From owner-freebsd-hardware@FreeBSD.ORG Sun Jan 14 18:40:24 2007 Return-Path: X-Original-To: freebsd-hardware@freebsd.org Delivered-To: freebsd-hardware@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id CD3AE16A417 for ; Sun, 14 Jan 2007 18:40:24 +0000 (UTC) (envelope-from sam@errno.com) Received: from ebb.errno.com (ebb.errno.com [69.12.149.25]) by mx1.freebsd.org (Postfix) with ESMTP id A475E13C45E for ; Sun, 14 Jan 2007 18:40:24 +0000 (UTC) (envelope-from sam@errno.com) Received: from [10.0.0.199] ([10.0.0.199]) (authenticated bits=0) by ebb.errno.com (8.13.6/8.12.6) with ESMTP id l0EIeOeV025788 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Sun, 14 Jan 2007 10:40:24 -0800 (PST) (envelope-from sam@errno.com) Message-ID: <45AA7918.2020603@errno.com> Date: Sun, 14 Jan 2007 10:40:24 -0800 From: Sam Leffler Organization: Errno Consulting User-Agent: Thunderbird 1.5.0.9 (Macintosh/20061207) MIME-Version: 1.0 To: Trigve Siver References: <20070114142837.13776.qmail@web52714.mail.yahoo.com> In-Reply-To: <20070114142837.13776.qmail@web52714.mail.yahoo.com> X-Enigmail-Version: 0.94.1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: freebsd-hardware@freebsd.org Subject: Re: DWL-G520 low signal, low speed X-BeenThere: freebsd-hardware@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: General discussion of FreeBSD hardware List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 14 Jan 2007 18:40:24 -0000 Trigve Siver wrote: > Hi, > > One more thing I want to add is...that when I was using ath I have had low speed and also had a lot of interupts and I realised that ath was sharing IRQ with USB (IRQ 19 in my case)... so I disabled the USB and throw away ath from kernel and compile it as module (want to take some performance test with ndis and ath)...and after that I have still low signal but media is OFDM/54 Mbps and is stable... to the performance....I think that ndis is somehow a little faster...when dowloading/uploading about 100 kB/s faster...but haven't done some serious testing (don't know how maybe with kismet?) Downstream performance is mostly dependent on performance of the sender. 100 kB/s (kilobits?) is likely not significant which means the radio operation are likely similar. If you really want to compare what's going on collect a packet trace of both drivers and look at the tx rates and other characteristics. The interrupt rate you cited original is too high. I believe your card has on-chip counters for phy errors which means you should get only interrupts for real frames and that's typically max's out at ~4K/sec when running full out in both directions. More typical packet rates are <1K and there are some interrupt mitigiation techniques used to bring the interrupt rate lower than that. BTW top-posting means all context is lost when I reply; it is discouraged. Sam