Date: Sat, 20 Aug 2016 10:54:53 -0400 From: Ryan Stone <rysto32@gmail.com> To: koobs@freebsd.org Cc: Ryan Stone <rstone@freebsd.org>, "src-committers@freebsd.org" <src-committers@freebsd.org>, "svn-src-all@freebsd.org" <svn-src-all@freebsd.org>, "svn-src-head@freebsd.org" <svn-src-head@freebsd.org> Subject: Re: svn commit: r304435 - head/sys/netinet Message-ID: <CAFMmRNwG3RpPMpJM_WaVodN8YKiXC4t=%2B_CrFz=tSJCg--4iZw@mail.gmail.com> In-Reply-To: <96b7428a-0a57-6d6d-4b2e-24010596ceaf@FreeBSD.org> References: <201608182259.u7IMx0bx001967@repo.freebsd.org> <96b7428a-0a57-6d6d-4b2e-24010596ceaf@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sat, Aug 20, 2016 at 6:08 AM, Kubilay Kocak <koobs@freebsd.org> wrote: > Hi Ryan, > > Could you elaborate on any of the potential performance implications of > this? > As if r304437, skipping the call to in_broadcast() means that we avoid an additional (potentially heavily contended) rlock acquire and release on the ifnet address list.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAFMmRNwG3RpPMpJM_WaVodN8YKiXC4t=%2B_CrFz=tSJCg--4iZw>