Date: Thu, 22 Mar 2007 18:39:26 +0300 From: Yar Tikhiy <yar@comp.chem.msu.su> To: Julian Elischer <julian@elischer.org> Cc: Luigi Rizzo <rizzo@icir.org>, net@freebsd.org Subject: Re: A dummy Ethernet driver Message-ID: <20070322153926.GY41715@comp.chem.msu.su> In-Reply-To: <4601B28B.7090204@elischer.org> References: <20070321201936.GN41715@comp.chem.msu.su> <20070321132306.A1936@xorpc.icir.org> <4601B28B.7090204@elischer.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, Mar 21, 2007 at 03:32:43PM -0700, Julian Elischer wrote: > Luigi Rizzo wrote: > >On Wed, Mar 21, 2007 at 11:19:36PM +0300, Yar Tikhiy wrote: > >>Hi folks, > >> > >>We have disc(4) for testing and benchmarking. However, it's a > >>loopback driver, so such things as vlan or bridge cannot attach to > >>it. I needed a similar dummy interface mimicing Ethernet and failed > >>to find a ready solution. I tried ng_eiface+ng_hole, but it just > >>couldn't keep up with gigabit rates. So I knocked up a new dummy > >>driver, edsc(4): Ethernet discard interface. I'd like to commit it > >>if there are no objections. Then it could also serve as the bones > >>of an Ethernet driver for those who study kernel internals or want > >>to write a new driver. > > > >seems like a good idea. > > > > I think the question I have is why eiface couldn't cope..? > I think it should have so I'll look at it.. I believe that it can't keep pace because the packet isn't passed straight from if_start() to Netgraph due to some reservations about locking. A callback is used intead, which may introduce a delay. -- Yar
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20070322153926.GY41715>