Ahmed moreilp
This commit is contained in:
@@ -5,15 +5,15 @@ 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): 16951060.50 cycles
|
||||
P(n): 1.13 flops/cycle
|
||||
T(n): 16580100.12 cycles
|
||||
P(n): 1.15 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): 65369333.50 cycles
|
||||
T(n): 65024621.00 cycles
|
||||
P(n): 1.14 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): 264705723.00 cycles
|
||||
P(n): 1.10 flops/cycle
|
||||
T(n): 261750350.00 cycles
|
||||
P(n): 1.12 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): 1782045217.00 cycles
|
||||
P(n): 1.02 flops/cycle
|
||||
T(n): 1758395916.00 cycles
|
||||
P(n): 1.03 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): 7925808103.00 cycles
|
||||
P(n): 0.91 flops/cycle
|
||||
T(n): 7434680291.00 cycles
|
||||
P(n): 0.97 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): 35824664520.00 cycles
|
||||
P(n): 0.81 flops/cycle
|
||||
T(n): 34379672055.00 cycles
|
||||
P(n): 0.84 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): 247389191708.00 cycles
|
||||
P(n): 0.73 flops/cycle
|
||||
T(n): 239178588140.00 cycles
|
||||
P(n): 0.75 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_ilp_2.cpp -o main_inlining_7_symmetry_2_ilp_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): 16547697.88 cycles
|
||||
P(n): 1.15 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): 65340354.00 cycles
|
||||
P(n): 1.14 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): 262140162.00 cycles
|
||||
P(n): 1.12 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): 1791946939.00 cycles
|
||||
P(n): 1.01 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): 7559703643.00 cycles
|
||||
P(n): 0.96 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): 34850899212.00 cycles
|
||||
P(n): 0.83 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): 239389045900.00 cycles
|
||||
P(n): 0.75 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_ilp_2.cpp -o main_inlining_7_symmetry_2_ilp_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): 16525790.75 cycles
|
||||
P(n): 1.16 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): 65160000.00 cycles
|
||||
P(n): 1.14 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): 263049074.00 cycles
|
||||
P(n): 1.11 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): 1763111279.00 cycles
|
||||
P(n): 1.03 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): 7502606429.00 cycles
|
||||
P(n): 0.96 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): 35210679018.00 cycles
|
||||
P(n): 0.82 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): 242369036471.00 cycles
|
||||
P(n): 0.74 flops/cycle
|
||||
|
||||
Reference in New Issue
Block a user