bug reported:
[impact]
Add Support for flow counters offset for bulk counters on kernel side.
Currently Ubuntu 20.04 supports 'flow counters offset for bulk counters' on
user space side only with these 3 rdma-core patches from v27.0
3956cf74 mlx5: Add support for bulk flow counters
** Description changed:
[Impact]
On mlx5 device, When creating geneve tunnel in a Virtual Function and
setting it up error are logged in kernel buffer.
[test case]
using two setups connected back to back
on both sides create vf's
$ echo 1 >
Public bug reported:
[Impact]
On mlx5 device, When creating geneve tunnel in a Virtual Function and
setting it up error are logged in kernel buffer.
[test case]
using two setups connected back to back
on both sides create vf's
$ echo 1 > /sys/class/net/ens5f0/device/sriov_numvfs
add ip-es to
I reviewed the groovy patch and It looks fine!
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1921104
Title:
net/mlx5e: Add missing capability check for uplink follow for Ubuntu
20.04
To manage
The test kernel was verified successfully.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1915717
Title:
Ubuntu 21.04, kernel feature request, TLS device offload for Bond
To manage notifications
Hi,
I retried and All 8 commits were cleanly applied above
tag: Ubuntu-unstable-5.11.0-10.11
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1915717
Title:
Ubuntu 21.04, kernel feature request, TLS
Public bug reported:
The request here is to pull needed patches to support tls rx/tx offload
for bond interfaces.
patch set:
4e5a73329051 net/tls: Except bond interface from some TLS checks
153cbd137f0a net/tls: Device offload to use lowest netdevice in chain
dc5809f9e2b6 net/bonding: Declare
** Description changed:
Hi,
[Impact]
Currently in focal, devices reporter recovery is enabled even if state is
healthy.
[test case]
1)
display devlink health status
# devlink health show pci/:05:00.0 reporter fw_fatal
pci/:05:00.0:
reporter fw_fatal
I tested the kernel It works as expected and meets our needs
thanks
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1915403
Title:
devlink: don't do reporter recovery if the state is healthy
To
** Description changed:
Hi,
[Impact]
- Currently in focal, mlx5 devices reporter recovery is enabled even if state
is healthy.
+ Currently in focal, devices reporter recovery is enabled even if state is
healthy.
[test case]
1)
display devlink health status
# devlink health
Public bug reported:
Hi,
[Impact]
Currently in focal, mlx5 devices reporter recovery is enabled even if state is
healthy.
[test case]
1)
display devlink health status
# devlink health show pci/:05:00.0 reporter fw_fatal
pci/:05:00.0:
reporter fw_fatal
state healthy error 0
Hi Jeff,
upstream commit
50b2412b7e78 net/mlx5: Avoid possible free of command entry while timeout comp
handler
was picked to Ubuntu-5.4.0-56.62 kernel
(hash bcd6e98bef76cc8a49a1b736b0fefffbffb75c30)
(v5.4.71 upstream stable release,
I also tested with kernel Ubuntu-4.15.0-129.132 downloaded from
http://archive.ubuntu.com/ubuntu/pool/main/l/linux/
which has the commit and the issue is resolved
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
William,
yes, I built a bionic kernel with this commit included and the bug was resolved.
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1904848
Title:
Ubuntu 18.04- call trace in kernel buffer when
Public bug reported:
[Impact]
commit
d43b7007dbd1 net/mlx5: Fix a race when moving command interface to events mode
from upstream v5.7-rc1 (and in groovy) fixes
e126ba97dba9 mlx5: Add driver for Mellanox Connect-IB adapters
this fix should come with four more patches from v5.9.
410bd754cd73
** Description changed:
[Impact]
- unloading ib_ipoib causes a call trace to be logged in kernel buffer.
+ unloading ib_ipoib causes a call trace to be logged in kernel buffer.
- bisecting the bionic kernel reveals that this issue was introduced by
- 616e695435e3 workqueue: Try to catch
** Patch added:
"0001-net-mlx5e-IPoIB-initialize-update_stat_work-for-ipoi.patch"
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1904848/+attachment/5435933/+files/0001-net-mlx5e-IPoIB-initialize-update_stat_work-for-ipoi.patch
--
You received this bug notification because you are a
Public bug reported:
[Impact]
unloading ib_ipoib causes a call trace to be logged in kernel buffer.
bisecting the bionic kernel reveals that this issue was introduced by
616e695435e3 workqueue: Try to catch flush_work() without INIT_WORK()
in version 4.15.0-59.66
[test case]
# modprobe
Hi Jeff,
Six additional bug fixes are needed.
all applied cleanly.
94c4fed710c3 net/mlx5e: kTLS, Avoid kzalloc(GFP_KERNEL) under spinlock
581642f32f33 net/mlx5e: kTLS, Fix leak on resync error flow
66ce5fc05713 net/mlx5e: kTLS, Add missing dma_unmap in RX resync
8f0bcd19b1da net/mlx5e: TLS, Do
the three bug fixes commits was accepted upstream.
6e8de0b6b469 net/mlx5e: kTLS, Fix napi sync and possible use-after-free
9c25a22dfb00 net/mlx5e: Use synchronize_rcu to sync with NAPI
fe45386a2082 net/mlx5e: Use RCU to protect rq->xdp_prog
** Description changed:
Hi,
[Impact]
The
Public bug reported:
Hi,
The request is to backport
09689703d29a RDMA/mlx5: Do not race with mlx5_ib_invalidate_range during create
and destroy
with it the below list (divided to four sets) is required.
most of the patches applied cleanly above focal tree four exceptions.
1. 806b101b2bfa
** Description changed:
[IMPACT]
Clearing the sock TX queue in sk_set_socket() might cause unexpected
out-of-order transmit when called from sock_orphan(), as outstanding
packets can pick a different TX queue and bypass the ones already queued.
This is undesired in general. More
Public bug reported:
The bellow upstream commit fixes a bug in Ktls feature.
It is applied cleanly above the ubuntu-focal tree and passed basic sanity
testing.
We would like it to be backported ubuntu-focal
Thanks,
Amir
commit 41b14fb8724d5a4b382a63cb4a1a61880347ccb8
Author: Tariq Toukan
Hi Jeff,
the following patch need to be included in the patch set.
4cca96a8d9da IB/mlx5: Do reverse sequence during device removal
without it devlink dev reload operation causes a double
free of cache memory and leaves the kernel in a state that requires a
restart.
the patch is applied cleanly
43 matches
Mail list logo
