22.11.11 patches review and test

[Kevin Traynor]

On 24/12/2025 10:12, Minggang(Gavin) Li wrote:
> Sorry @Luca Boccassi, @Shani Peretz and @Kevin Traynor,
> I double checked the first commit of the three and found that I did tag it with CC stable but missed the following commits to fix a bug introduce by it. This is the root cause of the mess made by me. The first commit introduced behavior change that should not be backported to stable branches.
> Please revert the commits mentioned.
> 
>> -----Original Message-----
>> From: Maayan Kashani <mkashani at nvidia.com>
>> Sent: Wednesday, December 24, 2025 3:33 PM
>> To: Minggang(Gavin) Li <gavinl at nvidia.com>; Luca Boccassi
>> <luca.boccassi at gmail.com>; Ali Alnubani <alialnu at nvidia.com>; Shani Peretz
>> <shperetz at nvidia.com>; Kevin Traynor <ktraynor at redhat.com>
>> Cc: stable at dpdk.org; dev at dpdk.org; John McNamara
>> <john.mcnamara at intel.com>; Raslan Darawsheh <rasland at nvidia.com>; NBU-
>> Contact-Thomas Monjalon (EXTERNAL) <thomas at monjalon.net>; Salaheddin
>> Badawi <sbadawi at nvidia.com>; Wael Abualrub <waela at nvidia.com>
>> Subject: RE: 22.11.11 patches review and test
>>
>> ++ at Shani Peretz
>> ++ at Kevin Traynor
>>
>> Please see below,
>>
>> Regards,
>> Maayan Kashani
>>
>>> -----Original Message-----
>>> From: Minggang(Gavin) Li <gavinl at nvidia.com>
>>> Sent: Wednesday, 24 December 2025 5:09
>>> To: Maayan Kashani <mkashani at nvidia.com>; Luca Boccassi
>>> <luca.boccassi at gmail.com>; Ali Alnubani <alialnu at nvidia.com>
>>> Cc: stable at dpdk.org; dev at dpdk.org; John McNamara
>>> <john.mcnamara at intel.com>; Raslan Darawsheh <rasland at nvidia.com>;
>>> NBU-Contact-Thomas Monjalon (EXTERNAL) <thomas at monjalon.net>;
>>> Salaheddin Badawi <sbadawi at nvidia.com>; Wael Abualrub
>>> <waela at nvidia.com>
>>> Subject: RE: 22.11.11 patches review and test
>>>
>>> Hi Luca,
>>>
>>> The issue(https://bugs.dpdk.org/show_bug.cgi?id=1851) was introduced
>>> by cherry-picking some of the 3 related commits of refactoring
>>> multicast handling of PF in NV pmd. I don't understand why it/they
>>> were chosen because they included behavior change in PF. They should
>>> not be backported to stable ones from my point of view.
>>>
>>> The commit(s) polluted 22.11, 23.11 and 24.11 so far. Please revert
>>> below commits to fix the issue.
>>>
>>> 24.11
>>> ddcaa1671f net/mlx5: fix multicast
>>> adce6d130b net/mlx5: align PF and VF/SF MAC address handling
>>>

Hi Gavin/Maayan/Ali,

Thanks for the notification.

Unfortunately 24.11.4 was already released just before Christmas. If we
revert them now it is scheduled to be April before this is fixed in a
24.11 release.

It seems like a serious regression, so we could consider a new release
for it now, but we need to confirm if it's really needed first.

The commits were present but this bug was not reported during 24.114-rc1
testing. Was this test case covered during 24.11.4-rc1 testing, or just
specifically for 22.11 ?

Maybe it has been fixed/masked by other changes in 24.11.4 ?

If we want to consider a new release, then we'd need testing to confirm
the bug (not just the commits) is present in 24.11.4 ? and that it is
fixed by revert of those patches.

thanks,
Kevin.

>>> 23.11
>>> c070455e8b net/mlx5: align PF and VF/SF MAC address handling
>>>
>>> 22.11
>>> 58a6421bc3 net/mlx5: fix multicast
>>> ed3411217b net/mlx5: align PF and VF/SF MAC address handling
>>>
>>> Regards.
>>>
>>>> -----Original Message-----
>>>> From: Maayan Kashani <mkashani at nvidia.com>
>>>> Sent: Tuesday, December 23, 2025 10:09 PM
>>>> To: Luca Boccassi <luca.boccassi at gmail.com>; Ali Alnubani
>>>> <alialnu at nvidia.com>; Minggang(Gavin) Li <gavinl at nvidia.com>
>>>> Cc: stable at dpdk.org; dev at dpdk.org; John McNamara
>>>> <john.mcnamara at intel.com>; Raslan Darawsheh <rasland at nvidia.com>;
>>> NBU-
>>>> Contact-Thomas Monjalon (EXTERNAL) <thomas at monjalon.net>;
>>> Salaheddin
>>>> Badawi <sbadawi at nvidia.com>; Wael Abualrub <waela at nvidia.com>
>>>> Subject: RE: 22.11.11 patches review and test
>>>>
>>>> ++ at Minggang(Gavin) Li
>>>>
>>>> Regards,
>>>> Maayan Kashani
>>>>
>>>>> -----Original Message-----
>>>>> From: Luca Boccassi <luca.boccassi at gmail.com>
>>>>> Sent: Monday, 22 December 2025 13:27
>>>>> To: Ali Alnubani <alialnu at nvidia.com>
>>>>> Cc: stable at dpdk.org; dev at dpdk.org; John McNamara
>>>>> <john.mcnamara at intel.com>; Raslan Darawsheh <rasland at nvidia.com>;
>>>>> NBU-Contact-Thomas Monjalon (EXTERNAL) <thomas at monjalon.net>;
>>>>> Salaheddin Badawi <sbadawi at nvidia.com>; Wael Abualrub
>>>>> <waela at nvidia.com>
>>>>> Subject: Re: 22.11.11 patches review and test
>>>>>
>>>>> External email: Use caution opening links or attachments
>>>>>
>>>>>
>>>>> On Thu, 18 Dec 2025 at 19:06, Luca Boccassi
>>>>> <luca.boccassi at gmail.com>
>>>>> wrote:
>>>>>>
>>>>>> On Thu, 18 Dec 2025 at 18:58, Ali Alnubani <alialnu at nvidia.com>
>>> wrote:
>>>>>>>
>>>>>>>
>>>>>>> On 12/18/25 8:27 PM, Luca Boccassi wrote:
>>>>>>>> On Thu, 18 Dec 2025 at 18:24, Ali Alnubani
>>>>>>>> <alialnu at nvidia.com>
>>>>> wrote:
>>>>>>>>> On 12/4/25 2:56 AM, luca.boccassi at gmail.com wrote:
>>>>>>>>>
>>>>>>>>>> Hi all,
>>>>>>>>>>
>>>>>>>>>> Here is a list of patches targeted for stable release 22.11.11.
>>>>>>>>>>
>>>>>>>>>> The planned date for the final release is December 18th.
>>>>>>>>>>
>>>>>>>>>> Please help with testing and validation of your use cases
>>>>>>>>>> and report any issues/results with reply-all to this mail.
>>>>>>>>>> For the final release the fixes and reported validations
>>>>>>>>>> will be added to the
>>>>> release notes.
>>>>>>>>>>
>>>>>>>>>> A release candidate tarball can be found at:
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> https://dpdk.org/browse/dpdk-stable/tag/?id=v22.11.11-rc1
>>>>>>>>>>
>>>>>>>>>> These patches are located at branch 22.11 of dpdk-stable repo:
>>>>>>>>>>       https://dpdk.org/browse/dpdk-stable/
>>>>>>>>>>
>>>>>>>>>> Thanks.
>>>>>>>>>>
>>>>>>>>>> Luca Boccassi
>>>>>>>>> Hello,
>>>>>>>>>
>>>>>>>>> We ran the following functional tests with Nvidia hardware
>>>>>>>>> on
>>>>> v22.11.11-rc1:
>>>>>>>>> - Basic functionality:
>>>>>>>>>     Send and receive multiple types of traffic.
>>>>>>>>> - testpmd xstats counter test.
>>>>>>>>> - testpmd timestamp test.
>>>>>>>>> - Changing/checking link status through testpmd.
>>>>>>>>> - rte_flow tests
>>>>>>>>> (https://doc.dpdk.org/guides/nics/mlx5.html#supported-hardw
>>>>>>>>> ar
>>>>>>>>> e-
>>>>>>>>> of
>>>>>>>>> floads)
>>>>>>>>> - RSS tests.
>>>>>>>>> - VLAN filtering, stripping, and insertion tests.
>>>>>>>>> - Checksum and TSO tests.
>>>>>>>>> - ptype tests.
>>>>>>>>> - link_status_interrupt example application tests.
>>>>>>>>> - l3fwd-power example application tests.
>>>>>>>>> - Multi-process example applications tests.
>>>>>>>>> - Hardware LRO tests.
>>>>>>>>> - Buffer Split tests.
>>>>>>>>> - Tx scheduling tests.
>>>>>>>>>
>>>>>>>>> Functional tests ran on ConnectX-6 Dx (OS: Ubuntu 22.04 | Driver:
>>>>>>>>> MLNX_OFED_LINUX-5.8-7.0.6.1 | Firmware: 22.35.8002).
>>>>>>>>>
>>>>>>>>> An engineer from my team reported 2 issues:
>>>>>>>>>
>>>>>>>>> https://bugs.dpdk.org/show_bug.cgi?id=1851
>>>>>>>>> Did not reproduce in 22.11.10, git-bisect points to
>>>>>>>>> 58a6421bc396
>>>>>>>>> ("net/mlx5: fix multicast")..
>>>>>>>> Thanks, are you planning to send a fix, or should I just
>>>>>>>> revert that
>>>>> commit?
>>>>>>>
>>>>>>> I reached out to the engineer who authored the patch and am
>>>>>>> waiting for their response. I may not be able to get back to
>>>>>>> you with an answer today, though.
>>>>>>
>>>>>> Thank you, no problem it can wait a while
>>>>>
>>>>> Hi,
>>>>>
>>>>> Any update?
>>>>>
>>>>> Thanks