The idea is to backport the driver for the 2.4 linux kernel. With some
reverse engineering I have found I can extract the firmware from the
driver.
I think there are many benefits and it deserves the effort.
bye
About the fact that monitor mode would be useful, in that case I think
the best thing to do is simply to install openwrt on it.
On Aug 28, 11:56 am, "Marco Vedovati" <marco....@gmail.com> wrote:
> I have uploaded the sources of the wireless drivers on google groups:http://groups.google.it/group/neptune354-dev/web/AP-DK5.7.0.4.tar.gz
>
> About the fact that monitor mode would be useful, in that case I think
> the best thing to do is simply to install openwrt on it.
>
> On 8/28/07, Jim Halfpenny <jim.halfpe...@gmail.com> wrote:
>
> > Monitor mode is particularly interesting to me. If you wanted to deploy a
> > WAG354G as something other than an access point e.g. a kismet drone, monitor
> > mode would be essential. If WPA is not supported it's pretty useless for
> > normal use though.
>
> > Jim
>
> > On 27/08/07, Marco Vedovati <marco....@gmail.com> wrote:
>
> > > hi, i dont see what are the benefits of using the acx100 drivers,
> > > apart the fact that they are open source. I had some experiences with
> > > those drivers and openwrt, they only support open and wep encryption.
> > > The good thing is that they fully support linux wireless extension and
> > > they can be used in client/monitor mode.
> > > I have sources of the apdk drivers used in wag354g, but i'm not able
> > > to compile them...
> > > If you need ask, or maybe i'll post them later....
>
> > > bye
>
Wirelesstools, as expected, don't do something special. The original
driver and mine come from the same codebase. We have to implement the
wireless extensions to the driver to give this functionality. Look
here: AP-DK5.7.0.4/src/os/linux/dda_wext_ioctl.c:123
I take the challenge but firstly I have to know the driver I compile
_really_ works.
Running ifconfig wlan0 up gives:
wlan0 Link encap:Ethernet HWaddr 00:00:00:00:00:00
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:100
RX bytes:0 (0.0 B) TX bytes:0 (0.0 B)
No hardware address. Trying to set it manually fails like this
~ # ifconfig wlan0 hw ether 00:01:02:03:04:05
SIOCSIFHWADDR: Device or resource busy
wlcfg segfaults with my driver. In my try to reverse engineer wlcfg I
compiled strace (download here http://l1nux.h4ck3r.googlepages.com/strace
) but wlcfg forks much and running strace -f wlcfg start (also trace
forks) freezes the whole process.
Help or ideas needed.
On Sep 1, 3:51 pm, johnath...@gmail.com wrote:
> Ok I 'll try them. I have uploaded the wireless driver I compiled
> here:http://l1nux.h4ck3r.googlepages.com/tiap.oif anyone wants to
wlcfg stop
insmod tiap.o
ifconfig wlan0 hw ether 00:01:02:03:04:05
ifconfig wlan0 up
brctl addif br0 wlan0
But wireless still don't work... Wireless led is off. Still something
missing. Does anyone know what?
On Sep 3, 1:44 pm, johnath...@gmail.com wrote:
> Ok something strange happens. I have tested the wireless driver
> unstripped and commands like ifconfig wlan0 up work. The driver I
> uploaded was stripped and doesn't work!! I reupload the unstripped so
> download it again.
>
> Wirelesstools, as expected, don't do something special. The original
> driver and mine come from the same codebase. We have to implement the
> wireless extensions to the driver to give this functionality. Look
> here: AP-DK5.7.0.4/src/os/linux/dda_wext_ioctl.c:123
>
> I take the challenge but firstly I have to know the driver I compile
> _really_ works.
> Running ifconfig wlan0 up gives:
>
> wlan0 Link encap:Ethernet HWaddr 00:00:00:00:00:00
> UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
> RX packets:0 errors:0 dropped:0 overruns:0 frame:0
> TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
> collisions:0 txqueuelen:100
> RX bytes:0 (0.0 B) TX bytes:0 (0.0 B)
>
> No hardware address. Trying to set it manually fails like this
>
> ~ # ifconfig wlan0 hw ether 00:01:02:03:04:05
> SIOCSIFHWADDR: Device or resource busy
>
> wlcfg segfaults with my driver. In my try to reverse engineer wlcfg I
> compiled strace (download herehttp://l1nux.h4ck3r.googlepages.com/strace
> ) but wlcfg forks much and running strace -f wlcfg start (also trace
> forks) freezes the whole process.
>
> Help or ideas needed.
>
> On Sep 1, 3:51 pm, johnath...@gmail.com wrote:
>
> > Ok I 'll try them. I have uploaded the wireless driver I compiled
> > here:http://l1nux.h4ck3r.googlepages.com/tiap.oifanyone wants to
Maybe I have found something interesting. Look at the file cfg/cli.c .
It is described as "Demo application console commands". And yes, it
contains a int main(int argc, char **argv) !
Try to compile it and let's see what happens...in the main Makefile
there also reference to other programs, take a look...
good luck
wlcfg stop
./init /usr/sbin/wpa_authenticator /lib/modules/2.4.17_mvl21-malta-
mips_fp_le/kernel/drivers/net/tiap.o wlan0 br0
Now you can run cli to configure and test everything. It rocks!
Unfortunately when I try to use my driver I get a segmentation fault.
I think it has to do with some compilation flags, maybe -fomit-frame-
pointer. I will keep trying to make it work.
For now I have uploaded the tools if somebody wants to experiment:
http://l1nux.h4ck3r.googlepages.com/init
http://l1nux.h4ck3r.googlepages.com/cli
On Sep 3, 8:34 pm, "Marco Vedovati" <marco....@gmail.com> wrote:
> yes, what you do is ok for the layer 3. It lacks of the configuration
> of the network level, eg channel to use, mode (managed or master) and
> encryption.... that is, the work that wlcfg does.
>
> Maybe I have found something interesting. Look at the file cfg/cli.c .
> It is described as "Demo application console commands". And yes, it
> contains a int main(int argc, char **argv) !
>
> Try to compile it and let's see what happens...in the main Makefile
> there also reference to other programs, take a look...
>
> good luck
>
> > > > here:http://l1nux.h4ck3r.googlepages.com/tiap.oifanyonewants to
On 6 Set, 11:48, johnath...@gmail.com wrote:
> Marco was completely right. Insmoding the driver is not enough.
> Initialization is needed. Compiling cfg folder creates 2 tools. Init
> and cli. Init initialize the wireless driver and wireless works for
> the original driver doing this:
>
> wlcfg stop
> ./init /usr/sbin/wpa_authenticator /lib/modules/2.4.17_mvl21-malta-
> mips_fp_le/kernel/drivers/net/tiap.o wlan0 br0
>
> Now you can run cli to configure and test everything. It rocks!
>
> Unfortunately when I try to use my driver I get a segmentation fault.
> I think it has to do with some compilation flags, maybe -fomit-frame-
> pointer. I will keep trying to make it work.
>
> For now I have uploaded the tools if somebody wants to experiment:
>
> http://l1nux.h4ck3r.googlepages.com/inithttp://l1nux.h4ck3r.googlepages.com/cli
Have fun,
John
On Sep 7, 6:23 pm, cyberstorm <marco....@gmail.com> wrote:
> i see there a lot of parameters to configure in the cli app. However
> the only useful I have seen so far is the possibility to use a radius
> server...
> I hope there is at least a way to set the wireless in client mode.
>
> On 6 Set, 11:48, johnath...@gmail.com wrote:
>
> > Marco was completely right. Insmoding the driver is not enough.
> > Initialization is needed. Compiling cfg folder creates 2 tools. Init
> > and cli. Init initialize the wireless driver and wireless works for
> > the original driver doing this:
>
> > wlcfg stop
> > ./init /usr/sbin/wpa_authenticator /lib/modules/2.4.17_mvl21-malta-
> > mips_fp_le/kernel/drivers/net/tiap.o wlan0 br0
>
> > Now you can run cli to configure and test everything. It rocks!
>
> > Unfortunately when I try to use my driver I get a segmentation fault.
> > I think it has to do with some compilation flags, maybe -fomit-frame-
> > pointer. I will keep trying to make it work.
>
> > For now I have uploaded the tools if somebody wants to experiment:
>
> >http://l1nux.h4ck3r.googlepages.com/inithttp://l1nux.h4ck3r.googlepag...
best regards,
marco
To compile successfully you have to edit make/Config.make. Change the
variables KERNEL_DIR_PRE, UCLIBCDIR and GCCINCLUDEDIR to what fits
for you.
Have fun,
John
On Sep 19, 10:18 am, "Marco Vedovati" <marco....@gmail.com> wrote:
> Great work johnathana!
> If you have done some changes in the source code, please make it
> available so that others can go through for changes.
>
> best regards,
> marco
>