Re: [Madwifi-devel] ath_rate in DadWifi - broken or unneeded?
Status: Beta
Brought to you by:
otaku
From: Nick K. <mic...@gm...> - 2007-01-18 11:01:53
|
Sorry ;-( i couldn't check madwifi.org because for some reason it didn't load where i am. I just did a c/p to have register names handy (i've also placed a comment there) didn't realy implemented it. It seems that set_bssid_mask is more complex, the function gets an 8 bit integer while the 2 registers are 32 bits (lower) and 16 bits (higher), a bssid is 48 bit long, so that makes 32 (first register) + 16 bits (second register) = 48bits. I guess your implementation is right, haven't realy looked into it. Since it's a VAP issue, it doesn't have anything to do with dadwifi (since dadwifi doesn't support VAPs, these are handled on hostapd), a normal bssid mask is written by the HAL when you use set_associd so set_bssid_mask should not be used. It's also a functionality that only 5212+ chips have. 2007/1/18, Pavel Roskin <pr...@gn...>: > On Thu, 2007-01-18 at 09:25 +0200, Nick Kossifidis wrote: > > > My mistake ! I'll fix it asap ;-) > > Please check my fix :) > > Yes, dadwifi-openhal compiles now without any warnings. I haven't tried > to run it yet. > > It looks like HAL needs to be heavily checked by sparse for endianess, > but sparse doesn't understand more than one -include, which is > unfortunately the case for MadWifi. A patch for sparse was posted > today. > > -- > Regards, > Pavel Roskin > > > -- As you read this post global entropy rises. Have Fun ;-) Nick |