Can you explain what you ran?Still about 10% slower than the 4GB pi5
Using single threaded runs (non OMP).
After the busybox command I'm not really seeing a difference in 4G and 8G for either:
Code:
gcc -O3 -march=native -mtune=cortex-a76 -fopenmp stream.c -o streamomp.10Mtaskset -c 0-0 ./streamomp.10M Code:
gcc -O3 -march=native -mtune=cortex-a76 stream.c -o stream.10M./stream.10M Statistics: Posted by dom — Wed May 01, 2024 6:12 pm