Created by: wangchaochaohu
修正ParallelEvent Record 记录不正确的问题
This PR
Total time: 723.26
Computation time Total: 99.9073 Ratio: 13.8135%
Framework overhead Total: 623.353 Ratio: 86.1865%
------------------------- GpuMemCpy Summary -------------------------
GpuMemcpy Calls: 510 Total: 14.4704 Ratio: 2.00071%
GpuMemcpyAsync Calls: 510 Total: 14.4704 Ratio: 2.00071%
------------------------- Event Summary -------------------------
Event Calls Total CPU Time (Ratio) GPU Time (Ratio) Min. Max. Ave. Ratio.
GpuMemcpyAsync:CPU->GPU 3 0.807003 0.794587 (0.984615) 0.012416 (0.015385) 0.018948 0.762712 0.269001 0.00111579
ParallelExecutor::Run 1 722.453 722.360589 (0.999872) 0.092511 (0.000128) 722.453 722.453 722.453 0.998884
BufferedReader:MemoryCopy 2 556.998 556.988530 (0.999983) 0.009216 (0.000017) 0.061644 556.936 278.499 0.770121
GpuMemcpyAsync:CUDAPinned->GPU 4 0.120022 0.110806 (0.923214) 0.009216 (0.076786) 0.011017 0.073242 0.0300055 0.000165946
transpose2 6 0.482985 0.466345 (0.965548) 0.016640 (0.034452) 0.045219 0.210003 0.0804975 0.000667789
transpose20 1 0.196926 0.193022 (0.980175) 0.003904 (0.019825) 0.196926 0.196926 0.196926 0.000272276
transpose20/prepare_data 1 0.010146 0.010146 (1.000000) 0.000000 (0.000000) 0.010146 0.010146 0.010146 1.40281e-05
transpose20/infer_shape 1 0.015754 0.015754 (1.000000) 0.000000 (0.000000) 0.015754 0.015754 0.015754 2.17819e-05
transpose20/compute 1 0.119195 0.115291 (0.967247) 0.003904 (0.032753) 0.119195 0.119195 0.119195 0.000164802
transpose21 1 0.048696 0.046072 (0.946115) 0.002624 (0.053885) 0.048696 0.048696 0.048696 6.73285e-05
transpose21/prepare_data 1 0.001609 0.001609 (1.000000) 0.000000 (0.000000) 0.001609 0.001609 0.001609 2.22465e-06
transpose21/infer_shape 1 0.009051 0.009051 (1.000000) 0.000000 (0.000000) 0.009051 0.009051 0.009051 1.25142e-05
transpose21/compute 1 0.027426 0.024802 (0.904324) 0.002624 (0.095676) 0.027426 0.027426 0.027426 3.792e-05
transpose22 1 0.064922 0.062362 (0.960568) 0.002560 (0.039432) 0.064922 0.064922 0.064922 8.9763e-05
transpose22/prepare_data 1 0.001643 0.001643 (1.000000) 0.000000 (0.000000) 0.001643 0.001643 0.001643 2.27166e-06
transpose22/infer_shape 1 0.005114 0.005114 (1.000000) 0.000000 (0.000000) 0.005114 0.005114 0.005114 7.07076e-06
transpose22/compute 1 0.0444 0.041840 (0.942342) 0.002560 (0.057658) 0.0444 0.0444 0.0444 6.13887e-05
transpose23 1 0.054668 0.052364 (0.957855) 0.002304 (0.042145) 0.054668 0.054668 0.054668 7.55855e-05
transpose23/prepare_data 1 0.001619 0.001619 (1.000000) 0.000000 (0.000000) 0.001619 0.001619 0.001619 2.23848e-06
transpose23/infer_shape 1 0.005287 0.005287 (1.000000) 0.000000 (0.000000) 0.005287 0.005287 0.005287 7.30996e-06
transpose23/compute 1 0.032872 0.030568 (0.929910) 0.002304 (0.070090) 0.032872 0.032872 0.032872 4.54498e-05
transpose24 1 0.045496 0.043256 (0.950765) 0.002240 (0.049235) 0.045496 0.045496 0.045496 6.29041e-05
transpose24/prepare_data 1 0.001593 0.001593 (1.000000) 0.000000 (0.000000) 0.001593 0.001593 0.001593 2.20253e-06
transpose24/infer_shape 1 0.004263 0.004263 (1.000000) 0.000000 (0.000000) 0.004263 0.004263 0.004263 5.89415e-06
transpose24/compute 1 0.028585 0.026345 (0.921637) 0.002240 (0.078363) 0.028585 0.028585 0.028585 3.95224e-05
transpose25 1 0.042416 0.039408 (0.929083) 0.003008 (0.070917) 0.042416 0.042416 0.042416 5.86456e-05
transpose25/prepare_data 1 0.002254 0.002254 (1.000000) 0.000000 (0.000000) 0.002254 0.002254 0.002254 3.11644e-06
transpose25/infer_shape 1 0.004733 0.004733 (1.000000) 0.000000 (0.000000) 0.004733 0.004733 0.004733 6.54398e-06
transpose25/compute 1 0.025669 0.022661 (0.882816) 0.003008 (0.117184) 0.025669 0.025669 0.025669 3.54907e-05
eager_deletion 60 0.215596 0.215596 (1.000000) 0.000000 (0.000000) 0.001481 0.016605 0.00359327 0.000298089
reshape2 10 0.279478 0.279478 (1.000000) 0.000000 (0.000000) 0.017544 0.052239 0.0279478 0.000386414
reshape20 1 0.038727 0.038727 (1.000000) 0.000000 (0.000000) 0.038727 0.038727 0.038727 5.35451e-05
reshape20/prepare_data 1 0.0071 0.007100 (1.000000) 0.000000 (0.000000) 0.0071 0.0071 0.0071 9.81666e-06
reshape20/infer_shape 1 0.011978 0.011978 (1.000000) 0.000000 (0.000000) 0.011978 0.011978 0.011978 1.65611e-05
reshape20/compute 1 0.006778 0.006778 (1.000000) 0.000000 (0.000000) 0.006778 0.006778 0.006778 9.37146e-06
reshape21 1 0.016051 0.016051 (1.000000) 0.000000 (0.000000) 0.016051 0.016051 0.016051 2.21926e-05
reshape21/prepare_data 1 0.001835 0.001835 (1.000000) 0.000000 (0.000000) 0.001835 0.001835 0.001835 2.53712e-06
reshape21/infer_shape 1 0.004135 0.004135 (1.000000) 0.000000 (0.000000) 0.004135 0.004135 0.004135 5.71717e-06
reshape21/compute 1 0.002453 0.002453 (1.000000) 0.000000 (0.000000) 0.002453 0.002453 0.002453 3.39159e-06
reshape22 1 0.014828 0.014828 (1.000000) 0.000000 (0.000000) 0.014828 0.014828 0.014828 2.05016e-05
reshape22/prepare_data 1 0.001693 0.001693 (1.000000) 0.000000 (0.000000) 0.001693 0.001693 0.001693 2.34079e-06
reshape22/infer_shape 1 0.003959 0.003959 (1.000000) 0.000000 (0.000000) 0.003959 0.003959 0.003959 5.47383e-06
reshape22/compute 1 0.002237 0.002237 (1.000000) 0.000000 (0.000000) 0.002237 0.002237 0.002237 3.09294e-06develop
Total time: 1717.3
Computation time Total: 53.8552 Ratio: 3.13604%
Framework overhead Total: 1663.44 Ratio: 96.864%
------------------------- GpuMemCpy Summary -------------------------
GpuMemcpy Calls: 510 Total: 12.8937 Ratio: 0.750814%
GpuMemcpyAsync Calls: 510 Total: 12.8937 Ratio: 0.750814%
------------------------- Event Summary -------------------------
Event Calls Total CPU Time (Ratio) GPU Time (Ratio) Min. Max. Ave. Ratio.
BufferedReader:MemoryCopy 2 512.126 512.116729 (0.999982) 0.009312 (0.000018) 0.061072 512.065 256.063 0.298216
GpuMemcpyAsync:CUDAPinned->GPU 4 0.125798 0.116486 (0.925977) 0.009312 (0.074023) 0.012689 0.068574 0.0314495 7.32535e-05
create_double_buffer_reader 1 0.13001 0.130010 (1.000000) 0.000000 (0.000000) 0.13001 0.13001 0.13001 7.57061e-05
create_double_buffer_reader0 1 0.121924 0.121924 (1.000000) 0.000000 (0.000000) 0.121924 0.121924 0.121924 7.09976e-05
read 1 512.231 512.230721 (1.000000) 0.000000 (0.000000) 512.231 512.231 512.231 0.298277
read0 1 512.228 512.228016 (1.000000) 0.000000 (0.000000) 512.228 512.228 512.228 0.298276
read 1 512.215 512.214651 (1.000000) 0.000000 (0.000000) 512.215 512.215 512.215 0.298268
lookup_table 1 0.30013 0.289570 (0.964815) 0.010560 (0.035185) 0.30013 0.30013 0.30013 0.000174769
lookup_table0 1 0.295008 0.284448 (0.964204) 0.010560 (0.035796) 0.295008 0.295008 0.295008 0.000171786
lookup_table0/prepare_data 1 0.00334 0.003340 (1.000000) 0.000000 (0.000000) 0.00334 0.00334 0.00334 1.94492e-06
lookup_table0/infer_shape 1 0.019978 0.019978 (1.000000) 0.000000 (0.000000) 0.019978 0.019978 0.019978 1.16334e-05
lookup_table0/compute 1 0.235805 0.225245 (0.955217) 0.010560 (0.044783) 0.235805 0.235805 0.235805 0.000137312
reshape2 10 0.197373 0.197373 (1.000000) 0.000000 (0.000000) 0.014707 0.04142 0.0197373 0.000114932
reshape20 1 0.035708 0.035708 (1.000000) 0.000000 (0.000000) 0.035708 0.035708 0.035708 2.07931e-05
reshape20/prepare_data 1 0.003174 0.003174 (1.000000) 0.000000 (0.000000) 0.003174 0.003174 0.003174 1.84825e-06
reshape20/infer_shape 1 0.01177 0.011770 (1.000000) 0.000000 (0.000000) 0.01177 0.01177 0.01177 6.85379e-06
reshape20/compute 1 0.006402 0.006402 (1.000000) 0.000000 (0.000000) 0.006402 0.006402 0.006402 3.72795e-06
reshape21 1 0.01533 0.015330 (1.000000) 0.000000 (0.000000) 0.01533 0.01533 0.01533 8.92681e-06
reshape21/prepare_data 1 0.001872 0.001872 (1.000000) 0.000000 (0.000000) 0.001872 0.001872 0.001872 1.09008e-06
reshape21/infer_shape 1 0.003927 0.003927 (1.000000) 0.000000 (0.000000) 0.003927 0.003927 0.003927 2.28673e-06
reshape21/compute 1 0.002166 0.002166 (1.000000) 0.000000 (0.000000) 0.002166 0.002166 0.002166 1.26128e-06
reshape22 1 0.017729 0.017729 (1.000000) 0.000000 (0.000000) 0.017729 0.017729 0.017729 1.03238e-05
reshape22/prepare_data 1 0.001475 0.001475 (1.000000) 0.000000 (0.000000) 0.001475 0.001475 0.001475 8.58907e-07
reshape22/infer_shape 1 0.003906 0.003906 (1.000000) 0.000000 (0.000000) 0.003906 0.003906 0.003906 2.2745e-06
reshape22/compute 1 0.002105 0.002105 (1.000000) 0.000000 (0.000000) 0.002105 0.002105 0.002105 1.22576e-06
reshape23 1 0.012867 0.012867 (1.000000) 0.000000 (0.000000) 0.012867 0.012867 0.012867 7.49258e-06
reshape23/prepare_data 1 0.001364 0.001364 (1.000000) 0.000000 (0.000000) 0.001364 0.001364 0.001364 7.94271e-07
reshape23/infer_shape 1 0.003345 0.003345 (1.000000) 0.000000 (0.000000) 0.003345 0.003345 0.003345 1.94783e-06
reshape23/compute 1 0.001864 0.001864 (1.000000) 0.000000 (0.000000) 0.001864 0.001864 0.001864 1.08543e-06
reshape24 1 0.012956 0.012956 (1.000000) 0.000000 (0.000000) 0.012956 0.012956 0.012956 7.54441e-06
reshape24/prepare_data 1 0.00137 0.001370 (1.000000) 0.000000 (0.000000) 0.00137 0.00137 0.00137 7.97765e-07
reshape24/infer_shape 1 0.003227 0.003227 (1.000000) 0.000000 (0.000000) 0.003227 0.003227 0.003227 1.87911e-06
reshape24/compute 1 0.001894 0.001894 (1.000000) 0.000000 (0.000000) 0.001894 0.001894 0.001894 1.1029e-06
reshape25 1 0.016259 0.016259 (1.000000) 0.000000 (0.000000) 0.016259 0.016259 0.016259 9.46778e-06
reshape25/prepare_data 1 0.001582 0.001582 (1.000000) 0.000000 (0.000000) 0.001582 0.001582 0.001582 9.21215e-07
reshape25/infer_shape 1 0.00446 0.004460 (1.000000) 0.000000 (0.000000) 0.00446 0.00446 0.00446 2.5971e-06
reshape25/compute 1 0.002128 0.002128 (1.000000) 0.000000 (0.000000) 0.002128 0.002128 0.002128 1.23916e-06
reshape26 1 0.013421 0.013421 (1.000000) 0.000000 (0.000000) 0.013421 0.013421 0.013421 7.81518e-06
reshape26/prepare_data 1 0.001382 0.001382 (1.000000) 0.000000 (0.000000) 0.001382 0.001382 0.001382 8.04753e-07
reshape26/infer_shape 1 0.003506 0.003506 (1.000000) 0.000000 (0.000000) 0.003506 0.003506 0.003506 2.04158e-06
reshape26/compute 1 0.001882 0.001882 (1.000000) 0.000000 (0.000000) 0.001882 0.001882 0.001882 1.09591e-06
reshape27 1 0.019242 0.019242 (1.000000) 0.000000 (0.000000) 0.019242 0.019242 0.019242 1.12048e-05
reshape27/prepare_data 1 0.002062 0.002062 (1.000000) 0.000000 (0.000000) 0.002062 0.002062 0.002062 1.20072e-06
reshape27/infer_shape 1 0.006968 0.006968 (1.000000) 0.000000 (0.000000) 0.006968 0.006968 0.006968 4.05754e-06
reshape27/compute 1 0.003008 0.003008 (1.000000) 0.000000 (0.000000) 0.003008 0.003008 0.003008 1.75159e-06
reshape28