Ahmed moreilp
This commit is contained in:
@@ -5,47 +5,47 @@ 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): 9097177.00 cycles
|
||||
P(n): 2.10 flops/cycle
|
||||
T(n): 9042754.00 cycles
|
||||
P(n): 2.11 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): 33503178.00 cycles
|
||||
P(n): 2.21 flops/cycle
|
||||
T(n): 33334258.00 cycles
|
||||
P(n): 2.23 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): 118530416.00 cycles
|
||||
P(n): 2.47 flops/cycle
|
||||
T(n): 118737600.00 cycles
|
||||
P(n): 2.46 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): 684755776.00 cycles
|
||||
P(n): 2.64 flops/cycle
|
||||
T(n): 683585216.00 cycles
|
||||
P(n): 2.65 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): 2936483584.00 cycles
|
||||
P(n): 2.46 flops/cycle
|
||||
T(n): 2916843264.00 cycles
|
||||
P(n): 2.48 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): 12939599872.00 cycles
|
||||
T(n): 12947107840.00 cycles
|
||||
P(n): 2.23 flops/cycle
|
||||
|
||||
=== N_P=5000 N_ITER=200 ===
|
||||
@@ -53,6 +53,6 @@ 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): 156191883264.00 cycles
|
||||
P(n): 1.15 flops/cycle
|
||||
T(n): 154301726720.00 cycles
|
||||
P(n): 1.17 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_ILP_16.cpp -o main_float_inlining_SoA_symmetry_ILP_16 -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): 9068705.00 cycles
|
||||
P(n): 2.11 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): 33619560.00 cycles
|
||||
P(n): 2.21 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): 119648512.00 cycles
|
||||
P(n): 2.44 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): 688923456.00 cycles
|
||||
P(n): 2.63 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): 2882365184.00 cycles
|
||||
P(n): 2.51 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): 12853307392.00 cycles
|
||||
P(n): 2.24 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): 157621452800.00 cycles
|
||||
P(n): 1.14 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_ILP_16.cpp -o main_float_inlining_SoA_symmetry_ILP_16 -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): 9045719.00 cycles
|
||||
P(n): 2.11 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): 33367980.00 cycles
|
||||
P(n): 2.22 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): 118491888.00 cycles
|
||||
P(n): 2.47 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): 683712000.00 cycles
|
||||
P(n): 2.65 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): 2850022656.00 cycles
|
||||
P(n): 2.53 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): 12636773376.00 cycles
|
||||
P(n): 2.28 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): 158985289728.00 cycles
|
||||
P(n): 1.13 flops/cycle
|
||||
|
||||
Reference in New Issue
Block a user