From owner-freebsd-current@FreeBSD.ORG Wed Jan 21 00:19:26 2004 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 97EF916A4CE for ; Wed, 21 Jan 2004 00:19:26 -0800 (PST) Received: from Espresso.NEEBU.Net (espresso.neebu.net [66.166.158.130]) by mx1.FreeBSD.org (Postfix) with SMTP id 6907D43D31 for ; Wed, 21 Jan 2004 00:19:24 -0800 (PST) (envelope-from khuon@Espresso.NEEBU.Net) Received: from Espresso.NEEBU.Net (khuon@localhost [127.0.0.1]) by Espresso.NEEBU.Net (8.12.10/8.12.10) with ESMTP id i0L8JN7v022325 for ; Wed, 21 Jan 2004 00:19:23 -0800 (PST) Received: from Espresso.NEEBU.Net (khuon@localhost)i0L8JMcf022324 for ; Wed, 21 Jan 2004 00:19:22 -0800 (PST) Message-Id: <200401210819.i0L8JMcf022324@Espresso.NEEBU.Net> From: "Jake Khuon" To: freebsd-current@freebsd.org Dcc: In-reply-to: Marco Wertejuk's message of Tue, 20 Jan 2004 15:07:56 +0100. <20040120140756.GA19095@maeko.hayai.de> Action: Expires: Precedence: Priority: Normal X-Face: "(e&e|OIYrcV1x8y?txN%k1E2f[qWLjRjOn+a30)3>x`Wx%_9XiXs\IO2#G5L1m=c/|^h|z29wJ#]D/.?Ks,Mw1 X-URI: http://Espresso.NEEBU.Net/~khuon/ X-Organisation: Network Engineers for Effective Bandwidth Utilisation X-Header: /usr/include gives great headers X-System: Sun UltraSPARCstation2/2300MP running SunOS Release 5.8 X-Shell: tcsh 6.07.02 (Astron) 1996-10-27 (sparc-sun-solaris) options 8b,nls,dl,al,ng,rh X-Chtorr: History is full of revisionists. Where it used to say "THOU SHALT NOT KILL" it now says, "except as specified in section III-B, Paragraph 12, Sub-section D, Schedule 3." If that still doesn't suit you, wait till next year's commandments come out and trade it in for something that does. X-Mailer: MH 6.8.4 #4[UCI] (Espresso.NEEBU.Net) of Mon Feb 19 15:14:03 EST 1996 Date: Wed, 21 Jan 2004 00:19:22 -0800 Sender: khuon@Espresso.NEEBU.Net Subject: Re: unstable aironet driver? X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.1 List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 21 Jan 2004 08:19:26 -0000 ### On Tue, 20 Jan 2004 15:07:56 +0100, Marco Wertejuk ### casually decided to expound upon Jake Khuon the ### following thoughts about "Re: unstable aironet driver?": MW> I do not have a mini pci aironet but at least I can tell you MW> those pcmcia cards do well with latest current. MW> There was a firmware issue some time ago with a newer cisco MW> firmware but it seems Doug fixed them at least the error MW> messages disappeared (too small packet received or sth like that) MW> so the problems seems to only exist on minipci cards. I just tried upgrading the firmware and the kernel hangs so I downgraded again. I originally downgraded the firmware under 5.1R to get it to work in the first place. I've noticed a bunch of things that don't work like they should with the miniPCI version of the 350. MW> | Jan 19 06:53:20 Mocha kernel: Jan 19 06:53:20 Mocha dhclient: send_packet: MW> | No buffer space available MW> MW> This looks as if your card is not associated and has to MW> much data in its send buffer. The card loses association and seems to simply go out to lunch. Or maybe the driver does. I tried restarting dhclient to cycle through my media settings but it never reassociates. I also tried configuring the interface manually with the same results. I recently ported wmwave to wmairo and looking at I notice some pecularities as well. Things will be going fine when all of a sudden, the channel will flip and the noise level will jump from near zero to 50%. Nothing short of a reboot will bring the card back to life. This phenomenon seems to occur when I'm sending packets. I'm not sure if it's high levels of traffic which trigger it or what. It seems really random. Sometimes it will happen with very little traffic havig been sent and other times I can be pushing a constant 50Kbps for a half-hour straight (tested by scp'ing a big file) without anytihng happening. I thought that this might be related to the problem for which Jeffrey Hsu just commited a an update but after just doing a cvsup and recompile of the kernel, I still see the problem. -- /*===================[ Jake Khuon ]======================+ | Packet Plumber, Network Engineers /| / [~ [~ |) | | --------------- | | for Effective Bandwidth Utilisation / |/ [_ [_ |) |_| N E T W O R K S | +=========================================================================*/