Date: Sun, 06 Jul 2003 10:49:29 +0800 From: Eugene Grosbein <eugen@kuzbass.ru> To: Luigi Rizzo <rizzo@icir.org> Cc: net@freebsd.org Subject: Re: ipprecedence Message-ID: <3F078E39.ABC0822F@kuzbass.ru> References: <20030703002247.A2097@grosbein.pp.ru> <3F0310CE.5070302@tenebras.com> <3F03867A.79F82968@kuzbass.ru> <20030705123332.A60972@xorpc.icir.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Luigi Rizzo wrote: > Bottom line -- the whole architecture has been designed with > FIFO in mind, and implementing any different queueing policy > will involve some significant rewriting of the device drivers, > plus, potentially, some significant performance loss. Thank you for detailed explanation. I hope that dummynet's WFQ would be sufficient but not sure: will it correctly process weights whith zero-bandwidth pipe? I mean one dummynet pipe (bw 0) and two dummynet queues for that pipe assigning weight 1 and weight 100. That's for each direction of traffic, of course. Eugene
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3F078E39.ABC0822F>