Re: mlx5e:zero-prefix corruption during TCP DMA-BUF RX - help requested

From: jiabin deng

Date: Fri Nov 06 2026 - 03:54:00 EST


Hi Dragos,

Thank you for the confirmation. I have now submitted the fix to netdev
and copied you on the patch.

Subject: [PATCH net] net/mlx5e: SHAMPO, Disable end padding when
relaxed ordering is supported

Message-ID:
<20261106083719.22798-1-dengjiabin2023@xxxxxxxxx>

The patch is available in the netdev archive:
https://lore.kernel.org/netdev/20261106083719.22798-1-dengjiabin2023@xxxxxxxxx/

Thank you again for your help with the investigation and for suggesting
the fix.

Best regards,
Jiabin Deng

On Thu, Nov 5, 2026 at 10:14 PM Dragos Tatulea <dtatulea@xxxxxxxxxx> wrote:
>
> On 23.10.26 15:14, Dragos Tatulea wrote:
> > On 21.10.26 07:19, jiabin deng wrote:
> >> Hi Dragos,
> >>
> >> [...]
> >> Thank you for continuing to investigate this issue and for suggesting
> >> the fix. I also greatly appreciate the time and effort you and the
> >> other maintainers dedicate to the open-source community. As a student,
> >> I am excited to have the opportunity to contribute to the community
> >> and to learn from this process.
> >>
> > Thank you for your good analysis and fast response!
> >
> >> I would be happy to help prepare a formal patch with the test results.
> >> Would you prefer to submit the fix yourself, or would it be helpful
> >> for me to prepare and submit it based on your suggested change?
> >>
> > I am still waiting for input from the HW folks to understand if this
> > change is needed for all SHAMPO cases or only specific P2P DMA
> > cases.
> >
> > I think you should submit the fix as you did most of the work. Just
> > give me some time to check the above. I'll ping you hopefully next
> > week.
> >
> I got my answers. Feel free to go ahead send a fix. If you won't get
> to, let me know so that we do it.
>
> Thanks,
> Dragos