Date: Sun, 17 Jul 2011 10:15:05 -0400 From: Ryan Stone <rysto32@gmail.com> To: vadim_nuclight@mail.ru Cc: freebsd-net@freebsd.org Subject: Re: m_pkthdr.rcvif dangling pointer problem Message-ID: <CAFMmRNxbw=akdBSn7jZi_S3iuk7GV7-P8naNPvSiXcs-SyCkKQ@mail.gmail.com> In-Reply-To: <slrnj25jkk.r6m.vadim_nuclight@kernblitz.nuclight.avtf.net> References: <20110714154457.GI70776@FreeBSD.org> <slrnj25jkk.r6m.vadim_nuclight@kernblitz.nuclight.avtf.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, Jul 17, 2011 at 7:59 AM, Vadim Goncharov <vadim_nuclight@mail.ru> wrote: > Ways to improve are to be found from this starting point. However, > are that +2 atomic ops per packet really so expensive? How many of > atomic ops are already on that path? Any measures? On high-performance multiqueue NICs, those two atomic ops are pretty well guaranteed to be contended every time. It's a scalability nightmare.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAFMmRNxbw=akdBSn7jZi_S3iuk7GV7-P8naNPvSiXcs-SyCkKQ>