Skip to content

Instantly share code, notes, and snippets.

@twmht
Created August 28, 2023 12:46
Show Gist options
  • Save twmht/e44e596b89f48be872588eaff621c845 to your computer and use it in GitHub Desktop.
Save twmht/e44e596b89f48be872588eaff621c845 to your computer and use it in GitHub Desktop.
[20:42:10] /home/acer/tvm/src/relay/transforms/convert_layout.cc:99: Warning: Desired layout(s) not specified for op: nn.max_pool2d
[20:42:10] /home/acer/tvm/src/relay/transforms/convert_layout.cc:99: Warning: Desired layout(s) not specified for op: nn.global_avg_pool2d
[20:42:14] /home/acer/tvm/src/relay/transforms/to_mixed_precision.cc:528: Warning: Op "layout_transform" not registered FTVMMixedPrecisionConversionType appears 2 times in graph.
[20:42:20] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:20] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 1.27395 ms, Memory: 19071503
[20:42:20] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 1.27405 ms, Memory: 19071503
[20:42:20] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 8.1408 ms, Memory: 0
[20:42:20] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:20] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:20] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:20] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:20] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
[20:42:20] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:20] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.243712 ms, Memory: 0
[20:42:20] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.866304 ms, Memory: 8033615
[20:42:20] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 73.6074 ms, Memory: 0
[20:42:20] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:20] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:20] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:20] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:20] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.272544 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.644576 ms, Memory: 8033615
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 47.2273 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.235776 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.631264 ms, Memory: 8033615
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 38.5931 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.255968 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.652928 ms, Memory: 8033615
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 38.5687 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.225984 ms, Memory: 1966719
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.392096 ms, Memory: 6071423
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 19.0812 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.243712 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.677888 ms, Memory: 294928
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 37.8674 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 0.127072 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.135776 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.236672 ms, Memory: 5923967
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.155968 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.622592 ms, Memory: 294928
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 37.0555 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.259872 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.694272 ms, Memory: 294928
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 38.2345 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.202336 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.401952 ms, Memory: 3547271
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 19.4576 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.154176 ms, Memory: 6912
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.361824 ms, Memory: 3154055
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 38.9446 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 0.091712 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.103424 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.173536 ms, Memory: 2957447
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.247808 ms, Memory: 6912
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.505792 ms, Memory: 3154055
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 38.9531 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:21] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.276288 ms, Memory: 6912
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.522944 ms, Memory: 3154055
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 38.9761 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.237312 ms, Memory: 3584
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.4912 ms, Memory: 3837775
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 18.8894 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.27968 ms, Memory: 524575
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.714752 ms, Memory: 5705551
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 36.0858 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 0.114752 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.161888 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.271296 ms, Memory: 1478479
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.2864 ms, Memory: 524575
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.731392 ms, Memory: 5705551
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 35.9467 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.284672 ms, Memory: 524575
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.735232 ms, Memory: 5705551
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 36.0818 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.752256 ms, Memory: 19071503
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.761184 ms, Memory: 19071503
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 4.96464 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:22] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
Get model...
Extract tasks...
Failed to create a ComputeDAG for auto_scheduler: Traceback (most recent call last):
4: TVMFuncCall
3: tvm::runtime::PackedFuncObj::Extractor<tvm::runtime::PackedFuncSubObj<tvm::runtime::TypedPackedFunc<tvm::auto_scheduler::ComputeDAG (tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)>::AssignTypedLambda<tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}>(tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >)::{lambda(tvm::runtime::TVMArgs const&, tvm::runtime::TVMRetValue*)#1}> >::Call(tvm::runtime::PackedFuncObj const*, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, tvm::runtime::TVMRetValue)
2: tvm::auto_scheduler::ComputeDAG::ComputeDAG(tvm::runtime::Array<tvm::te::Tensor, void>)
1: tvm::auto_scheduler::AccessAnalyzer::AccessAnalyzer(tvm::runtime::Array<tvm::te::Tensor, void> const&)
0: tvm::auto_scheduler::TopoSortOps(tvm::runtime::Array<tvm::te::Tensor, void> const&)
File "/home/acer/tvm/src/auto_scheduler/compute_dag.cc", line 97
TVMError: Unsupported op extern(y, 0x7f4324004980)
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.147872 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.5024 ms, Memory: 8033615
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 38.5011 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
Failed to create a ComputeDAG for auto_scheduler: Traceback (most recent call last):
4: TVMFuncCall
3: tvm::runtime::PackedFuncObj::Extractor<tvm::runtime::PackedFuncSubObj<tvm::runtime::TypedPackedFunc<tvm::auto_scheduler::ComputeDAG (tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)>::AssignTypedLambda<tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}>(tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >)::{lambda(tvm::runtime::TVMArgs const&, tvm::runtime::TVMRetValue*)#1}> >::Call(tvm::runtime::PackedFuncObj const*, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, tvm::runtime::TVMRetValue)
2: tvm::auto_scheduler::ComputeDAG::ComputeDAG(tvm::runtime::Array<tvm::te::Tensor, void>)
1: tvm::auto_scheduler::AccessAnalyzer::AccessAnalyzer(tvm::runtime::Array<tvm::te::Tensor, void> const&)
0: tvm::auto_scheduler::TopoSortOps(tvm::runtime::Array<tvm::te::Tensor, void> const&)
File "/home/acer/tvm/src/auto_scheduler/compute_dag.cc", line 97
TVMError: Unsupported op extern(y, 0x7f432422b8e0)
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.262176 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.651264 ms, Memory: 8033615
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 38.6261 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
Failed to create a ComputeDAG for auto_scheduler: Traceback (most recent call last):
4: TVMFuncCall
3: tvm::runtime::PackedFuncObj::Extractor<tvm::runtime::PackedFuncSubObj<tvm::runtime::TypedPackedFunc<tvm::auto_scheduler::ComputeDAG (tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)>::AssignTypedLambda<tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}>(tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >)::{lambda(tvm::runtime::TVMArgs const&, tvm::runtime::TVMRetValue*)#1}> >::Call(tvm::runtime::PackedFuncObj const*, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, tvm::runtime::TVMRetValue)
2: tvm::auto_scheduler::ComputeDAG::ComputeDAG(tvm::runtime::Array<tvm::te::Tensor, void>)
1: tvm::auto_scheduler::AccessAnalyzer::AccessAnalyzer(tvm::runtime::Array<tvm::te::Tensor, void> const&)
0: tvm::auto_scheduler::TopoSortOps(tvm::runtime::Array<tvm::te::Tensor, void> const&)
File "/home/acer/tvm/src/auto_scheduler/compute_dag.cc", line 97
TVMError: Unsupported op extern(y, 0x7f4324006510)
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.153152 ms, Memory: 1966719
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.250752 ms, Memory: 6071423
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 19.3181 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
Failed to create a ComputeDAG for auto_scheduler: Traceback (most recent call last):
4: TVMFuncCall
3: tvm::runtime::PackedFuncObj::Extractor<tvm::runtime::PackedFuncSubObj<tvm::runtime::TypedPackedFunc<tvm::auto_scheduler::ComputeDAG (tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)>::AssignTypedLambda<tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}>(tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >)::{lambda(tvm::runtime::TVMArgs const&, tvm::runtime::TVMRetValue*)#1}> >::Call(tvm::runtime::PackedFuncObj const*, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, tvm::runtime::TVMRetValue)
2: tvm::auto_scheduler::ComputeDAG::ComputeDAG(tvm::runtime::Array<tvm::te::Tensor, void>)
1: tvm::auto_scheduler::AccessAnalyzer::AccessAnalyzer(tvm::runtime::Array<tvm::te::Tensor, void> const&)
0: tvm::auto_scheduler::TopoSortOps(tvm::runtime::Array<tvm::te::Tensor, void> const&)
File "/home/acer/tvm/src/auto_scheduler/compute_dag.cc", line 97
TVMError: Unsupported op extern(y, 0x7f426cd6a760)
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.065536 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 0.103168 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.14304 ms, Memory: 5923967
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
Failed to create a ComputeDAG for auto_scheduler: Traceback (most recent call last):
4: TVMFuncCall
3: tvm::runtime::PackedFuncObj::Extractor<tvm::runtime::PackedFuncSubObj<tvm::runtime::TypedPackedFunc<tvm::auto_scheduler::ComputeDAG (tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)>::AssignTypedLambda<tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}>(tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >)::{lambda(tvm::runtime::TVMArgs const&, tvm::runtime::TVMRetValue*)#1}> >::Call(tvm::runtime::PackedFuncObj const*, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, tvm::runtime::TVMRetValue)
2: tvm::auto_scheduler::ComputeDAG::ComputeDAG(tvm::runtime::Array<tvm::te::Tensor, void>)
1: tvm::auto_scheduler::AccessAnalyzer::AccessAnalyzer(tvm::runtime::Array<tvm::te::Tensor, void> const&)
0: tvm::auto_scheduler::TopoSortOps(tvm::runtime::Array<tvm::te::Tensor, void> const&)
File "/home/acer/tvm/src/auto_scheduler/compute_dag.cc", line 97
TVMError: Unsupported op extern(y, 0x7f423567ca10)
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.16112 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.622048 ms, Memory: 294928
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 38.1946 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
Failed to create a ComputeDAG for auto_scheduler: Traceback (most recent call last):
4: TVMFuncCall
3: tvm::runtime::PackedFuncObj::Extractor<tvm::runtime::PackedFuncSubObj<tvm::runtime::TypedPackedFunc<tvm::auto_scheduler::ComputeDAG (tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)>::AssignTypedLambda<tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}>(tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >)::{lambda(tvm::runtime::TVMArgs const&, tvm::runtime::TVMRetValue*)#1}> >::Call(tvm::runtime::PackedFuncObj const*, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, tvm::runtime::TVMRetValue)
2: tvm::auto_scheduler::ComputeDAG::ComputeDAG(tvm::runtime::Array<tvm::te::Tensor, void>)
1: tvm::auto_scheduler::AccessAnalyzer::AccessAnalyzer(tvm::runtime::Array<tvm::te::Tensor, void> const&)
0: tvm::auto_scheduler::TopoSortOps(tvm::runtime::Array<tvm::te::Tensor, void> const&)
File "/home/acer/tvm/src/auto_scheduler/compute_dag.cc", line 97
TVMError: Unsupported op extern(y, 0x7f42356bb410)
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.266688 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.6992 ms, Memory: 294928
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 38.1591 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
Failed to create a ComputeDAG for auto_scheduler: Traceback (most recent call last):
4: TVMFuncCall
3: tvm::runtime::PackedFuncObj::Extractor<tvm::runtime::PackedFuncSubObj<tvm::runtime::TypedPackedFunc<tvm::auto_scheduler::ComputeDAG (tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)>::AssignTypedLambda<tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}>(tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >)::{lambda(tvm::runtime::TVMArgs const&, tvm::runtime::TVMRetValue*)#1}> >::Call(tvm::runtime::PackedFuncObj const*, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, tvm::runtime::TVMRetValue)
2: tvm::auto_scheduler::ComputeDAG::ComputeDAG(tvm::runtime::Array<tvm::te::Tensor, void>)
1: tvm::auto_scheduler::AccessAnalyzer::AccessAnalyzer(tvm::runtime::Array<tvm::te::Tensor, void> const&)
0: tvm::auto_scheduler::TopoSortOps(tvm::runtime::Array<tvm::te::Tensor, void> const&)
File "/home/acer/tvm/src/auto_scheduler/compute_dag.cc", line 97
TVMError: Unsupported op extern(y, 0x7f42356a54e0)
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.133312 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.242208 ms, Memory: 3547271
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 19.3645 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
Failed to create a ComputeDAG for auto_scheduler: Traceback (most recent call last):
4: TVMFuncCall
3: tvm::runtime::PackedFuncObj::Extractor<tvm::runtime::PackedFuncSubObj<tvm::runtime::TypedPackedFunc<tvm::auto_scheduler::ComputeDAG (tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)>::AssignTypedLambda<tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}>(tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >)::{lambda(tvm::runtime::TVMArgs const&, tvm::runtime::TVMRetValue*)#1}> >::Call(tvm::runtime::PackedFuncObj const*, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, tvm::runtime::TVMRetValue)
2: tvm::auto_scheduler::ComputeDAG::ComputeDAG(tvm::runtime::Array<tvm::te::Tensor, void>)
1: tvm::auto_scheduler::AccessAnalyzer::AccessAnalyzer(tvm::runtime::Array<tvm::te::Tensor, void> const&)
0: tvm::auto_scheduler::TopoSortOps(tvm::runtime::Array<tvm::te::Tensor, void> const&)
File "/home/acer/tvm/src/auto_scheduler/compute_dag.cc", line 97
TVMError: Unsupported op extern(y, 0x7f42356d4830)
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.066912 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 0.074816 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.1264 ms, Memory: 2957447
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
Failed to create a ComputeDAG for auto_scheduler: Traceback (most recent call last):
4: TVMFuncCall
3: tvm::runtime::PackedFuncObj::Extractor<tvm::runtime::PackedFuncSubObj<tvm::runtime::TypedPackedFunc<tvm::auto_scheduler::ComputeDAG (tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)>::AssignTypedLambda<tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}>(tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >)::{lambda(tvm::runtime::TVMArgs const&, tvm::runtime::TVMRetValue*)#1}> >::Call(tvm::runtime::PackedFuncObj const*, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, tvm::runtime::TVMRetValue)
2: tvm::auto_scheduler::ComputeDAG::ComputeDAG(tvm::runtime::Array<tvm::te::Tensor, void>)
1: tvm::auto_scheduler::AccessAnalyzer::AccessAnalyzer(tvm::runtime::Array<tvm::te::Tensor, void> const&)
0: tvm::auto_scheduler::TopoSortOps(tvm::runtime::Array<tvm::te::Tensor, void> const&)
File "/home/acer/tvm/src/auto_scheduler/compute_dag.cc", line 97
TVMError: Unsupported op extern(y, 0x7f42356d5680)
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.155168 ms, Memory: 6912
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.358144 ms, Memory: 3154055
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 38.9448 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
Failed to create a ComputeDAG for auto_scheduler: Traceback (most recent call last):
4: TVMFuncCall
3: tvm::runtime::PackedFuncObj::Extractor<tvm::runtime::PackedFuncSubObj<tvm::runtime::TypedPackedFunc<tvm::auto_scheduler::ComputeDAG (tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)>::AssignTypedLambda<tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}>(tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >)::{lambda(tvm::runtime::TVMArgs const&, tvm::runtime::TVMRetValue*)#1}> >::Call(tvm::runtime::PackedFuncObj const*, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, tvm::runtime::TVMRetValue)
2: tvm::auto_scheduler::ComputeDAG::ComputeDAG(tvm::runtime::Array<tvm::te::Tensor, void>)
1: tvm::auto_scheduler::AccessAnalyzer::AccessAnalyzer(tvm::runtime::Array<tvm::te::Tensor, void> const&)
0: tvm::auto_scheduler::TopoSortOps(tvm::runtime::Array<tvm::te::Tensor, void> const&)
File "/home/acer/tvm/src/auto_scheduler/compute_dag.cc", line 97
TVMError: Unsupported op extern(y, 0x7f42356cb960)
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.19456 ms, Memory: 6912
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.421184 ms, Memory: 3154055
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 38.9294 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
Failed to create a ComputeDAG for auto_scheduler: Traceback (most recent call last):
4: TVMFuncCall
3: tvm::runtime::PackedFuncObj::Extractor<tvm::runtime::PackedFuncSubObj<tvm::runtime::TypedPackedFunc<tvm::auto_scheduler::ComputeDAG (tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)>::AssignTypedLambda<tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}>(tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >)::{lambda(tvm::runtime::TVMArgs const&, tvm::runtime::TVMRetValue*)#1}> >::Call(tvm::runtime::PackedFuncObj const*, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, tvm::runtime::TVMRetValue)
2: tvm::auto_scheduler::ComputeDAG::ComputeDAG(tvm::runtime::Array<tvm::te::Tensor, void>)
1: tvm::auto_scheduler::AccessAnalyzer::AccessAnalyzer(tvm::runtime::Array<tvm::te::Tensor, void> const&)
0: tvm::auto_scheduler::TopoSortOps(tvm::runtime::Array<tvm::te::Tensor, void> const&)
File "/home/acer/tvm/src/auto_scheduler/compute_dag.cc", line 97
TVMError: Unsupported op extern(y, 0x7f42356ead50)
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.151552 ms, Memory: 3584
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.362816 ms, Memory: 3837775
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 18.8822 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
Failed to create a ComputeDAG for auto_scheduler: Traceback (most recent call last):
4: TVMFuncCall
3: tvm::runtime::PackedFuncObj::Extractor<tvm::runtime::PackedFuncSubObj<tvm::runtime::TypedPackedFunc<tvm::auto_scheduler::ComputeDAG (tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)>::AssignTypedLambda<tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}>(tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >)::{lambda(tvm::runtime::TVMArgs const&, tvm::runtime::TVMRetValue*)#1}> >::Call(tvm::runtime::PackedFuncObj const*, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, tvm::runtime::TVMRetValue)
2: tvm::auto_scheduler::ComputeDAG::ComputeDAG(tvm::runtime::Array<tvm::te::Tensor, void>)
1: tvm::auto_scheduler::AccessAnalyzer::AccessAnalyzer(tvm::runtime::Array<tvm::te::Tensor, void> const&)
0: tvm::auto_scheduler::TopoSortOps(tvm::runtime::Array<tvm::te::Tensor, void> const&)
File "/home/acer/tvm/src/auto_scheduler/compute_dag.cc", line 97
TVMError: Unsupported op extern(y, 0x7f42356bcaf0)
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.069792 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 0.076928 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.127328 ms, Memory: 1478479
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:23] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
Failed to create a ComputeDAG for auto_scheduler: Traceback (most recent call last):
4: TVMFuncCall
3: tvm::runtime::PackedFuncObj::Extractor<tvm::runtime::PackedFuncSubObj<tvm::runtime::TypedPackedFunc<tvm::auto_scheduler::ComputeDAG (tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)>::AssignTypedLambda<tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}>(tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >)::{lambda(tvm::runtime::TVMArgs const&, tvm::runtime::TVMRetValue*)#1}> >::Call(tvm::runtime::PackedFuncObj const*, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, tvm::runtime::TVMRetValue)
2: tvm::auto_scheduler::ComputeDAG::ComputeDAG(tvm::runtime::Array<tvm::te::Tensor, void>)
1: tvm::auto_scheduler::AccessAnalyzer::AccessAnalyzer(tvm::runtime::Array<tvm::te::Tensor, void> const&)
0: tvm::auto_scheduler::TopoSortOps(tvm::runtime::Array<tvm::te::Tensor, void> const&)
File "/home/acer/tvm/src/auto_scheduler/compute_dag.cc", line 97
TVMError: Unsupported op extern(y, 0x7f42356f07d0)
[20:42:24] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:24] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.153632 ms, Memory: 524575
[20:42:24] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.577312 ms, Memory: 5705551
[20:42:24] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 35.9759 ms, Memory: 0
[20:42:24] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:24] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:24] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:24] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:24] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
Failed to create a ComputeDAG for auto_scheduler: Traceback (most recent call last):
4: TVMFuncCall
3: tvm::runtime::PackedFuncObj::Extractor<tvm::runtime::PackedFuncSubObj<tvm::runtime::TypedPackedFunc<tvm::auto_scheduler::ComputeDAG (tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)>::AssignTypedLambda<tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}>(tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >)::{lambda(tvm::runtime::TVMArgs const&, tvm::runtime::TVMRetValue*)#1}> >::Call(tvm::runtime::PackedFuncObj const*, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, tvm::runtime::TVMRetValue)
2: tvm::auto_scheduler::ComputeDAG::ComputeDAG(tvm::runtime::Array<tvm::te::Tensor, void>)
1: tvm::auto_scheduler::AccessAnalyzer::AccessAnalyzer(tvm::runtime::Array<tvm::te::Tensor, void> const&)
0: tvm::auto_scheduler::TopoSortOps(tvm::runtime::Array<tvm::te::Tensor, void> const&)
File "/home/acer/tvm/src/auto_scheduler/compute_dag.cc", line 97
TVMError: Unsupported op extern(y, 0x7f42356fba40)
[20:42:24] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:24] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.201568 ms, Memory: 524575
[20:42:24] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.612256 ms, Memory: 5705551
[20:42:24] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 35.7139 ms, Memory: 0
[20:42:24] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:24] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:24] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:24] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:24] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
Failed to create a ComputeDAG for auto_scheduler: Traceback (most recent call last):
4: TVMFuncCall
3: tvm::runtime::PackedFuncObj::Extractor<tvm::runtime::PackedFuncSubObj<tvm::runtime::TypedPackedFunc<tvm::auto_scheduler::ComputeDAG (tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)>::AssignTypedLambda<tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}>(tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >)::{lambda(tvm::runtime::TVMArgs const&, tvm::runtime::TVMRetValue*)#1}> >::Call(tvm::runtime::PackedFuncObj const*, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, tvm::runtime::TVMRetValue)
2: tvm::auto_scheduler::ComputeDAG::ComputeDAG(tvm::runtime::Array<tvm::te::Tensor, void>)
1: tvm::auto_scheduler::AccessAnalyzer::AccessAnalyzer(tvm::runtime::Array<tvm::te::Tensor, void> const&)
0: tvm::auto_scheduler::TopoSortOps(tvm::runtime::Array<tvm::te::Tensor, void> const&)
File "/home/acer/tvm/src/auto_scheduler/compute_dag.cc", line 97
TVMError: Unsupported op extern(y, 0x7f423570c9d0)
[20:42:24] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:135: CUDNN Found 8 fwd algorithms, choosing CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM
[20:42:24] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 0) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.262144 ms, Memory: 524575
[20:42:24] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 1) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_PRECOMP_GEMM - time: 0.690176 ms, Memory: 5705551
[20:42:24] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 2) CUDNN_CONVOLUTION_FWD_ALGO_IMPLICIT_GEMM - time: 36.057 ms, Memory: 0
[20:42:24] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 3) CUDNN_CONVOLUTION_FWD_ALGO_GEMM - time: -1 ms, Memory: 0
[20:42:24] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 4) CUDNN_CONVOLUTION_FWD_ALGO_DIRECT - time: -1 ms, Memory: 0
[20:42:24] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 5) CUDNN_CONVOLUTION_FWD_ALGO_FFT - time: -1 ms, Memory: 0
[20:42:24] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 6) CUDNN_CONVOLUTION_FWD_ALGO_FFT_TILING - time: -1 ms, Memory: 0
[20:42:24] /home/acer/tvm/src/runtime/contrib/cudnn/conv_forward.cc:138: 7) CUDNN_CONVOLUTION_FWD_ALGO_WINOGRAD - time: -1 ms, Memory: 0
Failed to create a ComputeDAG for auto_scheduler: Traceback (most recent call last):
4: TVMFuncCall
3: tvm::runtime::PackedFuncObj::Extractor<tvm::runtime::PackedFuncSubObj<tvm::runtime::TypedPackedFunc<tvm::auto_scheduler::ComputeDAG (tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)>::AssignTypedLambda<tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}>(tvm::auto_scheduler::__mk_TVM4::{lambda(tvm::runtime::Optional<tvm::runtime::Array<tvm::te::Tensor, void> >, tvm::runtime::Optional<tvm::te::Schedule>)#1}, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >)::{lambda(tvm::runtime::TVMArgs const&, tvm::runtime::TVMRetValue*)#1}> >::Call(tvm::runtime::PackedFuncObj const*, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, tvm::runtime::TVMRetValue)
2: tvm::auto_scheduler::ComputeDAG::ComputeDAG(tvm::runtime::Array<tvm::te::Tensor, void>)
1: tvm::auto_scheduler::AccessAnalyzer::AccessAnalyzer(tvm::runtime::Array<tvm::te::Tensor, void> const&)
0: tvm::auto_scheduler::TopoSortOps(tvm::runtime::Array<tvm::te::Tensor, void> const&)
File "/home/acer/tvm/src/auto_scheduler/compute_dag.cc", line 97
TVMError: Unsupported op extern(y, 0x7f423571c0d0)
----------------------------------------------------------------------
------------------------------ [ Task Scheduler ]
----------------------------------------------------------------------
----------------------------------------------------------------------
------------------------------ [ Search ]
----------------------------------------------------------------------
Generate Sketches #s: 1
Sample Iter: 5 #Pop: 5 #Target: 50 fail_ct: 10235 Time elapsed: 9.17
#Target has been reduced to 25 due to too many failures or duplications
Sample Iter: 10 #Pop: 5 #Target: 25 fail_ct: 20475 Time elapsed: 18.39
#Target has been reduced to 12 due to too many failures or duplications
Sample Iter: 15 #Pop: 5 #Target: 12 fail_ct: 30715 Time elapsed: 27.65
Exception ignored in: <module 'threading' from '/home/acer/.pyenv/versions/3.8.17/lib/python3.8/threading.py'>
Traceback (most recent call last):
File "/home/acer/.pyenv/versions/3.8.17/lib/python3.8/threading.py", line 1388, in _shutdown
lock.acquire()
KeyboardInterrupt:
Exception ignored in: <module 'threading' from '/home/acer/.pyenv/versions/3.8.17/lib/python3.8/threading.py'>
Traceback (most recent call last):
File "/home/acer/.pyenv/versions/3.8.17/lib/python3.8/threading.py", line 1388, in _shutdown
lock.acquire()
KeyboardInterrupt:
Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 208, in tvm._ffi._cy3.core.make_ret
File "tvm/_ffi/_cython/./object.pxi", line 46, in tvm._ffi._cy3.core.make_ret_object
File "tvm/_ffi/_cython/./base.pxi", line 187, in tvm._ffi._cy3.core.ctypes_handle
File "/home/acer/.pyenv/versions/3.8.17/lib/python3.8/ctypes/__init__.py", line 506, in cast
def cast(obj, typ):
KeyboardInterrupt
Exception ignored in: 'tvm._ffi._cy3.core.tvm_callback'
Traceback (most recent call last):
File "tvm/_ffi/_cython/./packed_func.pxi", line 208, in tvm._ffi._cy3.core.make_ret
File "tvm/_ffi/_cython/./object.pxi", line 46, in tvm._ffi._cy3.core.make_ret_object
File "tvm/_ffi/_cython/./base.pxi", line 187, in tvm._ffi._cy3.core.ctypes_handle
File "/home/acer/.pyenv/versions/3.8.17/lib/python3.8/ctypes/__init__.py", line 506, in cast
def cast(obj, typ):
KeyboardInterrupt:
#Target has been reduced to 6 due to too many failures or duplications
Sample Iter: 20 #Pop: 5 #Target: 6 fail_ct: 40955 Time elapsed: 35.55
#Target has been reduced to 3 due to too many failures or duplications
Sample Initial Population #s: 5 fail_ct: 43003 Time elapsed: 37.41
GA Iter: 0 Max score: 0.6982 Min score: 0.3983 #Pop: 5 #M+: 0 #M-: 0
import numpy as np
import os
import tvm
from tvm import relay, auto_scheduler
import tvm.relay.testing
from tvm.contrib import graph_executor
import onnx
import logging
from tvm.relay.transform import mixed_precision
from tvm.autotvm.measure.measure_methods import set_cuda_target_arch
from tvm.contrib import rpc
class OnnxWrapper(object):
def __init__(self, session):
super(OnnxWrapper, self).__init__()
# import onnxruntime as ort
self.session = session
self.input_name = session.get_inputs()[0].name
self.output_names = []
for output in session.get_outputs():
self.output_names.append(output.name)
try:
seqs = [int(output_name) for output_name in self.output_names]
sorted_seq_inds = sorted(range(len(seqs)), key=lambda k: seqs[k])
self.output_names = [self.output_names[ind] for ind in sorted_seq_inds]
except Exception as e:
warnings.warn(str(e))
def forward(self, x, **kwargs):
outputs = self.session.run(self.output_names, {self.input_name: x})
return outputs
import sys
logging.getLogger("auto_scheduler").setLevel(logging.DEBUG)
logging.getLogger("auto_scheduler").addHandler(logging.StreamHandler(sys.stdout))
def get_network(weight, batch_size, layout="NHWC", dtype="float32", use_sparse=False):
"""Get the symbol definition and random weight of a network"""
input_shape = (batch_size, 3, 384, 640)
onnx_model = onnx.load(weight)
input_name = "input"
shape_dict = {input_name: input_shape}
mod, params = relay.frontend.from_onnx(onnx_model, shape_dict)
# desired_layouts = {'nn.conv2d': ['NHWC', 'default'], 'image.resize2d': ['NHWC'], 'nn.upsampling': ['NHWC']}
desired_layouts = {'nn.conv2d': ['NHWC', 'OHWI'], 'image.resize2d': ['NHWC'], 'nn.upsampling': ['NHWC']}
seq = tvm.transform.Sequential([relay.transform.RemoveUnusedFunctions(), relay.transform.ConvertLayout(desired_layouts)])
# seq = tvm.transform.Sequential([relay.transform.RemoveUnusedFunctions()])
with tvm.transform.PassContext(opt_level=3):
mod = seq(mod)
mod = tvm.IRModule.from_expr(mod["main"])
mod = tvm.relay.transform.FastMath()(mod)
mod = tvm.relay.transform.EliminateCommonSubexpr()(mod)
BindPass = tvm.relay.transform.function_pass(lambda fn, new_mod, ctx: tvm.relay.build_module.bind_params_by_name(fn, params), opt_level=1)
mod = BindPass(mod)
mod = tvm.relay.transform.FoldConstant()(mod)
mod = tvm.relay.transform.CombineParallelBatchMatmul()(mod)
mod = tvm.relay.transform.FoldConstant()(mod)
mod = tvm.relay.transform.InferType()(mod)
mod = tvm.relay.transform.ToMixedPrecision()(mod)
mod = tvm.relay.transform.EliminateCommonSubexpr()(mod)
mod = tvm.relay.transform.FoldConstant()(mod)
mod = tvm.relay.transform.CombineParallelBatchMatmul()(mod)
mod = tvm.relay.transform.FoldConstant()(mod)
return mod, params, input_shape
# weight = '/home/acer/nfs-share/autoslim_repvgg_a0_gfl_dsl_v2_320_320_epoch_150_calibrate_0.5.onnx'
# weight = '/home/acer/rtmdet_m_syncbn_fast_8xb32-300e_coco_640_640_best_coco_crocs_precision_epoch_218.onnx'
# weight = '/home/acer/group_fisher_act_deploy_retinanet_r50_fpn_1x_coco_random_384_640_random.onnx'
weight = '/home/acer/tvm_experiment/resnet18.onnx'
network = os.path.basename(weight).replace('.onnx', '')
use_sparse = False
batch_size = 2
layout = "NHWC"
# target = tvm.target.Target("llvm -mtriple=aarch64-linux-gnu cuda -arch=sm_53")
# target = tvm.target.Target("cuda -arch=sm_75", host='llvm -mtriple=aarch64-linux-gnu')
# https://gt3rs.medium.com/compile-with-nvcc-3566fbdfdbf
# target = tvm.target.Target("cuda -libs=cudnn -arch=sm_53", host='llvm -mtriple=aarch64-linux-gnu')
# set_cuda_target_arch('sm_53')
target = tvm.target.Target('cuda -arch=sm_75 -libs=cudnn')
# target = tvm.target.Target("nvidia/rtx-3000")
dtype = "float16"
# log_file = "%s-%s-B%d-%s.json" % (network, layout, batch_size, target.kind.name)
log_file = "rtmdet_m_syncbn_fast_8xb32-300e_coco_640_640_best_coco_crocs_precision_epoch_218_rtx2070.json"
#################################################################
# Extract Search Tasks
# --------------------
# Next, we extract the search tasks and their weights from a network.
# The weight of a task is the number of appearances of the task's subgraph
# in the whole network.
# By using the weight, we can approximate the end-to-end latency of the network
# as :code:`sum(latency[t] * weight[t])`, where :code:`latency[t]` is the
# latency of a task and :code:`weight[t]` is the weight of the task.
# The task scheduler will just optimize this objective.
# Extract tasks from the network
print("Get model...")
#runner = auto_scheduler.RPCRunner('nx', '127.0.0.1', 9190, timeout=1000, n_parallel=1),
#assert(0)
mod, params, input_shape = get_network(
weight,
batch_size,
layout,
dtype=dtype,
use_sparse=use_sparse,
)
print("Extract tasks...")
tasks, task_weights = auto_scheduler.extract_tasks(mod["main"], params, target)
for idx, task in enumerate(tasks):
print("========== Task %d (workload key: %s) ==========" % (idx, task.workload_key))
# print(task.compute_dag)
print(task.hardware_params)
#################################################################
# Begin Tuning
# ------------
# Now, we set some options for tuning and launch the search tasks
#
# * :code:`num_measure_trials` is the number of measurement trials we can use during the tuning.
# You can set it to a small number (e.g., 200) for a fast demonstrative run.
# In practice, we recommend setting it around :code:`800 * len(tasks)`,
# which is typically enough for the search to converge.
# For example, there are 29 tasks in resnet-50, so we can set it as 20000.
# You can adjust this parameter according to your time budget.
# * In addition, we use :code:`RecordToFile` to dump measurement records into a log file,
# The measurement records can be used to query the history best, resume the search,
# and do more analyses later.
# * see :any:`auto_scheduler.TuningOptions`,
# :any:`auto_scheduler.LocalRunner` for more parameters.
#
def run_tuning():
print("Begin tuning...")
measure_ctx = auto_scheduler.LocalRPCMeasureContext(repeat=1, min_repeat_ms=300, timeout=10)
tuner = auto_scheduler.TaskScheduler(tasks, task_weights)
tune_option = auto_scheduler.TuningOptions(
num_measure_trials=35600, # change this to 20000 to achieve the best performance
builder = auto_scheduler.LocalBuilder(timeout=1000),
runner=measure_ctx.runner,
# runner = auto_scheduler.RPCRunner('nano', '127.0.0.1', 9190, timeout=1000, n_parallel=1),
measure_callbacks=[auto_scheduler.RecordToFile(log_file)],
)
tuner.tune(tune_option)
# We do not run the tuning in our webpage server since it takes too long.
# Uncomment the following line to run it by yourself.
run_tuning()
#################################################################
# Compile and Evaluate
# --------------------
# After auto-tuning, we can compile the network with the best schedules we found.
# All measurement records are dumped into the log file during auto-tuning,
# so we can read the log file and load the best schedules.
# Compile with the history best
print("Compile...")
with auto_scheduler.ApplyHistoryBest(log_file):
with tvm.transform.PassContext(opt_level=3, config={"relay.backend.use_auto_scheduler": True}):
lib = relay.build(mod, target=target, params=params)
lib.export_library('/home/acer/rtmdet_m_syncbn_fast_8xb32-300e_coco_640_640_best_coco_crocs_precision_epoch_218_fp16_rtx2070.tar')
# Create graph executor
# dev = tvm.device(str(target), 0)
# module = graph_executor.GraphModule(lib["default"](dev))
# data_tvm = tvm.nd.array((np.random.uniform(size=input_shape)).astype(dtype))
# module.set_input("data", data_tvm)
# # Evaluate
# print("Evaluate inference time cost...")
# print(module.benchmark(dev, repeat=3, min_repeat_ms=500))
#################################################################
# Other Tips
# ----------
# 1. During the tuning, the auto-scheduler needs to compile many programs and
# extract feature from them. This part is CPU-intensive,
# so a high-performance CPU with many cores is recommended for faster search.
# 2. You can use :code:`python3 -m tvm.auto_scheduler.measure_record --mode distill -i log.json`
# to distill the large log file and only save the best useful records.
# 3. You can resume a search from the previous log file. You just need to
# add a new argument :code:`load_log_file` when creating the task scheduler
# in function :code:`run_tuning`. Say,
# :code:`tuner = auto_scheduler.TaskScheduler(tasks, task_weights, load_log_file=log_file)`
# 4. If you have multiple target CPUs, you can use all of them for measurements to
# parallelize the measurements. Check this :ref:`section <tutorials-autotvm-scale-up-rpc-tracker>`
# to learn how to use the RPC Tracker and RPC Server.
# To use the RPC Tracker in auto-scheduler, replace the runner in :code:`TuningOptions`
# with :any:`auto_scheduler.RPCRunner`.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment