Ahmed moreilp
This commit is contained in:
@@ -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): 16213663.88 cycles
|
||||
T(n): 16143845.50 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): 64706500.00 cycles
|
||||
T(n): 64419141.00 cycles
|
||||
P(n): 1.15 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): 266479325.00 cycles
|
||||
P(n): 1.10 flops/cycle
|
||||
T(n): 262269150.00 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): 1860823343.00 cycles
|
||||
P(n): 0.97 flops/cycle
|
||||
T(n): 1824103468.00 cycles
|
||||
P(n): 0.99 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): 8047713694.00 cycles
|
||||
P(n): 0.90 flops/cycle
|
||||
T(n): 7918236003.00 cycles
|
||||
P(n): 0.91 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): 35415664148.00 cycles
|
||||
P(n): 0.81 flops/cycle
|
||||
T(n): 34961836207.00 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): 188828167800.00 cycles
|
||||
P(n): 0.95 flops/cycle
|
||||
T(n): 187037217832.00 cycles
|
||||
P(n): 0.96 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_1.cpp -o main_inlining_7_symmetry_2_ilp_1 -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): 16226613.25 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): 64525677.00 cycles
|
||||
P(n): 1.15 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): 265742655.00 cycles
|
||||
P(n): 1.10 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): 1838249013.00 cycles
|
||||
P(n): 0.99 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): 8064528015.00 cycles
|
||||
P(n): 0.90 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): 35467168991.00 cycles
|
||||
P(n): 0.81 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): 188195572787.00 cycles
|
||||
P(n): 0.96 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_1.cpp -o main_inlining_7_symmetry_2_ilp_1 -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): 16192093.00 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): 64696094.50 cycles
|
||||
P(n): 1.15 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): 262498790.00 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): 1847962807.00 cycles
|
||||
P(n): 0.98 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): 8023346263.00 cycles
|
||||
P(n): 0.90 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): 35945565456.00 cycles
|
||||
P(n): 0.80 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): 188284363447.00 cycles
|
||||
P(n): 0.96 flops/cycle
|
||||
|
||||
Reference in New Issue
Block a user