[PATCH] net: fix signed shift overflow in IPv6 phdr cksum

[Aleksandr Khromov]

On 28/09/2026 16:18, Thomas Monjalon wrote:
> Should we merge this patch and expect another one for the endian issue?

Yes, please apply it as is. The byte order problem is a separate,
pre-existing bug that only shows up on big-endian builds, so I would
rather fix it in its own commit.

I will send a follow-up series with two patches:
  - net: use rte_cpu_to_be_32() for the next header in the IPv6
    pseudo-header, as Stephen suggested;
  - net/hinic: same fix for the copy in hinic_ipv6_phdr_cksum().

Thanks Stephen for the review.