Ahmed moreilp

This commit is contained in:
Ahmed Kallala
2026-06-17 03:01:19 +02:00
parent 42b3c7097e
commit 816626664c
167 changed files with 6415 additions and 269 deletions
@@ -5,54 +5,54 @@ W(n) per iteration: 95500.00 flops
Q(n) per iteration: 3256 bytes
I(n): 29.33 flops/byte
W(n): 19100000.00 flops
T(n): 12284355.00 cycles
P(n): 1.55 flops/cycle
T(n): 12254479.00 cycles
P(n): 1.56 flops/cycle
=== N_P=100 N_ITER=200 ===
W(n) per iteration: 371000.00 flops
Q(n) per iteration: 6456 bytes
I(n): 57.47 flops/byte
W(n): 74200000.00 flops
T(n): 50450768.00 cycles
P(n): 1.47 flops/cycle
T(n): 50135688.00 cycles
P(n): 1.48 flops/cycle
=== N_P=200 N_ITER=200 ===
W(n) per iteration: 1462000.00 flops
Q(n) per iteration: 12856 bytes
I(n): 113.72 flops/byte
W(n): 292400000.00 flops
T(n): 206580544.00 cycles
P(n): 1.42 flops/cycle
T(n): 205089712.00 cycles
P(n): 1.43 flops/cycle
=== N_P=500 N_ITER=200 ===
W(n) per iteration: 9055000.00 flops
Q(n) per iteration: 32056 bytes
I(n): 282.47 flops/byte
W(n): 1811000064.00 flops
T(n): 1437377152.00 cycles
P(n): 1.26 flops/cycle
T(n): 1389706496.00 cycles
P(n): 1.30 flops/cycle
=== N_P=1000 N_ITER=200 ===
W(n) per iteration: 36110000.00 flops
Q(n) per iteration: 64056 bytes
I(n): 563.73 flops/byte
W(n): 7222000128.00 flops
T(n): 6982579200.00 cycles
P(n): 1.03 flops/cycle
T(n): 6575198208.00 cycles
P(n): 1.10 flops/cycle
=== N_P=2000 N_ITER=200 ===
W(n) per iteration: 144220000.00 flops
Q(n) per iteration: 128056 bytes
I(n): 1126.23 flops/byte
W(n): 28843999232.00 flops
T(n): 32683665408.00 cycles
P(n): 0.88 flops/cycle
T(n): 28691492864.00 cycles
P(n): 1.01 flops/cycle
=== N_P=5000 N_ITER=200 ===
W(n) per iteration: 900550016.00 flops
Q(n) per iteration: 320056 bytes
I(n): 2813.73 flops/byte
W(n): 180110000128.00 flops
T(n): 232878047232.00 cycles
P(n): 0.77 flops/cycle
T(n): 219295383552.00 cycles
P(n): 0.82 flops/cycle
@@ -0,0 +1,58 @@
rm -f main_* run_tests *.o *.s variants/*.s variants/*.o headers/*.s headers/*.o
sudo /usr/bin/g++ main-float_symmetry.cpp ./variants/variant_float_inlining_SoA_symmetry.cpp -o main_float_inlining_SoA_symmetry -O3 -march=native -std=c++23 -Wall -Wextra -flto -DFLOPCOUNT=0 -DRANDOM_SEED=0 -DROOFLINE=0 -DBENCHMARK=1
=== N_P=50 N_ITER=200 ===
W(n) per iteration: 95500.00 flops
Q(n) per iteration: 3256 bytes
I(n): 29.33 flops/byte
W(n): 19100000.00 flops
T(n): 12256982.00 cycles
P(n): 1.56 flops/cycle
=== N_P=100 N_ITER=200 ===
W(n) per iteration: 371000.00 flops
Q(n) per iteration: 6456 bytes
I(n): 57.47 flops/byte
W(n): 74200000.00 flops
T(n): 50311960.00 cycles
P(n): 1.47 flops/cycle
=== N_P=200 N_ITER=200 ===
W(n) per iteration: 1462000.00 flops
Q(n) per iteration: 12856 bytes
I(n): 113.72 flops/byte
W(n): 292400000.00 flops
T(n): 205116992.00 cycles
P(n): 1.43 flops/cycle
=== N_P=500 N_ITER=200 ===
W(n) per iteration: 9055000.00 flops
Q(n) per iteration: 32056 bytes
I(n): 282.47 flops/byte
W(n): 1811000064.00 flops
T(n): 1430680320.00 cycles
P(n): 1.27 flops/cycle
=== N_P=1000 N_ITER=200 ===
W(n) per iteration: 36110000.00 flops
Q(n) per iteration: 64056 bytes
I(n): 563.73 flops/byte
W(n): 7222000128.00 flops
T(n): 6433722368.00 cycles
P(n): 1.12 flops/cycle
=== N_P=2000 N_ITER=200 ===
W(n) per iteration: 144220000.00 flops
Q(n) per iteration: 128056 bytes
I(n): 1126.23 flops/byte
W(n): 28843999232.00 flops
T(n): 28671172608.00 cycles
P(n): 1.01 flops/cycle
=== N_P=5000 N_ITER=200 ===
W(n) per iteration: 900550016.00 flops
Q(n) per iteration: 320056 bytes
I(n): 2813.73 flops/byte
W(n): 180110000128.00 flops
T(n): 226077196288.00 cycles
P(n): 0.80 flops/cycle
@@ -0,0 +1,58 @@
rm -f main_* run_tests *.o *.s variants/*.s variants/*.o headers/*.s headers/*.o
sudo /usr/bin/g++ main-float_symmetry.cpp ./variants/variant_float_inlining_SoA_symmetry.cpp -o main_float_inlining_SoA_symmetry -O3 -march=native -std=c++23 -Wall -Wextra -flto -DFLOPCOUNT=0 -DRANDOM_SEED=0 -DROOFLINE=0 -DBENCHMARK=1
=== N_P=50 N_ITER=200 ===
W(n) per iteration: 95500.00 flops
Q(n) per iteration: 3256 bytes
I(n): 29.33 flops/byte
W(n): 19100000.00 flops
T(n): 12241891.00 cycles
P(n): 1.56 flops/cycle
=== N_P=100 N_ITER=200 ===
W(n) per iteration: 371000.00 flops
Q(n) per iteration: 6456 bytes
I(n): 57.47 flops/byte
W(n): 74200000.00 flops
T(n): 50166764.00 cycles
P(n): 1.48 flops/cycle
=== N_P=200 N_ITER=200 ===
W(n) per iteration: 1462000.00 flops
Q(n) per iteration: 12856 bytes
I(n): 113.72 flops/byte
W(n): 292400000.00 flops
T(n): 205702832.00 cycles
P(n): 1.42 flops/cycle
=== N_P=500 N_ITER=200 ===
W(n) per iteration: 9055000.00 flops
Q(n) per iteration: 32056 bytes
I(n): 282.47 flops/byte
W(n): 1811000064.00 flops
T(n): 1447650560.00 cycles
P(n): 1.25 flops/cycle
=== N_P=1000 N_ITER=200 ===
W(n) per iteration: 36110000.00 flops
Q(n) per iteration: 64056 bytes
I(n): 563.73 flops/byte
W(n): 7222000128.00 flops
T(n): 6788238336.00 cycles
P(n): 1.06 flops/cycle
=== N_P=2000 N_ITER=200 ===
W(n) per iteration: 144220000.00 flops
Q(n) per iteration: 128056 bytes
I(n): 1126.23 flops/byte
W(n): 28843999232.00 flops
T(n): 28329756672.00 cycles
P(n): 1.02 flops/cycle
=== N_P=5000 N_ITER=200 ===
W(n) per iteration: 900550016.00 flops
Q(n) per iteration: 320056 bytes
I(n): 2813.73 flops/byte
W(n): 180110000128.00 flops
T(n): 226648129536.00 cycles
P(n): 0.79 flops/cycle