From: Jakub Kicinski <kuba@xxxxxxxxxx> Date: Fri, 1 Aug 2025 15:33:43 -0700 > On Wed, 30 Jul 2025 18:07:15 +0200 Alexander Lobakin wrote: >> Use __LIBETH_WORD_ACCESS to parse descriptors more efficiently when >> applicable. It really gives some good boosts and code size reduction >> on x86_64. > > Could you perhaps quantify the goodness of the boost with a number? :) Sure, only a matter of switching this definition and running the tests (and bloat-o-meter). Intel doesn't allow us to publish raw numbers (Gbps/Mpps), I hope the diff in percents (+ bloat-o-meter output) would be enough? Thanks, Olek
