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
@@ -5,7 +5,7 @@ W(n) per iteration: 95500.00 flops
Q(n) per iteration: 6512 bytes
I(n): 14.67 flops/byte
W(n): 19100000.00 flops
T(n): 16243406.75 cycles
T(n): 16142806.62 cycles
P(n): 1.18 flops/cycle
=== N_P=100 N_ITER=200 ===
@@ -13,7 +13,7 @@ W(n) per iteration: 371000.00 flops
Q(n) per iteration: 12912 bytes
I(n): 28.73 flops/byte
W(n): 74200000.00 flops
T(n): 66097465.00 cycles
T(n): 66409907.00 cycles
P(n): 1.12 flops/cycle
=== N_P=200 N_ITER=200 ===
@@ -21,38 +21,38 @@ W(n) per iteration: 1462000.00 flops
Q(n) per iteration: 25712 bytes
I(n): 56.86 flops/byte
W(n): 292400000.00 flops
T(n): 285031402.00 cycles
P(n): 1.03 flops/cycle
T(n): 278127300.00 cycles
P(n): 1.05 flops/cycle
=== N_P=500 N_ITER=200 ===
W(n) per iteration: 9055000.00 flops
Q(n) per iteration: 64112 bytes
I(n): 141.24 flops/byte
W(n): 1811000000.00 flops
T(n): 2157696662.00 cycles
P(n): 0.84 flops/cycle
T(n): 1983862454.00 cycles
P(n): 0.91 flops/cycle
=== N_P=1000 N_ITER=200 ===
W(n) per iteration: 36110000.00 flops
Q(n) per iteration: 128112 bytes
I(n): 281.86 flops/byte
W(n): 7222000000.00 flops
T(n): 9630862508.00 cycles
P(n): 0.75 flops/cycle
T(n): 8764024847.00 cycles
P(n): 0.82 flops/cycle
=== N_P=2000 N_ITER=200 ===
W(n) per iteration: 144220000.00 flops
Q(n) per iteration: 256112 bytes
I(n): 563.11 flops/byte
W(n): 28844000000.00 flops
T(n): 41773155034.00 cycles
P(n): 0.69 flops/cycle
T(n): 39723210337.00 cycles
P(n): 0.73 flops/cycle
=== N_P=5000 N_ITER=200 ===
W(n) per iteration: 900550000.00 flops
Q(n) per iteration: 640112 bytes
I(n): 1406.86 flops/byte
W(n): 180110000000.00 flops
T(n): 276672447552.00 cycles
T(n): 277797538652.00 cycles
P(n): 0.65 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-symmetry.cpp ./variants/variant_inlining_7_symmetry_2.cpp -o main_inlining_7_symmetry_2 -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: 6512 bytes
I(n): 14.67 flops/byte
W(n): 19100000.00 flops
T(n): 16168869.00 cycles
P(n): 1.18 flops/cycle
=== N_P=100 N_ITER=200 ===
W(n) per iteration: 371000.00 flops
Q(n) per iteration: 12912 bytes
I(n): 28.73 flops/byte
W(n): 74200000.00 flops
T(n): 65986593.50 cycles
P(n): 1.12 flops/cycle
=== N_P=200 N_ITER=200 ===
W(n) per iteration: 1462000.00 flops
Q(n) per iteration: 25712 bytes
I(n): 56.86 flops/byte
W(n): 292400000.00 flops
T(n): 276843557.00 cycles
P(n): 1.06 flops/cycle
=== N_P=500 N_ITER=200 ===
W(n) per iteration: 9055000.00 flops
Q(n) per iteration: 64112 bytes
I(n): 141.24 flops/byte
W(n): 1811000000.00 flops
T(n): 1995847484.00 cycles
P(n): 0.91 flops/cycle
=== N_P=1000 N_ITER=200 ===
W(n) per iteration: 36110000.00 flops
Q(n) per iteration: 128112 bytes
I(n): 281.86 flops/byte
W(n): 7222000000.00 flops
T(n): 8724469474.00 cycles
P(n): 0.83 flops/cycle
=== N_P=2000 N_ITER=200 ===
W(n) per iteration: 144220000.00 flops
Q(n) per iteration: 256112 bytes
I(n): 563.11 flops/byte
W(n): 28844000000.00 flops
T(n): 40376577856.00 cycles
P(n): 0.71 flops/cycle
=== N_P=5000 N_ITER=200 ===
W(n) per iteration: 900550000.00 flops
Q(n) per iteration: 640112 bytes
I(n): 1406.86 flops/byte
W(n): 180110000000.00 flops
T(n): 277173820474.00 cycles
P(n): 0.65 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-symmetry.cpp ./variants/variant_inlining_7_symmetry_2.cpp -o main_inlining_7_symmetry_2 -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: 6512 bytes
I(n): 14.67 flops/byte
W(n): 19100000.00 flops
T(n): 16206126.50 cycles
P(n): 1.18 flops/cycle
=== N_P=100 N_ITER=200 ===
W(n) per iteration: 371000.00 flops
Q(n) per iteration: 12912 bytes
I(n): 28.73 flops/byte
W(n): 74200000.00 flops
T(n): 65943577.00 cycles
P(n): 1.13 flops/cycle
=== N_P=200 N_ITER=200 ===
W(n) per iteration: 1462000.00 flops
Q(n) per iteration: 25712 bytes
I(n): 56.86 flops/byte
W(n): 292400000.00 flops
T(n): 278774560.00 cycles
P(n): 1.05 flops/cycle
=== N_P=500 N_ITER=200 ===
W(n) per iteration: 9055000.00 flops
Q(n) per iteration: 64112 bytes
I(n): 141.24 flops/byte
W(n): 1811000000.00 flops
T(n): 1997709210.00 cycles
P(n): 0.91 flops/cycle
=== N_P=1000 N_ITER=200 ===
W(n) per iteration: 36110000.00 flops
Q(n) per iteration: 128112 bytes
I(n): 281.86 flops/byte
W(n): 7222000000.00 flops
T(n): 9295948552.00 cycles
P(n): 0.78 flops/cycle
=== N_P=2000 N_ITER=200 ===
W(n) per iteration: 144220000.00 flops
Q(n) per iteration: 256112 bytes
I(n): 563.11 flops/byte
W(n): 28844000000.00 flops
T(n): 39953243724.00 cycles
P(n): 0.72 flops/cycle
=== N_P=5000 N_ITER=200 ===
W(n) per iteration: 900550000.00 flops
Q(n) per iteration: 640112 bytes
I(n): 1406.86 flops/byte
W(n): 180110000000.00 flops
T(n): 272359437585.00 cycles
P(n): 0.66 flops/cycle