Date: Thu, 17 Apr 2008 16:07:16 +0200 (CEST) From: Christopher Arnold <chris@arnold.se> To: Alexander Sack <pisymbol@gmail.com> Cc: freebsd-net@freebsd.org Subject: Re: bge dropping packets issue Message-ID: <20080417160712.S5510@localhost> In-Reply-To: <3c0b01820804170643w6b771ce9jdfc2dc5b240922b@mail.gmail.com> References: <3c0b01820804160929i76cc04fdy975929e2a04c0368@mail.gmail.com> <200804161456.20823.jkim@FreeBSD.org> <3c0b01820804161328m77704ca0g43077a9718d446d4@mail.gmail.com> <200804161654.22452.jkim@FreeBSD.org> <3c0b01820804161402u3aac4425n41172294ad33a667@mail.gmail.com> <20080417112329.G47027@delplex.bde.org> <3c0b01820804170643w6b771ce9jdfc2dc5b240922b@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, 17 Apr 2008, Alexander Sack wrote: > For my own edification, when do you want use DEVICE_POLLING versus > interrupt driven network I/O? With all question like these I suppose > the answer depends on the workload and the interrupt bandwidth of the > machine (which depends on the type of hardware)... > > But why was it added to begin with if standard interrupt driven I/O is > faster? (was it the fact that historically hardware didn't do > interrupt coalescing initially) > The ability to reserve cpu is one of the great features. If your host is being DDOS'ed it is good to have a portion of the CPU reserved to applications so a) the machine dosn't die. and b) so you can continue to login and investigate and perhaps solve the problem. /Chris -- http://www.arnold.se/ http://www.mbit.us/
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20080417160712.S5510>