[Quy Anh] «Elliot» Nguyen
42 lines
1.2 KiB
Plaintext
42 lines
1.2 KiB
Plaintext
rm -f main_* sizes run_tests *.o *.s
|
|
/usr/bin/g++ -O3 -march=native -std=c++23 -Wall -Wextra -flto -DFLOPCOUNT=0 -DRANDOM_SEED=0 -DROOFLINE=0 -DBENCHMARK=1 main.cpp ./variants/variant_branchless.cpp -o main_branchless
|
|
=== N_P=50 N_ITER=1000 ===
|
|
W(n) per iteration: 184500.00 flops
|
|
Q(n) per iteration: 6512 bytes
|
|
I(n): 28.33 flops/byte
|
|
W(n): 184500000.00 flops
|
|
T(n): 164628520.00 cycles
|
|
P(n): 1.12 flops/cycle
|
|
|
|
=== N_P=100 N_ITER=1000 ===
|
|
W(n) per iteration: 729000.00 flops
|
|
Q(n) per iteration: 12912 bytes
|
|
I(n): 56.46 flops/byte
|
|
W(n): 729000000.00 flops
|
|
T(n): 668671300.00 cycles
|
|
P(n): 1.09 flops/cycle
|
|
|
|
=== N_P=200 N_ITER=1000 ===
|
|
W(n) per iteration: 2898000.00 flops
|
|
Q(n) per iteration: 25712 bytes
|
|
I(n): 112.71 flops/byte
|
|
W(n): 2898000000.00 flops
|
|
T(n): 2797909140.00 cycles
|
|
P(n): 1.04 flops/cycle
|
|
|
|
=== N_P=500 N_ITER=1000 ===
|
|
W(n) per iteration: 18045000.00 flops
|
|
Q(n) per iteration: 64112 bytes
|
|
I(n): 281.46 flops/byte
|
|
W(n): 18045000000.00 flops
|
|
T(n): 18473174460.00 cycles
|
|
P(n): 0.98 flops/cycle
|
|
|
|
=== N_P=1000 N_ITER=1000 ===
|
|
W(n) per iteration: 72090000.00 flops
|
|
Q(n) per iteration: 128112 bytes
|
|
I(n): 562.71 flops/byte
|
|
W(n): 72090000000.00 flops
|
|
T(n): 72557284580.00 cycles
|
|
P(n): 0.99 flops/cycle
|