Where To Find Linux 802.11g Support Resources? 28
Matey-O writes "I just purchased a Linksys firewall/switch router (WAP54g) and two of their 802.11g cards for home use. Like MANY things linux, I found out after the fact that Linksys' 54g products use a different chipset (broadcom) rather than the intersil Prism chipsets of their linux supported 11b products. So my _main_ question is: does anybody know if/when the broadcom chipset will be supported? Google wasn't very helpful, it came up with an online petition to another Broadcom product, but neither Broadcom, not Linksys plan support for linux soon.
My secondary question is: Just where does one go to get info on 'cutting edge' linux development? I've used Linux off and on for years, but never needed to see what was going on this deep into the kernel, nor do I know where to look to see if any development is occurring for this chipset.
(meanwhile, RH8 is dormant on the laptop as the XP support is adequate.)"
mailing lists (Score:5, Informative)
Search google for lists or list plus:
linux-wlan, linux-usb, gatos, v4l
etc...
Main developers for a hardware series have mailing lists. More often than not, you can actually get their attention that way (or at least, get an issue out there).
I have had good success with the linux-wlan mailing lists for doing weird things with wireless cards in the past (granted they were all prism2 cookie-cutter). But the spirit is the same.
What I find is that because creating drivers for devices in linux can be so difficult, you really need a web page dedicated to the task. Because of this, we won't see a central site for driver development. The only centralization happens when they get chosen for bundling with the kernel.
Kernel Traffic (Score:2, Informative)
Huzzah!
Now give me a cookie.
Re:Kernel Traffic (Score:2)
The kernel Traffic pointed to by this list includes nothing on 802.11g, nor does a search on the same site come up with no results. So, I guess I should mod it -1 Offtopic, though sine it *could* be right at some point in the future, I suppose I'll just point it out in the comments.
Re:Kernel Traffic (Score:1)
Linux and G (Score:2)
My understanding is that it can be awfully tricky to build Linux support for network devices that don't release any sort of API unless you have a dedicated team willing to reverse engineer calls, decompile software, etc. I don't know if there will be a critical mass to make that happen.
Since the first company that offers Linux support in any way will obviously be the one that gets the sales from the hundreds of thousands of Linux users who want G added to their laptops and handhelds, there's a clear financial motivation, no?
Google sez... (Score:3, Informative)
And this FAQ [newswireless.net] says "the OneDriver software that ships with AirForce solutions uses the same driver for 802.11a, 802.11b, and 802.11g, so that customers of our 802.11b solution will be able to upgrade to the newer high-speed standards without changing their PCâ(TM)s software image."
Mailing list discussions [linux-wlan.com] seem to indicate that there are developers willing to write drivers to support Broadcom chipsets, but Broadcom is not "forthcoming with specs."
Just picked up 802.11g stuff today... (Score:1)
I, too, forgot to check linux compatibility before I ordered the equipment. It has been so long since I have had hardware issues with Linux that I didn't even think of checking.
I ended up breaking out an old compaq presario 1675 laptop and loading windows98 on it. After all, I got the Linksys WPC54G and WRT54G just to learn and play. My primary linux laptop (Thinkpad A21p) is tethered to so many gadgets that wireless is next to useless anyway.
But now, I find that the battery on the compaq is shot and a new one costs as much as the laptop is worth. So, I have a wireless laptop with about 20 minutes of battery until I force myself to shell out more money for
I dropped an email to linksys asking for status on the Linux support. I'm not holding my breath.
Strike at the heart of the problem (Score:1)
I've been on the wlan list since then and only get frustrated when another person comes on asking if there are drivers for it, just to get the reply I did- "No".
I did three things to try to help:
* I sent email to LinkSys, requesting information on Linux driver availability (they replied with a casual "We don't support linux" (even though you can download the wlan-ng drivers from their site). In a second letter, I explained that I would be returning the card to the retailor.
* I wrote a terrible epionion.com review of the card, explaining my situation and the way I was treated by the company (and of course posted a link to this on my website)
* Ban on LinkSys products - I will not buy a LinkSys product in the future and advise friends and family to use competing products.
I believe these are the only things we can (and must) do to get both hardware and software companies to understand that Linux is not a toy and that they should and start supporting it.
I'm no kernel hacker, but I'm sure there's no excuse for not releasing drivers for linux. If they can write a driver for Windows, it would seem to me that creating one for Linux would only be an easier task. Again, I don't know anything about driver programming, the lowest level I've worked is writing TSRs and accessing screen memory and keyboard buffers back in high school (using C, of course).
LinkSys and all these other hardware companies are not tiny firms. Hiring an open-source developer for a few months would not affect their profit. (and for the conspiracy theorists: Unless they happen to be getting subsidies by Microsoft for ignoring Linux...)
Common mistake. (Score:2)
This is a very common mistake that is made by SO many Linux users. They take it as an afront that Linux drivers are ignored.
The reality of the matter is far different. While it is true that they could develop a Linux driver rather inexpensively, the real cost is in the support.
First, remember that the Linux "market" is tiny when compared to the Windows market so, there isn't a lot of money to be made there in the first place. But manufacturers such as Linksys spend a great deal of money on the support for their products. The complexities of Linux would require more support than the Windows support does and this means more money spent.
Just imagine the ordeal that Linksys would have to go through to talk Joe Sixpack through the installation and configuration of the drivers on his Gentoo install. And immediately after that they will have to do the same thing for the next person but, this time it will be SuSE.
Simply put it just isn't worth it for Linksys et al. They have nothing against Linux, it is simply a matter of economics. Even now, with a clear policy of "We do not support Linux!" they must still dedicate resources to answering questions about Linux drivers.
Re:Common mistake. (Score:2)
Linksys wouldn't need to spend a dime on
Linux support or writing drivers, just release
their specs, and they'll get the code and the
support for free.
While the Linux market may be small, we've got
160 laptops to fit with 11g, and I can guarantee
you that we won't buy the device that only works
with Windows. The 8 or 10 laps running Linux
just cost Linksys 100% of this little market slice.
Re:Common mistake. (Score:3, Insightful)
Re:Common mistake. (Score:2)
I realize that Linksys is most likely bound by NDs, but that shouldn't preclude them from providing an unsupported, binary driver. Sure, its not the best solution, but it beats any alternative that is nonexistent.
Re:uhm... (Score:1)
Re:uhm... (Score:2)
Same deal for the WPC11, except that the switch to Broadcom happens with v3.0.
The computer stores in Halifax, NS (Canada) all have the 2.7 version, so I was out of luck. I ended up buying a D-Link 614+ AP/router, and plugging it into my ethernet instead of my PCI bus. (The 802.11b+ that the TI chipset does uses PBCC modulation at 22Mb/s, and is an optional part of the 802.11g standard, so if I get some G hardware, it can talk to my AP at 22 (if it supports PBCC22, probably have to be another TI chipset.))
Re:uhm... (Score:1)
Re:uhm... (Score:2)
you can't have bought an 802.11g card (Score:2)
Re:you can't have bought an 802.11g card (Score:1)
At this stage the spec is close to complete. The odds of these cards not being firmware upgradeable to the final spec are pretty slim.
Re:you can't have bought an 802.11g card (Score:1)
Drivers DO exist.... (Score:2, Interesting)
Consensus seems to be that if enough (potential) customers ask the product vendors (in your example, that would be Linksys, NOT Broadcom), they will produce and release drivers for linux, as Linksys has done in the past.
Note that there will probably NEVER be totally open source drivers for these. As software controlled radios, there are FCC issues. I don't pretend to understand all the issues, or agree, but I have been told essentially the same thing by several different engineers working for at least three companies.
I also don't give a rats ass if released drivers contain a closed binary only module for the software radio functions. I just want to be able to use the hardware.
A related question: is anyone working on implementing the 802.1x supplicant code which will be required to support operations using TKIP and/or CCMP, the new ciphers specified in the 802.11i DRAFT? Microsoft has devoted major resources to this, and has promised support for XP and I believe 2000. 802.1x is used for key management and distributed authentication, BTW.
Relevant standards schtuff is here [ieee.org]
Note to Slashdot Editors: Bite me. I asked virtually the same question on the 4th of February and you rejected it.
Re:Drivers DO exist.... (Score:2)
Re:Drivers DO exist.... (Score:2)