Re: [Rt2400-devel] rt73usb: support for wireless in Kohjinsha subnotebook

[Ivo van Doorn <ivdoorn@gmail.com>]

Hi,

> Kohjinsha subnotebook seems to contain wifi with USB IDs 0x18e8,
> 0x6206... I hope that to be compatible with
> net/wireless/rt2x00/rt73usb.c .

Is there any reason you assume it is a rt73usb device,
or are you just making wild guesses about what the chipset is?

> [Sidenote: could we add some real author info into rt73usb.c?]

You mean expanding the project name to the full list of developers in
the project? I don't see any need for that.

> I did this on 2.6.23-rc8-mm1 kernel (The RT hack may not be
> neccessary), but then it oopses somewhere in softmac layer. I'll try
> to play with it some more.
> 
> If you have any ideas, let me know.
> 								Pavel
> 
> --- clean-mm/drivers/net/wireless/rt2x00/rt73usb.c	2007-10-09 09:32:08.000000000 +0200
> +++ linux-mm/drivers/net/wireless/rt2x00/rt73usb.c	2007-10-19 22:19:45.000000000 +0200
> @@ -1580,7 +1580,6 @@
>  
>  	if (!rt2x00_rev(&rt2x00dev->chip, 0x25730)) {
>  		ERROR(rt2x00dev, "Invalid RT chipset detected.\n");
> -		return -ENODEV;
>  	}
>  
>  	if (!rt2x00_rf(&rt2x00dev->chip, RF5226) &&
> @@ -1588,7 +1587,6 @@
>  	    !rt2x00_rf(&rt2x00dev->chip, RF5225) &&
>  	    !rt2x00_rf(&rt2x00dev->chip, RF2527)) {
>  		ERROR(rt2x00dev, "Invalid RF chipset detected.\n");
> -		return -ENODEV;
>  	}

Both removals of -ENODEV are completely wrong. If the device does not pass the above
2 checks it is _not_ a rt73usb device. And forcing the driver to work with the device will
result in all kinds of problems.

If rt2x00 is loaded and detected the device, it should print out a debug message that starts with:
"Chipset detected - " What is in your case the complete line?

Ivo