[PATCH 0/7] P2PS fixes based on http://w1.fi/p/p2ps-pending/

Jouni Malinen j at w1.fi
Thu Aug 6 14:17:37 EDT 2015


On Thu, Jul 30, 2015 at 08:59:59AM +0300, Ilan Peer wrote:
> The only fix for this was in "Fix PD PIN event notifications" patch.
> 
> Tested this series on hwsim (with and without VMs) using
> the latest mac80211-next tree.

Thanks, partially applied with some changes.

> Andrei Otcheretianski (5):
>   tests: Use p2ps_provision() and p2ps_connect_pd() in
>     p2ps_connect_p2ps_method()

Applied.

>   tests: Clear scan cache on reset
>   tests: Few trivial fixes to p2ps tests

I dropped these two since I don't want to add the latency and extra
command between every single test case when only a small minority of
test cases depending on this. If you can identify test case sequences
where you see issues with old scan results, the affected test cases can
be addressed on case by case basis.

>   tests: Re-factor PD and connection flows in p2ps tests

Applied.

>   P2PS: Authorize any peer for p2ps method

I agree with the P2P Device Address being incorrect in the earlier
implementation, but I don't see why authorizing any peer would be a good
resolution for it. I modified this patch to fix the address selection,
i.e., use P2P Interface Address, not P2P Device Address and only if the
intended interface address is not available, authorize any address. That
said, I don't really like that either, but I wanted to get this patch in
to get this set finally through. I think it would be good to try to see
if the intended interface address can be made available in more cases
and if not, add code to make WPS Registrar reject peer based on P2P
Device Address in cases where any interface address had to be
authorized.

> Max Stepanov (2):
>   P2PS: Fix PD PIN event notifications
>   tests: Add P2PS PD feature capability CPT tests

Applied.

-- 
Jouni Malinen                                            PGP id EFC895FA


More information about the HostAP mailing list