[PATCH 00/24] Modify tests to properly work with P2P Device as well

[Peer, Ilan]

> -----Original Message-----
> From: hostap-bounces at lists.shmoo.com [mailto:hostap-
> bounces at lists.shmoo.com] On Behalf Of Jouni Malinen
> Sent: Thursday, February 05, 2015 14:11
> To: hostap at lists.shmoo.com
> Subject: Re: [PATCH 00/24] Modify tests to properly work with P2P Device as
> well
> 
> On Wed, Feb 04, 2015 at 04:30:12AM -0500, Ilan Peer wrote:
> > In our local setups we are using hwsim with P2P Device enabled. The
> > following patch set includes modifications for the tests cases to
> > allow them to pass also when P2P Device is supported. These mostly
> > consist of:
> >
> > 1. Using the global control interface for P2P commands and
> >    events.
> > 2. Using the relevant group interface name when needed.
> > 3. Resolve cases that need to handle with multi interface
> >    scenarios, e.g., assuming that the p2p_dev_addr == own_addr
> >
> > Verified that the tests are passing also when the P2P Device is not
> > enabled.
> 
> Thanks. I applied patches 1, 5, 8-15, 18-24. Patch 16 ('tests: Use global control
> interface in test_nfc_p2p', I'd assume) did not make it to the mailing list.
> Patches 2-4, 6-7, and 17 need changes as indicated (or more justification to
> make me understand the reasoning for couple of the items).
> 

Thanks. I'll handle the comments and send another set.

> As a side note, it would be good to add new test cases that use P2P Device
> enabled for areas that do not get full test coverage today. I understand that
> this can be done by modifying support_p2p_device=0 parameter in the
> scripts, but that does not get executed automatically at least in my test
> setups and running the full set of tests twice with two values for that would
> result in significant duplication of testing, so that's not going to happen either
> due to the extra time it would take.
> 

Sure, will try to get to it as well. Regardless, we are taking the path of using support_p2p_device=1 as this is the only configuration supported by our devices, and we would like to stress issues that can be found when using the P2P Device interface as well as other possible issues that might rise since we are always using multi interface configuration. FWIW, we are running the testing framework as part of our daily regression so in case any issues pop up with the P2P Device we'll fix them. 

Regards,

Ilan.