Wireless networking using the ndiswrapper module
Posted by Steve on Fri 2 Jun 2006 at 12:35
Getting wireless networking working with the ndiswrapper driver is fairly straightfoward if your card has an associated Windows driver. Here we'll look at getting wireless networking working for a Dell Inspiron 1300, you should be able to follow the recipe for most other wireless networking cards which are supported ndiswrapper.
ndiswrapper is a collection of utilities which essentially allows you to load and run a network card driver written for Microsoft Windows upon your Linux kernel. This means that a card which isn't supported natively may be used indirectly.
Note:
Running ndiswrapper isn't recommended unless it is the only means you have of using your wireless card, since it allows potentially buggy Windows driver code to interfere with your Linux kernel.
I've been given a Dell Inspiron 1300 laptop and discovered that the wireless network card isn't supported by the Linux kernel. Mostly this isn't a big deal, but it would be nice to be able to surf wirelessly on trains, and in bed.
Enter ndiswrapper! It turns out that there is a Windows driver which can be used to make this card work, and installing that driver is very straightfoward.
Since ndiswrapper isn't included in the stock kernels we have to build it seperately with the module-assistant package.
First of all we'll need to install the module-assistant package, and then prepare the build environment. (This means downloading the Kernel headers which match your currently running kernel):
root@steve:~# apt-get install module-assistant
Once installed the preparation is carried out by running "module-assistant prepare". (You can use m-a as a shortcut for module-assistant. I'll type the longer version in these examples.)
root@steve:~# module-assistant prepare Getting source for kernel version: 2.6.16-2-xen-686 apt-get install linux-headers-2.6.16-2-xen-686 Reading package lists... Done Building dependency tree... Done The following extra packages will be installed: linux-headers-2.6.16-2-xen The following NEW packages will be installed linux-headers-2.6.16-2-xen linux-headers-2.6.16-2-xen-686 0 upgraded, 2 newly installed, 0 to remove and 0 not upgraded. Need to get 3433kB of archives. After unpacking 28.0MB of additional disk space will be used. Do you want to continue [Y/n]?
Here we've downloaded the Kernel header package, and setup the symbolic link /usr/src/linux to point to the correct place.
Now that the module assistant configuration is complete we need to get the ndiswrapper kernel modules source code:
root@steve:~# apt-get install ndiswrapper-source
Build the module by executing:
root@steve:~# module-assistant build ndiswrapper
If there are no errors you should be ready to install the relevent tools:
root@steve:~# apt-get install ndiswrapper-utils-1.8
With the tools installed we can now install our freshly built module package:
root@steve:~# module-assistant install ndiswrapper Selecting previously deselected package ndiswrapper-modules-2.6.16-2-xen-686. (Reading database ... 68278 files and directories currently installed.) Unpacking ndiswrapper-modules-2.6.16-2-xen-686 (from .../ndiswrapper-modules-2.6.16-2-xen-686_1.16-1+2.6.16-14_i386.deb) ... Setting up ndiswrapper-modules-2.6.16-2-xen-686 (1.16-1+2.6.16-14) ...
Before installing your new kernel module you'll need to download and install the Windows Device Driver. In my case I could fetch these by running:
root@steve:~# wget http://biginoz.free.fr/linux/bcmwl5.sys root@steve:~# wget http://biginoz.free.fr/linux/bcmwl5a.inf
Download your drivers from wherever you can find them, or from your CD-ROMs. With the drivers downloaded you can configure ndiswrapper to use them by running:
root@steve:~# ndiswrapper -i bcmwl5a.inf Installing bcmwl5a Forcing parameter IBSSGMode|0 to IBSSGMode|2 Forcing parameter IBSSGMode|0 to IBSSGMode|2
At this point we're almost completely finished. We just need to load the kernel module we built. (Loading it before installing a driver will fail):
root@steve:~# depmod -a root@steve:~# modprobe ndiswrapper
All being well you should now have a wireless network interface:
root@steve:~# ifconfig wlan0 up
If you, your neighbours, or your train company, have no security you can now get an IP address:
root@steve:~# dhclient wlan0 Internet Software Consortium DHCP Client 2.0pl5 Copyright 1995, 1996, 1997, 1998, 1999 The Internet Software Consortium. All rights reserved. Please contribute if you find this software useful. For info, please visit http://www.isc.org/dhcp-contrib.html sit0: unknown hardware address type 776 sit0: unknown hardware address type 776 Listening on LPF/wlan0/00:16:ce:32:57:67 Sending on LPF/wlan0/00:16:ce:32:57:67 Sending on Socket/fallback/fallback-net DHCPREQUEST on wlan0 to 255.255.255.255 port 67 DHCPACK from 192.168.0.1 bound to 192.168.0.9 -- renewal in 43200 seconds.
If you're using WEP encryption you can use something like the following to configure it:
iwconfig wlan0 key open 1234abcd56 iwconfig wlan0 essid my_essid_name dhclient wlan0
(iwconfig is contained in the wireless-tools package.)
(Note: TCP over DNS works fine on British Rail trains ;)
[ Parent | Reply to this comment ]
[ Send Message | View Steve's Scratchpad | View Weblogs ]
Definitely true.
[ Parent | Reply to this comment ]
[ Parent | Reply to this comment ]
I may sound fanatic, but I really don't think one should advertise these solutions. People's systems are now tainted and they cannot receive proper help. People don't get to bug manufacturers to release drivers or specs, manufacturers publicize Linux support through this bad bridging technology.
All in all, not very positive for Linux.
ndiswrapper is a really short sighted solution to my point of view. Better to get another card (whenever possible).
jerome (at coffeebreaks.org)
[ Parent | Reply to this comment ]
[ Send Message | View Steve's Scratchpad | View Weblogs ]
On the one hand I agree. I'm a Debian developer and I utterly support free software, including sharing my own and trying to do useful work writing documentation and generally useful articles here and elsewhere.
On the other hand I disagree. There are practical reasons why such steps might be useful, even if they are not desirable.
If I had the choice I'd rather choose hardware which is natively supported, but refusing to use a non-free solution as a stop-gap measure seems shortsighted.
As for getting another card? Yes that would be a solution, and it would be what I'd do if I were buying wireless equipment of my own, but in this case the laptop has it integrated so buying another card is a pricy option..
I'll post a poll early next week free vs non-free and see what other people feel most comfortable with.
[ Parent | Reply to this comment ]
On the other hand I disagree. There are practical reasons why such steps might be useful, even if they are not desirable.'practical' in this case is unethical. Your failure to resist the minor temptation to use proprietary drivers means one less person pushing for the release of Free drivers for that hardware, and one less person buying hardware supporting Free drivers.
That's why the state of wireless drivers is poor. You've become part of the problem even if you can't bring yourself to admit it.
As for getting another card? Yes that would be a solution, and it would be what I'd do if I were buying wireless equipment of my own, but in this case the laptop has it integrated so buying another card is a pricy option..So you won't spend money to support Free drivers? Exactly.
[ Parent | Reply to this comment ]
No it isn't. "Practical" is a stable, secure wireless networking connection with consistent available bandwdith.
"You've become part of the problem even if you can't bring yourself to admit it."
Ouch. If your nappy/diaper rash is hurting, why not get some lotion, rather than taking it out on someone who's trying to help others?
So which 802.11g PCMCICA/Cardbus card with __stable__ native drivers and WPA TKIP support would you recommend, exactly? By "__stable__" I mean a card/driver whose throughput doesn't drop to 30kB/s whenever you look at it sideways. I've spent over a year dealing with various rt2500/rt2570 cards that crash and burn horribly every other hour with serialmonkey drivers (recommended by the FSF as "working well with GNU/Linux at least a year ago - see http://www.fsf.org/resources/hw/net/wireless/cards.html) and I'm now a devoted ndiswrapper user (consistent high bandwith available using WPA TKIP). Don't get me wrong: I'm not disparaging all the hard work put in by the serialmonkey team, and I'm looking forward to the day when rt2x00 is fit for public consumption (and, in fact, I've been looking forward to that day for at least a year). Until then, the only thing I _can_ use is ndiswrapper.
I'm looking forward with eager anticipation to the long list of cards/chipsets that you have personal experience of getting to work under Linux, dear newly-found teacher of ideology. Please - no lame, cheap "just go and google for it" cop-out - how about sharing _your_ solution that works so much better than ndiswrapper. And don't forget, I want something I can test to saturation for a day, without available bandwidth dropping under, say, 1Mbps. And WPA too.
PS While we're kindling the fire for a full-on flame-fest, may I suggest you change to an operating system that's released under a licence less restrictive than the GPL - perhaps one of the BSDs, for instance?
Over to you, Mr Brown.
[ Parent | Reply to this comment ]
Don't get me wrong: I'm not disparaging all the hard work put in by the serialmonkey team, and I'm looking forward to the day when rt2x00 is fit for public consumption (and, in fact, I've been looking forward to that day for at least a year). Until then, the only thing I _can_ use is ndiswrapper.As above, you are also unable to resist temptation and are part of the problem by sponsoring the companies that refuse to provide Free drivers or specs. You're literally paying them to fight against Free Software, and as you seem happy with your solution, are unlikely to be adding significant pressure to the development of the Free drivers.
There are always options available - I don't buy the argument from the tempted that using a slower 3D card with Free drivers is so unacceptible they must use a card with non-Free drivers, nor do I buy arguments about the speed or feature set of wireless cards in the same vein. There's always at least one option available regardless: to abstain.
may I suggest you change to an operating system that's released under a licence less restrictive than the GPL - perhaps one of the BSDs, for instance?How is this at all relevant to the article or your comment?
[ Parent | Reply to this comment ]
"As above, you are also unable to resist temptation and are part of the problem by sponsoring the companies that refuse to provide Free drivers or specs. You're literally paying them to fight against Free Software, and as you seem happy with your solution, are unlikely to be adding significant pressure to the development of the Free drivers."
A long time ago (March 2005?), I decided that I wanted (for the obvious reasons) to start using wireless, so I did what I thought was the responsible thing (from a "Free software" pov) and researched which cards to buy. I found the FSF's wireless page (http://www.fsf.org/resources/hw/net/wireless/cards.html) and, since everybody seemed to be heaping praise on Ralink Tech for handing over hardware documentation and specs for their rt2500 chipset, I dutifully trundled off to my local hardware store and lashed out on half-a-dozen different cards based on the rt2500.
Since then, I've spent countless hours trying drivers to get these things to behave properly: neither the ralink-released original drivers, nor the serialmonkey-maintained versions nor the ural-linux offshoot have ever worked reliably (they'll stay up and running for minutes, possibly an hour, but certainly not days). And no, I don't want your sympathy. I'm simply trying to explain why I resent your "You are part of the problem/You're paying them to fight against free software" comments.
Like I said above, I'm looking forward to the day when rt2x00 is fit for use, but until then, the only way I can use these cards reliably is with ndiswrapper. And as such, I greatly appreciate Steve's article.
"There's always at least one option available regardless: to abstain."
Do you expect me to abstain when the FSF tells me these cards "work well"? Besides which, why should I abstain when I can take Mr Brown's advice and "spend money to support free drivers"? All I'm waiting for now is to find out which free drivers he personally recommends...
"How is this at all relevant to the article or your comment?"
In my opinion, freedom isn't an "on/off" thing - it's "by degrees" (hence my black/grey/white comment below). You appear to prefer gpl licencing over closed, proprietary licencing, presumably because it affords user/developers more freedom (and if you do, we share the same opinion). But my point is that if we're on some freedom-crusade here, why not "go the whole hog" and embrace a licence which doesn't restrict developers in the way that the gpl does?
And while we're on the subject of false dichotomies ("so either do what's right or go back to Windows where you have all the proprietary drivers you'll ever need" - swbrown, Sun 4 Jun 2006 at 21:48), I'd be interested in knowing if you've ever purchased any products from Nike, Adidas, Wal Mart, Ikea, Reebok, Nestle, BP, Shell, (long list omitted for brevity), etc, etc. Surely if using ndiswrapper makes me "part of the problem", providing support (directly or indirectly) to any one of these companies makes you an advocate of unethical business practices.
I should probably temper this entire post by explaining that I'm both an avid Debian user and a huge fan of Richard Stallman.
Yours sincerely, Mr Grey.
[ Parent | Reply to this comment ]
On the positive side, while I am using ndiswrapper, at the same time, I will email to the adapter manufacturer that their lack of Linux support will lead me to switch to a different manufacturer in the future.
[ Parent | Reply to this comment ]
[ Parent | Reply to this comment ]
My history
I bought a Buffalo WLI Card and there was no support for linux. I wrote to the manufacturer and they said that in a short while they would release a linux driver for my card. This never happend... and im still waitting :(
So if i wanted to use my buffalo card i must use ndiwrapper.
Of corse the drivers are buggy (i had to dig in the inf files to put my card working correctly ) but is the best we've got.
Every article is a good article if it is usefull. We should not critisize other people good will.
[ Parent | Reply to this comment ]
I'm using ndiswrapper to run drive my D-Link DWL-G650+ PCMCIA wireless adaptor, without problems.
If ndiswrapper isn't recommended please could you supply a few hints on (a) how to find a card which definitely will run natively on Linux and (b) how to install such a card (and drivers, etc). I ended up with this D-Link because after a scout round the web there didn't seem to be much to choose between in terms of linux support.
Best wishes
Ivan
ps. I'm just about to register now;)
[ Parent | Reply to this comment ]
You can always find PCMCIA cards on ebay that will work under linux and will work straight off.
Personlly i use an old senoa and an equally as old cisco PCMCIA cards.
The only time ie had to do anything as ´busy´ as the above is patching the kernal for an orinoco silver card and kismet.
So im to lazy to even atempt running a windoze based wireless card if the above is anything to go by. Maybe the article has that in its merit
[ Parent | Reply to this comment ]
I used ndiswrapper for a while because centrino support was broken in the distro I was using. It was a real pain, though, because I had to reboot everytime I needed to switch essid's. (That is, boot up, set up the essid, reboot. Ugh). I'm glad I'm back on native drivers now. I learned a good bit about wireless drivers, too.
[ Parent | Reply to this comment ]
These days every bit of hardware is checked for native drivers where possible.
[ Parent | Reply to this comment ]
[ Parent | Reply to this comment ]
And what about no 3D acceleration on video cards, unless you install closed drivers?Buy hardware with Free 3D drivers. The FSF recommends the Intel GMA chipsets, and there are many versions of ATI cards with Free drivers (although the FSF doesn't recommend those as they see ATI as openly hostile to Free Software).
No Free 3D drivers for your card? Fight to get them, or buy a card that has them. GNU/Linux didn't get to where it is by discarding the philosophy whenever it was convenient. As a user, you're part of that, so either do what's right or go back to Windows where you have all the proprietary drivers you'll ever need.
[ Parent | Reply to this comment ]
And in other news, several billion shades of grey have just been discovered between the colours black and white.
(BTW, the above post at Sat 8 Jul 2006 at 17:30 is mine too).
Jaime
[ Parent | Reply to this comment ]
Yes, having working free BIOSes (and firmwares, for that matter) able to replace those on modern motherboard would be very nice and welcomed.
But beware mixing different things: BIOS and firmware run on the hardware chipset (in opposition to the main system cpus/mem). They don't affect the Linux runtime environement, neither depend on it.
On the other hand, non-free binary drivers (like ATI/NVIDIA GPU modules, Nvidia mb nic & sata chipsets or ndiswrapper Windows drivers ...) runs on the main system.
So, by opposition to non free firmwares, non free drivers:
- makes something free lesser free. The Linux kernel is free, we shouldn't accept regression there ! At some point, if this become a generaly adopted solution, we may end-up with a very small kernel emulating Windows, and everything up being Windows drivers. Linux would then become a quasi non free OS. And, yes it's already possible to reach this point (having binary proprietaries drivers for network cards, sata chip, graphic chip, scsi chips, raid management software ...: that's a complete driver set for an hypothetic linux-based-non-free-machine). On the other hand, i386 chipped bioses had always been non free, so we don't give up anything by using them.
- presents security and stability risk for the whole operating system. Can't be audited. Q&A can't be assured to the usual extent requiered in kernel devlopment. A non free bios is not so risky.
- slows down free softwares evolution, imposing the maintainance of compability API, while a free driver can be updated in tree to adapt for API changes. Or when the kernel/xorg/wireless-tools/... developers refuses to slow down their plans of evolution, they tend to break in newer versions, introducing a buggy linux experience.
- can't be redistributed freely (most of the time), so lack in distros install medias. This add pain for Linux installation & upgrades, and refrain newcommers to adopt Linux (most non free firmwares/bios don't have this effect, since they are mostly distributed in the hardware).
- slows down the implementation of free alternatives.. They tend to become considered as an acceptable status quo for many users and vendors, getting on the way of developers to obtain chipset documentation and to write free drivers. And diminushing the presure on vendors (to either provide docs or loose sales). Look at the OpenBSD case: they don't have ndiswrapper and vendors don't try to support them with binary drivers ; so, with a fraction of the Linux kernel ressources (in number of developers, in entreprise support etc) they wrote free drivers for Atheros wifi chipsets (that had been backported to Linux since), free drivers for the recent Intel 3945ABG, or for nvidia etherne chipsets, a free raid management solution for ami/lsi raid chipsets, an so on. On all those fronts, Linux had non free drivers and took much time to implement an alternative (if ever). Also note that those OpenBSD developers reported that when they asked for hardware specs, they'd been told: "rather than docs, we can provide a binary blob ; that's an acceptable solution since it has been accepted by several linux distribution (Suse, Mandriva)". Shame on those distributions. Supporting ndiswrapper (as using it, or publishing this doc) is really counterproductive: in fact, it slows down the progress of free alternatives, we have cases to proof this.
- breaks long term hardware support, since the vendor will at some point preffer you to buy they newer hardware than to spend themself ressource in maintaining old drivers. Long term support of non free bios isn't such a problem, since bios is independant of the software installed on your system.
- makes Linux perform bad in face of other proprietary OS (ease of installation, stability and performance): ndiswrapper can't perform as good as the original Windows implementation NDIS was designed for. There we loose an occasion to comment on the superiority of Linux (and for some users, to choose Linux rather than windows). In the other hand, non free BIOS is vendor neutral, and don't favor Windows nor Linux.
- break portability of the kernel, eg. if you want to run a ndiswrapped driver on a sparc64 host. While by definition, firmwares and bios arent' to run on the main CPU/chipset board, so they (mostly) don't prevent the Linux driver to support the device on other platforms. The world is not i386.
- divides our community. Costs us anger and stress and mailing lists flamewars, costs us valuable time going on endless discussions, documenting special procedure to use those drivers (like here), etc. while we all want to be supportive to each others.
We must not support non free drivers, in any way.
ps: sorry for my bad english.
[ Parent | Reply to this comment ]
Actually I believe that open source is the non-plus-ultra for the software development, but I'm not so shure that integralism is *the* only answer to increase its diffusion.
Mainly this is due to the forces I (believe to) see behind the market (what a dirty word...) moves.
If a software house, an hardware vendor or whatever releases his software under terms of GPL or BSD or else open licence, reasons are rarely ethical, but rather economical.
NdisWrapper is not a good solution, in terms of ethical meanings, and nothing more than a patch, a roast of code making nasty operations, resolving in an hybrid between a penguin and a faceless blob behind a window. Of course implications of stability and more are clear, and I agree completely with those doesn't like closed code (since I doesn't like it, me too).
Moreover it is not an emergency plug that could be used each time you want, but preferrably to be chosed whenever there are no more choices. At the end of the day, NdisWrapper can be a good thing to save all those that doesn't have a penguin-friendly hardware. By other hand, when you have the choice to buy linux compatible hardware instead to hack with ndiswrapper, *this* is a force that can attract vendors to the penguin world.
You can happily stop to use NdisW, whenever companies starts to release their hw drivers.
This, of course, doesn't imply that they will be open.
[ Parent | Reply to this comment ]
[ Parent | Reply to this comment ]
Prerequisites:
compile your own kernel, then install it (copying, editing boot loader config, rebooting, etc.). (needs make, gcc, kernel-source, kernel-headers, and so on) so at this step you are running your arbitrary kernel, with
its source in: /usr/src/linux
and its properly installed modules in: /lib/modules/YOUR_KERNEL_VERSION/
The steps to do (debian sid, version 2.6 kernel):
# apt-get install ndiswrapper-utils-1.8 wireless-tools
# apt-get source ndiswrapper-utils-1.8
change your working directory to the dir created right now (by apt-get source): eg: # cd ndiswrapper-1.16
# make (properly installed modules needed, for further info see the file named INSTALL)
# mkdir /lib/modules/YOUR_KERNEL_VERSION/misc/
# cp driver/ndiswrapper.ko /lib/modules/YOUR_KERNEL_VERSION/misc/
# update-modules (or run depmod -a)
get the windows driver files and run ndiswrapper -i YOUR_DRIVERS_NAME (as in the article)
# modprobe ndiswrapper
If I haven't forgotten to write down something, the card now works. Enjoy.
Bence Romsics
PS: There IS an open-source driver replacing the bcmwl5(a) windows driver (still buggy, but developing) its name is bcm43xx. It is included is the kernel.org source tree since 2.6.17-rc2.
[ Parent | Reply to this comment ]
[ Send Message | View Steve's Scratchpad | View Weblogs ]
Great!
My prior websearching didn't manage to help me find it, but once 2.6.17 comes out I'll certainly switch to it and see if it works. It would definitely be easier than having to keep rebuilding the ndiswrapper module.
[ Parent | Reply to this comment ]
[ Send Message | View dkg's Scratchpad | View Weblogs ]
A friend and i tested this with his HP laptop (integrated Broadcom wireless) this weekend after a Dapper install, and it worked just fine in the environment we were in. So there's hope!
[ Parent | Reply to this comment ]
http://www.hants.lug.org.uk/cgi-bin/wiki.pl?TechTalks/3rdJune2006
--
"It's Not Magic, It's Work"
Adam
[ Parent | Reply to this comment ]
There was an impression though that we are losing sight of the fact that this is the Debian administration site, and that the idea of software freedom underpins Debian.
Understandably, a disproportionate number of Debian users are quite passionate about the freedom of software - it is one of the reasons they have chosen Debian after all.
More blurb or link to a blurb explaining the evils of using ndiswrapper according to Debian reasoning would have been good. That kind of blurb would be appropriate for any future article about usage of proprietary software in Debian.
(Aside: Some discussion seemed unnecessarily confrontational about the freedom issue - I think polite reasoning is actually more persuasive than abrasiveness with most readers here. Still, whatever the style, I find it all valuable and welcome).
PJ
[ Parent | Reply to this comment ]
I get along just fine since I've been using linux (Debian) long enough to find my way. I do think we could use tools like knetworkmanager SuSe offers. It makes connecting (including WPA) so easy. People new to linux shouldn't have to be able to edit various network files. Wifi is getting the standard and if Debian wants to keep getting more users (which is very important imho, new users could become developers) it should work on those points.
[ Parent | Reply to this comment ]
This means I don't like this way of working.
Instead I'm using the acx100 driver now for my D-Link DWL-G520+ PCI WiFi card. And even if this driver is quite young, there is still a lot of work to do, but at least it just works!
I wrote a post about the use of my D-Link DWL-G520+ PCI WiFi card with the acx100 driver on Debian. Try a real driver until you can't do something else ...
Fred
http://Linox.BE
[ Parent | Reply to this comment ]
I often end up with discarded hardware and try to utilize them as linux clusters, or install Debian and give them to relatives, charity etc. I have lots of unused video cards, printers, webcams, Pentium ones etc. from the 90's which are not supported anymore by their makers and sometimes no open source drivers or linux drivers of any kind exist for parts of them.
So I was wondering, if any solutions similar to Ndiswrapper exist for those kind of hardware? It would be a shame to throw these machines away or install Windows 95 on them..
[ Parent | Reply to this comment ]
[ Parent | Reply to this comment ]
This page was helpful to me, but it was a step in a long journey. I wrote up my findings here: shinyblue.net/itybits/wirelessnetworking/
Hope they are of some use
[ Parent | Reply to this comment ]
[ Parent | Reply to this comment ]
[ Parent | Reply to this comment ]
Being 100% linux ignorant,it has taken me 2 weeks to master ndis and apt-get alone, and apt-get is only the download side, still working on the install bit........
Without tools such as ndis and people such as Steve willing to DONATE brain use to strangers FREELY, people such as myself would be forever stuck on a Windows box, if mobile computing is part of your life.
And GOD help me if i ever wanted to use some of my testing software and equipment with Window's, $6,000 for the software alone and the linux opensource works better and was FREE, also being maintained by users, all the tips and tricks are FREE too !!
How can ANYTHING that help's people such as me, use and learn linux as against Windows, be a BAD thing?
Well Done Steve, more of the same-- us total NOOB's need all the people like you we can get.
Azoix
http://azoic.mine.nu
[ Parent | Reply to this comment ]
[ Parent | Reply to this comment ]
[ Parent | Reply to this comment ]
[ Parent | Reply to this comment ]
YOU ARE A F*CKING GOD
not even kidding, spent 7+ hours today trying to get this damned broadcom to work. NOTHING worked, and ive gotten it to work on suse and ubuntu.
i just wanna say I <3 U, and ty ty ty ty much
[ Parent | Reply to this comment ]
[ Parent | Reply to this comment ]
[ Send Message | View Steve's Scratchpad | View Weblogs ]
In June 2006 when this article was written it did ..
[ Parent | Reply to this comment ]
Thanks.
[ Parent | Reply to this comment ]