on 08.12.2012 00:15, KP Kirchdoerfer wrote: > Hi Erich; > > Am 06.12.2012 19:13, schrieb Erich Titl: >> Hi Folks >> >>> >>> on 26.11.2012 16:09, KP Kirchdoerfer wrote: >>> ... >>>> >>>> Hi Erich; >>>> >>>> would be great. Altough it works for me, I understand it is an issue. >>> >>> I have looked some more into the atheros wifi driver. >> >> I pushed a change to maint for the regulatory db inclusion in the kernel. > > Great; I confirm it works (besides the pb you describe below). > I hope we don't have to take too much about changes to static regdb.
To make it dynamic some more work is required like the udev stuff. The actual regdb ist a few years old too. I will push the code to my patch to maint, leave it out of the automatic build though, so it is up to the individual user to apply it. It is actually just a few lines of code. I tried to be as little intrusive as possible. > > >>> >>> For my card, it is a CM9 bought at PCengines the EEPROM returns a value >>> which tells the driver to use it's default settings. For reasons >>> whatever the driver uses the US settings as default. >>> >>> The driver code ANDs the settings, so for any (Atheros) card with an >>> EEPROM value 'default' the channels are restricted to US settings. I >>> already complained about this at linux-wireless. >>> >>> I am playing with the driver code to make WORLD the default setting. >>> This is a tricky operation, as the init code is called at multiple >>> locations. >>> >> >> I also investigated in the Atheros US default regulatory domain issue. >> It appears there is a patch for this issue in the OpenWRT code, but I >> was unable to locate it, so I wrote my own patch against the ath.ko >> module, which allows to set a different country code at load time by >> means of a module parameter. I did not push this patch though, as it >> modifies the regulatory handling of all atheros drivers that depend on >> ath.ko >> >> I can push this if you think there is demand. > >> Someone might need to port this to master. > > > I looked into to your conversation at linux-wireless to get an idea what > you have done. I understand that the way the subsystem/driver works does > not satisfy you, but also that your patch is beyond the specification. The driver patch may be, the regdb stuff, I believe, should be ported. For any card adhering to regulatory standards, e.g. setting a real country code, it works as designed. In todays world with people moving from one continent to the other, being restricted to the legislation of one single regulatory domain is indeed not satisfacctory to me. The unspecific setting on my card sends ambivalent signals, to me it means, 'I don't care, set me to what is legal in your country'. I was unable to find the reference at Atheros which says 'default is US'. I found many references in the net to get around this problem, up to the description on how to change the code on the EEPROM, IMHO a wrong aproach. cheers Erich
smime.p7s
Description: S/MIME Kryptografische Unterschrift
------------------------------------------------------------------------------ LogMeIn Rescue: Anywhere, Anytime Remote support for IT. Free Trial Remotely access PCs and mobile devices and provide instant support Improve your efficiency, and focus on delivering more value-add services Discover what IT Professionals Know. Rescue delivers http://p.sf.net/sfu/logmein_12329d2d
_______________________________________________ leaf-devel mailing list leaf-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/leaf-devel