[ofa-general] Re: [PATCHv3 for-2.6.21] IB/mthca: fix race in QP destroy

[Roland Dreier]

 > Yes but here we also must make sure completion events and async events are
 > flushed out: once QP is in reset no events should be generated.

Completion events are fine -- at worst the consumer gets a spurious
event but doesn't find any CQEs.  I see the point about async events.
It does make sense to merge the async event and command EQs to get
good ordering -- splitting them doesn't make much performance
difference I guess.

 > > I don't get it -- how do you handle having multiple completion EQs
 > > with this idea?
 > 
 > You are right here.
 > But the point is that multiple completion EQs are not 2.6.21 material.
 > OK?

I don't want to put a fix in that we have to back out later.  Let's
figure out how to fix this properly in the first place.  But I think
just merging async events and command completions into one EQ suffices.

 - R.