APPLIED: [SRU][N/Q/R][PATCH v2 0/1] IB/coACK: Fix zero dmac race in neighbor resolution

[Mehmet Basaran]

Applied to questing:linux master-next branch. Patch was already applied
to noble/resolute linux master-next branches.

Thanks.

-------------- next part --------------
Gabriela Bittencourt via kernel-team <kernel-team at lists.ubuntu.com>
writes:

> From: Gabriela Bittencourt <gabriela.bittencourt at canonical.com>
>
> SRU Justification
>
> [Impact]
>
> NVIDIA runs into an issue building DGX Cloud GB300 clusters. The dst_fetch_ha() checks nud_state without holding the neighbor lock may result on the incorrect/zero mac being read and programmed in the device QP while it was not yet updated. This causes silent packet loss and eventual RETRY_EXC_ERR.
>
> The upstream commit 5e6de34d82b4 (“IB/core: Fix zero dmac race in neighbor resolution”) solved this problem.
>
> [Fix]
>
> Questing-aws, Resolute-aws and Noble-aws-6.17:
>
> * cherry-pick 5e6de34d82b4 ("IB/core: Fix zero dmac race in neighbor resolution")
>
> [Test Plan]
>
> * Compile and boot test.
>
> * AWS test.
>
> [Regression potential]
>
> * Regression likelihood is slim: the commit adds a lock in one line to hold nud_state until the copy of ha under seqlock.
>
> [Other info]
>
> * SF#00437493
>
> Chen Zhao (1):
>   IB/core: Fix zero dmac race in neighbor resolution
>
>  drivers/infiniband/core/addr.c | 3 +++
>  1 file changed, 3 insertions(+)
>
> --
> 2.53.0
>
>
> --
> kernel-team mailing list
> kernel-team at lists.ubuntu.com
> https://lists.ubuntu.com/mailman/listinfo/kernel-team
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 873 bytes
Desc: not available
URL: <https://lists.ubuntu.com/archives/kernel-team/attachments/20260619/86fb2ff6/attachment.sig>