Pavel Roskin wrote:
> On Mon, 2007-04-09 at 21:12 -0500, Larry Finger wrote:
>> Pavel Roskin wrote:
>>> I realize that it's an issue with softmac, and that softmac in the
>>> kernel is not the latest.  Perhaps it should be upgraded, but I'm not
>>> very good at convincing people, and I'm vary of starting another massive
>>> tread before I catch up with my e-mail queue.
>>>
>> Would you care to elaborate? If there is any SoftMAC development outside the 
>> kernel, I know nothing 
>> about it.
> 
> More specifically:
> 
> Problem 1.
> 
> I set ESSID to one value.  The driver associates to the AP with that
> ESSID (let's call it AP1).  I set ESSID to another value.  The driver
> associates to AP2.  dhclient is not working.  I'm looking at the packets
> in Wireshark.  I see packets with ToDS set, with addr1 of AP1, addr2 of
> the Broadcom card and addr3 of AP2.  It's like the driver is sending the
> packets to the new AP via the old one!
> 
> Problem 2.
> 
> iwconfig eth1 ap some_mac_addr
> iwconfig eth1 ap auto
> iwconfig eth1 shows AP as Invalid
> 
> I know that SoftMAC in the kernel is 1.1.13 (it calls itself
> "git-1.1.13" in include/net/ieee80211.h).  The latest version released
> by Intel is 1.2.17: http://ieee80211.sourceforge.net/downloads.php

That is ieee80211, not ieee80211_softmac - a critical difference. I looked at 
the latest ieee80211 
on Intel's website. The only differences are local patches in my version, and 
code needed to handle 
older kernel versions.

> If I'll fix the kernel version, I'm likely to be fixing fixed bugs.  I
> was going to ask why SoftMAC is not being updated in the kernel, but I'm
> afraid of triggering a flamewar (e.g. if Intel wants is and others
> resist).

There will be no flamewar. Intel issued ieee80211 with a GPL and made it 
available.

There is little development on SoftMAC as it is EOL'd and will be pulled from 
the kernel as soon as 
mac80211 enters mainline.

Larry
_______________________________________________
Bcm43xx-dev mailing list
[email protected]
https://lists.berlios.de/mailman/listinfo/bcm43xx-dev

Reply via email to