From owner-freebsd-hardware@FreeBSD.ORG Tue Apr 26 07:15:57 2005 Return-Path: Delivered-To: freebsd-hardware@freebsd.org Received: by hub.freebsd.org (Postfix, from userid 618) id 2650716A4CF; Tue, 26 Apr 2005 07:15:57 +0000 (GMT) In-Reply-To: <17005.52150.590534.365619@szamoca.krvarr.bc.ca> from Sandy Rutherford at "Apr 25, 2005 10:03:50 pm" To: sandy@krvarr.bc.ca (Sandy Rutherford) Date: Tue, 26 Apr 2005 07:15:57 +0000 (GMT) X-Mailer: ELM [version 2.4ME+ PL54 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Message-Id: <20050426071557.2650716A4CF@hub.freebsd.org> From: wpaul@FreeBSD.ORG (Bill Paul) cc: freebsd-current@freebsd.org cc: freebsd-hardware@freebsd.org Subject: Re: Too Evil, Too Furious X-BeenThere: freebsd-hardware@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: General discussion of FreeBSD hardware List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 26 Apr 2005 07:15:57 -0000 > >>>>> On Mon, 25 Apr 2005 17:15:28 +0000 (GMT), > >>>>> wpaul@freebsd.org (Bill Paul) said: > > > - RayLink RT2500 wireless -- this one shows up on some PCI cards, but > > I haven't had any luck finding one locally yet > > I assume that you mean RaLink (no "y"). Have you tried contacting the > people at RaLink? I understand that they are very supportive of open > source development. As a result, there is a Linux driver for both the > rt2400 and rt2500. If they were really supportive of open source, there would be manuals for the RT2400 and RT2500 chipsets on their website that people could download. Then people could develop drivers for _any_ OS, not just Linux. "But Linux support means open source support, right?" No, it just means Linux support. I'm sure the marketing people would like you to think the two things are equal, but they really aren't. "But you can easily port the Linux driver, right?" Any given network chipset has a number of different features. A driver written for OS A will take advantage of those features that produce good performance on OS A. The chip might also have additional features that would benefit OS B, but if all you have as a reference is the driver for OS A, you might be completely unaware those features exist. That constitutes an unfair handicap, nevermind the time wasted trying to decipher someone else's code before you can write your own. Releasing the manuals insures a level playing field for everyone. > The project page is > http://sourceforge.net/projects/rt2400. BTW, this driver seems solid. > I flogged it hard with flood pings and it held up well. My rt2500 is > a Belkin model F5D7010.PCMCIA card, which I have stuck in a laptop > running Slackware. Unfort., I need it so I can't send it to you. I am > not aware of any PCI cards using the rt2500. You've just provided a perfect example of model number confusion. Your Belkin F5D7010 card has a RaLink chipset. But Belkin has another card also called the F5D7010 which has a Broadcom chipset. I found a Belkin F5D7010 card at CompUSA this past weekend, but I couldn't buy it since there was no way to tell which revision it was, and I didn't want to end up with yet another Broadcom cardbus card. > Given RaLink's willingness to work with the open source community, > wouldn't the rt2500 be a better candidate for a native driver, rather > than Project Evil? Right now my goal is to make Project Evil support the NDIS spec as well as possible, and for that I need to test as many NDIS drivers as possible. If someone else wants to create a native driver, then more power to them. I've written more than enough drivers for one lifetime. -Bill -- ============================================================================= -Bill Paul (510) 749-2329 | Senior Engineer, Master of Unix-Fu wpaul@windriver.com | Wind River Systems ============================================================================= you're just BEGGING to face the moose =============================================================================