Merging

llvm/projects/hpvm-tensor-rt/PPoPP_results/knobs.txt

+perf_fp16,120,Baseline
+perf_fp16,151,Col perf 50% offset=0
+perf_fp16,152,Col perf 50% offset=1
+perf_fp16,153,Row perf 50% offset=0
+perf_fp16,154,Row perf 50% offset=1
+perf_fp16,155,Col perf 33% offset=0
+perf_fp16,156,Col perf 33% offset=1
+perf_fp16,157,Col perf 33% offset=2
+perf_fp16,158,Row perf 33% offset=0
+perf_fp16,159,Row perf 33% offset=1
+perf_fp16,160,Row perf 33% offset=2
+perf_fp16,161,Col perf 25% offset=0
+perf_fp16,162,Col perf 25% offset=1
+perf_fp16,163,Col perf 25% offset=2
+perf_fp16,164,Col perf 25% offset=3
+perf_fp16,165,Row perf 25% offset=0
+perf_fp16,166,Row perf 25% offset=1
+perf_fp16,167,Row perf 25% offset=2
+perf_fp16,168,Row perf 25% offset=3
+samp_fp16,261,Samp 50% offset=0
+samp_fp16,262,Samp 50% offset=1
+samp_fp16,263,Samp 33% offset=0
+samp_fp16,264,Samp 33% offset=1
+samp_fp16,265,Samp 33% offset=2
+samp_fp16,266,Samp 25% offset=0
+samp_fp16,267,Samp 25% offset=1
+samp_fp16,268,Samp 25% offset=2
+samp_fp16,269,Samp 25% offset=3

llvm/projects/hpvm-tensor-rt/PPoPP_results/profiling_results/alexnet2_cifar10/alexnet2_cifar10_fp16_perf_fp16_120.txt

+Conv1,65.2289,635838
+Conv1_f2h,0,0
+Conv1_h2f,0,0
+Add1,0.206437,13534.8
+Add1_f2h,0,0
+Add1_h2f,0,0
+Tanh1,0.171279,13528.5
+Tanh1_f2h,0,0
+Tanh1_h2f,0,0
+Conv2,257.082,2.53801e+06
+Conv2_f2h,0,0
+Conv2_h2f,0,0
+Add2,0.220024,13871.8
+Add2_f2h,0,0
+Add2_h2f,0,0
+Tanh2,0.191798,13871.6
+Tanh2_f2h,0,0
+Tanh2_h2f,0,0
+Pool1,2.78132,31392.7
+Pool1_f2h,0,0
+Pool1_h2f,0,0
+Conv3,79.5717,772941
+Conv3_f2h,0,0
+Conv3_h2f,0,0
+Add3,0.208738,14141.6
+Add3_f2h,0,0
+Add3_h2f,0,0
+Tanh3,0.179234,14145.7
+Tanh3_f2h,0,0
+Tanh3_h2f,0,0
+Conv4,132.214,1.32889e+06
+Conv4_f2h,0,0
+Conv4_h2f,0,0
+Add4,0.213516,14286.1
+Add4_f2h,0,0
+Add4_h2f,0,0
+Tanh4,0.178101,14283.1
+Tanh4_f2h,0,0
+Tanh4_h2f,0,0
+Pool2,2.07068,28582.4
+Pool2_f2h,0,0
+Pool2_h2f,0,0
+Conv5,44.624,449041
+Conv5_f2h,0,0
+Conv5_h2f,0,0
+Add5,0.199375,14430.7
+Add5_f2h,0,0
+Add5_h2f,0,0
+Tanh5,0.155295,14431.9
+Tanh5_f2h,0,0
+Tanh5_h2f,0,0
+Conv6,75.4301,795035
+Conv6_f2h,0,0
+Conv6_h2f,0,0
+Add6,0.205183,14622.6
+Add6_f2h,0,0
+Add6_h2f,0,0
+Tanh6,0.157695,14624.5
+Tanh6_f2h,0,0
+Tanh6_h2f,0,0
+Pool3,1.24782,14616.9
+Pool3_f2h,0,0
+Pool3_h2f,0,0
+Mul1,1.20811,15614.1
+Mul1_f2h,0,0
+Mul1_h2f,0,0
+Add7,0.164572,14641.2
+Add7_f2h,0,0
+Add7_h2f,0,0
+Softmax1,0.996439,15121.6
+Softmax1_f2h,0,0
+Softmax1_h2f,0,0

llvm/projects/hpvm-tensor-rt/PPoPP_results/profiling_results/alexnet2_cifar10/alexnet2_cifar10_fp16_perf_fp16_151.txt

+Conv1,58.0027,548113
+Conv1_f2h,0,0
+Conv1_h2f,0,0
+Add1,0.189544,13050.8
+Add1_f2h,0,0
+Add1_h2f,0,0
+Tanh1,0.17103,13052
+Tanh1_f2h,0,0
+Tanh1_h2f,0,0
+Conv2,128.221,1.22437e+06
+Conv2_f2h,0,0
+Conv2_h2f,0,0
+Add2,0.217308,13335.4
+Add2_f2h,0,0
+Add2_h2f,0,0
+Tanh2,0.17494,13336.5
+Tanh2_f2h,0,0
+Tanh2_h2f,0,0
+Pool1,2.54381,29480.6
+Pool1_f2h,0,0
+Pool1_h2f,0,0
+Conv3,48.9069,459648
+Conv3_f2h,0,0
+Conv3_h2f,0,0
+Add3,0.213516,13612.2
+Add3_f2h,0,0
+Add3_h2f,0,0
+Tanh3,0.19278,13600.7
+Tanh3_f2h,0,0
+Tanh3_h2f,0,0
+Conv4,66.2285,651716
+Conv4_f2h,0,0
+Conv4_h2f,0,0
+Add4,0.205122,13690.4
+Add4_f2h,0,0
+Add4_h2f,0,0
+Tanh4,0.174924,13694.9
+Tanh4_f2h,0,0
+Tanh4_h2f,0,0
+Pool2,1.83244,27400
+Pool2_f2h,0,0
+Pool2_h2f,0,0
+Conv5,27.4047,267152
+Conv5_f2h,0,0
+Conv5_h2f,0,0
+Add5,0.190466,13844.8
+Add5_f2h,0,0
+Add5_h2f,0,0
+Tanh5,0.149922,13851.3
+Tanh5_f2h,0,0
+Tanh5_h2f,0,0
+Conv6,39.169,396057
+Conv6_f2h,0,0
+Conv6_h2f,0,0
+Add6,0.186793,13947
+Add6_f2h,0,0
+Add6_h2f,0,0
+Tanh6,0.147929,13947.8
+Tanh6_f2h,0,0
+Tanh6_h2f,0,0
+Pool3,1.12431,14343.1
+Pool3_f2h,0,0
+Pool3_h2f,0,0
+Mul1,1.09979,14347.5
+Mul1_f2h,0,0
+Mul1_h2f,0,0
+Add7,0.145526,13961.3
+Add7_f2h,0,0
+Add7_h2f,0,0
+Softmax1,0.939092,13969.2
+Softmax1_f2h,0,0
+Softmax1_h2f,0,0

llvm/projects/hpvm-tensor-rt/PPoPP_results/profiling_results/alexnet2_cifar10/alexnet2_cifar10_fp16_perf_fp16_152.txt

+Conv1,57.9254,550132
+Conv1_f2h,0,0
+Conv1_h2f,0,0
+Add1,0.188185,13098.3
+Add1_f2h,0,0
+Add1_h2f,0,0
+Tanh1,0.168668,13100.2
+Tanh1_f2h,0,0
+Tanh1_h2f,0,0
+Conv2,128,1.22987e+06
+Conv2_f2h,0,0
+Conv2_h2f,0,0
+Add2,0.217096,13371.7
+Add2_f2h,0,0
+Add2_h2f,0,0
+Tanh2,0.17532,13354.5
+Tanh2_f2h,0,0
+Tanh2_h2f,0,0
+Pool1,2.45412,26727.6
+Pool1_f2h,0,0
+Pool1_h2f,0,0
+Conv3,48.745,462457
+Conv3_f2h,0,0
+Conv3_h2f,0,0
+Add3,0.20812,13623.3
+Add3_f2h,0,0
+Add3_h2f,0,0
+Tanh3,0.176805,13623.3
+Tanh3_f2h,0,0
+Tanh3_h2f,0,0
+Conv4,66.1231,650723
+Conv4_f2h,0,0
+Conv4_h2f,0,0
+Add4,0.204978,13775.7
+Add4_f2h,0,0
+Add4_h2f,0,0
+Tanh4,0.185167,13766.3
+Tanh4_f2h,0,0
+Tanh4_h2f,0,0
+Pool2,1.89592,25380.2
+Pool2_f2h,0,0
+Pool2_h2f,0,0
+Conv5,27.2569,267804
+Conv5_f2h,0,0
+Conv5_h2f,0,0
+Add5,0.187973,13877.8
+Add5_f2h,0,0
+Add5_h2f,0,0
+Tanh5,0.146757,13869.7
+Tanh5_f2h,0,0
+Tanh5_h2f,0,0
+Conv6,39.3148,398486
+Conv6_f2h,0,0
+Conv6_h2f,0,0
+Add6,0.18485,13995
+Add6_f2h,0,0
+Add6_h2f,0,0
+Tanh6,0.147977,14000.1
+Tanh6_f2h,0,0
+Tanh6_h2f,0,0
+Pool3,1.16044,13997.9
+Pool3_f2h,0,0
+Pool3_h2f,0,0
+Mul1,1.14272,14765.5
+Mul1_f2h,0,0
+Mul1_h2f,0,0
+Add7,0.154716,14008.5
+Add7_f2h,0,0
+Add7_h2f,0,0
+Softmax1,0.967361,14016.6
+Softmax1_f2h,0,0
+Softmax1_h2f,0,0

llvm/projects/hpvm-tensor-rt/PPoPP_results/profiling_results/alexnet2_cifar10/alexnet2_cifar10_fp16_perf_fp16_153.txt

+Conv1,52.3167,486159
+Conv1_f2h,0,0
+Conv1_h2f,0,0
+Add1,0.200114,12880.8
+Add1_f2h,0,0
+Add1_h2f,0,0
+Tanh1,0.175215,12886.3
+Tanh1_f2h,0,0
+Tanh1_h2f,0,0
+Conv2,122.54,1.15844e+06
+Conv2_f2h,0,0
+Conv2_h2f,0,0
+Add2,0.212952,13161.5
+Add2_f2h,0,0
+Add2_h2f,0,0
+Tanh2,0.179759,13157.4
+Tanh2_f2h,0,0
+Tanh2_h2f,0,0
+Pool1,2.5318,27617.3
+Pool1_f2h,0,0
+Pool1_h2f,0,0
+Conv3,47.8415,444086
+Conv3_f2h,0,0
+Conv3_h2f,0,0
+Add3,0.203436,13433.4
+Add3_f2h,0,0
+Add3_h2f,0,0
+Tanh3,0.170578,13433.7
+Tanh3_f2h,0,0
+Tanh3_h2f,0,0
+Conv4,65.082,630988
+Conv4_f2h,0,0
+Conv4_h2f,0,0
+Add4,0.213602,12821.4
+Add4_f2h,0,0
+Add4_h2f,0,0
+Tanh4,0.166886,12826.1
+Tanh4_f2h,0,0
+Tanh4_h2f,0,0
+Pool2,1.80028,25660.3
+Pool2_f2h,0,0
+Pool2_h2f,0,0
+Conv5,27.811,255106
+Conv5_f2h,0,0
+Conv5_h2f,0,0
+Add5,0.186975,13659.3
+Add5_f2h,0,0
+Add5_h2f,0,0
+Tanh5,0.158729,13644.4
+Tanh5_f2h,0,0
+Tanh5_h2f,0,0
+Conv6,39.012,391835
+Conv6_f2h,0,0
+Conv6_h2f,0,0
+Add6,0.183039,13762.1
+Add6_f2h,0,0
+Add6_h2f,0,0
+Tanh6,0.147983,13761.3
+Tanh6_f2h,0,0
+Tanh6_h2f,0,0
+Pool3,1.16698,13761.9
+Pool3_f2h,0,0
+Pool3_h2f,0,0
+Mul1,1.16997,14528.8
+Mul1_f2h,0,0
+Mul1_h2f,0,0
+Add7,0.16397,13782.7
+Add7_f2h,0,0
+Add7_h2f,0,0
+Softmax1,0.946711,13787
+Softmax1_f2h,0,0
+Softmax1_h2f,0,0

llvm/projects/hpvm-tensor-rt/PPoPP_results/profiling_results/alexnet2_cifar10/alexnet2_cifar10_fp16_perf_fp16_154.txt

+Conv1,52.9971,496465
+Conv1_f2h,0,0
+Conv1_h2f,0,0
+Add1,0.192665,12930.4
+Add1_f2h,0,0
+Add1_h2f,0,0
+Tanh1,0.170856,12926.5
+Tanh1_f2h,0,0
+Tanh1_h2f,0,0
+Conv2,122.764,1.15342e+06
+Conv2_f2h,0,0
+Conv2_h2f,0,0
+Add2,0.215928,13205.1
+Add2_f2h,0,0
+Add2_h2f,0,0
+Tanh2,0.175704,13204.8
+Tanh2_f2h,0,0
+Tanh2_h2f,0,0
+Pool1,2.49985,26407.8
+Pool1_f2h,0,0
+Pool1_h2f,0,0
+Conv3,47.9188,445564
+Conv3_f2h,0,0
+Conv3_h2f,0,0
+Add3,0.204233,13461.2
+Add3_f2h,0,0
+Add3_h2f,0,0
+Tanh3,0.177759,13459.2
+Tanh3_f2h,0,0
+Tanh3_h2f,0,0
+Conv4,64.6805,628375
+Conv4_f2h,0,0
+Conv4_h2f,0,0
+Add4,0.210421,13606
+Add4_f2h,0,0
+Add4_h2f,0,0
+Tanh4,0.173551,13591.1
+Tanh4_f2h,0,0
+Tanh4_h2f,0,0
+Pool2,1.87541,26494.5
+Pool2_f2h,0,0
+Pool2_h2f,0,0
+Conv5,27.3168,264238
+Conv5_f2h,0,0
+Conv5_h2f,0,0
+Add5,0.190476,13723.3
+Add5_f2h,0,0
+Add5_h2f,0,0
+Tanh5,0.150409,13732.8
+Tanh5_f2h,0,0
+Tanh5_h2f,0,0
+Conv6,38.9734,395354
+Conv6_f2h,0,0
+Conv6_h2f,0,0
+Add6,0.185871,13830.2
+Add6_f2h,0,0
+Add6_h2f,0,0
+Tanh6,0.147122,13835.5
+Tanh6_f2h,0,0
+Tanh6_h2f,0,0
+Pool3,1.13866,13834.3
+Pool3_f2h,0,0
+Pool3_h2f,0,0
+Mul1,1.1186,14601.5
+Mul1_f2h,0,0
+Mul1_h2f,0,0
+Add7,0.15205,13856.2
+Add7_f2h,0,0
+Add7_h2f,0,0
+Softmax1,0.938171,13864.3
+Softmax1_f2h,0,0
+Softmax1_h2f,0,0

llvm/projects/hpvm-tensor-rt/PPoPP_results/profiling_results/alexnet2_cifar10/alexnet2_cifar10_fp16_perf_fp16_155.txt

+Conv1,64.8669,622301
+Conv1_f2h,0,0
+Conv1_h2f,0,0
+Add1,0.20949,13239.5
+Add1_f2h,0,0
+Add1_h2f,0,0
+Tanh1,0.176294,13243.4
+Tanh1_f2h,0,0
+Tanh1_h2f,0,0
+Conv2,162.006,1.5735e+06
+Conv2_f2h,0,0
+Conv2_h2f,0,0
+Add2,0.220239,13511.2
+Add2_f2h,0,0
+Add2_h2f,0,0
+Tanh2,0.178239,13511.2
+Tanh2_f2h,0,0
+Tanh2_h2f,0,0
+Pool1,2.75893,29646.5
+Pool1_f2h,0,0
+Pool1_h2f,0,0
+Conv3,57.036,545033
+Conv3_f2h,0,0
+Conv3_h2f,0,0
+Add3,0.208872,13815.5
+Add3_f2h,0,0
+Add3_h2f,0,0
+Tanh3,0.181833,13813.6
+Tanh3_f2h,0,0
+Tanh3_h2f,0,0
+Conv4,84.4746,841306
+Conv4_f2h,0,0
+Conv4_h2f,0,0
+Add4,0.216168,13970.9
+Add4_f2h,0,0
+Add4_h2f,0,0
+Tanh4,0.181836,13976.3
+Tanh4_f2h,0,0
+Tanh4_h2f,0,0
+Pool2,2.06704,27968.2
+Pool2_f2h,0,0
+Pool2_h2f,0,0
+Conv5,32.4425,322697
+Conv5_f2h,0,0
+Conv5_h2f,0,0
+Add5,0.199647,14092.9
+Add5_f2h,0,0
+Add5_h2f,0,0
+Tanh5,0.160076,14084.6
+Tanh5_f2h,0,0
+Tanh5_h2f,0,0
+Conv6,48.8451,506372
+Conv6_f2h,0,0
+Conv6_h2f,0,0
+Add6,0.196623,14181.5
+Add6_f2h,0,0
+Add6_h2f,0,0
+Tanh6,0.159896,14172.9
+Tanh6_f2h,0,0
+Tanh6_h2f,0,0
+Pool3,1.22106,14178.1
+Pool3_f2h,0,0
+Pool3_h2f,0,0
+Mul1,1.21282,15029.7
+Mul1_f2h,0,0
+Mul1_h2f,0,0
+Add7,0.16867,14202.3
+Add7_f2h,0,0
+Add7_h2f,0,0
+Softmax1,0.963022,14204.2
+Softmax1_f2h,0,0
+Softmax1_h2f,0,0

llvm/projects/hpvm-tensor-rt/PPoPP_results/profiling_results/alexnet2_cifar10/alexnet2_cifar10_fp16_perf_fp16_156.txt

+Conv1,63.1272,608633
+Conv1_f2h,0,0
+Conv1_h2f,0,0
+Add1,0.203714,13382.6
+Add1_f2h,0,0
+Add1_h2f,0,0
+Tanh1,0.179644,13369.4
+Tanh1_f2h,0,0
+Tanh1_h2f,0,0
+Conv2,159.062,1.56322e+06
+Conv2_f2h,0,0
+Conv2_h2f,0,0
+Add2,0.219154,13650.7
+Add2_f2h,0,0
+Add2_h2f,0,0
+Tanh2,0.176476,13654.5
+Tanh2_f2h,0,0
+Tanh2_h2f,0,0
+Pool1,2.6318,29500.9
+Pool1_f2h,0,0
+Pool1_h2f,0,0
+Conv3,58.7054,568593
+Conv3_f2h,0,0
+Conv3_h2f,0,0
+Add3,0.212313,13955.8
+Add3_f2h,0,0
+Add3_h2f,0,0
+Tanh3,0.186098,13945.6
+Tanh3_f2h,0,0
+Tanh3_h2f,0,0
+Conv4,88.1808,883262
+Conv4_f2h,0,0
+Conv4_h2f,0,0
+Add4,0.211765,14119.8
+Add4_f2h,0,0
+Add4_h2f,0,0
+Tanh4,0.179279,14113.7
+Tanh4_f2h,0,0
+Tanh4_h2f,0,0
+Pool2,1.87673,27502.6
+Pool2_f2h,0,0
+Pool2_h2f,0,0
+Conv5,31.7267,314842
+Conv5_f2h,0,0
+Conv5_h2f,0,0
+Add5,0.199883,14261.2
+Add5_f2h,0,0
+Add5_h2f,0,0
+Tanh5,0.168866,14253.9
+Tanh5_f2h,0,0
+Tanh5_h2f,0,0
+Conv6,47.7887,497470
+Conv6_f2h,0,0
+Conv6_h2f,0,0
+Add6,0.194706,14369.6
+Add6_f2h,0,0
+Add6_h2f,0,0
+Tanh6,0.161791,14355.9
+Tanh6_f2h,0,0
+Tanh6_h2f,0,0
+Pool3,1.18783,14357
+Pool3_f2h,0,0
+Pool3_h2f,0,0
+Mul1,1.19429,15205.7
+Mul1_f2h,0,0
+Mul1_h2f,0,0
+Add7,0.174949,14370.7
+Add7_f2h,0,0
+Add7_h2f,0,0
+Softmax1,1.00356,15533.1
+Softmax1_f2h,0,0
+Softmax1_h2f,0,0

llvm/projects/hpvm-tensor-rt/PPoPP_results/profiling_results/alexnet2_cifar10/alexnet2_cifar10_fp16_perf_fp16_157.txt

+Conv1,63.9795,627055
+Conv1_f2h,0,0
+Conv1_h2f,0,0
+Add1,0.207468,13560.2
+Add1_f2h,0,0
+Add1_h2f,0,0
+Tanh1,0.172213,13546.7
+Tanh1_f2h,0,0
+Tanh1_h2f,0,0
+Conv2,164.125,1.62846e+06
+Conv2_f2h,0,0
+Conv2_h2f,0,0
+Add2,0.21318,13828.3
+Add2_f2h,0,0
+Add2_h2f,0,0
+Tanh2,0.174274,13824.4
+Tanh2_f2h,0,0
+Tanh2_h2f,0,0
+Pool1,2.56693,28880.1
+Pool1_f2h,0,0
+Pool1_h2f,0,0
+Conv3,58.2621,572924
+Conv3_f2h,0,0
+Conv3_h2f,0,0
+Add3,0.205416,14092.8
+Add3_f2h,0,0
+Add3_h2f,0,0
+Tanh3,0.182895,14103
+Tanh3_f2h,0,0
+Tanh3_h2f,0,0
+Conv4,88.3811,890106
+Conv4_f2h,0,0
+Conv4_h2f,0,0
+Add4,0.213208,14246.5
+Add4_f2h,0,0
+Add4_h2f,0,0
+Tanh4,0.179445,14255.8
+Tanh4_f2h,0,0
+Tanh4_h2f,0,0
+Pool2,1.93471,28531.2
+Pool2_f2h,0,0
+Pool2_h2f,0,0
+Conv5,34.5703,350902
+Conv5_f2h,0,0
+Conv5_h2f,0,0
+Add5,0.201976,14415.1
+Add5_f2h,0,0
+Add5_h2f,0,0
+Tanh5,0.164367,14414.3
+Tanh5_f2h,0,0
+Tanh5_h2f,0,0
+Conv6,53.1299,560793
+Conv6_f2h,0,0
+Conv6_h2f,0,0
+Add6,0.195474,14557
+Add6_f2h,0,0
+Add6_h2f,0,0
+Tanh6,0.158092,14560
+Tanh6_f2h,0,0
+Tanh6_h2f,0,0
+Pool3,1.15914,14565.3
+Pool3_f2h,0,0
+Pool3_h2f,0,0
+Mul1,1.14588,15448.4
+Mul1_f2h,0,0
+Mul1_h2f,0,0
+Add7,0.158146,14582.8
+Add7_f2h,0,0
+Add7_h2f,0,0
+Softmax1,0.947485,15022.4
+Softmax1_f2h,0,0
+Softmax1_h2f,0,0

llvm/projects/hpvm-tensor-rt/PPoPP_results/profiling_results/alexnet2_cifar10/alexnet2_cifar10_fp16_perf_fp16_158.txt

+Conv1,59.0766,557171
+Conv1_f2h,0,0
+Conv1_h2f,0,0
+Add1,0.203615,13064.7
+Add1_f2h,0,0
+Add1_h2f,0,0
+Tanh1,0.17004,13070.4
+Tanh1_f2h,0,0
+Tanh1_h2f,0,0
+Conv2,156.515,1.48722e+06
+Conv2_f2h,0,0
+Conv2_h2f,0,0
+Add2,0.242274,13321.9
+Add2_f2h,0,0
+Add2_h2f,0,0
+Tanh2,0.179781,13329.7
+Tanh2_f2h,0,0
+Tanh2_h2f,0,0
+Pool1,2.60261,28731.1
+Pool1_f2h,0,0
+Pool1_h2f,0,0
+Conv3,55.4282,523293
+Conv3_f2h,0,0
+Conv3_h2f,0,0
+Add3,0.211531,13627
+Add3_f2h,0,0
+Add3_h2f,0,0
+Tanh3,0.179788,13630.8
+Tanh3_f2h,0,0
+Tanh3_h2f,0,0
+Conv4,82.562,806140
+Conv4_f2h,0,0
+Conv4_h2f,0,0
+Add4,0.20909,13773.5
+Add4_f2h,0,0
+Add4_h2f,0,0
+Tanh4,0.179433,13771.6
+Tanh4_f2h,0,0
+Tanh4_h2f,0,0
+Pool2,1.89329,27557.9
+Pool2_f2h,0,0
+Pool2_h2f,0,0
+Conv5,32.1691,313581
+Conv5_f2h,0,0
+Conv5_h2f,0,0
+Add5,0.201173,13897.1
+Add5_f2h,0,0
+Add5_h2f,0,0
+Tanh5,0.15526,13902.6
+Tanh5_f2h,0,0
+Tanh5_h2f,0,0
+Conv6,48.4451,495873
+Conv6_f2h,0,0
+Conv6_h2f,0,0
+Add6,0.200427,14025.9
+Add6_f2h,0,0
+Add6_h2f,0,0
+Tanh6,0.15853,14018.5
+Tanh6_f2h,0,0
+Tanh6_h2f,0,0
+Pool3,1.21556,14018.5
+Pool3_f2h,0,0
+Pool3_h2f,0,0
+Mul1,1.20271,14854.8
+Mul1_f2h,0,0
+Mul1_h2f,0,0
+Add7,0.135273,14049.7
+Add7_f2h,0,0
+Add7_h2f,0,0
+Softmax1,0.970621,14465.4
+Softmax1_f2h,0,0
+Softmax1_h2f,0,0

llvm/projects/hpvm-tensor-rt/PPoPP_results/profiling_results/alexnet2_cifar10/alexnet2_cifar10_fp16_perf_fp16_159.txt

+Conv1,59.2911,555019
+Conv1_f2h,0,0
+Conv1_h2f,0,0
+Add1,0.201084,12973.3
+Add1_f2h,0,0
+Add1_h2f,0,0
+Tanh1,0.170233,12969.6
+Tanh1_f2h,0,0
+Tanh1_h2f,0,0
+Conv2,156.931,1.49121e+06
+Conv2_f2h,0,0
+Conv2_h2f,0,0
+Add2,0.217797,13251.4
+Add2_f2h,0,0
+Add2_h2f,0,0
+Tanh2,0.174917,13245.8
+Tanh2_f2h,0,0
+Tanh2_h2f,0,0
+Pool1,2.71397,29190.6
+Pool1_f2h,0,0
+Pool1_h2f,0,0
+Conv3,57.8418,542994
+Conv3_f2h,0,0
+Conv3_h2f,0,0
+Add3,0.21652,13546.6
+Add3_f2h,0,0
+Add3_h2f,0,0
+Tanh3,0.178658,13552.4
+Tanh3_f2h,0,0
+Tanh3_h2f,0,0
+Conv4,87.9539,849490
+Conv4_f2h,0,0
+Conv4_h2f,0,0
+Add4,0.217259,13704.8
+Add4_f2h,0,0
+Add4_h2f,0,0
+Tanh4,0.179077,13693.4
+Tanh4_f2h,0,0
+Tanh4_h2f,0,0
+Pool2,2.03865,27430.8
+Pool2_f2h,0,0
+Pool2_h2f,0,0
+Conv5,32.5739,308590
+Conv5_f2h,0,0
+Conv5_h2f,0,0
+Add5,0.199641,13812
+Add5_f2h,0,0
+Add5_h2f,0,0
+Tanh5,0.161615,13811.6
+Tanh5_f2h,0,0
+Tanh5_h2f,0,0
+Conv6,49.129,497072
+Conv6_f2h,0,0
+Conv6_h2f,0,0
+Add6,0.208169,13936
+Add6_f2h,0,0
+Add6_h2f,0,0
+Tanh6,0.161935,13937.9
+Tanh6_f2h,0,0
+Tanh6_h2f,0,0
+Pool3,1.28389,13935.6
+Pool3_f2h,0,0
+Pool3_h2f,0,0
+Mul1,1.25362,14764.7
+Mul1_f2h,0,0
+Mul1_h2f,0,0
+Add7,0.163733,13937.3
+Add7_f2h,0,0
+Add7_h2f,0,0
+Softmax1,1.00077,14665.8
+Softmax1_f2h,0,0
+Softmax1_h2f,0,0

llvm/projects/hpvm-tensor-rt/PPoPP_results/profiling_results/alexnet2_cifar10/alexnet2_cifar10_fp16_perf_fp16_160.txt

+Conv1,58.8135,556846
+Conv1_f2h,0,0
+Conv1_h2f,0,0
+Add1,0.203772,13190.9
+Add1_f2h,0,0
+Add1_h2f,0,0
+Tanh1,0.176348,13196.6
+Tanh1_f2h,0,0
+Tanh1_h2f,0,0
+Conv2,158.962,1.53725e+06
+Conv2_f2h,0,0
+Conv2_h2f,0,0
+Add2,0.219541,13440.8
+Add2_f2h,0,0
+Add2_h2f,0,0
+Tanh2,0.178588,13431.4
+Tanh2_f2h,0,0
+Tanh2_h2f,0,0
+Pool1,2.68237,28875.2
+Pool1_f2h,0,0
+Pool1_h2f,0,0
+Conv3,56.3609,533809
+Conv3_f2h,0,0
+Conv3_h2f,0,0
+Add3,0.210882,13724.2
+Add3_f2h,0,0
+Add3_h2f,0,0
+Tanh3,0.176328,13726.1
+Tanh3_f2h,0,0
+Tanh3_h2f,0,0
+Conv4,86.1982,845559
+Conv4_f2h,0,0
+Conv4_h2f,0,0
+Add4,0.214652,13882.4
+Add4_f2h,0,0
+Add4_h2f,0,0
+Tanh4,0.173688,13888.1
+Tanh4_f2h,0,0
+Tanh4_h2f,0,0
+Pool2,1.96952,27069.4
+Pool2_f2h,0,0
+Pool2_h2f,0,0
+Conv5,34.5895,340452
+Conv5_f2h,0,0
+Conv5_h2f,0,0
+Add5,0.200648,14023.1
+Add5_f2h,0,0
+Add5_h2f,0,0
+Tanh5,0.159231,14025
+Tanh5_f2h,0,0
+Tanh5_h2f,0,0
+Conv6,52.939,546006
+Conv6_f2h,0,0
+Conv6_h2f,0,0
+Add6,0.197173,14157.7
+Add6_f2h,0,0
+Add6_h2f,0,0
+Tanh6,0.150853,14157.7
+Tanh6_f2h,0,0
+Tanh6_h2f,0,0
+Pool3,1.16124,14167.2
+Pool3_f2h,0,0
+Pool3_h2f,0,0
+Mul1,1.17195,15013
+Mul1_f2h,0,0
+Mul1_h2f,0,0
+Add7,0.157372,14168.5
+Add7_f2h,0,0
+Add7_h2f,0,0
+Softmax1,0.983837,14180.1
+Softmax1_f2h,0,0
+Softmax1_h2f,0,0

llvm/projects/hpvm-tensor-rt/PPoPP_results/profiling_results/alexnet2_cifar10/alexnet2_cifar10_fp16_perf_fp16_161.txt

+Conv1,65.8243,634762
+Conv1_f2h,0,0
+Conv1_h2f,0,0
+Add1,0.205215,13383
+Add1_f2h,0,0
+Add1_h2f,0,0
+Tanh1,0.179957,13375.5
+Tanh1_f2h,0,0
+Tanh1_h2f,0,0
+Conv2,174.316,1.70456e+06
+Conv2_f2h,0,0
+Conv2_h2f,0,0
+Add2,0.219689,13640.9
+Add2_f2h,0,0
+Add2_h2f,0,0
+Tanh2,0.182165,13619.9
+Tanh2_f2h,0,0
+Tanh2_h2f,0,0
+Pool1,2.68031,29935.2
+Pool1_f2h,0,0
+Pool1_h2f,0,0
+Conv3,61.4199,594851
+Conv3_f2h,0,0
+Conv3_h2f,0,0
+Add3,0.215746,13926
+Add3_f2h,0,0
+Add3_h2f,0,0
+Tanh3,0.185436,13916.6
+Tanh3_f2h,0,0
+Tanh3_h2f,0,0
+Conv4,93.149,925082
+Conv4_f2h,0,0
+Conv4_h2f,0,0
+Add4,0.213054,14081
+Add4_f2h,0,0
+Add4_h2f,0,0
+Tanh4,0.180697,14075.4
+Tanh4_f2h,0,0
+Tanh4_h2f,0,0
+Pool2,1.99828,28166.8
+Pool2_f2h,0,0
+Pool2_h2f,0,0
+Conv5,35.2346,352384
+Conv5_f2h,0,0
+Conv5_h2f,0,0
+Add5,0.204687,14205.3
+Add5_f2h,0,0
+Add5_h2f,0,0
+Tanh5,0.157567,14209.1
+Tanh5_f2h,0,0
+Tanh5_h2f,0,0
+Conv6,53.4125,557191
+Conv6_f2h,0,0
+Conv6_h2f,0,0
+Add6,0.20316,14374.6
+Add6_f2h,0,0
+Add6_h2f,0,0
+Tanh6,0.159049,14361.3
+Tanh6_f2h,0,0
+Tanh6_h2f,0,0
+Pool3,1.19671,14363.3
+Pool3_f2h,0,0
+Pool3_h2f,0,0
+Mul1,1.18429,15228.9
+Mul1_f2h,0,0
+Mul1_h2f,0,0
+Add7,0.17158,14372.9
+Add7_f2h,0,0
+Add7_h2f,0,0
+Softmax1,0.963239,14813.3
+Softmax1_f2h,0,0
+Softmax1_h2f,0,0

llvm/projects/hpvm-tensor-rt/PPoPP_results/profiling_results/alexnet2_cifar10/alexnet2_cifar10_fp16_perf_fp16_162.txt

+Conv1,66.1421,634983
+Conv1_f2h,0,0
+Conv1_h2f,0,0
+Add1,0.21741,13337.1
+Add1_f2h,0,0
+Add1_h2f,0,0
+Tanh1,0.185688,13331.4
+Tanh1_f2h,0,0
+Tanh1_h2f,0,0
+Conv2,176.119,1.69722e+06
+Conv2_f2h,0,0
+Conv2_h2f,0,0
+Add2,0.227336,13574
+Add2_f2h,0,0
+Add2_h2f,0,0
+Tanh2,0.188882,13570.3
+Tanh2_f2h,0,0
+Tanh2_h2f,0,0
+Pool1,2.6811,30306.7
+Pool1_f2h,0,0
+Pool1_h2f,0,0
+Conv3,61.7431,583447
+Conv3_f2h,0,0
+Conv3_h2f,0,0
+Add3,0.219186,13881.5
+Add3_f2h,0,0
+Add3_h2f,0,0
+Tanh3,0.181445,13877.7
+Tanh3_f2h,0,0
+Tanh3_h2f,0,0
+Conv4,93.914,932514
+Conv4_f2h,0,0
+Conv4_h2f,0,0
+Add4,0.219256,14007.1
+Add4_f2h,0,0
+Add4_h2f,0,0
+Tanh4,0.1794,13997.6
+Tanh4_f2h,0,0
+Tanh4_h2f,0,0
+Pool2,1.93367,28014.6
+Pool2_f2h,0,0
+Pool2_h2f,0,0
+Conv5,35.3477,350762
+Conv5_f2h,0,0
+Conv5_h2f,0,0
+Add5,0.208718,14161.3
+Add5_f2h,0,0
+Add5_h2f,0,0
+Tanh5,0.161401,14163.3
+Tanh5_f2h,0,0
+Tanh5_h2f,0,0
+Conv6,53.9521,560676
+Conv6_f2h,0,0
+Conv6_h2f,0,0
+Add6,0.204236,14322.6
+Add6_f2h,0,0
+Add6_h2f,0,0
+Tanh6,0.157525,14318.9
+Tanh6_f2h,0,0
+Tanh6_h2f,0,0
+Pool3,1.23967,14322.8
+Pool3_f2h,0,0
+Pool3_h2f,0,0
+Mul1,1.22098,15186.2
+Mul1_f2h,0,0
+Mul1_h2f,0,0
+Add7,0.169871,14326.6
+Add7_f2h,0,0
+Add7_h2f,0,0
+Softmax1,0.978561,14755.4
+Softmax1_f2h,0,0
+Softmax1_h2f,0,0

llvm/projects/hpvm-tensor-rt/PPoPP_results/profiling_results/alexnet2_cifar10/alexnet2_cifar10_fp16_perf_fp16_163.txt

+Conv1,66.2346,634503
+Conv1_f2h,0,0
+Conv1_h2f,0,0
+Add1,0.209573,13306.7
+Add1_f2h,0,0
+Add1_h2f,0,0
+Tanh1,0.174533,13306.7
+Tanh1_f2h,0,0
+Tanh1_h2f,0,0
+Conv2,176.314,1.70236e+06
+Conv2_f2h,0,0
+Conv2_h2f,0,0
+Add2,0.234539,13560.7
+Add2_f2h,0,0
+Add2_h2f,0,0
+Tanh2,0.184031,13556.8
+Tanh2_f2h,0,0
+Tanh2_h2f,0,0
+Pool1,2.71175,30262.4
+Pool1_f2h,0,0
+Pool1_h2f,0,0
+Conv3,61.7066,582837
+Conv3_f2h,0,0
+Conv3_h2f,0,0
+Add3,0.221103,13124.2
+Add3_f2h,0,0
+Add3_h2f,0,0
+Tanh3,0.177237,13110.9
+Tanh3_f2h,0,0
+Tanh3_h2f,0,0
+Conv4,94.1784,928952
+Conv4_f2h,0,0
+Conv4_h2f,0,0
+Add4,0.219455,13997.5
+Add4_f2h,0,0
+Add4_h2f,0,0
+Tanh4,0.18468,13999.4
+Tanh4_f2h,0,0
+Tanh4_h2f,0,0
+Pool2,1.97442,28016.3
+Pool2_f2h,0,0
+Pool2_h2f,0,0
+Conv5,35.4772,352075
+Conv5_f2h,0,0
+Conv5_h2f,0,0
+Add5,0.209503,14157.7
+Add5_f2h,0,0
+Add5_h2f,0,0
+Tanh5,0.161119,14161.6
+Tanh5_f2h,0,0
+Tanh5_h2f,0,0
+Conv6,54.1191,562078
+Conv6_f2h,0,0
+Conv6_h2f,0,0
+Add6,0.207996,14315.5
+Add6_f2h,0,0
+Add6_h2f,0,0
+Tanh6,0.160258,14309.8
+Tanh6_f2h,0,0
+Tanh6_h2f,0,0
+Pool3,1.23714,14309.8
+Pool3_f2h,0,0
+Pool3_h2f,0,0
+Mul1,1.21279,15183.1
+Mul1_f2h,0,0
+Mul1_h2f,0,0
+Add7,0.175522,14321.3
+Add7_f2h,0,0
+Add7_h2f,0,0
+Softmax1,0.977623,14744.6
+Softmax1_f2h,0,0
+Softmax1_h2f,0,0

llvm/projects/hpvm-tensor-rt/PPoPP_results/profiling_results/alexnet2_cifar10/alexnet2_cifar10_fp16_perf_fp16_164.txt

+Conv1,65.9398,633363
+Conv1_f2h,0,0
+Conv1_h2f,0,0
+Add1,0.205851,13291
+Add1_f2h,0,0
+Add1_h2f,0,0
+Tanh1,0.172959,13287.3
+Tanh1_f2h,0,0
+Tanh1_h2f,0,0
+Conv2,174.804,1.69553e+06
+Conv2_f2h,0,0
+Conv2_h2f,0,0
+Add2,0.220789,13560.8
+Add2_f2h,0,0
+Add2_h2f,0,0
+Tanh2,0.177087,13564.6
+Tanh2_f2h,0,0
+Tanh2_h2f,0,0
+Pool1,2.81027,31205.1
+Pool1_f2h,0,0
+Pool1_h2f,0,0
+Conv3,61.5282,590104
+Conv3_f2h,0,0
+Conv3_h2f,0,0
+Add3,0.209931,13849.5
+Add3_f2h,0,0
+Add3_h2f,0,0
+Tanh3,0.18996,13845.8
+Tanh3_f2h,0,0
+Tanh3_h2f,0,0
+Conv4,93.4214,924284
+Conv4_f2h,0,0
+Conv4_h2f,0,0
+Add4,0.21342,13978.9
+Add4_f2h,0,0
+Add4_h2f,0,0
+Tanh4,0.184242,13988.7
+Tanh4_f2h,0,0
+Tanh4_h2f,0,0
+Pool2,2.0622,27996.2
+Pool2_f2h,0,0
+Pool2_h2f,0,0
+Conv5,34.9337,342861
+Conv5_f2h,0,0
+Conv5_h2f,0,0
+Add5,0.202178,14110.7
+Add5_f2h,0,0
+Add5_h2f,0,0
+Tanh5,0.160732,14112.6
+Tanh5_f2h,0,0
+Tanh5_h2f,0,0
+Conv6,53.8579,556438
+Conv6_f2h,0,0
+Conv6_h2f,0,0
+Add6,0.199122,14251.5
+Add6_f2h,0,0
+Add6_h2f,0,0
+Tanh6,0.157836,14257.4
+Tanh6_f2h,0,0
+Tanh6_h2f,0,0
+Pool3,1.22002,14266.9
+Pool3_f2h,0,0
+Pool3_h2f,0,0
+Mul1,1.23049,15130.6
+Mul1_f2h,0,0
+Mul1_h2f,0,0
+Add7,0.172575,14268.8
+Add7_f2h,0,0
+Add7_h2f,0,0
+Softmax1,1.02197,15447.3
+Softmax1_f2h,0,0
+Softmax1_h2f,0,0

llvm/projects/hpvm-tensor-rt/PPoPP_results/profiling_results/alexnet2_cifar10/alexnet2_cifar10_fp16_perf_fp16_165.txt

+Conv1,61.5774,582474
+Conv1_f2h,0,0
+Conv1_h2f,0,0
+Add1,0.210591,13093.8
+Add1_f2h,0,0
+Add1_h2f,0,0
+Tanh1,0.16853,13084.1
+Tanh1_f2h,0,0
+Tanh1_h2f,0,0
+Conv2,171.588,1.6167e+06
+Conv2_f2h,0,0
+Conv2_h2f,0,0
+Add2,0.230504,13322.4
+Add2_f2h,0,0
+Add2_h2f,0,0
+Tanh2,0.176787,13316.6
+Tanh2_f2h,0,0
+Tanh2_h2f,0,0
+Pool1,2.79893,31381.1
+Pool1_f2h,0,0
+Pool1_h2f,0,0
+Conv3,60.6244,573649
+Conv3_f2h,0,0
+Conv3_h2f,0,0
+Add3,0.223733,13607.6
+Add3_f2h,0,0
+Add3_h2f,0,0
+Tanh3,0.179596,13617.2
+Tanh3_f2h,0,0
+Tanh3_h2f,0,0
+Conv4,92.717,894994
+Conv4_f2h,0,0
+Conv4_h2f,0,0
+Add4,0.221378,13754.4
+Add4_f2h,0,0
+Add4_h2f,0,0
+Tanh4,0.177762,13744.8
+Tanh4_f2h,0,0
+Tanh4_h2f,0,0
+Pool2,2.00882,27528.1
+Pool2_f2h,0,0
+Pool2_h2f,0,0
+Conv5,34.7619,342294
+Conv5_f2h,0,0
+Conv5_h2f,0,0
+Add5,0.204523,13901.8
+Add5_f2h,0,0
+Add5_h2f,0,0
+Tanh5,0.15829,13901.8
+Tanh5_f2h,0,0
+Tanh5_h2f,0,0
+Conv6,53.3066,543043
+Conv6_f2h,0,0
+Conv6_h2f,0,0
+Add6,0.203224,14037
+Add6_f2h,0,0
+Add6_h2f,0,0
+Tanh6,0.161542,14042.7
+Tanh6_f2h,0,0
+Tanh6_h2f,0,0
+Pool3,1.26148,14044.6
+Pool3_f2h,0,0
+Pool3_h2f,0,0
+Mul1,1.21381,14900.5
+Mul1_f2h,0,0
+Mul1_h2f,0,0
+Add7,0.171011,14056
+Add7_f2h,0,0
+Add7_h2f,0,0
+Softmax1,0.964679,14479
+Softmax1_f2h,0,0
+Softmax1_h2f,0,0

llvm/projects/hpvm-tensor-rt/PPoPP_results/profiling_results/alexnet2_cifar10/alexnet2_cifar10_fp16_perf_fp16_166.txt

+Conv1,61.7188,579843
+Conv1_f2h,0,0
+Conv1_h2f,0,0
+Add1,0.205929,13080.3
+Add1_f2h,0,0
+Add1_h2f,0,0
+Tanh1,0.170476,13046.2
+Tanh1_f2h,0,0
+Tanh1_h2f,0,0
+Conv2,170.797,1.62614e+06
+Conv2_f2h,0,0
+Conv2_h2f,0,0
+Add2,0.221976,13303.2
+Add2_f2h,0,0
+Add2_h2f,0,0
+Tanh2,0.172012,13305.2
+Tanh2_f2h,0,0
+Tanh2_h2f,0,0
+Pool1,2.66512,28575.5
+Pool1_f2h,0,0
+Pool1_h2f,0,0
+Conv3,60.6009,571260
+Conv3_f2h,0,0
+Conv3_h2f,0,0
+Add3,0.212117,13600.1
+Add3_f2h,0,0
+Add3_h2f,0,0
+Tanh3,0.198482,13602
+Tanh3_f2h,0,0
+Tanh3_h2f,0,0
+Conv4,92.7343,897664
+Conv4_f2h,0,0
+Conv4_h2f,0,0
+Add4,0.214779,13754.5
+Add4_f2h,0,0
+Add4_h2f,0,0
+Tanh4,0.180697,13735.4
+Tanh4_f2h,0,0
+Tanh4_h2f,0,0
+Pool2,1.93166,27486
+Pool2_f2h,0,0
+Pool2_h2f,0,0
+Conv5,35.0139,341225
+Conv5_f2h,0,0
+Conv5_h2f,0,0
+Add5,0.198821,13911.2
+Add5_f2h,0,0
+Add5_h2f,0,0
+Tanh5,0.15717,13913.2
+Tanh5_f2h,0,0
+Tanh5_h2f,0,0
+Conv6,53.8398,547490
+Conv6_f2h,0,0
+Conv6_h2f,0,0
+Add6,0.200568,14061.1
+Add6_f2h,0,0
+Add6_h2f,0,0
+Tanh6,0.154255,14051.6
+Tanh6_f2h,0,0
+Tanh6_h2f,0,0
+Pool3,1.22431,14057.3
+Pool3_f2h,0,0
+Pool3_h2f,0,0
+Mul1,1.21994,14922.9
+Mul1_f2h,0,0
+Mul1_h2f,0,0
+Add7,0.143932,14068.7
+Add7_f2h,0,0
+Add7_h2f,0,0
+Softmax1,0.950046,14491.9
+Softmax1_f2h,0,0
+Softmax1_h2f,0,0

llvm/projects/hpvm-tensor-rt/PPoPP_results/profiling_results/alexnet2_cifar10/alexnet2_cifar10_fp16_perf_fp16_167.txt

+Conv1,60.3008,566150
+Conv1_f2h,0,0
+Conv1_h2f,0,0
+Add1,0.212953,13192.9
+Add1_f2h,0,0
+Add1_h2f,0,0
+Tanh1,0.16919,13192.9
+Tanh1_f2h,0,0
+Tanh1_h2f,0,0
+Conv2,168.206,1.62008e+06
+Conv2_f2h,0,0
+Conv2_h2f,0,0
+Add2,0.216709,13429.6
+Add2_f2h,0,0
+Add2_h2f,0,0
+Tanh2,0.176031,13418.5
+Tanh2_f2h,0,0
+Tanh2_h2f,0,0
+Pool1,2.62383,28235.3
+Pool1_f2h,0,0
+Pool1_h2f,0,0
+Conv3,59.0124,557003
+Conv3_f2h,0,0
+Conv3_h2f,0,0
+Add3,0.2075,13707.2
+Add3_f2h,0,0
+Add3_h2f,0,0
+Tanh3,0.183634,13707.1
+Tanh3_f2h,0,0
+Tanh3_h2f,0,0
+Conv4,90.5798,881313
+Conv4_f2h,0,0
+Conv4_h2f,0,0
+Add4,0.214488,13884.5
+Add4_f2h,0,0
+Add4_h2f,0,0
+Tanh4,0.174665,13865.5
+Tanh4_f2h,0,0
+Tanh4_h2f,0,0
+Pool2,1.87406,27761.5
+Pool2_f2h,0,0
+Pool2_h2f,0,0
+Conv5,34.5362,338970
+Conv5_f2h,0,0
+Conv5_h2f,0,0
+Add5,0.197125,14010.3
+Add5_f2h,0,0
+Add5_h2f,0,0
+Tanh5,0.156981,14014.1
+Tanh5_f2h,0,0
+Tanh5_h2f,0,0
+Conv6,53.1529,543496
+Conv6_f2h,0,0
+Conv6_h2f,0,0
+Add6,0.199279,14147.7
+Add6_f2h,0,0
+Add6_h2f,0,0
+Tanh6,0.152859,14148
+Tanh6_f2h,0,0
+Tanh6_h2f,0,0
+Pool3,1.22611,14878.4
+Pool3_f2h,0,0
+Pool3_h2f,0,0
+Mul1,1.14545,15004
+Mul1_f2h,0,0
+Mul1_h2f,0,0
+Add7,0.159065,14156.8
+Add7_f2h,0,0
+Add7_h2f,0,0
+Softmax1,0.967012,15316.6
+Softmax1_f2h,0,0
+Softmax1_h2f,0,0

llvm/projects/hpvm-tensor-rt/PPoPP_results/profiling_results/alexnet2_cifar10/alexnet2_cifar10_fp16_perf_fp16_168.txt

+Conv1,60.0049,566283
+Conv1_f2h,0,0
+Conv1_h2f,0,0
+Add1,0.198763,13133.8
+Add1_f2h,0,0
+Add1_h2f,0,0
+Tanh1,0.171007,13131.9
+Tanh1_f2h,0,0
+Tanh1_h2f,0,0
+Conv2,168.986,1.60868e+06
+Conv2_f2h,0,0
+Conv2_h2f,0,0
+Add2,0.214533,13381.9
+Add2_f2h,0,0
+Add2_h2f,0,0
+Tanh2,0.17733,13378.1
+Tanh2_f2h,0,0
+Tanh2_h2f,0,0
+Pool1,2.72244,29920.6
+Pool1_f2h,0,0
+Pool1_h2f,0,0
+Conv3,59.1517,556478
+Conv3_f2h,0,0
+Conv3_h2f,0,0
+Add3,0.209778,13645.8
+Add3_f2h,0,0
+Add3_h2f,0,0
+Tanh3,0.183487,13653.4
+Tanh3_f2h,0,0
+Tanh3_h2f,0,0
+Conv4,91.8964,875586
+Conv4_f2h,0,0
+Conv4_h2f,0,0
+Add4,0.218393,13825.4
+Add4_f2h,0,0
+Add4_h2f,0,0
+Tanh4,0.178373,13823.5
+Tanh4_f2h,0,0
+Tanh4_h2f,0,0
+Pool2,2.097,27656.5
+Pool2_f2h,0,0
+Pool2_h2f,0,0
+Conv5,34.1944,334924
+Conv5_f2h,0,0
+Conv5_h2f,0,0
+Add5,0.199813,13966.6
+Add5_f2h,0,0
+Add5_h2f,0,0
+Tanh5,0.157509,13945.6
+Tanh5_f2h,0,0
+Tanh5_h2f,0,0
+Conv6,52.856,538205
+Conv6_f2h,0,0
+Conv6_h2f,0,0
+Add6,0.201836,14090.9
+Add6_f2h,0,0
+Add6_h2f,0,0
+Tanh6,0.156297,14091.1
+Tanh6_f2h,0,0
+Tanh6_h2f,0,0
+Pool3,1.21402,14100.6
+Pool3_f2h,0,0
+Pool3_h2f,0,0
+Mul1,1.20897,14950.6
+Mul1_f2h,0,0
+Mul1_h2f,0,0
+Add7,0.173667,14119.4
+Add7_f2h,0,0
+Add7_h2f,0,0
+Softmax1,0.958647,14548.3
+Softmax1_f2h,0,0
+Softmax1_h2f,0,0