> If the protocol allow more thant one transitions taken in the same
> cycle, could the second transtion see the new state of the cacheline changed
> by the first transition?
Yes.
> If so, the protocol need to take many special conditions into account.
>
Not really. Most the time it is no different then if the message causing
the transition arrived 1 cycle later.
> If a new event putting into the EventQueue with zero delay and zero latency,
> does it means it will happen in the same cycle which generate it ?
>
In theory, yes. But we currently don't allow that. If you remove the
asserts, things might work or they might be entirely unpredictable. I've
never tried it.
|