On 12/10/2014 05:12 PM, Robbie King wrote: > v1 - Illustrates abstracting completion event out to be a unique data > type. What is here still uses packet to serve as completion event > however application is no longer aware of the fact and simply converts > the received buffer to a completion context and uses accessor > functions from there on out. > > Subsequent version of patch will (hopefully) demonstrate how > a synchronous only implementation can be achieved relatively > effeciently using the same API. > > v2 - Added support for using thread local storage instead of an ODP buffer > for the completion event when operating synchronously. Code is not > performance oriented but meant to demonstrate how we might enable > more effecient synchronous support while allowing the application > to be implementation agnostic.
Hi Robby Can you describe what advantage does it have over my RFC? As per my understanding they both try to abstracts completion event handling in a portable way. But this one is more complex and it seems to have higher overhead on both implementation and application sides. -- Taras Kondratiuk _______________________________________________ lng-odp mailing list lng-odp@lists.linaro.org http://lists.linaro.org/mailman/listinfo/lng-odp
