lats: Memory Latency Benchmarking

The value of STRIDE should be chosen to defeat prefetching on the test architecture.

Example run on Intel PVC:

# Source your compilers
make sycl-usm
ONEAPI_DEVICE_SELECTOR=level_zero:0.0 ./run.sycl
# Results are in lat.csv

Name		Name	Last commit message	Last commit date
Latest commit History 30 Commits
.gitignore		.gitignore
Makefile		Makefile
README.md		README.md
cpu.c		cpu.c
cuda.cpp		cuda.cpp
hip.cpp		hip.cpp
profiler.c		profiler.c
profiler.h		profiler.h
sycl-acc.cpp		sycl-acc.cpp
sycl-usm.cpp		sycl-usm.cpp

Provide feedback