Hermes driver missing?

[Donald Teed]

I never said I used 0.8 against WPA.

I used 0.8 against WEP.  And it worked but it was unstable
with truemobile 1350.

Not all broadcom devices are the same.  The chipset might
be the same, but after that there is plenty of room for
chaos.  I've seen it with these Broadcom ethernet
devices, even on Windows.

The wired broadcom devices had these failures:

5702: required 2.4.26 kernel (tg3 fixes) before udpcast
       could clone a system identically.  The maker
       of udpcast had never heard of such a failure
       associated with ethernet before.
5702: PXE boot could not obtain DHCP renewal on bootp
5705M: Program on first Windows XP boot failed to see
        network unless it was a DHCP renewal - fixed by
        sleeping (pause), and forcing a DHCP renewal
        in the Windows program that joins the notebooks
        to the domain on first boot.

Each one of the above problems was very strange to determine
why we had the symptoms we experienced.  In total I've probably
spent one man month on stupid problems introduced specifically
by Broadcom hardware.  This wireless is probably going to
boost it to two man months.  So I have the right to 
be frustrated with this company's crap.

The 1300 and 1350 truemobile report as the same
chipset (vendor ID and device ID) and yet the
1300 is stable under ndiswrapper, while the 1350
is unstable.  There are differences in how these things work.

The truemobile 1350 only, will lock up the system
while ndiswrapper is loaded, and the system is on battery
power, or when the winmodem is used.  This is
not reproduced on the truemobile 1300.  Same chipset,
different behaviours.

Anyway, I don't think that the latest problem
is on the ndiswrapper side.  It looked like that
for a moment when my networking guy gave me the
wrong info on the AP MAC address I should have been
seeing.  Now I believe the problem is on the wep_supplicant.

But thanks for your help.

--Donald Teed


On Thu, 19 Aug 2004, Giridhar Pemmasani wrote:

>
> You are unnecessarily casting aspersions against everyone and
> everything. If you have read documentation little more closely, you
> would know that ndiswrapper-0.8 doesn't support WPA.
>
> FYI, all the cards you mentioned use the same chipset. If pciid is
> same, then they are the same chipset. If you see the ndiswrapper wiki,
> you will see that many other cards use the same chipset and all of
> them work with the same driver (e.g., from Dell).
>
> If you have issues with ndiswrapper about kernel crashes, you should
> take them to ndiswrapper mailing list and give required information to
> identify the problem instead of whining here.
>
> In short: Instead of doing haphazard testing and drawing conclusions,
> read the documentation. It will save your (as well as others') time.
>
> --
> Giri
>
>