You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@tvm.apache.org by tq...@apache.org on 2023/03/30 16:02:45 UTC

[tvm-site] branch asf-site updated: deploying docs (apache/tvm@8e2382eea52bcead6e6b4835fb0e9381e38d76cd)

This is an automated email from the ASF dual-hosted git repository.

tqchen pushed a commit to branch asf-site
in repository https://gitbox.apache.org/repos/asf/tvm-site.git


The following commit(s) were added to refs/heads/asf-site by this push:
     new 2717c55755 deploying docs (apache/tvm@8e2382eea52bcead6e6b4835fb0e9381e38d76cd)
2717c55755 is described below

commit 2717c557557bc23f6c118af1dc9502406fa88db4
Author: tvm-bot <95...@users.noreply.github.com>
AuthorDate: Thu Mar 30 16:02:38 2023 +0000

    deploying docs (apache/tvm@8e2382eea52bcead6e6b4835fb0e9381e38d76cd)
---
 docs/_images/sphx_glr_micro_train_001.png          |  Bin 312563 -> 348666 bytes
 docs/_images/sphx_glr_micro_train_thumb.png        |  Bin 22858 -> 24386 bytes
 .../how_to/compile_models/from_darknet.rst.txt     |    2 +-
 .../how_to/compile_models/from_mxnet.rst.txt       |    2 +-
 .../how_to/compile_models/from_oneflow.rst.txt     |    2 +-
 .../how_to/compile_models/from_pytorch.rst.txt     |    2 +-
 .../how_to/compile_models/from_tensorflow.rst.txt  |    4 +-
 .../compile_models/sg_execution_times.rst.txt      |   22 +-
 .../deploy_models/deploy_model_on_adreno.rst.txt   |    7 +-
 .../deploy_model_on_adreno_tvmc.rst.txt            |    2 +-
 .../deploy_models/deploy_model_on_android.rst.txt  |    2 +-
 .../deploy_object_detection_pytorch.rst.txt        |    4 +-
 .../deploy_models/deploy_prequantized.rst.txt      |    6 +-
 .../deploy_prequantized_tflite.rst.txt             |    4 +-
 .../how_to/deploy_models/deploy_quantized.rst.txt  |    2 +-
 .../deploy_models/deploy_ssd_gluoncv.rst.txt       |    4 +-
 .../deploy_models/sg_execution_times.rst.txt       |   22 +-
 .../extend_tvm/bring_your_own_datatypes.rst.txt    |    2 +-
 .../how_to/extend_tvm/sg_execution_times.rst.txt   |    8 +-
 .../how_to/extend_tvm/use_pass_instrument.rst.txt  |   16 +-
 .../optimize_operators/opt_conv_cuda.rst.txt       |    2 +-
 .../optimize_operators/opt_conv_tensorcore.rst.txt |    2 +-
 .../how_to/optimize_operators/opt_gemm.rst.txt     |   16 +-
 .../optimize_operators/sg_execution_times.rst.txt  |    8 +-
 .../sg_execution_times.rst.txt                     |   14 +-
 .../tune_conv2d_layer_cuda.rst.txt                 | 1159 +++++++++++++++++---
 .../tune_network_cuda.rst.txt                      |    4 +-
 .../tune_network_x86.rst.txt                       |    4 +-
 .../tune_sparse_x86.rst.txt                        |  125 ++-
 .../tune_with_autotvm/sg_execution_times.rst.txt   |    8 +-
 .../tune_with_autotvm/tune_conv2d_cuda.rst.txt     |  483 ++++++--
 .../work_with_microtvm/micro_autotune.rst.txt      |   18 +-
 .../work_with_microtvm/micro_pytorch.rst.txt       |    4 +-
 .../how_to/work_with_microtvm/micro_train.rst.txt  |   18 +-
 .../work_with_microtvm/sg_execution_times.rst.txt  |   14 +-
 .../work_with_relay/sg_execution_times.rst.txt     |    8 +-
 .../how_to/work_with_schedules/intrin_math.rst.txt |    2 +-
 .../work_with_schedules/sg_execution_times.rst.txt |   18 +-
 .../tutorials/autotvm/sg_execution_times.rst.txt   |    4 +-
 .../frontend/deploy_classification.rst.txt         |    2 +-
 .../tutorials/frontend/deploy_detection.rst.txt    |    2 +-
 .../tutorials/frontend/sg_execution_times.rst.txt  |    6 +-
 .../tutorials/optimize/sg_execution_times.rst.txt  |    6 +-
 .../topic/vta/tutorials/sg_execution_times.rst.txt |    6 +-
 .../tutorial/auto_scheduler_matmul_x86.rst.txt     |    4 +-
 docs/_sources/tutorial/autotvm_matmul_x86.rst.txt  |   20 +-
 docs/_sources/tutorial/autotvm_relay_x86.rst.txt   |   56 +-
 .../tutorial/cross_compilation_and_rpc.rst.txt     |    2 +-
 docs/_sources/tutorial/intro_topi.rst.txt          |    2 +-
 docs/_sources/tutorial/sg_execution_times.rst.txt  |   18 +-
 .../tutorial/tensor_expr_get_started.rst.txt       |   42 +-
 docs/commit_hash                                   |    2 +-
 docs/how_to/compile_models/from_darknet.html       |    2 +-
 docs/how_to/compile_models/from_mxnet.html         |    2 +-
 docs/how_to/compile_models/from_oneflow.html       |   12 +-
 docs/how_to/compile_models/from_pytorch.html       |   12 +-
 docs/how_to/compile_models/from_tensorflow.html    |    4 +-
 docs/how_to/compile_models/sg_execution_times.html |   22 +-
 .../deploy_models/deploy_model_on_adreno.html      |    3 +-
 .../deploy_models/deploy_model_on_adreno_tvmc.html |   39 +-
 .../deploy_models/deploy_model_on_android.html     |    2 +-
 .../deploy_object_detection_pytorch.html           |   41 +-
 docs/how_to/deploy_models/deploy_prequantized.html |    8 +-
 .../deploy_models/deploy_prequantized_tflite.html  |    4 +-
 docs/how_to/deploy_models/deploy_quantized.html    |    2 +-
 docs/how_to/deploy_models/deploy_ssd_gluoncv.html  |   40 +-
 docs/how_to/deploy_models/sg_execution_times.html  |   22 +-
 .../extend_tvm/bring_your_own_datatypes.html       |    2 +-
 docs/how_to/extend_tvm/sg_execution_times.html     |    8 +-
 docs/how_to/extend_tvm/use_pass_instrument.html    |   16 +-
 docs/how_to/optimize_operators/opt_conv_cuda.html  |    2 +-
 .../optimize_operators/opt_conv_tensorcore.html    |    2 +-
 docs/how_to/optimize_operators/opt_gemm.html       |   16 +-
 .../optimize_operators/sg_execution_times.html     |    8 +-
 .../sg_execution_times.html                        |   14 +-
 .../tune_conv2d_layer_cuda.html                    | 1159 +++++++++++++++++---
 .../tune_with_autoscheduler/tune_network_cuda.html |    4 +-
 .../tune_with_autoscheduler/tune_network_x86.html  |    4 +-
 .../tune_with_autoscheduler/tune_sparse_x86.html   |  125 ++-
 .../tune_with_autotvm/sg_execution_times.html      |    8 +-
 .../how_to/tune_with_autotvm/tune_conv2d_cuda.html |  483 ++++++--
 docs/how_to/work_with_microtvm/micro_autotune.html |   18 +-
 docs/how_to/work_with_microtvm/micro_pytorch.html  |    5 +-
 docs/how_to/work_with_microtvm/micro_train.html    |   16 +-
 .../work_with_microtvm/sg_execution_times.html     |   14 +-
 .../how_to/work_with_relay/sg_execution_times.html |    8 +-
 docs/how_to/work_with_schedules/intrin_math.html   |    2 +-
 .../work_with_schedules/sg_execution_times.html    |   18 +-
 docs/install/nnpack.html                           |   12 +-
 .../api/doxygen/relay_2attrs_2nn_8h_source.html    |    2 +-
 docs/reference/api/python/auto_scheduler.html      |    4 +-
 docs/reference/api/python/relay/nn.html            |    2 +-
 .../api/typedoc/classes/bytestreamreader.html      |   12 +-
 .../api/typedoc/classes/cachedcallstack.html       |   34 +-
 docs/reference/api/typedoc/classes/dldatatype.html |   12 +-
 docs/reference/api/typedoc/classes/dldevice.html   |   10 +-
 .../reference/api/typedoc/classes/environment.html |   12 +-
 docs/reference/api/typedoc/classes/ffilibrary.html |   20 +-
 docs/reference/api/typedoc/classes/instance.html   |   58 +-
 docs/reference/api/typedoc/classes/memory.html     |   34 +-
 docs/reference/api/typedoc/classes/module.html     |   10 +-
 docs/reference/api/typedoc/classes/ndarray.html    |   22 +-
 .../api/typedoc/classes/packedfunccell.html        |    6 +-
 docs/reference/api/typedoc/classes/rpcserver.html  |   14 +-
 .../api/typedoc/classes/runtimecontext.html        |   22 +-
 docs/reference/api/typedoc/classes/scalar.html     |    6 +-
 docs/reference/api/typedoc/classes/tvmarray.html   |   16 +-
 docs/reference/api/typedoc/classes/tvmobject.html  |   12 +-
 .../api/typedoc/classes/webgpucontext.html         |   12 +-
 docs/reference/api/typedoc/enums/argtypecode.html  |   30 +-
 .../api/typedoc/enums/aynccallbackcode.html        |    4 +-
 .../api/typedoc/enums/dldatatypecode.html          |    8 +-
 .../api/typedoc/enums/rpcserverstate.html          |   12 +-
 docs/reference/api/typedoc/enums/sizeof.html       |   18 +-
 docs/reference/api/typedoc/index.html              |  124 +--
 .../api/typedoc/interfaces/disposable.html         |    2 +-
 .../api/typedoc/interfaces/functioninfo.html       |    6 +-
 .../api/typedoc/interfaces/libraryprovider.html    |    4 +-
 docs/searchindex.js                                |    2 +-
 .../vta/tutorials/autotvm/sg_execution_times.html  |    4 +-
 .../tutorials/frontend/deploy_classification.html  |    2 +-
 .../vta/tutorials/frontend/deploy_detection.html   |    2 +-
 .../vta/tutorials/frontend/sg_execution_times.html |    6 +-
 .../vta/tutorials/optimize/sg_execution_times.html |    6 +-
 docs/topic/vta/tutorials/sg_execution_times.html   |    6 +-
 docs/tutorial/auto_scheduler_matmul_x86.html       |    4 +-
 docs/tutorial/autotvm_matmul_x86.html              |   20 +-
 docs/tutorial/autotvm_relay_x86.html               |  270 ++---
 docs/tutorial/cross_compilation_and_rpc.html       |    2 +-
 docs/tutorial/intro_topi.html                      |    2 +-
 docs/tutorial/sg_execution_times.html              |   18 +-
 docs/tutorial/tensor_expr_get_started.html         |   42 +-
 132 files changed, 3959 insertions(+), 1306 deletions(-)

diff --git a/docs/_images/sphx_glr_micro_train_001.png b/docs/_images/sphx_glr_micro_train_001.png
index 793f3e36e6..b353dd858a 100644
Binary files a/docs/_images/sphx_glr_micro_train_001.png and b/docs/_images/sphx_glr_micro_train_001.png differ
diff --git a/docs/_images/sphx_glr_micro_train_thumb.png b/docs/_images/sphx_glr_micro_train_thumb.png
index 94f0e4545f..280b5c012d 100644
Binary files a/docs/_images/sphx_glr_micro_train_thumb.png and b/docs/_images/sphx_glr_micro_train_thumb.png differ
diff --git a/docs/_sources/how_to/compile_models/from_darknet.rst.txt b/docs/_sources/how_to/compile_models/from_darknet.rst.txt
index 0c7438306e..5ad0d08f53 100644
--- a/docs/_sources/how_to/compile_models/from_darknet.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_darknet.rst.txt
@@ -318,7 +318,7 @@ The process is no different from other examples.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  18.482 seconds)
+   **Total running time of the script:** ( 1 minutes  24.383 seconds)
 
 
 .. _sphx_glr_download_how_to_compile_models_from_darknet.py:
diff --git a/docs/_sources/how_to/compile_models/from_mxnet.rst.txt b/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
index e097cf187a..950d3e119f 100644
--- a/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
@@ -116,7 +116,7 @@ In this section, we download a pretrained imagenet model and classify an image.
 
  .. code-block:: none
 
-    Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zipbca01480-a548-4f52-8ffa-ec5dd7621c6a from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
+    Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip4a39d133-01d1-457a-b748-cc514c61ab7e from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
     x (1, 3, 224, 224)
 
 
diff --git a/docs/_sources/how_to/compile_models/from_oneflow.rst.txt b/docs/_sources/how_to/compile_models/from_oneflow.rst.txt
index 12855b3a87..7cb58473f7 100644
--- a/docs/_sources/how_to/compile_models/from_oneflow.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_oneflow.rst.txt
@@ -121,7 +121,7 @@ Load a pretrained OneFlow model and save model
  .. code-block:: none
 
     Downloading: "https://oneflow-public.oss-cn-beijing.aliyuncs.com/model_zoo/flowvision/classification/ResNet/resnet18.zip" to /workspace/.oneflow/flowvision_cache/resnet18.zip
-
      0%|          | 0.00/41.5M [00:00<?, ?B/s]
      5%|5         | 2.09M/41.5M [00:00<00:02, 15.7MB/s]
     39%|###8      | 16.0M/41.5M [00:00<00:00, 71.4MB/s]
     55%|#####5    | 23.0M/41.5M [00:00<00:00, 68.7MB/s]
     77%|#######7  | 32.0M/41.5M [00:00<00:00, 76.9MB/s]
    100%|##########| 41.5M/41.5M [00:00<00:00, 77.3MB/s]
+
      0%|          | 0.00/41.5M [00:00<?, ?B/s]
     15%|#5        | 6.33M/41.5M [00:00<00:00, 45.7MB/s]
     26%|##5       | 10.7M/41.5M [00:00<00:00, 36.3MB/s]
     39%|###8      | 16.0M/41.5M [00:00<00:00, 40.1MB/s]
     54%|#####3    | 22.3M/41.5M [00:00<00:00, 47.5MB/s]
     65%|######5   | 27.0M/41.5M [00:00<00:00, 42.0MB/s]
     86%|########6 | 35.8M/41.5M [00:00<00:00, 56.1MB/s]
    100%|##########| 41.5M/41.5M [00:00<00:00, 51.3MB/s]
 
 
 
diff --git a/docs/_sources/how_to/compile_models/from_pytorch.rst.txt b/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
index 391959b065..e65fc7497f 100644
--- a/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
@@ -101,7 +101,7 @@ Load a pretrained PyTorch model
     /venv/apache-tvm-py3.7/lib/python3.7/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and will be removed in 0.15. The current behavior is equivalent to passing `weights=ResNet18_Weights.IMAGENET1K_V1`. You can also use `weights=ResNet18_Weights.DEFAULT` to get the most up-to-date weights.
       warnings.warn(msg)
     Downloading: "https://download.pytorch.org/models/resnet18-f37072fd.pth" to /workspace/.cache/torch/hub/checkpoints/resnet18-f37072fd.pth
-
      0%|          | 0.00/44.7M [00:00<?, ?B/s]
     27%|##6       | 11.9M/44.7M [00:00<00:00, 125MB/s]
     53%|#####3    | 23.9M/44.7M [00:00<00:00, 109MB/s]
     77%|#######7  | 34.5M/44.7M [00:00<00:00, 104MB/s]
     99%|#########9| 44.4M/44.7M [00:00<00:00, 104MB/s]
    100%|##########| 44.7M/44.7M [00:00<00:00, 106MB/s]
+
      0%|          | 0.00/44.7M [00:00<?, ?B/s]
     14%|#4        | 6.30M/44.7M [00:00<00:00, 52.3MB/s]
     25%|##5       | 11.3M/44.7M [00:00<00:00, 44.2MB/s]
     36%|###5      | 16.0M/44.7M [00:00<00:00, 37.7MB/s]
     54%|#####3    | 24.0M/44.7M [00:00<00:00, 49.1MB/s]
     72%|#######1  | 32.0M/44.7M [00:00<00:00, 59.0MB/s]
     99%|#########8| 44.2M/44.7M [00:00<00:00, 72.5MB/s]
    100%|##########| 44.7M/44.7M [00:00<00:00, 60.4MB/s]
 
 
 
diff --git a/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt b/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
index e963bce495..672df00153 100644
--- a/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
@@ -244,7 +244,7 @@ Results:
 
     /workspace/python/tvm/relay/frontend/tensorflow.py:539: UserWarning: Ignore the passed shape. Shape in graphdef will be used for operator DecodeJpeg/contents.
       "will be used for operator %s." % node.name
-    /workspace/python/tvm/relay/frontend/tensorflow_ops.py:1009: UserWarning: DecodeJpeg: It's a pass through, please handle preprocessing before input
+    /workspace/python/tvm/relay/frontend/tensorflow_ops.py:1012: UserWarning: DecodeJpeg: It's a pass through, please handle preprocessing before input
       warnings.warn("DecodeJpeg: It's a pass through, please handle preprocessing before input")
     Tensorflow protobuf imported to relay frontend.
 
@@ -430,7 +430,7 @@ Run the corresponding model on tensorflow
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  31.026 seconds)
+   **Total running time of the script:** ( 1 minutes  34.630 seconds)
 
 
 .. _sphx_glr_download_how_to_compile_models_from_tensorflow.py:
diff --git a/docs/_sources/how_to/compile_models/sg_execution_times.rst.txt b/docs/_sources/how_to/compile_models/sg_execution_times.rst.txt
index 520e92eba8..a48bc01f28 100644
--- a/docs/_sources/how_to/compile_models/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/compile_models/sg_execution_times.rst.txt
@@ -5,26 +5,26 @@
 
 Computation times
 =================
-**06:47.913** total execution time for **how_to_compile_models** files:
+**07:06.436** total execution time for **how_to_compile_models** files:
 
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``) | 01:31.026 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``) | 01:34.630 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``)       | 01:18.482 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``)       | 01:24.383 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``)         | 00:57.297 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``)         | 00:59.418 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_oneflow.py` (``from_oneflow.py``)       | 00:38.207 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_oneflow.py` (``from_oneflow.py``)       | 00:39.250 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``)         | 00:32.606 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``)         | 00:33.876 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``)           | 00:31.049 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``)           | 00:31.554 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``)         | 00:27.114 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``)         | 00:28.065 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``)       | 00:26.129 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``)       | 00:27.296 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``)           | 00:23.268 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``)           | 00:25.134 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``)             | 00:02.735 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``)             | 00:02.828 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/deploy_models/deploy_model_on_adreno.rst.txt b/docs/_sources/how_to/deploy_models/deploy_model_on_adreno.rst.txt
index 23e8cb1e04..dce80d2072 100644
--- a/docs/_sources/how_to/deploy_models/deploy_model_on_adreno.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_model_on_adreno.rst.txt
@@ -637,18 +637,13 @@ well as provides information about the model's performance
     Evaluate inference time cost...
     Execution time summary:
      mean (ms)   median (ms)    max (ms)     min (ms)     std (ms)  
-     3336.1227    3335.3472    3341.7500    3332.7434      2.7839   
+     2683.2965    2681.7872    2694.2407    2679.4584      4.2230   
                
 
 
 
 
 
-.. rst-class:: sphx-glr-timing
-
-   **Total running time of the script:** ( 1 minutes  4.862 seconds)
-
-
 .. _sphx_glr_download_how_to_deploy_models_deploy_model_on_adreno.py:
 
 .. only:: html
diff --git a/docs/_sources/how_to/deploy_models/deploy_model_on_adreno_tvmc.rst.txt b/docs/_sources/how_to/deploy_models/deploy_model_on_adreno_tvmc.rst.txt
index 665b34318e..17befc7e02 100644
--- a/docs/_sources/how_to/deploy_models/deploy_model_on_adreno_tvmc.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_model_on_adreno_tvmc.rst.txt
@@ -127,7 +127,7 @@ Make a Keras Resnet50 Model
  .. code-block:: none
 
     Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/resnet/resnet50_weights_tf_dim_ordering_tf_kernels.h5
-
         8192/102967424 [..............................] - ETA: 0s
      7757824/102967424 [=>............................] - ETA: 0s
     12623872/102967424 [==>...........................] - ETA: 0s
     17793024/102967424 [====>.........................] - ETA: 0s
     22847488/102967424 [=====>........................] - ETA: 0s
     27820032/102967424 [=======>......................] - ETA: 0s
     32825344/102967424 [========>.....................] - ETA: 0s
     33546240/102967424 [========>.....................] - ETA: 0s
 
     33685504/102967424 [========>.....................] - ETA: 1s
     41402368/102967424 [===========>..................] - ETA: 0s
     50462720/102967424 [=============>................] - ETA: 0s
     64430080/102967424 [=================>............] - ETA: 0s
     72998912/102967424 [====================>.........] - ETA: 0s
     78110720/102967424 [=====================>........] - ETA: 0s
     83124224/102967424 [=======================>......] - ETA: 0s
     88227840/102967424 [========================>.....] -
  ETA: 0s
     93241344/102967424 [==========================>...] - ETA: 0s
     95518720/102967424 [==========================>...] - ETA: 0s
    102967424/102967424 [==============================] - 1s 0us/step
+
         8192/102967424 [..............................] - ETA: 0s
      8380416/102967424 [=>............................] - ETA: 0s
     16769024/102967424 [===>..........................] - ETA: 0s
     19955712/102967424 [====>.........................] - ETA: 1s
     22429696/102967424 [=====>........................] - ETA: 1s
     25567232/102967424 [======>.......................] - ETA: 1s
     32661504/102967424 [========>.....................] - ETA: 0s
     34480128/102967424 [=========>....................] - ETA: 0s
 
     39092224/102967424 [==========>...................] - ETA: 0s
     41934848/102967424 [===========>..................] - ETA: 0s
     48578560/102967424 [=============>................] - ETA: 0s
     52944896/102967424 [==============>...............] - ETA: 0s
     58712064/102967424 [================>.............] - ETA: 0s
     60907520/102967424 [================>.............] - ETA: 0s
     61505536/102967424 [================>.............] - ETA: 0s
     67239936/102967424 [==================>...........] -
  ETA: 0s
     75489280/102967424 [====================>.........] - ETA: 0s
     83877888/102967424 [=======================>......] - ETA: 0s
     92266496/102967424 [=========================>....] - ETA: 0s
     98910208/102967424 [===========================>..] - ETA: 0s
    102850560/102967424 [============================>.] - ETA: 0s
    102967424/102967424 [==============================] - 2s 0us/step
 
 
 
diff --git a/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt b/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt
index 34543fd69e..4f43fa6ca7 100644
--- a/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt
@@ -437,7 +437,7 @@ Execute on TVM
     Evaluate inference time cost...
     Execution time summary:
      mean (ms)   median (ms)    max (ms)     min (ms)     std (ms)  
-      15.9276      15.9283      16.4350      15.6684       0.1924   
+      16.8834      16.8946      17.5389      16.2270       0.3879   
                
 
 
diff --git a/docs/_sources/how_to/deploy_models/deploy_object_detection_pytorch.rst.txt b/docs/_sources/how_to/deploy_models/deploy_object_detection_pytorch.rst.txt
index 63b62ef1c6..c88c305398 100644
--- a/docs/_sources/how_to/deploy_models/deploy_object_detection_pytorch.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_object_detection_pytorch.rst.txt
@@ -130,7 +130,7 @@ Load pre-trained maskrcnn from torchvision and do tracing
     /venv/apache-tvm-py3.7/lib/python3.7/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and will be removed in 0.15. The current behavior is equivalent to passing `weights=MaskRCNN_ResNet50_FPN_Weights.COCO_V1`. You can also use `weights=MaskRCNN_ResNet50_FPN_Weights.DEFAULT` to get the most up-to-date weights.
       warnings.warn(msg)
     Downloading: "https://download.pytorch.org/models/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth" to /workspace/.cache/torch/hub/checkpoints/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth
-
      0%|          | 0.00/170M [00:00<?, ?B/s]
      7%|7         | 12.1M/170M [00:00<00:01, 127MB/s]
     14%|#4        | 24.2M/170M [00:00<00:01, 104MB/s]
     20%|##        | 34.7M/170M [00:00<00:01, 107MB/s]
     27%|##6       | 45.1M/170M [00:00<00:01, 104MB/s]
     32%|###2      | 55.1M/170M [00:00<00:01, 103MB/s]
     38%|###8      | 64.9M/170M [00:00<00:01, 85.1MB/s]
     47%|####7     | 80.0M/170M [00:00<00:00, 105MB/s] 
     53%|#####3    | 90.7M/170M [00:00<00:00, 105MB/s]
     59%|#####9    | 101M/170M [00:01<00:00, 103MB/s] 
     65%|######5   | 111M/170M [00:01<00:00, 102MB/s]
     71%|#######1  | 121M/170M [00:01<00:00, 101MB/s]
     77%|#######7  | 131M/170M [00:01<00:00, 101MB/s]
     83%|########2 | 141M/170M [00:01<00:00, 101MB/s]
     89%|########8 | 150M/170M [00:01<00:00, 101MB/s]
     94%|#########4| 160M/170M [00:01<00:00, 94.6MB/s]
    100%|##########| 170M/170M [00:01<00:00, 102MB/s] 
+
      0%|          | 0.00/170M [00:00<?, ?B/s]
      5%|4         | 7.99M/170M [00:00<00:02, 73.2MB/s]
      9%|9         | 16.0M/170M [00:00<00:02, 74.1MB/s]
     14%|#4        | 24.0M/170M [00:00<00:02, 59.7MB/s]
     18%|#7        | 30.3M/170M [00:00<00:02, 51.2MB/s]
     21%|##        | 35.4M/170M [00:00<00:03, 46.4MB/s]
     24%|##3       | 40.6M/170M [00:00<00:02, 48.5MB/s]
     28%|##8       | 48.0M/170M [00:00<00:02, 46.8MB/s]
     33%|###2      | 56.0M/170M [00:01<00:02, 54.8MB/s]
     38%|###7      | 64.0M/170M [00:01<00:02, 55.3MB/s]
     42%|####2     | 72.1M/170M [00:01<00:01, 61.7MB/s]
     50%|####9     | 84.4M/170M [00:01<00:01, 78.9MB/s]
     54%|#####4    | 92.5M/170M [00:01<00:01, 63.1MB/s]
     58%|#####8    | 99.3M/170M [00:01<00:01, 58.7MB/s]
     62%|######2   | 105M/170M [00:01<00:01, 58.4MB/s] 
     68%|######7   | 115M/170M [00:02<00:00, 68.6MB/s]
     72%|#######1  | 122M/170M [00:02<00:00, 67.0MB/s]
     76%|#######5  | 129M/170M [00:02<00:00, 49.2MB/s]
      80%|########  | 136M/170M [00:02<00:00, 49.0MB/s]
     85%|########4 | 144M/170M [00:02<00:00, 55.9MB/s]
     88%|########8 | 150M/170M [00:02<00:00, 56.0MB/s]
     93%|#########3| 158M/170M [00:02<00:00, 59.8MB/s]
     97%|#########6| 164M/170M [00:02<00:00, 56.6MB/s]
    100%|##########| 170M/170M [00:03<00:00, 58.1MB/s]
     /venv/apache-tvm-py3.7/lib/python3.7/site-packages/torch/nn/functional.py:3897: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
       for i in range(dim)
     /venv/apache-tvm-py3.7/lib/python3.7/site-packages/torchvision/models/detection/anchor_utils.py:124: UserWarning: __floordiv__ is deprecated, and its behavior will change in a future version of pytorch. It currently rounds toward 0 (like the 'trunc' function NOT 'floor'). This results in incorrect rounding for negative values. To keep the current behavior, use torch.div(a, b, rounding_mode='trunc'), or for actual floor division, use torch.div(a, b, rounding_mode='floor').
@@ -299,7 +299,7 @@ Get boxes with score larger than 0.9
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 3 minutes  37.088 seconds)
+   **Total running time of the script:** ( 3 minutes  50.861 seconds)
 
 
 .. _sphx_glr_download_how_to_deploy_models_deploy_object_detection_pytorch.py:
diff --git a/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt b/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt
index 550ec53318..a5f57415df 100644
--- a/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt
@@ -227,7 +227,7 @@ training. Other models require a full post training calibration.
     /venv/apache-tvm-py3.7/lib/python3.7/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and will be removed in 0.15. The current behavior is equivalent to passing `weights=MobileNet_V2_Weights.IMAGENET1K_V1`. You can also use `weights=MobileNet_V2_Weights.DEFAULT` to get the most up-to-date weights.
       warnings.warn(msg)
     Downloading: "https://download.pytorch.org/models/mobilenet_v2-b0353104.pth" to /workspace/.cache/torch/hub/checkpoints/mobilenet_v2-b0353104.pth
-
      0%|          | 0.00/13.6M [00:00<?, ?B/s]
     59%|#####8    | 7.99M/13.6M [00:00<00:00, 57.9MB/s]
    100%|##########| 13.6M/13.6M [00:00<00:00, 77.3MB/s]
+
      0%|          | 0.00/13.6M [00:00<?, ?B/s]
     59%|#####8    | 7.99M/13.6M [00:00<00:00, 39.4MB/s]
    100%|##########| 13.6M/13.6M [00:00<00:00, 57.4MB/s]
 
 
 
@@ -409,7 +409,7 @@ Here we give an example of how to measure performance of TVM compiled models.
 
     Execution time summary:
      mean (ms)   median (ms)    max (ms)     min (ms)     std (ms)  
-      90.3702      90.3248      91.0267      89.9968       0.2241   
+      90.5573      90.4678      91.6773      90.2090       0.2931   
                
 
 
@@ -458,7 +458,7 @@ TODO
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  17.916 seconds)
+   **Total running time of the script:** ( 1 minutes  20.761 seconds)
 
 
 .. _sphx_glr_download_how_to_deploy_models_deploy_prequantized.py:
diff --git a/docs/_sources/how_to/deploy_models/deploy_prequantized_tflite.rst.txt b/docs/_sources/how_to/deploy_models/deploy_prequantized_tflite.rst.txt
index 85ebe76ca3..321fbf0825 100644
--- a/docs/_sources/how_to/deploy_models/deploy_prequantized_tflite.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_prequantized_tflite.rst.txt
@@ -423,7 +423,7 @@ Here we give an example of how to measure performance of TVM compiled models.
 
     Execution time summary:
      mean (ms)   median (ms)    max (ms)     min (ms)     std (ms)  
-      121.2262     121.1393     125.6978     120.5065      0.5704   
+      121.6151     121.5017     127.3501     120.7108      0.7841   
                
 
 
@@ -460,7 +460,7 @@ Here we give an example of how to measure performance of TVM compiled models.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 2 minutes  31.789 seconds)
+   **Total running time of the script:** ( 2 minutes  31.371 seconds)
 
 
 .. _sphx_glr_download_how_to_deploy_models_deploy_prequantized_tflite.py:
diff --git a/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt b/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
index cfa0e5d243..037fd04f3d 100644
--- a/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
@@ -257,7 +257,7 @@ We create a Relay VM to build and execute the model.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  21.232 seconds)
+   **Total running time of the script:** ( 1 minutes  51.610 seconds)
 
 
 .. _sphx_glr_download_how_to_deploy_models_deploy_quantized.py:
diff --git a/docs/_sources/how_to/deploy_models/deploy_ssd_gluoncv.rst.txt b/docs/_sources/how_to/deploy_models/deploy_ssd_gluoncv.rst.txt
index de6bfd6897..777114bb5a 100644
--- a/docs/_sources/how_to/deploy_models/deploy_ssd_gluoncv.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_ssd_gluoncv.rst.txt
@@ -170,7 +170,7 @@ Convert and compile model for CPU.
             data: None
       input_sym_arg_type = in_param.infer_type()[0]
     Downloading /workspace/.mxnet/models/ssd_512_resnet50_v1_voc-9c8b225a.zip from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/ssd_512_resnet50_v1_voc-9c8b225a.zip...
-
      0%|          | 0/132723 [00:00<?, ?KB/s]
      1%|1         | 1381/132723 [00:00<00:09, 13807.77KB/s]
      3%|3         | 4239/132723 [00:00<00:05, 22495.87KB/s]
      7%|6         | 8896/132723 [00:00<00:03, 33349.47KB/s]
     12%|#2        | 16162/132723 [00:00<00:02, 48828.34KB/s]
     19%|#8        | 24802/132723 [00:00<00:01, 62353.63KB/s]
     25%|##5       | 33523/132723 [00:00<00:01, 70793.29KB/s]
     32%|###1      | 42144/132723 [00:00<00:01, 75825.85KB/s]
     38%|###8      | 50931/132723 [00:00<00:01, 79655.93KB/s]
     45%|####4     | 59697/132723 [00:00<00:00, 82153.61KB/s]
     52%|#####1    | 68472/132723 [00:01<00:00, 83878.61KB/s]
     58%|#####8    | 77221/132723 [00:01<00:00, 84979.87KB/s]
     65%|######4   | 85915/132723 [00:01<00:00, 85574.27KB/s]
     71%|#######1  | 94677/132723 [00:01<00:00, 86192.75KB/s]
     78%|#######7  | 103474/132723 [00:01<00:00, 86720.72KB/s]
     85%|########4 | 112259/132723 [00:01<00:00, 87059.16KB/s]
     91%|#########1|
  120978/132723 [00:01<00:00, 87093.75KB/s]
     98%|#########7| 129688/132723 [00:01<00:00, 87002.67KB/s]
    100%|##########| 132723/132723 [00:01<00:00, 76381.89KB/s]
+
      0%|          | 0/132723 [00:00<?, ?KB/s]
      1%|          | 713/132723 [00:00<00:18, 7084.57KB/s]
      1%|1         | 1897/132723 [00:00<00:13, 9826.07KB/s]
      3%|2         | 3785/132723 [00:00<00:09, 13866.95KB/s]
      5%|5         | 6777/132723 [00:00<00:06, 20141.75KB/s]
      9%|8         | 11501/132723 [00:00<00:04, 29867.33KB/s]
     14%|#4        | 18953/132723 [00:00<00:02, 45001.73KB/s]
     21%|##        | 27537/132723 [00:00<00:01, 58323.32KB/s]
     27%|##7       | 36167/132723 [00:00<00:01, 67214.27KB/s]
     34%|###3      | 44691/132723 [00:00<00:01, 72842.78KB/s]
     40%|####      | 53295/132723 [00:01<00:01, 76911.62KB/s]
     47%|####6     | 61981/132723 [00:01<00:00, 79951.06KB/s]
     53%|#####3    | 70644/132723 [00:01<00:00, 81980.55KB/s]
     60%|#####9    | 79348/132723 [00:01<00:00, 83511.56KB/s]
     66%|######6   | 88016/132723 [00:01<00:00, 84466.28KB/s]
     73%|#######2  | 96745/132723 [00:01<00:00, 85314.01KB/s]
     79%|#######9  | 10535
 8/132723 [00:01<00:00, 85557.28KB/s]
     86%|########5 | 113993/132723 [00:01<00:00, 85783.90KB/s]
     92%|#########2| 122654/132723 [00:01<00:00, 86029.53KB/s]
     99%|#########8| 131258/132723 [00:01<00:00, 85792.40KB/s]
    100%|##########| 132723/132723 [00:01<00:00, 69006.57KB/s]
 
 
 
@@ -246,7 +246,7 @@ Display result
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 3 minutes  54.132 seconds)
+   **Total running time of the script:** ( 4 minutes  1.036 seconds)
 
 
 .. _sphx_glr_download_how_to_deploy_models_deploy_ssd_gluoncv.py:
diff --git a/docs/_sources/how_to/deploy_models/sg_execution_times.rst.txt b/docs/_sources/how_to/deploy_models/sg_execution_times.rst.txt
index 32e77286f6..59d327ac21 100644
--- a/docs/_sources/how_to/deploy_models/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/deploy_models/sg_execution_times.rst.txt
@@ -5,28 +5,28 @@
 
 Computation times
 =================
-**16:19.722** total execution time for **how_to_deploy_models** files:
+**17:10.847** total execution time for **how_to_deploy_models** files:
 
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``)                           | 03:54.132 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``)                           | 04:01.036 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``) | 03:37.088 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``) | 03:50.861 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``)           | 02:31.789 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``)           | 02:31.371 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``)                               | 01:21.232 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``)                               | 01:51.610 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``)                         | 01:17.916 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``)                         | 01:20.761 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_adreno.py` (``deploy_model_on_adreno.py``)                   | 01:04.862 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_adreno.py` (``deploy_model_on_adreno.py``)                   | 00:57.822 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_adreno_tvmc.py` (``deploy_model_on_adreno_tvmc.py``)         | 00:51.934 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_adreno_tvmc.py` (``deploy_model_on_adreno_tvmc.py``)         | 00:53.536 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``)                 | 00:43.679 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``)                 | 00:45.312 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_nano.py` (``deploy_model_on_nano.py``)                       | 00:28.984 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_nano.py` (``deploy_model_on_nano.py``)                       | 00:29.511 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``)                       | 00:28.101 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``)                       | 00:29.020 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
 | :ref:`sphx_glr_how_to_deploy_models_deploy_sparse.py` (``deploy_sparse.py``)                                     | 00:00.006 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/extend_tvm/bring_your_own_datatypes.rst.txt b/docs/_sources/how_to/extend_tvm/bring_your_own_datatypes.rst.txt
index 57e520c727..6414306580 100644
--- a/docs/_sources/how_to/extend_tvm/bring_your_own_datatypes.rst.txt
+++ b/docs/_sources/how_to/extend_tvm/bring_your_own_datatypes.rst.txt
@@ -463,7 +463,7 @@ First let us define two helper functions to get the mobilenet model and a cat im
 
  .. code-block:: none
 
-    Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zip26bff2dd-e83e-4c22-9a2c-bcabb5b4c5ce from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
+    Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zipfb42a426-6471-4b98-9fbb-2ad9cea9a496 from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
 
 
 
diff --git a/docs/_sources/how_to/extend_tvm/sg_execution_times.rst.txt b/docs/_sources/how_to/extend_tvm/sg_execution_times.rst.txt
index ac294ebc8d..892d1c6220 100644
--- a/docs/_sources/how_to/extend_tvm/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/extend_tvm/sg_execution_times.rst.txt
@@ -5,14 +5,14 @@
 
 Computation times
 =================
-**00:56.188** total execution time for **how_to_extend_tvm** files:
+**00:57.766** total execution time for **how_to_extend_tvm** files:
 
 +-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_bring_your_own_datatypes.py` (``bring_your_own_datatypes.py``) | 00:52.233 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_bring_your_own_datatypes.py` (``bring_your_own_datatypes.py``) | 00:53.696 | 0.0 MB |
 +-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``)           | 00:02.840 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``)           | 00:02.922 | 0.0 MB |
 +-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``)                     | 00:01.108 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``)                     | 00:01.140 | 0.0 MB |
 +-------------------------------------------------------------------------------------------------+-----------+--------+
 | :ref:`sphx_glr_how_to_extend_tvm_low_level_custom_pass.py` (``low_level_custom_pass.py``)       | 00:00.007 | 0.0 MB |
 +-------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt b/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt
index dd926a2752..b115841303 100644
--- a/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt
+++ b/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt
@@ -220,10 +220,10 @@ profile the execution time of each passes.
  .. code-block:: none
 
     Printing results of timing profile...
-    InferType: 22434us [22434us] (48.67%; 48.67%)
-    FoldScaleAxis: 23657us [9us] (51.33%; 51.33%)
-            FoldConstant: 23648us [1726us] (51.31%; 99.96%)
-                    InferType: 21922us [21922us] (47.56%; 92.70%)
+    InferType: 22934us [22934us] (48.32%; 48.32%)
+    FoldScaleAxis: 24529us [11us] (51.68%; 51.68%)
+            FoldConstant: 24519us [2123us] (51.66%; 99.96%)
+                    InferType: 22396us [22396us] (47.18%; 91.34%)
 
 
 
@@ -262,10 +262,10 @@ Refer to following sections and :py:func:`tvm.instrument.pass_instrument` for th
  .. code-block:: none
 
     Printing results of timing profile...
-    InferType: 22129us [22129us] (48.31%; 48.31%)
-    FoldScaleAxis: 23675us [8us] (51.69%; 51.69%)
-            FoldConstant: 23667us [1774us] (51.67%; 99.97%)
-                    InferType: 21893us [21893us] (47.80%; 92.50%)
+    InferType: 22070us [22070us] (48.53%; 48.53%)
+    FoldScaleAxis: 23405us [6us] (51.47%; 51.47%)
+            FoldConstant: 23399us [1771us] (51.46%; 99.98%)
+                    InferType: 21629us [21629us] (47.56%; 92.43%)
 
 
 
diff --git a/docs/_sources/how_to/optimize_operators/opt_conv_cuda.rst.txt b/docs/_sources/how_to/optimize_operators/opt_conv_cuda.rst.txt
index 7d20adbadb..48a9adb571 100644
--- a/docs/_sources/how_to/optimize_operators/opt_conv_cuda.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_conv_cuda.rst.txt
@@ -331,7 +331,7 @@ latency of convolution.
 
  .. code-block:: none
 
-    Convolution: 53.503616 ms
+    Convolution: 45.397857 ms
 
 
 
diff --git a/docs/_sources/how_to/optimize_operators/opt_conv_tensorcore.rst.txt b/docs/_sources/how_to/optimize_operators/opt_conv_tensorcore.rst.txt
index bce9c5db2c..052e31a358 100644
--- a/docs/_sources/how_to/optimize_operators/opt_conv_tensorcore.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_conv_tensorcore.rst.txt
@@ -598,7 +598,7 @@ be able to run on our build server
 
  .. code-block:: none
 
-    conv2d with tensor core: 12.245091 ms
+    conv2d with tensor core: 12.237002 ms
 
 
 
diff --git a/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt b/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
index a6b0dfb153..8f14dd17f7 100644
--- a/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
@@ -134,8 +134,8 @@ Then we write a baseline implementation, the simplest way to write a matrix mult
 
  .. code-block:: none
 
-    Numpy running time: 0.018616
-    Baseline: 3.518981
+    Numpy running time: 0.020305
+    Baseline: 3.363956
 
 
 
@@ -227,7 +227,7 @@ fill 32 * 32 * sizeof(float) which is 4KB in the cache whose total size is 32KB
 
  .. code-block:: none
 
-    Opt1: 0.298253
+    Opt1: 0.332628
 
 
 
@@ -318,7 +318,7 @@ In this tutorial, we chose to vectorize the inner loop row data since it is cach
 
  .. code-block:: none
 
-    Opt2: 0.327582
+    Opt2: 0.359120
 
 
 
@@ -406,7 +406,7 @@ the access pattern for A matrix is more cache friendly.
 
  .. code-block:: none
 
-    Opt3: 0.122265
+    Opt3: 0.127020
 
 
 
@@ -523,7 +523,7 @@ flattening.
 
  .. code-block:: none
 
-    Opt4: 0.110673
+    Opt4: 0.109359
 
 
 
@@ -635,7 +635,7 @@ write to C when all the block results are ready.
 
  .. code-block:: none
 
-    Opt5: 0.111837
+    Opt5: 0.111824
 
 
 
@@ -748,7 +748,7 @@ Furthermore, we can also utilize multi-core processors to do the thread-level pa
 
  .. code-block:: none
 
-    Opt6: 0.147249
+    Opt6: 0.148218
 
 
 
diff --git a/docs/_sources/how_to/optimize_operators/sg_execution_times.rst.txt b/docs/_sources/how_to/optimize_operators/sg_execution_times.rst.txt
index f6ae50af1e..693fc14ed0 100644
--- a/docs/_sources/how_to/optimize_operators/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/sg_execution_times.rst.txt
@@ -5,12 +5,12 @@
 
 Computation times
 =================
-**00:35.809** total execution time for **how_to_optimize_operators** files:
+**00:36.430** total execution time for **how_to_optimize_operators** files:
 
 +-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``)                       | 00:32.718 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``)                       | 00:33.330 | 0.0 MB |
 +-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``) | 00:01.912 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``) | 00:01.904 | 0.0 MB |
 +-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``)             | 00:01.179 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``)             | 00:01.196 | 0.0 MB |
 +-----------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/tune_with_autoscheduler/sg_execution_times.rst.txt b/docs/_sources/how_to/tune_with_autoscheduler/sg_execution_times.rst.txt
index d21eb0ec32..6c0498d3f8 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/sg_execution_times.rst.txt
@@ -5,18 +5,18 @@
 
 Computation times
 =================
-**10:07.766** total execution time for **how_to_tune_with_autoscheduler** files:
+**10:22.168** total execution time for **how_to_tune_with_autoscheduler** files:
 
 +----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_conv2d_layer_cuda.py` (``tune_conv2d_layer_cuda.py``) | 06:10.669 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_conv2d_layer_cuda.py` (``tune_conv2d_layer_cuda.py``) | 06:18.523 | 0.0 MB |
 +----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``)             | 01:43.939 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``)             | 01:45.674 | 0.0 MB |
 +----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``)           | 01:12.321 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``)           | 01:13.936 | 0.0 MB |
 +----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``)               | 00:32.355 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``)               | 00:34.755 | 0.0 MB |
 +----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``)             | 00:14.666 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``)             | 00:14.952 | 0.0 MB |
 +----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``)           | 00:13.817 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``)           | 00:14.327 | 0.0 MB |
 +----------------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.rst.txt b/docs/_sources/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.rst.txt
index 6fea7ebe27..7677dee803 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.rst.txt
@@ -243,71 +243,551 @@ cooperative fetching, unrolling and operator fusion.
         @T.prim_func
         def main(data: T.Buffer((1, 512, 7, 7), "float32"), kernel: T.Buffer((512, 512, 3, 3), "float32"), bias: T.Buffer((1, 512, 1, 1), "float32"), compute: T.Buffer((1, 512, 7, 7), "float32")):
             T.func_attr({"from_legacy_te_schedule": T.bool(True), "global_symbol": "main", "tir.noalias": T.bool(True)})
-            blockIdx_x = T.launch_thread("blockIdx.x", 32)
+            blockIdx_x = T.launch_thread("blockIdx.x", 112)
             conv2d_nchw = T.allocate([4], "float32", "local")
-            pad_temp_shared = T.allocate([324], "float32", "shared")
-            kernel_shared = T.allocate([576], "float32", "shared")
-            threadIdx_x = T.launch_thread("threadIdx.x", 196)
-            conv2d_nchw_1 = T.Buffer((1,), data=conv2d_nchw, scope="local", align=4)
+            pad_temp_shared = T.allocate([672], "float32", "shared")
+            kernel_shared = T.allocate([3072], "float32", "shared")
+            threadIdx_x = T.launch_thread("threadIdx.x", 56)
+            conv2d_nchw_1 = T.Buffer((4,), data=conv2d_nchw, scope="local", align=8)
             conv2d_nchw_1[0] = T.float32(0)
-            conv2d_nchw_1[1] = T.float32(0)
             conv2d_nchw_1[2] = T.float32(0)
+            conv2d_nchw_1[1] = T.float32(0)
             conv2d_nchw_1[3] = T.float32(0)
-            for rc_outer_outer in range(128):
+            for rc_outer_outer, ry_outer_outer in T.grid(16, 3):
+                cse_var_4: T.int32 = rc_outer_outer * 1568
+                cse_var_3: T.int32 = ry_outer_outer * 7
+                cse_var_2: T.int32 = rc_outer_outer * 288
+                cse_var_1: T.int32 = ry_outer_outer * 3
                 threadIdx_x_1 = T.env_thread("threadIdx.x")
-                pad_temp_shared_1 = T.Buffer((324,), data=pad_temp_shared, scope="shared")
+                pad_temp_shared_1 = T.Buffer((672,), data=pad_temp_shared, scope="shared")
                 data_1 = T.Buffer((25088,), data=data.data)
-                with T.launch_thread(threadIdx_x_1, 196):
-                    pad_temp_shared_1[threadIdx_x_1] = T.if_then_else(9 <= threadIdx_x_1 % 81 and threadIdx_x_1 % 81 < 72 and 1 <= threadIdx_x_1 % 9 and threadIdx_x_1 % 9 < 8, data_1[rc_outer_outer * 196 + threadIdx_x_1 // 81 * 49 + threadIdx_x_1 % 81 // 9 * 7 + threadIdx_x_1 % 9 - 8], T.float32(0))
-                with T.launch_thread(threadIdx_x_1, 196):
-                    if T.likely(threadIdx_x_1 < 128):
-                        pad_temp_shared_1[threadIdx_x_1 + 196] = T.if_then_else(9 <= (threadIdx_x_1 + 34) % 81 and (threadIdx_x_1 + 34) % 81 < 72 and 1 <= (threadIdx_x_1 + 7) % 9 and (threadIdx_x_1 + 7) % 9 < 8, data_1[rc_outer_outer * 196 + (threadIdx_x_1 + 196) // 81 * 49 + (threadIdx_x_1 + 34) % 81 // 9 * 7 + (threadIdx_x_1 + 7) % 9 - 8], T.float32(0))
-                kernel_shared_1 = T.Buffer((576,), data=kernel_shared, scope="shared")
-                with T.launch_thread("threadIdx.x", 196) as threadIdx_x_2:
-                    kernel_1 = T.Buffer((2359296,), data=kernel.data)
-                    if T.likely(threadIdx_x_2 < 96):
-                        kernel_shared_1[threadIdx_x_2 * 6] = kernel_1[blockIdx_x * 73728 + threadIdx_x_2 // 6 * 4608 + rc_outer_outer * 36 + threadIdx_x_2 % 6 * 6]
-                    if T.likely(threadIdx_x_2 < 96):
-                        kernel_shared_1[threadIdx_x_2 * 6 + 1] = kernel_1[blockIdx_x * 73728 + threadIdx_x_2 // 6 * 4608 + rc_outer_outer * 36 + threadIdx_x_2 % 6 * 6 + 1]
-                    if T.likely(threadIdx_x_2 < 96):
-                        kernel_shared_1[threadIdx_x_2 * 6 + 2] = kernel_1[blockIdx_x * 73728 + threadIdx_x_2 // 6 * 4608 + rc_outer_outer * 36 + threadIdx_x_2 % 6 * 6 + 2]
-                    if T.likely(threadIdx_x_2 < 96):
-                        kernel_shared_1[threadIdx_x_2 * 6 + 3] = kernel_1[blockIdx_x * 73728 + threadIdx_x_2 // 6 * 4608 + rc_outer_outer * 36 + threadIdx_x_2 % 6 * 6 + 3]
-                    if T.likely(threadIdx_x_2 < 96):
-                        kernel_shared_1[threadIdx_x_2 * 6 + 4] = kernel_1[blockIdx_x * 73728 + threadIdx_x_2 // 6 * 4608 + rc_outer_outer * 36 + threadIdx_x_2 % 6 * 6 + 4]
-                    if T.likely(threadIdx_x_2 < 96):
-                        kernel_shared_1[threadIdx_x_2 * 6 + 5] = kernel_1[blockIdx_x * 73728 + threadIdx_x_2 // 6 * 4608 + rc_outer_outer * 36 + threadIdx_x_2 % 6 * 6 + 5]
-                for rc_outer_inner, ry_outer_inner in T.grid(2, 3):
-                    conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3]
-                    conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 144]
-                    conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 288]
-                    conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 432]
-                    conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 1]
-                    conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 145]
-                    conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 289]
-                    conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 433]
-                    conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 2]
-                    conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 146]
-                    conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 290]
-                    conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 434]
-                    conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 9]
-                    conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 153]
-                    conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 297]
-                    conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 441]
-                    conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 10]
-                    conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 154]
-                    conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 298]
-                    conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 442]
-                    conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 11]
-                    conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 155]
-                    conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 299]
-                    conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 443]
-            compute_1 = T.Buffer((25088,), data=compute.data)
-            bias_1 = T.Buffer((512,), data=bias.data)
-            compute_1[blockIdx_x * 784 + threadIdx_x] = T.max(conv2d_nchw_1[0] + bias_1[blockIdx_x * 16 + threadIdx_x // 49], T.float32(0))
-            compute_1[blockIdx_x * 784 + threadIdx_x + 196] = T.max(conv2d_nchw_1[1] + bias_1[blockIdx_x * 16 + threadIdx_x // 49 + 4], T.float32(0))
-            compute_1[blockIdx_x * 784 + threadIdx_x + 392] = T.max(conv2d_nchw_1[2] + bias_1[blockIdx_x * 16 + threadIdx_x // 49 + 8], T.float32(0))
-            compute_1[blockIdx_x * 784 + threadIdx_x + 588] = T.max(conv2d_nchw_1[3] + bias_1[blockIdx_x * 16 + threadIdx_x // 49 + 12], T.float32(0))
+                with T.launch_thread(threadIdx_x_1, 56):
+                    pad_temp_shared_1[threadIdx_x_1] = T.if_then_else(1 <= threadIdx_x_1 % 21 // 3 + ry_outer_outer and threadIdx_x_1 % 21 // 3 + ry_outer_outer < 8 and 1 <= blockIdx_x % 7 + threadIdx_x_1 % 3 and blockIdx_x % 7 + threadIdx_x_1 % 3 < 8, data_1[cse_var_4 + threadIdx_x_1 // 3 * 7 + cse_var_3 + blockIdx_x % 7 + threadIdx_x_1 % 3 - 8], T.float32(0))
+                with T.launch_thread(threadIdx_x_1, 56):
+                    pad_temp_shared_1[threadIdx_x_1 + 56] = T.if_then_else(1 <= (threadIdx_x_1 + 14) % 21 // 3 + ry_outer_outer and (threadIdx_x_1 + 14) % 21 // 3 + ry_outer_outer < 8 and 1 <= blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 and blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 < 8, data_1[cse_var_4 + (threadIdx_x_1 + 56) // 3 * 7 + cse_var_3 + blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 - 8], T.float32(0))
+                with T.launch_thread(threadIdx_x_1, 56):
+                    pad_temp_shared_1[threadIdx_x_1 + 112] = T.if_then_else(1 <= (threadIdx_x_1 + 7) % 21 // 3 + ry_outer_outer and (threadIdx_x_1 + 7) % 21 // 3 + ry_outer_outer < 8 and 1 <= blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 and blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 < 8, data_1[cse_var_4 + (threadIdx_x_1 + 112) // 3 * 7 + cse_var_3 + blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 - 8], T.float32(0))
+                with T.launch_thread(threadIdx_x_1, 56):
+                    pad_temp_shared_1[threadIdx_x_1 + 168] = T.if_then_else(1 <= threadIdx_x_1 % 21 // 3 + ry_outer_outer and threadIdx_x_1 % 21 // 3 + ry_outer_outer < 8 and 1 <= blockIdx_x % 7 + threadIdx_x_1 % 3 and blockIdx_x % 7 + threadIdx_x_1 % 3 < 8, data_1[cse_var_4 + threadIdx_x_1 // 3 * 7 + cse_var_3 + blockIdx_x % 7 + threadIdx_x_1 % 3 + 384], T.float32(0))
+                with T.launch_thread(threadIdx_x_1, 56):
+                    pad_temp_shared_1[threadIdx_x_1 + 224] = T.if_then_else(1 <= (threadIdx_x_1 + 14) % 21 // 3 + ry_outer_outer and (threadIdx_x_1 + 14) % 21 // 3 + ry_outer_outer < 8 and 1 <= blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 and blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 < 8, data_1[cse_var_4 + (threadIdx_x_1 + 224) // 3 * 7 + cse_var_3 + blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 - 8], T.float32(0))
+                with T.launch_thread(threadIdx_x_1, 56):
+                    pad_temp_shared_1[threadIdx_x_1 + 280] = T.if_then_else(1 <= (threadIdx_x_1 + 7) % 21 // 3 + ry_outer_outer and (threadIdx_x_1 + 7) % 21 // 3 + ry_outer_outer < 8 and 1 <= blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 and blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 < 8, data_1[cse_var_4 + (threadIdx_x_1 + 280) // 3 * 7 + cse_var_3 + blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 - 8], T.float32(0))
+                with T.launch_thread(threadIdx_x_1, 56):
+                    pad_temp_shared_1[threadIdx_x_1 + 336] = T.if_then_else(1 <= threadIdx_x_1 % 21 // 3 + ry_outer_outer and threadIdx_x_1 % 21 // 3 + ry_outer_outer < 8 and 1 <= blockIdx_x % 7 + threadIdx_x_1 % 3 and blockIdx_x % 7 + threadIdx_x_1 % 3 < 8, data_1[cse_var_4 + threadIdx_x_1 // 3 * 7 + cse_var_3 + blockIdx_x % 7 + threadIdx_x_1 % 3 + 776], T.float32(0))
+                with T.launch_thread(threadIdx_x_1, 56):
+                    pad_temp_shared_1[threadIdx_x_1 + 392] = T.if_then_else(1 <= (threadIdx_x_1 + 14) % 21 // 3 + ry_outer_outer and (threadIdx_x_1 + 14) % 21 // 3 + ry_outer_outer < 8 and 1 <= blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 and blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 < 8, data_1[cse_var_4 + (threadIdx_x_1 + 392) // 3 * 7 + cse_var_3 + blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 - 8], T.float32(0))
+                with T.launch_thread(threadIdx_x_1, 56):
+                    pad_temp_shared_1[threadIdx_x_1 + 448] = T.if_then_else(1 <= (threadIdx_x_1 + 7) % 21 // 3 + ry_outer_outer and (threadIdx_x_1 + 7) % 21 // 3 + ry_outer_outer < 8 and 1 <= blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 and blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 < 8, data_1[cse_var_4 + (threadIdx_x_1 + 448) // 3 * 7 + cse_var_3 + blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 - 8], T.float32(0))
+                with T.launch_thread(threadIdx_x_1, 56):
+                    pad_temp_shared_1[threadIdx_x_1 + 504] = T.if_then_else(1 <= threadIdx_x_1 % 21 // 3 + ry_outer_outer and threadIdx_x_1 % 21 // 3 + ry_outer_outer < 8 and 1 <= blockIdx_x % 7 + threadIdx_x_1 % 3 and blockIdx_x % 7 + threadIdx_x_1 % 3 < 8, data_1[cse_var_4 + threadIdx_x_1 // 3 * 7 + cse_var_3 + blockIdx_x % 7 + threadIdx_x_1 % 3 + 1168], T.float32(0))
+                with T.launch_thread(threadIdx_x_1, 56):
+                    pad_temp_shared_1[threadIdx_x_1 + 560] = T.if_then_else(1 <= (threadIdx_x_1 + 14) % 21 // 3 + ry_outer_outer and (threadIdx_x_1 + 14) % 21 // 3 + ry_outer_outer < 8 and 1 <= blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 and blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 < 8, data_1[cse_var_4 + (threadIdx_x_1 + 560) // 3 * 7 + cse_var_3 + blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 - 8], T.float32(0))
+                with T.launch_thread(threadIdx_x_1, 56):
+                    pad_temp_shared_1[threadIdx_x_1 + 616] = T.if_then_else(1 <= (threadIdx_x_1 + 7) % 21 // 3 + ry_outer_outer and (threadIdx_x_1 + 7) % 21 // 3 + ry_outer_outer < 8 and 1 <= blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 and blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 < 8, data_1[cse_var_4 + (threadIdx_x_1 + 616) // 3 * 7 + cse_var_3 + blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 - 8], T.float32(0))
+                threadIdx_x_2 = T.env_thread("threadIdx.x")
+                kernel_shared_1 = T.Buffer((3072,), data=kernel_shared, scope="shared")
+                kernel_1 = T.Buffer((2359296,), data=kernel.data)
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2] = kernel_1[blockIdx_x // 7 * 147456 + cse_var_2 + threadIdx_x_2 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 56] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 56) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 56) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 112] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 112) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 168] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 168) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 // 3 + 24) % 32 * 9 + cse_var_1 + threadIdx_x_2 % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 224] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 224) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 280] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 280) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 88) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 336] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 336) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 // 3 + 16) % 32 * 9 + cse_var_1 + threadIdx_x_2 % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 392] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 392) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 8) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 448] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 448) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 64) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 504] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 504) // 96 * 4608 + cse_var_2 + threadIdx_x_2 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 72]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 560] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 560) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 80) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 616] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 616) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 40) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 672] = kernel_1[blockIdx_x // 7 * 147456 + cse_var_2 + threadIdx_x_2 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 32256]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 728] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 728) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 56) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 784] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 784) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 840] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 840) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 // 3 + 24) % 32 * 9 + cse_var_1 + threadIdx_x_2 % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 896] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 896) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 952] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 952) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 88) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 1008] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1008) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 // 3 + 16) % 32 * 9 + cse_var_1 + threadIdx_x_2 % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 1064] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1064) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 8) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 1120] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1120) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 64) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 1176] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1176) // 96 * 4608 + cse_var_2 + threadIdx_x_2 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 72]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 1232] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1232) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 80) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 1288] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1288) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 40) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 1344] = kernel_1[blockIdx_x // 7 * 147456 + cse_var_2 + threadIdx_x_2 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 64512]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 1400] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1400) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 56) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 1456] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1456) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 1512] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1512) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 // 3 + 24) % 32 * 9 + cse_var_1 + threadIdx_x_2 % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 1568] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1568) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 1624] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1624) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 88) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 1680] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1680) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 // 3 + 16) % 32 * 9 + cse_var_1 + threadIdx_x_2 % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 1736] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1736) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 8) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 1792] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1792) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 64) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 1848] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1848) // 96 * 4608 + cse_var_2 + threadIdx_x_2 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 72]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 1904] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1904) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 80) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 1960] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1960) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 40) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 2016] = kernel_1[blockIdx_x // 7 * 147456 + cse_var_2 + threadIdx_x_2 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 96768]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 2072] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2072) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 56) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 2128] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2128) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 2184] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2184) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 // 3 + 24) % 32 * 9 + cse_var_1 + threadIdx_x_2 % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 2240] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2240) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 2296] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2296) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 88) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 2352] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2352) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 // 3 + 16) % 32 * 9 + cse_var_1 + threadIdx_x_2 % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 2408] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2408) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 8) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 2464] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2464) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 64) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 2520] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2520) // 96 * 4608 + cse_var_2 + threadIdx_x_2 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 72]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 2576] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2576) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 80) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 2632] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2632) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 40) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 2688] = kernel_1[blockIdx_x // 7 * 147456 + cse_var_2 + threadIdx_x_2 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 129024]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 2744] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2744) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 56) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 2800] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2800) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 2856] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2856) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 // 3 + 24) % 32 * 9 + cse_var_1 + threadIdx_x_2 % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 2912] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2912) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    kernel_shared_1[threadIdx_x_2 + 2968] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2968) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 88) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+                with T.launch_thread(threadIdx_x_2, 56):
+                    if T.likely(threadIdx_x_2 < 48):
+                        kernel_shared_1[threadIdx_x_2 + 3024] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 3024) // 96 * 4608 + cse_var_2 + threadIdx_x_2 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 144]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3] * kernel_shared_1[threadIdx_x // 7 * 192]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3] * kernel_shared_1[threadIdx_x // 7 * 192 + 1536]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3] * kernel_shared_1[threadIdx_x // 7 * 192 + 96]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3] * kernel_shared_1[threadIdx_x // 7 * 192 + 1632]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 21] * kernel_shared_1[threadIdx_x // 7 * 192 + 3]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 21] * kernel_shared_1[threadIdx_x // 7 * 192 + 1539]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 21] * kernel_shared_1[threadIdx_x // 7 * 192 + 99]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 21] * kernel_shared_1[threadIdx_x // 7 * 192 + 1635]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 42] * kernel_shared_1[threadIdx_x // 7 * 192 + 6]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 42] * kernel_shared_1[threadIdx_x // 7 * 192 + 1542]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 42] * kernel_shared_1[threadIdx_x // 7 * 192 + 102]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 42] * kernel_shared_1[threadIdx_x // 7 * 192 + 1638]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 63] * kernel_shared_1[threadIdx_x // 7 * 192 + 9]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 63] * kernel_shared_1[threadIdx_x // 7 * 192 + 1545]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 63] * kernel_shared_1[threadIdx_x // 7 * 192 + 105]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 63] * kernel_shared_1[threadIdx_x // 7 * 192 + 1641]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 1] * kernel_shared_1[threadIdx_x // 7 * 192 + 1]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 1] * kernel_shared_1[threadIdx_x // 7 * 192 + 1537]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 1] * kernel_shared_1[threadIdx_x // 7 * 192 + 97]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 1] * kernel_shared_1[threadIdx_x // 7 * 192 + 1633]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 22] * kernel_shared_1[threadIdx_x // 7 * 192 + 4]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 22] * kernel_shared_1[threadIdx_x // 7 * 192 + 1540]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 22] * kernel_shared_1[threadIdx_x // 7 * 192 + 100]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 22] * kernel_shared_1[threadIdx_x // 7 * 192 + 1636]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 43] * kernel_shared_1[threadIdx_x // 7 * 192 + 7]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 43] * kernel_shared_1[threadIdx_x // 7 * 192 + 1543]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 43] * kernel_shared_1[threadIdx_x // 7 * 192 + 103]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 43] * kernel_shared_1[threadIdx_x // 7 * 192 + 1639]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 64] * kernel_shared_1[threadIdx_x // 7 * 192 + 10]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 64] * kernel_shared_1[threadIdx_x // 7 * 192 + 1546]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 64] * kernel_shared_1[threadIdx_x // 7 * 192 + 106]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 64] * kernel_shared_1[threadIdx_x // 7 * 192 + 1642]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 2] * kernel_shared_1[threadIdx_x // 7 * 192 + 2]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 2] * kernel_shared_1[threadIdx_x // 7 * 192 + 1538]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 2] * kernel_shared_1[threadIdx_x // 7 * 192 + 98]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 2] * kernel_shared_1[threadIdx_x // 7 * 192 + 1634]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 23] * kernel_shared_1[threadIdx_x // 7 * 192 + 5]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 23] * kernel_shared_1[threadIdx_x // 7 * 192 + 1541]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 23] * kernel_shared_1[threadIdx_x // 7 * 192 + 101]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 23] * kernel_shared_1[threadIdx_x // 7 * 192 + 1637]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 44] * kernel_shared_1[threadIdx_x // 7 * 192 + 8]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 44] * kernel_shared_1[threadIdx_x // 7 * 192 + 1544]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 44] * kernel_shared_1[threadIdx_x // 7 * 192 + 104]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 44] * kernel_shared_1[threadIdx_x // 7 * 192 + 1640]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 65] * kernel_shared_1[threadIdx_x // 7 * 192 + 11]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 65] * kernel_shared_1[threadIdx_x // 7 * 192 + 1547]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 65] * kernel_shared_1[threadIdx_x // 7 * 192 + 107]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 65] * kernel_shared_1[threadIdx_x // 7 * 192 + 1643]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 84] * kernel_shared_1[threadIdx_x // 7 * 192 + 12]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 84] * kernel_shared_1[threadIdx_x // 7 * 192 + 1548]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 84] * kernel_shared_1[threadIdx_x // 7 * 192 + 108]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 84] * kernel_shared_1[threadIdx_x // 7 * 192 + 1644]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 105] * kernel_shared_1[threadIdx_x // 7 * 192 + 15]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 105] * kernel_shared_1[threadIdx_x // 7 * 192 + 1551]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 105] * kernel_shared_1[threadIdx_x // 7 * 192 + 111]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 105] * kernel_shared_1[threadIdx_x // 7 * 192 + 1647]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 126] * kernel_shared_1[threadIdx_x // 7 * 192 + 18]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 126] * kernel_shared_1[threadIdx_x // 7 * 192 + 1554]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 126] * kernel_shared_1[threadIdx_x // 7 * 192 + 114]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 126] * kernel_shared_1[threadIdx_x // 7 * 192 + 1650]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 147] * kernel_shared_1[threadIdx_x // 7 * 192 + 21]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 147] * kernel_shared_1[threadIdx_x // 7 * 192 + 1557]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 147] * kernel_shared_1[threadIdx_x // 7 * 192 + 117]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 147] * kernel_shared_1[threadIdx_x // 7 * 192 + 1653]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 85] * kernel_shared_1[threadIdx_x // 7 * 192 + 13]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 85] * kernel_shared_1[threadIdx_x // 7 * 192 + 1549]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 85] * kernel_shared_1[threadIdx_x // 7 * 192 + 109]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 85] * kernel_shared_1[threadIdx_x // 7 * 192 + 1645]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 106] * kernel_shared_1[threadIdx_x // 7 * 192 + 16]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 106] * kernel_shared_1[threadIdx_x // 7 * 192 + 1552]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 106] * kernel_shared_1[threadIdx_x // 7 * 192 + 112]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 106] * kernel_shared_1[threadIdx_x // 7 * 192 + 1648]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 127] * kernel_shared_1[threadIdx_x // 7 * 192 + 19]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 127] * kernel_shared_1[threadIdx_x // 7 * 192 + 1555]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 127] * kernel_shared_1[threadIdx_x // 7 * 192 + 115]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 127] * kernel_shared_1[threadIdx_x // 7 * 192 + 1651]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 148] * kernel_shared_1[threadIdx_x // 7 * 192 + 22]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 148] * kernel_shared_1[threadIdx_x // 7 * 192 + 1558]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 148] * kernel_shared_1[threadIdx_x // 7 * 192 + 118]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 148] * kernel_shared_1[threadIdx_x // 7 * 192 + 1654]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 86] * kernel_shared_1[threadIdx_x // 7 * 192 + 14]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 86] * kernel_shared_1[threadIdx_x // 7 * 192 + 1550]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 86] * kernel_shared_1[threadIdx_x // 7 * 192 + 110]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 86] * kernel_shared_1[threadIdx_x // 7 * 192 + 1646]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 107] * kernel_shared_1[threadIdx_x // 7 * 192 + 17]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 107] * kernel_shared_1[threadIdx_x // 7 * 192 + 1553]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 107] * kernel_shared_1[threadIdx_x // 7 * 192 + 113]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 107] * kernel_shared_1[threadIdx_x // 7 * 192 + 1649]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 128] * kernel_shared_1[threadIdx_x // 7 * 192 + 20]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 128] * kernel_shared_1[threadIdx_x // 7 * 192 + 1556]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 128] * kernel_shared_1[threadIdx_x // 7 * 192 + 116]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 128] * kernel_shared_1[threadIdx_x // 7 * 192 + 1652]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 149] * kernel_shared_1[threadIdx_x // 7 * 192 + 23]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 149] * kernel_shared_1[threadIdx_x // 7 * 192 + 1559]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 149] * kernel_shared_1[threadIdx_x // 7 * 192 + 119]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 149] * kernel_shared_1[threadIdx_x // 7 * 192 + 1655]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 168] * kernel_shared_1[threadIdx_x // 7 * 192 + 24]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 168] * kernel_shared_1[threadIdx_x // 7 * 192 + 1560]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 168] * kernel_shared_1[threadIdx_x // 7 * 192 + 120]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 168] * kernel_shared_1[threadIdx_x // 7 * 192 + 1656]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 189] * kernel_shared_1[threadIdx_x // 7 * 192 + 27]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 189] * kernel_shared_1[threadIdx_x // 7 * 192 + 1563]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 189] * kernel_shared_1[threadIdx_x // 7 * 192 + 123]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 189] * kernel_shared_1[threadIdx_x // 7 * 192 + 1659]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 210] * kernel_shared_1[threadIdx_x // 7 * 192 + 30]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 210] * kernel_shared_1[threadIdx_x // 7 * 192 + 1566]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 210] * kernel_shared_1[threadIdx_x // 7 * 192 + 126]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 210] * kernel_shared_1[threadIdx_x // 7 * 192 + 1662]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 231] * kernel_shared_1[threadIdx_x // 7 * 192 + 33]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 231] * kernel_shared_1[threadIdx_x // 7 * 192 + 1569]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 231] * kernel_shared_1[threadIdx_x // 7 * 192 + 129]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 231] * kernel_shared_1[threadIdx_x // 7 * 192 + 1665]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 169] * kernel_shared_1[threadIdx_x // 7 * 192 + 25]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 169] * kernel_shared_1[threadIdx_x // 7 * 192 + 1561]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 169] * kernel_shared_1[threadIdx_x // 7 * 192 + 121]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 169] * kernel_shared_1[threadIdx_x // 7 * 192 + 1657]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 190] * kernel_shared_1[threadIdx_x // 7 * 192 + 28]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 190] * kernel_shared_1[threadIdx_x // 7 * 192 + 1564]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 190] * kernel_shared_1[threadIdx_x // 7 * 192 + 124]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 190] * kernel_shared_1[threadIdx_x // 7 * 192 + 1660]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 211] * kernel_shared_1[threadIdx_x // 7 * 192 + 31]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 211] * kernel_shared_1[threadIdx_x // 7 * 192 + 1567]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 211] * kernel_shared_1[threadIdx_x // 7 * 192 + 127]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 211] * kernel_shared_1[threadIdx_x // 7 * 192 + 1663]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 232] * kernel_shared_1[threadIdx_x // 7 * 192 + 34]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 232] * kernel_shared_1[threadIdx_x // 7 * 192 + 1570]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 232] * kernel_shared_1[threadIdx_x // 7 * 192 + 130]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 232] * kernel_shared_1[threadIdx_x // 7 * 192 + 1666]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 170] * kernel_shared_1[threadIdx_x // 7 * 192 + 26]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 170] * kernel_shared_1[threadIdx_x // 7 * 192 + 1562]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 170] * kernel_shared_1[threadIdx_x // 7 * 192 + 122]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 170] * kernel_shared_1[threadIdx_x // 7 * 192 + 1658]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 191] * kernel_shared_1[threadIdx_x // 7 * 192 + 29]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 191] * kernel_shared_1[threadIdx_x // 7 * 192 + 1565]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 191] * kernel_shared_1[threadIdx_x // 7 * 192 + 125]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 191] * kernel_shared_1[threadIdx_x // 7 * 192 + 1661]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 212] * kernel_shared_1[threadIdx_x // 7 * 192 + 32]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 212] * kernel_shared_1[threadIdx_x // 7 * 192 + 1568]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 212] * kernel_shared_1[threadIdx_x // 7 * 192 + 128]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 212] * kernel_shared_1[threadIdx_x // 7 * 192 + 1664]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 233] * kernel_shared_1[threadIdx_x // 7 * 192 + 35]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 233] * kernel_shared_1[threadIdx_x // 7 * 192 + 1571]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 233] * kernel_shared_1[threadIdx_x // 7 * 192 + 131]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 233] * kernel_shared_1[threadIdx_x // 7 * 192 + 1667]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 252] * kernel_shared_1[threadIdx_x // 7 * 192 + 36]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 252] * kernel_shared_1[threadIdx_x // 7 * 192 + 1572]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 252] * kernel_shared_1[threadIdx_x // 7 * 192 + 132]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 252] * kernel_shared_1[threadIdx_x // 7 * 192 + 1668]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 273] * kernel_shared_1[threadIdx_x // 7 * 192 + 39]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 273] * kernel_shared_1[threadIdx_x // 7 * 192 + 1575]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 273] * kernel_shared_1[threadIdx_x // 7 * 192 + 135]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 273] * kernel_shared_1[threadIdx_x // 7 * 192 + 1671]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 294] * kernel_shared_1[threadIdx_x // 7 * 192 + 42]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 294] * kernel_shared_1[threadIdx_x // 7 * 192 + 1578]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 294] * kernel_shared_1[threadIdx_x // 7 * 192 + 138]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 294] * kernel_shared_1[threadIdx_x // 7 * 192 + 1674]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 315] * kernel_shared_1[threadIdx_x // 7 * 192 + 45]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 315] * kernel_shared_1[threadIdx_x // 7 * 192 + 1581]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 315] * kernel_shared_1[threadIdx_x // 7 * 192 + 141]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 315] * kernel_shared_1[threadIdx_x // 7 * 192 + 1677]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 253] * kernel_shared_1[threadIdx_x // 7 * 192 + 37]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 253] * kernel_shared_1[threadIdx_x // 7 * 192 + 1573]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 253] * kernel_shared_1[threadIdx_x // 7 * 192 + 133]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 253] * kernel_shared_1[threadIdx_x // 7 * 192 + 1669]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 274] * kernel_shared_1[threadIdx_x // 7 * 192 + 40]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 274] * kernel_shared_1[threadIdx_x // 7 * 192 + 1576]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 274] * kernel_shared_1[threadIdx_x // 7 * 192 + 136]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 274] * kernel_shared_1[threadIdx_x // 7 * 192 + 1672]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 295] * kernel_shared_1[threadIdx_x // 7 * 192 + 43]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 295] * kernel_shared_1[threadIdx_x // 7 * 192 + 1579]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 295] * kernel_shared_1[threadIdx_x // 7 * 192 + 139]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 295] * kernel_shared_1[threadIdx_x // 7 * 192 + 1675]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 316] * kernel_shared_1[threadIdx_x // 7 * 192 + 46]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 316] * kernel_shared_1[threadIdx_x // 7 * 192 + 1582]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 316] * kernel_shared_1[threadIdx_x // 7 * 192 + 142]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 316] * kernel_shared_1[threadIdx_x // 7 * 192 + 1678]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 254] * kernel_shared_1[threadIdx_x // 7 * 192 + 38]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 254] * kernel_shared_1[threadIdx_x // 7 * 192 + 1574]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 254] * kernel_shared_1[threadIdx_x // 7 * 192 + 134]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 254] * kernel_shared_1[threadIdx_x // 7 * 192 + 1670]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 275] * kernel_shared_1[threadIdx_x // 7 * 192 + 41]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 275] * kernel_shared_1[threadIdx_x // 7 * 192 + 1577]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 275] * kernel_shared_1[threadIdx_x // 7 * 192 + 137]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 275] * kernel_shared_1[threadIdx_x // 7 * 192 + 1673]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 296] * kernel_shared_1[threadIdx_x // 7 * 192 + 44]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 296] * kernel_shared_1[threadIdx_x // 7 * 192 + 1580]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 296] * kernel_shared_1[threadIdx_x // 7 * 192 + 140]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 296] * kernel_shared_1[threadIdx_x // 7 * 192 + 1676]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 317] * kernel_shared_1[threadIdx_x // 7 * 192 + 47]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 317] * kernel_shared_1[threadIdx_x // 7 * 192 + 1583]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 317] * kernel_shared_1[threadIdx_x // 7 * 192 + 143]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 317] * kernel_shared_1[threadIdx_x // 7 * 192 + 1679]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 336] * kernel_shared_1[threadIdx_x // 7 * 192 + 48]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 336] * kernel_shared_1[threadIdx_x // 7 * 192 + 1584]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 336] * kernel_shared_1[threadIdx_x // 7 * 192 + 144]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 336] * kernel_shared_1[threadIdx_x // 7 * 192 + 1680]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 357] * kernel_shared_1[threadIdx_x // 7 * 192 + 51]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 357] * kernel_shared_1[threadIdx_x // 7 * 192 + 1587]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 357] * kernel_shared_1[threadIdx_x // 7 * 192 + 147]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 357] * kernel_shared_1[threadIdx_x // 7 * 192 + 1683]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 378] * kernel_shared_1[threadIdx_x // 7 * 192 + 54]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 378] * kernel_shared_1[threadIdx_x // 7 * 192 + 1590]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 378] * kernel_shared_1[threadIdx_x // 7 * 192 + 150]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 378] * kernel_shared_1[threadIdx_x // 7 * 192 + 1686]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 399] * kernel_shared_1[threadIdx_x // 7 * 192 + 57]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 399] * kernel_shared_1[threadIdx_x // 7 * 192 + 1593]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 399] * kernel_shared_1[threadIdx_x // 7 * 192 + 153]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 399] * kernel_shared_1[threadIdx_x // 7 * 192 + 1689]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 337] * kernel_shared_1[threadIdx_x // 7 * 192 + 49]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 337] * kernel_shared_1[threadIdx_x // 7 * 192 + 1585]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 337] * kernel_shared_1[threadIdx_x // 7 * 192 + 145]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 337] * kernel_shared_1[threadIdx_x // 7 * 192 + 1681]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 358] * kernel_shared_1[threadIdx_x // 7 * 192 + 52]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 358] * kernel_shared_1[threadIdx_x // 7 * 192 + 1588]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 358] * kernel_shared_1[threadIdx_x // 7 * 192 + 148]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 358] * kernel_shared_1[threadIdx_x // 7 * 192 + 1684]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 379] * kernel_shared_1[threadIdx_x // 7 * 192 + 55]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 379] * kernel_shared_1[threadIdx_x // 7 * 192 + 1591]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 379] * kernel_shared_1[threadIdx_x // 7 * 192 + 151]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 379] * kernel_shared_1[threadIdx_x // 7 * 192 + 1687]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 400] * kernel_shared_1[threadIdx_x // 7 * 192 + 58]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 400] * kernel_shared_1[threadIdx_x // 7 * 192 + 1594]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 400] * kernel_shared_1[threadIdx_x // 7 * 192 + 154]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 400] * kernel_shared_1[threadIdx_x // 7 * 192 + 1690]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 338] * kernel_shared_1[threadIdx_x // 7 * 192 + 50]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 338] * kernel_shared_1[threadIdx_x // 7 * 192 + 1586]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 338] * kernel_shared_1[threadIdx_x // 7 * 192 + 146]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 338] * kernel_shared_1[threadIdx_x // 7 * 192 + 1682]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 359] * kernel_shared_1[threadIdx_x // 7 * 192 + 53]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 359] * kernel_shared_1[threadIdx_x // 7 * 192 + 1589]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 359] * kernel_shared_1[threadIdx_x // 7 * 192 + 149]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 359] * kernel_shared_1[threadIdx_x // 7 * 192 + 1685]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 380] * kernel_shared_1[threadIdx_x // 7 * 192 + 56]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 380] * kernel_shared_1[threadIdx_x // 7 * 192 + 1592]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 380] * kernel_shared_1[threadIdx_x // 7 * 192 + 152]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 380] * kernel_shared_1[threadIdx_x // 7 * 192 + 1688]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 401] * kernel_shared_1[threadIdx_x // 7 * 192 + 59]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 401] * kernel_shared_1[threadIdx_x // 7 * 192 + 1595]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 401] * kernel_shared_1[threadIdx_x // 7 * 192 + 155]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 401] * kernel_shared_1[threadIdx_x // 7 * 192 + 1691]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 420] * kernel_shared_1[threadIdx_x // 7 * 192 + 60]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 420] * kernel_shared_1[threadIdx_x // 7 * 192 + 1596]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 420] * kernel_shared_1[threadIdx_x // 7 * 192 + 156]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 420] * kernel_shared_1[threadIdx_x // 7 * 192 + 1692]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 441] * kernel_shared_1[threadIdx_x // 7 * 192 + 63]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 441] * kernel_shared_1[threadIdx_x // 7 * 192 + 1599]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 441] * kernel_shared_1[threadIdx_x // 7 * 192 + 159]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 441] * kernel_shared_1[threadIdx_x // 7 * 192 + 1695]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 462] * kernel_shared_1[threadIdx_x // 7 * 192 + 66]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 462] * kernel_shared_1[threadIdx_x // 7 * 192 + 1602]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 462] * kernel_shared_1[threadIdx_x // 7 * 192 + 162]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 462] * kernel_shared_1[threadIdx_x // 7 * 192 + 1698]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 483] * kernel_shared_1[threadIdx_x // 7 * 192 + 69]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 483] * kernel_shared_1[threadIdx_x // 7 * 192 + 1605]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 483] * kernel_shared_1[threadIdx_x // 7 * 192 + 165]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 483] * kernel_shared_1[threadIdx_x // 7 * 192 + 1701]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 421] * kernel_shared_1[threadIdx_x // 7 * 192 + 61]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 421] * kernel_shared_1[threadIdx_x // 7 * 192 + 1597]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 421] * kernel_shared_1[threadIdx_x // 7 * 192 + 157]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 421] * kernel_shared_1[threadIdx_x // 7 * 192 + 1693]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 442] * kernel_shared_1[threadIdx_x // 7 * 192 + 64]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 442] * kernel_shared_1[threadIdx_x // 7 * 192 + 1600]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 442] * kernel_shared_1[threadIdx_x // 7 * 192 + 160]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 442] * kernel_shared_1[threadIdx_x // 7 * 192 + 1696]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 463] * kernel_shared_1[threadIdx_x // 7 * 192 + 67]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 463] * kernel_shared_1[threadIdx_x // 7 * 192 + 1603]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 463] * kernel_shared_1[threadIdx_x // 7 * 192 + 163]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 463] * kernel_shared_1[threadIdx_x // 7 * 192 + 1699]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 484] * kernel_shared_1[threadIdx_x // 7 * 192 + 70]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 484] * kernel_shared_1[threadIdx_x // 7 * 192 + 1606]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 484] * kernel_shared_1[threadIdx_x // 7 * 192 + 166]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 484] * kernel_shared_1[threadIdx_x // 7 * 192 + 1702]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 422] * kernel_shared_1[threadIdx_x // 7 * 192 + 62]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 422] * kernel_shared_1[threadIdx_x // 7 * 192 + 1598]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 422] * kernel_shared_1[threadIdx_x // 7 * 192 + 158]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 422] * kernel_shared_1[threadIdx_x // 7 * 192 + 1694]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 443] * kernel_shared_1[threadIdx_x // 7 * 192 + 65]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 443] * kernel_shared_1[threadIdx_x // 7 * 192 + 1601]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 443] * kernel_shared_1[threadIdx_x // 7 * 192 + 161]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 443] * kernel_shared_1[threadIdx_x // 7 * 192 + 1697]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 464] * kernel_shared_1[threadIdx_x // 7 * 192 + 68]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 464] * kernel_shared_1[threadIdx_x // 7 * 192 + 1604]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 464] * kernel_shared_1[threadIdx_x // 7 * 192 + 164]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 464] * kernel_shared_1[threadIdx_x // 7 * 192 + 1700]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 485] * kernel_shared_1[threadIdx_x // 7 * 192 + 71]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 485] * kernel_shared_1[threadIdx_x // 7 * 192 + 1607]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 485] * kernel_shared_1[threadIdx_x // 7 * 192 + 167]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 485] * kernel_shared_1[threadIdx_x // 7 * 192 + 1703]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 504] * kernel_shared_1[threadIdx_x // 7 * 192 + 72]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 504] * kernel_shared_1[threadIdx_x // 7 * 192 + 1608]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 504] * kernel_shared_1[threadIdx_x // 7 * 192 + 168]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 504] * kernel_shared_1[threadIdx_x // 7 * 192 + 1704]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 525] * kernel_shared_1[threadIdx_x // 7 * 192 + 75]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 525] * kernel_shared_1[threadIdx_x // 7 * 192 + 1611]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 525] * kernel_shared_1[threadIdx_x // 7 * 192 + 171]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 525] * kernel_shared_1[threadIdx_x // 7 * 192 + 1707]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 546] * kernel_shared_1[threadIdx_x // 7 * 192 + 78]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 546] * kernel_shared_1[threadIdx_x // 7 * 192 + 1614]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 546] * kernel_shared_1[threadIdx_x // 7 * 192 + 174]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 546] * kernel_shared_1[threadIdx_x // 7 * 192 + 1710]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 567] * kernel_shared_1[threadIdx_x // 7 * 192 + 81]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 567] * kernel_shared_1[threadIdx_x // 7 * 192 + 1617]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 567] * kernel_shared_1[threadIdx_x // 7 * 192 + 177]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 567] * kernel_shared_1[threadIdx_x // 7 * 192 + 1713]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 505] * kernel_shared_1[threadIdx_x // 7 * 192 + 73]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 505] * kernel_shared_1[threadIdx_x // 7 * 192 + 1609]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 505] * kernel_shared_1[threadIdx_x // 7 * 192 + 169]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 505] * kernel_shared_1[threadIdx_x // 7 * 192 + 1705]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 526] * kernel_shared_1[threadIdx_x // 7 * 192 + 76]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 526] * kernel_shared_1[threadIdx_x // 7 * 192 + 1612]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 526] * kernel_shared_1[threadIdx_x // 7 * 192 + 172]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 526] * kernel_shared_1[threadIdx_x // 7 * 192 + 1708]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 547] * kernel_shared_1[threadIdx_x // 7 * 192 + 79]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 547] * kernel_shared_1[threadIdx_x // 7 * 192 + 1615]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 547] * kernel_shared_1[threadIdx_x // 7 * 192 + 175]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 547] * kernel_shared_1[threadIdx_x // 7 * 192 + 1711]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 568] * kernel_shared_1[threadIdx_x // 7 * 192 + 82]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 568] * kernel_shared_1[threadIdx_x // 7 * 192 + 1618]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 568] * kernel_shared_1[threadIdx_x // 7 * 192 + 178]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 568] * kernel_shared_1[threadIdx_x // 7 * 192 + 1714]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 506] * kernel_shared_1[threadIdx_x // 7 * 192 + 74]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 506] * kernel_shared_1[threadIdx_x // 7 * 192 + 1610]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 506] * kernel_shared_1[threadIdx_x // 7 * 192 + 170]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 506] * kernel_shared_1[threadIdx_x // 7 * 192 + 1706]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 527] * kernel_shared_1[threadIdx_x // 7 * 192 + 77]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 527] * kernel_shared_1[threadIdx_x // 7 * 192 + 1613]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 527] * kernel_shared_1[threadIdx_x // 7 * 192 + 173]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 527] * kernel_shared_1[threadIdx_x // 7 * 192 + 1709]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 548] * kernel_shared_1[threadIdx_x // 7 * 192 + 80]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 548] * kernel_shared_1[threadIdx_x // 7 * 192 + 1616]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 548] * kernel_shared_1[threadIdx_x // 7 * 192 + 176]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 548] * kernel_shared_1[threadIdx_x // 7 * 192 + 1712]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 569] * kernel_shared_1[threadIdx_x // 7 * 192 + 83]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 569] * kernel_shared_1[threadIdx_x // 7 * 192 + 1619]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 569] * kernel_shared_1[threadIdx_x // 7 * 192 + 179]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 569] * kernel_shared_1[threadIdx_x // 7 * 192 + 1715]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 588] * kernel_shared_1[threadIdx_x // 7 * 192 + 84]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 588] * kernel_shared_1[threadIdx_x // 7 * 192 + 1620]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 588] * kernel_shared_1[threadIdx_x // 7 * 192 + 180]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 588] * kernel_shared_1[threadIdx_x // 7 * 192 + 1716]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 609] * kernel_shared_1[threadIdx_x // 7 * 192 + 87]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 609] * kernel_shared_1[threadIdx_x // 7 * 192 + 1623]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 609] * kernel_shared_1[threadIdx_x // 7 * 192 + 183]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 609] * kernel_shared_1[threadIdx_x // 7 * 192 + 1719]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 630] * kernel_shared_1[threadIdx_x // 7 * 192 + 90]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 630] * kernel_shared_1[threadIdx_x // 7 * 192 + 1626]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 630] * kernel_shared_1[threadIdx_x // 7 * 192 + 186]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 630] * kernel_shared_1[threadIdx_x // 7 * 192 + 1722]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 651] * kernel_shared_1[threadIdx_x // 7 * 192 + 93]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 651] * kernel_shared_1[threadIdx_x // 7 * 192 + 1629]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 651] * kernel_shared_1[threadIdx_x // 7 * 192 + 189]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 651] * kernel_shared_1[threadIdx_x // 7 * 192 + 1725]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 589] * kernel_shared_1[threadIdx_x // 7 * 192 + 85]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 589] * kernel_shared_1[threadIdx_x // 7 * 192 + 1621]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 589] * kernel_shared_1[threadIdx_x // 7 * 192 + 181]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 589] * kernel_shared_1[threadIdx_x // 7 * 192 + 1717]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 610] * kernel_shared_1[threadIdx_x // 7 * 192 + 88]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 610] * kernel_shared_1[threadIdx_x // 7 * 192 + 1624]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 610] * kernel_shared_1[threadIdx_x // 7 * 192 + 184]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 610] * kernel_shared_1[threadIdx_x // 7 * 192 + 1720]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 631] * kernel_shared_1[threadIdx_x // 7 * 192 + 91]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 631] * kernel_shared_1[threadIdx_x // 7 * 192 + 1627]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 631] * kernel_shared_1[threadIdx_x // 7 * 192 + 187]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 631] * kernel_shared_1[threadIdx_x // 7 * 192 + 1723]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 652] * kernel_shared_1[threadIdx_x // 7 * 192 + 94]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 652] * kernel_shared_1[threadIdx_x // 7 * 192 + 1630]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 652] * kernel_shared_1[threadIdx_x // 7 * 192 + 190]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 652] * kernel_shared_1[threadIdx_x // 7 * 192 + 1726]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 590] * kernel_shared_1[threadIdx_x // 7 * 192 + 86]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 590] * kernel_shared_1[threadIdx_x // 7 * 192 + 1622]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 590] * kernel_shared_1[threadIdx_x // 7 * 192 + 182]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 590] * kernel_shared_1[threadIdx_x // 7 * 192 + 1718]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 611] * kernel_shared_1[threadIdx_x // 7 * 192 + 89]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 611] * kernel_shared_1[threadIdx_x // 7 * 192 + 1625]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 611] * kernel_shared_1[threadIdx_x // 7 * 192 + 185]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 611] * kernel_shared_1[threadIdx_x // 7 * 192 + 1721]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 632] * kernel_shared_1[threadIdx_x // 7 * 192 + 92]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 632] * kernel_shared_1[threadIdx_x // 7 * 192 + 1628]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 632] * kernel_shared_1[threadIdx_x // 7 * 192 + 188]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 632] * kernel_shared_1[threadIdx_x // 7 * 192 + 1724]
+                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 653] * kernel_shared_1[threadIdx_x // 7 * 192 + 95]
+                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 653] * kernel_shared_1[threadIdx_x // 7 * 192 + 1631]
+                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 653] * kernel_shared_1[threadIdx_x // 7 * 192 + 191]
+                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 653] * kernel_shared_1[threadIdx_x // 7 * 192 + 1727]
+            for i1_inner in range(2):
+                compute_1 = T.Buffer((25088,), data=compute.data)
+                bias_1 = T.Buffer((512,), data=bias.data)
+                compute_1[blockIdx_x // 7 * 1568 + threadIdx_x // 7 * 98 + i1_inner * 49 + threadIdx_x % 7 * 7 + blockIdx_x % 7] = T.max(conv2d_nchw_1[i1_inner] + bias_1[blockIdx_x // 7 * 32 + threadIdx_x // 7 * 2 + i1_inner], T.float32(0))
+                compute_1[blockIdx_x // 7 * 1568 + threadIdx_x // 7 * 98 + i1_inner * 49 + threadIdx_x % 7 * 7 + blockIdx_x % 7 + 784] = T.max(conv2d_nchw_1[i1_inner + 2] + bias_1[blockIdx_x // 7 * 32 + threadIdx_x // 7 * 2 + i1_inner + 16], T.float32(0))
 
 
 
@@ -357,7 +837,7 @@ We build the binary and check its correctness and performance.
 
  .. code-block:: none
 
-    Execution time of this operator: 0.257 ms
+    Execution time of this operator: 0.419 ms
 
 
 
@@ -405,36 +885,36 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
     conv2d_nchw_nn_o_o_i, conv2d_nchw_nn_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_i, factor=1)
     conv2d_nchw_nn_o_o_o_i, conv2d_nchw_nn_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_i, factor=1)
     conv2d_nchw_nn_o_o_o_o, conv2d_nchw_nn_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_o_i, factor=1)
-    conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=1)
+    conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=2)
     conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=1)
-    conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=4)
-    conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=4)
+    conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=8)
+    conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=2)
     conv2d_nchw_yy_o_i, conv2d_nchw_yy_i = s[conv2d_nchw].split(conv2d_nchw_yy, factor=1)
     conv2d_nchw_yy_o_o_i, conv2d_nchw_yy_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_i, factor=1)
     conv2d_nchw_yy_o_o_o_i, conv2d_nchw_yy_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_i, factor=7)
     conv2d_nchw_yy_o_o_o_o, conv2d_nchw_yy_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_o_i, factor=1)
     conv2d_nchw_xx_o_i, conv2d_nchw_xx_i = s[conv2d_nchw].split(conv2d_nchw_xx, factor=1)
     conv2d_nchw_xx_o_o_i, conv2d_nchw_xx_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_i, factor=1)
-    conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, factor=7)
+    conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, factor=1)
     conv2d_nchw_xx_o_o_o_o, conv2d_nchw_xx_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_o_i, factor=1)
-    conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=2)
-    conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=2)
+    conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=4)
+    conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=8)
     conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=1)
-    conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=3)
-    conv2d_nchw_rx_o_i, conv2d_nchw_rx_i = s[conv2d_nchw].split(conv2d_nchw_rx, factor=3)
-    conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=1)
+    conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=1)
+    conv2d_nchw_rx_o_i, conv2d_nchw_rx_i = s[conv2d_nchw].split(conv2d_nchw_rx, factor=1)
+    conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=3)
     s[conv2d_nchw].reorder(conv2d_nchw_nn_o_o_o_o, conv2d_nchw_ff_o_o_o_o, conv2d_nchw_yy_o_o_o_o, conv2d_nchw_xx_o_o_o_o, conv2d_nchw_nn_o_o_o_i, conv2d_nchw_ff_o_o_o_i, conv2d_nchw_yy_o_o_o_i, conv2d_nchw_xx_o_o_o_i, conv2d_nchw_nn_o_o_i, conv2d_nchw_ff_o_o_i, conv2d_nchw_yy_o_o_i, conv2d_nchw_xx_o_o_i, conv2d_nchw_rc_o_o, conv2d_nchw_ry_o_o, conv2d_nchw_rx_o_o, conv2d_nchw_rc_o_i, conv2d_nchw_ry_o_i, conv2d_nchw_rx_o_i, conv2d_nchw_nn_o_i, conv2d_nchw_ff_o_i, conv2d_nchw_yy_o_i, conv2 [...]
     compute_i0_o_i, compute_i0_i = s[compute].split(compute_i0, factor=1)
     compute_i0_o_o_i, compute_i0_o_i = s[compute].split(compute_i0_o_i, factor=1)
     compute_i0_o_o_o, compute_i0_o_o_i = s[compute].split(compute_i0_o_o_i, factor=1)
-    compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=1)
-    compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=4)
-    compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=4)
+    compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=2)
+    compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=8)
+    compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=2)
     compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, factor=1)
     compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=7)
     compute_i2_o_o_o, compute_i2_o_o_i = s[compute].split(compute_i2_o_o_i, factor=1)
     compute_i3_o_i, compute_i3_i = s[compute].split(compute_i3, factor=1)
-    compute_i3_o_o_i, compute_i3_o_i = s[compute].split(compute_i3_o_i, factor=7)
+    compute_i3_o_o_i, compute_i3_o_i = s[compute].split(compute_i3_o_i, factor=1)
     compute_i3_o_o_o, compute_i3_o_o_i = s[compute].split(compute_i3_o_o_i, factor=1)
     s[compute].reorder(compute_i0_o_o_o, compute_i1_o_o_o, compute_i2_o_o_o, compute_i3_o_o_o, compute_i0_o_o_i, compute_i1_o_o_i, compute_i2_o_o_i, compute_i3_o_o_i, compute_i0_o_i, compute_i1_o_i, compute_i2_o_i, compute_i3_o_i, compute_i0_i, compute_i1_i, compute_i2_i, compute_i3_i)
     s[conv2d_nchw].compute_at(s[compute], compute_i3_o_i)
@@ -452,16 +932,16 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
     compute_i0_o_i_i1_o_i_fused_i2_o_i_fused_i3_o_i_fused = s[compute].fuse(compute_i0_o_i, compute_i1_o_i, compute_i2_o_i, compute_i3_o_i)
     s[compute].bind(compute_i0_o_i_i1_o_i_fused_i2_o_i_fused_i3_o_i_fused, te.thread_axis("threadIdx.x"))
     kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[kernel_shared].fuse(kernel_shared_ax0, kernel_shared_ax1, kernel_shared_ax2, kernel_shared_ax3)
-    kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=6)
+    kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
     s[kernel_shared].vectorize(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
-    kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=196)
+    kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=56)
     s[kernel_shared].bind(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
     pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[pad_temp_shared].fuse(pad_temp_shared_ax0, pad_temp_shared_ax1, pad_temp_shared_ax2, pad_temp_shared_ax3)
     pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
     s[pad_temp_shared].vectorize(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
-    pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=196)
+    pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=56)
     s[pad_temp_shared].bind(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis("threadIdx.x"))
-    s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "auto_unroll_max_step", 64)
+    s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "auto_unroll_max_step", 1024)
     s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "unroll_explicit", True)
 
     CUDA source code:
@@ -479,72 +959,477 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
       #define int64_t long long
       #define uint64_t unsigned long long
     #endif
-    extern "C" __global__ void __launch_bounds__(196) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
+    extern "C" __global__ void __launch_bounds__(56) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
       float conv2d_nchw[4];
-      __shared__ float pad_temp_shared[324];
-      __shared__ float kernel_shared[576];
+      __shared__ float pad_temp_shared[672];
+      __shared__ float kernel_shared[3072];
       conv2d_nchw[0] = 0.000000e+00f;
-      conv2d_nchw[1] = 0.000000e+00f;
       conv2d_nchw[2] = 0.000000e+00f;
+      conv2d_nchw[1] = 0.000000e+00f;
       conv2d_nchw[3] = 0.000000e+00f;
-      for (int rc_outer_outer = 0; rc_outer_outer < 128; ++rc_outer_outer) {
-        __syncthreads();
-        pad_temp_shared[((int)threadIdx.x)] = (((((9 <= (((int)threadIdx.x) % 81)) && ((((int)threadIdx.x) % 81) < 72)) && (1 <= (((int)threadIdx.x) % 9))) && ((((int)threadIdx.x) % 9) < 8)) ? data[(((((rc_outer_outer * 196) + ((((int)threadIdx.x) / 81) * 49)) + (((((int)threadIdx.x) % 81) / 9) * 7)) + (((int)threadIdx.x) % 9)) - 8)] : 0.000000e+00f);
-        if (((int)threadIdx.x) < 128) {
-          pad_temp_shared[(((int)threadIdx.x) + 196)] = (((((9 <= ((((int)threadIdx.x) + 34) % 81)) && (((((int)threadIdx.x) + 34) % 81) < 72)) && (1 <= ((((int)threadIdx.x) + 7) % 9))) && (((((int)threadIdx.x) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 196) / 81) * 49)) + ((((((int)threadIdx.x) + 34) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 7) % 9)) - 8)] : 0.000000e+00f);
-        }
-        if (((int)threadIdx.x) < 96) {
-          kernel_shared[(((int)threadIdx.x) * 6)] = kernel[((((((int)blockIdx.x) * 73728) + ((((int)threadIdx.x) / 6) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) % 6) * 6))];
-        }
-        if (((int)threadIdx.x) < 96) {
-          kernel_shared[((((int)threadIdx.x) * 6) + 1)] = kernel[(((((((int)blockIdx.x) * 73728) + ((((int)threadIdx.x) / 6) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) % 6) * 6)) + 1)];
-        }
-        if (((int)threadIdx.x) < 96) {
-          kernel_shared[((((int)threadIdx.x) * 6) + 2)] = kernel[(((((((int)blockIdx.x) * 73728) + ((((int)threadIdx.x) / 6) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) % 6) * 6)) + 2)];
-        }
-        if (((int)threadIdx.x) < 96) {
-          kernel_shared[((((int)threadIdx.x) * 6) + 3)] = kernel[(((((((int)blockIdx.x) * 73728) + ((((int)threadIdx.x) / 6) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) % 6) * 6)) + 3)];
-        }
-        if (((int)threadIdx.x) < 96) {
-          kernel_shared[((((int)threadIdx.x) * 6) + 4)] = kernel[(((((((int)blockIdx.x) * 73728) + ((((int)threadIdx.x) / 6) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) % 6) * 6)) + 4)];
-        }
-        if (((int)threadIdx.x) < 96) {
-          kernel_shared[((((int)threadIdx.x) * 6) + 5)] = kernel[(((((((int)blockIdx.x) * 73728) + ((((int)threadIdx.x) / 6) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) % 6) * 6)) + 5)];
-        }
-        __syncthreads();
-        for (int rc_outer_inner = 0; rc_outer_inner < 2; ++rc_outer_inner) {
-          for (int ry_outer_inner = 0; ry_outer_inner < 3; ++ry_outer_inner) {
-            conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3))]));
-            conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 144)]));
-            conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 288)]));
-            conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 432)]));
-            conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 1)]));
-            conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 145)]));
-            conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 289)]));
-            conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 433)]));
-            conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 2)]));
-            conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 146)]));
-            conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 290)]));
-            conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 434)]));
-            conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 9)]));
-            conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 153)]));
-            conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 297)]));
-            conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 441)]));
-            conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 10)]));
-            conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 154)]));
-            conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 298)]));
-            conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 442)]));
-            conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 11)]));
-            conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 155)]));
-            conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 299)]));
-            conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 443)]));
+      for (int rc_outer_outer = 0; rc_outer_outer < 16; ++rc_outer_outer) {
+        for (int ry_outer_outer = 0; ry_outer_outer < 3; ++ry_outer_outer) {
+          __syncthreads();
+          pad_temp_shared[((int)threadIdx.x)] = (((((1 <= (((((int)threadIdx.x) % 21) / 3) + ry_outer_outer)) && ((((((int)threadIdx.x) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((int)threadIdx.x) % 3)))) && (((((int)blockIdx.x) % 7) + (((int)threadIdx.x) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((int)threadIdx.x) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + (((int)threadIdx.x) % 3)) - 8)] : 0.000000e+00f);
+          pad_temp_shared[(((int)threadIdx.x) + 56)] = (((((1 <= ((((((int)threadIdx.x) + 14) % 21) / 3) + ry_outer_outer)) && (((((((int)threadIdx.x) + 14) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 2) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 56) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + ((((int)threadIdx.x) + 2) % 3)) [...]
+          pad_temp_shared[(((int)threadIdx.x) + 112)] = (((((1 <= ((((((int)threadIdx.x) + 7) % 21) / 3) + ry_outer_outer)) && (((((((int)threadIdx.x) + 7) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 1) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 1) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 112) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + ((((int)threadIdx.x) + 1) % 3)) [...]
+          pad_temp_shared[(((int)threadIdx.x) + 168)] = (((((1 <= (((((int)threadIdx.x) % 21) / 3) + ry_outer_outer)) && ((((((int)threadIdx.x) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((int)threadIdx.x) % 3)))) && (((((int)blockIdx.x) % 7) + (((int)threadIdx.x) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((int)threadIdx.x) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + (((int)threadIdx.x) % 3)) + 384)] : 0.000000e+00f);
+          pad_temp_shared[(((int)threadIdx.x) + 224)] = (((((1 <= ((((((int)threadIdx.x) + 14) % 21) / 3) + ry_outer_outer)) && (((((((int)threadIdx.x) + 14) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 2) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 224) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + ((((int)threadIdx.x) + 2) % 3 [...]
+          pad_temp_shared[(((int)threadIdx.x) + 280)] = (((((1 <= ((((((int)threadIdx.x) + 7) % 21) / 3) + ry_outer_outer)) && (((((((int)threadIdx.x) + 7) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 1) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 1) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 280) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + ((((int)threadIdx.x) + 1) % 3)) [...]
+          pad_temp_shared[(((int)threadIdx.x) + 336)] = (((((1 <= (((((int)threadIdx.x) % 21) / 3) + ry_outer_outer)) && ((((((int)threadIdx.x) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((int)threadIdx.x) % 3)))) && (((((int)blockIdx.x) % 7) + (((int)threadIdx.x) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((int)threadIdx.x) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + (((int)threadIdx.x) % 3)) + 776)] : 0.000000e+00f);
+          pad_temp_shared[(((int)threadIdx.x) + 392)] = (((((1 <= ((((((int)threadIdx.x) + 14) % 21) / 3) + ry_outer_outer)) && (((((((int)threadIdx.x) + 14) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 2) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 392) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + ((((int)threadIdx.x) + 2) % 3 [...]
+          pad_temp_shared[(((int)threadIdx.x) + 448)] = (((((1 <= ((((((int)threadIdx.x) + 7) % 21) / 3) + ry_outer_outer)) && (((((((int)threadIdx.x) + 7) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 1) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 1) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 448) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + ((((int)threadIdx.x) + 1) % 3)) [...]
+          pad_temp_shared[(((int)threadIdx.x) + 504)] = (((((1 <= (((((int)threadIdx.x) % 21) / 3) + ry_outer_outer)) && ((((((int)threadIdx.x) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + (((int)threadIdx.x) % 3)))) && (((((int)blockIdx.x) % 7) + (((int)threadIdx.x) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + ((((int)threadIdx.x) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + (((int)threadIdx.x) % 3)) + 1168)] : 0.000000e+00f);
+          pad_temp_shared[(((int)threadIdx.x) + 560)] = (((((1 <= ((((((int)threadIdx.x) + 14) % 21) / 3) + ry_outer_outer)) && (((((((int)threadIdx.x) + 14) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 2) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 2) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 560) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + ((((int)threadIdx.x) + 2) % 3 [...]
+          pad_temp_shared[(((int)threadIdx.x) + 616)] = (((((1 <= ((((((int)threadIdx.x) + 7) % 21) / 3) + ry_outer_outer)) && (((((((int)threadIdx.x) + 7) % 21) / 3) + ry_outer_outer) < 8)) && (1 <= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 1) % 3)))) && (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 1) % 3)) < 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 616) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + ((((int)threadIdx.x) + 1) % 3)) [...]
+          kernel_shared[((int)threadIdx.x)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 56)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 56) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 56) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 112)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 112) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 16) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 168)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 168) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 24) & 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 224)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 224) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 280)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 280) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 88) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 336)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 336) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 16) & 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 392)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 392) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 8) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 448)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 448) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 64) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 504)] = kernel[((((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 504) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 72)];
+          kernel_shared[(((int)threadIdx.x) + 560)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 560) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 80) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 616)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 616) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 40) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 672)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 32256)];
+          kernel_shared[(((int)threadIdx.x) + 728)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 728) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 56) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 784)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 784) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 16) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 840)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 840) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 24) & 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 896)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 896) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 952)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 952) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 88) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 1008)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1008) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 16) & 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 1064)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1064) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 8) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 1120)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1120) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 64) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 1176)] = kernel[((((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1176) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 72)];
+          kernel_shared[(((int)threadIdx.x) + 1232)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1232) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 80) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 1288)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1288) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 40) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 1344)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 64512)];
+          kernel_shared[(((int)threadIdx.x) + 1400)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1400) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 56) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 1456)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1456) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 16) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 1512)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1512) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 24) & 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 1568)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1568) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 1624)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1624) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 88) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 1680)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1680) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 16) & 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 1736)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1736) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 8) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 1792)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1792) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 64) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 1848)] = kernel[((((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1848) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 72)];
+          kernel_shared[(((int)threadIdx.x) + 1904)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1904) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 80) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 1960)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1960) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 40) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 2016)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 96768)];
+          kernel_shared[(((int)threadIdx.x) + 2072)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2072) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 56) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 2128)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2128) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 16) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 2184)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2184) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 24) & 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 2240)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2240) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 2296)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2296) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 88) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 2352)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2352) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 16) & 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 2408)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2408) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 8) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 2464)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2464) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 64) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 2520)] = kernel[((((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2520) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 72)];
+          kernel_shared[(((int)threadIdx.x) + 2576)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2576) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 80) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 2632)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2632) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 40) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 2688)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 129024)];
+          kernel_shared[(((int)threadIdx.x) + 2744)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2744) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 56) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 2800)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2800) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 16) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 2856)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2856) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 24) & 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 2912)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2912) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+          kernel_shared[(((int)threadIdx.x) + 2968)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2968) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 88) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+          if (((int)threadIdx.x) < 48) {
+            kernel_shared[(((int)threadIdx.x) + 3024)] = kernel[((((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 3024) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 144)];
           }
+          __syncthreads();
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) * 3)] * kernel_shared[((((int)threadIdx.x) / 7) * 192)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) * 3)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1536)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) * 3)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 96)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) * 3)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1632)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 21)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 3)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 21)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1539)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 21)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 99)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 21)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1635)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 42)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 6)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 42)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1542)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 42)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 102)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 42)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1638)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 9)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1545)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 105)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1641)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1537)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 97)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1633)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 22)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 4)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 22)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1540)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 22)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 100)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 22)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1636)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 43)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 7)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 43)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1543)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 43)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 103)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 43)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1639)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 10)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1546)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 106)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1642)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 2)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1538)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 98)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1634)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 23)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 5)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 23)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1541)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 23)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 101)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 23)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1637)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 44)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 8)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 44)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1544)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 44)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 104)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 44)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1640)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 11)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1547)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 107)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1643)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 84)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 12)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 84)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1548)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 84)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 108)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 84)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1644)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 105)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 15)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 105)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1551)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 105)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 111)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 105)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1647)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 18)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1554)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 114)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1650)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 147)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 21)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 147)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1557)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 147)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 117)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 147)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1653)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 85)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 13)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 85)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1549)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 85)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 109)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 85)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1645)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 106)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 16)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 106)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1552)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 106)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 112)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 106)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1648)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 19)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1555)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 115)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1651)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 148)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 22)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 148)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1558)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 148)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 118)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 148)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1654)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 86)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 14)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 86)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1550)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 86)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 110)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 86)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1646)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 107)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 17)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 107)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1553)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 107)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 113)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 107)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1649)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 20)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1556)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 116)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1652)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 149)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 23)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 149)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1559)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 149)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 119)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 149)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1655)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 168)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 24)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 168)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1560)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 168)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 120)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 168)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1656)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 27)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1563)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 123)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1659)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 210)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 30)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 210)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1566)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 210)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 126)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 210)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1662)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 231)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 33)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 231)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1569)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 231)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 129)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 231)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1665)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 169)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 25)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 169)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1561)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 169)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 121)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 169)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1657)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 28)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1564)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 124)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1660)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 211)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 31)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 211)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1567)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 211)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 127)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 211)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1663)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 232)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 34)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 232)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1570)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 232)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 130)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 232)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1666)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 170)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 26)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 170)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1562)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 170)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 122)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 170)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1658)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 29)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1565)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 125)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1661)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 212)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 32)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 212)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1568)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 212)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 128)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 212)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1664)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 233)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 35)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 233)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1571)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 233)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 131)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 233)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1667)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 36)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1572)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 132)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1668)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 273)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 39)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 273)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1575)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 273)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 135)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 273)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1671)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 294)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 42)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 294)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1578)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 294)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 138)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 294)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1674)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 315)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 45)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 315)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1581)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 315)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 141)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 315)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1677)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 37)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1573)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 133)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1669)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 274)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 40)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 274)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1576)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 274)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 136)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 274)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1672)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 295)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 43)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 295)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1579)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 295)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 139)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 295)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1675)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 316)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 46)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 316)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1582)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 316)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 142)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 316)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1678)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 38)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1574)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 134)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1670)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 275)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 41)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 275)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1577)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 275)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 137)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 275)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1673)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 296)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 44)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 296)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1580)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 296)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 140)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 296)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1676)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 317)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 47)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 317)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1583)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 317)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 143)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 317)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1679)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 336)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 48)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 336)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1584)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 336)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 144)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 336)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1680)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 357)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 51)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 357)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1587)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 357)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 147)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 357)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1683)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 378)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 54)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 378)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1590)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 378)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 150)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 378)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1686)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 399)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 57)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 399)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1593)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 399)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 153)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 399)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1689)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 337)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 49)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 337)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1585)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 337)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 145)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 337)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1681)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 358)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 52)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 358)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1588)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 358)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 148)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 358)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1684)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 379)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 55)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 379)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1591)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 379)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 151)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 379)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1687)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 400)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 58)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 400)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1594)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 400)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 154)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 400)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1690)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 338)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 50)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 338)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1586)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 338)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 146)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 338)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1682)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 359)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 53)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 359)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1589)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 359)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 149)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 359)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1685)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 380)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 56)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 380)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1592)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 380)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 152)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 380)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1688)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 401)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 59)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 401)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1595)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 401)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 155)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 401)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1691)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 420)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 60)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 420)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1596)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 420)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 156)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 420)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1692)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 441)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 63)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 441)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1599)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 441)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 159)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 441)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1695)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 462)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 66)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 462)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1602)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 462)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 162)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 462)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1698)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 483)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 69)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 483)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1605)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 483)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 165)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 483)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1701)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 421)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 61)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 421)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1597)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 421)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 157)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 421)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1693)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 442)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 64)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 442)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1600)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 442)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 160)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 442)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1696)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 463)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 67)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 463)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1603)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 463)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 163)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 463)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1699)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 484)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 70)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 484)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1606)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 484)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 166)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 484)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1702)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 422)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 62)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 422)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1598)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 422)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 158)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 422)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1694)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 443)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 65)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 443)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1601)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 443)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 161)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 443)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1697)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 464)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 68)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 464)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1604)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 464)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 164)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 464)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1700)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 485)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 71)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 485)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1607)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 485)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 167)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 485)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1703)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 504)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 72)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 504)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1608)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 504)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 168)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 504)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1704)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 525)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 75)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 525)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1611)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 525)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 171)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 525)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1707)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 546)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 78)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 546)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1614)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 546)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 174)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 546)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1710)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 567)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 81)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 567)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1617)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 567)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 177)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 567)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1713)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 505)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 73)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 505)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1609)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 505)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 169)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 505)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1705)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 526)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 76)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 526)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1612)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 526)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 172)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 526)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1708)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 547)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 79)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 547)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1615)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 547)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 175)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 547)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1711)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 568)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 82)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 568)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1618)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 568)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 178)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 568)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1714)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 506)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 74)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 506)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1610)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 506)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 170)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 506)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1706)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 527)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 77)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 527)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1613)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 527)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 173)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 527)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1709)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 548)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 80)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 548)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1616)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 548)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 176)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 548)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1712)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 569)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 83)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 569)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1619)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 569)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 179)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 569)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1715)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 588)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 84)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 588)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1620)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 588)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 180)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 588)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1716)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 609)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 87)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 609)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1623)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 609)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 183)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 609)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1719)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 630)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 90)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 630)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1626)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 630)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 186)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 630)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1722)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 651)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 93)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 651)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1629)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 651)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 189)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 651)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1725)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 589)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 85)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 589)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1621)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 589)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 181)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 589)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1717)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 610)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 88)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 610)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1624)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 610)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 184)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 610)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1720)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 631)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 91)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 631)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1627)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 631)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 187)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 631)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1723)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 652)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 94)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 652)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1630)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 652)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 190)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 652)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1726)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 590)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 86)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 590)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1622)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 590)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 182)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 590)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1718)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 611)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 89)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 611)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1625)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 611)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 185)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 611)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1721)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 632)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 92)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 632)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1628)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 632)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 188)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 632)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1724)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 653)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 95)]));
+          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 653)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1631)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 653)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 191)]));
+          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 653)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1727)]));
         }
       }
-      compute[((((int)blockIdx.x) * 784) + ((int)threadIdx.x))] = max((conv2d_nchw[0] + bias[((((int)blockIdx.x) * 16) + (((int)threadIdx.x) / 49))]), 0.000000e+00f);
-      compute[(((((int)blockIdx.x) * 784) + ((int)threadIdx.x)) + 196)] = max((conv2d_nchw[1] + bias[(((((int)blockIdx.x) * 16) + (((int)threadIdx.x) / 49)) + 4)]), 0.000000e+00f);
-      compute[(((((int)blockIdx.x) * 784) + ((int)threadIdx.x)) + 392)] = max((conv2d_nchw[2] + bias[(((((int)blockIdx.x) * 16) + (((int)threadIdx.x) / 49)) + 8)]), 0.000000e+00f);
-      compute[(((((int)blockIdx.x) * 784) + ((int)threadIdx.x)) + 588)] = max((conv2d_nchw[3] + bias[(((((int)blockIdx.x) * 16) + (((int)threadIdx.x) / 49)) + 12)]), 0.000000e+00f);
+      for (int i1_inner = 0; i1_inner < 2; ++i1_inner) {
+        compute[((((((((int)blockIdx.x) / 7) * 1568) + ((((int)threadIdx.x) / 7) * 98)) + (i1_inner * 49)) + ((((int)threadIdx.x) % 7) * 7)) + (((int)blockIdx.x) % 7))] = max((conv2d_nchw[i1_inner] + bias[((((((int)blockIdx.x) / 7) * 32) + ((((int)threadIdx.x) / 7) * 2)) + i1_inner)]), 0.000000e+00f);
+        compute[(((((((((int)blockIdx.x) / 7) * 1568) + ((((int)threadIdx.x) / 7) * 98)) + (i1_inner * 49)) + ((((int)threadIdx.x) % 7) * 7)) + (((int)blockIdx.x) % 7)) + 784)] = max((conv2d_nchw[(i1_inner + 2)] + bias[(((((((int)blockIdx.x) / 7) * 32) + ((((int)threadIdx.x) / 7) * 2)) + i1_inner) + 16)]), 0.000000e+00f);
+      }
     }
 
 
@@ -603,7 +1488,7 @@ In the example below we resume the status and do more 5 trials.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 6 minutes  10.669 seconds)
+   **Total running time of the script:** ( 6 minutes  18.523 seconds)
 
 
 .. _sphx_glr_download_how_to_tune_with_autoscheduler_tune_conv2d_layer_cuda.py:
diff --git a/docs/_sources/how_to/tune_with_autoscheduler/tune_network_cuda.rst.txt b/docs/_sources/how_to/tune_with_autoscheduler/tune_network_cuda.rst.txt
index 93b5dce11d..d1e8a856cf 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_network_cuda.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_network_cuda.rst.txt
@@ -647,7 +647,7 @@ so we can read the log file and load the best schedules.
     Evaluate inference time cost...
     Execution time summary:
      mean (ms)   median (ms)    max (ms)     min (ms)     std (ms)  
-       8.1053       8.1077       8.1103       8.0979       0.0053   
+       8.1199       8.1181       8.1249       8.1165       0.0036   
                
 
 
@@ -675,7 +675,7 @@ Other Tips
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  12.321 seconds)
+   **Total running time of the script:** ( 1 minutes  13.936 seconds)
 
 
 .. _sphx_glr_download_how_to_tune_with_autoscheduler_tune_network_cuda.py:
diff --git a/docs/_sources/how_to/tune_with_autoscheduler/tune_network_x86.rst.txt b/docs/_sources/how_to/tune_with_autoscheduler/tune_network_x86.rst.txt
index c986092127..9acc42933b 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_network_x86.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_network_x86.rst.txt
@@ -666,7 +666,7 @@ so we can read the log file and load the best schedules.
     Evaluate inference time cost...
     Execution time summary:
      mean (ms)   median (ms)    max (ms)     min (ms)     std (ms)  
-      758.2928     758.2217     758.4373     758.2193      0.1022   
+      763.8272     765.0162     766.1968     760.2686      2.5621   
                
 
 
@@ -694,7 +694,7 @@ Other Tips
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  43.939 seconds)
+   **Total running time of the script:** ( 1 minutes  45.674 seconds)
 
 
 .. _sphx_glr_download_how_to_tune_with_autoscheduler_tune_network_x86.py:
diff --git a/docs/_sources/how_to/tune_with_autoscheduler/tune_sparse_x86.rst.txt b/docs/_sources/how_to/tune_with_autoscheduler/tune_sparse_x86.rst.txt
index e62c3893dd..219536fe0e 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_sparse_x86.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_sparse_x86.rst.txt
@@ -389,26 +389,117 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
         @T.prim_func
         def main(placeholder: T.Buffer((128, 256), "float32"), placeholder_1: T.Buffer((4916, 16, 1), "float32"), placeholder_2: T.Buffer((4916,), "int32"), placeholder_3: T.Buffer((33,), "int32"), placeholder_4: T.Buffer((128, 512), "float32"), compute: T.Buffer((128, 512), "float32")):
             T.func_attr({"from_legacy_te_schedule": T.bool(True), "global_symbol": "main", "tir.noalias": T.bool(True)})
-            for i0_outer_i1_outer_fused in T.parallel(32):
-                compute_1 = T.allocate([2048], "float32", "global")
-                compute_2 = T.Buffer((2048,), data=compute_1)
-                for i_outer_inner, nb_j_inner in T.grid(4, 2):
-                    for i_inner_init, j_init in T.grid(16, 16):
-                        compute_2[i_outer_inner * 512 + i_inner_init * 32 + nb_j_inner * 16 + j_init] = T.float32(0)
-                    for elem_idx, i_inner, j in T.grid(T.Let(placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1], where={cse_var_1: i0_outer_i1_outer_fused % 16 * 2 + nb_j_inner}), 16, 16):
-                        cse_var_1 = T.int32()
+            for i0_outer in T.parallel(32):
+                compute_1 = T.allocate([32], "float32", "global")
+                for i1_outer in range(64):
+                    cse_var_1: T.int32 = i1_outer // 2
+                    compute_2 = T.Buffer((32,), data=compute_1)
+                    compute_2[0] = T.float32(0)
+                    compute_2[1] = T.float32(0)
+                    compute_2[2] = T.float32(0)
+                    compute_2[3] = T.float32(0)
+                    compute_2[4] = T.float32(0)
+                    compute_2[5] = T.float32(0)
+                    compute_2[6] = T.float32(0)
+                    compute_2[7] = T.float32(0)
+                    compute_2[8] = T.float32(0)
+                    compute_2[9] = T.float32(0)
+                    compute_2[10] = T.float32(0)
+                    compute_2[11] = T.float32(0)
+                    compute_2[12] = T.float32(0)
+                    compute_2[13] = T.float32(0)
+                    compute_2[14] = T.float32(0)
+                    compute_2[15] = T.float32(0)
+                    compute_2[16] = T.float32(0)
+                    compute_2[17] = T.float32(0)
+                    compute_2[18] = T.float32(0)
+                    compute_2[19] = T.float32(0)
+                    compute_2[20] = T.float32(0)
+                    compute_2[21] = T.float32(0)
+                    compute_2[22] = T.float32(0)
+                    compute_2[23] = T.float32(0)
+                    compute_2[24] = T.float32(0)
+                    compute_2[25] = T.float32(0)
+                    compute_2[26] = T.float32(0)
+                    compute_2[27] = T.float32(0)
+                    compute_2[28] = T.float32(0)
+                    compute_2[29] = T.float32(0)
+                    compute_2[30] = T.float32(0)
+                    compute_2[31] = T.float32(0)
+                    for elem_idx in range(placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
                         placeholder_5 = T.Buffer((33,), "int32", data=placeholder_3.data)
-                        cse_var_3: T.int32 = i0_outer_i1_outer_fused % 16 * 2 + nb_j_inner
-                        cse_var_2: T.int32 = i_outer_inner * 512 + i_inner * 32 + nb_j_inner * 16 + j
                         placeholder_6 = T.Buffer((78656,), data=placeholder_1.data)
                         placeholder_7 = T.Buffer((32768,), data=placeholder.data)
                         placeholder_8 = T.Buffer((4916,), "int32", data=placeholder_2.data)
-                        compute_2[cse_var_2] = compute_2[cse_var_2] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + j] * T.max(placeholder_7[i0_outer_i1_outer_fused // 16 * 16384 + i_outer_inner * 4096 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
-                for i0_inner, i1_inner in T.grid(64, 32):
-                    cse_var_4: T.int32 = i0_outer_i1_outer_fused // 16 * 32768 + i0_inner * 512 + i0_outer_i1_outer_fused % 16 * 32 + i1_inner
-                    compute_3 = T.Buffer((65536,), data=compute.data)
-                    placeholder_5 = T.Buffer((65536,), data=placeholder_4.data)
-                    compute_3[cse_var_4] = T.max(compute_2[i0_inner * 32 + i1_inner] + placeholder_5[cse_var_4], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[0] = compute_2[0] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[1] = compute_2[1] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 1] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[2] = compute_2[2] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 2] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[3] = compute_2[3] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 3] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[4] = compute_2[4] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 4] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[5] = compute_2[5] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 5] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[6] = compute_2[6] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 6] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[7] = compute_2[7] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 7] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[8] = compute_2[8] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[9] = compute_2[9] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 1] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[10] = compute_2[10] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 2] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[11] = compute_2[11] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 3] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[12] = compute_2[12] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 4] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[13] = compute_2[13] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 5] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[14] = compute_2[14] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 6] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[15] = compute_2[15] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 7] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[16] = compute_2[16] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[17] = compute_2[17] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 1] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[18] = compute_2[18] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 2] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[19] = compute_2[19] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 3] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[20] = compute_2[20] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 4] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[21] = compute_2[21] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 5] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[22] = compute_2[22] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 6] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[23] = compute_2[23] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 7] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[24] = compute_2[24] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[25] = compute_2[25] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 1] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[26] = compute_2[26] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 2] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[27] = compute_2[27] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 3] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[28] = compute_2[28] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 4] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[29] = compute_2[29] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 5] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[30] = compute_2[30] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 6] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+                        if T.likely(elem_idx < placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                            compute_2[31] = compute_2[31] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 7] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+                    for i0_inner in range(4):
+                        cse_var_2: T.int32 = i0_outer * 2048 + i0_inner * 512 + i1_outer * 8
+                        compute_3 = T.Buffer((65536,), data=compute.data)
+                        placeholder_5 = T.Buffer((65536,), data=placeholder_4.data)
+                        compute_3[cse_var_2:cse_var_2 + 8] = T.max(compute_2[i0_inner * 8:i0_inner * 8 + 8] + placeholder_5[cse_var_2:cse_var_2 + 8], T.Broadcast(T.float32(0), 8))
 
 
 
@@ -458,7 +549,7 @@ We build the binary and check its correctness and performance.
 
  .. code-block:: none
 
-    Execution time of this operator: 1.519 ms
+    Execution time of this operator: 1.841 ms
 
 
 
diff --git a/docs/_sources/how_to/tune_with_autotvm/sg_execution_times.rst.txt b/docs/_sources/how_to/tune_with_autotvm/sg_execution_times.rst.txt
index 9201e88272..d790391a04 100644
--- a/docs/_sources/how_to/tune_with_autotvm/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/tune_with_autotvm/sg_execution_times.rst.txt
@@ -5,16 +5,16 @@
 
 Computation times
 =================
-**00:49.351** total execution time for **how_to_tune_with_autotvm** files:
+**00:38.513** total execution time for **how_to_tune_with_autotvm** files:
 
 +--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_conv2d_cuda.py` (``tune_conv2d_cuda.py``)           | 00:49.318 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_conv2d_cuda.py` (``tune_conv2d_cuda.py``)           | 00:38.479 | 0.0 MB |
 +--------------------------------------------------------------------------------------------------+-----------+--------+
 | :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_x86.py` (``tune_relay_x86.py``)               | 00:00.020 | 0.0 MB |
 +--------------------------------------------------------------------------------------------------+-----------+--------+
 | :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_cuda.py` (``tune_relay_cuda.py``)             | 00:00.005 | 0.0 MB |
 +--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_arm.py` (``tune_relay_arm.py``)               | 00:00.004 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_arm.py` (``tune_relay_arm.py``)               | 00:00.005 | 0.0 MB |
 +--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_mobile_gpu.py` (``tune_relay_mobile_gpu.py``) | 00:00.004 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_mobile_gpu.py` (``tune_relay_mobile_gpu.py``) | 00:00.005 | 0.0 MB |
 +--------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/tune_with_autotvm/tune_conv2d_cuda.rst.txt b/docs/_sources/how_to/tune_with_autotvm/tune_conv2d_cuda.rst.txt
index 72e6414b85..d90bdda9d5 100644
--- a/docs/_sources/how_to/tune_with_autotvm/tune_conv2d_cuda.rst.txt
+++ b/docs/_sources/how_to/tune_with_autotvm/tune_conv2d_cuda.rst.txt
@@ -268,26 +268,131 @@ for this template
     waiting for device...
     device available
     Get devices for measurement successfully!
-    No: 1   GFLOPS: 7.51/7.51       result: MeasureResult(costs=(0.0308341445,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.836233854293823, timestamp=1680177386.5471036)        [('tile_f', [-1, 32, 1, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3552261
-    No: 2   GFLOPS: 0.00/7.51       result: Traceback (most recent call last):
-      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 142, in build
-        res = future.result()
-      File "/usr/lib/python3.7/concurrent/futures/_base.py", line 435, in result
-        return self.__get_result()
-      File "/usr/lib/python3.7/concurrent/futures/_base.py", line 384, in __get_result
-        raise self._exception
-      File "/usr/lib/python3.7/concurrent/futures/thread.py", line 57, in run
-        result = self.fn(*self.args, **self.kwargs)
-      File "/workspace/python/tvm/contrib/popen_pool.py", line 432, in <lambda>
-        worker = lambda *args: self._worker_run(*args)
-      File "/workspace/python/tvm/contrib/popen_pool.py", line 401, in _worker_run
-        return proc.recv()
-      File "/workspace/python/tvm/contrib/popen_pool.py", line 309, in recv
-        raise TimeoutError()
-    TimeoutError
+    No: 1   GFLOPS: 345.53/345.53   result: MeasureResult(costs=(0.00066998061875,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.27738881111145, timestamp=1680189104.0109434)     [('tile_f', [-1, 1, 16, 4]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 8]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2033146
+    No: 2   GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
+      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
+        func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
+      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
+        func = build(s, args, target=target, runtime=runtime)
+      File "/workspace/python/tvm/driver/build_module.py", line 227, in build
+        input_mod = lower(inputs, args, name=name, binds=binds)
+      File "/workspace/python/tvm/driver/build_module.py", line 134, in lower
+        return ffi.lower_schedule(inp, args, name, binds, simple_mode)
+      File "tvm/_ffi/_cython/./packed_func.pxi", line 331, in tvm._ffi._cy3.core.PackedFuncBase.__call__
+      File "tvm/_ffi/_cython/./packed_func.pxi", line 276, in tvm._ffi._cy3.core.FuncCall
+      File "tvm/_ffi/_cython/./base.pxi", line 181, in tvm._ffi._cy3.core.CHECK_CALL
+    tvm._ffi.base.TVMError: Traceback (most recent call last):
+      24: TVMFuncCall
+            at ../src/runtime/c_runtime_api.cc:477
+      23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+            at ../include/tvm/runtime/packed_func.h:1217
+      22: Call
+            at ../include/tvm/runtime/packed_func.h:1213
+      21: operator()
+            at ../include/tvm/runtime/packed_func.h:1734
+      20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
+            at ../include/tvm/runtime/packed_func.h:1674
+      19: run<>
+            at ../include/tvm/runtime/packed_func.h:1634
+      18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1634
+      17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1634
+      16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1634
+      15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1634
+      14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1649
+      13: operator()
+            at ../src/driver/driver_api.cc:402
+      12: tvm::LowerSchedule(tvm::te::Schedule, tvm::runtime::Array<tvm::runtime::ObjectRef, void> const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::unordered_map<tvm::te::Tensor, tvm::tir::Buffer, std::hash<tvm::te::Tensor>, std::equal_to<tvm::te::Tensor>, std::allocator<std::pair<tvm::te::Tensor const, tvm::tir::Buffer> > > const&, tvm::GlobalVarSupply, bool)
+            at ../src/driver/driver_api.cc:388
+      11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+            at ../src/driver/driver_api.cc:283
+      10: tvm::transform::Pass::operator()(tvm::IRModule) const
+            at ../src/ir/transform.cc:258
+      9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+            at ../src/ir/transform.cc:274
+      8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+            at ../src/ir/transform.cc:451
+      7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+            at ../src/ir/transform.cc:274
+      6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+            at ../src/tir/ir/transform.cc:100
+      5: tvm::runtime::TypedPackedFunc<tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)>::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
+            at ../include/tvm/runtime/packed_func.h:1753
+      4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher<tvm::tir::PrimFunc>::run<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::runtime::PackedFunc const&, tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&)
+            at ../include/tvm/runtime/packed_func.h:1697
+      3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&) const
+            at ../include/tvm/runtime/packed_func.h:1621
+      2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+            at ../include/tvm/runtime/packed_func.h:1217
+      1: Call
+            at ../include/tvm/runtime/packed_func.h:1213
+      0: operator()
+            at ../src/runtime/c_runtime_api.cc:534
+      File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
+      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
+        raise InstantiationError("Skipped because of invalid gpu kernel")
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel
 
-            [('tile_f', [-1, 16, 2, 4]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 64]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1902012
-    No: 3   GFLOPS: 0.00/7.51       result: Traceback (most recent call last):
+    Traceback (most recent call last):
+      24: TVMFuncCall
+            at ../src/runtime/c_runtime_api.cc:477
+      23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+            at ../include/tvm/runtime/packed_func.h:1217
+      22: Call
+            at ../include/tvm/runtime/packed_func.h:1213
+      21: operator()
+            at ../include/tvm/runtime/packed_func.h:1734
+      20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
+            at ../include/tvm/runtime/packed_func.h:1674
+      19: run<>
+            at ../include/tvm/runtime/packed_func.h:1634
+      18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1634
+      17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1634
+      16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1634
+      15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1634
+      14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1649
+      13: operator()
+            at ../src/driver/driver_api.cc:402
+      12: tvm::LowerSchedule(tvm::te::Schedule, tvm::runtime::Array<tvm::runtime::ObjectRef, void> const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::unordered_map<tvm::te::Tensor, tvm::tir::Buffer, std::hash<tvm::te::Tensor>, std::equal_to<tvm::te::Tensor>, std::allocator<std::pair<tvm::te::Tensor const, tvm::tir::Buffer> > > const&, tvm::GlobalVarSupply, bool)
+            at ../src/driver/driver_api.cc:388
+      11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+            at ../src/driver/driver_api.cc:283
+      10: tvm::transform::Pass::operator()(tvm::IRModule) const
+            at ../src/ir/transform.cc:258
+      9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+            at ../src/ir/transform.cc:274
+      8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+            at ../src/ir/transform.cc:451
+      7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+            at ../src/ir/transform.cc:274
+      6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+            at ../src/tir/ir/transform.cc:100
+      5: tvm::runtime::TypedPackedFunc<tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)>::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
+            at ../include/tvm/runtime/packed_func.h:1753
+      4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher<tvm::tir::PrimFunc>::run<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::runtime::PackedFunc const&, tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&)
+            at ../include/tvm/runtime/packed_func.h:1697
+      3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&) const
+            at ../include/tvm/runtime/packed_func.h:1621
+      2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+            at ../include/tvm/runtime/packed_func.h:1217
+      1: Call
+            at ../include/tvm/runtime/packed_func.h:1213
+      0: operator()
+            at ../src/runtime/c_runtime_api.cc:534
+      File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
+      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
+        raise InstantiationError("Skipped because of invalid gpu kernel")
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 64, 4, 2]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 8, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2816738
+    No: 3   GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -409,26 +514,131 @@ for this template
       File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
-    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 4, 2, 4]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 16, 32]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,736010
-    No: 4   GFLOPS: 0.00/7.51       result: Traceback (most recent call last):
-      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 142, in build
-        res = future.result()
-      File "/usr/lib/python3.7/concurrent/futures/_base.py", line 435, in result
-        return self.__get_result()
-      File "/usr/lib/python3.7/concurrent/futures/_base.py", line 384, in __get_result
-        raise self._exception
-      File "/usr/lib/python3.7/concurrent/futures/thread.py", line 57, in run
-        result = self.fn(*self.args, **self.kwargs)
-      File "/workspace/python/tvm/contrib/popen_pool.py", line 432, in <lambda>
-        worker = lambda *args: self._worker_run(*args)
-      File "/workspace/python/tvm/contrib/popen_pool.py", line 401, in _worker_run
-        return proc.recv()
-      File "/workspace/python/tvm/contrib/popen_pool.py", line 309, in recv
-        raise TimeoutError()
-    TimeoutError
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 8, 8, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 128, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1574570
+    No: 4   GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
+      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
+        func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
+      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
+        func = build(s, args, target=target, runtime=runtime)
+      File "/workspace/python/tvm/driver/build_module.py", line 227, in build
+        input_mod = lower(inputs, args, name=name, binds=binds)
+      File "/workspace/python/tvm/driver/build_module.py", line 134, in lower
+        return ffi.lower_schedule(inp, args, name, binds, simple_mode)
+      File "tvm/_ffi/_cython/./packed_func.pxi", line 331, in tvm._ffi._cy3.core.PackedFuncBase.__call__
+      File "tvm/_ffi/_cython/./packed_func.pxi", line 276, in tvm._ffi._cy3.core.FuncCall
+      File "tvm/_ffi/_cython/./base.pxi", line 181, in tvm._ffi._cy3.core.CHECK_CALL
+    tvm._ffi.base.TVMError: Traceback (most recent call last):
+      24: TVMFuncCall
+            at ../src/runtime/c_runtime_api.cc:477
+      23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+            at ../include/tvm/runtime/packed_func.h:1217
+      22: Call
+            at ../include/tvm/runtime/packed_func.h:1213
+      21: operator()
+            at ../include/tvm/runtime/packed_func.h:1734
+      20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
+            at ../include/tvm/runtime/packed_func.h:1674
+      19: run<>
+            at ../include/tvm/runtime/packed_func.h:1634
+      18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1634
+      17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1634
+      16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1634
+      15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1634
+      14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1649
+      13: operator()
+            at ../src/driver/driver_api.cc:402
+      12: tvm::LowerSchedule(tvm::te::Schedule, tvm::runtime::Array<tvm::runtime::ObjectRef, void> const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::unordered_map<tvm::te::Tensor, tvm::tir::Buffer, std::hash<tvm::te::Tensor>, std::equal_to<tvm::te::Tensor>, std::allocator<std::pair<tvm::te::Tensor const, tvm::tir::Buffer> > > const&, tvm::GlobalVarSupply, bool)
+            at ../src/driver/driver_api.cc:388
+      11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+            at ../src/driver/driver_api.cc:283
+      10: tvm::transform::Pass::operator()(tvm::IRModule) const
+            at ../src/ir/transform.cc:258
+      9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+            at ../src/ir/transform.cc:274
+      8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+            at ../src/ir/transform.cc:451
+      7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+            at ../src/ir/transform.cc:274
+      6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+            at ../src/tir/ir/transform.cc:100
+      5: tvm::runtime::TypedPackedFunc<tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)>::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
+            at ../include/tvm/runtime/packed_func.h:1753
+      4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher<tvm::tir::PrimFunc>::run<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::runtime::PackedFunc const&, tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&)
+            at ../include/tvm/runtime/packed_func.h:1697
+      3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&) const
+            at ../include/tvm/runtime/packed_func.h:1621
+      2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+            at ../include/tvm/runtime/packed_func.h:1217
+      1: Call
+            at ../include/tvm/runtime/packed_func.h:1213
+      0: operator()
+            at ../src/runtime/c_runtime_api.cc:534
+      File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
+      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
+        raise InstantiationError("Skipped because of invalid gpu kernel")
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel
 
-            [('tile_f', [-1, 4, 1, 16]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4912986
-    No: 5   GFLOPS: 0.00/7.51       result: Traceback (most recent call last):
+    Traceback (most recent call last):
+      24: TVMFuncCall
+            at ../src/runtime/c_runtime_api.cc:477
+      23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+            at ../include/tvm/runtime/packed_func.h:1217
+      22: Call
+            at ../include/tvm/runtime/packed_func.h:1213
+      21: operator()
+            at ../include/tvm/runtime/packed_func.h:1734
+      20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
+            at ../include/tvm/runtime/packed_func.h:1674
+      19: run<>
+            at ../include/tvm/runtime/packed_func.h:1634
+      18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1634
+      17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1634
+      16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1634
+      15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1634
+      14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1649
+      13: operator()
+            at ../src/driver/driver_api.cc:402
+      12: tvm::LowerSchedule(tvm::te::Schedule, tvm::runtime::Array<tvm::runtime::ObjectRef, void> const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::unordered_map<tvm::te::Tensor, tvm::tir::Buffer, std::hash<tvm::te::Tensor>, std::equal_to<tvm::te::Tensor>, std::allocator<std::pair<tvm::te::Tensor const, tvm::tir::Buffer> > > const&, tvm::GlobalVarSupply, bool)
+            at ../src/driver/driver_api.cc:388
+      11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+            at ../src/driver/driver_api.cc:283
+      10: tvm::transform::Pass::operator()(tvm::IRModule) const
+            at ../src/ir/transform.cc:258
+      9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+            at ../src/ir/transform.cc:274
+      8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+            at ../src/ir/transform.cc:451
+      7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+            at ../src/ir/transform.cc:274
+      6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+            at ../src/tir/ir/transform.cc:100
+      5: tvm::runtime::TypedPackedFunc<tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)>::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
+            at ../include/tvm/runtime/packed_func.h:1753
+      4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher<tvm::tir::PrimFunc>::run<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::runtime::PackedFunc const&, tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&)
+            at ../include/tvm/runtime/packed_func.h:1697
+      3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&) const
+            at ../include/tvm/runtime/packed_func.h:1621
+      2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+            at ../include/tvm/runtime/packed_func.h:1217
+      1: Call
+            at ../include/tvm/runtime/packed_func.h:1213
+      0: operator()
+            at ../src/runtime/c_runtime_api.cc:534
+      File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
+      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
+        raise InstantiationError("Skipped because of invalid gpu kernel")
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 4, 32, 2]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 8, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4305272
+    No: 5   GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -550,8 +760,8 @@ for this template
       File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
-    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 2, 32, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 32]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1497482
-    No: 6   GFLOPS: 0.00/7.51       result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 8, 8, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 128]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8304464
+    No: 6   GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -673,8 +883,9 @@ for this template
       File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
-    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 16, 32, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 8, 64]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4818704
-    No: 7   GFLOPS: 0.00/7.51       result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 16, 8, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 128, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1963363
+    No: 7   GFLOPS: 6.48/345.53     result: MeasureResult(costs=(0.03575116725,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.9435207843780518, timestamp=1680189110.9070928)      [('tile_f', [-1, 4, 1, 16]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,970586
+    No: 8   GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -796,10 +1007,9 @@ for this template
       File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
-    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 8, 64, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 8, 32]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4220088
-    No: 8   GFLOPS: 119.61/119.61   result: MeasureResult(costs=(0.001935420724137931,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.9431655406951904, timestamp=1680177394.8302236)       [('tile_f', [-1, 4, 8, 2]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 8]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5525381
-    No: 9   GFLOPS: 7.99/119.61     result: MeasureResult(costs=(0.02897599975,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.5775856971740723, timestamp=1680177397.6259804)      [('tile_f', [-1, 4, 1, 2]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 8, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,13917
-    No: 10  GFLOPS: 0.00/119.61     result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 2, 64, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 128, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4902746
+    No: 9   GFLOPS: 218.08/345.53   result: MeasureResult(costs=(0.0010615198210526316,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.275648593902588, timestamp=1680189112.3862677)       [('tile_f', [-1, 4, 64, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,5100967
+    No: 10  GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -921,8 +1131,8 @@ for this template
       File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
-    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 4, 1, 32]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 16, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2822127
-    No: 11  GFLOPS: 0.00/119.61     result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 2, 1, 256]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 32]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8080377
+    No: 11  GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1044,8 +1254,8 @@ for this template
       File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
-    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 16, 4, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 4, 128]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6182779
-    No: 12  GFLOPS: 0.00/119.61     result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 16, 4, 4]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1283159
+    No: 12  GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1167,8 +1377,8 @@ for this template
       File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
-    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 16, 4, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1431836
-    No: 13  GFLOPS: 0.00/119.61     result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 1, 2, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 32, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4593523
+    No: 13  GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1290,9 +1500,8 @@ for this template
       File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
-    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 1, 128, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 8, 16]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4392655
-    No: 14  GFLOPS: 4.04/119.61     result: MeasureResult(costs=(0.05735276025,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.262774705886841, timestamp=1680177403.1119318)       [('tile_f', [-1, 2, 8, 32]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 2]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,8747618
-    No: 15  GFLOPS: 0.00/119.61     result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 64, 4, 2]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 8, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6909178
+    No: 14  GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1414,8 +1623,8 @@ for this template
       File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
-    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 1, 1, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 16, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6885284
-    No: 16  GFLOPS: 0.00/119.61     result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 2, 8, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 64, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3601555
+    No: 15  GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1537,8 +1746,149 @@ for this template
       File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
-    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 8, 4, 16]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 16]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9029418
-    No: 17  GFLOPS: 0.00/119.61     result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 4, 2, 16]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 32, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6669472
+    No: 16  GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
+      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
+        func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
+      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
+        func = build(s, args, target=target, runtime=runtime)
+      File "/workspace/python/tvm/driver/build_module.py", line 227, in build
+        input_mod = lower(inputs, args, name=name, binds=binds)
+      File "/workspace/python/tvm/driver/build_module.py", line 134, in lower
+        return ffi.lower_schedule(inp, args, name, binds, simple_mode)
+      File "tvm/_ffi/_cython/./packed_func.pxi", line 331, in tvm._ffi._cy3.core.PackedFuncBase.__call__
+      File "tvm/_ffi/_cython/./packed_func.pxi", line 276, in tvm._ffi._cy3.core.FuncCall
+      File "tvm/_ffi/_cython/./base.pxi", line 181, in tvm._ffi._cy3.core.CHECK_CALL
+    tvm._ffi.base.TVMError: Traceback (most recent call last):
+      24: TVMFuncCall
+            at ../src/runtime/c_runtime_api.cc:477
+      23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+            at ../include/tvm/runtime/packed_func.h:1217
+      22: Call
+            at ../include/tvm/runtime/packed_func.h:1213
+      21: operator()
+            at ../include/tvm/runtime/packed_func.h:1734
+      20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
+            at ../include/tvm/runtime/packed_func.h:1674
+      19: run<>
+            at ../include/tvm/runtime/packed_func.h:1634
+      18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1634
+      17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1634
+      16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1634
+      15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1634
+      14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1649
+      13: operator()
+            at ../src/driver/driver_api.cc:402
+      12: tvm::LowerSchedule(tvm::te::Schedule, tvm::runtime::Array<tvm::runtime::ObjectRef, void> const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::unordered_map<tvm::te::Tensor, tvm::tir::Buffer, std::hash<tvm::te::Tensor>, std::equal_to<tvm::te::Tensor>, std::allocator<std::pair<tvm::te::Tensor const, tvm::tir::Buffer> > > const&, tvm::GlobalVarSupply, bool)
+            at ../src/driver/driver_api.cc:388
+      11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+            at ../src/driver/driver_api.cc:283
+      10: tvm::transform::Pass::operator()(tvm::IRModule) const
+            at ../src/ir/transform.cc:258
+      9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+            at ../src/ir/transform.cc:274
+      8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+            at ../src/ir/transform.cc:451
+      7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+            at ../src/ir/transform.cc:274
+      6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+            at ../src/tir/ir/transform.cc:100
+      5: tvm::runtime::TypedPackedFunc<tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)>::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
+            at ../include/tvm/runtime/packed_func.h:1753
+      4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher<tvm::tir::PrimFunc>::run<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::runtime::PackedFunc const&, tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&)
+            at ../include/tvm/runtime/packed_func.h:1697
+      3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&) const
+            at ../include/tvm/runtime/packed_func.h:1621
+      2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+            at ../include/tvm/runtime/packed_func.h:1217
+      1: Call
+            at ../include/tvm/runtime/packed_func.h:1213
+      0: operator()
+            at ../src/runtime/c_runtime_api.cc:534
+      File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
+      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
+        raise InstantiationError("Skipped because of invalid gpu kernel")
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel
+
+    Traceback (most recent call last):
+      24: TVMFuncCall
+            at ../src/runtime/c_runtime_api.cc:477
+      23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+            at ../include/tvm/runtime/packed_func.h:1217
+      22: Call
+            at ../include/tvm/runtime/packed_func.h:1213
+      21: operator()
+            at ../include/tvm/runtime/packed_func.h:1734
+      20: unpack_call<tvm::IRModule, 5, tvm::<lambda(tvm::te::Schedule, const tvm::runtime::Array<tvm::runtime::ObjectRef>&, const tvm::runtime::String&, const tvm::runtime::Map<tvm::te::Tensor, tvm::tir::Buffer>&, bool)> >
+            at ../include/tvm/runtime/packed_func.h:1674
+      19: run<>
+            at ../include/tvm/runtime/packed_func.h:1634
+      18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1634
+      17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1634
+      16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1634
+      15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1634
+      14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1649
+      13: operator()
+            at ../src/driver/driver_api.cc:402
+      12: tvm::LowerSchedule(tvm::te::Schedule, tvm::runtime::Array<tvm::runtime::ObjectRef, void> const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::unordered_map<tvm::te::Tensor, tvm::tir::Buffer, std::hash<tvm::te::Tensor>, std::equal_to<tvm::te::Tensor>, std::allocator<std::pair<tvm::te::Tensor const, tvm::tir::Buffer> > > const&, tvm::GlobalVarSupply, bool)
+            at ../src/driver/driver_api.cc:388
+      11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+            at ../src/driver/driver_api.cc:283
+      10: tvm::transform::Pass::operator()(tvm::IRModule) const
+            at ../src/ir/transform.cc:258
+      9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+            at ../src/ir/transform.cc:274
+      8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+            at ../src/ir/transform.cc:451
+      7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+            at ../src/ir/transform.cc:274
+      6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&) const
+            at ../src/tir/ir/transform.cc:100
+      5: tvm::runtime::TypedPackedFunc<tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)>::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
+            at ../include/tvm/runtime/packed_func.h:1753
+      4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher<tvm::tir::PrimFunc>::run<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::runtime::PackedFunc const&, tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&)
+            at ../include/tvm/runtime/packed_func.h:1697
+      3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext>(tvm::tir::PrimFunc&&, tvm::IRModule&&, tvm::transform::PassContext&&) const
+            at ../include/tvm/runtime/packed_func.h:1621
+      2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+            at ../include/tvm/runtime/packed_func.h:1217
+      1: Call
+            at ../include/tvm/runtime/packed_func.h:1213
+      0: operator()
+            at ../src/runtime/c_runtime_api.cc:534
+      File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
+      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
+        raise InstantiationError("Skipped because of invalid gpu kernel")
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 2, 64, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 128, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6643694
+    No: 17  GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
+      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 142, in build
+        res = future.result()
+      File "/usr/lib/python3.7/concurrent/futures/_base.py", line 435, in result
+        return self.__get_result()
+      File "/usr/lib/python3.7/concurrent/futures/_base.py", line 384, in __get_result
+        raise self._exception
+      File "/usr/lib/python3.7/concurrent/futures/thread.py", line 57, in run
+        result = self.fn(*self.args, **self.kwargs)
+      File "/workspace/python/tvm/contrib/popen_pool.py", line 432, in <lambda>
+        worker = lambda *args: self._worker_run(*args)
+      File "/workspace/python/tvm/contrib/popen_pool.py", line 401, in _worker_run
+        return proc.recv()
+      File "/workspace/python/tvm/contrib/popen_pool.py", line 309, in recv
+        raise TimeoutError()
+    TimeoutError
+
+            [('tile_f', [-1, 64, 1, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9363701
+    No: 18  GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1660,8 +2010,8 @@ for this template
       File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
-    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 2, 64, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 32]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,725914
-    No: 18  GFLOPS: 0.00/119.61     result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 1, 1, 128]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 256]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,3093190
+    No: 19  GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1783,8 +2133,8 @@ for this template
       File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
-    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 1, 32, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4553281
-    No: 19  GFLOPS: 0.00/119.61     result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 16, 16, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 128, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4478849
+    No: 20  GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1906,8 +2256,7 @@ for this template
       File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
-    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 1, 8, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 16, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,857901
-    No: 20  GFLOPS: 17.43/119.61    result: MeasureResult(costs=(0.013282744444444446,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.1519412994384766, timestamp=1680177406.5303857)       [('tile_f', [-1, 4, 16, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 32]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,529136
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 1, 4, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 256]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8508615
 
 
 
@@ -1962,9 +2311,9 @@ and measure running time.
     Finish loading 20 records
 
     Best config:
-    [('tile_f', [-1, 4, 8, 2]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 8]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5525381
+    [('tile_f', [-1, 1, 16, 4]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 8]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2033146
     Finish loading 20 records
-    Time cost of this operator: 0.002253
+    Time cost of this operator: 0.001035
 
 
 
diff --git a/docs/_sources/how_to/work_with_microtvm/micro_autotune.rst.txt b/docs/_sources/how_to/work_with_microtvm/micro_autotune.rst.txt
index 0ba7fee730..ef35e8b400 100644
--- a/docs/_sources/how_to/work_with_microtvm/micro_autotune.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/micro_autotune.rst.txt
@@ -360,10 +360,10 @@ Timing the untuned program
     ########## Build without Autotuning ##########
     Node Name                                     Ops                                           Time(us)  Time(%)  Shape              Inputs  Outputs  Measurements(us)  
     ---------                                     ---                                           --------  -------  -----              ------  -------  ----------------  
-    tvmgen_default_fused_nn_contrib_conv2d_NCHWc  tvmgen_default_fused_nn_contrib_conv2d_NCHWc  315.2     98.742   (1, 2, 10, 10, 3)  2       1        [315.2]           
-    tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       3.048     0.955    (1, 6, 10, 10)     1       1        [3.048]           
-    tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.967     0.303    (1, 1, 10, 10, 3)  1       1        [0.967]           
-    Total_time                                    -                                             319.215   -        -                  -       -        -                 
+    tvmgen_default_fused_nn_contrib_conv2d_NCHWc  tvmgen_default_fused_nn_contrib_conv2d_NCHWc  316.0     98.731   (1, 2, 10, 10, 3)  2       1        [316.0]           
+    tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       3.103     0.969    (1, 6, 10, 10)     1       1        [3.103]           
+    tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.96      0.3      (1, 1, 10, 10, 3)  1       1        [0.96]            
+    Total_time                                    -                                             320.063   -        -                  -       -        -                 
 
 
 
@@ -428,10 +428,10 @@ Timing the tuned program
     ########## Build with Autotuning ##########
     Node Name                                     Ops                                           Time(us)  Time(%)  Shape              Inputs  Outputs  Measurements(us)  
     ---------                                     ---                                           --------  -------  -----              ------  -------  ----------------  
-    tvmgen_default_fused_nn_contrib_conv2d_NCHWc  tvmgen_default_fused_nn_contrib_conv2d_NCHWc  137.9     98.101   (1, 6, 10, 10, 1)  2       1        [137.9]           
-    tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       1.81      1.288    (1, 6, 10, 10)     1       1        [1.81]            
-    tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.859     0.611    (1, 3, 10, 10, 1)  1       1        [0.859]           
-    Total_time                                    -                                             140.57    -        -                  -       -        -                 
+    tvmgen_default_fused_nn_contrib_conv2d_NCHWc  tvmgen_default_fused_nn_contrib_conv2d_NCHWc  101.3     97.346   (1, 6, 10, 10, 1)  2       1        [101.3]           
+    tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       1.789     1.72     (1, 6, 10, 10)     1       1        [1.789]           
+    tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.972     0.934    (1, 1, 10, 10, 3)  1       1        [0.972]           
+    Total_time                                    -                                             104.062   -        -                  -       -        -                 
 
 
 
@@ -439,7 +439,7 @@ Timing the tuned program
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  23.369 seconds)
+   **Total running time of the script:** ( 1 minutes  26.505 seconds)
 
 
 .. _sphx_glr_download_how_to_work_with_microtvm_micro_autotune.py:
diff --git a/docs/_sources/how_to/work_with_microtvm/micro_pytorch.rst.txt b/docs/_sources/how_to/work_with_microtvm/micro_pytorch.rst.txt
index d7620d771c..7c0b4b37a2 100644
--- a/docs/_sources/how_to/work_with_microtvm/micro_pytorch.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/micro_pytorch.rst.txt
@@ -118,7 +118,7 @@ download a cat image and preprocess it to use as the model input.
     /venv/apache-tvm-py3.7/lib/python3.7/site-packages/torch/ao/quantization/utils.py:281: UserWarning: must run observer before calling calculate_qparams. Returning default values.
       "must run observer before calling calculate_qparams. " +
     Downloading: "https://download.pytorch.org/models/quantized/mobilenet_v2_qnnpack_37f702c5.pth" to /workspace/.cache/torch/hub/checkpoints/mobilenet_v2_qnnpack_37f702c5.pth
-
      0%|          | 0.00/3.42M [00:00<?, ?B/s]
     61%|######    | 2.09M/3.42M [00:00<00:00, 19.0MB/s]
    100%|##########| 3.42M/3.42M [00:00<00:00, 29.5MB/s]
+
      0%|          | 0.00/3.42M [00:00<?, ?B/s]
    100%|##########| 3.42M/3.42M [00:00<00:00, 74.9MB/s]
     /workspace/python/tvm/relay/frontend/pytorch_utils.py:47: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead.
       return LooseVersion(torch_ver) > ver
     /venv/apache-tvm-py3.7/lib/python3.7/site-packages/setuptools/_distutils/version.py:346: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead.
@@ -324,7 +324,7 @@ Look up prediction top 1 index in 1000 class synset.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  19.859 seconds)
+   **Total running time of the script:** ( 1 minutes  22.711 seconds)
 
 
 .. _sphx_glr_download_how_to_work_with_microtvm_micro_pytorch.py:
diff --git a/docs/_sources/how_to/work_with_microtvm/micro_train.rst.txt b/docs/_sources/how_to/work_with_microtvm/micro_train.rst.txt
index bd9af37b1c..9a8b37257a 100644
--- a/docs/_sources/how_to/work_with_microtvm/micro_train.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/micro_train.rst.txt
@@ -217,7 +217,7 @@ take about **2 minutes** to download the Stanford Cars, while COCO 2017 validati
  .. code-block:: none
 
 
-    '/tmp/tmp29az3utf/images/random'
+    '/tmp/tmpdtgsg1v_/images/random'
 
 
 
@@ -308,7 +308,7 @@ objects to other stuff? We can display some examples from our datasets using ``m
 
 
 .. image-sg:: /how_to/work_with_microtvm/images/sphx_glr_micro_train_001.png
-   :alt: [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0]
+   :alt: [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [1.0, 0.0]
    :srcset: /how_to/work_with_microtvm/images/sphx_glr_micro_train_001.png
    :class: sphx-glr-single-img
 
@@ -317,8 +317,8 @@ objects to other stuff? We can display some examples from our datasets using ``m
 
  .. code-block:: none
 
-    /tmp/tmp29az3utf/images/target contains 8144 images
-    /tmp/tmp29az3utf/images/random contains 5000 images
+    /tmp/tmpdtgsg1v_/images/target contains 8144 images
+    /tmp/tmpdtgsg1v_/images/random contains 5000 images
 
 
 
@@ -493,13 +493,13 @@ the time on our validation set).
  .. code-block:: none
 
     Epoch 1/3
-    328/328 - 41s - loss: 0.2038 - accuracy: 0.9279 - val_loss: 0.1232 - val_accuracy: 0.9551 - 41s/epoch - 124ms/step
+    328/328 - 40s - loss: 0.2212 - accuracy: 0.9237 - val_loss: 0.1385 - val_accuracy: 0.9535 - 40s/epoch - 122ms/step
     Epoch 2/3
-    328/328 - 35s - loss: 0.1043 - accuracy: 0.9628 - val_loss: 0.0954 - val_accuracy: 0.9660 - 35s/epoch - 106ms/step
+    328/328 - 35s - loss: 0.0964 - accuracy: 0.9647 - val_loss: 0.1278 - val_accuracy: 0.9588 - 35s/epoch - 107ms/step
     Epoch 3/3
-    328/328 - 35s - loss: 0.0742 - accuracy: 0.9718 - val_loss: 0.0743 - val_accuracy: 0.9713 - 35s/epoch - 105ms/step
+    328/328 - 35s - loss: 0.0650 - accuracy: 0.9745 - val_loss: 0.1614 - val_accuracy: 0.9535 - 35s/epoch - 107ms/step
 
-    <keras.callbacks.History object at 0x7f82e72d6dd0>
+    <keras.callbacks.History object at 0x7f5d233527d0>
 
 
 
@@ -860,7 +860,7 @@ Arduino tutorial for how to do that `on GitHub <https://github.com/guberti/tvm-a
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 3 minutes  51.843 seconds)
+   **Total running time of the script:** ( 4 minutes  22.021 seconds)
 
 
 .. _sphx_glr_download_how_to_work_with_microtvm_micro_train.py:
diff --git a/docs/_sources/how_to/work_with_microtvm/sg_execution_times.rst.txt b/docs/_sources/how_to/work_with_microtvm/sg_execution_times.rst.txt
index f266446e0f..9cb3fc2f89 100644
--- a/docs/_sources/how_to/work_with_microtvm/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/sg_execution_times.rst.txt
@@ -5,20 +5,20 @@
 
 Computation times
 =================
-**07:01.210** total execution time for **how_to_work_with_microtvm** files:
+**07:38.176** total execution time for **how_to_work_with_microtvm** files:
 
 +-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_train.py` (``micro_train.py``)           | 03:51.843 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_train.py` (``micro_train.py``)           | 04:22.021 | 0.0 MB |
 +-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``)     | 01:23.369 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``)     | 01:26.505 | 0.0 MB |
 +-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_pytorch.py` (``micro_pytorch.py``)       | 01:19.859 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_pytorch.py` (``micro_pytorch.py``)       | 01:22.711 | 0.0 MB |
 +-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_aot.py` (``micro_aot.py``)               | 00:10.468 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_aot.py` (``micro_aot.py``)               | 00:10.806 | 0.0 MB |
 +-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_custom_ide.py` (``micro_custom_ide.py``) | 00:08.123 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_custom_ide.py` (``micro_custom_ide.py``) | 00:08.436 | 0.0 MB |
 +-----------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``)         | 00:07.549 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``)         | 00:07.697 | 0.0 MB |
 +-----------------------------------------------------------------------------------------+-----------+--------+
 | :ref:`sphx_glr_how_to_work_with_microtvm_micro_ethosu.py` (``micro_ethosu.py``)         | 00:00.000 | 0.0 MB |
 +-----------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/work_with_relay/sg_execution_times.rst.txt b/docs/_sources/how_to/work_with_relay/sg_execution_times.rst.txt
index ec2f04859f..739118e556 100644
--- a/docs/_sources/how_to/work_with_relay/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/work_with_relay/sg_execution_times.rst.txt
@@ -5,14 +5,14 @@
 
 Computation times
 =================
-**00:37.273** total execution time for **how_to_work_with_relay** files:
+**00:38.943** total execution time for **how_to_work_with_relay** files:
 
 +----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_using_pipeline_executor.py` (``using_pipeline_executor.py``) | 00:32.622 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_using_pipeline_executor.py` (``using_pipeline_executor.py``) | 00:33.769 | 0.0 MB |
 +----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``)           | 00:02.930 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``)           | 00:03.042 | 0.0 MB |
 +----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``)                             | 00:01.714 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``)                             | 00:02.125 | 0.0 MB |
 +----------------------------------------------------------------------------------------------------+-----------+--------+
 | :ref:`sphx_glr_how_to_work_with_relay_using_relay_viz.py` (``using_relay_viz.py``)                 | 00:00.007 | 0.0 MB |
 +----------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/work_with_schedules/intrin_math.rst.txt b/docs/_sources/how_to/work_with_schedules/intrin_math.rst.txt
index 439a1b4f14..7d2616ada7 100644
--- a/docs/_sources/how_to/work_with_schedules/intrin_math.rst.txt
+++ b/docs/_sources/how_to/work_with_schedules/intrin_math.rst.txt
@@ -264,7 +264,7 @@ The following example customizes CUDA lowering rule for :code:`exp`.
  .. code-block:: none
 
 
-    <function my_cuda_math_rule at 0x7f832c24e560>
+    <function my_cuda_math_rule at 0x7f5971ece170>
 
 
 
diff --git a/docs/_sources/how_to/work_with_schedules/sg_execution_times.rst.txt b/docs/_sources/how_to/work_with_schedules/sg_execution_times.rst.txt
index 0e9abd8f74..e0b4c63487 100644
--- a/docs/_sources/how_to/work_with_schedules/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/work_with_schedules/sg_execution_times.rst.txt
@@ -5,22 +5,22 @@
 
 Computation times
 =================
-**00:06.679** total execution time for **how_to_work_with_schedules** files:
+**00:08.778** total execution time for **how_to_work_with_schedules** files:
 
 +------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``)                 | 00:03.896 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``)                 | 00:05.950 | 0.0 MB |
 +------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``)                     | 00:01.254 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``)                     | 00:01.257 | 0.0 MB |
 +------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``)                     | 00:00.629 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``)                     | 00:00.643 | 0.0 MB |
 +------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``)                               | 00:00.618 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``)                               | 00:00.632 | 0.0 MB |
 +------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``)                     | 00:00.132 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``)                     | 00:00.137 | 0.0 MB |
 +------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_tedd.py` (``tedd.py``)                               | 00:00.065 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tedd.py` (``tedd.py``)                               | 00:00.068 | 0.0 MB |
 +------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``) | 00:00.055 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``) | 00:00.059 | 0.0 MB |
 +------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_tuple_inputs.py` (``tuple_inputs.py``)               | 00:00.030 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tuple_inputs.py` (``tuple_inputs.py``)               | 00:00.032 | 0.0 MB |
 +------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/topic/vta/tutorials/autotvm/sg_execution_times.rst.txt b/docs/_sources/topic/vta/tutorials/autotvm/sg_execution_times.rst.txt
index 0131ba0980..143a4fbf3d 100644
--- a/docs/_sources/topic/vta/tutorials/autotvm/sg_execution_times.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/autotvm/sg_execution_times.rst.txt
@@ -5,10 +5,10 @@
 
 Computation times
 =================
-**00:31.991** total execution time for **topic_vta_tutorials_autotvm** files:
+**00:33.267** total execution time for **topic_vta_tutorials_autotvm** files:
 
 +---------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``) | 00:31.984 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``) | 00:33.260 | 0.0 MB |
 +---------------------------------------------------------------------------------------+-----------+--------+
 | :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_alu_vta.py` (``tune_alu_vta.py``)     | 00:00.007 | 0.0 MB |
 +---------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt b/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
index 726eccd8d9..7ebd3a4510 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
@@ -293,7 +293,7 @@ The compilation steps are:
       DeprecationWarning,
     /workspace/vta/tutorials/frontend/deploy_classification.py:213: DeprecationWarning: legacy graph executor behavior of producing json / lib / params will be removed in the next release. Please see documents of tvm.contrib.graph_executor.GraphModule for the  new recommended usage.
       relay_prog, target=tvm.target.Target(target, host=env.target_host), params=params
-    resnet18_v1 inference graph built in 34.58s!
+    resnet18_v1 inference graph built in 35.76s!
 
 
 
diff --git a/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt b/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
index 8c15e1558c..f48aeecc4a 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
@@ -337,7 +337,7 @@ The compilation steps are:
 
     /workspace/python/tvm/relay/build_module.py:348: DeprecationWarning: Please use input parameter mod (tvm.IRModule) instead of deprecated parameter mod (tvm.relay.function.Function)
       DeprecationWarning,
-    yolov3-tiny inference graph built in 23.38s!
+    yolov3-tiny inference graph built in 24.06s!
 
 
 
diff --git a/docs/_sources/topic/vta/tutorials/frontend/sg_execution_times.rst.txt b/docs/_sources/topic/vta/tutorials/frontend/sg_execution_times.rst.txt
index 0dd865a4eb..a7582c1cb9 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/sg_execution_times.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/sg_execution_times.rst.txt
@@ -5,10 +5,10 @@
 
 Computation times
 =================
-**01:41.715** total execution time for **topic_vta_tutorials_frontend** files:
+**01:44.128** total execution time for **topic_vta_tutorials_frontend** files:
 
 +------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``) | 00:51.486 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``) | 00:52.788 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``)           | 00:50.228 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``)           | 00:51.340 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/topic/vta/tutorials/optimize/sg_execution_times.rst.txt b/docs/_sources/topic/vta/tutorials/optimize/sg_execution_times.rst.txt
index 82555c8ea3..661322e6d2 100644
--- a/docs/_sources/topic/vta/tutorials/optimize/sg_execution_times.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/optimize/sg_execution_times.rst.txt
@@ -5,10 +5,10 @@
 
 Computation times
 =================
-**00:03.243** total execution time for **topic_vta_tutorials_optimize** files:
+**00:03.267** total execution time for **topic_vta_tutorials_optimize** files:
 
 +--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``)         | 00:02.756 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``)         | 00:02.749 | 0.0 MB |
 +--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``) | 00:00.488 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``) | 00:00.518 | 0.0 MB |
 +--------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/topic/vta/tutorials/sg_execution_times.rst.txt b/docs/_sources/topic/vta/tutorials/sg_execution_times.rst.txt
index ebb1bb4736..a091421599 100644
--- a/docs/_sources/topic/vta/tutorials/sg_execution_times.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/sg_execution_times.rst.txt
@@ -5,10 +5,10 @@
 
 Computation times
 =================
-**00:00.837** total execution time for **topic_vta_tutorials** files:
+**00:00.874** total execution time for **topic_vta_tutorials** files:
 
 +---------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``) | 00:00.431 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``) | 00:00.450 | 0.0 MB |
 +---------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``) | 00:00.406 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``) | 00:00.424 | 0.0 MB |
 +---------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt b/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
index bc22cbb5f3..6d8b6f03e7 100644
--- a/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
+++ b/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
@@ -318,7 +318,7 @@ We build the binary and check its correctness and performance.
 
  .. code-block:: none
 
-    Execution time of this operator: 94.776 ms
+    Execution time of this operator: 95.152 ms
 
 
 
@@ -434,7 +434,7 @@ operations.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  22.908 seconds)
+   **Total running time of the script:** ( 1 minutes  29.430 seconds)
 
 
 .. _sphx_glr_download_tutorial_auto_scheduler_matmul_x86.py:
diff --git a/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt b/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt
index 2a66b2ff72..092f907ae1 100644
--- a/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt
+++ b/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt
@@ -454,16 +454,16 @@ reduce variance, we take 5 measurements and average them.
     waiting for device...
     device available
     Get devices for measurement successfully!
-    No: 1   GFLOPS: 4.54/4.54       result: MeasureResult(costs=(0.05908292340000001,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.2322745323181152, timestamp=1680175745.71282)  [('tile_y', [-1, 16]), ('tile_x', [-1, 16])],None,44
-    No: 2   GFLOPS: 10.74/10.74     result: MeasureResult(costs=(0.0250013296,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.65120530128479, timestamp=1680175747.6591613) [('tile_y', [-1, 256]), ('tile_x', [-1, 512])],None,98
-    No: 3   GFLOPS: 3.91/10.74      result: MeasureResult(costs=(0.0687092798,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.3654131889343262, timestamp=1680175750.292137)        [('tile_y', [-1, 32]), ('tile_x', [-1, 16])],None,45
-    No: 4   GFLOPS: 3.56/10.74      result: MeasureResult(costs=(0.0754346876,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.436528205871582, timestamp=1680175753.0419655)        [('tile_y', [-1, 256]), ('tile_x', [-1, 16])],None,48
-    No: 5   GFLOPS: 2.42/10.74      result: MeasureResult(costs=(0.1108373208,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.0261006355285645, timestamp=1680175755.2605486)       [('tile_y', [-1, 512]), ('tile_x', [-1, 16])],None,49
-    No: 6   GFLOPS: 3.31/10.74      result: MeasureResult(costs=(0.08105388,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.5578992366790771, timestamp=1680175756.8147922) [('tile_y', [-1, 32]), ('tile_x', [-1, 8])],None,35
-    No: 7   GFLOPS: 8.17/10.74      result: MeasureResult(costs=(0.0328537718,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.8625633716583252, timestamp=1680175757.6071165)       [('tile_y', [-1, 1]), ('tile_x', [-1, 128])],None,70
-    No: 8   GFLOPS: 2.64/10.74      result: MeasureResult(costs=(0.1016909336,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.8725430965423584, timestamp=1680175759.4908597)       [('tile_y', [-1, 2]), ('tile_x', [-1, 8])],None,31
-    No: 9   GFLOPS: 1.52/10.74      result: MeasureResult(costs=(0.17685110340000001,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.046210527420044, timestamp=1680175762.658594)  [('tile_y', [-1, 16]), ('tile_x', [-1, 1])],None,4
-    No: 10  GFLOPS: 2.74/10.74      result: MeasureResult(costs=(0.0979635238,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.789787769317627, timestamp=1680175764.4825878)        [('tile_y', [-1, 2]), ('tile_x', [-1, 16])],None,41
+    No: 1   GFLOPS: 11.30/11.30     result: MeasureResult(costs=(0.023755678,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6285767555236816, timestamp=1680187381.321088) [('tile_y', [-1, 32]), ('tile_x', [-1, 512])],None,95
+    No: 2   GFLOPS: 0.89/11.30      result: MeasureResult(costs=(0.3003585704,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.058050632476807, timestamp=1680187387.6999588)        [('tile_y', [-1, 256]), ('tile_x', [-1, 2])],None,18
+    No: 3   GFLOPS: 4.53/11.30      result: MeasureResult(costs=(0.05927983779999999,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.1981897354125977, timestamp=1680187388.9131694)        [('tile_y', [-1, 16]), ('tile_x', [-1, 16])],None,44
+    No: 4   GFLOPS: 3.63/11.30      result: MeasureResult(costs=(0.0739912782,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.4480412006378174, timestamp=1680187391.6941915)       [('tile_y', [-1, 128]), ('tile_x', [-1, 16])],None,47
+    No: 5   GFLOPS: 9.27/11.30      result: MeasureResult(costs=(0.028957669800000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.8281950950622559, timestamp=1680187392.6537077)       [('tile_y', [-1, 1]), ('tile_x', [-1, 128])],None,70
+    No: 6   GFLOPS: 0.46/11.30      result: MeasureResult(costs=(0.5841573652000001,), error_no=MeasureErrorNo.NO_ERROR, all_cost=9.609525918960571, timestamp=1680187403.6005287)  [('tile_y', [-1, 256]), ('tile_x', [-1, 1])],None,8
+    No: 7   GFLOPS: 9.89/11.30      result: MeasureResult(costs=(0.027142059200000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6920754909515381, timestamp=1680187404.3035996)       [('tile_y', [-1, 2]), ('tile_x', [-1, 256])],None,81
+    No: 8   GFLOPS: 8.73/11.30      result: MeasureResult(costs=(0.030733858800000003,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.8724057674407959, timestamp=1680187405.0673347)       [('tile_y', [-1, 1]), ('tile_x', [-1, 64])],None,60
+    No: 9   GFLOPS: 6.43/11.30      result: MeasureResult(costs=(0.041733574399999994,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.8913915157318115, timestamp=1680187406.0791385)       [('tile_y', [-1, 512]), ('tile_x', [-1, 32])],None,59
+    No: 10  GFLOPS: 1.86/11.30      result: MeasureResult(costs=(0.144045639,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.5245437622070312, timestamp=1680187408.6476486)        [('tile_y', [-1, 256]), ('tile_x', [-1, 4])],None,28
 
 
 
diff --git a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
index c629b10b36..d9a8edec03 100644
--- a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
+++ b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
@@ -311,7 +311,7 @@ standard deviation.
 
  .. code-block:: none
 
-    {'mean': 515.2749678299995, 'median': 515.1268733999927, 'std': 1.803341233836743}
+    {'mean': 520.3476530599983, 'median': 519.627120499996, 'std': 2.1229252106345986}
 
 
 
@@ -545,29 +545,31 @@ the tuning data to.
 
  .. code-block:: none
 
-
    [Task  1/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  1/25]  Current/Best:   11.21/  14.82 GFLOPS | Progress: (4/20) | 15.47 s
    [Task  1/25]  Current/Best:   15.17/  22.57 GFLOPS | Progress: (8/20) | 18.91 s
    [Task  1/25]  Current/Best:   22.53/  22.57 GFLOPS | Progress: (12/20) | 20.83 s
    [Task  1/25]  Current/Best:    9.77/  22.57 GFLOPS | Progress: (16/20) | 24.75 s
    [Task  1/25]  Current/Best:   21.97/  22.98 GFLOPS | Progress: (20/20) | 27.26 s Done.
-
    [Task  2/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  2/25]  Current/Best:   12.75/  12.75 GFLOPS | Progress: (4/20) | 4.44 s
    [Task  2/25]  Current/Best:    4.73/  20.70 GFLOPS | Progress: (8/20) | 6.06 s
    [Task  2/25]  Current/Best:   18.18/  20.70 GFLOPS | Progress: (12/20) | 7.79 s
    [Task  2/25]  Current/Best:   19.72/  20.70 GFLOPS | Progress: (16/20) | 9.66 s
    [Task  2/25]  Current/Best:   12.96/  20.70 GFLOPS | Progress: (20/20) | 12.09 s Done.
-
    [Task  3/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  3/25]  Current/Best:    9.42/  20.24 GFLOPS | Progress: (4/20) | 5.01 s
    [Task  3/25]  Current/Best:   16.99/  20.24 GFLOPS | Progress: (8/20) | 7.40 s
    [Task  3/25]  Current/Best:   11.58/  20.24 GFLOPS | Progress: (12/20) | 12.34 s
    [Task  3/25]  Current/Best:   11.01/  20.24 GFLOPS | Progress: (16/20) | 15.66 s
    [Task  3/25]  Current/Best:   17.43/  20.24 GFLOPS | Progress: (20/20) | 17.74 s Done.
-
    [Task  4/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  4/25]  Current/Best:   17.27/  18.50 GFLOPS | Progress: (4/20) | 4.33 s
    [Task  4/25]  Current/Best:   13.84/  18.50 GFLOPS | Progress: (8/20) | 6.61 s
    [Task  4/25]  Current/Best:   16.81/  18.50 GFLOPS | Progress: (12/20) | 8.57 s
    [Task  4/25]  Current/Best:    2.04/  18.50 GFLOPS | Progress: (16/20) | 10.93 s
    [Task  4/25]  Current/Best:    5.64/  18.50 GFLOPS | Progress: (20/20) | 14.61 s Done.
-
    [Task  5/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  5/25]  Current/Best:   10.82/  19.56 GFLOPS | Progress: (4/20) | 6.15 s
    [Task  5/25]  Current/Best:    5.63/  19.99 GFLOPS | Progress: (8/20) | 8.31 s
    [Task  5/25]  Current/Best:    3.02/  19.99 GFLOPS | Progress: (12/20) | 10.46 s
    [Task  5/25]  Current/Best:    3.21/  19.99 GFLOPS | Progress: (16/20) | 13.35 s
    [Task  5/25]  Current/Best:    5.58/  19.99 GFLOPS | Progress: (20/20) | 15.33 s Done.
-
    [Task  6/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  6/25]  Current/Best:   12.02/  20.42 GFLOPS | Progress: (4/20) | 7.19 s
    [Task  6/25]  Current/Best:   17.60/  20.42 GFLOPS | Progress: (8/20) | 9.78 s
    [Task  6/25]  Current/Best:    5.71/  20.42 GFLOPS | Progress: (12/20) | 14.28 s
    [Task  6/25]  Current/Best:   12.57/  20.42 GFLOPS | Progress: (16/20) | 17.47 s
    [Task  6/25]  Current/Best:   15.83/  20.42 GFLOPS | Progress: (20/20) | 20.24 s Done.
-
    [Task  7/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  7/25]  Current/Best:   20.60/  20.60 GFLOPS | Progress: (4/20) | 4.85 s
    [Task  7/25]  Current/Best:   19.54/  20.60 GFLOPS | Progress: (8/20) | 7.26 s
    [Task  7/25]  Current/Best:   11.79/  20.60 GFLOPS | Progress: (12/20) | 10.34 s
    [Task  7/25]  Current/Best:   21.15/  23.07 GFLOPS | Progress: (16/20) | 12.35 s
    [Task  7/25]  Current/Best:   13.24/  23.07 GFLOPS | Progress: (20/20) | 15.07 s Done.
-
    [Task  8/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  8/25]  Current/Best:   13.61/  14.76 GFLOPS | Progress: (4/20) | 4.98 s
    [Task  8/25]  Current/Best:    3.64/  17.56 GFLOPS | Progress: (8/20) | 12.93 s
    [Task  8/25]  Current/Best:   14.74/  17.56 GFLOPS | Progress: (12/20) | 19.04 s
    [Task  8/25]  Current/Best:   10.15/  17.56 GFLOPS | Progress: (16/20) | 21.87 s
    [Task  8/25]  Current/Best:    2.92/  17.56 GFLOPS | Progress: (20/20) | 25.01 s Done.
-
    [Task  9/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  9/25]  Current/Best:   13.54/  21.14 GFLOPS | Progress: (4/20) | 4.60 s
    [Task  9/25]  Current/Best:   10.00/  21.32 GFLOPS | Progress: (8/20) | 6.85 s
    [Task  9/25]  Current/Best:   13.94/  21.32 GFLOPS | Progress: (12/20) | 10.82 s
    [Task  9/25]  Current/Best:    8.14/  21.32 GFLOPS | Progress: (16/20) | 14.00 s
    [Task  9/25]  Current/Best:    1.90/  22.85 GFLOPS | Progress: (20/20) | 19.83 s Done.
-
    [Task 10/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 10/25]  Current/Best:   14.16/  16.76 GFLOPS | Progress: (4/20) | 4.96 s
    [Task 10/25]  Current/Best:   14.94/  17.64 GFLOPS | Progress: (8/20) | 7.20 s
    [Task 10/25]  Current/Best:   13.53/  17.64 GFLOPS | Progress: (12/20) | 10.07 s
    [Task 10/25]  Current/Best:   13.16/  17.88 GFLOPS | Progress: (16/20) | 11.96 s
    [Task 10/25]  Current/Best:    9.53/  17.88 GFLOPS | Progress: (20/20) | 14.77 s Done.
-
    [Task 11/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 11/25]  Current/Best:   13.51/  22.64 GFLOPS | Progress: (4/20) | 5.21 s
    [Task 11/25]  Current/Best:    5.80/  22.64 GFLOPS | Progress: (8/20) | 8.56 s
    [Task 11/25]  Current/Best:   19.48/  22.64 GFLOPS | Progress: (12/20) | 10.43 s
    [Task 11/25]  Current/Best:   11.95/  22.64 GFLOPS | Progress: (16/20) | 13.09 s
    [Task 11/25]  Current/Best:   19.65/  22.64 GFLOPS | Progress: (20/20) | 17.10 s Done.
-
    [Task 12/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 12/25]  Current/Best:   14.38/  14.38 GFLOPS | Progress: (4/20) | 5.29 s
    [Task 12/25]  Current/Best:   10.14/  14.38 GFLOPS | Progress: (8/20) | 14.12 s
    [Task 12/25]  Current/Best:   16.32/  16.32 GFLOPS | Progress: (12/20) | 17.19 s
    [Task 12/25]  Current/Best:    5.38/  16.32 GFLOPS | Progress: (16/20) | 19.57 s
    [Task 12/25]  Current/Best:   11.76/  16.32 GFLOPS | Progress: (20/20) | 23.56 s Done.
-
    [Task 13/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 13/25]  Current/Best:   18.28/  18.28 GFLOPS | Progress: (4/20) | 6.14 s
    [Task 13/25]  Current/Best:   12.58/  18.28 GFLOPS | Progress: (8/20) | 10.95 s
    [Task 13/25]  Current/Best:    8.55/  18.28 GFLOPS | Progress: (12/20) | 14.18 s
    [Task 13/25]  Current/Best:   15.65/  18.67 GFLOPS | Progress: (16/20) | 17.19 s
    [Task 13/25]  Current/Best:   11.07/  18.67 GFLOPS | Progress: (20/20) | 20.69 s Done.
-
    [Task 14/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 14/25]  Current/Best:   12.61/  12.61 GFLOPS | Progress: (4/20) | 6.32 s
    [Task 14/25]  Current/Best:   16.90/  16.90 GFLOPS | Progress: (8/20) | 9.73 s
    [Task 14/25]  Current/Best:    9.15/  17.41 GFLOPS | Progress: (12/20) | 13.08 s
    [Task 14/25]  Current/Best:   12.20/  19.44 GFLOPS | Progress: (16/20) | 15.08 s
    [Task 14/25]  Current/Best:   14.17/  19.44 GFLOPS | Progress: (20/20) | 17.97 s
    [Task 15/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
-
    [Task 15/25]  Current/Best:   22.24/  22.24 GFLOPS | Progress: (4/20) | 5.33 s
    [Task 15/25]  Current/Best:    7.32/  22.24 GFLOPS | Progress: (8/20) | 9.56 s
    [Task 15/25]  Current/Best:    9.93/  22.24 GFLOPS | Progress: (12/20) | 13.06 s
    [Task 15/25]  Current/Best:   13.60/  22.24 GFLOPS | Progress: (16/20) | 15.53 s
    [Task 15/25]  Current/Best:   13.43/  22.24 GFLOPS | Progress: (20/20) | 18.74 s
    [Task 16/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 16/25]  Current/Best:   11.52/  16.17 GFLOPS | Progress: (4/20) | 5.32 s
    [Task 16/25]  Current/Best:   11.62/  16.17 GFLOPS | Progress: (8/20) | 9.02 s
    [Task 16/25]  Current/Best:   13.37/  21.83 GFLOPS | Progress: (12/20) | 11.14 s
    [Task 16/25]  Current/Best:   11.47/  21.83 GFLOPS | Progress: (16/20) | 13.57 s
    [Task 16/25]  Current/Best:   19.79/  21.83 GFLOPS | Progress: (20/20) | 15.42 s Done.
-
    [Task 17/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 17/25]  Current/Best:   10.83/  10.83 GFLOPS | Progress: (4/20) | 6.71 s
    [Task 17/25]  Current/Best:   16.14/  16.14 GFLOPS | Progress: (8/20) | 9.48 s
    [Task 17/25]  Current/Best:    6.07/  16.14 GFLOPS | Progress: (12/20) | 12.42 s
    [Task 17/25]  Current/Best:   12.11/  18.04 GFLOPS | Progress: (16/20) | 14.45 s
    [Task 17/25]  Current/Best:   14.37/  18.04 GFLOPS | Progress: (20/20) | 16.95 s Done.
-
    [Task 18/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 18/25]  Current/Best:    5.29/  17.50 GFLOPS | Progress: (4/20) | 5.10 s
    [Task 18/25]  Current/Best:   10.74/  17.50 GFLOPS | Progress: (8/20) | 9.73 s
    [Task 18/25]  Current/Best:   18.62/  18.62 GFLOPS | Progress: (12/20) | 12.40 s
    [Task 18/25]  Current/Best:   16.72/  18.62 GFLOPS | Progress: (16/20) | 14.29 s
    [Task 18/25]  Current/Best:   13.63/  20.73 GFLOPS | Progress: (20/20) | 16.73 s Done.
-
    [Task 19/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 19/25]  Current/Best:    9.45/  20.66 GFLOPS | Progress: (4/20) | 5.56 s
    [Task 19/25]  Current/Best:   10.25/  20.66 GFLOPS | Progress: (8/20) | 9.71 s
    [Task 19/25]  Current/Best:    9.55/  20.66 GFLOPS | Progress: (12/20) | 13.11 s
    [Task 19/25]  Current/Best:    7.82/  20.66 GFLOPS | Progress: (16/20) | 16.91 s
    [Task 19/25]  Current/Best:   17.32/  20.66 GFLOPS | Progress: (20/20) | 23.14 s Done.
-
    [Task 20/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 20/25]  Current/Best:   10.12/  10.61 GFLOPS | Progress: (4/20) | 7.13 s
    [Task 20/25]  Current/Best:   15.49/  15.49 GFLOPS | Progress: (8/20) | 9.30 s
    [Task 20/25]  Current/Best:   10.71/  15.49 GFLOPS | Progress: (12/20) | 13.26 s
    [Task 20/25]  Current/Best:    8.80/  15.49 GFLOPS | Progress: (16/20) | 16.33 s
    [Task 20/25]  Current/Best:    2.24/  19.24 GFLOPS | Progress: (20/20) | 21.05 s
    [Task 21/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
+
    [Task  1/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  1/25]  Current/Best:   12.38/  22.93 GFLOPS | Progress: (4/20) | 10.26 s
    [Task  1/25]  Current/Best:   17.83/  22.93 GFLOPS | Progress: (8/20) | 16.37 s
    [Task  1/25]  Current/Best:   10.84/  22.93 GFLOPS | Progress: (12/20) | 19.20 s
    [Task  1/25]  Current/Best:   12.84/  22.93 GFLOPS | Progress: (16/20) | 21.61 s
    [Task  1/25]  Current/Best:   14.35/  22.93 GFLOPS | Progress: (20/20) | 24.98 s Done.
+
    [Task  2/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  2/25]  Current/Best:   12.00/  16.23 GFLOPS | Progress: (4/20) | 4.82 s
    [Task  2/25]  Current/Best:   14.97/  16.90 GFLOPS | Progress: (8/20) | 6.37 s
    [Task  2/25]  Current/Best:   10.01/  16.90 GFLOPS | Progress: (12/20) | 8.40 s
    [Task  2/25]  Current/Best:   18.77/  18.93 GFLOPS | Progress: (16/20) | 10.39 s
    [Task  2/25]  Current/Best:   14.74/  19.97 GFLOPS | Progress: (20/20) | 11.75 s Done.
+
    [Task  3/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  3/25]  Current/Best:    5.27/  13.70 GFLOPS | Progress: (4/20) | 6.22 s
    [Task  3/25]  Current/Best:   12.61/  13.70 GFLOPS | Progress: (8/20) | 8.64 s
    [Task  3/25]  Current/Best:   16.17/  17.86 GFLOPS | Progress: (12/20) | 10.97 s
    [Task  3/25]  Current/Best:   14.72/  17.86 GFLOPS | Progress: (16/20) | 14.44 s
    [Task  3/25]  Current/Best:    6.73/  17.86 GFLOPS | Progress: (20/20) | 17.77 s Done.
+
    [Task  4/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  4/25]  Current/Best:   13.45/  15.03 GFLOPS | Progress: (4/20) | 5.08 s
    [Task  4/25]  Current/Best:   15.01/  17.78 GFLOPS | Progress: (8/20) | 9.54 s
    [Task  4/25]  Current/Best:    9.71/  18.15 GFLOPS | Progress: (12/20) | 16.12 s
    [Task  4/25]  Current/Best:   11.43/  18.15 GFLOPS | Progress: (16/20) | 21.18 s
    [Task  4/25]  Current/Best:   11.59/  18.15 GFLOPS | Progress: (20/20) | 24.16 s Done.
+
    [Task  5/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  5/25]  Current/Best:   10.40/  18.25 GFLOPS | Progress: (4/20) | 4.99 s
    [Task  5/25]  Current/Best:   13.46/  18.25 GFLOPS | Progress: (8/20) | 7.20 s
    [Task  5/25]  Current/Best:    5.68/  18.25 GFLOPS | Progress: (12/20) | 9.85 s
    [Task  5/25]  Current/Best:   12.96/  18.25 GFLOPS | Progress: (16/20) | 11.88 s
    [Task  5/25]  Current/Best:   12.47/  18.25 GFLOPS | Progress: (20/20) | 14.55 s Done.
+
    [Task  6/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  6/25]  Current/Best:   18.33/  18.33 GFLOPS | Progress: (4/20) | 5.82 s
    [Task  6/25]  Current/Best:   18.51/  18.51 GFLOPS | Progress: (8/20) | 8.32 s
    [Task  6/25]  Current/Best:    4.86/  19.95 GFLOPS | Progress: (12/20) | 11.45 s
    [Task  6/25]  Current/Best:   12.35/  21.03 GFLOPS | Progress: (16/20) | 13.79 s
    [Task  6/25]  Current/Best:   17.60/  21.03 GFLOPS | Progress: (20/20) | 16.44 s Done.
+
    [Task  7/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  7/25]  Current/Best:    6.01/  18.23 GFLOPS | Progress: (4/20) | 6.68 s
    [Task  7/25]  Current/Best:   11.81/  18.23 GFLOPS | Progress: (8/20) | 9.80 s
    [Task  7/25]  Current/Best:   12.21/  18.23 GFLOPS | Progress: (12/20) | 13.77 s
    [Task  7/25]  Current/Best:    5.76/  18.23 GFLOPS | Progress: (16/20) | 17.57 s
    [Task  7/25]  Current/Best:   13.81/  18.23 GFLOPS | Progress: (20/20) | 20.02 s Done.
+
    [Task  8/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  8/25]  Current/Best:    7.24/  20.20 GFLOPS | Progress: (4/20) | 6.22 s
    [Task  8/25]  Current/Best:    4.24/  20.20 GFLOPS | Progress: (8/20) | 18.53 s
    [Task  8/25]  Current/Best:   10.51/  20.20 GFLOPS | Progress: (12/20) | 25.65 s
    [Task  8/25]  Current/Best:    6.02/  20.20 GFLOPS | Progress: (16/20) | 34.59 s
    [Task  8/25]  Current/Best:   11.59/  20.20 GFLOPS | Progress: (20/20) | 40.20 s
    [Task  9/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  9/25]  Current/Best:   10.45/  13.95 GFLOPS | Progress: (4/20) | 4.79 s Done.
+
    [Task  9/25]  Current/Best:   18.76/  18.99 GFLOPS | Progress: (8/20) | 7.90 s
    [Task  9/25]  Current/Best:   18.28/  19.80 GFLOPS | Progress: (12/20) | 10.97 s
    [Task  9/25]  Current/Best:   15.67/  19.80 GFLOPS | Progress: (16/20) | 12.72 s
    [Task  9/25]  Current/Best:   13.60/  19.80 GFLOPS | Progress: (20/20) | 17.99 s Done.
+
    [Task 10/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 10/25]  Current/Best:   13.30/  15.66 GFLOPS | Progress: (4/20) | 4.74 s
    [Task 10/25]  Current/Best:    5.46/  17.95 GFLOPS | Progress: (8/20) | 7.44 s
    [Task 10/25]  Current/Best:   16.08/  17.95 GFLOPS | Progress: (12/20) | 9.13 s
    [Task 10/25]  Current/Best:   18.58/  18.58 GFLOPS | Progress: (16/20) | 10.92 s
    [Task 10/25]  Current/Best:    4.14/  18.58 GFLOPS | Progress: (20/20) | 13.62 s Done.
+
    [Task 11/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 11/25]  Current/Best:   16.91/  16.91 GFLOPS | Progress: (4/20) | 6.61 s
    [Task 11/25]  Current/Best:    9.46/  18.47 GFLOPS | Progress: (8/20) | 9.76 s
    [Task 11/25]  Current/Best:   10.56/  18.47 GFLOPS | Progress: (12/20) | 12.67 s
    [Task 11/25]  Current/Best:    3.09/  20.65 GFLOPS | Progress: (16/20) | 15.62 s
    [Task 11/25]  Current/Best:    3.11/  20.65 GFLOPS | Progress: (20/20) | 18.43 s Done.
+
    [Task 12/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 12/25]  Current/Best:   11.67/  18.03 GFLOPS | Progress: (4/20) | 7.04 s
    [Task 12/25]  Current/Best:   14.54/  18.03 GFLOPS | Progress: (8/20) | 15.64 s
    [Task 12/25]  Current/Best:    9.91/  18.03 GFLOPS | Progress: (12/20) | 19.94 s
    [Task 12/25]  Current/Best:   17.20/  18.03 GFLOPS | Progress: (16/20) | 22.46 s
    [Task 12/25]  Current/Best:   11.48/  18.20 GFLOPS | Progress: (20/20) | 24.45 s Done.
+
    [Task 13/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 13/25]  Current/Best:    6.20/  16.90 GFLOPS | Progress: (4/20) | 5.59 s
    [Task 13/25]  Current/Best:   15.80/  18.54 GFLOPS | Progress: (8/20) | 9.44 s
    [Task 13/25]  Current/Best:   11.27/  22.05 GFLOPS | Progress: (12/20) | 12.73 s
    [Task 13/25]  Current/Best:   22.30/  22.30 GFLOPS | Progress: (16/20) | 16.05 s
    [Task 13/25]  Current/Best:   15.56/  22.30 GFLOPS | Progress: (20/20) | 20.37 s Done.
+
    [Task 14/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 14/25]  Current/Best:    3.89/  15.61 GFLOPS | Progress: (4/20) | 6.53 s
    [Task 14/25]  Current/Best:   10.21/  18.86 GFLOPS | Progress: (8/20) | 9.80 s
    [Task 14/25]  Current/Best:   10.63/  18.86 GFLOPS | Progress: (12/20) | 14.23 s
    [Task 14/25]  Current/Best:   14.35/  18.86 GFLOPS | Progress: (16/20) | 18.09 s
    [Task 14/25]  Current/Best:    8.97/  18.86 GFLOPS | Progress: (20/20) | 20.66 s
    [Task 15/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 15/25]  Current/Best:   19.69/  19.69 GFLOPS | Progress: (4/20) | 9.91 s
    [Task 15/25]  Current/Best:    6.42/  22.02 GFLOPS | Progress: (8/20) | 12.88 s
    [Task 15/25]  Current/Best:    8.97/  22.02 GFLOPS | Progress: (12/20) | 17.93 s
    [Task 15/25]  Current/Best:   20.66/  22.74 GFLOPS | Progress: (16/20) | 21.40 s
    [Task 15/25]  Current/Best:   12.49/  22.74 GFLOPS | Progress: (20/2
 0) | 23.08 s
    [Task 16/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 16/25]  Current/Best:   13.84/  19.81 GFLOPS | Progress: (4/20) | 4.56 s
    [Task 16/25]  Current/Best:   13.69/  19.81 GFLOPS | Progress: (8/20) | 6.31 s
    [Task 16/25]  Current/Best:   16.07/  19.81 GFLOPS | Progress: (12/20) | 7.88 s
    [Task 16/25]  Current/Best:   11.98/  19.81 GFLOPS | Progress: (16/20) | 10.19 s
    [Task 16/25]  Current/Best:   14.89/  20.59 GFLOPS | Progress: (20/20) | 11.84 s Done.
+
    [Task 17/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 17/25]  Current/Best:   18.35/  18.35 GFLOPS | Progress: (4/20) | 5.60 s
    [Task 17/25]  Current/Best:   12.43/  18.35 GFLOPS | Progress: (8/20) | 8.35 s
    [Task 17/25]  Current/Best:   22.04/  22.04 GFLOPS | Progress: (12/20) | 11.06 s
    [Task 17/25]  Current/Best:   11.32/  22.04 GFLOPS | Progress: (16/20) | 13.83 s
    [Task 17/25]  Current/Best:   23.31/  23.31 GFLOPS | Progress: (20/20) | 16.50 s Done.
+
    [Task 18/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 18/25]  Current/Best:    9.73/  16.58 GFLOPS | Progress: (4/20) | 8.15 s
    [Task 18/25]  Current/Best:    7.08/  17.35 GFLOPS | Progress: (8/20) | 10.23 s
    [Task 18/25]  Current/Best:   12.50/  19.45 GFLOPS | Progress: (12/20) | 12.66 s
    [Task 18/25]  Current/Best:    1.58/  19.45 GFLOPS | Progress: (16/20) | 18.66 s
    [Task 18/25]  Current/Best:    9.25/  19.45 GFLOPS | Progress: (20/20) | 20.95 s Done.
+
    [Task 19/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 19/25]  Current/Best:   11.29/  20.77 GFLOPS | Progress: (4/20) | 7.52 s
    [Task 19/25]  Current/Best:   21.42/  21.42 GFLOPS | Progress: (8/20) | 10.10 s
    [Task 19/25]  Current/Best:   20.74/  21.97 GFLOPS | Progress: (12/20) | 13.38 s
    [Task 19/25]  Current/Best:   18.78/  21.97 GFLOPS | Progress: (16/20) | 16.86 s
    [Task 19/25]  Current/Best:   18.08/  21.97 GFLOPS | Progress: (20/20) | 19.44 s Done.
+
    [Task 20/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 20/25]  Current/Best:   11.51/  12.75 GFLOPS | Progress: (4/20) | 5.98 s
    [Task 20/25]  Current/Best:    2.52/  17.02 GFLOPS | Progress: (8/20) | 13.09 s Done.
      Done.
-
    [Task 21/25]  Current/Best:   18.27/  18.27 GFLOPS | Progress: (4/20) | 4.68 s
    [Task 21/25]  Current/Best:    9.49/  18.27 GFLOPS | Progress: (8/20) | 8.89 s
    [Task 21/25]  Current/Best:   20.42/  20.42 GFLOPS | Progress: (12/20) | 11.62 s
    [Task 21/25]  Current/Best:   10.14/  20.42 GFLOPS | Progress: (16/20) | 15.23 s
    [Task 21/25]  Current/Best:   12.06/  20.42 GFLOPS | Progress: (20/20) | 17.91 s
    [Task 22/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 22/25]  Current/Best:    2.70/  18.94 GFLOPS | Progress: (4/20) | 5.98 s
    [Task 22/25]  Current/Best:   15.03/  20.31 GFLOPS | Progress: (8/20) | 7.58 s
    [Task 22/25]  Current/Best:    6.67/  20.31 GFLOPS | Progress: (12/20) | 9.45 s
    [Task 22/25]  Current/Best:   12.32/  20.31 GFLOPS | Progress: (16/20) | 11.03 s
    [Task 22/25]  Current/Best:    9.96/  20.31 GFLOPS | Progress: (20/20) | 14.49 s Done.
-
    [Task 23/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 23/25]  Current/Best:   12.88/  16.66 GFLOPS | Progress: (4/20) | 5.36 s
    [Task 23/25]  Current/Best:   23.26/  23.26 GFLOPS | Progress: (8/20) | 10.57 s
    [Task 23/25]  Current/Best:    4.86/  23.26 GFLOPS | Progress: (12/20) | 14.84 s
    [Task 23/25]  Current/Best:   21.54/  23.26 GFLOPS | Progress: (16/20) | 17.32 s
    [Task 23/25]  Current/Best:   22.71/  23.26 GFLOPS | Progress: (20/20) | 19.88 s Done.
-
    [Task 24/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 24/25]  Current/Best:    3.47/   5.21 GFLOPS | Progress: (4/20) | 13.79 s
    [Task 24/25]  Current/Best:    3.84/   5.59 GFLOPS | Progress: (8/20) | 24.47 s
    [Task 24/25]  Current/Best:    4.41/   5.85 GFLOPS | Progress: (12/20) | 28.96 s
    [Task 24/25]  Current/Best:    5.61/   7.44 GFLOPS | Progress: (16/20) | 32.35 s
    [Task 24/25]  Current/Best:   10.06/  10.06 GFLOPS | Progress: (20/20) | 39.38 s
    [Task 25/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 25/25]  Current/Best:    3.00/   9.08 GFLOPS | Progress: (4/20) | 8.27 s
    [Task 25/25]  Current/Best:    7.48/   9.08 GFLOPS | Progress: (8/20) | 10.85 s
    [Task 25/25]  Current/Best:    1.54/   9.08 GFLOPS | Progress: (12/20) | 13.48 s
    [Task 25/25]  Current/Best:    7.91/   9.08 GFLOPS | Progress: (16/20) | 14.94 s
    [Task 25/25]  Current/Best:    2.95/   9.08 GFLOPS | Progress: (20
 /20) | 25.64 s
+
    [Task 20/25]  Current/Best:    8.03/  17.02 GFLOPS | Progress: (12/20) | 16.73 s
    [Task 20/25]  Current/Best:   11.44/  19.26 GFLOPS | Progress: (16/20) | 20.19 s
    [Task 20/25]  Current/Best:    2.38/  19.26 GFLOPS | Progress: (20/20) | 23.59 s
    [Task 21/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 21/25]  Current/Best:    8.38/  12.25 GFLOPS | Progress: (4/20) | 9.37 s
    [Task 21/25]  Current/Best:    9.89/  12.25 GFLOPS | Progress: (8/20) | 11.83 s
    [Task 21/25]  Current/Best:    9.57/  17.19 GFLOPS | Progress: (12/20) | 14.05 s
    [Task 21/25]  Current/Best:    5.72/  19.12 GFLOPS | Progress: (16/20) | 17.01 s
    [Task 21/25]  Current/Best:    8.47/  19.12 GFLOPS | Progress: (20/20) | 22.14 s
    [Task 22/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 22/25]  Current/Best:   12.55/  15.87 GFLOPS | Progress: (4/20) | 5.01 s
    [Task 22/25]  Current/Best:    2.62/  17.84 GFLOPS | Progress: (8/
 20) | 7.17 s
    [Task 22/25]  Current/Best:   14.81/  17.84 GFLOPS | Progress: (12/20) | 9.36 s Done.
+
    [Task 22/25]  Current/Best:   10.15/  17.84 GFLOPS | Progress: (16/20) | 11.66 s
    [Task 22/25]  Current/Best:   11.49/  17.84 GFLOPS | Progress: (20/20) | 13.67 s Done.
+
    [Task 23/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 23/25]  Current/Best:   11.97/  21.36 GFLOPS | Progress: (4/20) | 6.88 s
    [Task 23/25]  Current/Best:    9.81/  22.79 GFLOPS | Progress: (8/20) | 9.41 s
    [Task 23/25]  Current/Best:   12.87/  22.79 GFLOPS | Progress: (12/20) | 14.11 s
    [Task 23/25]  Current/Best:   11.84/  22.79 GFLOPS | Progress: (16/20) | 17.68 s
    [Task 23/25]  Current/Best:   21.52/  22.79 GFLOPS | Progress: (20/20) | 20.43 s Done.
+
    [Task 24/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 24/25]  Current/Best:    3.10/   3.10 GFLOPS | Progress: (4/20) | 13.97 s
    [Task 24/25]  Current/Best:    7.39/   7.81 GFLOPS | Progress: (8/20) | 24.96 s
    [Task 24/25]  Current/Best:    1.40/   7.81 GFLOPS | Progress: (12/20) | 34.01 s
    [Task 24/25]  Current/Best:    4.38/   7.81 GFLOPS | Progress: (16/20) | 37.14 s
    [Task 24/25]  Current/Best:    1.77/   7.81 GFLOPS | Progress: (20/20) | 47.54 s
    [Task 25/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 25/25]  Current/Best:    3.06/   5.62 GFLOPS | Progress: (4/20) | 15.51 s Done.
+     Done.
+
    [Task 25/25]  Current/Best:    9.71/   9.71 GFLOPS | Progress: (8/20) | 27.60 s
    [Task 25/25]  Current/Best:    7.90/   9.71 GFLOPS | Progress: (12/20) | 31.09 s
    [Task 25/25]  Current/Best:    7.48/   9.71 GFLOPS | Progress: (16/20) | 42.07 s
    [Task 25/25]  Current/Best:    7.25/   9.71 GFLOPS | Progress: (20/20) | 53.04 s
 
 
 
@@ -663,7 +665,7 @@ Verify that the optimized model runs and produces the same results:
 
  .. code-block:: none
 
-    class='n02123045 tabby, tabby cat' with probability=0.621102
+    class='n02123045 tabby, tabby cat' with probability=0.621103
     class='n02123159 tiger cat' with probability=0.356379
     class='n02124075 Egyptian cat' with probability=0.019712
     class='n02129604 tiger, Panthera tigris' with probability=0.001215
@@ -721,8 +723,8 @@ improvement in comparing the optimized model to the unoptimized model.
 
  .. code-block:: none
 
-    optimized: {'mean': 431.3350053699992, 'median': 429.80177999999114, 'std': 3.522132704584328}
-    unoptimized: {'mean': 515.2749678299995, 'median': 515.1268733999927, 'std': 1.803341233836743}
+    optimized: {'mean': 424.97096428000077, 'median': 423.9221155999985, 'std': 3.647045798278026}
+    unoptimized: {'mean': 520.3476530599983, 'median': 519.627120499996, 'std': 2.1229252106345986}
 
 
 
@@ -745,7 +747,7 @@ profiling/benchmarking.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 12 minutes  4.440 seconds)
+   **Total running time of the script:** ( 13 minutes  17.978 seconds)
 
 
 .. _sphx_glr_download_tutorial_autotvm_relay_x86.py:
diff --git a/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt b/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
index a9addc8ec2..be0b867357 100644
--- a/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
+++ b/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
@@ -274,7 +274,7 @@ device and returns the measured cost. Network overhead is excluded.
 
  .. code-block:: none
 
-    1.243e-07 secs/op
+    1.249e-07 secs/op
 
 
 
diff --git a/docs/_sources/tutorial/intro_topi.rst.txt b/docs/_sources/tutorial/intro_topi.rst.txt
index a48f2dedbd..b3fc780d9a 100644
--- a/docs/_sources/tutorial/intro_topi.rst.txt
+++ b/docs/_sources/tutorial/intro_topi.rst.txt
@@ -270,7 +270,7 @@ As you can see, scheduled stages of computation have been accumulated and we can
 
  .. code-block:: none
 
-    [stage(a, placeholder(a, 0x23253170)), stage(b, placeholder(b, 0x236e0390)), stage(T_add, compute(T_add, body=[a[ax0, ax1, ax2] + b[ax1, ax2]], axis=[T.iter_var(ax0, T.Range(0, 100), "DataPar", ""), T.iter_var(ax1, T.Range(0, 10), "DataPar", ""), T.iter_var(ax2, T.Range(0, 10), "DataPar", "")], reduce_axis=[], tag=broadcast, attrs={})), stage(T_multiply, compute(T_multiply, body=[a[ax0, ax1, ax2] * b[ax1, ax2]], axis=[T.iter_var(ax0, T.Range(0, 100), "DataPar", ""), T.iter_var(ax1, T [...]
+    [stage(a, placeholder(a, 0xe517b00)), stage(b, placeholder(b, 0xe573c30)), stage(T_add, compute(T_add, body=[a[ax0, ax1, ax2] + b[ax1, ax2]], axis=[T.iter_var(ax0, T.Range(0, 100), "DataPar", ""), T.iter_var(ax1, T.Range(0, 10), "DataPar", ""), T.iter_var(ax2, T.Range(0, 10), "DataPar", "")], reduce_axis=[], tag=broadcast, attrs={})), stage(T_multiply, compute(T_multiply, body=[a[ax0, ax1, ax2] * b[ax1, ax2]], axis=[T.iter_var(ax0, T.Range(0, 100), "DataPar", ""), T.iter_var(ax1, T.R [...]
 
 
 
diff --git a/docs/_sources/tutorial/sg_execution_times.rst.txt b/docs/_sources/tutorial/sg_execution_times.rst.txt
index 3ac6de4c98..e8a4915cc7 100644
--- a/docs/_sources/tutorial/sg_execution_times.rst.txt
+++ b/docs/_sources/tutorial/sg_execution_times.rst.txt
@@ -5,24 +5,24 @@
 
 Computation times
 =================
-**15:35.429** total execution time for **tutorial** files:
+**17:04.432** total execution time for **tutorial** files:
 
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``)                 | 12:04.440 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``)                 | 13:17.978 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``) | 01:22.908 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``) | 01:29.430 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``)     | 01:01.737 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``)     | 01:01.229 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``)                 | 00:37.310 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``)                 | 00:38.194 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``)               | 00:26.597 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``)               | 00:34.809 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``)       | 00:01.395 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``)       | 00:01.684 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``)                               | 00:00.859 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``)                               | 00:00.902 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``) | 00:00.184 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``) | 00:00.206 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
 | :ref:`sphx_glr_tutorial_uma.py` (``uma.py``)                                             | 00:00.000 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/tutorial/tensor_expr_get_started.rst.txt b/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
index 386c4ac90c..b167369796 100644
--- a/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
+++ b/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
@@ -285,7 +285,7 @@ helper function to run a profile of the TVM generated code.
 
  .. code-block:: none
 
-    Numpy running time: 0.000007
+    Numpy running time: 0.000008
     naive: 0.000007
 
 
@@ -498,10 +498,10 @@ We can now compare the different schedules
  .. code-block:: none
 
                 Operator                  Timing             Performance
-                   numpy    6.844819999969331e-06                    1.0
-                   naive              6.7166e-06       0.981267586295928
-                parallel              8.2675e-06      1.2078476862849634
-                  vector              2.4757e-05      3.6168956963237786
+                   numpy    7.6065899997956875e-06                   1.0
+                   naive              6.9944e-06      0.9195184701933283
+                parallel              8.1905e-06      1.0767637009777045
+                  vector             2.45968e-05       3.233617166254612
 
 
 
@@ -922,7 +922,7 @@ matrix multiplication.
 
  .. code-block:: none
 
-    Numpy running time: 0.018900
+    Numpy running time: 0.019549
 
 
 
@@ -980,7 +980,7 @@ optimizations.
 
  .. code-block:: none
 
-    none: 3.451678
+    none: 3.368996
 
 
 
@@ -1080,7 +1080,7 @@ schedule.
 
  .. code-block:: none
 
-    blocking: 0.309195
+    blocking: 0.322118
 
 
 
@@ -1164,7 +1164,7 @@ already cache friendly from our previous optimizations.
 
  .. code-block:: none
 
-    vectorization: 0.338413
+    vectorization: 0.351828
     # from tvm.script import ir as I
     # from tvm.script import tir as T
 
@@ -1230,7 +1230,7 @@ more cache friendly.
 
  .. code-block:: none
 
-    loop permutation: 0.117266
+    loop permutation: 0.124595
     # from tvm.script import ir as I
     # from tvm.script import tir as T
 
@@ -1321,7 +1321,7 @@ optimized schedule.
 
  .. code-block:: none
 
-    array packing: 0.108719
+    array packing: 0.110738
     # from tvm.script import ir as I
     # from tvm.script import tir as T
 
@@ -1404,7 +1404,7 @@ to `C` when all the block results are ready.
 
  .. code-block:: none
 
-    block caching: 0.110787
+    block caching: 0.111938
     # from tvm.script import ir as I
     # from tvm.script import tir as T
 
@@ -1478,7 +1478,7 @@ of thread-level parallelization.
 
  .. code-block:: none
 
-    parallelization: 0.146727
+    parallelization: 0.148113
     # from tvm.script import ir as I
     # from tvm.script import tir as T
 
@@ -1548,13 +1548,13 @@ working, we can compare the results.
  .. code-block:: none
 
                 Operator                  Timing             Performance
-                    none      3.4516783755999993                     1.0
-                blocking            0.3091947298     0.08957808235718173
-           vectorization            0.3384131855     0.09804308185034019
-        loop permutation            0.1172662061     0.03397367695928964
-           array packing     0.10871941569999999     0.03149755100838486
-           block caching            0.1107873264     0.03209665395917487
-         parallelization            0.1467269098    0.042508859121178896
+                    none      3.3689961033999998                     1.0
+                blocking            0.3221179141     0.09561243296628268
+           vectorization            0.3518277945       0.104431048211939
+        loop permutation     0.12459456070000001    0.036982696588535334
+           array packing            0.1107379905     0.03286972946874083
+           block caching     0.11193792529999999     0.03322589930781812
+         parallelization     0.14811343380000003     0.04396367026086007
 
 
 
@@ -1596,7 +1596,7 @@ the computation for specific platforms.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  1.737 seconds)
+   **Total running time of the script:** ( 1 minutes  1.229 seconds)
 
 
 .. _sphx_glr_download_tutorial_tensor_expr_get_started.py:
diff --git a/docs/commit_hash b/docs/commit_hash
index 88daaf6d15..9a541877dd 100644
--- a/docs/commit_hash
+++ b/docs/commit_hash
@@ -1 +1 @@
-1d1dbebc733c173096afbb20f379210ef54acffa
+8e2382eea52bcead6e6b4835fb0e9381e38d76cd
diff --git a/docs/how_to/compile_models/from_darknet.html b/docs/how_to/compile_models/from_darknet.html
index 582ca14964..97836a4a0b 100644
--- a/docs/how_to/compile_models/from_darknet.html
+++ b/docs/how_to/compile_models/from_darknet.html
@@ -590,7 +590,7 @@ class:[&#39;truck 0.9266&#39;] left:471 top:83 right:689 bottom:169
 class:[&#39;bicycle 0.9984&#39;] left:111 top:113 right:577 bottom:447
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  18.482 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  24.383 seconds)</p>
 <div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-compile-models-from-darknet-py">
 <div class="sphx-glr-download sphx-glr-download-python docutils container">
 <p><a class="reference download internal" download="" href="../../_downloads/7716f96385bd5abb6e822041e285be54/from_darknet.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">from_darknet.py</span></code></a></p>
diff --git a/docs/how_to/compile_models/from_mxnet.html b/docs/how_to/compile_models/from_mxnet.html
index 87bf330c12..6be4b378b2 100644
--- a/docs/how_to/compile_models/from_mxnet.html
+++ b/docs/how_to/compile_models/from_mxnet.html
@@ -444,7 +444,7 @@
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;x&quot;</span><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#tuple" title="builtins.tuple" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">x</span><span class="o">.</span><span class="n">shape</span></a><span class="p">)</span>
 </pre></div>
 </div>
-<img src="../../_images/sphx_glr_from_mxnet_001.png" srcset="../../_images/sphx_glr_from_mxnet_001.png" alt="from mxnet" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zipbca01480-a548-4f52-8ffa-ec5dd7621c6a from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
+<img src="../../_images/sphx_glr_from_mxnet_001.png" srcset="../../_images/sphx_glr_from_mxnet_001.png" alt="from mxnet" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip4a39d133-01d1-457a-b748-cc514c61ab7e from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
 x (1, 3, 224, 224)
 </pre></div>
 </div>
diff --git a/docs/how_to/compile_models/from_oneflow.html b/docs/how_to/compile_models/from_oneflow.html
index 92ce04e2fd..75008a107f 100644
--- a/docs/how_to/compile_models/from_oneflow.html
+++ b/docs/how_to/compile_models/from_oneflow.html
@@ -454,11 +454,13 @@ Deprecated in NumPy 1.20; for more details and guidance: https://numpy.org/devdo
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading: &quot;https://oneflow-public.oss-cn-beijing.aliyuncs.com/model_zoo/flowvision/classification/ResNet/resnet18.zip&quot; to /workspace/.oneflow/flowvision_cache/resnet18.zip
 
   0%|          | 0.00/41.5M [00:00&lt;?, ?B/s]
-  5%|5         | 2.09M/41.5M [00:00&lt;00:02, 15.7MB/s]
- 39%|###8      | 16.0M/41.5M [00:00&lt;00:00, 71.4MB/s]
- 55%|#####5    | 23.0M/41.5M [00:00&lt;00:00, 68.7MB/s]
- 77%|#######7  | 32.0M/41.5M [00:00&lt;00:00, 76.9MB/s]
-100%|##########| 41.5M/41.5M [00:00&lt;00:00, 77.3MB/s]
+ 15%|#5        | 6.33M/41.5M [00:00&lt;00:00, 45.7MB/s]
+ 26%|##5       | 10.7M/41.5M [00:00&lt;00:00, 36.3MB/s]
+ 39%|###8      | 16.0M/41.5M [00:00&lt;00:00, 40.1MB/s]
+ 54%|#####3    | 22.3M/41.5M [00:00&lt;00:00, 47.5MB/s]
+ 65%|######5   | 27.0M/41.5M [00:00&lt;00:00, 42.0MB/s]
+ 86%|########6 | 35.8M/41.5M [00:00&lt;00:00, 56.1MB/s]
+100%|##########| 41.5M/41.5M [00:00&lt;00:00, 51.3MB/s]
 </pre></div>
 </div>
 </div>
diff --git a/docs/how_to/compile_models/from_pytorch.html b/docs/how_to/compile_models/from_pytorch.html
index 561cd6d050..af6e2a309f 100644
--- a/docs/how_to/compile_models/from_pytorch.html
+++ b/docs/how_to/compile_models/from_pytorch.html
@@ -437,11 +437,13 @@ be unstable.</p>
 Downloading: &quot;https://download.pytorch.org/models/resnet18-f37072fd.pth&quot; to /workspace/.cache/torch/hub/checkpoints/resnet18-f37072fd.pth
 
   0%|          | 0.00/44.7M [00:00&lt;?, ?B/s]
- 27%|##6       | 11.9M/44.7M [00:00&lt;00:00, 125MB/s]
- 53%|#####3    | 23.9M/44.7M [00:00&lt;00:00, 109MB/s]
- 77%|#######7  | 34.5M/44.7M [00:00&lt;00:00, 104MB/s]
- 99%|#########9| 44.4M/44.7M [00:00&lt;00:00, 104MB/s]
-100%|##########| 44.7M/44.7M [00:00&lt;00:00, 106MB/s]
+ 14%|#4        | 6.30M/44.7M [00:00&lt;00:00, 52.3MB/s]
+ 25%|##5       | 11.3M/44.7M [00:00&lt;00:00, 44.2MB/s]
+ 36%|###5      | 16.0M/44.7M [00:00&lt;00:00, 37.7MB/s]
+ 54%|#####3    | 24.0M/44.7M [00:00&lt;00:00, 49.1MB/s]
+ 72%|#######1  | 32.0M/44.7M [00:00&lt;00:00, 59.0MB/s]
+ 99%|#########8| 44.2M/44.7M [00:00&lt;00:00, 72.5MB/s]
+100%|##########| 44.7M/44.7M [00:00&lt;00:00, 60.4MB/s]
 </pre></div>
 </div>
 </div>
diff --git a/docs/how_to/compile_models/from_tensorflow.html b/docs/how_to/compile_models/from_tensorflow.html
index 914a563082..e34d81ad75 100644
--- a/docs/how_to/compile_models/from_tensorflow.html
+++ b/docs/how_to/compile_models/from_tensorflow.html
@@ -532,7 +532,7 @@ params: params converted from tensorflow params (tensor protobuf).</p>
 </div>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/workspace/python/tvm/relay/frontend/tensorflow.py:539: UserWarning: Ignore the passed shape. Shape in graphdef will be used for operator DecodeJpeg/contents.
   &quot;will be used for operator %s.&quot; % node.name
-/workspace/python/tvm/relay/frontend/tensorflow_ops.py:1009: UserWarning: DecodeJpeg: It&#39;s a pass through, please handle preprocessing before input
+/workspace/python/tvm/relay/frontend/tensorflow_ops.py:1012: UserWarning: DecodeJpeg: It&#39;s a pass through, please handle preprocessing before input
   warnings.warn(&quot;DecodeJpeg: It&#39;s a pass through, please handle preprocessing before input&quot;)
 Tensorflow protobuf imported to relay frontend.
 </pre></div>
@@ -657,7 +657,7 @@ banana (score = 0.00022)
 desk (score = 0.00019)
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  31.026 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  34.630 seconds)</p>
 <div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-compile-models-from-tensorflow-py">
 <div class="sphx-glr-download sphx-glr-download-python docutils container">
 <p><a class="reference download internal" download="" href="../../_downloads/7f1d3d1b878694c201c614c807cdebc8/from_tensorflow.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">from_tensorflow.py</span></code></a></p>
diff --git a/docs/how_to/compile_models/sg_execution_times.html b/docs/how_to/compile_models/sg_execution_times.html
index dd3cff1a9a..1d2dc754be 100644
--- a/docs/how_to/compile_models/sg_execution_times.html
+++ b/docs/how_to/compile_models/sg_execution_times.html
@@ -345,7 +345,7 @@
             
   <div class="section" id="computation-times">
 <span id="sphx-glr-how-to-compile-models-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>06:47.913</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
+<p><strong>07:06.436</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 81%" />
@@ -354,43 +354,43 @@
 </colgroup>
 <tbody>
 <tr class="row-odd"><td><p><a class="reference internal" href="from_tensorflow.html#sphx-glr-how-to-compile-models-from-tensorflow-py"><span class="std std-ref">Compile Tensorflow Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_tensorflow.py</span></code>)</p></td>
-<td><p>01:31.026</p></td>
+<td><p>01:34.630</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="from_darknet.html#sphx-glr-how-to-compile-models-from-darknet-py"><span class="std std-ref">Compile YOLO-V2 and YOLO-V3 in DarkNet Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_darknet.py</span></code>)</p></td>
-<td><p>01:18.482</p></td>
+<td><p>01:24.383</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="from_paddle.html#sphx-glr-how-to-compile-models-from-paddle-py"><span class="std std-ref">Compile PaddlePaddle Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_paddle.py</span></code>)</p></td>
-<td><p>00:57.297</p></td>
+<td><p>00:59.418</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="from_oneflow.html#sphx-glr-how-to-compile-models-from-oneflow-py"><span class="std std-ref">Compile OneFlow Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_oneflow.py</span></code>)</p></td>
-<td><p>00:38.207</p></td>
+<td><p>00:39.250</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="from_coreml.html#sphx-glr-how-to-compile-models-from-coreml-py"><span class="std std-ref">Compile CoreML Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_coreml.py</span></code>)</p></td>
-<td><p>00:32.606</p></td>
+<td><p>00:33.876</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="from_mxnet.html#sphx-glr-how-to-compile-models-from-mxnet-py"><span class="std std-ref">Compile MXNet Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_mxnet.py</span></code>)</p></td>
-<td><p>00:31.049</p></td>
+<td><p>00:31.554</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="from_tflite.html#sphx-glr-how-to-compile-models-from-tflite-py"><span class="std std-ref">Compile TFLite Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_tflite.py</span></code>)</p></td>
-<td><p>00:27.114</p></td>
+<td><p>00:28.065</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="from_pytorch.html#sphx-glr-how-to-compile-models-from-pytorch-py"><span class="std std-ref">Compile PyTorch Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_pytorch.py</span></code>)</p></td>
-<td><p>00:26.129</p></td>
+<td><p>00:27.296</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="from_keras.html#sphx-glr-how-to-compile-models-from-keras-py"><span class="std std-ref">Compile Keras Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_keras.py</span></code>)</p></td>
-<td><p>00:23.268</p></td>
+<td><p>00:25.134</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="from_onnx.html#sphx-glr-how-to-compile-models-from-onnx-py"><span class="std std-ref">Compile ONNX Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_onnx.py</span></code>)</p></td>
-<td><p>00:02.735</p></td>
+<td><p>00:02.828</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 </tbody>
diff --git a/docs/how_to/deploy_models/deploy_model_on_adreno.html b/docs/how_to/deploy_models/deploy_model_on_adreno.html
index db5b475e03..6f2b4110d1 100644
--- a/docs/how_to/deploy_models/deploy_model_on_adreno.html
+++ b/docs/how_to/deploy_models/deploy_model_on_adreno.html
@@ -799,10 +799,9 @@ Top5 predictions:
 Evaluate inference time cost...
 Execution time summary:
  mean (ms)   median (ms)    max (ms)     min (ms)     std (ms)
- 3336.1227    3335.3472    3341.7500    3332.7434      2.7839
+ 2683.2965    2681.7872    2694.2407    2679.4584      4.2230
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  4.862 seconds)</p>
 <div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-model-on-adreno-py">
 <div class="sphx-glr-download sphx-glr-download-python docutils container">
 <p><a class="reference download internal" download="" href="../../_downloads/2387d8448da213eb625e6b3d916327d4/deploy_model_on_adreno.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">deploy_model_on_adreno.py</span></code></a></p>
diff --git a/docs/how_to/deploy_models/deploy_model_on_adreno_tvmc.html b/docs/how_to/deploy_models/deploy_model_on_adreno_tvmc.html
index c086a7474d..c6aafce918 100644
--- a/docs/how_to/deploy_models/deploy_model_on_adreno_tvmc.html
+++ b/docs/how_to/deploy_models/deploy_model_on_adreno_tvmc.html
@@ -443,24 +443,27 @@ to run this tutorial with a real device over rpc.</p>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/resnet/resnet50_weights_tf_dim_ordering_tf_kernels.h5
 
      8192/102967424 [..............................] - ETA: 0s
-  7757824/102967424 [=&gt;............................] - ETA: 0s
- 12623872/102967424 [==&gt;...........................] - ETA: 0s
- 17793024/102967424 [====&gt;.........................] - ETA: 0s
- 22847488/102967424 [=====&gt;........................] - ETA: 0s
- 27820032/102967424 [=======&gt;......................] - ETA: 0s
- 32825344/102967424 [========&gt;.....................] - ETA: 0s
- 33546240/102967424 [========&gt;.....................] - ETA: 0s
- 33685504/102967424 [========&gt;.....................] - ETA: 1s
- 41402368/102967424 [===========&gt;..................] - ETA: 0s
- 50462720/102967424 [=============&gt;................] - ETA: 0s
- 64430080/102967424 [=================&gt;............] - ETA: 0s
- 72998912/102967424 [====================&gt;.........] - ETA: 0s
- 78110720/102967424 [=====================&gt;........] - ETA: 0s
- 83124224/102967424 [=======================&gt;......] - ETA: 0s
- 88227840/102967424 [========================&gt;.....] - ETA: 0s
- 93241344/102967424 [==========================&gt;...] - ETA: 0s
- 95518720/102967424 [==========================&gt;...] - ETA: 0s
-102967424/102967424 [==============================] - 1s 0us/step
+  8380416/102967424 [=&gt;............................] - ETA: 0s
+ 16769024/102967424 [===&gt;..........................] - ETA: 0s
+ 19955712/102967424 [====&gt;.........................] - ETA: 1s
+ 22429696/102967424 [=====&gt;........................] - ETA: 1s
+ 25567232/102967424 [======&gt;.......................] - ETA: 1s
+ 32661504/102967424 [========&gt;.....................] - ETA: 0s
+ 34480128/102967424 [=========&gt;....................] - ETA: 0s
+ 39092224/102967424 [==========&gt;...................] - ETA: 0s
+ 41934848/102967424 [===========&gt;..................] - ETA: 0s
+ 48578560/102967424 [=============&gt;................] - ETA: 0s
+ 52944896/102967424 [==============&gt;...............] - ETA: 0s
+ 58712064/102967424 [================&gt;.............] - ETA: 0s
+ 60907520/102967424 [================&gt;.............] - ETA: 0s
+ 61505536/102967424 [================&gt;.............] - ETA: 0s
+ 67239936/102967424 [==================&gt;...........] - ETA: 0s
+ 75489280/102967424 [====================&gt;.........] - ETA: 0s
+ 83877888/102967424 [=======================&gt;......] - ETA: 0s
+ 92266496/102967424 [=========================&gt;....] - ETA: 0s
+ 98910208/102967424 [===========================&gt;..] - ETA: 0s
+102850560/102967424 [============================&gt;.] - ETA: 0s
+102967424/102967424 [==============================] - 2s 0us/step
 </pre></div>
 </div>
 </div>
diff --git a/docs/how_to/deploy_models/deploy_model_on_android.html b/docs/how_to/deploy_models/deploy_model_on_android.html
index 0c34d79fbd..dbd026c84d 100644
--- a/docs/how_to/deploy_models/deploy_model_on_android.html
+++ b/docs/how_to/deploy_models/deploy_model_on_android.html
@@ -667,7 +667,7 @@ to the remote android device.</p>
 Evaluate inference time cost...
 Execution time summary:
  mean (ms)   median (ms)    max (ms)     min (ms)     std (ms)
-  15.9276      15.9283      16.4350      15.6684       0.1924
+  16.8834      16.8946      17.5389      16.2270       0.3879
 </pre></div>
 </div>
 </div>
diff --git a/docs/how_to/deploy_models/deploy_object_detection_pytorch.html b/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
index e20c0aada8..eeec251c47 100644
--- a/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
+++ b/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
@@ -459,22 +459,29 @@ be unstable.</p>
 Downloading: &quot;https://download.pytorch.org/models/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth&quot; to /workspace/.cache/torch/hub/checkpoints/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth
 
   0%|          | 0.00/170M [00:00&lt;?, ?B/s]
-  7%|7         | 12.1M/170M [00:00&lt;00:01, 127MB/s]
- 14%|#4        | 24.2M/170M [00:00&lt;00:01, 104MB/s]
- 20%|##        | 34.7M/170M [00:00&lt;00:01, 107MB/s]
- 27%|##6       | 45.1M/170M [00:00&lt;00:01, 104MB/s]
- 32%|###2      | 55.1M/170M [00:00&lt;00:01, 103MB/s]
- 38%|###8      | 64.9M/170M [00:00&lt;00:01, 85.1MB/s]
- 47%|####7     | 80.0M/170M [00:00&lt;00:00, 105MB/s]
- 53%|#####3    | 90.7M/170M [00:00&lt;00:00, 105MB/s]
- 59%|#####9    | 101M/170M [00:01&lt;00:00, 103MB/s]
- 65%|######5   | 111M/170M [00:01&lt;00:00, 102MB/s]
- 71%|#######1  | 121M/170M [00:01&lt;00:00, 101MB/s]
- 77%|#######7  | 131M/170M [00:01&lt;00:00, 101MB/s]
- 83%|########2 | 141M/170M [00:01&lt;00:00, 101MB/s]
- 89%|########8 | 150M/170M [00:01&lt;00:00, 101MB/s]
- 94%|#########4| 160M/170M [00:01&lt;00:00, 94.6MB/s]
-100%|##########| 170M/170M [00:01&lt;00:00, 102MB/s]
+  5%|4         | 7.99M/170M [00:00&lt;00:02, 73.2MB/s]
+  9%|9         | 16.0M/170M [00:00&lt;00:02, 74.1MB/s]
+ 14%|#4        | 24.0M/170M [00:00&lt;00:02, 59.7MB/s]
+ 18%|#7        | 30.3M/170M [00:00&lt;00:02, 51.2MB/s]
+ 21%|##        | 35.4M/170M [00:00&lt;00:03, 46.4MB/s]
+ 24%|##3       | 40.6M/170M [00:00&lt;00:02, 48.5MB/s]
+ 28%|##8       | 48.0M/170M [00:00&lt;00:02, 46.8MB/s]
+ 33%|###2      | 56.0M/170M [00:01&lt;00:02, 54.8MB/s]
+ 38%|###7      | 64.0M/170M [00:01&lt;00:02, 55.3MB/s]
+ 42%|####2     | 72.1M/170M [00:01&lt;00:01, 61.7MB/s]
+ 50%|####9     | 84.4M/170M [00:01&lt;00:01, 78.9MB/s]
+ 54%|#####4    | 92.5M/170M [00:01&lt;00:01, 63.1MB/s]
+ 58%|#####8    | 99.3M/170M [00:01&lt;00:01, 58.7MB/s]
+ 62%|######2   | 105M/170M [00:01&lt;00:01, 58.4MB/s]
+ 68%|######7   | 115M/170M [00:02&lt;00:00, 68.6MB/s]
+ 72%|#######1  | 122M/170M [00:02&lt;00:00, 67.0MB/s]
+ 76%|#######5  | 129M/170M [00:02&lt;00:00, 49.2MB/s]
+ 80%|########  | 136M/170M [00:02&lt;00:00, 49.0MB/s]
+ 85%|########4 | 144M/170M [00:02&lt;00:00, 55.9MB/s]
+ 88%|########8 | 150M/170M [00:02&lt;00:00, 56.0MB/s]
+ 93%|#########3| 158M/170M [00:02&lt;00:00, 59.8MB/s]
+ 97%|#########6| 164M/170M [00:02&lt;00:00, 56.6MB/s]
+100%|##########| 170M/170M [00:03&lt;00:00, 58.1MB/s]
 /venv/apache-tvm-py3.7/lib/python3.7/site-packages/torch/nn/functional.py:3897: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
   for i in range(dim)
 /venv/apache-tvm-py3.7/lib/python3.7/site-packages/torchvision/models/detection/anchor_utils.py:124: UserWarning: __floordiv__ is deprecated, and its behavior will change in a future version of pytorch. It currently rounds toward 0 (like the &#39;trunc&#39; function NOT &#39;floor&#39;). This results in incorrect rounding for negative values. To keep the current behavior, use torch.div(a, b, rounding_mode=&#39;trunc&#39;), or for actual floor division, use torch.div(a, b, rounding_mode=& [...]
@@ -572,7 +579,7 @@ torchvision rcnn models.</p>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Get 9 valid boxes
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 3 minutes  37.088 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 3 minutes  50.861 seconds)</p>
 <div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-object-detection-pytorch-py">
 <div class="sphx-glr-download sphx-glr-download-python docutils container">
 <p><a class="reference download internal" download="" href="../../_downloads/7795da4b258c8feff986668b95ef57ad/deploy_object_detection_pytorch.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">deploy_object_detection_pytorch.py</span></code></a></p>
diff --git a/docs/how_to/deploy_models/deploy_prequantized.html b/docs/how_to/deploy_models/deploy_prequantized.html
index 79426f69c9..47079a07cd 100644
--- a/docs/how_to/deploy_models/deploy_prequantized.html
+++ b/docs/how_to/deploy_models/deploy_prequantized.html
@@ -500,8 +500,8 @@ training. Other models require a full post training calibration.</p>
 Downloading: &quot;https://download.pytorch.org/models/mobilenet_v2-b0353104.pth&quot; to /workspace/.cache/torch/hub/checkpoints/mobilenet_v2-b0353104.pth
 
   0%|          | 0.00/13.6M [00:00&lt;?, ?B/s]
- 59%|#####8    | 7.99M/13.6M [00:00&lt;00:00, 57.9MB/s]
-100%|##########| 13.6M/13.6M [00:00&lt;00:00, 77.3MB/s]
+ 59%|#####8    | 7.99M/13.6M [00:00&lt;00:00, 39.4MB/s]
+100%|##########| 13.6M/13.6M [00:00&lt;00:00, 57.4MB/s]
 </pre></div>
 </div>
 </div>
@@ -592,7 +592,7 @@ output values are identical out of 1000 outputs from mobilenet v2.</p>
 </div>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time summary:
  mean (ms)   median (ms)    max (ms)     min (ms)     std (ms)
-  90.3702      90.3248      91.0267      89.9968       0.2241
+  90.5573      90.4678      91.6773      90.2090       0.2931
 </pre></div>
 </div>
 <div class="admonition note">
@@ -631,7 +631,7 @@ This includes support for the VNNI 8 bit dot product instruction (CascadeLake or
 <div class="section" id="deploy-a-quantized-tflite-model">
 <h2>Deploy a quantized TFLite Model<a class="headerlink" href="#deploy-a-quantized-tflite-model" title="Permalink to this headline">¶</a></h2>
 <p>TODO</p>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  17.916 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  20.761 seconds)</p>
 <div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-prequantized-py">
 <div class="sphx-glr-download sphx-glr-download-python docutils container">
 <p><a class="reference download internal" download="" href="../../_downloads/fb8217c13f4351224c6cf3aacf1a87fc/deploy_prequantized.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">deploy_prequantized.py</span></code></a></p>
diff --git a/docs/how_to/deploy_models/deploy_prequantized_tflite.html b/docs/how_to/deploy_models/deploy_prequantized_tflite.html
index f52e2ffc8b..a937e85899 100644
--- a/docs/how_to/deploy_models/deploy_prequantized_tflite.html
+++ b/docs/how_to/deploy_models/deploy_prequantized_tflite.html
@@ -585,7 +585,7 @@ TFLite Top-5 labels: [387 102 386 341 349]
 </div>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time summary:
  mean (ms)   median (ms)    max (ms)     min (ms)     std (ms)
-  121.2262     121.1393     125.6978     120.5065      0.5704
+  121.6151     121.5017     127.3501     120.7108      0.7841
 </pre></div>
 </div>
 <div class="admonition note">
@@ -613,7 +613,7 @@ network for ARM CPU</span></a>.</p></li>
 </ul>
 </div></blockquote>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes  31.789 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes  31.371 seconds)</p>
 <div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-prequantized-tflite-py">
 <div class="sphx-glr-download sphx-glr-download-python docutils container">
 <p><a class="reference download internal" download="" href="../../_downloads/56691c7a27d45da61d112276334640d3/deploy_prequantized_tflite.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">deploy_prequantized_tflite.py</span></code></a></p>
diff --git a/docs/how_to/deploy_models/deploy_quantized.html b/docs/how_to/deploy_models/deploy_quantized.html
index cc47c9d5b8..abda133784 100644
--- a/docs/how_to/deploy_models/deploy_quantized.html
+++ b/docs/how_to/deploy_models/deploy_quantized.html
@@ -526,7 +526,7 @@ for calibration. But the accuracy might be impacted.</p>
   DeprecationWarning,
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  21.232 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  51.610 seconds)</p>
 <div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-quantized-py">
 <div class="sphx-glr-download sphx-glr-download-python docutils container">
 <p><a class="reference download internal" download="" href="../../_downloads/7810ecf51bfc05f7d5e8a400ac3e815d/deploy_quantized.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">deploy_quantized.py</span></code></a></p>
diff --git a/docs/how_to/deploy_models/deploy_ssd_gluoncv.html b/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
index e283bc798c..8903582e6d 100644
--- a/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
+++ b/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
@@ -468,24 +468,26 @@ to your device.</p>
 Downloading /workspace/.mxnet/models/ssd_512_resnet50_v1_voc-9c8b225a.zip from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/ssd_512_resnet50_v1_voc-9c8b225a.zip...
 
   0%|          | 0/132723 [00:00&lt;?, ?KB/s]
-  1%|1         | 1381/132723 [00:00&lt;00:09, 13807.77KB/s]
-  3%|3         | 4239/132723 [00:00&lt;00:05, 22495.87KB/s]
-  7%|6         | 8896/132723 [00:00&lt;00:03, 33349.47KB/s]
- 12%|#2        | 16162/132723 [00:00&lt;00:02, 48828.34KB/s]
- 19%|#8        | 24802/132723 [00:00&lt;00:01, 62353.63KB/s]
- 25%|##5       | 33523/132723 [00:00&lt;00:01, 70793.29KB/s]
- 32%|###1      | 42144/132723 [00:00&lt;00:01, 75825.85KB/s]
- 38%|###8      | 50931/132723 [00:00&lt;00:01, 79655.93KB/s]
- 45%|####4     | 59697/132723 [00:00&lt;00:00, 82153.61KB/s]
- 52%|#####1    | 68472/132723 [00:01&lt;00:00, 83878.61KB/s]
- 58%|#####8    | 77221/132723 [00:01&lt;00:00, 84979.87KB/s]
- 65%|######4   | 85915/132723 [00:01&lt;00:00, 85574.27KB/s]
- 71%|#######1  | 94677/132723 [00:01&lt;00:00, 86192.75KB/s]
- 78%|#######7  | 103474/132723 [00:01&lt;00:00, 86720.72KB/s]
- 85%|########4 | 112259/132723 [00:01&lt;00:00, 87059.16KB/s]
- 91%|#########1| 120978/132723 [00:01&lt;00:00, 87093.75KB/s]
- 98%|#########7| 129688/132723 [00:01&lt;00:00, 87002.67KB/s]
-100%|##########| 132723/132723 [00:01&lt;00:00, 76381.89KB/s]
+  1%|          | 713/132723 [00:00&lt;00:18, 7084.57KB/s]
+  1%|1         | 1897/132723 [00:00&lt;00:13, 9826.07KB/s]
+  3%|2         | 3785/132723 [00:00&lt;00:09, 13866.95KB/s]
+  5%|5         | 6777/132723 [00:00&lt;00:06, 20141.75KB/s]
+  9%|8         | 11501/132723 [00:00&lt;00:04, 29867.33KB/s]
+ 14%|#4        | 18953/132723 [00:00&lt;00:02, 45001.73KB/s]
+ 21%|##        | 27537/132723 [00:00&lt;00:01, 58323.32KB/s]
+ 27%|##7       | 36167/132723 [00:00&lt;00:01, 67214.27KB/s]
+ 34%|###3      | 44691/132723 [00:00&lt;00:01, 72842.78KB/s]
+ 40%|####      | 53295/132723 [00:01&lt;00:01, 76911.62KB/s]
+ 47%|####6     | 61981/132723 [00:01&lt;00:00, 79951.06KB/s]
+ 53%|#####3    | 70644/132723 [00:01&lt;00:00, 81980.55KB/s]
+ 60%|#####9    | 79348/132723 [00:01&lt;00:00, 83511.56KB/s]
+ 66%|######6   | 88016/132723 [00:01&lt;00:00, 84466.28KB/s]
+ 73%|#######2  | 96745/132723 [00:01&lt;00:00, 85314.01KB/s]
+ 79%|#######9  | 105358/132723 [00:01&lt;00:00, 85557.28KB/s]
+ 86%|########5 | 113993/132723 [00:01&lt;00:00, 85783.90KB/s]
+ 92%|#########2| 122654/132723 [00:01&lt;00:00, 86029.53KB/s]
+ 99%|#########8| 131258/132723 [00:01&lt;00:00, 85792.40KB/s]
+100%|##########| 132723/132723 [00:01&lt;00:00, 69006.57KB/s]
 </pre></div>
 </div>
 <p>Create TVM runtime and do inference
@@ -524,7 +526,7 @@ Downloading /workspace/.mxnet/models/ssd_512_resnet50_v1_voc-9c8b225a.zip from h
 <span class="n">plt</span><span class="o">.</span><span class="n">show</span><span class="p">()</span>
 </pre></div>
 </div>
-<img src="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" srcset="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" alt="deploy ssd gluoncv" class = "sphx-glr-single-img"/><p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 3 minutes  54.132 seconds)</p>
+<img src="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" srcset="../../_images/sphx_glr_deploy_ssd_gluoncv_001.png" alt="deploy ssd gluoncv" class = "sphx-glr-single-img"/><p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 4 minutes  1.036 seconds)</p>
 <div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-ssd-gluoncv-py">
 <div class="sphx-glr-download sphx-glr-download-python docutils container">
 <p><a class="reference download internal" download="" href="../../_downloads/cccb17d28e5e8b2e94ea8cd5ec59f6ed/deploy_ssd_gluoncv.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">deploy_ssd_gluoncv.py</span></code></a></p>
diff --git a/docs/how_to/deploy_models/sg_execution_times.html b/docs/how_to/deploy_models/sg_execution_times.html
index 487a95faae..55d19b7a2b 100644
--- a/docs/how_to/deploy_models/sg_execution_times.html
+++ b/docs/how_to/deploy_models/sg_execution_times.html
@@ -345,7 +345,7 @@
             
   <div class="section" id="computation-times">
 <span id="sphx-glr-how-to-deploy-models-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>16:19.722</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
+<p><strong>17:10.847</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 86%" />
@@ -354,43 +354,43 @@
 </colgroup>
 <tbody>
 <tr class="row-odd"><td><p><a class="reference internal" href="deploy_ssd_gluoncv.html#sphx-glr-how-to-deploy-models-deploy-ssd-gluoncv-py"><span class="std std-ref">Deploy Single Shot Multibox Detector(SSD) model</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_ssd_gluoncv.py</span></code>)</p></td>
-<td><p>03:54.132</p></td>
+<td><p>04:01.036</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="deploy_object_detection_pytorch.html#sphx-glr-how-to-deploy-models-deploy-object-detection-pytorch-py"><span class="std std-ref">Compile PyTorch Object Detection Models</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_object_detection_pytorch.py</span></code>)</p></td>
-<td><p>03:37.088</p></td>
+<td><p>03:50.861</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="deploy_prequantized_tflite.html#sphx-glr-how-to-deploy-models-deploy-prequantized-tflite-py"><span class="std std-ref">Deploy a Framework-prequantized Model with TVM - Part 3 (TFLite)</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_prequantized_tflite.py</span></code>)</p></td>
-<td><p>02:31.789</p></td>
+<td><p>02:31.371</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="deploy_quantized.html#sphx-glr-how-to-deploy-models-deploy-quantized-py"><span class="std std-ref">Deploy a Quantized Model on Cuda</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_quantized.py</span></code>)</p></td>
-<td><p>01:21.232</p></td>
+<td><p>01:51.610</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="deploy_prequantized.html#sphx-glr-how-to-deploy-models-deploy-prequantized-py"><span class="std std-ref">Deploy a Framework-prequantized Model with TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_prequantized.py</span></code>)</p></td>
-<td><p>01:17.916</p></td>
+<td><p>01:20.761</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="deploy_model_on_adreno.html#sphx-glr-how-to-deploy-models-deploy-model-on-adreno-py"><span class="std std-ref">Deploy the Pretrained Model on Adreno™</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_adreno.py</span></code>)</p></td>
-<td><p>01:04.862</p></td>
+<td><p>00:57.822</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="deploy_model_on_adreno_tvmc.html#sphx-glr-how-to-deploy-models-deploy-model-on-adreno-tvmc-py"><span class="std std-ref">Deploy the Pretrained Model on Adreno™ with tvmc Interface</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_adreno_tvmc.py</span></code>)</p></td>
-<td><p>00:51.934</p></td>
+<td><p>00:53.536</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="deploy_model_on_android.html#sphx-glr-how-to-deploy-models-deploy-model-on-android-py"><span class="std std-ref">Deploy the Pretrained Model on Android</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_android.py</span></code>)</p></td>
-<td><p>00:43.679</p></td>
+<td><p>00:45.312</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="deploy_model_on_nano.html#sphx-glr-how-to-deploy-models-deploy-model-on-nano-py"><span class="std std-ref">Deploy the Pretrained Model on Jetson Nano</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_nano.py</span></code>)</p></td>
-<td><p>00:28.984</p></td>
+<td><p>00:29.511</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="deploy_model_on_rasp.html#sphx-glr-how-to-deploy-models-deploy-model-on-rasp-py"><span class="std std-ref">Deploy the Pretrained Model on Raspberry Pi</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_rasp.py</span></code>)</p></td>
-<td><p>00:28.101</p></td>
+<td><p>00:29.020</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="deploy_sparse.html#sphx-glr-how-to-deploy-models-deploy-sparse-py"><span class="std std-ref">Deploy a Hugging Face Pruned Model on CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_sparse.py</span></code>)</p></td>
diff --git a/docs/how_to/extend_tvm/bring_your_own_datatypes.html b/docs/how_to/extend_tvm/bring_your_own_datatypes.html
index 23520bb568..fd4915c5dd 100644
--- a/docs/how_to/extend_tvm/bring_your_own_datatypes.html
+++ b/docs/how_to/extend_tvm/bring_your_own_datatypes.html
@@ -624,7 +624,7 @@ In this alpha state of the Bring Your Own Datatypes framework, we have not imple
 <span class="n">module</span><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">params</span></a> <span class="o">=</span> <span class="n">get_mobilenet</span><span class="p">()</span>
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zip26bff2dd-e83e-4c22-9a2c-bcabb5b4c5ce from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zipfb42a426-6471-4b98-9fbb-2ad9cea9a496 from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
 </pre></div>
 </div>
 <p>It’s easy to execute MobileNet with native TVM:</p>
diff --git a/docs/how_to/extend_tvm/sg_execution_times.html b/docs/how_to/extend_tvm/sg_execution_times.html
index 8c6cce8ca5..f3744a24c1 100644
--- a/docs/how_to/extend_tvm/sg_execution_times.html
+++ b/docs/how_to/extend_tvm/sg_execution_times.html
@@ -345,7 +345,7 @@
             
   <div class="section" id="computation-times">
 <span id="sphx-glr-how-to-extend-tvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:56.188</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
+<p><strong>00:57.766</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 84%" />
@@ -354,15 +354,15 @@
 </colgroup>
 <tbody>
 <tr class="row-odd"><td><p><a class="reference internal" href="bring_your_own_datatypes.html#sphx-glr-how-to-extend-tvm-bring-your-own-datatypes-py"><span class="std std-ref">Bring Your Own Datatypes to TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">bring_your_own_datatypes.py</span></code>)</p></td>
-<td><p>00:52.233</p></td>
+<td><p>00:53.696</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="use_pass_instrument.html#sphx-glr-how-to-extend-tvm-use-pass-instrument-py"><span class="std std-ref">How to Use TVM Pass Instrument</span></a> (<code class="docutils literal notranslate"><span class="pre">use_pass_instrument.py</span></code>)</p></td>
-<td><p>00:02.840</p></td>
+<td><p>00:02.922</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="use_pass_infra.html#sphx-glr-how-to-extend-tvm-use-pass-infra-py"><span class="std std-ref">How to Use TVM Pass Infra</span></a> (<code class="docutils literal notranslate"><span class="pre">use_pass_infra.py</span></code>)</p></td>
-<td><p>00:01.108</p></td>
+<td><p>00:01.140</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="low_level_custom_pass.html#sphx-glr-how-to-extend-tvm-low-level-custom-pass-py"><span class="std std-ref">Writing a Customized Pass</span></a> (<code class="docutils literal notranslate"><span class="pre">low_level_custom_pass.py</span></code>)</p></td>
diff --git a/docs/how_to/extend_tvm/use_pass_instrument.html b/docs/how_to/extend_tvm/use_pass_instrument.html
index 2c27d19236..d4f1073bae 100644
--- a/docs/how_to/extend_tvm/use_pass_instrument.html
+++ b/docs/how_to/extend_tvm/use_pass_instrument.html
@@ -531,10 +531,10 @@ profile the execution time of each passes.</p>
 </pre></div>
 </div>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Printing results of timing profile...
-InferType: 22434us [22434us] (48.67%; 48.67%)
-FoldScaleAxis: 23657us [9us] (51.33%; 51.33%)
-        FoldConstant: 23648us [1726us] (51.31%; 99.96%)
-                InferType: 21922us [21922us] (47.56%; 92.70%)
+InferType: 22934us [22934us] (48.32%; 48.32%)
+FoldScaleAxis: 24529us [11us] (51.68%; 51.68%)
+        FoldConstant: 24519us [2123us] (51.66%; 99.96%)
+                InferType: 22396us [22396us] (47.18%; 91.34%)
 </pre></div>
 </div>
 </div>
@@ -556,10 +556,10 @@ Refer to following sections and <a class="reference internal" href="../../refere
 </pre></div>
 </div>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Printing results of timing profile...
-InferType: 22129us [22129us] (48.31%; 48.31%)
-FoldScaleAxis: 23675us [8us] (51.69%; 51.69%)
-        FoldConstant: 23667us [1774us] (51.67%; 99.97%)
-                InferType: 21893us [21893us] (47.80%; 92.50%)
+InferType: 22070us [22070us] (48.53%; 48.53%)
+FoldScaleAxis: 23405us [6us] (51.47%; 51.47%)
+        FoldConstant: 23399us [1771us] (51.46%; 99.98%)
+                InferType: 21629us [21629us] (47.56%; 92.43%)
 </pre></div>
 </div>
 <p>Register empty list to clear existing instruments.</p>
diff --git a/docs/how_to/optimize_operators/opt_conv_cuda.html b/docs/how_to/optimize_operators/opt_conv_cuda.html
index f2110937d5..f5113a7dd9 100644
--- a/docs/how_to/optimize_operators/opt_conv_cuda.html
+++ b/docs/how_to/optimize_operators/opt_conv_cuda.html
@@ -580,7 +580,7 @@ latency of convolution.</p>
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Convolution: </span><span class="si">%f</span><span class="s2"> ms&quot;</span> <span class="o">%</span> <span class="p">(</span><span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">w</span><span class="p">,</span> <span class="n">b</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span> <span class="o">*</span> <span cl [...]
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Convolution: 53.503616 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Convolution: 45.397857 ms
 </pre></div>
 </div>
 <div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-optimize-operators-opt-conv-cuda-py">
diff --git a/docs/how_to/optimize_operators/opt_conv_tensorcore.html b/docs/how_to/optimize_operators/opt_conv_tensorcore.html
index e40b4c0b87..d249573163 100644
--- a/docs/how_to/optimize_operators/opt_conv_tensorcore.html
+++ b/docs/how_to/optimize_operators/opt_conv_tensorcore.html
@@ -862,7 +862,7 @@ be able to run on our build server</p>
     <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;conv2d with tensor core: </span><span class="si">%f</span><span class="s2"> ms&quot;</span> <span class="o">%</span> <span class="p">(</span><span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">w</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span> <span class="o">* [...]
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>conv2d with tensor core: 12.245091 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>conv2d with tensor core: 12.237002 ms
 </pre></div>
 </div>
 </div>
diff --git a/docs/how_to/optimize_operators/opt_gemm.html b/docs/how_to/optimize_operators/opt_gemm.html
index d2741a1f6a..cb967700e9 100644
--- a/docs/how_to/optimize_operators/opt_gemm.html
+++ b/docs/how_to/optimize_operators/opt_gemm.html
@@ -477,8 +477,8 @@ Then we write a baseline implementation, the simplest way to write a matrix mult
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Baseline: </span><span class="si">%f</span><span class="s2">&quot;</span> <span class="o">%</span> <span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">)</span>
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.018616
-Baseline: 3.518981
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.020305
+Baseline: 3.363956
 </pre></div>
 </div>
 <p>In TVM, we can always inspect lower level IR to debug or optimize our schedule.
@@ -537,7 +537,7 @@ fill 32 * 32 * sizeof(float) which is 4KB in the cache whose total size is 32KB
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Opt1: </span><span class="si">%f</span><span class="s2">&quot;</span> <span class="o">%</span> <span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">)</span>
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt1: 0.298253
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt1: 0.332628
 </pre></div>
 </div>
 <p>Here is the generated IR after blocking.</p>
@@ -594,7 +594,7 @@ vastly.</p>
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Opt2: </span><span class="si">%f</span><span class="s2">&quot;</span> <span class="o">%</span> <span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">)</span>
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt2: 0.327582
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt2: 0.359120
 </pre></div>
 </div>
 <p>Here is the generated IR after vectorization.</p>
@@ -649,7 +649,7 @@ the access pattern for A matrix is more cache friendly.</p>
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Opt3: </span><span class="si">%f</span><span class="s2">&quot;</span> <span class="o">%</span> <span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">)</span>
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt3: 0.122265
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt3: 0.127020
 </pre></div>
 </div>
 <p>Here is the generated IR after loop permutation.</p>
@@ -726,7 +726,7 @@ flattening.</p>
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Opt4: </span><span class="si">%f</span><span class="s2">&quot;</span> <span class="o">%</span> <span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">)</span>
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt4: 0.110673
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt4: 0.109359
 </pre></div>
 </div>
 <p>Here is the generated IR after array packing.</p>
@@ -804,7 +804,7 @@ write to C when all the block results are ready.</p>
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Opt5: </span><span class="si">%f</span><span class="s2">&quot;</span> <span class="o">%</span> <span class="n">evaluator</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="n">c</span><span class="p">)</span><span class="o">.</span><span class="n">mean</span><span class="p">)</span>
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt5: 0.111837
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt5: 0.111824
 </pre></div>
 </div>
 <p>Here is the generated IR after blocking.</p>
@@ -884,7 +884,7 @@ class Module:
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;Opt6: </span><span class="si">%f</span><span class="s2">&quot;</span> <span class="o">%</span> <span class="n">opt6_time</span><span class="p">)</span>
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt6: 0.147249
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt6: 0.148218
 </pre></div>
 </div>
 <p>Here is the generated IR after parallelization.</p>
diff --git a/docs/how_to/optimize_operators/sg_execution_times.html b/docs/how_to/optimize_operators/sg_execution_times.html
index 7cb096627c..2faacc371c 100644
--- a/docs/how_to/optimize_operators/sg_execution_times.html
+++ b/docs/how_to/optimize_operators/sg_execution_times.html
@@ -345,7 +345,7 @@
             
   <div class="section" id="computation-times">
 <span id="sphx-glr-how-to-optimize-operators-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:35.809</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
+<p><strong>00:36.430</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 83%" />
@@ -354,15 +354,15 @@
 </colgroup>
 <tbody>
 <tr class="row-odd"><td><p><a class="reference internal" href="opt_gemm.html#sphx-glr-how-to-optimize-operators-opt-gemm-py"><span class="std std-ref">How to optimize GEMM on CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">opt_gemm.py</span></code>)</p></td>
-<td><p>00:32.718</p></td>
+<td><p>00:33.330</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="opt_conv_tensorcore.html#sphx-glr-how-to-optimize-operators-opt-conv-tensorcore-py"><span class="std std-ref">How to optimize convolution using TensorCores</span></a> (<code class="docutils literal notranslate"><span class="pre">opt_conv_tensorcore.py</span></code>)</p></td>
-<td><p>00:01.912</p></td>
+<td><p>00:01.904</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="opt_conv_cuda.html#sphx-glr-how-to-optimize-operators-opt-conv-cuda-py"><span class="std std-ref">How to optimize convolution on GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">opt_conv_cuda.py</span></code>)</p></td>
-<td><p>00:01.179</p></td>
+<td><p>00:01.196</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 </tbody>
diff --git a/docs/how_to/tune_with_autoscheduler/sg_execution_times.html b/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
index d0d95854e3..dd81fe5bf8 100644
--- a/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
+++ b/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
@@ -345,7 +345,7 @@
             
   <div class="section" id="computation-times">
 <span id="sphx-glr-how-to-tune-with-autoscheduler-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>10:07.766</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
+<p><strong>10:22.168</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 85%" />
@@ -354,27 +354,27 @@
 </colgroup>
 <tbody>
 <tr class="row-odd"><td><p><a class="reference internal" href="tune_conv2d_layer_cuda.html#sphx-glr-how-to-tune-with-autoscheduler-tune-conv2d-layer-cuda-py"><span class="std std-ref">Auto-scheduling a Convolution Layer for GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_conv2d_layer_cuda.py</span></code>)</p></td>
-<td><p>06:10.669</p></td>
+<td><p>06:18.523</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="tune_network_x86.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-x86-py"><span class="std std-ref">Auto-scheduling a Neural Network for x86 CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_x86.py</span></code>)</p></td>
-<td><p>01:43.939</p></td>
+<td><p>01:45.674</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="tune_network_cuda.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-cuda-py"><span class="std std-ref">Auto-scheduling a Neural Network for NVIDIA GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_cuda.py</span></code>)</p></td>
-<td><p>01:12.321</p></td>
+<td><p>01:13.936</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="tune_sparse_x86.html#sphx-glr-how-to-tune-with-autoscheduler-tune-sparse-x86-py"><span class="std std-ref">Auto-scheduling Sparse Matrix Multiplication on CPU with Custom Sketch Rule</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_sparse_x86.py</span></code>)</p></td>
-<td><p>00:32.355</p></td>
+<td><p>00:34.755</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="tune_network_arm.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-arm-py"><span class="std std-ref">Auto-scheduling a Neural Network for ARM CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_arm.py</span></code>)</p></td>
-<td><p>00:14.666</p></td>
+<td><p>00:14.952</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="tune_network_mali.html#sphx-glr-how-to-tune-with-autoscheduler-tune-network-mali-py"><span class="std std-ref">Auto-scheduling a Neural Network for mali GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_network_mali.py</span></code>)</p></td>
-<td><p>00:13.817</p></td>
+<td><p>00:14.327</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 </tbody>
diff --git a/docs/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.html b/docs/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.html
index ad922998a8..3f87bce697 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_conv2d_layer_cuda.html
@@ -510,71 +510,551 @@ class Module:
     @T.prim_func
     def main(data: T.Buffer((1, 512, 7, 7), &quot;float32&quot;), kernel: T.Buffer((512, 512, 3, 3), &quot;float32&quot;), bias: T.Buffer((1, 512, 1, 1), &quot;float32&quot;), compute: T.Buffer((1, 512, 7, 7), &quot;float32&quot;)):
         T.func_attr({&quot;from_legacy_te_schedule&quot;: T.bool(True), &quot;global_symbol&quot;: &quot;main&quot;, &quot;tir.noalias&quot;: T.bool(True)})
-        blockIdx_x = T.launch_thread(&quot;blockIdx.x&quot;, 32)
+        blockIdx_x = T.launch_thread(&quot;blockIdx.x&quot;, 112)
         conv2d_nchw = T.allocate([4], &quot;float32&quot;, &quot;local&quot;)
-        pad_temp_shared = T.allocate([324], &quot;float32&quot;, &quot;shared&quot;)
-        kernel_shared = T.allocate([576], &quot;float32&quot;, &quot;shared&quot;)
-        threadIdx_x = T.launch_thread(&quot;threadIdx.x&quot;, 196)
-        conv2d_nchw_1 = T.Buffer((1,), data=conv2d_nchw, scope=&quot;local&quot;, align=4)
+        pad_temp_shared = T.allocate([672], &quot;float32&quot;, &quot;shared&quot;)
+        kernel_shared = T.allocate([3072], &quot;float32&quot;, &quot;shared&quot;)
+        threadIdx_x = T.launch_thread(&quot;threadIdx.x&quot;, 56)
+        conv2d_nchw_1 = T.Buffer((4,), data=conv2d_nchw, scope=&quot;local&quot;, align=8)
         conv2d_nchw_1[0] = T.float32(0)
-        conv2d_nchw_1[1] = T.float32(0)
         conv2d_nchw_1[2] = T.float32(0)
+        conv2d_nchw_1[1] = T.float32(0)
         conv2d_nchw_1[3] = T.float32(0)
-        for rc_outer_outer in range(128):
+        for rc_outer_outer, ry_outer_outer in T.grid(16, 3):
+            cse_var_4: T.int32 = rc_outer_outer * 1568
+            cse_var_3: T.int32 = ry_outer_outer * 7
+            cse_var_2: T.int32 = rc_outer_outer * 288
+            cse_var_1: T.int32 = ry_outer_outer * 3
             threadIdx_x_1 = T.env_thread(&quot;threadIdx.x&quot;)
-            pad_temp_shared_1 = T.Buffer((324,), data=pad_temp_shared, scope=&quot;shared&quot;)
+            pad_temp_shared_1 = T.Buffer((672,), data=pad_temp_shared, scope=&quot;shared&quot;)
             data_1 = T.Buffer((25088,), data=data.data)
-            with T.launch_thread(threadIdx_x_1, 196):
-                pad_temp_shared_1[threadIdx_x_1] = T.if_then_else(9 &lt;= threadIdx_x_1 % 81 and threadIdx_x_1 % 81 &lt; 72 and 1 &lt;= threadIdx_x_1 % 9 and threadIdx_x_1 % 9 &lt; 8, data_1[rc_outer_outer * 196 + threadIdx_x_1 // 81 * 49 + threadIdx_x_1 % 81 // 9 * 7 + threadIdx_x_1 % 9 - 8], T.float32(0))
-            with T.launch_thread(threadIdx_x_1, 196):
-                if T.likely(threadIdx_x_1 &lt; 128):
-                    pad_temp_shared_1[threadIdx_x_1 + 196] = T.if_then_else(9 &lt;= (threadIdx_x_1 + 34) % 81 and (threadIdx_x_1 + 34) % 81 &lt; 72 and 1 &lt;= (threadIdx_x_1 + 7) % 9 and (threadIdx_x_1 + 7) % 9 &lt; 8, data_1[rc_outer_outer * 196 + (threadIdx_x_1 + 196) // 81 * 49 + (threadIdx_x_1 + 34) % 81 // 9 * 7 + (threadIdx_x_1 + 7) % 9 - 8], T.float32(0))
-            kernel_shared_1 = T.Buffer((576,), data=kernel_shared, scope=&quot;shared&quot;)
-            with T.launch_thread(&quot;threadIdx.x&quot;, 196) as threadIdx_x_2:
-                kernel_1 = T.Buffer((2359296,), data=kernel.data)
-                if T.likely(threadIdx_x_2 &lt; 96):
-                    kernel_shared_1[threadIdx_x_2 * 6] = kernel_1[blockIdx_x * 73728 + threadIdx_x_2 // 6 * 4608 + rc_outer_outer * 36 + threadIdx_x_2 % 6 * 6]
-                if T.likely(threadIdx_x_2 &lt; 96):
-                    kernel_shared_1[threadIdx_x_2 * 6 + 1] = kernel_1[blockIdx_x * 73728 + threadIdx_x_2 // 6 * 4608 + rc_outer_outer * 36 + threadIdx_x_2 % 6 * 6 + 1]
-                if T.likely(threadIdx_x_2 &lt; 96):
-                    kernel_shared_1[threadIdx_x_2 * 6 + 2] = kernel_1[blockIdx_x * 73728 + threadIdx_x_2 // 6 * 4608 + rc_outer_outer * 36 + threadIdx_x_2 % 6 * 6 + 2]
-                if T.likely(threadIdx_x_2 &lt; 96):
-                    kernel_shared_1[threadIdx_x_2 * 6 + 3] = kernel_1[blockIdx_x * 73728 + threadIdx_x_2 // 6 * 4608 + rc_outer_outer * 36 + threadIdx_x_2 % 6 * 6 + 3]
-                if T.likely(threadIdx_x_2 &lt; 96):
-                    kernel_shared_1[threadIdx_x_2 * 6 + 4] = kernel_1[blockIdx_x * 73728 + threadIdx_x_2 // 6 * 4608 + rc_outer_outer * 36 + threadIdx_x_2 % 6 * 6 + 4]
-                if T.likely(threadIdx_x_2 &lt; 96):
-                    kernel_shared_1[threadIdx_x_2 * 6 + 5] = kernel_1[blockIdx_x * 73728 + threadIdx_x_2 // 6 * 4608 + rc_outer_outer * 36 + threadIdx_x_2 % 6 * 6 + 5]
-            for rc_outer_inner, ry_outer_inner in T.grid(2, 3):
-                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3]
-                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 144]
-                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 288]
-                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 432]
-                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 1]
-                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 145]
-                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 289]
-                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 1] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 433]
-                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 2]
-                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 146]
-                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 290]
-                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 2] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 434]
-                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 9]
-                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 153]
-                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 297]
-                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 81] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 441]
-                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 10]
-                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 154]
-                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 298]
-                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 82] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 442]
-                conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 11]
-                conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 155]
-                conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 299]
-                conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[rc_outer_inner * 162 + threadIdx_x % 49 // 7 * 9 + ry_outer_inner * 9 + threadIdx_x % 7 + 83] * kernel_shared_1[threadIdx_x // 49 * 36 + rc_outer_inner * 18 + ry_outer_inner * 3 + 443]
-        compute_1 = T.Buffer((25088,), data=compute.data)
-        bias_1 = T.Buffer((512,), data=bias.data)
-        compute_1[blockIdx_x * 784 + threadIdx_x] = T.max(conv2d_nchw_1[0] + bias_1[blockIdx_x * 16 + threadIdx_x // 49], T.float32(0))
-        compute_1[blockIdx_x * 784 + threadIdx_x + 196] = T.max(conv2d_nchw_1[1] + bias_1[blockIdx_x * 16 + threadIdx_x // 49 + 4], T.float32(0))
-        compute_1[blockIdx_x * 784 + threadIdx_x + 392] = T.max(conv2d_nchw_1[2] + bias_1[blockIdx_x * 16 + threadIdx_x // 49 + 8], T.float32(0))
-        compute_1[blockIdx_x * 784 + threadIdx_x + 588] = T.max(conv2d_nchw_1[3] + bias_1[blockIdx_x * 16 + threadIdx_x // 49 + 12], T.float32(0))
+            with T.launch_thread(threadIdx_x_1, 56):
+                pad_temp_shared_1[threadIdx_x_1] = T.if_then_else(1 &lt;= threadIdx_x_1 % 21 // 3 + ry_outer_outer and threadIdx_x_1 % 21 // 3 + ry_outer_outer &lt; 8 and 1 &lt;= blockIdx_x % 7 + threadIdx_x_1 % 3 and blockIdx_x % 7 + threadIdx_x_1 % 3 &lt; 8, data_1[cse_var_4 + threadIdx_x_1 // 3 * 7 + cse_var_3 + blockIdx_x % 7 + threadIdx_x_1 % 3 - 8], T.float32(0))
+            with T.launch_thread(threadIdx_x_1, 56):
+                pad_temp_shared_1[threadIdx_x_1 + 56] = T.if_then_else(1 &lt;= (threadIdx_x_1 + 14) % 21 // 3 + ry_outer_outer and (threadIdx_x_1 + 14) % 21 // 3 + ry_outer_outer &lt; 8 and 1 &lt;= blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 and blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 &lt; 8, data_1[cse_var_4 + (threadIdx_x_1 + 56) // 3 * 7 + cse_var_3 + blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 - 8], T.float32(0))
+            with T.launch_thread(threadIdx_x_1, 56):
+                pad_temp_shared_1[threadIdx_x_1 + 112] = T.if_then_else(1 &lt;= (threadIdx_x_1 + 7) % 21 // 3 + ry_outer_outer and (threadIdx_x_1 + 7) % 21 // 3 + ry_outer_outer &lt; 8 and 1 &lt;= blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 and blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 &lt; 8, data_1[cse_var_4 + (threadIdx_x_1 + 112) // 3 * 7 + cse_var_3 + blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 - 8], T.float32(0))
+            with T.launch_thread(threadIdx_x_1, 56):
+                pad_temp_shared_1[threadIdx_x_1 + 168] = T.if_then_else(1 &lt;= threadIdx_x_1 % 21 // 3 + ry_outer_outer and threadIdx_x_1 % 21 // 3 + ry_outer_outer &lt; 8 and 1 &lt;= blockIdx_x % 7 + threadIdx_x_1 % 3 and blockIdx_x % 7 + threadIdx_x_1 % 3 &lt; 8, data_1[cse_var_4 + threadIdx_x_1 // 3 * 7 + cse_var_3 + blockIdx_x % 7 + threadIdx_x_1 % 3 + 384], T.float32(0))
+            with T.launch_thread(threadIdx_x_1, 56):
+                pad_temp_shared_1[threadIdx_x_1 + 224] = T.if_then_else(1 &lt;= (threadIdx_x_1 + 14) % 21 // 3 + ry_outer_outer and (threadIdx_x_1 + 14) % 21 // 3 + ry_outer_outer &lt; 8 and 1 &lt;= blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 and blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 &lt; 8, data_1[cse_var_4 + (threadIdx_x_1 + 224) // 3 * 7 + cse_var_3 + blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 - 8], T.float32(0))
+            with T.launch_thread(threadIdx_x_1, 56):
+                pad_temp_shared_1[threadIdx_x_1 + 280] = T.if_then_else(1 &lt;= (threadIdx_x_1 + 7) % 21 // 3 + ry_outer_outer and (threadIdx_x_1 + 7) % 21 // 3 + ry_outer_outer &lt; 8 and 1 &lt;= blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 and blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 &lt; 8, data_1[cse_var_4 + (threadIdx_x_1 + 280) // 3 * 7 + cse_var_3 + blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 - 8], T.float32(0))
+            with T.launch_thread(threadIdx_x_1, 56):
+                pad_temp_shared_1[threadIdx_x_1 + 336] = T.if_then_else(1 &lt;= threadIdx_x_1 % 21 // 3 + ry_outer_outer and threadIdx_x_1 % 21 // 3 + ry_outer_outer &lt; 8 and 1 &lt;= blockIdx_x % 7 + threadIdx_x_1 % 3 and blockIdx_x % 7 + threadIdx_x_1 % 3 &lt; 8, data_1[cse_var_4 + threadIdx_x_1 // 3 * 7 + cse_var_3 + blockIdx_x % 7 + threadIdx_x_1 % 3 + 776], T.float32(0))
+            with T.launch_thread(threadIdx_x_1, 56):
+                pad_temp_shared_1[threadIdx_x_1 + 392] = T.if_then_else(1 &lt;= (threadIdx_x_1 + 14) % 21 // 3 + ry_outer_outer and (threadIdx_x_1 + 14) % 21 // 3 + ry_outer_outer &lt; 8 and 1 &lt;= blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 and blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 &lt; 8, data_1[cse_var_4 + (threadIdx_x_1 + 392) // 3 * 7 + cse_var_3 + blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 - 8], T.float32(0))
+            with T.launch_thread(threadIdx_x_1, 56):
+                pad_temp_shared_1[threadIdx_x_1 + 448] = T.if_then_else(1 &lt;= (threadIdx_x_1 + 7) % 21 // 3 + ry_outer_outer and (threadIdx_x_1 + 7) % 21 // 3 + ry_outer_outer &lt; 8 and 1 &lt;= blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 and blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 &lt; 8, data_1[cse_var_4 + (threadIdx_x_1 + 448) // 3 * 7 + cse_var_3 + blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 - 8], T.float32(0))
+            with T.launch_thread(threadIdx_x_1, 56):
+                pad_temp_shared_1[threadIdx_x_1 + 504] = T.if_then_else(1 &lt;= threadIdx_x_1 % 21 // 3 + ry_outer_outer and threadIdx_x_1 % 21 // 3 + ry_outer_outer &lt; 8 and 1 &lt;= blockIdx_x % 7 + threadIdx_x_1 % 3 and blockIdx_x % 7 + threadIdx_x_1 % 3 &lt; 8, data_1[cse_var_4 + threadIdx_x_1 // 3 * 7 + cse_var_3 + blockIdx_x % 7 + threadIdx_x_1 % 3 + 1168], T.float32(0))
+            with T.launch_thread(threadIdx_x_1, 56):
+                pad_temp_shared_1[threadIdx_x_1 + 560] = T.if_then_else(1 &lt;= (threadIdx_x_1 + 14) % 21 // 3 + ry_outer_outer and (threadIdx_x_1 + 14) % 21 // 3 + ry_outer_outer &lt; 8 and 1 &lt;= blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 and blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 &lt; 8, data_1[cse_var_4 + (threadIdx_x_1 + 560) // 3 * 7 + cse_var_3 + blockIdx_x % 7 + (threadIdx_x_1 + 2) % 3 - 8], T.float32(0))
+            with T.launch_thread(threadIdx_x_1, 56):
+                pad_temp_shared_1[threadIdx_x_1 + 616] = T.if_then_else(1 &lt;= (threadIdx_x_1 + 7) % 21 // 3 + ry_outer_outer and (threadIdx_x_1 + 7) % 21 // 3 + ry_outer_outer &lt; 8 and 1 &lt;= blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 and blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 &lt; 8, data_1[cse_var_4 + (threadIdx_x_1 + 616) // 3 * 7 + cse_var_3 + blockIdx_x % 7 + (threadIdx_x_1 + 1) % 3 - 8], T.float32(0))
+            threadIdx_x_2 = T.env_thread(&quot;threadIdx.x&quot;)
+            kernel_shared_1 = T.Buffer((3072,), data=kernel_shared, scope=&quot;shared&quot;)
+            kernel_1 = T.Buffer((2359296,), data=kernel.data)
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2] = kernel_1[blockIdx_x // 7 * 147456 + cse_var_2 + threadIdx_x_2 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 56] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 56) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 56) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 112] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 112) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 168] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 168) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 // 3 + 24) % 32 * 9 + cse_var_1 + threadIdx_x_2 % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 224] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 224) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 280] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 280) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 88) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 336] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 336) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 // 3 + 16) % 32 * 9 + cse_var_1 + threadIdx_x_2 % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 392] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 392) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 8) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 448] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 448) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 64) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 504] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 504) // 96 * 4608 + cse_var_2 + threadIdx_x_2 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 72]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 560] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 560) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 80) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 616] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 616) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 40) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 672] = kernel_1[blockIdx_x // 7 * 147456 + cse_var_2 + threadIdx_x_2 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 32256]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 728] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 728) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 56) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 784] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 784) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 840] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 840) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 // 3 + 24) % 32 * 9 + cse_var_1 + threadIdx_x_2 % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 896] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 896) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 952] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 952) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 88) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 1008] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1008) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 // 3 + 16) % 32 * 9 + cse_var_1 + threadIdx_x_2 % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 1064] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1064) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 8) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 1120] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1120) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 64) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 1176] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1176) // 96 * 4608 + cse_var_2 + threadIdx_x_2 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 72]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 1232] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1232) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 80) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 1288] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1288) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 40) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 1344] = kernel_1[blockIdx_x // 7 * 147456 + cse_var_2 + threadIdx_x_2 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 64512]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 1400] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1400) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 56) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 1456] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1456) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 1512] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1512) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 // 3 + 24) % 32 * 9 + cse_var_1 + threadIdx_x_2 % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 1568] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1568) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 1624] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1624) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 88) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 1680] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1680) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 // 3 + 16) % 32 * 9 + cse_var_1 + threadIdx_x_2 % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 1736] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1736) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 8) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 1792] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1792) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 64) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 1848] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1848) // 96 * 4608 + cse_var_2 + threadIdx_x_2 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 72]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 1904] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1904) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 80) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 1960] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 1960) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 40) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 2016] = kernel_1[blockIdx_x // 7 * 147456 + cse_var_2 + threadIdx_x_2 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 96768]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 2072] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2072) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 56) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 2128] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2128) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 2184] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2184) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 // 3 + 24) % 32 * 9 + cse_var_1 + threadIdx_x_2 % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 2240] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2240) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 2296] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2296) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 88) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 2352] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2352) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 // 3 + 16) % 32 * 9 + cse_var_1 + threadIdx_x_2 % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 2408] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2408) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 8) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 2464] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2464) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 64) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 2520] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2520) // 96 * 4608 + cse_var_2 + threadIdx_x_2 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 72]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 2576] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2576) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 80) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 2632] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2632) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 40) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 2688] = kernel_1[blockIdx_x // 7 * 147456 + cse_var_2 + threadIdx_x_2 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 129024]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 2744] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2744) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 56) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 2800] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2800) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 16) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 2856] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2856) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 // 3 + 24) % 32 * 9 + cse_var_1 + threadIdx_x_2 % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 2912] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2912) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 32) // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 2) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                kernel_shared_1[threadIdx_x_2 + 2968] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 2968) // 96 * 4608 + cse_var_2 + (threadIdx_x_2 + 88) % 96 // 3 * 9 + cse_var_1 + (threadIdx_x_2 + 1) % 3]
+            with T.launch_thread(threadIdx_x_2, 56):
+                if T.likely(threadIdx_x_2 &lt; 48):
+                    kernel_shared_1[threadIdx_x_2 + 3024] = kernel_1[blockIdx_x // 7 * 147456 + (threadIdx_x_2 + 3024) // 96 * 4608 + cse_var_2 + threadIdx_x_2 // 3 * 9 + cse_var_1 + threadIdx_x_2 % 3 + 144]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3] * kernel_shared_1[threadIdx_x // 7 * 192]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3] * kernel_shared_1[threadIdx_x // 7 * 192 + 1536]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3] * kernel_shared_1[threadIdx_x // 7 * 192 + 96]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3] * kernel_shared_1[threadIdx_x // 7 * 192 + 1632]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 21] * kernel_shared_1[threadIdx_x // 7 * 192 + 3]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 21] * kernel_shared_1[threadIdx_x // 7 * 192 + 1539]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 21] * kernel_shared_1[threadIdx_x // 7 * 192 + 99]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 21] * kernel_shared_1[threadIdx_x // 7 * 192 + 1635]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 42] * kernel_shared_1[threadIdx_x // 7 * 192 + 6]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 42] * kernel_shared_1[threadIdx_x // 7 * 192 + 1542]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 42] * kernel_shared_1[threadIdx_x // 7 * 192 + 102]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 42] * kernel_shared_1[threadIdx_x // 7 * 192 + 1638]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 63] * kernel_shared_1[threadIdx_x // 7 * 192 + 9]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 63] * kernel_shared_1[threadIdx_x // 7 * 192 + 1545]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 63] * kernel_shared_1[threadIdx_x // 7 * 192 + 105]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 63] * kernel_shared_1[threadIdx_x // 7 * 192 + 1641]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 1] * kernel_shared_1[threadIdx_x // 7 * 192 + 1]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 1] * kernel_shared_1[threadIdx_x // 7 * 192 + 1537]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 1] * kernel_shared_1[threadIdx_x // 7 * 192 + 97]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 1] * kernel_shared_1[threadIdx_x // 7 * 192 + 1633]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 22] * kernel_shared_1[threadIdx_x // 7 * 192 + 4]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 22] * kernel_shared_1[threadIdx_x // 7 * 192 + 1540]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 22] * kernel_shared_1[threadIdx_x // 7 * 192 + 100]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 22] * kernel_shared_1[threadIdx_x // 7 * 192 + 1636]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 43] * kernel_shared_1[threadIdx_x // 7 * 192 + 7]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 43] * kernel_shared_1[threadIdx_x // 7 * 192 + 1543]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 43] * kernel_shared_1[threadIdx_x // 7 * 192 + 103]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 43] * kernel_shared_1[threadIdx_x // 7 * 192 + 1639]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 64] * kernel_shared_1[threadIdx_x // 7 * 192 + 10]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 64] * kernel_shared_1[threadIdx_x // 7 * 192 + 1546]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 64] * kernel_shared_1[threadIdx_x // 7 * 192 + 106]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 64] * kernel_shared_1[threadIdx_x // 7 * 192 + 1642]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 2] * kernel_shared_1[threadIdx_x // 7 * 192 + 2]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 2] * kernel_shared_1[threadIdx_x // 7 * 192 + 1538]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 2] * kernel_shared_1[threadIdx_x // 7 * 192 + 98]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 2] * kernel_shared_1[threadIdx_x // 7 * 192 + 1634]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 23] * kernel_shared_1[threadIdx_x // 7 * 192 + 5]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 23] * kernel_shared_1[threadIdx_x // 7 * 192 + 1541]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 23] * kernel_shared_1[threadIdx_x // 7 * 192 + 101]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 23] * kernel_shared_1[threadIdx_x // 7 * 192 + 1637]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 44] * kernel_shared_1[threadIdx_x // 7 * 192 + 8]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 44] * kernel_shared_1[threadIdx_x // 7 * 192 + 1544]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 44] * kernel_shared_1[threadIdx_x // 7 * 192 + 104]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 44] * kernel_shared_1[threadIdx_x // 7 * 192 + 1640]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 65] * kernel_shared_1[threadIdx_x // 7 * 192 + 11]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 65] * kernel_shared_1[threadIdx_x // 7 * 192 + 1547]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 65] * kernel_shared_1[threadIdx_x // 7 * 192 + 107]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 65] * kernel_shared_1[threadIdx_x // 7 * 192 + 1643]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 84] * kernel_shared_1[threadIdx_x // 7 * 192 + 12]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 84] * kernel_shared_1[threadIdx_x // 7 * 192 + 1548]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 84] * kernel_shared_1[threadIdx_x // 7 * 192 + 108]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 84] * kernel_shared_1[threadIdx_x // 7 * 192 + 1644]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 105] * kernel_shared_1[threadIdx_x // 7 * 192 + 15]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 105] * kernel_shared_1[threadIdx_x // 7 * 192 + 1551]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 105] * kernel_shared_1[threadIdx_x // 7 * 192 + 111]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 105] * kernel_shared_1[threadIdx_x // 7 * 192 + 1647]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 126] * kernel_shared_1[threadIdx_x // 7 * 192 + 18]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 126] * kernel_shared_1[threadIdx_x // 7 * 192 + 1554]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 126] * kernel_shared_1[threadIdx_x // 7 * 192 + 114]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 126] * kernel_shared_1[threadIdx_x // 7 * 192 + 1650]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 147] * kernel_shared_1[threadIdx_x // 7 * 192 + 21]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 147] * kernel_shared_1[threadIdx_x // 7 * 192 + 1557]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 147] * kernel_shared_1[threadIdx_x // 7 * 192 + 117]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 147] * kernel_shared_1[threadIdx_x // 7 * 192 + 1653]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 85] * kernel_shared_1[threadIdx_x // 7 * 192 + 13]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 85] * kernel_shared_1[threadIdx_x // 7 * 192 + 1549]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 85] * kernel_shared_1[threadIdx_x // 7 * 192 + 109]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 85] * kernel_shared_1[threadIdx_x // 7 * 192 + 1645]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 106] * kernel_shared_1[threadIdx_x // 7 * 192 + 16]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 106] * kernel_shared_1[threadIdx_x // 7 * 192 + 1552]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 106] * kernel_shared_1[threadIdx_x // 7 * 192 + 112]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 106] * kernel_shared_1[threadIdx_x // 7 * 192 + 1648]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 127] * kernel_shared_1[threadIdx_x // 7 * 192 + 19]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 127] * kernel_shared_1[threadIdx_x // 7 * 192 + 1555]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 127] * kernel_shared_1[threadIdx_x // 7 * 192 + 115]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 127] * kernel_shared_1[threadIdx_x // 7 * 192 + 1651]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 148] * kernel_shared_1[threadIdx_x // 7 * 192 + 22]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 148] * kernel_shared_1[threadIdx_x // 7 * 192 + 1558]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 148] * kernel_shared_1[threadIdx_x // 7 * 192 + 118]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 148] * kernel_shared_1[threadIdx_x // 7 * 192 + 1654]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 86] * kernel_shared_1[threadIdx_x // 7 * 192 + 14]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 86] * kernel_shared_1[threadIdx_x // 7 * 192 + 1550]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 86] * kernel_shared_1[threadIdx_x // 7 * 192 + 110]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 86] * kernel_shared_1[threadIdx_x // 7 * 192 + 1646]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 107] * kernel_shared_1[threadIdx_x // 7 * 192 + 17]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 107] * kernel_shared_1[threadIdx_x // 7 * 192 + 1553]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 107] * kernel_shared_1[threadIdx_x // 7 * 192 + 113]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 107] * kernel_shared_1[threadIdx_x // 7 * 192 + 1649]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 128] * kernel_shared_1[threadIdx_x // 7 * 192 + 20]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 128] * kernel_shared_1[threadIdx_x // 7 * 192 + 1556]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 128] * kernel_shared_1[threadIdx_x // 7 * 192 + 116]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 128] * kernel_shared_1[threadIdx_x // 7 * 192 + 1652]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 149] * kernel_shared_1[threadIdx_x // 7 * 192 + 23]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 149] * kernel_shared_1[threadIdx_x // 7 * 192 + 1559]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 149] * kernel_shared_1[threadIdx_x // 7 * 192 + 119]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 149] * kernel_shared_1[threadIdx_x // 7 * 192 + 1655]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 168] * kernel_shared_1[threadIdx_x // 7 * 192 + 24]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 168] * kernel_shared_1[threadIdx_x // 7 * 192 + 1560]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 168] * kernel_shared_1[threadIdx_x // 7 * 192 + 120]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 168] * kernel_shared_1[threadIdx_x // 7 * 192 + 1656]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 189] * kernel_shared_1[threadIdx_x // 7 * 192 + 27]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 189] * kernel_shared_1[threadIdx_x // 7 * 192 + 1563]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 189] * kernel_shared_1[threadIdx_x // 7 * 192 + 123]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 189] * kernel_shared_1[threadIdx_x // 7 * 192 + 1659]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 210] * kernel_shared_1[threadIdx_x // 7 * 192 + 30]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 210] * kernel_shared_1[threadIdx_x // 7 * 192 + 1566]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 210] * kernel_shared_1[threadIdx_x // 7 * 192 + 126]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 210] * kernel_shared_1[threadIdx_x // 7 * 192 + 1662]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 231] * kernel_shared_1[threadIdx_x // 7 * 192 + 33]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 231] * kernel_shared_1[threadIdx_x // 7 * 192 + 1569]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 231] * kernel_shared_1[threadIdx_x // 7 * 192 + 129]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 231] * kernel_shared_1[threadIdx_x // 7 * 192 + 1665]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 169] * kernel_shared_1[threadIdx_x // 7 * 192 + 25]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 169] * kernel_shared_1[threadIdx_x // 7 * 192 + 1561]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 169] * kernel_shared_1[threadIdx_x // 7 * 192 + 121]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 169] * kernel_shared_1[threadIdx_x // 7 * 192 + 1657]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 190] * kernel_shared_1[threadIdx_x // 7 * 192 + 28]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 190] * kernel_shared_1[threadIdx_x // 7 * 192 + 1564]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 190] * kernel_shared_1[threadIdx_x // 7 * 192 + 124]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 190] * kernel_shared_1[threadIdx_x // 7 * 192 + 1660]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 211] * kernel_shared_1[threadIdx_x // 7 * 192 + 31]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 211] * kernel_shared_1[threadIdx_x // 7 * 192 + 1567]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 211] * kernel_shared_1[threadIdx_x // 7 * 192 + 127]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 211] * kernel_shared_1[threadIdx_x // 7 * 192 + 1663]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 232] * kernel_shared_1[threadIdx_x // 7 * 192 + 34]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 232] * kernel_shared_1[threadIdx_x // 7 * 192 + 1570]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 232] * kernel_shared_1[threadIdx_x // 7 * 192 + 130]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 232] * kernel_shared_1[threadIdx_x // 7 * 192 + 1666]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 170] * kernel_shared_1[threadIdx_x // 7 * 192 + 26]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 170] * kernel_shared_1[threadIdx_x // 7 * 192 + 1562]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 170] * kernel_shared_1[threadIdx_x // 7 * 192 + 122]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 170] * kernel_shared_1[threadIdx_x // 7 * 192 + 1658]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 191] * kernel_shared_1[threadIdx_x // 7 * 192 + 29]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 191] * kernel_shared_1[threadIdx_x // 7 * 192 + 1565]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 191] * kernel_shared_1[threadIdx_x // 7 * 192 + 125]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 191] * kernel_shared_1[threadIdx_x // 7 * 192 + 1661]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 212] * kernel_shared_1[threadIdx_x // 7 * 192 + 32]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 212] * kernel_shared_1[threadIdx_x // 7 * 192 + 1568]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 212] * kernel_shared_1[threadIdx_x // 7 * 192 + 128]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 212] * kernel_shared_1[threadIdx_x // 7 * 192 + 1664]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 233] * kernel_shared_1[threadIdx_x // 7 * 192 + 35]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 233] * kernel_shared_1[threadIdx_x // 7 * 192 + 1571]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 233] * kernel_shared_1[threadIdx_x // 7 * 192 + 131]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 233] * kernel_shared_1[threadIdx_x // 7 * 192 + 1667]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 252] * kernel_shared_1[threadIdx_x // 7 * 192 + 36]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 252] * kernel_shared_1[threadIdx_x // 7 * 192 + 1572]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 252] * kernel_shared_1[threadIdx_x // 7 * 192 + 132]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 252] * kernel_shared_1[threadIdx_x // 7 * 192 + 1668]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 273] * kernel_shared_1[threadIdx_x // 7 * 192 + 39]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 273] * kernel_shared_1[threadIdx_x // 7 * 192 + 1575]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 273] * kernel_shared_1[threadIdx_x // 7 * 192 + 135]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 273] * kernel_shared_1[threadIdx_x // 7 * 192 + 1671]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 294] * kernel_shared_1[threadIdx_x // 7 * 192 + 42]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 294] * kernel_shared_1[threadIdx_x // 7 * 192 + 1578]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 294] * kernel_shared_1[threadIdx_x // 7 * 192 + 138]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 294] * kernel_shared_1[threadIdx_x // 7 * 192 + 1674]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 315] * kernel_shared_1[threadIdx_x // 7 * 192 + 45]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 315] * kernel_shared_1[threadIdx_x // 7 * 192 + 1581]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 315] * kernel_shared_1[threadIdx_x // 7 * 192 + 141]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 315] * kernel_shared_1[threadIdx_x // 7 * 192 + 1677]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 253] * kernel_shared_1[threadIdx_x // 7 * 192 + 37]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 253] * kernel_shared_1[threadIdx_x // 7 * 192 + 1573]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 253] * kernel_shared_1[threadIdx_x // 7 * 192 + 133]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 253] * kernel_shared_1[threadIdx_x // 7 * 192 + 1669]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 274] * kernel_shared_1[threadIdx_x // 7 * 192 + 40]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 274] * kernel_shared_1[threadIdx_x // 7 * 192 + 1576]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 274] * kernel_shared_1[threadIdx_x // 7 * 192 + 136]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 274] * kernel_shared_1[threadIdx_x // 7 * 192 + 1672]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 295] * kernel_shared_1[threadIdx_x // 7 * 192 + 43]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 295] * kernel_shared_1[threadIdx_x // 7 * 192 + 1579]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 295] * kernel_shared_1[threadIdx_x // 7 * 192 + 139]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 295] * kernel_shared_1[threadIdx_x // 7 * 192 + 1675]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 316] * kernel_shared_1[threadIdx_x // 7 * 192 + 46]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 316] * kernel_shared_1[threadIdx_x // 7 * 192 + 1582]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 316] * kernel_shared_1[threadIdx_x // 7 * 192 + 142]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 316] * kernel_shared_1[threadIdx_x // 7 * 192 + 1678]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 254] * kernel_shared_1[threadIdx_x // 7 * 192 + 38]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 254] * kernel_shared_1[threadIdx_x // 7 * 192 + 1574]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 254] * kernel_shared_1[threadIdx_x // 7 * 192 + 134]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 254] * kernel_shared_1[threadIdx_x // 7 * 192 + 1670]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 275] * kernel_shared_1[threadIdx_x // 7 * 192 + 41]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 275] * kernel_shared_1[threadIdx_x // 7 * 192 + 1577]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 275] * kernel_shared_1[threadIdx_x // 7 * 192 + 137]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 275] * kernel_shared_1[threadIdx_x // 7 * 192 + 1673]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 296] * kernel_shared_1[threadIdx_x // 7 * 192 + 44]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 296] * kernel_shared_1[threadIdx_x // 7 * 192 + 1580]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 296] * kernel_shared_1[threadIdx_x // 7 * 192 + 140]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 296] * kernel_shared_1[threadIdx_x // 7 * 192 + 1676]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 317] * kernel_shared_1[threadIdx_x // 7 * 192 + 47]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 317] * kernel_shared_1[threadIdx_x // 7 * 192 + 1583]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 317] * kernel_shared_1[threadIdx_x // 7 * 192 + 143]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 317] * kernel_shared_1[threadIdx_x // 7 * 192 + 1679]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 336] * kernel_shared_1[threadIdx_x // 7 * 192 + 48]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 336] * kernel_shared_1[threadIdx_x // 7 * 192 + 1584]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 336] * kernel_shared_1[threadIdx_x // 7 * 192 + 144]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 336] * kernel_shared_1[threadIdx_x // 7 * 192 + 1680]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 357] * kernel_shared_1[threadIdx_x // 7 * 192 + 51]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 357] * kernel_shared_1[threadIdx_x // 7 * 192 + 1587]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 357] * kernel_shared_1[threadIdx_x // 7 * 192 + 147]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 357] * kernel_shared_1[threadIdx_x // 7 * 192 + 1683]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 378] * kernel_shared_1[threadIdx_x // 7 * 192 + 54]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 378] * kernel_shared_1[threadIdx_x // 7 * 192 + 1590]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 378] * kernel_shared_1[threadIdx_x // 7 * 192 + 150]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 378] * kernel_shared_1[threadIdx_x // 7 * 192 + 1686]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 399] * kernel_shared_1[threadIdx_x // 7 * 192 + 57]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 399] * kernel_shared_1[threadIdx_x // 7 * 192 + 1593]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 399] * kernel_shared_1[threadIdx_x // 7 * 192 + 153]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 399] * kernel_shared_1[threadIdx_x // 7 * 192 + 1689]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 337] * kernel_shared_1[threadIdx_x // 7 * 192 + 49]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 337] * kernel_shared_1[threadIdx_x // 7 * 192 + 1585]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 337] * kernel_shared_1[threadIdx_x // 7 * 192 + 145]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 337] * kernel_shared_1[threadIdx_x // 7 * 192 + 1681]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 358] * kernel_shared_1[threadIdx_x // 7 * 192 + 52]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 358] * kernel_shared_1[threadIdx_x // 7 * 192 + 1588]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 358] * kernel_shared_1[threadIdx_x // 7 * 192 + 148]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 358] * kernel_shared_1[threadIdx_x // 7 * 192 + 1684]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 379] * kernel_shared_1[threadIdx_x // 7 * 192 + 55]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 379] * kernel_shared_1[threadIdx_x // 7 * 192 + 1591]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 379] * kernel_shared_1[threadIdx_x // 7 * 192 + 151]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 379] * kernel_shared_1[threadIdx_x // 7 * 192 + 1687]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 400] * kernel_shared_1[threadIdx_x // 7 * 192 + 58]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 400] * kernel_shared_1[threadIdx_x // 7 * 192 + 1594]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 400] * kernel_shared_1[threadIdx_x // 7 * 192 + 154]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 400] * kernel_shared_1[threadIdx_x // 7 * 192 + 1690]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 338] * kernel_shared_1[threadIdx_x // 7 * 192 + 50]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 338] * kernel_shared_1[threadIdx_x // 7 * 192 + 1586]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 338] * kernel_shared_1[threadIdx_x // 7 * 192 + 146]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 338] * kernel_shared_1[threadIdx_x // 7 * 192 + 1682]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 359] * kernel_shared_1[threadIdx_x // 7 * 192 + 53]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 359] * kernel_shared_1[threadIdx_x // 7 * 192 + 1589]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 359] * kernel_shared_1[threadIdx_x // 7 * 192 + 149]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 359] * kernel_shared_1[threadIdx_x // 7 * 192 + 1685]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 380] * kernel_shared_1[threadIdx_x // 7 * 192 + 56]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 380] * kernel_shared_1[threadIdx_x // 7 * 192 + 1592]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 380] * kernel_shared_1[threadIdx_x // 7 * 192 + 152]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 380] * kernel_shared_1[threadIdx_x // 7 * 192 + 1688]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 401] * kernel_shared_1[threadIdx_x // 7 * 192 + 59]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 401] * kernel_shared_1[threadIdx_x // 7 * 192 + 1595]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 401] * kernel_shared_1[threadIdx_x // 7 * 192 + 155]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 401] * kernel_shared_1[threadIdx_x // 7 * 192 + 1691]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 420] * kernel_shared_1[threadIdx_x // 7 * 192 + 60]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 420] * kernel_shared_1[threadIdx_x // 7 * 192 + 1596]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 420] * kernel_shared_1[threadIdx_x // 7 * 192 + 156]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 420] * kernel_shared_1[threadIdx_x // 7 * 192 + 1692]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 441] * kernel_shared_1[threadIdx_x // 7 * 192 + 63]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 441] * kernel_shared_1[threadIdx_x // 7 * 192 + 1599]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 441] * kernel_shared_1[threadIdx_x // 7 * 192 + 159]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 441] * kernel_shared_1[threadIdx_x // 7 * 192 + 1695]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 462] * kernel_shared_1[threadIdx_x // 7 * 192 + 66]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 462] * kernel_shared_1[threadIdx_x // 7 * 192 + 1602]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 462] * kernel_shared_1[threadIdx_x // 7 * 192 + 162]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 462] * kernel_shared_1[threadIdx_x // 7 * 192 + 1698]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 483] * kernel_shared_1[threadIdx_x // 7 * 192 + 69]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 483] * kernel_shared_1[threadIdx_x // 7 * 192 + 1605]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 483] * kernel_shared_1[threadIdx_x // 7 * 192 + 165]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 483] * kernel_shared_1[threadIdx_x // 7 * 192 + 1701]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 421] * kernel_shared_1[threadIdx_x // 7 * 192 + 61]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 421] * kernel_shared_1[threadIdx_x // 7 * 192 + 1597]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 421] * kernel_shared_1[threadIdx_x // 7 * 192 + 157]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 421] * kernel_shared_1[threadIdx_x // 7 * 192 + 1693]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 442] * kernel_shared_1[threadIdx_x // 7 * 192 + 64]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 442] * kernel_shared_1[threadIdx_x // 7 * 192 + 1600]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 442] * kernel_shared_1[threadIdx_x // 7 * 192 + 160]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 442] * kernel_shared_1[threadIdx_x // 7 * 192 + 1696]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 463] * kernel_shared_1[threadIdx_x // 7 * 192 + 67]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 463] * kernel_shared_1[threadIdx_x // 7 * 192 + 1603]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 463] * kernel_shared_1[threadIdx_x // 7 * 192 + 163]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 463] * kernel_shared_1[threadIdx_x // 7 * 192 + 1699]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 484] * kernel_shared_1[threadIdx_x // 7 * 192 + 70]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 484] * kernel_shared_1[threadIdx_x // 7 * 192 + 1606]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 484] * kernel_shared_1[threadIdx_x // 7 * 192 + 166]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 484] * kernel_shared_1[threadIdx_x // 7 * 192 + 1702]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 422] * kernel_shared_1[threadIdx_x // 7 * 192 + 62]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 422] * kernel_shared_1[threadIdx_x // 7 * 192 + 1598]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 422] * kernel_shared_1[threadIdx_x // 7 * 192 + 158]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 422] * kernel_shared_1[threadIdx_x // 7 * 192 + 1694]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 443] * kernel_shared_1[threadIdx_x // 7 * 192 + 65]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 443] * kernel_shared_1[threadIdx_x // 7 * 192 + 1601]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 443] * kernel_shared_1[threadIdx_x // 7 * 192 + 161]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 443] * kernel_shared_1[threadIdx_x // 7 * 192 + 1697]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 464] * kernel_shared_1[threadIdx_x // 7 * 192 + 68]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 464] * kernel_shared_1[threadIdx_x // 7 * 192 + 1604]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 464] * kernel_shared_1[threadIdx_x // 7 * 192 + 164]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 464] * kernel_shared_1[threadIdx_x // 7 * 192 + 1700]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 485] * kernel_shared_1[threadIdx_x // 7 * 192 + 71]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 485] * kernel_shared_1[threadIdx_x // 7 * 192 + 1607]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 485] * kernel_shared_1[threadIdx_x // 7 * 192 + 167]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 485] * kernel_shared_1[threadIdx_x // 7 * 192 + 1703]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 504] * kernel_shared_1[threadIdx_x // 7 * 192 + 72]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 504] * kernel_shared_1[threadIdx_x // 7 * 192 + 1608]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 504] * kernel_shared_1[threadIdx_x // 7 * 192 + 168]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 504] * kernel_shared_1[threadIdx_x // 7 * 192 + 1704]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 525] * kernel_shared_1[threadIdx_x // 7 * 192 + 75]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 525] * kernel_shared_1[threadIdx_x // 7 * 192 + 1611]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 525] * kernel_shared_1[threadIdx_x // 7 * 192 + 171]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 525] * kernel_shared_1[threadIdx_x // 7 * 192 + 1707]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 546] * kernel_shared_1[threadIdx_x // 7 * 192 + 78]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 546] * kernel_shared_1[threadIdx_x // 7 * 192 + 1614]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 546] * kernel_shared_1[threadIdx_x // 7 * 192 + 174]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 546] * kernel_shared_1[threadIdx_x // 7 * 192 + 1710]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 567] * kernel_shared_1[threadIdx_x // 7 * 192 + 81]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 567] * kernel_shared_1[threadIdx_x // 7 * 192 + 1617]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 567] * kernel_shared_1[threadIdx_x // 7 * 192 + 177]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 567] * kernel_shared_1[threadIdx_x // 7 * 192 + 1713]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 505] * kernel_shared_1[threadIdx_x // 7 * 192 + 73]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 505] * kernel_shared_1[threadIdx_x // 7 * 192 + 1609]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 505] * kernel_shared_1[threadIdx_x // 7 * 192 + 169]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 505] * kernel_shared_1[threadIdx_x // 7 * 192 + 1705]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 526] * kernel_shared_1[threadIdx_x // 7 * 192 + 76]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 526] * kernel_shared_1[threadIdx_x // 7 * 192 + 1612]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 526] * kernel_shared_1[threadIdx_x // 7 * 192 + 172]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 526] * kernel_shared_1[threadIdx_x // 7 * 192 + 1708]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 547] * kernel_shared_1[threadIdx_x // 7 * 192 + 79]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 547] * kernel_shared_1[threadIdx_x // 7 * 192 + 1615]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 547] * kernel_shared_1[threadIdx_x // 7 * 192 + 175]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 547] * kernel_shared_1[threadIdx_x // 7 * 192 + 1711]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 568] * kernel_shared_1[threadIdx_x // 7 * 192 + 82]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 568] * kernel_shared_1[threadIdx_x // 7 * 192 + 1618]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 568] * kernel_shared_1[threadIdx_x // 7 * 192 + 178]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 568] * kernel_shared_1[threadIdx_x // 7 * 192 + 1714]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 506] * kernel_shared_1[threadIdx_x // 7 * 192 + 74]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 506] * kernel_shared_1[threadIdx_x // 7 * 192 + 1610]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 506] * kernel_shared_1[threadIdx_x // 7 * 192 + 170]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 506] * kernel_shared_1[threadIdx_x // 7 * 192 + 1706]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 527] * kernel_shared_1[threadIdx_x // 7 * 192 + 77]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 527] * kernel_shared_1[threadIdx_x // 7 * 192 + 1613]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 527] * kernel_shared_1[threadIdx_x // 7 * 192 + 173]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 527] * kernel_shared_1[threadIdx_x // 7 * 192 + 1709]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 548] * kernel_shared_1[threadIdx_x // 7 * 192 + 80]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 548] * kernel_shared_1[threadIdx_x // 7 * 192 + 1616]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 548] * kernel_shared_1[threadIdx_x // 7 * 192 + 176]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 548] * kernel_shared_1[threadIdx_x // 7 * 192 + 1712]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 569] * kernel_shared_1[threadIdx_x // 7 * 192 + 83]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 569] * kernel_shared_1[threadIdx_x // 7 * 192 + 1619]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 569] * kernel_shared_1[threadIdx_x // 7 * 192 + 179]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 569] * kernel_shared_1[threadIdx_x // 7 * 192 + 1715]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 588] * kernel_shared_1[threadIdx_x // 7 * 192 + 84]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 588] * kernel_shared_1[threadIdx_x // 7 * 192 + 1620]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 588] * kernel_shared_1[threadIdx_x // 7 * 192 + 180]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 588] * kernel_shared_1[threadIdx_x // 7 * 192 + 1716]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 609] * kernel_shared_1[threadIdx_x // 7 * 192 + 87]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 609] * kernel_shared_1[threadIdx_x // 7 * 192 + 1623]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 609] * kernel_shared_1[threadIdx_x // 7 * 192 + 183]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 609] * kernel_shared_1[threadIdx_x // 7 * 192 + 1719]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 630] * kernel_shared_1[threadIdx_x // 7 * 192 + 90]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 630] * kernel_shared_1[threadIdx_x // 7 * 192 + 1626]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 630] * kernel_shared_1[threadIdx_x // 7 * 192 + 186]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 630] * kernel_shared_1[threadIdx_x // 7 * 192 + 1722]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 651] * kernel_shared_1[threadIdx_x // 7 * 192 + 93]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 651] * kernel_shared_1[threadIdx_x // 7 * 192 + 1629]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 651] * kernel_shared_1[threadIdx_x // 7 * 192 + 189]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 651] * kernel_shared_1[threadIdx_x // 7 * 192 + 1725]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 589] * kernel_shared_1[threadIdx_x // 7 * 192 + 85]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 589] * kernel_shared_1[threadIdx_x // 7 * 192 + 1621]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 589] * kernel_shared_1[threadIdx_x // 7 * 192 + 181]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 589] * kernel_shared_1[threadIdx_x // 7 * 192 + 1717]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 610] * kernel_shared_1[threadIdx_x // 7 * 192 + 88]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 610] * kernel_shared_1[threadIdx_x // 7 * 192 + 1624]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 610] * kernel_shared_1[threadIdx_x // 7 * 192 + 184]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 610] * kernel_shared_1[threadIdx_x // 7 * 192 + 1720]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 631] * kernel_shared_1[threadIdx_x // 7 * 192 + 91]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 631] * kernel_shared_1[threadIdx_x // 7 * 192 + 1627]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 631] * kernel_shared_1[threadIdx_x // 7 * 192 + 187]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 631] * kernel_shared_1[threadIdx_x // 7 * 192 + 1723]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 652] * kernel_shared_1[threadIdx_x // 7 * 192 + 94]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 652] * kernel_shared_1[threadIdx_x // 7 * 192 + 1630]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 652] * kernel_shared_1[threadIdx_x // 7 * 192 + 190]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 652] * kernel_shared_1[threadIdx_x // 7 * 192 + 1726]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 590] * kernel_shared_1[threadIdx_x // 7 * 192 + 86]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 590] * kernel_shared_1[threadIdx_x // 7 * 192 + 1622]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 590] * kernel_shared_1[threadIdx_x // 7 * 192 + 182]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 590] * kernel_shared_1[threadIdx_x // 7 * 192 + 1718]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 611] * kernel_shared_1[threadIdx_x // 7 * 192 + 89]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 611] * kernel_shared_1[threadIdx_x // 7 * 192 + 1625]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 611] * kernel_shared_1[threadIdx_x // 7 * 192 + 185]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 611] * kernel_shared_1[threadIdx_x // 7 * 192 + 1721]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 632] * kernel_shared_1[threadIdx_x // 7 * 192 + 92]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 632] * kernel_shared_1[threadIdx_x // 7 * 192 + 1628]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 632] * kernel_shared_1[threadIdx_x // 7 * 192 + 188]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 632] * kernel_shared_1[threadIdx_x // 7 * 192 + 1724]
+            conv2d_nchw_1[0] = conv2d_nchw_1[0] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 653] * kernel_shared_1[threadIdx_x // 7 * 192 + 95]
+            conv2d_nchw_1[2] = conv2d_nchw_1[2] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 653] * kernel_shared_1[threadIdx_x // 7 * 192 + 1631]
+            conv2d_nchw_1[1] = conv2d_nchw_1[1] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 653] * kernel_shared_1[threadIdx_x // 7 * 192 + 191]
+            conv2d_nchw_1[3] = conv2d_nchw_1[3] + pad_temp_shared_1[threadIdx_x % 7 * 3 + 653] * kernel_shared_1[threadIdx_x // 7 * 192 + 1727]
+        for i1_inner in range(2):
+            compute_1 = T.Buffer((25088,), data=compute.data)
+            bias_1 = T.Buffer((512,), data=bias.data)
+            compute_1[blockIdx_x // 7 * 1568 + threadIdx_x // 7 * 98 + i1_inner * 49 + threadIdx_x % 7 * 7 + blockIdx_x % 7] = T.max(conv2d_nchw_1[i1_inner] + bias_1[blockIdx_x // 7 * 32 + threadIdx_x // 7 * 2 + i1_inner], T.float32(0))
+            compute_1[blockIdx_x // 7 * 1568 + threadIdx_x // 7 * 98 + i1_inner * 49 + threadIdx_x % 7 * 7 + blockIdx_x % 7 + 784] = T.max(conv2d_nchw_1[i1_inner + 2] + bias_1[blockIdx_x // 7 * 32 + threadIdx_x // 7 * 2 + i1_inner + 16], T.float32(0))
 </pre></div>
 </div>
 </div>
@@ -608,7 +1088,7 @@ class Module:
 <span class="p">)</span>
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 0.257 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 0.419 ms
 </pre></div>
 </div>
 </div>
@@ -637,36 +1117,36 @@ conv2d_nchw_nn_o_i, conv2d_nchw_nn_i = s[conv2d_nchw].split(conv2d_nchw_nn, fact
 conv2d_nchw_nn_o_o_i, conv2d_nchw_nn_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_i, factor=1)
 conv2d_nchw_nn_o_o_o_i, conv2d_nchw_nn_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_i, factor=1)
 conv2d_nchw_nn_o_o_o_o, conv2d_nchw_nn_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_nn_o_o_o_i, factor=1)
-conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=1)
+conv2d_nchw_ff_o_i, conv2d_nchw_ff_i = s[conv2d_nchw].split(conv2d_nchw_ff, factor=2)
 conv2d_nchw_ff_o_o_i, conv2d_nchw_ff_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_i, factor=1)
-conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=4)
-conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=4)
+conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=8)
+conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=2)
 conv2d_nchw_yy_o_i, conv2d_nchw_yy_i = s[conv2d_nchw].split(conv2d_nchw_yy, factor=1)
 conv2d_nchw_yy_o_o_i, conv2d_nchw_yy_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_i, factor=1)
 conv2d_nchw_yy_o_o_o_i, conv2d_nchw_yy_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_i, factor=7)
 conv2d_nchw_yy_o_o_o_o, conv2d_nchw_yy_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_o_i, factor=1)
 conv2d_nchw_xx_o_i, conv2d_nchw_xx_i = s[conv2d_nchw].split(conv2d_nchw_xx, factor=1)
 conv2d_nchw_xx_o_o_i, conv2d_nchw_xx_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_i, factor=1)
-conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, factor=7)
+conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, factor=1)
 conv2d_nchw_xx_o_o_o_o, conv2d_nchw_xx_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_o_i, factor=1)
-conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=2)
-conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=2)
+conv2d_nchw_rc_o_i, conv2d_nchw_rc_i = s[conv2d_nchw].split(conv2d_nchw_rc, factor=4)
+conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=8)
 conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=1)
-conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=3)
-conv2d_nchw_rx_o_i, conv2d_nchw_rx_i = s[conv2d_nchw].split(conv2d_nchw_rx, factor=3)
-conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=1)
+conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=1)
+conv2d_nchw_rx_o_i, conv2d_nchw_rx_i = s[conv2d_nchw].split(conv2d_nchw_rx, factor=1)
+conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=3)
 s[conv2d_nchw].reorder(conv2d_nchw_nn_o_o_o_o, conv2d_nchw_ff_o_o_o_o, conv2d_nchw_yy_o_o_o_o, conv2d_nchw_xx_o_o_o_o, conv2d_nchw_nn_o_o_o_i, conv2d_nchw_ff_o_o_o_i, conv2d_nchw_yy_o_o_o_i, conv2d_nchw_xx_o_o_o_i, conv2d_nchw_nn_o_o_i, conv2d_nchw_ff_o_o_i, conv2d_nchw_yy_o_o_i, conv2d_nchw_xx_o_o_i, conv2d_nchw_rc_o_o, conv2d_nchw_ry_o_o, conv2d_nchw_rx_o_o, conv2d_nchw_rc_o_i, conv2d_nchw_ry_o_i, conv2d_nchw_rx_o_i, conv2d_nchw_nn_o_i, conv2d_nchw_ff_o_i, conv2d_nchw_yy_o_i, conv2d_nc [...]
 compute_i0_o_i, compute_i0_i = s[compute].split(compute_i0, factor=1)
 compute_i0_o_o_i, compute_i0_o_i = s[compute].split(compute_i0_o_i, factor=1)
 compute_i0_o_o_o, compute_i0_o_o_i = s[compute].split(compute_i0_o_o_i, factor=1)
-compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=1)
-compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=4)
-compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=4)
+compute_i1_o_i, compute_i1_i = s[compute].split(compute_i1, factor=2)
+compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=8)
+compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=2)
 compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, factor=1)
 compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=7)
 compute_i2_o_o_o, compute_i2_o_o_i = s[compute].split(compute_i2_o_o_i, factor=1)
 compute_i3_o_i, compute_i3_i = s[compute].split(compute_i3, factor=1)
-compute_i3_o_o_i, compute_i3_o_i = s[compute].split(compute_i3_o_i, factor=7)
+compute_i3_o_o_i, compute_i3_o_i = s[compute].split(compute_i3_o_i, factor=1)
 compute_i3_o_o_o, compute_i3_o_o_i = s[compute].split(compute_i3_o_o_i, factor=1)
 s[compute].reorder(compute_i0_o_o_o, compute_i1_o_o_o, compute_i2_o_o_o, compute_i3_o_o_o, compute_i0_o_o_i, compute_i1_o_o_i, compute_i2_o_o_i, compute_i3_o_o_i, compute_i0_o_i, compute_i1_o_i, compute_i2_o_i, compute_i3_o_i, compute_i0_i, compute_i1_i, compute_i2_i, compute_i3_i)
 s[conv2d_nchw].compute_at(s[compute], compute_i3_o_i)
@@ -684,16 +1164,16 @@ s[compute].bind(compute_i0_o_o_i_i1_o_o_i_fused_i2_o_o_i_fused_i3_o_o_i_fused, t
 compute_i0_o_i_i1_o_i_fused_i2_o_i_fused_i3_o_i_fused = s[compute].fuse(compute_i0_o_i, compute_i1_o_i, compute_i2_o_i, compute_i3_o_i)
 s[compute].bind(compute_i0_o_i_i1_o_i_fused_i2_o_i_fused_i3_o_i_fused, te.thread_axis(&quot;threadIdx.x&quot;))
 kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[kernel_shared].fuse(kernel_shared_ax0, kernel_shared_ax1, kernel_shared_ax2, kernel_shared_ax3)
-kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=6)
+kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
 s[kernel_shared].vectorize(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
-kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=196)
+kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[kernel_shared].split(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=56)
 s[kernel_shared].bind(kernel_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis(&quot;threadIdx.x&quot;))
 pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused = s[pad_temp_shared].fuse(pad_temp_shared_ax0, pad_temp_shared_ax1, pad_temp_shared_ax2, pad_temp_shared_ax3)
 pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused, factor=1)
 s[pad_temp_shared].vectorize(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_i)
-pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=196)
+pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_o, pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i = s[pad_temp_shared].split(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o, factor=56)
 s[pad_temp_shared].bind(pad_temp_shared_ax0_ax1_fused_ax2_fused_ax3_fused_o_i, te.thread_axis(&quot;threadIdx.x&quot;))
-s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, &quot;auto_unroll_max_step&quot;, 64)
+s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, &quot;auto_unroll_max_step&quot;, 1024)
 s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, &quot;unroll_explicit&quot;, True)
 
 CUDA source code:
@@ -711,72 +1191,477 @@ CUDA source code:
   #define int64_t long long
   #define uint64_t unsigned long long
 #endif
-extern &quot;C&quot; __global__ void __launch_bounds__(196) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
+extern &quot;C&quot; __global__ void __launch_bounds__(56) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
   float conv2d_nchw[4];
-  __shared__ float pad_temp_shared[324];
-  __shared__ float kernel_shared[576];
+  __shared__ float pad_temp_shared[672];
+  __shared__ float kernel_shared[3072];
   conv2d_nchw[0] = 0.000000e+00f;
-  conv2d_nchw[1] = 0.000000e+00f;
   conv2d_nchw[2] = 0.000000e+00f;
+  conv2d_nchw[1] = 0.000000e+00f;
   conv2d_nchw[3] = 0.000000e+00f;
-  for (int rc_outer_outer = 0; rc_outer_outer &lt; 128; ++rc_outer_outer) {
-    __syncthreads();
-    pad_temp_shared[((int)threadIdx.x)] = (((((9 &lt;= (((int)threadIdx.x) % 81)) &amp;&amp; ((((int)threadIdx.x) % 81) &lt; 72)) &amp;&amp; (1 &lt;= (((int)threadIdx.x) % 9))) &amp;&amp; ((((int)threadIdx.x) % 9) &lt; 8)) ? data[(((((rc_outer_outer * 196) + ((((int)threadIdx.x) / 81) * 49)) + (((((int)threadIdx.x) % 81) / 9) * 7)) + (((int)threadIdx.x) % 9)) - 8)] : 0.000000e+00f);
-    if (((int)threadIdx.x) &lt; 128) {
-      pad_temp_shared[(((int)threadIdx.x) + 196)] = (((((9 &lt;= ((((int)threadIdx.x) + 34) % 81)) &amp;&amp; (((((int)threadIdx.x) + 34) % 81) &lt; 72)) &amp;&amp; (1 &lt;= ((((int)threadIdx.x) + 7) % 9))) &amp;&amp; (((((int)threadIdx.x) + 7) % 9) &lt; 8)) ? data[(((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 196) / 81) * 49)) + ((((((int)threadIdx.x) + 34) % 81) / 9) * 7)) + ((((int)threadIdx.x) + 7) % 9)) - 8)] : 0.000000e+00f);
-    }
-    if (((int)threadIdx.x) &lt; 96) {
-      kernel_shared[(((int)threadIdx.x) * 6)] = kernel[((((((int)blockIdx.x) * 73728) + ((((int)threadIdx.x) / 6) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) % 6) * 6))];
-    }
-    if (((int)threadIdx.x) &lt; 96) {
-      kernel_shared[((((int)threadIdx.x) * 6) + 1)] = kernel[(((((((int)blockIdx.x) * 73728) + ((((int)threadIdx.x) / 6) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) % 6) * 6)) + 1)];
-    }
-    if (((int)threadIdx.x) &lt; 96) {
-      kernel_shared[((((int)threadIdx.x) * 6) + 2)] = kernel[(((((((int)blockIdx.x) * 73728) + ((((int)threadIdx.x) / 6) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) % 6) * 6)) + 2)];
-    }
-    if (((int)threadIdx.x) &lt; 96) {
-      kernel_shared[((((int)threadIdx.x) * 6) + 3)] = kernel[(((((((int)blockIdx.x) * 73728) + ((((int)threadIdx.x) / 6) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) % 6) * 6)) + 3)];
-    }
-    if (((int)threadIdx.x) &lt; 96) {
-      kernel_shared[((((int)threadIdx.x) * 6) + 4)] = kernel[(((((((int)blockIdx.x) * 73728) + ((((int)threadIdx.x) / 6) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) % 6) * 6)) + 4)];
-    }
-    if (((int)threadIdx.x) &lt; 96) {
-      kernel_shared[((((int)threadIdx.x) * 6) + 5)] = kernel[(((((((int)blockIdx.x) * 73728) + ((((int)threadIdx.x) / 6) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) % 6) * 6)) + 5)];
-    }
-    __syncthreads();
-    for (int rc_outer_inner = 0; rc_outer_inner &lt; 2; ++rc_outer_inner) {
-      for (int ry_outer_inner = 0; ry_outer_inner &lt; 3; ++ry_outer_inner) {
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3))]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 144)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 288)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7))] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 432)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 1)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 145)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 289)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 433)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 2)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 146)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 290)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 434)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 9)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 153)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 297)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 81)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 441)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 10)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 154)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 298)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 82)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 442)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 11)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 155)]));
-        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 299)]));
-        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((rc_outer_inner * 162) + (((((int)threadIdx.x) % 49) / 7) * 9)) + (ry_outer_inner * 9)) + (((int)threadIdx.x) % 7)) + 83)] * kernel_shared[(((((((int)threadIdx.x) / 49) * 36) + (rc_outer_inner * 18)) + (ry_outer_inner * 3)) + 443)]));
+  for (int rc_outer_outer = 0; rc_outer_outer &lt; 16; ++rc_outer_outer) {
+    for (int ry_outer_outer = 0; ry_outer_outer &lt; 3; ++ry_outer_outer) {
+      __syncthreads();
+      pad_temp_shared[((int)threadIdx.x)] = (((((1 &lt;= (((((int)threadIdx.x) % 21) / 3) + ry_outer_outer)) &amp;&amp; ((((((int)threadIdx.x) % 21) / 3) + ry_outer_outer) &lt; 8)) &amp;&amp; (1 &lt;= ((((int)blockIdx.x) % 7) + (((int)threadIdx.x) % 3)))) &amp;&amp; (((((int)blockIdx.x) % 7) + (((int)threadIdx.x) % 3)) &lt; 8)) ? data[((((((rc_outer_outer * 1568) + ((((int)threadIdx.x) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + (((int)threadIdx.x) % 3)) - 8)] : 0.000 [...]
+      pad_temp_shared[(((int)threadIdx.x) + 56)] = (((((1 &lt;= ((((((int)threadIdx.x) + 14) % 21) / 3) + ry_outer_outer)) &amp;&amp; (((((((int)threadIdx.x) + 14) % 21) / 3) + ry_outer_outer) &lt; 8)) &amp;&amp; (1 &lt;= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 2) % 3)))) &amp;&amp; (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 2) % 3)) &lt; 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 56) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + [...]
+      pad_temp_shared[(((int)threadIdx.x) + 112)] = (((((1 &lt;= ((((((int)threadIdx.x) + 7) % 21) / 3) + ry_outer_outer)) &amp;&amp; (((((((int)threadIdx.x) + 7) % 21) / 3) + ry_outer_outer) &lt; 8)) &amp;&amp; (1 &lt;= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 1) % 3)))) &amp;&amp; (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 1) % 3)) &lt; 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 112) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + [...]
+      pad_temp_shared[(((int)threadIdx.x) + 168)] = (((((1 &lt;= (((((int)threadIdx.x) % 21) / 3) + ry_outer_outer)) &amp;&amp; ((((((int)threadIdx.x) % 21) / 3) + ry_outer_outer) &lt; 8)) &amp;&amp; (1 &lt;= ((((int)blockIdx.x) % 7) + (((int)threadIdx.x) % 3)))) &amp;&amp; (((((int)blockIdx.x) % 7) + (((int)threadIdx.x) % 3)) &lt; 8)) ? data[((((((rc_outer_outer * 1568) + ((((int)threadIdx.x) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + (((int)threadIdx.x) % 3)) + 384 [...]
+      pad_temp_shared[(((int)threadIdx.x) + 224)] = (((((1 &lt;= ((((((int)threadIdx.x) + 14) % 21) / 3) + ry_outer_outer)) &amp;&amp; (((((((int)threadIdx.x) + 14) % 21) / 3) + ry_outer_outer) &lt; 8)) &amp;&amp; (1 &lt;= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 2) % 3)))) &amp;&amp; (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 2) % 3)) &lt; 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 224) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) [...]
+      pad_temp_shared[(((int)threadIdx.x) + 280)] = (((((1 &lt;= ((((((int)threadIdx.x) + 7) % 21) / 3) + ry_outer_outer)) &amp;&amp; (((((((int)threadIdx.x) + 7) % 21) / 3) + ry_outer_outer) &lt; 8)) &amp;&amp; (1 &lt;= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 1) % 3)))) &amp;&amp; (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 1) % 3)) &lt; 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 280) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + [...]
+      pad_temp_shared[(((int)threadIdx.x) + 336)] = (((((1 &lt;= (((((int)threadIdx.x) % 21) / 3) + ry_outer_outer)) &amp;&amp; ((((((int)threadIdx.x) % 21) / 3) + ry_outer_outer) &lt; 8)) &amp;&amp; (1 &lt;= ((((int)blockIdx.x) % 7) + (((int)threadIdx.x) % 3)))) &amp;&amp; (((((int)blockIdx.x) % 7) + (((int)threadIdx.x) % 3)) &lt; 8)) ? data[((((((rc_outer_outer * 1568) + ((((int)threadIdx.x) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + (((int)threadIdx.x) % 3)) + 776 [...]
+      pad_temp_shared[(((int)threadIdx.x) + 392)] = (((((1 &lt;= ((((((int)threadIdx.x) + 14) % 21) / 3) + ry_outer_outer)) &amp;&amp; (((((((int)threadIdx.x) + 14) % 21) / 3) + ry_outer_outer) &lt; 8)) &amp;&amp; (1 &lt;= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 2) % 3)))) &amp;&amp; (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 2) % 3)) &lt; 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 392) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) [...]
+      pad_temp_shared[(((int)threadIdx.x) + 448)] = (((((1 &lt;= ((((((int)threadIdx.x) + 7) % 21) / 3) + ry_outer_outer)) &amp;&amp; (((((((int)threadIdx.x) + 7) % 21) / 3) + ry_outer_outer) &lt; 8)) &amp;&amp; (1 &lt;= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 1) % 3)))) &amp;&amp; (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 1) % 3)) &lt; 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 448) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + [...]
+      pad_temp_shared[(((int)threadIdx.x) + 504)] = (((((1 &lt;= (((((int)threadIdx.x) % 21) / 3) + ry_outer_outer)) &amp;&amp; ((((((int)threadIdx.x) % 21) / 3) + ry_outer_outer) &lt; 8)) &amp;&amp; (1 &lt;= ((((int)blockIdx.x) % 7) + (((int)threadIdx.x) % 3)))) &amp;&amp; (((((int)blockIdx.x) % 7) + (((int)threadIdx.x) % 3)) &lt; 8)) ? data[((((((rc_outer_outer * 1568) + ((((int)threadIdx.x) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + (((int)threadIdx.x) % 3)) + 116 [...]
+      pad_temp_shared[(((int)threadIdx.x) + 560)] = (((((1 &lt;= ((((((int)threadIdx.x) + 14) % 21) / 3) + ry_outer_outer)) &amp;&amp; (((((((int)threadIdx.x) + 14) % 21) / 3) + ry_outer_outer) &lt; 8)) &amp;&amp; (1 &lt;= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 2) % 3)))) &amp;&amp; (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 2) % 3)) &lt; 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 560) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) [...]
+      pad_temp_shared[(((int)threadIdx.x) + 616)] = (((((1 &lt;= ((((((int)threadIdx.x) + 7) % 21) / 3) + ry_outer_outer)) &amp;&amp; (((((((int)threadIdx.x) + 7) % 21) / 3) + ry_outer_outer) &lt; 8)) &amp;&amp; (1 &lt;= ((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 1) % 3)))) &amp;&amp; (((((int)blockIdx.x) % 7) + ((((int)threadIdx.x) + 1) % 3)) &lt; 8)) ? data[((((((rc_outer_outer * 1568) + (((((int)threadIdx.x) + 616) / 3) * 7)) + (ry_outer_outer * 7)) + (((int)blockIdx.x) % 7)) + [...]
+      kernel_shared[((int)threadIdx.x)] = kernel[((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 56)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 56) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 56) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 112)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 112) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 16) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 168)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 168) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 24) &amp; 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 224)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 224) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 280)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 280) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 88) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 336)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 336) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 16) &amp; 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 392)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 392) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 8) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 448)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 448) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 64) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 504)] = kernel[((((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 504) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 72)];
+      kernel_shared[(((int)threadIdx.x) + 560)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 560) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 80) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 616)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 616) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 40) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 672)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 32256)];
+      kernel_shared[(((int)threadIdx.x) + 728)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 728) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 56) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 784)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 784) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 16) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 840)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 840) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 24) &amp; 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 896)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 896) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 952)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 952) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 88) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 1008)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1008) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 16) &amp; 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 1064)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1064) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 8) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 1120)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1120) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 64) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 1176)] = kernel[((((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1176) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 72)];
+      kernel_shared[(((int)threadIdx.x) + 1232)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1232) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 80) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 1288)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1288) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 40) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 1344)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 64512)];
+      kernel_shared[(((int)threadIdx.x) + 1400)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1400) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 56) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 1456)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1456) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 16) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 1512)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1512) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 24) &amp; 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 1568)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1568) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 1624)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1624) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 88) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 1680)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1680) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 16) &amp; 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 1736)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1736) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 8) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 1792)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1792) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 64) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 1848)] = kernel[((((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1848) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 72)];
+      kernel_shared[(((int)threadIdx.x) + 1904)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1904) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 80) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 1960)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 1960) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 40) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 2016)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 96768)];
+      kernel_shared[(((int)threadIdx.x) + 2072)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2072) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 56) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 2128)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2128) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 16) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 2184)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2184) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 24) &amp; 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 2240)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2240) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 2296)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2296) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 88) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 2352)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2352) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 16) &amp; 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 2408)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2408) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 8) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 2464)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2464) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 64) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 2520)] = kernel[((((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2520) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 72)];
+      kernel_shared[(((int)threadIdx.x) + 2576)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2576) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 80) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 2632)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2632) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 40) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 2688)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 129024)];
+      kernel_shared[(((int)threadIdx.x) + 2744)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2744) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 56) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 2800)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2800) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 16) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 2856)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2856) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) / 3) + 24) &amp; 31) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 2912)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2912) / 96) * 4608)) + (rc_outer_outer * 288)) + (((((int)threadIdx.x) + 32) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+      kernel_shared[(((int)threadIdx.x) + 2968)] = kernel[(((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 2968) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((((int)threadIdx.x) + 88) % 96) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+      if (((int)threadIdx.x) &lt; 48) {
+        kernel_shared[(((int)threadIdx.x) + 3024)] = kernel[((((((((((int)blockIdx.x) / 7) * 147456) + (((((int)threadIdx.x) + 3024) / 96) * 4608)) + (rc_outer_outer * 288)) + ((((int)threadIdx.x) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 144)];
       }
+      __syncthreads();
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) * 3)] * kernel_shared[((((int)threadIdx.x) / 7) * 192)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((((int)threadIdx.x) % 7) * 3)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1536)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 7) * 3)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 96)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((((int)threadIdx.x) % 7) * 3)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1632)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 21)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 3)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 21)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1539)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 21)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 99)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 21)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1635)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 42)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 6)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 42)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1542)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 42)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 102)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 42)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1638)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 9)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1545)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 105)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1641)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1537)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 97)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1633)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 22)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 4)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 22)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1540)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 22)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 100)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 22)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1636)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 43)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 7)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 43)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1543)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 43)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 103)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 43)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1639)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 10)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1546)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 106)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1642)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 2)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1538)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 98)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1634)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 23)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 5)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 23)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1541)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 23)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 101)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 23)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1637)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 44)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 8)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 44)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1544)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 44)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 104)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 44)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1640)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 11)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1547)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 107)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1643)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 84)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 12)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 84)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1548)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 84)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 108)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 84)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1644)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 105)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 15)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 105)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1551)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 105)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 111)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 105)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1647)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 18)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1554)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 114)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1650)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 147)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 21)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 147)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1557)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 147)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 117)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 147)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1653)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 85)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 13)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 85)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1549)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 85)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 109)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 85)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1645)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 106)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 16)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 106)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1552)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 106)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 112)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 106)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1648)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 19)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1555)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 115)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1651)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 148)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 22)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 148)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1558)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 148)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 118)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 148)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1654)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 86)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 14)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 86)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1550)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 86)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 110)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 86)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1646)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 107)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 17)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 107)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1553)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 107)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 113)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 107)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1649)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 20)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1556)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 116)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1652)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 149)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 23)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 149)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1559)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 149)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 119)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 149)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1655)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 168)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 24)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 168)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1560)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 168)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 120)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 168)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1656)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 27)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1563)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 123)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1659)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 210)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 30)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 210)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1566)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 210)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 126)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 210)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1662)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 231)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 33)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 231)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1569)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 231)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 129)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 231)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1665)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 169)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 25)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 169)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1561)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 169)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 121)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 169)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1657)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 28)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1564)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 124)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1660)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 211)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 31)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 211)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1567)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 211)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 127)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 211)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1663)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 232)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 34)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 232)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1570)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 232)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 130)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 232)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1666)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 170)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 26)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 170)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1562)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 170)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 122)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 170)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1658)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 29)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1565)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 125)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1661)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 212)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 32)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 212)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1568)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 212)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 128)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 212)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1664)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 233)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 35)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 233)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1571)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 233)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 131)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 233)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1667)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 36)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1572)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 132)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1668)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 273)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 39)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 273)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1575)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 273)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 135)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 273)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1671)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 294)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 42)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 294)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1578)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 294)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 138)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 294)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1674)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 315)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 45)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 315)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1581)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 315)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 141)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 315)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1677)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 37)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1573)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 133)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1669)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 274)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 40)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 274)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1576)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 274)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 136)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 274)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1672)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 295)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 43)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 295)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1579)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 295)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 139)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 295)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1675)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 316)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 46)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 316)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1582)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 316)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 142)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 316)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1678)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 38)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1574)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 134)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1670)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 275)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 41)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 275)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1577)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 275)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 137)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 275)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1673)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 296)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 44)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 296)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1580)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 296)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 140)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 296)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1676)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 317)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 47)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 317)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1583)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 317)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 143)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 317)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1679)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 336)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 48)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 336)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1584)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 336)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 144)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 336)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1680)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 357)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 51)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 357)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1587)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 357)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 147)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 357)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1683)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 378)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 54)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 378)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1590)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 378)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 150)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 378)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1686)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 399)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 57)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 399)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1593)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 399)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 153)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 399)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1689)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 337)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 49)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 337)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1585)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 337)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 145)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 337)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1681)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 358)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 52)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 358)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1588)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 358)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 148)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 358)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1684)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 379)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 55)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 379)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1591)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 379)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 151)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 379)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1687)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 400)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 58)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 400)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1594)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 400)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 154)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 400)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1690)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 338)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 50)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 338)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1586)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 338)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 146)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 338)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1682)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 359)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 53)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 359)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1589)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 359)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 149)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 359)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1685)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 380)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 56)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 380)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1592)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 380)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 152)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 380)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1688)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 401)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 59)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 401)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1595)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 401)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 155)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 401)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1691)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 420)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 60)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 420)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1596)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 420)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 156)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 420)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1692)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 441)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 63)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 441)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1599)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 441)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 159)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 441)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1695)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 462)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 66)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 462)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1602)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 462)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 162)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 462)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1698)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 483)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 69)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 483)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1605)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 483)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 165)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 483)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1701)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 421)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 61)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 421)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1597)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 421)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 157)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 421)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1693)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 442)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 64)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 442)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1600)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 442)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 160)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 442)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1696)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 463)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 67)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 463)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1603)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 463)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 163)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 463)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1699)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 484)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 70)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 484)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1606)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 484)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 166)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 484)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1702)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 422)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 62)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 422)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1598)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 422)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 158)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 422)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1694)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 443)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 65)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 443)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1601)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 443)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 161)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 443)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1697)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 464)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 68)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 464)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1604)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 464)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 164)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 464)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1700)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 485)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 71)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 485)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1607)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 485)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 167)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 485)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1703)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 504)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 72)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 504)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1608)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 504)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 168)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 504)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1704)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 525)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 75)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 525)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1611)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 525)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 171)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 525)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1707)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 546)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 78)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 546)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1614)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 546)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 174)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 546)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1710)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 567)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 81)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 567)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1617)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 567)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 177)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 567)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1713)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 505)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 73)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 505)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1609)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 505)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 169)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 505)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1705)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 526)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 76)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 526)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1612)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 526)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 172)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 526)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1708)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 547)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 79)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 547)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1615)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 547)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 175)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 547)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1711)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 568)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 82)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 568)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1618)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 568)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 178)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 568)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1714)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 506)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 74)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 506)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1610)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 506)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 170)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 506)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1706)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 527)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 77)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 527)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1613)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 527)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 173)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 527)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1709)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 548)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 80)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 548)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1616)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 548)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 176)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 548)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1712)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 569)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 83)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 569)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1619)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 569)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 179)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 569)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1715)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 588)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 84)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 588)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1620)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 588)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 180)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 588)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1716)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 609)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 87)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 609)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1623)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 609)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 183)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 609)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1719)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 630)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 90)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 630)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1626)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 630)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 186)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 630)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1722)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 651)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 93)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 651)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1629)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 651)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 189)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 651)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1725)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 589)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 85)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 589)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1621)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 589)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 181)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 589)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1717)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 610)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 88)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 610)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1624)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 610)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 184)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 610)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1720)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 631)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 91)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 631)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1627)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 631)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 187)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 631)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1723)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 652)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 94)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 652)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1630)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 652)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 190)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 652)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1726)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 590)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 86)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 590)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1622)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 590)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 182)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 590)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1718)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 611)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 89)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 611)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1625)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 611)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 185)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 611)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1721)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 632)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 92)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 632)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1628)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 632)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 188)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 632)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1724)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 653)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 95)]));
+      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 653)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1631)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 653)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 191)]));
+      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 3) + 653)] * kernel_shared[(((((int)threadIdx.x) / 7) * 192) + 1727)]));
     }
   }
-  compute[((((int)blockIdx.x) * 784) + ((int)threadIdx.x))] = max((conv2d_nchw[0] + bias[((((int)blockIdx.x) * 16) + (((int)threadIdx.x) / 49))]), 0.000000e+00f);
-  compute[(((((int)blockIdx.x) * 784) + ((int)threadIdx.x)) + 196)] = max((conv2d_nchw[1] + bias[(((((int)blockIdx.x) * 16) + (((int)threadIdx.x) / 49)) + 4)]), 0.000000e+00f);
-  compute[(((((int)blockIdx.x) * 784) + ((int)threadIdx.x)) + 392)] = max((conv2d_nchw[2] + bias[(((((int)blockIdx.x) * 16) + (((int)threadIdx.x) / 49)) + 8)]), 0.000000e+00f);
-  compute[(((((int)blockIdx.x) * 784) + ((int)threadIdx.x)) + 588)] = max((conv2d_nchw[3] + bias[(((((int)blockIdx.x) * 16) + (((int)threadIdx.x) / 49)) + 12)]), 0.000000e+00f);
+  for (int i1_inner = 0; i1_inner &lt; 2; ++i1_inner) {
+    compute[((((((((int)blockIdx.x) / 7) * 1568) + ((((int)threadIdx.x) / 7) * 98)) + (i1_inner * 49)) + ((((int)threadIdx.x) % 7) * 7)) + (((int)blockIdx.x) % 7))] = max((conv2d_nchw[i1_inner] + bias[((((((int)blockIdx.x) / 7) * 32) + ((((int)threadIdx.x) / 7) * 2)) + i1_inner)]), 0.000000e+00f);
+    compute[(((((((((int)blockIdx.x) / 7) * 1568) + ((((int)threadIdx.x) / 7) * 98)) + (i1_inner * 49)) + ((((int)threadIdx.x) % 7) * 7)) + (((int)blockIdx.x) % 7)) + 784)] = max((conv2d_nchw[(i1_inner + 2)] + bias[(((((((int)blockIdx.x) / 7) * 32) + ((((int)threadIdx.x) / 7) * 2)) + i1_inner) + 16)]), 0.000000e+00f);
+  }
 }
 </pre></div>
 </div>
@@ -810,7 +1695,7 @@ In the example below we resume the status and do more 5 trials.</p>
 Get devices for measurement successfully!
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 6 minutes  10.669 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 6 minutes  18.523 seconds)</p>
 <div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-tune-with-autoscheduler-tune-conv2d-layer-cuda-py">
 <div class="sphx-glr-download sphx-glr-download-python docutils container">
 <p><a class="reference download internal" download="" href="../../_downloads/e3e540f3b477c0c52d8eb73e674e8ffd/tune_conv2d_layer_cuda.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">tune_conv2d_layer_cuda.py</span></code></a></p>
diff --git a/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html b/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
index 8c7aed0471..fad7fda335 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
@@ -921,7 +921,7 @@ so we can read the log file and load the best schedules.</p>
 Evaluate inference time cost...
 Execution time summary:
  mean (ms)   median (ms)    max (ms)     min (ms)     std (ms)
-   8.1053       8.1077       8.1103       8.0979       0.0053
+   8.1199       8.1181       8.1249       8.1165       0.0036
 </pre></div>
 </div>
 </div>
@@ -943,7 +943,7 @@ to learn how to use the RPC Tracker and RPC Server.
 To use the RPC Tracker in auto-scheduler, replace the runner in <code class="code docutils literal notranslate"><span class="pre">TuningOptions</span></code>
 with <a class="reference internal" href="../../reference/api/python/auto_scheduler.html#tvm.auto_scheduler.RPCRunner" title="tvm.auto_scheduler.RPCRunner"><code class="xref any py py-class docutils literal notranslate"><span class="pre">auto_scheduler.RPCRunner</span></code></a>.</p></li>
 </ol>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  12.321 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  13.936 seconds)</p>
 <div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-tune-with-autoscheduler-tune-network-cuda-py">
 <div class="sphx-glr-download sphx-glr-download-python docutils container">
 <p><a class="reference download internal" download="" href="../../_downloads/eafe360d52540634c9eea0fa89e804bd/tune_network_cuda.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">tune_network_cuda.py</span></code></a></p>
diff --git a/docs/how_to/tune_with_autoscheduler/tune_network_x86.html b/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
index 5b3d61be6c..d25486892d 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
@@ -940,7 +940,7 @@ so we can read the log file and load the best schedules.</p>
 Evaluate inference time cost...
 Execution time summary:
  mean (ms)   median (ms)    max (ms)     min (ms)     std (ms)
-  758.2928     758.2217     758.4373     758.2193      0.1022
+  763.8272     765.0162     766.1968     760.2686      2.5621
 </pre></div>
 </div>
 </div>
@@ -962,7 +962,7 @@ to learn how to use the RPC Tracker and RPC Server.
 To use the RPC Tracker in auto-scheduler, replace the runner in <code class="code docutils literal notranslate"><span class="pre">TuningOptions</span></code>
 with <a class="reference internal" href="../../reference/api/python/auto_scheduler.html#tvm.auto_scheduler.RPCRunner" title="tvm.auto_scheduler.RPCRunner"><code class="xref any py py-class docutils literal notranslate"><span class="pre">auto_scheduler.RPCRunner</span></code></a>.</p></li>
 </ol>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  43.939 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  45.674 seconds)</p>
 <div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-tune-with-autoscheduler-tune-network-x86-py">
 <div class="sphx-glr-download sphx-glr-download-python docutils container">
 <p><a class="reference download internal" download="" href="../../_downloads/e416b94ca1090b0897c0f6e0df95b911/tune_network_x86.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">tune_network_x86.py</span></code></a></p>
diff --git a/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html b/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
index 0f05202234..169769ba55 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
@@ -637,26 +637,117 @@ class Module:
     @T.prim_func
     def main(placeholder: T.Buffer((128, 256), &quot;float32&quot;), placeholder_1: T.Buffer((4916, 16, 1), &quot;float32&quot;), placeholder_2: T.Buffer((4916,), &quot;int32&quot;), placeholder_3: T.Buffer((33,), &quot;int32&quot;), placeholder_4: T.Buffer((128, 512), &quot;float32&quot;), compute: T.Buffer((128, 512), &quot;float32&quot;)):
         T.func_attr({&quot;from_legacy_te_schedule&quot;: T.bool(True), &quot;global_symbol&quot;: &quot;main&quot;, &quot;tir.noalias&quot;: T.bool(True)})
-        for i0_outer_i1_outer_fused in T.parallel(32):
-            compute_1 = T.allocate([2048], &quot;float32&quot;, &quot;global&quot;)
-            compute_2 = T.Buffer((2048,), data=compute_1)
-            for i_outer_inner, nb_j_inner in T.grid(4, 2):
-                for i_inner_init, j_init in T.grid(16, 16):
-                    compute_2[i_outer_inner * 512 + i_inner_init * 32 + nb_j_inner * 16 + j_init] = T.float32(0)
-                for elem_idx, i_inner, j in T.grid(T.Let(placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1], where={cse_var_1: i0_outer_i1_outer_fused % 16 * 2 + nb_j_inner}), 16, 16):
-                    cse_var_1 = T.int32()
+        for i0_outer in T.parallel(32):
+            compute_1 = T.allocate([32], &quot;float32&quot;, &quot;global&quot;)
+            for i1_outer in range(64):
+                cse_var_1: T.int32 = i1_outer // 2
+                compute_2 = T.Buffer((32,), data=compute_1)
+                compute_2[0] = T.float32(0)
+                compute_2[1] = T.float32(0)
+                compute_2[2] = T.float32(0)
+                compute_2[3] = T.float32(0)
+                compute_2[4] = T.float32(0)
+                compute_2[5] = T.float32(0)
+                compute_2[6] = T.float32(0)
+                compute_2[7] = T.float32(0)
+                compute_2[8] = T.float32(0)
+                compute_2[9] = T.float32(0)
+                compute_2[10] = T.float32(0)
+                compute_2[11] = T.float32(0)
+                compute_2[12] = T.float32(0)
+                compute_2[13] = T.float32(0)
+                compute_2[14] = T.float32(0)
+                compute_2[15] = T.float32(0)
+                compute_2[16] = T.float32(0)
+                compute_2[17] = T.float32(0)
+                compute_2[18] = T.float32(0)
+                compute_2[19] = T.float32(0)
+                compute_2[20] = T.float32(0)
+                compute_2[21] = T.float32(0)
+                compute_2[22] = T.float32(0)
+                compute_2[23] = T.float32(0)
+                compute_2[24] = T.float32(0)
+                compute_2[25] = T.float32(0)
+                compute_2[26] = T.float32(0)
+                compute_2[27] = T.float32(0)
+                compute_2[28] = T.float32(0)
+                compute_2[29] = T.float32(0)
+                compute_2[30] = T.float32(0)
+                compute_2[31] = T.float32(0)
+                for elem_idx in range(placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
                     placeholder_5 = T.Buffer((33,), &quot;int32&quot;, data=placeholder_3.data)
-                    cse_var_3: T.int32 = i0_outer_i1_outer_fused % 16 * 2 + nb_j_inner
-                    cse_var_2: T.int32 = i_outer_inner * 512 + i_inner * 32 + nb_j_inner * 16 + j
                     placeholder_6 = T.Buffer((78656,), data=placeholder_1.data)
                     placeholder_7 = T.Buffer((32768,), data=placeholder.data)
                     placeholder_8 = T.Buffer((4916,), &quot;int32&quot;, data=placeholder_2.data)
-                    compute_2[cse_var_2] = compute_2[cse_var_2] + placeholder_6[placeholder_5[cse_var_3] * 16 + elem_idx * 16 + j] * T.max(placeholder_7[i0_outer_i1_outer_fused // 16 * 16384 + i_outer_inner * 4096 + i_inner * 256 + placeholder_8[placeholder_5[cse_var_3] + elem_idx]], T.float32(0))
-            for i0_inner, i1_inner in T.grid(64, 32):
-                cse_var_4: T.int32 = i0_outer_i1_outer_fused // 16 * 32768 + i0_inner * 512 + i0_outer_i1_outer_fused % 16 * 32 + i1_inner
-                compute_3 = T.Buffer((65536,), data=compute.data)
-                placeholder_5 = T.Buffer((65536,), data=placeholder_4.data)
-                compute_3[cse_var_4] = T.max(compute_2[i0_inner * 32 + i1_inner] + placeholder_5[cse_var_4], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[0] = compute_2[0] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[1] = compute_2[1] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 1] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[2] = compute_2[2] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 2] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[3] = compute_2[3] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 3] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[4] = compute_2[4] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 4] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[5] = compute_2[5] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 5] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[6] = compute_2[6] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 6] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[7] = compute_2[7] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 7] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx]], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[8] = compute_2[8] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[9] = compute_2[9] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 1] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[10] = compute_2[10] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 2] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[11] = compute_2[11] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 3] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[12] = compute_2[12] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 4] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[13] = compute_2[13] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 5] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[14] = compute_2[14] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 6] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[15] = compute_2[15] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 7] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 256], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[16] = compute_2[16] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[17] = compute_2[17] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 1] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[18] = compute_2[18] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 2] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[19] = compute_2[19] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 3] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[20] = compute_2[20] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 4] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[21] = compute_2[21] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 5] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[22] = compute_2[22] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 6] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[23] = compute_2[23] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 7] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 512], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[24] = compute_2[24] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[25] = compute_2[25] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 1] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[26] = compute_2[26] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 2] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[27] = compute_2[27] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 3] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[28] = compute_2[28] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 4] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[29] = compute_2[29] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 5] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[30] = compute_2[30] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 6] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+                    if T.likely(elem_idx &lt; placeholder_5[cse_var_1 + 1] - placeholder_5[cse_var_1]):
+                        compute_2[31] = compute_2[31] + placeholder_6[placeholder_5[cse_var_1] * 16 + elem_idx * 16 + i1_outer % 2 * 8 + 7] * T.max(placeholder_7[i0_outer * 1024 + placeholder_8[placeholder_5[cse_var_1] + elem_idx] + 768], T.float32(0))
+                for i0_inner in range(4):
+                    cse_var_2: T.int32 = i0_outer * 2048 + i0_inner * 512 + i1_outer * 8
+                    compute_3 = T.Buffer((65536,), data=compute.data)
+                    placeholder_5 = T.Buffer((65536,), data=placeholder_4.data)
+                    compute_3[cse_var_2:cse_var_2 + 8] = T.max(compute_2[i0_inner * 8:i0_inner * 8 + 8] + placeholder_5[cse_var_2:cse_var_2 + 8], T.Broadcast(T.float32(0), 8))
 </pre></div>
 </div>
 </div>
@@ -690,7 +781,7 @@ class Module:
 <span class="p">)</span>
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 1.519 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 1.841 ms
 </pre></div>
 </div>
 <div class="admonition note">
diff --git a/docs/how_to/tune_with_autotvm/sg_execution_times.html b/docs/how_to/tune_with_autotvm/sg_execution_times.html
index 7e832575ed..4e4b47c160 100644
--- a/docs/how_to/tune_with_autotvm/sg_execution_times.html
+++ b/docs/how_to/tune_with_autotvm/sg_execution_times.html
@@ -345,7 +345,7 @@
             
   <div class="section" id="computation-times">
 <span id="sphx-glr-how-to-tune-with-autotvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:49.351</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
+<p><strong>00:38.513</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 84%" />
@@ -354,7 +354,7 @@
 </colgroup>
 <tbody>
 <tr class="row-odd"><td><p><a class="reference internal" href="tune_conv2d_cuda.html#sphx-glr-how-to-tune-with-autotvm-tune-conv2d-cuda-py"><span class="std std-ref">Tuning High Performance Convolution on NVIDIA GPUs</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_conv2d_cuda.py</span></code>)</p></td>
-<td><p>00:49.318</p></td>
+<td><p>00:38.479</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="tune_relay_x86.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-x86-py"><span class="std std-ref">Auto-tuning a Convolutional Network for x86 CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_x86.py</span></code>)</p></td>
@@ -366,11 +366,11 @@
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="tune_relay_arm.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-arm-py"><span class="std std-ref">Auto-tuning a Convolutional Network for ARM CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_arm.py</span></code>)</p></td>
-<td><p>00:00.004</p></td>
+<td><p>00:00.005</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="tune_relay_mobile_gpu.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-mobile-gpu-py"><span class="std std-ref">Auto-tuning a Convolutional Network for Mobile GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_mobile_gpu.py</span></code>)</p></td>
-<td><p>00:00.004</p></td>
+<td><p>00:00.005</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 </tbody>
diff --git a/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html b/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
index 6fb78ceee2..e02b82c629 100644
--- a/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
+++ b/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
@@ -573,26 +573,131 @@ for this template</p>
 waiting for device...
 device available
 Get devices for measurement successfully!
-No: 1   GFLOPS: 7.51/7.51       result: MeasureResult(costs=(0.0308341445,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.836233854293823, timestamp=1680177386.5471036)        [(&#39;tile_f&#39;, [-1, 32, 1, 8]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 1, 4]), (&#39;tile_ry&#39;, [-1, 1, 1]), (&#39;tile_rx&#39;, [-1, 1, 1]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 0)],None,3552261
-No: 2   GFLOPS: 0.00/7.51       result: Traceback (most recent call last):
-  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 142, in build
-    res = future.result()
-  File &quot;/usr/lib/python3.7/concurrent/futures/_base.py&quot;, line 435, in result
-    return self.__get_result()
-  File &quot;/usr/lib/python3.7/concurrent/futures/_base.py&quot;, line 384, in __get_result
-    raise self._exception
-  File &quot;/usr/lib/python3.7/concurrent/futures/thread.py&quot;, line 57, in run
-    result = self.fn(*self.args, **self.kwargs)
-  File &quot;/workspace/python/tvm/contrib/popen_pool.py&quot;, line 432, in &lt;lambda&gt;
-    worker = lambda *args: self._worker_run(*args)
-  File &quot;/workspace/python/tvm/contrib/popen_pool.py&quot;, line 401, in _worker_run
-    return proc.recv()
-  File &quot;/workspace/python/tvm/contrib/popen_pool.py&quot;, line 309, in recv
-    raise TimeoutError()
-TimeoutError
+No: 1   GFLOPS: 345.53/345.53   result: MeasureResult(costs=(0.00066998061875,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.27738881111145, timestamp=1680189104.0109434)     [(&#39;tile_f&#39;, [-1, 1, 16, 4]), (&#39;tile_y&#39;, [-1, 7, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 1, 8]), (&#39;tile_ry&#39;, [-1, 3, 1]), (&#39;tile_rx&#39;, [-1, 1, 1]), (&#39;auto_unroll_max_step&#39;, 512), (&#39;unroll_explicit&#39;, 0)],None,2033146
+No: 2   GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
+  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 592, in __call__
+    func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
+  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 544, in _build_func_common
+    func = build(s, args, target=target, runtime=runtime)
+  File &quot;/workspace/python/tvm/driver/build_module.py&quot;, line 227, in build
+    input_mod = lower(inputs, args, name=name, binds=binds)
+  File &quot;/workspace/python/tvm/driver/build_module.py&quot;, line 134, in lower
+    return ffi.lower_schedule(inp, args, name, binds, simple_mode)
+  File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 331, in tvm._ffi._cy3.core.PackedFuncBase.__call__
+  File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 276, in tvm._ffi._cy3.core.FuncCall
+  File &quot;tvm/_ffi/_cython/./base.pxi&quot;, line 181, in tvm._ffi._cy3.core.CHECK_CALL
+tvm._ffi.base.TVMError: Traceback (most recent call last):
+  24: TVMFuncCall
+        at ../src/runtime/c_runtime_api.cc:477
+  23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+        at ../include/tvm/runtime/packed_func.h:1217
+  22: Call
+        at ../include/tvm/runtime/packed_func.h:1213
+  21: operator()
+        at ../include/tvm/runtime/packed_func.h:1734
+  20: unpack_call&lt;tvm::IRModule, 5, tvm::&lt;lambda(tvm::te::Schedule, const tvm::runtime::Array&lt;tvm::runtime::ObjectRef&gt;&amp;, const tvm::runtime::String&amp;, const tvm::runtime::Map&lt;tvm::te::Tensor, tvm::tir::Buffer&gt;&amp;, bool)&gt; &gt;
+        at ../include/tvm/runtime/packed_func.h:1674
+  19: run&lt;&gt;
+        at ../include/tvm/runtime/packed_func.h:1634
+  18: run&lt;tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1634
+  17: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1634
+  16: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1634
+  15: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1634
+  14: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1649
+  13: operator()
+        at ../src/driver/driver_api.cc:402
+  12: tvm::LowerSchedule(tvm::te::Schedule, tvm::runtime::Array&lt;tvm::runtime::ObjectRef, void&gt; const&amp;, std::__cxx11::basic_string&lt;char, std::char_traits&lt;char&gt;, std::allocator&lt;char&gt; &gt; const&amp;, std::unordered_map&lt;tvm::te::Tensor, tvm::tir::Buffer, std::hash&lt;tvm::te::Tensor&gt;, std::equal_to&lt;tvm::te::Tensor&gt;, std::allocator&lt;std::pair&lt;tvm::te::Tensor const, tvm::tir::Buffer&gt; &gt; &gt; const&amp;, tvm::GlobalVarSupply, bool)
+        at ../src/driver/driver_api.cc:388
+  11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array&lt;tvm::transform::Pass, void&gt;)
+        at ../src/driver/driver_api.cc:283
+  10: tvm::transform::Pass::operator()(tvm::IRModule) const
+        at ../src/ir/transform.cc:258
+  9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&amp;) const
+        at ../src/ir/transform.cc:274
+  8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&amp;) const
+        at ../src/ir/transform.cc:451
+  7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&amp;) const
+        at ../src/ir/transform.cc:274
+  6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&amp;) const
+        at ../src/tir/ir/transform.cc:100
+  5: tvm::runtime::TypedPackedFunc&lt;tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)&gt;::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
+        at ../include/tvm/runtime/packed_func.h:1753
+  4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher&lt;tvm::tir::PrimFunc&gt;::run&lt;tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext&gt;(tvm::runtime::PackedFunc const&amp;, tvm::tir::PrimFunc&amp;&amp;, tvm::IRModule&amp;&amp;, tvm::transform::PassContext&amp;&amp;)
+        at ../include/tvm/runtime/packed_func.h:1697
+  3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()&lt;tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext&gt;(tvm::tir::PrimFunc&amp;&amp;, tvm::IRModule&amp;&amp;, tvm::transform::PassContext&amp;&amp;) const
+        at ../include/tvm/runtime/packed_func.h:1621
+  2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+        at ../include/tvm/runtime/packed_func.h:1217
+  1: Call
+        at ../include/tvm/runtime/packed_func.h:1213
+  0: operator()
+        at ../src/runtime/c_runtime_api.cc:534
+  File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 56, in tvm._ffi._cy3.core.tvm_callback
+  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 875, in verify_pass
+    raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel
 
-        [(&#39;tile_f&#39;, [-1, 16, 2, 4]), (&#39;tile_y&#39;, [-1, 7, 1, 1]), (&#39;tile_x&#39;, [-1, 7, 1, 1]), (&#39;tile_rc&#39;, [-1, 1, 64]), (&#39;tile_ry&#39;, [-1, 1, 1]), (&#39;tile_rx&#39;, [-1, 1, 1]), (&#39;auto_unroll_max_step&#39;, 512), (&#39;unroll_explicit&#39;, 0)],None,1902012
-No: 3   GFLOPS: 0.00/7.51       result: Traceback (most recent call last):
+Traceback (most recent call last):
+  24: TVMFuncCall
+        at ../src/runtime/c_runtime_api.cc:477
+  23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+        at ../include/tvm/runtime/packed_func.h:1217
+  22: Call
+        at ../include/tvm/runtime/packed_func.h:1213
+  21: operator()
+        at ../include/tvm/runtime/packed_func.h:1734
+  20: unpack_call&lt;tvm::IRModule, 5, tvm::&lt;lambda(tvm::te::Schedule, const tvm::runtime::Array&lt;tvm::runtime::ObjectRef&gt;&amp;, const tvm::runtime::String&amp;, const tvm::runtime::Map&lt;tvm::te::Tensor, tvm::tir::Buffer&gt;&amp;, bool)&gt; &gt;
+        at ../include/tvm/runtime/packed_func.h:1674
+  19: run&lt;&gt;
+        at ../include/tvm/runtime/packed_func.h:1634
+  18: run&lt;tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1634
+  17: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1634
+  16: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1634
+  15: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1634
+  14: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1649
+  13: operator()
+        at ../src/driver/driver_api.cc:402
+  12: tvm::LowerSchedule(tvm::te::Schedule, tvm::runtime::Array&lt;tvm::runtime::ObjectRef, void&gt; const&amp;, std::__cxx11::basic_string&lt;char, std::char_traits&lt;char&gt;, std::allocator&lt;char&gt; &gt; const&amp;, std::unordered_map&lt;tvm::te::Tensor, tvm::tir::Buffer, std::hash&lt;tvm::te::Tensor&gt;, std::equal_to&lt;tvm::te::Tensor&gt;, std::allocator&lt;std::pair&lt;tvm::te::Tensor const, tvm::tir::Buffer&gt; &gt; &gt; const&amp;, tvm::GlobalVarSupply, bool)
+        at ../src/driver/driver_api.cc:388
+  11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array&lt;tvm::transform::Pass, void&gt;)
+        at ../src/driver/driver_api.cc:283
+  10: tvm::transform::Pass::operator()(tvm::IRModule) const
+        at ../src/ir/transform.cc:258
+  9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&amp;) const
+        at ../src/ir/transform.cc:274
+  8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&amp;) const
+        at ../src/ir/transform.cc:451
+  7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&amp;) const
+        at ../src/ir/transform.cc:274
+  6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&amp;) const
+        at ../src/tir/ir/transform.cc:100
+  5: tvm::runtime::TypedPackedFunc&lt;tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)&gt;::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
+        at ../include/tvm/runtime/packed_func.h:1753
+  4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher&lt;tvm::tir::PrimFunc&gt;::run&lt;tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext&gt;(tvm::runtime::PackedFunc const&amp;, tvm::tir::PrimFunc&amp;&amp;, tvm::IRModule&amp;&amp;, tvm::transform::PassContext&amp;&amp;)
+        at ../include/tvm/runtime/packed_func.h:1697
+  3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()&lt;tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext&gt;(tvm::tir::PrimFunc&amp;&amp;, tvm::IRModule&amp;&amp;, tvm::transform::PassContext&amp;&amp;) const
+        at ../include/tvm/runtime/packed_func.h:1621
+  2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+        at ../include/tvm/runtime/packed_func.h:1217
+  1: Call
+        at ../include/tvm/runtime/packed_func.h:1213
+  0: operator()
+        at ../src/runtime/c_runtime_api.cc:534
+  File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 56, in tvm._ffi._cy3.core.tvm_callback
+  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 875, in verify_pass
+    raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 64, 4, 2]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 8, 8]), (&#39;tile_ry&#39;, [-1, 1, 3]), (&#39;tile_rx&#39;, [-1, 3, 1]), (&#39;auto_unroll_max_step&#39;, 512), (&#39;unroll_explicit&#39;, 0)],None,2816738
+No: 3   GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 592, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 544, in _build_func_common
@@ -714,26 +819,131 @@ Traceback (most recent call last):
   File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 56, in tvm._ffi._cy3.core.tvm_callback
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 875, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 4, 2, 4]), (&#39;tile_y&#39;, [-1, 7, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 16, 32]), (&#39;tile_ry&#39;, [-1, 1, 1]), (&#39;tile_rx&#39;, [-1, 3, 1]), (&#39;auto_unroll_max_step&#39;, 0), (&#39;unroll_explicit&#39;, 0)],None,736010
-No: 4   GFLOPS: 0.00/7.51       result: Traceback (most recent call last):
-  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 142, in build
-    res = future.result()
-  File &quot;/usr/lib/python3.7/concurrent/futures/_base.py&quot;, line 435, in result
-    return self.__get_result()
-  File &quot;/usr/lib/python3.7/concurrent/futures/_base.py&quot;, line 384, in __get_result
-    raise self._exception
-  File &quot;/usr/lib/python3.7/concurrent/futures/thread.py&quot;, line 57, in run
-    result = self.fn(*self.args, **self.kwargs)
-  File &quot;/workspace/python/tvm/contrib/popen_pool.py&quot;, line 432, in &lt;lambda&gt;
-    worker = lambda *args: self._worker_run(*args)
-  File &quot;/workspace/python/tvm/contrib/popen_pool.py&quot;, line 401, in _worker_run
-    return proc.recv()
-  File &quot;/workspace/python/tvm/contrib/popen_pool.py&quot;, line 309, in recv
-    raise TimeoutError()
-TimeoutError
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 8, 8, 1]), (&#39;tile_y&#39;, [-1, 7, 1, 1]), (&#39;tile_x&#39;, [-1, 7, 1, 1]), (&#39;tile_rc&#39;, [-1, 128, 1]), (&#39;tile_ry&#39;, [-1, 1, 3]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 0), (&#39;unroll_explicit&#39;, 0)],None,1574570
+No: 4   GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
+  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 592, in __call__
+    func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
+  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 544, in _build_func_common
+    func = build(s, args, target=target, runtime=runtime)
+  File &quot;/workspace/python/tvm/driver/build_module.py&quot;, line 227, in build
+    input_mod = lower(inputs, args, name=name, binds=binds)
+  File &quot;/workspace/python/tvm/driver/build_module.py&quot;, line 134, in lower
+    return ffi.lower_schedule(inp, args, name, binds, simple_mode)
+  File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 331, in tvm._ffi._cy3.core.PackedFuncBase.__call__
+  File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 276, in tvm._ffi._cy3.core.FuncCall
+  File &quot;tvm/_ffi/_cython/./base.pxi&quot;, line 181, in tvm._ffi._cy3.core.CHECK_CALL
+tvm._ffi.base.TVMError: Traceback (most recent call last):
+  24: TVMFuncCall
+        at ../src/runtime/c_runtime_api.cc:477
+  23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+        at ../include/tvm/runtime/packed_func.h:1217
+  22: Call
+        at ../include/tvm/runtime/packed_func.h:1213
+  21: operator()
+        at ../include/tvm/runtime/packed_func.h:1734
+  20: unpack_call&lt;tvm::IRModule, 5, tvm::&lt;lambda(tvm::te::Schedule, const tvm::runtime::Array&lt;tvm::runtime::ObjectRef&gt;&amp;, const tvm::runtime::String&amp;, const tvm::runtime::Map&lt;tvm::te::Tensor, tvm::tir::Buffer&gt;&amp;, bool)&gt; &gt;
+        at ../include/tvm/runtime/packed_func.h:1674
+  19: run&lt;&gt;
+        at ../include/tvm/runtime/packed_func.h:1634
+  18: run&lt;tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1634
+  17: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1634
+  16: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1634
+  15: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1634
+  14: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1649
+  13: operator()
+        at ../src/driver/driver_api.cc:402
+  12: tvm::LowerSchedule(tvm::te::Schedule, tvm::runtime::Array&lt;tvm::runtime::ObjectRef, void&gt; const&amp;, std::__cxx11::basic_string&lt;char, std::char_traits&lt;char&gt;, std::allocator&lt;char&gt; &gt; const&amp;, std::unordered_map&lt;tvm::te::Tensor, tvm::tir::Buffer, std::hash&lt;tvm::te::Tensor&gt;, std::equal_to&lt;tvm::te::Tensor&gt;, std::allocator&lt;std::pair&lt;tvm::te::Tensor const, tvm::tir::Buffer&gt; &gt; &gt; const&amp;, tvm::GlobalVarSupply, bool)
+        at ../src/driver/driver_api.cc:388
+  11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array&lt;tvm::transform::Pass, void&gt;)
+        at ../src/driver/driver_api.cc:283
+  10: tvm::transform::Pass::operator()(tvm::IRModule) const
+        at ../src/ir/transform.cc:258
+  9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&amp;) const
+        at ../src/ir/transform.cc:274
+  8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&amp;) const
+        at ../src/ir/transform.cc:451
+  7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&amp;) const
+        at ../src/ir/transform.cc:274
+  6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&amp;) const
+        at ../src/tir/ir/transform.cc:100
+  5: tvm::runtime::TypedPackedFunc&lt;tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)&gt;::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
+        at ../include/tvm/runtime/packed_func.h:1753
+  4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher&lt;tvm::tir::PrimFunc&gt;::run&lt;tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext&gt;(tvm::runtime::PackedFunc const&amp;, tvm::tir::PrimFunc&amp;&amp;, tvm::IRModule&amp;&amp;, tvm::transform::PassContext&amp;&amp;)
+        at ../include/tvm/runtime/packed_func.h:1697
+  3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()&lt;tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext&gt;(tvm::tir::PrimFunc&amp;&amp;, tvm::IRModule&amp;&amp;, tvm::transform::PassContext&amp;&amp;) const
+        at ../include/tvm/runtime/packed_func.h:1621
+  2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+        at ../include/tvm/runtime/packed_func.h:1217
+  1: Call
+        at ../include/tvm/runtime/packed_func.h:1213
+  0: operator()
+        at ../src/runtime/c_runtime_api.cc:534
+  File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 56, in tvm._ffi._cy3.core.tvm_callback
+  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 875, in verify_pass
+    raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel
 
-        [(&#39;tile_f&#39;, [-1, 4, 1, 16]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 2, 4]), (&#39;tile_ry&#39;, [-1, 3, 1]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 0)],None,4912986
-No: 5   GFLOPS: 0.00/7.51       result: Traceback (most recent call last):
+Traceback (most recent call last):
+  24: TVMFuncCall
+        at ../src/runtime/c_runtime_api.cc:477
+  23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+        at ../include/tvm/runtime/packed_func.h:1217
+  22: Call
+        at ../include/tvm/runtime/packed_func.h:1213
+  21: operator()
+        at ../include/tvm/runtime/packed_func.h:1734
+  20: unpack_call&lt;tvm::IRModule, 5, tvm::&lt;lambda(tvm::te::Schedule, const tvm::runtime::Array&lt;tvm::runtime::ObjectRef&gt;&amp;, const tvm::runtime::String&amp;, const tvm::runtime::Map&lt;tvm::te::Tensor, tvm::tir::Buffer&gt;&amp;, bool)&gt; &gt;
+        at ../include/tvm/runtime/packed_func.h:1674
+  19: run&lt;&gt;
+        at ../include/tvm/runtime/packed_func.h:1634
+  18: run&lt;tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1634
+  17: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1634
+  16: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1634
+  15: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1634
+  14: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1649
+  13: operator()
+        at ../src/driver/driver_api.cc:402
+  12: tvm::LowerSchedule(tvm::te::Schedule, tvm::runtime::Array&lt;tvm::runtime::ObjectRef, void&gt; const&amp;, std::__cxx11::basic_string&lt;char, std::char_traits&lt;char&gt;, std::allocator&lt;char&gt; &gt; const&amp;, std::unordered_map&lt;tvm::te::Tensor, tvm::tir::Buffer, std::hash&lt;tvm::te::Tensor&gt;, std::equal_to&lt;tvm::te::Tensor&gt;, std::allocator&lt;std::pair&lt;tvm::te::Tensor const, tvm::tir::Buffer&gt; &gt; &gt; const&amp;, tvm::GlobalVarSupply, bool)
+        at ../src/driver/driver_api.cc:388
+  11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array&lt;tvm::transform::Pass, void&gt;)
+        at ../src/driver/driver_api.cc:283
+  10: tvm::transform::Pass::operator()(tvm::IRModule) const
+        at ../src/ir/transform.cc:258
+  9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&amp;) const
+        at ../src/ir/transform.cc:274
+  8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&amp;) const
+        at ../src/ir/transform.cc:451
+  7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&amp;) const
+        at ../src/ir/transform.cc:274
+  6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&amp;) const
+        at ../src/tir/ir/transform.cc:100
+  5: tvm::runtime::TypedPackedFunc&lt;tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)&gt;::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
+        at ../include/tvm/runtime/packed_func.h:1753
+  4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher&lt;tvm::tir::PrimFunc&gt;::run&lt;tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext&gt;(tvm::runtime::PackedFunc const&amp;, tvm::tir::PrimFunc&amp;&amp;, tvm::IRModule&amp;&amp;, tvm::transform::PassContext&amp;&amp;)
+        at ../include/tvm/runtime/packed_func.h:1697
+  3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()&lt;tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext&gt;(tvm::tir::PrimFunc&amp;&amp;, tvm::IRModule&amp;&amp;, tvm::transform::PassContext&amp;&amp;) const
+        at ../include/tvm/runtime/packed_func.h:1621
+  2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+        at ../include/tvm/runtime/packed_func.h:1217
+  1: Call
+        at ../include/tvm/runtime/packed_func.h:1213
+  0: operator()
+        at ../src/runtime/c_runtime_api.cc:534
+  File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 56, in tvm._ffi._cy3.core.tvm_callback
+  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 875, in verify_pass
+    raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 4, 32, 2]), (&#39;tile_y&#39;, [-1, 7, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 8, 2]), (&#39;tile_ry&#39;, [-1, 3, 1]), (&#39;tile_rx&#39;, [-1, 3, 1]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 0)],None,4305272
+No: 5   GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 592, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 544, in _build_func_common
@@ -855,8 +1065,8 @@ Traceback (most recent call last):
   File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 56, in tvm._ffi._cy3.core.tvm_callback
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 875, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 2, 32, 8]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 7, 1, 1]), (&#39;tile_rc&#39;, [-1, 1, 32]), (&#39;tile_ry&#39;, [-1, 3, 1]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 0), (&#39;unroll_explicit&#39;, 0)],None,1497482
-No: 6   GFLOPS: 0.00/7.51       result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 8, 8, 4]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 1, 128]), (&#39;tile_ry&#39;, [-1, 1, 1]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 512), (&#39;unroll_explicit&#39;, 1)],None,8304464
+No: 6   GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 592, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 544, in _build_func_common
@@ -978,8 +1188,9 @@ Traceback (most recent call last):
   File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 56, in tvm._ffi._cy3.core.tvm_callback
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 875, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 16, 32, 1]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 8, 64]), (&#39;tile_ry&#39;, [-1, 1, 1]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 0)],None,4818704
-No: 7   GFLOPS: 0.00/7.51       result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 16, 8, 2]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 128, 1]), (&#39;tile_ry&#39;, [-1, 3, 1]), (&#39;tile_rx&#39;, [-1, 1, 1]), (&#39;auto_unroll_max_step&#39;, 512), (&#39;unroll_explicit&#39;, 0)],None,1963363
+No: 7   GFLOPS: 6.48/345.53     result: MeasureResult(costs=(0.03575116725,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.9435207843780518, timestamp=1680189110.9070928)      [(&#39;tile_f&#39;, [-1, 4, 1, 16]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 1, 1]), (&#39;tile_ry&#39;, [-1, 1, 3]), (&#39;tile_rx&#39;, [-1, 3, 1]), (&#39;auto_unroll_max_step&#39;, 0), (&#39;unroll_explicit&#39;, 0)],None,970586
+No: 8   GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 592, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 544, in _build_func_common
@@ -1101,10 +1312,9 @@ Traceback (most recent call last):
   File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 56, in tvm._ffi._cy3.core.tvm_callback
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 875, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 8, 64, 1]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 8, 32]), (&#39;tile_ry&#39;, [-1, 1, 1]), (&#39;tile_rx&#39;, [-1, 3, 1]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 0)],None,4220088
-No: 8   GFLOPS: 119.61/119.61   result: MeasureResult(costs=(0.001935420724137931,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.9431655406951904, timestamp=1680177394.8302236)       [(&#39;tile_f&#39;, [-1, 4, 8, 2]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 4, 8]), (&#39;tile_ry&#39;, [-1, 3, 1]), (&#39;tile_rx&#39;, [-1, 1, 1]), (&#39;auto_unroll_max_step&#39;, 0), (&#39;unroll_explicit&#39;, 1)],None,5525381
-No: 9   GFLOPS: 7.99/119.61     result: MeasureResult(costs=(0.02897599975,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.5775856971740723, timestamp=1680177397.6259804)      [(&#39;tile_f&#39;, [-1, 4, 1, 2]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 8, 1]), (&#39;tile_ry&#39;, [-1, 1, 1]), (&#39;tile_rx&#39;, [-1, 1, 1]), (&#39;auto_unroll_max_step&#39;, 0), (&#39;unroll_explicit&#39;, 0)],None,13917
-No: 10  GFLOPS: 0.00/119.61     result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 2, 64, 1]), (&#39;tile_y&#39;, [-1, 7, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 128, 2]), (&#39;tile_ry&#39;, [-1, 3, 1]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 0)],None,4902746
+No: 9   GFLOPS: 218.08/345.53   result: MeasureResult(costs=(0.0010615198210526316,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.275648593902588, timestamp=1680189112.3862677)       [(&#39;tile_f&#39;, [-1, 4, 64, 1]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 1, 4]), (&#39;tile_ry&#39;, [-1, 1, 3]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 0)],None,5100967
+No: 10  GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 592, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 544, in _build_func_common
@@ -1226,8 +1436,8 @@ Traceback (most recent call last):
   File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 56, in tvm._ffi._cy3.core.tvm_callback
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 875, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 4, 1, 32]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 16, 8]), (&#39;tile_ry&#39;, [-1, 1, 3]), (&#39;tile_rx&#39;, [-1, 3, 1]), (&#39;auto_unroll_max_step&#39;, 512), (&#39;unroll_explicit&#39;, 0)],None,2822127
-No: 11  GFLOPS: 0.00/119.61     result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 2, 1, 256]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 1, 32]), (&#39;tile_ry&#39;, [-1, 1, 3]), (&#39;tile_rx&#39;, [-1, 3, 1]), (&#39;auto_unroll_max_step&#39;, 512), (&#39;unroll_explicit&#39;, 1)],None,8080377
+No: 11  GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 592, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 544, in _build_func_common
@@ -1349,8 +1559,8 @@ Traceback (most recent call last):
   File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 56, in tvm._ffi._cy3.core.tvm_callback
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 875, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 16, 4, 4]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 7, 1, 1]), (&#39;tile_rc&#39;, [-1, 4, 128]), (&#39;tile_ry&#39;, [-1, 3, 1]), (&#39;tile_rx&#39;, [-1, 3, 1]), (&#39;auto_unroll_max_step&#39;, 0), (&#39;unroll_explicit&#39;, 1)],None,6182779
-No: 12  GFLOPS: 0.00/119.61     result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 16, 4, 4]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 1, 16]), (&#39;tile_ry&#39;, [-1, 1, 1]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 0), (&#39;unroll_explicit&#39;, 0)],None,1283159
+No: 12  GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 592, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 544, in _build_func_common
@@ -1472,8 +1682,8 @@ Traceback (most recent call last):
   File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 56, in tvm._ffi._cy3.core.tvm_callback
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 875, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 16, 4, 2]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 4, 4]), (&#39;tile_ry&#39;, [-1, 3, 1]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 0), (&#39;unroll_explicit&#39;, 0)],None,1431836
-No: 13  GFLOPS: 0.00/119.61     result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 1, 2, 8]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 32, 16]), (&#39;tile_ry&#39;, [-1, 1, 3]), (&#39;tile_rx&#39;, [-1, 3, 1]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 0)],None,4593523
+No: 13  GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 592, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 544, in _build_func_common
@@ -1595,9 +1805,8 @@ Traceback (most recent call last):
   File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 56, in tvm._ffi._cy3.core.tvm_callback
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 875, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 1, 128, 4]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 8, 16]), (&#39;tile_ry&#39;, [-1, 3, 1]), (&#39;tile_rx&#39;, [-1, 3, 1]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 0)],None,4392655
-No: 14  GFLOPS: 4.04/119.61     result: MeasureResult(costs=(0.05735276025,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.262774705886841, timestamp=1680177403.1119318)       [(&#39;tile_f&#39;, [-1, 2, 8, 32]), (&#39;tile_y&#39;, [-1, 7, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 1, 2]), (&#39;tile_ry&#39;, [-1, 1, 1]), (&#39;tile_rx&#39;, [-1, 1, 1]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 1)],None,8747618
-No: 15  GFLOPS: 0.00/119.61     result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 64, 4, 2]), (&#39;tile_y&#39;, [-1, 7, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 8, 16]), (&#39;tile_ry&#39;, [-1, 1, 3]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 0), (&#39;unroll_explicit&#39;, 1)],None,6909178
+No: 14  GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 592, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 544, in _build_func_common
@@ -1719,8 +1928,8 @@ Traceback (most recent call last):
   File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 56, in tvm._ffi._cy3.core.tvm_callback
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 875, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 1, 1, 16]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 16, 8]), (&#39;tile_ry&#39;, [-1, 1, 3]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 0), (&#39;unroll_explicit&#39;, 1)],None,6885284
-No: 16  GFLOPS: 0.00/119.61     result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 2, 8, 8]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 64, 8]), (&#39;tile_ry&#39;, [-1, 1, 1]), (&#39;tile_rx&#39;, [-1, 1, 1]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 0)],None,3601555
+No: 15  GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 592, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 544, in _build_func_common
@@ -1842,8 +2051,149 @@ Traceback (most recent call last):
   File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 56, in tvm._ffi._cy3.core.tvm_callback
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 875, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 8, 4, 16]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 2, 16]), (&#39;tile_ry&#39;, [-1, 3, 1]), (&#39;tile_rx&#39;, [-1, 1, 1]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 1)],None,9029418
-No: 17  GFLOPS: 0.00/119.61     result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 4, 2, 16]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 32, 4]), (&#39;tile_ry&#39;, [-1, 3, 1]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 0), (&#39;unroll_explicit&#39;, 1)],None,6669472
+No: 16  GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
+  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 592, in __call__
+    func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
+  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 544, in _build_func_common
+    func = build(s, args, target=target, runtime=runtime)
+  File &quot;/workspace/python/tvm/driver/build_module.py&quot;, line 227, in build
+    input_mod = lower(inputs, args, name=name, binds=binds)
+  File &quot;/workspace/python/tvm/driver/build_module.py&quot;, line 134, in lower
+    return ffi.lower_schedule(inp, args, name, binds, simple_mode)
+  File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 331, in tvm._ffi._cy3.core.PackedFuncBase.__call__
+  File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 276, in tvm._ffi._cy3.core.FuncCall
+  File &quot;tvm/_ffi/_cython/./base.pxi&quot;, line 181, in tvm._ffi._cy3.core.CHECK_CALL
+tvm._ffi.base.TVMError: Traceback (most recent call last):
+  24: TVMFuncCall
+        at ../src/runtime/c_runtime_api.cc:477
+  23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+        at ../include/tvm/runtime/packed_func.h:1217
+  22: Call
+        at ../include/tvm/runtime/packed_func.h:1213
+  21: operator()
+        at ../include/tvm/runtime/packed_func.h:1734
+  20: unpack_call&lt;tvm::IRModule, 5, tvm::&lt;lambda(tvm::te::Schedule, const tvm::runtime::Array&lt;tvm::runtime::ObjectRef&gt;&amp;, const tvm::runtime::String&amp;, const tvm::runtime::Map&lt;tvm::te::Tensor, tvm::tir::Buffer&gt;&amp;, bool)&gt; &gt;
+        at ../include/tvm/runtime/packed_func.h:1674
+  19: run&lt;&gt;
+        at ../include/tvm/runtime/packed_func.h:1634
+  18: run&lt;tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1634
+  17: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1634
+  16: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1634
+  15: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1634
+  14: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1649
+  13: operator()
+        at ../src/driver/driver_api.cc:402
+  12: tvm::LowerSchedule(tvm::te::Schedule, tvm::runtime::Array&lt;tvm::runtime::ObjectRef, void&gt; const&amp;, std::__cxx11::basic_string&lt;char, std::char_traits&lt;char&gt;, std::allocator&lt;char&gt; &gt; const&amp;, std::unordered_map&lt;tvm::te::Tensor, tvm::tir::Buffer, std::hash&lt;tvm::te::Tensor&gt;, std::equal_to&lt;tvm::te::Tensor&gt;, std::allocator&lt;std::pair&lt;tvm::te::Tensor const, tvm::tir::Buffer&gt; &gt; &gt; const&amp;, tvm::GlobalVarSupply, bool)
+        at ../src/driver/driver_api.cc:388
+  11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array&lt;tvm::transform::Pass, void&gt;)
+        at ../src/driver/driver_api.cc:283
+  10: tvm::transform::Pass::operator()(tvm::IRModule) const
+        at ../src/ir/transform.cc:258
+  9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&amp;) const
+        at ../src/ir/transform.cc:274
+  8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&amp;) const
+        at ../src/ir/transform.cc:451
+  7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&amp;) const
+        at ../src/ir/transform.cc:274
+  6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&amp;) const
+        at ../src/tir/ir/transform.cc:100
+  5: tvm::runtime::TypedPackedFunc&lt;tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)&gt;::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
+        at ../include/tvm/runtime/packed_func.h:1753
+  4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher&lt;tvm::tir::PrimFunc&gt;::run&lt;tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext&gt;(tvm::runtime::PackedFunc const&amp;, tvm::tir::PrimFunc&amp;&amp;, tvm::IRModule&amp;&amp;, tvm::transform::PassContext&amp;&amp;)
+        at ../include/tvm/runtime/packed_func.h:1697
+  3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()&lt;tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext&gt;(tvm::tir::PrimFunc&amp;&amp;, tvm::IRModule&amp;&amp;, tvm::transform::PassContext&amp;&amp;) const
+        at ../include/tvm/runtime/packed_func.h:1621
+  2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+        at ../include/tvm/runtime/packed_func.h:1217
+  1: Call
+        at ../include/tvm/runtime/packed_func.h:1213
+  0: operator()
+        at ../src/runtime/c_runtime_api.cc:534
+  File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 56, in tvm._ffi._cy3.core.tvm_callback
+  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 875, in verify_pass
+    raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel
+
+Traceback (most recent call last):
+  24: TVMFuncCall
+        at ../src/runtime/c_runtime_api.cc:477
+  23: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+        at ../include/tvm/runtime/packed_func.h:1217
+  22: Call
+        at ../include/tvm/runtime/packed_func.h:1213
+  21: operator()
+        at ../include/tvm/runtime/packed_func.h:1734
+  20: unpack_call&lt;tvm::IRModule, 5, tvm::&lt;lambda(tvm::te::Schedule, const tvm::runtime::Array&lt;tvm::runtime::ObjectRef&gt;&amp;, const tvm::runtime::String&amp;, const tvm::runtime::Map&lt;tvm::te::Tensor, tvm::tir::Buffer&gt;&amp;, bool)&gt; &gt;
+        at ../include/tvm/runtime/packed_func.h:1674
+  19: run&lt;&gt;
+        at ../include/tvm/runtime/packed_func.h:1634
+  18: run&lt;tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1634
+  17: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1634
+  16: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1634
+  15: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1634
+  14: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1649
+  13: operator()
+        at ../src/driver/driver_api.cc:402
+  12: tvm::LowerSchedule(tvm::te::Schedule, tvm::runtime::Array&lt;tvm::runtime::ObjectRef, void&gt; const&amp;, std::__cxx11::basic_string&lt;char, std::char_traits&lt;char&gt;, std::allocator&lt;char&gt; &gt; const&amp;, std::unordered_map&lt;tvm::te::Tensor, tvm::tir::Buffer, std::hash&lt;tvm::te::Tensor&gt;, std::equal_to&lt;tvm::te::Tensor&gt;, std::allocator&lt;std::pair&lt;tvm::te::Tensor const, tvm::tir::Buffer&gt; &gt; &gt; const&amp;, tvm::GlobalVarSupply, bool)
+        at ../src/driver/driver_api.cc:388
+  11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array&lt;tvm::transform::Pass, void&gt;)
+        at ../src/driver/driver_api.cc:283
+  10: tvm::transform::Pass::operator()(tvm::IRModule) const
+        at ../src/ir/transform.cc:258
+  9: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&amp;) const
+        at ../src/ir/transform.cc:274
+  8: tvm::transform::SequentialNode::operator()(tvm::IRModule, tvm::transform::PassContext const&amp;) const
+        at ../src/ir/transform.cc:451
+  7: tvm::transform::Pass::operator()(tvm::IRModule, tvm::transform::PassContext const&amp;) const
+        at ../src/ir/transform.cc:274
+  6: tvm::tir::transform::PrimFuncPassNode::operator()(tvm::IRModule, tvm::transform::PassContext const&amp;) const
+        at ../src/tir/ir/transform.cc:100
+  5: tvm::runtime::TypedPackedFunc&lt;tvm::tir::PrimFunc (tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext)&gt;::operator()(tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext) const
+        at ../include/tvm/runtime/packed_func.h:1753
+  4: tvm::tir::PrimFunc tvm::runtime::detail::typed_packed_call_dispatcher&lt;tvm::tir::PrimFunc&gt;::run&lt;tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext&gt;(tvm::runtime::PackedFunc const&amp;, tvm::tir::PrimFunc&amp;&amp;, tvm::IRModule&amp;&amp;, tvm::transform::PassContext&amp;&amp;)
+        at ../include/tvm/runtime/packed_func.h:1697
+  3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()&lt;tvm::tir::PrimFunc, tvm::IRModule, tvm::transform::PassContext&gt;(tvm::tir::PrimFunc&amp;&amp;, tvm::IRModule&amp;&amp;, tvm::transform::PassContext&amp;&amp;) const
+        at ../include/tvm/runtime/packed_func.h:1621
+  2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+        at ../include/tvm/runtime/packed_func.h:1217
+  1: Call
+        at ../include/tvm/runtime/packed_func.h:1213
+  0: operator()
+        at ../src/runtime/c_runtime_api.cc:534
+  File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 56, in tvm._ffi._cy3.core.tvm_callback
+  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 875, in verify_pass
+    raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 2, 64, 4]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 7, 1, 1]), (&#39;tile_rc&#39;, [-1, 128, 2]), (&#39;tile_ry&#39;, [-1, 3, 1]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 0), (&#39;unroll_explicit&#39;, 1)],None,6643694
+No: 17  GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
+  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 142, in build
+    res = future.result()
+  File &quot;/usr/lib/python3.7/concurrent/futures/_base.py&quot;, line 435, in result
+    return self.__get_result()
+  File &quot;/usr/lib/python3.7/concurrent/futures/_base.py&quot;, line 384, in __get_result
+    raise self._exception
+  File &quot;/usr/lib/python3.7/concurrent/futures/thread.py&quot;, line 57, in run
+    result = self.fn(*self.args, **self.kwargs)
+  File &quot;/workspace/python/tvm/contrib/popen_pool.py&quot;, line 432, in &lt;lambda&gt;
+    worker = lambda *args: self._worker_run(*args)
+  File &quot;/workspace/python/tvm/contrib/popen_pool.py&quot;, line 401, in _worker_run
+    return proc.recv()
+  File &quot;/workspace/python/tvm/contrib/popen_pool.py&quot;, line 309, in recv
+    raise TimeoutError()
+TimeoutError
+
+        [(&#39;tile_f&#39;, [-1, 64, 1, 2]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 2, 4]), (&#39;tile_ry&#39;, [-1, 1, 1]), (&#39;tile_rx&#39;, [-1, 3, 1]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 1)],None,9363701
+No: 18  GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 592, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 544, in _build_func_common
@@ -1965,8 +2315,8 @@ Traceback (most recent call last):
   File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 56, in tvm._ffi._cy3.core.tvm_callback
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 875, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 2, 64, 4]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 2, 32]), (&#39;tile_ry&#39;, [-1, 1, 1]), (&#39;tile_rx&#39;, [-1, 3, 1]), (&#39;auto_unroll_max_step&#39;, 0), (&#39;unroll_explicit&#39;, 0)],None,725914
-No: 18  GFLOPS: 0.00/119.61     result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 1, 1, 128]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 2, 256]), (&#39;tile_ry&#39;, [-1, 1, 1]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 512), (&#39;unroll_explicit&#39;, 0)],None,3093190
+No: 19  GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 592, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 544, in _build_func_common
@@ -2088,8 +2438,8 @@ Traceback (most recent call last):
   File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 56, in tvm._ffi._cy3.core.tvm_callback
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 875, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 1, 32, 8]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 2, 8]), (&#39;tile_ry&#39;, [-1, 1, 3]), (&#39;tile_rx&#39;, [-1, 3, 1]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 0)],None,4553281
-No: 19  GFLOPS: 0.00/119.61     result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 16, 16, 2]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 7, 1, 1]), (&#39;tile_rc&#39;, [-1, 128, 1]), (&#39;tile_ry&#39;, [-1, 1, 3]), (&#39;tile_rx&#39;, [-1, 3, 1]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 0)],None,4478849
+No: 20  GFLOPS: 0.00/345.53     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 592, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 544, in _build_func_common
@@ -2211,8 +2561,7 @@ Traceback (most recent call last):
   File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 56, in tvm._ffi._cy3.core.tvm_callback
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 875, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 1, 8, 4]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 16, 4]), (&#39;tile_ry&#39;, [-1, 3, 1]), (&#39;tile_rx&#39;, [-1, 3, 1]), (&#39;auto_unroll_max_step&#39;, 0), (&#39;unroll_explicit&#39;, 0)],None,857901
-No: 20  GFLOPS: 17.43/119.61    result: MeasureResult(costs=(0.013282744444444446,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.1519412994384766, timestamp=1680177406.5303857)       [(&#39;tile_f&#39;, [-1, 4, 16, 1]), (&#39;tile_y&#39;, [-1, 7, 1, 1]), (&#39;tile_x&#39;, [-1, 7, 1, 1]), (&#39;tile_rc&#39;, [-1, 1, 32]), (&#39;tile_ry&#39;, [-1, 1, 3]), (&#39;tile_rx&#39;, [-1, 1, 1]), (&#39;auto_unroll_max_step&#39;, 0), (&#39;unroll_explicit&#39;, 0)],None,529136
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 1, 4, 4]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 1, 256]), (&#39;tile_ry&#39;, [-1, 3, 1]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 512), (&#39;unroll_explicit&#39;, 1)],None,8508615
 </pre></div>
 </div>
 <p>Finally we can inspect the best config from log file, check correctness,
@@ -2251,9 +2600,9 @@ and measure running time.</p>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Finish loading 20 records
 
 Best config:
-[(&#39;tile_f&#39;, [-1, 4, 8, 2]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 4, 8]), (&#39;tile_ry&#39;, [-1, 3, 1]), (&#39;tile_rx&#39;, [-1, 1, 1]), (&#39;auto_unroll_max_step&#39;, 0), (&#39;unroll_explicit&#39;, 1)],None,5525381
+[(&#39;tile_f&#39;, [-1, 1, 16, 4]), (&#39;tile_y&#39;, [-1, 7, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 1, 8]), (&#39;tile_ry&#39;, [-1, 3, 1]), (&#39;tile_rx&#39;, [-1, 1, 1]), (&#39;auto_unroll_max_step&#39;, 512), (&#39;unroll_explicit&#39;, 0)],None,2033146
 Finish loading 20 records
-Time cost of this operator: 0.002253
+Time cost of this operator: 0.001035
 </pre></div>
 </div>
 <div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-tune-with-autotvm-tune-conv2d-cuda-py">
diff --git a/docs/how_to/work_with_microtvm/micro_autotune.html b/docs/how_to/work_with_microtvm/micro_autotune.html
index a5a2fef3fe..016f3b28ed 100644
--- a/docs/how_to/work_with_microtvm/micro_autotune.html
+++ b/docs/how_to/work_with_microtvm/micro_autotune.html
@@ -649,10 +649,10 @@ the tuned operator.</p>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>########## Build without Autotuning ##########
 Node Name                                     Ops                                           Time(us)  Time(%)  Shape              Inputs  Outputs  Measurements(us)
 ---------                                     ---                                           --------  -------  -----              ------  -------  ----------------
-tvmgen_default_fused_nn_contrib_conv2d_NCHWc  tvmgen_default_fused_nn_contrib_conv2d_NCHWc  315.2     98.742   (1, 2, 10, 10, 3)  2       1        [315.2]
-tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       3.048     0.955    (1, 6, 10, 10)     1       1        [3.048]
-tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.967     0.303    (1, 1, 10, 10, 3)  1       1        [0.967]
-Total_time                                    -                                             319.215   -        -                  -       -        -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc  tvmgen_default_fused_nn_contrib_conv2d_NCHWc  316.0     98.731   (1, 2, 10, 10, 3)  2       1        [316.0]
+tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       3.103     0.969    (1, 6, 10, 10)     1       1        [3.103]
+tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.96      0.3      (1, 1, 10, 10, 3)  1       1        [0.96]
+Total_time                                    -                                             320.063   -        -                  -       -        -
 </pre></div>
 </div>
 </div>
@@ -704,13 +704,13 @@ Total_time                                    -
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>########## Build with Autotuning ##########
 Node Name                                     Ops                                           Time(us)  Time(%)  Shape              Inputs  Outputs  Measurements(us)
 ---------                                     ---                                           --------  -------  -----              ------  -------  ----------------
-tvmgen_default_fused_nn_contrib_conv2d_NCHWc  tvmgen_default_fused_nn_contrib_conv2d_NCHWc  137.9     98.101   (1, 6, 10, 10, 1)  2       1        [137.9]
-tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       1.81      1.288    (1, 6, 10, 10)     1       1        [1.81]
-tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.859     0.611    (1, 3, 10, 10, 1)  1       1        [0.859]
-Total_time                                    -                                             140.57    -        -                  -       -        -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc  tvmgen_default_fused_nn_contrib_conv2d_NCHWc  101.3     97.346   (1, 6, 10, 10, 1)  2       1        [101.3]
+tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       1.789     1.72     (1, 6, 10, 10)     1       1        [1.789]
+tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.972     0.934    (1, 1, 10, 10, 3)  1       1        [0.972]
+Total_time                                    -                                             104.062   -        -                  -       -        -
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  23.369 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  26.505 seconds)</p>
 <div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-work-with-microtvm-micro-autotune-py">
 <div class="sphx-glr-download sphx-glr-download-python docutils container">
 <p><a class="reference download internal" download="" href="../../_downloads/9ccca8fd489a1486ac71b55a55c320c5/micro_autotune.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">micro_autotune.py</span></code></a></p>
diff --git a/docs/how_to/work_with_microtvm/micro_pytorch.html b/docs/how_to/work_with_microtvm/micro_pytorch.html
index caabf68591..0e5ae91077 100644
--- a/docs/how_to/work_with_microtvm/micro_pytorch.html
+++ b/docs/how_to/work_with_microtvm/micro_pytorch.html
@@ -460,8 +460,7 @@ download a cat image and preprocess it to use as the model input.</p>
 Downloading: &quot;https://download.pytorch.org/models/quantized/mobilenet_v2_qnnpack_37f702c5.pth&quot; to /workspace/.cache/torch/hub/checkpoints/mobilenet_v2_qnnpack_37f702c5.pth
 
   0%|          | 0.00/3.42M [00:00&lt;?, ?B/s]
- 61%|######    | 2.09M/3.42M [00:00&lt;00:00, 19.0MB/s]
-100%|##########| 3.42M/3.42M [00:00&lt;00:00, 29.5MB/s]
+100%|##########| 3.42M/3.42M [00:00&lt;00:00, 74.9MB/s]
 /workspace/python/tvm/relay/frontend/pytorch_utils.py:47: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead.
   return LooseVersion(torch_ver) &gt; ver
 /venv/apache-tvm-py3.7/lib/python3.7/site-packages/setuptools/_distutils/version.py:346: DeprecationWarning: distutils Version classes are deprecated. Use packaging.version instead.
@@ -587,7 +586,7 @@ via the host <cite>main.cc`</cite> or if a Zephyr emulated board is selected as
 Torch top-1 id: 282, class name: tiger cat
 </pre></div>
 </div>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  19.859 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  22.711 seconds)</p>
 <div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-work-with-microtvm-micro-pytorch-py">
 <div class="sphx-glr-download sphx-glr-download-python docutils container">
 <p><a class="reference download internal" download="" href="../../_downloads/12b9ecc04c41abaa12022061771821d1/micro_pytorch.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">micro_pytorch.py</span></code></a></p>
diff --git a/docs/how_to/work_with_microtvm/micro_train.html b/docs/how_to/work_with_microtvm/micro_train.html
index 9fc4bff4c0..4f67ea5376 100644
--- a/docs/how_to/work_with_microtvm/micro_train.html
+++ b/docs/how_to/work_with_microtvm/micro_train.html
@@ -528,7 +528,7 @@ take about <strong>2 minutes</strong> to download the Stanford Cars, while COCO
 <a href="https://docs.python.org/3/library/shutil.html#shutil.move" title="shutil.move" class="sphx-glr-backref-module-shutil sphx-glr-backref-type-py-function"><span class="n">shutil</span><span class="o">.</span><span class="n">move</span></a><span class="p">(</span><span class="sa">f</span><span class="s2">&quot;</span><span class="si">{</span><a href="https://docs.python.org/3/library/stdtypes.html#str" title="builtins.str" class="sphx-glr-backref-module-builtins sphx-glr-backref-typ [...]
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>&#39;/tmp/tmp29az3utf/images/random&#39;
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>&#39;/tmp/tmpdtgsg1v_/images/random&#39;
 </pre></div>
 </div>
 </div>
@@ -588,8 +588,8 @@ objects to other stuff? We can display some examples from our datasets using <co
     <span class="n">plt</span><span class="o">.</span><span class="n">axis</span><span class="p">(</span><span class="s2">&quot;off&quot;</span><span class="p">)</span>
 </pre></div>
 </div>
-<img src="../../_images/sphx_glr_micro_train_001.png" srcset="../../_images/sphx_glr_micro_train_001.png" alt="[0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0]" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/tmp/tmp29az3utf/images/target contains 8144 images
-/tmp/tmp29az3utf/images/random contains 5000 images
+<img src="../../_images/sphx_glr_micro_train_001.png" srcset="../../_images/sphx_glr_micro_train_001.png" alt="[1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [1.0, 0.0]" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/tmp/tmpdtgsg1v_/images/target contains 8144 images
+/tmp/tmpdtgsg1v_/images/random contains 5000 images
 </pre></div>
 </div>
 </div>
@@ -701,13 +701,13 @@ the time on our validation set).</p>
 </pre></div>
 </div>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Epoch 1/3
-328/328 - 41s - loss: 0.2038 - accuracy: 0.9279 - val_loss: 0.1232 - val_accuracy: 0.9551 - 41s/epoch - 124ms/step
+328/328 - 40s - loss: 0.2212 - accuracy: 0.9237 - val_loss: 0.1385 - val_accuracy: 0.9535 - 40s/epoch - 122ms/step
 Epoch 2/3
-328/328 - 35s - loss: 0.1043 - accuracy: 0.9628 - val_loss: 0.0954 - val_accuracy: 0.9660 - 35s/epoch - 106ms/step
+328/328 - 35s - loss: 0.0964 - accuracy: 0.9647 - val_loss: 0.1278 - val_accuracy: 0.9588 - 35s/epoch - 107ms/step
 Epoch 3/3
-328/328 - 35s - loss: 0.0742 - accuracy: 0.9718 - val_loss: 0.0743 - val_accuracy: 0.9713 - 35s/epoch - 105ms/step
+328/328 - 35s - loss: 0.0650 - accuracy: 0.9745 - val_loss: 0.1614 - val_accuracy: 0.9535 - 35s/epoch - 107ms/step
 
-&lt;keras.callbacks.History object at 0x7f82e72d6dd0&gt;
+&lt;keras.callbacks.History object at 0x7f5d233527d0&gt;
 </pre></div>
 </div>
 </div>
@@ -971,7 +971,7 @@ as intended.</p>
 <p>From here, we could modify the model to read live images from the camera - we have another
 Arduino tutorial for how to do that <a class="reference external" href="https://github.com/guberti/tvm-arduino-demos/tree/master/examples/person_detection">on GitHub</a>. Alternatively, we could also
 <a class="reference external" href="https://tvm.apache.org/docs/how_to/work_with_microtvm/micro_autotune.html">use TVM’s autotuning capabilities</a> to dramatically improve the model’s performance.</p>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 3 minutes  51.843 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 4 minutes  22.021 seconds)</p>
 <div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-work-with-microtvm-micro-train-py">
 <div class="sphx-glr-download sphx-glr-download-python docutils container">
 <p><a class="reference download internal" download="" href="../../_downloads/b52cec46baf4f78d6bcd94cbe269c8a6/micro_train.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">micro_train.py</span></code></a></p>
diff --git a/docs/how_to/work_with_microtvm/sg_execution_times.html b/docs/how_to/work_with_microtvm/sg_execution_times.html
index c6aaeeb0ec..c16b04cd4b 100644
--- a/docs/how_to/work_with_microtvm/sg_execution_times.html
+++ b/docs/how_to/work_with_microtvm/sg_execution_times.html
@@ -345,7 +345,7 @@
             
   <div class="section" id="computation-times">
 <span id="sphx-glr-how-to-work-with-microtvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>07:01.210</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
+<p><strong>07:38.176</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 82%" />
@@ -354,27 +354,27 @@
 </colgroup>
 <tbody>
 <tr class="row-odd"><td><p><a class="reference internal" href="micro_train.html#sphx-glr-how-to-work-with-microtvm-micro-train-py"><span class="std std-ref">5. Training Vision Models for microTVM on Arduino</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_train.py</span></code>)</p></td>
-<td><p>03:51.843</p></td>
+<td><p>04:22.021</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="micro_autotune.html#sphx-glr-how-to-work-with-microtvm-micro-autotune-py"><span class="std std-ref">6. Model Tuning with microTVM</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_autotune.py</span></code>)</p></td>
-<td><p>01:23.369</p></td>
+<td><p>01:26.505</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="micro_pytorch.html#sphx-glr-how-to-work-with-microtvm-micro-pytorch-py"><span class="std std-ref">4. microTVM PyTorch Tutorial</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_pytorch.py</span></code>)</p></td>
-<td><p>01:19.859</p></td>
+<td><p>01:22.711</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="micro_aot.html#sphx-glr-how-to-work-with-microtvm-micro-aot-py"><span class="std std-ref">3. microTVM Ahead-of-Time (AOT) Compilation</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_aot.py</span></code>)</p></td>
-<td><p>00:10.468</p></td>
+<td><p>00:10.806</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="micro_custom_ide.html#sphx-glr-how-to-work-with-microtvm-micro-custom-ide-py"><span class="std std-ref">9. Bring microTVM to your own development environment</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_custom_ide.py</span></code>)</p></td>
-<td><p>00:08.123</p></td>
+<td><p>00:08.436</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="micro_tflite.html#sphx-glr-how-to-work-with-microtvm-micro-tflite-py"><span class="std std-ref">2. microTVM TFLite Tutorial</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_tflite.py</span></code>)</p></td>
-<td><p>00:07.549</p></td>
+<td><p>00:07.697</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="micro_ethosu.html#sphx-glr-how-to-work-with-microtvm-micro-ethosu-py"><span class="std std-ref">7. Running TVM on bare metal Arm(R) Cortex(R)-M55 CPU and Ethos(TM)-U55 NPU with CMSIS-NN</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_ethosu.py</span></code>)</p></td>
diff --git a/docs/how_to/work_with_relay/sg_execution_times.html b/docs/how_to/work_with_relay/sg_execution_times.html
index 975e286436..8909a48020 100644
--- a/docs/how_to/work_with_relay/sg_execution_times.html
+++ b/docs/how_to/work_with_relay/sg_execution_times.html
@@ -345,7 +345,7 @@
             
   <div class="section" id="computation-times">
 <span id="sphx-glr-how-to-work-with-relay-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:37.273</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
+<p><strong>00:38.943</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 84%" />
@@ -354,15 +354,15 @@
 </colgroup>
 <tbody>
 <tr class="row-odd"><td><p><a class="reference internal" href="using_pipeline_executor.html#sphx-glr-how-to-work-with-relay-using-pipeline-executor-py"><span class="std std-ref">Using Pipeline Executor in Relay</span></a> (<code class="docutils literal notranslate"><span class="pre">using_pipeline_executor.py</span></code>)</p></td>
-<td><p>00:32.622</p></td>
+<td><p>00:33.769</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="using_external_lib.html#sphx-glr-how-to-work-with-relay-using-external-lib-py"><span class="std std-ref">Using External Libraries in Relay</span></a> (<code class="docutils literal notranslate"><span class="pre">using_external_lib.py</span></code>)</p></td>
-<td><p>00:02.930</p></td>
+<td><p>00:03.042</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="build_gcn.html#sphx-glr-how-to-work-with-relay-build-gcn-py"><span class="std std-ref">Building a Graph Convolutional Network</span></a> (<code class="docutils literal notranslate"><span class="pre">build_gcn.py</span></code>)</p></td>
-<td><p>00:01.714</p></td>
+<td><p>00:02.125</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="using_relay_viz.html#sphx-glr-how-to-work-with-relay-using-relay-viz-py"><span class="std std-ref">Use Relay Visualizer to Visualize Relay</span></a> (<code class="docutils literal notranslate"><span class="pre">using_relay_viz.py</span></code>)</p></td>
diff --git a/docs/how_to/work_with_schedules/intrin_math.html b/docs/how_to/work_with_schedules/intrin_math.html
index b896cde593..2352613a61 100644
--- a/docs/how_to/work_with_schedules/intrin_math.html
+++ b/docs/how_to/work_with_schedules/intrin_math.html
@@ -540,7 +540,7 @@ The following example customizes CUDA lowering rule for <code class="code docuti
 <a href="../../reference/api/python/ir.html#tvm.ir.register_intrin_lowering" title="tvm.ir.register_intrin_lowering" class="sphx-glr-backref-module-tvm-ir sphx-glr-backref-type-py-function"><span class="n">register_intrin_lowering</span></a><span class="p">(</span><span class="s2">&quot;tir.exp&quot;</span><span class="p">,</span> <span class="n">target</span><span class="o">=</span><span class="s2">&quot;cuda&quot;</span><span class="p">,</span> <span class="n">f</span><span class="o">= [...]
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>&lt;function my_cuda_math_rule at 0x7f832c24e560&gt;
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>&lt;function my_cuda_math_rule at 0x7f5971ece170&gt;
 </pre></div>
 </div>
 <p>Register the rule to TVM with override option to override existing rule.
diff --git a/docs/how_to/work_with_schedules/sg_execution_times.html b/docs/how_to/work_with_schedules/sg_execution_times.html
index b900237d50..950e0a6255 100644
--- a/docs/how_to/work_with_schedules/sg_execution_times.html
+++ b/docs/how_to/work_with_schedules/sg_execution_times.html
@@ -345,7 +345,7 @@
             
   <div class="section" id="computation-times">
 <span id="sphx-glr-how-to-work-with-schedules-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:06.679</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
+<p><strong>00:08.778</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 83%" />
@@ -354,35 +354,35 @@
 </colgroup>
 <tbody>
 <tr class="row-odd"><td><p><a class="reference internal" href="intrin_math.html#sphx-glr-how-to-work-with-schedules-intrin-math-py"><span class="std std-ref">Intrinsics and Math Functions</span></a> (<code class="docutils literal notranslate"><span class="pre">intrin_math.py</span></code>)</p></td>
-<td><p>00:03.896</p></td>
+<td><p>00:05.950</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="tensorize.html#sphx-glr-how-to-work-with-schedules-tensorize-py"><span class="std std-ref">Use Tensorize to Leverage Hardware Intrinsics</span></a> (<code class="docutils literal notranslate"><span class="pre">tensorize.py</span></code>)</p></td>
-<td><p>00:01.254</p></td>
+<td><p>00:01.257</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="reduction.html#sphx-glr-how-to-work-with-schedules-reduction-py"><span class="std std-ref">Reduction</span></a> (<code class="docutils literal notranslate"><span class="pre">reduction.py</span></code>)</p></td>
-<td><p>00:00.629</p></td>
+<td><p>00:00.643</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="scan.html#sphx-glr-how-to-work-with-schedules-scan-py"><span class="std std-ref">Scan and Recurrent Kernel</span></a> (<code class="docutils literal notranslate"><span class="pre">scan.py</span></code>)</p></td>
-<td><p>00:00.618</p></td>
+<td><p>00:00.632</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="extern_op.html#sphx-glr-how-to-work-with-schedules-extern-op-py"><span class="std std-ref">External Tensor Functions</span></a> (<code class="docutils literal notranslate"><span class="pre">extern_op.py</span></code>)</p></td>
-<td><p>00:00.132</p></td>
+<td><p>00:00.137</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="tedd.html#sphx-glr-how-to-work-with-schedules-tedd-py"><span class="std std-ref">Use Tensor Expression Debug Display (TEDD) for Visualization</span></a> (<code class="docutils literal notranslate"><span class="pre">tedd.py</span></code>)</p></td>
-<td><p>00:00.065</p></td>
+<td><p>00:00.068</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="schedule_primitives.html#sphx-glr-how-to-work-with-schedules-schedule-primitives-py"><span class="std std-ref">Schedule Primitives in TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">schedule_primitives.py</span></code>)</p></td>
-<td><p>00:00.055</p></td>
+<td><p>00:00.059</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="tuple_inputs.html#sphx-glr-how-to-work-with-schedules-tuple-inputs-py"><span class="std std-ref">Compute and Reduce with Tuple Inputs</span></a> (<code class="docutils literal notranslate"><span class="pre">tuple_inputs.py</span></code>)</p></td>
-<td><p>00:00.030</p></td>
+<td><p>00:00.032</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 </tbody>
diff --git a/docs/install/nnpack.html b/docs/install/nnpack.html
index e285b427a5..1515c4b747 100644
--- a/docs/install/nnpack.html
+++ b/docs/install/nnpack.html
@@ -234,17 +234,7 @@
               <p class="caption" role="heading"><span class="caption-text">Getting Started</span></p>
 <ul class="current">
 <li class="toctree-l1 current"><a class="reference internal" href="index.html">Installing TVM</a><ul class="current">
-<li class="toctree-l2 current"><a class="reference internal" href="from_source.html">Install from Source</a><ul class="current">
-<li class="toctree-l3"><a class="reference internal" href="from_source.html#developers-get-source-from-github">Developers: Get Source from Github</a></li>
-<li class="toctree-l3"><a class="reference internal" href="from_source.html#build-the-shared-library">Build the Shared Library</a></li>
-<li class="toctree-l3"><a class="reference internal" href="from_source.html#python-package-installation">Python Package Installation</a></li>
-<li class="toctree-l3 current"><a class="reference internal" href="from_source.html#install-contrib-libraries">Install Contrib Libraries</a><ul class="current">
-<li class="toctree-l4 current"><a class="current reference internal" href="#">NNPACK Contrib Installation</a></li>
-</ul>
-</li>
-<li class="toctree-l3"><a class="reference internal" href="from_source.html#enable-c-tests">Enable C++ Tests</a></li>
-</ul>
-</li>
+<li class="toctree-l2"><a class="reference internal" href="from_source.html">Install from Source</a></li>
 <li class="toctree-l2"><a class="reference internal" href="docker.html">Docker Images</a></li>
 <li class="toctree-l2 current"><a class="current reference internal" href="#">NNPACK Contrib Installation</a><ul>
 <li class="toctree-l3"><a class="reference internal" href="#conditions">Conditions</a></li>
diff --git a/docs/reference/api/doxygen/relay_2attrs_2nn_8h_source.html b/docs/reference/api/doxygen/relay_2attrs_2nn_8h_source.html
index ee0b9e38e3..a9538086b8 100644
--- a/docs/reference/api/doxygen/relay_2attrs_2nn_8h_source.html
+++ b/docs/reference/api/doxygen/relay_2attrs_2nn_8h_source.html
@@ -66,7 +66,7 @@ $(function() {
 <div class="title">nn.h</div>  </div>
 </div><!--header-->
 <div class="contents">
-<a href="relay_2attrs_2nn_8h.html">Go to the documentation of this file.</a><div class="fragment"><div class="line"><a name="l00001"></a><span class="lineno">    1</span>&#160;<span class="comment">/*</span></div><div class="line"><a name="l00002"></a><span class="lineno">    2</span>&#160;<span class="comment"> * Licensed to the Apache Software Foundation (ASF) under one</span></div><div class="line"><a name="l00003"></a><span class="lineno">    3</span>&#160;<span class="comment"> * or [...]
+<a href="relay_2attrs_2nn_8h.html">Go to the documentation of this file.</a><div class="fragment"><div class="line"><a name="l00001"></a><span class="lineno">    1</span>&#160;<span class="comment">/*</span></div><div class="line"><a name="l00002"></a><span class="lineno">    2</span>&#160;<span class="comment"> * Licensed to the Apache Software Foundation (ASF) under one</span></div><div class="line"><a name="l00003"></a><span class="lineno">    3</span>&#160;<span class="comment"> * or [...]
 <div class="ttc" id="structtvm_1_1relay_1_1Conv3DAttrs_html_aed6a9346217ee4817b028734c660cfb0"><div class="ttname"><a href="structtvm_1_1relay_1_1Conv3DAttrs.html#aed6a9346217ee4817b028734c660cfb0">tvm::relay::Conv3DAttrs::dilation</a></div><div class="ttdeci">Array&lt; IndexExpr &gt; dilation</div><div class="ttdef"><b>Definition:</b> nn.h:306</div></div>
 <div class="ttc" id="structtvm_1_1relay_1_1UpSampling3DAttrs_html_af8891c263011a72ee4989a601b860f08"><div class="ttname"><a href="structtvm_1_1relay_1_1UpSampling3DAttrs.html#af8891c263011a72ee4989a601b860f08">tvm::relay::UpSampling3DAttrs::layout</a></div><div class="ttdeci">std::string layout</div><div class="ttdef"><b>Definition:</b> nn.h:1215</div></div>
 <div class="ttc" id="structtvm_1_1relay_1_1MaxPool2DAttrs_html"><div class="ttname"><a href="structtvm_1_1relay_1_1MaxPool2DAttrs.html">tvm::relay::MaxPool2DAttrs</a></div><div class="ttdoc">Attributes for max pool operator. </div><div class="ttdef"><b>Definition:</b> nn.h:692</div></div>
diff --git a/docs/reference/api/python/auto_scheduler.html b/docs/reference/api/python/auto_scheduler.html
index 725447251c..951eac5cdf 100644
--- a/docs/reference/api/python/auto_scheduler.html
+++ b/docs/reference/api/python/auto_scheduler.html
@@ -1621,7 +1621,7 @@ history states as starting point to perform Evolutionary Search).</p></li>
 
 <dl class="py class">
 <dt class="sig sig-object py" id="tvm.auto_scheduler.SketchPolicy">
-<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">SketchPolicy</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">program_cost_model</span></span><span class="o"><span class="pre">=</span></span><span class="defau [...]
+<em class="property"><span class="pre">class</span> </em><span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">SketchPolicy</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">program_cost_model</span></span><span class="o"><span class="pre">=</span></span><span class="defau [...]
 <dd><p>The search policy that searches in a hierarchical search space defined by sketches.
 The policy randomly samples programs from the space defined by sketches and use evolutionary
 search to fine-tune them.</p>
@@ -1905,7 +1905,7 @@ Candidates:
 
 <dl class="py function">
 <dt class="sig sig-object py" id="tvm.auto_scheduler.auto_schedule">
-<span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">auto_schedule</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">search_policy</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em clas [...]
+<span class="sig-prename descclassname"><span class="pre">tvm.auto_scheduler.</span></span><span class="sig-name descname"><span class="pre">auto_schedule</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">task</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">search_policy</span></span><span class="o"><span class="pre">=</span></span><span class="default_value"><span class="pre">None</span></span></em>, <em clas [...]
 <dd><p>THIS API IS DEPRECATED.</p>
 <p>Run auto scheduling search for a task.</p>
 <dl class="field-list simple">
diff --git a/docs/reference/api/python/relay/nn.html b/docs/reference/api/python/relay/nn.html
index e1e149815d..88adcb345b 100644
--- a/docs/reference/api/python/relay/nn.html
+++ b/docs/reference/api/python/relay/nn.html
@@ -1825,7 +1825,7 @@ then convert to the out_layout.</p>
 
 <dl class="py function">
 <dt class="sig sig-object py" id="tvm.relay.nn.conv3d_transpose">
-<span class="sig-prename descclassname"><span class="pre">tvm.relay.nn.</span></span><span class="sig-name descname"><span class="pre">conv3d_transpose</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">data</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weight</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">strides</span></span><span class="o"><span class="pre">=</span></span><span [...]
+<span class="sig-prename descclassname"><span class="pre">tvm.relay.nn.</span></span><span class="sig-name descname"><span class="pre">conv3d_transpose</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">data</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">weight</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">strides</span></span><span class="o"><span class="pre">=</span></span><span [...]
 <dd><p>3D transpose convolution.</p>
 <dl class="field-list simple">
 <dt class="field-odd">Parameters</dt>
diff --git a/docs/reference/api/typedoc/classes/bytestreamreader.html b/docs/reference/api/typedoc/classes/bytestreamreader.html
index 9cf42e96d3..b32602cd85 100644
--- a/docs/reference/api/typedoc/classes/bytestreamreader.html
+++ b/docs/reference/api/typedoc/classes/bytestreamreader.html
@@ -119,7 +119,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/rpc_server.ts#L45">rpc_server.ts:45</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/rpc_server.ts#L45">rpc_server.ts:45</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -141,7 +141,7 @@
 					<div class="tsd-signature tsd-kind-icon">bytes<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Uint8Array</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/rpc_server.ts#L45">rpc_server.ts:45</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/rpc_server.ts#L45">rpc_server.ts:45</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -151,7 +151,7 @@
 					<div class="tsd-signature tsd-kind-icon">offset<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 0</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/rpc_server.ts#L44">rpc_server.ts:44</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/rpc_server.ts#L44">rpc_server.ts:44</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -168,7 +168,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/rpc_server.ts#L65">rpc_server.ts:65</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/rpc_server.ts#L65">rpc_server.ts:65</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">Uint8Array</span></h4>
@@ -185,7 +185,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/rpc_server.ts#L51">rpc_server.ts:51</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/rpc_server.ts#L51">rpc_server.ts:51</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
@@ -202,7 +202,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/rpc_server.ts#L59">rpc_server.ts:59</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/rpc_server.ts#L59">rpc_server.ts:59</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
diff --git a/docs/reference/api/typedoc/classes/cachedcallstack.html b/docs/reference/api/typedoc/classes/cachedcallstack.html
index 5e7898fc9b..fc33bd06c7 100644
--- a/docs/reference/api/typedoc/classes/cachedcallstack.html
+++ b/docs/reference/api/typedoc/classes/cachedcallstack.html
@@ -144,7 +144,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L223">memory.ts:223</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L223">memory.ts:223</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -172,7 +172,7 @@
 					<div class="tsd-signature tsd-kind-icon">temp<wbr>Args<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Array</span><span class="tsd-signature-symbol">&lt;</span><a href="../interfaces/disposable.html" class="tsd-signature-type">Disposable</a><span class="tsd-signature-symbol">&gt;</span><span class="tsd-signature-symbol"> = []</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L208">memory.ts:208</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L208">memory.ts:208</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -194,7 +194,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L312">memory.ts:312</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L312">memory.ts:312</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -226,7 +226,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L284">memory.ts:284</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L284">memory.ts:284</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -262,7 +262,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L388">memory.ts:388</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L388">memory.ts:388</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -300,7 +300,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L376">memory.ts:376</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L376">memory.ts:376</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -340,7 +340,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L267">memory.ts:267</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L267">memory.ts:267</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -373,7 +373,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L243">memory.ts:243</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L243">memory.ts:243</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -390,7 +390,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L321">memory.ts:321</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L321">memory.ts:321</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -422,7 +422,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L252">memory.ts:252</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L252">memory.ts:252</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -444,7 +444,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L359">memory.ts:359</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L359">memory.ts:359</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -470,7 +470,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L342">memory.ts:342</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L342">memory.ts:342</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -496,7 +496,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L350">memory.ts:350</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L350">memory.ts:350</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -522,7 +522,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L326">memory.ts:326</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L326">memory.ts:326</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -548,7 +548,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L363">memory.ts:363</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L363">memory.ts:363</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -574,7 +574,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L346">memory.ts:346</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L346">memory.ts:346</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -600,7 +600,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L334">memory.ts:334</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L334">memory.ts:334</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
diff --git a/docs/reference/api/typedoc/classes/dldatatype.html b/docs/reference/api/typedoc/classes/dldatatype.html
index 59bd6ed09a..0cada838e9 100644
--- a/docs/reference/api/typedoc/classes/dldatatype.html
+++ b/docs/reference/api/typedoc/classes/dldatatype.html
@@ -119,7 +119,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L359">runtime.ts:359</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L359">runtime.ts:359</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -147,7 +147,7 @@
 					<div class="tsd-signature tsd-kind-icon">bits<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L357">runtime.ts:357</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L357">runtime.ts:357</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -162,7 +162,7 @@
 					<div class="tsd-signature tsd-kind-icon">code<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L355">runtime.ts:355</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L355">runtime.ts:355</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -177,7 +177,7 @@
 					<div class="tsd-signature tsd-kind-icon">lanes<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L359">runtime.ts:359</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L359">runtime.ts:359</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -199,7 +199,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L376">runtime.ts:376</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L376">runtime.ts:376</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
@@ -216,7 +216,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L367">runtime.ts:367</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L367">runtime.ts:367</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">string</span></h4>
diff --git a/docs/reference/api/typedoc/classes/dldevice.html b/docs/reference/api/typedoc/classes/dldevice.html
index fc5939971c..8554f6a3c9 100644
--- a/docs/reference/api/typedoc/classes/dldevice.html
+++ b/docs/reference/api/typedoc/classes/dldevice.html
@@ -118,7 +118,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L299">runtime.ts:299</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L299">runtime.ts:299</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -146,7 +146,7 @@
 					<div class="tsd-signature tsd-kind-icon">device<wbr>Id<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L297">runtime.ts:297</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L297">runtime.ts:297</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -161,7 +161,7 @@
 					<div class="tsd-signature tsd-kind-icon">device<wbr>Type<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L295">runtime.ts:295</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L295">runtime.ts:295</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -183,7 +183,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L320">runtime.ts:320</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L320">runtime.ts:320</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -205,7 +205,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L327">runtime.ts:327</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L327">runtime.ts:327</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">string</span></h4>
diff --git a/docs/reference/api/typedoc/classes/environment.html b/docs/reference/api/typedoc/classes/environment.html
index 05024ab305..49d6b4e457 100644
--- a/docs/reference/api/typedoc/classes/environment.html
+++ b/docs/reference/api/typedoc/classes/environment.html
@@ -125,7 +125,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/environment.ts#L86">environment.ts:86</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/environment.ts#L86">environment.ts:86</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -169,7 +169,7 @@
 					<aside class="tsd-sources">
 						<p>Implementation of <a href="../interfaces/libraryprovider.html">LibraryProvider</a>.<a href="../interfaces/libraryprovider.html#imports">imports</a></p>
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/environment.ts#L70">environment.ts:70</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/environment.ts#L70">environment.ts:70</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -179,7 +179,7 @@
 					<div class="tsd-signature tsd-kind-icon">logger<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>msg<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> =&gt; </span><span class="tsd-signature-type">void</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/environment.ts#L69">environment.ts:69</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/environment.ts#L69">environment.ts:69</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-type-declaration">
@@ -210,7 +210,7 @@
 					<div class="tsd-signature tsd-kind-icon">packedCFunc<wbr>Table<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Array</span><span class="tsd-signature-symbol">&lt;</span><span class="tsd-signature-type">ctypes.FTVMWasmPackedCFunc</span><span class="tsd-signature-symbol"> | </span><span class="tsd-signature-type">undefined</span><span class="tsd-signature-symbol">&gt;</span><span class="tsd-signature-symbol"> = [undefined,]</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/environment.ts#L78">environment.ts:78</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/environment.ts#L78">environment.ts:78</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -228,7 +228,7 @@
 					<div class="tsd-signature tsd-kind-icon">packedCFunc<wbr>Table<wbr>Free<wbr>Id<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Array</span><span class="tsd-signature-symbol">&lt;</span><span class="tsd-signature-type">number</span><span class="tsd-signature-symbol">&gt;</span><span class="tsd-signature-symbol"> = []</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/environment.ts#L84">environment.ts:84</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/environment.ts#L84">environment.ts:84</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -250,7 +250,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/environment.ts#L105">environment.ts:105</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/environment.ts#L105">environment.ts:105</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/ffilibrary.html b/docs/reference/api/typedoc/classes/ffilibrary.html
index 66b2aed412..105550d470 100644
--- a/docs/reference/api/typedoc/classes/ffilibrary.html
+++ b/docs/reference/api/typedoc/classes/ffilibrary.html
@@ -131,7 +131,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L50">runtime.ts:50</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L50">runtime.ts:50</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -156,7 +156,7 @@
 					<div class="tsd-signature tsd-kind-icon">exports<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Record</span><span class="tsd-signature-symbol">&lt;</span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">, </span><span class="tsd-signature-type">Function</span><span class="tsd-signature-symbol">&gt;</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L47">runtime.ts:47</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L47">runtime.ts:47</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -166,7 +166,7 @@
 					<div class="tsd-signature tsd-kind-icon">memory<span class="tsd-signature-symbol">:</span> <a href="memory.html" class="tsd-signature-type">Memory</a></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L46">runtime.ts:46</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L46">runtime.ts:46</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -176,7 +176,7 @@
 					<div class="tsd-signature tsd-kind-icon">wasm32<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">boolean</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L45">runtime.ts:45</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L45">runtime.ts:45</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -186,7 +186,7 @@
 					<div class="tsd-signature tsd-kind-icon">webGPUContext<span class="tsd-signature-symbol">:</span> <a href="webgpucontext.html" class="tsd-signature-type">WebGPUContext</a></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L48">runtime.ts:48</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L48">runtime.ts:48</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -203,7 +203,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L77">runtime.ts:77</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L77">runtime.ts:77</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -226,7 +226,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L67">runtime.ts:67</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L67">runtime.ts:67</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -243,7 +243,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L85">runtime.ts:85</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L85">runtime.ts:85</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-returns-title">Returns <a href="cachedcallstack.html" class="tsd-signature-type">CachedCallStack</a></h4>
@@ -260,7 +260,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L96">runtime.ts:96</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L96">runtime.ts:96</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -283,7 +283,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L73">runtime.ts:73</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L73">runtime.ts:73</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
diff --git a/docs/reference/api/typedoc/classes/instance.html b/docs/reference/api/typedoc/classes/instance.html
index 7b7bdd7109..92b093f30e 100644
--- a/docs/reference/api/typedoc/classes/instance.html
+++ b/docs/reference/api/typedoc/classes/instance.html
@@ -161,7 +161,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L844">runtime.ts:844</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L844">runtime.ts:844</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -224,7 +224,7 @@
 					<div class="tsd-signature tsd-kind-icon">exports<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Record</span><span class="tsd-signature-symbol">&lt;</span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">, </span><span class="tsd-signature-type">Function</span><span class="tsd-signature-symbol">&gt;</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L834">runtime.ts:834</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L834">runtime.ts:834</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -234,7 +234,7 @@
 					<div class="tsd-signature tsd-kind-icon">memory<span class="tsd-signature-symbol">:</span> <a href="memory.html" class="tsd-signature-type">Memory</a></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L833">runtime.ts:833</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L833">runtime.ts:833</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -251,7 +251,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L973">runtime.ts:973</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L973">runtime.ts:973</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -296,7 +296,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L932">runtime.ts:932</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L932">runtime.ts:932</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -318,7 +318,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L901">runtime.ts:901</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L901">runtime.ts:901</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -381,7 +381,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L1215">runtime.ts:1215</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L1215">runtime.ts:1215</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -412,7 +412,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L1000">runtime.ts:1000</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L1000">runtime.ts:1000</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -453,7 +453,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L1207">runtime.ts:1207</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L1207">runtime.ts:1207</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -491,7 +491,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L922">runtime.ts:922</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L922">runtime.ts:922</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -508,7 +508,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L1235">runtime.ts:1235</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L1235">runtime.ts:1235</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -552,7 +552,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L943">runtime.ts:943</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L943">runtime.ts:943</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -577,7 +577,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L1088">runtime.ts:1088</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L1088">runtime.ts:1088</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -609,7 +609,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L1363">runtime.ts:1363</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L1363">runtime.ts:1363</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -640,7 +640,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L1123">runtime.ts:1123</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L1123">runtime.ts:1123</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -672,7 +672,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L1016">runtime.ts:1016</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L1016">runtime.ts:1016</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -695,7 +695,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L1281">runtime.ts:1281</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L1281">runtime.ts:1281</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -729,7 +729,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L986">runtime.ts:986</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L986">runtime.ts:986</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -769,7 +769,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L1341">runtime.ts:1341</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L1341">runtime.ts:1341</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -817,7 +817,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L1055">runtime.ts:1055</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L1055">runtime.ts:1055</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -857,7 +857,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L1320">runtime.ts:1320</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L1320">runtime.ts:1320</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -900,7 +900,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L1197">runtime.ts:1197</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L1197">runtime.ts:1197</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -938,7 +938,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L1491">runtime.ts:1491</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L1491">runtime.ts:1491</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -990,7 +990,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L1009">runtime.ts:1009</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L1009">runtime.ts:1009</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -1014,7 +1014,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L1151">runtime.ts:1151</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L1151">runtime.ts:1151</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -1046,7 +1046,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L1134">runtime.ts:1134</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L1134">runtime.ts:1134</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -1078,7 +1078,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L1292">runtime.ts:1292</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L1292">runtime.ts:1292</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -1110,7 +1110,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L1223">runtime.ts:1223</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L1223">runtime.ts:1223</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -1141,7 +1141,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L957">runtime.ts:957</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L957">runtime.ts:957</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/memory.html b/docs/reference/api/typedoc/classes/memory.html
index 3f766d63c2..7dcd89cb9b 100644
--- a/docs/reference/api/typedoc/classes/memory.html
+++ b/docs/reference/api/typedoc/classes/memory.html
@@ -130,7 +130,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L40">memory.ts:40</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L40">memory.ts:40</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -152,7 +152,7 @@
 					<div class="tsd-signature tsd-kind-icon">memory<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Memory</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L32">memory.ts:32</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L32">memory.ts:32</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -162,7 +162,7 @@
 					<div class="tsd-signature tsd-kind-icon">wasm32<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">boolean</span><span class="tsd-signature-symbol"> = true</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L33">memory.ts:33</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L33">memory.ts:33</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -179,7 +179,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L154">memory.ts:154</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L154">memory.ts:154</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -210,7 +210,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L90">memory.ts:90</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L90">memory.ts:90</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -233,7 +233,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L97">memory.ts:97</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L97">memory.ts:97</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -256,7 +256,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L74">memory.ts:74</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L74">memory.ts:74</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -279,7 +279,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L81">memory.ts:81</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L81">memory.ts:81</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -302,7 +302,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L104">memory.ts:104</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L104">memory.ts:104</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -325,7 +325,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L132">memory.ts:132</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L132">memory.ts:132</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -362,7 +362,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L145">memory.ts:145</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L145">memory.ts:145</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -393,7 +393,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L60">memory.ts:60</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L60">memory.ts:60</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -416,7 +416,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L67">memory.ts:67</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L67">memory.ts:67</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -439,7 +439,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L53">memory.ts:53</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L53">memory.ts:53</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -462,7 +462,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L114">memory.ts:114</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L114">memory.ts:114</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -485,7 +485,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L124">memory.ts:124</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L124">memory.ts:124</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">number</span></h4>
@@ -502,7 +502,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/memory.ts#L175">memory.ts:175</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/memory.ts#L175">memory.ts:175</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/module.html b/docs/reference/api/typedoc/classes/module.html
index 8148532618..deab67c9a9 100644
--- a/docs/reference/api/typedoc/classes/module.html
+++ b/docs/reference/api/typedoc/classes/module.html
@@ -119,7 +119,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L614">runtime.ts:614</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L614">runtime.ts:614</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -169,7 +169,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L626">runtime.ts:626</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L626">runtime.ts:626</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -186,7 +186,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L653">runtime.ts:653</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L653">runtime.ts:653</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -218,7 +218,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L641">runtime.ts:641</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L641">runtime.ts:641</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -250,7 +250,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L687">runtime.ts:687</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L687">runtime.ts:687</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/ndarray.html b/docs/reference/api/typedoc/classes/ndarray.html
index 8828ec86ff..22d9a99545 100644
--- a/docs/reference/api/typedoc/classes/ndarray.html
+++ b/docs/reference/api/typedoc/classes/ndarray.html
@@ -130,7 +130,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L401">runtime.ts:401</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L401">runtime.ts:401</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -158,7 +158,7 @@
 					<div class="tsd-signature tsd-kind-icon">device<span class="tsd-signature-symbol">:</span> <a href="dldevice.html" class="tsd-signature-type">DLDevice</a></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L394">runtime.ts:394</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L394">runtime.ts:394</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -173,7 +173,7 @@
 					<div class="tsd-signature tsd-kind-icon">dtype<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L390">runtime.ts:390</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L390">runtime.ts:390</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -188,7 +188,7 @@
 					<div class="tsd-signature tsd-kind-icon">ndim<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L388">runtime.ts:388</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L388">runtime.ts:388</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -203,7 +203,7 @@
 					<div class="tsd-signature tsd-kind-icon">shape<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Array</span><span class="tsd-signature-symbol">&lt;</span><span class="tsd-signature-type">number</span><span class="tsd-signature-symbol">&gt;</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L392">runtime.ts:392</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L392">runtime.ts:392</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -225,7 +225,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L480">runtime.ts:480</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L480">runtime.ts:480</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -258,7 +258,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L524">runtime.ts:524</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L524">runtime.ts:524</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -290,7 +290,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L465">runtime.ts:465</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L465">runtime.ts:465</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -307,7 +307,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L458">runtime.ts:458</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L458">runtime.ts:458</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -339,7 +339,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L584">runtime.ts:584</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L584">runtime.ts:584</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -363,7 +363,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L553">runtime.ts:553</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L553">runtime.ts:553</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/packedfunccell.html b/docs/reference/api/typedoc/classes/packedfunccell.html
index 7fad1f0ced..830c970fb0 100644
--- a/docs/reference/api/typedoc/classes/packedfunccell.html
+++ b/docs/reference/api/typedoc/classes/packedfunccell.html
@@ -117,7 +117,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L248">runtime.ts:248</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L248">runtime.ts:248</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -146,7 +146,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L255">runtime.ts:255</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L255">runtime.ts:255</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -163,7 +163,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L264">runtime.ts:264</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L264">runtime.ts:264</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
diff --git a/docs/reference/api/typedoc/classes/rpcserver.html b/docs/reference/api/typedoc/classes/rpcserver.html
index 21c0fdc851..483adb7748 100644
--- a/docs/reference/api/typedoc/classes/rpcserver.html
+++ b/docs/reference/api/typedoc/classes/rpcserver.html
@@ -115,7 +115,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/rpc_server.ts#L95">rpc_server.ts:95</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/rpc_server.ts#L95">rpc_server.ts:95</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -176,7 +176,7 @@
 					<div class="tsd-signature tsd-kind-icon">get<wbr>Imports<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> =&gt; </span><span class="tsd-signature-type">Record</span><span class="tsd-signature-symbol">&lt;</span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">, </span><span class="tsd-signature-type">unknown</span><span class="tsd-signat [...]
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/rpc_server.ts#L84">rpc_server.ts:84</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/rpc_server.ts#L84">rpc_server.ts:84</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-type-declaration">
@@ -201,7 +201,7 @@
 					<div class="tsd-signature tsd-kind-icon">key<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/rpc_server.ts#L80">rpc_server.ts:80</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/rpc_server.ts#L80">rpc_server.ts:80</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -211,7 +211,7 @@
 					<div class="tsd-signature tsd-kind-icon">logger<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>msg<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> =&gt; </span><span class="tsd-signature-type">void</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/rpc_server.ts#L83">rpc_server.ts:83</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/rpc_server.ts#L83">rpc_server.ts:83</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-type-declaration">
@@ -242,7 +242,7 @@
 					<div class="tsd-signature tsd-kind-icon">socket<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">WebSocket</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/rpc_server.ts#L81">rpc_server.ts:81</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/rpc_server.ts#L81">rpc_server.ts:81</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -252,7 +252,7 @@
 					<div class="tsd-signature tsd-kind-icon">state<span class="tsd-signature-symbol">:</span> <a href="../enums/rpcserverstate.html" class="tsd-signature-type">RPCServerState</a><span class="tsd-signature-symbol"> = RPCServerState.InitHeader</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/rpc_server.ts#L82">rpc_server.ts:82</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/rpc_server.ts#L82">rpc_server.ts:82</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -262,7 +262,7 @@
 					<div class="tsd-signature tsd-kind-icon">url<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/rpc_server.ts#L79">rpc_server.ts:79</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/rpc_server.ts#L79">rpc_server.ts:79</a></li>
 						</ul>
 					</aside>
 				</section>
diff --git a/docs/reference/api/typedoc/classes/runtimecontext.html b/docs/reference/api/typedoc/classes/runtimecontext.html
index cf97a7f77f..073b8f7741 100644
--- a/docs/reference/api/typedoc/classes/runtimecontext.html
+++ b/docs/reference/api/typedoc/classes/runtimecontext.html
@@ -132,7 +132,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L148">runtime.ts:148</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L148">runtime.ts:148</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -172,7 +172,7 @@
 					<div class="tsd-signature tsd-kind-icon">array<wbr>Get<wbr>Item<span class="tsd-signature-symbol">:</span> <a href="../index.html#packedfunc" class="tsd-signature-type">PackedFunc</a></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L143">runtime.ts:143</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L143">runtime.ts:143</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -182,7 +182,7 @@
 					<div class="tsd-signature tsd-kind-icon">array<wbr>Get<wbr>Size<span class="tsd-signature-symbol">:</span> <a href="../index.html#packedfunc" class="tsd-signature-type">PackedFunc</a></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L144">runtime.ts:144</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L144">runtime.ts:144</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -192,7 +192,7 @@
 					<div class="tsd-signature tsd-kind-icon">array<wbr>Make<span class="tsd-signature-symbol">:</span> <a href="../index.html#packedfunc" class="tsd-signature-type">PackedFunc</a></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L145">runtime.ts:145</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L145">runtime.ts:145</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -202,7 +202,7 @@
 					<div class="tsd-signature tsd-kind-icon">get<wbr>Sys<wbr>Lib<span class="tsd-signature-symbol">:</span> <a href="../index.html#packedfunc" class="tsd-signature-type">PackedFunc</a></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L146">runtime.ts:146</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L146">runtime.ts:146</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -219,7 +219,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L189">runtime.ts:189</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L189">runtime.ts:189</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -263,7 +263,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L163">runtime.ts:163</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L163">runtime.ts:163</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -280,7 +280,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L208">runtime.ts:208</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L208">runtime.ts:208</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-type-parameters-title">Type parameters</h4>
@@ -309,7 +309,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L157">runtime.ts:157</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L157">runtime.ts:157</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -326,7 +326,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L167">runtime.ts:167</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L167">runtime.ts:167</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -343,7 +343,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L198">runtime.ts:198</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L198">runtime.ts:198</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-type-parameters-title">Type parameters</h4>
diff --git a/docs/reference/api/typedoc/classes/scalar.html b/docs/reference/api/typedoc/classes/scalar.html
index d0358d9f23..d075055961 100644
--- a/docs/reference/api/typedoc/classes/scalar.html
+++ b/docs/reference/api/typedoc/classes/scalar.html
@@ -112,7 +112,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L235">runtime.ts:235</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L235">runtime.ts:235</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -137,7 +137,7 @@
 					<div class="tsd-signature tsd-kind-icon">dtype<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L235">runtime.ts:235</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L235">runtime.ts:235</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -152,7 +152,7 @@
 					<div class="tsd-signature tsd-kind-icon">value<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L233">runtime.ts:233</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L233">runtime.ts:233</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/tvmarray.html b/docs/reference/api/typedoc/classes/tvmarray.html
index f6b99a7ef6..9dcba9a0e7 100644
--- a/docs/reference/api/typedoc/classes/tvmarray.html
+++ b/docs/reference/api/typedoc/classes/tvmarray.html
@@ -133,7 +133,7 @@
 							<aside class="tsd-sources">
 								<p>Overrides <a href="tvmobject.html">TVMObject</a>.<a href="tvmobject.html#constructor">constructor</a></p>
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L784">runtime.ts:784</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L784">runtime.ts:784</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -162,7 +162,7 @@
 					<aside class="tsd-sources">
 						<p>Inherited from <a href="tvmobject.html">TVMObject</a>.<a href="tvmobject.html#ctx">ctx</a></p>
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L703">runtime.ts:703</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L703">runtime.ts:703</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -180,7 +180,7 @@
 							<aside class="tsd-sources">
 								<p>Inherited from <a href="tvmobject.html">TVMObject</a>.<a href="tvmobject.html#dispose">dispose</a></p>
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L715">runtime.ts:715</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L715">runtime.ts:715</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -197,7 +197,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L804">runtime.ts:804</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L804">runtime.ts:804</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -230,7 +230,7 @@
 							<aside class="tsd-sources">
 								<p>Inherited from <a href="tvmobject.html">TVMObject</a>.<a href="tvmobject.html#gethandle">getHandle</a></p>
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L730">runtime.ts:730</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L730">runtime.ts:730</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -262,7 +262,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L796">runtime.ts:796</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L796">runtime.ts:796</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -283,7 +283,7 @@
 							<aside class="tsd-sources">
 								<p>Inherited from <a href="tvmobject.html">TVMObject</a>.<a href="tvmobject.html#typeindex">typeIndex</a></p>
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L738">runtime.ts:738</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L738">runtime.ts:738</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -306,7 +306,7 @@
 							<aside class="tsd-sources">
 								<p>Inherited from <a href="tvmobject.html">TVMObject</a>.<a href="tvmobject.html#typekey">typeKey</a></p>
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L758">runtime.ts:758</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L758">runtime.ts:758</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/tvmobject.html b/docs/reference/api/typedoc/classes/tvmobject.html
index 50d48dd61f..5a10da7014 100644
--- a/docs/reference/api/typedoc/classes/tvmobject.html
+++ b/docs/reference/api/typedoc/classes/tvmobject.html
@@ -130,7 +130,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L703">runtime.ts:703</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L703">runtime.ts:703</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -158,7 +158,7 @@
 					<div class="tsd-signature tsd-kind-icon">ctx<span class="tsd-signature-symbol">:</span> <a href="runtimecontext.html" class="tsd-signature-type">RuntimeContext</a></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L703">runtime.ts:703</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L703">runtime.ts:703</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -175,7 +175,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L715">runtime.ts:715</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L715">runtime.ts:715</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -192,7 +192,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L730">runtime.ts:730</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L730">runtime.ts:730</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -224,7 +224,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L738">runtime.ts:738</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L738">runtime.ts:738</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -246,7 +246,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L758">runtime.ts:758</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L758">runtime.ts:758</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/webgpucontext.html b/docs/reference/api/typedoc/classes/webgpucontext.html
index 2610e3b2ef..936c6c47d1 100644
--- a/docs/reference/api/typedoc/classes/webgpucontext.html
+++ b/docs/reference/api/typedoc/classes/webgpucontext.html
@@ -120,7 +120,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/webgpu.ts#L57">webgpu.ts:57</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/webgpu.ts#L57">webgpu.ts:57</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -145,7 +145,7 @@
 					<div class="tsd-signature tsd-kind-icon">device<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">GPUDevice</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/webgpu.ts#L50">webgpu.ts:50</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/webgpu.ts#L50">webgpu.ts:50</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -155,7 +155,7 @@
 					<div class="tsd-signature tsd-kind-icon">memory<span class="tsd-signature-symbol">:</span> <a href="memory.html" class="tsd-signature-type">Memory</a></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/webgpu.ts#L51">webgpu.ts:51</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/webgpu.ts#L51">webgpu.ts:51</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -172,7 +172,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/webgpu.ts#L84">webgpu.ts:84</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/webgpu.ts#L84">webgpu.ts:84</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -209,7 +209,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/webgpu.ts#L172">webgpu.ts:172</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/webgpu.ts#L172">webgpu.ts:172</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -238,7 +238,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/webgpu.ts#L67">webgpu.ts:67</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/webgpu.ts#L67">webgpu.ts:67</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/enums/argtypecode.html b/docs/reference/api/typedoc/enums/argtypecode.html
index dcc53cd2b2..3256f81ce6 100644
--- a/docs/reference/api/typedoc/enums/argtypecode.html
+++ b/docs/reference/api/typedoc/enums/argtypecode.html
@@ -106,7 +106,7 @@
 					<div class="tsd-signature tsd-kind-icon">DLDevice<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 6</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L242">ctypes.ts:242</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L242">ctypes.ts:242</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -116,7 +116,7 @@
 					<div class="tsd-signature tsd-kind-icon">Float<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 2</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L238">ctypes.ts:238</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L238">ctypes.ts:238</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -126,7 +126,7 @@
 					<div class="tsd-signature tsd-kind-icon">Int<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 0</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L236">ctypes.ts:236</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L236">ctypes.ts:236</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -136,7 +136,7 @@
 					<div class="tsd-signature tsd-kind-icon">Null<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 4</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L240">ctypes.ts:240</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L240">ctypes.ts:240</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -146,7 +146,7 @@
 					<div class="tsd-signature tsd-kind-icon">TVMBytes<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 12</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L248">ctypes.ts:248</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L248">ctypes.ts:248</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -156,7 +156,7 @@
 					<div class="tsd-signature tsd-kind-icon">TVMDLTensor<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 7</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L243">ctypes.ts:243</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L243">ctypes.ts:243</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -166,7 +166,7 @@
 					<div class="tsd-signature tsd-kind-icon">TVMData<wbr>Type<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 5</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L241">ctypes.ts:241</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L241">ctypes.ts:241</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -176,7 +176,7 @@
 					<div class="tsd-signature tsd-kind-icon">TVMModule<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 9</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L245">ctypes.ts:245</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L245">ctypes.ts:245</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -186,7 +186,7 @@
 					<div class="tsd-signature tsd-kind-icon">TVMNDArray<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 13</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L249">ctypes.ts:249</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L249">ctypes.ts:249</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -196,7 +196,7 @@
 					<div class="tsd-signature tsd-kind-icon">TVMObject<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 8</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L244">ctypes.ts:244</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L244">ctypes.ts:244</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -206,7 +206,7 @@
 					<div class="tsd-signature tsd-kind-icon">TVMObjectRValue<wbr>Ref<wbr>Arg<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 14</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L250">ctypes.ts:250</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L250">ctypes.ts:250</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -216,7 +216,7 @@
 					<div class="tsd-signature tsd-kind-icon">TVMOpaque<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 3</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L239">ctypes.ts:239</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L239">ctypes.ts:239</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -226,7 +226,7 @@
 					<div class="tsd-signature tsd-kind-icon">TVMPacked<wbr>Func<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 10</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L246">ctypes.ts:246</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L246">ctypes.ts:246</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -236,7 +236,7 @@
 					<div class="tsd-signature tsd-kind-icon">TVMStr<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 11</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L247">ctypes.ts:247</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L247">ctypes.ts:247</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -246,7 +246,7 @@
 					<div class="tsd-signature tsd-kind-icon">UInt<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 1</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L237">ctypes.ts:237</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L237">ctypes.ts:237</a></li>
 						</ul>
 					</aside>
 				</section>
diff --git a/docs/reference/api/typedoc/enums/aynccallbackcode.html b/docs/reference/api/typedoc/enums/aynccallbackcode.html
index 8322e58ee6..afd9d20ea6 100644
--- a/docs/reference/api/typedoc/enums/aynccallbackcode.html
+++ b/docs/reference/api/typedoc/enums/aynccallbackcode.html
@@ -93,7 +93,7 @@
 					<div class="tsd-signature tsd-kind-icon">k<wbr>Exception<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 5</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L812">runtime.ts:812</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L812">runtime.ts:812</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -103,7 +103,7 @@
 					<div class="tsd-signature tsd-kind-icon">k<wbr>Return<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 4</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L811">runtime.ts:811</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L811">runtime.ts:811</a></li>
 						</ul>
 					</aside>
 				</section>
diff --git a/docs/reference/api/typedoc/enums/dldatatypecode.html b/docs/reference/api/typedoc/enums/dldatatypecode.html
index ee79829461..1781ebbdb1 100644
--- a/docs/reference/api/typedoc/enums/dldatatypecode.html
+++ b/docs/reference/api/typedoc/enums/dldatatypecode.html
@@ -95,7 +95,7 @@
 					<div class="tsd-signature tsd-kind-icon">Float<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 2</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L339">runtime.ts:339</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L339">runtime.ts:339</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -105,7 +105,7 @@
 					<div class="tsd-signature tsd-kind-icon">Int<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 0</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L337">runtime.ts:337</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L337">runtime.ts:337</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -115,7 +115,7 @@
 					<div class="tsd-signature tsd-kind-icon">Opaque<wbr>Handle<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 3</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L340">runtime.ts:340</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L340">runtime.ts:340</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -125,7 +125,7 @@
 					<div class="tsd-signature tsd-kind-icon">UInt<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 1</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L338">runtime.ts:338</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L338">runtime.ts:338</a></li>
 						</ul>
 					</aside>
 				</section>
diff --git a/docs/reference/api/typedoc/enums/rpcserverstate.html b/docs/reference/api/typedoc/enums/rpcserverstate.html
index 413d9af220..506d4cdc31 100644
--- a/docs/reference/api/typedoc/enums/rpcserverstate.html
+++ b/docs/reference/api/typedoc/enums/rpcserverstate.html
@@ -90,7 +90,7 @@
 					<div class="tsd-signature tsd-kind-icon">Init<wbr>Header<span class="tsd-signature-symbol">:</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/rpc_server.ts#L29">rpc_server.ts:29</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/rpc_server.ts#L29">rpc_server.ts:29</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -100,7 +100,7 @@
 					<div class="tsd-signature tsd-kind-icon">Init<wbr>Header<wbr>Key<span class="tsd-signature-symbol">:</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/rpc_server.ts#L30">rpc_server.ts:30</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/rpc_server.ts#L30">rpc_server.ts:30</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -110,7 +110,7 @@
 					<div class="tsd-signature tsd-kind-icon">Init<wbr>Server<span class="tsd-signature-symbol">:</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/rpc_server.ts#L31">rpc_server.ts:31</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/rpc_server.ts#L31">rpc_server.ts:31</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -120,7 +120,7 @@
 					<div class="tsd-signature tsd-kind-icon">Receive<wbr>Packet<wbr>Body<span class="tsd-signature-symbol">:</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/rpc_server.ts#L34">rpc_server.ts:34</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/rpc_server.ts#L34">rpc_server.ts:34</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -130,7 +130,7 @@
 					<div class="tsd-signature tsd-kind-icon">Receive<wbr>Packet<wbr>Header<span class="tsd-signature-symbol">:</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/rpc_server.ts#L33">rpc_server.ts:33</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/rpc_server.ts#L33">rpc_server.ts:33</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -140,7 +140,7 @@
 					<div class="tsd-signature tsd-kind-icon">Wait<wbr>For<wbr>Callback<span class="tsd-signature-symbol">:</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/rpc_server.ts#L32">rpc_server.ts:32</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/rpc_server.ts#L32">rpc_server.ts:32</a></li>
 						</ul>
 					</aside>
 				</section>
diff --git a/docs/reference/api/typedoc/enums/sizeof.html b/docs/reference/api/typedoc/enums/sizeof.html
index 286f8c32bd..0d2e9b6ac8 100644
--- a/docs/reference/api/typedoc/enums/sizeof.html
+++ b/docs/reference/api/typedoc/enums/sizeof.html
@@ -100,7 +100,7 @@
 					<div class="tsd-signature tsd-kind-icon">DLData<wbr>Type<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = I32</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L228">ctypes.ts:228</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L228">ctypes.ts:228</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -110,7 +110,7 @@
 					<div class="tsd-signature tsd-kind-icon">DLDevice<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = I32 + I32</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L229">ctypes.ts:229</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L229">ctypes.ts:229</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -120,7 +120,7 @@
 					<div class="tsd-signature tsd-kind-icon">F32<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 4</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L225">ctypes.ts:225</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L225">ctypes.ts:225</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -130,7 +130,7 @@
 					<div class="tsd-signature tsd-kind-icon">F64<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 8</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L226">ctypes.ts:226</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L226">ctypes.ts:226</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -140,7 +140,7 @@
 					<div class="tsd-signature tsd-kind-icon">I32<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 4</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L223">ctypes.ts:223</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L223">ctypes.ts:223</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -150,7 +150,7 @@
 					<div class="tsd-signature tsd-kind-icon">I64<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 8</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L224">ctypes.ts:224</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L224">ctypes.ts:224</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -160,7 +160,7 @@
 					<div class="tsd-signature tsd-kind-icon">TVMValue<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 8</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L227">ctypes.ts:227</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L227">ctypes.ts:227</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -170,7 +170,7 @@
 					<div class="tsd-signature tsd-kind-icon">U16<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 2</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L222">ctypes.ts:222</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L222">ctypes.ts:222</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -180,7 +180,7 @@
 					<div class="tsd-signature tsd-kind-icon">U8<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol"> = 1</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L221">ctypes.ts:221</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L221">ctypes.ts:221</a></li>
 						</ul>
 					</aside>
 				</section>
diff --git a/docs/reference/api/typedoc/index.html b/docs/reference/api/typedoc/index.html
index 9ac77d1520..a2773a44a6 100644
--- a/docs/reference/api/typedoc/index.html
+++ b/docs/reference/api/typedoc/index.html
@@ -182,7 +182,7 @@
 					<div class="tsd-signature tsd-kind-icon">FObject<wbr>Constructor<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>handle<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, lib<span class="tsd-signature-symbol">: </span><a href="classes/ffilibrary.html" class="tsd-signature-type">FFILibrary</a>, ctx<span class="tsd-signature-symbol">: </span><a href="classes/runtimecontext.html" class="t [...]
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L778">runtime.ts:778</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L778">runtime.ts:778</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -224,7 +224,7 @@
 					<div class="tsd-signature tsd-kind-icon">FTVMArray<wbr>Alloc<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>shape<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, ndim<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span>, dtypeCode<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span>, dtypeBits<span class="tsd [...]
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L113">ctypes.ts:113</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L113">ctypes.ts:113</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -288,7 +288,7 @@
 					<div class="tsd-signature tsd-kind-icon">FTVMArray<wbr>Copy<wbr>From<wbr>Bytes<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>handle<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, data<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, nbytes<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">num [...]
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L129">ctypes.ts:129</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L129">ctypes.ts:129</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -332,7 +332,7 @@
 					<div class="tsd-signature tsd-kind-icon">FTVMArray<wbr>Copy<wbr>From<wbr>To<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>from<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, to<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, stream<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-sig [...]
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L145">ctypes.ts:145</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L145">ctypes.ts:145</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -376,7 +376,7 @@
 					<div class="tsd-signature tsd-kind-icon">FTVMArray<wbr>Copy<wbr>ToBytes<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>handle<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, data<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, nbytes<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</sp [...]
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L137">ctypes.ts:137</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L137">ctypes.ts:137</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -420,7 +420,7 @@
 					<div class="tsd-signature tsd-kind-icon">FTVMArray<wbr>Free<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>handle<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> =&gt; </span><span class="tsd-signature-type">number</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L122">ctypes.ts:122</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L122">ctypes.ts:122</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -456,7 +456,7 @@
 					<div class="tsd-signature tsd-kind-icon">FTVMBackend<wbr>PackedCFunc<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>argValues<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, argCodes<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, nargs<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number< [...]
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L161">ctypes.ts:161</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L161">ctypes.ts:161</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -508,7 +508,7 @@
 					<div class="tsd-signature tsd-kind-icon">FTVMCFunc<wbr>Set<wbr>Return<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>ret<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, value<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, typeCode<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signa [...]
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L78">ctypes.ts:78</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L78">ctypes.ts:78</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -556,7 +556,7 @@
 					<div class="tsd-signature tsd-kind-icon">FTVMCb<wbr>Arg<wbr>ToReturn<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>value<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, code<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> =&gt; </span><span c [...]
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L84">ctypes.ts:84</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L84">ctypes.ts:84</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -595,7 +595,7 @@
 					<div class="tsd-signature tsd-kind-icon">FTVMFunc<wbr>Call<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>func<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, argValues<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, typeCode<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-t [...]
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L68">ctypes.ts:68</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L68">ctypes.ts:68</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -651,7 +651,7 @@
 					<div class="tsd-signature tsd-kind-icon">FTVMFunc<wbr>Free<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>func<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> =&gt; </span><span class="tsd-signature-type">number</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L58">ctypes.ts:58</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L58">ctypes.ts:58</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -687,7 +687,7 @@
 					<div class="tsd-signature tsd-kind-icon">FTVMFunc<wbr>Get<wbr>Global<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>name<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, out<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> =&gt; </span><span cla [...]
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L101">ctypes.ts:101</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L101">ctypes.ts:101</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -726,7 +726,7 @@
 					<div class="tsd-signature tsd-kind-icon">FTVMFunc<wbr>List<wbr>Global<wbr>Names<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>outSize<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, outArray<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> =&g [...]
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L89">ctypes.ts:89</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L89">ctypes.ts:89</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -765,7 +765,7 @@
 					<div class="tsd-signature tsd-kind-icon">FTVMFunc<wbr>Register<wbr>Global<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>name<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, f<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, override<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</spa [...]
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L95">ctypes.ts:95</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L95">ctypes.ts:95</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -808,7 +808,7 @@
 					<div class="tsd-signature tsd-kind-icon">FTVMGet<wbr>Last<wbr>Error<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> =&gt; </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L34">ctypes.ts:34</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L34">ctypes.ts:34</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -838,7 +838,7 @@
 					<div class="tsd-signature tsd-kind-icon">FTVMMod<wbr>Free<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>mod<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> =&gt; </span><span class="tsd-signature-type">number</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L53">ctypes.ts:53</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L53">ctypes.ts:53</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -874,7 +874,7 @@
 					<div class="tsd-signature tsd-kind-icon">FTVMMod<wbr>Get<wbr>Function<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>mod<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, funcName<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, queryImports<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">numbe [...]
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L42">ctypes.ts:42</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L42">ctypes.ts:42</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -922,7 +922,7 @@
 					<div class="tsd-signature tsd-kind-icon">FTVMMod<wbr>Import<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>mod<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, dep<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> =&gt; </span><span class="tsd-si [...]
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L48">ctypes.ts:48</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L48">ctypes.ts:48</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -962,7 +962,7 @@
 					<div class="tsd-signature tsd-kind-icon">FTVMObject<wbr>Free<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>obj<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> =&gt; </span><span class="tsd-signature-type">number</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L169">ctypes.ts:169</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L169">ctypes.ts:169</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -998,7 +998,7 @@
 					<div class="tsd-signature tsd-kind-icon">FTVMObject<wbr>Get<wbr>Type<wbr>Index<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>obj<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, out_tindex<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> =&gt;  [...]
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L174">ctypes.ts:174</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L174">ctypes.ts:174</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -1037,7 +1037,7 @@
 					<div class="tsd-signature tsd-kind-icon">FTVMObject<wbr>Type<wbr>Index2<wbr>Key<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>type_index<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span>, out_type_key<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> =&gt; </span><spa [...]
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L179">ctypes.ts:179</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L179">ctypes.ts:179</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -1076,7 +1076,7 @@
 					<div class="tsd-signature tsd-kind-icon">FTVMObject<wbr>Type<wbr>Key2<wbr>Index<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>type_key<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, out_tindex<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol">  [...]
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L184">ctypes.ts:184</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L184">ctypes.ts:184</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -1115,7 +1115,7 @@
 					<div class="tsd-signature tsd-kind-icon">FTVMSynchronize<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>deviceType<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span>, deviceId<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span>, stream<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signatur [...]
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L151">ctypes.ts:151</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L151">ctypes.ts:151</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -1157,7 +1157,7 @@
 					<div class="tsd-signature tsd-kind-icon">FTVMWasm<wbr>Alloc<wbr>Space<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>size<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> =&gt; </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L189">ctypes.ts:189</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L189">ctypes.ts:189</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -1193,7 +1193,7 @@
 					<div class="tsd-signature tsd-kind-icon">FTVMWasm<wbr>Free<wbr>Space<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>ptr<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> =&gt; </span><span class="tsd-signature-type">void</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L192">ctypes.ts:192</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L192">ctypes.ts:192</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -1229,7 +1229,7 @@
 					<div class="tsd-signature tsd-kind-icon">FTVMWasm<wbr>Func<wbr>Create<wbr>FromCFunc<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>resource<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, out<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> =&g [...]
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L209">ctypes.ts:209</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L209">ctypes.ts:209</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -1269,7 +1269,7 @@
 					<div class="tsd-signature tsd-kind-icon">FTVMWasm<wbr>PackedCFunc<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>args<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, typeCodes<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a>, nargs<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">number</span>, [...]
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L201">ctypes.ts:201</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L201">ctypes.ts:201</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -1321,7 +1321,7 @@
 					<div class="tsd-signature tsd-kind-icon">FTVMWasm<wbr>PackedCFunc<wbr>Finalizer<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>resourceHandle<span class="tsd-signature-symbol">: </span><a href="index.html#pointer" class="tsd-signature-type">Pointer</a><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> =&gt; </span><span class="tsd-signature-type">void</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L215">ctypes.ts:215</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L215">ctypes.ts:215</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -1357,7 +1357,7 @@
 					<div class="tsd-signature tsd-kind-icon">GPUPointer<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/webgpu.ts#L25">webgpu.ts:25</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/webgpu.ts#L25">webgpu.ts:25</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -1372,7 +1372,7 @@
 					<div class="tsd-signature tsd-kind-icon">Packed<wbr>Func<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span><span class="tsd-signature-symbol">...</span>args<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">any</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> =&gt; </span><span class="tsd-signature-type">any</span><span class="tsd-signature-symbol"> &amp; </span><a href="interfaces/disp [...]
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L37">runtime.ts:37</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L37">runtime.ts:37</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -1387,7 +1387,7 @@
 					<div class="tsd-signature tsd-kind-icon">Pointer<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L25">ctypes.ts:25</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L25">ctypes.ts:25</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -1402,7 +1402,7 @@
 					<div class="tsd-signature tsd-kind-icon">Ptr<wbr>Offset<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/ctypes.ts#L28">ctypes.ts:28</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/ctypes.ts#L28">ctypes.ts:28</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -1417,7 +1417,7 @@
 					<div class="tsd-signature tsd-kind-icon">TVMObject<wbr>Base<span class="tsd-signature-symbol">:</span> <a href="classes/tvmobject.html" class="tsd-signature-type">TVMObject</a><span class="tsd-signature-symbol"> | </span><a href="classes/ndarray.html" class="tsd-signature-type">NDArray</a><span class="tsd-signature-symbol"> | </span><a href="classes/module.html" class="tsd-signature-type">Module</a><span class="tsd-signature-symbol"> | </span><a href="index.html#packedfunc" class="t [...]
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L781">runtime.ts:781</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L781">runtime.ts:781</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -1435,7 +1435,7 @@
 					<div class="tsd-signature tsd-kind-icon">RPC_<wbr>MAGIC<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">1045105</span><span class="tsd-signature-symbol"> = 1045105</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/rpc_server.ts#L38">rpc_server.ts:38</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/rpc_server.ts#L38">rpc_server.ts:38</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -1457,7 +1457,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/support.ts#L25">support.ts:25</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/support.ts#L25">support.ts:25</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -1489,7 +1489,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/support.ts#L39">support.ts:39</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/support.ts#L39">support.ts:39</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -1518,7 +1518,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/support.ts#L52">support.ts:52</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/support.ts#L52">support.ts:52</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -1555,7 +1555,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/compact.ts#L38">compact.ts:38</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/compact.ts#L38">compact.ts:38</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -1586,7 +1586,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/webgpu.ts#L30">webgpu.ts:30</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/webgpu.ts#L30">webgpu.ts:30</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -1608,7 +1608,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/environment.ts#L32">environment.ts:32</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/environment.ts#L32">environment.ts:32</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -1639,7 +1639,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/compact.ts#L24">compact.ts:24</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/compact.ts#L24">compact.ts:24</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -1661,7 +1661,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L1749">runtime.ts:1749</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L1749">runtime.ts:1749</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -1726,7 +1726,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/support.ts#L62">support.ts:62</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/support.ts#L62">support.ts:62</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -1748,7 +1748,7 @@
 					<div class="tsd-signature tsd-kind-icon">DLData<wbr>Type<wbr>Code<wbr>ToStr<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">object</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L343">runtime.ts:343</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L343">runtime.ts:343</a></li>
 						</ul>
 					</aside>
 					<section class="tsd-panel tsd-member tsd-kind-variable tsd-parent-kind-object-literal">
@@ -1757,7 +1757,7 @@
 						<div class="tsd-signature tsd-kind-icon">0<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = &quot;int&quot;</span></div>
 						<aside class="tsd-sources">
 							<ul>
-								<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L344">runtime.ts:344</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L344">runtime.ts:344</a></li>
 							</ul>
 						</aside>
 					</section>
@@ -1767,7 +1767,7 @@
 						<div class="tsd-signature tsd-kind-icon">1<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = &quot;uint&quot;</span></div>
 						<aside class="tsd-sources">
 							<ul>
-								<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L345">runtime.ts:345</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L345">runtime.ts:345</a></li>
 							</ul>
 						</aside>
 					</section>
@@ -1777,7 +1777,7 @@
 						<div class="tsd-signature tsd-kind-icon">2<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = &quot;float&quot;</span></div>
 						<aside class="tsd-sources">
 							<ul>
-								<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L346">runtime.ts:346</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L346">runtime.ts:346</a></li>
 							</ul>
 						</aside>
 					</section>
@@ -1787,7 +1787,7 @@
 						<div class="tsd-signature tsd-kind-icon">3<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = &quot;handle&quot;</span></div>
 						<aside class="tsd-sources">
 							<ul>
-								<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L347">runtime.ts:347</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L347">runtime.ts:347</a></li>
 							</ul>
 						</aside>
 					</section>
@@ -1798,7 +1798,7 @@
 					<div class="tsd-signature tsd-kind-icon">Device<wbr>Enum<wbr>ToStr<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">object</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L272">runtime.ts:272</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L272">runtime.ts:272</a></li>
 						</ul>
 					</aside>
 					<section class="tsd-panel tsd-member tsd-kind-variable tsd-parent-kind-object-literal">
@@ -1807,7 +1807,7 @@
 						<div class="tsd-signature tsd-kind-icon">1<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = &quot;cpu&quot;</span></div>
 						<aside class="tsd-sources">
 							<ul>
-								<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L273">runtime.ts:273</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L273">runtime.ts:273</a></li>
 							</ul>
 						</aside>
 					</section>
@@ -1817,7 +1817,7 @@
 						<div class="tsd-signature tsd-kind-icon">15<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = &quot;webgpu&quot;</span></div>
 						<aside class="tsd-sources">
 							<ul>
-								<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L277">runtime.ts:277</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L277">runtime.ts:277</a></li>
 							</ul>
 						</aside>
 					</section>
@@ -1827,7 +1827,7 @@
 						<div class="tsd-signature tsd-kind-icon">2<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = &quot;cuda&quot;</span></div>
 						<aside class="tsd-sources">
 							<ul>
-								<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L274">runtime.ts:274</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L274">runtime.ts:274</a></li>
 							</ul>
 						</aside>
 					</section>
@@ -1837,7 +1837,7 @@
 						<div class="tsd-signature tsd-kind-icon">4<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = &quot;opencl&quot;</span></div>
 						<aside class="tsd-sources">
 							<ul>
-								<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L275">runtime.ts:275</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L275">runtime.ts:275</a></li>
 							</ul>
 						</aside>
 					</section>
@@ -1847,7 +1847,7 @@
 						<div class="tsd-signature tsd-kind-icon">8<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span><span class="tsd-signature-symbol"> = &quot;metal&quot;</span></div>
 						<aside class="tsd-sources">
 							<ul>
-								<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L276">runtime.ts:276</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L276">runtime.ts:276</a></li>
 							</ul>
 						</aside>
 					</section>
@@ -1858,7 +1858,7 @@
 					<div class="tsd-signature tsd-kind-icon">Device<wbr>Str<wbr>ToEnum<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">object</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L280">runtime.ts:280</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L280">runtime.ts:280</a></li>
 						</ul>
 					</aside>
 					<section class="tsd-panel tsd-member tsd-kind-variable tsd-parent-kind-object-literal">
@@ -1867,7 +1867,7 @@
 						<div class="tsd-signature tsd-kind-icon">cl<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 4</span></div>
 						<aside class="tsd-sources">
 							<ul>
-								<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L283">runtime.ts:283</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L283">runtime.ts:283</a></li>
 							</ul>
 						</aside>
 					</section>
@@ -1877,7 +1877,7 @@
 						<div class="tsd-signature tsd-kind-icon">cpu<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 1</span></div>
 						<aside class="tsd-sources">
 							<ul>
-								<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L281">runtime.ts:281</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L281">runtime.ts:281</a></li>
 							</ul>
 						</aside>
 					</section>
@@ -1887,7 +1887,7 @@
 						<div class="tsd-signature tsd-kind-icon">cuda<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 2</span></div>
 						<aside class="tsd-sources">
 							<ul>
-								<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L282">runtime.ts:282</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L282">runtime.ts:282</a></li>
 							</ul>
 						</aside>
 					</section>
@@ -1897,7 +1897,7 @@
 						<div class="tsd-signature tsd-kind-icon">metal<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 8</span></div>
 						<aside class="tsd-sources">
 							<ul>
-								<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L286">runtime.ts:286</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L286">runtime.ts:286</a></li>
 							</ul>
 						</aside>
 					</section>
@@ -1907,7 +1907,7 @@
 						<div class="tsd-signature tsd-kind-icon">opencl<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 4</span></div>
 						<aside class="tsd-sources">
 							<ul>
-								<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L284">runtime.ts:284</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L284">runtime.ts:284</a></li>
 							</ul>
 						</aside>
 					</section>
@@ -1917,7 +1917,7 @@
 						<div class="tsd-signature tsd-kind-icon">vulkan<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 7</span></div>
 						<aside class="tsd-sources">
 							<ul>
-								<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L285">runtime.ts:285</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L285">runtime.ts:285</a></li>
 							</ul>
 						</aside>
 					</section>
@@ -1927,7 +1927,7 @@
 						<div class="tsd-signature tsd-kind-icon">webgpu<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">number</span><span class="tsd-signature-symbol"> = 15</span></div>
 						<aside class="tsd-sources">
 							<ul>
-								<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/runtime.ts#L287">runtime.ts:287</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/runtime.ts#L287">runtime.ts:287</a></li>
 							</ul>
 						</aside>
 					</section>
diff --git a/docs/reference/api/typedoc/interfaces/disposable.html b/docs/reference/api/typedoc/interfaces/disposable.html
index 057d833213..7a73fdb992 100644
--- a/docs/reference/api/typedoc/interfaces/disposable.html
+++ b/docs/reference/api/typedoc/interfaces/disposable.html
@@ -115,7 +115,7 @@
 					<div class="tsd-signature tsd-kind-icon">dispose<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> =&gt; </span><span class="tsd-signature-type">void</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/types.ts#L52">types.ts:52</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/types.ts#L52">types.ts:52</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/interfaces/functioninfo.html b/docs/reference/api/typedoc/interfaces/functioninfo.html
index c20545478d..d4ea9588f9 100644
--- a/docs/reference/api/typedoc/interfaces/functioninfo.html
+++ b/docs/reference/api/typedoc/interfaces/functioninfo.html
@@ -95,7 +95,7 @@
 					<div class="tsd-signature tsd-kind-icon">arg_<wbr>types<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Array</span><span class="tsd-signature-symbol">&lt;</span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">&gt;</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/webgpu.ts#L41">webgpu.ts:41</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/webgpu.ts#L41">webgpu.ts:41</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -105,7 +105,7 @@
 					<div class="tsd-signature tsd-kind-icon">launch_<wbr>param_<wbr>tags<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Array</span><span class="tsd-signature-symbol">&lt;</span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">&gt;</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/webgpu.ts#L42">webgpu.ts:42</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/webgpu.ts#L42">webgpu.ts:42</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -115,7 +115,7 @@
 					<div class="tsd-signature tsd-kind-icon">name<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">string</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/webgpu.ts#L40">webgpu.ts:40</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/webgpu.ts#L40">webgpu.ts:40</a></li>
 						</ul>
 					</aside>
 				</section>
diff --git a/docs/reference/api/typedoc/interfaces/libraryprovider.html b/docs/reference/api/typedoc/interfaces/libraryprovider.html
index d5f0643050..459b4ff3c3 100644
--- a/docs/reference/api/typedoc/interfaces/libraryprovider.html
+++ b/docs/reference/api/typedoc/interfaces/libraryprovider.html
@@ -112,7 +112,7 @@
 					<div class="tsd-signature tsd-kind-icon">imports<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-type">Record</span><span class="tsd-signature-symbol">&lt;</span><span class="tsd-signature-type">string</span><span class="tsd-signature-symbol">, </span><span class="tsd-signature-type">any</span><span class="tsd-signature-symbol">&gt;</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/types.ts#L34">types.ts:34</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/types.ts#L34">types.ts:34</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -127,7 +127,7 @@
 					<div class="tsd-signature tsd-kind-icon">start<span class="tsd-signature-symbol">:</span> <span class="tsd-signature-symbol">(</span>inst<span class="tsd-signature-symbol">: </span><span class="tsd-signature-type">Instance</span><span class="tsd-signature-symbol">)</span><span class="tsd-signature-symbol"> =&gt; </span><span class="tsd-signature-type">void</span></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/1d1dbebc7/web/src/types.ts#L39">types.ts:39</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/8e2382eea/web/src/types.ts#L39">types.ts:39</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
diff --git a/docs/searchindex.js b/docs/searchindex.js
index cd247f1b0e..0cfe4f704c 100644
--- a/docs/searchindex.js
+++ b/docs/searchindex.js
@@ -1 +1 @@
-Search.setIndex({docnames:["arch/benchmark","arch/convert_layout","arch/debugger","arch/device_target_interactions","arch/frontend/tensorflow","arch/hybrid_script","arch/index","arch/inferbound","arch/introduction_to_module_serialization","arch/microtvm_design","arch/microtvm_project_api","arch/model_library_format","arch/pass_infra","arch/relay_intro","arch/relay_op_strategy","arch/runtime","arch/runtimes/vulkan","arch/security","arch/virtual_machine","contribute/ci","contribute/code_gu [...]
\ No newline at end of file
+Search.setIndex({docnames:["arch/benchmark","arch/convert_layout","arch/debugger","arch/device_target_interactions","arch/frontend/tensorflow","arch/hybrid_script","arch/index","arch/inferbound","arch/introduction_to_module_serialization","arch/microtvm_design","arch/microtvm_project_api","arch/model_library_format","arch/pass_infra","arch/relay_intro","arch/relay_op_strategy","arch/runtime","arch/runtimes/vulkan","arch/security","arch/virtual_machine","contribute/ci","contribute/code_gu [...]
\ No newline at end of file
diff --git a/docs/topic/vta/tutorials/autotvm/sg_execution_times.html b/docs/topic/vta/tutorials/autotvm/sg_execution_times.html
index 536e6fb9d7..8ef119ed5e 100644
--- a/docs/topic/vta/tutorials/autotvm/sg_execution_times.html
+++ b/docs/topic/vta/tutorials/autotvm/sg_execution_times.html
@@ -345,7 +345,7 @@
             
   <div class="section" id="computation-times">
 <span id="sphx-glr-topic-vta-tutorials-autotvm-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:31.991</strong> total execution time for <strong>topic_vta_tutorials_autotvm</strong> files:</p>
+<p><strong>00:33.267</strong> total execution time for <strong>topic_vta_tutorials_autotvm</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 82%" />
@@ -354,7 +354,7 @@
 </colgroup>
 <tbody>
 <tr class="row-odd"><td><p><a class="reference internal" href="tune_relay_vta.html#sphx-glr-topic-vta-tutorials-autotvm-tune-relay-vta-py"><span class="std std-ref">Auto-tuning a convolutional network on VTA</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_vta.py</span></code>)</p></td>
-<td><p>00:31.984</p></td>
+<td><p>00:33.260</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="tune_alu_vta.html#sphx-glr-topic-vta-tutorials-autotvm-tune-alu-vta-py"><span class="std std-ref">Auto-tuning a ALU fused op on VTA</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_alu_vta.py</span></code>)</p></td>
diff --git a/docs/topic/vta/tutorials/frontend/deploy_classification.html b/docs/topic/vta/tutorials/frontend/deploy_classification.html
index db865ff013..aef7263589 100644
--- a/docs/topic/vta/tutorials/frontend/deploy_classification.html
+++ b/docs/topic/vta/tutorials/frontend/deploy_classification.html
@@ -588,7 +588,7 @@ and dense layer which will both be executed in fp32 on the CPU.</p></li>
   DeprecationWarning,
 /workspace/vta/tutorials/frontend/deploy_classification.py:213: DeprecationWarning: legacy graph executor behavior of producing json / lib / params will be removed in the next release. Please see documents of tvm.contrib.graph_executor.GraphModule for the  new recommended usage.
   relay_prog, target=tvm.target.Target(target, host=env.target_host), params=params
-resnet18_v1 inference graph built in 34.58s!
+resnet18_v1 inference graph built in 35.76s!
 </pre></div>
 </div>
 </div>
diff --git a/docs/topic/vta/tutorials/frontend/deploy_detection.html b/docs/topic/vta/tutorials/frontend/deploy_detection.html
index 84343e16c3..686c90d9bc 100644
--- a/docs/topic/vta/tutorials/frontend/deploy_detection.html
+++ b/docs/topic/vta/tutorials/frontend/deploy_detection.html
@@ -606,7 +606,7 @@ and dense layer which will both be executed in fp32 on the CPU.</p></li>
 </div>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/workspace/python/tvm/relay/build_module.py:348: DeprecationWarning: Please use input parameter mod (tvm.IRModule) instead of deprecated parameter mod (tvm.relay.function.Function)
   DeprecationWarning,
-yolov3-tiny inference graph built in 23.38s!
+yolov3-tiny inference graph built in 24.06s!
 </pre></div>
 </div>
 </div>
diff --git a/docs/topic/vta/tutorials/frontend/sg_execution_times.html b/docs/topic/vta/tutorials/frontend/sg_execution_times.html
index 49f44d9ecd..310fdf1866 100644
--- a/docs/topic/vta/tutorials/frontend/sg_execution_times.html
+++ b/docs/topic/vta/tutorials/frontend/sg_execution_times.html
@@ -345,7 +345,7 @@
             
   <div class="section" id="computation-times">
 <span id="sphx-glr-topic-vta-tutorials-frontend-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>01:41.715</strong> total execution time for <strong>topic_vta_tutorials_frontend</strong> files:</p>
+<p><strong>01:44.128</strong> total execution time for <strong>topic_vta_tutorials_frontend</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 84%" />
@@ -354,11 +354,11 @@
 </colgroup>
 <tbody>
 <tr class="row-odd"><td><p><a class="reference internal" href="deploy_classification.html#sphx-glr-topic-vta-tutorials-frontend-deploy-classification-py"><span class="std std-ref">Deploy Pretrained Vision Model from MxNet on VTA</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_classification.py</span></code>)</p></td>
-<td><p>00:51.486</p></td>
+<td><p>00:52.788</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="deploy_detection.html#sphx-glr-topic-vta-tutorials-frontend-deploy-detection-py"><span class="std std-ref">Deploy Pretrained Vision Detection Model from Darknet on VTA</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_detection.py</span></code>)</p></td>
-<td><p>00:50.228</p></td>
+<td><p>00:51.340</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 </tbody>
diff --git a/docs/topic/vta/tutorials/optimize/sg_execution_times.html b/docs/topic/vta/tutorials/optimize/sg_execution_times.html
index e901dc5b07..122a8dcb7d 100644
--- a/docs/topic/vta/tutorials/optimize/sg_execution_times.html
+++ b/docs/topic/vta/tutorials/optimize/sg_execution_times.html
@@ -345,7 +345,7 @@
             
   <div class="section" id="computation-times">
 <span id="sphx-glr-topic-vta-tutorials-optimize-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:03.243</strong> total execution time for <strong>topic_vta_tutorials_optimize</strong> files:</p>
+<p><strong>00:03.267</strong> total execution time for <strong>topic_vta_tutorials_optimize</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 84%" />
@@ -354,11 +354,11 @@
 </colgroup>
 <tbody>
 <tr class="row-odd"><td><p><a class="reference internal" href="convolution_opt.html#sphx-glr-topic-vta-tutorials-optimize-convolution-opt-py"><span class="std std-ref">2D Convolution Optimization</span></a> (<code class="docutils literal notranslate"><span class="pre">convolution_opt.py</span></code>)</p></td>
-<td><p>00:02.756</p></td>
+<td><p>00:02.749</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="matrix_multiply_opt.html#sphx-glr-topic-vta-tutorials-optimize-matrix-multiply-opt-py"><span class="std std-ref">Matrix Multiply Blocking</span></a> (<code class="docutils literal notranslate"><span class="pre">matrix_multiply_opt.py</span></code>)</p></td>
-<td><p>00:00.488</p></td>
+<td><p>00:00.518</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 </tbody>
diff --git a/docs/topic/vta/tutorials/sg_execution_times.html b/docs/topic/vta/tutorials/sg_execution_times.html
index e684e1cde3..0c70a3dc99 100644
--- a/docs/topic/vta/tutorials/sg_execution_times.html
+++ b/docs/topic/vta/tutorials/sg_execution_times.html
@@ -345,7 +345,7 @@
             
   <div class="section" id="computation-times">
 <span id="sphx-glr-topic-vta-tutorials-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>00:00.837</strong> total execution time for <strong>topic_vta_tutorials</strong> files:</p>
+<p><strong>00:00.874</strong> total execution time for <strong>topic_vta_tutorials</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 81%" />
@@ -354,11 +354,11 @@
 </colgroup>
 <tbody>
 <tr class="row-odd"><td><p><a class="reference internal" href="matrix_multiply.html#sphx-glr-topic-vta-tutorials-matrix-multiply-py"><span class="std std-ref">Simple Matrix Multiply</span></a> (<code class="docutils literal notranslate"><span class="pre">matrix_multiply.py</span></code>)</p></td>
-<td><p>00:00.431</p></td>
+<td><p>00:00.450</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="vta_get_started.html#sphx-glr-topic-vta-tutorials-vta-get-started-py"><span class="std std-ref">Get Started with VTA</span></a> (<code class="docutils literal notranslate"><span class="pre">vta_get_started.py</span></code>)</p></td>
-<td><p>00:00.406</p></td>
+<td><p>00:00.424</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 </tbody>
diff --git a/docs/tutorial/auto_scheduler_matmul_x86.html b/docs/tutorial/auto_scheduler_matmul_x86.html
index ed51a55393..95455c868d 100644
--- a/docs/tutorial/auto_scheduler_matmul_x86.html
+++ b/docs/tutorial/auto_scheduler_matmul_x86.html
@@ -574,7 +574,7 @@ class Module:
 <span class="p">)</span>
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 94.776 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 95.152 ms
 </pre></div>
 </div>
 </div>
@@ -646,7 +646,7 @@ automatically optimize a matrix multiplication, without the need to specify a
 search template.  It ends a series of examples that starts from the Tensor
 Expression (TE) language that demonstrates how TVM can optimize computational
 operations.</p>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  22.908 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  29.430 seconds)</p>
 <div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-tutorial-auto-scheduler-matmul-x86-py">
 <div class="sphx-glr-download sphx-glr-download-python docutils container">
 <p><a class="reference download internal" download="" href="../_downloads/eac4389b114db015e95cb3cdf8b86b83/auto_scheduler_matmul_x86.py"><code class="xref download docutils literal notranslate"><span class="pre">Download</span> <span class="pre">Python</span> <span class="pre">source</span> <span class="pre">code:</span> <span class="pre">auto_scheduler_matmul_x86.py</span></code></a></p>
diff --git a/docs/tutorial/autotvm_matmul_x86.html b/docs/tutorial/autotvm_matmul_x86.html
index 15e8937178..953ac836c0 100644
--- a/docs/tutorial/autotvm_matmul_x86.html
+++ b/docs/tutorial/autotvm_matmul_x86.html
@@ -685,16 +685,16 @@ reduce variance, we take 5 measurements and average them.</p>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>waiting for device...
 device available
 Get devices for measurement successfully!
-No: 1   GFLOPS: 4.54/4.54       result: MeasureResult(costs=(0.05908292340000001,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.2322745323181152, timestamp=1680175745.71282)  [(&#39;tile_y&#39;, [-1, 16]), (&#39;tile_x&#39;, [-1, 16])],None,44
-No: 2   GFLOPS: 10.74/10.74     result: MeasureResult(costs=(0.0250013296,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.65120530128479, timestamp=1680175747.6591613) [(&#39;tile_y&#39;, [-1, 256]), (&#39;tile_x&#39;, [-1, 512])],None,98
-No: 3   GFLOPS: 3.91/10.74      result: MeasureResult(costs=(0.0687092798,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.3654131889343262, timestamp=1680175750.292137)        [(&#39;tile_y&#39;, [-1, 32]), (&#39;tile_x&#39;, [-1, 16])],None,45
-No: 4   GFLOPS: 3.56/10.74      result: MeasureResult(costs=(0.0754346876,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.436528205871582, timestamp=1680175753.0419655)        [(&#39;tile_y&#39;, [-1, 256]), (&#39;tile_x&#39;, [-1, 16])],None,48
-No: 5   GFLOPS: 2.42/10.74      result: MeasureResult(costs=(0.1108373208,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.0261006355285645, timestamp=1680175755.2605486)       [(&#39;tile_y&#39;, [-1, 512]), (&#39;tile_x&#39;, [-1, 16])],None,49
-No: 6   GFLOPS: 3.31/10.74      result: MeasureResult(costs=(0.08105388,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.5578992366790771, timestamp=1680175756.8147922) [(&#39;tile_y&#39;, [-1, 32]), (&#39;tile_x&#39;, [-1, 8])],None,35
-No: 7   GFLOPS: 8.17/10.74      result: MeasureResult(costs=(0.0328537718,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.8625633716583252, timestamp=1680175757.6071165)       [(&#39;tile_y&#39;, [-1, 1]), (&#39;tile_x&#39;, [-1, 128])],None,70
-No: 8   GFLOPS: 2.64/10.74      result: MeasureResult(costs=(0.1016909336,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.8725430965423584, timestamp=1680175759.4908597)       [(&#39;tile_y&#39;, [-1, 2]), (&#39;tile_x&#39;, [-1, 8])],None,31
-No: 9   GFLOPS: 1.52/10.74      result: MeasureResult(costs=(0.17685110340000001,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.046210527420044, timestamp=1680175762.658594)  [(&#39;tile_y&#39;, [-1, 16]), (&#39;tile_x&#39;, [-1, 1])],None,4
-No: 10  GFLOPS: 2.74/10.74      result: MeasureResult(costs=(0.0979635238,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.789787769317627, timestamp=1680175764.4825878)        [(&#39;tile_y&#39;, [-1, 2]), (&#39;tile_x&#39;, [-1, 16])],None,41
+No: 1   GFLOPS: 11.30/11.30     result: MeasureResult(costs=(0.023755678,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6285767555236816, timestamp=1680187381.321088) [(&#39;tile_y&#39;, [-1, 32]), (&#39;tile_x&#39;, [-1, 512])],None,95
+No: 2   GFLOPS: 0.89/11.30      result: MeasureResult(costs=(0.3003585704,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.058050632476807, timestamp=1680187387.6999588)        [(&#39;tile_y&#39;, [-1, 256]), (&#39;tile_x&#39;, [-1, 2])],None,18
+No: 3   GFLOPS: 4.53/11.30      result: MeasureResult(costs=(0.05927983779999999,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.1981897354125977, timestamp=1680187388.9131694)        [(&#39;tile_y&#39;, [-1, 16]), (&#39;tile_x&#39;, [-1, 16])],None,44
+No: 4   GFLOPS: 3.63/11.30      result: MeasureResult(costs=(0.0739912782,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.4480412006378174, timestamp=1680187391.6941915)       [(&#39;tile_y&#39;, [-1, 128]), (&#39;tile_x&#39;, [-1, 16])],None,47
+No: 5   GFLOPS: 9.27/11.30      result: MeasureResult(costs=(0.028957669800000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.8281950950622559, timestamp=1680187392.6537077)       [(&#39;tile_y&#39;, [-1, 1]), (&#39;tile_x&#39;, [-1, 128])],None,70
+No: 6   GFLOPS: 0.46/11.30      result: MeasureResult(costs=(0.5841573652000001,), error_no=MeasureErrorNo.NO_ERROR, all_cost=9.609525918960571, timestamp=1680187403.6005287)  [(&#39;tile_y&#39;, [-1, 256]), (&#39;tile_x&#39;, [-1, 1])],None,8
+No: 7   GFLOPS: 9.89/11.30      result: MeasureResult(costs=(0.027142059200000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6920754909515381, timestamp=1680187404.3035996)       [(&#39;tile_y&#39;, [-1, 2]), (&#39;tile_x&#39;, [-1, 256])],None,81
+No: 8   GFLOPS: 8.73/11.30      result: MeasureResult(costs=(0.030733858800000003,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.8724057674407959, timestamp=1680187405.0673347)       [(&#39;tile_y&#39;, [-1, 1]), (&#39;tile_x&#39;, [-1, 64])],None,60
+No: 9   GFLOPS: 6.43/11.30      result: MeasureResult(costs=(0.041733574399999994,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.8913915157318115, timestamp=1680187406.0791385)       [(&#39;tile_y&#39;, [-1, 512]), (&#39;tile_x&#39;, [-1, 32])],None,59
+No: 10  GFLOPS: 1.86/11.30      result: MeasureResult(costs=(0.144045639,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.5245437622070312, timestamp=1680187408.6476486)        [(&#39;tile_y&#39;, [-1, 256]), (&#39;tile_x&#39;, [-1, 4])],None,28
 </pre></div>
 </div>
 <p>With tuning completed, we can choose the configuration from the log file that
diff --git a/docs/tutorial/autotvm_relay_x86.html b/docs/tutorial/autotvm_relay_x86.html
index f93b96f162..4266be215a 100644
--- a/docs/tutorial/autotvm_relay_x86.html
+++ b/docs/tutorial/autotvm_relay_x86.html
@@ -563,7 +563,7 @@ standard deviation.</p>
 <span class="nb">print</span><span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">unoptimized</span></a><span class="p">)</span>
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>{&#39;mean&#39;: 515.2749678299995, &#39;median&#39;: 515.1268733999927, &#39;std&#39;: 1.803341233836743}
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>{&#39;mean&#39;: 520.3476530599983, &#39;median&#39;: 519.627120499996, &#39;std&#39;: 2.1229252106345986}
 </pre></div>
 </div>
 </div>
@@ -715,177 +715,179 @@ depending on the specifics of the model and the target platform.</p>
 </pre></div>
 </div>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>[Task  1/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task  1/25]  Current/Best:   11.21/  14.82 GFLOPS | Progress: (4/20) | 15.47 s
-[Task  1/25]  Current/Best:   15.17/  22.57 GFLOPS | Progress: (8/20) | 18.91 s
-[Task  1/25]  Current/Best:   22.53/  22.57 GFLOPS | Progress: (12/20) | 20.83 s
-[Task  1/25]  Current/Best:    9.77/  22.57 GFLOPS | Progress: (16/20) | 24.75 s
-[Task  1/25]  Current/Best:   21.97/  22.98 GFLOPS | Progress: (20/20) | 27.26 s Done.
+[Task  1/25]  Current/Best:   12.38/  22.93 GFLOPS | Progress: (4/20) | 10.26 s
+[Task  1/25]  Current/Best:   17.83/  22.93 GFLOPS | Progress: (8/20) | 16.37 s
+[Task  1/25]  Current/Best:   10.84/  22.93 GFLOPS | Progress: (12/20) | 19.20 s
+[Task  1/25]  Current/Best:   12.84/  22.93 GFLOPS | Progress: (16/20) | 21.61 s
+[Task  1/25]  Current/Best:   14.35/  22.93 GFLOPS | Progress: (20/20) | 24.98 s Done.
 
 [Task  2/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task  2/25]  Current/Best:   12.75/  12.75 GFLOPS | Progress: (4/20) | 4.44 s
-[Task  2/25]  Current/Best:    4.73/  20.70 GFLOPS | Progress: (8/20) | 6.06 s
-[Task  2/25]  Current/Best:   18.18/  20.70 GFLOPS | Progress: (12/20) | 7.79 s
-[Task  2/25]  Current/Best:   19.72/  20.70 GFLOPS | Progress: (16/20) | 9.66 s
-[Task  2/25]  Current/Best:   12.96/  20.70 GFLOPS | Progress: (20/20) | 12.09 s Done.
+[Task  2/25]  Current/Best:   12.00/  16.23 GFLOPS | Progress: (4/20) | 4.82 s
+[Task  2/25]  Current/Best:   14.97/  16.90 GFLOPS | Progress: (8/20) | 6.37 s
+[Task  2/25]  Current/Best:   10.01/  16.90 GFLOPS | Progress: (12/20) | 8.40 s
+[Task  2/25]  Current/Best:   18.77/  18.93 GFLOPS | Progress: (16/20) | 10.39 s
+[Task  2/25]  Current/Best:   14.74/  19.97 GFLOPS | Progress: (20/20) | 11.75 s Done.
 
 [Task  3/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task  3/25]  Current/Best:    9.42/  20.24 GFLOPS | Progress: (4/20) | 5.01 s
-[Task  3/25]  Current/Best:   16.99/  20.24 GFLOPS | Progress: (8/20) | 7.40 s
-[Task  3/25]  Current/Best:   11.58/  20.24 GFLOPS | Progress: (12/20) | 12.34 s
-[Task  3/25]  Current/Best:   11.01/  20.24 GFLOPS | Progress: (16/20) | 15.66 s
-[Task  3/25]  Current/Best:   17.43/  20.24 GFLOPS | Progress: (20/20) | 17.74 s Done.
+[Task  3/25]  Current/Best:    5.27/  13.70 GFLOPS | Progress: (4/20) | 6.22 s
+[Task  3/25]  Current/Best:   12.61/  13.70 GFLOPS | Progress: (8/20) | 8.64 s
+[Task  3/25]  Current/Best:   16.17/  17.86 GFLOPS | Progress: (12/20) | 10.97 s
+[Task  3/25]  Current/Best:   14.72/  17.86 GFLOPS | Progress: (16/20) | 14.44 s
+[Task  3/25]  Current/Best:    6.73/  17.86 GFLOPS | Progress: (20/20) | 17.77 s Done.
 
 [Task  4/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task  4/25]  Current/Best:   17.27/  18.50 GFLOPS | Progress: (4/20) | 4.33 s
-[Task  4/25]  Current/Best:   13.84/  18.50 GFLOPS | Progress: (8/20) | 6.61 s
-[Task  4/25]  Current/Best:   16.81/  18.50 GFLOPS | Progress: (12/20) | 8.57 s
-[Task  4/25]  Current/Best:    2.04/  18.50 GFLOPS | Progress: (16/20) | 10.93 s
-[Task  4/25]  Current/Best:    5.64/  18.50 GFLOPS | Progress: (20/20) | 14.61 s Done.
+[Task  4/25]  Current/Best:   13.45/  15.03 GFLOPS | Progress: (4/20) | 5.08 s
+[Task  4/25]  Current/Best:   15.01/  17.78 GFLOPS | Progress: (8/20) | 9.54 s
+[Task  4/25]  Current/Best:    9.71/  18.15 GFLOPS | Progress: (12/20) | 16.12 s
+[Task  4/25]  Current/Best:   11.43/  18.15 GFLOPS | Progress: (16/20) | 21.18 s
+[Task  4/25]  Current/Best:   11.59/  18.15 GFLOPS | Progress: (20/20) | 24.16 s Done.
 
 [Task  5/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task  5/25]  Current/Best:   10.82/  19.56 GFLOPS | Progress: (4/20) | 6.15 s
-[Task  5/25]  Current/Best:    5.63/  19.99 GFLOPS | Progress: (8/20) | 8.31 s
-[Task  5/25]  Current/Best:    3.02/  19.99 GFLOPS | Progress: (12/20) | 10.46 s
-[Task  5/25]  Current/Best:    3.21/  19.99 GFLOPS | Progress: (16/20) | 13.35 s
-[Task  5/25]  Current/Best:    5.58/  19.99 GFLOPS | Progress: (20/20) | 15.33 s Done.
+[Task  5/25]  Current/Best:   10.40/  18.25 GFLOPS | Progress: (4/20) | 4.99 s
+[Task  5/25]  Current/Best:   13.46/  18.25 GFLOPS | Progress: (8/20) | 7.20 s
+[Task  5/25]  Current/Best:    5.68/  18.25 GFLOPS | Progress: (12/20) | 9.85 s
+[Task  5/25]  Current/Best:   12.96/  18.25 GFLOPS | Progress: (16/20) | 11.88 s
+[Task  5/25]  Current/Best:   12.47/  18.25 GFLOPS | Progress: (20/20) | 14.55 s Done.
 
 [Task  6/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task  6/25]  Current/Best:   12.02/  20.42 GFLOPS | Progress: (4/20) | 7.19 s
-[Task  6/25]  Current/Best:   17.60/  20.42 GFLOPS | Progress: (8/20) | 9.78 s
-[Task  6/25]  Current/Best:    5.71/  20.42 GFLOPS | Progress: (12/20) | 14.28 s
-[Task  6/25]  Current/Best:   12.57/  20.42 GFLOPS | Progress: (16/20) | 17.47 s
-[Task  6/25]  Current/Best:   15.83/  20.42 GFLOPS | Progress: (20/20) | 20.24 s Done.
+[Task  6/25]  Current/Best:   18.33/  18.33 GFLOPS | Progress: (4/20) | 5.82 s
+[Task  6/25]  Current/Best:   18.51/  18.51 GFLOPS | Progress: (8/20) | 8.32 s
+[Task  6/25]  Current/Best:    4.86/  19.95 GFLOPS | Progress: (12/20) | 11.45 s
+[Task  6/25]  Current/Best:   12.35/  21.03 GFLOPS | Progress: (16/20) | 13.79 s
+[Task  6/25]  Current/Best:   17.60/  21.03 GFLOPS | Progress: (20/20) | 16.44 s Done.
 
 [Task  7/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task  7/25]  Current/Best:   20.60/  20.60 GFLOPS | Progress: (4/20) | 4.85 s
-[Task  7/25]  Current/Best:   19.54/  20.60 GFLOPS | Progress: (8/20) | 7.26 s
-[Task  7/25]  Current/Best:   11.79/  20.60 GFLOPS | Progress: (12/20) | 10.34 s
-[Task  7/25]  Current/Best:   21.15/  23.07 GFLOPS | Progress: (16/20) | 12.35 s
-[Task  7/25]  Current/Best:   13.24/  23.07 GFLOPS | Progress: (20/20) | 15.07 s Done.
+[Task  7/25]  Current/Best:    6.01/  18.23 GFLOPS | Progress: (4/20) | 6.68 s
+[Task  7/25]  Current/Best:   11.81/  18.23 GFLOPS | Progress: (8/20) | 9.80 s
+[Task  7/25]  Current/Best:   12.21/  18.23 GFLOPS | Progress: (12/20) | 13.77 s
+[Task  7/25]  Current/Best:    5.76/  18.23 GFLOPS | Progress: (16/20) | 17.57 s
+[Task  7/25]  Current/Best:   13.81/  18.23 GFLOPS | Progress: (20/20) | 20.02 s Done.
 
 [Task  8/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task  8/25]  Current/Best:   13.61/  14.76 GFLOPS | Progress: (4/20) | 4.98 s
-[Task  8/25]  Current/Best:    3.64/  17.56 GFLOPS | Progress: (8/20) | 12.93 s
-[Task  8/25]  Current/Best:   14.74/  17.56 GFLOPS | Progress: (12/20) | 19.04 s
-[Task  8/25]  Current/Best:   10.15/  17.56 GFLOPS | Progress: (16/20) | 21.87 s
-[Task  8/25]  Current/Best:    2.92/  17.56 GFLOPS | Progress: (20/20) | 25.01 s Done.
-
+[Task  8/25]  Current/Best:    7.24/  20.20 GFLOPS | Progress: (4/20) | 6.22 s
+[Task  8/25]  Current/Best:    4.24/  20.20 GFLOPS | Progress: (8/20) | 18.53 s
+[Task  8/25]  Current/Best:   10.51/  20.20 GFLOPS | Progress: (12/20) | 25.65 s
+[Task  8/25]  Current/Best:    6.02/  20.20 GFLOPS | Progress: (16/20) | 34.59 s
+[Task  8/25]  Current/Best:   11.59/  20.20 GFLOPS | Progress: (20/20) | 40.20 s
 [Task  9/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task  9/25]  Current/Best:   13.54/  21.14 GFLOPS | Progress: (4/20) | 4.60 s
-[Task  9/25]  Current/Best:   10.00/  21.32 GFLOPS | Progress: (8/20) | 6.85 s
-[Task  9/25]  Current/Best:   13.94/  21.32 GFLOPS | Progress: (12/20) | 10.82 s
-[Task  9/25]  Current/Best:    8.14/  21.32 GFLOPS | Progress: (16/20) | 14.00 s
-[Task  9/25]  Current/Best:    1.90/  22.85 GFLOPS | Progress: (20/20) | 19.83 s Done.
+[Task  9/25]  Current/Best:   10.45/  13.95 GFLOPS | Progress: (4/20) | 4.79 s Done.
+
+[Task  9/25]  Current/Best:   18.76/  18.99 GFLOPS | Progress: (8/20) | 7.90 s
+[Task  9/25]  Current/Best:   18.28/  19.80 GFLOPS | Progress: (12/20) | 10.97 s
+[Task  9/25]  Current/Best:   15.67/  19.80 GFLOPS | Progress: (16/20) | 12.72 s
+[Task  9/25]  Current/Best:   13.60/  19.80 GFLOPS | Progress: (20/20) | 17.99 s Done.
 
 [Task 10/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 10/25]  Current/Best:   14.16/  16.76 GFLOPS | Progress: (4/20) | 4.96 s
-[Task 10/25]  Current/Best:   14.94/  17.64 GFLOPS | Progress: (8/20) | 7.20 s
-[Task 10/25]  Current/Best:   13.53/  17.64 GFLOPS | Progress: (12/20) | 10.07 s
-[Task 10/25]  Current/Best:   13.16/  17.88 GFLOPS | Progress: (16/20) | 11.96 s
-[Task 10/25]  Current/Best:    9.53/  17.88 GFLOPS | Progress: (20/20) | 14.77 s Done.
+[Task 10/25]  Current/Best:   13.30/  15.66 GFLOPS | Progress: (4/20) | 4.74 s
+[Task 10/25]  Current/Best:    5.46/  17.95 GFLOPS | Progress: (8/20) | 7.44 s
+[Task 10/25]  Current/Best:   16.08/  17.95 GFLOPS | Progress: (12/20) | 9.13 s
+[Task 10/25]  Current/Best:   18.58/  18.58 GFLOPS | Progress: (16/20) | 10.92 s
+[Task 10/25]  Current/Best:    4.14/  18.58 GFLOPS | Progress: (20/20) | 13.62 s Done.
 
 [Task 11/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 11/25]  Current/Best:   13.51/  22.64 GFLOPS | Progress: (4/20) | 5.21 s
-[Task 11/25]  Current/Best:    5.80/  22.64 GFLOPS | Progress: (8/20) | 8.56 s
-[Task 11/25]  Current/Best:   19.48/  22.64 GFLOPS | Progress: (12/20) | 10.43 s
-[Task 11/25]  Current/Best:   11.95/  22.64 GFLOPS | Progress: (16/20) | 13.09 s
-[Task 11/25]  Current/Best:   19.65/  22.64 GFLOPS | Progress: (20/20) | 17.10 s Done.
+[Task 11/25]  Current/Best:   16.91/  16.91 GFLOPS | Progress: (4/20) | 6.61 s
+[Task 11/25]  Current/Best:    9.46/  18.47 GFLOPS | Progress: (8/20) | 9.76 s
+[Task 11/25]  Current/Best:   10.56/  18.47 GFLOPS | Progress: (12/20) | 12.67 s
+[Task 11/25]  Current/Best:    3.09/  20.65 GFLOPS | Progress: (16/20) | 15.62 s
+[Task 11/25]  Current/Best:    3.11/  20.65 GFLOPS | Progress: (20/20) | 18.43 s Done.
 
 [Task 12/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 12/25]  Current/Best:   14.38/  14.38 GFLOPS | Progress: (4/20) | 5.29 s
-[Task 12/25]  Current/Best:   10.14/  14.38 GFLOPS | Progress: (8/20) | 14.12 s
-[Task 12/25]  Current/Best:   16.32/  16.32 GFLOPS | Progress: (12/20) | 17.19 s
-[Task 12/25]  Current/Best:    5.38/  16.32 GFLOPS | Progress: (16/20) | 19.57 s
-[Task 12/25]  Current/Best:   11.76/  16.32 GFLOPS | Progress: (20/20) | 23.56 s Done.
+[Task 12/25]  Current/Best:   11.67/  18.03 GFLOPS | Progress: (4/20) | 7.04 s
+[Task 12/25]  Current/Best:   14.54/  18.03 GFLOPS | Progress: (8/20) | 15.64 s
+[Task 12/25]  Current/Best:    9.91/  18.03 GFLOPS | Progress: (12/20) | 19.94 s
+[Task 12/25]  Current/Best:   17.20/  18.03 GFLOPS | Progress: (16/20) | 22.46 s
+[Task 12/25]  Current/Best:   11.48/  18.20 GFLOPS | Progress: (20/20) | 24.45 s Done.
 
 [Task 13/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 13/25]  Current/Best:   18.28/  18.28 GFLOPS | Progress: (4/20) | 6.14 s
-[Task 13/25]  Current/Best:   12.58/  18.28 GFLOPS | Progress: (8/20) | 10.95 s
-[Task 13/25]  Current/Best:    8.55/  18.28 GFLOPS | Progress: (12/20) | 14.18 s
-[Task 13/25]  Current/Best:   15.65/  18.67 GFLOPS | Progress: (16/20) | 17.19 s
-[Task 13/25]  Current/Best:   11.07/  18.67 GFLOPS | Progress: (20/20) | 20.69 s Done.
+[Task 13/25]  Current/Best:    6.20/  16.90 GFLOPS | Progress: (4/20) | 5.59 s
+[Task 13/25]  Current/Best:   15.80/  18.54 GFLOPS | Progress: (8/20) | 9.44 s
+[Task 13/25]  Current/Best:   11.27/  22.05 GFLOPS | Progress: (12/20) | 12.73 s
+[Task 13/25]  Current/Best:   22.30/  22.30 GFLOPS | Progress: (16/20) | 16.05 s
+[Task 13/25]  Current/Best:   15.56/  22.30 GFLOPS | Progress: (20/20) | 20.37 s Done.
 
 [Task 14/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 14/25]  Current/Best:   12.61/  12.61 GFLOPS | Progress: (4/20) | 6.32 s
-[Task 14/25]  Current/Best:   16.90/  16.90 GFLOPS | Progress: (8/20) | 9.73 s
-[Task 14/25]  Current/Best:    9.15/  17.41 GFLOPS | Progress: (12/20) | 13.08 s
-[Task 14/25]  Current/Best:   12.20/  19.44 GFLOPS | Progress: (16/20) | 15.08 s
-[Task 14/25]  Current/Best:   14.17/  19.44 GFLOPS | Progress: (20/20) | 17.97 s
-[Task 15/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
-
-[Task 15/25]  Current/Best:   22.24/  22.24 GFLOPS | Progress: (4/20) | 5.33 s
-[Task 15/25]  Current/Best:    7.32/  22.24 GFLOPS | Progress: (8/20) | 9.56 s
-[Task 15/25]  Current/Best:    9.93/  22.24 GFLOPS | Progress: (12/20) | 13.06 s
-[Task 15/25]  Current/Best:   13.60/  22.24 GFLOPS | Progress: (16/20) | 15.53 s
-[Task 15/25]  Current/Best:   13.43/  22.24 GFLOPS | Progress: (20/20) | 18.74 s
+[Task 14/25]  Current/Best:    3.89/  15.61 GFLOPS | Progress: (4/20) | 6.53 s
+[Task 14/25]  Current/Best:   10.21/  18.86 GFLOPS | Progress: (8/20) | 9.80 s
+[Task 14/25]  Current/Best:   10.63/  18.86 GFLOPS | Progress: (12/20) | 14.23 s
+[Task 14/25]  Current/Best:   14.35/  18.86 GFLOPS | Progress: (16/20) | 18.09 s
+[Task 14/25]  Current/Best:    8.97/  18.86 GFLOPS | Progress: (20/20) | 20.66 s
+[Task 15/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
+[Task 15/25]  Current/Best:   19.69/  19.69 GFLOPS | Progress: (4/20) | 9.91 s
+[Task 15/25]  Current/Best:    6.42/  22.02 GFLOPS | Progress: (8/20) | 12.88 s
+[Task 15/25]  Current/Best:    8.97/  22.02 GFLOPS | Progress: (12/20) | 17.93 s
+[Task 15/25]  Current/Best:   20.66/  22.74 GFLOPS | Progress: (16/20) | 21.40 s
+[Task 15/25]  Current/Best:   12.49/  22.74 GFLOPS | Progress: (20/20) | 23.08 s
 [Task 16/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 16/25]  Current/Best:   11.52/  16.17 GFLOPS | Progress: (4/20) | 5.32 s
-[Task 16/25]  Current/Best:   11.62/  16.17 GFLOPS | Progress: (8/20) | 9.02 s
-[Task 16/25]  Current/Best:   13.37/  21.83 GFLOPS | Progress: (12/20) | 11.14 s
-[Task 16/25]  Current/Best:   11.47/  21.83 GFLOPS | Progress: (16/20) | 13.57 s
-[Task 16/25]  Current/Best:   19.79/  21.83 GFLOPS | Progress: (20/20) | 15.42 s Done.
+[Task 16/25]  Current/Best:   13.84/  19.81 GFLOPS | Progress: (4/20) | 4.56 s
+[Task 16/25]  Current/Best:   13.69/  19.81 GFLOPS | Progress: (8/20) | 6.31 s
+[Task 16/25]  Current/Best:   16.07/  19.81 GFLOPS | Progress: (12/20) | 7.88 s
+[Task 16/25]  Current/Best:   11.98/  19.81 GFLOPS | Progress: (16/20) | 10.19 s
+[Task 16/25]  Current/Best:   14.89/  20.59 GFLOPS | Progress: (20/20) | 11.84 s Done.
 
 [Task 17/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 17/25]  Current/Best:   10.83/  10.83 GFLOPS | Progress: (4/20) | 6.71 s
-[Task 17/25]  Current/Best:   16.14/  16.14 GFLOPS | Progress: (8/20) | 9.48 s
-[Task 17/25]  Current/Best:    6.07/  16.14 GFLOPS | Progress: (12/20) | 12.42 s
-[Task 17/25]  Current/Best:   12.11/  18.04 GFLOPS | Progress: (16/20) | 14.45 s
-[Task 17/25]  Current/Best:   14.37/  18.04 GFLOPS | Progress: (20/20) | 16.95 s Done.
+[Task 17/25]  Current/Best:   18.35/  18.35 GFLOPS | Progress: (4/20) | 5.60 s
+[Task 17/25]  Current/Best:   12.43/  18.35 GFLOPS | Progress: (8/20) | 8.35 s
+[Task 17/25]  Current/Best:   22.04/  22.04 GFLOPS | Progress: (12/20) | 11.06 s
+[Task 17/25]  Current/Best:   11.32/  22.04 GFLOPS | Progress: (16/20) | 13.83 s
+[Task 17/25]  Current/Best:   23.31/  23.31 GFLOPS | Progress: (20/20) | 16.50 s Done.
 
 [Task 18/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 18/25]  Current/Best:    5.29/  17.50 GFLOPS | Progress: (4/20) | 5.10 s
-[Task 18/25]  Current/Best:   10.74/  17.50 GFLOPS | Progress: (8/20) | 9.73 s
-[Task 18/25]  Current/Best:   18.62/  18.62 GFLOPS | Progress: (12/20) | 12.40 s
-[Task 18/25]  Current/Best:   16.72/  18.62 GFLOPS | Progress: (16/20) | 14.29 s
-[Task 18/25]  Current/Best:   13.63/  20.73 GFLOPS | Progress: (20/20) | 16.73 s Done.
+[Task 18/25]  Current/Best:    9.73/  16.58 GFLOPS | Progress: (4/20) | 8.15 s
+[Task 18/25]  Current/Best:    7.08/  17.35 GFLOPS | Progress: (8/20) | 10.23 s
+[Task 18/25]  Current/Best:   12.50/  19.45 GFLOPS | Progress: (12/20) | 12.66 s
+[Task 18/25]  Current/Best:    1.58/  19.45 GFLOPS | Progress: (16/20) | 18.66 s
+[Task 18/25]  Current/Best:    9.25/  19.45 GFLOPS | Progress: (20/20) | 20.95 s Done.
 
 [Task 19/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 19/25]  Current/Best:    9.45/  20.66 GFLOPS | Progress: (4/20) | 5.56 s
-[Task 19/25]  Current/Best:   10.25/  20.66 GFLOPS | Progress: (8/20) | 9.71 s
-[Task 19/25]  Current/Best:    9.55/  20.66 GFLOPS | Progress: (12/20) | 13.11 s
-[Task 19/25]  Current/Best:    7.82/  20.66 GFLOPS | Progress: (16/20) | 16.91 s
-[Task 19/25]  Current/Best:   17.32/  20.66 GFLOPS | Progress: (20/20) | 23.14 s Done.
+[Task 19/25]  Current/Best:   11.29/  20.77 GFLOPS | Progress: (4/20) | 7.52 s
+[Task 19/25]  Current/Best:   21.42/  21.42 GFLOPS | Progress: (8/20) | 10.10 s
+[Task 19/25]  Current/Best:   20.74/  21.97 GFLOPS | Progress: (12/20) | 13.38 s
+[Task 19/25]  Current/Best:   18.78/  21.97 GFLOPS | Progress: (16/20) | 16.86 s
+[Task 19/25]  Current/Best:   18.08/  21.97 GFLOPS | Progress: (20/20) | 19.44 s Done.
 
 [Task 20/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 20/25]  Current/Best:   10.12/  10.61 GFLOPS | Progress: (4/20) | 7.13 s
-[Task 20/25]  Current/Best:   15.49/  15.49 GFLOPS | Progress: (8/20) | 9.30 s
-[Task 20/25]  Current/Best:   10.71/  15.49 GFLOPS | Progress: (12/20) | 13.26 s
-[Task 20/25]  Current/Best:    8.80/  15.49 GFLOPS | Progress: (16/20) | 16.33 s
-[Task 20/25]  Current/Best:    2.24/  19.24 GFLOPS | Progress: (20/20) | 21.05 s
-[Task 21/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
+[Task 20/25]  Current/Best:   11.51/  12.75 GFLOPS | Progress: (4/20) | 5.98 s
+[Task 20/25]  Current/Best:    2.52/  17.02 GFLOPS | Progress: (8/20) | 13.09 s Done.
  Done.
 
-[Task 21/25]  Current/Best:   18.27/  18.27 GFLOPS | Progress: (4/20) | 4.68 s
-[Task 21/25]  Current/Best:    9.49/  18.27 GFLOPS | Progress: (8/20) | 8.89 s
-[Task 21/25]  Current/Best:   20.42/  20.42 GFLOPS | Progress: (12/20) | 11.62 s
-[Task 21/25]  Current/Best:   10.14/  20.42 GFLOPS | Progress: (16/20) | 15.23 s
-[Task 21/25]  Current/Best:   12.06/  20.42 GFLOPS | Progress: (20/20) | 17.91 s
+[Task 20/25]  Current/Best:    8.03/  17.02 GFLOPS | Progress: (12/20) | 16.73 s
+[Task 20/25]  Current/Best:   11.44/  19.26 GFLOPS | Progress: (16/20) | 20.19 s
+[Task 20/25]  Current/Best:    2.38/  19.26 GFLOPS | Progress: (20/20) | 23.59 s
+[Task 21/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
+[Task 21/25]  Current/Best:    8.38/  12.25 GFLOPS | Progress: (4/20) | 9.37 s
+[Task 21/25]  Current/Best:    9.89/  12.25 GFLOPS | Progress: (8/20) | 11.83 s
+[Task 21/25]  Current/Best:    9.57/  17.19 GFLOPS | Progress: (12/20) | 14.05 s
+[Task 21/25]  Current/Best:    5.72/  19.12 GFLOPS | Progress: (16/20) | 17.01 s
+[Task 21/25]  Current/Best:    8.47/  19.12 GFLOPS | Progress: (20/20) | 22.14 s
 [Task 22/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 22/25]  Current/Best:    2.70/  18.94 GFLOPS | Progress: (4/20) | 5.98 s
-[Task 22/25]  Current/Best:   15.03/  20.31 GFLOPS | Progress: (8/20) | 7.58 s
-[Task 22/25]  Current/Best:    6.67/  20.31 GFLOPS | Progress: (12/20) | 9.45 s
-[Task 22/25]  Current/Best:   12.32/  20.31 GFLOPS | Progress: (16/20) | 11.03 s
-[Task 22/25]  Current/Best:    9.96/  20.31 GFLOPS | Progress: (20/20) | 14.49 s Done.
+[Task 22/25]  Current/Best:   12.55/  15.87 GFLOPS | Progress: (4/20) | 5.01 s
+[Task 22/25]  Current/Best:    2.62/  17.84 GFLOPS | Progress: (8/20) | 7.17 s
+[Task 22/25]  Current/Best:   14.81/  17.84 GFLOPS | Progress: (12/20) | 9.36 s Done.
+
+[Task 22/25]  Current/Best:   10.15/  17.84 GFLOPS | Progress: (16/20) | 11.66 s
+[Task 22/25]  Current/Best:   11.49/  17.84 GFLOPS | Progress: (20/20) | 13.67 s Done.
 
 [Task 23/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 23/25]  Current/Best:   12.88/  16.66 GFLOPS | Progress: (4/20) | 5.36 s
-[Task 23/25]  Current/Best:   23.26/  23.26 GFLOPS | Progress: (8/20) | 10.57 s
-[Task 23/25]  Current/Best:    4.86/  23.26 GFLOPS | Progress: (12/20) | 14.84 s
-[Task 23/25]  Current/Best:   21.54/  23.26 GFLOPS | Progress: (16/20) | 17.32 s
-[Task 23/25]  Current/Best:   22.71/  23.26 GFLOPS | Progress: (20/20) | 19.88 s Done.
+[Task 23/25]  Current/Best:   11.97/  21.36 GFLOPS | Progress: (4/20) | 6.88 s
+[Task 23/25]  Current/Best:    9.81/  22.79 GFLOPS | Progress: (8/20) | 9.41 s
+[Task 23/25]  Current/Best:   12.87/  22.79 GFLOPS | Progress: (12/20) | 14.11 s
+[Task 23/25]  Current/Best:   11.84/  22.79 GFLOPS | Progress: (16/20) | 17.68 s
+[Task 23/25]  Current/Best:   21.52/  22.79 GFLOPS | Progress: (20/20) | 20.43 s Done.
 
 [Task 24/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 24/25]  Current/Best:    3.47/   5.21 GFLOPS | Progress: (4/20) | 13.79 s
-[Task 24/25]  Current/Best:    3.84/   5.59 GFLOPS | Progress: (8/20) | 24.47 s
-[Task 24/25]  Current/Best:    4.41/   5.85 GFLOPS | Progress: (12/20) | 28.96 s
-[Task 24/25]  Current/Best:    5.61/   7.44 GFLOPS | Progress: (16/20) | 32.35 s
-[Task 24/25]  Current/Best:   10.06/  10.06 GFLOPS | Progress: (20/20) | 39.38 s
+[Task 24/25]  Current/Best:    3.10/   3.10 GFLOPS | Progress: (4/20) | 13.97 s
+[Task 24/25]  Current/Best:    7.39/   7.81 GFLOPS | Progress: (8/20) | 24.96 s
+[Task 24/25]  Current/Best:    1.40/   7.81 GFLOPS | Progress: (12/20) | 34.01 s
+[Task 24/25]  Current/Best:    4.38/   7.81 GFLOPS | Progress: (16/20) | 37.14 s
+[Task 24/25]  Current/Best:    1.77/   7.81 GFLOPS | Progress: (20/20) | 47.54 s
 [Task 25/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 25/25]  Current/Best:    3.00/   9.08 GFLOPS | Progress: (4/20) | 8.27 s
-[Task 25/25]  Current/Best:    7.48/   9.08 GFLOPS | Progress: (8/20) | 10.85 s
-[Task 25/25]  Current/Best:    1.54/   9.08 GFLOPS | Progress: (12/20) | 13.48 s
-[Task 25/25]  Current/Best:    7.91/   9.08 GFLOPS | Progress: (16/20) | 14.94 s
-[Task 25/25]  Current/Best:    2.95/   9.08 GFLOPS | Progress: (20/20) | 25.64 s
+[Task 25/25]  Current/Best:    3.06/   5.62 GFLOPS | Progress: (4/20) | 15.51 s Done.
+ Done.
+
+[Task 25/25]  Current/Best:    9.71/   9.71 GFLOPS | Progress: (8/20) | 27.60 s
+[Task 25/25]  Current/Best:    7.90/   9.71 GFLOPS | Progress: (12/20) | 31.09 s
+[Task 25/25]  Current/Best:    7.48/   9.71 GFLOPS | Progress: (16/20) | 42.07 s
+[Task 25/25]  Current/Best:    7.25/   9.71 GFLOPS | Progress: (20/20) | 53.04 s
 </pre></div>
 </div>
 <p>The output from this tuning process will look something like this:</p>
@@ -946,7 +948,7 @@ model using optimized operators to speed up our computations.</p>
     <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;class=&#39;</span><span class="si">%s</span><span class="s2">&#39; with probability=</span><span class="si">%f</span><span class="s2">&quot;</span> <span class="o">%</span> <span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#list" title="builtins.list" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">labels</span></a [...]
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>class=&#39;n02123045 tabby, tabby cat&#39; with probability=0.621102
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>class=&#39;n02123045 tabby, tabby cat&#39; with probability=0.621103
 class=&#39;n02123159 tiger cat&#39; with probability=0.356379
 class=&#39;n02124075 Egyptian cat&#39; with probability=0.019712
 class=&#39;n02129604 tiger, Panthera tigris&#39; with probability=0.001215
@@ -984,8 +986,8 @@ improvement in comparing the optimized model to the unoptimized model.</p>
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;unoptimized: </span><span class="si">%s</span><span class="s2">&quot;</span> <span class="o">%</span> <span class="p">(</span><a href="https://docs.python.org/3/library/stdtypes.html#dict" title="builtins.dict" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">unoptimized</span></a><span class="p">))</span>
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>optimized: {&#39;mean&#39;: 431.3350053699992, &#39;median&#39;: 429.80177999999114, &#39;std&#39;: 3.522132704584328}
-unoptimized: {&#39;mean&#39;: 515.2749678299995, &#39;median&#39;: 515.1268733999927, &#39;std&#39;: 1.803341233836743}
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>optimized: {&#39;mean&#39;: 424.97096428000077, &#39;median&#39;: 423.9221155999985, &#39;std&#39;: 3.647045798278026}
+unoptimized: {&#39;mean&#39;: 520.3476530599983, &#39;median&#39;: 519.627120499996, &#39;std&#39;: 2.1229252106345986}
 </pre></div>
 </div>
 </div>
@@ -999,7 +1001,7 @@ models.</p>
 <p>Here we presented a simple example using ResNet-50 v2 locally. However, TVM
 supports many more features including cross-compilation, remote execution and
 profiling/benchmarking.</p>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 12 minutes  4.440 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 13 minutes  17.978 seconds)</p>
 <div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-tutorial-autotvm-relay-x86-py">
 <div class="sphx-glr-download sphx-glr-download-python docutils container">
... 214 lines suppressed ...