From owner-freebsd-net Fri Dec 7 12:44:34 2001 Delivered-To: freebsd-net@freebsd.org Received: from inje.iskon.hr (inje.iskon.hr [213.191.128.16]) by hub.freebsd.org (Postfix) with ESMTP id 8360B37B41C; Fri, 7 Dec 2001 12:44:25 -0800 (PST) Received: from tel.fer.hr (zg07-209.dialin.iskon.hr [213.191.150.210]) by mail.iskon.hr (8.11.4/8.11.4/Iskon 8.11.3-1) with ESMTP id fB7Ki9O17119; Fri, 7 Dec 2001 21:44:12 +0100 (MET) Message-ID: <3C112A14.21F08D50@tel.fer.hr> Date: Fri, 07 Dec 2001 21:44:04 +0100 From: Marko Zec X-Mailer: Mozilla 4.78 [en] (Windows NT 5.0; U) X-Accept-Language: en MIME-Version: 1.0 To: Mike Smith Cc: Poul-Henning Kamp , arch@freebsd.org, net@freebsd.org Subject: Re: Request to back out Luigis polled-net patch in -stable. References: <200112071926.fB7JQx301437@mass.dis.org> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org Mike Smith wrote: > > Subject: Request to back out Luigis polled-net patch in -stable. > > I'm entirely in agreement with this; the decision to commit this code was > extremely ill-advised, and the best thing we can do now for everyone's > sake is to pull it as quickly as possible. > > > I would also like to point to the parallel piece of code: Jun-Itohs > > ALTQ for which he reliably has maintained a patch relative to the > > 4.X branch and which despite various peoples requests have not > > haphazardly been committed into -stable. And in that context one > > should not forget that ALTQ has a lot longer and better trackrecord > > of high quality than Luigis poll-code, or any of Luigis code for > > that matter. > > Yes; this is an excellent example of how it can be done better. Sorry guys, but aren't you comparing apples with oranges? As far as I understand, ALTQ is focused on implementing various new queuing disciplines, but on outgoing traffic if I am not mistaking. Luigi's code is aimed on achieving something completely different - making the system more susceptible to huge *incoming* traffic loads, by reducing interrupt processing and some PCI bus overhead. What do these two things have in common? Concerning the request for removal of the polling code, I personally as a BSD rookie cannot judge your arguments properly, but I must admit that the wording and intonation of pkh's note wasn't very pleasant... Marko To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message