wpa_supplicant: TNCC support reentrant ?

[r.ooba at ictec.co.jp]

>Do you mean concurrent negotiations for multiple network interfaces? 

Yes, Thank you very much for answer my questions.

-- 
Regards,
Ryuji


>On Mon, Apr 13, 2009 at 10:04:56AM +0900, r.ooba at ictec.co.jp wrote:
>
>> - Is TNCC support the multi interface ?
>>   (Wireless LAN and Wired LAN etc, for instance.)
>
>Do you mean concurrent negotiations for multiple network interfaces? I
>don't think there is any explicit support for this and in practice, this
>may show up as multiple TNCC instances trying to use the IMC at the same
>time.
>
>> - Is TNCC support reentrant ? 
>>   (Please, refer to the TNC_IFIMC_v1_2_r8.pdf, p.17,
>>   3.3 Threading and Reentrancy.)
>> 
>> - Is TNCC support thread-safe ? 
>>   (Please, refer to the TNC_IFIMC_v1_2_r8.pdf, p.40,
>>   4.2.4 Threading.)
>
>I don't think the current version would handle multiple IMC threads
>nicely, i.e., there is currently no synchronization or protection on
>data since the code in wpa_supplicant assumes everything to live in a
>single thread. Eventually, the TNCC code would need to handle this in a
>way that isolates the possibility of calls from other threads from
>confusing the core wpa_supplicant code.
>
>-- 
>Jouni Malinen                                            PGP id EFC895FA
>_______________________________________________
>HostAP mailing list
>HostAP at lists.shmoo.com
>http://lists.shmoo.com/mailman/listinfo/hostap