off-chip memory access

Hello All,

I was wondering if there is a way to say which off-chip memory address and controller was used to access data when there is a cache miss. Any lights in this regard will be really helpful.
Off-Chip memory = GDDR RAM on GPGPU.


I guess there is no such function in cuda-gdb, cuda profiler can tell you how many times cache miss and hit happens, but I don’t know if there is a way to find out from which real memory address the app reads when cache miss happens.

Best Regards