[PATCH v9] net: optimize raw checksum computation

Scott Mitchell scott.k.mitch1 at gmail.com
Sat Jan 10 04:46:42 CET 2026


Cross posting from v11 thread:

https://godbolt.org/z/qr39hf76s
rte_ipv4_phdr_cksum and rte_ipv6_phdr_cksum are both fully unrolled
(-O2 or higher). Vectorization also happens (clang chooses
not to vectorize ipv4). yay compilers :)


More information about the dev mailing list