hi, i’m recently studying l2 cache persistence. I found out that persisting accesses have prioritized use of this set-aside portion of L2 cache, whereas normal or streaming, accesses to global memory can only utilize this portion of L2 when it is unused by persisting accesses. The senetence is from nvidia doucment blow.
here is my questions.
Is there any specific policy for it? For instance, if persisting acceses occupy 50% of l2 cache set-aside area, I think the other 50% of l2 cache set-aside area will be occupied by streaming accesses. However if persisting acceses occupy 90% of l2 cache set-aside area, will straming accesses will still occupy the other 10% of set-aside area?
thank you in advance
1. Introduction — CUDA C++ Programming Guide.