[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Approaches to enabling scanning for AirPort (not Extreme)



Barry Hawkins wrote:
Luca Bigliardi - shammash wrote:
[...]
| You can find the original work here:
| http://www.kismetwireless.net/download.shtml
|
| The patch for the vanilla kernel is under "Orinoco Driver Patches for
| 2.6.9 & 2.6.10 & 2.6.8.1-mdk" section.
|
| I'm using the patch for the CVS version of the drivers and is seems
| good.
[...]
I applied this patch, and kismet seems to work now, but after using
kismet my wifi card sometimes goes into a seizure.  During the kernel
build, I noticed some warnings when the wireless drive was being
compiled; the lines are attached to the end of this message.  The
PowerBook won't even shut down, it just closes X and goes to console
mode printing a message I failed to write down message endlessly.  The
next time it happens I'll write it down.

What kernel version? Could you possibly try my 2.6.10 kernel and see if the problem
persists? I know a few people who use it (including myself) and we don't have any problems.
Also what are you running? Sarge, Sid? What version of kismet?


Also, I noticed that (prior to running kismet and the wifi freaking out)
iwlist scanning does not work:

$ sudo iwlist eth1 scanning
eth1      Interface doesn't support scanning : Operation not supported

Is there any additional tweaking or configuration in
/etc/network/interfaces in order to enable the use of iwlist in this
manner?  With that support compiled in, does it have to be explicitly
switched on?  I think waproamd needs it to be.

Kernel build warnings for driver:
~  CC [M]  drivers/net/wireless/orinoco.o
In file included from drivers/net/wireless/orinoco.c:448:
drivers/net/wireless/hermes.h: In function `hermes_present':
drivers/net/wireless/hermes.h:450: warning: passing arg 1 of `readw'
makes pointer from integer without a cast
drivers/net/wireless/hermes.h: In function `hermes_set_irqmask':

The prototype for readw()

static inline __u16 readw(volatile void __iomem *addr)

And the call itslef...

*p++ = cpu_to_le16(readw(hw->iobase + off));

iobase is an unsigned long so it seems like just casting issue and shouldn't cause any
problems other than compiler warnings (iobase + off produces an address which is all a
pointer is anyway).

--
"Imagination is more important than knowledge."
	- Albert Einstein

Attachment: signature.asc
Description: OpenPGP digital signature


Reply to: