[Prism54-devel] Re: Prism54 development update

[Jean-Baptiste Note]

Johannes Steingraeber <Jo_Stein at web.de> said :
> On Wed, Jan 05, 2005 at 10:35:31PM +0100, Jean-Baptiste Note wrote:
>> [...] From what we understand of the windows usb drivers, in the
>> softmac version, the net2280 is not orthogonal to the prism chipset
>> operation
>
> I would like to learn what is the rationale for this?

Bad design, maybe. I'm not saying this is how i'd do it (in fact i've
said i would've designed it orthogonal) but maybe they had no choice.

>
>> : softmac firmware uploaded to the device is not the same as
>> in the pci cards.
>
> Of course it's different, the hardware is different, too. Obviously
> the softmac left most of (or all?) the 802.11 related work to be done
> by the PC. But from my point of view this does not mean that the
> net2280 chip is used for anything else than bridging between USB and
> PCI.
>

Well first it seems firmwares for fullmac chipsets are universal. Why
would it be different for softmac chips ?

Secondly, you may think that for my linksys device, which is a net2280 +
fullmac chip, the "softmac firmware" uploaded to it would be found in
PCI softmac drivers if the net2280 only did bridging. This does not seem
to be the case.

For full understanding of the issue, i'm copying a mail by Sebastien
Bourdeauducq, sent off-line when the list was down (i hope you don't
mind :) :

<quote sebastien>

Hello,
Sorry for sending you this message this way, but the prism54-devel mailing 
list seems down.
Alexandre Strube reported his Sitecom WL-121 PCI card to be ISL3886-based. The 
firmware of this device is the same version (2.4.6.0) as my USB device and 
the two firmware's build dates differ only by a few minutes. But there are 
notorious differences between the contents of the files (maybe for the 
communicating with the NET2280 firmware ?). My USB device doesn't boot at all 
using the PCI firmware. They are really different.
As usual, download the firmware from 
http://c.bourdeauducq.free.fr/prism54.html
Sebastien

</quote>

But we'd really, really be happy to be proven wrong, but in this case,
i'd say we're defendent (there's no reason a priori to think that the
net2280 only does bridging, except faith in good design and possibility
to do so).

Regards,
JB
-- 
Jean-Baptiste Note
+33 (0)6 83 03 42 38
jean-baptiste.note at wanadoo.fr