[openib-general] Handling SM class (SMInfo vs. other queries)

[Michael S. Tsirkin]

Hello!
Quoting r. Roland Dreier (roland at topspin.com) "Re: [openib-general] Handling SM class (SMInfo vs. other queries)":
>     Michael> I think the difficulty with the last one is that (at
>     Michael> least for Tavor) process local mad can block, since there
>     Michael> is a limited number of outstanding commands. Of course
>     Michael> you could always make it non-blocking by dropping the MAD
>     Michael> if the command interface is busy ... no idea if this will
>     Michael> improve or decease performance, though.
> 
> Not sure if I understand what the issue is... if the interface is
> asynchronous, if there is not room for another outstanding command,
> it seems one can just queue the command until a slot becomes open.
> 
>  - Roland

I was thinking in terms of gen1 code I guess.
What I was trying to say I think is that you'll need some kind of limit
to avoid exhausing kernel memory if there are lots of MADs.

MST