On Fri, 10 Jan 2025 13:40:28 +1100 David Gibson <david(a)gibson.dropbear.id.au> wrote:That said, I think we can do a bit better long term: I believe it should be possible to correct the value from the of-by-one csum_avx2(), I think with just an unconditional byteswap.See also https://bugs.passt.top/show_bug.cgi?id=108#c16, but I'm not so sure it's "a bit better". -- Stefano