Thread: [Madwifi-devel] Madwifi-NG + hostapd + RSN
Status: Beta
Brought to you by:
otaku
From: Stijn T. <st...@li...> - 2006-01-30 08:02:14
|
Since madwifi-ng SVN r1357, the problem reported at http://hostap.epitest.fi/bugz/show_bug.cgi?id=75 and http://madwifi.org/ticket/132 should be solved. Madwifi-ng now reports both wpa_ie and rsn_ie. However, because of this change in the driver, RSN does not work anymore with hostapd. This is reported by me at http://madwifi.org/ticket/241. Can someone please look into this problem, and fix hostapd or maybe suggest another way too fix madwifi-ng for this matter? TIA Stijn |
From: thomas s. <t.s...@gm...> - 2006-01-30 10:29:55
|
Stijn Tintel wrote: > Since madwifi-ng SVN r1357, the problem reported at > http://hostap.epitest.fi/bugz/show_bug.cgi?id=75 and > http://madwifi.org/ticket/132 should be solved. Madwifi-ng now reports > both wpa_ie and rsn_ie. > > However, because of this change in the driver, RSN does not work anymore > with hostapd. This is reported by me at http://madwifi.org/ticket/241. > > Can someone please look into this problem, and fix hostapd or maybe > suggest another way too fix madwifi-ng for this matter? > i suggest You all put Your ressources better to a real oss driver, namely http://www.ath-driver.org/ . madwifi has surely no chance to get into the official linux kernel and the binary blob hal is suspected by at least 2 persons to damage some hardware. (indeed, my hostapd 0.4.7 is working fine with madwifi-old on a Netgear WG311T pci, but if you try to do dev work on the driver and get kernel crashes, you have to power off the laptop which has no RESET button and then some PCMCIA devices could get damaged.) my second (new and generously warranty exchanged!) D-Link DWL-G650 Cardbus is damaged now -HAL Status 3- and kernel crashs after some time using it, even with madwifi-old code :( > > TIA > Stijn tom schorpp |
From: Michael R. <ma...@no...> - 2006-01-30 11:08:05
|
Hi. On Mon, 2006-01-30 at 11:29 +0100, thomas schorpp wrote: > madwifi has surely no chance to get into the official linux kernel and > the binary blob hal is suspected by at least 2 persons to damage some > hardware. Do you have anything that proofs your accusation? I mean, something that is a bit more detailed than "MadWifi broke my card"? There could be dozens of other reasons why your card went to Nirvana, making your suspected link to MadWifi a red herring... Serious bug reports are always appreciated, suggestions for bug fixes (or, even better yet, patches that implement the suggested fix) as well. Trolling is not. Bye, Mike |
From: Stijn T. <st...@li...> - 2006-01-30 11:19:29
|
thomas schorpp wrote: > i suggest You all put Your ressources better to a real oss driver, > namely http://www.ath-driver.org/ . > I suggest that in the future, you think twice before sending a useless and destructive mail like this. This really helps NOBODY. Also, I would like to remind you that hostapd supports both madwifi-old and madwifi-ng, but ath-driver doesn't even support master mode afaik, so I wouldn't even be able to use it in any way. I guess you never thought of that... Again, your mail isn't of any use for solving this problem. Anyway, I'm don't want to start a flamewar, but I repeat, you really shouldn't send mails like this. It's like stabbing a knife in the back of every involved developer. IMO, the only thing you should do is being grateful for the work they do... and if you don't like there work? Try to do it better yourself... Just my two cents ... Stijn |
From: Paolo <oo...@us...> - 2006-01-30 14:36:46
|
On Mon, Jan 30, 2006 at 12:19:13PM +0100, Stijn Tintel wrote: > thomas schorpp wrote: > >i suggest You all put Your ressources better to a real oss driver, > >namely http://www.ath-driver.org/ . ... > shouldn't send mails like this. It's like stabbing a knife in the back > of every involved developer. IMO, the only thing you should do is being com'on, don't be so tragic ;) - also, that poster seems misinformed, as it looks like the ath-driver.org might be more a complement (the HAL stuff, like openHAL) to madwifi, than a replacement. That is, it'd offer an alternative OS HAL. If I got it right :) -- paolo |
From: Rasto S. <ra...@ks...> - 2006-01-30 15:30:04
|
> > thomas schorpp wrote: > > >i suggest You all put Your ressources better to a real oss driver, > > >namely http://www.ath-driver.org/ . > ... > > shouldn't send mails like this. It's like stabbing a knife in the > > back > > of every involved developer. IMO, the only thing you should do is > > being > > com'on, don't be so tragic ;) - also, that poster seems misinformed, > as > it looks like the ath-driver.org might be more a complement (the HAL > stuff, like openHAL) to madwifi, than a replacement. That is, it'd > offer > an alternative OS HAL. If I got it right :) > > -- paolo Ath-driver is not much more than a basic open-source HAL so far (I believe), and I'm not very happy about the closed-source HAL in Madwifi either. I'm probably not going to help much, so I should mind my own business, but would it be a fundamental problem to use ath-driver as HAL in Madwifi? As a complement to the closed-source HAL of course. If both projects cooperate? Rasto |
From: Jouni M. <jkm...@cc...> - 2006-01-30 14:17:56
|
On Mon, Jan 30, 2006 at 09:02:07AM +0100, Stijn Tintel wrote: > Since madwifi-ng SVN r1357, the problem reported at > http://hostap.epitest.fi/bugz/show_bug.cgi?id=75 and > http://madwifi.org/ticket/132 should be solved. Madwifi-ng now reports > both wpa_ie and rsn_ie. > > However, because of this change in the driver, RSN does not work anymore > with hostapd. This is reported by me at http://madwifi.org/ticket/241. Thanks for the reminder. This is on my to-do list, but I have not yet had a chance to take a closer look at the problem. Both IEs must be synchronized between the driver and hostapd to get all security combinations working properly. -- Jouni Malinen PGP id EFC895FA |