[ofiwg] event queue overflow handling?
sean.hefty at intel.com
Mon Oct 20 16:21:07 PDT 2014
> Since there is a eq_readerr call, may not really need to be an explicit
> entry reserved? Errors could be essentially be "out-of-band" with same
> semantics as cq_read / cq_readerr ? i.e. cq_read returns ERRAVAIL == need
> to call eq_readerr. So, the EQ could still be "in hardware" but with
> metadata in host possibly indicating overflow.
That makes sense.
A really smart provider could even fail the call that might result in the EQ being overrun in the first place. Or back off whatever operations were trying to fill up the EQ (such as connection requests).
In any case, an EQ is currently defined as having a size, but the EQ size is a number of entries, which may or may not be what is wanted.
More information about the ofiwg