Date: Tue, 10 Feb 2015 18:55:28 +0000 (GMT) From: Rui Paulo <rpaulo@me.com> To: Slawa Olhovchenkov <slw@zxy.spb.ru> Cc: "svn-src-head@freebsd.org" <svn-src-head@freebsd.org>, Adrian Chadd <adrian@freebsd.org>, "src-committers@freebsd.org" <src-committers@freebsd.org>, "svn-src-all@freebsd.org" <svn-src-all@freebsd.org>, John Baldwin <jhb@freebsd.org> Subject: Re: svn commit: r278479 - in head: etc sys/kern Message-ID: <fea3f464-c674-4ccc-80c4-8bfb98803d80@me.com>
next in thread | raw e-mail | index | archive | help
On Feb 10, 2015, at 10:52 AM, Slawa Olhovchenkov <slw@zxy.spb.ru> wrote:=0A= For common case (I am not talk about current devd implementation -- I=0Aam= don't have any inforamtion/metrics/etc) routing and processing=0Aevents m= ay be sensitive to delay and ordering: may be exist=0Arequirement 'delay n= ot more then 700ns', may be exist requirement=0A'next event process only a= fter complete process previos event'. And=0Asome event handling may be ver= y CPU/disk/etc consumption. Need to good=0Athink over and design API and a= rchitecure.=0A=C2=A0=0Arouting events do not come over devd (or any other = /dev device): there's a routing socket specifically tailored for that purp= ose. =C2=A0For that type of latency, you probably don't want to be crossin= g the kernel/userland barrier too often (or never).=
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?fea3f464-c674-4ccc-80c4-8bfb98803d80>