eap Mailing List: Re: Issue 193: Peer SM review

[Jari Arkko (jari.arkko]

Pasi.Eronen [at] nokia.com wrote:

> Of these, methodState, selectedMethod, and decision are already 
> initialized (though the exact value for "decision" was not
> exactly right). 
>
> However, eapSuccess and eapFail must be initialized by 
> the lower layer; otherwise there's a race condition.
>
> I'm not not quite sure about eapKeyData and eapKeyAvailable.
> I would prefer if they were also initialized by the lower layer
> (in general, variables that trigger a transition in some 
> other state machine must be reset in the state machine that 
> reads them; otherwise some extra synchronization is needed),
> but possibly eapSuccess provides the needed "extra synchronization".

Right. I support this approach. Do not rely on eapSuccess
to provide the extra sync.

--Jari