Extending HostAP to support WPA

khoh at etri.re.kr khoh at etri.re.kr
Tue Jun 3 03:58:06 EDT 2003


Thank you for your reply.

I have a card with STA firmware 1.7.4. After I read the reply,
I tried some more tests, and got the right beacons.
The first byte is the length IE, not frame type.
I still don't know what the second byte means. But zero value
of the second byte made tests successful.
The rest 26 bytes from the third are WPA IE.
My successful codes are something like followings.

char buf[28] = { 0x1a, 0x00, 0xdd, 0x18, ....}; // 0x1a==26
local->func->set_rid(local->dev, HFA384X_RID_GENERICELEMENT, 
	   buf, 28); 

> I have checked the codes you mentioned. 
> And I think you can write the codes like this: 
> char buf[26] = {0xdd, 0x18, ...} 
> local->func->set_rid(local->dev, HFA384X_RID_GENERICELEMENT,buf, 26); 
>              >			 ~~~I think 26 
> should be (26+1)*2. I donot know about it exactly, pls just try it.^_^
> You need not to tell firmware about the frame type of WPA, because RID
FC48 is 
> just for WPA IE, firmware knew about it. I set IE in AP and sniffed
frames, 
> it's good.

>>char buf[28] = { 0x80, 0x00, 0xdd, 0x18, ....}; 
>>local->func->set_rid(local->dev, HFA384X_RID_GENERICELEMENT, 
>>	   buf, 28); 
>> 
>>The first byte of 'buf' is the frame type of beacon, and 26 bytes from 
>>the third to the end are WPA IE. 
>> 
>>I caught the beacon frame in the air, and saw the IE is attached. But 
>>102 meaningless bytes are attached in the trail also. 
>>I don't know the right procedures to configure IEs in management 
>>frames. I just tried, but it is not successful. 
>>I guess, if we know the right configuration procedures of f/w 1.7.x, 
>>we can make HostAP to support WPA. 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.shmoo.com/pipermail/hostap/attachments/20030603/d2af01c1/attachment.htm 


More information about the HostAP mailing list