[tvm-site] branch asf-site updated: deploying docs (apache/tvm@3aa16f72dd3f1807b11ec61cf372af07d32099c4)

[tq...@apache.org]

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 f4b1c6e5c0 deploying docs (apache/tvm@3aa16f72dd3f1807b11ec61cf372af07d32099c4)
f4b1c6e5c0 is described below

commit f4b1c6e5c0efdeeb812e9c51b5cdceb18a497142
Author: tvm-bot <95...@users.noreply.github.com>
AuthorDate: Tue Nov 15 04:24:42 2022 +0000

    deploying docs (apache/tvm@3aa16f72dd3f1807b11ec61cf372af07d32099c4)
---
 docs/_images/sphx_glr_micro_train_001.png          | Bin 337505 -> 324292 bytes
 docs/_images/sphx_glr_micro_train_thumb.png        | Bin 23675 -> 23851 bytes
 .../how_to/compile_models/from_darknet.rst.txt     |   2 +-
 .../how_to/compile_models/from_keras.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  |   2 +-
 .../compile_models/sg_execution_times.rst.txt      |  22 +-
 .../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       |  18 +-
 .../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                 | 975 +++++++++++++++------
 .../tune_network_cuda.rst.txt                      |   4 +-
 .../tune_network_x86.rst.txt                       |   4 +-
 .../tune_sparse_x86.rst.txt                        | 114 +--
 .../tune_with_autotvm/sg_execution_times.rst.txt   |   6 +-
 .../tune_with_autotvm/tune_conv2d_cuda.rst.txt     | 699 ++++++++-------
 .../work_with_microtvm/micro_autotune.rst.txt      |  16 +-
 .../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     |  10 +-
 .../how_to/work_with_schedules/intrin_math.rst.txt |   2 +-
 .../work_with_schedules/sg_execution_times.rst.txt |  18 +-
 .../how_to/work_with_schedules/tensorize.rst.txt   |   2 +-
 .../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  |   4 +-
 .../topic/vta/tutorials/sg_execution_times.rst.txt |   6 +-
 .../tutorial/auto_scheduler_matmul_x86.rst.txt     |  11 +-
 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  |  24 +-
 .../tutorial/tensor_expr_get_started.rst.txt       |  40 +-
 docs/commit_hash                                   |   2 +-
 docs/how_to/compile_models/from_darknet.html       |   2 +-
 docs/how_to/compile_models/from_keras.html         |   2 +-
 docs/how_to/compile_models/from_mxnet.html         |   2 +-
 docs/how_to/compile_models/from_oneflow.html       |  11 +-
 docs/how_to/compile_models/from_pytorch.html       |   8 +-
 docs/how_to/compile_models/from_tensorflow.html    |   2 +-
 docs/how_to/compile_models/sg_execution_times.html |  26 +-
 .../deploy_models/deploy_model_on_android.html     |   2 +-
 .../deploy_object_detection_pytorch.html           |  31 +-
 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  |  38 +-
 docs/how_to/deploy_models/sg_execution_times.html  |  18 +-
 .../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                    | 975 +++++++++++++++------
 .../tune_with_autoscheduler/tune_network_cuda.html |   4 +-
 .../tune_with_autoscheduler/tune_network_x86.html  |   4 +-
 .../tune_with_autoscheduler/tune_sparse_x86.html   | 114 +--
 .../tune_with_autotvm/sg_execution_times.html      |   6 +-
 .../how_to/tune_with_autotvm/tune_conv2d_cuda.html | 699 ++++++++-------
 docs/how_to/work_with_microtvm/micro_autotune.html |  16 +-
 docs/how_to/work_with_microtvm/micro_pytorch.html  |   8 +-
 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 |  10 +-
 docs/how_to/work_with_schedules/intrin_math.html   |   2 +-
 .../work_with_schedules/sg_execution_times.html    |  18 +-
 docs/how_to/work_with_schedules/tensorize.html     |   2 +-
 docs/reference/api/python/auto_scheduler.html      |   4 +-
 .../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 +-
 .../api/typedoc/classes/graphexecutor.html         |  16 +-
 docs/reference/api/typedoc/classes/instance.html   |  40 +-
 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 +-
 docs/reference/api/typedoc/classes/scalar.html     |   6 +-
 .../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              | 112 +--
 .../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 |   4 +-
 docs/topic/vta/tutorials/sg_execution_times.html   |   6 +-
 docs/tutorial/auto_scheduler_matmul_x86.html       |   7 +-
 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              |  28 +-
 docs/tutorial/tensor_expr_get_started.html         |  40 +-
 127 files changed, 3041 insertions(+), 2135 deletions(-)

diff --git a/docs/_images/sphx_glr_micro_train_001.png b/docs/_images/sphx_glr_micro_train_001.png
index cdece017f1..fd04aec899 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 3ea3b2a601..176d23232e 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 45ec43a01a..fc8da9bfc1 100644
--- a/docs/_sources/how_to/compile_models/from_darknet.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_darknet.rst.txt
@@ -315,7 +315,7 @@ The process is no different from other examples.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  12.906 seconds)
+   **Total running time of the script:** ( 1 minutes  10.959 seconds)
 
 
 .. _sphx_glr_download_how_to_compile_models_from_darknet.py:
diff --git a/docs/_sources/how_to/compile_models/from_keras.rst.txt b/docs/_sources/how_to/compile_models/from_keras.rst.txt
index d7e4cf10da..4a5cc0c401 100644
--- a/docs/_sources/how_to/compile_models/from_keras.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_keras.rst.txt
@@ -228,7 +228,7 @@ Look up prediction top 1 index in 1000 class synset.
  .. code-block:: none
 
     Relay top-1 id: 285, class name: Egyptian cat
-
    1/1 [==============================] - ETA: 0s
    1/1 [==============================] - 1s 938ms/step
+
    1/1 [==============================] - ETA: 0s
    1/1 [==============================] - 1s 988ms/step
     Keras top-1 id: 285, class name: Egyptian cat
 
 
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 3957002c5c..9f49d2b207 100644
--- a/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
@@ -115,7 +115,7 @@ In this section, we download a pretrained imagenet model and classify an image.
 
  .. code-block:: none
 
-    Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip7e4d0e69-aa5a-4589-a37a-99f69c911e3e from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
+    Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip85dc4430-f38f-45c8-95f0-1610ffdf492a 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 b2aaf9d12a..82d8a3d48a 100644
--- a/docs/_sources/how_to/compile_models/from_oneflow.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_oneflow.rst.txt
@@ -116,7 +116,7 @@ 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]
     19%|#9        | 7.99M/41.5M [00:00<00:00, 42.0MB/s]
     29%|##8       | 12.0M/41.5M [00:00<00:00, 35.8MB/s]
     39%|###8      | 16.0M/41.5M [00:00<00:00, 31.1MB/s]
     58%|#####7    | 24.0M/41.5M [00:00<00:00, 35.4MB/s]
     77%|#######7  | 32.0M/41.5M [00:00<00:00, 39.6MB/s]
     92%|#########2| 38.3M/41.5M [00:01<00:00, 33.9MB/s]
    100%|##########| 41.5M/41.5M [00:01<00:00, 35.7MB/s]
+
      0%|          | 0.00/41.5M [00:00<?, ?B/s]
     39%|###8      | 16.0M/41.5M [00:00<00:00, 101MB/s]
     62%|######1   | 25.6M/41.5M [00:00<00:00, 75.4MB/s]
     82%|########2 | 34.1M/41.5M [00:00<00:00, 69.4MB/s]
    100%|##########| 41.5M/41.5M [00:00<00:00, 75.2MB/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 8e5ea169f9..ab085682f2 100644
--- a/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
@@ -98,7 +98,7 @@ 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]
     28%|##7       | 12.4M/44.7M [00:00<00:00, 129MB/s]
     55%|#####5    | 24.7M/44.7M [00:00<00:00, 110MB/s]
     79%|#######9  | 35.3M/44.7M [00:00<00:00, 106MB/s]
    100%|##########| 44.7M/44.7M [00:00<00:00, 107MB/s]
+
      0%|          | 0.00/44.7M [00:00<?, ?B/s]
     34%|###4      | 15.3M/44.7M [00:00<00:00, 160MB/s]
     69%|######8   | 30.6M/44.7M [00:00<00:00, 118MB/s]
     95%|#########5| 42.5M/44.7M [00:00<00:00, 97.3MB/s]
    100%|##########| 44.7M/44.7M [00:00<00:00, 109MB/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 f9fc94ca9c..38119170f4 100644
--- a/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
@@ -416,7 +416,7 @@ Run the corresponding model on tensorflow
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  11.987 seconds)
+   **Total running time of the script:** ( 1 minutes  12.549 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 19634f9336..960430c9fe 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
 =================
-**05:49.415** total execution time for **how_to_compile_models** files:
+**05:51.913** total execution time for **how_to_compile_models** files:
 
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``)       | 01:12.906 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``) | 01:12.549 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``) | 01:11.987 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``)       | 01:10.959 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``)         | 00:46.883 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``)         | 00:47.690 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_oneflow.py` (``from_oneflow.py``)       | 00:33.450 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_oneflow.py` (``from_oneflow.py``)       | 00:33.105 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``)           | 00:30.046 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``)           | 00:31.248 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``)         | 00:26.740 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``)         | 00:27.261 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``)         | 00:24.927 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``)         | 00:25.089 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``)       | 00:22.933 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``)       | 00:22.844 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``)           | 00:17.078 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``)           | 00:18.732 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``)             | 00:02.466 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``)             | 00:02.436 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
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 17e9b8c1fc..8dd01d3f24 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
@@ -434,7 +434,7 @@ Execute on TVM
     Evaluate inference time cost...
     Execution time summary:
      mean (ms)   median (ms)    max (ms)     min (ms)     std (ms)  
-      16.4142      16.3849      16.9731      15.9958       0.3440   
+      16.8910      16.2437      22.1541      16.0129       1.7825   
                
 
 
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 d0efce0934..9a67af78a7 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
@@ -127,7 +127,7 @@ 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]
      5%|4         | 7.99M/170M [00:00<00:02, 61.2MB/s]
     11%|#1        | 18.8M/170M [00:00<00:01, 87.8MB/s]
     19%|#8        | 32.0M/170M [00:00<00:01, 88.1MB/s]
     27%|##7       | 46.3M/170M [00:00<00:01, 102MB/s] 
     33%|###3      | 56.8M/170M [00:00<00:01, 105MB/s]
     42%|####2     | 72.0M/170M [00:00<00:00, 117MB/s]
     49%|####9     | 83.3M/170M [00:00<00:00, 116MB/s]
     56%|#####5    | 94.5M/170M [00:00<00:00, 97.8MB/s]
     61%|######1   | 104M/170M [00:01<00:00, 98.8MB/s] 
     70%|######9   | 118M/170M [00:01<00:00, 111MB/s] 
     76%|#######6  | 129M/170M [00:01<00:00, 94.2MB/s]
     82%|########1 | 139M/170M [00:01<00:00, 94.3MB/s]
     87%|########7 | 148M/170M [00:01<00:00, 93.5MB/s]
     95%|#########5| 161M/170M [00:01<00:00, 105MB/s] 
    100%|##########| 170M/170M [00:01<00:00, 102MB/s]
+
      0%|          | 0.00/170M [00:00<?, ?B/s]
      8%|8         | 13.9M/170M [00:00<00:01, 146MB/s]
     16%|#6        | 27.8M/170M [00:00<00:01, 113MB/s]
     23%|##2       | 39.0M/170M [00:00<00:01, 109MB/s]
     29%|##9       | 49.5M/170M [00:00<00:01, 106MB/s]
     35%|###5      | 59.7M/170M [00:00<00:01, 104MB/s]
     41%|####1     | 69.7M/170M [00:00<00:01, 102MB/s]
     47%|####6     | 79.5M/170M [00:00<00:00, 102MB/s]
     53%|#####2    | 89.2M/170M [00:00<00:00, 96.2MB/s]
     59%|#####9    | 100M/170M [00:01<00:00, 102MB/s]  
     65%|######4   | 110M/170M [00:01<00:00, 102MB/s]
     71%|#######   | 120M/170M [00:01<00:00, 101MB/s]
     76%|#######6  | 130M/170M [00:01<00:00, 101MB/s]
     82%|########1 | 139M/170M [00:01<00:00, 100MB/s]
     88%|########7 | 149M/170M [00:01<00:00, 100MB/s]
     93%|#########3| 158M/170M [00:01<00:00, 82.0MB/s]
    100%|##########| 170M/170M [00:01<00:00, 102MB/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').
@@ -296,7 +296,7 @@ Get boxes with score larger than 0.9
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 3 minutes  20.869 seconds)
+   **Total running time of the script:** ( 3 minutes  21.087 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 26a3172683..7c90146c94 100644
--- a/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt
@@ -236,7 +236,7 @@ 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, 80.2MB/s]
    100%|##########| 13.6M/13.6M [00:00<00:00, 58.5MB/s]
+
      0%|          | 0.00/13.6M [00:00<?, ?B/s]
     69%|######9   | 9.38M/13.6M [00:00<00:00, 98.3MB/s]
    100%|##########| 13.6M/13.6M [00:00<00:00, 124MB/s] 
 
 
 
@@ -418,7 +418,7 @@ 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)  
-      90.5781      90.5102      92.8514      90.1945       0.3491   
+      90.3989      90.3041      95.5078      90.1724       0.5412   
                
 
 
@@ -467,7 +467,7 @@ TODO
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  7.730 seconds)
+   **Total running time of the script:** ( 1 minutes  7.793 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 1f86d8baed..f07111c51a 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
@@ -432,7 +432,7 @@ 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)  
-      120.9755     120.8276     126.3660     119.9609      0.7430   
+      121.0398     120.9786     126.2886     120.1887      0.6482   
                
 
 
@@ -469,7 +469,7 @@ 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  27.444 seconds)
+   **Total running time of the script:** ( 2 minutes  27.687 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 9b831bee42..4fba063930 100644
--- a/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
@@ -253,7 +253,7 @@ We create a Relay VM to build and execute the model.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  40.803 seconds)
+   **Total running time of the script:** ( 1 minutes  16.908 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 3c52ae928b..037247d268 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
@@ -166,7 +166,7 @@ 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]
      2%|1         | 2188/132723 [00:00<00:06, 21489.31KB/s]
      5%|4         | 6078/132723 [00:00<00:04, 31652.43KB/s]
      9%|9         | 12209/132723 [00:00<00:02, 45135.88KB/s]
     15%|#5        | 20190/132723 [00:00<00:01, 58786.19KB/s]
     21%|##1       | 28449/132723 [00:00<00:01, 67353.36KB/s]
     28%|##7       | 36599/132723 [00:00<00:01, 72154.59KB/s]
     34%|###3      | 44881/132723 [00:00<00:01, 75635.00KB/s]
     40%|####      | 53214/132723 [00:00<00:01, 78080.10KB/s]
     46%|####6     | 61563/132723 [00:00<00:00, 79768.85KB/s]
     53%|#####2    | 69908/132723 [00:01<00:00, 80902.06KB/s]
     59%|#####8    | 78165/132723 [00:01<00:00, 81409.82KB/s]
     65%|######5   | 86482/132723 [00:01<00:00, 81943.49KB/s]
     71%|#######1  | 94821/132723 [00:01<00:00, 82378.61KB/s]
     78%|#######7  | 103060/132723 [00:01<00:00, 60350.65KB/s]
     84%|########3 | 111129/132723 [00:01<00:00, 65232.74KB/s]
     90%|########9 
 | 119373/132723 [00:01<00:00, 69625.24KB/s]
     96%|#########6| 127617/132723 [00:01<00:00, 73050.40KB/s]
    100%|##########| 132723/132723 [00:01<00:00, 70374.98KB/s]
+
      0%|          | 0/132723 [00:00<?, ?KB/s]
      4%|4         | 5496/132723 [00:00<00:02, 54954.42KB/s]
     10%|#         | 13327/132723 [00:00<00:01, 68687.35KB/s]
     16%|#5        | 21173/132723 [00:00<00:01, 73144.09KB/s]
     22%|##1       | 29079/132723 [00:00<00:01, 75475.98KB/s]
     28%|##7       | 36842/132723 [00:00<00:01, 76245.75KB/s]
     34%|###3      | 44898/132723 [00:00<00:01, 77709.24KB/s]
     40%|###9      | 52743/132723 [00:00<00:01, 77949.23KB/s]
     46%|####5     | 60631/132723 [00:00<00:00, 78242.27KB/s]
     52%|#####1    | 68457/132723 [00:00<00:00, 78246.20KB/s]
     58%|#####7    | 76349/132723 [00:01<00:00, 78450.47KB/s]
     63%|######3   | 84237/132723 [00:01<00:00, 78579.71KB/s]
     70%|######9   | 92262/132723 [00:01<00:00, 79085.11KB/s]
     75%|#######5  | 100171/132723 [00:01<00:00, 78619.05KB/s]
     81%|########1 | 108090/132723 [00:01<00:00, 78788.20KB/s]
     87%|########7 | 116026/132723 [00:01<00:00, 78952.27KB/s]
     93%|########
 #3| 123945/132723 [00:01<00:00, 79020.19KB/s]
     99%|#########9| 131990/132723 [00:01<00:00, 79447.00KB/s]
    100%|##########| 132723/132723 [00:01<00:00, 77469.79KB/s]
 
 
 
@@ -242,7 +242,7 @@ Display result
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 3 minutes  4.107 seconds)
+   **Total running time of the script:** ( 3 minutes  4.805 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 0ac22eda8b..f1c15c731c 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
@@ -5,24 +5,24 @@
 
 Computation times
 =================
-**13:09.839** total execution time for **how_to_deploy_models** files:
+**12:47.044** total execution time for **how_to_deploy_models** files:
 
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``) | 03:20.869 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``) | 03:21.087 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``)                           | 03:04.107 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``)                           | 03:04.805 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``)           | 02:27.444 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``)           | 02:27.687 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``)                               | 01:40.803 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``)                               | 01:16.908 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``)                         | 01:07.730 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``)                         | 01:07.793 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``)                 | 00:37.004 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``)                 | 00:37.100 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_nano.py` (``deploy_model_on_nano.py``)                       | 00:26.263 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_nano.py` (``deploy_model_on_nano.py``)                       | 00:26.195 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``)                       | 00:25.614 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``)                       | 00:25.462 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
 | :ref:`sphx_glr_how_to_deploy_models_deploy_sparse.py` (``deploy_sparse.py``)                                     | 00:00.007 | 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 107b05adf4..41ff9c4d17 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
@@ -472,7 +472,7 @@ 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.zip52070d7f-f159-486c-82a3-217adcb347fe from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
+    Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zip178162a8-b0e0-4095-8df6-151f912a796c 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 e886baebac..33234e3ac3 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
 =================
-**00:48.735** total execution time for **how_to_extend_tvm** files:
+**00:49.273** 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``) | 00:45.195 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_bring_your_own_datatypes.py` (``bring_your_own_datatypes.py``) | 00:45.717 | 0.0 MB |
 +-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``)           | 00:02.472 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``)           | 00:02.479 | 0.0 MB |
 +-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``)                     | 00:01.060 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``)                     | 00:01.068 | 0.0 MB |
 +-------------------------------------------------------------------------------------------------+-----------+--------+
 | :ref:`sphx_glr_how_to_extend_tvm_low_level_custom_pass.py` (``low_level_custom_pass.py``)       | 00:00.008 | 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 4c0ab619a9..662954305f 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
@@ -216,10 +216,10 @@ profile the execution time of each passes.
  .. code-block:: none
 
     Printing results of timing profile...
-    InferType: 7264us [7264us] (46.43%; 46.43%)
-    FoldScaleAxis: 8382us [7us] (53.57%; 53.57%)
-            FoldConstant: 8375us [1696us] (53.53%; 99.92%)
-                    InferType: 6679us [6679us] (42.69%; 79.75%)
+    InferType: 7318us [7318us] (45.35%; 45.35%)
+    FoldScaleAxis: 8820us [8us] (54.65%; 54.65%)
+            FoldConstant: 8812us [1704us] (54.61%; 99.91%)
+                    InferType: 7108us [7108us] (44.04%; 80.66%)
 
 
 
@@ -258,10 +258,10 @@ Refer to following sections and :py:func:`tvm.instrument.pass_instrument` for th
  .. code-block:: none
 
     Printing results of timing profile...
-    InferType: 6710us [6710us] (44.83%; 44.83%)
-    FoldScaleAxis: 8258us [6us] (55.17%; 55.17%)
-            FoldConstant: 8253us [1716us] (55.13%; 99.93%)
-                    InferType: 6537us [6537us] (43.67%; 79.21%)
+    InferType: 6940us [6940us] (45.33%; 45.33%)
+    FoldScaleAxis: 8370us [6us] (54.67%; 54.67%)
+            FoldConstant: 8364us [1716us] (54.63%; 99.93%)
+                    InferType: 6648us [6648us] (43.42%; 79.48%)
 
 
 
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 5a5418053a..8451828495 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
@@ -340,7 +340,7 @@ latency of convolution.
 
  .. code-block:: none
 
-    Convolution: 51.707935 ms
+    Convolution: 54.220928 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 76141b8079..a8f7e395e7 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
@@ -659,7 +659,7 @@ be able to run on our build server
 
  .. code-block:: none
 
-    conv2d with tensor core: 13.361002 ms
+    conv2d with tensor core: 13.369524 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 ee317c908b..ca7896519f 100644
--- a/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
@@ -143,8 +143,8 @@ Then we write a baseline implementation, the simplest way to write a matrix mult
 
  .. code-block:: none
 
-    Numpy running time: 0.019154
-    Baseline: 3.370971
+    Numpy running time: 0.019707
+    Baseline: 3.518368
 
 
 
@@ -239,7 +239,7 @@ fill 32 * 32 * sizeof(float) which is 4KB in the cache whose total size is 32KB
 
  .. code-block:: none
 
-    Opt1: 0.310276
+    Opt1: 0.326323
 
 
 
@@ -342,7 +342,7 @@ In this tutorial, we chose to vectorize the inner loop row data since it is cach
 
  .. code-block:: none
 
-    Opt2: 0.348115
+    Opt2: 0.350061
 
 
 
@@ -438,7 +438,7 @@ the access pattern for A matrix is more cache friendly.
 
  .. code-block:: none
 
-    Opt3: 0.120739
+    Opt3: 0.124679
 
 
 
@@ -563,7 +563,7 @@ flattening.
 
  .. code-block:: none
 
-    Opt4: 0.110055
+    Opt4: 0.109649
 
 
 
@@ -685,7 +685,7 @@ write to C when all the block results are ready.
 
  .. code-block:: none
 
-    Opt5: 0.111622
+    Opt5: 0.111123
 
 
 
@@ -810,7 +810,7 @@ Furthermore, we can also utilize multi-core processors to do the thread-level pa
 
  .. code-block:: none
 
-    Opt6: 0.147616
+    Opt6: 0.147065
 
 
 
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 81fbb7eb6c..3f02674896 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
 =================
-**00:35.320** total execution time for **how_to_optimize_operators** files:
+**00:35.790** total execution time for **how_to_optimize_operators** files:
 
 +-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``)                       | 00:32.617 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``)                       | 00:33.287 | 0.0 MB |
 +-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``) | 00:01.541 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``) | 00:01.430 | 0.0 MB |
 +-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``)             | 00:01.162 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``)             | 00:01.073 | 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 1f9e4089d5..48aae17ebf 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
 =================
-**09:16.216** total execution time for **how_to_tune_with_autoscheduler** files:
+**09:27.284** 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``) | 05:45.244 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_conv2d_layer_cuda.py` (``tune_conv2d_layer_cuda.py``) | 05:56.080 | 0.0 MB |
 +----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``)             | 01:33.722 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``)             | 01:34.623 | 0.0 MB |
 +----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``)           | 01:04.412 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``)           | 01:04.140 | 0.0 MB |
 +----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``)               | 00:29.278 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``)               | 00:28.953 | 0.0 MB |
 +----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``)             | 00:12.155 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``)             | 00:12.164 | 0.0 MB |
 +----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``)           | 00:11.404 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``)           | 00:11.323 | 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 4cc9dc35e2..cb737a2a5d 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
@@ -241,142 +241,375 @@ cooperative fetching, unrolling and operator fusion.
       buffer_map = {data_1: data, kernel_1: kernel, bias_1: bias, compute_1: compute}
       preflattened_buffer_map = {data_1: data_3: Buffer(data_2, float32, [1, 512, 7, 7], []), kernel_1: kernel_3: Buffer(kernel_2, float32, [512, 512, 3, 3], []), bias_1: bias_3: Buffer(bias_2, float32, [1, 512, 1, 1], []), compute_1: compute_3: Buffer(compute_2, float32, [1, 512, 7, 7], [])} {
       attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 64;
-      allocate(conv2d_nchw: Pointer(local float32), float32, [2]), storage_scope = local;
-      allocate(pad_temp.shared: Pointer(shared float32), float32, [196]), storage_scope = shared;
-      allocate(kernel.shared: Pointer(shared float32), float32, [32]), storage_scope = shared;
-      attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196 {
-        conv2d_nchw_1: Buffer(conv2d_nchw, float32, [1], [], scope="local", align=4)[0] = 0f32
+      allocate(conv2d_nchw: Pointer(local float32), float32, [8]), storage_scope = local;
+      allocate(pad_temp.shared: Pointer(shared float32), float32, [252]), storage_scope = shared;
+      allocate(kernel.shared: Pointer(shared float32), float32, [96]), storage_scope = shared;
+      attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49 {
+        conv2d_nchw_1: Buffer(conv2d_nchw, float32, [16], [], scope="local", align=16)[0] = 0f32
+        conv2d_nchw_1[4] = 0f32
         conv2d_nchw_1[1] = 0f32
+        conv2d_nchw_1[5] = 0f32
+        conv2d_nchw_1[2] = 0f32
+        conv2d_nchw_1[6] = 0f32
+        conv2d_nchw_1[3] = 0f32
+        conv2d_nchw_1[7] = 0f32
         for (rc.outer.outer: int32, 0, 128) {
-          attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
-          pad_temp.shared_1: Buffer(pad_temp.shared, float32, [196], [], scope="shared")[threadIdx.x_1] = @tir.if_then_else(((7 <= floormod(threadIdx.x_1, 49)) && (1 <= floormod(threadIdx.x_1, 7))), data[(((rc.outer.outer*196) + threadIdx.x_1) - 8)], 0f32, dtype=float32)
-          attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
-          if @tir.likely((threadIdx.x_2 < 32), dtype=bool) {
-            kernel.shared_1: Buffer(kernel.shared, float32, [32], [], scope="shared")[threadIdx.x_2] = kernel[((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 4)*4608)) + (rc.outer.outer*36)) + (floormod(threadIdx.x_2, 4)*9))]
+          let cse_var_1: int32 = (rc.outer.outer*196)
+           {
+            attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+            pad_temp.shared_1: Buffer(pad_temp.shared, float32, [252], [], scope="shared")[threadIdx.x_1] = @tir.if_then_else(((7 <= threadIdx.x_1) && (1 <= floormod(threadIdx.x_1, 7))), data[((cse_var_1 + threadIdx.x_1) - 8)], 0f32, dtype=float32)
+            attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+            pad_temp.shared_1[(threadIdx.x_1 + 49)] = @tir.if_then_else((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 7), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 7), 9) < 8)) && (1 <= floormod(threadIdx.x_1, 7))), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 49), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 7), 9)*7)) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
+            attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+            pad_temp.shared_1[(threadIdx.x_1 + 98)] = @tir.if_then_else((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 5), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 5), 9) < 8)) && (1 <= floormod(threadIdx.x_1, 7))), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 98), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 5), 9)*7)) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
+            attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+            pad_temp.shared_1[(threadIdx.x_1 + 147)] = @tir.if_then_else((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 3), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 3), 9) < 8)) && (1 <= floormod(threadIdx.x_1, 7))), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 147), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 3), 9)*7)) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
+            attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+            pad_temp.shared_1[(threadIdx.x_1 + 196)] = @tir.if_then_else((1 <= floormod(threadIdx.x_1, 7)), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 196), 63)*49)) + ((floordiv(threadIdx.x_1, 7) + 1)*7)) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
+            attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+            if @tir.likely((threadIdx.x_1 < 7), dtype=bool) {
+              pad_temp.shared_1[(threadIdx.x_1 + 245)] = 0f32
+            }
+            attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+            kernel.shared_1: Buffer(kernel.shared, float32, [96], [], scope="shared")[threadIdx.x_2] = kernel[((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 12)*4608)) + (rc.outer.outer*36)) + (floormod(threadIdx.x_2, 12)*3))]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+            if @tir.likely((threadIdx.x_2 < 47), dtype=bool) {
+              kernel.shared_1[(threadIdx.x_2 + 49)] = kernel[((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 49), 12)*4608)) + (rc.outer.outer*36)) + (floormod((threadIdx.x_2 + 1), 12)*3))]
+            }
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[0]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[48]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[12]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[60]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[3]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[51]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[15]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[63]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[24]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[72]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[36]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[84]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[27]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[75]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[39]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[87]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[1]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[49]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[13]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[61]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[4]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[52]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[16]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[64]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[25]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[73]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[37]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[85]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[28]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[76]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[40]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[88]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[2]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[50]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[14]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[62]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[5]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[53]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[17]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[65]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[26]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[74]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[38]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[86]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[29]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[77]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[41]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[89]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[6]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[54]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[18]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[66]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[9]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[57]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[21]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[69]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[30]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[78]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[42]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[90]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[33]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[81]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[45]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[93]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[7]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[55]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[19]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[67]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[10]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[58]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[22]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[70]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[31]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[79]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[43]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[91]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[34]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[82]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[46]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[94]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[8]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[56]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[20]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[68]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[11]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[59]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[23]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[71]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[32]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[80]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[44]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[92]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[35]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[83]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[47]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[95]))
+            attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+            pad_temp.shared_1[threadIdx.x_1] = @tir.if_then_else((7 <= threadIdx.x_1), data[((cse_var_1 + threadIdx.x_1) - 7)], 0f32, dtype=float32)
+            attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+            pad_temp.shared_1[(threadIdx.x_1 + 49)] = @tir.if_then_else(((1 <= floormod((floordiv(threadIdx.x_1, 7) + 7), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 7), 9) < 8)), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 49), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 7), 9)*7)) + floormod(threadIdx.x_1, 7)) - 7)], 0f32, dtype=float32)
+            attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+            pad_temp.shared_1[(threadIdx.x_1 + 98)] = @tir.if_then_else(((1 <= floormod((floordiv(threadIdx.x_1, 7) + 5), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 5), 9) < 8)), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 98), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 5), 9)*7)) + floormod(threadIdx.x_1, 7)) - 7)], 0f32, dtype=float32)
+            attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+            pad_temp.shared_1[(threadIdx.x_1 + 147)] = @tir.if_then_else(((1 <= floormod((floordiv(threadIdx.x_1, 7) + 3), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 3), 9) < 8)), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 147), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 3), 9)*7)) + floormod(threadIdx.x_1, 7)) - 7)], 0f32, dtype=float32)
+            attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+            pad_temp.shared_1[(threadIdx.x_1 + 196)] = data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 196), 63)*49)) + ((floordiv(threadIdx.x_1, 7) + 1)*7)) + floormod(threadIdx.x_1, 7)) - 7)]
+            attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+            if @tir.likely((threadIdx.x_1 < 7), dtype=bool) {
+              pad_temp.shared_1[(threadIdx.x_1 + 245)] = 0f32
+            }
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+            kernel.shared_1[threadIdx.x_2] = kernel[(((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 12)*4608)) + (rc.outer.outer*36)) + (floormod(threadIdx.x_2, 12)*3)) + 1)]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+            if @tir.likely((threadIdx.x_2 < 47), dtype=bool) {
+              kernel.shared_1[(threadIdx.x_2 + 49)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 49), 12)*4608)) + (rc.outer.outer*36)) + (floormod((threadIdx.x_2 + 1), 12)*3)) + 1)]
+            }
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[0]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[48]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[12]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[60]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[3]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[51]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[15]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[63]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[24]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[72]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[36]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[84]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[27]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[75]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[39]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[87]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[1]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[49]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[13]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[61]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[4]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[52]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[16]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[64]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[25]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[73]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[37]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[85]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[28]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[76]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[40]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[88]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[2]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[50]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[14]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[62]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[5]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[53]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[17]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[65]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[26]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[74]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[38]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[86]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[29]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[77]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[41]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[89]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[6]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[54]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[18]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[66]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[9]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[57]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[21]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[69]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[30]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[78]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[42]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[90]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[33]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[81]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[45]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[93]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[7]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[55]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[19]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[67]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[10]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[58]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[22]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[70]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[31]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[79]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[43]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[91]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[34]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[82]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[46]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[94]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[8]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[56]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[20]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[68]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[11]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[59]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[23]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[71]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[32]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[80]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[44]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[92]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[35]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[83]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[47]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[95]))
+            attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+            pad_temp.shared_1[threadIdx.x_1] = @tir.if_then_else(((7 <= threadIdx.x_1) && (floormod(threadIdx.x_1, 7) < 6)), data[((cse_var_1 + threadIdx.x_1) - 6)], 0f32, dtype=float32)
+            attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+            pad_temp.shared_1[(threadIdx.x_1 + 49)] = @tir.if_then_else((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 7), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 7), 9) < 8)) && (floormod(threadIdx.x_1, 7) < 6)), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 49), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 7), 9)*7)) + floormod(threadIdx.x_1, 7)) - 6)], 0f32, dtype=float32)
+            attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+            pad_temp.shared_1[(threadIdx.x_1 + 98)] = @tir.if_then_else((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 5), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 5), 9) < 8)) && (floormod(threadIdx.x_1, 7) < 6)), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 98), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 5), 9)*7)) + floormod(threadIdx.x_1, 7)) - 6)], 0f32, dtype=float32)
+            attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+            pad_temp.shared_1[(threadIdx.x_1 + 147)] = @tir.if_then_else((((1 <= floormod((floordiv(threadIdx.x_1, 7) + 3), 9)) && (floormod((floordiv(threadIdx.x_1, 7) + 3), 9) < 8)) && (floormod(threadIdx.x_1, 7) < 6)), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 147), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 3), 9)*7)) + floormod(threadIdx.x_1, 7)) - 6)], 0f32, dtype=float32)
+            attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+            pad_temp.shared_1[(threadIdx.x_1 + 196)] = @tir.if_then_else((floormod(threadIdx.x_1, 7) < 6), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 196), 63)*49)) + ((floordiv(threadIdx.x_1, 7) + 1)*7)) + floormod(threadIdx.x_1, 7)) - 6)], 0f32, dtype=float32)
+            attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+            if @tir.likely((threadIdx.x_1 < 7), dtype=bool) {
+              pad_temp.shared_1[(threadIdx.x_1 + 245)] = 0f32
+            }
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+            kernel.shared_1[threadIdx.x_2] = kernel[(((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 12)*4608)) + (rc.outer.outer*36)) + (floormod(threadIdx.x_2, 12)*3)) + 2)]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 49;
+            if @tir.likely((threadIdx.x_2 < 47), dtype=bool) {
+              kernel.shared_1[(threadIdx.x_2 + 49)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 49), 12)*4608)) + (rc.outer.outer*36)) + (floormod((threadIdx.x_2 + 1), 12)*3)) + 2)]
+            }
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[0]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[48]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[12]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[60]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[3]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[51]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[15]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[63]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[24]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[72]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[36]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[84]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[27]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[75]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[39]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[87]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[1]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[49]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[13]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[61]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[4]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[52]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[16]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[64]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[25]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[73]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[37]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[85]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[28]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[76]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[40]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[88]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[2]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[50]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[14]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[62]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[5]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[53]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[17]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[65]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[26]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[74]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[38]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[86]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[29]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[77]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[41]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[89]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[6]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[54]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[18]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[66]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[9]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[57]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[21]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[69]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[30]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[78]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[42]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[90]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[33]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[81]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[45]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[93]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[7]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[55]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[19]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[67]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[10]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[58]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[22]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[70]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[31]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[79]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[43]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[91]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[34]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[82]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[46]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[94]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[8]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[56]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[20]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[68]))
+            conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[11]))
+            conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[59]))
+            conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[23]))
+            conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[71]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[32]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[80]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[44]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[92]))
+            conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[35]))
+            conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[83]))
+            conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[47]))
+            conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[95]))
           }
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*4)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 16)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 1)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 17)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 2)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 18)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 3)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 19)]))
-          attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
-          pad_temp.shared_1[threadIdx.x_1] = @tir.if_then_else((7 <= floormod(threadIdx.x_1, 49)), data[(((rc.outer.outer*196) + threadIdx.x_1) - 7)], 0f32, dtype=float32)
-          attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
-          if @tir.likely((threadIdx.x_2 < 32), dtype=bool) {
-            kernel.shared_1[threadIdx.x_2] = kernel[(((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 4)*4608)) + (rc.outer.outer*36)) + (floormod(threadIdx.x_2, 4)*9)) + 1)]
-          }
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*4)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 16)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 1)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 17)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 2)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 18)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 3)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 19)]))
-          attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
-          pad_temp.shared_1[threadIdx.x_1] = @tir.if_then_else(((7 <= floormod(threadIdx.x_1, 49)) && (floormod(threadIdx.x_1, 7) < 6)), data[(((rc.outer.outer*196) + threadIdx.x_1) - 6)], 0f32, dtype=float32)
-          attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
-          if @tir.likely((threadIdx.x_2 < 32), dtype=bool) {
-            kernel.shared_1[threadIdx.x_2] = kernel[(((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 4)*4608)) + (rc.outer.outer*36)) + (floormod(threadIdx.x_2, 4)*9)) + 2)]
-          }
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*4)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 16)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 1)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 17)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 2)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 18)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 3)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 19)]))
-          attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
-          pad_temp.shared_1[threadIdx.x_1] = @tir.if_then_else((1 <= floormod(threadIdx.x_1, 7)), data[(((rc.outer.outer*196) + threadIdx.x_1) - 1)], 0f32, dtype=float32)
-          attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
-          if @tir.likely((threadIdx.x_2 < 32), dtype=bool) {
-            kernel.shared_1[threadIdx.x_2] = kernel[(((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 4)*4608)) + (rc.outer.outer*36)) + (floormod(threadIdx.x_2, 4)*9)) + 3)]
-          }
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*4)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 16)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 1)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 17)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 2)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 18)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 3)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 19)]))
-          attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
-          pad_temp.shared_1[threadIdx.x_1] = data[((rc.outer.outer*196) + threadIdx.x_1)]
-          attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
-          if @tir.likely((threadIdx.x_2 < 32), dtype=bool) {
-            kernel.shared_1[threadIdx.x_2] = kernel[(((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 4)*4608)) + (rc.outer.outer*36)) + (floormod(threadIdx.x_2, 4)*9)) + 4)]
-          }
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*4)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 16)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 1)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 17)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 2)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 18)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 3)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 19)]))
-          attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
-          pad_temp.shared_1[threadIdx.x_1] = @tir.if_then_else((floormod(threadIdx.x_1, 7) < 6), data[(((rc.outer.outer*196) + threadIdx.x_1) + 1)], 0f32, dtype=float32)
-          attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
-          if @tir.likely((threadIdx.x_2 < 32), dtype=bool) {
-            kernel.shared_1[threadIdx.x_2] = kernel[(((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 4)*4608)) + (rc.outer.outer*36)) + (floormod(threadIdx.x_2, 4)*9)) + 5)]
-          }
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*4)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 16)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 1)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 17)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 2)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 18)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 3)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 19)]))
-          attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
-          pad_temp.shared_1[threadIdx.x_1] = @tir.if_then_else(((floormod(threadIdx.x_1, 49) < 42) && (1 <= floormod(threadIdx.x_1, 7))), data[(((rc.outer.outer*196) + threadIdx.x_1) + 6)], 0f32, dtype=float32)
-          attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
-          if @tir.likely((threadIdx.x_2 < 32), dtype=bool) {
-            kernel.shared_1[threadIdx.x_2] = kernel[(((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 4)*4608)) + (rc.outer.outer*36)) + (floormod(threadIdx.x_2, 4)*9)) + 6)]
-          }
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*4)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 16)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 1)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 17)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 2)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 18)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 3)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 19)]))
-          attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
-          pad_temp.shared_1[threadIdx.x_1] = @tir.if_then_else((floormod(threadIdx.x_1, 49) < 42), data[(((rc.outer.outer*196) + threadIdx.x_1) + 7)], 0f32, dtype=float32)
-          attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
-          if @tir.likely((threadIdx.x_2 < 32), dtype=bool) {
-            kernel.shared_1[threadIdx.x_2] = kernel[(((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 4)*4608)) + (rc.outer.outer*36)) + (floormod(threadIdx.x_2, 4)*9)) + 7)]
-          }
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*4)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 16)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 1)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 17)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 2)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 18)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 3)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 19)]))
-          attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
-          pad_temp.shared_1[threadIdx.x_1] = @tir.if_then_else(((floormod(threadIdx.x_1, 49) < 42) && (floormod(threadIdx.x_1, 7) < 6)), data[(((rc.outer.outer*196) + threadIdx.x_1) + 8)], 0f32, dtype=float32)
-          attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 196;
-          if @tir.likely((threadIdx.x_2 < 32), dtype=bool) {
-            kernel.shared_1[threadIdx.x_2] = kernel[(((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 4)*4608)) + (rc.outer.outer*36)) + (floormod(threadIdx.x_2, 4)*9)) + 8)]
-          }
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*4)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 16)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 1)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 17)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 2)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 18)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 3)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 19)]))
         }
-        compute[((blockIdx.x*392) + threadIdx.x)] = max((conv2d_nchw_1[0] + bias[((blockIdx.x*8) + floordiv(threadIdx.x, 49))]), 0f32)
-        compute[(((blockIdx.x*392) + threadIdx.x) + 196)] = max((conv2d_nchw_1[1] + bias[(((blockIdx.x*8) + floordiv(threadIdx.x, 49)) + 4)]), 0f32)
+        for (i1.inner: int32, 0, 4) {
+          compute[(((blockIdx.x*392) + (i1.inner*49)) + threadIdx.x)] = max((conv2d_nchw_1[i1.inner] + bias[((blockIdx.x*8) + i1.inner)]), 0f32)
+          compute[((((blockIdx.x*392) + (i1.inner*49)) + threadIdx.x) + 196)] = max((conv2d_nchw_1[(i1.inner + 4)] + bias[(((blockIdx.x*8) + i1.inner) + 4)]), 0f32)
+        }
       }
     }
 
@@ -430,7 +663,7 @@ We build the binary and check its correctness and performance.
 
  .. code-block:: none
 
-    Execution time of this operator: 0.381 ms
+    Execution time of this operator: 0.228 ms
 
 
 
@@ -478,9 +711,9 @@ 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_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_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=2)
+    conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=1)
     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)
@@ -490,18 +723,18 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
     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_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=1)
-    conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=4)
+    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_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=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=1)
     conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=1)
     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_i, compute_i1_i = s[compute].split(compute_i1, factor=4)
+    compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=1)
     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)
@@ -527,14 +760,14 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
     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=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=49)
     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=49)
     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", 512)
+    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:
@@ -552,142 +785,354 @@ 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) {
-      float conv2d_nchw[2];
-      __shared__ float pad_temp_shared[196];
-      __shared__ float kernel_shared[32];
+    extern "C" __global__ void __launch_bounds__(49) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
+      float conv2d_nchw[8];
+      __shared__ float pad_temp_shared[252];
+      __shared__ float kernel_shared[96];
       conv2d_nchw[0] = 0.000000e+00f;
+      conv2d_nchw[4] = 0.000000e+00f;
       conv2d_nchw[1] = 0.000000e+00f;
+      conv2d_nchw[5] = 0.000000e+00f;
+      conv2d_nchw[2] = 0.000000e+00f;
+      conv2d_nchw[6] = 0.000000e+00f;
+      conv2d_nchw[3] = 0.000000e+00f;
+      conv2d_nchw[7] = 0.000000e+00f;
       for (int rc_outer_outer = 0; rc_outer_outer < 128; ++rc_outer_outer) {
         __syncthreads();
-        pad_temp_shared[((int)threadIdx.x)] = (((7 <= (((int)threadIdx.x) % 49)) && (1 <= (((int)threadIdx.x) % 7))) ? data[(((rc_outer_outer * 196) + ((int)threadIdx.x)) - 8)] : 0.000000e+00f);
-        if (((int)threadIdx.x) < 32) {
-          kernel_shared[((int)threadIdx.x)] = kernel[((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) >> 2) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) & 3) * 9))];
+        pad_temp_shared[((int)threadIdx.x)] = (((7 <= ((int)threadIdx.x)) && (1 <= (((int)threadIdx.x) % 7))) ? data[(((rc_outer_outer * 196) + ((int)threadIdx.x)) - 8)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 49)] = ((((1 <= (((((int)threadIdx.x) / 7) + 7) % 9)) && ((((((int)threadIdx.x) / 7) + 7) % 9) < 8)) && (1 <= (((int)threadIdx.x) % 7))) ? data[(((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 49) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 7) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 98)] = ((((1 <= (((((int)threadIdx.x) / 7) + 5) % 9)) && ((((((int)threadIdx.x) / 7) + 5) % 9) < 8)) && (1 <= (((int)threadIdx.x) % 7))) ? data[(((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 98) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 5) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 147)] = ((((1 <= (((((int)threadIdx.x) / 7) + 3) % 9)) && ((((((int)threadIdx.x) / 7) + 3) % 9) < 8)) && (1 <= (((int)threadIdx.x) % 7))) ? data[(((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 147) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 3) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 196)] = ((1 <= (((int)threadIdx.x) % 7)) ? data[((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 196) / 63) * 49)) + ((int)threadIdx.x)) - 1)] : 0.000000e+00f);
+        if (((int)threadIdx.x) < 7) {
+          pad_temp_shared[(((int)threadIdx.x) + 245)] = 0.000000e+00f;
         }
-        __syncthreads();
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 4)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 16)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 1)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 17)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 2)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 18)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 3)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 19)]));
-        __syncthreads();
-        pad_temp_shared[((int)threadIdx.x)] = ((7 <= (((int)threadIdx.x) % 49)) ? data[(((rc_outer_outer * 196) + ((int)threadIdx.x)) - 7)] : 0.000000e+00f);
-        if (((int)threadIdx.x) < 32) {
-          kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) >> 2) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) & 3) * 9)) + 1)];
+        kernel_shared[((int)threadIdx.x)] = kernel[((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) / 12) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) % 12) * 3))];
+        if (((int)threadIdx.x) < 47) {
+          kernel_shared[(((int)threadIdx.x) + 49)] = kernel[((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 49) / 12) * 4608)) + (rc_outer_outer * 36)) + (((((int)threadIdx.x) + 1) % 12) * 3))];
         }
         __syncthreads();
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 4)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 16)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 1)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 17)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 2)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 18)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 3)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 19)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[0]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[48]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[12]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[60]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[3]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[51]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[15]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[63]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[24]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[72]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[36]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[84]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[27]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[75]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[39]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[87]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[1]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[49]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[13]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[61]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[4]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[52]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[16]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[64]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[25]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[73]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[37]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[85]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[28]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[76]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[40]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[88]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[2]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[50]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[14]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[62]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[5]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[53]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[17]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[65]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[26]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[74]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[38]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[86]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[29]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[77]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[41]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[89]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[6]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[54]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[18]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[66]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[9]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[57]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[21]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[69]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[30]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[78]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[42]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[90]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[33]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[81]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[45]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[93]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[7]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[55]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[19]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[67]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[10]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[58]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[22]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[70]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[31]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[79]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[43]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[91]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[34]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[82]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[46]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[94]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[8]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[56]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[20]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[68]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[11]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[59]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[23]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[71]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[32]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[80]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[44]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[92]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[35]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[83]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[47]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[95]));
         __syncthreads();
-        pad_temp_shared[((int)threadIdx.x)] = (((7 <= (((int)threadIdx.x) % 49)) && ((((int)threadIdx.x) % 7) < 6)) ? data[(((rc_outer_outer * 196) + ((int)threadIdx.x)) - 6)] : 0.000000e+00f);
-        if (((int)threadIdx.x) < 32) {
-          kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) >> 2) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) & 3) * 9)) + 2)];
+        pad_temp_shared[((int)threadIdx.x)] = ((7 <= ((int)threadIdx.x)) ? data[(((rc_outer_outer * 196) + ((int)threadIdx.x)) - 7)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 49)] = (((1 <= (((((int)threadIdx.x) / 7) + 7) % 9)) && ((((((int)threadIdx.x) / 7) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 49) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 7) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 7)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 98)] = (((1 <= (((((int)threadIdx.x) / 7) + 5) % 9)) && ((((((int)threadIdx.x) / 7) + 5) % 9) < 8)) ? data[(((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 98) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 5) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 7)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 147)] = (((1 <= (((((int)threadIdx.x) / 7) + 3) % 9)) && ((((((int)threadIdx.x) / 7) + 3) % 9) < 8)) ? data[(((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 147) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 3) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 7)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 196)] = data[(((rc_outer_outer * 196) + (((((int)threadIdx.x) + 196) / 63) * 49)) + ((int)threadIdx.x))];
+        if (((int)threadIdx.x) < 7) {
+          pad_temp_shared[(((int)threadIdx.x) + 245)] = 0.000000e+00f;
         }
-        __syncthreads();
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 4)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 16)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 1)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 17)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 2)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 18)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 3)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 19)]));
-        __syncthreads();
-        pad_temp_shared[((int)threadIdx.x)] = ((1 <= (((int)threadIdx.x) % 7)) ? data[(((rc_outer_outer * 196) + ((int)threadIdx.x)) - 1)] : 0.000000e+00f);
-        if (((int)threadIdx.x) < 32) {
-          kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) >> 2) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) & 3) * 9)) + 3)];
+        kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) / 12) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) % 12) * 3)) + 1)];
+        if (((int)threadIdx.x) < 47) {
+          kernel_shared[(((int)threadIdx.x) + 49)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 49) / 12) * 4608)) + (rc_outer_outer * 36)) + (((((int)threadIdx.x) + 1) % 12) * 3)) + 1)];
         }
         __syncthreads();
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 4)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 16)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 1)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 17)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 2)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 18)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 3)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 19)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[0]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[48]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[12]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[60]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[3]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[51]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[15]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[63]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[24]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[72]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[36]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[84]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[27]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[75]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[39]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[87]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[1]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[49]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[13]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[61]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[4]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[52]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[16]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[64]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[25]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[73]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[37]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[85]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[28]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[76]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[40]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[88]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[2]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[50]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[14]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[62]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[5]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[53]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[17]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[65]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[26]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[74]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[38]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[86]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[29]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[77]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[41]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[89]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[6]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[54]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[18]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[66]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[9]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[57]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[21]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[69]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[30]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[78]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[42]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[90]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[33]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[81]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[45]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[93]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[7]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[55]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[19]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[67]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[10]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[58]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[22]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[70]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[31]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[79]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[43]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[91]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[34]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[82]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[46]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[94]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[8]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[56]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[20]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[68]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[11]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[59]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[23]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[71]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[32]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[80]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[44]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[92]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[35]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[83]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[47]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[95]));
         __syncthreads();
-        pad_temp_shared[((int)threadIdx.x)] = data[((rc_outer_outer * 196) + ((int)threadIdx.x))];
-        if (((int)threadIdx.x) < 32) {
-          kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) >> 2) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) & 3) * 9)) + 4)];
+        pad_temp_shared[((int)threadIdx.x)] = (((7 <= ((int)threadIdx.x)) && ((((int)threadIdx.x) % 7) < 6)) ? data[(((rc_outer_outer * 196) + ((int)threadIdx.x)) - 6)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 49)] = ((((1 <= (((((int)threadIdx.x) / 7) + 7) % 9)) && ((((((int)threadIdx.x) / 7) + 7) % 9) < 8)) && ((((int)threadIdx.x) % 7) < 6)) ? data[(((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 49) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 7) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 6)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 98)] = ((((1 <= (((((int)threadIdx.x) / 7) + 5) % 9)) && ((((((int)threadIdx.x) / 7) + 5) % 9) < 8)) && ((((int)threadIdx.x) % 7) < 6)) ? data[(((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 98) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 5) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 6)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 147)] = ((((1 <= (((((int)threadIdx.x) / 7) + 3) % 9)) && ((((((int)threadIdx.x) / 7) + 3) % 9) < 8)) && ((((int)threadIdx.x) % 7) < 6)) ? data[(((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 147) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 3) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 6)] : 0.000000e+00f);
+        pad_temp_shared[(((int)threadIdx.x) + 196)] = (((((int)threadIdx.x) % 7) < 6) ? data[((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 196) / 63) * 49)) + ((int)threadIdx.x)) + 1)] : 0.000000e+00f);
+        if (((int)threadIdx.x) < 7) {
+          pad_temp_shared[(((int)threadIdx.x) + 245)] = 0.000000e+00f;
         }
-        __syncthreads();
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 4)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 16)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 1)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 17)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 2)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 18)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 3)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 19)]));
-        __syncthreads();
-        pad_temp_shared[((int)threadIdx.x)] = (((((int)threadIdx.x) % 7) < 6) ? data[(((rc_outer_outer * 196) + ((int)threadIdx.x)) + 1)] : 0.000000e+00f);
-        if (((int)threadIdx.x) < 32) {
-          kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) >> 2) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) & 3) * 9)) + 5)];
-        }
-        __syncthreads();
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 4)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 16)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 1)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 17)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 2)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 18)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 3)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 19)]));
-        __syncthreads();
-        pad_temp_shared[((int)threadIdx.x)] = ((((((int)threadIdx.x) % 49) < 42) && (1 <= (((int)threadIdx.x) % 7))) ? data[(((rc_outer_outer * 196) + ((int)threadIdx.x)) + 6)] : 0.000000e+00f);
-        if (((int)threadIdx.x) < 32) {
-          kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) >> 2) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) & 3) * 9)) + 6)];
-        }
-        __syncthreads();
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 4)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 16)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 1)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 17)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 2)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 18)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 3)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 19)]));
-        __syncthreads();
-        pad_temp_shared[((int)threadIdx.x)] = (((((int)threadIdx.x) % 49) < 42) ? data[(((rc_outer_outer * 196) + ((int)threadIdx.x)) + 7)] : 0.000000e+00f);
-        if (((int)threadIdx.x) < 32) {
-          kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) >> 2) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) & 3) * 9)) + 7)];
-        }
-        __syncthreads();
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 4)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 16)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 1)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 17)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 2)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 18)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 3)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 19)]));
-        __syncthreads();
-        pad_temp_shared[((int)threadIdx.x)] = ((((((int)threadIdx.x) % 49) < 42) && ((((int)threadIdx.x) % 7) < 6)) ? data[(((rc_outer_outer * 196) + ((int)threadIdx.x)) + 8)] : 0.000000e+00f);
-        if (((int)threadIdx.x) < 32) {
-          kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) >> 2) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) & 3) * 9)) + 8)];
+        kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) / 12) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) % 12) * 3)) + 2)];
+        if (((int)threadIdx.x) < 47) {
+          kernel_shared[(((int)threadIdx.x) + 49)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 49) / 12) * 4608)) + (rc_outer_outer * 36)) + (((((int)threadIdx.x) + 1) % 12) * 3)) + 2)];
         }
         __syncthreads();
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 4)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 16)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 1)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 17)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 2)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 18)]));
-        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 3)]));
-        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 19)]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[0]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[48]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[12]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[60]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[3]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[51]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[15]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[63]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[24]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[72]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[36]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[84]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[27]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[75]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[39]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[87]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[1]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[49]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[13]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[61]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[4]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[52]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[16]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[64]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[25]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[73]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[37]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[85]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[28]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[76]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[40]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[88]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[2]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[50]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[14]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[62]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[5]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[53]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[17]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[65]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[26]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[74]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[38]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[86]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[29]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[77]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[41]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[89]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[6]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[54]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[18]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[66]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[9]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[57]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[21]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[69]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[30]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[78]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[42]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[90]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[33]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[81]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[45]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[93]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[7]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[55]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[19]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[67]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[10]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[58]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[22]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[70]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[31]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[79]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[43]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[91]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[34]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[82]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[46]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[94]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[8]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[56]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[20]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[68]));
+        conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[11]));
+        conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[59]));
+        conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[23]));
+        conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[71]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[32]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[80]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[44]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[92]));
+        conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[35]));
+        conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[83]));
+        conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[47]));
+        conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[95]));
+      }
+      for (int i1_inner = 0; i1_inner < 4; ++i1_inner) {
+        compute[(((((int)blockIdx.x) * 392) + (i1_inner * 49)) + ((int)threadIdx.x))] = max((conv2d_nchw[i1_inner] + bias[((((int)blockIdx.x) * 8) + i1_inner)]), 0.000000e+00f);
+        compute[((((((int)blockIdx.x) * 392) + (i1_inner * 49)) + ((int)threadIdx.x)) + 196)] = max((conv2d_nchw[(i1_inner + 4)] + bias[(((((int)blockIdx.x) * 8) + i1_inner) + 4)]), 0.000000e+00f);
       }
-      compute[((((int)blockIdx.x) * 392) + ((int)threadIdx.x))] = max((conv2d_nchw[0] + bias[((((int)blockIdx.x) * 8) + (((int)threadIdx.x) / 49))]), 0.000000e+00f);
-      compute[(((((int)blockIdx.x) * 392) + ((int)threadIdx.x)) + 196)] = max((conv2d_nchw[1] + bias[(((((int)blockIdx.x) * 8) + (((int)threadIdx.x) / 49)) + 4)]), 0.000000e+00f);
     }
 
 
@@ -748,7 +1193,7 @@ In the example below we resume the status and do more 5 trials.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 5 minutes  45.244 seconds)
+   **Total running time of the script:** ( 5 minutes  56.080 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 fab2847c1c..52db7c0b4a 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
@@ -643,7 +643,7 @@ 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)  
-       8.2434       8.2416       8.2486       8.2401       0.0037   
+       8.2229       8.2286       8.2300       8.2102       0.0090   
                
 
 
@@ -671,7 +671,7 @@ Other Tips
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  4.412 seconds)
+   **Total running time of the script:** ( 1 minutes  4.140 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 faae3ad2da..bcd8cd290b 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
@@ -662,7 +662,7 @@ 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)  
-      752.3158     750.6468     756.5245     749.7761      2.9972   
+      770.9234     770.6793     775.3688     766.7221      3.5342   
                
 
 
@@ -690,7 +690,7 @@ Other Tips
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  33.722 seconds)
+   **Total running time of the script:** ( 1 minutes  34.623 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 d0f9308b1c..1a17cfc5ec 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
@@ -386,104 +386,30 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
                  placeholder_4: Buffer(placeholder_14: Pointer(float32), float32, [65536], []),
                  compute: Buffer(compute_2: Pointer(float32), float32, [65536], [])}
       buffer_map = {placeholder_5: placeholder, placeholder_6: placeholder_1, placeholder_7: placeholder_2, placeholder_8: placeholder_3, placeholder_9: placeholder_4, compute_1: compute}
-      preflattened_buffer_map = {placeholder_7: placeholder_15: Buffer(placeholder_12, int32, [4916], []), placeholder_5: placeholder_16: Buffer(placeholder_10, float32, [128, 256], []), placeholder_8: placeholder_17: Buffer(placeholder_13, int32, [33], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], []), placeholder_9: placeholder_18: Buffer(placeholder_14, float32, [128, 512], []), placeholder_6: placeholder_19: Buffer(placeholder_11, float32, [4916, 16, 1], [])} {
-      for (i0.outer: int32, 0, 2) "parallel" {
-        allocate(compute_4: Pointer(global float32), float32, [1024]), storage_scope = global;
-        for (i1.outer: int32, 0, 32) {
-          for (i.outer.inner: int32, 0, 2) {
-            for (i.inner.init: int32, 0, 32) {
-              let cse_var_1: int32 = ((i.outer.inner*512) + (i.inner.init*16))
-               {
-                compute_5: Buffer(compute_4, float32, [1024], [])[cse_var_1] = 0f32
-                compute_5[(cse_var_1 + 1)] = 0f32
-                compute_5[(cse_var_1 + 2)] = 0f32
-                compute_5[(cse_var_1 + 3)] = 0f32
-                compute_5[(cse_var_1 + 4)] = 0f32
-                compute_5[(cse_var_1 + 5)] = 0f32
-                compute_5[(cse_var_1 + 6)] = 0f32
-                compute_5[(cse_var_1 + 7)] = 0f32
-                compute_5[(cse_var_1 + 8)] = 0f32
-                compute_5[(cse_var_1 + 9)] = 0f32
-                compute_5[(cse_var_1 + 10)] = 0f32
-                compute_5[(cse_var_1 + 11)] = 0f32
-                compute_5[(cse_var_1 + 12)] = 0f32
-                compute_5[(cse_var_1 + 13)] = 0f32
-                compute_5[(cse_var_1 + 14)] = 0f32
-                compute_5[(cse_var_1 + 15)] = 0f32
-              }
-            }
-            for (elem_idx: int32, 0, (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])) {
-              for (i.inner: int32, 0, 32) {
-                if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-                  let cse_var_2: int32 = ((i.outer.inner*512) + (i.inner*16))
-                  compute_5[cse_var_2] = (compute_5[cse_var_2] + (placeholder_1[((placeholder_3[i1.outer]*16) + (elem_idx*16))]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
-                }
-                if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-                  let cse_var_3: int32 = (((i.outer.inner*512) + (i.inner*16)) + 1)
-                  compute_5[cse_var_3] = (compute_5[cse_var_3] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 1)]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
-                }
-                if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-                  let cse_var_4: int32 = (((i.outer.inner*512) + (i.inner*16)) + 2)
-                  compute_5[cse_var_4] = (compute_5[cse_var_4] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 2)]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
-                }
-                if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-                  let cse_var_5: int32 = (((i.outer.inner*512) + (i.inner*16)) + 3)
-                  compute_5[cse_var_5] = (compute_5[cse_var_5] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 3)]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
-                }
-                if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-                  let cse_var_6: int32 = (((i.outer.inner*512) + (i.inner*16)) + 4)
-                  compute_5[cse_var_6] = (compute_5[cse_var_6] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 4)]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
-                }
-                if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-                  let cse_var_7: int32 = (((i.outer.inner*512) + (i.inner*16)) + 5)
-                  compute_5[cse_var_7] = (compute_5[cse_var_7] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 5)]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
-                }
-                if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-                  let cse_var_8: int32 = (((i.outer.inner*512) + (i.inner*16)) + 6)
-                  compute_5[cse_var_8] = (compute_5[cse_var_8] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 6)]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
-                }
-                if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-                  let cse_var_9: int32 = (((i.outer.inner*512) + (i.inner*16)) + 7)
-                  compute_5[cse_var_9] = (compute_5[cse_var_9] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 7)]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+      preflattened_buffer_map = {placeholder_7: placeholder_15: Buffer(placeholder_12, int32, [4916], []), placeholder_6: placeholder_16: Buffer(placeholder_11, float32, [4916, 16, 1], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], []), placeholder_5: placeholder_17: Buffer(placeholder_10, float32, [128, 256], []), placeholder_8: placeholder_18: Buffer(placeholder_13, int32, [33], []), placeholder_9: placeholder_19: Buffer(placeholder_14, float32, [128, 512], [])} {
+      for (i0.outer.i1.outer.fused: int32, 0, 64) "parallel" {
+        allocate(compute_4: Pointer(global float32), float32, [1024]), storage_scope = global {
+          for (i.outer.inner: int32, 0, 8) {
+            for (nb_j.inner: int32, 0, 2) {
+              for (i.inner.init: int32, 0, 4) {
+                for (j.init: int32, 0, 16) {
+                  compute_5: Buffer(compute_4, float32, [1024], [])[((((i.outer.inner*128) + (i.inner.init*32)) + (nb_j.inner*16)) + j.init)] = 0f32
                 }
-                if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-                  let cse_var_10: int32 = (((i.outer.inner*512) + (i.inner*16)) + 8)
-                  compute_5[cse_var_10] = (compute_5[cse_var_10] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 8)]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
-                }
-                if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-                  let cse_var_11: int32 = (((i.outer.inner*512) + (i.inner*16)) + 9)
-                  compute_5[cse_var_11] = (compute_5[cse_var_11] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 9)]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
-                }
-                if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-                  let cse_var_12: int32 = (((i.outer.inner*512) + (i.inner*16)) + 10)
-                  compute_5[cse_var_12] = (compute_5[cse_var_12] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 10)]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
-                }
-                if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-                  let cse_var_13: int32 = (((i.outer.inner*512) + (i.inner*16)) + 11)
-                  compute_5[cse_var_13] = (compute_5[cse_var_13] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 11)]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
-                }
-                if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-                  let cse_var_14: int32 = (((i.outer.inner*512) + (i.inner*16)) + 12)
-                  compute_5[cse_var_14] = (compute_5[cse_var_14] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 12)]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
-                }
-                if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-                  let cse_var_15: int32 = (((i.outer.inner*512) + (i.inner*16)) + 13)
-                  compute_5[cse_var_15] = (compute_5[cse_var_15] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 13)]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
-                }
-                if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-                  let cse_var_16: int32 = (((i.outer.inner*512) + (i.inner*16)) + 14)
-                  compute_5[cse_var_16] = (compute_5[cse_var_16] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 14)]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
-                }
-                if @tir.likely((elem_idx < (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-                  let cse_var_17: int32 = (((i.outer.inner*512) + (i.inner*16)) + 15)
-                  compute_5[cse_var_17] = (compute_5[cse_var_17] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 15)]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+              }
+              for (elem_idx: int32, 0, let cse_var_1: int32 = ((floormod(i0.outer.i1.outer.fused, 16)*2) + nb_j.inner) in (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])) {
+                for (i.inner: int32, 0, 4) {
+                  for (j: int32, 0, 16) {
+                    let cse_var_3: int32 = ((floormod(i0.outer.i1.outer.fused, 16)*2) + nb_j.inner)
+                    let cse_var_2: int32 = ((((i.outer.inner*128) + (i.inner*32)) + (nb_j.inner*16)) + j)
+                    compute_5[cse_var_2] = (compute_5[cse_var_2] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + j)]*max(placeholder[((((floordiv(i0.outer.i1.outer.fused, 16)*8192) + (i.outer.inner*1024)) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+                  }
                 }
               }
             }
           }
-          for (i0.inner: int32, 0, 64) {
-            let cse_var_18: int32 = (((i0.outer*32768) + (i0.inner*512)) + (i1.outer*16))
-            compute[ramp(cse_var_18, 1, 16)] = max((compute_5[ramp((i0.inner*16), 1, 16)] + placeholder_4[ramp(cse_var_18, 1, 16)]), broadcast(0f32, 16))
+          for (i0.inner: int32, 0, 32) {
+            let cse_var_4: int32 = (((floordiv(i0.outer.i1.outer.fused, 16)*16384) + (i0.inner*512)) + (floormod(i0.outer.i1.outer.fused, 16)*32))
+            compute[ramp(cse_var_4, 1, 32)] = max((compute_5[ramp((i0.inner*32), 1, 32)] + placeholder_4[ramp(cse_var_4, 1, 32)]), broadcast(0f32, 32))
           }
         }
       }
@@ -539,7 +465,7 @@ We build the binary and check its correctness and performance.
 
  .. code-block:: none
 
-    Execution time of this operator: 1.871 ms
+    Execution time of this operator: 1.443 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 2665a07690..dbaff6ec37 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
@@ -5,12 +5,12 @@
 
 Computation times
 =================
-**00:38.715** total execution time for **how_to_tune_with_autotvm** files:
+**00:33.132** 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``)           | 00:38.679 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_conv2d_cuda.py` (``tune_conv2d_cuda.py``)           | 00:33.097 | 0.0 MB |
 +--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_x86.py` (``tune_relay_x86.py``)               | 00:00.021 | 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 |
 +--------------------------------------------------------------------------------------------------+-----------+--------+
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 1983b350ce..cb73b04d99 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
@@ -387,7 +387,7 @@ 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 871, 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, 8, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 64, 2]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6060643
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 64, 2, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 8, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3615716
     No: 2   GFLOPS: 0.00/0.00       result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
@@ -510,9 +510,8 @@ 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 871, 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, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 512]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9483670
-    No: 3   GFLOPS: 7.79/7.79       result: MeasureResult(costs=(0.0297286395,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.632194995880127, timestamp=1668475003.5101364)        [('tile_f', [-1, 4, 1, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 16, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,470058
-    No: 4   GFLOPS: 0.00/7.79       result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 1, 64, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 16, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8020043
+    No: 3   GFLOPS: 0.00/0.00       result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -634,8 +633,8 @@ 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 871, 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, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 32, 2]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,634592
-    No: 5   GFLOPS: 0.00/7.79       result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 4, 1, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 32, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6863342
+    No: 4   GFLOPS: 0.00/0.00       result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -757,8 +756,8 @@ 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 871, 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, 64, 4]), ('tile_y', [-1, 1, 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', 512), ('unroll_explicit', 1)],None,8544053
-    No: 6   GFLOPS: 0.00/7.79       result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 4, 4, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1072397
+    No: 5   GFLOPS: 0.00/0.00       result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -880,8 +879,9 @@ 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 871, 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, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 16, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9956499
-    No: 7   GFLOPS: 0.00/7.79       result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 16, 4, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 16, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,3426743
+    No: 6   GFLOPS: 11.64/11.64     result: MeasureResult(costs=(0.019892005833333334,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.3444783687591553, timestamp=1668481233.3563986)       [('tile_f', [-1, 1, 4, 2]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 2, 32]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2662732
+    No: 7   GFLOPS: 0.00/11.64      result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1003,8 +1003,8 @@ 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 871, 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, 7, 1, 1]), ('tile_rc', [-1, 8, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5820058
-    No: 8   GFLOPS: 0.00/7.79       result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 4, 32, 2]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 64, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,21872
+    No: 8   GFLOPS: 0.00/11.64      result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1126,9 +1126,8 @@ 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 871, 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, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 2, 64]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6553931
-    No: 9   GFLOPS: 24.56/24.56     result: MeasureResult(costs=(0.009426631357142857,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.1752407550811768, timestamp=1668475009.0452948)       [('tile_f', [-1, 4, 1, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 8, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,8789442
-    No: 10  GFLOPS: 0.00/24.56      result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 64, 8, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 32, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1054493
+    No: 9   GFLOPS: 0.00/11.64      result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1250,8 +1249,8 @@ 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 871, 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, 32]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 512, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4294152
-    No: 11  GFLOPS: 0.00/24.56      result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 2, 16, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 1, 512]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8711719
+    No: 10  GFLOPS: 0.00/11.64      result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1373,8 +1372,8 @@ 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 871, 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, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 256, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7422823
-    No: 12  GFLOPS: 0.00/24.56      result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 4, 1, 1]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 128]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2117502
+    No: 11  GFLOPS: 0.00/11.64      result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1496,315 +1495,395 @@ 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 871, 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, 16, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2792842
-    No: 13  GFLOPS: 0.00/24.56      result: Traceback (most recent call last):
-      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 738, in __call__
-        yield remote, remote.load_module(os.path.split(build_result.filename)[1])
-      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 702, in run_through_rpc
-        costs = time_f(*args).results
-      File "/workspace/python/tvm/runtime/module.py", line 357, in evaluator
-        blob = feval(*args)
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 4, 2, 32]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4634932
+    No: 12  GFLOPS: 0.00/11.64      result: Traceback (most recent call last):
+      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
+        func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
+      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
+        func = build(s, args, target_host=task.target_host, 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 262, in tvm._ffi._cy3.core.FuncCall
-      File "tvm/_ffi/_cython/./packed_func.pxi", line 251, in tvm._ffi._cy3.core.FuncCall3
+      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):
-      4: TVMFuncCall
+      24: TVMFuncCall
             at ../src/runtime/c_runtime_api.cc:477
-      3: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
-            at ../include/tvm/runtime/packed_func.h:1217
-      2: tvm::runtime::RPCWrappedFunc::operator()(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
-            at ../src/runtime/rpc/rpc_module.cc:129
-      1: tvm::runtime::RPCClientSession::CallFunc(void*, TVMValue const*, int const*, int, std::function<void (tvm::runtime::TVMArgs)> const&)
-            at ../src/runtime/rpc/rpc_endpoint.cc:1012
-      0: tvm::runtime::RPCEndpoint::CallFunc(void*, TVMValue const*, int const*, int, std::function<void (tvm::runtime::TVMArgs)>)
-            at ../src/runtime/rpc/rpc_endpoint.cc:804
-      File "../src/runtime/rpc/rpc_endpoint.cc", line 804
-    TVMError: 
-    ---------------------------------------------------------------
-    An error occurred during the execution of TVM.
-    For more information, please see: https://tvm.apache.org/docs/errors.html
-    ---------------------------------------------------------------
-      Check failed: (code == RPCCode::kReturn) is false: code=kShutdown
-
-    During handling of the above exception, another exception occurred:
-
-    Traceback (most recent call last):
-      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 702, in run_through_rpc
-        costs = time_f(*args).results
-      File "/usr/lib/python3.7/contextlib.py", line 130, in __exit__
-        self.gen.throw(type, value, traceback)
-      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 742, in __call__
-        remote.remove(build_result.filename)
-      File "/workspace/python/tvm/rpc/client.py", line 144, in remove
-        self._remote_funcs["remove"] = self.get_function("tvm.rpc.server.remove")
-      File "/workspace/python/tvm/rpc/client.py", line 72, in get_function
-        return self._sess.get_function(name)
-      File "/workspace/python/tvm/runtime/module.py", line 171, in get_function
-        self.handle, c_str(name), ctypes.c_int(query_imports), ctypes.byref(ret_handle)
-      File "/workspace/python/tvm/_ffi/base.py", line 348, in check_call
-        raise get_last_ffi_error()
-    tvm._ffi.base.TVMError: Traceback (most recent call last):
-      52: 0xffffffffffffffff
-      51: _start
-      50: __libc_start_main
-      49: _Py_UnixMain
-      48: 0x0000000000650da0
-      47: 0x0000000000650afa
-      46: _PyFunction_FastCallDict
-      45: _PyEval_EvalCodeWithName
-      44: _PyEval_EvalFrameDefault
-      43: _PyFunction_FastCallKeywords
-      42: _PyEval_EvalCodeWithName
-      41: _PyEval_EvalFrameDefault
-      40: _PyMethodDef_RawFastCallKeywords
-      39: 0x0000000000546369
-      38: _PyEval_EvalCodeWithName
-      37: _PyEval_EvalFrameDefault
-      36: _PyFunction_FastCallKeywords
-      35: _PyEval_EvalCodeWithName
-      34: _PyEval_EvalFrameDefault
-      33: _PyFunction_FastCallDict
-      32: _PyEval_EvalCodeWithName
-      31: _PyEval_EvalFrameDefault
-      30: _PyObject_FastCallDict
-      29: 0x00000000004c06e1
-      28: _PyFunction_FastCallDict
-      27: _PyEval_EvalFrameDefault
-      26: _PyMethodDescr_FastCallKeywords
-      25: 0x00000000005dcb58
-      24: 0x00000000005dc83f
-      23: 0x00000000004ba127
-      22: _PyEval_EvalFrameDefault
-      21: _PyFunction_FastCallKeywords
-      20: _PyEval_EvalFrameDefault
-      19: _PyFunction_FastCallKeywords
-      18: _PyEval_EvalFrameDefault
-      17: _PyFunction_FastCallKeywords
-      16: _PyEval_EvalCodeWithName
-      15: _PyEval_EvalFrameDefault
-      14: 0x0000000000537c30
-      13: _PyObject_FastCallKeywords
-      12: 0x00007f3866235fa2
-      11: _ctypes_callproc
-      10: ffi_call
-      9: ffi_call_unix64
-      8: TVMModGetFunction
-            at ../src/runtime/c_runtime_api.cc:408
-      7: tvm::runtime::ModuleNode::GetFunction(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, bool)
-            at ../src/runtime/module.cc:66
-      6: tvm::runtime::RPCModuleNode::GetFunction(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, tvm::runtime::ObjectPtr<tvm::runtime::Object> const&)
-            at ../src/runtime/rpc/rpc_module.cc:185
-      5: tvm::runtime::RPCClientSession::GetFunction(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)
-            at ../src/runtime/rpc/rpc_endpoint.cc:1007
-      4: tvm::runtime::TVMRetValue tvm::runtime::RPCEndpoint::SysCallRemote<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&>(tvm::runtime::RPCCode, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)
-            at ../src/runtime/rpc/rpc_endpoint.h:223
-      3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<int, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&>(int&&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) const
+      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:1731
+      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:1671
+      19: run<>
+            at ../include/tvm/runtime/packed_func.h:1631
+      18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1631
+      17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1631
+      16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1631
+      15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1631
+      14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1646
+      13: operator()
+            at ../src/driver/driver_api.cc:388
+      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:374
+      11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+            at ../src/driver/driver_api.cc:269
+      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:453
+      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:1750
+      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:1694
+      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:1618
       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/rpc/rpc_endpoint.cc:684
-      File "../src/runtime/rpc/rpc_endpoint.cc", line 684
-    TVMError: 
-    ---------------------------------------------------------------
-    An error occurred during the execution of TVM.
-    For more information, please see: https://tvm.apache.org/docs/errors.html
-    ---------------------------------------------------------------
-      Check failed: (code == RPCCode::kReturn) is false: code=1
+            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 871, 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):
-      52: 0xffffffffffffffff
-      51: _start
-      50: __libc_start_main
-      49: _Py_UnixMain
-      48: 0x0000000000650da0
-      47: 0x0000000000650afa
-      46: _PyFunction_FastCallDict
-      45: _PyEval_EvalCodeWithName
-      44: _PyEval_EvalFrameDefault
-      43: _PyFunction_FastCallKeywords
-      42: _PyEval_EvalCodeWithName
-      41: _PyEval_EvalFrameDefault
-      40: _PyMethodDef_RawFastCallKeywords
-      39: 0x0000000000546369
-      38: _PyEval_EvalCodeWithName
-      37: _PyEval_EvalFrameDefault
-      36: _PyFunction_FastCallKeywords
-      35: _PyEval_EvalCodeWithName
-      34: _PyEval_EvalFrameDefault
-      33: _PyFunction_FastCallDict
-      32: _PyEval_EvalCodeWithName
-      31: _PyEval_EvalFrameDefault
-      30: _PyObject_FastCallDict
-      29: 0x00000000004c06e1
-      28: _PyFunction_FastCallDict
-      27: _PyEval_EvalFrameDefault
-      26: _PyMethodDescr_FastCallKeywords
-      25: 0x00000000005dcb58
-      24: 0x00000000005dc83f
-      23: 0x00000000004ba127
-      22: _PyEval_EvalFrameDefault
-      21: _PyFunction_FastCallKeywords
-      20: _PyEval_EvalFrameDefault
-      19: _PyFunction_FastCall      [('tile_f', [-1, 32, 4, 4]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 16, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1758580
-    No: 14  GFLOPS: 0.00/24.56      result: Traceback (most recent call last):
-      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 738, in __call__
-        yield remote, remote.load_module(os.path.split(build_result.filename)[1])
-      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 702, in run_through_rpc
-        costs = time_f(*args).results
-      File "/workspace/python/tvm/runtime/module.py", line 357, in evaluator
-        blob = feval(*args)
+      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:1731
+      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:1671
+      19: run<>
+            at ../include/tvm/runtime/packed_func.h:1631
+      18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1631
+      17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1631
+      16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1631
+      15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1631
+      14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1646
+      13: operator()
+            at ../src/driver/driver_api.cc:388
+      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:374
+      11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+            at ../src/driver/driver_api.cc:269
+      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:453
+      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:1750
+      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:1694
+      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:1618
+      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 871, 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, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 32]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,925278
+    No: 13  GFLOPS: 0.00/11.64      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, 32, 2, 2]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 32]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1884369
+    No: 14  GFLOPS: 0.00/11.64      result: Traceback (most recent call last):
+      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
+        func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
+      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
+        func = build(s, args, target_host=task.target_host, 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 262, in tvm._ffi._cy3.core.FuncCall
-      File "tvm/_ffi/_cython/./packed_func.pxi", line 251, in tvm._ffi._cy3.core.FuncCall3
+      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):
-      4: TVMFuncCall
+      24: TVMFuncCall
             at ../src/runtime/c_runtime_api.cc:477
-      3: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
-            at ../include/tvm/runtime/packed_func.h:1217
-      2: tvm::runtime::RPCWrappedFunc::operator()(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
-            at ../src/runtime/rpc/rpc_module.cc:129
-      1: tvm::runtime::RPCClientSession::CallFunc(void*, TVMValue const*, int const*, int, std::function<void (tvm::runtime::TVMArgs)> const&)
-            at ../src/runtime/rpc/rpc_endpoint.cc:1012
-      0: tvm::runtime::RPCEndpoint::CallFunc(void*, TVMValue const*, int const*, int, std::function<void (tvm::runtime::TVMArgs)>)
-            at ../src/runtime/rpc/rpc_endpoint.cc:804
-      File "../src/runtime/rpc/rpc_endpoint.cc", line 804
-    TVMError: 
-    ---------------------------------------------------------------
-    An error occurred during the execution of TVM.
-    For more information, please see: https://tvm.apache.org/docs/errors.html
-    ---------------------------------------------------------------
-      Check failed: (code == RPCCode::kReturn) is false: code=kShutdown
-
-    During handling of the above exception, another exception occurred:
+      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:1731
+      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:1671
+      19: run<>
+            at ../include/tvm/runtime/packed_func.h:1631
+      18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1631
+      17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1631
+      16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1631
+      15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1631
+      14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1646
+      13: operator()
+            at ../src/driver/driver_api.cc:388
+      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:374
+      11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+            at ../src/driver/driver_api.cc:269
+      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:453
+      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:1750
+      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:1694
+      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:1618
+      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 871, 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):
-      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 702, in run_through_rpc
-        costs = time_f(*args).results
-      File "/usr/lib/python3.7/contextlib.py", line 130, in __exit__
-        self.gen.throw(type, value, traceback)
-      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 742, in __call__
-        remote.remove(build_result.filename)
-      File "/workspace/python/tvm/rpc/client.py", line 144, in remove
-        self._remote_funcs["remove"] = self.get_function("tvm.rpc.server.remove")
-      File "/workspace/python/tvm/rpc/client.py", line 72, in get_function
-        return self._sess.get_function(name)
-      File "/workspace/python/tvm/runtime/module.py", line 171, in get_function
-        self.handle, c_str(name), ctypes.c_int(query_imports), ctypes.byref(ret_handle)
-      File "/workspace/python/tvm/_ffi/base.py", line 348, in check_call
-        raise get_last_ffi_error()
+      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:1731
+      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:1671
+      19: run<>
+            at ../include/tvm/runtime/packed_func.h:1631
+      18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1631
+      17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1631
+      16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1631
+      15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1631
+      14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1646
+      13: operator()
+            at ../src/driver/driver_api.cc:388
+      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:374
+      11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+            at ../src/driver/driver_api.cc:269
+      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:453
+      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:1750
+      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:1694
+      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:1618
+      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 871, 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, 16, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 128, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9900475
+    No: 15  GFLOPS: 0.00/11.64      result: Traceback (most recent call last):
+      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
+        func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
+      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
+        func = build(s, args, target_host=task.target_host, 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):
-      52: 0xffffffffffffffff
-      51: _start
-      50: __libc_start_main
-      49: _Py_UnixMain
-      48: 0x0000000000650da0
-      47: 0x0000000000650afa
-      46: _PyFunction_FastCallDict
-      45: _PyEval_EvalCodeWithName
-      44: _PyEval_EvalFrameDefault
-      43: _PyFunction_FastCallKeywords
-      42: _PyEval_EvalCodeWithName
-      41: _PyEval_EvalFrameDefault
-      40: _PyMethodDef_RawFastCallKeywords
-      39: 0x0000000000546369
-      38: _PyEval_EvalCodeWithName
-      37: _PyEval_EvalFrameDefault
-      36: _PyFunction_FastCallKeywords
-      35: _PyEval_EvalCodeWithName
-      34: _PyEval_EvalFrameDefault
-      33: _PyFunction_FastCallDict
-      32: _PyEval_EvalCodeWithName
-      31: _PyEval_EvalFrameDefault
-      30: _PyObject_FastCallDict
-      29: 0x00000000004c06e1
-      28: _PyFunction_FastCallDict
-      27: _PyEval_EvalFrameDefault
-      26: _PyMethodDescr_FastCallKeywords
-      25: 0x00000000005dcb58
-      24: 0x00000000005dc83f
-      23: 0x00000000004ba127
-      22: _PyEval_EvalFrameDefault
-      21: _PyFunction_FastCallKeywords
-      20: _PyEval_EvalFrameDefault
-      19: _PyFunction_FastCallKeywords
-      18: _PyEval_EvalFrameDefault
-      17: _PyFunction_FastCallKeywords
-      16: _PyEval_EvalCodeWithName
-      15: _PyEval_EvalFrameDefault
-      14: 0x0000000000537c30
-      13: _PyObject_FastCallKeywords
-      12: 0x00007fcfb46e6fa2
-      11: _ctypes_callproc
-      10: ffi_call
-      9: ffi_call_unix64
-      8: TVMModGetFunction
-            at ../src/runtime/c_runtime_api.cc:408
-      7: tvm::runtime::ModuleNode::GetFunction(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, bool)
-            at ../src/runtime/module.cc:66
-      6: tvm::runtime::RPCModuleNode::GetFunction(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, tvm::runtime::ObjectPtr<tvm::runtime::Object> const&)
-            at ../src/runtime/rpc/rpc_module.cc:185
-      5: tvm::runtime::RPCClientSession::GetFunction(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)
-            at ../src/runtime/rpc/rpc_endpoint.cc:1007
-      4: tvm::runtime::TVMRetValue tvm::runtime::RPCEndpoint::SysCallRemote<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&>(tvm::runtime::RPCCode, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)
-            at ../src/runtime/rpc/rpc_endpoint.h:223
-      3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()<int, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&>(int&&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) const
+      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:1731
+      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:1671
+      19: run<>
+            at ../include/tvm/runtime/packed_func.h:1631
+      18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1631
+      17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1631
+      16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1631
+      15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1631
+      14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1646
+      13: operator()
+            at ../src/driver/driver_api.cc:388
+      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:374
+      11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+            at ../src/driver/driver_api.cc:269
+      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:453
+      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:1750
+      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:1694
+      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:1618
       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/rpc/rpc_endpoint.cc:684
-      File "../src/runtime/rpc/rpc_endpoint.cc", line 684
-    TVMError: 
-    ---------------------------------------------------------------
-    An error occurred during the execution of TVM.
-    For more information, please see: https://tvm.apache.org/docs/errors.html
-    ---------------------------------------------------------------
-      Check failed: (code == RPCCode::kReturn) is false: code=1
+            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 871, 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):
-      52: 0xffffffffffffffff
-      51: _start
-      50: __libc_start_main
-      49: _Py_UnixMain
-      48: 0x0000000000650da0
-      47: 0x0000000000650afa
-      46: _PyFunction_FastCallDict
-      45: _PyEval_EvalCodeWithName
-      44: _PyEval_EvalFrameDefault
-      43: _PyFunction_FastCallKeywords
-      42: _PyEval_EvalCodeWithName
-      41: _PyEval_EvalFrameDefault
-      40: _PyMethodDef_RawFastCallKeywords
-      39: 0x0000000000546369
-      38: _PyEval_EvalCodeWithName
-      37: _PyEval_EvalFrameDefault
-      36: _PyFunction_FastCallKeywords
-      35: _PyEval_EvalCodeWithName
-      34: _PyEval_EvalFrameDefault
-      33: _PyFunction_FastCallDict
-      32: _PyEval_EvalCodeWithName
-      31: _PyEval_EvalFrameDefault
-      30: _PyObject_FastCallDict
-      29: 0x00000000004c06e1
-      28: _PyFunction_FastCallDict
-      27: _PyEval_EvalFrameDefault
-      26: _PyMethodDescr_FastCallKeywords
-      25: 0x00000000005dcb58
-      24: 0x00000000005dc83f
-      23: 0x00000000004ba127
-      22: _PyEval_EvalFrameDefault
-      21: _PyFunction_FastCallKeywords
-      20: _PyEval_EvalFrameDefault
-      19: _PyFunction_FastCall      [('tile_f', [-1, 64, 1, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5659946
-    No: 15  GFLOPS: 6.77/24.56      result: MeasureResult(costs=(0.034179073250000004,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.2322049140930176, timestamp=1668475021.8653607)       [('tile_f', [-1, 2, 1, 8]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 32, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1566537
-    No: 16  GFLOPS: 0.00/24.56      result: 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:1731
+      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:1671
+      19: run<>
+            at ../include/tvm/runtime/packed_func.h:1631
+      18: run<tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1631
+      17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1631
+      16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1631
+      15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1631
+      14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
+            at ../include/tvm/runtime/packed_func.h:1646
+      13: operator()
+            at ../src/driver/driver_api.cc:388
+      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:374
+      11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
+            at ../src/driver/driver_api.cc:269
+      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:453
+      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:1750
+      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:1694
+      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:1618
+      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 871, 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, 2, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 128]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,3465890
+    No: 16  GFLOPS: 0.00/11.64      result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -1926,8 +2005,10 @@ 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 871, 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, 1, 64]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 16, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1049823
-    No: 17  GFLOPS: 0.00/24.56      result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 1, 32, 16]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 64, 8]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2636444
+    No: 17  GFLOPS: 232.51/232.51   result: MeasureResult(costs=(0.0009956469262295082,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.6534688472747803, timestamp=1668481246.2235832)      [('tile_f', [-1, 2, 32, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7065171
+    No: 18  GFLOPS: 189.37/232.51   result: MeasureResult(costs=(0.0012224727938931297,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.5868277549743652, timestamp=1668481247.1970925)      [('tile_f', [-1, 8, 8, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 2, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,680710
+    No: 19  GFLOPS: 0.00/232.51     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -2049,8 +2130,8 @@ 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 871, 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, 2, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 1, 512]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9483830
-    No: 18  GFLOPS: 0.00/24.56      result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 2, 64, 4]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 4, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8562754
+    No: 20  GFLOPS: 0.00/232.51     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 588, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 540, in _build_func_common
@@ -2172,9 +2253,7 @@ 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 871, 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, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 16, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9375323
-    No: 19  GFLOPS: 290.01/290.01   result: MeasureResult(costs=(0.0007982536534653466,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.7852540016174316, timestamp=1668475023.8795164)      [('tile_f', [-1, 1, 8, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 32, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1761567
-    No: 20  GFLOPS: 38.19/290.01    result: MeasureResult(costs=(0.006061600176470588,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.1658580303192139, timestamp=1668475024.541543)        [('tile_f', [-1, 1, 4, 32]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 16]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,315674
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 2, 8, 32]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 16, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2181938
 
 
 
@@ -2229,9 +2308,9 @@ and measure running time.
     Finish loading 20 records
 
     Best config:
-    [('tile_f', [-1, 1, 8, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 32, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1761567
+    [('tile_f', [-1, 2, 32, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 8]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7065171
     Finish loading 20 records
-    Time cost of this operator: 0.001224
+    Time cost of this operator: 0.001205
 
 
 
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 c1a0f5c6d1..c763f1f4b7 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
@@ -327,10 +327,10 @@ 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  312.0     98.705   (1, 2, 10, 10, 3)  2       1        [312.0]           
-    tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       3.094     0.979    (1, 6, 10, 10)     1       1        [3.094]           
-    tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         1.001     0.317    (1, 1, 10, 10, 3)  1       1        [1.001]           
-    Total_time                                    -                                             316.095   -        -                  -       -        -                 
+    tvmgen_default_fused_nn_contrib_conv2d_NCHWc  tvmgen_default_fused_nn_contrib_conv2d_NCHWc  310.6     98.728   (1, 2, 10, 10, 3)  2       1        [310.6]           
+    tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       3.021     0.96     (1, 6, 10, 10)     1       1        [3.021]           
+    tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.979     0.311    (1, 1, 10, 10, 3)  1       1        [0.979]           
+    Total_time                                    -                                             314.6     -        -                  -       -        -                 
 
 
 
@@ -394,10 +394,10 @@ 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  100.2     97.322   (1, 6, 10, 10, 1)  2       1        [100.2]           
-    tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       1.792     1.74     (1, 6, 10, 10)     1       1        [1.792]           
-    tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.966     0.938    (1, 1, 10, 10, 3)  1       1        [0.966]           
-    Total_time                                    -                                             102.957   -        -                  -       -        -                 
+    tvmgen_default_fused_nn_contrib_conv2d_NCHWc  tvmgen_default_fused_nn_contrib_conv2d_NCHWc  137.1     98.095   (1, 6, 10, 10, 1)  2       1        [137.1]           
+    tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       1.818     1.301    (1, 6, 10, 10)     1       1        [1.818]           
+    tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.844     0.604    (1, 3, 10, 10, 1)  1       1        [0.844]           
+    Total_time                                    -                                             139.762   -        -                  -       -        -                 
 
 
 
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 e6394fb7fe..661be5f9a3 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
@@ -109,7 +109,7 @@ 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]
      2%|1         | 64.0k/3.42M [00:00<00:06, 537kB/s]
      9%|8         | 304k/3.42M [00:00<00:02, 1.41MB/s]
     37%|###7      | 1.27M/3.42M [00:00<00:00, 4.58MB/s]
    100%|##########| 3.42M/3.42M [00:00<00:00, 8.25MB/s]
+
      0%|          | 0.00/3.42M [00:00<?, ?B/s]
     37%|###7      | 1.28M/3.42M [00:00<00:00, 13.2MB/s]
    100%|##########| 3.42M/3.42M [00:00<00:00, 19.5MB/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.
@@ -314,7 +314,7 @@ Look up prediction top 1 index in 1000 class synset.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  4.460 seconds)
+   **Total running time of the script:** ( 1 minutes  4.661 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 e3bf580733..530f58d35c 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
@@ -225,7 +225,7 @@ take about **2 minutes** to download the Stanford Cars, while COCO 2017 validati
  .. code-block:: none
 
 
-    '/tmp/tmpkpo3z5js/images/random'
+    '/tmp/tmp9x_81ijo/images/random'
 
 
 
@@ -316,7 +316,7 @@ 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], [1.0, 0.0], [0.0, 1.0], [1.0, 0.0], [1.0, 0.0], [1.0, 0.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [1.0, 0.0]
+   :alt: [0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0]
    :srcset: /how_to/work_with_microtvm/images/sphx_glr_micro_train_001.png
    :class: sphx-glr-single-img
 
@@ -325,8 +325,8 @@ objects to other stuff? We can display some examples from our datasets using ``m
 
  .. code-block:: none
 
-    /tmp/tmpkpo3z5js/images/target contains 8144 images
-    /tmp/tmpkpo3z5js/images/random contains 5000 images
+    /tmp/tmp9x_81ijo/images/target contains 8144 images
+    /tmp/tmp9x_81ijo/images/random contains 5000 images
 
 
 
@@ -501,13 +501,13 @@ the time on our validation set).
  .. code-block:: none
 
     Epoch 1/3
-    328/328 - 47s - loss: 0.2287 - accuracy: 0.9221 - val_loss: 0.1266 - val_accuracy: 0.9502 - 47s/epoch - 145ms/step
+    328/328 - 47s - loss: 0.2268 - accuracy: 0.9215 - val_loss: 0.1660 - val_accuracy: 0.9452 - 47s/epoch - 144ms/step
     Epoch 2/3
-    328/328 - 44s - loss: 0.1137 - accuracy: 0.9588 - val_loss: 0.0845 - val_accuracy: 0.9732 - 44s/epoch - 134ms/step
+    328/328 - 43s - loss: 0.1011 - accuracy: 0.9637 - val_loss: 0.1315 - val_accuracy: 0.9524 - 43s/epoch - 133ms/step
     Epoch 3/3
-    328/328 - 44s - loss: 0.0703 - accuracy: 0.9748 - val_loss: 0.0871 - val_accuracy: 0.9698 - 44s/epoch - 133ms/step
+    328/328 - 43s - loss: 0.0697 - accuracy: 0.9728 - val_loss: 0.1259 - val_accuracy: 0.9573 - 43s/epoch - 132ms/step
 
-    <keras.callbacks.History object at 0x7f6f10c5db90>
+    <keras.callbacks.History object at 0x7f8789ef85d0>
 
 
 
@@ -864,7 +864,7 @@ 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:** ( 4 minutes  35.858 seconds)
+   **Total running time of the script:** ( 4 minutes  21.227 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 665b01343c..0e717c0246 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
 =================
-**06:43.515** total execution time for **how_to_work_with_microtvm** files:
+**06:29.210** total execution time for **how_to_work_with_microtvm** files:
 
 +---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_train.py` (``micro_train.py``)               | 04:35.858 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_train.py` (``micro_train.py``)               | 04:21.227 | 0.0 MB |
 +---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_pytorch.py` (``micro_pytorch.py``)           | 01:04.460 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_pytorch.py` (``micro_pytorch.py``)           | 01:04.661 | 0.0 MB |
 +---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``)         | 00:50.784 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``)         | 00:51.140 | 0.0 MB |
 +---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_aot.py` (``micro_aot.py``)                   | 00:08.577 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_aot.py` (``micro_aot.py``)                   | 00:08.327 | 0.0 MB |
 +---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``)             | 00:03.835 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``)             | 00:03.853 | 0.0 MB |
 +---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_reference_vm.py` (``micro_reference_vm.py``) | 00:00.001 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_reference_vm.py` (``micro_reference_vm.py``) | 00:00.002 | 0.0 MB |
 +---------------------------------------------------------------------------------------------+-----------+--------+
 | :ref:`sphx_glr_how_to_work_with_microtvm_micro_ethosu.py` (``micro_ethosu.py``)             | 00:00.001 | 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 273a06a6e5..b59c8a83a1 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
 =================
-**00:45.224** total execution time for **how_to_work_with_relay** files:
+**00:44.332** 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``) | 00:32.617 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_using_pipeline_executor.py` (``using_pipeline_executor.py``) | 00:32.582 | 0.0 MB |
 +----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``)           | 00:10.366 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``)           | 00:10.216 | 0.0 MB |
 +----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``)                             | 00:02.234 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``)                             | 00:01.528 | 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 |
+| :ref:`sphx_glr_how_to_work_with_relay_using_relay_viz.py` (``using_relay_viz.py``)                 | 00:00.006 | 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 5966281da8..566ea4e9e9 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
@@ -261,7 +261,7 @@ The following example customizes CUDA lowering rule for :code:`exp`.
  .. code-block:: none
 
 
-    <function my_cuda_math_rule at 0x7f6f0c303050>
+    <function my_cuda_math_rule at 0x7f8785916a70>
 
 
 
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 80fd808716..4383edfafa 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
 =================
-**00:06.555** total execution time for **how_to_work_with_schedules** files:
+**00:06.518** total execution time for **how_to_work_with_schedules** files:
 
 +------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``)                 | 00:04.116 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``)                 | 00:04.113 | 0.0 MB |
 +------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``)                     | 00:01.082 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``)                     | 00:01.053 | 0.0 MB |
 +------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``)                     | 00:00.579 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``)                     | 00:00.574 | 0.0 MB |
 +------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``)                               | 00:00.562 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``)                               | 00:00.561 | 0.0 MB |
 +------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``)                     | 00:00.116 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``)                     | 00:00.118 | 0.0 MB |
 +------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``) | 00:00.049 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``) | 00:00.050 | 0.0 MB |
 +------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_tedd.py` (``tedd.py``)                               | 00:00.030 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tedd.py` (``tedd.py``)                               | 00:00.029 | 0.0 MB |
 +------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_tuple_inputs.py` (``tuple_inputs.py``)               | 00:00.019 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_tuple_inputs.py` (``tuple_inputs.py``)               | 00:00.020 | 0.0 MB |
 +------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt b/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt
index 0fc4b0f654..458bb58198 100644
--- a/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt
+++ b/docs/_sources/how_to/work_with_schedules/tensorize.rst.txt
@@ -347,7 +347,7 @@ The importing needs to happen before the tensorized GEMV being executed.
                  C: Buffer(C_2: Pointer(float32), float32, [524288], [])}
       buffer_map = {A_1: A, B_1: B, C_1: C}
       preflattened_buffer_map = {A_1: A_3: Buffer(A_2, float32, [1024, 64], []), B_1: B_3: Buffer(B_2, float32, [512, 64], []), C_1: C_3: Buffer(C_2, float32, [1024, 512], [])} {
-      attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmpy_fnrd3r/input0.cc'\nsource_filename = \"/tmp/tmpy_fnrd3r/input0.cc\"\ntarget datalayout = \"e-m:e-i64:64-f80:128-n8:16:32:64-S128\"\ntarget triple = \"x86_64-pc-linux-gnu\"\n\n; Function Attrs: noinline nounwind optnone uwtable\ndefine dso_local i32 @gemv_update(float*, float*, float*, i32, i32, i32) #0 {\n  %7 = alloca float*, align 8\n  %8 = alloca float*, align 8\n  %9 = alloca floa [...]
+      attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmp7g8rvtkr/input0.cc'\nsource_filename = \"/tmp/tmp7g8rvtkr/input0.cc\"\ntarget datalayout = \"e-m:e-i64:64-f80:128-n8:16:32:64-S128\"\ntarget triple = \"x86_64-pc-linux-gnu\"\n\n; Function Attrs: noinline nounwind optnone uwtable\ndefine dso_local i32 @gemv_update(float*, float*, float*, i32, i32, i32) #0 {\n  %7 = alloca float*, align 8\n  %8 = alloca float*, align 8\n  %9 = alloca floa [...]
       for (i, 0, 1024) {
         for (j.outer: int32, 0, 32) {
           @tir.call_extern("gemv_update", @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), C_2, ((i*512) + (j.outer*16)), 16, 2, dtype=handle), @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), A_2, (i*64), 64, 1, dtype=handle), @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), B_2, (j.outer*1024), 1024, 1, dtype=handle), 16, 64, 64, dtype=int32)
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 bf024247d1..8d706c4ef7 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
 =================
-**00:26.919** total execution time for **topic_vta_tutorials_autotvm** files:
+**00:26.929** total execution time for **topic_vta_tutorials_autotvm** files:
 
 +---------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``) | 00:26.912 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``) | 00:26.922 | 0.0 MB |
 +---------------------------------------------------------------------------------------+-----------+--------+
 | :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_alu_vta.py` (``tune_alu_vta.py``)     | 00:00.006 | 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 d249c03d4f..20985a6a06 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
@@ -289,7 +289,7 @@ 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 29.79s!
+    resnet18_v1 inference graph built in 30.06s!
 
 
 
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 199e348350..1f09a273f3 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
@@ -333,7 +333,7 @@ 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 20.01s!
+    yolov3-tiny inference graph built in 20.30s!
 
 
 
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 a9a2e8b763..938e6d929e 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
 =================
-**01:41.564** total execution time for **topic_vta_tutorials_frontend** files:
+**01:42.643** total execution time for **topic_vta_tutorials_frontend** files:
 
 +------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``)           | 00:51.897 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``)           | 00:52.575 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``) | 00:49.667 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``) | 00:50.068 | 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 b7990df781..77ca9dde30 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
 =================
-**00:03.131** total execution time for **topic_vta_tutorials_optimize** files:
+**00:03.112** total execution time for **topic_vta_tutorials_optimize** files:
 
 +--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``)         | 00:02.681 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``)         | 00:02.662 | 0.0 MB |
 +--------------------------------------------------------------------------------------------------+-----------+--------+
 | :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``) | 00:00.450 | 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 a5cf47be91..9c7b50fa62 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
 =================
-**00:00.801** total execution time for **topic_vta_tutorials** files:
+**00:00.791** total execution time for **topic_vta_tutorials** files:
 
 +---------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``) | 00:00.429 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``) | 00:00.420 | 0.0 MB |
 +---------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``) | 00:00.372 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``) | 00:00.370 | 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 8f8ef6dd15..169ebfa914 100644
--- a/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
+++ b/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
@@ -203,6 +203,13 @@ trials, we can load the best schedule from the log file and apply it.
 
 
 
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+
+    .T
+
 
 
 
@@ -326,7 +333,7 @@ We build the binary and check its correctness and performance.
 
  .. code-block:: none
 
-    Execution time of this operator: 95.070 ms
+    Execution time of this operator: 96.686 ms
 
 
 
@@ -444,7 +451,7 @@ operations.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  30.940 seconds)
+   **Total running time of the script:** ( 1 minutes  42.599 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 564440b7cd..3bb72724f0 100644
--- a/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt
+++ b/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt
@@ -450,16 +450,16 @@ reduce variance, we take 5 measurements and average them.
     waiting for device...
     device available
     Get devices for measurement successfully!
-    No: 1   GFLOPS: 1.58/1.58       result: MeasureResult(costs=(0.169607917,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.8739066123962402, timestamp=1668473609.7446082)        [('tile_y', [-1, 32]), ('tile_x', [-1, 4])],None,25
-    No: 2   GFLOPS: 9.28/9.28       result: MeasureResult(costs=(0.028912128200000005,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.7243156433105469, timestamp=1668473610.4245136)       [('tile_y', [-1, 512]), ('tile_x', [-1, 128])],None,79
-    No: 3   GFLOPS: 2.30/9.28       result: MeasureResult(costs=(0.11654239640000001,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.0009708404541016, timestamp=1668473613.2164934)        [('tile_y', [-1, 1]), ('tile_x', [-1, 16])],None,40
-    No: 4   GFLOPS: 12.58/12.58     result: MeasureResult(costs=(0.0213367548,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6361486911773682, timestamp=1668473614.4972048)       [('tile_y', [-1, 64]), ('tile_x', [-1, 128])],None,76
-    No: 5   GFLOPS: 10.59/12.58     result: MeasureResult(costs=(0.025336678600000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5550379753112793, timestamp=1668473615.9628344)       [('tile_y', [-1, 1]), ('tile_x', [-1, 256])],None,80
-    No: 6   GFLOPS: 13.69/13.69     result: MeasureResult(costs=(0.019606959599999998,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.48113441467285156, timestamp=1668473616.440613)       [('tile_y', [-1, 128]), ('tile_x', [-1, 64])],None,67
-    No: 7   GFLOPS: 2.51/13.69      result: MeasureResult(costs=(0.10688991080000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.8739430904388428, timestamp=1668473618.3178763)        [('tile_y', [-1, 8]), ('tile_x', [-1, 4])],None,23
-    No: 8   GFLOPS: 4.16/13.69      result: MeasureResult(costs=(0.0645883018,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.2422730922698975, timestamp=1668473619.5611193)       [('tile_y', [-1, 16]), ('tile_x', [-1, 16])],None,44
-    No: 9   GFLOPS: 9.08/13.69      result: MeasureResult(costs=(0.029561042,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6742880344390869, timestamp=1668473620.3487682)        [('tile_y', [-1, 1]), ('tile_x', [-1, 128])],None,70
-    No: 10  GFLOPS: 11.65/13.69     result: MeasureResult(costs=(0.0230363428,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.528548002243042, timestamp=1668473620.8892539)        [('tile_y', [-1, 16]), ('tile_x', [-1, 256])],None,84
+    No: 1   GFLOPS: 0.51/0.51       result: MeasureResult(costs=(0.5258266778,), error_no=MeasureErrorNo.NO_ERROR, all_cost=8.5500328540802, timestamp=1668479848.6790347)  [('tile_y', [-1, 256]), ('tile_x', [-1, 1])],None,8
+    No: 2   GFLOPS: 11.67/11.67     result: MeasureResult(costs=(0.0229983434,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5694522857666016, timestamp=1668479849.2623055)       [('tile_y', [-1, 2]), ('tile_x', [-1, 256])],None,81
+    No: 3   GFLOPS: 11.76/11.76     result: MeasureResult(costs=(0.022829464,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5173089504241943, timestamp=1668479850.55224)  [('tile_y', [-1, 256]), ('tile_x', [-1, 256])],None,88
+    No: 4   GFLOPS: 7.39/11.76      result: MeasureResult(costs=(0.036339880000000005,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.7135937213897705, timestamp=1668479852.0477846)       [('tile_y', [-1, 512]), ('tile_x', [-1, 32])],None,59
+    No: 5   GFLOPS: 2.70/11.76      result: MeasureResult(costs=(0.0992417416,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.7401914596557617, timestamp=1668479854.8115833)       [('tile_y', [-1, 4]), ('tile_x', [-1, 4])],None,22
+    No: 6   GFLOPS: 8.73/11.76      result: MeasureResult(costs=(0.030753971999999997,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.8017129898071289, timestamp=1668479855.4690278)       [('tile_y', [-1, 16]), ('tile_x', [-1, 64])],None,64
+    No: 7   GFLOPS: 11.80/11.80     result: MeasureResult(costs=(0.022747619599999998,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5359385013580322, timestamp=1668479856.0026648)       [('tile_y', [-1, 32]), ('tile_x', [-1, 256])],None,85
+    No: 8   GFLOPS: 9.91/11.80      result: MeasureResult(costs=(0.027076459999999997,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.7049639225006104, timestamp=1668479856.6463385)       [('tile_y', [-1, 1]), ('tile_x', [-1, 128])],None,70
+    No: 9   GFLOPS: 12.69/12.69     result: MeasureResult(costs=(0.021145623800000003,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.4574155807495117, timestamp=1668479857.2369988)       [('tile_y', [-1, 32]), ('tile_x', [-1, 512])],None,95
+    No: 10  GFLOPS: 12.29/12.69     result: MeasureResult(costs=(0.0218373526,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.4883308410644531, timestamp=1668479857.7563374)       [('tile_y', [-1, 8]), ('tile_x', [-1, 256])],None,83
 
 
 
diff --git a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
index 9f1e9404cd..e63b68d6db 100644
--- a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
+++ b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
@@ -320,7 +320,7 @@ standard deviation.
 
  .. code-block:: none
 
-    {'mean': 517.1077569899899, 'median': 516.9921277999947, 'std': 1.3162579716279146}
+    {'mean': 510.4595590599991, 'median': 509.4302212499997, 'std': 3.3041499135301597}
 
 
 
@@ -554,31 +554,31 @@ 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:   21.50/  21.50 GFLOPS | Progress: (4/20) | 8.66 s
    [Task  1/25]  Current/Best:    5.70/  21.50 GFLOPS | Progress: (8/20) | 11.62 s
    [Task  1/25]  Current/Best:   23.34/  23.34 GFLOPS | Progress: (12/20) | 15.11 s
    [Task  1/25]  Current/Best:   11.92/  23.34 GFLOPS | Progress: (16/20) | 17.45 s
    [Task  1/25]  Current/Best:   12.54/  23.34 GFLOPS | Progress: (20/20) | 20.20 s Done.
-
    [Task  2/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  2/25]  Current/Best:    6.58/  12.38 GFLOPS | Progress: (4/20) | 3.09 s
    [Task  2/25]  Current/Best:    2.15/  17.86 GFLOPS | Progress: (8/20) | 4.84 s
    [Task  2/25]  Current/Best:    6.11/  20.20 GFLOPS | Progress: (12/20) | 6.09 s
    [Task  2/25]  Current/Best:   15.42/  20.20 GFLOPS | Progress: (16/20) | 7.42 s
    [Task  2/25]  Current/Best:   10.15/  20.20 GFLOPS | Progress: (20/20) | 9.04 s Done.
-
    [Task  3/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  3/25]  Current/Best:   10.91/  18.23 GFLOPS | Progress: (4/20) | 3.53 s
    [Task  3/25]  Current/Best:    4.54/  18.23 GFLOPS | Progress: (8/20) | 5.59 s
    [Task  3/25]  Current/Best:   11.77/  18.23 GFLOPS | Progress: (12/20) | 7.76 s
    [Task  3/25]  Current/Best:    6.57/  22.34 GFLOPS | Progress: (16/20) | 9.54 s
    [Task  3/25]  Current/Best:   22.99/  22.99 GFLOPS | Progress: (20/20) | 11.57 s Done.
-
    [Task  4/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  4/25]  Current/Best:    5.93/   9.78 GFLOPS | Progress: (4/20) | 3.43 s
    [Task  4/25]  Current/Best:   15.49/  20.30 GFLOPS | Progress: (8/20) | 4.78 s
    [Task  4/25]  Current/Best:   12.71/  20.30 GFLOPS | Progress: (12/20) | 6.54 s
    [Task  4/25]  Current/Best:   12.28/  20.30 GFLOPS | Progress: (16/20) | 8.62 s
    [Task  4/25]  Current/Best:   10.79/  20.30 GFLOPS | Progress: (20/20) | 19.63 s
    [Task  5/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
-
    [Task  5/25]  Current/Best:    8.78/  22.55 GFLOPS | Progress: (4/20) | 3.60 s
    [Task  5/25]  Current/Best:   16.45/  22.55 GFLOPS | Progress: (8/20) | 5.00 s
    [Task  5/25]  Current/Best:   19.75/  22.55 GFLOPS | Progress: (12/20) | 6.75 s
    [Task  5/25]  Current/Best:   20.60/  22.55 GFLOPS | Progress: (16/20) | 8.36 s
    [Task  5/25]  Current/Best:    9.46/  22.55 GFLOPS | Progress: (20/20) | 10.39 s Done.
-
    [Task  6/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  6/25]  Current/Best:   10.04/  18.64 GFLOPS | Progress: (4/20) | 5.29 s
    [Task  6/25]  Current/Best:   22.65/  22.65 GFLOPS | Progress: (8/20) | 9.32 s
    [Task  6/25]  Current/Best:    1.46/  22.65 GFLOPS | Progress: (12/20) | 12.60 s
    [Task  6/25]  Current/Best:   15.57/  22.65 GFLOPS | Progress: (16/20) | 15.05 s
    [Task  6/25]  Current/Best:   17.95/  22.65 GFLOPS | Progress: (20/20) | 16.99 s Done.
-
    [Task  7/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  7/25]  Current/Best:    7.49/  18.95 GFLOPS | Progress: (4/20) | 3.50 s
    [Task  7/25]  Current/Best:   11.41/  18.95 GFLOPS | Progress: (8/20) | 7.24 s
    [Task  7/25]  Current/Best:   17.22/  18.95 GFLOPS | Progress: (12/20) | 9.20 s
    [Task  7/25]  Current/Best:    1.58/  18.95 GFLOPS | Progress: (16/20) | 13.60 s
    [Task  7/25]  Current/Best:    6.26/  18.95 GFLOPS | Progress: (20/20) | 16.71 s Done.
-
    [Task  8/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  8/25]  Current/Best:   12.83/  16.56 GFLOPS | Progress: (4/20) | 4.11 s
    [Task  8/25]  Current/Best:    8.48/  16.56 GFLOPS | Progress: (8/20) | 6.78 s
    [Task  8/25]  Current/Best:    3.27/  16.56 GFLOPS | Progress: (12/20) | 12.79 s
    [Task  8/25]  Current/Best:   16.12/  16.56 GFLOPS | Progress: (16/20) | 19.38 s
    [Task  8/25]  Current/Best:    3.44/  16.56 GFLOPS | Progress: (20/20) | 22.67 s Done.
-
    [Task  9/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  9/25]  Current/Best:   12.46/  20.88 GFLOPS | Progress: (4/20) | 2.77 s
    [Task  9/25]  Current/Best:   12.62/  20.88 GFLOPS | Progress: (8/20) | 4.25 s
    [Task  9/25]  Current/Best:    8.35/  20.88 GFLOPS | Progress: (12/20) | 8.81 s
    [Task  9/25]  Current/Best:    6.59/  20.88 GFLOPS | Progress: (16/20) | 19.70 s
    [Task  9/25]  Current/Best:   15.70/  20.88 GFLOPS | Progress: (20/20) | 22.33 s
    [Task 10/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 10/25]  Current/Best:   17.55/  17.55 GFLOPS | Progress: (4/20) | 3.36 s
    [Task 10/25]  Current/Best:   11.88/  17.55 GFLOPS | Progress: (8/20) | 6.18 s
    [Task 10/25]  Current/Best:   11.52/  17.55 GFLOPS | Progress: (12/20) | 7.67 s
    [Task 10/25]  Current/Best:   10.39/  17.55 GFLOPS | Progress: (16/20) | 10.13 s
    [Task 10/25]  Current/Best:    1.61/  17.55 GFLOPS | Progress: (20/20) 
 | 12.34 s Done.
-
    [Task 11/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 11/25]  Current/Best:   22.13/  24.00 GFLOPS | Progress: (4/20) | 3.75 s
    [Task 11/25]  Current/Best:   12.05/  24.00 GFLOPS | Progress: (8/20) | 6.45 s
    [Task 11/25]  Current/Best:    9.81/  24.00 GFLOPS | Progress: (12/20) | 8.66 s
    [Task 11/25]  Current/Best:   13.03/  24.00 GFLOPS | Progress: (16/20) | 10.78 s
    [Task 11/25]  Current/Best:   18.50/  24.00 GFLOPS | Progress: (20/20) | 13.63 s Done.
-
    [Task 12/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 12/25]  Current/Best:    6.76/  20.68 GFLOPS | Progress: (4/20) | 3.48 s
    [Task 12/25]  Current/Best:   15.02/  20.68 GFLOPS | Progress: (8/20) | 5.13 s
    [Task 12/25]  Current/Best:   16.30/  20.68 GFLOPS | Progress: (12/20) | 7.10 s
    [Task 12/25]  Current/Best:   14.75/  20.68 GFLOPS | Progress: (16/20) | 9.51 s
    [Task 12/25]  Current/Best:   16.67/  20.68 GFLOPS | Progress: (20/20) | 11.51 s Done.
-
    [Task 13/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 13/25]  Current/Best:    1.57/   9.69 GFLOPS | Progress: (4/20) | 6.96 s
    [Task 13/25]  Current/Best:   11.73/  17.74 GFLOPS | Progress: (8/20) | 10.85 s
    [Task 13/25]  Current/Best:   18.98/  20.09 GFLOPS | Progress: (12/20) | 13.81 s
    [Task 13/25]  Current/Best:    7.14/  20.09 GFLOPS | Progress: (16/20) | 16.43 s
    [Task 13/25]  Current/Best:    4.69/  21.30 GFLOPS | Progress: (20/20) | 18.83 s Done.
-
    [Task 14/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 14/25]  Current/Best:    9.38/  17.04 GFLOPS | Progress: (4/20) | 3.52 s
    [Task 14/25]  Current/Best:   14.23/  17.04 GFLOPS | Progress: (8/20) | 6.22 s
    [Task 14/25]  Current/Best:   21.33/  21.33 GFLOPS | Progress: (12/20) | 7.77 s
    [Task 14/25]  Current/Best:   11.77/  21.33 GFLOPS | Progress: (16/20) | 11.56 s Done.
-
    [Task 14/25]  Current/Best:    8.75/  21.33 GFLOPS | Progress: (20/20) | 14.22 s Done.
-
    [Task 15/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 15/25]  Current/Best:    4.96/  17.69 GFLOPS | Progress: (4/20) | 6.10 s
    [Task 15/25]  Current/Best:   11.85/  17.69 GFLOPS | Progress: (8/20) | 8.24 s
    [Task 15/25]  Current/Best:   12.71/  19.69 GFLOPS | Progress: (12/20) | 12.49 s
    [Task 15/25]  Current/Best:   16.93/  19.69 GFLOPS | Progress: (16/20) | 13.93 s
    [Task 15/25]  Current/Best:   22.77/  22.77 GFLOPS | Progress: (20/20) | 15.63 s
    [Task 16/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 16/25]  Current/Best:   14.80/  14.80 GFLOPS | Progress: (4/20) | 4.21 s
    [Task 16/25]  Current/Best:   17.28/  18.67 GFLOPS | Progress: (8/20) | 5.55 s
    [Task 16/25]  Current/Best:   11.45/  18.67 GFLOPS | Progress: (12/20) | 8.13 s
    [Task 16/25]  Current/Best:    4.37/  18.67 GFLOPS | Progress: (16/20) | 10.35 s
    [Task 16/25]  Current/Best:   19.64/  23.22 GFLOPS | Progress: (20/20)
  | 11.56 s Done.
-
    [Task 17/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 17/25]  Current/Best:    8.53/  19.47 GFLOPS | Progress: (4/20) | 4.14 s
    [Task 17/25]  Current/Best:    6.16/  19.47 GFLOPS | Progress: (8/20) | 6.75 s
    [Task 17/25]  Current/Best:   19.70/  23.06 GFLOPS | Progress: (12/20) | 9.14 s
    [Task 17/25]  Current/Best:    6.10/  23.06 GFLOPS | Progress: (16/20) | 12.36 s
    [Task 17/25]  Current/Best:   10.13/  23.06 GFLOPS | Progress: (20/20) | 14.24 s Done.
-
    [Task 18/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 18/25]  Current/Best:   11.38/  15.02 GFLOPS | Progress: (4/20) | 5.47 s
    [Task 18/25]  Current/Best:   12.48/  18.84 GFLOPS | Progress: (8/20) | 7.55 s
    [Task 18/25]  Current/Best:   20.26/  20.26 GFLOPS | Progress: (12/20) | 9.43 s
    [Task 18/25]  Current/Best:   10.94/  20.26 GFLOPS | Progress: (16/20) | 13.81 s
    [Task 18/25]  Current/Best:   11.31/  20.26 GFLOPS | Progress: (20/20) | 16.04 s Done.
-
    [Task 19/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 19/25]  Current/Best:    4.41/  10.48 GFLOPS | Progress: (4/20) | 5.26 s
    [Task 19/25]  Current/Best:    6.63/  17.88 GFLOPS | Progress: (8/20) | 7.59 s
    [Task 19/25]  Current/Best:    9.05/  17.88 GFLOPS | Progress: (12/20) | 10.50 s
    [Task 19/25]  Current/Best:   10.49/  18.01 GFLOPS | Progress: (16/20) | 13.87 s
    [Task 19/25]  Current/Best:   10.42/  20.98 GFLOPS | Progress: (20/20) | 17.08 s Done.
-
    [Task 20/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 20/25]  Current/Best:   18.00/  18.00 GFLOPS | Progress: (4/20) | 3.85 s
    [Task 20/25]  Current/Best:    4.93/  18.92 GFLOPS | Progress: (8/20) | 7.97 s
    [Task 20/25]  Current/Best:    2.26/  18.92 GFLOPS | Progress: (12/20) | 10.59 s
    [Task 20/25]  Current/Best:   16.09/  18.92 GFLOPS | Progress: (16/20) | 13.97 s
    [Task 20/25]  Current/Best:    6.68/  18.92 GFLOPS | Progress: (20/20) | 16.09 s
    [Task 21/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 21/25]  Current/Best:    1.62/  19.59 GFLOPS | Progress: (4/20) | 3.93 s
    [Task 21/25]  Current/Best:   10.36/  19.59 GFLOPS | Progress: (8/20) | 6.21 s Done.
+
    [Task  1/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  1/25]  Current/Best:   13.77/  23.62 GFLOPS | Progress: (4/20) | 8.04 s
    [Task  1/25]  Current/Best:   14.45/  23.62 GFLOPS | Progress: (8/20) | 10.84 s
    [Task  1/25]  Current/Best:    8.66/  23.62 GFLOPS | Progress: (12/20) | 12.80 s
    [Task  1/25]  Current/Best:    9.09/  23.62 GFLOPS | Progress: (16/20) | 16.11 s
    [Task  1/25]  Current/Best:   12.47/  23.62 GFLOPS | Progress: (20/20) | 19.01 s Done.
+
    [Task  2/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  2/25]  Current/Best:   15.75/  16.13 GFLOPS | Progress: (4/20) | 2.89 s
    [Task  2/25]  Current/Best:   15.36/  16.13 GFLOPS | Progress: (8/20) | 4.21 s
    [Task  2/25]  Current/Best:   12.12/  20.99 GFLOPS | Progress: (12/20) | 6.92 s
    [Task  2/25]  Current/Best:    7.47/  20.99 GFLOPS | Progress: (16/20) | 8.36 s
    [Task  2/25]  Current/Best:    9.97/  20.99 GFLOPS | Progress: (20/20) | 11.24 s Done.
+
    [Task  3/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  3/25]  Current/Best:   16.00/  16.00 GFLOPS | Progress: (4/20) | 3.50 s
    [Task  3/25]  Current/Best:   18.67/  18.67 GFLOPS | Progress: (8/20) | 5.70 s
    [Task  3/25]  Current/Best:   12.08/  18.67 GFLOPS | Progress: (12/20) | 8.89 s
    [Task  3/25]  Current/Best:   16.40/  19.83 GFLOPS | Progress: (16/20) | 11.43 s
    [Task  3/25]  Current/Best:   14.89/  19.83 GFLOPS | Progress: (20/20) | 13.02 s Done.
+
    [Task  4/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  4/25]  Current/Best:    9.43/  11.94 GFLOPS | Progress: (4/20) | 4.20 s
    [Task  4/25]  Current/Best:   17.68/  17.68 GFLOPS | Progress: (8/20) | 8.22 s
    [Task  4/25]  Current/Best:   13.76/  17.68 GFLOPS | Progress: (12/20) | 9.88 s
    [Task  4/25]  Current/Best:   16.36/  17.68 GFLOPS | Progress: (16/20) | 15.00 s
    [Task  4/25]  Current/Best:   18.56/  20.03 GFLOPS | Progress: (20/20) | 16.71 s Done.
+
    [Task  5/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  5/25]  Current/Best:    3.62/  10.48 GFLOPS | Progress: (4/20) | 3.57 s
    [Task  5/25]  Current/Best:   18.20/  18.20 GFLOPS | Progress: (8/20) | 5.22 s
    [Task  5/25]  Current/Best:   15.86/  19.98 GFLOPS | Progress: (12/20) | 7.04 s
    [Task  5/25]  Current/Best:   13.96/  19.98 GFLOPS | Progress: (16/20) | 9.21 s
    [Task  5/25]  Current/Best:   19.68/  19.98 GFLOPS | Progress: (20/20) | 11.06 s Done.
+
    [Task  6/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  6/25]  Current/Best:    3.48/  19.87 GFLOPS | Progress: (4/20) | 4.11 s
    [Task  6/25]  Current/Best:   16.29/  19.87 GFLOPS | Progress: (8/20) | 6.64 s
    [Task  6/25]  Current/Best:    4.20/  19.87 GFLOPS | Progress: (12/20) | 10.53 s
    [Task  6/25]  Current/Best:   13.42/  19.87 GFLOPS | Progress: (16/20) | 13.12 s
    [Task  6/25]  Current/Best:   18.79/  19.87 GFLOPS | Progress: (20/20) | 16.57 s Done.
+
    [Task  7/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  7/25]  Current/Best:   11.69/  20.81 GFLOPS | Progress: (4/20) | 4.27 s
    [Task  7/25]  Current/Best:   15.93/  20.81 GFLOPS | Progress: (8/20) | 6.68 s
    [Task  7/25]  Current/Best:   12.81/  20.81 GFLOPS | Progress: (12/20) | 9.28 s
    [Task  7/25]  Current/Best:   19.42/  20.81 GFLOPS | Progress: (16/20) | 11.14 s
    [Task  7/25]  Current/Best:   15.32/  20.81 GFLOPS | Progress: (20/20) | 13.39 s Done.
+
    [Task  8/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  8/25]  Current/Best:   10.79/  17.18 GFLOPS | Progress: (4/20) | 7.52 s
    [Task  8/25]  Current/Best:    5.81/  17.18 GFLOPS | Progress: (8/20) | 9.82 s
    [Task  8/25]  Current/Best:    4.03/  22.09 GFLOPS | Progress: (12/20) | 13.01 s
    [Task  8/25]  Current/Best:   16.68/  22.09 GFLOPS | Progress: (16/20) | 19.09 s
    [Task  8/25]  Current/Best:   20.21/  22.09 GFLOPS | Progress: (20/20) | 21.80 s Done.
+
    [Task  9/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  9/25]  Current/Best:    6.98/  15.94 GFLOPS | Progress: (4/20) | 3.50 s
    [Task  9/25]  Current/Best:   14.25/  15.94 GFLOPS | Progress: (8/20) | 5.56 s
    [Task  9/25]  Current/Best:   23.09/  23.09 GFLOPS | Progress: (12/20) | 8.70 s
    [Task  9/25]  Current/Best:    6.91/  23.09 GFLOPS | Progress: (16/20) | 10.27 s
    [Task  9/25]  Current/Best:    3.40/  23.09 GFLOPS | Progress: (20/20) | 13.67 s Done.
+
    [Task 10/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 10/25]  Current/Best:   14.30/  18.61 GFLOPS | Progress: (4/20) | 3.11 s
    [Task 10/25]  Current/Best:   13.06/  18.61 GFLOPS | Progress: (8/20) | 4.59 s
    [Task 10/25]  Current/Best:    4.76/  18.61 GFLOPS | Progress: (12/20) | 7.81 s
    [Task 10/25]  Current/Best:    9.82/  20.08 GFLOPS | Progress: (16/20) | 9.46 s
    [Task 10/25]  Current/Best:   14.06/  20.08 GFLOPS | Progress: (20/20) | 11.99 s Done.
+
    [Task 11/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 11/25]  Current/Best:   15.89/  15.89 GFLOPS | Progress: (4/20) | 5.06 s
    [Task 11/25]  Current/Best:   16.23/  16.23 GFLOPS | Progress: (8/20) | 7.06 s
    [Task 11/25]  Current/Best:   12.32/  20.44 GFLOPS | Progress: (12/20) | 9.33 s
    [Task 11/25]  Current/Best:   23.59/  23.59 GFLOPS | Progress: (16/20) | 10.98 s
    [Task 11/25]  Current/Best:   13.06/  23.59 GFLOPS | Progress: (20/20) | 13.55 s Done.
+
    [Task 12/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 12/25]  Current/Best:    9.96/  14.34 GFLOPS | Progress: (4/20) | 4.60 s
    [Task 12/25]  Current/Best:    5.28/  18.11 GFLOPS | Progress: (8/20) | 11.03 s
    [Task 12/25]  Current/Best:    9.19/  18.11 GFLOPS | Progress: (12/20) | 19.82 s
    [Task 12/25]  Current/Best:    9.69/  18.11 GFLOPS | Progress: (16/20) | 26.30 s
    [Task 12/25]  Current/Best:   20.94/  20.94 GFLOPS | Progress: (20/20) | 31.31 s Done.
+
    [Task 13/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 13/25]  Current/Best:   11.26/  11.86 GFLOPS | Progress: (4/20) | 4.08 s
    [Task 13/25]  Current/Best:   10.13/  15.40 GFLOPS | Progress: (8/20) | 8.06 s
    [Task 13/25]  Current/Best:   11.77/  20.87 GFLOPS | Progress: (12/20) | 11.12 s
    [Task 13/25]  Current/Best:   11.84/  20.87 GFLOPS | Progress: (16/20) | 14.26 s
    [Task 13/25]  Current/Best:   11.54/  20.87 GFLOPS | Progress: (20/20) | 17.49 s Done.
+
    [Task 14/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 14/25]  Current/Best:    6.42/  18.34 GFLOPS | Progress: (4/20) | 7.42 s
    [Task 14/25]  Current/Best:   12.82/  18.34 GFLOPS | Progress: (8/20) | 11.32 s
    [Task 14/25]  Current/Best:   18.86/  18.86 GFLOPS | Progress: (12/20) | 12.48 s
    [Task 14/25]  Current/Best:    3.01/  18.86 GFLOPS | Progress: (16/20) | 15.45 s
    [Task 14/25]  Current/Best:   10.44/  18.86 GFLOPS | Progress: (20/20) | 20.26 s
    [Task 15/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 15/25]  Current/Best:   18.48/  18.48 GFLOPS | Progress: (4/20) | 3.09 s
    [Task 15/25]  Current/Best:   18.87/  18.87 GFLOPS | Progress: (8/20) | 5.51 s
    [Task 15/25]  Current/Best:   23.60/  23.60 GFLOPS | Progress: (12/20) | 7.08 s
    [Task 15/25]  Current/Best:   12.11/  23.60 GFLOPS | Progress: (16/20) | 9.40 s
    [Task 15/25]  Current/Best:   18.75/  23.60 GFLOPS | Progress: (20/20)
  | 10.69 s Done.
+
    [Task 16/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 16/25]  Current/Best:    5.96/  17.77 GFLOPS | Progress: (4/20) | 3.11 s
    [Task 16/25]  Current/Best:   11.78/  21.12 GFLOPS | Progress: (8/20) | 6.83 s
    [Task 16/25]  Current/Best:   17.44/  21.12 GFLOPS | Progress: (12/20) | 10.12 s
    [Task 16/25]  Current/Best:   13.60/  21.12 GFLOPS | Progress: (16/20) | 11.65 s
    [Task 16/25]  Current/Best:   13.57/  21.12 GFLOPS | Progress: (20/20) | 15.39 s Done.
+
    [Task 17/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 17/25]  Current/Best:    1.56/  20.25 GFLOPS | Progress: (4/20) | 5.17 s Done.
+
    [Task 17/25]  Current/Best:    6.35/  20.25 GFLOPS | Progress: (8/20) | 7.95 s
    [Task 17/25]  Current/Best:   10.03/  21.03 GFLOPS | Progress: (12/20) | 9.95 s
    [Task 17/25]  Current/Best:   15.80/  21.03 GFLOPS | Progress: (16/20) | 12.17 s
    [Task 17/25]  Current/Best:   18.62/  21.03 GFLOPS | Progress: (20/20) | 13.78 s Done.
+
    [Task 18/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 18/25]  Current/Best:    7.34/  14.22 GFLOPS | Progress: (4/20) | 5.10 s
    [Task 18/25]  Current/Best:   14.22/  16.20 GFLOPS | Progress: (8/20) | 7.24 s
    [Task 18/25]  Current/Best:    9.32/  16.38 GFLOPS | Progress: (12/20) | 14.36 s
    [Task 18/25]  Current/Best:   12.04/  19.89 GFLOPS | Progress: (16/20) | 16.00 s
    [Task 18/25]  Current/Best:   12.06/  22.30 GFLOPS | Progress: (20/20) | 17.75 s Done.
+
    [Task 19/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 19/25]  Current/Best:   11.94/  19.08 GFLOPS | Progress: (4/20) | 4.51 s
    [Task 19/25]  Current/Best:    8.61/  19.08 GFLOPS | Progress: (8/20) | 7.48 s
    [Task 19/25]  Current/Best:   18.35/  19.08 GFLOPS | Progress: (12/20) | 11.00 s
    [Task 19/25]  Current/Best:   18.63/  19.08 GFLOPS | Progress: (16/20) | 14.26 s
    [Task 19/25]  Current/Best:   18.77/  19.08 GFLOPS | Progress: (20/20) | 20.38 s Done.
+
    [Task 20/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 20/25]  Current/Best:   16.36/  16.36 GFLOPS | Progress: (4/20) | 3.95 s
    [Task 20/25]  Current/Best:   13.96/  16.36 GFLOPS | Progress: (8/20) | 6.15 s
    [Task 20/25]  Current/Best:    9.67/  16.36 GFLOPS | Progress: (12/20) | 9.30 s
    [Task 20/25]  Current/Best:   15.77/  17.66 GFLOPS | Progress: (16/20) | 11.23 s
    [Task 20/25]  Current/Best:   13.44/  18.37 GFLOPS | Progress: (20/20) | 13.29 s Done.
+
    [Task 21/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 21/25]  Current/Best:   10.58/  19.75 GFLOPS | Progress: (4/20) | 4.55 s
    [Task 21/25]  Current/Best:   18.44/  19.75 GFLOPS | Progress: (8/20) | 6.51 s
    [Task 21/25]  Current/Best:    7.69/  19.75 GFLOPS | Progress: (12/20) | 7.96 s
    [Task 21/25]  Current/Best:    2.26/  19.75 GFLOPS | Progress: (16/20) | 12.10 s
    [Task 21/25]  Current/Best:    6.52/  19.75 GFLOPS | Progress: (20/20) | 13.62 s
    [Task 22/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 22/25]  Current/Best:   11.05/  13.18 GFLOPS | Progress: (4/20) | 3.10 s
    [Task 22/25]  Current/Best:   19.84/  19.84 GFLOPS | Progress: (8/20) | 4.34 s
    [Task 22/25]  Current/Best:   10.94/  19.84 GFLOPS | Progress: (12/20) | 5.79 s
    [Task 22/25]  Current/Best:    4.44/  19.84 GFLOPS | Progress: (16/20) | 9.21 s
    [Task 22/25]  Current/Best:    9.85/  19.84 GFLOPS | Progress: (20/20) |
  11.60 s Done.
+
    [Task 23/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 23/25]  Current/Best:    3.08/  13.93 GFLOPS | Progress: (4/20) | 4.81 s
    [Task 23/25]  Current/Best:    9.72/  22.13 GFLOPS | Progress: (8/20) | 7.49 s
    [Task 23/25]  Current/Best:   22.97/  22.97 GFLOPS | Progress: (12/20) | 10.22 s
    [Task 23/25]  Current/Best:    8.76/  22.97 GFLOPS | Progress: (16/20) | 12.63 s
    [Task 23/25]  Current/Best:   10.78/  22.97 GFLOPS | Progress: (20/20) | 16.24 s Done.
+
    [Task 24/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 24/25]  Current/Best:    4.70/   9.67 GFLOPS | Progress: (4/20) | 4.68 s
    [Task 24/25]  Current/Best:    7.15/   9.67 GFLOPS | Progress: (8/20) | 12.14 s
    [Task 24/25]  Current/Best:    3.32/   9.67 GFLOPS | Progress: (12/20) | 23.75 s
    [Task 24/25]  Current/Best:    2.21/   9.97 GFLOPS | Progress: (16/20) | 35.47 s
    [Task 24/25]  Current/Best:    5.98/   9.97 GFLOPS | Progress: (20/20) | 41.58 s
    [Task 25/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
      Done.
-
    [Task 21/25]  Current/Best:    3.13/  19.59 GFLOPS | Progress: (12/20) | 8.62 s
    [Task 21/25]  Current/Best:   18.18/  22.09 GFLOPS | Progress: (16/20) | 10.03 s
    [Task 21/25]  Current/Best:    9.40/  22.09 GFLOPS | Progress: (20/20) | 11.31 s
    [Task 22/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 22/25]  Current/Best:   14.53/  19.59 GFLOPS | Progress: (4/20) | 5.00 s
    [Task 22/25]  Current/Best:    5.30/  19.59 GFLOPS | Progress: (8/20) | 9.72 s
    [Task 22/25]  Current/Best:   15.46/  19.59 GFLOPS | Progress: (12/20) | 11.62 s
    [Task 22/25]  Current/Best:   18.21/  19.59 GFLOPS | Progress: (16/20) | 13.79 s
    [Task 22/25]  Current/Best:    6.88/  19.59 GFLOPS | Progress: (20/20) | 17.23 s Done.
-
    [Task 23/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 23/25]  Current/Best:   18.72/  18.72 GFLOPS | Progress: (4/20) | 5.50 s
    [Task 23/25]  Current/Best:    3.08/  18.72 GFLOPS | Progress: (8/20) | 11.20 s
    [Task 23/25]  Current/Best:   11.90/  18.72 GFLOPS | Progress: (12/20) | 13.33 s
    [Task 23/25]  Current/Best:   20.95/  20.95 GFLOPS | Progress: (16/20) | 16.14 s
    [Task 23/25]  Current/Best:   10.15/  20.95 GFLOPS | Progress: (20/20) | 19.25 s Done.
-
    [Task 24/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 24/25]  Current/Best:    8.11/   9.02 GFLOPS | Progress: (4/20) | 12.27 s
    [Task 24/25]  Current/Best:    5.15/   9.02 GFLOPS | Progress: (8/20) | 16.70 s
    [Task 24/25]  Current/Best:    6.74/   9.02 GFLOPS | Progress: (12/20) | 26.77 s
    [Task 24/25]  Current/Best:    1.62/   9.02 GFLOPS | Progress: (16/20) | 37.51 s
    [Task 24/25]  Current/Best:    2.90/   9.02 GFLOPS | Progress: (20/20) | 49.12 s
    [Task 25/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
-     Done.
-
    [Task 25/25]  Current/Best:    1.53/   7.76 GFLOPS | Progress: (4/20) | 13.03 s
    [Task 25/25]  Current/Best:    6.16/   7.76 GFLOPS | Progress: (8/20) | 18.28 s
    [Task 25/25]  Current/Best:    2.98/   8.84 GFLOPS | Progress: (12/20) | 28.78 s
    [Task 25/25]  Current/Best:    3.91/   8.84 GFLOPS | Progress: (16/20) | 40.36 s
    [Task 25/25]  Current/Best:    9.01/   9.01 GFLOPS | Progress: (20/20) | 51.07 s
+
    [Task 25/25]  Current/Best:    7.95/   8.73 GFLOPS | Progress: (4/20) | 6.61 s
    [Task 25/25]  Current/Best:    8.01/   8.73 GFLOPS | Progress: (8/20) | 17.36 s
    [Task 25/25]  Current/Best:    5.71/   8.73 GFLOPS | Progress: (12/20) | 19.57 s
    [Task 25/25]  Current/Best:    1.54/   9.40 GFLOPS | Progress: (16/20) | 24.44 s
    [Task 25/25]  Current/Best:    5.76/   9.40 GFLOPS | Progress: (20/20) | 26.00 s
 
 
 
@@ -732,8 +732,8 @@ improvement in comparing the optimized model to the unoptimized model.
 
  .. code-block:: none
 
-    optimized: {'mean': 405.79506909000884, 'median': 405.99190305001684, 'std': 0.8678081614214388}
-    unoptimized: {'mean': 517.1077569899899, 'median': 516.9921277999947, 'std': 1.3162579716279146}
+    optimized: {'mean': 407.88420194999844, 'median': 405.84936035000965, 'std': 3.4246055305415424}
+    unoptimized: {'mean': 510.4595590599991, 'median': 509.4302212499997, 'std': 3.3041499135301597}
 
 
 
@@ -756,7 +756,7 @@ profiling/benchmarking.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 11 minutes  8.794 seconds)
+   **Total running time of the script:** ( 10 minutes  38.131 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 461651c82f..a99e6768d5 100644
--- a/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
+++ b/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
@@ -270,7 +270,7 @@ device and returns the measured cost. Network overhead is excluded.
 
  .. code-block:: none
 
-    1.299e-07 secs/op
+    1.269e-07 secs/op
 
 
 
diff --git a/docs/_sources/tutorial/intro_topi.rst.txt b/docs/_sources/tutorial/intro_topi.rst.txt
index 229c7bf668..e8f2bf965c 100644
--- a/docs/_sources/tutorial/intro_topi.rst.txt
+++ b/docs/_sources/tutorial/intro_topi.rst.txt
@@ -263,7 +263,7 @@ As you can see, scheduled stages of computation have been accumulated and we can
 
  .. code-block:: none
 
-    [stage(a, placeholder(a, 0x174fabe0)), stage(b, placeholder(b, 0xc477070)), stage(T_add, compute(T_add, body=[(a[ax0, ax1, ax2] + b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(min=0, ext=10))], reduce_axis=[], tag=broadcast, attrs={})), stage(T_multiply, compute(T_multiply, body=[(a[ax0, ax1, ax2]*b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(min [...]
+    [stage(a, placeholder(a, 0xc4ba520)), stage(b, placeholder(b, 0xec67740)), stage(T_add, compute(T_add, body=[(a[ax0, ax1, ax2] + b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(min=0, ext=10))], reduce_axis=[], tag=broadcast, attrs={})), stage(T_multiply, compute(T_multiply, body=[(a[ax0, ax1, ax2]*b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(min= [...]
 
 
 
diff --git a/docs/_sources/tutorial/sg_execution_times.rst.txt b/docs/_sources/tutorial/sg_execution_times.rst.txt
index ebaa29d417..bf98b2cf16 100644
--- a/docs/_sources/tutorial/sg_execution_times.rst.txt
+++ b/docs/_sources/tutorial/sg_execution_times.rst.txt
@@ -5,32 +5,32 @@
 
 Computation times
 =================
-**14:36.316** total execution time for **tutorial** files:
+**14:21.829** total execution time for **tutorial** files:
 
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``)                 | 11:08.794 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``)                 | 10:38.131 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``) | 01:30.940 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``) | 01:42.599 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``)     | 01:00.696 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``)     | 01:01.598 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``)                 | 00:36.288 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``)                 | 00:35.843 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``)               | 00:18.007 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``)               | 00:21.416 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``)                               | 00:00.776 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``)       | 00:01.293 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``)       | 00:00.643 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``)                               | 00:00.772 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``) | 00:00.163 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``) | 00:00.169 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
 | :ref:`sphx_glr_tutorial_introduction.py` (``introduction.py``)                           | 00:00.005 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_uma.py` (``uma.py``)                                             | 00:00.002 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_uma.py` (``uma.py``)                                             | 00:00.001 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
 | :ref:`sphx_glr_tutorial_install.py` (``install.py``)                                     | 00:00.001 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_tvmc_command_line_driver.py` (``tvmc_command_line_driver.py``)   | 00:00.001 | 0.0 MB |
-+------------------------------------------------------------------------------------------+-----------+--------+
 | :ref:`sphx_glr_tutorial_tvmc_python.py` (``tvmc_python.py``)                             | 00:00.001 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
+| :ref:`sphx_glr_tutorial_tvmc_command_line_driver.py` (``tvmc_command_line_driver.py``)   | 00:00.001 | 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 864e030126..f33c639281 100644
--- a/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
+++ b/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
@@ -501,10 +501,10 @@ We can now compare the different schedules
  .. code-block:: none
 
                 Operator                  Timing             Performance
-                   numpy    6.728879998263437e-06                    1.0
-                   naive    6.689199999999999e-06      0.994103030775749
-                parallel              7.1519e-06      1.0628663316697182
-                  vector    2.4517499999999997e-05    3.6436227137840738
+                   numpy    7.337909999023395e-06                    1.0
+                   naive    6.675599999999999e-06     0.9097413297367312
+                parallel    6.990699999999999e-06     0.9526827122341911
+                  vector    2.4616599999999997e-05    3.3547154439446967
 
 
 
@@ -925,7 +925,7 @@ matrix multiplication.
 
  .. code-block:: none
 
-    Numpy running time: 0.018565
+    Numpy running time: 0.018753
 
 
 
@@ -983,7 +983,7 @@ optimizations.
 
  .. code-block:: none
 
-    none: 3.368203
+    none: 3.448494
 
 
 
@@ -1086,7 +1086,7 @@ schedule.
 
  .. code-block:: none
 
-    blocking: 0.310311
+    blocking: 0.301517
 
 
 
@@ -1182,7 +1182,7 @@ already cache friendly from our previous optimizations.
 
  .. code-block:: none
 
-    vectorization: 0.342435
+    vectorization: 0.345078
     @main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
       attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
       buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1256,7 +1256,7 @@ more cache friendly.
 
  .. code-block:: none
 
-    loop permutation: 0.122163
+    loop permutation: 0.116288
     @main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
       attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
       buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1355,7 +1355,7 @@ optimized schedule.
 
  .. code-block:: none
 
-    array packing: 0.109077
+    array packing: 0.107839
     @main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
       attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
       buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1448,7 +1448,7 @@ to `C` when all the block results are ready.
 
  .. code-block:: none
 
-    block caching: 0.111902
+    block caching: 0.110582
     @main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
       attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
       buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1534,7 +1534,7 @@ of thread-level parallelization.
 
  .. code-block:: none
 
-    parallelization: 0.146945
+    parallelization: 0.146665
     @main = primfn(A_1: handle, B_1: handle, C_1: handle) -> ()
       attr = {"from_legacy_te_schedule": True, "global_symbol": "main", "tir.noalias": True}
       buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1615,13 +1615,13 @@ working, we can compare the results.
  .. code-block:: none
 
                 Operator                  Timing             Performance
-                    none             3.368203356                     1.0
-                blocking            0.3103111295     0.09212957078355218
-           vectorization            0.3424345915     0.10166683994599048
-        loop permutation     0.12216339170000001     0.03626960096764419
-           array packing            0.1090769235     0.03238430461916564
-           block caching     0.11190187370000002    0.033223015914600855
-         parallelization     0.14694490659999998     0.04362708870835766
+                    none            3.4484936041                     1.0
+                blocking     0.30151734750000003     0.08743450970635948
+           vectorization            0.3450780444     0.10006631416967923
+        loop permutation     0.11628793479999999     0.03372137174960753
+           array packing            0.1078393406     0.03127143413338135
+           block caching     0.11058221309999998     0.03206681693378408
+         parallelization            0.1466654998     0.04253030935757739
 
 
 
@@ -1663,7 +1663,7 @@ the computation for specific platforms.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  0.696 seconds)
+   **Total running time of the script:** ( 1 minutes  1.598 seconds)
 
 
 .. _sphx_glr_download_tutorial_tensor_expr_get_started.py:
diff --git a/docs/commit_hash b/docs/commit_hash
index 344a77844a..f3aa1d1179 100644
--- a/docs/commit_hash
+++ b/docs/commit_hash
@@ -1 +1 @@
-b6fae9b35eff4ad1f7cc2e83d8d7da5d701d8e44
+3aa16f72dd3f1807b11ec61cf372af07d32099c4
diff --git a/docs/how_to/compile_models/from_darknet.html b/docs/how_to/compile_models/from_darknet.html
index d235ec3e02..5c68851228 100644
--- a/docs/how_to/compile_models/from_darknet.html
+++ b/docs/how_to/compile_models/from_darknet.html
@@ -585,7 +585,7 @@ 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  12.906 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  10.959 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_keras.html b/docs/how_to/compile_models/from_keras.html
index 38054e2b72..4bdb2e7bb9 100644
--- a/docs/how_to/compile_models/from_keras.html
+++ b/docs/how_to/compile_models/from_keras.html
@@ -506,7 +506,7 @@ pip install -U tensorflow --user
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Relay top-1 id: 285, class name: Egyptian cat
 
 1/1 [==============================] - ETA: 0s
-1/1 [==============================] - 1s 938ms/step
+1/1 [==============================] - 1s 988ms/step
 Keras top-1 id: 285, class name: Egyptian cat
 </pre></div>
 </div>
diff --git a/docs/how_to/compile_models/from_mxnet.html b/docs/how_to/compile_models/from_mxnet.html
index 3a0a9f0a44..3f66af943e 100644
--- a/docs/how_to/compile_models/from_mxnet.html
+++ b/docs/how_to/compile_models/from_mxnet.html
@@ -440,7 +440,7 @@ to download the full example code</p>
 <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.zip7e4d0e69-aa5a-4589-a37a-99f69c911e3e 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.zip85dc4430-f38f-45c8-95f0-1610ffdf492a 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 f136772e2c..27eb72ae7f 100644
--- a/docs/how_to/compile_models/from_oneflow.html
+++ b/docs/how_to/compile_models/from_oneflow.html
@@ -448,13 +448,10 @@ 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]
- 19%|#9        | 7.99M/41.5M [00:00&lt;00:00, 42.0MB/s]
- 29%|##8       | 12.0M/41.5M [00:00&lt;00:00, 35.8MB/s]
- 39%|###8      | 16.0M/41.5M [00:00&lt;00:00, 31.1MB/s]
- 58%|#####7    | 24.0M/41.5M [00:00&lt;00:00, 35.4MB/s]
- 77%|#######7  | 32.0M/41.5M [00:00&lt;00:00, 39.6MB/s]
- 92%|#########2| 38.3M/41.5M [00:01&lt;00:00, 33.9MB/s]
-100%|##########| 41.5M/41.5M [00:01&lt;00:00, 35.7MB/s]
+ 39%|###8      | 16.0M/41.5M [00:00&lt;00:00, 101MB/s]
+ 62%|######1   | 25.6M/41.5M [00:00&lt;00:00, 75.4MB/s]
+ 82%|########2 | 34.1M/41.5M [00:00&lt;00:00, 69.4MB/s]
+100%|##########| 41.5M/41.5M [00:00&lt;00:00, 75.2MB/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 75750f8475..e6d525dcc4 100644
--- a/docs/how_to/compile_models/from_pytorch.html
+++ b/docs/how_to/compile_models/from_pytorch.html
@@ -431,10 +431,10 @@ 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]
- 28%|##7       | 12.4M/44.7M [00:00&lt;00:00, 129MB/s]
- 55%|#####5    | 24.7M/44.7M [00:00&lt;00:00, 110MB/s]
- 79%|#######9  | 35.3M/44.7M [00:00&lt;00:00, 106MB/s]
-100%|##########| 44.7M/44.7M [00:00&lt;00:00, 107MB/s]
+ 34%|###4      | 15.3M/44.7M [00:00&lt;00:00, 160MB/s]
+ 69%|######8   | 30.6M/44.7M [00:00&lt;00:00, 118MB/s]
+ 95%|#########5| 42.5M/44.7M [00:00&lt;00:00, 97.3MB/s]
+100%|##########| 44.7M/44.7M [00:00&lt;00:00, 109MB/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 470067cf5b..d2b845b4e5 100644
--- a/docs/how_to/compile_models/from_tensorflow.html
+++ b/docs/how_to/compile_models/from_tensorflow.html
@@ -645,7 +645,7 @@ 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  11.987 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  12.549 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 48bb5ea16a..6409836fbb 100644
--- a/docs/how_to/compile_models/sg_execution_times.html
+++ b/docs/how_to/compile_models/sg_execution_times.html
@@ -340,7 +340,7 @@
             
   <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>05:49.415</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
+<p><strong>05:51.913</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 81%" />
@@ -348,44 +348,44 @@
 <col style="width: 8%" />
 </colgroup>
 <tbody>
-<tr class="row-odd"><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>01:12.906</p></td>
+<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>01:12.549</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
-<tr class="row-even"><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>01:11.987</p></td>
+<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>01:10.959</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>00:46.883</p></td>
+<td><p>00:47.690</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>00:33.450</p></td>
+<td><p>00:33.105</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><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>00:30.046</p></td>
+<td><p>00:31.248</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><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>00:26.740</p></td>
+<td><p>00:27.261</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>00:24.927</p></td>
+<td><p>00:25.089</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>00:22.933</p></td>
+<td><p>00:22.844</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>00:17.078</p></td>
+<td><p>00:18.732</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>00:02.466</p></td>
+<td><p>00:02.436</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 </tbody>
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 59b327060a..4ab178c8bb 100644
--- a/docs/how_to/deploy_models/deploy_model_on_android.html
+++ b/docs/how_to/deploy_models/deploy_model_on_android.html
@@ -662,7 +662,7 @@ to the remote android device.</p>
 Evaluate inference time cost...
 Execution time summary:
  mean (ms)   median (ms)    max (ms)     min (ms)     std (ms)
-  16.4142      16.3849      16.9731      15.9958       0.3440
+  16.8910      16.2437      22.1541      16.0129       1.7825
 </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 864cc7ae08..357488bd7d 100644
--- a/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
+++ b/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
@@ -453,20 +453,21 @@ 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]
-  5%|4         | 7.99M/170M [00:00&lt;00:02, 61.2MB/s]
- 11%|#1        | 18.8M/170M [00:00&lt;00:01, 87.8MB/s]
- 19%|#8        | 32.0M/170M [00:00&lt;00:01, 88.1MB/s]
- 27%|##7       | 46.3M/170M [00:00&lt;00:01, 102MB/s]
- 33%|###3      | 56.8M/170M [00:00&lt;00:01, 105MB/s]
- 42%|####2     | 72.0M/170M [00:00&lt;00:00, 117MB/s]
- 49%|####9     | 83.3M/170M [00:00&lt;00:00, 116MB/s]
- 56%|#####5    | 94.5M/170M [00:00&lt;00:00, 97.8MB/s]
- 61%|######1   | 104M/170M [00:01&lt;00:00, 98.8MB/s]
- 70%|######9   | 118M/170M [00:01&lt;00:00, 111MB/s]
- 76%|#######6  | 129M/170M [00:01&lt;00:00, 94.2MB/s]
- 82%|########1 | 139M/170M [00:01&lt;00:00, 94.3MB/s]
- 87%|########7 | 148M/170M [00:01&lt;00:00, 93.5MB/s]
- 95%|#########5| 161M/170M [00:01&lt;00:00, 105MB/s]
+  8%|8         | 13.9M/170M [00:00&lt;00:01, 146MB/s]
+ 16%|#6        | 27.8M/170M [00:00&lt;00:01, 113MB/s]
+ 23%|##2       | 39.0M/170M [00:00&lt;00:01, 109MB/s]
+ 29%|##9       | 49.5M/170M [00:00&lt;00:01, 106MB/s]
+ 35%|###5      | 59.7M/170M [00:00&lt;00:01, 104MB/s]
+ 41%|####1     | 69.7M/170M [00:00&lt;00:01, 102MB/s]
+ 47%|####6     | 79.5M/170M [00:00&lt;00:00, 102MB/s]
+ 53%|#####2    | 89.2M/170M [00:00&lt;00:00, 96.2MB/s]
+ 59%|#####9    | 100M/170M [00:01&lt;00:00, 102MB/s]
+ 65%|######4   | 110M/170M [00:01&lt;00:00, 102MB/s]
+ 71%|#######   | 120M/170M [00:01&lt;00:00, 101MB/s]
+ 76%|#######6  | 130M/170M [00:01&lt;00:00, 101MB/s]
+ 82%|########1 | 139M/170M [00:01&lt;00:00, 100MB/s]
+ 88%|########7 | 149M/170M [00:01&lt;00:00, 100MB/s]
+ 93%|#########3| 158M/170M [00:01&lt;00:00, 82.0MB/s]
 100%|##########| 170M/170M [00:01&lt;00:00, 102MB/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)
@@ -565,7 +566,7 @@ 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  20.869 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 3 minutes  21.087 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 6a40655efe..cb8e88ad07 100644
--- a/docs/how_to/deploy_models/deploy_prequantized.html
+++ b/docs/how_to/deploy_models/deploy_prequantized.html
@@ -497,8 +497,8 @@ 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, 80.2MB/s]
-100%|##########| 13.6M/13.6M [00:00&lt;00:00, 58.5MB/s]
+ 69%|######9   | 9.38M/13.6M [00:00&lt;00:00, 98.3MB/s]
+100%|##########| 13.6M/13.6M [00:00&lt;00:00, 124MB/s]
 </pre></div>
 </div>
 </div>
@@ -589,7 +589,7 @@ 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)
-  90.5781      90.5102      92.8514      90.1945       0.3491
+  90.3989      90.3041      95.5078      90.1724       0.5412
 </pre></div>
 </div>
 <div class="admonition note">
@@ -628,7 +628,7 @@ 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  7.730 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  7.793 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 3cc723e49d..e80a632007 100644
--- a/docs/how_to/deploy_models/deploy_prequantized_tflite.html
+++ b/docs/how_to/deploy_models/deploy_prequantized_tflite.html
@@ -582,7 +582,7 @@ 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)
-  120.9755     120.8276     126.3660     119.9609      0.7430
+  121.0398     120.9786     126.2886     120.1887      0.6482
 </pre></div>
 </div>
 <div class="admonition note">
@@ -610,7 +610,7 @@ 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  27.444 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes  27.687 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 f897dfa6c0..fb42de54b3 100644
--- a/docs/how_to/deploy_models/deploy_quantized.html
+++ b/docs/how_to/deploy_models/deploy_quantized.html
@@ -520,7 +520,7 @@ 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  40.803 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  16.908 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 6df4d34646..aab0d53fda 100644
--- a/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
+++ b/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
@@ -462,24 +462,24 @@ 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]
-  2%|1         | 2188/132723 [00:00&lt;00:06, 21489.31KB/s]
-  5%|4         | 6078/132723 [00:00&lt;00:04, 31652.43KB/s]
-  9%|9         | 12209/132723 [00:00&lt;00:02, 45135.88KB/s]
- 15%|#5        | 20190/132723 [00:00&lt;00:01, 58786.19KB/s]
- 21%|##1       | 28449/132723 [00:00&lt;00:01, 67353.36KB/s]
- 28%|##7       | 36599/132723 [00:00&lt;00:01, 72154.59KB/s]
- 34%|###3      | 44881/132723 [00:00&lt;00:01, 75635.00KB/s]
- 40%|####      | 53214/132723 [00:00&lt;00:01, 78080.10KB/s]
- 46%|####6     | 61563/132723 [00:00&lt;00:00, 79768.85KB/s]
- 53%|#####2    | 69908/132723 [00:01&lt;00:00, 80902.06KB/s]
- 59%|#####8    | 78165/132723 [00:01&lt;00:00, 81409.82KB/s]
- 65%|######5   | 86482/132723 [00:01&lt;00:00, 81943.49KB/s]
- 71%|#######1  | 94821/132723 [00:01&lt;00:00, 82378.61KB/s]
- 78%|#######7  | 103060/132723 [00:01&lt;00:00, 60350.65KB/s]
- 84%|########3 | 111129/132723 [00:01&lt;00:00, 65232.74KB/s]
- 90%|########9 | 119373/132723 [00:01&lt;00:00, 69625.24KB/s]
- 96%|#########6| 127617/132723 [00:01&lt;00:00, 73050.40KB/s]
-100%|##########| 132723/132723 [00:01&lt;00:00, 70374.98KB/s]
+  4%|4         | 5496/132723 [00:00&lt;00:02, 54954.42KB/s]
+ 10%|#         | 13327/132723 [00:00&lt;00:01, 68687.35KB/s]
+ 16%|#5        | 21173/132723 [00:00&lt;00:01, 73144.09KB/s]
+ 22%|##1       | 29079/132723 [00:00&lt;00:01, 75475.98KB/s]
+ 28%|##7       | 36842/132723 [00:00&lt;00:01, 76245.75KB/s]
+ 34%|###3      | 44898/132723 [00:00&lt;00:01, 77709.24KB/s]
+ 40%|###9      | 52743/132723 [00:00&lt;00:01, 77949.23KB/s]
+ 46%|####5     | 60631/132723 [00:00&lt;00:00, 78242.27KB/s]
+ 52%|#####1    | 68457/132723 [00:00&lt;00:00, 78246.20KB/s]
+ 58%|#####7    | 76349/132723 [00:01&lt;00:00, 78450.47KB/s]
+ 63%|######3   | 84237/132723 [00:01&lt;00:00, 78579.71KB/s]
+ 70%|######9   | 92262/132723 [00:01&lt;00:00, 79085.11KB/s]
+ 75%|#######5  | 100171/132723 [00:01&lt;00:00, 78619.05KB/s]
+ 81%|########1 | 108090/132723 [00:01&lt;00:00, 78788.20KB/s]
+ 87%|########7 | 116026/132723 [00:01&lt;00:00, 78952.27KB/s]
+ 93%|#########3| 123945/132723 [00:01&lt;00:00, 79020.19KB/s]
+ 99%|#########9| 131990/132723 [00:01&lt;00:00, 79447.00KB/s]
+100%|##########| 132723/132723 [00:01&lt;00:00, 77469.79KB/s]
 </pre></div>
 </div>
 <p>Create TVM runtime and do inference
@@ -518,7 +518,7 @@ 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  4.107 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> ( 3 minutes  4.805 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 e612f2b9e5..f3aa39f4d9 100644
--- a/docs/how_to/deploy_models/sg_execution_times.html
+++ b/docs/how_to/deploy_models/sg_execution_times.html
@@ -340,7 +340,7 @@
             
   <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>13:09.839</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
+<p><strong>12:47.044</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 86%" />
@@ -349,35 +349,35 @@
 </colgroup>
 <tbody>
 <tr class="row-odd"><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>03:20.869</p></td>
+<td><p>03:21.087</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><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>03:04.107</p></td>
+<td><p>03:04.805</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>02:27.444</p></td>
+<td><p>02:27.687</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>01:40.803</p></td>
+<td><p>01:16.908</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>01:07.730</p></td>
+<td><p>01:07.793</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>00:37.004</p></td>
+<td><p>00:37.100</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>00:26.263</p></td>
+<td><p>00:26.195</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>00:25.614</p></td>
+<td><p>00:25.462</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 ea47caa224..b29a45bddf 100644
--- a/docs/how_to/extend_tvm/bring_your_own_datatypes.html
+++ b/docs/how_to/extend_tvm/bring_your_own_datatypes.html
@@ -621,7 +621,7 @@ 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.zip52070d7f-f159-486c-82a3-217adcb347fe 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.zip178162a8-b0e0-4095-8df6-151f912a796c 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 fbdde0e075..b7bf3980ec 100644
--- a/docs/how_to/extend_tvm/sg_execution_times.html
+++ b/docs/how_to/extend_tvm/sg_execution_times.html
@@ -340,7 +340,7 @@
             
   <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>00:48.735</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
+<p><strong>00:49.273</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 84%" />
@@ -349,15 +349,15 @@
 </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>00:45.195</p></td>
+<td><p>00:45.717</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>00:02.472</p></td>
+<td><p>00:02.479</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>00:01.060</p></td>
+<td><p>00:01.068</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 aff2ede3be..4dc85c2f0a 100644
--- a/docs/how_to/extend_tvm/use_pass_instrument.html
+++ b/docs/how_to/extend_tvm/use_pass_instrument.html
@@ -525,10 +525,10 @@ 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: 7264us [7264us] (46.43%; 46.43%)
-FoldScaleAxis: 8382us [7us] (53.57%; 53.57%)
-        FoldConstant: 8375us [1696us] (53.53%; 99.92%)
-                InferType: 6679us [6679us] (42.69%; 79.75%)
+InferType: 7318us [7318us] (45.35%; 45.35%)
+FoldScaleAxis: 8820us [8us] (54.65%; 54.65%)
+        FoldConstant: 8812us [1704us] (54.61%; 99.91%)
+                InferType: 7108us [7108us] (44.04%; 80.66%)
 </pre></div>
 </div>
 </div>
@@ -550,10 +550,10 @@ 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: 6710us [6710us] (44.83%; 44.83%)
-FoldScaleAxis: 8258us [6us] (55.17%; 55.17%)
-        FoldConstant: 8253us [1716us] (55.13%; 99.93%)
-                InferType: 6537us [6537us] (43.67%; 79.21%)
+InferType: 6940us [6940us] (45.33%; 45.33%)
+FoldScaleAxis: 8370us [6us] (54.67%; 54.67%)
+        FoldConstant: 8364us [1716us] (54.63%; 99.93%)
+                InferType: 6648us [6648us] (43.42%; 79.48%)
 </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 c886bd8d61..f44f450be1 100644
--- a/docs/how_to/optimize_operators/opt_conv_cuda.html
+++ b/docs/how_to/optimize_operators/opt_conv_cuda.html
@@ -577,7 +577,7 @@ 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: 51.707935 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Convolution: 54.220928 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 dccf9a1fa4..389060737e 100644
--- a/docs/how_to/optimize_operators/opt_conv_tensorcore.html
+++ b/docs/how_to/optimize_operators/opt_conv_tensorcore.html
@@ -916,7 +916,7 @@ 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: 13.361002 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>conv2d with tensor core: 13.369524 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 88af04c7aa..e6eaf962a0 100644
--- a/docs/how_to/optimize_operators/opt_gemm.html
+++ b/docs/how_to/optimize_operators/opt_gemm.html
@@ -474,8 +474,8 @@ 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: 0.019154
-Baseline: 3.370971
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.019707
+Baseline: 3.518368
 </pre></div>
 </div>
 <p>In TVM, we can always inspect lower level IR to debug or optimize our schedule.
@@ -535,7 +535,7 @@ 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: 0.310276
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt1: 0.326323
 </pre></div>
 </div>
 <p>Here is the generated IR after blocking.</p>
@@ -602,7 +602,7 @@ 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: 0.348115
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt2: 0.350061
 </pre></div>
 </div>
 <p>Here is the generated IR after vectorization.</p>
@@ -663,7 +663,7 @@ 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: 0.120739
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt3: 0.124679
 </pre></div>
 </div>
 <p>Here is the generated IR after loop permutation.</p>
@@ -746,7 +746,7 @@ 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: 0.110055
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt4: 0.109649
 </pre></div>
 </div>
 <p>Here is the generated IR after array packing.</p>
@@ -832,7 +832,7 @@ 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: 0.111622
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt5: 0.111123
 </pre></div>
 </div>
 <p>Here is the generated IR after blocking.</p>
@@ -922,7 +922,7 @@ write to C when all the block results are ready.</p>
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;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: 0.147616
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt6: 0.147065
 </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 0c0f904ba2..a2e11f1068 100644
--- a/docs/how_to/optimize_operators/sg_execution_times.html
+++ b/docs/how_to/optimize_operators/sg_execution_times.html
@@ -340,7 +340,7 @@
             
   <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>00:35.320</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
+<p><strong>00:35.790</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 83%" />
@@ -349,15 +349,15 @@
 </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>00:32.617</p></td>
+<td><p>00:33.287</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>00:01.541</p></td>
+<td><p>00:01.430</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>00:01.162</p></td>
+<td><p>00:01.073</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 c5e2feee48..14539cc558 100644
--- a/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
+++ b/docs/how_to/tune_with_autoscheduler/sg_execution_times.html
@@ -340,7 +340,7 @@
             
   <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>09:16.216</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
+<p><strong>09:27.284</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 85%" />
@@ -349,27 +349,27 @@
 </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>05:45.244</p></td>
+<td><p>05:56.080</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>01:33.722</p></td>
+<td><p>01:34.623</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>01:04.412</p></td>
+<td><p>01:04.140</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>00:29.278</p></td>
+<td><p>00:28.953</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>00:12.155</p></td>
+<td><p>00:12.164</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>00:11.404</p></td>
+<td><p>00:11.323</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 7172e89492..1ad41bf941 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
@@ -505,142 +505,375 @@ cooperative fetching, unrolling and operator fusion.</p>
   buffer_map = {data_1: data, kernel_1: kernel, bias_1: bias, compute_1: compute}
   preflattened_buffer_map = {data_1: data_3: Buffer(data_2, float32, [1, 512, 7, 7], []), kernel_1: kernel_3: Buffer(kernel_2, float32, [512, 512, 3, 3], []), bias_1: bias_3: Buffer(bias_2, float32, [1, 512, 1, 1], []), compute_1: compute_3: Buffer(compute_2, float32, [1, 512, 7, 7], [])} {
   attr [IterVar(blockIdx.x: int32, (nullptr), &quot;ThreadIndex&quot;, &quot;blockIdx.x&quot;)] &quot;thread_extent&quot; = 64;
-  allocate(conv2d_nchw: Pointer(local float32), float32, [2]), storage_scope = local;
-  allocate(pad_temp.shared: Pointer(shared float32), float32, [196]), storage_scope = shared;
-  allocate(kernel.shared: Pointer(shared float32), float32, [32]), storage_scope = shared;
-  attr [IterVar(threadIdx.x: int32, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 196 {
-    conv2d_nchw_1: Buffer(conv2d_nchw, float32, [1], [], scope=&quot;local&quot;, align=4)[0] = 0f32
+  allocate(conv2d_nchw: Pointer(local float32), float32, [8]), storage_scope = local;
+  allocate(pad_temp.shared: Pointer(shared float32), float32, [252]), storage_scope = shared;
+  allocate(kernel.shared: Pointer(shared float32), float32, [96]), storage_scope = shared;
+  attr [IterVar(threadIdx.x: int32, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 49 {
+    conv2d_nchw_1: Buffer(conv2d_nchw, float32, [16], [], scope=&quot;local&quot;, align=16)[0] = 0f32
+    conv2d_nchw_1[4] = 0f32
     conv2d_nchw_1[1] = 0f32
+    conv2d_nchw_1[5] = 0f32
+    conv2d_nchw_1[2] = 0f32
+    conv2d_nchw_1[6] = 0f32
+    conv2d_nchw_1[3] = 0f32
+    conv2d_nchw_1[7] = 0f32
     for (rc.outer.outer: int32, 0, 128) {
-      attr [IterVar(threadIdx.x_1: int32, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 196;
-      pad_temp.shared_1: Buffer(pad_temp.shared, float32, [196], [], scope=&quot;shared&quot;)[threadIdx.x_1] = @tir.if_then_else(((7 &lt;= floormod(threadIdx.x_1, 49)) &amp;&amp; (1 &lt;= floormod(threadIdx.x_1, 7))), data[(((rc.outer.outer*196) + threadIdx.x_1) - 8)], 0f32, dtype=float32)
-      attr [IterVar(threadIdx.x_2: int32, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 196;
-      if @tir.likely((threadIdx.x_2 &lt; 32), dtype=bool) {
-        kernel.shared_1: Buffer(kernel.shared, float32, [32], [], scope=&quot;shared&quot;)[threadIdx.x_2] = kernel[((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 4)*4608)) + (rc.outer.outer*36)) + (floormod(threadIdx.x_2, 4)*9))]
+      let cse_var_1: int32 = (rc.outer.outer*196)
+       {
+        attr [IterVar(threadIdx.x_1: int32, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 49;
+        pad_temp.shared_1: Buffer(pad_temp.shared, float32, [252], [], scope=&quot;shared&quot;)[threadIdx.x_1] = @tir.if_then_else(((7 &lt;= threadIdx.x_1) &amp;&amp; (1 &lt;= floormod(threadIdx.x_1, 7))), data[((cse_var_1 + threadIdx.x_1) - 8)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 49;
+        pad_temp.shared_1[(threadIdx.x_1 + 49)] = @tir.if_then_else((((1 &lt;= floormod((floordiv(threadIdx.x_1, 7) + 7), 9)) &amp;&amp; (floormod((floordiv(threadIdx.x_1, 7) + 7), 9) &lt; 8)) &amp;&amp; (1 &lt;= floormod(threadIdx.x_1, 7))), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 49), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 7), 9)*7)) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 49;
+        pad_temp.shared_1[(threadIdx.x_1 + 98)] = @tir.if_then_else((((1 &lt;= floormod((floordiv(threadIdx.x_1, 7) + 5), 9)) &amp;&amp; (floormod((floordiv(threadIdx.x_1, 7) + 5), 9) &lt; 8)) &amp;&amp; (1 &lt;= floormod(threadIdx.x_1, 7))), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 98), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 5), 9)*7)) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 49;
+        pad_temp.shared_1[(threadIdx.x_1 + 147)] = @tir.if_then_else((((1 &lt;= floormod((floordiv(threadIdx.x_1, 7) + 3), 9)) &amp;&amp; (floormod((floordiv(threadIdx.x_1, 7) + 3), 9) &lt; 8)) &amp;&amp; (1 &lt;= floormod(threadIdx.x_1, 7))), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 147), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 3), 9)*7)) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 49;
+        pad_temp.shared_1[(threadIdx.x_1 + 196)] = @tir.if_then_else((1 &lt;= floormod(threadIdx.x_1, 7)), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 196), 63)*49)) + ((floordiv(threadIdx.x_1, 7) + 1)*7)) + floormod(threadIdx.x_1, 7)) - 8)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 49;
+        if @tir.likely((threadIdx.x_1 &lt; 7), dtype=bool) {
+          pad_temp.shared_1[(threadIdx.x_1 + 245)] = 0f32
+        }
+        attr [IterVar(threadIdx.x_2: int32, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 49;
+        kernel.shared_1: Buffer(kernel.shared, float32, [96], [], scope=&quot;shared&quot;)[threadIdx.x_2] = kernel[((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 12)*4608)) + (rc.outer.outer*36)) + (floormod(threadIdx.x_2, 12)*3))]
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 49;
+        if @tir.likely((threadIdx.x_2 &lt; 47), dtype=bool) {
+          kernel.shared_1[(threadIdx.x_2 + 49)] = kernel[((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 49), 12)*4608)) + (rc.outer.outer*36)) + (floormod((threadIdx.x_2 + 1), 12)*3))]
+        }
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[0]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[48]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[12]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[60]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[3]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[51]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[15]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[63]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[24]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[72]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[36]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[84]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[27]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[75]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[39]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[87]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[1]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[49]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[13]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[61]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[4]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[52]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[16]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[64]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[25]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[73]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[37]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[85]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[28]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[76]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[40]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[88]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[2]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[50]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[14]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[62]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[5]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[53]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[17]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[65]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[26]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[74]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[38]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[86]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[29]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[77]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[41]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[89]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[6]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[54]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[18]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[66]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[9]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[57]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[21]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[69]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[30]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[78]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[42]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[90]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[33]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[81]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[45]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[93]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[7]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[55]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[19]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[67]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[10]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[58]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[22]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[70]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[31]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[79]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[43]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[91]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[34]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[82]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[46]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[94]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[8]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[56]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[20]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[68]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[11]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[59]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[23]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[71]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[32]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[80]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[44]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[92]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[35]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[83]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[47]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[95]))
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 49;
+        pad_temp.shared_1[threadIdx.x_1] = @tir.if_then_else((7 &lt;= threadIdx.x_1), data[((cse_var_1 + threadIdx.x_1) - 7)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 49;
+        pad_temp.shared_1[(threadIdx.x_1 + 49)] = @tir.if_then_else(((1 &lt;= floormod((floordiv(threadIdx.x_1, 7) + 7), 9)) &amp;&amp; (floormod((floordiv(threadIdx.x_1, 7) + 7), 9) &lt; 8)), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 49), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 7), 9)*7)) + floormod(threadIdx.x_1, 7)) - 7)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 49;
+        pad_temp.shared_1[(threadIdx.x_1 + 98)] = @tir.if_then_else(((1 &lt;= floormod((floordiv(threadIdx.x_1, 7) + 5), 9)) &amp;&amp; (floormod((floordiv(threadIdx.x_1, 7) + 5), 9) &lt; 8)), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 98), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 5), 9)*7)) + floormod(threadIdx.x_1, 7)) - 7)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 49;
+        pad_temp.shared_1[(threadIdx.x_1 + 147)] = @tir.if_then_else(((1 &lt;= floormod((floordiv(threadIdx.x_1, 7) + 3), 9)) &amp;&amp; (floormod((floordiv(threadIdx.x_1, 7) + 3), 9) &lt; 8)), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 147), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 3), 9)*7)) + floormod(threadIdx.x_1, 7)) - 7)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 49;
+        pad_temp.shared_1[(threadIdx.x_1 + 196)] = data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 196), 63)*49)) + ((floordiv(threadIdx.x_1, 7) + 1)*7)) + floormod(threadIdx.x_1, 7)) - 7)]
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 49;
+        if @tir.likely((threadIdx.x_1 &lt; 7), dtype=bool) {
+          pad_temp.shared_1[(threadIdx.x_1 + 245)] = 0f32
+        }
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 49;
+        kernel.shared_1[threadIdx.x_2] = kernel[(((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 12)*4608)) + (rc.outer.outer*36)) + (floormod(threadIdx.x_2, 12)*3)) + 1)]
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 49;
+        if @tir.likely((threadIdx.x_2 &lt; 47), dtype=bool) {
+          kernel.shared_1[(threadIdx.x_2 + 49)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 49), 12)*4608)) + (rc.outer.outer*36)) + (floormod((threadIdx.x_2 + 1), 12)*3)) + 1)]
+        }
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[0]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[48]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[12]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[60]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[3]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[51]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[15]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[63]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[24]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[72]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[36]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[84]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[27]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[75]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[39]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[87]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[1]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[49]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[13]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[61]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[4]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[52]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[16]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[64]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[25]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[73]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[37]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[85]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[28]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[76]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[40]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[88]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[2]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[50]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[14]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[62]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[5]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[53]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[17]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[65]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[26]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[74]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[38]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[86]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[29]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[77]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[41]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[89]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[6]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[54]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[18]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[66]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[9]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[57]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[21]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[69]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[30]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[78]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[42]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[90]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[33]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[81]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[45]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[93]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[7]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[55]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[19]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[67]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[10]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[58]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[22]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[70]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[31]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[79]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[43]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[91]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[34]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[82]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[46]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[94]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[8]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[56]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[20]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[68]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[11]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[59]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[23]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[71]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[32]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[80]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[44]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[92]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[35]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[83]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[47]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[95]))
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 49;
+        pad_temp.shared_1[threadIdx.x_1] = @tir.if_then_else(((7 &lt;= threadIdx.x_1) &amp;&amp; (floormod(threadIdx.x_1, 7) &lt; 6)), data[((cse_var_1 + threadIdx.x_1) - 6)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 49;
+        pad_temp.shared_1[(threadIdx.x_1 + 49)] = @tir.if_then_else((((1 &lt;= floormod((floordiv(threadIdx.x_1, 7) + 7), 9)) &amp;&amp; (floormod((floordiv(threadIdx.x_1, 7) + 7), 9) &lt; 8)) &amp;&amp; (floormod(threadIdx.x_1, 7) &lt; 6)), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 49), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 7), 9)*7)) + floormod(threadIdx.x_1, 7)) - 6)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 49;
+        pad_temp.shared_1[(threadIdx.x_1 + 98)] = @tir.if_then_else((((1 &lt;= floormod((floordiv(threadIdx.x_1, 7) + 5), 9)) &amp;&amp; (floormod((floordiv(threadIdx.x_1, 7) + 5), 9) &lt; 8)) &amp;&amp; (floormod(threadIdx.x_1, 7) &lt; 6)), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 98), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 5), 9)*7)) + floormod(threadIdx.x_1, 7)) - 6)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 49;
+        pad_temp.shared_1[(threadIdx.x_1 + 147)] = @tir.if_then_else((((1 &lt;= floormod((floordiv(threadIdx.x_1, 7) + 3), 9)) &amp;&amp; (floormod((floordiv(threadIdx.x_1, 7) + 3), 9) &lt; 8)) &amp;&amp; (floormod(threadIdx.x_1, 7) &lt; 6)), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 147), 63)*49)) + (floormod((floordiv(threadIdx.x_1, 7) + 3), 9)*7)) + floormod(threadIdx.x_1, 7)) - 6)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 49;
+        pad_temp.shared_1[(threadIdx.x_1 + 196)] = @tir.if_then_else((floormod(threadIdx.x_1, 7) &lt; 6), data[((((cse_var_1 + (floordiv((threadIdx.x_1 + 196), 63)*49)) + ((floordiv(threadIdx.x_1, 7) + 1)*7)) + floormod(threadIdx.x_1, 7)) - 6)], 0f32, dtype=float32)
+        attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 49;
+        if @tir.likely((threadIdx.x_1 &lt; 7), dtype=bool) {
+          pad_temp.shared_1[(threadIdx.x_1 + 245)] = 0f32
+        }
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 49;
+        kernel.shared_1[threadIdx.x_2] = kernel[(((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 12)*4608)) + (rc.outer.outer*36)) + (floormod(threadIdx.x_2, 12)*3)) + 2)]
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 49;
+        if @tir.likely((threadIdx.x_2 &lt; 47), dtype=bool) {
+          kernel.shared_1[(threadIdx.x_2 + 49)] = kernel[(((((blockIdx.x*36864) + (floordiv((threadIdx.x_2 + 49), 12)*4608)) + (rc.outer.outer*36)) + (floormod((threadIdx.x_2 + 1), 12)*3)) + 2)]
+        }
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[0]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[48]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[12]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[60]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[3]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[51]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[15]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[63]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[24]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[72]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[36]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[threadIdx.x]*kernel.shared_1[84]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[27]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[75]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[39]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 63)]*kernel.shared_1[87]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[1]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[49]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[13]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[61]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[4]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[52]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[16]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[64]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[25]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[73]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[37]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 7)]*kernel.shared_1[85]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[28]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[76]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[40]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 70)]*kernel.shared_1[88]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[2]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[50]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[14]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[62]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[5]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[53]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[17]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[65]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[26]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[74]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[38]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 14)]*kernel.shared_1[86]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[29]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[77]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[41]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 77)]*kernel.shared_1[89]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[6]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[54]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[18]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[66]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[9]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[57]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[21]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[69]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[30]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[78]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[42]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 126)]*kernel.shared_1[90]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[33]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[81]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[45]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 189)]*kernel.shared_1[93]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[7]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[55]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[19]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[67]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[10]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[58]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[22]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[70]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[31]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[79]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[43]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 133)]*kernel.shared_1[91]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[34]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[82]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[46]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 196)]*kernel.shared_1[94]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[8]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[56]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[20]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[68]))
+        conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[11]))
+        conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[59]))
+        conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[23]))
+        conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[71]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[32]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[80]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[44]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 140)]*kernel.shared_1[92]))
+        conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[35]))
+        conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[83]))
+        conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[47]))
+        conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(threadIdx.x + 203)]*kernel.shared_1[95]))
       }
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*4)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 16)]))
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 1)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 17)]))
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 2)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 18)]))
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 3)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 19)]))
-      attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 196;
-      pad_temp.shared_1[threadIdx.x_1] = @tir.if_then_else((7 &lt;= floormod(threadIdx.x_1, 49)), data[(((rc.outer.outer*196) + threadIdx.x_1) - 7)], 0f32, dtype=float32)
-      attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 196;
-      if @tir.likely((threadIdx.x_2 &lt; 32), dtype=bool) {
-        kernel.shared_1[threadIdx.x_2] = kernel[(((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 4)*4608)) + (rc.outer.outer*36)) + (floormod(threadIdx.x_2, 4)*9)) + 1)]
-      }
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*4)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 16)]))
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 1)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 17)]))
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 2)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 18)]))
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 3)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 19)]))
-      attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 196;
-      pad_temp.shared_1[threadIdx.x_1] = @tir.if_then_else(((7 &lt;= floormod(threadIdx.x_1, 49)) &amp;&amp; (floormod(threadIdx.x_1, 7) &lt; 6)), data[(((rc.outer.outer*196) + threadIdx.x_1) - 6)], 0f32, dtype=float32)
-      attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 196;
-      if @tir.likely((threadIdx.x_2 &lt; 32), dtype=bool) {
-        kernel.shared_1[threadIdx.x_2] = kernel[(((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 4)*4608)) + (rc.outer.outer*36)) + (floormod(threadIdx.x_2, 4)*9)) + 2)]
-      }
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*4)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 16)]))
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 1)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 17)]))
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 2)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 18)]))
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 3)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 19)]))
-      attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 196;
-      pad_temp.shared_1[threadIdx.x_1] = @tir.if_then_else((1 &lt;= floormod(threadIdx.x_1, 7)), data[(((rc.outer.outer*196) + threadIdx.x_1) - 1)], 0f32, dtype=float32)
-      attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 196;
-      if @tir.likely((threadIdx.x_2 &lt; 32), dtype=bool) {
-        kernel.shared_1[threadIdx.x_2] = kernel[(((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 4)*4608)) + (rc.outer.outer*36)) + (floormod(threadIdx.x_2, 4)*9)) + 3)]
-      }
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*4)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 16)]))
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 1)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 17)]))
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 2)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 18)]))
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 3)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 19)]))
-      attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 196;
-      pad_temp.shared_1[threadIdx.x_1] = data[((rc.outer.outer*196) + threadIdx.x_1)]
-      attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 196;
-      if @tir.likely((threadIdx.x_2 &lt; 32), dtype=bool) {
-        kernel.shared_1[threadIdx.x_2] = kernel[(((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 4)*4608)) + (rc.outer.outer*36)) + (floormod(threadIdx.x_2, 4)*9)) + 4)]
-      }
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*4)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 16)]))
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 1)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 17)]))
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 2)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 18)]))
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 3)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 19)]))
-      attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 196;
-      pad_temp.shared_1[threadIdx.x_1] = @tir.if_then_else((floormod(threadIdx.x_1, 7) &lt; 6), data[(((rc.outer.outer*196) + threadIdx.x_1) + 1)], 0f32, dtype=float32)
-      attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 196;
-      if @tir.likely((threadIdx.x_2 &lt; 32), dtype=bool) {
-        kernel.shared_1[threadIdx.x_2] = kernel[(((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 4)*4608)) + (rc.outer.outer*36)) + (floormod(threadIdx.x_2, 4)*9)) + 5)]
-      }
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*4)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 16)]))
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 1)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 17)]))
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 2)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 18)]))
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 3)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 19)]))
-      attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 196;
-      pad_temp.shared_1[threadIdx.x_1] = @tir.if_then_else(((floormod(threadIdx.x_1, 49) &lt; 42) &amp;&amp; (1 &lt;= floormod(threadIdx.x_1, 7))), data[(((rc.outer.outer*196) + threadIdx.x_1) + 6)], 0f32, dtype=float32)
-      attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 196;
-      if @tir.likely((threadIdx.x_2 &lt; 32), dtype=bool) {
-        kernel.shared_1[threadIdx.x_2] = kernel[(((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 4)*4608)) + (rc.outer.outer*36)) + (floormod(threadIdx.x_2, 4)*9)) + 6)]
-      }
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*4)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 16)]))
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 1)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 17)]))
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 2)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 18)]))
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 3)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 19)]))
-      attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 196;
-      pad_temp.shared_1[threadIdx.x_1] = @tir.if_then_else((floormod(threadIdx.x_1, 49) &lt; 42), data[(((rc.outer.outer*196) + threadIdx.x_1) + 7)], 0f32, dtype=float32)
-      attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 196;
-      if @tir.likely((threadIdx.x_2 &lt; 32), dtype=bool) {
-        kernel.shared_1[threadIdx.x_2] = kernel[(((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 4)*4608)) + (rc.outer.outer*36)) + (floormod(threadIdx.x_2, 4)*9)) + 7)]
-      }
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*4)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 16)]))
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 1)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 17)]))
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 2)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 18)]))
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 3)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 19)]))
-      attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 196;
-      pad_temp.shared_1[threadIdx.x_1] = @tir.if_then_else(((floormod(threadIdx.x_1, 49) &lt; 42) &amp;&amp; (floormod(threadIdx.x_1, 7) &lt; 6)), data[(((rc.outer.outer*196) + threadIdx.x_1) + 8)], 0f32, dtype=float32)
-      attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 196;
-      if @tir.likely((threadIdx.x_2 &lt; 32), dtype=bool) {
-        kernel.shared_1[threadIdx.x_2] = kernel[(((((blockIdx.x*36864) + (floordiv(threadIdx.x_2, 4)*4608)) + (rc.outer.outer*36)) + (floormod(threadIdx.x_2, 4)*9)) + 8)]
-      }
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[(floordiv(threadIdx.x, 49)*4)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[floormod(threadIdx.x, 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 16)]))
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 1)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 49)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 17)]))
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 2)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 98)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 18)]))
-      conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 3)]))
-      conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(floormod(threadIdx.x, 49) + 147)]*kernel.shared_1[((floordiv(threadIdx.x, 49)*4) + 19)]))
     }
-    compute[((blockIdx.x*392) + threadIdx.x)] = max((conv2d_nchw_1[0] + bias[((blockIdx.x*8) + floordiv(threadIdx.x, 49))]), 0f32)
-    compute[(((blockIdx.x*392) + threadIdx.x) + 196)] = max((conv2d_nchw_1[1] + bias[(((blockIdx.x*8) + floordiv(threadIdx.x, 49)) + 4)]), 0f32)
+    for (i1.inner: int32, 0, 4) {
+      compute[(((blockIdx.x*392) + (i1.inner*49)) + threadIdx.x)] = max((conv2d_nchw_1[i1.inner] + bias[((blockIdx.x*8) + i1.inner)]), 0f32)
+      compute[((((blockIdx.x*392) + (i1.inner*49)) + threadIdx.x) + 196)] = max((conv2d_nchw_1[(i1.inner + 4)] + bias[(((blockIdx.x*8) + i1.inner) + 4)]), 0f32)
+    }
   }
 }
 </pre></div>
@@ -676,7 +909,7 @@ cooperative fetching, unrolling and operator fusion.</p>
 <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: 0.381 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 0.228 ms
 </pre></div>
 </div>
 </div>
@@ -705,9 +938,9 @@ 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_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_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=2)
+conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_i, factor=1)
 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)
@@ -717,18 +950,18 @@ conv2d_nchw_xx_o_i, conv2d_nchw_xx_i = s[conv2d_nchw].split(conv2d_nchw_xx, fact
 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_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=1)
-conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_o_i, factor=4)
+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_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=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=1)
 conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=1)
 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_i, compute_i1_i = s[compute].split(compute_i1, factor=4)
+compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=1)
 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)
@@ -754,14 +987,14 @@ s[compute].bind(compute_i0_o_i_i1_o_i_fused_i2_o_i_fused_i3_o_i_fused, te.thread
 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=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=49)
 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=49)
 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;, 512)
+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:
@@ -779,142 +1012,354 @@ 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) {
-  float conv2d_nchw[2];
-  __shared__ float pad_temp_shared[196];
-  __shared__ float kernel_shared[32];
+extern &quot;C&quot; __global__ void __launch_bounds__(49) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
+  float conv2d_nchw[8];
+  __shared__ float pad_temp_shared[252];
+  __shared__ float kernel_shared[96];
   conv2d_nchw[0] = 0.000000e+00f;
+  conv2d_nchw[4] = 0.000000e+00f;
   conv2d_nchw[1] = 0.000000e+00f;
+  conv2d_nchw[5] = 0.000000e+00f;
+  conv2d_nchw[2] = 0.000000e+00f;
+  conv2d_nchw[6] = 0.000000e+00f;
+  conv2d_nchw[3] = 0.000000e+00f;
+  conv2d_nchw[7] = 0.000000e+00f;
   for (int rc_outer_outer = 0; rc_outer_outer &lt; 128; ++rc_outer_outer) {
     __syncthreads();
-    pad_temp_shared[((int)threadIdx.x)] = (((7 &lt;= (((int)threadIdx.x) % 49)) &amp;&amp; (1 &lt;= (((int)threadIdx.x) % 7))) ? data[(((rc_outer_outer * 196) + ((int)threadIdx.x)) - 8)] : 0.000000e+00f);
-    if (((int)threadIdx.x) &lt; 32) {
-      kernel_shared[((int)threadIdx.x)] = kernel[((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) &gt;&gt; 2) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) &amp; 3) * 9))];
+    pad_temp_shared[((int)threadIdx.x)] = (((7 &lt;= ((int)threadIdx.x)) &amp;&amp; (1 &lt;= (((int)threadIdx.x) % 7))) ? data[(((rc_outer_outer * 196) + ((int)threadIdx.x)) - 8)] : 0.000000e+00f);
+    pad_temp_shared[(((int)threadIdx.x) + 49)] = ((((1 &lt;= (((((int)threadIdx.x) / 7) + 7) % 9)) &amp;&amp; ((((((int)threadIdx.x) / 7) + 7) % 9) &lt; 8)) &amp;&amp; (1 &lt;= (((int)threadIdx.x) % 7))) ? data[(((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 49) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 7) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
+    pad_temp_shared[(((int)threadIdx.x) + 98)] = ((((1 &lt;= (((((int)threadIdx.x) / 7) + 5) % 9)) &amp;&amp; ((((((int)threadIdx.x) / 7) + 5) % 9) &lt; 8)) &amp;&amp; (1 &lt;= (((int)threadIdx.x) % 7))) ? data[(((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 98) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 5) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
+    pad_temp_shared[(((int)threadIdx.x) + 147)] = ((((1 &lt;= (((((int)threadIdx.x) / 7) + 3) % 9)) &amp;&amp; ((((((int)threadIdx.x) / 7) + 3) % 9) &lt; 8)) &amp;&amp; (1 &lt;= (((int)threadIdx.x) % 7))) ? data[(((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 147) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 3) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 8)] : 0.000000e+00f);
+    pad_temp_shared[(((int)threadIdx.x) + 196)] = ((1 &lt;= (((int)threadIdx.x) % 7)) ? data[((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 196) / 63) * 49)) + ((int)threadIdx.x)) - 1)] : 0.000000e+00f);
+    if (((int)threadIdx.x) &lt; 7) {
+      pad_temp_shared[(((int)threadIdx.x) + 245)] = 0.000000e+00f;
     }
-    __syncthreads();
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 4)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 16)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 1)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 17)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 2)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 18)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 3)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 19)]));
-    __syncthreads();
-    pad_temp_shared[((int)threadIdx.x)] = ((7 &lt;= (((int)threadIdx.x) % 49)) ? data[(((rc_outer_outer * 196) + ((int)threadIdx.x)) - 7)] : 0.000000e+00f);
-    if (((int)threadIdx.x) &lt; 32) {
-      kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) &gt;&gt; 2) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) &amp; 3) * 9)) + 1)];
+    kernel_shared[((int)threadIdx.x)] = kernel[((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) / 12) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) % 12) * 3))];
+    if (((int)threadIdx.x) &lt; 47) {
+      kernel_shared[(((int)threadIdx.x) + 49)] = kernel[((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 49) / 12) * 4608)) + (rc_outer_outer * 36)) + (((((int)threadIdx.x) + 1) % 12) * 3))];
     }
     __syncthreads();
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 4)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 16)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 1)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 17)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 2)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 18)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 3)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 19)]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[0]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[48]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[12]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[60]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[3]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[51]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[15]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[63]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[24]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[72]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[36]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[84]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[27]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[75]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[39]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[87]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[1]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[49]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[13]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[61]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[4]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[52]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[16]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[64]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[25]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[73]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[37]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[85]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[28]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[76]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[40]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[88]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[2]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[50]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[14]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[62]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[5]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[53]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[17]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[65]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[26]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[74]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[38]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[86]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[29]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[77]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[41]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[89]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[6]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[54]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[18]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[66]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[9]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[57]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[21]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[69]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[30]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[78]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[42]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[90]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[33]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[81]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[45]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[93]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[7]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[55]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[19]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[67]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[10]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[58]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[22]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[70]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[31]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[79]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[43]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[91]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[34]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[82]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[46]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[94]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[8]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[56]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[20]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[68]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[11]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[59]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[23]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[71]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[32]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[80]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[44]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[92]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[35]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[83]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[47]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[95]));
     __syncthreads();
-    pad_temp_shared[((int)threadIdx.x)] = (((7 &lt;= (((int)threadIdx.x) % 49)) &amp;&amp; ((((int)threadIdx.x) % 7) &lt; 6)) ? data[(((rc_outer_outer * 196) + ((int)threadIdx.x)) - 6)] : 0.000000e+00f);
-    if (((int)threadIdx.x) &lt; 32) {
-      kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) &gt;&gt; 2) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) &amp; 3) * 9)) + 2)];
+    pad_temp_shared[((int)threadIdx.x)] = ((7 &lt;= ((int)threadIdx.x)) ? data[(((rc_outer_outer * 196) + ((int)threadIdx.x)) - 7)] : 0.000000e+00f);
+    pad_temp_shared[(((int)threadIdx.x) + 49)] = (((1 &lt;= (((((int)threadIdx.x) / 7) + 7) % 9)) &amp;&amp; ((((((int)threadIdx.x) / 7) + 7) % 9) &lt; 8)) ? data[(((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 49) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 7) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 7)] : 0.000000e+00f);
+    pad_temp_shared[(((int)threadIdx.x) + 98)] = (((1 &lt;= (((((int)threadIdx.x) / 7) + 5) % 9)) &amp;&amp; ((((((int)threadIdx.x) / 7) + 5) % 9) &lt; 8)) ? data[(((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 98) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 5) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 7)] : 0.000000e+00f);
+    pad_temp_shared[(((int)threadIdx.x) + 147)] = (((1 &lt;= (((((int)threadIdx.x) / 7) + 3) % 9)) &amp;&amp; ((((((int)threadIdx.x) / 7) + 3) % 9) &lt; 8)) ? data[(((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 147) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 3) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 7)] : 0.000000e+00f);
+    pad_temp_shared[(((int)threadIdx.x) + 196)] = data[(((rc_outer_outer * 196) + (((((int)threadIdx.x) + 196) / 63) * 49)) + ((int)threadIdx.x))];
+    if (((int)threadIdx.x) &lt; 7) {
+      pad_temp_shared[(((int)threadIdx.x) + 245)] = 0.000000e+00f;
     }
-    __syncthreads();
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 4)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 16)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 1)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 17)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 2)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 18)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 3)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 19)]));
-    __syncthreads();
-    pad_temp_shared[((int)threadIdx.x)] = ((1 &lt;= (((int)threadIdx.x) % 7)) ? data[(((rc_outer_outer * 196) + ((int)threadIdx.x)) - 1)] : 0.000000e+00f);
-    if (((int)threadIdx.x) &lt; 32) {
-      kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) &gt;&gt; 2) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) &amp; 3) * 9)) + 3)];
+    kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) / 12) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) % 12) * 3)) + 1)];
+    if (((int)threadIdx.x) &lt; 47) {
+      kernel_shared[(((int)threadIdx.x) + 49)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 49) / 12) * 4608)) + (rc_outer_outer * 36)) + (((((int)threadIdx.x) + 1) % 12) * 3)) + 1)];
     }
     __syncthreads();
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 4)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 16)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 1)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 17)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 2)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 18)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 3)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 19)]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[0]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[48]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[12]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[60]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[3]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[51]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[15]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[63]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[24]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[72]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[36]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[84]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[27]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[75]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[39]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[87]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[1]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[49]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[13]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[61]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[4]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[52]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[16]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[64]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[25]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[73]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[37]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[85]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[28]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[76]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[40]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[88]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[2]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[50]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[14]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[62]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[5]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[53]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[17]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[65]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[26]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[74]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[38]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[86]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[29]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[77]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[41]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[89]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[6]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[54]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[18]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[66]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[9]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[57]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[21]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[69]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[30]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[78]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[42]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[90]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[33]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[81]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[45]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[93]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[7]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[55]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[19]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[67]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[10]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[58]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[22]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[70]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[31]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[79]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[43]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[91]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[34]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[82]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[46]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[94]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[8]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[56]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[20]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[68]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[11]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[59]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[23]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[71]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[32]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[80]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[44]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[92]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[35]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[83]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[47]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[95]));
     __syncthreads();
-    pad_temp_shared[((int)threadIdx.x)] = data[((rc_outer_outer * 196) + ((int)threadIdx.x))];
-    if (((int)threadIdx.x) &lt; 32) {
-      kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) &gt;&gt; 2) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) &amp; 3) * 9)) + 4)];
+    pad_temp_shared[((int)threadIdx.x)] = (((7 &lt;= ((int)threadIdx.x)) &amp;&amp; ((((int)threadIdx.x) % 7) &lt; 6)) ? data[(((rc_outer_outer * 196) + ((int)threadIdx.x)) - 6)] : 0.000000e+00f);
+    pad_temp_shared[(((int)threadIdx.x) + 49)] = ((((1 &lt;= (((((int)threadIdx.x) / 7) + 7) % 9)) &amp;&amp; ((((((int)threadIdx.x) / 7) + 7) % 9) &lt; 8)) &amp;&amp; ((((int)threadIdx.x) % 7) &lt; 6)) ? data[(((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 49) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 7) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 6)] : 0.000000e+00f);
+    pad_temp_shared[(((int)threadIdx.x) + 98)] = ((((1 &lt;= (((((int)threadIdx.x) / 7) + 5) % 9)) &amp;&amp; ((((((int)threadIdx.x) / 7) + 5) % 9) &lt; 8)) &amp;&amp; ((((int)threadIdx.x) % 7) &lt; 6)) ? data[(((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 98) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 5) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 6)] : 0.000000e+00f);
+    pad_temp_shared[(((int)threadIdx.x) + 147)] = ((((1 &lt;= (((((int)threadIdx.x) / 7) + 3) % 9)) &amp;&amp; ((((((int)threadIdx.x) / 7) + 3) % 9) &lt; 8)) &amp;&amp; ((((int)threadIdx.x) % 7) &lt; 6)) ? data[(((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 147) / 63) * 49)) + ((((((int)threadIdx.x) / 7) + 3) % 9) * 7)) + (((int)threadIdx.x) % 7)) - 6)] : 0.000000e+00f);
+    pad_temp_shared[(((int)threadIdx.x) + 196)] = (((((int)threadIdx.x) % 7) &lt; 6) ? data[((((rc_outer_outer * 196) + (((((int)threadIdx.x) + 196) / 63) * 49)) + ((int)threadIdx.x)) + 1)] : 0.000000e+00f);
+    if (((int)threadIdx.x) &lt; 7) {
+      pad_temp_shared[(((int)threadIdx.x) + 245)] = 0.000000e+00f;
     }
-    __syncthreads();
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 4)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 16)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 1)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 17)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 2)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 18)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 3)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 19)]));
-    __syncthreads();
-    pad_temp_shared[((int)threadIdx.x)] = (((((int)threadIdx.x) % 7) &lt; 6) ? data[(((rc_outer_outer * 196) + ((int)threadIdx.x)) + 1)] : 0.000000e+00f);
-    if (((int)threadIdx.x) &lt; 32) {
-      kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) &gt;&gt; 2) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) &amp; 3) * 9)) + 5)];
-    }
-    __syncthreads();
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 4)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 16)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 1)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 17)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 2)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 18)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 3)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 19)]));
-    __syncthreads();
-    pad_temp_shared[((int)threadIdx.x)] = ((((((int)threadIdx.x) % 49) &lt; 42) &amp;&amp; (1 &lt;= (((int)threadIdx.x) % 7))) ? data[(((rc_outer_outer * 196) + ((int)threadIdx.x)) + 6)] : 0.000000e+00f);
-    if (((int)threadIdx.x) &lt; 32) {
-      kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) &gt;&gt; 2) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) &amp; 3) * 9)) + 6)];
-    }
-    __syncthreads();
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 4)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 16)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 1)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 17)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 2)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 18)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 3)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 19)]));
-    __syncthreads();
-    pad_temp_shared[((int)threadIdx.x)] = (((((int)threadIdx.x) % 49) &lt; 42) ? data[(((rc_outer_outer * 196) + ((int)threadIdx.x)) + 7)] : 0.000000e+00f);
-    if (((int)threadIdx.x) &lt; 32) {
-      kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) &gt;&gt; 2) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) &amp; 3) * 9)) + 7)];
-    }
-    __syncthreads();
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 4)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 16)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 1)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 17)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 2)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 18)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 3)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 19)]));
-    __syncthreads();
-    pad_temp_shared[((int)threadIdx.x)] = ((((((int)threadIdx.x) % 49) &lt; 42) &amp;&amp; ((((int)threadIdx.x) % 7) &lt; 6)) ? data[(((rc_outer_outer * 196) + ((int)threadIdx.x)) + 8)] : 0.000000e+00f);
-    if (((int)threadIdx.x) &lt; 32) {
-      kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) &gt;&gt; 2) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) &amp; 3) * 9)) + 8)];
+    kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) * 36864) + ((((int)threadIdx.x) / 12) * 4608)) + (rc_outer_outer * 36)) + ((((int)threadIdx.x) % 12) * 3)) + 2)];
+    if (((int)threadIdx.x) &lt; 47) {
+      kernel_shared[(((int)threadIdx.x) + 49)] = kernel[(((((((int)blockIdx.x) * 36864) + (((((int)threadIdx.x) + 49) / 12) * 4608)) + (rc_outer_outer * 36)) + (((((int)threadIdx.x) + 1) % 12) * 3)) + 2)];
     }
     __syncthreads();
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[((((int)threadIdx.x) / 49) * 4)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) % 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 16)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 1)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 49)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 17)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 2)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 98)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 18)]));
-    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 3)]));
-    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((((int)threadIdx.x) % 49) + 147)] * kernel_shared[(((((int)threadIdx.x) / 49) * 4) + 19)]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[0]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[48]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[12]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[60]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[3]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[51]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[15]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[63]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[24]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[72]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[36]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((int)threadIdx.x)] * kernel_shared[84]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[27]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[75]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[39]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 63)] * kernel_shared[87]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[1]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[49]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[13]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[61]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[4]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[52]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[16]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[64]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[25]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[73]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[37]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 7)] * kernel_shared[85]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[28]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[76]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[40]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 70)] * kernel_shared[88]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[2]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[50]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[14]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[62]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[5]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[53]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[17]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[65]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[26]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[74]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[38]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 14)] * kernel_shared[86]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[29]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[77]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[41]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 77)] * kernel_shared[89]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[6]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[54]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[18]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[66]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[9]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[57]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[21]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[69]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[30]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[78]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[42]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 126)] * kernel_shared[90]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[33]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[81]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[45]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 189)] * kernel_shared[93]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[7]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[55]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[19]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[67]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[10]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[58]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[22]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[70]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[31]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[79]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[43]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 133)] * kernel_shared[91]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[34]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[82]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[46]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 196)] * kernel_shared[94]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[8]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[56]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[20]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[68]));
+    conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[11]));
+    conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[59]));
+    conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[23]));
+    conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[71]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[32]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[80]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[44]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 140)] * kernel_shared[92]));
+    conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[35]));
+    conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[83]));
+    conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[47]));
+    conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((int)threadIdx.x) + 203)] * kernel_shared[95]));
+  }
+  for (int i1_inner = 0; i1_inner &lt; 4; ++i1_inner) {
+    compute[(((((int)blockIdx.x) * 392) + (i1_inner * 49)) + ((int)threadIdx.x))] = max((conv2d_nchw[i1_inner] + bias[((((int)blockIdx.x) * 8) + i1_inner)]), 0.000000e+00f);
+    compute[((((((int)blockIdx.x) * 392) + (i1_inner * 49)) + ((int)threadIdx.x)) + 196)] = max((conv2d_nchw[(i1_inner + 4)] + bias[(((((int)blockIdx.x) * 8) + i1_inner) + 4)]), 0.000000e+00f);
   }
-  compute[((((int)blockIdx.x) * 392) + ((int)threadIdx.x))] = max((conv2d_nchw[0] + bias[((((int)blockIdx.x) * 8) + (((int)threadIdx.x) / 49))]), 0.000000e+00f);
-  compute[(((((int)blockIdx.x) * 392) + ((int)threadIdx.x)) + 196)] = max((conv2d_nchw[1] + bias[(((((int)blockIdx.x) * 8) + (((int)threadIdx.x) / 49)) + 4)]), 0.000000e+00f);
 }
 </pre></div>
 </div>
@@ -950,7 +1395,7 @@ 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> ( 5 minutes  45.244 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 5 minutes  56.080 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 20f689bdad..3537d3dffc 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
@@ -915,7 +915,7 @@ 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)
-   8.2434       8.2416       8.2486       8.2401       0.0037
+   8.2229       8.2286       8.2300       8.2102       0.0090
 </pre></div>
 </div>
 </div>
@@ -937,7 +937,7 @@ 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  4.412 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  4.140 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 ed5d619395..0ffdeb099b 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
@@ -934,7 +934,7 @@ 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)
-  752.3158     750.6468     756.5245     749.7761      2.9972
+  770.9234     770.6793     775.3688     766.7221      3.5342
 </pre></div>
 </div>
 </div>
@@ -956,7 +956,7 @@ 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  33.722 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  34.623 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 066a91f7ec..afdaecd9ce 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
@@ -632,104 +632,30 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
              placeholder_4: Buffer(placeholder_14: Pointer(float32), float32, [65536], []),
              compute: Buffer(compute_2: Pointer(float32), float32, [65536], [])}
   buffer_map = {placeholder_5: placeholder, placeholder_6: placeholder_1, placeholder_7: placeholder_2, placeholder_8: placeholder_3, placeholder_9: placeholder_4, compute_1: compute}
-  preflattened_buffer_map = {placeholder_7: placeholder_15: Buffer(placeholder_12, int32, [4916], []), placeholder_5: placeholder_16: Buffer(placeholder_10, float32, [128, 256], []), placeholder_8: placeholder_17: Buffer(placeholder_13, int32, [33], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], []), placeholder_9: placeholder_18: Buffer(placeholder_14, float32, [128, 512], []), placeholder_6: placeholder_19: Buffer(placeholder_11, float32, [4916, 16, 1], [])} {
-  for (i0.outer: int32, 0, 2) &quot;parallel&quot; {
-    allocate(compute_4: Pointer(global float32), float32, [1024]), storage_scope = global;
-    for (i1.outer: int32, 0, 32) {
-      for (i.outer.inner: int32, 0, 2) {
-        for (i.inner.init: int32, 0, 32) {
-          let cse_var_1: int32 = ((i.outer.inner*512) + (i.inner.init*16))
-           {
-            compute_5: Buffer(compute_4, float32, [1024], [])[cse_var_1] = 0f32
-            compute_5[(cse_var_1 + 1)] = 0f32
-            compute_5[(cse_var_1 + 2)] = 0f32
-            compute_5[(cse_var_1 + 3)] = 0f32
-            compute_5[(cse_var_1 + 4)] = 0f32
-            compute_5[(cse_var_1 + 5)] = 0f32
-            compute_5[(cse_var_1 + 6)] = 0f32
-            compute_5[(cse_var_1 + 7)] = 0f32
-            compute_5[(cse_var_1 + 8)] = 0f32
-            compute_5[(cse_var_1 + 9)] = 0f32
-            compute_5[(cse_var_1 + 10)] = 0f32
-            compute_5[(cse_var_1 + 11)] = 0f32
-            compute_5[(cse_var_1 + 12)] = 0f32
-            compute_5[(cse_var_1 + 13)] = 0f32
-            compute_5[(cse_var_1 + 14)] = 0f32
-            compute_5[(cse_var_1 + 15)] = 0f32
-          }
-        }
-        for (elem_idx: int32, 0, (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])) {
-          for (i.inner: int32, 0, 32) {
-            if @tir.likely((elem_idx &lt; (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-              let cse_var_2: int32 = ((i.outer.inner*512) + (i.inner*16))
-              compute_5[cse_var_2] = (compute_5[cse_var_2] + (placeholder_1[((placeholder_3[i1.outer]*16) + (elem_idx*16))]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
-            }
-            if @tir.likely((elem_idx &lt; (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-              let cse_var_3: int32 = (((i.outer.inner*512) + (i.inner*16)) + 1)
-              compute_5[cse_var_3] = (compute_5[cse_var_3] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 1)]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
-            }
-            if @tir.likely((elem_idx &lt; (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-              let cse_var_4: int32 = (((i.outer.inner*512) + (i.inner*16)) + 2)
-              compute_5[cse_var_4] = (compute_5[cse_var_4] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 2)]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
-            }
-            if @tir.likely((elem_idx &lt; (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-              let cse_var_5: int32 = (((i.outer.inner*512) + (i.inner*16)) + 3)
-              compute_5[cse_var_5] = (compute_5[cse_var_5] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 3)]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
-            }
-            if @tir.likely((elem_idx &lt; (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-              let cse_var_6: int32 = (((i.outer.inner*512) + (i.inner*16)) + 4)
-              compute_5[cse_var_6] = (compute_5[cse_var_6] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 4)]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
-            }
-            if @tir.likely((elem_idx &lt; (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-              let cse_var_7: int32 = (((i.outer.inner*512) + (i.inner*16)) + 5)
-              compute_5[cse_var_7] = (compute_5[cse_var_7] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 5)]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
-            }
-            if @tir.likely((elem_idx &lt; (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-              let cse_var_8: int32 = (((i.outer.inner*512) + (i.inner*16)) + 6)
-              compute_5[cse_var_8] = (compute_5[cse_var_8] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 6)]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
-            }
-            if @tir.likely((elem_idx &lt; (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-              let cse_var_9: int32 = (((i.outer.inner*512) + (i.inner*16)) + 7)
-              compute_5[cse_var_9] = (compute_5[cse_var_9] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 7)]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+  preflattened_buffer_map = {placeholder_7: placeholder_15: Buffer(placeholder_12, int32, [4916], []), placeholder_6: placeholder_16: Buffer(placeholder_11, float32, [4916, 16, 1], []), compute_1: compute_3: Buffer(compute_2, float32, [128, 512], []), placeholder_5: placeholder_17: Buffer(placeholder_10, float32, [128, 256], []), placeholder_8: placeholder_18: Buffer(placeholder_13, int32, [33], []), placeholder_9: placeholder_19: Buffer(placeholder_14, float32, [128, 512], [])} {
+  for (i0.outer.i1.outer.fused: int32, 0, 64) &quot;parallel&quot; {
+    allocate(compute_4: Pointer(global float32), float32, [1024]), storage_scope = global {
+      for (i.outer.inner: int32, 0, 8) {
+        for (nb_j.inner: int32, 0, 2) {
+          for (i.inner.init: int32, 0, 4) {
+            for (j.init: int32, 0, 16) {
+              compute_5: Buffer(compute_4, float32, [1024], [])[((((i.outer.inner*128) + (i.inner.init*32)) + (nb_j.inner*16)) + j.init)] = 0f32
             }
-            if @tir.likely((elem_idx &lt; (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-              let cse_var_10: int32 = (((i.outer.inner*512) + (i.inner*16)) + 8)
-              compute_5[cse_var_10] = (compute_5[cse_var_10] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 8)]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
-            }
-            if @tir.likely((elem_idx &lt; (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-              let cse_var_11: int32 = (((i.outer.inner*512) + (i.inner*16)) + 9)
-              compute_5[cse_var_11] = (compute_5[cse_var_11] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 9)]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
-            }
-            if @tir.likely((elem_idx &lt; (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-              let cse_var_12: int32 = (((i.outer.inner*512) + (i.inner*16)) + 10)
-              compute_5[cse_var_12] = (compute_5[cse_var_12] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 10)]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
-            }
-            if @tir.likely((elem_idx &lt; (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-              let cse_var_13: int32 = (((i.outer.inner*512) + (i.inner*16)) + 11)
-              compute_5[cse_var_13] = (compute_5[cse_var_13] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 11)]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
-            }
-            if @tir.likely((elem_idx &lt; (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-              let cse_var_14: int32 = (((i.outer.inner*512) + (i.inner*16)) + 12)
-              compute_5[cse_var_14] = (compute_5[cse_var_14] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 12)]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
-            }
-            if @tir.likely((elem_idx &lt; (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-              let cse_var_15: int32 = (((i.outer.inner*512) + (i.inner*16)) + 13)
-              compute_5[cse_var_15] = (compute_5[cse_var_15] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 13)]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
-            }
-            if @tir.likely((elem_idx &lt; (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-              let cse_var_16: int32 = (((i.outer.inner*512) + (i.inner*16)) + 14)
-              compute_5[cse_var_16] = (compute_5[cse_var_16] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 14)]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
-            }
-            if @tir.likely((elem_idx &lt; (placeholder_3[(i1.outer + 1)] - placeholder_3[i1.outer])), dtype=bool) {
-              let cse_var_17: int32 = (((i.outer.inner*512) + (i.inner*16)) + 15)
-              compute_5[cse_var_17] = (compute_5[cse_var_17] + (placeholder_1[(((placeholder_3[i1.outer]*16) + (elem_idx*16)) + 15)]*max(placeholder[((((i0.outer*16384) + (i.outer.inner*8192)) + (i.inner*256)) + placeholder_2[(placeholder_3[i1.outer] + elem_idx)])], 0f32)))
+          }
+          for (elem_idx: int32, 0, let cse_var_1: int32 = ((floormod(i0.outer.i1.outer.fused, 16)*2) + nb_j.inner) in (placeholder_3[(cse_var_1 + 1)] - placeholder_3[cse_var_1])) {
+            for (i.inner: int32, 0, 4) {
+              for (j: int32, 0, 16) {
+                let cse_var_3: int32 = ((floormod(i0.outer.i1.outer.fused, 16)*2) + nb_j.inner)
+                let cse_var_2: int32 = ((((i.outer.inner*128) + (i.inner*32)) + (nb_j.inner*16)) + j)
+                compute_5[cse_var_2] = (compute_5[cse_var_2] + (placeholder_1[(((placeholder_3[cse_var_3]*16) + (elem_idx*16)) + j)]*max(placeholder[((((floordiv(i0.outer.i1.outer.fused, 16)*8192) + (i.outer.inner*1024)) + (i.inner*256)) + placeholder_2[(placeholder_3[cse_var_3] + elem_idx)])], 0f32)))
+              }
             }
           }
         }
       }
-      for (i0.inner: int32, 0, 64) {
-        let cse_var_18: int32 = (((i0.outer*32768) + (i0.inner*512)) + (i1.outer*16))
-        compute[ramp(cse_var_18, 1, 16)] = max((compute_5[ramp((i0.inner*16), 1, 16)] + placeholder_4[ramp(cse_var_18, 1, 16)]), broadcast(0f32, 16))
+      for (i0.inner: int32, 0, 32) {
+        let cse_var_4: int32 = (((floordiv(i0.outer.i1.outer.fused, 16)*16384) + (i0.inner*512)) + (floormod(i0.outer.i1.outer.fused, 16)*32))
+        compute[ramp(cse_var_4, 1, 32)] = max((compute_5[ramp((i0.inner*32), 1, 32)] + placeholder_4[ramp(cse_var_4, 1, 32)]), broadcast(0f32, 32))
       }
     }
   }
@@ -767,7 +693,7 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
 <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: 1.871 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 1.443 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 a6aca13df8..6141236900 100644
--- a/docs/how_to/tune_with_autotvm/sg_execution_times.html
+++ b/docs/how_to/tune_with_autotvm/sg_execution_times.html
@@ -340,7 +340,7 @@
             
   <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>00:38.715</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
+<p><strong>00:33.132</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 84%" />
@@ -349,11 +349,11 @@
 </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>00:38.679</p></td>
+<td><p>00:33.097</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>
-<td><p>00:00.021</p></td>
+<td><p>00:00.020</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="tune_relay_cuda.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-cuda-py"><span class="std std-ref">Auto-tuning a Convolutional Network for NVIDIA GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_cuda.py</span></code>)</p></td>
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 7da040a304..6b63b9c0dd 100644
--- a/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
+++ b/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
@@ -689,7 +689,7 @@ 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 871, 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, 8, 2]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 64, 2]), (&#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,6060643
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 64, 2, 1]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 8, 16]), (&#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,3615716
 No: 2   GFLOPS: 0.00/0.00       result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 588, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
@@ -812,9 +812,8 @@ 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 871, 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, 4]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 1, 512]), (&#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,9483670
-No: 3   GFLOPS: 7.79/7.79       result: MeasureResult(costs=(0.0297286395,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.632194995880127, timestamp=1668475003.5101364)        [(&#39;tile_f&#39;, [-1, 4, 1, 8]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 16, 4]), (&#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,470058
-No: 4   GFLOPS: 0.00/7.79       result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 1, 64, 8]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 7, 1, 1]), (&#39;tile_rc&#39;, [-1, 16, 4]), (&#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,8020043
+No: 3   GFLOPS: 0.00/0.00       result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 588, 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 540, in _build_func_common
@@ -936,8 +935,8 @@ 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 871, 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, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 7, 1, 1]), (&#39;tile_rc&#39;, [-1, 32, 2]), (&#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,634592
-No: 5   GFLOPS: 0.00/7.79       result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 4, 1, 1]), (&#39;tile_y&#39;, [-1, 7, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 32, 4]), (&#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,6863342
+No: 4   GFLOPS: 0.00/0.00       result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 588, 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 540, in _build_func_common
@@ -1059,8 +1058,8 @@ 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 871, 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, 64, 4]), (&#39;tile_y&#39;, [-1, 1, 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;, 512), (&#39;unroll_explicit&#39;, 1)],None,8544053
-No: 6   GFLOPS: 0.00/7.79       result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 4, 4, 4]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 4, 8]), (&#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,1072397
+No: 5   GFLOPS: 0.00/0.00       result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 588, 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 540, in _build_func_common
@@ -1182,8 +1181,9 @@ 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 871, 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, 16]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 16, 4]), (&#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;, 1)],None,9956499
-No: 7   GFLOPS: 0.00/7.79       result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 16, 4, 1]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 16, 16]), (&#39;tile_ry&#39;, [-1, 1, 3]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 512), (&#39;unroll_explicit&#39;, 0)],None,3426743
+No: 6   GFLOPS: 11.64/11.64     result: MeasureResult(costs=(0.019892005833333334,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.3444783687591553, timestamp=1668481233.3563986)       [(&#39;tile_f&#39;, [-1, 1, 4, 2]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 7, 1, 1]), (&#39;tile_rc&#39;, [-1, 2, 32]), (&#39;tile_ry&#39;, [-1, 3, 1]), (&#39;tile_rx&#39;, [-1, 3, 1]), (&#39;auto_unroll_max_step&#39;, 512), (&#39;unroll_explicit&#39;, 0)],None,2662732
+No: 7   GFLOPS: 0.00/11.64      result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 588, 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 540, in _build_func_common
@@ -1305,8 +1305,8 @@ 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 871, 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, 7, 1, 1]), (&#39;tile_rc&#39;, [-1, 8, 1]), (&#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;, 1)],None,5820058
-No: 8   GFLOPS: 0.00/7.79       result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 4, 32, 2]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 64, 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,21872
+No: 8   GFLOPS: 0.00/11.64      result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 588, 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 540, in _build_func_common
@@ -1428,9 +1428,8 @@ 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 871, 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, 4]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 2, 64]), (&#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;, 1)],None,6553931
-No: 9   GFLOPS: 24.56/24.56     result: MeasureResult(costs=(0.009426631357142857,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.1752407550811768, timestamp=1668475009.0452948)       [(&#39;tile_f&#39;, [-1, 4, 1, 1]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 8, 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;, 1)],None,8789442
-No: 10  GFLOPS: 0.00/24.56      result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 64, 8, 1]), (&#39;tile_y&#39;, [-1, 7, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 32, 4]), (&#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,1054493
+No: 9   GFLOPS: 0.00/11.64      result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 588, 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 540, in _build_func_common
@@ -1552,8 +1551,8 @@ 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 871, 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, 32]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 512, 1]), (&#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,4294152
-No: 11  GFLOPS: 0.00/24.56      result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 2, 16, 8]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 1, 512]), (&#39;tile_ry&#39;, [-1, 1, 3]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 512), (&#39;unroll_explicit&#39;, 1)],None,8711719
+No: 10  GFLOPS: 0.00/11.64      result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 588, 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 540, in _build_func_common
@@ -1675,8 +1674,8 @@ 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 871, 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, 1]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 256, 2]), (&#39;tile_ry&#39;, [-1, 1, 3]), (&#39;tile_rx&#39;, [-1, 1, 1]), (&#39;auto_unroll_max_step&#39;, 512), (&#39;unroll_explicit&#39;, 1)],None,7422823
-No: 12  GFLOPS: 0.00/24.56      result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 4, 1, 1]), (&#39;tile_y&#39;, [-1, 7, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 4, 128]), (&#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,2117502
+No: 11  GFLOPS: 0.00/11.64      result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 588, 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 540, in _build_func_common
@@ -1798,315 +1797,395 @@ 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 871, 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, 16, 4]), (&#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,2792842
-No: 13  GFLOPS: 0.00/24.56      result: Traceback (most recent call last):
-  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 738, in __call__
-    yield remote, remote.load_module(os.path.split(build_result.filename)[1])
-  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 702, in run_through_rpc
-    costs = time_f(*args).results
-  File &quot;/workspace/python/tvm/runtime/module.py&quot;, line 357, in evaluator
-    blob = feval(*args)
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 4, 2, 32]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 4, 128]), (&#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,4634932
+No: 12  GFLOPS: 0.00/11.64      result: Traceback (most recent call last):
+  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 588, 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 540, in _build_func_common
+    func = build(s, args, target_host=task.target_host, 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 262, in tvm._ffi._cy3.core.FuncCall
-  File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 251, in tvm._ffi._cy3.core.FuncCall3
+  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):
-  4: TVMFuncCall
+  24: TVMFuncCall
         at ../src/runtime/c_runtime_api.cc:477
-  3: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
-        at ../include/tvm/runtime/packed_func.h:1217
-  2: tvm::runtime::RPCWrappedFunc::operator()(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
-        at ../src/runtime/rpc/rpc_module.cc:129
-  1: tvm::runtime::RPCClientSession::CallFunc(void*, TVMValue const*, int const*, int, std::function&lt;void (tvm::runtime::TVMArgs)&gt; const&amp;)
-        at ../src/runtime/rpc/rpc_endpoint.cc:1012
-  0: tvm::runtime::RPCEndpoint::CallFunc(void*, TVMValue const*, int const*, int, std::function&lt;void (tvm::runtime::TVMArgs)&gt;)
-        at ../src/runtime/rpc/rpc_endpoint.cc:804
-  File &quot;../src/runtime/rpc/rpc_endpoint.cc&quot;, line 804
-TVMError:
----------------------------------------------------------------
-An error occurred during the execution of TVM.
-For more information, please see: https://tvm.apache.org/docs/errors.html
----------------------------------------------------------------
-  Check failed: (code == RPCCode::kReturn) is false: code=kShutdown
-
-During handling of the above exception, another exception occurred:
-
-Traceback (most recent call last):
-  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 702, in run_through_rpc
-    costs = time_f(*args).results
-  File &quot;/usr/lib/python3.7/contextlib.py&quot;, line 130, in __exit__
-    self.gen.throw(type, value, traceback)
-  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 742, in __call__
-    remote.remove(build_result.filename)
-  File &quot;/workspace/python/tvm/rpc/client.py&quot;, line 144, in remove
-    self._remote_funcs[&quot;remove&quot;] = self.get_function(&quot;tvm.rpc.server.remove&quot;)
-  File &quot;/workspace/python/tvm/rpc/client.py&quot;, line 72, in get_function
-    return self._sess.get_function(name)
-  File &quot;/workspace/python/tvm/runtime/module.py&quot;, line 171, in get_function
-    self.handle, c_str(name), ctypes.c_int(query_imports), ctypes.byref(ret_handle)
-  File &quot;/workspace/python/tvm/_ffi/base.py&quot;, line 348, in check_call
-    raise get_last_ffi_error()
-tvm._ffi.base.TVMError: Traceback (most recent call last):
-  52: 0xffffffffffffffff
-  51: _start
-  50: __libc_start_main
-  49: _Py_UnixMain
-  48: 0x0000000000650da0
-  47: 0x0000000000650afa
-  46: _PyFunction_FastCallDict
-  45: _PyEval_EvalCodeWithName
-  44: _PyEval_EvalFrameDefault
-  43: _PyFunction_FastCallKeywords
-  42: _PyEval_EvalCodeWithName
-  41: _PyEval_EvalFrameDefault
-  40: _PyMethodDef_RawFastCallKeywords
-  39: 0x0000000000546369
-  38: _PyEval_EvalCodeWithName
-  37: _PyEval_EvalFrameDefault
-  36: _PyFunction_FastCallKeywords
-  35: _PyEval_EvalCodeWithName
-  34: _PyEval_EvalFrameDefault
-  33: _PyFunction_FastCallDict
-  32: _PyEval_EvalCodeWithName
-  31: _PyEval_EvalFrameDefault
-  30: _PyObject_FastCallDict
-  29: 0x00000000004c06e1
-  28: _PyFunction_FastCallDict
-  27: _PyEval_EvalFrameDefault
-  26: _PyMethodDescr_FastCallKeywords
-  25: 0x00000000005dcb58
-  24: 0x00000000005dc83f
-  23: 0x00000000004ba127
-  22: _PyEval_EvalFrameDefault
-  21: _PyFunction_FastCallKeywords
-  20: _PyEval_EvalFrameDefault
-  19: _PyFunction_FastCallKeywords
-  18: _PyEval_EvalFrameDefault
-  17: _PyFunction_FastCallKeywords
-  16: _PyEval_EvalCodeWithName
-  15: _PyEval_EvalFrameDefault
-  14: 0x0000000000537c30
-  13: _PyObject_FastCallKeywords
-  12: 0x00007f3866235fa2
-  11: _ctypes_callproc
-  10: ffi_call
-  9: ffi_call_unix64
-  8: TVMModGetFunction
-        at ../src/runtime/c_runtime_api.cc:408
-  7: tvm::runtime::ModuleNode::GetFunction(std::__cxx11::basic_string&lt;char, std::char_traits&lt;char&gt;, std::allocator&lt;char&gt; &gt; const&amp;, bool)
-        at ../src/runtime/module.cc:66
-  6: tvm::runtime::RPCModuleNode::GetFunction(std::__cxx11::basic_string&lt;char, std::char_traits&lt;char&gt;, std::allocator&lt;char&gt; &gt; const&amp;, tvm::runtime::ObjectPtr&lt;tvm::runtime::Object&gt; const&amp;)
-        at ../src/runtime/rpc/rpc_module.cc:185
-  5: tvm::runtime::RPCClientSession::GetFunction(std::__cxx11::basic_string&lt;char, std::char_traits&lt;char&gt;, std::allocator&lt;char&gt; &gt; const&amp;)
-        at ../src/runtime/rpc/rpc_endpoint.cc:1007
-  4: tvm::runtime::TVMRetValue tvm::runtime::RPCEndpoint::SysCallRemote&lt;std::__cxx11::basic_string&lt;char, std::char_traits&lt;char&gt;, std::allocator&lt;char&gt; &gt; const&amp;&gt;(tvm::runtime::RPCCode, std::__cxx11::basic_string&lt;char, std::char_traits&lt;char&gt;, std::allocator&lt;char&gt; &gt; const&amp;)
-        at ../src/runtime/rpc/rpc_endpoint.h:223
-  3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()&lt;int, std::__cxx11::basic_string&lt;char, std::char_traits&lt;char&gt;, std::allocator&lt;char&gt; &gt; const&amp;&gt;(int&amp;&amp;, std::__cxx11::basic_string&lt;char, std::char_traits&lt;char&gt;, std::allocator&lt;char&gt; &gt; const&amp;) const
+  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:1731
+  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:1671
+  19: run&lt;&gt;
+        at ../include/tvm/runtime/packed_func.h:1631
+  18: run&lt;tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1631
+  17: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1631
+  16: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1631
+  15: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1631
+  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:1646
+  13: operator()
+        at ../src/driver/driver_api.cc:388
+  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:374
+  11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array&lt;tvm::transform::Pass, void&gt;)
+        at ../src/driver/driver_api.cc:269
+  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:453
+  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:1750
+  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:1694
+  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:1618
   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/rpc/rpc_endpoint.cc:684
-  File &quot;../src/runtime/rpc/rpc_endpoint.cc&quot;, line 684
-TVMError:
----------------------------------------------------------------
-An error occurred during the execution of TVM.
-For more information, please see: https://tvm.apache.org/docs/errors.html
----------------------------------------------------------------
-  Check failed: (code == RPCCode::kReturn) is false: code=1
+        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 871, 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):
-  52: 0xffffffffffffffff
-  51: _start
-  50: __libc_start_main
-  49: _Py_UnixMain
-  48: 0x0000000000650da0
-  47: 0x0000000000650afa
-  46: _PyFunction_FastCallDict
-  45: _PyEval_EvalCodeWithName
-  44: _PyEval_EvalFrameDefault
-  43: _PyFunction_FastCallKeywords
-  42: _PyEval_EvalCodeWithName
-  41: _PyEval_EvalFrameDefault
-  40: _PyMethodDef_RawFastCallKeywords
-  39: 0x0000000000546369
-  38: _PyEval_EvalCodeWithName
-  37: _PyEval_EvalFrameDefault
-  36: _PyFunction_FastCallKeywords
-  35: _PyEval_EvalCodeWithName
-  34: _PyEval_EvalFrameDefault
-  33: _PyFunction_FastCallDict
-  32: _PyEval_EvalCodeWithName
-  31: _PyEval_EvalFrameDefault
-  30: _PyObject_FastCallDict
-  29: 0x00000000004c06e1
-  28: _PyFunction_FastCallDict
-  27: _PyEval_EvalFrameDefault
-  26: _PyMethodDescr_FastCallKeywords
-  25: 0x00000000005dcb58
-  24: 0x00000000005dc83f
-  23: 0x00000000004ba127
-  22: _PyEval_EvalFrameDefault
-  21: _PyFunction_FastCallKeywords
-  20: _PyEval_EvalFrameDefault
-  19: _PyFunction_FastCall      [(&#39;tile_f&#39;, [-1, 32, 4, 4]), (&#39;tile_y&#39;, [-1, 7, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 16, 1]), (&#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,1758580
-No: 14  GFLOPS: 0.00/24.56      result: Traceback (most recent call last):
-  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 738, in __call__
-    yield remote, remote.load_module(os.path.split(build_result.filename)[1])
-  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 702, in run_through_rpc
-    costs = time_f(*args).results
-  File &quot;/workspace/python/tvm/runtime/module.py&quot;, line 357, in evaluator
-    blob = feval(*args)
+  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:1731
+  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:1671
+  19: run&lt;&gt;
+        at ../include/tvm/runtime/packed_func.h:1631
+  18: run&lt;tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1631
+  17: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1631
+  16: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1631
+  15: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1631
+  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:1646
+  13: operator()
+        at ../src/driver/driver_api.cc:388
+  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:374
+  11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array&lt;tvm::transform::Pass, void&gt;)
+        at ../src/driver/driver_api.cc:269
+  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:453
+  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:1750
+  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:1694
+  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:1618
+  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 871, 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, 7, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 4, 32]), (&#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,925278
+No: 13  GFLOPS: 0.00/11.64      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, 32, 2, 2]), (&#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, 1]), (&#39;tile_rx&#39;, [-1, 1, 1]), (&#39;auto_unroll_max_step&#39;, 512), (&#39;unroll_explicit&#39;, 0)],None,1884369
+No: 14  GFLOPS: 0.00/11.64      result: Traceback (most recent call last):
+  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 588, 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 540, in _build_func_common
+    func = build(s, args, target_host=task.target_host, 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 262, in tvm._ffi._cy3.core.FuncCall
-  File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 251, in tvm._ffi._cy3.core.FuncCall3
+  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):
-  4: TVMFuncCall
+  24: TVMFuncCall
         at ../src/runtime/c_runtime_api.cc:477
-  3: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
-        at ../include/tvm/runtime/packed_func.h:1217
-  2: tvm::runtime::RPCWrappedFunc::operator()(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
-        at ../src/runtime/rpc/rpc_module.cc:129
-  1: tvm::runtime::RPCClientSession::CallFunc(void*, TVMValue const*, int const*, int, std::function&lt;void (tvm::runtime::TVMArgs)&gt; const&amp;)
-        at ../src/runtime/rpc/rpc_endpoint.cc:1012
-  0: tvm::runtime::RPCEndpoint::CallFunc(void*, TVMValue const*, int const*, int, std::function&lt;void (tvm::runtime::TVMArgs)&gt;)
-        at ../src/runtime/rpc/rpc_endpoint.cc:804
-  File &quot;../src/runtime/rpc/rpc_endpoint.cc&quot;, line 804
-TVMError:
----------------------------------------------------------------
-An error occurred during the execution of TVM.
-For more information, please see: https://tvm.apache.org/docs/errors.html
----------------------------------------------------------------
-  Check failed: (code == RPCCode::kReturn) is false: code=kShutdown
-
-During handling of the above exception, another exception occurred:
+  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:1731
+  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:1671
+  19: run&lt;&gt;
+        at ../include/tvm/runtime/packed_func.h:1631
+  18: run&lt;tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1631
+  17: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1631
+  16: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1631
+  15: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1631
+  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:1646
+  13: operator()
+        at ../src/driver/driver_api.cc:388
+  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:374
+  11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array&lt;tvm::transform::Pass, void&gt;)
+        at ../src/driver/driver_api.cc:269
+  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:453
+  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:1750
+  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:1694
+  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:1618
+  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 871, 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):
-  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 702, in run_through_rpc
-    costs = time_f(*args).results
-  File &quot;/usr/lib/python3.7/contextlib.py&quot;, line 130, in __exit__
-    self.gen.throw(type, value, traceback)
-  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 742, in __call__
-    remote.remove(build_result.filename)
-  File &quot;/workspace/python/tvm/rpc/client.py&quot;, line 144, in remove
-    self._remote_funcs[&quot;remove&quot;] = self.get_function(&quot;tvm.rpc.server.remove&quot;)
-  File &quot;/workspace/python/tvm/rpc/client.py&quot;, line 72, in get_function
-    return self._sess.get_function(name)
-  File &quot;/workspace/python/tvm/runtime/module.py&quot;, line 171, in get_function
-    self.handle, c_str(name), ctypes.c_int(query_imports), ctypes.byref(ret_handle)
-  File &quot;/workspace/python/tvm/_ffi/base.py&quot;, line 348, in check_call
-    raise get_last_ffi_error()
+  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:1731
+  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:1671
+  19: run&lt;&gt;
+        at ../include/tvm/runtime/packed_func.h:1631
+  18: run&lt;tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1631
+  17: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1631
+  16: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1631
+  15: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1631
+  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:1646
+  13: operator()
+        at ../src/driver/driver_api.cc:388
+  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:374
+  11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array&lt;tvm::transform::Pass, void&gt;)
+        at ../src/driver/driver_api.cc:269
+  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:453
+  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:1750
+  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:1694
+  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:1618
+  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 871, 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, 16, 1]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 128, 1]), (&#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;, 1)],None,9900475
+No: 15  GFLOPS: 0.00/11.64      result: Traceback (most recent call last):
+  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 588, 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 540, in _build_func_common
+    func = build(s, args, target_host=task.target_host, 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):
-  52: 0xffffffffffffffff
-  51: _start
-  50: __libc_start_main
-  49: _Py_UnixMain
-  48: 0x0000000000650da0
-  47: 0x0000000000650afa
-  46: _PyFunction_FastCallDict
-  45: _PyEval_EvalCodeWithName
-  44: _PyEval_EvalFrameDefault
-  43: _PyFunction_FastCallKeywords
-  42: _PyEval_EvalCodeWithName
-  41: _PyEval_EvalFrameDefault
-  40: _PyMethodDef_RawFastCallKeywords
-  39: 0x0000000000546369
-  38: _PyEval_EvalCodeWithName
-  37: _PyEval_EvalFrameDefault
-  36: _PyFunction_FastCallKeywords
-  35: _PyEval_EvalCodeWithName
-  34: _PyEval_EvalFrameDefault
-  33: _PyFunction_FastCallDict
-  32: _PyEval_EvalCodeWithName
-  31: _PyEval_EvalFrameDefault
-  30: _PyObject_FastCallDict
-  29: 0x00000000004c06e1
-  28: _PyFunction_FastCallDict
-  27: _PyEval_EvalFrameDefault
-  26: _PyMethodDescr_FastCallKeywords
-  25: 0x00000000005dcb58
-  24: 0x00000000005dc83f
-  23: 0x00000000004ba127
-  22: _PyEval_EvalFrameDefault
-  21: _PyFunction_FastCallKeywords
-  20: _PyEval_EvalFrameDefault
-  19: _PyFunction_FastCallKeywords
-  18: _PyEval_EvalFrameDefault
-  17: _PyFunction_FastCallKeywords
-  16: _PyEval_EvalCodeWithName
-  15: _PyEval_EvalFrameDefault
-  14: 0x0000000000537c30
-  13: _PyObject_FastCallKeywords
-  12: 0x00007fcfb46e6fa2
-  11: _ctypes_callproc
-  10: ffi_call
-  9: ffi_call_unix64
-  8: TVMModGetFunction
-        at ../src/runtime/c_runtime_api.cc:408
-  7: tvm::runtime::ModuleNode::GetFunction(std::__cxx11::basic_string&lt;char, std::char_traits&lt;char&gt;, std::allocator&lt;char&gt; &gt; const&amp;, bool)
-        at ../src/runtime/module.cc:66
-  6: tvm::runtime::RPCModuleNode::GetFunction(std::__cxx11::basic_string&lt;char, std::char_traits&lt;char&gt;, std::allocator&lt;char&gt; &gt; const&amp;, tvm::runtime::ObjectPtr&lt;tvm::runtime::Object&gt; const&amp;)
-        at ../src/runtime/rpc/rpc_module.cc:185
-  5: tvm::runtime::RPCClientSession::GetFunction(std::__cxx11::basic_string&lt;char, std::char_traits&lt;char&gt;, std::allocator&lt;char&gt; &gt; const&amp;)
-        at ../src/runtime/rpc/rpc_endpoint.cc:1007
-  4: tvm::runtime::TVMRetValue tvm::runtime::RPCEndpoint::SysCallRemote&lt;std::__cxx11::basic_string&lt;char, std::char_traits&lt;char&gt;, std::allocator&lt;char&gt; &gt; const&amp;&gt;(tvm::runtime::RPCCode, std::__cxx11::basic_string&lt;char, std::char_traits&lt;char&gt;, std::allocator&lt;char&gt; &gt; const&amp;)
-        at ../src/runtime/rpc/rpc_endpoint.h:223
-  3: tvm::runtime::TVMRetValue tvm::runtime::PackedFunc::operator()&lt;int, std::__cxx11::basic_string&lt;char, std::char_traits&lt;char&gt;, std::allocator&lt;char&gt; &gt; const&amp;&gt;(int&amp;&amp;, std::__cxx11::basic_string&lt;char, std::char_traits&lt;char&gt;, std::allocator&lt;char&gt; &gt; const&amp;) const
+  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:1731
+  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:1671
+  19: run&lt;&gt;
+        at ../include/tvm/runtime/packed_func.h:1631
+  18: run&lt;tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1631
+  17: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1631
+  16: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1631
+  15: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1631
+  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:1646
+  13: operator()
+        at ../src/driver/driver_api.cc:388
+  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:374
+  11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array&lt;tvm::transform::Pass, void&gt;)
+        at ../src/driver/driver_api.cc:269
+  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:453
+  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:1750
+  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:1694
+  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:1618
   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/rpc/rpc_endpoint.cc:684
-  File &quot;../src/runtime/rpc/rpc_endpoint.cc&quot;, line 684
-TVMError:
----------------------------------------------------------------
-An error occurred during the execution of TVM.
-For more information, please see: https://tvm.apache.org/docs/errors.html
----------------------------------------------------------------
-  Check failed: (code == RPCCode::kReturn) is false: code=1
+        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 871, 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):
-  52: 0xffffffffffffffff
-  51: _start
-  50: __libc_start_main
-  49: _Py_UnixMain
-  48: 0x0000000000650da0
-  47: 0x0000000000650afa
-  46: _PyFunction_FastCallDict
-  45: _PyEval_EvalCodeWithName
-  44: _PyEval_EvalFrameDefault
-  43: _PyFunction_FastCallKeywords
-  42: _PyEval_EvalCodeWithName
-  41: _PyEval_EvalFrameDefault
-  40: _PyMethodDef_RawFastCallKeywords
-  39: 0x0000000000546369
-  38: _PyEval_EvalCodeWithName
-  37: _PyEval_EvalFrameDefault
-  36: _PyFunction_FastCallKeywords
-  35: _PyEval_EvalCodeWithName
-  34: _PyEval_EvalFrameDefault
-  33: _PyFunction_FastCallDict
-  32: _PyEval_EvalCodeWithName
-  31: _PyEval_EvalFrameDefault
-  30: _PyObject_FastCallDict
-  29: 0x00000000004c06e1
-  28: _PyFunction_FastCallDict
-  27: _PyEval_EvalFrameDefault
-  26: _PyMethodDescr_FastCallKeywords
-  25: 0x00000000005dcb58
-  24: 0x00000000005dc83f
-  23: 0x00000000004ba127
-  22: _PyEval_EvalFrameDefault
-  21: _PyFunction_FastCallKeywords
-  20: _PyEval_EvalFrameDefault
-  19: _PyFunction_FastCall      [(&#39;tile_f&#39;, [-1, 64, 1, 1]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 4, 2]), (&#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;, 1)],None,5659946
-No: 15  GFLOPS: 6.77/24.56      result: MeasureResult(costs=(0.034179073250000004,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.2322049140930176, timestamp=1668475021.8653607)       [(&#39;tile_f&#39;, [-1, 2, 1, 8]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 32, 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,1566537
-No: 16  GFLOPS: 0.00/24.56      result: 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:1731
+  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:1671
+  19: run&lt;&gt;
+        at ../include/tvm/runtime/packed_func.h:1631
+  18: run&lt;tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1631
+  17: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1631
+  16: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1631
+  15: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
+        at ../include/tvm/runtime/packed_func.h:1631
+  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:1646
+  13: operator()
+        at ../src/driver/driver_api.cc:388
+  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:374
+  11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array&lt;tvm::transform::Pass, void&gt;)
+        at ../src/driver/driver_api.cc:269
+  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:453
+  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:1750
+  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:1694
+  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:1618
+  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 871, 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, 2, 1]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 1, 128]), (&#39;tile_ry&#39;, [-1, 1, 3]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 512), (&#39;unroll_explicit&#39;, 0)],None,3465890
+No: 16  GFLOPS: 0.00/11.64      result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 588, 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 540, in _build_func_common
@@ -2228,8 +2307,10 @@ 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 871, 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, 1, 64]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 16, 4]), (&#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,1049823
-No: 17  GFLOPS: 0.00/24.56      result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 1, 32, 16]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 64, 8]), (&#39;tile_ry&#39;, [-1, 3, 1]), (&#39;tile_rx&#39;, [-1, 3, 1]), (&#39;auto_unroll_max_step&#39;, 512), (&#39;unroll_explicit&#39;, 0)],None,2636444
+No: 17  GFLOPS: 232.51/232.51   result: MeasureResult(costs=(0.0009956469262295082,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.6534688472747803, timestamp=1668481246.2235832)      [(&#39;tile_f&#39;, [-1, 2, 32, 2]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 1, 8]), (&#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;, 1)],None,7065171
+No: 18  GFLOPS: 189.37/232.51   result: MeasureResult(costs=(0.0012224727938931297,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.5868277549743652, timestamp=1668481247.1970925)      [(&#39;tile_f&#39;, [-1, 8, 8, 1]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 7, 1, 1]), (&#39;tile_rc&#39;, [-1, 2, 8]), (&#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,680710
+No: 19  GFLOPS: 0.00/232.51     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 588, 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 540, in _build_func_common
@@ -2351,8 +2432,8 @@ 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 871, 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, 2, 2]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 7, 1, 1]), (&#39;tile_rc&#39;, [-1, 1, 512]), (&#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,9483830
-No: 18  GFLOPS: 0.00/24.56      result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 2, 64, 4]), (&#39;tile_y&#39;, [-1, 7, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 4, 2]), (&#39;tile_ry&#39;, [-1, 1, 3]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 512), (&#39;unroll_explicit&#39;, 1)],None,8562754
+No: 20  GFLOPS: 0.00/232.51     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 588, 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 540, in _build_func_common
@@ -2474,9 +2555,7 @@ 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 871, 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, 1]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 7, 1, 1]), (&#39;tile_rc&#39;, [-1, 16, 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,9375323
-No: 19  GFLOPS: 290.01/290.01   result: MeasureResult(costs=(0.0007982536534653466,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.7852540016174316, timestamp=1668475023.8795164)      [(&#39;tile_f&#39;, [-1, 1, 8, 1]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 7, 1, 1]), (&#39;tile_rc&#39;, [-1, 32, 1]), (&#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,1761567
-No: 20  GFLOPS: 38.19/290.01    result: MeasureResult(costs=(0.006061600176470588,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.1658580303192139, timestamp=1668475024.541543)        [(&#39;tile_f&#39;, [-1, 1, 4, 32]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 1, 16]), (&#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;, 0)],None,315674
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 2, 8, 32]), (&#39;tile_y&#39;, [-1, 7, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 16, 2]), (&#39;tile_ry&#39;, [-1, 1, 3]), (&#39;tile_rx&#39;, [-1, 1, 1]), (&#39;auto_unroll_max_step&#39;, 512), (&#39;unroll_explicit&#39;, 0)],None,2181938
 </pre></div>
 </div>
 <p>Finally we can inspect the best config from log file, check correctness,
@@ -2515,9 +2594,9 @@ 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, 1, 8, 1]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 7, 1, 1]), (&#39;tile_rc&#39;, [-1, 32, 1]), (&#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,1761567
+[(&#39;tile_f&#39;, [-1, 2, 32, 2]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 1, 8]), (&#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;, 1)],None,7065171
 Finish loading 20 records
-Time cost of this operator: 0.001224
+Time cost of this operator: 0.001205
 </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 1cc8692d1b..7d8dbc665d 100644
--- a/docs/how_to/work_with_microtvm/micro_autotune.html
+++ b/docs/how_to/work_with_microtvm/micro_autotune.html
@@ -596,10 +596,10 @@ 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  312.0     98.705   (1, 2, 10, 10, 3)  2       1        [312.0]
-tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       3.094     0.979    (1, 6, 10, 10)     1       1        [3.094]
-tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         1.001     0.317    (1, 1, 10, 10, 3)  1       1        [1.001]
-Total_time                                    -                                             316.095   -        -                  -       -        -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc  tvmgen_default_fused_nn_contrib_conv2d_NCHWc  310.6     98.728   (1, 2, 10, 10, 3)  2       1        [310.6]
+tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       3.021     0.96     (1, 6, 10, 10)     1       1        [3.021]
+tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.979     0.311    (1, 1, 10, 10, 3)  1       1        [0.979]
+Total_time                                    -                                             314.6     -        -                  -       -        -
 </pre></div>
 </div>
 </div>
@@ -650,10 +650,10 @@ 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  100.2     97.322   (1, 6, 10, 10, 1)  2       1        [100.2]
-tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       1.792     1.74     (1, 6, 10, 10)     1       1        [1.792]
-tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.966     0.938    (1, 1, 10, 10, 3)  1       1        [0.966]
-Total_time                                    -                                             102.957   -        -                  -       -        -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc  tvmgen_default_fused_nn_contrib_conv2d_NCHWc  137.1     98.095   (1, 6, 10, 10, 1)  2       1        [137.1]
+tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       1.818     1.301    (1, 6, 10, 10)     1       1        [1.818]
+tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.844     0.604    (1, 3, 10, 10, 1)  1       1        [0.844]
+Total_time                                    -                                             139.762   -        -                  -       -        -
 </pre></div>
 </div>
 <div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-work-with-microtvm-micro-autotune-py">
diff --git a/docs/how_to/work_with_microtvm/micro_pytorch.html b/docs/how_to/work_with_microtvm/micro_pytorch.html
index 2271db3a1d..90e683a2d6 100644
--- a/docs/how_to/work_with_microtvm/micro_pytorch.html
+++ b/docs/how_to/work_with_microtvm/micro_pytorch.html
@@ -440,10 +440,8 @@ 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]
-  2%|1         | 64.0k/3.42M [00:00&lt;00:06, 537kB/s]
-  9%|8         | 304k/3.42M [00:00&lt;00:02, 1.41MB/s]
- 37%|###7      | 1.27M/3.42M [00:00&lt;00:00, 4.58MB/s]
-100%|##########| 3.42M/3.42M [00:00&lt;00:00, 8.25MB/s]
+ 37%|###7      | 1.28M/3.42M [00:00&lt;00:00, 13.2MB/s]
+100%|##########| 3.42M/3.42M [00:00&lt;00:00, 19.5MB/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.
@@ -567,7 +565,7 @@ 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  4.460 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  4.661 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 49aa8e8f87..1ba9454275 100644
--- a/docs/how_to/work_with_microtvm/micro_train.html
+++ b/docs/how_to/work_with_microtvm/micro_train.html
@@ -530,7 +530,7 @@ 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/tmpkpo3z5js/images/random&#39;
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>&#39;/tmp/tmp9x_81ijo/images/random&#39;
 </pre></div>
 </div>
 </div>
@@ -590,8 +590,8 @@ 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], [1.0, 0.0], [0.0, 1.0], [1.0, 0.0], [1.0, 0.0], [1.0, 0.0], [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/tmpkpo3z5js/images/target contains 8144 images
-/tmp/tmpkpo3z5js/images/random contains 5000 images
+<img src="../../_images/sphx_glr_micro_train_001.png" srcset="../../_images/sphx_glr_micro_train_001.png" alt="[0.0, 1.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [0.0, 1.0]" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/tmp/tmp9x_81ijo/images/target contains 8144 images
+/tmp/tmp9x_81ijo/images/random contains 5000 images
 </pre></div>
 </div>
 </div>
@@ -703,13 +703,13 @@ 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 - 47s - loss: 0.2287 - accuracy: 0.9221 - val_loss: 0.1266 - val_accuracy: 0.9502 - 47s/epoch - 145ms/step
+328/328 - 47s - loss: 0.2268 - accuracy: 0.9215 - val_loss: 0.1660 - val_accuracy: 0.9452 - 47s/epoch - 144ms/step
 Epoch 2/3
-328/328 - 44s - loss: 0.1137 - accuracy: 0.9588 - val_loss: 0.0845 - val_accuracy: 0.9732 - 44s/epoch - 134ms/step
+328/328 - 43s - loss: 0.1011 - accuracy: 0.9637 - val_loss: 0.1315 - val_accuracy: 0.9524 - 43s/epoch - 133ms/step
 Epoch 3/3
-328/328 - 44s - loss: 0.0703 - accuracy: 0.9748 - val_loss: 0.0871 - val_accuracy: 0.9698 - 44s/epoch - 133ms/step
+328/328 - 43s - loss: 0.0697 - accuracy: 0.9728 - val_loss: 0.1259 - val_accuracy: 0.9573 - 43s/epoch - 132ms/step
 
-&lt;keras.callbacks.History object at 0x7f6f10c5db90&gt;
+&lt;keras.callbacks.History object at 0x7f8789ef85d0&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> ( 4 minutes  35.858 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 4 minutes  21.227 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 d4a616b853..32697f304c 100644
--- a/docs/how_to/work_with_microtvm/sg_execution_times.html
+++ b/docs/how_to/work_with_microtvm/sg_execution_times.html
@@ -340,7 +340,7 @@
             
   <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>06:43.515</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
+<p><strong>06:29.210</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 83%" />
@@ -349,27 +349,27 @@
 </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">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>04:35.858</p></td>
+<td><p>04:21.227</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><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">microTVM PyTorch Tutorial</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_pytorch.py</span></code>)</p></td>
-<td><p>01:04.460</p></td>
+<td><p>01:04.661</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><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">Autotuning with microTVM</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_autotune.py</span></code>)</p></td>
-<td><p>00:50.784</p></td>
+<td><p>00:51.140</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">microTVM Host-Driven AoT</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_aot.py</span></code>)</p></td>
-<td><p>00:08.577</p></td>
+<td><p>00:08.327</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><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">microTVM with TFLite Models</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_tflite.py</span></code>)</p></td>
-<td><p>00:03.835</p></td>
+<td><p>00:03.853</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="micro_reference_vm.html#sphx-glr-how-to-work-with-microtvm-micro-reference-vm-py"><span class="std std-ref">microTVM Reference Virtual Machines</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_reference_vm.py</span></code>)</p></td>
-<td><p>00:00.001</p></td>
+<td><p>00:00.002</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">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 448b9c5b09..04175ccbe8 100644
--- a/docs/how_to/work_with_relay/sg_execution_times.html
+++ b/docs/how_to/work_with_relay/sg_execution_times.html
@@ -340,7 +340,7 @@
             
   <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>00:45.224</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
+<p><strong>00:44.332</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 84%" />
@@ -349,19 +349,19 @@
 </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>00:32.617</p></td>
+<td><p>00:32.582</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>00:10.366</p></td>
+<td><p>00:10.216</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>00:02.234</p></td>
+<td><p>00:01.528</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>
-<td><p>00:00.007</p></td>
+<td><p>00:00.006</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 </tbody>
diff --git a/docs/how_to/work_with_schedules/intrin_math.html b/docs/how_to/work_with_schedules/intrin_math.html
index 7a66dddf5e..ff7a426c1e 100644
--- a/docs/how_to/work_with_schedules/intrin_math.html
+++ b/docs/how_to/work_with_schedules/intrin_math.html
@@ -535,7 +535,7 @@ 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 0x7f6f0c303050&gt;
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>&lt;function my_cuda_math_rule at 0x7f8785916a70&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 59158094d1..ab33ed0f65 100644
--- a/docs/how_to/work_with_schedules/sg_execution_times.html
+++ b/docs/how_to/work_with_schedules/sg_execution_times.html
@@ -340,7 +340,7 @@
             
   <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>00:06.555</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
+<p><strong>00:06.518</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 83%" />
@@ -349,35 +349,35 @@
 </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>00:04.116</p></td>
+<td><p>00:04.113</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>00:01.082</p></td>
+<td><p>00:01.053</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>00:00.579</p></td>
+<td><p>00:00.574</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>00:00.562</p></td>
+<td><p>00:00.561</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>00:00.116</p></td>
+<td><p>00:00.118</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><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>00:00.049</p></td>
+<td><p>00:00.050</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><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>00:00.030</p></td>
+<td><p>00:00.029</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>00:00.019</p></td>
+<td><p>00:00.020</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 </tbody>
diff --git a/docs/how_to/work_with_schedules/tensorize.html b/docs/how_to/work_with_schedules/tensorize.html
index fcb01fb5b8..2dc54f2a33 100644
--- a/docs/how_to/work_with_schedules/tensorize.html
+++ b/docs/how_to/work_with_schedules/tensorize.html
@@ -590,7 +590,7 @@ The importing needs to happen before the tensorized GEMV being executed.</p>
              C: Buffer(C_2: Pointer(float32), float32, [524288], [])}
   buffer_map = {A_1: A, B_1: B, C_1: C}
   preflattened_buffer_map = {A_1: A_3: Buffer(A_2, float32, [1024, 64], []), B_1: B_3: Buffer(B_2, float32, [512, 64], []), C_1: C_3: Buffer(C_2, float32, [1024, 512], [])} {
-  attr [IterVar(i: int32, (nullptr), &quot;DataPar&quot;, &quot;&quot;)] &quot;pragma_import_llvm&quot; = &quot;; ModuleID = &#39;/tmp/tmpy_fnrd3r/input0.cc&#39;\nsource_filename = \&quot;/tmp/tmpy_fnrd3r/input0.cc\&quot;\ntarget datalayout = \&quot;e-m:e-i64:64-f80:128-n8:16:32:64-S128\&quot;\ntarget triple = \&quot;x86_64-pc-linux-gnu\&quot;\n\n; Function Attrs: noinline nounwind optnone uwtable\ndefine dso_local i32 @gemv_update(float*, float*, float*, i32, i32, i32) #0 {\n  %7 = allo [...]
+  attr [IterVar(i: int32, (nullptr), &quot;DataPar&quot;, &quot;&quot;)] &quot;pragma_import_llvm&quot; = &quot;; ModuleID = &#39;/tmp/tmp7g8rvtkr/input0.cc&#39;\nsource_filename = \&quot;/tmp/tmp7g8rvtkr/input0.cc\&quot;\ntarget datalayout = \&quot;e-m:e-i64:64-f80:128-n8:16:32:64-S128\&quot;\ntarget triple = \&quot;x86_64-pc-linux-gnu\&quot;\n\n; Function Attrs: noinline nounwind optnone uwtable\ndefine dso_local i32 @gemv_update(float*, float*, float*, i32, i32, i32) #0 {\n  %7 = allo [...]
   for (i, 0, 1024) {
     for (j.outer: int32, 0, 32) {
       @tir.call_extern(&quot;gemv_update&quot;, @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), C_2, ((i*512) + (j.outer*16)), 16, 2, dtype=handle), @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), A_2, (i*64), 64, 1, dtype=handle), @tir.tvm_access_ptr(@tir.type_annotation(, dtype=float32), B_2, (j.outer*1024), 1024, 1, dtype=handle), 16, 64, 64, dtype=int32)
diff --git a/docs/reference/api/python/auto_scheduler.html b/docs/reference/api/python/auto_scheduler.html
index 5cabb478e2..41ed09bd68 100644
--- a/docs/reference/api/python/auto_scheduler.html
+++ b/docs/reference/api/python/auto_scheduler.html
@@ -1615,7 +1615,7 @@ 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>
@@ -1899,7 +1899,7 @@ 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/typedoc/classes/bytestreamreader.html b/docs/reference/api/typedoc/classes/bytestreamreader.html
index 80dc032a70..4a1f59a99d 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/b6fae9b35/web/src/rpc_server.ts#L43">rpc_server.ts:43</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/rpc_server.ts#L43">rpc_server.ts:43</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/b6fae9b35/web/src/rpc_server.ts#L43">rpc_server.ts:43</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/rpc_server.ts#L43">rpc_server.ts:43</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/b6fae9b35/web/src/rpc_server.ts#L42">rpc_server.ts:42</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/rpc_server.ts#L42">rpc_server.ts:42</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/b6fae9b35/web/src/rpc_server.ts#L63">rpc_server.ts:63</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/rpc_server.ts#L63">rpc_server.ts:63</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/b6fae9b35/web/src/rpc_server.ts#L49">rpc_server.ts:49</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/rpc_server.ts#L49">rpc_server.ts:49</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/b6fae9b35/web/src/rpc_server.ts#L57">rpc_server.ts:57</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/rpc_server.ts#L57">rpc_server.ts:57</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 2302242091..438beb82d7 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/b6fae9b35/web/src/memory.ts#L223">memory.ts:223</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L208">memory.ts:208</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L312">memory.ts:312</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L284">memory.ts:284</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L388">memory.ts:388</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L376">memory.ts:376</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L267">memory.ts:267</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L243">memory.ts:243</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L321">memory.ts:321</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L252">memory.ts:252</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L359">memory.ts:359</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L342">memory.ts:342</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L350">memory.ts:350</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L326">memory.ts:326</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L363">memory.ts:363</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L346">memory.ts:346</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L334">memory.ts:334</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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 88d975e8b3..48e3fb2f64 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/b6fae9b35/web/src/runtime.ts#L262">runtime.ts:262</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L262">runtime.ts:262</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/b6fae9b35/web/src/runtime.ts#L260">runtime.ts:260</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L260">runtime.ts:260</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/b6fae9b35/web/src/runtime.ts#L258">runtime.ts:258</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L258">runtime.ts:258</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/b6fae9b35/web/src/runtime.ts#L262">runtime.ts:262</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L262">runtime.ts:262</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/b6fae9b35/web/src/runtime.ts#L279">runtime.ts:279</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L279">runtime.ts:279</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/b6fae9b35/web/src/runtime.ts#L270">runtime.ts:270</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L270">runtime.ts:270</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 d91f1104e9..65467035a0 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/b6fae9b35/web/src/runtime.ts#L202">runtime.ts:202</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L202">runtime.ts:202</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/b6fae9b35/web/src/runtime.ts#L200">runtime.ts:200</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L200">runtime.ts:200</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/b6fae9b35/web/src/runtime.ts#L198">runtime.ts:198</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L198">runtime.ts:198</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/b6fae9b35/web/src/runtime.ts#L223">runtime.ts:223</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L223">runtime.ts:223</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/b6fae9b35/web/src/runtime.ts#L230">runtime.ts:230</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L230">runtime.ts:230</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 981181139c..17f97f776e 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/b6fae9b35/web/src/environment.ts#L86">environment.ts:86</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/environment.ts#L70">environment.ts:70</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/environment.ts#L69">environment.ts:69</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/environment.ts#L78">environment.ts:78</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/environment.ts#L84">environment.ts:84</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/environment.ts#L105">environment.ts:105</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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 90dbc22761..e95e189c26 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/b6fae9b35/web/src/runtime.ts#L49">runtime.ts:49</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L49">runtime.ts:49</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/b6fae9b35/web/src/runtime.ts#L46">runtime.ts:46</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L46">runtime.ts:46</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/b6fae9b35/web/src/runtime.ts#L45">runtime.ts:45</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L45">runtime.ts:45</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/b6fae9b35/web/src/runtime.ts#L44">runtime.ts:44</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L44">runtime.ts:44</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/b6fae9b35/web/src/runtime.ts#L47">runtime.ts:47</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L47">runtime.ts:47</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/b6fae9b35/web/src/runtime.ts#L76">runtime.ts:76</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L76">runtime.ts:76</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/b6fae9b35/web/src/runtime.ts#L66">runtime.ts:66</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L66">runtime.ts:66</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/b6fae9b35/web/src/runtime.ts#L84">runtime.ts:84</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L84">runtime.ts:84</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/b6fae9b35/web/src/runtime.ts#L95">runtime.ts:95</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L95">runtime.ts:95</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/b6fae9b35/web/src/runtime.ts#L72">runtime.ts:72</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L72">runtime.ts:72</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/graphexecutor.html b/docs/reference/api/typedoc/classes/graphexecutor.html
index e48ba46929..c815262517 100644
--- a/docs/reference/api/typedoc/classes/graphexecutor.html
+++ b/docs/reference/api/typedoc/classes/graphexecutor.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/b6fae9b35/web/src/runtime.ts#L583">runtime.ts:583</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L583">runtime.ts:583</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -162,7 +162,7 @@
 					<div class="tsd-signature tsd-kind-icon">module<span class="tsd-signature-symbol">:</span> <a href="module.html" class="tsd-signature-type">Module</a></div>
 					<aside class="tsd-sources">
 						<ul>
-							<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L579">runtime.ts:579</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L579">runtime.ts:579</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/b6fae9b35/web/src/runtime.ts#L654">runtime.ts:654</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L654">runtime.ts:654</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/b6fae9b35/web/src/runtime.ts#L597">runtime.ts:597</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L597">runtime.ts:597</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -241,7 +241,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L631">runtime.ts:631</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L631">runtime.ts:631</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -279,7 +279,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L644">runtime.ts:644</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L644">runtime.ts:644</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -310,7 +310,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L621">runtime.ts:621</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L621">runtime.ts:621</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -332,7 +332,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L609">runtime.ts:609</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L609">runtime.ts:609</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
diff --git a/docs/reference/api/typedoc/classes/instance.html b/docs/reference/api/typedoc/classes/instance.html
index 9d648b9162..a08948a3df 100644
--- a/docs/reference/api/typedoc/classes/instance.html
+++ b/docs/reference/api/typedoc/classes/instance.html
@@ -139,7 +139,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L692">runtime.ts:692</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L692">runtime.ts:692</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -202,7 +202,7 @@
 					<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/b6fae9b35/web/src/runtime.ts#L684">runtime.ts:684</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L684">runtime.ts:684</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -212,7 +212,7 @@
 					<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/b6fae9b35/web/src/runtime.ts#L683">runtime.ts:683</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L683">runtime.ts:683</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -229,7 +229,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L932">runtime.ts:932</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L932">runtime.ts:932</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -260,7 +260,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L994">runtime.ts:994</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L994">runtime.ts:994</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -303,7 +303,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L924">runtime.ts:924</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L924">runtime.ts:924</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -341,7 +341,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L732">runtime.ts:732</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L732">runtime.ts:732</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -358,7 +358,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L952">runtime.ts:952</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L952">runtime.ts:952</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -402,7 +402,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L816">runtime.ts:816</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L816">runtime.ts:816</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -434,7 +434,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L1033">runtime.ts:1033</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L1033">runtime.ts:1033</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -465,7 +465,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L846">runtime.ts:846</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L846">runtime.ts:846</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -497,7 +497,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L750">runtime.ts:750</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L750">runtime.ts:750</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -520,7 +520,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L1013">runtime.ts:1013</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L1013">runtime.ts:1013</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -568,7 +568,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L789">runtime.ts:789</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L789">runtime.ts:789</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -608,7 +608,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L914">runtime.ts:914</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L914">runtime.ts:914</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -646,7 +646,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L1145">runtime.ts:1145</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L1145">runtime.ts:1145</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -698,7 +698,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L740">runtime.ts:740</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L740">runtime.ts:740</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -722,7 +722,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L868">runtime.ts:868</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L868">runtime.ts:868</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -754,7 +754,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L857">runtime.ts:857</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L857">runtime.ts:857</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -786,7 +786,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L940">runtime.ts:940</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L940">runtime.ts:940</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 201ddef6e8..05bc73fa77 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/b6fae9b35/web/src/memory.ts#L40">memory.ts:40</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L32">memory.ts:32</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L33">memory.ts:33</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L154">memory.ts:154</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L90">memory.ts:90</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L97">memory.ts:97</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L74">memory.ts:74</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L81">memory.ts:81</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L104">memory.ts:104</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L132">memory.ts:132</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L145">memory.ts:145</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L60">memory.ts:60</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L67">memory.ts:67</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L53">memory.ts:53</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L114">memory.ts:114</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L124">memory.ts:124</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/memory.ts#L175">memory.ts:175</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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 0f812e7abe..98717eb9eb 100644
--- a/docs/reference/api/typedoc/classes/module.html
+++ b/docs/reference/api/typedoc/classes/module.html
@@ -124,7 +124,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L504">runtime.ts:504</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L504">runtime.ts:504</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -170,7 +170,7 @@
 					<div class="tsd-signature tsd-kind-icon">handle<span class="tsd-signature-symbol">:</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/b6fae9b35/web/src/runtime.ts#L502">runtime.ts:502</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L502">runtime.ts:502</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -187,7 +187,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L516">runtime.ts:516</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L516">runtime.ts:516</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -204,7 +204,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L530">runtime.ts:530</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L530">runtime.ts:530</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -236,7 +236,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L561">runtime.ts:561</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L561">runtime.ts:561</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 9c85284d88..2e85708f97 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/b6fae9b35/web/src/runtime.ts#L304">runtime.ts:304</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L304">runtime.ts:304</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/b6fae9b35/web/src/runtime.ts#L297">runtime.ts:297</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L297">runtime.ts:297</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/b6fae9b35/web/src/runtime.ts#L293">runtime.ts:293</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L293">runtime.ts:293</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -188,7 +188,7 @@
 					<div class="tsd-signature tsd-kind-icon">handle<span class="tsd-signature-symbol">:</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/b6fae9b35/web/src/runtime.ts#L289">runtime.ts:289</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L289">runtime.ts:289</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -203,7 +203,7 @@
 					<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/b6fae9b35/web/src/runtime.ts#L291">runtime.ts:291</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L291">runtime.ts:291</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -218,7 +218,7 @@
 					<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/b6fae9b35/web/src/runtime.ts#L295">runtime.ts:295</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L295">runtime.ts:295</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -240,7 +240,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L370">runtime.ts:370</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L370">runtime.ts:370</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -273,7 +273,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L414">runtime.ts:414</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L414">runtime.ts:414</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -305,7 +305,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L355">runtime.ts:355</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L355">runtime.ts:355</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
@@ -322,7 +322,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L474">runtime.ts:474</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L474">runtime.ts:474</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -346,7 +346,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L443">runtime.ts:443</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L443">runtime.ts:443</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 8e3143860a..e4a060d25c 100644
--- a/docs/reference/api/typedoc/classes/packedfunccell.html
+++ b/docs/reference/api/typedoc/classes/packedfunccell.html
@@ -122,7 +122,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L158">runtime.ts:158</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L158">runtime.ts:158</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-parameters-title">Parameters</h4>
@@ -147,7 +147,7 @@
 					<div class="tsd-signature tsd-kind-icon">handle<span class="tsd-signature-symbol">:</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/b6fae9b35/web/src/runtime.ts#L157">runtime.ts:157</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L157">runtime.ts:157</a></li>
 						</ul>
 					</aside>
 				</section>
@@ -164,7 +164,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L165">runtime.ts:165</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L165">runtime.ts:165</a></li>
 								</ul>
 							</aside>
 							<h4 class="tsd-returns-title">Returns <span class="tsd-signature-type">void</span></h4>
diff --git a/docs/reference/api/typedoc/classes/rpcserver.html b/docs/reference/api/typedoc/classes/rpcserver.html
index fa150f296b..be2d54ee8e 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/b6fae9b35/web/src/rpc_server.ts#L92">rpc_server.ts:92</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/rpc_server.ts#L92">rpc_server.ts:92</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/b6fae9b35/web/src/rpc_server.ts#L82">rpc_server.ts:82</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/rpc_server.ts#L82">rpc_server.ts:82</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/b6fae9b35/web/src/rpc_server.ts#L78">rpc_server.ts:78</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/rpc_server.ts#L78">rpc_server.ts:78</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/b6fae9b35/web/src/rpc_server.ts#L81">rpc_server.ts:81</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/rpc_server.ts#L81">rpc_server.ts:81</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/b6fae9b35/web/src/rpc_server.ts#L79">rpc_server.ts:79</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/rpc_server.ts#L79">rpc_server.ts:79</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/b6fae9b35/web/src/rpc_server.ts#L80">rpc_server.ts:80</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/rpc_server.ts#L80">rpc_server.ts:80</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/b6fae9b35/web/src/rpc_server.ts#L77">rpc_server.ts:77</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/rpc_server.ts#L77">rpc_server.ts:77</a></li>
 						</ul>
 					</aside>
 				</section>
diff --git a/docs/reference/api/typedoc/classes/scalar.html b/docs/reference/api/typedoc/classes/scalar.html
index b7ce071202..8592413b5a 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/b6fae9b35/web/src/runtime.ts#L145">runtime.ts:145</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L145">runtime.ts:145</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/b6fae9b35/web/src/runtime.ts#L145">runtime.ts:145</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L145">runtime.ts:145</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/b6fae9b35/web/src/runtime.ts#L143">runtime.ts:143</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L143">runtime.ts:143</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 3b1230fe87..0825f9c5e4 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/b6fae9b35/web/src/webgpu.ts#L57">webgpu.ts:57</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/webgpu.ts#L50">webgpu.ts:50</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/webgpu.ts#L51">webgpu.ts:51</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/webgpu.ts#L84">webgpu.ts:84</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/webgpu.ts#L170">webgpu.ts:170</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/webgpu.ts#L170">webgpu.ts:170</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/b6fae9b35/web/src/webgpu.ts#L67">webgpu.ts:67</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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 6449193bfa..4369a63b65 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/b6fae9b35/web/src/ctypes.ts#L220">ctypes.ts:220</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L220">ctypes.ts:220</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/b6fae9b35/web/src/ctypes.ts#L216">ctypes.ts:216</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L216">ctypes.ts:216</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/b6fae9b35/web/src/ctypes.ts#L214">ctypes.ts:214</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L214">ctypes.ts:214</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/b6fae9b35/web/src/ctypes.ts#L218">ctypes.ts:218</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L218">ctypes.ts:218</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/b6fae9b35/web/src/ctypes.ts#L226">ctypes.ts:226</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L226">ctypes.ts:226</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/b6fae9b35/web/src/ctypes.ts#L221">ctypes.ts:221</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L221">ctypes.ts:221</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/b6fae9b35/web/src/ctypes.ts#L219">ctypes.ts:219</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L219">ctypes.ts:219</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/b6fae9b35/web/src/ctypes.ts#L223">ctypes.ts:223</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L223">ctypes.ts:223</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/b6fae9b35/web/src/ctypes.ts#L227">ctypes.ts:227</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L227">ctypes.ts:227</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/b6fae9b35/web/src/ctypes.ts#L222">ctypes.ts:222</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L222">ctypes.ts:222</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/b6fae9b35/web/src/ctypes.ts#L228">ctypes.ts:228</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L228">ctypes.ts:228</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/b6fae9b35/web/src/ctypes.ts#L217">ctypes.ts:217</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L217">ctypes.ts:217</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/b6fae9b35/web/src/ctypes.ts#L224">ctypes.ts:224</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L224">ctypes.ts:224</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/b6fae9b35/web/src/ctypes.ts#L225">ctypes.ts:225</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L225">ctypes.ts:225</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/b6fae9b35/web/src/ctypes.ts#L215">ctypes.ts:215</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L215">ctypes.ts:215</a></li>
 						</ul>
 					</aside>
 				</section>
diff --git a/docs/reference/api/typedoc/enums/aynccallbackcode.html b/docs/reference/api/typedoc/enums/aynccallbackcode.html
index 0848567f79..082e6e0738 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/b6fae9b35/web/src/runtime.ts#L676">runtime.ts:676</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L676">runtime.ts:676</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/b6fae9b35/web/src/runtime.ts#L675">runtime.ts:675</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L675">runtime.ts:675</a></li>
 						</ul>
 					</aside>
 				</section>
diff --git a/docs/reference/api/typedoc/enums/dldatatypecode.html b/docs/reference/api/typedoc/enums/dldatatypecode.html
index 87f98f9553..8e38992596 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/b6fae9b35/web/src/runtime.ts#L242">runtime.ts:242</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L242">runtime.ts:242</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/b6fae9b35/web/src/runtime.ts#L240">runtime.ts:240</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L240">runtime.ts:240</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/b6fae9b35/web/src/runtime.ts#L243">runtime.ts:243</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L243">runtime.ts:243</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/b6fae9b35/web/src/runtime.ts#L241">runtime.ts:241</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L241">runtime.ts:241</a></li>
 						</ul>
 					</aside>
 				</section>
diff --git a/docs/reference/api/typedoc/enums/rpcserverstate.html b/docs/reference/api/typedoc/enums/rpcserverstate.html
index 680ec4cf67..7cbdc36baa 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/b6fae9b35/web/src/rpc_server.ts#L27">rpc_server.ts:27</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/rpc_server.ts#L27">rpc_server.ts:27</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/b6fae9b35/web/src/rpc_server.ts#L28">rpc_server.ts:28</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/rpc_server.ts#L28">rpc_server.ts:28</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/b6fae9b35/web/src/rpc_server.ts#L29">rpc_server.ts:29</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/rpc_server.ts#L29">rpc_server.ts:29</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/b6fae9b35/web/src/rpc_server.ts#L32">rpc_server.ts:32</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/rpc_server.ts#L32">rpc_server.ts:32</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/b6fae9b35/web/src/rpc_server.ts#L31">rpc_server.ts:31</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/rpc_server.ts#L31">rpc_server.ts:31</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/b6fae9b35/web/src/rpc_server.ts#L30">rpc_server.ts:30</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/rpc_server.ts#L30">rpc_server.ts:30</a></li>
 						</ul>
 					</aside>
 				</section>
diff --git a/docs/reference/api/typedoc/enums/sizeof.html b/docs/reference/api/typedoc/enums/sizeof.html
index c800148ef7..f771b023e6 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/b6fae9b35/web/src/ctypes.ts#L206">ctypes.ts:206</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L206">ctypes.ts:206</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/b6fae9b35/web/src/ctypes.ts#L207">ctypes.ts:207</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L207">ctypes.ts:207</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/b6fae9b35/web/src/ctypes.ts#L203">ctypes.ts:203</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L203">ctypes.ts:203</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/b6fae9b35/web/src/ctypes.ts#L204">ctypes.ts:204</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L204">ctypes.ts:204</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/b6fae9b35/web/src/ctypes.ts#L201">ctypes.ts:201</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L201">ctypes.ts:201</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/b6fae9b35/web/src/ctypes.ts#L202">ctypes.ts:202</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L202">ctypes.ts:202</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/b6fae9b35/web/src/ctypes.ts#L205">ctypes.ts:205</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L205">ctypes.ts:205</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/b6fae9b35/web/src/ctypes.ts#L200">ctypes.ts:200</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L200">ctypes.ts:200</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/b6fae9b35/web/src/ctypes.ts#L199">ctypes.ts:199</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L199">ctypes.ts:199</a></li>
 						</ul>
 					</aside>
 				</section>
diff --git a/docs/reference/api/typedoc/index.html b/docs/reference/api/typedoc/index.html
index 7af3d96daf..32683ad9bb 100644
--- a/docs/reference/api/typedoc/index.html
+++ b/docs/reference/api/typedoc/index.html
@@ -174,7 +174,7 @@
 					<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/b6fae9b35/web/src/ctypes.ts#L112">ctypes.ts:112</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L112">ctypes.ts:112</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -238,7 +238,7 @@
 					<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/b6fae9b35/web/src/ctypes.ts#L128">ctypes.ts:128</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L128">ctypes.ts:128</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -282,7 +282,7 @@
 					<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/b6fae9b35/web/src/ctypes.ts#L144">ctypes.ts:144</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L144">ctypes.ts:144</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -326,7 +326,7 @@
 					<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/b6fae9b35/web/src/ctypes.ts#L136">ctypes.ts:136</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L136">ctypes.ts:136</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -370,7 +370,7 @@
 					<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/b6fae9b35/web/src/ctypes.ts#L121">ctypes.ts:121</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L121">ctypes.ts:121</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -406,7 +406,7 @@
 					<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/b6fae9b35/web/src/ctypes.ts#L160">ctypes.ts:160</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L160">ctypes.ts:160</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -458,7 +458,7 @@
 					<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/b6fae9b35/web/src/ctypes.ts#L77">ctypes.ts:77</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L77">ctypes.ts:77</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -506,7 +506,7 @@
 					<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/b6fae9b35/web/src/ctypes.ts#L83">ctypes.ts:83</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L83">ctypes.ts:83</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -545,7 +545,7 @@
 					<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/b6fae9b35/web/src/ctypes.ts#L67">ctypes.ts:67</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L67">ctypes.ts:67</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -601,7 +601,7 @@
 					<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/b6fae9b35/web/src/ctypes.ts#L57">ctypes.ts:57</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L57">ctypes.ts:57</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -637,7 +637,7 @@
 					<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/b6fae9b35/web/src/ctypes.ts#L100">ctypes.ts:100</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L100">ctypes.ts:100</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -676,7 +676,7 @@
 					<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/b6fae9b35/web/src/ctypes.ts#L88">ctypes.ts:88</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L88">ctypes.ts:88</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -715,7 +715,7 @@
 					<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/b6fae9b35/web/src/ctypes.ts#L94">ctypes.ts:94</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L94">ctypes.ts:94</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -758,7 +758,7 @@
 					<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/b6fae9b35/web/src/ctypes.ts#L34">ctypes.ts:34</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L34">ctypes.ts:34</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -788,7 +788,7 @@
 					<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/b6fae9b35/web/src/ctypes.ts#L52">ctypes.ts:52</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L52">ctypes.ts:52</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -824,7 +824,7 @@
 					<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/b6fae9b35/web/src/ctypes.ts#L42">ctypes.ts:42</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L42">ctypes.ts:42</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -872,7 +872,7 @@
 					<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/b6fae9b35/web/src/ctypes.ts#L48">ctypes.ts:48</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L48">ctypes.ts:48</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -912,7 +912,7 @@
 					<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/b6fae9b35/web/src/ctypes.ts#L150">ctypes.ts:150</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L150">ctypes.ts:150</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -954,7 +954,7 @@
 					<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/b6fae9b35/web/src/ctypes.ts#L167">ctypes.ts:167</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L167">ctypes.ts:167</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -990,7 +990,7 @@
 					<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/b6fae9b35/web/src/ctypes.ts#L170">ctypes.ts:170</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L170">ctypes.ts:170</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -1026,7 +1026,7 @@
 					<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/b6fae9b35/web/src/ctypes.ts#L187">ctypes.ts:187</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L187">ctypes.ts:187</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -1066,7 +1066,7 @@
 					<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/b6fae9b35/web/src/ctypes.ts#L179">ctypes.ts:179</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L179">ctypes.ts:179</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -1118,7 +1118,7 @@
 					<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/b6fae9b35/web/src/ctypes.ts#L193">ctypes.ts:193</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L193">ctypes.ts:193</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -1154,7 +1154,7 @@
 					<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/b6fae9b35/web/src/webgpu.ts#L25">webgpu.ts:25</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/webgpu.ts#L25">webgpu.ts:25</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -1169,7 +1169,7 @@
 					<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/b6fae9b35/web/src/runtime.ts#L36">runtime.ts:36</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L36">runtime.ts:36</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -1184,7 +1184,7 @@
 					<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/b6fae9b35/web/src/ctypes.ts#L25">ctypes.ts:25</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L25">ctypes.ts:25</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -1199,7 +1199,7 @@
 					<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/b6fae9b35/web/src/ctypes.ts#L28">ctypes.ts:28</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/ctypes.ts#L28">ctypes.ts:28</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -1217,7 +1217,7 @@
 					<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/b6fae9b35/web/src/rpc_server.ts#L36">rpc_server.ts:36</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/rpc_server.ts#L36">rpc_server.ts:36</a></li>
 						</ul>
 					</aside>
 					<div class="tsd-comment tsd-typography">
@@ -1239,7 +1239,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/support.ts#L25">support.ts:25</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/support.ts#L25">support.ts:25</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -1271,7 +1271,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/support.ts#L39">support.ts:39</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/support.ts#L39">support.ts:39</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -1300,7 +1300,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/support.ts#L52">support.ts:52</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/support.ts#L52">support.ts:52</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -1337,7 +1337,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/compact.ts#L38">compact.ts:38</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/compact.ts#L38">compact.ts:38</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -1368,7 +1368,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/webgpu.ts#L30">webgpu.ts:30</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/webgpu.ts#L30">webgpu.ts:30</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -1390,7 +1390,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/environment.ts#L32">environment.ts:32</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/environment.ts#L32">environment.ts:32</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -1421,7 +1421,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/compact.ts#L24">compact.ts:24</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/compact.ts#L24">compact.ts:24</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -1443,7 +1443,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/runtime.ts#L1367">runtime.ts:1367</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L1367">runtime.ts:1367</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -1508,7 +1508,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/b6fae9b35/web/src/support.ts#L62">support.ts:62</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/support.ts#L62">support.ts:62</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -1530,7 +1530,7 @@
 					<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/b6fae9b35/web/src/runtime.ts#L246">runtime.ts:246</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L246">runtime.ts:246</a></li>
 						</ul>
 					</aside>
 					<section class="tsd-panel tsd-member tsd-kind-variable tsd-parent-kind-object-literal">
@@ -1539,7 +1539,7 @@
 						<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/b6fae9b35/web/src/runtime.ts#L247">runtime.ts:247</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L247">runtime.ts:247</a></li>
 							</ul>
 						</aside>
 					</section>
@@ -1549,7 +1549,7 @@
 						<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/b6fae9b35/web/src/runtime.ts#L248">runtime.ts:248</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L248">runtime.ts:248</a></li>
 							</ul>
 						</aside>
 					</section>
@@ -1559,7 +1559,7 @@
 						<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/b6fae9b35/web/src/runtime.ts#L249">runtime.ts:249</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L249">runtime.ts:249</a></li>
 							</ul>
 						</aside>
 					</section>
@@ -1569,7 +1569,7 @@
 						<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/b6fae9b35/web/src/runtime.ts#L250">runtime.ts:250</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L250">runtime.ts:250</a></li>
 							</ul>
 						</aside>
 					</section>
@@ -1580,7 +1580,7 @@
 					<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/b6fae9b35/web/src/runtime.ts#L175">runtime.ts:175</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L175">runtime.ts:175</a></li>
 						</ul>
 					</aside>
 					<section class="tsd-panel tsd-member tsd-kind-variable tsd-parent-kind-object-literal">
@@ -1589,7 +1589,7 @@
 						<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/b6fae9b35/web/src/runtime.ts#L176">runtime.ts:176</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L176">runtime.ts:176</a></li>
 							</ul>
 						</aside>
 					</section>
@@ -1599,7 +1599,7 @@
 						<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/b6fae9b35/web/src/runtime.ts#L180">runtime.ts:180</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L180">runtime.ts:180</a></li>
 							</ul>
 						</aside>
 					</section>
@@ -1609,7 +1609,7 @@
 						<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/b6fae9b35/web/src/runtime.ts#L177">runtime.ts:177</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L177">runtime.ts:177</a></li>
 							</ul>
 						</aside>
 					</section>
@@ -1619,7 +1619,7 @@
 						<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/b6fae9b35/web/src/runtime.ts#L178">runtime.ts:178</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L178">runtime.ts:178</a></li>
 							</ul>
 						</aside>
 					</section>
@@ -1629,7 +1629,7 @@
 						<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/b6fae9b35/web/src/runtime.ts#L179">runtime.ts:179</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L179">runtime.ts:179</a></li>
 							</ul>
 						</aside>
 					</section>
@@ -1640,7 +1640,7 @@
 					<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/b6fae9b35/web/src/runtime.ts#L183">runtime.ts:183</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L183">runtime.ts:183</a></li>
 						</ul>
 					</aside>
 					<section class="tsd-panel tsd-member tsd-kind-variable tsd-parent-kind-object-literal">
@@ -1649,7 +1649,7 @@
 						<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/b6fae9b35/web/src/runtime.ts#L186">runtime.ts:186</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L186">runtime.ts:186</a></li>
 							</ul>
 						</aside>
 					</section>
@@ -1659,7 +1659,7 @@
 						<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/b6fae9b35/web/src/runtime.ts#L184">runtime.ts:184</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L184">runtime.ts:184</a></li>
 							</ul>
 						</aside>
 					</section>
@@ -1669,7 +1669,7 @@
 						<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/b6fae9b35/web/src/runtime.ts#L185">runtime.ts:185</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L185">runtime.ts:185</a></li>
 							</ul>
 						</aside>
 					</section>
@@ -1679,7 +1679,7 @@
 						<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/b6fae9b35/web/src/runtime.ts#L189">runtime.ts:189</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L189">runtime.ts:189</a></li>
 							</ul>
 						</aside>
 					</section>
@@ -1689,7 +1689,7 @@
 						<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/b6fae9b35/web/src/runtime.ts#L187">runtime.ts:187</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L187">runtime.ts:187</a></li>
 							</ul>
 						</aside>
 					</section>
@@ -1699,7 +1699,7 @@
 						<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/b6fae9b35/web/src/runtime.ts#L188">runtime.ts:188</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L188">runtime.ts:188</a></li>
 							</ul>
 						</aside>
 					</section>
@@ -1709,7 +1709,7 @@
 						<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/b6fae9b35/web/src/runtime.ts#L190">runtime.ts:190</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/web/src/runtime.ts#L190">runtime.ts:190</a></li>
 							</ul>
 						</aside>
 					</section>
diff --git a/docs/reference/api/typedoc/interfaces/disposable.html b/docs/reference/api/typedoc/interfaces/disposable.html
index 9879f222ab..0cbcc6e43b 100644
--- a/docs/reference/api/typedoc/interfaces/disposable.html
+++ b/docs/reference/api/typedoc/interfaces/disposable.html
@@ -113,7 +113,7 @@
 					<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/b6fae9b35/web/src/types.ts#L52">types.ts:52</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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 733b2f3826..01e20f9240 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/b6fae9b35/web/src/webgpu.ts#L41">webgpu.ts:41</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/webgpu.ts#L42">webgpu.ts:42</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/webgpu.ts#L40">webgpu.ts:40</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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 71f5ba19c0..8ba516c381 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/b6fae9b35/web/src/types.ts#L34">types.ts:34</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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/b6fae9b35/web/src/types.ts#L39">types.ts:39</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/3aa16f72d/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 0f384a6d20..7845584466 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 954bb297a7..171ac9089d 100644
--- a/docs/topic/vta/tutorials/autotvm/sg_execution_times.html
+++ b/docs/topic/vta/tutorials/autotvm/sg_execution_times.html
@@ -340,7 +340,7 @@
             
   <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>00:26.919</strong> total execution time for <strong>topic_vta_tutorials_autotvm</strong> files:</p>
+<p><strong>00:26.929</strong> total execution time for <strong>topic_vta_tutorials_autotvm</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 82%" />
@@ -349,7 +349,7 @@
 </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>00:26.912</p></td>
+<td><p>00:26.922</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 d5b415dace..f45ccfdfa3 100644
--- a/docs/topic/vta/tutorials/frontend/deploy_classification.html
+++ b/docs/topic/vta/tutorials/frontend/deploy_classification.html
@@ -582,7 +582,7 @@ 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 29.79s!
+resnet18_v1 inference graph built in 30.06s!
 </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 0b90f4197d..bf75d134c1 100644
--- a/docs/topic/vta/tutorials/frontend/deploy_detection.html
+++ b/docs/topic/vta/tutorials/frontend/deploy_detection.html
@@ -600,7 +600,7 @@ 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 20.01s!
+yolov3-tiny inference graph built in 20.30s!
 </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 248dd21ca0..2b5a906fcc 100644
--- a/docs/topic/vta/tutorials/frontend/sg_execution_times.html
+++ b/docs/topic/vta/tutorials/frontend/sg_execution_times.html
@@ -340,7 +340,7 @@
             
   <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>01:41.564</strong> total execution time for <strong>topic_vta_tutorials_frontend</strong> files:</p>
+<p><strong>01:42.643</strong> total execution time for <strong>topic_vta_tutorials_frontend</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 84%" />
@@ -349,11 +349,11 @@
 </colgroup>
 <tbody>
 <tr class="row-odd"><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>00:51.897</p></td>
+<td><p>00:52.575</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><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>00:49.667</p></td>
+<td><p>00:50.068</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 cec5da3b49..985378b4b6 100644
--- a/docs/topic/vta/tutorials/optimize/sg_execution_times.html
+++ b/docs/topic/vta/tutorials/optimize/sg_execution_times.html
@@ -340,7 +340,7 @@
             
   <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>00:03.131</strong> total execution time for <strong>topic_vta_tutorials_optimize</strong> files:</p>
+<p><strong>00:03.112</strong> total execution time for <strong>topic_vta_tutorials_optimize</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 84%" />
@@ -349,7 +349,7 @@
 </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>00:02.681</p></td>
+<td><p>00:02.662</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>
diff --git a/docs/topic/vta/tutorials/sg_execution_times.html b/docs/topic/vta/tutorials/sg_execution_times.html
index 45b01f1638..ac6821c2a6 100644
--- a/docs/topic/vta/tutorials/sg_execution_times.html
+++ b/docs/topic/vta/tutorials/sg_execution_times.html
@@ -340,7 +340,7 @@
             
   <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>00:00.801</strong> total execution time for <strong>topic_vta_tutorials</strong> files:</p>
+<p><strong>00:00.791</strong> total execution time for <strong>topic_vta_tutorials</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 81%" />
@@ -349,11 +349,11 @@
 </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>00:00.429</p></td>
+<td><p>00:00.420</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>00:00.372</p></td>
+<td><p>00:00.370</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 015dd50c7f..0b6c473b6f 100644
--- a/docs/tutorial/auto_scheduler_matmul_x86.html
+++ b/docs/tutorial/auto_scheduler_matmul_x86.html
@@ -491,6 +491,9 @@ trials, we can load the best schedule from the log file and apply it.</p>
 <a href="../reference/api/python/te.html#tvm.te.Schedule" title="tvm.te.Schedule" class="sphx-glr-backref-module-tvm-te sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">sch</span></a><span class="p">,</span> <a href="../reference/api/python/ir.html#tvm.ir.Array" title="tvm.ir.Array" class="sphx-glr-backref-module-tvm-ir sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">args</span></a> <span class="o">=</span> <a href="../reference/api/pyth [...]
 </pre></div>
 </div>
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>.T
+</pre></div>
+</div>
 </div>
 <div class="section" id="inspecting-the-optimized-schedule">
 <h2>Inspecting the Optimized Schedule<a class="headerlink" href="#inspecting-the-optimized-schedule" title="Permalink to this headline">¶</a></h2>
@@ -578,7 +581,7 @@ operator fusion.</p>
 <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: 95.070 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 96.686 ms
 </pre></div>
 </div>
 </div>
@@ -652,7 +655,7 @@ 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  30.940 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  42.599 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 1b7b7afe31..622b9f6d79 100644
--- a/docs/tutorial/autotvm_matmul_x86.html
+++ b/docs/tutorial/autotvm_matmul_x86.html
@@ -679,16 +679,16 @@ 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: 1.58/1.58       result: MeasureResult(costs=(0.169607917,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.8739066123962402, timestamp=1668473609.7446082)        [(&#39;tile_y&#39;, [-1, 32]), (&#39;tile_x&#39;, [-1, 4])],None,25
-No: 2   GFLOPS: 9.28/9.28       result: MeasureResult(costs=(0.028912128200000005,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.7243156433105469, timestamp=1668473610.4245136)       [(&#39;tile_y&#39;, [-1, 512]), (&#39;tile_x&#39;, [-1, 128])],None,79
-No: 3   GFLOPS: 2.30/9.28       result: MeasureResult(costs=(0.11654239640000001,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.0009708404541016, timestamp=1668473613.2164934)        [(&#39;tile_y&#39;, [-1, 1]), (&#39;tile_x&#39;, [-1, 16])],None,40
-No: 4   GFLOPS: 12.58/12.58     result: MeasureResult(costs=(0.0213367548,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6361486911773682, timestamp=1668473614.4972048)       [(&#39;tile_y&#39;, [-1, 64]), (&#39;tile_x&#39;, [-1, 128])],None,76
-No: 5   GFLOPS: 10.59/12.58     result: MeasureResult(costs=(0.025336678600000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5550379753112793, timestamp=1668473615.9628344)       [(&#39;tile_y&#39;, [-1, 1]), (&#39;tile_x&#39;, [-1, 256])],None,80
-No: 6   GFLOPS: 13.69/13.69     result: MeasureResult(costs=(0.019606959599999998,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.48113441467285156, timestamp=1668473616.440613)       [(&#39;tile_y&#39;, [-1, 128]), (&#39;tile_x&#39;, [-1, 64])],None,67
-No: 7   GFLOPS: 2.51/13.69      result: MeasureResult(costs=(0.10688991080000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.8739430904388428, timestamp=1668473618.3178763)        [(&#39;tile_y&#39;, [-1, 8]), (&#39;tile_x&#39;, [-1, 4])],None,23
-No: 8   GFLOPS: 4.16/13.69      result: MeasureResult(costs=(0.0645883018,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.2422730922698975, timestamp=1668473619.5611193)       [(&#39;tile_y&#39;, [-1, 16]), (&#39;tile_x&#39;, [-1, 16])],None,44
-No: 9   GFLOPS: 9.08/13.69      result: MeasureResult(costs=(0.029561042,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6742880344390869, timestamp=1668473620.3487682)        [(&#39;tile_y&#39;, [-1, 1]), (&#39;tile_x&#39;, [-1, 128])],None,70
-No: 10  GFLOPS: 11.65/13.69     result: MeasureResult(costs=(0.0230363428,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.528548002243042, timestamp=1668473620.8892539)        [(&#39;tile_y&#39;, [-1, 16]), (&#39;tile_x&#39;, [-1, 256])],None,84
+No: 1   GFLOPS: 0.51/0.51       result: MeasureResult(costs=(0.5258266778,), error_no=MeasureErrorNo.NO_ERROR, all_cost=8.5500328540802, timestamp=1668479848.6790347)  [(&#39;tile_y&#39;, [-1, 256]), (&#39;tile_x&#39;, [-1, 1])],None,8
+No: 2   GFLOPS: 11.67/11.67     result: MeasureResult(costs=(0.0229983434,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5694522857666016, timestamp=1668479849.2623055)       [(&#39;tile_y&#39;, [-1, 2]), (&#39;tile_x&#39;, [-1, 256])],None,81
+No: 3   GFLOPS: 11.76/11.76     result: MeasureResult(costs=(0.022829464,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5173089504241943, timestamp=1668479850.55224)  [(&#39;tile_y&#39;, [-1, 256]), (&#39;tile_x&#39;, [-1, 256])],None,88
+No: 4   GFLOPS: 7.39/11.76      result: MeasureResult(costs=(0.036339880000000005,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.7135937213897705, timestamp=1668479852.0477846)       [(&#39;tile_y&#39;, [-1, 512]), (&#39;tile_x&#39;, [-1, 32])],None,59
+No: 5   GFLOPS: 2.70/11.76      result: MeasureResult(costs=(0.0992417416,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.7401914596557617, timestamp=1668479854.8115833)       [(&#39;tile_y&#39;, [-1, 4]), (&#39;tile_x&#39;, [-1, 4])],None,22
+No: 6   GFLOPS: 8.73/11.76      result: MeasureResult(costs=(0.030753971999999997,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.8017129898071289, timestamp=1668479855.4690278)       [(&#39;tile_y&#39;, [-1, 16]), (&#39;tile_x&#39;, [-1, 64])],None,64
+No: 7   GFLOPS: 11.80/11.80     result: MeasureResult(costs=(0.022747619599999998,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5359385013580322, timestamp=1668479856.0026648)       [(&#39;tile_y&#39;, [-1, 32]), (&#39;tile_x&#39;, [-1, 256])],None,85
+No: 8   GFLOPS: 9.91/11.80      result: MeasureResult(costs=(0.027076459999999997,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.7049639225006104, timestamp=1668479856.6463385)       [(&#39;tile_y&#39;, [-1, 1]), (&#39;tile_x&#39;, [-1, 128])],None,70
+No: 9   GFLOPS: 12.69/12.69     result: MeasureResult(costs=(0.021145623800000003,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.4574155807495117, timestamp=1668479857.2369988)       [(&#39;tile_y&#39;, [-1, 32]), (&#39;tile_x&#39;, [-1, 512])],None,95
+No: 10  GFLOPS: 12.29/12.69     result: MeasureResult(costs=(0.0218373526,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.4883308410644531, timestamp=1668479857.7563374)       [(&#39;tile_y&#39;, [-1, 8]), (&#39;tile_x&#39;, [-1, 256])],None,83
 </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 1a531ad957..fcea632a88 100644
--- a/docs/tutorial/autotvm_relay_x86.html
+++ b/docs/tutorial/autotvm_relay_x86.html
@@ -560,7 +560,7 @@ 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;: 517.1077569899899, &#39;median&#39;: 516.9921277999947, &#39;std&#39;: 1.3162579716279146}
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>{&#39;mean&#39;: 510.4595590599991, &#39;median&#39;: 509.4302212499997, &#39;std&#39;: 3.3041499135301597}
 </pre></div>
 </div>
 </div>
@@ -712,179 +712,179 @@ 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:   21.50/  21.50 GFLOPS | Progress: (4/20) | 8.66 s
-[Task  1/25]  Current/Best:    5.70/  21.50 GFLOPS | Progress: (8/20) | 11.62 s
-[Task  1/25]  Current/Best:   23.34/  23.34 GFLOPS | Progress: (12/20) | 15.11 s
-[Task  1/25]  Current/Best:   11.92/  23.34 GFLOPS | Progress: (16/20) | 17.45 s
-[Task  1/25]  Current/Best:   12.54/  23.34 GFLOPS | Progress: (20/20) | 20.20 s Done.
+[Task  1/25]  Current/Best:   13.77/  23.62 GFLOPS | Progress: (4/20) | 8.04 s
+[Task  1/25]  Current/Best:   14.45/  23.62 GFLOPS | Progress: (8/20) | 10.84 s
+[Task  1/25]  Current/Best:    8.66/  23.62 GFLOPS | Progress: (12/20) | 12.80 s
+[Task  1/25]  Current/Best:    9.09/  23.62 GFLOPS | Progress: (16/20) | 16.11 s
+[Task  1/25]  Current/Best:   12.47/  23.62 GFLOPS | Progress: (20/20) | 19.01 s Done.
 
 [Task  2/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task  2/25]  Current/Best:    6.58/  12.38 GFLOPS | Progress: (4/20) | 3.09 s
-[Task  2/25]  Current/Best:    2.15/  17.86 GFLOPS | Progress: (8/20) | 4.84 s
-[Task  2/25]  Current/Best:    6.11/  20.20 GFLOPS | Progress: (12/20) | 6.09 s
-[Task  2/25]  Current/Best:   15.42/  20.20 GFLOPS | Progress: (16/20) | 7.42 s
-[Task  2/25]  Current/Best:   10.15/  20.20 GFLOPS | Progress: (20/20) | 9.04 s Done.
+[Task  2/25]  Current/Best:   15.75/  16.13 GFLOPS | Progress: (4/20) | 2.89 s
+[Task  2/25]  Current/Best:   15.36/  16.13 GFLOPS | Progress: (8/20) | 4.21 s
+[Task  2/25]  Current/Best:   12.12/  20.99 GFLOPS | Progress: (12/20) | 6.92 s
+[Task  2/25]  Current/Best:    7.47/  20.99 GFLOPS | Progress: (16/20) | 8.36 s
+[Task  2/25]  Current/Best:    9.97/  20.99 GFLOPS | Progress: (20/20) | 11.24 s Done.
 
 [Task  3/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task  3/25]  Current/Best:   10.91/  18.23 GFLOPS | Progress: (4/20) | 3.53 s
-[Task  3/25]  Current/Best:    4.54/  18.23 GFLOPS | Progress: (8/20) | 5.59 s
-[Task  3/25]  Current/Best:   11.77/  18.23 GFLOPS | Progress: (12/20) | 7.76 s
-[Task  3/25]  Current/Best:    6.57/  22.34 GFLOPS | Progress: (16/20) | 9.54 s
-[Task  3/25]  Current/Best:   22.99/  22.99 GFLOPS | Progress: (20/20) | 11.57 s Done.
+[Task  3/25]  Current/Best:   16.00/  16.00 GFLOPS | Progress: (4/20) | 3.50 s
+[Task  3/25]  Current/Best:   18.67/  18.67 GFLOPS | Progress: (8/20) | 5.70 s
+[Task  3/25]  Current/Best:   12.08/  18.67 GFLOPS | Progress: (12/20) | 8.89 s
+[Task  3/25]  Current/Best:   16.40/  19.83 GFLOPS | Progress: (16/20) | 11.43 s
+[Task  3/25]  Current/Best:   14.89/  19.83 GFLOPS | Progress: (20/20) | 13.02 s Done.
 
 [Task  4/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task  4/25]  Current/Best:    5.93/   9.78 GFLOPS | Progress: (4/20) | 3.43 s
-[Task  4/25]  Current/Best:   15.49/  20.30 GFLOPS | Progress: (8/20) | 4.78 s
-[Task  4/25]  Current/Best:   12.71/  20.30 GFLOPS | Progress: (12/20) | 6.54 s
-[Task  4/25]  Current/Best:   12.28/  20.30 GFLOPS | Progress: (16/20) | 8.62 s
-[Task  4/25]  Current/Best:   10.79/  20.30 GFLOPS | Progress: (20/20) | 19.63 s
-[Task  5/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
-
-[Task  5/25]  Current/Best:    8.78/  22.55 GFLOPS | Progress: (4/20) | 3.60 s
-[Task  5/25]  Current/Best:   16.45/  22.55 GFLOPS | Progress: (8/20) | 5.00 s
-[Task  5/25]  Current/Best:   19.75/  22.55 GFLOPS | Progress: (12/20) | 6.75 s
-[Task  5/25]  Current/Best:   20.60/  22.55 GFLOPS | Progress: (16/20) | 8.36 s
-[Task  5/25]  Current/Best:    9.46/  22.55 GFLOPS | Progress: (20/20) | 10.39 s Done.
+[Task  4/25]  Current/Best:    9.43/  11.94 GFLOPS | Progress: (4/20) | 4.20 s
+[Task  4/25]  Current/Best:   17.68/  17.68 GFLOPS | Progress: (8/20) | 8.22 s
+[Task  4/25]  Current/Best:   13.76/  17.68 GFLOPS | Progress: (12/20) | 9.88 s
+[Task  4/25]  Current/Best:   16.36/  17.68 GFLOPS | Progress: (16/20) | 15.00 s
+[Task  4/25]  Current/Best:   18.56/  20.03 GFLOPS | Progress: (20/20) | 16.71 s Done.
+
+[Task  5/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
+[Task  5/25]  Current/Best:    3.62/  10.48 GFLOPS | Progress: (4/20) | 3.57 s
+[Task  5/25]  Current/Best:   18.20/  18.20 GFLOPS | Progress: (8/20) | 5.22 s
+[Task  5/25]  Current/Best:   15.86/  19.98 GFLOPS | Progress: (12/20) | 7.04 s
+[Task  5/25]  Current/Best:   13.96/  19.98 GFLOPS | Progress: (16/20) | 9.21 s
+[Task  5/25]  Current/Best:   19.68/  19.98 GFLOPS | Progress: (20/20) | 11.06 s Done.
 
 [Task  6/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task  6/25]  Current/Best:   10.04/  18.64 GFLOPS | Progress: (4/20) | 5.29 s
-[Task  6/25]  Current/Best:   22.65/  22.65 GFLOPS | Progress: (8/20) | 9.32 s
-[Task  6/25]  Current/Best:    1.46/  22.65 GFLOPS | Progress: (12/20) | 12.60 s
-[Task  6/25]  Current/Best:   15.57/  22.65 GFLOPS | Progress: (16/20) | 15.05 s
-[Task  6/25]  Current/Best:   17.95/  22.65 GFLOPS | Progress: (20/20) | 16.99 s Done.
+[Task  6/25]  Current/Best:    3.48/  19.87 GFLOPS | Progress: (4/20) | 4.11 s
+[Task  6/25]  Current/Best:   16.29/  19.87 GFLOPS | Progress: (8/20) | 6.64 s
+[Task  6/25]  Current/Best:    4.20/  19.87 GFLOPS | Progress: (12/20) | 10.53 s
+[Task  6/25]  Current/Best:   13.42/  19.87 GFLOPS | Progress: (16/20) | 13.12 s
+[Task  6/25]  Current/Best:   18.79/  19.87 GFLOPS | Progress: (20/20) | 16.57 s Done.
 
 [Task  7/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task  7/25]  Current/Best:    7.49/  18.95 GFLOPS | Progress: (4/20) | 3.50 s
-[Task  7/25]  Current/Best:   11.41/  18.95 GFLOPS | Progress: (8/20) | 7.24 s
-[Task  7/25]  Current/Best:   17.22/  18.95 GFLOPS | Progress: (12/20) | 9.20 s
-[Task  7/25]  Current/Best:    1.58/  18.95 GFLOPS | Progress: (16/20) | 13.60 s
-[Task  7/25]  Current/Best:    6.26/  18.95 GFLOPS | Progress: (20/20) | 16.71 s Done.
+[Task  7/25]  Current/Best:   11.69/  20.81 GFLOPS | Progress: (4/20) | 4.27 s
+[Task  7/25]  Current/Best:   15.93/  20.81 GFLOPS | Progress: (8/20) | 6.68 s
+[Task  7/25]  Current/Best:   12.81/  20.81 GFLOPS | Progress: (12/20) | 9.28 s
+[Task  7/25]  Current/Best:   19.42/  20.81 GFLOPS | Progress: (16/20) | 11.14 s
+[Task  7/25]  Current/Best:   15.32/  20.81 GFLOPS | Progress: (20/20) | 13.39 s Done.
 
 [Task  8/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task  8/25]  Current/Best:   12.83/  16.56 GFLOPS | Progress: (4/20) | 4.11 s
-[Task  8/25]  Current/Best:    8.48/  16.56 GFLOPS | Progress: (8/20) | 6.78 s
-[Task  8/25]  Current/Best:    3.27/  16.56 GFLOPS | Progress: (12/20) | 12.79 s
-[Task  8/25]  Current/Best:   16.12/  16.56 GFLOPS | Progress: (16/20) | 19.38 s
-[Task  8/25]  Current/Best:    3.44/  16.56 GFLOPS | Progress: (20/20) | 22.67 s Done.
+[Task  8/25]  Current/Best:   10.79/  17.18 GFLOPS | Progress: (4/20) | 7.52 s
+[Task  8/25]  Current/Best:    5.81/  17.18 GFLOPS | Progress: (8/20) | 9.82 s
+[Task  8/25]  Current/Best:    4.03/  22.09 GFLOPS | Progress: (12/20) | 13.01 s
+[Task  8/25]  Current/Best:   16.68/  22.09 GFLOPS | Progress: (16/20) | 19.09 s
+[Task  8/25]  Current/Best:   20.21/  22.09 GFLOPS | Progress: (20/20) | 21.80 s Done.
 
 [Task  9/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task  9/25]  Current/Best:   12.46/  20.88 GFLOPS | Progress: (4/20) | 2.77 s
-[Task  9/25]  Current/Best:   12.62/  20.88 GFLOPS | Progress: (8/20) | 4.25 s
-[Task  9/25]  Current/Best:    8.35/  20.88 GFLOPS | Progress: (12/20) | 8.81 s
-[Task  9/25]  Current/Best:    6.59/  20.88 GFLOPS | Progress: (16/20) | 19.70 s
-[Task  9/25]  Current/Best:   15.70/  20.88 GFLOPS | Progress: (20/20) | 22.33 s
+[Task  9/25]  Current/Best:    6.98/  15.94 GFLOPS | Progress: (4/20) | 3.50 s
+[Task  9/25]  Current/Best:   14.25/  15.94 GFLOPS | Progress: (8/20) | 5.56 s
+[Task  9/25]  Current/Best:   23.09/  23.09 GFLOPS | Progress: (12/20) | 8.70 s
+[Task  9/25]  Current/Best:    6.91/  23.09 GFLOPS | Progress: (16/20) | 10.27 s
+[Task  9/25]  Current/Best:    3.40/  23.09 GFLOPS | Progress: (20/20) | 13.67 s Done.
+
 [Task 10/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 10/25]  Current/Best:   17.55/  17.55 GFLOPS | Progress: (4/20) | 3.36 s
-[Task 10/25]  Current/Best:   11.88/  17.55 GFLOPS | Progress: (8/20) | 6.18 s
-[Task 10/25]  Current/Best:   11.52/  17.55 GFLOPS | Progress: (12/20) | 7.67 s
-[Task 10/25]  Current/Best:   10.39/  17.55 GFLOPS | Progress: (16/20) | 10.13 s
-[Task 10/25]  Current/Best:    1.61/  17.55 GFLOPS | Progress: (20/20) | 12.34 s Done.
+[Task 10/25]  Current/Best:   14.30/  18.61 GFLOPS | Progress: (4/20) | 3.11 s
+[Task 10/25]  Current/Best:   13.06/  18.61 GFLOPS | Progress: (8/20) | 4.59 s
+[Task 10/25]  Current/Best:    4.76/  18.61 GFLOPS | Progress: (12/20) | 7.81 s
+[Task 10/25]  Current/Best:    9.82/  20.08 GFLOPS | Progress: (16/20) | 9.46 s
+[Task 10/25]  Current/Best:   14.06/  20.08 GFLOPS | Progress: (20/20) | 11.99 s Done.
 
 [Task 11/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 11/25]  Current/Best:   22.13/  24.00 GFLOPS | Progress: (4/20) | 3.75 s
-[Task 11/25]  Current/Best:   12.05/  24.00 GFLOPS | Progress: (8/20) | 6.45 s
-[Task 11/25]  Current/Best:    9.81/  24.00 GFLOPS | Progress: (12/20) | 8.66 s
-[Task 11/25]  Current/Best:   13.03/  24.00 GFLOPS | Progress: (16/20) | 10.78 s
-[Task 11/25]  Current/Best:   18.50/  24.00 GFLOPS | Progress: (20/20) | 13.63 s Done.
+[Task 11/25]  Current/Best:   15.89/  15.89 GFLOPS | Progress: (4/20) | 5.06 s
+[Task 11/25]  Current/Best:   16.23/  16.23 GFLOPS | Progress: (8/20) | 7.06 s
+[Task 11/25]  Current/Best:   12.32/  20.44 GFLOPS | Progress: (12/20) | 9.33 s
+[Task 11/25]  Current/Best:   23.59/  23.59 GFLOPS | Progress: (16/20) | 10.98 s
+[Task 11/25]  Current/Best:   13.06/  23.59 GFLOPS | Progress: (20/20) | 13.55 s Done.
 
 [Task 12/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 12/25]  Current/Best:    6.76/  20.68 GFLOPS | Progress: (4/20) | 3.48 s
-[Task 12/25]  Current/Best:   15.02/  20.68 GFLOPS | Progress: (8/20) | 5.13 s
-[Task 12/25]  Current/Best:   16.30/  20.68 GFLOPS | Progress: (12/20) | 7.10 s
-[Task 12/25]  Current/Best:   14.75/  20.68 GFLOPS | Progress: (16/20) | 9.51 s
-[Task 12/25]  Current/Best:   16.67/  20.68 GFLOPS | Progress: (20/20) | 11.51 s Done.
+[Task 12/25]  Current/Best:    9.96/  14.34 GFLOPS | Progress: (4/20) | 4.60 s
+[Task 12/25]  Current/Best:    5.28/  18.11 GFLOPS | Progress: (8/20) | 11.03 s
+[Task 12/25]  Current/Best:    9.19/  18.11 GFLOPS | Progress: (12/20) | 19.82 s
+[Task 12/25]  Current/Best:    9.69/  18.11 GFLOPS | Progress: (16/20) | 26.30 s
+[Task 12/25]  Current/Best:   20.94/  20.94 GFLOPS | Progress: (20/20) | 31.31 s Done.
 
 [Task 13/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 13/25]  Current/Best:    1.57/   9.69 GFLOPS | Progress: (4/20) | 6.96 s
-[Task 13/25]  Current/Best:   11.73/  17.74 GFLOPS | Progress: (8/20) | 10.85 s
-[Task 13/25]  Current/Best:   18.98/  20.09 GFLOPS | Progress: (12/20) | 13.81 s
-[Task 13/25]  Current/Best:    7.14/  20.09 GFLOPS | Progress: (16/20) | 16.43 s
-[Task 13/25]  Current/Best:    4.69/  21.30 GFLOPS | Progress: (20/20) | 18.83 s Done.
+[Task 13/25]  Current/Best:   11.26/  11.86 GFLOPS | Progress: (4/20) | 4.08 s
+[Task 13/25]  Current/Best:   10.13/  15.40 GFLOPS | Progress: (8/20) | 8.06 s
+[Task 13/25]  Current/Best:   11.77/  20.87 GFLOPS | Progress: (12/20) | 11.12 s
+[Task 13/25]  Current/Best:   11.84/  20.87 GFLOPS | Progress: (16/20) | 14.26 s
+[Task 13/25]  Current/Best:   11.54/  20.87 GFLOPS | Progress: (20/20) | 17.49 s Done.
 
 [Task 14/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 14/25]  Current/Best:    9.38/  17.04 GFLOPS | Progress: (4/20) | 3.52 s
-[Task 14/25]  Current/Best:   14.23/  17.04 GFLOPS | Progress: (8/20) | 6.22 s
-[Task 14/25]  Current/Best:   21.33/  21.33 GFLOPS | Progress: (12/20) | 7.77 s
-[Task 14/25]  Current/Best:   11.77/  21.33 GFLOPS | Progress: (16/20) | 11.56 s Done.
-
-[Task 14/25]  Current/Best:    8.75/  21.33 GFLOPS | Progress: (20/20) | 14.22 s Done.
-
+[Task 14/25]  Current/Best:    6.42/  18.34 GFLOPS | Progress: (4/20) | 7.42 s
+[Task 14/25]  Current/Best:   12.82/  18.34 GFLOPS | Progress: (8/20) | 11.32 s
+[Task 14/25]  Current/Best:   18.86/  18.86 GFLOPS | Progress: (12/20) | 12.48 s
+[Task 14/25]  Current/Best:    3.01/  18.86 GFLOPS | Progress: (16/20) | 15.45 s
+[Task 14/25]  Current/Best:   10.44/  18.86 GFLOPS | Progress: (20/20) | 20.26 s
 [Task 15/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 15/25]  Current/Best:    4.96/  17.69 GFLOPS | Progress: (4/20) | 6.10 s
-[Task 15/25]  Current/Best:   11.85/  17.69 GFLOPS | Progress: (8/20) | 8.24 s
-[Task 15/25]  Current/Best:   12.71/  19.69 GFLOPS | Progress: (12/20) | 12.49 s
-[Task 15/25]  Current/Best:   16.93/  19.69 GFLOPS | Progress: (16/20) | 13.93 s
-[Task 15/25]  Current/Best:   22.77/  22.77 GFLOPS | Progress: (20/20) | 15.63 s
+[Task 15/25]  Current/Best:   18.48/  18.48 GFLOPS | Progress: (4/20) | 3.09 s
+[Task 15/25]  Current/Best:   18.87/  18.87 GFLOPS | Progress: (8/20) | 5.51 s
+[Task 15/25]  Current/Best:   23.60/  23.60 GFLOPS | Progress: (12/20) | 7.08 s
+[Task 15/25]  Current/Best:   12.11/  23.60 GFLOPS | Progress: (16/20) | 9.40 s
+[Task 15/25]  Current/Best:   18.75/  23.60 GFLOPS | Progress: (20/20) | 10.69 s Done.
+
 [Task 16/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 16/25]  Current/Best:   14.80/  14.80 GFLOPS | Progress: (4/20) | 4.21 s
-[Task 16/25]  Current/Best:   17.28/  18.67 GFLOPS | Progress: (8/20) | 5.55 s
-[Task 16/25]  Current/Best:   11.45/  18.67 GFLOPS | Progress: (12/20) | 8.13 s
-[Task 16/25]  Current/Best:    4.37/  18.67 GFLOPS | Progress: (16/20) | 10.35 s
-[Task 16/25]  Current/Best:   19.64/  23.22 GFLOPS | Progress: (20/20) | 11.56 s Done.
+[Task 16/25]  Current/Best:    5.96/  17.77 GFLOPS | Progress: (4/20) | 3.11 s
+[Task 16/25]  Current/Best:   11.78/  21.12 GFLOPS | Progress: (8/20) | 6.83 s
+[Task 16/25]  Current/Best:   17.44/  21.12 GFLOPS | Progress: (12/20) | 10.12 s
+[Task 16/25]  Current/Best:   13.60/  21.12 GFLOPS | Progress: (16/20) | 11.65 s
+[Task 16/25]  Current/Best:   13.57/  21.12 GFLOPS | Progress: (20/20) | 15.39 s Done.
 
 [Task 17/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 17/25]  Current/Best:    8.53/  19.47 GFLOPS | Progress: (4/20) | 4.14 s
-[Task 17/25]  Current/Best:    6.16/  19.47 GFLOPS | Progress: (8/20) | 6.75 s
-[Task 17/25]  Current/Best:   19.70/  23.06 GFLOPS | Progress: (12/20) | 9.14 s
-[Task 17/25]  Current/Best:    6.10/  23.06 GFLOPS | Progress: (16/20) | 12.36 s
-[Task 17/25]  Current/Best:   10.13/  23.06 GFLOPS | Progress: (20/20) | 14.24 s Done.
+[Task 17/25]  Current/Best:    1.56/  20.25 GFLOPS | Progress: (4/20) | 5.17 s Done.
+
+[Task 17/25]  Current/Best:    6.35/  20.25 GFLOPS | Progress: (8/20) | 7.95 s
+[Task 17/25]  Current/Best:   10.03/  21.03 GFLOPS | Progress: (12/20) | 9.95 s
+[Task 17/25]  Current/Best:   15.80/  21.03 GFLOPS | Progress: (16/20) | 12.17 s
+[Task 17/25]  Current/Best:   18.62/  21.03 GFLOPS | Progress: (20/20) | 13.78 s Done.
 
 [Task 18/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 18/25]  Current/Best:   11.38/  15.02 GFLOPS | Progress: (4/20) | 5.47 s
-[Task 18/25]  Current/Best:   12.48/  18.84 GFLOPS | Progress: (8/20) | 7.55 s
-[Task 18/25]  Current/Best:   20.26/  20.26 GFLOPS | Progress: (12/20) | 9.43 s
-[Task 18/25]  Current/Best:   10.94/  20.26 GFLOPS | Progress: (16/20) | 13.81 s
-[Task 18/25]  Current/Best:   11.31/  20.26 GFLOPS | Progress: (20/20) | 16.04 s Done.
+[Task 18/25]  Current/Best:    7.34/  14.22 GFLOPS | Progress: (4/20) | 5.10 s
+[Task 18/25]  Current/Best:   14.22/  16.20 GFLOPS | Progress: (8/20) | 7.24 s
+[Task 18/25]  Current/Best:    9.32/  16.38 GFLOPS | Progress: (12/20) | 14.36 s
+[Task 18/25]  Current/Best:   12.04/  19.89 GFLOPS | Progress: (16/20) | 16.00 s
+[Task 18/25]  Current/Best:   12.06/  22.30 GFLOPS | Progress: (20/20) | 17.75 s Done.
 
 [Task 19/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 19/25]  Current/Best:    4.41/  10.48 GFLOPS | Progress: (4/20) | 5.26 s
-[Task 19/25]  Current/Best:    6.63/  17.88 GFLOPS | Progress: (8/20) | 7.59 s
-[Task 19/25]  Current/Best:    9.05/  17.88 GFLOPS | Progress: (12/20) | 10.50 s
-[Task 19/25]  Current/Best:   10.49/  18.01 GFLOPS | Progress: (16/20) | 13.87 s
-[Task 19/25]  Current/Best:   10.42/  20.98 GFLOPS | Progress: (20/20) | 17.08 s Done.
+[Task 19/25]  Current/Best:   11.94/  19.08 GFLOPS | Progress: (4/20) | 4.51 s
+[Task 19/25]  Current/Best:    8.61/  19.08 GFLOPS | Progress: (8/20) | 7.48 s
+[Task 19/25]  Current/Best:   18.35/  19.08 GFLOPS | Progress: (12/20) | 11.00 s
+[Task 19/25]  Current/Best:   18.63/  19.08 GFLOPS | Progress: (16/20) | 14.26 s
+[Task 19/25]  Current/Best:   18.77/  19.08 GFLOPS | Progress: (20/20) | 20.38 s Done.
 
 [Task 20/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 20/25]  Current/Best:   18.00/  18.00 GFLOPS | Progress: (4/20) | 3.85 s
-[Task 20/25]  Current/Best:    4.93/  18.92 GFLOPS | Progress: (8/20) | 7.97 s
-[Task 20/25]  Current/Best:    2.26/  18.92 GFLOPS | Progress: (12/20) | 10.59 s
-[Task 20/25]  Current/Best:   16.09/  18.92 GFLOPS | Progress: (16/20) | 13.97 s
-[Task 20/25]  Current/Best:    6.68/  18.92 GFLOPS | Progress: (20/20) | 16.09 s
-[Task 21/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 21/25]  Current/Best:    1.62/  19.59 GFLOPS | Progress: (4/20) | 3.93 s
-[Task 21/25]  Current/Best:   10.36/  19.59 GFLOPS | Progress: (8/20) | 6.21 s Done.
- Done.
+[Task 20/25]  Current/Best:   16.36/  16.36 GFLOPS | Progress: (4/20) | 3.95 s
+[Task 20/25]  Current/Best:   13.96/  16.36 GFLOPS | Progress: (8/20) | 6.15 s
+[Task 20/25]  Current/Best:    9.67/  16.36 GFLOPS | Progress: (12/20) | 9.30 s
+[Task 20/25]  Current/Best:   15.77/  17.66 GFLOPS | Progress: (16/20) | 11.23 s
+[Task 20/25]  Current/Best:   13.44/  18.37 GFLOPS | Progress: (20/20) | 13.29 s Done.
 
-[Task 21/25]  Current/Best:    3.13/  19.59 GFLOPS | Progress: (12/20) | 8.62 s
-[Task 21/25]  Current/Best:   18.18/  22.09 GFLOPS | Progress: (16/20) | 10.03 s
-[Task 21/25]  Current/Best:    9.40/  22.09 GFLOPS | Progress: (20/20) | 11.31 s
+[Task 21/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
+[Task 21/25]  Current/Best:   10.58/  19.75 GFLOPS | Progress: (4/20) | 4.55 s
+[Task 21/25]  Current/Best:   18.44/  19.75 GFLOPS | Progress: (8/20) | 6.51 s
+[Task 21/25]  Current/Best:    7.69/  19.75 GFLOPS | Progress: (12/20) | 7.96 s
+[Task 21/25]  Current/Best:    2.26/  19.75 GFLOPS | Progress: (16/20) | 12.10 s
+[Task 21/25]  Current/Best:    6.52/  19.75 GFLOPS | Progress: (20/20) | 13.62 s
 [Task 22/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 22/25]  Current/Best:   14.53/  19.59 GFLOPS | Progress: (4/20) | 5.00 s
-[Task 22/25]  Current/Best:    5.30/  19.59 GFLOPS | Progress: (8/20) | 9.72 s
-[Task 22/25]  Current/Best:   15.46/  19.59 GFLOPS | Progress: (12/20) | 11.62 s
-[Task 22/25]  Current/Best:   18.21/  19.59 GFLOPS | Progress: (16/20) | 13.79 s
-[Task 22/25]  Current/Best:    6.88/  19.59 GFLOPS | Progress: (20/20) | 17.23 s Done.
+[Task 22/25]  Current/Best:   11.05/  13.18 GFLOPS | Progress: (4/20) | 3.10 s
+[Task 22/25]  Current/Best:   19.84/  19.84 GFLOPS | Progress: (8/20) | 4.34 s
+[Task 22/25]  Current/Best:   10.94/  19.84 GFLOPS | Progress: (12/20) | 5.79 s
+[Task 22/25]  Current/Best:    4.44/  19.84 GFLOPS | Progress: (16/20) | 9.21 s
+[Task 22/25]  Current/Best:    9.85/  19.84 GFLOPS | Progress: (20/20) | 11.60 s Done.
 
 [Task 23/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 23/25]  Current/Best:   18.72/  18.72 GFLOPS | Progress: (4/20) | 5.50 s
-[Task 23/25]  Current/Best:    3.08/  18.72 GFLOPS | Progress: (8/20) | 11.20 s
-[Task 23/25]  Current/Best:   11.90/  18.72 GFLOPS | Progress: (12/20) | 13.33 s
-[Task 23/25]  Current/Best:   20.95/  20.95 GFLOPS | Progress: (16/20) | 16.14 s
-[Task 23/25]  Current/Best:   10.15/  20.95 GFLOPS | Progress: (20/20) | 19.25 s Done.
+[Task 23/25]  Current/Best:    3.08/  13.93 GFLOPS | Progress: (4/20) | 4.81 s
+[Task 23/25]  Current/Best:    9.72/  22.13 GFLOPS | Progress: (8/20) | 7.49 s
+[Task 23/25]  Current/Best:   22.97/  22.97 GFLOPS | Progress: (12/20) | 10.22 s
+[Task 23/25]  Current/Best:    8.76/  22.97 GFLOPS | Progress: (16/20) | 12.63 s
+[Task 23/25]  Current/Best:   10.78/  22.97 GFLOPS | Progress: (20/20) | 16.24 s Done.
 
 [Task 24/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 24/25]  Current/Best:    8.11/   9.02 GFLOPS | Progress: (4/20) | 12.27 s
-[Task 24/25]  Current/Best:    5.15/   9.02 GFLOPS | Progress: (8/20) | 16.70 s
-[Task 24/25]  Current/Best:    6.74/   9.02 GFLOPS | Progress: (12/20) | 26.77 s
-[Task 24/25]  Current/Best:    1.62/   9.02 GFLOPS | Progress: (16/20) | 37.51 s
-[Task 24/25]  Current/Best:    2.90/   9.02 GFLOPS | Progress: (20/20) | 49.12 s
+[Task 24/25]  Current/Best:    4.70/   9.67 GFLOPS | Progress: (4/20) | 4.68 s
+[Task 24/25]  Current/Best:    7.15/   9.67 GFLOPS | Progress: (8/20) | 12.14 s
+[Task 24/25]  Current/Best:    3.32/   9.67 GFLOPS | Progress: (12/20) | 23.75 s
+[Task 24/25]  Current/Best:    2.21/   9.97 GFLOPS | Progress: (16/20) | 35.47 s
+[Task 24/25]  Current/Best:    5.98/   9.97 GFLOPS | Progress: (20/20) | 41.58 s
 [Task 25/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
  Done.
 
-[Task 25/25]  Current/Best:    1.53/   7.76 GFLOPS | Progress: (4/20) | 13.03 s
-[Task 25/25]  Current/Best:    6.16/   7.76 GFLOPS | Progress: (8/20) | 18.28 s
-[Task 25/25]  Current/Best:    2.98/   8.84 GFLOPS | Progress: (12/20) | 28.78 s
-[Task 25/25]  Current/Best:    3.91/   8.84 GFLOPS | Progress: (16/20) | 40.36 s
-[Task 25/25]  Current/Best:    9.01/   9.01 GFLOPS | Progress: (20/20) | 51.07 s
+[Task 25/25]  Current/Best:    7.95/   8.73 GFLOPS | Progress: (4/20) | 6.61 s
+[Task 25/25]  Current/Best:    8.01/   8.73 GFLOPS | Progress: (8/20) | 17.36 s
+[Task 25/25]  Current/Best:    5.71/   8.73 GFLOPS | Progress: (12/20) | 19.57 s
+[Task 25/25]  Current/Best:    1.54/   9.40 GFLOPS | Progress: (16/20) | 24.44 s
+[Task 25/25]  Current/Best:    5.76/   9.40 GFLOPS | Progress: (20/20) | 26.00 s
 </pre></div>
 </div>
 <p>The output from this tuning process will look something like this:</p>
@@ -983,8 +983,8 @@ 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;: 405.79506909000884, &#39;median&#39;: 405.99190305001684, &#39;std&#39;: 0.8678081614214388}
-unoptimized: {&#39;mean&#39;: 517.1077569899899, &#39;median&#39;: 516.9921277999947, &#39;std&#39;: 1.3162579716279146}
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>optimized: {&#39;mean&#39;: 407.88420194999844, &#39;median&#39;: 405.84936035000965, &#39;std&#39;: 3.4246055305415424}
+unoptimized: {&#39;mean&#39;: 510.4595590599991, &#39;median&#39;: 509.4302212499997, &#39;std&#39;: 3.3041499135301597}
 </pre></div>
 </div>
 </div>
@@ -998,7 +998,7 @@ 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> ( 11 minutes  8.794 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 10 minutes  38.131 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">
 <p><a class="reference download internal" download="" href="../_downloads/57a45d9bef1af358191e7d50043e652c/autotvm_relay_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">autotvm_relay_x86.py</span></code></a></p>
diff --git a/docs/tutorial/cross_compilation_and_rpc.html b/docs/tutorial/cross_compilation_and_rpc.html
index b55ecc1e6a..e4411eca8d 100644
--- a/docs/tutorial/cross_compilation_and_rpc.html
+++ b/docs/tutorial/cross_compilation_and_rpc.html
@@ -537,7 +537,7 @@ device and returns the measured cost. Network overhead is excluded.</p>
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;</span><span class="si">%g</span><span class="s2"> secs/op&quot;</span> <span class="o">%</span> <span class="n">cost</span><span class="p">)</span>
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>1.299e-07 secs/op
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>1.269e-07 secs/op
 </pre></div>
 </div>
 </div>
diff --git a/docs/tutorial/intro_topi.html b/docs/tutorial/intro_topi.html
index 4e8e23d0ee..de5213a9bb 100644
--- a/docs/tutorial/intro_topi.html
+++ b/docs/tutorial/intro_topi.html
@@ -497,7 +497,7 @@ we can schedule the following series of operations ending with <code class="code
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="nb">print</span><span class="p">(</span><a href="../reference/api/python/ir.html#tvm.ir.Array" title="tvm.ir.Array" class="sphx-glr-backref-module-tvm-ir sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">sg</span><span class="o">.</span><span class="n">stages</span></a><span class="p">)</span>
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>[stage(a, placeholder(a, 0x174fabe0)), stage(b, placeholder(b, 0xc477070)), stage(T_add, compute(T_add, body=[(a[ax0, ax1, ax2] + b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(min=0, ext=10))], reduce_axis=[], tag=broadcast, attrs={})), stage(T_multiply, compute(T_multiply, body=[(a[ax0, ax1, ax2]*b[ax1, ax2])], axis=[i [...]
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>[stage(a, placeholder(a, 0xc4ba520)), stage(b, placeholder(b, 0xec67740)), stage(T_add, compute(T_add, body=[(a[ax0, ax1, ax2] + b[ax1, ax2])], axis=[iter_var(ax0, range(min=0, ext=100)), iter_var(ax1, range(min=0, ext=10)), iter_var(ax2, range(min=0, ext=10))], reduce_axis=[], tag=broadcast, attrs={})), stage(T_multiply, compute(T_multiply, body=[(a[ax0, ax1, ax2]*b[ax1, ax2])], axis=[it [...]
 </pre></div>
 </div>
 <p>We can test the correctness by comparing with <code class="code docutils literal notranslate"><span class="pre">numpy</span></code> result as follows</p>
diff --git a/docs/tutorial/sg_execution_times.html b/docs/tutorial/sg_execution_times.html
index 91ae03ffff..d7568e3ee1 100644
--- a/docs/tutorial/sg_execution_times.html
+++ b/docs/tutorial/sg_execution_times.html
@@ -340,7 +340,7 @@
             
   <div class="section" id="computation-times">
 <span id="sphx-glr-tutorial-sg-execution-times"></span><h1>Computation times<a class="headerlink" href="#computation-times" title="Permalink to this headline">¶</a></h1>
-<p><strong>14:36.316</strong> total execution time for <strong>tutorial</strong> files:</p>
+<p><strong>14:21.829</strong> total execution time for <strong>tutorial</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 83%" />
@@ -349,35 +349,35 @@
 </colgroup>
 <tbody>
 <tr class="row-odd"><td><p><a class="reference internal" href="autotvm_relay_x86.html#sphx-glr-tutorial-autotvm-relay-x86-py"><span class="std std-ref">Compiling and Optimizing a Model with the Python Interface (AutoTVM)</span></a> (<code class="docutils literal notranslate"><span class="pre">autotvm_relay_x86.py</span></code>)</p></td>
-<td><p>11:08.794</p></td>
+<td><p>10:38.131</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="auto_scheduler_matmul_x86.html#sphx-glr-tutorial-auto-scheduler-matmul-x86-py"><span class="std std-ref">Optimizing Operators with Auto-scheduling</span></a> (<code class="docutils literal notranslate"><span class="pre">auto_scheduler_matmul_x86.py</span></code>)</p></td>
-<td><p>01:30.940</p></td>
+<td><p>01:42.599</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="tensor_expr_get_started.html#sphx-glr-tutorial-tensor-expr-get-started-py"><span class="std std-ref">Working with Operators Using Tensor Expression</span></a> (<code class="docutils literal notranslate"><span class="pre">tensor_expr_get_started.py</span></code>)</p></td>
-<td><p>01:00.696</p></td>
+<td><p>01:01.598</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="relay_quick_start.html#sphx-glr-tutorial-relay-quick-start-py"><span class="std std-ref">Quick Start Tutorial for Compiling Deep Learning Models</span></a> (<code class="docutils literal notranslate"><span class="pre">relay_quick_start.py</span></code>)</p></td>
-<td><p>00:36.288</p></td>
+<td><p>00:35.843</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="autotvm_matmul_x86.html#sphx-glr-tutorial-autotvm-matmul-x86-py"><span class="std std-ref">Optimizing Operators with Schedule Templates and AutoTVM</span></a> (<code class="docutils literal notranslate"><span class="pre">autotvm_matmul_x86.py</span></code>)</p></td>
-<td><p>00:18.007</p></td>
+<td><p>00:21.416</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
-<tr class="row-even"><td><p><a class="reference internal" href="intro_topi.html#sphx-glr-tutorial-intro-topi-py"><span class="std std-ref">Introduction to TOPI</span></a> (<code class="docutils literal notranslate"><span class="pre">intro_topi.py</span></code>)</p></td>
-<td><p>00:00.776</p></td>
+<tr class="row-even"><td><p><a class="reference internal" href="tensor_ir_blitz_course.html#sphx-glr-tutorial-tensor-ir-blitz-course-py"><span class="std std-ref">Blitz Course to TensorIR</span></a> (<code class="docutils literal notranslate"><span class="pre">tensor_ir_blitz_course.py</span></code>)</p></td>
+<td><p>00:01.293</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
-<tr class="row-odd"><td><p><a class="reference internal" href="tensor_ir_blitz_course.html#sphx-glr-tutorial-tensor-ir-blitz-course-py"><span class="std std-ref">Blitz Course to TensorIR</span></a> (<code class="docutils literal notranslate"><span class="pre">tensor_ir_blitz_course.py</span></code>)</p></td>
-<td><p>00:00.643</p></td>
+<tr class="row-odd"><td><p><a class="reference internal" href="intro_topi.html#sphx-glr-tutorial-intro-topi-py"><span class="std std-ref">Introduction to TOPI</span></a> (<code class="docutils literal notranslate"><span class="pre">intro_topi.py</span></code>)</p></td>
+<td><p>00:00.772</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="cross_compilation_and_rpc.html#sphx-glr-tutorial-cross-compilation-and-rpc-py"><span class="std std-ref">Cross Compilation and RPC</span></a> (<code class="docutils literal notranslate"><span class="pre">cross_compilation_and_rpc.py</span></code>)</p></td>
-<td><p>00:00.163</p></td>
+<td><p>00:00.169</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="introduction.html#sphx-glr-tutorial-introduction-py"><span class="std std-ref">Introduction</span></a> (<code class="docutils literal notranslate"><span class="pre">introduction.py</span></code>)</p></td>
@@ -385,18 +385,18 @@
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="uma.html#sphx-glr-tutorial-uma-py"><span class="std std-ref">Making your Hardware Accelerator TVM-ready with UMA</span></a> (<code class="docutils literal notranslate"><span class="pre">uma.py</span></code>)</p></td>
-<td><p>00:00.002</p></td>
+<td><p>00:00.001</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="install.html#sphx-glr-tutorial-install-py"><span class="std std-ref">Installing TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">install.py</span></code>)</p></td>
 <td><p>00:00.001</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
-<tr class="row-even"><td><p><a class="reference internal" href="tvmc_command_line_driver.html#sphx-glr-tutorial-tvmc-command-line-driver-py"><span class="std std-ref">Compiling and Optimizing a Model with TVMC</span></a> (<code class="docutils literal notranslate"><span class="pre">tvmc_command_line_driver.py</span></code>)</p></td>
+<tr class="row-even"><td><p><a class="reference internal" href="tvmc_python.html#sphx-glr-tutorial-tvmc-python-py"><span class="std std-ref">Getting Starting using TVMC Python: a high-level API for TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">tvmc_python.py</span></code>)</p></td>
 <td><p>00:00.001</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
-<tr class="row-odd"><td><p><a class="reference internal" href="tvmc_python.html#sphx-glr-tutorial-tvmc-python-py"><span class="std std-ref">Getting Starting using TVMC Python: a high-level API for TVM</span></a> (<code class="docutils literal notranslate"><span class="pre">tvmc_python.py</span></code>)</p></td>
+<tr class="row-odd"><td><p><a class="reference internal" href="tvmc_command_line_driver.html#sphx-glr-tutorial-tvmc-command-line-driver-py"><span class="std std-ref">Compiling and Optimizing a Model with TVMC</span></a> (<code class="docutils literal notranslate"><span class="pre">tvmc_command_line_driver.py</span></code>)</p></td>
 <td><p>00:00.001</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
diff --git a/docs/tutorial/tensor_expr_get_started.html b/docs/tutorial/tensor_expr_get_started.html
index bee9c5e8ec..b4d9d72352 100644
--- a/docs/tutorial/tensor_expr_get_started.html
+++ b/docs/tutorial/tensor_expr_get_started.html
@@ -673,10 +673,10 @@ factor to be the number of threads on your CPU.</p>
 </pre></div>
 </div>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Operator                  Timing             Performance
-   numpy    6.728879998263437e-06                    1.0
-   naive    6.689199999999999e-06      0.994103030775749
-parallel              7.1519e-06      1.0628663316697182
-  vector    2.4517499999999997e-05    3.6436227137840738
+   numpy    7.337909999023395e-06                    1.0
+   naive    6.675599999999999e-06     0.9097413297367312
+parallel    6.990699999999999e-06     0.9526827122341911
+  vector    2.4616599999999997e-05    3.3547154439446967
 </pre></div>
 </div>
 <div class="admonition-code-specialization admonition">
@@ -992,7 +992,7 @@ matrix multiplication.</p>
 <span class="n">answer</span> <span class="o">=</span> <span class="n">numpy</span><span class="o">.</span><span class="n">dot</span><span class="p">(</span><span class="n">a</span><span class="o">.</span><span class="n">numpy</span><span class="p">(),</span> <span class="n">b</span><span class="o">.</span><span class="n">numpy</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: 0.018565
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.018753
 </pre></div>
 </div>
 <p>Now we write a basic matrix multiplication using TVM TE and verify that it
@@ -1033,7 +1033,7 @@ optimizations.</p>
 <span class="n">evaluate_operation</span><span class="p">(</span><a href="../reference/api/python/te.html#tvm.te.Schedule" title="tvm.te.Schedule" class="sphx-glr-backref-module-tvm-te sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">s</span></a><span class="p">,</span> <span class="p">[</span><a href="../reference/api/python/te.html#tvm.te.Tensor" title="tvm.te.Tensor" class="sphx-glr-backref-module-tvm-te sphx-glr-backref-type-py-class sphx-glr-backref-instance [...]
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>none: 3.368203
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>none: 3.448494
 </pre></div>
 </div>
 <p>Let’s take a look at the intermediate representation of the operator and
@@ -1098,7 +1098,7 @@ schedule.</p>
 <span class="n">evaluate_operation</span><span class="p">(</span><a href="../reference/api/python/te.html#tvm.te.Schedule" title="tvm.te.Schedule" class="sphx-glr-backref-module-tvm-te sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">s</span></a><span class="p">,</span> <span class="p">[</span><a href="../reference/api/python/te.html#tvm.te.Tensor" title="tvm.te.Tensor" class="sphx-glr-backref-module-tvm-te sphx-glr-backref-type-py-class sphx-glr-backref-instance [...]
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>blocking: 0.310311
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>blocking: 0.301517
 </pre></div>
 </div>
 <p>By reordering the computation to take advantage of caching, you should see a
@@ -1157,7 +1157,7 @@ already cache friendly from our previous optimizations.</p>
 <span class="nb">print</span><span class="p">(</span><a href="../reference/api/python/driver.html#tvm.lower" title="tvm.lower" class="sphx-glr-backref-module-tvm sphx-glr-backref-type-py-function"><span class="n">tvm</span><span class="o">.</span><span class="n">lower</span></a><span class="p">(</span><a href="../reference/api/python/te.html#tvm.te.Schedule" title="tvm.te.Schedule" class="sphx-glr-backref-module-tvm-te sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class [...]
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>vectorization: 0.342435
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>vectorization: 0.345078
 @main = primfn(A_1: handle, B_1: handle, C_1: handle) -&gt; ()
   attr = {&quot;from_legacy_te_schedule&quot;: True, &quot;global_symbol&quot;: &quot;main&quot;, &quot;tir.noalias&quot;: True}
   buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1212,7 +1212,7 @@ more cache friendly.</p>
 <span class="nb">print</span><span class="p">(</span><a href="../reference/api/python/driver.html#tvm.lower" title="tvm.lower" class="sphx-glr-backref-module-tvm sphx-glr-backref-type-py-function"><span class="n">tvm</span><span class="o">.</span><span class="n">lower</span></a><span class="p">(</span><a href="../reference/api/python/te.html#tvm.te.Schedule" title="tvm.te.Schedule" class="sphx-glr-backref-module-tvm-te sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class [...]
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>loop permutation: 0.122163
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>loop permutation: 0.116288
 @main = primfn(A_1: handle, B_1: handle, C_1: handle) -&gt; ()
   attr = {&quot;from_legacy_te_schedule&quot;: True, &quot;global_symbol&quot;: &quot;main&quot;, &quot;tir.noalias&quot;: True}
   buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1288,7 +1288,7 @@ optimized schedule.</p>
 <span class="nb">print</span><span class="p">(</span><a href="../reference/api/python/driver.html#tvm.lower" title="tvm.lower" class="sphx-glr-backref-module-tvm sphx-glr-backref-type-py-function"><span class="n">tvm</span><span class="o">.</span><span class="n">lower</span></a><span class="p">(</span><a href="../reference/api/python/te.html#tvm.te.Schedule" title="tvm.te.Schedule" class="sphx-glr-backref-module-tvm-te sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class [...]
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>array packing: 0.109077
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>array packing: 0.107839
 @main = primfn(A_1: handle, B_1: handle, C_1: handle) -&gt; ()
   attr = {&quot;from_legacy_te_schedule&quot;: True, &quot;global_symbol&quot;: &quot;main&quot;, &quot;tir.noalias&quot;: True}
   buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1362,7 +1362,7 @@ to `C</cite> when all the block results are ready.</p>
 <span class="nb">print</span><span class="p">(</span><a href="../reference/api/python/driver.html#tvm.lower" title="tvm.lower" class="sphx-glr-backref-module-tvm sphx-glr-backref-type-py-function"><span class="n">tvm</span><span class="o">.</span><span class="n">lower</span></a><span class="p">(</span><a href="../reference/api/python/te.html#tvm.te.Schedule" title="tvm.te.Schedule" class="sphx-glr-backref-module-tvm-te sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class [...]
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>block caching: 0.111902
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>block caching: 0.110582
 @main = primfn(A_1: handle, B_1: handle, C_1: handle) -&gt; ()
   attr = {&quot;from_legacy_te_schedule&quot;: True, &quot;global_symbol&quot;: &quot;main&quot;, &quot;tir.noalias&quot;: True}
   buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1429,7 +1429,7 @@ of thread-level parallelization.</p>
 <span class="nb">print</span><span class="p">(</span><a href="../reference/api/python/driver.html#tvm.lower" title="tvm.lower" class="sphx-glr-backref-module-tvm sphx-glr-backref-type-py-function"><span class="n">tvm</span><span class="o">.</span><span class="n">lower</span></a><span class="p">(</span><a href="../reference/api/python/te.html#tvm.te.Schedule" title="tvm.te.Schedule" class="sphx-glr-backref-module-tvm-te sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class [...]
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>parallelization: 0.146945
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>parallelization: 0.146665
 @main = primfn(A_1: handle, B_1: handle, C_1: handle) -&gt; ()
   attr = {&quot;from_legacy_te_schedule&quot;: True, &quot;global_symbol&quot;: &quot;main&quot;, &quot;tir.noalias&quot;: True}
   buffers = {A: Buffer(A_2: Pointer(float32), float32, [1048576], []),
@@ -1491,13 +1491,13 @@ working, we can compare the results.</p>
 </pre></div>
 </div>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>        Operator                  Timing             Performance
-            none             3.368203356                     1.0
-        blocking            0.3103111295     0.09212957078355218
-   vectorization            0.3424345915     0.10166683994599048
-loop permutation     0.12216339170000001     0.03626960096764419
-   array packing            0.1090769235     0.03238430461916564
-   block caching     0.11190187370000002    0.033223015914600855
- parallelization     0.14694490659999998     0.04362708870835766
+            none            3.4484936041                     1.0
+        blocking     0.30151734750000003     0.08743450970635948
+   vectorization            0.3450780444     0.10006631416967923
+loop permutation     0.11628793479999999     0.03372137174960753
+   array packing            0.1078393406     0.03127143413338135
+   block caching     0.11058221309999998     0.03206681693378408
+ parallelization            0.1466654998     0.04253030935757739
 </pre></div>
 </div>
 <p>Note that the outputs on the web page reflect the running times on a
@@ -1529,7 +1529,7 @@ is</p>
 you can build generic templates of the matrix multiplication and other
 operations with tunable parameters that allows you to automatically optimize
 the computation for specific platforms.</p>
-<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  0.696 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  1.598 seconds)</p>
 <div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-tutorial-tensor-expr-get-started-py">
 <div class="sphx-glr-download sphx-glr-download-python docutils container">
 <p><a class="reference download internal" download="" href="../_downloads/40a01cffb015a67aaec0fad7e27cf80d/tensor_expr_get_started.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">tensor_expr_get_started.py</span></code></a></p>