Re: [RFC] cfg80211 API for channels/bitrates, mac80211 and driver conversion

[Michael Wu <flamingice@sourmilk.net>]

[-- Attachment #1: Type: text/plain, Size: 1498 bytes --]

On Friday 04 January 2008 17:01:20 Johannes Berg wrote:
> > > + * @IEEE80211_HW_2GHZ_SHORT_PREAMBLE_INCAPABLE:
> > > + *	Hardware is not capable of receiving frames with short preamble on
> > > + *	the 2.4 GHz band.
> >
> > Let's add it when we actually find hardware that needs it.
>
> I disagree. It's literally two lines of code (one defining it and one
> using it) and makes things more easy to discover when writing a driver.
>
Yeah but it's never ever going to get used. Bet you 10 us pesos on it. So 
what's the point? If you feel this makes mac80211 better fit the theory, 
well.. whatever..

> > > +	strcpy(name, "mac80211");
> > >
> > >  	return 0;
> > >  }
> >
> > It's simple, sure, but no. Let's not change userspace facing behavior if
> > we can. If you really need to, use "IEEE 802.11".
>
> It's pretty dumb to indicate the mode that way, but I changed it to just
> "IEEE 802.11" now.
>
I'm not even sure what the intention of this ioctl is suppose to be (verify 
presence of WE? there are WE implemented ioctls that can be used for that 
trick..) but it does indicate it's not for saying what the name of the driver 
is and I don't think driver subsystem names count either. 

Indicating mode this way is dumb for sure but it fits with the examples given 
for the ioctl. Doubt anyone is counting on that though since this ioctl has 
been abused already (eg. association status in ipw2200) so "IEEE 802.11" 
should be fine.

Thanks,
-Michael Wu

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 194 bytes --]