I am not sure where to go from here. I have wired internet but I am unable to get my wireless card to work at all. I can find it in sys. info and that is about it. I am very new to linux and so I am sure this is to some degree operator error as much as anything else. Please give me any suggestions that you have regarding how to repair this. I am running openSUSE 11.0 as a dual install with Vista and I have wireless through Vista. I am not sure what other information would be helpful so I wait to post anymore. Thanks ahead of time for all of the help!!!
Just posting the fact that it is a BCM4310 is enough. Unfortunately, your card
is one of the few that the open-source driver does not support. In your case,
you will need to use ndiswrapper. There are a number of places to learn how to
do that.
FYI, we of the bcm43xx team are currently doing the reverse engineering of this
device, but it will not be ready for some time.
In the future when seeking help for a device that is not working, you should
post the appropriate section of the output of ‘/sbin/lspci -v’ and the pertinent
data from the output of the ‘dmesg’ command.
I actually have read quite a few of your posts and I saw that you were developing exactly what I am looking for. I did not want to sound like the noob that I am but I tried to get the output that you are talking about from the konsole and I could not get it to output the information that I saw everyone else posting when they have a problem like this. I would love to know what resource you think would be the best to learn how to work more in depth with linux, or more importantly to me in the short term, to solve my current problem. I have already installed the ndiswrapper however I cannot seem to get it to do what others have. I am reading the posts that many other people with this problem have put up and I am simply missing something in my process. I also tried a few other things but I think that perhaps they are a lost cause. ANY help that you can offer would be great, and once again thank you for your time and your patience!
The BCM4310 is not recognized by the driver, thus it doesn’t even provide the
log output that we ask others to provide.
As to implementing ndiswrapper, I have not used it for so long that I cannot
help you. I suggest that you go to the ndiswrapper home site
(http://ndiswrapper.sourceforge.net/joomla/) and look at the documentation. If
something doesn’t work, their mailing list is the proper place to ask.
Larry Finger wrote:
> As to implementing ndiswrapper, I have not used it for so long that I
> cannot help you. I suggest that you go to the ndiswrapper home site
> (http://ndiswrapper.sourceforge.net/joomla/) and look at the
> documentation. If something doesn’t work, their mailing list is the proper
> place to ask.
I would add again that I’ve been unable to get ndiswrapper+bcmwl5.inf to
work in 11.0 but it’s fine in 10.3.
–
Graham P Davis, Bracknell, Berks., UK. E-mail: newsman not newsboy
I just completed some edits to SUSE’s ndiswrapper documentation at:
Ndiswrapper - openSUSE
concerning 32-bit vs 64-bit drivers and blacklisting conflicting native drivers.
If you weren’t aware of those concerns, take another look there.
Hope this helps…
Dave
cookdav wrote:
>
> Graham P Davis;1835347 Wrote:
>>
>> I would add again that I’ve been unable to get ndiswrapper+bcmwl5.inf
>> to
>> work in 11.0 but it’s fine in 10.3.
>> –
>> Graham P Davis, Bracknell, Berks., UK. E-mail: newsman not newsboy
>
> I just completed some edits to SUSE’s ndiswrapper documentation at:
> ‘Ndiswrapper - openSUSE’ (http://en.opensuse.org/Ndiswrapper)
> concerning 32-bit vs 64-bit drivers and blacklisting conflicting native
> drivers.
> If you weren’t aware of those concerns, take another look there.
>
> Hope this helps…
>
Thanks Dave, but I’ve already checked that there were no conflicting
drivers. As I say, the machine is working fine with 10.3 (64-bit). When
I’ve been booting the machine with 11.0 and ndiswrapper, the machine will
sometimes re-boot directly after loading the driver - it won’t boot
successfully unless it fails to boot the driver. Here is what Larry had to
say on the subject a couple of weeks ago:
"Welcome to the Linux equivalent of “Blue Screens of Death” while
running the Windows driver with ndiswrapper. Unfortunately, I don’t
have much to offer you. Broadcom does a really lousy job of error
checking in their drivers, which is why you are getting the crashes. I
know this because I’m currently working on the reverse engineering for
the BCM4310.
At the moment, roughly 1/3 of the routines needed to make the BCM4310
and BCM4328 interfaces work have been translated. Once that is done,
we will write the specifications for the LP and N PHYs and post them
at http://bcm-v4.sipsolutions.net/. Finally, the code-writing part of
the team will prepare the Linux driver. This is an involved process,
but it works - just takes a long time."
I should probably wait for the work on the driver to be completed but I just
like tinkering, so I’m also playing with the ipw2200 driver at the moment.
Unfortunately, I haven’t got the device (an Intel Pro/Wireless 2200BG)
recognised so far.
–
Graham P Davis, Bracknell, Berks., UK. E-mail: newsman not newsboy
Maybe I am missing something but if I understand the above posts then there is currently no solution for my problem? At least not with ndiswrapper. I am debating on trying UBUNTO but when I run the live CD I still have the problem with not being able to use the wireless on my laptop. It seems however that there are a few solutions people have suggested and that they have said worked in UBUNTO. Does anyone have a new solution for me if ndiswrapper won’t work. If not should I try UBUNTO? Thanks again guys, even if this is not working I am learning a ton just by fighting with it.
If ndiswrapper will not work for you with openSUSE 11.0, it is unlikely to work
with Ubuntu (note spelling). I don’t know if your system is 32- or 64-bit. With
ndiswrapper, 32-bit is more likely to work as the 32-bit Windows drivers are
used a lot more than the 64-bit versions. The same is likely true for ndiswrapper.
As I wrote earlier, the open-source driver is a ways off.
Larry
I got the impression from what others were posting that ndiswrapper was not an option with openSUSE 11.0, is this correct? I know that you initially said and since you seem to be the best authority on the topic that I have found I am going to give that another go in the next couple of evenings unless you know of a distro that would be more fit as my first Linux OS. The system is 32-bit. Thanks for the Ubuntu tip as well, I would have been disappointed if I had gone through the trouble of installing it on the system and then still had it not work properly. Thanks again for all of the help.
I don’t use ndiswrapper as a matter of principle. I do a lot of
latest-and-not-the-greatest kernel testing to help find and fix problems during
the pre-release phase. If I were to taint my kernel with ndiswrapper, none of my
bug reports would be worth anything, so I don’t use it. I have a BCM4310 on the
shelf that I cannot use at the moment, but for me that is no big deal as I have
two other PCIe cards - a BCM4311 and a BCM4312 that I can use instead. When the
reverse engineering is completed, I’ll pop that card back in for testing.
With a 32-bit system, ndiswrapper should work. The main difficulty happens with
64-bit code. Neither ndiswrapper nor the Windows drivers have been used very
much at 64 bits.
Larry
In theory, one should NOT have to go to the trouble of installing it to find out
whether the wireless is going to work. That’s what a LiveCD is for!
[openSUSE just doesn’t quite ‘get that’ yet.]
That’s the philosopy of the MEPIS Linux distro, which has, by far, the best record of
having wireless work right from the LiveCD. [They pre-install a dozen wireless
drivers with ndiswrapper, and at least twice that many native drivers, firmware and all.]
So, if you wanna try, go to
MEPIS | Now Shipping Version 7.0 and click on the
download link, burn it, and take their v7.0 LiveCD
for a test-drive on your hardwire and see.
My 2-cents…
Dave [a true Linux distro-junkie]
cookdav wrote:
>
> In theory, one should NOT have to go to the trouble of installing it to
> find out
> whether the wireless is going to work. That’s what a LiveCD is for!
> [openSUSE just doesn’t quite ‘get that’ yet.]
>
> That’s the philosopy of the MEPIS Linux distro, which has, by far, the
> best record of
> having wireless work right from the LiveCD. So, if you wanna try, go
> to
> ‘MEPIS | Now Shipping Version 7.0’ (http://www.mepis.org) and click on
> the
> download link, burn it, and take their v7.0 LiveCD
> for a test-drive on your hardwire and see.
If Mepis is distribution Broadcom firmware, they are asking for trouble.
Broadcom is very insistent that this is their IP and very stubborn about it.
Larry
In my view, Mepis is in no more trouble than you or Novell is.
When you post the cmd ‘install_bcm43xx_firmware’, and a user
executes it, absolutely NO attribution is given to ‘Broadcom’,
and no license-statement is shown. So, how is that distribution any different
than someone downloading a distro containing that same firmware?
[Contrast that to what happens when one installs the Nvidia driver…each
user is presented a license-agreement with usage restrictions that they
agree to, which is necessary to make any usage restriction legally binding
(in my opinion…I am not a lawyer).]
I’ll quote once more from overiew section of:
First-sale doctrine - Wikipedia, the free encyclopedia
“So, for example, if the copyright owner licenses someone to make a copy (such as by downloading), then that copy (meaning the tangible medium of expression onto which it was copied under license, be it a hard drive or removable storage medium) may lawfully be sold, lent, traded, or given away.”
So, you’re either both innocent or both guilty…I’ll let you decide which.
My 2-cents worth…
Dave
Well, I got it to work!!! It took all of last night from 6-2 but once again I learned a ton and it ended up working. I ended up using ndiswrapper as was initially suggested, thank you very much, and also a driver I found on a walk through of sorts (the walk through was on a very similar system and later I will go through my history and post the link, it was great!) but I had to try two different drivers before I realized I had made a few errors along the way. Anyway I won’t go into details but thank you everyone for all of your help especially lwfinger!!!
What I used: ndiswrapper 1.53
ndiswrapper -l (shows output below)
bcmwl5 : driver installed
device (14E4:4315) present
Driver gotten from: Linux on the HP Pavillion dv6835nr
then extracted along with bcm5wl.sys into tmp folder under root.
I would be more than happy to go into more detail if anyone needs help or has questions.
cookdav wrote:
> lwfinger;1836340 Wrote:
>> If Mepis is distribution Broadcom firmware, they are asking for
>> trouble.
>> Broadcom is very insistent that this is their IP and very stubborn
>> about it.
>>
>> Larry
>
> In my view, Mepis is in no more trouble than you or Novell is.
> When you post the cmd ‘install_bcm43xx_firmware’, and a user
> executes it, absolutely NO attribution is given to ‘Broadcom’,
> and no license-statement is shown. So, how is that distribution any
> different
> than someone downloading a distro containing that same firmware?
>
> [Contrast that to what happens when one installs the Nvidia
> driver…each
> user is presented a license-agreement with usage restrictions that they
> agree to, which is necessary to make any usage restriction legally
> binding
> (in my opinion…I am not a lawyer).]
First of all, Mepis is using ndiswrapper. Whether they have the right to
redistribute the Windows driver(s) is between their lawyers and those of the
various vendors.
As for the firmware extracted from the Broadcom drivers using
‘install_bcm43xx_firmware’, the script downloads a driver from a public web site
and extracts the firmware. There is absolutely no redistribution of any
copyrighted material. That is the essential difference! If we were to extract
the firmware and put it on a site somewhere, then that would not be legal. The
basis for the script is GPL licensed, thus Novell can include it if they choose.
One more thing, there is a very useful and informative walk through here:
http://wiki.linuxquestions.org/wiki/NDIS_Wrapper
, it helped me when I was lost at a couple of points and someone might find it useful! Thanks again lwfinger and everyone else, I am ecstatic to finally have this resolved!
Those vendors PUBLISH a website, where one can download their Windows
driver, with no restrictions!
So, therefore, the downloader can legally redistribute.
Ditto for the firmware. Downloading is redistributing, no matter who pushes the
button, or when it is pushed.
[One of us is still missing something.]
cookdav wrote:
> Those vendors PUBLISH a website, where one can download their Windows
> driver, with no restrictions!
> So, therefore, the downloader can legally redistribute.
Not true. You can download things for your personal use, but you are not allowed
to distribute it to anyone else. Check copyright law.
> Ditto for the firmware. Downloading is redistributing, no matter who
> pushes the
> button, or when it is pushed.
When it is on a website, there is a presumption of having the right to
distribute. In the case of Windows drivers, there is no question. The Broadcom
version of the Linux drivers are available because Broadcom got caught violating
the Linux GPL on the operating software for the home routers (Linksys, etc.).
That is why the OpenWRT site has the right to redistribute, but once again,
after you do the download, you can only have the right for personal use. Making
it available to the general public is not allowed.
> [One of us is still missing something.]
Agreed.
>
Hmmm…I guess you aren’t reading the first sale doctrine (the same way I do).
As stated on Wikipedia, it is a LIMITATION of copyright law.
‘Personal copy’ is a temporary concept. I can download my one (personal) copy. Then, according to first sale doctrine, I can sell or give it away. (i.e. re-distribute it).
So, now, if you want Ubuntu, SUSE, and Mepis, etc, to stress your download-server to get their next copy(ies), they’ll all gladly comply. [But, I’m betting you’ll forego
that requirement, and just let the cloning proceed.:)]