[tvm-site] branch asf-site updated: deploying docs (apache/tvm@bd3d93b6a4f7883c89711b9feda44d205f1e3146)

[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 4671d4bd91 deploying docs (apache/tvm@bd3d93b6a4f7883c89711b9feda44d205f1e3146)
4671d4bd91 is described below

commit 4671d4bd91c832d63106291b85e8080378816075
Author: tvm-bot <95...@users.noreply.github.com>
AuthorDate: Fri Jan 13 01:07:34 2023 +0000

    deploying docs (apache/tvm@bd3d93b6a4f7883c89711b9feda44d205f1e3146)
---
 .../micro_tflite.py                                |   8 +-
 .../micro_tflite.ipynb                             |  22 +-
 .../micro_autotune.py                              |   6 +-
 .../micro_aot.ipynb                                |  22 +-
 .../micro_autotune.ipynb                           |  20 +-
 .../f8a7209a0e66b246185bfc41bbc82f54/micro_aot.py  |   8 +-
 docs/_images/sphx_glr_micro_train_001.png          | Bin 332942 -> 322550 bytes
 docs/_images/sphx_glr_micro_train_thumb.png        | Bin 23901 -> 23209 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_adreno.rst.txt   |   2 +-
 .../deploy_models/deploy_model_on_android.rst.txt  |   2 +-
 .../deploy_object_detection_pytorch.rst.txt        |   4 +-
 .../deploy_models/deploy_prequantized.rst.txt      |   6 +-
 .../deploy_prequantized_tflite.rst.txt             |   4 +-
 .../how_to/deploy_models/deploy_quantized.rst.txt  |   2 +-
 .../deploy_models/deploy_ssd_gluoncv.rst.txt       |   4 +-
 .../deploy_models/sg_execution_times.rst.txt       |  20 +-
 .../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                 | 951 ++++-----------------
 .../tune_network_cuda.rst.txt                      |   4 +-
 .../tune_network_x86.rst.txt                       |   4 +-
 .../tune_sparse_x86.rst.txt                        | 129 +--
 .../tune_with_autotvm/sg_execution_times.rst.txt   |   8 +-
 .../tune_with_autotvm/tune_conv2d_cuda.rst.txt     | 312 ++++---
 .../how_to/work_with_microtvm/micro_aot.rst.txt    |  32 +-
 .../work_with_microtvm/micro_autotune.rst.txt      |  55 +-
 .../work_with_microtvm/micro_pytorch.rst.txt       |   4 +-
 .../how_to/work_with_microtvm/micro_tflite.rst.txt |  36 +-
 .../how_to/work_with_microtvm/micro_train.rst.txt  |  18 +-
 .../work_with_microtvm/sg_execution_times.rst.txt  |  12 +-
 .../work_with_relay/sg_execution_times.rst.txt     |   8 +-
 .../how_to/work_with_schedules/intrin_math.rst.txt |   2 +-
 .../work_with_schedules/sg_execution_times.rst.txt |  10 +-
 .../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  |   6 +-
 .../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   |  59 +-
 .../tutorial/cross_compilation_and_rpc.rst.txt     |   2 +-
 docs/_sources/tutorial/intro_topi.rst.txt          |   2 +-
 docs/_sources/tutorial/sg_execution_times.rst.txt  |  18 +-
 .../tutorial/tensor_expr_get_started.rst.txt       |  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       |  16 +-
 docs/how_to/compile_models/from_pytorch.html       |  12 +-
 docs/how_to/compile_models/from_tensorflow.html    |   2 +-
 docs/how_to/compile_models/sg_execution_times.html |  26 +-
 .../deploy_models/deploy_model_on_adreno.html      |   2 +-
 .../deploy_models/deploy_model_on_android.html     |   2 +-
 .../deploy_object_detection_pytorch.html           |  52 +-
 docs/how_to/deploy_models/deploy_prequantized.html |  10 +-
 .../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  |  36 +-
 docs/how_to/deploy_models/sg_execution_times.html  |  20 +-
 .../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                    | 947 ++++----------------
 .../tune_with_autoscheduler/tune_network_cuda.html |   4 +-
 .../tune_with_autoscheduler/tune_network_x86.html  |   4 +-
 .../tune_with_autoscheduler/tune_sparse_x86.html   | 129 +--
 .../tune_with_autotvm/sg_execution_times.html      |   8 +-
 .../how_to/tune_with_autotvm/tune_conv2d_cuda.html | 312 ++++---
 docs/how_to/work_with_microtvm/micro_aot.html      |  20 +-
 docs/how_to/work_with_microtvm/micro_autotune.html |  35 +-
 docs/how_to/work_with_microtvm/micro_pytorch.html  |   6 +-
 docs/how_to/work_with_microtvm/micro_tflite.html   |  20 +-
 docs/how_to/work_with_microtvm/micro_train.html    |  16 +-
 .../work_with_microtvm/sg_execution_times.html     |  12 +-
 .../how_to/work_with_relay/sg_execution_times.html |   8 +-
 docs/how_to/work_with_schedules/intrin_math.html   |   2 +-
 .../work_with_schedules/sg_execution_times.html    |  10 +-
 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 |   6 +-
 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               | 268 +++---
 docs/tutorial/cross_compilation_and_rpc.html       |   2 +-
 docs/tutorial/intro_topi.html                      |   2 +-
 docs/tutorial/sg_execution_times.html              |  18 +-
 docs/tutorial/tensor_expr_get_started.html         |  40 +-
 139 files changed, 1593 insertions(+), 3044 deletions(-)

diff --git a/docs/_downloads/2fb9ae7bf124f72614a43137cf2919cb/micro_tflite.py b/docs/_downloads/2fb9ae7bf124f72614a43137cf2919cb/micro_tflite.py
index 86e5d6b4b1..0770d472c9 100644
--- a/docs/_downloads/2fb9ae7bf124f72614a43137cf2919cb/micro_tflite.py
+++ b/docs/_downloads/2fb9ae7bf124f72614a43137cf2919cb/micro_tflite.py
@@ -36,7 +36,7 @@ import os
 # By default, this tutorial runs on x86 CPU using TVM's C runtime. If you would like
 # to run on real Zephyr hardware, you must export the `TVM_MICRO_USE_HW` environment
 # variable. Otherwise (if you are using the C runtime), you can skip installing
-# Zephyr and CMSIS-NN. It takes ~20 minutes to install both of them.
+# Zephyr. It takes ~20 minutes to install Zephyr.
 use_physical_hw = bool(os.getenv("TVM_MICRO_USE_HW"))
 
 ######################################################################
@@ -44,11 +44,6 @@ use_physical_hw = bool(os.getenv("TVM_MICRO_USE_HW"))
 #     .. include:: ../../../../gallery/how_to/work_with_microtvm/install_zephyr.rst
 #
 
-######################################################################
-#
-#     .. include:: ../../../../gallery/how_to/work_with_microtvm/install_cmsis.rst
-#
-
 ######################################################################
 # Import Python dependencies
 # -------------------------------
@@ -219,7 +214,6 @@ if use_physical_hw:
         "board": BOARD,
         "serial_number": SERIAL,
         "config_main_stack_size": 4096,
-        "cmsis_path": os.getenv("CMSIS_PATH", default="/content/cmsis"),
         "zephyr_base": os.getenv("ZEPHYR_BASE", default="/content/zephyrproject/zephyr"),
     }
 
diff --git a/docs/_downloads/5b279d8a8718816263fa65b0eef1a5c0/micro_tflite.ipynb b/docs/_downloads/5b279d8a8718816263fa65b0eef1a5c0/micro_tflite.ipynb
index e99e9f04b6..a2233b9b63 100644
--- a/docs/_downloads/5b279d8a8718816263fa65b0eef1a5c0/micro_tflite.ipynb
+++ b/docs/_downloads/5b279d8a8718816263fa65b0eef1a5c0/micro_tflite.ipynb
@@ -44,7 +44,7 @@
       },
       "outputs": [],
       "source": [
-        "import os\n\n# By default, this tutorial runs on x86 CPU using TVM's C runtime. If you would like\n# to run on real Zephyr hardware, you must export the `TVM_MICRO_USE_HW` environment\n# variable. Otherwise (if you are using the C runtime), you can skip installing\n# Zephyr and CMSIS-NN. It takes ~20 minutes to install both of them.\nuse_physical_hw = bool(os.getenv(\"TVM_MICRO_USE_HW\"))"
+        "import os\n\n# By default, this tutorial runs on x86 CPU using TVM's C runtime. If you would like\n# to run on real Zephyr hardware, you must export the `TVM_MICRO_USE_HW` environment\n# variable. Otherwise (if you are using the C runtime), you can skip installing\n# Zephyr. It takes ~20 minutes to install Zephyr.\nuse_physical_hw = bool(os.getenv(\"TVM_MICRO_USE_HW\"))"
       ]
     },
     {
@@ -65,24 +65,6 @@
         "%%shell\n# Install west and ninja\npython3 -m pip install west\napt-get install -y ninja-build\n\n# Install ZephyrProject\nZEPHYR_PROJECT_PATH=\"/content/zephyrproject\"\nexport ZEPHYR_BASE=${ZEPHYR_PROJECT_PATH}/zephyr\nwest init ${ZEPHYR_PROJECT_PATH}\ncd ${ZEPHYR_BASE}\ngit checkout v2.7-branch\ncd ..\nwest update\nwest zephyr-export\nchmod -R o+w ${ZEPHYR_PROJECT_PATH}\n\n# Install Zephyr SDK\nZEPHYR_SDK_VERSION=0.13.2\nZEPHYR_SDK_FILE=\"/content/zephyr-sdk-linux-setup.run\" [...]
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {},
-      "source": [
-        "## Install CMSIS-NN\n"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "collapsed": false
-      },
-      "outputs": [],
-      "source": [
-        "%%shell\nCMSIS_SHA=\"51263182d16c92649a48144ba56c0945f9fce60e\"\nCMSIS_URL=\"http://github.com/ARM-software/CMSIS_5/archive/${CMSIS_SHA}.tar.gz\"\nexport CMSIS_PATH=/content/cmsis\nDOWNLOAD_PATH=\"/content/${CMSIS_SHA}.tar.gz\"\nmkdir ${CMSIS_PATH}\nwget ${CMSIS_URL} -O \"${DOWNLOAD_PATH}\"\ntar -xf \"${DOWNLOAD_PATH}\" -C ${CMSIS_PATH} --strip-components=1\nrm ${DOWNLOAD_PATH}"
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {},
@@ -188,7 +170,7 @@
       },
       "outputs": [],
       "source": [
-        "with tvm.transform.PassContext(\n    opt_level=3, config={\"tir.disable_vectorize\": True}, disabled_pass=[\"AlterOpLayout\"]\n):\n    module = relay.build(mod, target=TARGET, runtime=RUNTIME, params=params)\n\n\n# Inspecting the compilation output\n# ---------------------------------\n#\n# The compilation process has produced some C code implementing the operators in this graph. We\n# can inspect it by printing the CSourceModule contents (for the purposes of this tutorial, let' [...]
+        "with tvm.transform.PassContext(\n    opt_level=3, config={\"tir.disable_vectorize\": True}, disabled_pass=[\"AlterOpLayout\"]\n):\n    module = relay.build(mod, target=TARGET, runtime=RUNTIME, params=params)\n\n\n# Inspecting the compilation output\n# ---------------------------------\n#\n# The compilation process has produced some C code implementing the operators in this graph. We\n# can inspect it by printing the CSourceModule contents (for the purposes of this tutorial, let' [...]
       ]
     },
     {
diff --git a/docs/_downloads/9ccca8fd489a1486ac71b55a55c320c5/micro_autotune.py b/docs/_downloads/9ccca8fd489a1486ac71b55a55c320c5/micro_autotune.py
index 9edb9ae75e..9be257a57a 100644
--- a/docs/_downloads/9ccca8fd489a1486ac71b55a55c320c5/micro_autotune.py
+++ b/docs/_downloads/9ccca8fd489a1486ac71b55a55c320c5/micro_autotune.py
@@ -33,7 +33,7 @@ This tutorial explains how to autotune a model using the C runtime.
 #
 
 
-# You can skip the following two sections (installing Zephyr and CMSIS-NN) if the following flag is False.
+# You can skip the following section (installing Zephyr) if the following flag is False.
 # Installing Zephyr takes ~20 min.
 import os
 
@@ -44,10 +44,6 @@ use_physical_hw = bool(os.getenv("TVM_MICRO_USE_HW"))
 #     .. include:: ../../../../gallery/how_to/work_with_microtvm/install_zephyr.rst
 #
 
-######################################################################
-#
-#     .. include:: ../../../../gallery/how_to/work_with_microtvm/install_cmsis.rst
-#
 
 ######################################################################
 # Import Python dependencies
diff --git a/docs/_downloads/c00933f3fbcf90c4f584d54607b33805/micro_aot.ipynb b/docs/_downloads/c00933f3fbcf90c4f584d54607b33805/micro_aot.ipynb
index 52125163f9..1a0e3f2787 100644
--- a/docs/_downloads/c00933f3fbcf90c4f584d54607b33805/micro_aot.ipynb
+++ b/docs/_downloads/c00933f3fbcf90c4f584d54607b33805/micro_aot.ipynb
@@ -44,7 +44,7 @@
       },
       "outputs": [],
       "source": [
-        "import os\n\n# By default, this tutorial runs on x86 CPU using TVM's C runtime. If you would like\n# to run on real Zephyr hardware, you must export the `TVM_MICRO_USE_HW` environment\n# variable. Otherwise (if you are using the C runtime), you can skip installing\n# Zephyr and CMSIS-NN. It takes ~20 minutes to install both of them.\nuse_physical_hw = bool(os.getenv(\"TVM_MICRO_USE_HW\"))"
+        "import os\n\n# By default, this tutorial runs on x86 CPU using TVM's C runtime. If you would like\n# to run on real Zephyr hardware, you must export the `TVM_MICRO_USE_HW` environment\n# variable. Otherwise (if you are using the C runtime), you can skip installing\n# Zephyr. It takes ~20 minutes to install Zephyr.\nuse_physical_hw = bool(os.getenv(\"TVM_MICRO_USE_HW\"))"
       ]
     },
     {
@@ -65,24 +65,6 @@
         "%%shell\n# Install west and ninja\npython3 -m pip install west\napt-get install -y ninja-build\n\n# Install ZephyrProject\nZEPHYR_PROJECT_PATH=\"/content/zephyrproject\"\nexport ZEPHYR_BASE=${ZEPHYR_PROJECT_PATH}/zephyr\nwest init ${ZEPHYR_PROJECT_PATH}\ncd ${ZEPHYR_BASE}\ngit checkout v2.7-branch\ncd ..\nwest update\nwest zephyr-export\nchmod -R o+w ${ZEPHYR_PROJECT_PATH}\n\n# Install Zephyr SDK\nZEPHYR_SDK_VERSION=0.13.2\nZEPHYR_SDK_FILE=\"/content/zephyr-sdk-linux-setup.run\" [...]
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {},
-      "source": [
-        "## Install CMSIS-NN\n"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "collapsed": false
-      },
-      "outputs": [],
-      "source": [
-        "%%shell\nCMSIS_SHA=\"51263182d16c92649a48144ba56c0945f9fce60e\"\nCMSIS_URL=\"http://github.com/ARM-software/CMSIS_5/archive/${CMSIS_SHA}.tar.gz\"\nexport CMSIS_PATH=/content/cmsis\nDOWNLOAD_PATH=\"/content/${CMSIS_SHA}.tar.gz\"\nmkdir ${CMSIS_PATH}\nwget ${CMSIS_URL} -O \"${DOWNLOAD_PATH}\"\ntar -xf \"${DOWNLOAD_PATH}\" -C ${CMSIS_PATH} --strip-components=1\nrm ${DOWNLOAD_PATH}"
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {},
@@ -170,7 +152,7 @@
       },
       "outputs": [],
       "source": [
-        "template_project_path = pathlib.Path(tvm.micro.get_microtvm_template_projects(\"crt\"))\nproject_options = {}  # You can use options to provide platform-specific options through TVM.\n\nif use_physical_hw:\n    template_project_path = pathlib.Path(tvm.micro.get_microtvm_template_projects(\"zephyr\"))\n    project_options = {\n        \"project_type\": \"host_driven\",\n        \"board\": BOARD,\n        \"serial_number\": SERIAL,\n        \"config_main_stack_size\": 4096,\n      [...]
+        "template_project_path = pathlib.Path(tvm.micro.get_microtvm_template_projects(\"crt\"))\nproject_options = {}  # You can use options to provide platform-specific options through TVM.\n\nif use_physical_hw:\n    template_project_path = pathlib.Path(tvm.micro.get_microtvm_template_projects(\"zephyr\"))\n    project_options = {\n        \"project_type\": \"host_driven\",\n        \"board\": BOARD,\n        \"serial_number\": SERIAL,\n        \"config_main_stack_size\": 4096,\n      [...]
       ]
     },
     {
diff --git a/docs/_downloads/f83ba3df2d52f9b54cf141114359481a/micro_autotune.ipynb b/docs/_downloads/f83ba3df2d52f9b54cf141114359481a/micro_autotune.ipynb
index 272e54ae35..d7bea7ffba 100644
--- a/docs/_downloads/f83ba3df2d52f9b54cf141114359481a/micro_autotune.ipynb
+++ b/docs/_downloads/f83ba3df2d52f9b54cf141114359481a/micro_autotune.ipynb
@@ -44,7 +44,7 @@
       },
       "outputs": [],
       "source": [
-        "# You can skip the following two sections (installing Zephyr and CMSIS-NN) if the following flag is False.\n# Installing Zephyr takes ~20 min.\nimport os\n\nuse_physical_hw = bool(os.getenv(\"TVM_MICRO_USE_HW\"))"
+        "# You can skip the following section (installing Zephyr) if the following flag is False.\n# Installing Zephyr takes ~20 min.\nimport os\n\nuse_physical_hw = bool(os.getenv(\"TVM_MICRO_USE_HW\"))"
       ]
     },
     {
@@ -65,24 +65,6 @@
         "%%shell\n# Install west and ninja\npython3 -m pip install west\napt-get install -y ninja-build\n\n# Install ZephyrProject\nZEPHYR_PROJECT_PATH=\"/content/zephyrproject\"\nexport ZEPHYR_BASE=${ZEPHYR_PROJECT_PATH}/zephyr\nwest init ${ZEPHYR_PROJECT_PATH}\ncd ${ZEPHYR_BASE}\ngit checkout v2.7-branch\ncd ..\nwest update\nwest zephyr-export\nchmod -R o+w ${ZEPHYR_PROJECT_PATH}\n\n# Install Zephyr SDK\nZEPHYR_SDK_VERSION=0.13.2\nZEPHYR_SDK_FILE=\"/content/zephyr-sdk-linux-setup.run\" [...]
       ]
     },
-    {
-      "cell_type": "markdown",
-      "metadata": {},
-      "source": [
-        "## Install CMSIS-NN\n"
-      ]
-    },
-    {
-      "cell_type": "code",
-      "execution_count": null,
-      "metadata": {
-        "collapsed": false
-      },
-      "outputs": [],
-      "source": [
-        "%%shell\nCMSIS_SHA=\"51263182d16c92649a48144ba56c0945f9fce60e\"\nCMSIS_URL=\"http://github.com/ARM-software/CMSIS_5/archive/${CMSIS_SHA}.tar.gz\"\nexport CMSIS_PATH=/content/cmsis\nDOWNLOAD_PATH=\"/content/${CMSIS_SHA}.tar.gz\"\nmkdir ${CMSIS_PATH}\nwget ${CMSIS_URL} -O \"${DOWNLOAD_PATH}\"\ntar -xf \"${DOWNLOAD_PATH}\" -C ${CMSIS_PATH} --strip-components=1\nrm ${DOWNLOAD_PATH}"
-      ]
-    },
     {
       "cell_type": "markdown",
       "metadata": {},
diff --git a/docs/_downloads/f8a7209a0e66b246185bfc41bbc82f54/micro_aot.py b/docs/_downloads/f8a7209a0e66b246185bfc41bbc82f54/micro_aot.py
index 81109b2965..c1b29ba5c5 100644
--- a/docs/_downloads/f8a7209a0e66b246185bfc41bbc82f54/micro_aot.py
+++ b/docs/_downloads/f8a7209a0e66b246185bfc41bbc82f54/micro_aot.py
@@ -41,7 +41,7 @@ import os
 # By default, this tutorial runs on x86 CPU using TVM's C runtime. If you would like
 # to run on real Zephyr hardware, you must export the `TVM_MICRO_USE_HW` environment
 # variable. Otherwise (if you are using the C runtime), you can skip installing
-# Zephyr and CMSIS-NN. It takes ~20 minutes to install both of them.
+# Zephyr. It takes ~20 minutes to install Zephyr.
 use_physical_hw = bool(os.getenv("TVM_MICRO_USE_HW"))
 
 ######################################################################
@@ -49,11 +49,6 @@ use_physical_hw = bool(os.getenv("TVM_MICRO_USE_HW"))
 #     .. include:: ../../../../gallery/how_to/work_with_microtvm/install_zephyr.rst
 #
 
-######################################################################
-#
-#     .. include:: ../../../../gallery/how_to/work_with_microtvm/install_cmsis.rst
-#
-
 ######################################################################
 # Import Python dependencies
 # -------------------------------
@@ -159,7 +154,6 @@ if use_physical_hw:
         "board": BOARD,
         "serial_number": SERIAL,
         "config_main_stack_size": 4096,
-        "cmsis_path": os.getenv("CMSIS_PATH", default="/content/cmsis"),
         "zephyr_base": os.getenv("ZEPHYR_BASE", default="/content/zephyrproject/zephyr"),
     }
 
diff --git a/docs/_images/sphx_glr_micro_train_001.png b/docs/_images/sphx_glr_micro_train_001.png
index 52fc7b9506..b961179557 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 7c8ae2aa15..ab1b531c46 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 84b7fae6b9..10c26b77cb 100644
--- a/docs/_sources/how_to/compile_models/from_darknet.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_darknet.rst.txt
@@ -318,7 +318,7 @@ The process is no different from other examples.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  10.285 seconds)
+   **Total running time of the script:** ( 1 minutes  11.325 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 6dd9420e81..081816dd0d 100644
--- a/docs/_sources/how_to/compile_models/from_keras.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_keras.rst.txt
@@ -232,7 +232,7 @@ 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 941ms/step
+
    1/1 [==============================] - ETA: 0s
    1/1 [==============================] - 1s 923ms/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 1b0d3b1b35..72ad224bea 100644
--- a/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_mxnet.rst.txt
@@ -116,7 +116,7 @@ In this section, we download a pretrained imagenet model and classify an image.
 
  .. code-block:: none
 
-    Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip9e6df392-f212-4196-b8bc-203b5e5d4f02 from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/resnet18_v1-a0666292.zip...
+    Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zipa0e9fe50-cc19-4c5d-a56c-bdaa5f904f8a 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 8be0712337..9cf14b7992 100644
--- a/docs/_sources/how_to/compile_models/from_oneflow.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_oneflow.rst.txt
@@ -121,7 +121,7 @@ Load a pretrained OneFlow model and save model
  .. code-block:: none
 
     Downloading: "https://oneflow-public.oss-cn-beijing.aliyuncs.com/model_zoo/flowvision/classification/ResNet/resnet18.zip" to /workspace/.oneflow/flowvision_cache/resnet18.zip
-
      0%|          | 0.00/41.5M [00:00<?, ?B/s]
     15%|#5        | 6.33M/41.5M [00:00<00:00, 65.9MB/s]
     30%|###       | 12.6M/41.5M [00:00<00:00, 48.3MB/s]
     42%|####2     | 17.5M/41.5M [00:00<00:00, 38.2MB/s]
     58%|#####7    | 24.0M/41.5M [00:00<00:00, 44.2MB/s]
     77%|#######7  | 32.0M/41.5M [00:00<00:00, 51.3MB/s]
     92%|#########2| 38.3M/41.5M [00:00<00:00, 37.2MB/s]
    100%|##########| 41.5M/41.5M [00:01<00:00, 42.5MB/s]
+
      0%|          | 0.00/41.5M [00:00<?, ?B/s]
     15%|#5        | 6.33M/41.5M [00:00<00:00, 53.7MB/s]
     28%|##7       | 11.5M/41.5M [00:00<00:00, 44.4MB/s]
     38%|###7      | 15.7M/41.5M [00:00<00:00, 35.3MB/s]
     46%|####6     | 19.2M/41.5M [00:00<00:00, 32.7MB/s]
     54%|#####3    | 22.4M/41.5M [00:00<00:00, 32.5MB/s]
     62%|######1   | 25.5M/41.5M [00:00<00:00, 30.2MB/s]
     80%|########  | 33.4M/41.5M [00:00<00:00, 44.6MB/s]
     96%|#########6| 40.0M/41.5M [00:00<00:00, 51.3MB/s]
    100%|##########| 41.5M/41.5M [00:01<00:00, 42.9MB/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 23cc5026a3..e154e50f8b 100644
--- a/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_pytorch.rst.txt
@@ -101,7 +101,7 @@ Load a pretrained PyTorch model
     /venv/apache-tvm-py3.7/lib/python3.7/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and will be removed in 0.15. The current behavior is equivalent to passing `weights=ResNet18_Weights.IMAGENET1K_V1`. You can also use `weights=ResNet18_Weights.DEFAULT` to get the most up-to-date weights.
       warnings.warn(msg)
     Downloading: "https://download.pytorch.org/models/resnet18-f37072fd.pth" to /workspace/.cache/torch/hub/checkpoints/resnet18-f37072fd.pth
-
      0%|          | 0.00/44.7M [00:00<?, ?B/s]
     18%|#7        | 7.99M/44.7M [00:00<00:00, 60.6MB/s]
     38%|###7      | 16.9M/44.7M [00:00<00:00, 77.1MB/s]
     63%|######2   | 28.0M/44.7M [00:00<00:00, 94.0MB/s]
     90%|########9 | 40.0M/44.7M [00:00<00:00, 98.6MB/s]
    100%|##########| 44.7M/44.7M [00:00<00:00, 96.5MB/s]
+
      0%|          | 0.00/44.7M [00:00<?, ?B/s]
     14%|#4        | 6.30M/44.7M [00:00<00:00, 45.1MB/s]
     24%|##3       | 10.6M/44.7M [00:00<00:00, 38.6MB/s]
     36%|###5      | 16.0M/44.7M [00:00<00:00, 35.1MB/s]
     65%|######5   | 29.0M/44.7M [00:00<00:00, 64.9MB/s]
     81%|########1 | 36.2M/44.7M [00:00<00:00, 48.3MB/s]
     94%|#########3| 41.9M/44.7M [00:00<00:00, 49.7MB/s]
    100%|##########| 44.7M/44.7M [00:00<00:00, 50.9MB/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 fd1cf3ae02..b05a3002ac 100644
--- a/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
+++ b/docs/_sources/how_to/compile_models/from_tensorflow.rst.txt
@@ -424,7 +424,7 @@ Run the corresponding model on tensorflow
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  11.018 seconds)
+   **Total running time of the script:** ( 1 minutes  10.912 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 98dd669bc9..c6b22e182b 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
 =================
-**05:37.857** total execution time for **how_to_compile_models** files:
+**05:39.585** total execution time for **how_to_compile_models** files:
 
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``) | 01:11.018 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``)       | 01:11.325 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_darknet.py` (``from_darknet.py``)       | 01:10.285 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_tensorflow.py` (``from_tensorflow.py``) | 01:10.912 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``)         | 00:45.279 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_paddle.py` (``from_paddle.py``)         | 00:45.558 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_oneflow.py` (``from_oneflow.py``)       | 00:31.605 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_oneflow.py` (``from_oneflow.py``)       | 00:31.415 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``)           | 00:27.623 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_mxnet.py` (``from_mxnet.py``)           | 00:27.592 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``)         | 00:26.413 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_coreml.py` (``from_coreml.py``)         | 00:26.393 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``)         | 00:24.887 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_tflite.py` (``from_tflite.py``)         | 00:25.287 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``)       | 00:21.849 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_pytorch.py` (``from_pytorch.py``)       | 00:22.227 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``)           | 00:16.456 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_keras.py` (``from_keras.py``)           | 00:16.443 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``)             | 00:02.442 | 0.0 MB |
+| :ref:`sphx_glr_how_to_compile_models_from_onnx.py` (``from_onnx.py``)             | 00:02.434 | 0.0 MB |
 +-----------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/deploy_models/deploy_model_on_adreno.rst.txt b/docs/_sources/how_to/deploy_models/deploy_model_on_adreno.rst.txt
index ea218ec65e..d3dc350f9e 100644
--- a/docs/_sources/how_to/deploy_models/deploy_model_on_adreno.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_model_on_adreno.rst.txt
@@ -727,7 +727,7 @@ well as provides information about the model's performance
     Evaluate inference time cost...
     Execution time summary:
      mean (ms)   median (ms)    max (ms)     min (ms)     std (ms)  
-     2541.6013    2541.3973    2545.4494    2538.4311      2.0426   
+     2542.3067    2540.6301    2552.0888    2538.5837      3.9541   
                
 
 
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 153afefed6..83f405d458 100644
--- a/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_model_on_android.rst.txt
@@ -437,7 +437,7 @@ Execute on TVM
     Evaluate inference time cost...
     Execution time summary:
      mean (ms)   median (ms)    max (ms)     min (ms)     std (ms)  
-      16.1475      15.7952      17.0500      15.6662       0.5307   
+      15.7059      15.6497      16.0923      15.5916       0.1471   
                
 
 
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 1f1cfecfbd..945b4d753e 100644
--- a/docs/_sources/how_to/deploy_models/deploy_object_detection_pytorch.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_object_detection_pytorch.rst.txt
@@ -130,7 +130,7 @@ Load pre-trained maskrcnn from torchvision and do tracing
     /venv/apache-tvm-py3.7/lib/python3.7/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and will be removed in 0.15. The current behavior is equivalent to passing `weights=MaskRCNN_ResNet50_FPN_Weights.COCO_V1`. You can also use `weights=MaskRCNN_ResNet50_FPN_Weights.DEFAULT` to get the most up-to-date weights.
       warnings.warn(msg)
     Downloading: "https://download.pytorch.org/models/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth" to /workspace/.cache/torch/hub/checkpoints/maskrcnn_resnet50_fpn_coco-bf2d0c1e.pth
-
      0%|          | 0.00/170M [00:00<?, ?B/s]
      5%|4         | 7.99M/170M [00:00<00:02, 81.6MB/s]
      9%|9         | 16.0M/170M [00:00<00:02, 64.4MB/s]
     13%|#3        | 22.4M/170M [00:00<00:02, 56.5MB/s]
     16%|#6        | 27.9M/170M [00:00<00:02, 52.2MB/s]
     19%|#9        | 32.9M/170M [00:00<00:03, 37.7MB/s]
     24%|##3       | 40.0M/170M [00:00<00:03, 45.3MB/s]
     28%|##8       | 48.0M/170M [00:00<00:02, 52.2MB/s]
     35%|###5      | 59.9M/170M [00:01<00:01, 70.5MB/s]
     40%|###9      | 67.4M/170M [00:01<00:01, 55.9MB/s]
     45%|####4     | 76.4M/170M [00:01<00:01, 64.6MB/s]
     49%|####9     | 83.5M/170M [00:01<00:01, 54.9MB/s]
     53%|#####2    | 89.6M/170M [00:01<00:02, 38.4MB/s]
     56%|#####5    | 94.9M/170M [00:01<00:01, 41.4MB/s]
     60%|######    | 102M/170M [00:02<00:01, 44.7MB/s] 
     66%|######5   | 112M/170M [00:02<00:01, 49.5MB/s]
     71%|#######   | 120M/170M [00:02<00:01, 47.0MB/s]
     79%|#######9  | 134M/170M [00:02<00:00, 67.6MB/s]
      84%|########3 | 142M/170M [00:02<00:00, 68.0MB/s]
     88%|########8 | 150M/170M [00:02<00:00, 63.7MB/s]
     92%|#########2| 157M/170M [00:03<00:00, 61.3MB/s]
     96%|#########6| 163M/170M [00:03<00:00, 56.9MB/s]
    100%|#########9| 169M/170M [00:03<00:00, 51.2MB/s]
    100%|##########| 170M/170M [00:03<00:00, 53.7MB/s]
+
      0%|          | 0.00/170M [00:00<?, ?B/s]
      4%|3         | 6.30M/170M [00:00<00:03, 51.8MB/s]
      7%|6         | 11.2M/170M [00:00<00:03, 44.9MB/s]
      9%|9         | 16.0M/170M [00:00<00:04, 36.3MB/s]
     14%|#4        | 24.2M/170M [00:00<00:02, 51.5MB/s]
     18%|#7        | 30.3M/170M [00:00<00:03, 45.0MB/s]
     21%|##        | 35.0M/170M [00:00<00:03, 44.7MB/s]
     24%|##3       | 40.0M/170M [00:00<00:02, 46.0MB/s]
     28%|##8       | 48.0M/170M [00:01<00:02, 46.2MB/s]
     33%|###2      | 56.0M/170M [00:01<00:02, 53.0MB/s]
     37%|###6      | 62.4M/170M [00:01<00:01, 56.5MB/s]
     40%|####      | 68.0M/170M [00:01<00:02, 51.3MB/s]
     44%|####3     | 74.3M/170M [00:01<00:01, 55.0MB/s]
     47%|####7     | 80.0M/170M [00:01<00:01, 56.1MB/s]
     52%|#####1    | 88.0M/170M [00:01<00:01, 55.3MB/s]
     57%|#####6    | 96.0M/170M [00:02<00:01, 51.0MB/s]
     60%|######    | 102M/170M [00:02<00:01, 53.3MB/s] 
     63%|######3   | 108M/170M [00:02<00:01, 52.2MB/s
 ]
     66%|######6   | 113M/170M [00:02<00:01, 51.6MB/s]
     71%|#######   | 120M/170M [00:02<00:00, 53.8MB/s]
     74%|#######4  | 126M/170M [00:02<00:01, 39.9MB/s]
     77%|#######6  | 131M/170M [00:02<00:01, 40.6MB/s]
     80%|########  | 136M/170M [00:03<00:01, 33.1MB/s]
     85%|########4 | 144M/170M [00:03<00:00, 37.3MB/s]
     89%|########9 | 152M/170M [00:03<00:00, 45.4MB/s]
     92%|#########2| 157M/170M [00:03<00:00, 43.6MB/s]
     95%|#########4| 161M/170M [00:03<00:00, 45.0MB/s]
    100%|##########| 170M/170M [00:03<00:00, 48.4MB/s]
     /venv/apache-tvm-py3.7/lib/python3.7/site-packages/torch/nn/functional.py:3897: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
       for i in range(dim)
     /venv/apache-tvm-py3.7/lib/python3.7/site-packages/torchvision/models/detection/anchor_utils.py:124: UserWarning: __floordiv__ is deprecated, and its behavior will change in a future version of pytorch. It currently rounds toward 0 (like the 'trunc' function NOT 'floor'). This results in incorrect rounding for negative values. To keep the current behavior, use torch.div(a, b, rounding_mode='trunc'), or for actual floor division, use torch.div(a, b, rounding_mode='floor').
@@ -299,7 +299,7 @@ Get boxes with score larger than 0.9
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 3 minutes  13.151 seconds)
+   **Total running time of the script:** ( 3 minutes  16.484 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 24ba7a89b1..8fa5b93cac 100644
--- a/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_prequantized.rst.txt
@@ -227,7 +227,7 @@ training. Other models require a full post training calibration.
     /venv/apache-tvm-py3.7/lib/python3.7/site-packages/torchvision/models/_utils.py:223: UserWarning: Arguments other than a weight enum or `None` for 'weights' are deprecated since 0.13 and will be removed in 0.15. The current behavior is equivalent to passing `weights=MobileNet_V2_Weights.IMAGENET1K_V1`. You can also use `weights=MobileNet_V2_Weights.DEFAULT` to get the most up-to-date weights.
       warnings.warn(msg)
     Downloading: "https://download.pytorch.org/models/mobilenet_v2-b0353104.pth" to /workspace/.cache/torch/hub/checkpoints/mobilenet_v2-b0353104.pth
-
      0%|          | 0.00/13.6M [00:00<?, ?B/s]
     59%|#####8    | 7.99M/13.6M [00:00<00:00, 47.8MB/s]
     93%|#########2| 12.6M/13.6M [00:00<00:00, 33.3MB/s]
    100%|##########| 13.6M/13.6M [00:00<00:00, 37.6MB/s]
+
      0%|          | 0.00/13.6M [00:00<?, ?B/s]
     59%|#####8    | 7.99M/13.6M [00:00<00:00, 50.5MB/s]
     95%|#########4| 12.8M/13.6M [00:00<00:00, 43.5MB/s]
    100%|##########| 13.6M/13.6M [00:00<00:00, 46.8MB/s]
 
 
 
@@ -409,7 +409,7 @@ Here we give an example of how to measure performance of TVM compiled models.
 
     Execution time summary:
      mean (ms)   median (ms)    max (ms)     min (ms)     std (ms)  
-      90.4012      90.2446      96.5908      90.1191       0.7391   
+      90.4182      90.2649      95.7405      90.1323       0.6007   
                
 
 
@@ -458,7 +458,7 @@ TODO
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  5.979 seconds)
+   **Total running time of the script:** ( 1 minutes  6.526 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 497f6df9f9..6b08cdcb11 100644
--- a/docs/_sources/how_to/deploy_models/deploy_prequantized_tflite.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_prequantized_tflite.rst.txt
@@ -423,7 +423,7 @@ Here we give an example of how to measure performance of TVM compiled models.
 
     Execution time summary:
      mean (ms)   median (ms)    max (ms)     min (ms)     std (ms)  
-      119.8170     119.7858     125.6149     119.0645      0.6544   
+      121.2631     121.1364     125.7362     120.5266      0.6134   
                
 
 
@@ -460,7 +460,7 @@ Here we give an example of how to measure performance of TVM compiled models.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 2 minutes  23.273 seconds)
+   **Total running time of the script:** ( 2 minutes  23.497 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 358953f0d9..9df49f9d89 100644
--- a/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_quantized.rst.txt
@@ -257,7 +257,7 @@ We create a Relay VM to build and execute the model.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  32.603 seconds)
+   **Total running time of the script:** ( 1 minutes  37.643 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 28678d7b16..7fb49e8e1b 100644
--- a/docs/_sources/how_to/deploy_models/deploy_ssd_gluoncv.rst.txt
+++ b/docs/_sources/how_to/deploy_models/deploy_ssd_gluoncv.rst.txt
@@ -170,7 +170,7 @@ Convert and compile model for CPU.
             data: None
       input_sym_arg_type = in_param.infer_type()[0]
     Downloading /workspace/.mxnet/models/ssd_512_resnet50_v1_voc-9c8b225a.zip from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/ssd_512_resnet50_v1_voc-9c8b225a.zip...
-
      0%|          | 0/132723 [00:00<?, ?KB/s]
      4%|4         | 5381/132723 [00:00<00:02, 53803.00KB/s]
     10%|#         | 13873/132723 [00:00<00:01, 72100.72KB/s]
     16%|#5        | 21084/132723 [00:00<00:01, 68158.10KB/s]
     21%|##1       | 27924/132723 [00:00<00:01, 66563.51KB/s]
     27%|##7       | 36329/132723 [00:00<00:01, 72625.91KB/s]
     34%|###3      | 44777/132723 [00:00<00:01, 76552.05KB/s]
     40%|###9      | 53028/132723 [00:00<00:01, 78463.47KB/s]
     46%|####6     | 61464/132723 [00:00<00:00, 80315.81KB/s]
     52%|#####2    | 69515/132723 [00:00<00:00, 73191.08KB/s]
     58%|#####8    | 77641/132723 [00:01<00:00, 75507.02KB/s]
     65%|######4   | 86065/132723 [00:01<00:00, 78046.65KB/s]
     71%|#######1  | 94519/132723 [00:01<00:00, 79952.30KB/s]
     78%|#######7  | 103014/132723 [00:01<00:00, 81426.76KB/s]
     84%|########4 | 111532/132723 [00:01<00:00, 82538.36KB/s]
     90%|######### | 120053/132723 [00:01<00:00, 83332.67KB/s]
     97%|########
 #6| 128544/132723 [00:01<00:00, 83801.60KB/s]
    100%|##########| 132723/132723 [00:01<00:00, 77987.16KB/s]
+
      0%|          | 0/132723 [00:00<?, ?KB/s]
      5%|4         | 6456/132723 [00:00<00:01, 64553.60KB/s]
     11%|#1        | 15152/132723 [00:00<00:01, 66245.19KB/s]
     16%|#6        | 21761/132723 [00:00<00:02, 47253.76KB/s]
     22%|##2       | 29611/132723 [00:00<00:01, 56581.10KB/s]
     28%|##8       | 37380/132723 [00:00<00:01, 62890.86KB/s]
     35%|###4      | 46037/132723 [00:00<00:01, 69976.68KB/s]
     41%|####1     | 54715/132723 [00:00<00:01, 75008.10KB/s]
     48%|####7     | 63417/132723 [00:00<00:00, 78604.10KB/s]
     54%|#####3    | 71521/132723 [00:01<00:00, 62891.87KB/s]
     60%|######    | 80225/132723 [00:01<00:00, 69004.12KB/s]
     66%|######6   | 87697/132723 [00:01<00:00, 68583.53KB/s]
     73%|#######2  | 96285/132723 [00:01<00:00, 73246.35KB/s]
     78%|#######8  | 103944/132723 [00:01<00:00, 53890.48KB/s]
     85%|########4 | 112611/132723 [00:01<00:00, 61231.42KB/s]
     91%|#########1| 121266/132723 [00:01<00:00, 67367.24KB/s]
     98%|########
 #7| 129994/132723 [00:01<00:00, 72488.40KB/s]
    100%|##########| 132723/132723 [00:01<00:00, 66630.50KB/s]
 
 
 
@@ -246,7 +246,7 @@ Display result
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 3 minutes  4.005 seconds)
+   **Total running time of the script:** ( 3 minutes  6.813 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 5aea8b8ef6..ea35494e07 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
@@ -5,26 +5,26 @@
 
 Computation times
 =================
-**13:34.566** total execution time for **how_to_deploy_models** files:
+**13:47.571** 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``) | 03:13.151 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_object_detection_pytorch.py` (``deploy_object_detection_pytorch.py``) | 03:16.484 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``)                           | 03:04.005 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_ssd_gluoncv.py` (``deploy_ssd_gluoncv.py``)                           | 03:06.813 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``)           | 02:23.273 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized_tflite.py` (``deploy_prequantized_tflite.py``)           | 02:23.497 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``)                               | 01:32.603 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_quantized.py` (``deploy_quantized.py``)                               | 01:37.643 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``)                         | 01:05.979 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_prequantized.py` (``deploy_prequantized.py``)                         | 01:06.526 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_adreno.py` (``deploy_model_on_adreno.py``)                   | 00:51.283 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_adreno.py` (``deploy_model_on_adreno.py``)                   | 00:51.450 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``)                 | 00:35.130 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_android.py` (``deploy_model_on_android.py``)                 | 00:35.591 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_nano.py` (``deploy_model_on_nano.py``)                       | 00:24.787 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_nano.py` (``deploy_model_on_nano.py``)                       | 00:25.014 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``)                       | 00:24.348 | 0.0 MB |
+| :ref:`sphx_glr_how_to_deploy_models_deploy_model_on_rasp.py` (``deploy_model_on_rasp.py``)                       | 00:24.545 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
 | :ref:`sphx_glr_how_to_deploy_models_deploy_sparse.py` (``deploy_sparse.py``)                                     | 00:00.006 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/extend_tvm/bring_your_own_datatypes.rst.txt b/docs/_sources/how_to/extend_tvm/bring_your_own_datatypes.rst.txt
index 5a3a7bd51b..914f171f15 100644
--- a/docs/_sources/how_to/extend_tvm/bring_your_own_datatypes.rst.txt
+++ b/docs/_sources/how_to/extend_tvm/bring_your_own_datatypes.rst.txt
@@ -463,7 +463,7 @@ First let us define two helper functions to get the mobilenet model and a cat im
 
  .. code-block:: none
 
-    Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zip6802e823-ec81-46cd-8187-6ab788d3cd23 from https://apache-mxnet.s3-accelerate.dualstack.amazonaws.com/gluon/models/mobilenet0.25-9f83e440.zip...
+    Downloading /workspace/.mxnet/models/mobilenet0.25-9f83e440.zip2bdf695d-1f09-432f-aef8-19c8b53f9e48 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 ddb6f1938b..ae488dfaa1 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
 =================
-**00:47.943** total execution time for **how_to_extend_tvm** files:
+**00:47.939** 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``) | 00:44.447 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_bring_your_own_datatypes.py` (``bring_your_own_datatypes.py``) | 00:44.449 | 0.0 MB |
 +-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``)           | 00:02.437 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_use_pass_instrument.py` (``use_pass_instrument.py``)           | 00:02.446 | 0.0 MB |
 +-------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``)                     | 00:01.052 | 0.0 MB |
+| :ref:`sphx_glr_how_to_extend_tvm_use_pass_infra.py` (``use_pass_infra.py``)                     | 00:01.038 | 0.0 MB |
 +-------------------------------------------------------------------------------------------------+-----------+--------+
 | :ref:`sphx_glr_how_to_extend_tvm_low_level_custom_pass.py` (``low_level_custom_pass.py``)       | 00:00.007 | 0.0 MB |
 +-------------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt b/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt
index 830672aca4..06d2e906d6 100644
--- a/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt
+++ b/docs/_sources/how_to/extend_tvm/use_pass_instrument.rst.txt
@@ -220,10 +220,10 @@ profile the execution time of each passes.
  .. code-block:: none
 
     Printing results of timing profile...
-    InferType: 7246us [7246us] (46.57%; 46.57%)
-    FoldScaleAxis: 8312us [7us] (53.43%; 53.43%)
-            FoldConstant: 8306us [1714us] (53.38%; 99.92%)
-                    InferType: 6592us [6592us] (42.37%; 79.37%)
+    InferType: 7769us [7769us] (47.25%; 47.25%)
+    FoldScaleAxis: 8674us [8us] (52.75%; 52.75%)
+            FoldConstant: 8666us [1910us] (52.71%; 99.91%)
+                    InferType: 6757us [6757us] (41.09%; 77.96%)
 
 
 
@@ -262,10 +262,10 @@ Refer to following sections and :py:func:`tvm.instrument.pass_instrument` for th
  .. code-block:: none
 
     Printing results of timing profile...
-    InferType: 6673us [6673us] (44.91%; 44.91%)
-    FoldScaleAxis: 8186us [5us] (55.09%; 55.09%)
-            FoldConstant: 8180us [1689us] (55.05%; 99.94%)
-                    InferType: 6492us [6492us] (43.69%; 79.36%)
+    InferType: 6741us [6741us] (44.15%; 44.15%)
+    FoldScaleAxis: 8526us [5us] (55.85%; 55.85%)
+            FoldConstant: 8521us [1900us] (55.81%; 99.94%)
+                    InferType: 6621us [6621us] (43.37%; 77.70%)
 
 
 
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 c9a371a7ed..6e9978d7e1 100644
--- a/docs/_sources/how_to/optimize_operators/opt_conv_cuda.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_conv_cuda.rst.txt
@@ -331,7 +331,7 @@ latency of convolution.
 
  .. code-block:: none
 
-    Convolution: 44.052543 ms
+    Convolution: 54.173633 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 1132256dea..6afd006f48 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
@@ -660,7 +660,7 @@ be able to run on our build server
 
  .. code-block:: none
 
-    conv2d with tensor core: 13.369545 ms
+    conv2d with tensor core: 6.566064 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 aa148ded98..a3d5c4695f 100644
--- a/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
+++ b/docs/_sources/how_to/optimize_operators/opt_gemm.rst.txt
@@ -134,8 +134,8 @@ Then we write a baseline implementation, the simplest way to write a matrix mult
 
  .. code-block:: none
 
-    Numpy running time: 0.018431
-    Baseline: 3.208171
+    Numpy running time: 0.018273
+    Baseline: 3.226653
 
 
 
@@ -229,7 +229,7 @@ fill 32 * 32 * sizeof(float) which is 4KB in the cache whose total size is 32KB
 
  .. code-block:: none
 
-    Opt1: 0.296398
+    Opt1: 0.298855
 
 
 
@@ -331,7 +331,7 @@ In this tutorial, we chose to vectorize the inner loop row data since it is cach
 
  .. code-block:: none
 
-    Opt2: 0.330635
+    Opt2: 0.337102
 
 
 
@@ -426,7 +426,7 @@ the access pattern for A matrix is more cache friendly.
 
  .. code-block:: none
 
-    Opt3: 0.114155
+    Opt3: 0.115017
 
 
 
@@ -550,7 +550,7 @@ flattening.
 
  .. code-block:: none
 
-    Opt4: 0.110279
+    Opt4: 0.108878
 
 
 
@@ -671,7 +671,7 @@ write to C when all the block results are ready.
 
  .. code-block:: none
 
-    Opt5: 0.111988
+    Opt5: 0.111888
 
 
 
@@ -795,7 +795,7 @@ Furthermore, we can also utilize multi-core processors to do the thread-level pa
 
  .. code-block:: none
 
-    Opt6: 0.146003
+    Opt6: 0.146227
 
 
 
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 ef54ff4875..52437dfdf7 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
 =================
-**00:34.116** total execution time for **how_to_optimize_operators** files:
+**00:34.226** total execution time for **how_to_optimize_operators** files:
 
 +-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``)                       | 00:31.510 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_gemm.py` (``opt_gemm.py``)                       | 00:31.682 | 0.0 MB |
 +-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``) | 00:01.537 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_tensorcore.py` (``opt_conv_tensorcore.py``) | 00:01.432 | 0.0 MB |
 +-----------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``)             | 00:01.069 | 0.0 MB |
+| :ref:`sphx_glr_how_to_optimize_operators_opt_conv_cuda.py` (``opt_conv_cuda.py``)             | 00:01.112 | 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 9186bda969..923c164641 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
 =================
-**09:10.972** total execution time for **how_to_tune_with_autoscheduler** files:
+**08:52.998** 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``) | 05:38.343 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_conv2d_layer_cuda.py` (``tune_conv2d_layer_cuda.py``) | 05:29.208 | 0.0 MB |
 +----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``)             | 01:30.980 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_x86.py` (``tune_network_x86.py``)             | 01:31.583 | 0.0 MB |
 +----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``)           | 01:01.132 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_cuda.py` (``tune_network_cuda.py``)           | 01:01.747 | 0.0 MB |
 +----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``)               | 00:37.743 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_sparse_x86.py` (``tune_sparse_x86.py``)               | 00:26.987 | 0.0 MB |
 +----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``)             | 00:11.847 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_arm.py` (``tune_network_arm.py``)             | 00:12.168 | 0.0 MB |
 +----------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``)           | 00:10.927 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autoscheduler_tune_network_mali.py` (``tune_network_mali.py``)           | 00:11.306 | 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 6f2af2656c..9b99a24e3a 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
@@ -209,13 +209,6 @@ file and apply it.
 
 
 
-.. rst-class:: sphx-glr-script-out
-
- .. code-block:: none
-
-    .T
-
-
 
 
 
@@ -249,399 +242,104 @@ cooperative fetching, unrolling and operator fusion.
                  bias: Buffer(bias_2: Pointer(float32), float32, [1, 512, 1, 1], []),
                  compute: Buffer(compute_2: Pointer(float32), float32, [1, 512, 7, 7], [])}
       buffer_map = {data_1: data, kernel_1: kernel, bias_1: bias, compute_1: compute} {
-      attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 8;
-      allocate(conv2d_nchw: Pointer(local float32), float32, [14]), storage_scope = local;
-      allocate(pad_temp.shared: Pointer(shared float32), float32, [504]), storage_scope = shared;
-      allocate(kernel.shared: Pointer(shared float32), float32, [1536]), storage_scope = shared;
+      attr [IterVar(blockIdx.x: int32, (nullptr), "ThreadIndex", "blockIdx.x")] "thread_extent" = 56;
+      allocate(conv2d_nchw: Pointer(local float32), float32, [2]), storage_scope = local;
+      allocate(pad_temp.shared: Pointer(shared float32), float32, [216]), storage_scope = shared;
+      allocate(kernel.shared: Pointer(shared float32), float32, [4608]), storage_scope = shared;
       attr [IterVar(threadIdx.x: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224 {
-        conv2d_nchw_1: Buffer(conv2d_nchw, float32, [49], [], scope="local", align=16)[0] = 0f32
-        conv2d_nchw_1[7] = 0f32
+        conv2d_nchw_1: Buffer(conv2d_nchw, float32, [1], [], scope="local", align=4)[0] = 0f32
         conv2d_nchw_1[1] = 0f32
-        conv2d_nchw_1[8] = 0f32
-        conv2d_nchw_1[2] = 0f32
-        conv2d_nchw_1[9] = 0f32
-        conv2d_nchw_1[3] = 0f32
-        conv2d_nchw_1[10] = 0f32
-        conv2d_nchw_1[4] = 0f32
-        conv2d_nchw_1[11] = 0f32
-        conv2d_nchw_1[5] = 0f32
-        conv2d_nchw_1[12] = 0f32
-        conv2d_nchw_1[6] = 0f32
-        conv2d_nchw_1[13] = 0f32
         for (rc.outer.outer: int32, 0, 64) {
-          for (ry.outer.outer: int32, 0, 3) {
-            let cse_var_4: int32 = (rc.outer.outer*392)
-            let cse_var_3: int32 = (ry.outer.outer*7)
-            let cse_var_2: int32 = (rc.outer.outer*72)
-            let cse_var_1: int32 = (ry.outer.outer*3)
-             {
-              attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
-              pad_temp.shared_1: Buffer(pad_temp.shared, float32, [504], [], scope="shared")[threadIdx.x_1] = @tir.if_then_else(((((1 <= (floordiv(floormod(threadIdx.x_1, 63), 9) + ry.outer.outer)) && ((floordiv(floormod(threadIdx.x_1, 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod(threadIdx.x_1, 9))) && (floormod(threadIdx.x_1, 9) < 8)), data_3: Buffer(data_2, float32, [25088], [])[((((cse_var_4 + (floordiv(threadIdx.x_1, 9)*7)) + cse_var_3) + floormod(threadIdx.x_1, 9)) - 8)], 0f32, [...]
-              attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
-              pad_temp.shared_1[(threadIdx.x_1 + 224)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 35), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 35), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 8), 9))) && (floormod((threadIdx.x_1 + 8), 9) < 8)), data_3[((((cse_var_4 + (floordiv((threadIdx.x_1 + 224), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 8), 9)) - 8)], 0f32, dtype=float32)
-              attr [IterVar(threadIdx.x_1, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
-              if @tir.likely((threadIdx.x_1 < 56), dtype=bool) {
-                pad_temp.shared_1[(threadIdx.x_1 + 448)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1 + 7), 63), 9) + ry.outer.outer)) && ((floordiv(floormod((threadIdx.x_1 + 7), 63), 9) + ry.outer.outer) < 8)) && (1 <= floormod((threadIdx.x_1 + 7), 9))) && (floormod((threadIdx.x_1 + 7), 9) < 8)), data_3[((((cse_var_4 + (floordiv((threadIdx.x_1 + 448), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 7), 9)) - 8)], 0f32, dtype=float32)
+          let cse_var_1: int32 = (rc.outer.outer*72)
+           {
+            attr [IterVar(threadIdx.x_1: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224 {
+              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
+                pad_temp.shared_1: Buffer(pad_temp.shared, float32, [216], [], scope="shared")[(threadIdx.x_1*6)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1*2), 9), 3) + floormod(blockIdx.x, 7))) && ((floordiv(floormod((threadIdx.x_1*2), 9), 3) + floormod(blockIdx.x, 7)) < 8)) && (1 < floormod((threadIdx.x_1*6), 9))) && (floormod((threadIdx.x_1*6), 9) < 8)), data_3: Buffer(data_2, float32, [25088], [])[((((((rc.outer.outer*392) + (floordiv((threadIdx.x_1*2), 9)*49)) + [...]
               }
-              attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
-              kernel.shared_1: Buffer(kernel.shared, float32, [1536], [], scope="shared")[threadIdx.x_2] = kernel_3: Buffer(kernel_2, float32, [2359296], [])[((((((blockIdx.x*294912) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
-              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
-              kernel.shared_1[(threadIdx.x_2 + 224)] = kernel_3[((((((blockIdx.x*294912) + (floordiv((threadIdx.x_2 + 224), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
-              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
-              kernel.shared_1[(threadIdx.x_2 + 448)] = kernel_3[((((((blockIdx.x*294912) + (floordiv((threadIdx.x_2 + 448), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
-              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
-              kernel.shared_1[(threadIdx.x_2 + 672)] = kernel_3[(((((((blockIdx.x*294912) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 129024)]
-              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
-              kernel.shared_1[(threadIdx.x_2 + 896)] = kernel_3[((((((blockIdx.x*294912) + (floordiv((threadIdx.x_2 + 896), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
-              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
-              kernel.shared_1[(threadIdx.x_2 + 1120)] = kernel_3[((((((blockIdx.x*294912) + (floordiv((threadIdx.x_2 + 1120), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
-              attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
-              if @tir.likely((threadIdx.x_2 < 192), dtype=bool) {
-                kernel.shared_1[(threadIdx.x_2 + 1344)] = kernel_3[(((((((blockIdx.x*294912) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 258048)]
+              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
+                pad_temp.shared_1[((threadIdx.x_1*6) + 1)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1*2), 9), 3) + floormod(blockIdx.x, 7))) && ((floordiv(floormod((threadIdx.x_1*2), 9), 3) + floormod(blockIdx.x, 7)) < 8)) && (1 <= floormod(((threadIdx.x_1*6) + 1), 9))) && (floormod(((threadIdx.x_1*6) + 1), 9) < 8)), data_3[((((((rc.outer.outer*392) + (floordiv((threadIdx.x_1*2), 9)*49)) + (floordiv(floormod((threadIdx.x_1*2), 9), 3)*7)) + (floormod(blockIdx.x, 7)*7)) [...]
               }
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7)*9)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
-              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7)*9)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 768)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 1)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
-              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 1)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 769)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 2)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
-              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 2)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 770)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
-              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 771)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 64)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
-              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 64)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 772)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
-              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 773)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
-              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 774)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
-              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 775)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
-              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 776)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
-              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 777)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
-              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 778)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
-              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 779)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
-              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 780)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
-              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 781)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
-              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 782)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
-              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 783)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 316)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
-              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 316)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 784)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 317)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
-              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 317)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 785)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
-              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 786)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 379)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
-              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 379)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 787)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 380)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
-              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 380)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 788)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
-              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 789)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 442)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
-              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 442)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 790)]))
-              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 443)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
-              conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 443)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 791)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 1)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
-              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 1)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 768)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 2)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
-              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 2)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 769)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 3)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
-              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 3)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 770)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 64)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
-              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 64)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 771)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
-              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 772)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 66)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
-              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 66)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 773)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
-              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 774)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
-              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 775)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 129)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
-              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 129)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 776)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
-              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 777)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
-              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 778)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 192)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
-              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 192)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 779)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
-              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 780)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
-              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 781)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 255)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
-              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 255)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 782)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 316)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
-              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 316)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 783)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 317)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
-              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 317)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 784)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 318)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
-              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 318)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 785)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 379)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
-              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 379)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 786)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 380)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
-              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 380)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 787)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 381)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
-              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 381)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 788)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 442)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
-              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 442)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 789)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 443)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
-              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 443)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 790)]))
-              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 444)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
-              conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 444)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 791)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 2)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
-              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 2)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 768)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 3)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
-              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 3)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 769)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 4)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
-              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 4)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 770)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
-              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 771)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 66)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
-              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 66)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 772)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 67)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
-              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 67)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 773)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
-              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 774)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 129)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
-              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 129)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 775)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 130)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
-              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 130)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 776)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
-              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 777)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 192)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
-              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 192)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 778)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 193)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
-              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 193)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 779)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
-              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 780)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 255)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
-              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 255)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 781)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 256)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
-              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 256)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 782)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 317)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
-              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 317)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 783)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 318)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
-              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 318)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 784)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 319)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
-              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 319)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 785)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 380)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
-              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 380)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 786)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 381)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
-              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 381)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 787)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 382)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
-              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 382)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 788)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 443)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
-              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 443)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 789)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 444)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
-              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 444)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 790)]))
-              conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 445)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
-              conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 445)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 791)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 3)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
-              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 3)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 768)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 4)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
-              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 4)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 769)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 5)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
-              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 5)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 770)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 66)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
-              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 66)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 771)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 67)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
-              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 67)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 772)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 68)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
-              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 68)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 773)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 129)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
-              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 129)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 774)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 130)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
-              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 130)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 775)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 131)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
-              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 131)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 776)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 192)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
-              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 192)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 777)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 193)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
-              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 193)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 778)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 194)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
-              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 194)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 779)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 255)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
-              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 255)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 780)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 256)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
-              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 256)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 781)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 257)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
-              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 257)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 782)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 318)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
-              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 318)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 783)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 319)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
-              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 319)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 784)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 320)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
-              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 320)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 785)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 381)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
-              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 381)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 786)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 382)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
-              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 382)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 787)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 383)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
-              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 383)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 788)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 444)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
-              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 444)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 789)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 445)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
-              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 445)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 790)]))
-              conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 446)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
-              conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 446)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 791)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 4)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
-              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 4)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 768)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 5)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
-              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 5)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 769)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 6)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
-              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 6)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 770)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 67)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
-              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 67)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 771)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 68)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
-              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 68)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 772)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 69)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
-              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 69)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 773)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 130)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
-              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 130)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 774)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 131)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
-              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 131)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 775)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 132)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
-              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 132)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 776)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 193)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
-              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 193)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 777)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 194)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
-              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 194)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 778)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 195)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
-              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 195)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 779)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 256)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
-              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 256)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 780)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 257)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
-              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 257)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 781)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 258)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
-              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 258)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 782)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 319)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
-              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 319)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 783)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 320)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
-              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 320)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 784)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 321)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
-              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 321)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 785)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 382)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
-              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 382)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 786)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 383)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
-              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 383)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 787)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 384)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
-              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 384)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 788)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 445)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
-              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 445)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 789)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 446)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
-              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 446)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 790)]))
-              conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 447)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
-              conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 447)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 791)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 5)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
-              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 5)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 768)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 6)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
-              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 6)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 769)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
-              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 770)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 68)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
-              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 68)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 771)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 69)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
-              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 69)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 772)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
-              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 773)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 131)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
-              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 131)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 774)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 132)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
-              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 132)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 775)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
-              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 776)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 194)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
-              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 194)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 777)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 195)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
-              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 195)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 778)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
-              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 779)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 257)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
-              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 257)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 780)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 258)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
-              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 258)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 781)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
-              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 782)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 320)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
-              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 320)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 783)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 321)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
-              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 321)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 784)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
-              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 785)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 383)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
-              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 383)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 786)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 384)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
-              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 384)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 787)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
-              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 788)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 446)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
-              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 446)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 789)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 447)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
-              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 447)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 790)]))
-              conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
-              conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 791)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 6)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
-              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 6)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 768)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
-              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 769)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 8)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
-              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 8)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 770)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 69)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
-              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 69)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 771)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
-              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 772)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 71)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
-              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 71)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 773)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 132)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
-              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 132)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 774)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
-              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 775)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 134)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
-              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 134)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 776)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 195)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
-              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 195)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 777)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
-              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 778)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 197)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
-              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 197)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 779)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 258)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
-              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 258)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 780)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
-              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 781)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 260)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
-              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 260)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 782)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 321)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
-              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 321)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 783)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
-              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 784)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 323)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
-              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 323)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 785)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 384)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
-              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 384)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 786)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
-              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 787)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 386)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
-              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 386)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 788)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 447)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
-              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 447)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 789)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
-              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 790)]))
-              conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 449)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
-              conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 449)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 791)]))
+              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
+                pad_temp.shared_1[((threadIdx.x_1*6) + 2)] = @tir.if_then_else(((((1 <= (floordiv(floormod((threadIdx.x_1*2), 9), 3) + floormod(blockIdx.x, 7))) && ((floordiv(floormod((threadIdx.x_1*2), 9), 3) + floormod(blockIdx.x, 7)) < 8)) && (1 < floormod(((threadIdx.x_1*6) + 2), 9))) && (floormod(((threadIdx.x_1*6) + 2), 9) < 8)), data_3[((((((rc.outer.outer*392) + (floordiv((threadIdx.x_1*2), 9)*49)) + (floordiv(floormod((threadIdx.x_1*2), 9), 3)*7)) + (floormod(blockIdx.x, 7)*7))  [...]
+              }
+              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
+                pad_temp.shared_1[((threadIdx.x_1*6) + 3)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*2) + 1), 9), 3) + floormod(blockIdx.x, 7))) && ((floordiv(floormod(((threadIdx.x_1*2) + 1), 9), 3) + floormod(blockIdx.x, 7)) < 8)) && (1 < floormod(((threadIdx.x_1*6) + 3), 9))) && (floormod(((threadIdx.x_1*6) + 3), 9) < 8)), data_3[((((((rc.outer.outer*392) + (floordiv(((threadIdx.x_1*2) + 1), 9)*49)) + (floordiv(floormod(((threadIdx.x_1*2) + 1), 9), 3)*7)) + (floo [...]
+              }
+              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
+                pad_temp.shared_1[((threadIdx.x_1*6) + 4)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*2) + 1), 9), 3) + floormod(blockIdx.x, 7))) && ((floordiv(floormod(((threadIdx.x_1*2) + 1), 9), 3) + floormod(blockIdx.x, 7)) < 8)) && (1 <= floormod(((threadIdx.x_1*6) + 4), 9))) && (floormod(((threadIdx.x_1*6) + 4), 9) < 8)), data_3[((((((rc.outer.outer*392) + (floordiv(((threadIdx.x_1*2) + 1), 9)*49)) + (floordiv(floormod(((threadIdx.x_1*2) + 1), 9), 3)*7)) + (flo [...]
+              }
+              if @tir.likely((threadIdx.x_1 < 36), dtype=bool) {
+                pad_temp.shared_1[((threadIdx.x_1*6) + 5)] = @tir.if_then_else(((((1 <= (floordiv(floormod(((threadIdx.x_1*2) + 1), 9), 3) + floormod(blockIdx.x, 7))) && ((floordiv(floormod(((threadIdx.x_1*2) + 1), 9), 3) + floormod(blockIdx.x, 7)) < 8)) && (1 < floormod(((threadIdx.x_1*6) + 5), 9))) && (floormod(((threadIdx.x_1*6) + 5), 9) < 8)), data_3[((((((rc.outer.outer*392) + (floordiv(((threadIdx.x_1*2) + 1), 9)*49)) + (floordiv(floormod(((threadIdx.x_1*2) + 1), 9), 3)*7)) + (floo [...]
+              }
+            }
+            attr [IterVar(threadIdx.x_2: int32, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
+            kernel.shared_1: Buffer(kernel.shared, float32, [4608], [], scope="shared")[threadIdx.x_2] = kernel_3: Buffer(kernel_2, float32, [2359296], [])[((((floordiv(blockIdx.x, 7)*294912) + (floordiv(threadIdx.x_2, 72)*4608)) + cse_var_1) + floormod(threadIdx.x_2, 72))]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
+            kernel.shared_1[(threadIdx.x_2 + 224)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 224), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 8), 72), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
+            kernel.shared_1[(threadIdx.x_2 + 448)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 448), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 16), 72), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
+            kernel.shared_1[(threadIdx.x_2 + 672)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 672), 72)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 8), 24)*3)) + floormod(threadIdx.x_2, 3))]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
+            kernel.shared_1[(threadIdx.x_2 + 896)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 896), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 32), 72), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
+            kernel.shared_1[(threadIdx.x_2 + 1120)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 1120), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 40), 72), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
+            kernel.shared_1[(threadIdx.x_2 + 1344)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 1344), 72)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 16), 24)*3)) + floormod(threadIdx.x_2, 3))]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
+            kernel.shared_1[(threadIdx.x_2 + 1568)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 1568), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 56), 72), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
+            kernel.shared_1[(threadIdx.x_2 + 1792)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 1792), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 64), 72), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
+            kernel.shared_1[(threadIdx.x_2 + 2016)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv(threadIdx.x_2, 72)*4608)) + cse_var_1) + floormod(threadIdx.x_2, 72)) + 129024)]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
+            kernel.shared_1[(threadIdx.x_2 + 2240)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 2240), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 8), 72), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
+            kernel.shared_1[(threadIdx.x_2 + 2464)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 2464), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 16), 72), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
+            kernel.shared_1[(threadIdx.x_2 + 2688)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 2688), 72)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 8), 24)*3)) + floormod(threadIdx.x_2, 3))]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
+            kernel.shared_1[(threadIdx.x_2 + 2912)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 2912), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 32), 72), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
+            kernel.shared_1[(threadIdx.x_2 + 3136)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 3136), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 40), 72), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
+            kernel.shared_1[(threadIdx.x_2 + 3360)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 3360), 72)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 16), 24)*3)) + floormod(threadIdx.x_2, 3))]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
+            kernel.shared_1[(threadIdx.x_2 + 3584)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 3584), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 56), 72), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
+            kernel.shared_1[(threadIdx.x_2 + 3808)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 3808), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 64), 72), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
+            kernel.shared_1[(threadIdx.x_2 + 4032)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv(threadIdx.x_2, 72)*4608)) + cse_var_1) + floormod(threadIdx.x_2, 72)) + 258048)]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
+            kernel.shared_1[(threadIdx.x_2 + 4256)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 4256), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 8), 72), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+            attr [IterVar(threadIdx.x_2, (nullptr), "ThreadIndex", "threadIdx.x")] "thread_extent" = 224;
+            if @tir.likely((threadIdx.x_2 < 128), dtype=bool) {
+              kernel.shared_1[(threadIdx.x_2 + 4480)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 4480), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 16), 72), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+            }
+            for (rc.outer.inner: int32, 0, 8) {
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((rc.outer.inner*27) + floormod(threadIdx.x, 7))]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9))]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((rc.outer.inner*27) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 2304)]))
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 1)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 2305)]))
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 2)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 2306)]))
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 3)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 2307)]))
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 10)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 4)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 10)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 2308)]))
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 11)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 5)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 11)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 2309)]))
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 6)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 2310)]))
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 19)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 7)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 19)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 2311)]))
+              conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 20)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 8)]))
+              conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 20)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 2312)]))
             }
           }
         }
-        for (i3.inner: int32, 0, 7) {
-          compute_3: Buffer(compute_2, float32, [25088], [])[(((blockIdx.x*3136) + (threadIdx.x*7)) + i3.inner)] = max((conv2d_nchw_1[i3.inner] + bias_3: Buffer(bias_2, float32, [512], [])[((blockIdx.x*64) + floordiv(threadIdx.x, 7))]), 0f32)
-          compute_3[((((blockIdx.x*3136) + (threadIdx.x*7)) + i3.inner) + 1568)] = max((conv2d_nchw_1[(i3.inner + 7)] + bias_3[(((blockIdx.x*64) + floordiv(threadIdx.x, 7)) + 32)]), 0f32)
-        }
+        compute_3: Buffer(compute_2, float32, [25088], [])[((((floordiv(blockIdx.x, 7)*3136) + (floordiv(threadIdx.x, 7)*49)) + (floormod(blockIdx.x, 7)*7)) + floormod(threadIdx.x, 7))] = max((conv2d_nchw_1[0] + bias_3: Buffer(bias_2, float32, [512], [])[((floordiv(blockIdx.x, 7)*64) + floordiv(threadIdx.x, 7))]), 0f32)
+        compute_3[(((((floordiv(blockIdx.x, 7)*3136) + (floordiv(threadIdx.x, 7)*49)) + (floormod(blockIdx.x, 7)*7)) + floormod(threadIdx.x, 7)) + 1568)] = max((conv2d_nchw_1[1] + bias_3[(((floordiv(blockIdx.x, 7)*64) + floordiv(threadIdx.x, 7)) + 32)]), 0f32)
       }
     }
 
@@ -695,7 +393,7 @@ We build the binary and check its correctness and performance.
 
  .. code-block:: none
 
-    Execution time of this operator: 0.335 ms
+    Execution time of this operator: 0.350 ms
 
 
 
@@ -749,18 +447,18 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
     conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=2)
     conv2d_nchw_yy_o_i, conv2d_nchw_yy_i = s[conv2d_nchw].split(conv2d_nchw_yy, factor=1)
     conv2d_nchw_yy_o_o_i, conv2d_nchw_yy_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_i, factor=1)
-    conv2d_nchw_yy_o_o_o_i, conv2d_nchw_yy_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_i, factor=7)
+    conv2d_nchw_yy_o_o_o_i, conv2d_nchw_yy_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_i, factor=1)
     conv2d_nchw_yy_o_o_o_o, conv2d_nchw_yy_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_o_i, factor=1)
     conv2d_nchw_xx_o_i, conv2d_nchw_xx_i = s[conv2d_nchw].split(conv2d_nchw_xx, factor=1)
-    conv2d_nchw_xx_o_o_i, conv2d_nchw_xx_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_i, factor=7)
-    conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, factor=1)
+    conv2d_nchw_xx_o_o_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=8)
-    conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_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=8)
     conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=1)
-    conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=1)
-    conv2d_nchw_rx_o_i, conv2d_nchw_rx_i = s[conv2d_nchw].split(conv2d_nchw_rx, factor=3)
-    conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=1)
+    conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=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=3)
     s[conv2d_nchw].reorder(conv2d_nchw_nn_o_o_o_o, conv2d_nchw_ff_o_o_o_o, conv2d_nchw_yy_o_o_o_o, conv2d_nchw_xx_o_o_o_o, conv2d_nchw_nn_o_o_o_i, conv2d_nchw_ff_o_o_o_i, conv2d_nchw_yy_o_o_o_i, conv2d_nchw_xx_o_o_o_i, conv2d_nchw_nn_o_o_i, conv2d_nchw_ff_o_o_i, conv2d_nchw_yy_o_o_i, conv2d_nchw_xx_o_o_i, conv2d_nchw_rc_o_o, conv2d_nchw_ry_o_o, conv2d_nchw_rx_o_o, conv2d_nchw_rc_o_i, conv2d_nchw_ry_o_i, conv2d_nchw_rx_o_i, conv2d_nchw_nn_o_i, conv2d_nchw_ff_o_i, conv2d_nchw_yy_o_i, conv2 [...]
     compute_i0_o_i, compute_i0_i = s[compute].split(compute_i0, factor=1)
     compute_i0_o_o_i, compute_i0_o_i = s[compute].split(compute_i0_o_i, factor=1)
@@ -769,10 +467,10 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
     compute_i1_o_o_i, compute_i1_o_i = s[compute].split(compute_i1_o_i, factor=32)
     compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=2)
     compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, factor=1)
-    compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=7)
+    compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=1)
     compute_i2_o_o_o, compute_i2_o_o_i = s[compute].split(compute_i2_o_o_i, factor=1)
-    compute_i3_o_i, compute_i3_i = s[compute].split(compute_i3, factor=7)
-    compute_i3_o_o_i, compute_i3_o_i = s[compute].split(compute_i3_o_i, factor=1)
+    compute_i3_o_i, compute_i3_i = s[compute].split(compute_i3, factor=1)
+    compute_i3_o_o_i, compute_i3_o_i = s[compute].split(compute_i3_o_i, factor=7)
     compute_i3_o_o_o, compute_i3_o_o_i = s[compute].split(compute_i3_o_o_i, factor=1)
     s[compute].reorder(compute_i0_o_o_o, compute_i1_o_o_o, compute_i2_o_o_o, compute_i3_o_o_o, compute_i0_o_o_i, compute_i1_o_o_i, compute_i2_o_o_i, compute_i3_o_o_i, compute_i0_o_i, compute_i1_o_i, compute_i2_o_i, compute_i3_o_i, compute_i0_i, compute_i1_i, compute_i2_i, compute_i3_i)
     s[conv2d_nchw].compute_at(s[compute], compute_i3_o_i)
@@ -795,11 +493,11 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
     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=224)
     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)
+    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=6)
     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=224)
     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", 64)
     s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, "unroll_explicit", True)
 
     CUDA source code:
@@ -818,383 +516,78 @@ They can be used for debugging and learning the behavior of the auto-scheduler.
       #define uint64_t unsigned long long
     #endif
     extern "C" __global__ void __launch_bounds__(224) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
-      float conv2d_nchw[14];
-      __shared__ float pad_temp_shared[504];
-      __shared__ float kernel_shared[1536];
+      float conv2d_nchw[2];
+      __shared__ float pad_temp_shared[216];
+      __shared__ float kernel_shared[4608];
       conv2d_nchw[0] = 0.000000e+00f;
-      conv2d_nchw[7] = 0.000000e+00f;
       conv2d_nchw[1] = 0.000000e+00f;
-      conv2d_nchw[8] = 0.000000e+00f;
-      conv2d_nchw[2] = 0.000000e+00f;
-      conv2d_nchw[9] = 0.000000e+00f;
-      conv2d_nchw[3] = 0.000000e+00f;
-      conv2d_nchw[10] = 0.000000e+00f;
-      conv2d_nchw[4] = 0.000000e+00f;
-      conv2d_nchw[11] = 0.000000e+00f;
-      conv2d_nchw[5] = 0.000000e+00f;
-      conv2d_nchw[12] = 0.000000e+00f;
-      conv2d_nchw[6] = 0.000000e+00f;
-      conv2d_nchw[13] = 0.000000e+00f;
       for (int rc_outer_outer = 0; rc_outer_outer < 64; ++rc_outer_outer) {
-        for (int ry_outer_outer = 0; ry_outer_outer < 3; ++ry_outer_outer) {
-          __syncthreads();
-          pad_temp_shared[((int)threadIdx.x)] = (((((1 <= (((((int)threadIdx.x) % 63) / 9) + ry_outer_outer)) && ((((((int)threadIdx.x) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= (((int)threadIdx.x) % 9))) && ((((int)threadIdx.x) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + ((((int)threadIdx.x) / 9) * 7)) + (ry_outer_outer * 7)) + (((int)threadIdx.x) % 9)) - 8)] : 0.000000e+00f);
-          pad_temp_shared[(((int)threadIdx.x) + 224)] = (((((1 <= ((((((int)threadIdx.x) + 35) % 63) / 9) + ry_outer_outer)) && (((((((int)threadIdx.x) + 35) % 63) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 8) % 9))) && (((((int)threadIdx.x) + 8) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 224) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 8) % 9)) - 8)] : 0.000000e+00f);
-          if (((int)threadIdx.x) < 56) {
-            pad_temp_shared[(((int)threadIdx.x) + 448)] = (((((1 <= (((((int)threadIdx.x) + 7) / 9) + ry_outer_outer)) && ((((((int)threadIdx.x) + 7) / 9) + ry_outer_outer) < 8)) && (1 <= ((((int)threadIdx.x) + 7) % 9))) && (((((int)threadIdx.x) + 7) % 9) < 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 448) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 7) % 9)) - 8)] : 0.000000e+00f);
-          }
-          kernel_shared[((int)threadIdx.x)] = kernel[((((((((int)blockIdx.x) * 294912) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
-          kernel_shared[(((int)threadIdx.x) + 224)] = kernel[((((((((int)blockIdx.x) * 294912) + (((((int)threadIdx.x) + 224) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
-          kernel_shared[(((int)threadIdx.x) + 448)] = kernel[((((((((int)blockIdx.x) * 294912) + (((((int)threadIdx.x) + 448) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
-          kernel_shared[(((int)threadIdx.x) + 672)] = kernel[(((((((((int)blockIdx.x) * 294912) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 129024)];
-          kernel_shared[(((int)threadIdx.x) + 896)] = kernel[((((((((int)blockIdx.x) * 294912) + (((((int)threadIdx.x) + 896) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
-          kernel_shared[(((int)threadIdx.x) + 1120)] = kernel[((((((((int)blockIdx.x) * 294912) + (((((int)threadIdx.x) + 1120) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
-          if (((int)threadIdx.x) < 192) {
-            kernel_shared[(((int)threadIdx.x) + 1344)] = kernel[(((((((((int)blockIdx.x) * 294912) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 258048)];
-          }
-          __syncthreads();
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) * 9)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
-          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) * 9)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 768)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
-          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 769)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
-          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 770)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
-          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 771)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
-          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 772)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
-          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 773)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
-          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 774)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
-          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 775)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
-          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 776)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
-          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 777)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
-          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 778)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
-          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 779)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
-          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 780)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
-          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 781)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
-          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 782)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 315)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
-          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 315)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 783)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 316)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
-          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 316)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 784)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 317)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
-          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 317)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 785)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 378)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
-          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 378)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 786)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 379)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
-          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 379)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 787)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 380)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
-          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 380)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 788)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 441)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
-          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 441)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 789)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 442)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
-          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 442)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 790)]));
-          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 443)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
-          conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 443)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 791)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 1)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
-          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 768)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
-          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 769)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 3)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
-          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 3)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 770)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
-          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 771)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
-          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 772)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 66)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
-          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 66)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 773)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
-          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 774)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
-          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 775)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 129)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
-          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 129)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 776)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
-          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 777)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
-          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 778)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 192)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
-          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 192)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 779)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
-          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 780)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
-          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 781)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 255)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
-          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 255)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 782)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 316)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
-          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 316)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 783)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 317)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
-          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 317)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 784)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 318)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
-          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 318)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 785)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 379)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
-          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 379)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 786)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 380)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
-          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 380)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 787)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 381)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
-          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 381)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 788)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 442)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
-          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 442)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 789)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 443)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
-          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 443)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 790)]));
-          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 444)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
-          conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 444)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 791)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 2)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
-          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 768)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 3)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
-          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 3)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 769)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 4)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
-          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 4)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 770)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
-          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 771)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 66)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
-          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 66)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 772)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 67)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
-          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 67)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 773)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
-          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 774)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 129)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
-          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 129)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 775)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 130)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
-          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 130)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 776)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
-          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 777)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 192)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
-          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 192)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 778)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 193)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
-          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 193)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 779)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
-          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 780)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 255)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
-          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 255)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 781)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 256)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
-          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 256)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 782)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 317)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
-          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 317)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 783)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 318)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
-          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 318)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 784)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 319)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
-          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 319)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 785)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 380)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
-          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 380)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 786)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 381)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
-          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 381)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 787)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 382)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
-          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 382)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 788)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 443)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
-          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 443)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 789)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 444)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
-          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 444)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 790)]));
-          conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 445)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
-          conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 445)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 791)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 3)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
-          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 3)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 768)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 4)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
-          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 4)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 769)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 5)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
-          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 5)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 770)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 66)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
-          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 66)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 771)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 67)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
-          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 67)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 772)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 68)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
-          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 68)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 773)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 129)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
-          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 129)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 774)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 130)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
-          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 130)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 775)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 131)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
-          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 131)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 776)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 192)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
-          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 192)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 777)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 193)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
-          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 193)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 778)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 194)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
-          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 194)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 779)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 255)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
-          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 255)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 780)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 256)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
-          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 256)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 781)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 257)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
-          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 257)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 782)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 318)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
-          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 318)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 783)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 319)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
-          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 319)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 784)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 320)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
-          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 320)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 785)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 381)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
-          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 381)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 786)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 382)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
-          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 382)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 787)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 383)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
-          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 383)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 788)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 444)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
-          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 444)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 789)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 445)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
-          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 445)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 790)]));
-          conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 446)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
-          conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 446)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 791)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 4)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
-          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 4)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 768)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 5)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
-          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 5)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 769)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 6)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
-          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 6)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 770)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 67)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
-          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 67)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 771)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 68)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
-          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 68)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 772)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 69)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
-          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 69)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 773)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 130)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
-          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 130)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 774)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 131)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
-          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 131)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 775)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 132)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
-          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 132)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 776)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 193)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
-          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 193)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 777)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 194)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
-          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 194)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 778)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 195)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
-          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 195)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 779)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 256)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
-          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 256)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 780)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 257)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
-          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 257)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 781)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 258)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
-          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 258)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 782)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 319)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
-          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 319)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 783)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 320)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
-          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 320)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 784)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 321)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
-          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 321)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 785)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 382)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
-          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 382)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 786)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 383)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
-          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 383)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 787)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 384)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
-          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 384)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 788)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 445)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
-          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 445)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 789)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 446)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
-          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 446)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 790)]));
-          conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 447)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
-          conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 447)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 791)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 5)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
-          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 5)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 768)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 6)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
-          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 6)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 769)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 7)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
-          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 7)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 770)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 68)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
-          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 68)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 771)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 69)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
-          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 69)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 772)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 70)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
-          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 70)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 773)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 131)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
-          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 131)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 774)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 132)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
-          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 132)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 775)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 133)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
-          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 133)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 776)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 194)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
-          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 194)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 777)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 195)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
-          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 195)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 778)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 196)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
-          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 196)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 779)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 257)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
-          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 257)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 780)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 258)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
-          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 258)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 781)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 259)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
-          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 259)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 782)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 320)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
-          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 320)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 783)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 321)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
-          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 321)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 784)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 322)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
-          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 322)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 785)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 383)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
-          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 383)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 786)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 384)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
-          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 384)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 787)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 385)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
-          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 385)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 788)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 446)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
-          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 446)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 789)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 447)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
-          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 447)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 790)]));
-          conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 448)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
-          conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 448)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 791)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 6)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
-          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 6)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 768)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 7)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
-          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 7)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 769)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 8)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
-          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 8)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 770)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 69)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
-          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 69)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 771)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 70)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
-          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 70)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 772)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 71)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
-          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 71)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 773)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 132)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
-          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 132)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 774)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 133)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
-          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 133)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 775)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 134)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
-          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 134)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 776)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 195)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
-          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 195)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 777)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 196)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
-          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 196)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 778)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 197)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
-          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 197)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 779)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 258)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
-          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 258)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 780)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 259)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
-          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 259)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 781)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 260)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
-          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 260)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 782)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 321)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
-          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 321)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 783)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 322)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
-          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 322)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 784)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 323)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
-          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 323)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 785)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 384)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
-          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 384)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 786)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 385)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
-          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 385)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 787)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 386)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
-          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 386)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 788)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 447)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
-          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 447)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 789)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 448)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
-          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 448)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 790)]));
-          conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 449)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
-          conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 449)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 791)]));
+        __syncthreads();
+        if (((int)threadIdx.x) < 36) {
+          pad_temp_shared[(((int)threadIdx.x) * 6)] = (((((1 <= ((((((int)threadIdx.x) * 2) % 9) / 3) + (((int)blockIdx.x) % 7))) && (((((((int)threadIdx.x) * 2) % 9) / 3) + (((int)blockIdx.x) % 7)) < 8)) && (1 < ((((int)threadIdx.x) * 6) % 9))) && (((((int)threadIdx.x) * 6) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + (((((int)threadIdx.x) * 2) / 9) * 49)) + ((((((int)threadIdx.x) * 2) % 9) / 3) * 7)) + ((((int)blockIdx.x) % 7) * 7)) + ((((int)threadIdx.x) * 6) % 9)) - 8)] : 0.000000e+00f);
+        }
+        if (((int)threadIdx.x) < 36) {
+          pad_temp_shared[((((int)threadIdx.x) * 6) + 1)] = (((((1 <= ((((((int)threadIdx.x) * 2) % 9) / 3) + (((int)blockIdx.x) % 7))) && (((((((int)threadIdx.x) * 2) % 9) / 3) + (((int)blockIdx.x) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 6) + 1) % 9))) && ((((((int)threadIdx.x) * 6) + 1) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + (((((int)threadIdx.x) * 2) / 9) * 49)) + ((((((int)threadIdx.x) * 2) % 9) / 3) * 7)) + ((((int)blockIdx.x) % 7) * 7)) + (((((int)threadIdx.x) * 6) +  [...]
+        }
+        if (((int)threadIdx.x) < 36) {
+          pad_temp_shared[((((int)threadIdx.x) * 6) + 2)] = (((((1 <= ((((((int)threadIdx.x) * 2) % 9) / 3) + (((int)blockIdx.x) % 7))) && (((((((int)threadIdx.x) * 2) % 9) / 3) + (((int)blockIdx.x) % 7)) < 8)) && (1 < (((((int)threadIdx.x) * 6) + 2) % 9))) && ((((((int)threadIdx.x) * 6) + 2) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + (((((int)threadIdx.x) * 2) / 9) * 49)) + ((((((int)threadIdx.x) * 2) % 9) / 3) * 7)) + ((((int)blockIdx.x) % 7) * 7)) + (((((int)threadIdx.x) * 6) + 2 [...]
+        }
+        if (((int)threadIdx.x) < 36) {
+          pad_temp_shared[((((int)threadIdx.x) * 6) + 3)] = (((((1 <= (((((((int)threadIdx.x) * 2) + 1) % 9) / 3) + (((int)blockIdx.x) % 7))) && ((((((((int)threadIdx.x) * 2) + 1) % 9) / 3) + (((int)blockIdx.x) % 7)) < 8)) && (1 < (((((int)threadIdx.x) * 6) + 3) % 9))) && ((((((int)threadIdx.x) * 6) + 3) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + ((((((int)threadIdx.x) * 2) + 1) / 9) * 49)) + (((((((int)threadIdx.x) * 2) + 1) % 9) / 3) * 7)) + ((((int)blockIdx.x) % 7) * 7)) + (((((i [...]
+        }
+        if (((int)threadIdx.x) < 36) {
+          pad_temp_shared[((((int)threadIdx.x) * 6) + 4)] = (((((1 <= (((((((int)threadIdx.x) * 2) + 1) % 9) / 3) + (((int)blockIdx.x) % 7))) && ((((((((int)threadIdx.x) * 2) + 1) % 9) / 3) + (((int)blockIdx.x) % 7)) < 8)) && (1 <= (((((int)threadIdx.x) * 6) + 4) % 9))) && ((((((int)threadIdx.x) * 6) + 4) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + ((((((int)threadIdx.x) * 2) + 1) / 9) * 49)) + (((((((int)threadIdx.x) * 2) + 1) % 9) / 3) * 7)) + ((((int)blockIdx.x) % 7) * 7)) + ((((( [...]
+        }
+        if (((int)threadIdx.x) < 36) {
+          pad_temp_shared[((((int)threadIdx.x) * 6) + 5)] = (((((1 <= (((((((int)threadIdx.x) * 2) + 1) % 9) / 3) + (((int)blockIdx.x) % 7))) && ((((((((int)threadIdx.x) * 2) + 1) % 9) / 3) + (((int)blockIdx.x) % 7)) < 8)) && (1 < (((((int)threadIdx.x) * 6) + 5) % 9))) && ((((((int)threadIdx.x) * 6) + 5) % 9) < 8)) ? data[((((((rc_outer_outer * 392) + ((((((int)threadIdx.x) * 2) + 1) / 9) * 49)) + (((((((int)threadIdx.x) * 2) + 1) % 9) / 3) * 7)) + ((((int)blockIdx.x) % 7) * 7)) + (((((i [...]
+        }
+        kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) % 72))];
+        kernel_shared[(((int)threadIdx.x) + 224)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 224) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+        kernel_shared[(((int)threadIdx.x) + 448)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 448) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+        kernel_shared[(((int)threadIdx.x) + 672)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 672) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) / 3) + 8) % 24) * 3)) + (((int)threadIdx.x) % 3))];
+        kernel_shared[(((int)threadIdx.x) + 896)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 896) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 32) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+        kernel_shared[(((int)threadIdx.x) + 1120)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 1120) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 40) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+        kernel_shared[(((int)threadIdx.x) + 1344)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 1344) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) / 3) + 16) % 24) * 3)) + (((int)threadIdx.x) % 3))];
+        kernel_shared[(((int)threadIdx.x) + 1568)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 1568) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 56) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+        kernel_shared[(((int)threadIdx.x) + 1792)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 1792) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 64) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+        kernel_shared[(((int)threadIdx.x) + 2016)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) % 72)) + 129024)];
+        kernel_shared[(((int)threadIdx.x) + 2240)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 2240) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+        kernel_shared[(((int)threadIdx.x) + 2464)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 2464) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+        kernel_shared[(((int)threadIdx.x) + 2688)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 2688) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) / 3) + 8) % 24) * 3)) + (((int)threadIdx.x) % 3))];
+        kernel_shared[(((int)threadIdx.x) + 2912)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 2912) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 32) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+        kernel_shared[(((int)threadIdx.x) + 3136)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 3136) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 40) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+        kernel_shared[(((int)threadIdx.x) + 3360)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 3360) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) / 3) + 16) % 24) * 3)) + (((int)threadIdx.x) % 3))];
+        kernel_shared[(((int)threadIdx.x) + 3584)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 3584) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 56) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+        kernel_shared[(((int)threadIdx.x) + 3808)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 3808) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 64) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+        kernel_shared[(((int)threadIdx.x) + 4032)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) % 72)) + 258048)];
+        kernel_shared[(((int)threadIdx.x) + 4256)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 4256) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+        if (((int)threadIdx.x) < 128) {
+          kernel_shared[(((int)threadIdx.x) + 4480)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 4480) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+        }
+        __syncthreads();
+        for (int rc_outer_inner = 0; rc_outer_inner < 8; ++rc_outer_inner) {
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((rc_outer_inner * 27) + (((int)threadIdx.x) % 7))] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9))]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((rc_outer_inner * 27) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 2304)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 1)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 2305)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 2)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 2306)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 3)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 2307)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 4)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 2308)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 5)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 2309)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 6)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 2310)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 7)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 2311)]));
+          conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 8)]));
+          conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 2312)]));
         }
       }
-      for (int i3_inner = 0; i3_inner < 7; ++i3_inner) {
-        compute[(((((int)blockIdx.x) * 3136) + (((int)threadIdx.x) * 7)) + i3_inner)] = max((conv2d_nchw[i3_inner] + bias[((((int)blockIdx.x) * 64) + (((int)threadIdx.x) / 7))]), 0.000000e+00f);
-        compute[((((((int)blockIdx.x) * 3136) + (((int)threadIdx.x) * 7)) + i3_inner) + 1568)] = max((conv2d_nchw[(i3_inner + 7)] + bias[(((((int)blockIdx.x) * 64) + (((int)threadIdx.x) / 7)) + 32)]), 0.000000e+00f);
-      }
+      compute[(((((((int)blockIdx.x) / 7) * 3136) + ((((int)threadIdx.x) / 7) * 49)) + ((((int)blockIdx.x) % 7) * 7)) + (((int)threadIdx.x) % 7))] = max((conv2d_nchw[0] + bias[(((((int)blockIdx.x) / 7) * 64) + (((int)threadIdx.x) / 7))]), 0.000000e+00f);
+      compute[((((((((int)blockIdx.x) / 7) * 3136) + ((((int)threadIdx.x) / 7) * 49)) + ((((int)blockIdx.x) % 7) * 7)) + (((int)threadIdx.x) % 7)) + 1568)] = max((conv2d_nchw[1] + bias[((((((int)blockIdx.x) / 7) * 64) + (((int)threadIdx.x) / 7)) + 32)]), 0.000000e+00f);
     }
 
 
@@ -1255,7 +648,7 @@ 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  38.343 seconds)
+   **Total running time of the script:** ( 5 minutes  29.208 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 266fecdc71..968666398d 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_network_cuda.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_network_cuda.rst.txt
@@ -647,7 +647,7 @@ so we can read the log file and load the best schedules.
     Evaluate inference time cost...
     Execution time summary:
      mean (ms)   median (ms)    max (ms)     min (ms)     std (ms)  
-       7.9053       7.9052       7.9161       7.8946       0.0088   
+       7.8731       7.8745       7.8789       7.8659       0.0054   
                
 
 
@@ -675,7 +675,7 @@ Other Tips
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  1.132 seconds)
+   **Total running time of the script:** ( 1 minutes  1.747 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 4877ff63eb..3a9b714c84 100644
--- a/docs/_sources/how_to/tune_with_autoscheduler/tune_network_x86.rst.txt
+++ b/docs/_sources/how_to/tune_with_autoscheduler/tune_network_x86.rst.txt
@@ -666,7 +666,7 @@ so we can read the log file and load the best schedules.
     Evaluate inference time cost...
     Execution time summary:
      mean (ms)   median (ms)    max (ms)     min (ms)     std (ms)  
-      755.9445     755.6418     757.1292     755.0625      0.8704   
+      752.6776     753.1605     753.6184     751.2538      1.0239   
                
 
 
@@ -694,7 +694,7 @@ Other Tips
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  30.980 seconds)
+   **Total running time of the script:** ( 1 minutes  31.583 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 cea4ebf70a..ac45f59d4d 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
@@ -390,121 +390,30 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
                  placeholder_4: Buffer(placeholder_14: Pointer(float32), float32, [128, 512], []),
                  compute: Buffer(compute_2: Pointer(float32), float32, [128, 512], [])}
       buffer_map = {placeholder_5: placeholder, placeholder_6: placeholder_1, placeholder_7: placeholder_2, placeholder_8: placeholder_3, placeholder_9: placeholder_4, compute_1: compute} {
-      for (i0.outer.i1.outer.fused: int32, 0, 64) "parallel" {
-        allocate(compute_3: Pointer(global float32), float32, [1024]), storage_scope = global {
-          for (i.outer.inner: int32, 0, 16) {
-            for (nb_j.inner: int32, 0, 2) {
-              let cse_var_2: int32 = ((i.outer.inner*64) + (nb_j.inner*16))
-              let cse_var_1: int32 = ((floormod(i0.outer.i1.outer.fused, 16)*2) + nb_j.inner)
-               {
-                compute_4: Buffer(compute_3, float32, [1024], [])[cse_var_2] = 0f32
-                compute_4[(cse_var_2 + 1)] = 0f32
-                compute_4[(cse_var_2 + 2)] = 0f32
-                compute_4[(cse_var_2 + 3)] = 0f32
-                compute_4[(cse_var_2 + 4)] = 0f32
-                compute_4[(cse_var_2 + 5)] = 0f32
-                compute_4[(cse_var_2 + 6)] = 0f32
-                compute_4[(cse_var_2 + 7)] = 0f32
-                compute_4[(cse_var_2 + 8)] = 0f32
-                compute_4[(cse_var_2 + 9)] = 0f32
-                compute_4[(cse_var_2 + 10)] = 0f32
-                compute_4[(cse_var_2 + 11)] = 0f32
-                compute_4[(cse_var_2 + 12)] = 0f32
-                compute_4[(cse_var_2 + 13)] = 0f32
-                compute_4[(cse_var_2 + 14)] = 0f32
-                compute_4[(cse_var_2 + 15)] = 0f32
-                compute_4[(cse_var_2 + 32)] = 0f32
-                compute_4[(cse_var_2 + 33)] = 0f32
-                compute_4[(cse_var_2 + 34)] = 0f32
-                compute_4[(cse_var_2 + 35)] = 0f32
-                compute_4[(cse_var_2 + 36)] = 0f32
-                compute_4[(cse_var_2 + 37)] = 0f32
-                compute_4[(cse_var_2 + 38)] = 0f32
-                compute_4[(cse_var_2 + 39)] = 0f32
-                compute_4[(cse_var_2 + 40)] = 0f32
-                compute_4[(cse_var_2 + 41)] = 0f32
-                compute_4[(cse_var_2 + 42)] = 0f32
-                compute_4[(cse_var_2 + 43)] = 0f32
-                compute_4[(cse_var_2 + 44)] = 0f32
-                compute_4[(cse_var_2 + 45)] = 0f32
-                compute_4[(cse_var_2 + 46)] = 0f32
-                compute_4[(cse_var_2 + 47)] = 0f32
-                for (elem_idx: int32, 0, (placeholder_15: Buffer(placeholder_13, int32, [33], [])[(cse_var_1 + 1)] - placeholder_15[cse_var_1])) {
-                  let cse_var_35: int32 = (elem_idx*16)
-                  let cse_var_34: int32 = (cse_var_2 + 9)
-                  let cse_var_33: int32 = (cse_var_2 + 8)
-                  let cse_var_32: int32 = (cse_var_2 + 7)
-                  let cse_var_31: int32 = (cse_var_2 + 6)
-                  let cse_var_30: int32 = (cse_var_2 + 5)
-                  let cse_var_29: int32 = (cse_var_2 + 47)
-                  let cse_var_28: int32 = (cse_var_2 + 46)
-                  let cse_var_27: int32 = (cse_var_2 + 45)
-                  let cse_var_26: int32 = (cse_var_2 + 44)
-                  let cse_var_25: int32 = (cse_var_2 + 43)
-                  let cse_var_24: int32 = (cse_var_2 + 42)
-                  let cse_var_23: int32 = (cse_var_2 + 41)
-                  let cse_var_22: int32 = (cse_var_2 + 40)
-                  let cse_var_21: int32 = (cse_var_2 + 4)
-                  let cse_var_20: int32 = (cse_var_2 + 39)
-                  let cse_var_19: int32 = (cse_var_2 + 38)
-                  let cse_var_18: int32 = (cse_var_2 + 37)
-                  let cse_var_17: int32 = (cse_var_2 + 36)
-                  let cse_var_16: int32 = (cse_var_2 + 35)
-                  let cse_var_15: int32 = (cse_var_2 + 34)
-                  let cse_var_14: int32 = (cse_var_2 + 33)
-                  let cse_var_13: int32 = (cse_var_2 + 32)
-                  let cse_var_12: int32 = (cse_var_2 + 3)
-                  let cse_var_11: int32 = (cse_var_2 + 2)
-                  let cse_var_10: int32 = (cse_var_2 + 15)
-                  let cse_var_9: int32 = (cse_var_2 + 14)
-                  let cse_var_8: int32 = (cse_var_2 + 13)
-                  let cse_var_7: int32 = (cse_var_2 + 12)
-                  let cse_var_6: int32 = (cse_var_2 + 11)
-                  let cse_var_5: int32 = (cse_var_2 + 10)
-                  let cse_var_4: int32 = (cse_var_2 + 1)
-                  let cse_var_3: int32 = ((floordiv(i0.outer.i1.outer.fused, 16)*8192) + (i.outer.inner*512))
-                   {
-                    compute_4[cse_var_2] = (compute_4[cse_var_2] + (placeholder_16: Buffer(placeholder_11, float32, [78656], [])[((placeholder_15[cse_var_1]*16) + cse_var_35)]*max(placeholder_17: Buffer(placeholder_10, float32, [32768], [])[(cse_var_3 + placeholder_18: Buffer(placeholder_12, int32, [4916], [])[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                    compute_4[cse_var_4] = (compute_4[cse_var_4] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 1)]*max(placeholder_17[(cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                    compute_4[cse_var_11] = (compute_4[cse_var_11] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 2)]*max(placeholder_17[(cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                    compute_4[cse_var_12] = (compute_4[cse_var_12] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 3)]*max(placeholder_17[(cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                    compute_4[cse_var_21] = (compute_4[cse_var_21] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 4)]*max(placeholder_17[(cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                    compute_4[cse_var_30] = (compute_4[cse_var_30] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 5)]*max(placeholder_17[(cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                    compute_4[cse_var_31] = (compute_4[cse_var_31] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 6)]*max(placeholder_17[(cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                    compute_4[cse_var_32] = (compute_4[cse_var_32] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 7)]*max(placeholder_17[(cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                    compute_4[cse_var_33] = (compute_4[cse_var_33] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 8)]*max(placeholder_17[(cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                    compute_4[cse_var_34] = (compute_4[cse_var_34] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 9)]*max(placeholder_17[(cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                    compute_4[cse_var_5] = (compute_4[cse_var_5] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 10)]*max(placeholder_17[(cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                    compute_4[cse_var_6] = (compute_4[cse_var_6] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 11)]*max(placeholder_17[(cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                    compute_4[cse_var_7] = (compute_4[cse_var_7] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 12)]*max(placeholder_17[(cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                    compute_4[cse_var_8] = (compute_4[cse_var_8] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 13)]*max(placeholder_17[(cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                    compute_4[cse_var_9] = (compute_4[cse_var_9] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 14)]*max(placeholder_17[(cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                    compute_4[cse_var_10] = (compute_4[cse_var_10] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 15)]*max(placeholder_17[(cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                    compute_4[cse_var_13] = (compute_4[cse_var_13] + (placeholder_16[((placeholder_15[cse_var_1]*16) + cse_var_35)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
-                    compute_4[cse_var_14] = (compute_4[cse_var_14] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 1)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
-                    compute_4[cse_var_15] = (compute_4[cse_var_15] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 2)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
-                    compute_4[cse_var_16] = (compute_4[cse_var_16] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 3)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
-                    compute_4[cse_var_17] = (compute_4[cse_var_17] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 4)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
-                    compute_4[cse_var_18] = (compute_4[cse_var_18] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 5)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
-                    compute_4[cse_var_19] = (compute_4[cse_var_19] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 6)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
-                    compute_4[cse_var_20] = (compute_4[cse_var_20] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 7)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
-                    compute_4[cse_var_22] = (compute_4[cse_var_22] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 8)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
-                    compute_4[cse_var_23] = (compute_4[cse_var_23] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 9)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
-                    compute_4[cse_var_24] = (compute_4[cse_var_24] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 10)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
-                    compute_4[cse_var_25] = (compute_4[cse_var_25] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 11)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
-                    compute_4[cse_var_26] = (compute_4[cse_var_26] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 12)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
-                    compute_4[cse_var_27] = (compute_4[cse_var_27] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 13)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
-                    compute_4[cse_var_28] = (compute_4[cse_var_28] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 14)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
-                    compute_4[cse_var_29] = (compute_4[cse_var_29] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 15)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+      for (i0.outer.i1.outer.fused: int32, 0, 128) "parallel" {
+        allocate(compute_3: Pointer(global float32), float32, [512]), storage_scope = global {
+          for (i.outer.inner: int32, 0, 2) {
+            for (i.inner.init: int32, 0, 16) {
+              for (j.init: int32, 0, 16) {
+                compute_4: Buffer(compute_3, float32, [512], [])[(((i.outer.inner*256) + (i.inner.init*16)) + j.init)] = 0f32
+              }
+            }
+            for (elem_idx: int32, 0, let cse_var_1: int32 = floormod(i0.outer.i1.outer.fused, 32) in (placeholder_15: Buffer(placeholder_13, int32, [33], [])[(cse_var_1 + 1)] - placeholder_15[cse_var_1])) {
+              for (i.inner: int32, 0, 16) {
+                for (j: int32, 0, 16) {
+                  let cse_var_2: int32 = floormod(i0.outer.i1.outer.fused, 32)
+                  if @tir.likely((elem_idx < (placeholder_15[(cse_var_2 + 1)] - placeholder_15[cse_var_2])), dtype=bool) {
+                    let cse_var_3: int32 = (((i.outer.inner*256) + (i.inner*16)) + j)
+                    compute_4[cse_var_3] = (compute_4[cse_var_3] + (placeholder_16: Buffer(placeholder_11, float32, [78656], [])[(((placeholder_15[cse_var_2]*16) + (elem_idx*16)) + j)]*max(placeholder_17: Buffer(placeholder_10, float32, [32768], [])[((((floordiv(i0.outer.i1.outer.fused, 32)*8192) + (i.outer.inner*4096)) + (i.inner*256)) + placeholder_18: Buffer(placeholder_12, int32, [4916], [])[(placeholder_15[cse_var_2] + elem_idx)])], 0f32)))
                   }
                 }
               }
             }
           }
           for (i0.inner: int32, 0, 32) {
-            for (i1.inner: int32, 0, 32) {
-              let cse_var_36: int32 = ((((floordiv(i0.outer.i1.outer.fused, 16)*16384) + (i0.inner*512)) + (floormod(i0.outer.i1.outer.fused, 16)*32)) + i1.inner)
-              compute_5: Buffer(compute_2, float32, [65536], [])[cse_var_36] = max((compute_4[((i0.inner*32) + i1.inner)] + placeholder_19: Buffer(placeholder_14, float32, [65536], [])[cse_var_36]), 0f32)
+            for (i1.inner: int32, 0, 16) {
+              let cse_var_4: int32 = ((((floordiv(i0.outer.i1.outer.fused, 32)*16384) + (i0.inner*512)) + (floormod(i0.outer.i1.outer.fused, 32)*16)) + i1.inner)
+              compute_5: Buffer(compute_2, float32, [65536], [])[cse_var_4] = max((compute_4[((i0.inner*16) + i1.inner)] + placeholder_19: Buffer(placeholder_14, float32, [65536], [])[cse_var_4]), 0f32)
             }
           }
         }
@@ -561,7 +470,7 @@ We build the binary and check its correctness and performance.
 
  .. code-block:: none
 
-    Execution time of this operator: 3.507 ms
+    Execution time of this operator: 1.669 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 4936fc1f0c..76b4446391 100644
--- a/docs/_sources/how_to/tune_with_autotvm/sg_execution_times.rst.txt
+++ b/docs/_sources/how_to/tune_with_autotvm/sg_execution_times.rst.txt
@@ -5,16 +5,16 @@
 
 Computation times
 =================
-**00:33.745** total execution time for **how_to_tune_with_autotvm** files:
+**00:43.740** 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``)           | 00:33.711 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_conv2d_cuda.py` (``tune_conv2d_cuda.py``)           | 00:43.708 | 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``)               | 00:00.019 | 0.0 MB |
 +--------------------------------------------------------------------------------------------------+-----------+--------+
 | :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_cuda.py` (``tune_relay_cuda.py``)             | 00:00.005 | 0.0 MB |
 +--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_arm.py` (``tune_relay_arm.py``)               | 00:00.004 | 0.0 MB |
+| :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_arm.py` (``tune_relay_arm.py``)               | 00:00.005 | 0.0 MB |
 +--------------------------------------------------------------------------------------------------+-----------+--------+
 | :ref:`sphx_glr_how_to_tune_with_autotvm_tune_relay_mobile_gpu.py` (``tune_relay_mobile_gpu.py``) | 00:00.004 | 0.0 MB |
 +--------------------------------------------------------------------------------------------------+-----------+--------+
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 00a36d356e..b8e58934ec 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
@@ -390,7 +390,7 @@ for this template
       File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
-    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 2, 8, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 32, 16]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,3818882
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 2, 1, 256]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 1, 8]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7841897
     No: 2   GFLOPS: 0.00/0.00       result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
@@ -513,9 +513,8 @@ for this template
       File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
-    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 4, 2, 8]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 32]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9821605
-    No: 3   GFLOPS: 36.11/36.11     result: MeasureResult(costs=(0.006411471315789473,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.768928050994873, timestamp=1673555737.4988983)        [('tile_f', [-1, 1, 1, 2]), ('tile_y', [-1, 7, 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', 0), ('unroll_explicit', 0)],None,77715
-    No: 4   GFLOPS: 0.00/36.11      result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 2, 256, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 8, 2]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,1788653
+    No: 3   GFLOPS: 0.00/0.00       result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -637,8 +636,9 @@ for this template
       File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
-    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 64, 4, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 512, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4291838
-    No: 5   GFLOPS: 0.00/36.11      result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 8, 2, 8]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 512]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8128046
+    No: 4   GFLOPS: 825.44/825.44   result: MeasureResult(costs=(0.00028045692893401017,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.2746119499206543, timestamp=1673569917.1928902)     [('tile_f', [-1, 1, 16, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 8, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,8919274
+    No: 5   GFLOPS: 0.00/825.44     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -760,8 +760,8 @@ for this template
       File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
-    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 8, 1, 64]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 8, 2]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1207783
-    No: 6   GFLOPS: 0.00/36.11      result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 4, 4, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2835951
+    No: 6   GFLOPS: 0.00/825.44     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -883,8 +883,8 @@ for this template
       File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
-    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 1, 32, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 256]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7156344
-    No: 7   GFLOPS: 0.00/36.11      result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 64, 1, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 16, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2920386
+    No: 7   GFLOPS: 0.00/825.44     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1006,11 +1006,8 @@ for this template
       File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
-    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 4, 4, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 2, 128]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,3081657
-    No: 8   GFLOPS: 443.05/443.05   result: MeasureResult(costs=(0.0005225226458333333,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.2629878520965576, timestamp=1673555740.8875287)      [('tile_f', [-1, 2, 8, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2983720
-    No: 9   GFLOPS: 1.01/443.05     result: MeasureResult(costs=(0.22923324575,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.9031102657318115, timestamp=1673555745.067556)       [('tile_f', [-1, 1, 2, 256]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,1359598
-    No: 10  GFLOPS: 366.19/443.05   result: MeasureResult(costs=(0.0006321807421383648,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.3297412395477295, timestamp=1673555745.7973425)      [('tile_f', [-1, 2, 8, 2]), ('tile_y', [-1, 1, 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,8562480
-    No: 11  GFLOPS: 0.00/443.05     result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 8, 16, 2]), ('tile_y', [-1, 7, 1, 1]), ('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', 0), ('unroll_explicit', 0)],None,1149588
+    No: 8   GFLOPS: 0.00/825.44     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1132,8 +1129,8 @@ for this template
       File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
-    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 16, 1, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 256]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,3091004
-    No: 12  GFLOPS: 0.00/443.05     result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 4, 8, 2]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 16, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9236341
+    No: 9   GFLOPS: 0.00/825.44     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1255,10 +1252,10 @@ 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 875, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
-    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 4, 1, 128]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 64, 8]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9022412
-    No: 13  GFLOPS: 323.70/443.05   result: MeasureResult(costs=(0.0007151712485875707,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.3592932224273682, timestamp=1673555748.9529142)      [('tile_f', [-1, 1, 4, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 1, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7393832
-    No: 14  GFLOPS: 0.83/443.05     result: MeasureResult(costs=(0.279625122,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.254995346069336, timestamp=1673555752.9961503) [('tile_f', [-1, 1, 4, 128]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 2, 2]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2363235
-    No: 15  GFLOPS: 0.00/443.05     result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 32, 4, 2]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 512, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5260717
+    No: 10  GFLOPS: 72.60/825.44    result: MeasureResult(costs=(0.003188784342857143,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.266326665878296, timestamp=1673569922.8719883)        [('tile_f', [-1, 2, 2, 4]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7627509
+    No: 11  GFLOPS: 369.71/825.44   result: MeasureResult(costs=(0.0006261778949416343,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.3138256072998047, timestamp=1673569923.858065)       [('tile_f', [-1, 8, 2, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9721153
+    No: 12  GFLOPS: 0.00/825.44     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1380,132 +1377,162 @@ for this template
       File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
-    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 2, 2, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 32, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,139545
-    No: 16  GFLOPS: 11.83/443.05    result: MeasureResult(costs=(0.019574101666666666,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.9008331298828125, timestamp=1673555753.8789542)       [('tile_f', [-1, 2, 2, 32]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 4]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,5106391
-    No: 17  GFLOPS: 0.00/443.05     result: Traceback (most recent call last):
-      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
-        func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
-      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
-        func = build(s, args, target_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)
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 64, 4, 2]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 8, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,400698
+    No: 13  GFLOPS: 0.00/825.44     result: Traceback (most recent call last):
+      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 742, in __call__
+        yield remote, remote.load_module(os.path.split(build_result.filename)[1])
+      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 706, in run_through_rpc
+        costs = time_f(*args).results
+      File "/workspace/python/tvm/runtime/module.py", line 357, in evaluator
+        blob = feval(*args)
       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/./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/./base.pxi", line 181, in tvm._ffi._cy3.core.CHECK_CALL
     tvm._ffi.base.TVMError: Traceback (most recent call last):
-      24: TVMFuncCall
+      4: 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:1730
-      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:1670
-      19: run<>
-            at ../include/tvm/runtime/packed_func.h:1630
-      18: run<tvm::runtime::TVMMovableArgValueWithContext_>
-            at ../include/tvm/runtime/packed_func.h:1630
-      17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
-            at ../include/tvm/runtime/packed_func.h:1630
-      16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
-            at ../include/tvm/runtime/packed_func.h:1630
-      15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
-            at ../include/tvm/runtime/packed_func.h:1630
-      14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
-            at ../include/tvm/runtime/packed_func.h:1645
-      13: operator()
-            at ../src/driver/driver_api.cc:395
-      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:381
-      11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
-            at ../src/driver/driver_api.cc:276
-      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:454
-      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:1749
-      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:1693
-      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:1617
-      2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+      3: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
             at ../include/tvm/runtime/packed_func.h:1217
-      1: Call
-            at ../include/tvm/runtime/packed_func.h:1213
-      0: operator()
-            at ../src/runtime/c_runtime_api.cc:534
-      File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
-      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
-        raise InstantiationError("Skipped because of invalid gpu kernel")
-    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel
+      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):
-      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:1730
-      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:1670
-      19: run<>
-            at ../include/tvm/runtime/packed_func.h:1630
-      18: run<tvm::runtime::TVMMovableArgValueWithContext_>
-            at ../include/tvm/runtime/packed_func.h:1630
-      17: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
-            at ../include/tvm/runtime/packed_func.h:1630
-      16: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
-            at ../include/tvm/runtime/packed_func.h:1630
-      15: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
-            at ../include/tvm/runtime/packed_func.h:1630
-      14: run<tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_>
-            at ../include/tvm/runtime/packed_func.h:1645
-      13: operator()
-            at ../src/driver/driver_api.cc:395
-      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:381
-      11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array<tvm::transform::Pass, void>)
-            at ../src/driver/driver_api.cc:276
-      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:454
-      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:1749
-      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:1693
-      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
+      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 706, 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 746, 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: 0x00007f839f345fa2
+      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
             at ../include/tvm/runtime/packed_func.h:1617
       2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
             at ../include/tvm/runtime/packed_func.h:1217
       1: Call
             at ../include/tvm/runtime/packed_func.h:1213
       0: operator()
-            at ../src/runtime/c_runtime_api.cc:534
-      File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
-      File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
-        raise InstantiationError("Skipped because of invalid gpu kernel")
-    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 2, 16, 4]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 128, 2]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8194467
-    No: 18  GFLOPS: 0.00/443.05     result: Traceback (most recent call last):
+            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
+
+    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, 8, 2, 8]), ('tile_y', [-1, 7, 1, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 4, 16]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,5356926
+    No: 14  GFLOPS: 40.92/825.44    result: MeasureResult(costs=(0.005657900444444444,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.81038236618042, timestamp=1673569933.718674)  [('tile_f', [-1, 1, 4, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 16, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7053879
+    No: 15  GFLOPS: 0.00/825.44     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1627,8 +1654,10 @@ 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 875, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
-    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 32, 1, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 7, 1, 1]), ('tile_rc', [-1, 64, 1]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,7960265
-    No: 19  GFLOPS: 0.00/443.05     result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 8, 8, 8]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 1, 4]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 1)],None,6069077
+    No: 16  GFLOPS: 14.39/825.44    result: MeasureResult(costs=(0.01609306,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.153165578842163, timestamp=1673569934.541408)   [('tile_f', [-1, 2, 2, 1]), ('tile_y', [-1, 1, 1, 7]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 2, 16]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 1)],None,8642271
+    No: 17  GFLOPS: 78.30/825.44    result: MeasureResult(costs=(0.002956405617647059,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.077221393585205, timestamp=1673569939.8245046)        [('tile_f', [-1, 1, 2, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 64, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 0)],None,4282530
+    No: 18  GFLOPS: 0.00/825.44     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1750,8 +1779,9 @@ for this template
       File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
-    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 32, 2, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 128, 2]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 3, 1]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2773335
-    No: 20  GFLOPS: 0.00/443.05     result: Traceback (most recent call last):
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 4, 16, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 32, 8]), ('tile_ry', [-1, 1, 3]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,10375676
+    No: 19  GFLOPS: 144.58/825.44   result: MeasureResult(costs=(0.0016012034600000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.168120622634888, timestamp=1673569940.8427498)       [('tile_f', [-1, 4, 1, 1]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 64, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,8928482
+    No: 20  GFLOPS: 0.00/825.44     result: Traceback (most recent call last):
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 592, in __call__
         func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 544, in _build_func_common
@@ -1873,7 +1903,7 @@ for this template
       File "tvm/_ffi/_cython/./packed_func.pxi", line 56, in tvm._ffi._cy3.core.tvm_callback
       File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 875, in verify_pass
         raise InstantiationError("Skipped because of invalid gpu kernel")
-    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 128, 2, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 2, 64]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 0), ('unroll_explicit', 0)],None,357737
+    tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [('tile_f', [-1, 8, 1, 16]), ('tile_y', [-1, 1, 1, 1]), ('tile_x', [-1, 1, 1, 1]), ('tile_rc', [-1, 512, 1]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,9905447
 
 
 
@@ -1928,9 +1958,9 @@ and measure running time.
     Finish loading 20 records
 
     Best config:
-    [('tile_f', [-1, 2, 8, 2]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 7, 1]), ('tile_rc', [-1, 8, 4]), ('tile_ry', [-1, 1, 1]), ('tile_rx', [-1, 1, 3]), ('auto_unroll_max_step', 512), ('unroll_explicit', 0)],None,2983720
+    [('tile_f', [-1, 1, 16, 1]), ('tile_y', [-1, 1, 7, 1]), ('tile_x', [-1, 1, 1, 7]), ('tile_rc', [-1, 8, 1]), ('tile_ry', [-1, 3, 1]), ('tile_rx', [-1, 1, 1]), ('auto_unroll_max_step', 1500), ('unroll_explicit', 1)],None,8919274
     Finish loading 20 records
-    Time cost of this operator: 0.000654
+    Time cost of this operator: 0.000570
 
 
 
diff --git a/docs/_sources/how_to/work_with_microtvm/micro_aot.rst.txt b/docs/_sources/how_to/work_with_microtvm/micro_aot.rst.txt
index 180afb8e9a..f6cff8f0dc 100644
--- a/docs/_sources/how_to/work_with_microtvm/micro_aot.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/micro_aot.rst.txt
@@ -52,7 +52,7 @@ or on Zephyr platform on a microcontroller/board supported by Zephyr.
     # By default, this tutorial runs on x86 CPU using TVM's C runtime. If you would like
     # to run on real Zephyr hardware, you must export the `TVM_MICRO_USE_HW` environment
     # variable. Otherwise (if you are using the C runtime), you can skip installing
-    # Zephyr and CMSIS-NN. It takes ~20 minutes to install both of them.
+    # Zephyr. It takes ~20 minutes to install Zephyr.
     use_physical_hw = bool(os.getenv("TVM_MICRO_USE_HW"))
 
 
@@ -67,18 +67,13 @@ or on Zephyr platform on a microcontroller/board supported by Zephyr.
 .. include:: ../../../../gallery/how_to/work_with_microtvm/install_zephyr.rst
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 53-55
-
-.. include:: ../../../../gallery/how_to/work_with_microtvm/install_cmsis.rst
-
-
-.. GENERATED FROM PYTHON SOURCE LINES 58-61
+.. GENERATED FROM PYTHON SOURCE LINES 53-56
 
 Import Python dependencies
 -------------------------------
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 61-70
+.. GENERATED FROM PYTHON SOURCE LINES 56-65
 
 .. code-block:: default
 
@@ -98,7 +93,7 @@ Import Python dependencies
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 71-81
+.. GENERATED FROM PYTHON SOURCE LINES 66-76
 
 Import a TFLite model
 ---------------------
@@ -111,7 +106,7 @@ To test this model, we use samples from `KWS dataset provided by Google <https:/
 you need to export `TVM_MICRO_USE_HW` environment variable.
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 81-102
+.. GENERATED FROM PYTHON SOURCE LINES 76-97
 
 .. code-block:: default
 
@@ -143,7 +138,7 @@ you need to export `TVM_MICRO_USE_HW` environment variable.
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 103-113
+.. GENERATED FROM PYTHON SOURCE LINES 98-108
 
 Defining the target
 -------------------
@@ -156,7 +151,7 @@ board (E.g. nucleo_l4r5zi) and pass it to `tvm.target.target.micro` to create a
 micro target.
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 113-132
+.. GENERATED FROM PYTHON SOURCE LINES 108-127
 
 .. code-block:: default
 
@@ -186,7 +181,7 @@ micro target.
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 133-138
+.. GENERATED FROM PYTHON SOURCE LINES 128-133
 
 Compile the model
 -----------------
@@ -194,7 +189,7 @@ Compile the model
 Now, we compile the model for the target:
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 138-143
+.. GENERATED FROM PYTHON SOURCE LINES 133-138
 
 .. code-block:: default
 
@@ -210,7 +205,7 @@ Now, we compile the model for the target:
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 144-152
+.. GENERATED FROM PYTHON SOURCE LINES 139-147
 
 Create a microTVM project
 -------------------------
@@ -221,7 +216,7 @@ CRT and Zephyr microTVM template projects which are used for x86 CPU and Zephyr
 respectively.
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 152-172
+.. GENERATED FROM PYTHON SOURCE LINES 147-166
 
 .. code-block:: default
 
@@ -235,7 +230,6 @@ respectively.
             "board": BOARD,
             "serial_number": SERIAL,
             "config_main_stack_size": 4096,
-            "cmsis_path": os.getenv("CMSIS_PATH", default="/content/cmsis"),
             "zephyr_base": os.getenv("ZEPHYR_BASE", default="/content/zephyrproject/zephyr"),
         }
 
@@ -252,7 +246,7 @@ respectively.
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 173-181
+.. GENERATED FROM PYTHON SOURCE LINES 167-175
 
 Build, flash and execute the model
 ----------------------------------
@@ -263,7 +257,7 @@ Next, we define the labels for the model output and execute the model with a
 sample with expected value of 6 (label: left).
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 181-205
+.. GENERATED FROM PYTHON SOURCE LINES 175-199
 
 .. code-block:: default
 
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 adc22f63b0..5c9f0d2448 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
@@ -43,7 +43,7 @@ This tutorial explains how to autotune a model using the C runtime.
 
 
 
-    # You can skip the following two sections (installing Zephyr and CMSIS-NN) if the following flag is False.
+    # You can skip the following section (installing Zephyr) if the following flag is False.
     # Installing Zephyr takes ~20 min.
     import os
 
@@ -61,18 +61,13 @@ This tutorial explains how to autotune a model using the C runtime.
 .. include:: ../../../../gallery/how_to/work_with_microtvm/install_zephyr.rst
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 48-50
-
-.. include:: ../../../../gallery/how_to/work_with_microtvm/install_cmsis.rst
-
-
-.. GENERATED FROM PYTHON SOURCE LINES 53-56
+.. GENERATED FROM PYTHON SOURCE LINES 49-52
 
 Import Python dependencies
 -------------------------------
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 56-63
+.. GENERATED FROM PYTHON SOURCE LINES 52-59
 
 .. code-block:: default
 
@@ -90,7 +85,7 @@ Import Python dependencies
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 64-70
+.. GENERATED FROM PYTHON SOURCE LINES 60-66
 
 Defining the model
 ###################
@@ -99,7 +94,7 @@ Defining the model
  fill parameters with random numbers.
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 70-95
+.. GENERATED FROM PYTHON SOURCE LINES 66-91
 
 .. code-block:: default
 
@@ -135,7 +130,7 @@ Defining the model
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 96-107
+.. GENERATED FROM PYTHON SOURCE LINES 92-103
 
 Defining the target
 ######################
@@ -149,7 +144,7 @@ Defining the target
  this tutorial.
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 107-125
+.. GENERATED FROM PYTHON SOURCE LINES 103-121
 
 .. code-block:: default
 
@@ -178,7 +173,7 @@ Defining the target
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 126-135
+.. GENERATED FROM PYTHON SOURCE LINES 122-131
 
 Extracting tuning tasks
 ########################
@@ -190,7 +185,7 @@ Extracting tuning tasks
  transformation passes; we'll apply the same configuration later on during autotuning.
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 135-141
+.. GENERATED FROM PYTHON SOURCE LINES 131-137
 
 .. code-block:: default
 
@@ -207,7 +202,7 @@ Extracting tuning tasks
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 142-152
+.. GENERATED FROM PYTHON SOURCE LINES 138-148
 
 Configuring microTVM
 #####################
@@ -220,7 +215,7 @@ Configuring microTVM
  choose other options by choosing from `PLATFORM` list.
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 152-190
+.. GENERATED FROM PYTHON SOURCE LINES 148-186
 
 .. code-block:: default
 
@@ -269,14 +264,14 @@ Configuring microTVM
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 191-195
+.. GENERATED FROM PYTHON SOURCE LINES 187-191
 
 Run Autotuning
 #########################
  Now we can run autotuning separately on each extracted task on microTVM device.
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 195-213
+.. GENERATED FROM PYTHON SOURCE LINES 191-209
 
 .. code-block:: default
 
@@ -305,7 +300,7 @@ Run Autotuning
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 214-220
+.. GENERATED FROM PYTHON SOURCE LINES 210-216
 
 Timing the untuned program
 ###########################
@@ -314,7 +309,7 @@ Timing the untuned program
  the tuned operator.
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 220-259
+.. GENERATED FROM PYTHON SOURCE LINES 216-255
 
 .. code-block:: default
 
@@ -368,21 +363,21 @@ 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  310.8     98.721   (1, 2, 10, 10, 3)  2       1        [310.8]           
-    tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       3.045     0.967    (1, 6, 10, 10)     1       1        [3.045]           
-    tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.981     0.312    (1, 1, 10, 10, 3)  1       1        [0.981]           
-    Total_time                                    -                                             314.826   -        -                  -       -        -                 
+    tvmgen_default_fused_nn_contrib_conv2d_NCHWc  tvmgen_default_fused_nn_contrib_conv2d_NCHWc  311.0     98.714   (1, 2, 10, 10, 3)  2       1        [311.0]           
+    tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       3.067     0.974    (1, 6, 10, 10)     1       1        [3.067]           
+    tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.984     0.312    (1, 1, 10, 10, 3)  1       1        [0.984]           
+    Total_time                                    -                                             315.051   -        -                  -       -        -                 
 
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 260-263
+.. GENERATED FROM PYTHON SOURCE LINES 256-259
 
 Timing the tuned program
 #########################
  Once autotuning completes, you can time execution of the entire program using the Debug Runtime:
 
-.. GENERATED FROM PYTHON SOURCE LINES 263-302
+.. GENERATED FROM PYTHON SOURCE LINES 259-298
 
 .. code-block:: default
 
@@ -436,10 +431,10 @@ 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  134.2     97.923   (1, 6, 10, 10, 1)  2       1        [134.2]           
-    tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       1.86      1.357    (1, 6, 10, 10)     1       1        [1.86]            
-    tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.986     0.72     (1, 1, 10, 10, 3)  1       1        [0.986]           
-    Total_time                                    -                                             137.046   -        -                  -       -        -                 
+    tvmgen_default_fused_nn_contrib_conv2d_NCHWc  tvmgen_default_fused_nn_contrib_conv2d_NCHWc  100.1     97.309   (1, 6, 10, 10, 1)  2       1        [100.1]           
+    tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       1.794     1.744    (1, 6, 10, 10)     1       1        [1.794]           
+    tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.975     0.948    (1, 1, 10, 10, 3)  1       1        [0.975]           
+    Total_time                                    -                                             102.869   -        -                  -       -        -                 
 
 
 
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 7bd5c66999..635985fac5 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
@@ -117,7 +117,7 @@ download a cat image and preprocess it to use as the model input.
     /venv/apache-tvm-py3.7/lib/python3.7/site-packages/torch/ao/quantization/utils.py:281: UserWarning: must run observer before calling calculate_qparams. Returning default values.
       "must run observer before calling calculate_qparams. " +
     Downloading: "https://download.pytorch.org/models/quantized/mobilenet_v2_qnnpack_37f702c5.pth" to /workspace/.cache/torch/hub/checkpoints/mobilenet_v2_qnnpack_37f702c5.pth
-
      0%|          | 0.00/3.42M [00:00<?, ?B/s]
     61%|######    | 2.09M/3.42M [00:00<00:00, 16.6MB/s]
    100%|##########| 3.42M/3.42M [00:00<00:00, 26.1MB/s]
+
      0%|          | 0.00/3.42M [00:00<?, ?B/s]
     61%|######    | 2.09M/3.42M [00:00<00:00, 20.7MB/s]
    100%|##########| 3.42M/3.42M [00:00<00:00, 32.5MB/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.
@@ -322,7 +322,7 @@ Look up prediction top 1 index in 1000 class synset.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  2.083 seconds)
+   **Total running time of the script:** ( 1 minutes  3.519 seconds)
 
 
 .. _sphx_glr_download_how_to_work_with_microtvm_micro_pytorch.py:
diff --git a/docs/_sources/how_to/work_with_microtvm/micro_tflite.rst.txt b/docs/_sources/how_to/work_with_microtvm/micro_tflite.rst.txt
index dac4528729..2637830335 100644
--- a/docs/_sources/how_to/work_with_microtvm/micro_tflite.rst.txt
+++ b/docs/_sources/how_to/work_with_microtvm/micro_tflite.rst.txt
@@ -47,7 +47,7 @@ model with Relay.
     # By default, this tutorial runs on x86 CPU using TVM's C runtime. If you would like
     # to run on real Zephyr hardware, you must export the `TVM_MICRO_USE_HW` environment
     # variable. Otherwise (if you are using the C runtime), you can skip installing
-    # Zephyr and CMSIS-NN. It takes ~20 minutes to install both of them.
+    # Zephyr. It takes ~20 minutes to install Zephyr.
     use_physical_hw = bool(os.getenv("TVM_MICRO_USE_HW"))
 
 
@@ -62,18 +62,13 @@ model with Relay.
 .. include:: ../../../../gallery/how_to/work_with_microtvm/install_zephyr.rst
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 48-50
-
-.. include:: ../../../../gallery/how_to/work_with_microtvm/install_cmsis.rst
-
-
-.. GENERATED FROM PYTHON SOURCE LINES 53-56
+.. GENERATED FROM PYTHON SOURCE LINES 48-51
 
 Import Python dependencies
 -------------------------------
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 56-73
+.. GENERATED FROM PYTHON SOURCE LINES 51-68
 
 .. code-block:: default
 
@@ -101,11 +96,11 @@ Import Python dependencies
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 74-75
+.. GENERATED FROM PYTHON SOURCE LINES 69-70
 
 Using the buffer, transform into a tflite model python object
 
-.. GENERATED FROM PYTHON SOURCE LINES 75-84
+.. GENERATED FROM PYTHON SOURCE LINES 70-79
 
 .. code-block:: default
 
@@ -125,11 +120,11 @@ Using the buffer, transform into a tflite model python object
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 85-86
+.. GENERATED FROM PYTHON SOURCE LINES 80-81
 
 Print out the version of the model
 
-.. GENERATED FROM PYTHON SOURCE LINES 86-89
+.. GENERATED FROM PYTHON SOURCE LINES 81-84
 
 .. code-block:: default
 
@@ -149,7 +144,7 @@ Print out the version of the model
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 90-98
+.. GENERATED FROM PYTHON SOURCE LINES 85-93
 
 Parse the python model object to convert it into a relay module
 and weights.
@@ -160,7 +155,7 @@ If you are unsure what that might be, this can be discovered by using
 the ``visualize.py`` script within the Tensorflow project.
 See `How do I inspect a .tflite file? <https://www.tensorflow.org/lite/guide/faq>`_
 
-.. GENERATED FROM PYTHON SOURCE LINES 98-107
+.. GENERATED FROM PYTHON SOURCE LINES 93-102
 
 .. code-block:: default
 
@@ -180,7 +175,7 @@ See `How do I inspect a .tflite file? <https://www.tensorflow.org/lite/guide/faq
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 108-117
+.. GENERATED FROM PYTHON SOURCE LINES 103-112
 
 Defining the target
 -------------------
@@ -192,7 +187,7 @@ TARGET, the C Runtime as the RUNTIME and a proper board/VM to run it (Zephyr wil
 QEMU VM based on BOARD. In the example below the x86 arch is selected and a x86 VM is picked up accordingly:
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 117-147
+.. GENERATED FROM PYTHON SOURCE LINES 112-142
 
 .. code-block:: default
 
@@ -233,11 +228,11 @@ QEMU VM based on BOARD. In the example below the x86 arch is selected and a x86
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 148-149
+.. GENERATED FROM PYTHON SOURCE LINES 143-144
 
 Now, compile the model for the target:
 
-.. GENERATED FROM PYTHON SOURCE LINES 149-239
+.. GENERATED FROM PYTHON SOURCE LINES 144-233
 
 .. code-block:: default
 
@@ -314,7 +309,6 @@ Now, compile the model for the target:
             "board": BOARD,
             "serial_number": SERIAL,
             "config_main_stack_size": 4096,
-            "cmsis_path": os.getenv("CMSIS_PATH", default="/content/cmsis"),
             "zephyr_base": os.getenv("ZEPHYR_BASE", default="/content/zephyrproject/zephyr"),
         }
 
@@ -368,14 +362,14 @@ Now, compile the model for the target:
 
 
 
-.. GENERATED FROM PYTHON SOURCE LINES 240-244
+.. GENERATED FROM PYTHON SOURCE LINES 234-238
 
 Next, establish a session with the simulated device and run the
 computation. The `with session` line would typically flash an attached
 microcontroller, but in this tutorial, it simply launches a subprocess
 to stand in for an attached microcontroller.
 
-.. GENERATED FROM PYTHON SOURCE LINES 244-261
+.. GENERATED FROM PYTHON SOURCE LINES 238-255
 
 .. code-block:: default
 
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 8fe16d0525..2040064119 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
@@ -218,7 +218,7 @@ take about **2 minutes** to download the Stanford Cars, while COCO 2017 validati
  .. code-block:: none
 
 
-    '/tmp/tmpbyatuwte/images/random'
+    '/tmp/tmpdg8c_6sp/images/random'
 
 
 
@@ -309,7 +309,7 @@ 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], [1.0, 0.0], [1.0, 0.0], [1.0, 0.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0]
+   :alt: [0.0, 1.0], [0.0, 1.0], [1.0, 0.0], [1.0, 0.0], [1.0, 0.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0]
    :srcset: /how_to/work_with_microtvm/images/sphx_glr_micro_train_001.png
    :class: sphx-glr-single-img
 
@@ -318,8 +318,8 @@ objects to other stuff? We can display some examples from our datasets using ``m
 
  .. code-block:: none
 
-    /tmp/tmpbyatuwte/images/target contains 8144 images
-    /tmp/tmpbyatuwte/images/random contains 5000 images
+    /tmp/tmpdg8c_6sp/images/target contains 8144 images
+    /tmp/tmpdg8c_6sp/images/random contains 5000 images
 
 
 
@@ -494,13 +494,13 @@ the time on our validation set).
  .. code-block:: none
 
     Epoch 1/3
-    328/328 - 46s - loss: 0.2104 - accuracy: 0.9268 - val_loss: 0.1214 - val_accuracy: 0.9585 - 46s/epoch - 141ms/step
+    328/328 - 47s - loss: 0.2108 - accuracy: 0.9265 - val_loss: 0.1686 - val_accuracy: 0.9400 - 47s/epoch - 143ms/step
     Epoch 2/3
-    328/328 - 42s - loss: 0.0854 - accuracy: 0.9678 - val_loss: 0.1452 - val_accuracy: 0.9551 - 42s/epoch - 129ms/step
+    328/328 - 43s - loss: 0.0950 - accuracy: 0.9641 - val_loss: 0.1165 - val_accuracy: 0.9588 - 43s/epoch - 132ms/step
     Epoch 3/3
-    328/328 - 43s - loss: 0.0654 - accuracy: 0.9766 - val_loss: 0.1017 - val_accuracy: 0.9634 - 43s/epoch - 131ms/step
+    328/328 - 43s - loss: 0.0598 - accuracy: 0.9775 - val_loss: 0.1400 - val_accuracy: 0.9573 - 43s/epoch - 131ms/step
 
-    <keras.callbacks.History object at 0x7f7004188410>
+    <keras.callbacks.History object at 0x7fcd419a44d0>
 
 
 
@@ -857,7 +857,7 @@ 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  24.588 seconds)
+   **Total running time of the script:** ( 4 minutes  29.829 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 986e1407be..43009c5da2 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
@@ -5,18 +5,18 @@
 
 Computation times
 =================
-**06:28.891** total execution time for **how_to_work_with_microtvm** files:
+**06:36.662** total execution time for **how_to_work_with_microtvm** files:
 
 +---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_train.py` (``micro_train.py``)               | 04:24.588 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_train.py` (``micro_train.py``)               | 04:29.829 | 0.0 MB |
 +---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_pytorch.py` (``micro_pytorch.py``)           | 01:02.083 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_pytorch.py` (``micro_pytorch.py``)           | 01:03.519 | 0.0 MB |
 +---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``)         | 00:50.692 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_autotune.py` (``micro_autotune.py``)         | 00:51.606 | 0.0 MB |
 +---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_aot.py` (``micro_aot.py``)                   | 00:07.742 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_aot.py` (``micro_aot.py``)                   | 00:07.890 | 0.0 MB |
 +---------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``)             | 00:03.786 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_microtvm_micro_tflite.py` (``micro_tflite.py``)             | 00:03.819 | 0.0 MB |
 +---------------------------------------------------------------------------------------------+-----------+--------+
 | :ref:`sphx_glr_how_to_work_with_microtvm_micro_tvmc.py` (``micro_tvmc.py``)                 | 00:00.000 | 0.0 MB |
 +---------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/how_to/work_with_relay/sg_execution_times.rst.txt b/docs/_sources/how_to/work_with_relay/sg_execution_times.rst.txt
index 31f3077a3b..b05f64238e 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
 =================
-**00:40.975** total execution time for **how_to_work_with_relay** files:
+**00:44.127** 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``) | 00:31.826 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_using_pipeline_executor.py` (``using_pipeline_executor.py``) | 00:32.385 | 0.0 MB |
 +----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``)           | 00:07.588 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_using_external_lib.py` (``using_external_lib.py``)           | 00:10.269 | 0.0 MB |
 +----------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``)                             | 00:01.555 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_relay_build_gcn.py` (``build_gcn.py``)                             | 00:01.467 | 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 27c0fe1b3f..adab0a09dd 100644
--- a/docs/_sources/how_to/work_with_schedules/intrin_math.rst.txt
+++ b/docs/_sources/how_to/work_with_schedules/intrin_math.rst.txt
@@ -264,7 +264,7 @@ The following example customizes CUDA lowering rule for :code:`exp`.
  .. code-block:: none
 
 
-    <function my_cuda_math_rule at 0x7f6e32240950>
+    <function my_cuda_math_rule at 0x7fcce21f5680>
 
 
 
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 5e0ae6cbc0..b0708f5ad7 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
@@ -5,18 +5,18 @@
 
 Computation times
 =================
-**00:07.381** total execution time for **how_to_work_with_schedules** files:
+**00:07.467** total execution time for **how_to_work_with_schedules** files:
 
 +------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``)                 | 00:04.854 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_intrin_math.py` (``intrin_math.py``)                 | 00:04.936 | 0.0 MB |
 +------------------------------------------------------------------------------------------------+-----------+--------+
 | :ref:`sphx_glr_how_to_work_with_schedules_tensorize.py` (``tensorize.py``)                     | 00:01.174 | 0.0 MB |
 +------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_reduction.py` (``reduction.py``)                     | 00:00.581 | 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_scan.py` (``scan.py``)                               | 00:00.559 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_scan.py` (``scan.py``)                               | 00:00.564 | 0.0 MB |
 +------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``)                     | 00:00.113 | 0.0 MB |
+| :ref:`sphx_glr_how_to_work_with_schedules_extern_op.py` (``extern_op.py``)                     | 00:00.114 | 0.0 MB |
 +------------------------------------------------------------------------------------------------+-----------+--------+
 | :ref:`sphx_glr_how_to_work_with_schedules_schedule_primitives.py` (``schedule_primitives.py``) | 00:00.049 | 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 6fb6d3213f..5f49d62b88 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.
                  B: Buffer(B_2: Pointer(float32), float32, [512, 64], []),
                  C: Buffer(C_2: Pointer(float32), float32, [1024, 512], [])}
       buffer_map = {A_1: A, B_1: B, C_1: C} {
-      attr [IterVar(i: int32, (nullptr), "DataPar", "")] "pragma_import_llvm" = "; ModuleID = '/tmp/tmpz_jf9bov/input0.cc'\nsource_filename = \"/tmp/tmpz_jf9bov/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/tmp5m3vhvsv/input0.cc'\nsource_filename = \"/tmp/tmp5m3vhvsv/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 a44c78a1ef..d62b51ad7d 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
 =================
-**00:25.777** total execution time for **topic_vta_tutorials_autotvm** files:
+**00:26.209** total execution time for **topic_vta_tutorials_autotvm** files:
 
 +---------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``) | 00:25.770 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_relay_vta.py` (``tune_relay_vta.py``) | 00:26.202 | 0.0 MB |
 +---------------------------------------------------------------------------------------+-----------+--------+
 | :ref:`sphx_glr_topic_vta_tutorials_autotvm_tune_alu_vta.py` (``tune_alu_vta.py``)     | 00:00.007 | 0.0 MB |
 +---------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt b/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
index 9149535ae1..29af50b59c 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/deploy_classification.rst.txt
@@ -293,7 +293,7 @@ The compilation steps are:
       DeprecationWarning,
     /workspace/vta/tutorials/frontend/deploy_classification.py:213: DeprecationWarning: legacy graph executor behavior of producing json / lib / params will be removed in the next release. Please see documents of tvm.contrib.graph_executor.GraphModule for the  new recommended usage.
       relay_prog, target=tvm.target.Target(target, host=env.target_host), params=params
-    resnet18_v1 inference graph built in 28.91s!
+    resnet18_v1 inference graph built in 29.13s!
 
 
 
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 0fc71fefd3..2f14f1fd70 100644
--- a/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
+++ b/docs/_sources/topic/vta/tutorials/frontend/deploy_detection.rst.txt
@@ -337,7 +337,7 @@ The compilation steps are:
 
     /workspace/python/tvm/relay/build_module.py:348: DeprecationWarning: Please use input parameter mod (tvm.IRModule) instead of deprecated parameter mod (tvm.relay.function.Function)
       DeprecationWarning,
-    yolov3-tiny inference graph built in 19.59s!
+    yolov3-tiny inference graph built in 19.60s!
 
 
 
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 b752e95269..e8f9edc838 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
 =================
-**01:32.127** total execution time for **topic_vta_tutorials_frontend** files:
+**01:32.450** total execution time for **topic_vta_tutorials_frontend** files:
 
 +------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``)           | 00:46.327 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_detection.py` (``deploy_detection.py``)           | 00:46.469 | 0.0 MB |
 +------------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``) | 00:45.800 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_frontend_deploy_classification.py` (``deploy_classification.py``) | 00:45.981 | 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 8c5d1e4566..5c4730b16a 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
 =================
-**00:03.127** total execution time for **topic_vta_tutorials_optimize** files:
+**00:03.308** total execution time for **topic_vta_tutorials_optimize** files:
 
 +--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``)         | 00:02.671 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_optimize_convolution_opt.py` (``convolution_opt.py``)         | 00:02.838 | 0.0 MB |
 +--------------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``) | 00:00.457 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_optimize_matrix_multiply_opt.py` (``matrix_multiply_opt.py``) | 00:00.470 | 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 a7aa86da77..8a87e202d2 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
 =================
-**00:00.843** total execution time for **topic_vta_tutorials** files:
+**00:00.819** total execution time for **topic_vta_tutorials** files:
 
 +---------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``) | 00:00.455 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_matrix_multiply.py` (``matrix_multiply.py``) | 00:00.434 | 0.0 MB |
 +---------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``) | 00:00.387 | 0.0 MB |
+| :ref:`sphx_glr_topic_vta_tutorials_vta_get_started.py` (``vta_get_started.py``) | 00:00.385 | 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 409ca5efa0..32137b0cf4 100644
--- a/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
+++ b/docs/_sources/tutorial/auto_scheduler_matmul_x86.rst.txt
@@ -207,6 +207,13 @@ trials, we can load the best schedule from the log file and apply it.
 
 
 
+.. rst-class:: sphx-glr-script-out
+
+ .. code-block:: none
+
+    *E
+
+
 
 
 
@@ -329,7 +336,7 @@ We build the binary and check its correctness and performance.
 
  .. code-block:: none
 
-    Execution time of this operator: 95.280 ms
+    Execution time of this operator: 97.624 ms
 
 
 
@@ -447,7 +454,7 @@ operations.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 1 minutes  16.612 seconds)
+   **Total running time of the script:** ( 1 minutes  20.637 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 569919c89d..05ac5a8f8d 100644
--- a/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt
+++ b/docs/_sources/tutorial/autotvm_matmul_x86.rst.txt
@@ -454,16 +454,16 @@ reduce variance, we take 5 measurements and average them.
     waiting for device...
     device available
     Get devices for measurement successfully!
-    No: 1   GFLOPS: 1.18/1.18       result: MeasureResult(costs=(0.22698483580000003,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.9494411945343018, timestamp=1673554357.5490327)        [('tile_y', [-1, 16]), ('tile_x', [-1, 1])],None,4
-    No: 2   GFLOPS: 12.32/12.32     result: MeasureResult(costs=(0.0217813936,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5843517780303955, timestamp=1673554358.9106257)       [('tile_y', [-1, 256]), ('tile_x', [-1, 512])],None,98
-    No: 3   GFLOPS: 9.90/12.32      result: MeasureResult(costs=(0.027110872399999996,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.671090841293335, timestamp=1673554359.6068351)        [('tile_y', [-1, 1]), ('tile_x', [-1, 256])],None,80
-    No: 4   GFLOPS: 2.04/12.32      result: MeasureResult(costs=(0.131435967,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.3466086387634277, timestamp=1673554362.718498) [('tile_y', [-1, 8]), ('tile_x', [-1, 2])],None,13
-    No: 5   GFLOPS: 9.15/12.32      result: MeasureResult(costs=(0.0293520404,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.7791948318481445, timestamp=1673554363.6109908)       [('tile_y', [-1, 16]), ('tile_x', [-1, 32])],None,54
-    No: 6   GFLOPS: 8.79/12.32      result: MeasureResult(costs=(0.0305245594,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.7350471019744873, timestamp=1673554365.1322844)       [('tile_y', [-1, 2]), ('tile_x', [-1, 64])],None,61
-    No: 7   GFLOPS: 11.79/12.32     result: MeasureResult(costs=(0.0227621998,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6264462471008301, timestamp=1673554365.7575874)       [('tile_y', [-1, 32]), ('tile_x', [-1, 256])],None,85
-    No: 8   GFLOPS: 10.55/12.32     result: MeasureResult(costs=(0.025449613399999998,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.700911283493042, timestamp=1673554366.4198508)        [('tile_y', [-1, 512]), ('tile_x', [-1, 64])],None,69
-    No: 9   GFLOPS: 1.62/12.32      result: MeasureResult(costs=(0.1657385202,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.8665077686309814, timestamp=1673554369.4074454)       [('tile_y', [-1, 2]), ('tile_x', [-1, 2])],None,11
-    No: 10  GFLOPS: 3.58/12.32      result: MeasureResult(costs=(0.0750699752,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.434582233428955, timestamp=1673554370.872784) [('tile_y', [-1, 16]), ('tile_x', [-1, 8])],None,34
+    No: 1   GFLOPS: 3.68/3.68       result: MeasureResult(costs=(0.0730355736,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.420715570449829, timestamp=1673568505.9202921)        [('tile_y', [-1, 128]), ('tile_x', [-1, 16])],None,47
+    No: 2   GFLOPS: 12.29/12.29     result: MeasureResult(costs=(0.0218502902,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6094512939453125, timestamp=1673568506.527237)        [('tile_y', [-1, 8]), ('tile_x', [-1, 256])],None,83
+    No: 3   GFLOPS: 7.71/12.29      result: MeasureResult(costs=(0.0348013332,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.8950433731079102, timestamp=1673568508.1103191)       [('tile_y', [-1, 1]), ('tile_x', [-1, 128])],None,70
+    No: 4   GFLOPS: 1.29/12.29      result: MeasureResult(costs=(0.2072951306,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.529980421066284, timestamp=1673568512.4440742)        [('tile_y', [-1, 1]), ('tile_x', [-1, 2])],None,10
+    No: 5   GFLOPS: 13.08/13.08     result: MeasureResult(costs=(0.0205211554,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5846850872039795, timestamp=1673568513.1419537)       [('tile_y', [-1, 64]), ('tile_x', [-1, 512])],None,96
+    No: 6   GFLOPS: 1.88/13.08      result: MeasureResult(costs=(0.142679848,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.5142734050750732, timestamp=1673568516.44907)  [('tile_y', [-1, 512]), ('tile_x', [-1, 4])],None,29
+    No: 7   GFLOPS: 0.94/13.08      result: MeasureResult(costs=(0.285880802,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.803407192230225, timestamp=1673568521.2701445) [('tile_y', [-1, 32]), ('tile_x', [-1, 2])],None,15
+    No: 8   GFLOPS: 1.89/13.08      result: MeasureResult(costs=(0.141984141,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.5025081634521484, timestamp=1673568523.7981641)        [('tile_y', [-1, 1]), ('tile_x', [-1, 8])],None,30
+    No: 9   GFLOPS: 2.05/13.08      result: MeasureResult(costs=(0.130725027,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.294708251953125, timestamp=1673568526.2706785) [('tile_y', [-1, 256]), ('tile_x', [-1, 4])],None,28
+    No: 10  GFLOPS: 11.74/13.08     result: MeasureResult(costs=(0.0228681012,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6322391033172607, timestamp=1673568526.8919046)       [('tile_y', [-1, 64]), ('tile_x', [-1, 32])],None,56
 
 
 
diff --git a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
index 09fe06cc8f..ef8952945f 100644
--- a/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
+++ b/docs/_sources/tutorial/autotvm_relay_x86.rst.txt
@@ -311,7 +311,7 @@ standard deviation.
 
  .. code-block:: none
 
-    {'mean': 505.9607368499974, 'median': 506.24863114999243, 'std': 1.0552857114823602}
+    {'mean': 512.5318330900063, 'median': 512.4503830500089, 'std': 1.829093237816327}
 
 
 
@@ -545,30 +545,31 @@ the tuning data to.
 
  .. code-block:: none
 
-
    [Task  1/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  1/25]  Current/Best:   22.18/  22.18 GFLOPS | Progress: (4/20) | 8.82 s
    [Task  1/25]  Current/Best:   13.47/  22.18 GFLOPS | Progress: (8/20) | 12.00 s
    [Task  1/25]  Current/Best:   19.22/  22.18 GFLOPS | Progress: (12/20) | 15.26 s
    [Task  1/25]  Current/Best:   23.51/  23.51 GFLOPS | Progress: (16/20) | 18.13 s
    [Task  1/25]  Current/Best:    3.38/  23.51 GFLOPS | Progress: (20/20) | 21.79 s Done.
-
    [Task  2/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  2/25]  Current/Best:   12.14/  16.55 GFLOPS | Progress: (4/20) | 3.59 s
    [Task  2/25]  Current/Best:   11.63/  22.11 GFLOPS | Progress: (8/20) | 5.07 s
    [Task  2/25]  Current/Best:   16.72/  22.11 GFLOPS | Progress: (12/20) | 6.53 s
    [Task  2/25]  Current/Best:   20.96/  22.11 GFLOPS | Progress: (16/20) | 8.89 s
    [Task  2/25]  Current/Best:   10.74/  22.11 GFLOPS | Progress: (20/20) | 10.86 s Done.
-
    [Task  3/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  3/25]  Current/Best:   18.92/  18.92 GFLOPS | Progress: (4/20) | 3.92 s
    [Task  3/25]  Current/Best:    9.03/  18.92 GFLOPS | Progress: (8/20) | 6.27 s
    [Task  3/25]  Current/Best:    1.63/  18.92 GFLOPS | Progress: (12/20) | 9.98 s
    [Task  3/25]  Current/Best:    3.17/  18.92 GFLOPS | Progress: (16/20) | 14.25 s
    [Task  3/25]  Current/Best:    5.63/  18.92 GFLOPS | Progress: (20/20) | 16.97 s Done.
-
    [Task  4/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  4/25]  Current/Best:    8.10/  12.09 GFLOPS | Progress: (4/20) | 6.44 s
    [Task  4/25]  Current/Best:   16.94/  16.94 GFLOPS | Progress: (8/20) | 8.46 s
    [Task  4/25]  Current/Best:    9.85/  16.94 GFLOPS | Progress: (12/20) | 14.77 s
    [Task  4/25]  Current/Best:   11.83/  21.54 GFLOPS | Progress: (16/20) | 16.68 s
    [Task  4/25]  Current/Best:   13.20/  21.54 GFLOPS | Progress: (20/20) | 18.85 s Done.
-
    [Task  5/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  5/25]  Current/Best:   14.11/  16.26 GFLOPS | Progress: (4/20) | 3.51 s
    [Task  5/25]  Current/Best:   11.25/  20.76 GFLOPS | Progress: (8/20) | 5.61 s
    [Task  5/25]  Current/Best:    5.33/  20.76 GFLOPS | Progress: (12/20) | 8.01 s
    [Task  5/25]  Current/Best:   18.35/  20.76 GFLOPS | Progress: (16/20) | 10.22 s
    [Task  5/25]  Current/Best:   16.04/  20.76 GFLOPS | Progress: (20/20) | 12.68 s Done.
-
    [Task  6/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  6/25]  Current/Best:   17.88/  17.88 GFLOPS | Progress: (4/20) | 4.55 s
    [Task  6/25]  Current/Best:   13.05/  17.88 GFLOPS | Progress: (8/20) | 8.21 s
    [Task  6/25]  Current/Best:    9.49/  18.40 GFLOPS | Progress: (12/20) | 10.65 s
    [Task  6/25]  Current/Best:   18.05/  18.40 GFLOPS | Progress: (16/20) | 13.67 s
    [Task  6/25]  Current/Best:   11.04/  18.40 GFLOPS | Progress: (20/20) | 16.50 s Done.
-
    [Task  7/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  7/25]  Current/Best:    7.62/  18.65 GFLOPS | Progress: (4/20) | 5.36 s
    [Task  7/25]  Current/Best:   19.22/  19.22 GFLOPS | Progress: (8/20) | 7.52 s
    [Task  7/25]  Current/Best:   20.00/  20.00 GFLOPS | Progress: (12/20) | 9.70 s
    [Task  7/25]  Current/Best:   17.74/  20.00 GFLOPS | Progress: (16/20) | 12.03 s
    [Task  7/25]  Current/Best:   15.30/  20.00 GFLOPS | Progress: (20/20) | 14.41 s Done.
-
    [Task  8/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  8/25]  Current/Best:   10.25/  16.61 GFLOPS | Progress: (4/20) | 5.86 s
    [Task  8/25]  Current/Best:   11.58/  16.61 GFLOPS | Progress: (8/20) | 11.19 s
    [Task  8/25]  Current/Best:    7.79/  16.61 GFLOPS | Progress: (12/20) | 20.85 s
    [Task  8/25]  Current/Best:    9.29/  16.61 GFLOPS | Progress: (16/20) | 29.58 s
    [Task  8/25]  Current/Best:   13.11/  16.61 GFLOPS | Progress: (20/20) | 32.76 s Done.
-
    [Task  9/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  9/25]  Current/Best:   12.94/  12.94 GFLOPS | Progress: (4/20) | 5.07 s
    [Task  9/25]  Current/Best:   18.42/  18.42 GFLOPS | Progress: (8/20) | 7.76 s
    [Task  9/25]  Current/Best:   14.00/  18.42 GFLOPS | Progress: (12/20) | 18.81 s
    [Task  9/25]  Current/Best:    9.97/  18.42 GFLOPS | Progress: (16/20) | 22.42 s
    [Task  9/25]  Current/Best:    7.47/  18.42 GFLOPS | Progress: (20/20) | 24.20 s
    [Task 10/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 10/25]  Current/Best:   15.69/  20.68 GFLOPS | Progress: (4/20) | 3.82 s
    [Task 10/25]  Current/Best:   12.88/  20.68 GFLOPS | Progress: (8/20) | 5.70 s
    [Task 10/25]  Current/Best:   16.44/  20.68 GFLOPS | Progress: (12/20) | 7.80 s
    [Task 10/25]  Current/Best:    9.40/  20.68 GFLOPS | Progress: (16/20) | 9.90 s
    [Task 10/25]  Current/Best:   11.94/  20.68 GFLOPS | Progress: (20/20) 
 | 12.26 s Done.
-
    [Task 11/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 11/25]  Current/Best:   18.09/  20.97 GFLOPS | Progress: (4/20) | 4.24 s
    [Task 11/25]  Current/Best:   21.82/  21.82 GFLOPS | Progress: (8/20) | 6.61 s
    [Task 11/25]  Current/Best:   19.04/  21.82 GFLOPS | Progress: (12/20) | 9.21 s
    [Task 11/25]  Current/Best:   11.93/  21.82 GFLOPS | Progress: (16/20) | 12.53 s
    [Task 11/25]  Current/Best:   22.68/  22.68 GFLOPS | Progress: (20/20) | 15.73 s Done.
-
    [Task 12/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 12/25]  Current/Best:   13.99/  17.74 GFLOPS | Progress: (4/20) | 5.12 s
    [Task 12/25]  Current/Best:    6.63/  17.74 GFLOPS | Progress: (8/20) | 11.38 s
    [Task 12/25]  Current/Best:    4.96/  17.74 GFLOPS | Progress: (12/20) | 13.79 s
    [Task 12/25]  Current/Best:    4.47/  18.45 GFLOPS | Progress: (16/20) | 18.19 s
    [Task 12/25]  Current/Best:   14.83/  18.70 GFLOPS | Progress: (20/20) | 22.38 s Done.
-
    [Task 13/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 13/25]  Current/Best:   11.65/  22.81 GFLOPS | Progress: (4/20) | 4.95 s
    [Task 13/25]  Current/Best:   12.58/  22.81 GFLOPS | Progress: (8/20) | 8.17 s
    [Task 13/25]  Current/Best:   20.06/  22.81 GFLOPS | Progress: (12/20) | 10.42 s
    [Task 13/25]  Current/Best:    9.65/  22.81 GFLOPS | Progress: (16/20) | 13.27 s
    [Task 13/25]  Current/Best:   16.39/  22.81 GFLOPS | Progress: (20/20) | 15.45 s Done.
-
    [Task 14/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 14/25]  Current/Best:   15.45/  15.59 GFLOPS | Progress: (4/20) | 4.57 s
    [Task 14/25]  Current/Best:   13.40/  15.59 GFLOPS | Progress: (8/20) | 7.00 s
    [Task 14/25]  Current/Best:    4.76/  15.59 GFLOPS | Progress: (12/20) | 9.51 s
    [Task 14/25]  Current/Best:   15.78/  15.78 GFLOPS | Progress: (16/20) | 12.48 s
    [Task 14/25]  Current/Best:   14.04/  15.78 GFLOPS | Progress: (20/20) | 15.88 s
    [Task 15/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 15/25]  Current/Best:   10.21/  21.19 GFLOPS | Progress: (4/20) | 3.86 s
    [Task 15/25]  Current/Best:    6.03/  21.19 GFLOPS | Progress: (8/20) | 5.80 s
    [Task 15/25]  Current/Best:   22.19/  22.19 GFLOPS | Progress: (12/20) | 7.96 s
    [Task 15/25]  Current/Best:    9.38/  22.19 GFLOPS | Progress: (16/20) | 10.24 s Done.
-     Done.
-
    [Task 15/25]  Current/Best:   18.45/  22.19 GFLOPS | Progress: (20/20) | 12.46 s Done.
-
    [Task 16/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 16/25]  Current/Best:   14.17/  20.16 GFLOPS | Progress: (4/20) | 4.32 s
    [Task 16/25]  Current/Best:   18.56/  20.16 GFLOPS | Progress: (8/20) | 6.92 s
    [Task 16/25]  Current/Best:   12.78/  20.16 GFLOPS | Progress: (12/20) | 9.67 s
    [Task 16/25]  Current/Best:   16.64/  20.16 GFLOPS | Progress: (16/20) | 11.17 s
    [Task 16/25]  Current/Best:   15.49/  20.16 GFLOPS | Progress: (20/20) | 13.40 s Done.
-
    [Task 17/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 17/25]  Current/Best:   11.49/  23.62 GFLOPS | Progress: (4/20) | 4.51 s
    [Task 17/25]  Current/Best:   10.41/  23.62 GFLOPS | Progress: (8/20) | 6.72 s
    [Task 17/25]  Current/Best:    8.97/  23.62 GFLOPS | Progress: (12/20) | 9.30 s
    [Task 17/25]  Current/Best:    7.08/  23.62 GFLOPS | Progress: (16/20) | 12.37 s
    [Task 17/25]  Current/Best:   17.45/  23.62 GFLOPS | Progress: (20/20) | 15.15 s Done.
-
    [Task 18/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 18/25]  Current/Best:    9.14/  16.07 GFLOPS | Progress: (4/20) | 7.36 s
    [Task 18/25]  Current/Best:    3.08/  16.07 GFLOPS | Progress: (8/20) | 10.87 s
    [Task 18/25]  Current/Best:   10.27/  18.23 GFLOPS | Progress: (12/20) | 16.16 s
    [Task 18/25]  Current/Best:   12.73/  18.23 GFLOPS | Progress: (16/20) | 21.78 s
    [Task 18/25]  Current/Best:   10.68/  18.23 GFLOPS | Progress: (20/20) | 24.11 s Done.
-
    [Task 19/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 19/25]  Current/Best:   15.02/  18.39 GFLOPS | Progress: (4/20) | 6.23 s
    [Task 19/25]  Current/Best:   10.02/  18.39 GFLOPS | Progress: (8/20) | 9.65 s
    [Task 19/25]  Current/Best:    4.19/  18.85 GFLOPS | Progress: (12/20) | 13.30 s
    [Task 19/25]  Current/Best:   22.20/  22.20 GFLOPS | Progress: (16/20) | 16.95 s
    [Task 19/25]  Current/Best:   11.44/  22.20 GFLOPS | Progress: (20/20) | 20.29 s Done.
-
    [Task 20/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 20/25]  Current/Best:   11.40/  13.88 GFLOPS | Progress: (4/20) | 5.26 s
    [Task 20/25]  Current/Best:   12.76/  15.84 GFLOPS | Progress: (8/20) | 7.53 s
    [Task 20/25]  Current/Best:   11.24/  16.41 GFLOPS | Progress: (12/20) | 9.65 s
    [Task 20/25]  Current/Best:   18.54/  18.54 GFLOPS | Progress: (16/20) | 12.10 s
    [Task 20/25]  Current/Best:   16.74/  18.54 GFLOPS | Progress: (20/20) | 15.29 s
    [Task 21/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 21/25]  Current/Best:   12.70/  15.40 GFLOPS | Progress: (4/20) | 4.31 s Done.
-
    [Task 21/25]  Current/Best:   20.95/  20.95 GFLOPS | Progress: (8/20) | 5.32 s
    [Task 21/25]  Current/Best:    2.28/  20.95 GFLOPS | Progress: (12/20) | 7.92 s
    [Task 21/25]  Current/Best:   17.84/  20.95 GFLOPS | Progress: (16/20) | 10.98 s
    [Task 21/25]  Current/Best:   10.74/  20.95 GFLOPS | Progress: (20/20) | 13.73 s
    [Task 22/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 22/25]  Current/Best:   12.02/  12.77 GFLOPS | Progress: (4/20) | 3.71 s
    [Task 22/25]  Current/Best:   10.33/  19.31 GFLOPS | Progress: (8/20) | 6.60 s
    [Task 22/25]  Current/Best:   15.37/  19.31 GFLOPS | Progress: (12/20) | 8.37 s
    [Task 22/25]  Current/Best:    6.91/  20.63 GFLOPS | Progress: (16/20) | 10.75 s
    [Task 22/25]  Current/Best:   15.80/  20.63 GFLOPS | Progress: (20/20) | 12.47 s Done.
-
    [Task 23/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 23/25]  Current/Best:   10.46/  20.90 GFLOPS | Progress: (4/20) | 4.71 s
    [Task 23/25]  Current/Best:   19.69/  20.90 GFLOPS | Progress: (8/20) | 12.68 s
    [Task 23/25]  Current/Best:    3.08/  22.58 GFLOPS | Progress: (12/20) | 16.74 s
    [Task 23/25]  Current/Best:    6.17/  22.58 GFLOPS | Progress: (16/20) | 21.47 s
    [Task 23/25]  Current/Best:   23.04/  23.04 GFLOPS | Progress: (20/20) | 26.96 s Done.
-
    [Task 24/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 24/25]  Current/Best:    8.31/   8.31 GFLOPS | Progress: (4/20) | 12.77 s
    [Task 24/25]  Current/Best:    5.96/   8.31 GFLOPS | Progress: (8/20) | 23.72 s
    [Task 24/25]  Current/Best:    8.88/   8.88 GFLOPS | Progress: (12/20) | 35.23 s
    [Task 24/25]  Current/Best:    1.37/  10.26 GFLOPS | Progress: (16/20) | 47.10 s
    [Task 24/25]  Current/Best:   10.11/  10.26 GFLOPS | Progress: (20/20) | 58.04 s
    [Task 25/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 25/25]  Current/Best:    3.46/   4.13 GFLOPS | Progress: (4/20) | 12.76 s
    [Task 25/25]  Current/Best:    8.56/   8.56 GFLOPS | Progress: (8/20) | 24.50 s
    [Task 25/25]  Current/Best:    5.64/   8.56 GFLOPS | Progress: (12/20) | 35.47 s
    [Task 25/25]  Current/Best:    2.99/   9.39 GFLOPS | Progress: (16/20) | 46.42 s
    [Task 25/25]  Current/Best:    7.84/   9.39 GFLOPS | Progress: (2
 0/20) | 48.59 s Done.
-
+
    [Task  1/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  1/25]  Current/Best:   18.99/  18.99 GFLOPS | Progress: (4/20) | 7.50 s
    [Task  1/25]  Current/Best:   14.97/  18.99 GFLOPS | Progress: (8/20) | 11.73 s
    [Task  1/25]  Current/Best:   19.09/  19.09 GFLOPS | Progress: (12/20) | 14.33 s
    [Task  1/25]  Current/Best:   12.91/  22.42 GFLOPS | Progress: (16/20) | 16.37 s
    [Task  1/25]  Current/Best:   11.11/  22.42 GFLOPS | Progress: (20/20) | 19.49 s Done.
+
    [Task  2/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  2/25]  Current/Best:   16.87/  17.11 GFLOPS | Progress: (4/20) | 3.96 s
    [Task  2/25]  Current/Best:    5.77/  17.97 GFLOPS | Progress: (8/20) | 5.48 s
    [Task  2/25]  Current/Best:   17.79/  17.97 GFLOPS | Progress: (12/20) | 7.11 s
    [Task  2/25]  Current/Best:   19.56/  20.75 GFLOPS | Progress: (16/20) | 8.55 s
    [Task  2/25]  Current/Best:    3.34/  20.75 GFLOPS | Progress: (20/20) | 10.01 s Done.
+
    [Task  3/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  3/25]  Current/Best:   13.70/  23.08 GFLOPS | Progress: (4/20) | 3.73 s
    [Task  3/25]  Current/Best:    7.54/  23.08 GFLOPS | Progress: (8/20) | 6.78 s
    [Task  3/25]  Current/Best:    9.82/  23.08 GFLOPS | Progress: (12/20) | 10.22 s
    [Task  3/25]  Current/Best:    9.19/  23.08 GFLOPS | Progress: (16/20) | 12.50 s
    [Task  3/25]  Current/Best:   12.74/  23.08 GFLOPS | Progress: (20/20) | 15.01 s Done.
+
    [Task  4/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  4/25]  Current/Best:   12.66/  18.10 GFLOPS | Progress: (4/20) | 4.54 s
    [Task  4/25]  Current/Best:   11.99/  18.10 GFLOPS | Progress: (8/20) | 10.14 s
    [Task  4/25]  Current/Best:    6.42/  18.10 GFLOPS | Progress: (12/20) | 12.24 s
    [Task  4/25]  Current/Best:    8.34/  18.10 GFLOPS | Progress: (16/20) | 14.48 s
    [Task  4/25]  Current/Best:   16.66/  18.10 GFLOPS | Progress: (20/20) | 18.45 s Done.
+
    [Task  5/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  5/25]  Current/Best:   15.51/  17.45 GFLOPS | Progress: (4/20) | 4.13 s
    [Task  5/25]  Current/Best:    8.46/  17.45 GFLOPS | Progress: (8/20) | 6.38 s
    [Task  5/25]  Current/Best:   18.06/  18.06 GFLOPS | Progress: (12/20) | 8.72 s
    [Task  5/25]  Current/Best:    3.07/  20.14 GFLOPS | Progress: (16/20) | 10.80 s
    [Task  5/25]  Current/Best:    4.86/  20.14 GFLOPS | Progress: (20/20) | 14.36 s Done.
+
    [Task  6/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  6/25]  Current/Best:   14.64/  17.80 GFLOPS | Progress: (4/20) | 4.66 s
    [Task  6/25]  Current/Best:   12.91/  22.51 GFLOPS | Progress: (8/20) | 7.25 s
    [Task  6/25]  Current/Best:   11.60/  22.51 GFLOPS | Progress: (12/20) | 11.12 s
    [Task  6/25]  Current/Best:   13.77/  22.51 GFLOPS | Progress: (16/20) | 13.38 s
    [Task  6/25]  Current/Best:    4.65/  22.51 GFLOPS | Progress: (20/20) | 17.60 s Done.
+
    [Task  7/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  7/25]  Current/Best:   13.58/  20.78 GFLOPS | Progress: (4/20) | 4.43 s
    [Task  7/25]  Current/Best:   19.29/  20.78 GFLOPS | Progress: (8/20) | 6.70 s
    [Task  7/25]  Current/Best:   16.13/  20.78 GFLOPS | Progress: (12/20) | 9.08 s
    [Task  7/25]  Current/Best:   18.73/  20.78 GFLOPS | Progress: (16/20) | 11.39 s
    [Task  7/25]  Current/Best:    6.22/  20.78 GFLOPS | Progress: (20/20) | 13.80 s Done.
+
    [Task  8/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  8/25]  Current/Best:   10.10/  19.90 GFLOPS | Progress: (4/20) | 7.59 s
    [Task  8/25]  Current/Best:   11.43/  19.90 GFLOPS | Progress: (8/20) | 10.37 s
    [Task  8/25]  Current/Best:    9.24/  20.66 GFLOPS | Progress: (12/20) | 21.22 s
    [Task  8/25]  Current/Best:    5.16/  20.66 GFLOPS | Progress: (16/20) | 25.70 s
    [Task  8/25]  Current/Best:   11.94/  20.66 GFLOPS | Progress: (20/20) | 28.56 s Done.
+
    [Task  9/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task  9/25]  Current/Best:   18.82/  18.82 GFLOPS | Progress: (4/20) | 3.72 s
    [Task  9/25]  Current/Best:   11.40/  18.82 GFLOPS | Progress: (8/20) | 11.05 s
    [Task  9/25]  Current/Best:   12.33/  18.82 GFLOPS | Progress: (12/20) | 13.71 s
    [Task  9/25]  Current/Best:    8.19/  18.82 GFLOPS | Progress: (16/20) | 17.19 s
    [Task  9/25]  Current/Best:   17.06/  18.82 GFLOPS | Progress: (20/20) | 19.19 s Done.
+
    [Task 10/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 10/25]  Current/Best:   14.24/  15.97 GFLOPS | Progress: (4/20) | 3.50 s
    [Task 10/25]  Current/Best:    6.23/  15.97 GFLOPS | Progress: (8/20) | 6.29 s
    [Task 10/25]  Current/Best:    9.59/  17.28 GFLOPS | Progress: (12/20) | 8.93 s
    [Task 10/25]  Current/Best:    3.95/  17.28 GFLOPS | Progress: (16/20) | 11.30 s
    [Task 10/25]  Current/Best:   20.11/  20.11 GFLOPS | Progress: (20/20) | 13.67 s Done.
+
    [Task 11/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 11/25]  Current/Best:   18.42/  18.42 GFLOPS | Progress: (4/20) | 4.05 s
    [Task 11/25]  Current/Best:   21.61/  23.89 GFLOPS | Progress: (8/20) | 6.37 s
    [Task 11/25]  Current/Best:   18.11/  23.89 GFLOPS | Progress: (12/20) | 8.58 s
    [Task 11/25]  Current/Best:   12.34/  23.89 GFLOPS | Progress: (16/20) | 10.91 s
    [Task 11/25]  Current/Best:   12.19/  23.89 GFLOPS | Progress: (20/20) | 13.78 s Done.
+
    [Task 12/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 12/25]  Current/Best:   14.94/  14.94 GFLOPS | Progress: (4/20) | 4.52 s
    [Task 12/25]  Current/Best:    2.97/  15.49 GFLOPS | Progress: (8/20) | 7.26 s
    [Task 12/25]  Current/Best:   11.94/  21.12 GFLOPS | Progress: (12/20) | 9.50 s
    [Task 12/25]  Current/Best:    4.84/  21.12 GFLOPS | Progress: (16/20) | 12.84 s
    [Task 12/25]  Current/Best:   11.86/  21.12 GFLOPS | Progress: (20/20) | 15.73 s Done.
+
    [Task 13/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 13/25]  Current/Best:   11.53/  20.90 GFLOPS | Progress: (4/20) | 5.20 s
    [Task 13/25]  Current/Best:   14.12/  20.90 GFLOPS | Progress: (8/20) | 9.15 s
    [Task 13/25]  Current/Best:   17.57/  20.90 GFLOPS | Progress: (12/20) | 12.37 s
    [Task 13/25]  Current/Best:   10.26/  20.90 GFLOPS | Progress: (16/20) | 15.33 s
    [Task 13/25]  Current/Best:    1.57/  20.90 GFLOPS | Progress: (20/20) | 19.26 s Done.
+
    [Task 14/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 14/25]  Current/Best:    2.99/   6.21 GFLOPS | Progress: (4/20) | 5.80 s
    [Task 14/25]  Current/Best:   14.21/  15.80 GFLOPS | Progress: (8/20) | 7.89 s
    [Task 14/25]  Current/Best:   13.34/  15.80 GFLOPS | Progress: (12/20) | 14.85 s
    [Task 14/25]  Current/Best:   14.44/  15.80 GFLOPS | Progress: (16/20) | 18.94 s
    [Task 14/25]  Current/Best:    4.42/  15.80 GFLOPS | Progress: (20/20) | 21.41 s Done.
+
    [Task 15/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 15/25]  Current/Best:    7.87/  14.65 GFLOPS | Progress: (4/20) | 4.48 s
    [Task 15/25]  Current/Best:   18.79/  20.68 GFLOPS | Progress: (8/20) | 6.07 s
    [Task 15/25]  Current/Best:   11.74/  20.68 GFLOPS | Progress: (12/20) | 10.44 s
    [Task 15/25]  Current/Best:   20.43/  20.68 GFLOPS | Progress: (16/20) | 16.39 s
    [Task 15/25]  Current/Best:    3.01/  20.68 GFLOPS | Progress: (20/20) | 18.37 s Done.
+
    [Task 16/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 16/25]  Current/Best:    9.90/  13.66 GFLOPS | Progress: (4/20) | 4.15 s
    [Task 16/25]  Current/Best:   13.27/  13.66 GFLOPS | Progress: (8/20) | 6.53 s
    [Task 16/25]  Current/Best:   16.19/  17.85 GFLOPS | Progress: (12/20) | 8.43 s
    [Task 16/25]  Current/Best:   12.24/  17.85 GFLOPS | Progress: (16/20) | 11.05 s
    [Task 16/25]  Current/Best:   14.57/  17.85 GFLOPS | Progress: (20/20) | 13.84 s Done.
+
    [Task 17/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 17/25]  Current/Best:   14.15/  19.09 GFLOPS | Progress: (4/20) | 4.42 s
    [Task 17/25]  Current/Best:    3.10/  19.09 GFLOPS | Progress: (8/20) | 8.62 s
    [Task 17/25]  Current/Best:   21.86/  21.86 GFLOPS | Progress: (12/20) | 11.16 s
    [Task 17/25]  Current/Best:   11.29/  21.86 GFLOPS | Progress: (16/20) | 14.36 s
    [Task 17/25]  Current/Best:   12.50/  21.86 GFLOPS | Progress: (20/20) | 17.27 s Done.
+
    [Task 18/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 18/25]  Current/Best:   21.60/  21.60 GFLOPS | Progress: (4/20) | 3.56 s
    [Task 18/25]  Current/Best:   19.15/  21.60 GFLOPS | Progress: (8/20) | 6.05 s
    [Task 18/25]  Current/Best:   18.63/  21.60 GFLOPS | Progress: (12/20) | 10.00 s
    [Task 18/25]  Current/Best:    9.16/  21.60 GFLOPS | Progress: (16/20) | 15.87 s
    [Task 18/25]  Current/Best:   16.30/  21.60 GFLOPS | Progress: (20/20) | 17.83 s Done.
+
    [Task 19/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 19/25]  Current/Best:    5.35/  19.37 GFLOPS | Progress: (4/20) | 4.78 s
    [Task 19/25]  Current/Best:   23.04/  23.04 GFLOPS | Progress: (8/20) | 8.27 s
    [Task 19/25]  Current/Best:    3.08/  23.04 GFLOPS | Progress: (12/20) | 11.26 s
    [Task 19/25]  Current/Best:   18.84/  23.04 GFLOPS | Progress: (16/20) | 15.97 s
    [Task 19/25]  Current/Best:   19.81/  23.04 GFLOPS | Progress: (20/20) | 19.20 s Done.
+
    [Task 20/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 20/25]  Current/Best:   12.14/  22.53 GFLOPS | Progress: (4/20) | 3.89 s
    [Task 20/25]  Current/Best:   12.72/  22.53 GFLOPS | Progress: (8/20) | 6.54 s
    [Task 20/25]  Current/Best:   14.48/  22.53 GFLOPS | Progress: (12/20) | 9.19 s
    [Task 20/25]  Current/Best:   13.78/  22.53 GFLOPS | Progress: (16/20) | 13.30 s
    [Task 20/25]  Current/Best:   14.14/  22.53 GFLOPS | Progress: (20/20) | 19.77 s
    [Task 21/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 21/25]  Current/Best:   16.36/  21.86 GFLOPS | Progress: (4/20) | 4.96 s Done.
+
    [Task 21/25]  Current/Best:    5.34/  21.86 GFLOPS | Progress: (8/20) | 6.64 s
    [Task 21/25]  Current/Best:    5.25/  21.86 GFLOPS | Progress: (12/20) | 10.04 s
    [Task 21/25]  Current/Best:   15.39/  21.86 GFLOPS | Progress: (16/20) | 11.79 s
    [Task 21/25]  Current/Best:    9.50/  21.86 GFLOPS | Progress: (20/20) | 15.04 s Done.
+
    [Task 22/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 22/25]  Current/Best:   17.80/  18.58 GFLOPS | Progress: (4/20) | 4.00 s
    [Task 22/25]  Current/Best:   10.84/  18.58 GFLOPS | Progress: (8/20) | 6.42 s
    [Task 22/25]  Current/Best:   14.81/  18.58 GFLOPS | Progress: (12/20) | 8.67 s
    [Task 22/25]  Current/Best:   17.91/  18.58 GFLOPS | Progress: (16/20) | 10.47 s
    [Task 22/25]  Current/Best:   12.37/  18.58 GFLOPS | Progress: (20/20) | 13.46 s Done.
+
    [Task 23/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 23/25]  Current/Best:    6.11/  11.95 GFLOPS | Progress: (4/20) | 7.70 s
    [Task 23/25]  Current/Best:   18.42/  18.42 GFLOPS | Progress: (8/20) | 11.12 s
    [Task 23/25]  Current/Best:   11.10/  19.30 GFLOPS | Progress: (12/20) | 13.87 s
    [Task 23/25]  Current/Best:   12.18/  23.83 GFLOPS | Progress: (16/20) | 16.29 s
    [Task 23/25]  Current/Best:   22.51/  23.83 GFLOPS | Progress: (20/20) | 18.84 s Done.
+
    [Task 24/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
    [Task 24/25]  Current/Best:    3.15/   6.49 GFLOPS | Progress: (4/20) | 9.85 s
    [Task 24/25]  Current/Best:    4.23/   8.54 GFLOPS | Progress: (8/20) | 20.80 s
    [Task 24/25]  Current/Best:    9.34/   9.34 GFLOPS | Progress: (12/20) | 31.93 s
    [Task 24/25]  Current/Best:    2.04/   9.34 GFLOPS | Progress: (16/20) | 43.83 s
    [Task 24/25]  Current/Best:    2.99/   9.34 GFLOPS | Progress: (20/20) | 54.51 s
    [Task 25/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
+
    [Task 25/25]  Current/Best:    1.53/   8.39 GFLOPS | Progress: (4/20) | 4.33 s
    [Task 25/25]  Current/Best:    3.71/   8.99 GFLOPS | Progress: (8/20) | 15.30 s
    [Task 25/25]  Current/Best:    8.83/   9.14 GFLOPS | Progress: (12/20) | 27.08 s
    [Task 25/25]  Current/Best:    3.00/   9.14 GFLOPS | Progress: (16/20) | 37.75 s
    [Task 25/25]  Current/Best:    7.87/   9.69 GFLOPS | Progress: (20/20) | 48.67 s
 
 
 
@@ -665,7 +666,7 @@ Verify that the optimized model runs and produces the same results:
  .. code-block:: none
 
     class='n02123045 tabby, tabby cat' with probability=0.621104
-    class='n02123159 tiger cat' with probability=0.356378
+    class='n02123159 tiger cat' with probability=0.356379
     class='n02124075 Egyptian cat' with probability=0.019712
     class='n02129604 tiger, Panthera tigris' with probability=0.001215
     class='n04040759 radiator' with probability=0.000262
@@ -722,8 +723,8 @@ improvement in comparing the optimized model to the unoptimized model.
 
  .. code-block:: none
 
-    optimized: {'mean': 419.9832372800074, 'median': 419.7767239500081, 'std': 1.768500945522016}
-    unoptimized: {'mean': 505.9607368499974, 'median': 506.24863114999243, 'std': 1.0552857114823602}
+    optimized: {'mean': 410.4258411699948, 'median': 408.519481050007, 'std': 3.4550342127633344}
+    unoptimized: {'mean': 512.5318330900063, 'median': 512.4503830500089, 'std': 1.829093237816327}
 
 
 
@@ -746,7 +747,7 @@ profiling/benchmarking.
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 11 minutes  58.606 seconds)
+   **Total running time of the script:** ( 11 minutes  45.042 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 b2959d824a..b7d29e03f4 100644
--- a/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
+++ b/docs/_sources/tutorial/cross_compilation_and_rpc.rst.txt
@@ -274,7 +274,7 @@ device and returns the measured cost. Network overhead is excluded.
 
  .. code-block:: none
 
-    1.282e-07 secs/op
+    1.317e-07 secs/op
 
 
 
diff --git a/docs/_sources/tutorial/intro_topi.rst.txt b/docs/_sources/tutorial/intro_topi.rst.txt
index f81d330718..d84cebc9b2 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, 0x1cdf7900)), stage(b, placeholder(b, 0x229ecbf0)), 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(mi [...]
+    [stage(a, placeholder(a, 0x1b0c3470)), stage(b, placeholder(b, 0x1fa44170)), 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(mi [...]
 
 
 
diff --git a/docs/_sources/tutorial/sg_execution_times.rst.txt b/docs/_sources/tutorial/sg_execution_times.rst.txt
index 5d3dcff68a..5d0d2120d9 100644
--- a/docs/_sources/tutorial/sg_execution_times.rst.txt
+++ b/docs/_sources/tutorial/sg_execution_times.rst.txt
@@ -5,24 +5,24 @@
 
 Computation times
 =================
-**15:10.668** total execution time for **tutorial** files:
+**15:06.103** total execution time for **tutorial** files:
 
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``)                 | 11:58.606 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_autotvm_relay_x86.py` (``autotvm_relay_x86.py``)                 | 11:45.042 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``) | 01:16.612 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_auto_scheduler_matmul_x86.py` (``auto_scheduler_matmul_x86.py``) | 01:20.637 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``)     | 00:58.752 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_tensor_expr_get_started.py` (``tensor_expr_get_started.py``)     | 00:58.450 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``)                 | 00:33.702 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_relay_quick_start.py` (``relay_quick_start.py``)                 | 00:33.661 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``)               | 00:21.427 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_autotvm_matmul_x86.py` (``autotvm_matmul_x86.py``)               | 00:26.705 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``)                               | 00:00.817 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_intro_topi.py` (``intro_topi.py``)                               | 00:00.820 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``)       | 00:00.598 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_tensor_ir_blitz_course.py` (``tensor_ir_blitz_course.py``)       | 00:00.608 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
-| :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``) | 00:00.153 | 0.0 MB |
+| :ref:`sphx_glr_tutorial_cross_compilation_and_rpc.py` (``cross_compilation_and_rpc.py``) | 00:00.179 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
 | :ref:`sphx_glr_tutorial_uma.py` (``uma.py``)                                             | 00:00.000 | 0.0 MB |
 +------------------------------------------------------------------------------------------+-----------+--------+
diff --git a/docs/_sources/tutorial/tensor_expr_get_started.rst.txt b/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
index d722bf5bc6..592c6ac3b8 100644
--- a/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
+++ b/docs/_sources/tutorial/tensor_expr_get_started.rst.txt
@@ -285,7 +285,7 @@ helper function to run a profile of the TVM generated code.
 
  .. code-block:: none
 
-    Numpy running time: 0.000007
+    Numpy running time: 0.000008
     naive: 0.000007
 
 
@@ -490,10 +490,10 @@ We can now compare the different schedules
  .. code-block:: none
 
                 Operator                  Timing             Performance
-                   numpy    7.031929999357089e-06                    1.0
-                   naive    6.6723000000000005e-06    0.9488575683503719
-                parallel    7.009799999999999e-06      0.996852926670329
-                  vector             2.46551e-05       3.506164026412969
+                   numpy    7.642879995728436e-06                    1.0
+                   naive    6.6804000000000004e-06     0.874068414489517
+                parallel    6.9625000000000015e-06    0.9109785844984227
+                  vector             2.47534e-05      3.2387529326424778
 
 
 
@@ -914,7 +914,7 @@ matrix multiplication.
 
  .. code-block:: none
 
-    Numpy running time: 0.018194
+    Numpy running time: 0.018027
 
 
 
@@ -972,7 +972,7 @@ optimizations.
 
  .. code-block:: none
 
-    none: 3.261212
+    none: 3.213367
 
 
 
@@ -1074,7 +1074,7 @@ schedule.
 
  .. code-block:: none
 
-    blocking: 0.294277
+    blocking: 0.301631
 
 
 
@@ -1169,7 +1169,7 @@ already cache friendly from our previous optimizations.
 
  .. code-block:: none
 
-    vectorization: 0.328194
+    vectorization: 0.332175
     @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, [1024, 1024], []),
@@ -1242,7 +1242,7 @@ more cache friendly.
 
  .. code-block:: none
 
-    loop permutation: 0.116499
+    loop permutation: 0.117709
     @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, [1024, 1024], []),
@@ -1340,7 +1340,7 @@ optimized schedule.
 
  .. code-block:: none
 
-    array packing: 0.109260
+    array packing: 0.109681
     @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, [1024, 1024], []),
@@ -1432,7 +1432,7 @@ to `C` when all the block results are ready.
 
  .. code-block:: none
 
-    block caching: 0.110412
+    block caching: 0.110728
     @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, [1024, 1024], []),
@@ -1517,7 +1517,7 @@ of thread-level parallelization.
 
  .. code-block:: none
 
-    parallelization: 0.145131
+    parallelization: 0.146554
     @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, [1024, 1024], []),
@@ -1597,13 +1597,13 @@ working, we can compare the results.
  .. code-block:: none
 
                 Operator                  Timing             Performance
-                    none             3.261211896                     1.0
-                blocking            0.2942766486     0.09023536586535254
-           vectorization            0.3281936381      0.1006354841592912
-        loop permutation            0.1164992826     0.03572269644388664
-           array packing            0.1092597523     0.03350280686575786
-           block caching     0.11041168139999999     0.03385602804142353
-         parallelization            0.1451309414     0.04450215012952964
+                    none              3.21336669                     1.0
+                blocking     0.30163059130000003     0.09386746686541399
+           vectorization            0.3321748137     0.10337283159551268
+        loop permutation     0.11770911390000001     0.03663108672480824
+           array packing            0.1096807973     0.03413267388416228
+           block caching     0.11072812300000001    0.034458601735241116
+         parallelization            0.1465540376      0.0456076295481858
 
 
 
diff --git a/docs/commit_hash b/docs/commit_hash
index bfd035e1a7..af1b065acc 100644
--- a/docs/commit_hash
+++ b/docs/commit_hash
@@ -1 +1 @@
-be25803178696efd8c38b228ead71cd2e2862808
+bd3d93b6a4f7883c89711b9feda44d205f1e3146
diff --git a/docs/how_to/compile_models/from_darknet.html b/docs/how_to/compile_models/from_darknet.html
index bc8e4fcee5..495746f5d5 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  10.285 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  11.325 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 cbdce2a81e..384f35c1a1 100644
--- a/docs/how_to/compile_models/from_keras.html
+++ b/docs/how_to/compile_models/from_keras.html
@@ -506,7 +506,7 @@ Tensorflow is also required since it’s used as the default backend of keras.</
 <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 941ms/step
+1/1 [==============================] - 1s 923ms/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 806ea622ee..4f1f8160a9 100644
--- a/docs/how_to/compile_models/from_mxnet.html
+++ b/docs/how_to/compile_models/from_mxnet.html
@@ -439,7 +439,7 @@
 <span class="nb">print</span><span class="p">(</span><span class="s2">&quot;x&quot;</span><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#tuple" title="builtins.tuple" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">x</span><span class="o">.</span><span class="n">shape</span></a><span class="p">)</span>
 </pre></div>
 </div>
-<img src="../../_images/sphx_glr_from_mxnet_001.png" srcset="../../_images/sphx_glr_from_mxnet_001.png" alt="from mxnet" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Downloading /workspace/.mxnet/models/resnet18_v1-a0666292.zip9e6df392-f212-4196-b8bc-203b5e5d4f02 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.zipa0e9fe50-cc19-4c5d-a56c-bdaa5f904f8a 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 7bf166fba8..96c7692bd8 100644
--- a/docs/how_to/compile_models/from_oneflow.html
+++ b/docs/how_to/compile_models/from_oneflow.html
@@ -449,13 +449,15 @@ 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]
- 15%|#5        | 6.33M/41.5M [00:00&lt;00:00, 65.9MB/s]
- 30%|###       | 12.6M/41.5M [00:00&lt;00:00, 48.3MB/s]
- 42%|####2     | 17.5M/41.5M [00:00&lt;00:00, 38.2MB/s]
- 58%|#####7    | 24.0M/41.5M [00:00&lt;00:00, 44.2MB/s]
- 77%|#######7  | 32.0M/41.5M [00:00&lt;00:00, 51.3MB/s]
- 92%|#########2| 38.3M/41.5M [00:00&lt;00:00, 37.2MB/s]
-100%|##########| 41.5M/41.5M [00:01&lt;00:00, 42.5MB/s]
+ 15%|#5        | 6.33M/41.5M [00:00&lt;00:00, 53.7MB/s]
+ 28%|##7       | 11.5M/41.5M [00:00&lt;00:00, 44.4MB/s]
+ 38%|###7      | 15.7M/41.5M [00:00&lt;00:00, 35.3MB/s]
+ 46%|####6     | 19.2M/41.5M [00:00&lt;00:00, 32.7MB/s]
+ 54%|#####3    | 22.4M/41.5M [00:00&lt;00:00, 32.5MB/s]
+ 62%|######1   | 25.5M/41.5M [00:00&lt;00:00, 30.2MB/s]
+ 80%|########  | 33.4M/41.5M [00:00&lt;00:00, 44.6MB/s]
+ 96%|#########6| 40.0M/41.5M [00:00&lt;00:00, 51.3MB/s]
+100%|##########| 41.5M/41.5M [00:01&lt;00:00, 42.9MB/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 4e519e4976..99383603dd 100644
--- a/docs/how_to/compile_models/from_pytorch.html
+++ b/docs/how_to/compile_models/from_pytorch.html
@@ -432,11 +432,13 @@ be unstable.</p>
 Downloading: &quot;https://download.pytorch.org/models/resnet18-f37072fd.pth&quot; to /workspace/.cache/torch/hub/checkpoints/resnet18-f37072fd.pth
 
   0%|          | 0.00/44.7M [00:00&lt;?, ?B/s]
- 18%|#7        | 7.99M/44.7M [00:00&lt;00:00, 60.6MB/s]
- 38%|###7      | 16.9M/44.7M [00:00&lt;00:00, 77.1MB/s]
- 63%|######2   | 28.0M/44.7M [00:00&lt;00:00, 94.0MB/s]
- 90%|########9 | 40.0M/44.7M [00:00&lt;00:00, 98.6MB/s]
-100%|##########| 44.7M/44.7M [00:00&lt;00:00, 96.5MB/s]
+ 14%|#4        | 6.30M/44.7M [00:00&lt;00:00, 45.1MB/s]
+ 24%|##3       | 10.6M/44.7M [00:00&lt;00:00, 38.6MB/s]
+ 36%|###5      | 16.0M/44.7M [00:00&lt;00:00, 35.1MB/s]
+ 65%|######5   | 29.0M/44.7M [00:00&lt;00:00, 64.9MB/s]
+ 81%|########1 | 36.2M/44.7M [00:00&lt;00:00, 48.3MB/s]
+ 94%|#########3| 41.9M/44.7M [00:00&lt;00:00, 49.7MB/s]
+100%|##########| 44.7M/44.7M [00:00&lt;00:00, 50.9MB/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 36b04b758b..766bdbbf17 100644
--- a/docs/how_to/compile_models/from_tensorflow.html
+++ b/docs/how_to/compile_models/from_tensorflow.html
@@ -649,7 +649,7 @@ 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  11.018 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  10.912 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 1e32d32024..13511e1433 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>05:37.857</strong> total execution time for <strong>how_to_compile_models</strong> files:</p>
+<p><strong>05:39.585</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_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>01:11.018</p></td>
+<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>01:11.325</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
-<tr class="row-even"><td><p><a class="reference internal" href="from_darknet.html#sphx-glr-how-to-compile-models-from-darknet-py"><span class="std std-ref">Compile YOLO-V2 and YOLO-V3 in DarkNet Models</span></a> (<code class="docutils literal notranslate"><span class="pre">from_darknet.py</span></code>)</p></td>
-<td><p>01:10.285</p></td>
+<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>01:10.912</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>00:45.279</p></td>
+<td><p>00:45.558</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>00:31.605</p></td>
+<td><p>00:31.415</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>00:27.623</p></td>
+<td><p>00:27.592</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>00:26.413</p></td>
+<td><p>00:26.393</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>00:24.887</p></td>
+<td><p>00:25.287</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>00:21.849</p></td>
+<td><p>00:22.227</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>00:16.456</p></td>
+<td><p>00:16.443</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>00:02.442</p></td>
+<td><p>00:02.434</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 </tbody>
diff --git a/docs/how_to/deploy_models/deploy_model_on_adreno.html b/docs/how_to/deploy_models/deploy_model_on_adreno.html
index 65d25bf122..234fbc9e59 100644
--- a/docs/how_to/deploy_models/deploy_model_on_adreno.html
+++ b/docs/how_to/deploy_models/deploy_model_on_adreno.html
@@ -920,7 +920,7 @@ Top5 predictions:
 Evaluate inference time cost...
 Execution time summary:
  mean (ms)   median (ms)    max (ms)     min (ms)     std (ms)
- 2541.6013    2541.3973    2545.4494    2538.4311      2.0426
+ 2542.3067    2540.6301    2552.0888    2538.5837      3.9541
 </pre></div>
 </div>
 <div class="sphx-glr-footer sphx-glr-footer-example docutils container" id="sphx-glr-download-how-to-deploy-models-deploy-model-on-adreno-py">
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 f5c6bced89..a7431cf92b 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)
-  16.1475      15.7952      17.0500      15.6662       0.5307
+  15.7059      15.6497      16.0923      15.5916       0.1471
 </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 0cd35cc26a..a6c5206446 100644
--- a/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
+++ b/docs/how_to/deploy_models/deploy_object_detection_pytorch.html
@@ -454,29 +454,33 @@ 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, 81.6MB/s]
-  9%|9         | 16.0M/170M [00:00&lt;00:02, 64.4MB/s]
- 13%|#3        | 22.4M/170M [00:00&lt;00:02, 56.5MB/s]
- 16%|#6        | 27.9M/170M [00:00&lt;00:02, 52.2MB/s]
- 19%|#9        | 32.9M/170M [00:00&lt;00:03, 37.7MB/s]
- 24%|##3       | 40.0M/170M [00:00&lt;00:03, 45.3MB/s]
- 28%|##8       | 48.0M/170M [00:00&lt;00:02, 52.2MB/s]
- 35%|###5      | 59.9M/170M [00:01&lt;00:01, 70.5MB/s]
- 40%|###9      | 67.4M/170M [00:01&lt;00:01, 55.9MB/s]
- 45%|####4     | 76.4M/170M [00:01&lt;00:01, 64.6MB/s]
- 49%|####9     | 83.5M/170M [00:01&lt;00:01, 54.9MB/s]
- 53%|#####2    | 89.6M/170M [00:01&lt;00:02, 38.4MB/s]
- 56%|#####5    | 94.9M/170M [00:01&lt;00:01, 41.4MB/s]
- 60%|######    | 102M/170M [00:02&lt;00:01, 44.7MB/s]
- 66%|######5   | 112M/170M [00:02&lt;00:01, 49.5MB/s]
- 71%|#######   | 120M/170M [00:02&lt;00:01, 47.0MB/s]
- 79%|#######9  | 134M/170M [00:02&lt;00:00, 67.6MB/s]
- 84%|########3 | 142M/170M [00:02&lt;00:00, 68.0MB/s]
- 88%|########8 | 150M/170M [00:02&lt;00:00, 63.7MB/s]
- 92%|#########2| 157M/170M [00:03&lt;00:00, 61.3MB/s]
- 96%|#########6| 163M/170M [00:03&lt;00:00, 56.9MB/s]
-100%|#########9| 169M/170M [00:03&lt;00:00, 51.2MB/s]
-100%|##########| 170M/170M [00:03&lt;00:00, 53.7MB/s]
+  4%|3         | 6.30M/170M [00:00&lt;00:03, 51.8MB/s]
+  7%|6         | 11.2M/170M [00:00&lt;00:03, 44.9MB/s]
+  9%|9         | 16.0M/170M [00:00&lt;00:04, 36.3MB/s]
+ 14%|#4        | 24.2M/170M [00:00&lt;00:02, 51.5MB/s]
+ 18%|#7        | 30.3M/170M [00:00&lt;00:03, 45.0MB/s]
+ 21%|##        | 35.0M/170M [00:00&lt;00:03, 44.7MB/s]
+ 24%|##3       | 40.0M/170M [00:00&lt;00:02, 46.0MB/s]
+ 28%|##8       | 48.0M/170M [00:01&lt;00:02, 46.2MB/s]
+ 33%|###2      | 56.0M/170M [00:01&lt;00:02, 53.0MB/s]
+ 37%|###6      | 62.4M/170M [00:01&lt;00:01, 56.5MB/s]
+ 40%|####      | 68.0M/170M [00:01&lt;00:02, 51.3MB/s]
+ 44%|####3     | 74.3M/170M [00:01&lt;00:01, 55.0MB/s]
+ 47%|####7     | 80.0M/170M [00:01&lt;00:01, 56.1MB/s]
+ 52%|#####1    | 88.0M/170M [00:01&lt;00:01, 55.3MB/s]
+ 57%|#####6    | 96.0M/170M [00:02&lt;00:01, 51.0MB/s]
+ 60%|######    | 102M/170M [00:02&lt;00:01, 53.3MB/s]
+ 63%|######3   | 108M/170M [00:02&lt;00:01, 52.2MB/s]
+ 66%|######6   | 113M/170M [00:02&lt;00:01, 51.6MB/s]
+ 71%|#######   | 120M/170M [00:02&lt;00:00, 53.8MB/s]
+ 74%|#######4  | 126M/170M [00:02&lt;00:01, 39.9MB/s]
+ 77%|#######6  | 131M/170M [00:02&lt;00:01, 40.6MB/s]
+ 80%|########  | 136M/170M [00:03&lt;00:01, 33.1MB/s]
+ 85%|########4 | 144M/170M [00:03&lt;00:00, 37.3MB/s]
+ 89%|########9 | 152M/170M [00:03&lt;00:00, 45.4MB/s]
+ 92%|#########2| 157M/170M [00:03&lt;00:00, 43.6MB/s]
+ 95%|#########4| 161M/170M [00:03&lt;00:00, 45.0MB/s]
+100%|##########| 170M/170M [00:03&lt;00:00, 48.4MB/s]
 /venv/apache-tvm-py3.7/lib/python3.7/site-packages/torch/nn/functional.py:3897: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
   for i in range(dim)
 /venv/apache-tvm-py3.7/lib/python3.7/site-packages/torchvision/models/detection/anchor_utils.py:124: UserWarning: __floordiv__ is deprecated, and its behavior will change in a future version of pytorch. It currently rounds toward 0 (like the &#39;trunc&#39; function NOT &#39;floor&#39;). This results in incorrect rounding for negative values. To keep the current behavior, use torch.div(a, b, rounding_mode=&#39;trunc&#39;), or for actual floor division, use torch.div(a, b, rounding_mode=& [...]
@@ -574,7 +578,7 @@ 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  13.151 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 3 minutes  16.484 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 300aca659b..d15eaa6997 100644
--- a/docs/how_to/deploy_models/deploy_prequantized.html
+++ b/docs/how_to/deploy_models/deploy_prequantized.html
@@ -495,9 +495,9 @@ 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, 47.8MB/s]
- 93%|#########2| 12.6M/13.6M [00:00&lt;00:00, 33.3MB/s]
-100%|##########| 13.6M/13.6M [00:00&lt;00:00, 37.6MB/s]
+ 59%|#####8    | 7.99M/13.6M [00:00&lt;00:00, 50.5MB/s]
+ 95%|#########4| 12.8M/13.6M [00:00&lt;00:00, 43.5MB/s]
+100%|##########| 13.6M/13.6M [00:00&lt;00:00, 46.8MB/s]
 </pre></div>
 </div>
 </div>
@@ -588,7 +588,7 @@ 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)
-  90.4012      90.2446      96.5908      90.1191       0.7391
+  90.4182      90.2649      95.7405      90.1323       0.6007
 </pre></div>
 </div>
 <div class="admonition note">
@@ -627,7 +627,7 @@ 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  5.979 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  6.526 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 c361188981..9c00e07670 100644
--- a/docs/how_to/deploy_models/deploy_prequantized_tflite.html
+++ b/docs/how_to/deploy_models/deploy_prequantized_tflite.html
@@ -580,7 +580,7 @@ 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)
-  119.8170     119.7858     125.6149     119.0645      0.6544
+  121.2631     121.1364     125.7362     120.5266      0.6134
 </pre></div>
 </div>
 <div class="admonition note">
@@ -608,7 +608,7 @@ 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  23.273 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 2 minutes  23.497 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 afe041ee62..956b960d25 100644
--- a/docs/how_to/deploy_models/deploy_quantized.html
+++ b/docs/how_to/deploy_models/deploy_quantized.html
@@ -521,7 +521,7 @@ 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  32.603 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  37.643 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 53b142e4da..a9382184e9 100644
--- a/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
+++ b/docs/how_to/deploy_models/deploy_ssd_gluoncv.html
@@ -463,23 +463,23 @@ 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]
-  4%|4         | 5381/132723 [00:00&lt;00:02, 53803.00KB/s]
- 10%|#         | 13873/132723 [00:00&lt;00:01, 72100.72KB/s]
- 16%|#5        | 21084/132723 [00:00&lt;00:01, 68158.10KB/s]
- 21%|##1       | 27924/132723 [00:00&lt;00:01, 66563.51KB/s]
- 27%|##7       | 36329/132723 [00:00&lt;00:01, 72625.91KB/s]
- 34%|###3      | 44777/132723 [00:00&lt;00:01, 76552.05KB/s]
- 40%|###9      | 53028/132723 [00:00&lt;00:01, 78463.47KB/s]
- 46%|####6     | 61464/132723 [00:00&lt;00:00, 80315.81KB/s]
- 52%|#####2    | 69515/132723 [00:00&lt;00:00, 73191.08KB/s]
- 58%|#####8    | 77641/132723 [00:01&lt;00:00, 75507.02KB/s]
- 65%|######4   | 86065/132723 [00:01&lt;00:00, 78046.65KB/s]
- 71%|#######1  | 94519/132723 [00:01&lt;00:00, 79952.30KB/s]
- 78%|#######7  | 103014/132723 [00:01&lt;00:00, 81426.76KB/s]
- 84%|########4 | 111532/132723 [00:01&lt;00:00, 82538.36KB/s]
- 90%|######### | 120053/132723 [00:01&lt;00:00, 83332.67KB/s]
- 97%|#########6| 128544/132723 [00:01&lt;00:00, 83801.60KB/s]
-100%|##########| 132723/132723 [00:01&lt;00:00, 77987.16KB/s]
+  5%|4         | 6456/132723 [00:00&lt;00:01, 64553.60KB/s]
+ 11%|#1        | 15152/132723 [00:00&lt;00:01, 66245.19KB/s]
+ 16%|#6        | 21761/132723 [00:00&lt;00:02, 47253.76KB/s]
+ 22%|##2       | 29611/132723 [00:00&lt;00:01, 56581.10KB/s]
+ 28%|##8       | 37380/132723 [00:00&lt;00:01, 62890.86KB/s]
+ 35%|###4      | 46037/132723 [00:00&lt;00:01, 69976.68KB/s]
+ 41%|####1     | 54715/132723 [00:00&lt;00:01, 75008.10KB/s]
+ 48%|####7     | 63417/132723 [00:00&lt;00:00, 78604.10KB/s]
+ 54%|#####3    | 71521/132723 [00:01&lt;00:00, 62891.87KB/s]
+ 60%|######    | 80225/132723 [00:01&lt;00:00, 69004.12KB/s]
+ 66%|######6   | 87697/132723 [00:01&lt;00:00, 68583.53KB/s]
+ 73%|#######2  | 96285/132723 [00:01&lt;00:00, 73246.35KB/s]
+ 78%|#######8  | 103944/132723 [00:01&lt;00:00, 53890.48KB/s]
+ 85%|########4 | 112611/132723 [00:01&lt;00:00, 61231.42KB/s]
+ 91%|#########1| 121266/132723 [00:01&lt;00:00, 67367.24KB/s]
+ 98%|#########7| 129994/132723 [00:01&lt;00:00, 72488.40KB/s]
+100%|##########| 132723/132723 [00:01&lt;00:00, 66630.50KB/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  4.005 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  6.813 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 7f3f775405..25fd7b15da 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>13:34.566</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
+<p><strong>13:47.571</strong> total execution time for <strong>how_to_deploy_models</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 86%" />
@@ -349,39 +349,39 @@
 </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>03:13.151</p></td>
+<td><p>03:16.484</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>03:04.005</p></td>
+<td><p>03:06.813</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>02:23.273</p></td>
+<td><p>02:23.497</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>01:32.603</p></td>
+<td><p>01:37.643</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>01:05.979</p></td>
+<td><p>01:06.526</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="deploy_model_on_adreno.html#sphx-glr-how-to-deploy-models-deploy-model-on-adreno-py"><span class="std std-ref">Deploy the Pretrained Model on Adreno</span></a> (<code class="docutils literal notranslate"><span class="pre">deploy_model_on_adreno.py</span></code>)</p></td>
-<td><p>00:51.283</p></td>
+<td><p>00:51.450</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><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>00:35.130</p></td>
+<td><p>00:35.591</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><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>00:24.787</p></td>
+<td><p>00:25.014</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><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>00:24.348</p></td>
+<td><p>00:24.545</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><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 73fd848801..beacdea142 100644
--- a/docs/how_to/extend_tvm/bring_your_own_datatypes.html
+++ b/docs/how_to/extend_tvm/bring_your_own_datatypes.html
@@ -619,7 +619,7 @@ 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.zip6802e823-ec81-46cd-8187-6ab788d3cd23 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.zip2bdf695d-1f09-432f-aef8-19c8b53f9e48 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 128ade55d2..766d5b5dd5 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>00:47.943</strong> total execution time for <strong>how_to_extend_tvm</strong> files:</p>
+<p><strong>00:47.939</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>00:44.447</p></td>
+<td><p>00:44.449</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>00:02.437</p></td>
+<td><p>00:02.446</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>00:01.052</p></td>
+<td><p>00:01.038</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 b05b94ca55..2d3b521d62 100644
--- a/docs/how_to/extend_tvm/use_pass_instrument.html
+++ b/docs/how_to/extend_tvm/use_pass_instrument.html
@@ -526,10 +526,10 @@ 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: 7246us [7246us] (46.57%; 46.57%)
-FoldScaleAxis: 8312us [7us] (53.43%; 53.43%)
-        FoldConstant: 8306us [1714us] (53.38%; 99.92%)
-                InferType: 6592us [6592us] (42.37%; 79.37%)
+InferType: 7769us [7769us] (47.25%; 47.25%)
+FoldScaleAxis: 8674us [8us] (52.75%; 52.75%)
+        FoldConstant: 8666us [1910us] (52.71%; 99.91%)
+                InferType: 6757us [6757us] (41.09%; 77.96%)
 </pre></div>
 </div>
 </div>
@@ -551,10 +551,10 @@ 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: 6673us [6673us] (44.91%; 44.91%)
-FoldScaleAxis: 8186us [5us] (55.09%; 55.09%)
-        FoldConstant: 8180us [1689us] (55.05%; 99.94%)
-                InferType: 6492us [6492us] (43.69%; 79.36%)
+InferType: 6741us [6741us] (44.15%; 44.15%)
+FoldScaleAxis: 8526us [5us] (55.85%; 55.85%)
+        FoldConstant: 8521us [1900us] (55.81%; 99.94%)
+                InferType: 6621us [6621us] (43.37%; 77.70%)
 </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 d037be62c6..a63830b2da 100644
--- a/docs/how_to/optimize_operators/opt_conv_cuda.html
+++ b/docs/how_to/optimize_operators/opt_conv_cuda.html
@@ -575,7 +575,7 @@ 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: 44.052543 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Convolution: 54.173633 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 20eeac4243..e46ca945aa 100644
--- a/docs/how_to/optimize_operators/opt_conv_tensorcore.html
+++ b/docs/how_to/optimize_operators/opt_conv_tensorcore.html
@@ -915,7 +915,7 @@ 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: 13.369545 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>conv2d with tensor core: 6.566064 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 b0090a3691..b1cbd5333a 100644
--- a/docs/how_to/optimize_operators/opt_gemm.html
+++ b/docs/how_to/optimize_operators/opt_gemm.html
@@ -472,8 +472,8 @@ 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: 0.018431
-Baseline: 3.208171
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.018273
+Baseline: 3.226653
 </pre></div>
 </div>
 <p>In TVM, we can always inspect lower level IR to debug or optimize our schedule.
@@ -532,7 +532,7 @@ 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: 0.296398
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt1: 0.298855
 </pre></div>
 </div>
 <p>Here is the generated IR after blocking.</p>
@@ -598,7 +598,7 @@ 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: 0.330635
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt2: 0.337102
 </pre></div>
 </div>
 <p>Here is the generated IR after vectorization.</p>
@@ -658,7 +658,7 @@ 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: 0.114155
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt3: 0.115017
 </pre></div>
 </div>
 <p>Here is the generated IR after loop permutation.</p>
@@ -740,7 +740,7 @@ 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: 0.110279
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt4: 0.108878
 </pre></div>
 </div>
 <p>Here is the generated IR after array packing.</p>
@@ -825,7 +825,7 @@ 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: 0.111988
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt5: 0.111888
 </pre></div>
 </div>
 <p>Here is the generated IR after blocking.</p>
@@ -914,7 +914,7 @@ 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: 0.146003
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Opt6: 0.146227
 </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 a407a0025c..70ca699260 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>00:34.116</strong> total execution time for <strong>how_to_optimize_operators</strong> files:</p>
+<p><strong>00:34.226</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>00:31.510</p></td>
+<td><p>00:31.682</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>00:01.537</p></td>
+<td><p>00:01.432</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>00:01.069</p></td>
+<td><p>00:01.112</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 bcc851e93f..1e656559d8 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>09:10.972</strong> total execution time for <strong>how_to_tune_with_autoscheduler</strong> files:</p>
+<p><strong>08:52.998</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>05:38.343</p></td>
+<td><p>05:29.208</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>01:30.980</p></td>
+<td><p>01:31.583</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>01:01.132</p></td>
+<td><p>01:01.747</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>00:37.743</p></td>
+<td><p>00:26.987</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>00:11.847</p></td>
+<td><p>00:12.168</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>00:10.927</p></td>
+<td><p>00:11.306</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 1378003fa2..8f97392578 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
@@ -489,9 +489,6 @@ file and apply it.</p>
 <span class="k">del</span> <span class="n">measure_ctx</span>
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>.T
-</pre></div>
-</div>
 <p>We can lower the schedule to see the IR after auto-scheduling.
 The auto-scheduler correctly performs optimizations including multi-level tiling,
 cooperative fetching, unrolling and operator fusion.</p>
@@ -507,399 +504,104 @@ cooperative fetching, unrolling and operator fusion.</p>
              bias: Buffer(bias_2: Pointer(float32), float32, [1, 512, 1, 1], []),
              compute: Buffer(compute_2: Pointer(float32), float32, [1, 512, 7, 7], [])}
   buffer_map = {data_1: data, kernel_1: kernel, bias_1: bias, compute_1: compute} {
-  attr [IterVar(blockIdx.x: int32, (nullptr), &quot;ThreadIndex&quot;, &quot;blockIdx.x&quot;)] &quot;thread_extent&quot; = 8;
-  allocate(conv2d_nchw: Pointer(local float32), float32, [14]), storage_scope = local;
-  allocate(pad_temp.shared: Pointer(shared float32), float32, [504]), storage_scope = shared;
-  allocate(kernel.shared: Pointer(shared float32), float32, [1536]), storage_scope = shared;
+  attr [IterVar(blockIdx.x: int32, (nullptr), &quot;ThreadIndex&quot;, &quot;blockIdx.x&quot;)] &quot;thread_extent&quot; = 56;
+  allocate(conv2d_nchw: Pointer(local float32), float32, [2]), storage_scope = local;
+  allocate(pad_temp.shared: Pointer(shared float32), float32, [216]), storage_scope = shared;
+  allocate(kernel.shared: Pointer(shared float32), float32, [4608]), storage_scope = shared;
   attr [IterVar(threadIdx.x: int32, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224 {
-    conv2d_nchw_1: Buffer(conv2d_nchw, float32, [49], [], scope=&quot;local&quot;, align=16)[0] = 0f32
-    conv2d_nchw_1[7] = 0f32
+    conv2d_nchw_1: Buffer(conv2d_nchw, float32, [1], [], scope=&quot;local&quot;, align=4)[0] = 0f32
     conv2d_nchw_1[1] = 0f32
-    conv2d_nchw_1[8] = 0f32
-    conv2d_nchw_1[2] = 0f32
-    conv2d_nchw_1[9] = 0f32
-    conv2d_nchw_1[3] = 0f32
-    conv2d_nchw_1[10] = 0f32
-    conv2d_nchw_1[4] = 0f32
-    conv2d_nchw_1[11] = 0f32
-    conv2d_nchw_1[5] = 0f32
-    conv2d_nchw_1[12] = 0f32
-    conv2d_nchw_1[6] = 0f32
-    conv2d_nchw_1[13] = 0f32
     for (rc.outer.outer: int32, 0, 64) {
-      for (ry.outer.outer: int32, 0, 3) {
-        let cse_var_4: int32 = (rc.outer.outer*392)
-        let cse_var_3: int32 = (ry.outer.outer*7)
-        let cse_var_2: int32 = (rc.outer.outer*72)
-        let cse_var_1: int32 = (ry.outer.outer*3)
-         {
-          attr [IterVar(threadIdx.x_1: int32, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
-          pad_temp.shared_1: Buffer(pad_temp.shared, float32, [504], [], scope=&quot;shared&quot;)[threadIdx.x_1] = @tir.if_then_else(((((1 &lt;= (floordiv(floormod(threadIdx.x_1, 63), 9) + ry.outer.outer)) &amp;&amp; ((floordiv(floormod(threadIdx.x_1, 63), 9) + ry.outer.outer) &lt; 8)) &amp;&amp; (1 &lt;= floormod(threadIdx.x_1, 9))) &amp;&amp; (floormod(threadIdx.x_1, 9) &lt; 8)), data_3: Buffer(data_2, float32, [25088], [])[((((cse_var_4 + (floordiv(threadIdx.x_1, 9)*7)) + cse_var_3)  [...]
-          attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
-          pad_temp.shared_1[(threadIdx.x_1 + 224)] = @tir.if_then_else(((((1 &lt;= (floordiv(floormod((threadIdx.x_1 + 35), 63), 9) + ry.outer.outer)) &amp;&amp; ((floordiv(floormod((threadIdx.x_1 + 35), 63), 9) + ry.outer.outer) &lt; 8)) &amp;&amp; (1 &lt;= floormod((threadIdx.x_1 + 8), 9))) &amp;&amp; (floormod((threadIdx.x_1 + 8), 9) &lt; 8)), data_3[((((cse_var_4 + (floordiv((threadIdx.x_1 + 224), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 8), 9)) - 8)], 0f32, dtype=float32)
-          attr [IterVar(threadIdx.x_1, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
-          if @tir.likely((threadIdx.x_1 &lt; 56), dtype=bool) {
-            pad_temp.shared_1[(threadIdx.x_1 + 448)] = @tir.if_then_else(((((1 &lt;= (floordiv(floormod((threadIdx.x_1 + 7), 63), 9) + ry.outer.outer)) &amp;&amp; ((floordiv(floormod((threadIdx.x_1 + 7), 63), 9) + ry.outer.outer) &lt; 8)) &amp;&amp; (1 &lt;= floormod((threadIdx.x_1 + 7), 9))) &amp;&amp; (floormod((threadIdx.x_1 + 7), 9) &lt; 8)), data_3[((((cse_var_4 + (floordiv((threadIdx.x_1 + 448), 9)*7)) + cse_var_3) + floormod((threadIdx.x_1 + 7), 9)) - 8)], 0f32, dtype=float32)
+      let cse_var_1: int32 = (rc.outer.outer*72)
+       {
+        attr [IterVar(threadIdx.x_1: int32, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224 {
+          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
+            pad_temp.shared_1: Buffer(pad_temp.shared, float32, [216], [], scope=&quot;shared&quot;)[(threadIdx.x_1*6)] = @tir.if_then_else(((((1 &lt;= (floordiv(floormod((threadIdx.x_1*2), 9), 3) + floormod(blockIdx.x, 7))) &amp;&amp; ((floordiv(floormod((threadIdx.x_1*2), 9), 3) + floormod(blockIdx.x, 7)) &lt; 8)) &amp;&amp; (1 &lt; floormod((threadIdx.x_1*6), 9))) &amp;&amp; (floormod((threadIdx.x_1*6), 9) &lt; 8)), data_3: Buffer(data_2, float32, [25088], [])[((((((rc.outer.outer*392 [...]
+          }
+          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
+            pad_temp.shared_1[((threadIdx.x_1*6) + 1)] = @tir.if_then_else(((((1 &lt;= (floordiv(floormod((threadIdx.x_1*2), 9), 3) + floormod(blockIdx.x, 7))) &amp;&amp; ((floordiv(floormod((threadIdx.x_1*2), 9), 3) + floormod(blockIdx.x, 7)) &lt; 8)) &amp;&amp; (1 &lt;= floormod(((threadIdx.x_1*6) + 1), 9))) &amp;&amp; (floormod(((threadIdx.x_1*6) + 1), 9) &lt; 8)), data_3[((((((rc.outer.outer*392) + (floordiv((threadIdx.x_1*2), 9)*49)) + (floordiv(floormod((threadIdx.x_1*2), 9), 3)*7) [...]
+          }
+          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
+            pad_temp.shared_1[((threadIdx.x_1*6) + 2)] = @tir.if_then_else(((((1 &lt;= (floordiv(floormod((threadIdx.x_1*2), 9), 3) + floormod(blockIdx.x, 7))) &amp;&amp; ((floordiv(floormod((threadIdx.x_1*2), 9), 3) + floormod(blockIdx.x, 7)) &lt; 8)) &amp;&amp; (1 &lt; floormod(((threadIdx.x_1*6) + 2), 9))) &amp;&amp; (floormod(((threadIdx.x_1*6) + 2), 9) &lt; 8)), data_3[((((((rc.outer.outer*392) + (floordiv((threadIdx.x_1*2), 9)*49)) + (floordiv(floormod((threadIdx.x_1*2), 9), 3)*7)) [...]
+          }
+          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
+            pad_temp.shared_1[((threadIdx.x_1*6) + 3)] = @tir.if_then_else(((((1 &lt;= (floordiv(floormod(((threadIdx.x_1*2) + 1), 9), 3) + floormod(blockIdx.x, 7))) &amp;&amp; ((floordiv(floormod(((threadIdx.x_1*2) + 1), 9), 3) + floormod(blockIdx.x, 7)) &lt; 8)) &amp;&amp; (1 &lt; floormod(((threadIdx.x_1*6) + 3), 9))) &amp;&amp; (floormod(((threadIdx.x_1*6) + 3), 9) &lt; 8)), data_3[((((((rc.outer.outer*392) + (floordiv(((threadIdx.x_1*2) + 1), 9)*49)) + (floordiv(floormod(((threadIdx [...]
           }
-          attr [IterVar(threadIdx.x_2: int32, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
-          kernel.shared_1: Buffer(kernel.shared, float32, [1536], [], scope=&quot;shared&quot;)[threadIdx.x_2] = kernel_3: Buffer(kernel_2, float32, [2359296], [])[((((((blockIdx.x*294912) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3))]
-          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
-          kernel.shared_1[(threadIdx.x_2 + 224)] = kernel_3[((((((blockIdx.x*294912) + (floordiv((threadIdx.x_2 + 224), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
-          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
-          kernel.shared_1[(threadIdx.x_2 + 448)] = kernel_3[((((((blockIdx.x*294912) + (floordiv((threadIdx.x_2 + 448), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
-          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
-          kernel.shared_1[(threadIdx.x_2 + 672)] = kernel_3[(((((((blockIdx.x*294912) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 129024)]
-          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
-          kernel.shared_1[(threadIdx.x_2 + 896)] = kernel_3[((((((blockIdx.x*294912) + (floordiv((threadIdx.x_2 + 896), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 8), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 2), 3))]
-          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
-          kernel.shared_1[(threadIdx.x_2 + 1120)] = kernel_3[((((((blockIdx.x*294912) + (floordiv((threadIdx.x_2 + 1120), 24)*4608)) + cse_var_2) + (floordiv(floormod((threadIdx.x_2 + 16), 24), 3)*9)) + cse_var_1) + floormod((threadIdx.x_2 + 1), 3))]
-          attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
-          if @tir.likely((threadIdx.x_2 &lt; 192), dtype=bool) {
-            kernel.shared_1[(threadIdx.x_2 + 1344)] = kernel_3[(((((((blockIdx.x*294912) + (floordiv(threadIdx.x_2, 24)*4608)) + cse_var_2) + (floordiv(floormod(threadIdx.x_2, 24), 3)*9)) + cse_var_1) + floormod(threadIdx.x_2, 3)) + 258048)]
+          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
+            pad_temp.shared_1[((threadIdx.x_1*6) + 4)] = @tir.if_then_else(((((1 &lt;= (floordiv(floormod(((threadIdx.x_1*2) + 1), 9), 3) + floormod(blockIdx.x, 7))) &amp;&amp; ((floordiv(floormod(((threadIdx.x_1*2) + 1), 9), 3) + floormod(blockIdx.x, 7)) &lt; 8)) &amp;&amp; (1 &lt;= floormod(((threadIdx.x_1*6) + 4), 9))) &amp;&amp; (floormod(((threadIdx.x_1*6) + 4), 9) &lt; 8)), data_3[((((((rc.outer.outer*392) + (floordiv(((threadIdx.x_1*2) + 1), 9)*49)) + (floordiv(floormod(((threadId [...]
           }
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(floormod(threadIdx.x, 7)*9)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
-          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[(floormod(threadIdx.x, 7)*9)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 768)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 1)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
-          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 1)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 769)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 2)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
-          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 2)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 770)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
-          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 63)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 771)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 64)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
-          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 64)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 772)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
-          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 773)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
-          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 126)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 774)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
-          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 775)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
-          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 776)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
-          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 189)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 777)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
-          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 778)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
-          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 779)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
-          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 252)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 780)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
-          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 781)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
-          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 782)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
-          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 315)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 783)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 316)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
-          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 316)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 784)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 317)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
-          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 317)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 785)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
-          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 378)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 786)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 379)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
-          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 379)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 787)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 380)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
-          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 380)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 788)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
-          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 441)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 789)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 442)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
-          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 442)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 790)]))
-          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 443)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
-          conv2d_nchw_1[7] = (conv2d_nchw_1[7] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 443)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 791)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 1)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
-          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 1)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 768)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 2)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
-          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 2)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 769)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 3)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
-          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 3)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 770)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 64)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
-          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 64)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 771)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
-          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 772)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 66)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
-          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 66)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 773)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
-          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 127)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 774)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
-          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 775)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 129)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
-          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 129)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 776)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
-          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 190)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 777)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
-          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 778)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 192)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
-          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 192)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 779)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
-          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 253)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 780)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
-          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 781)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 255)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
-          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 255)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 782)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 316)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
-          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 316)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 783)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 317)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
-          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 317)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 784)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 318)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
-          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 318)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 785)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 379)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
-          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 379)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 786)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 380)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
-          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 380)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 787)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 381)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
-          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 381)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 788)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 442)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
-          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 442)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 789)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 443)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
-          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 443)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 790)]))
-          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 444)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
-          conv2d_nchw_1[8] = (conv2d_nchw_1[8] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 444)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 791)]))
-          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 2)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
-          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 2)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 768)]))
-          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 3)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
-          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 3)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 769)]))
-          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 4)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
-          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 4)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 770)]))
-          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
-          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 65)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 771)]))
-          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 66)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
-          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 66)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 772)]))
-          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 67)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
-          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 67)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 773)]))
-          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
-          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 128)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 774)]))
-          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 129)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
-          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 129)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 775)]))
-          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 130)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
-          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 130)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 776)]))
-          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
-          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 191)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 777)]))
-          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 192)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
-          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 192)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 778)]))
-          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 193)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
-          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 193)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 779)]))
-          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
-          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 254)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 780)]))
-          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 255)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
-          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 255)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 781)]))
-          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 256)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
-          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 256)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 782)]))
-          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 317)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
-          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 317)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 783)]))
-          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 318)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
-          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 318)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 784)]))
-          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 319)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
-          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 319)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 785)]))
-          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 380)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
-          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 380)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 786)]))
-          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 381)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
-          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 381)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 787)]))
-          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 382)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
-          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 382)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 788)]))
-          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 443)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
-          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 443)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 789)]))
-          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 444)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
-          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 444)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 790)]))
-          conv2d_nchw_1[2] = (conv2d_nchw_1[2] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 445)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
-          conv2d_nchw_1[9] = (conv2d_nchw_1[9] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 445)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 791)]))
-          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 3)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
-          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 3)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 768)]))
-          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 4)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
-          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 4)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 769)]))
-          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 5)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
-          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 5)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 770)]))
-          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 66)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
-          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 66)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 771)]))
-          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 67)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
-          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 67)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 772)]))
-          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 68)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
-          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 68)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 773)]))
-          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 129)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
-          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 129)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 774)]))
-          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 130)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
-          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 130)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 775)]))
-          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 131)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
-          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 131)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 776)]))
-          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 192)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
-          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 192)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 777)]))
-          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 193)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
-          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 193)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 778)]))
-          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 194)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
-          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 194)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 779)]))
-          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 255)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
-          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 255)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 780)]))
-          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 256)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
-          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 256)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 781)]))
-          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 257)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
-          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 257)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 782)]))
-          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 318)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
-          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 318)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 783)]))
-          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 319)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
-          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 319)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 784)]))
-          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 320)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
-          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 320)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 785)]))
-          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 381)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
-          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 381)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 786)]))
-          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 382)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
-          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 382)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 787)]))
-          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 383)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
-          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 383)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 788)]))
-          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 444)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
-          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 444)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 789)]))
-          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 445)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
-          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 445)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 790)]))
-          conv2d_nchw_1[3] = (conv2d_nchw_1[3] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 446)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
-          conv2d_nchw_1[10] = (conv2d_nchw_1[10] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 446)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 791)]))
-          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 4)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
-          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 4)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 768)]))
-          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 5)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
-          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 5)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 769)]))
-          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 6)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
-          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 6)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 770)]))
-          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 67)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
-          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 67)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 771)]))
-          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 68)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
-          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 68)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 772)]))
-          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 69)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
-          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 69)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 773)]))
-          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 130)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
-          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 130)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 774)]))
-          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 131)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
-          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 131)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 775)]))
-          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 132)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
-          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 132)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 776)]))
-          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 193)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
-          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 193)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 777)]))
-          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 194)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
-          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 194)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 778)]))
-          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 195)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
-          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 195)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 779)]))
-          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 256)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
-          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 256)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 780)]))
-          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 257)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
-          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 257)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 781)]))
-          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 258)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
-          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 258)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 782)]))
-          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 319)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
-          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 319)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 783)]))
-          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 320)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
-          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 320)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 784)]))
-          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 321)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
-          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 321)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 785)]))
-          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 382)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
-          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 382)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 786)]))
-          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 383)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
-          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 383)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 787)]))
-          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 384)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
-          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 384)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 788)]))
-          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 445)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
-          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 445)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 789)]))
-          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 446)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
-          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 446)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 790)]))
-          conv2d_nchw_1[4] = (conv2d_nchw_1[4] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 447)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
-          conv2d_nchw_1[11] = (conv2d_nchw_1[11] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 447)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 791)]))
-          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 5)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
-          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 5)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 768)]))
-          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 6)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
-          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 6)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 769)]))
-          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
-          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 770)]))
-          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 68)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
-          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 68)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 771)]))
-          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 69)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
-          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 69)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 772)]))
-          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
-          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 773)]))
-          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 131)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
-          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 131)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 774)]))
-          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 132)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
-          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 132)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 775)]))
-          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
-          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 776)]))
-          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 194)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
-          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 194)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 777)]))
-          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 195)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
-          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 195)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 778)]))
-          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
-          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 779)]))
-          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 257)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
-          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 257)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 780)]))
-          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 258)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
-          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 258)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 781)]))
-          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
-          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 782)]))
-          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 320)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
-          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 320)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 783)]))
-          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 321)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
-          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 321)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 784)]))
-          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
-          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 785)]))
-          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 383)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
-          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 383)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 786)]))
-          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 384)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
-          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 384)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 787)]))
-          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
-          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 788)]))
-          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 446)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
-          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 446)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 789)]))
-          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 447)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
-          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 447)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 790)]))
-          conv2d_nchw_1[5] = (conv2d_nchw_1[5] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
-          conv2d_nchw_1[12] = (conv2d_nchw_1[12] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 791)]))
-          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 6)]*kernel.shared_1[(floordiv(threadIdx.x, 7)*24)]))
-          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 6)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 768)]))
-          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 1)]))
-          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 7)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 769)]))
-          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 8)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 2)]))
-          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 8)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 770)]))
-          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 69)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 3)]))
-          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 69)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 771)]))
-          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 4)]))
-          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 70)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 772)]))
-          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 71)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 5)]))
-          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 71)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 773)]))
-          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 132)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 6)]))
-          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 132)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 774)]))
-          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 7)]))
-          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 133)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 775)]))
-          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 134)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 8)]))
-          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 134)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 776)]))
-          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 195)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 9)]))
-          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 195)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 777)]))
-          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 10)]))
-          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 196)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 778)]))
-          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 197)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 11)]))
-          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 197)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 779)]))
-          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 258)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 12)]))
-          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 258)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 780)]))
-          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 13)]))
-          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 259)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 781)]))
-          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 260)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 14)]))
-          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 260)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 782)]))
-          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 321)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 15)]))
-          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 321)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 783)]))
-          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 16)]))
-          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 322)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 784)]))
-          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 323)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 17)]))
-          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 323)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 785)]))
-          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 384)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 18)]))
-          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 384)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 786)]))
-          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 19)]))
-          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 385)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 787)]))
-          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 386)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 20)]))
-          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 386)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 788)]))
-          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 447)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 21)]))
-          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 447)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 789)]))
-          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 22)]))
-          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 448)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 790)]))
-          conv2d_nchw_1[6] = (conv2d_nchw_1[6] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 449)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 23)]))
-          conv2d_nchw_1[13] = (conv2d_nchw_1[13] + (pad_temp.shared_1[((floormod(threadIdx.x, 7)*9) + 449)]*kernel.shared_1[((floordiv(threadIdx.x, 7)*24) + 791)]))
+          if @tir.likely((threadIdx.x_1 &lt; 36), dtype=bool) {
+            pad_temp.shared_1[((threadIdx.x_1*6) + 5)] = @tir.if_then_else(((((1 &lt;= (floordiv(floormod(((threadIdx.x_1*2) + 1), 9), 3) + floormod(blockIdx.x, 7))) &amp;&amp; ((floordiv(floormod(((threadIdx.x_1*2) + 1), 9), 3) + floormod(blockIdx.x, 7)) &lt; 8)) &amp;&amp; (1 &lt; floormod(((threadIdx.x_1*6) + 5), 9))) &amp;&amp; (floormod(((threadIdx.x_1*6) + 5), 9) &lt; 8)), data_3[((((((rc.outer.outer*392) + (floordiv(((threadIdx.x_1*2) + 1), 9)*49)) + (floordiv(floormod(((threadIdx [...]
+          }
+        }
+        attr [IterVar(threadIdx.x_2: int32, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
+        kernel.shared_1: Buffer(kernel.shared, float32, [4608], [], scope=&quot;shared&quot;)[threadIdx.x_2] = kernel_3: Buffer(kernel_2, float32, [2359296], [])[((((floordiv(blockIdx.x, 7)*294912) + (floordiv(threadIdx.x_2, 72)*4608)) + cse_var_1) + floormod(threadIdx.x_2, 72))]
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
+        kernel.shared_1[(threadIdx.x_2 + 224)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 224), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 8), 72), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
+        kernel.shared_1[(threadIdx.x_2 + 448)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 448), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 16), 72), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
+        kernel.shared_1[(threadIdx.x_2 + 672)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 672), 72)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 8), 24)*3)) + floormod(threadIdx.x_2, 3))]
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
+        kernel.shared_1[(threadIdx.x_2 + 896)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 896), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 32), 72), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
+        kernel.shared_1[(threadIdx.x_2 + 1120)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 1120), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 40), 72), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
+        kernel.shared_1[(threadIdx.x_2 + 1344)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 1344), 72)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 16), 24)*3)) + floormod(threadIdx.x_2, 3))]
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
+        kernel.shared_1[(threadIdx.x_2 + 1568)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 1568), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 56), 72), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
+        kernel.shared_1[(threadIdx.x_2 + 1792)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 1792), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 64), 72), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
+        kernel.shared_1[(threadIdx.x_2 + 2016)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv(threadIdx.x_2, 72)*4608)) + cse_var_1) + floormod(threadIdx.x_2, 72)) + 129024)]
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
+        kernel.shared_1[(threadIdx.x_2 + 2240)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 2240), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 8), 72), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
+        kernel.shared_1[(threadIdx.x_2 + 2464)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 2464), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 16), 72), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
+        kernel.shared_1[(threadIdx.x_2 + 2688)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 2688), 72)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 8), 24)*3)) + floormod(threadIdx.x_2, 3))]
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
+        kernel.shared_1[(threadIdx.x_2 + 2912)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 2912), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 32), 72), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
+        kernel.shared_1[(threadIdx.x_2 + 3136)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 3136), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 40), 72), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
+        kernel.shared_1[(threadIdx.x_2 + 3360)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 3360), 72)*4608)) + cse_var_1) + (floormod((floordiv(threadIdx.x_2, 3) + 16), 24)*3)) + floormod(threadIdx.x_2, 3))]
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
+        kernel.shared_1[(threadIdx.x_2 + 3584)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 3584), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 56), 72), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
+        kernel.shared_1[(threadIdx.x_2 + 3808)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 3808), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 64), 72), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
+        kernel.shared_1[(threadIdx.x_2 + 4032)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv(threadIdx.x_2, 72)*4608)) + cse_var_1) + floormod(threadIdx.x_2, 72)) + 258048)]
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
+        kernel.shared_1[(threadIdx.x_2 + 4256)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 4256), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 8), 72), 3)*3)) + floormod((threadIdx.x_2 + 2), 3))]
+        attr [IterVar(threadIdx.x_2, (nullptr), &quot;ThreadIndex&quot;, &quot;threadIdx.x&quot;)] &quot;thread_extent&quot; = 224;
+        if @tir.likely((threadIdx.x_2 &lt; 128), dtype=bool) {
+          kernel.shared_1[(threadIdx.x_2 + 4480)] = kernel_3[(((((floordiv(blockIdx.x, 7)*294912) + (floordiv((threadIdx.x_2 + 4480), 72)*4608)) + cse_var_1) + (floordiv(floormod((threadIdx.x_2 + 16), 72), 3)*3)) + floormod((threadIdx.x_2 + 1), 3))]
+        }
+        for (rc.outer.inner: int32, 0, 8) {
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[((rc.outer.inner*27) + floormod(threadIdx.x, 7))]*kernel.shared_1[((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9))]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[((rc.outer.inner*27) + floormod(threadIdx.x, 7))]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 2304)]))
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 1)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 1)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 2305)]))
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 2)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 2)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 2306)]))
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 3)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 9)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 2307)]))
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 10)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 4)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 10)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 2308)]))
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 11)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 5)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 11)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 2309)]))
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 6)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 18)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 2310)]))
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 19)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 7)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 19)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 2311)]))
+          conv2d_nchw_1[0] = (conv2d_nchw_1[0] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 20)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 8)]))
+          conv2d_nchw_1[1] = (conv2d_nchw_1[1] + (pad_temp.shared_1[(((rc.outer.inner*27) + floormod(threadIdx.x, 7)) + 20)]*kernel.shared_1[(((floordiv(threadIdx.x, 7)*72) + (rc.outer.inner*9)) + 2312)]))
         }
       }
     }
-    for (i3.inner: int32, 0, 7) {
-      compute_3: Buffer(compute_2, float32, [25088], [])[(((blockIdx.x*3136) + (threadIdx.x*7)) + i3.inner)] = max((conv2d_nchw_1[i3.inner] + bias_3: Buffer(bias_2, float32, [512], [])[((blockIdx.x*64) + floordiv(threadIdx.x, 7))]), 0f32)
-      compute_3[((((blockIdx.x*3136) + (threadIdx.x*7)) + i3.inner) + 1568)] = max((conv2d_nchw_1[(i3.inner + 7)] + bias_3[(((blockIdx.x*64) + floordiv(threadIdx.x, 7)) + 32)]), 0f32)
-    }
+    compute_3: Buffer(compute_2, float32, [25088], [])[((((floordiv(blockIdx.x, 7)*3136) + (floordiv(threadIdx.x, 7)*49)) + (floormod(blockIdx.x, 7)*7)) + floormod(threadIdx.x, 7))] = max((conv2d_nchw_1[0] + bias_3: Buffer(bias_2, float32, [512], [])[((floordiv(blockIdx.x, 7)*64) + floordiv(threadIdx.x, 7))]), 0f32)
+    compute_3[(((((floordiv(blockIdx.x, 7)*3136) + (floordiv(threadIdx.x, 7)*49)) + (floormod(blockIdx.x, 7)*7)) + floormod(threadIdx.x, 7)) + 1568)] = max((conv2d_nchw_1[1] + bias_3[(((floordiv(blockIdx.x, 7)*64) + floordiv(threadIdx.x, 7)) + 32)]), 0f32)
   }
 }
 </pre></div>
@@ -935,7 +637,7 @@ 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: 0.335 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 0.350 ms
 </pre></div>
 </div>
 </div>
@@ -970,18 +672,18 @@ conv2d_nchw_ff_o_o_o_i, conv2d_nchw_ff_o_o_i = s[conv2d_nchw].split(conv2d_nchw_
 conv2d_nchw_ff_o_o_o_o, conv2d_nchw_ff_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_ff_o_o_o_i, factor=2)
 conv2d_nchw_yy_o_i, conv2d_nchw_yy_i = s[conv2d_nchw].split(conv2d_nchw_yy, factor=1)
 conv2d_nchw_yy_o_o_i, conv2d_nchw_yy_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_i, factor=1)
-conv2d_nchw_yy_o_o_o_i, conv2d_nchw_yy_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_i, factor=7)
+conv2d_nchw_yy_o_o_o_i, conv2d_nchw_yy_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_i, factor=1)
 conv2d_nchw_yy_o_o_o_o, conv2d_nchw_yy_o_o_o_i = s[conv2d_nchw].split(conv2d_nchw_yy_o_o_o_i, factor=1)
 conv2d_nchw_xx_o_i, conv2d_nchw_xx_i = s[conv2d_nchw].split(conv2d_nchw_xx, factor=1)
-conv2d_nchw_xx_o_o_i, conv2d_nchw_xx_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_i, factor=7)
-conv2d_nchw_xx_o_o_o_i, conv2d_nchw_xx_o_o_i = s[conv2d_nchw].split(conv2d_nchw_xx_o_o_i, factor=1)
+conv2d_nchw_xx_o_o_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=8)
-conv2d_nchw_rc_o_o, conv2d_nchw_rc_o_i = s[conv2d_nchw].split(conv2d_nchw_rc_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=8)
 conv2d_nchw_ry_o_i, conv2d_nchw_ry_i = s[conv2d_nchw].split(conv2d_nchw_ry, factor=1)
-conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=1)
-conv2d_nchw_rx_o_i, conv2d_nchw_rx_i = s[conv2d_nchw].split(conv2d_nchw_rx, factor=3)
-conv2d_nchw_rx_o_o, conv2d_nchw_rx_o_i = s[conv2d_nchw].split(conv2d_nchw_rx_o_i, factor=1)
+conv2d_nchw_ry_o_o, conv2d_nchw_ry_o_i = s[conv2d_nchw].split(conv2d_nchw_ry_o_i, factor=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=3)
 s[conv2d_nchw].reorder(conv2d_nchw_nn_o_o_o_o, conv2d_nchw_ff_o_o_o_o, conv2d_nchw_yy_o_o_o_o, conv2d_nchw_xx_o_o_o_o, conv2d_nchw_nn_o_o_o_i, conv2d_nchw_ff_o_o_o_i, conv2d_nchw_yy_o_o_o_i, conv2d_nchw_xx_o_o_o_i, conv2d_nchw_nn_o_o_i, conv2d_nchw_ff_o_o_i, conv2d_nchw_yy_o_o_i, conv2d_nchw_xx_o_o_i, conv2d_nchw_rc_o_o, conv2d_nchw_ry_o_o, conv2d_nchw_rx_o_o, conv2d_nchw_rc_o_i, conv2d_nchw_ry_o_i, conv2d_nchw_rx_o_i, conv2d_nchw_nn_o_i, conv2d_nchw_ff_o_i, conv2d_nchw_yy_o_i, conv2d_nc [...]
 compute_i0_o_i, compute_i0_i = s[compute].split(compute_i0, factor=1)
 compute_i0_o_o_i, compute_i0_o_i = s[compute].split(compute_i0_o_i, factor=1)
@@ -990,10 +692,10 @@ 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=32)
 compute_i1_o_o_o, compute_i1_o_o_i = s[compute].split(compute_i1_o_o_i, factor=2)
 compute_i2_o_i, compute_i2_i = s[compute].split(compute_i2, factor=1)
-compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=7)
+compute_i2_o_o_i, compute_i2_o_i = s[compute].split(compute_i2_o_i, factor=1)
 compute_i2_o_o_o, compute_i2_o_o_i = s[compute].split(compute_i2_o_o_i, factor=1)
-compute_i3_o_i, compute_i3_i = s[compute].split(compute_i3, factor=7)
-compute_i3_o_o_i, compute_i3_o_i = s[compute].split(compute_i3_o_i, factor=1)
+compute_i3_o_i, compute_i3_i = s[compute].split(compute_i3, factor=1)
+compute_i3_o_o_i, compute_i3_o_i = s[compute].split(compute_i3_o_i, factor=7)
 compute_i3_o_o_o, compute_i3_o_o_i = s[compute].split(compute_i3_o_o_i, factor=1)
 s[compute].reorder(compute_i0_o_o_o, compute_i1_o_o_o, compute_i2_o_o_o, compute_i3_o_o_o, compute_i0_o_o_i, compute_i1_o_o_i, compute_i2_o_o_i, compute_i3_o_o_i, compute_i0_o_i, compute_i1_o_i, compute_i2_o_i, compute_i3_o_i, compute_i0_i, compute_i1_i, compute_i2_i, compute_i3_i)
 s[conv2d_nchw].compute_at(s[compute], compute_i3_o_i)
@@ -1016,11 +718,11 @@ 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=224)
 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)
+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=6)
 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=224)
 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;, 64)
 s[conv2d_nchw].pragma(conv2d_nchw_nn_o_o_o_o, &quot;unroll_explicit&quot;, True)
 
 CUDA source code:
@@ -1039,383 +741,78 @@ CUDA source code:
   #define uint64_t unsigned long long
 #endif
 extern &quot;C&quot; __global__ void __launch_bounds__(224) default_function_kernel0(float* __restrict__ data, float* __restrict__ kernel, float* __restrict__ compute, float* __restrict__ bias) {
-  float conv2d_nchw[14];
-  __shared__ float pad_temp_shared[504];
-  __shared__ float kernel_shared[1536];
+  float conv2d_nchw[2];
+  __shared__ float pad_temp_shared[216];
+  __shared__ float kernel_shared[4608];
   conv2d_nchw[0] = 0.000000e+00f;
-  conv2d_nchw[7] = 0.000000e+00f;
   conv2d_nchw[1] = 0.000000e+00f;
-  conv2d_nchw[8] = 0.000000e+00f;
-  conv2d_nchw[2] = 0.000000e+00f;
-  conv2d_nchw[9] = 0.000000e+00f;
-  conv2d_nchw[3] = 0.000000e+00f;
-  conv2d_nchw[10] = 0.000000e+00f;
-  conv2d_nchw[4] = 0.000000e+00f;
-  conv2d_nchw[11] = 0.000000e+00f;
-  conv2d_nchw[5] = 0.000000e+00f;
-  conv2d_nchw[12] = 0.000000e+00f;
-  conv2d_nchw[6] = 0.000000e+00f;
-  conv2d_nchw[13] = 0.000000e+00f;
   for (int rc_outer_outer = 0; rc_outer_outer &lt; 64; ++rc_outer_outer) {
-    for (int ry_outer_outer = 0; ry_outer_outer &lt; 3; ++ry_outer_outer) {
-      __syncthreads();
-      pad_temp_shared[((int)threadIdx.x)] = (((((1 &lt;= (((((int)threadIdx.x) % 63) / 9) + ry_outer_outer)) &amp;&amp; ((((((int)threadIdx.x) % 63) / 9) + ry_outer_outer) &lt; 8)) &amp;&amp; (1 &lt;= (((int)threadIdx.x) % 9))) &amp;&amp; ((((int)threadIdx.x) % 9) &lt; 8)) ? data[(((((rc_outer_outer * 392) + ((((int)threadIdx.x) / 9) * 7)) + (ry_outer_outer * 7)) + (((int)threadIdx.x) % 9)) - 8)] : 0.000000e+00f);
-      pad_temp_shared[(((int)threadIdx.x) + 224)] = (((((1 &lt;= ((((((int)threadIdx.x) + 35) % 63) / 9) + ry_outer_outer)) &amp;&amp; (((((((int)threadIdx.x) + 35) % 63) / 9) + ry_outer_outer) &lt; 8)) &amp;&amp; (1 &lt;= ((((int)threadIdx.x) + 8) % 9))) &amp;&amp; (((((int)threadIdx.x) + 8) % 9) &lt; 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 224) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 8) % 9)) - 8)] : 0.000000e+00f);
-      if (((int)threadIdx.x) &lt; 56) {
-        pad_temp_shared[(((int)threadIdx.x) + 448)] = (((((1 &lt;= (((((int)threadIdx.x) + 7) / 9) + ry_outer_outer)) &amp;&amp; ((((((int)threadIdx.x) + 7) / 9) + ry_outer_outer) &lt; 8)) &amp;&amp; (1 &lt;= ((((int)threadIdx.x) + 7) % 9))) &amp;&amp; (((((int)threadIdx.x) + 7) % 9) &lt; 8)) ? data[(((((rc_outer_outer * 392) + (((((int)threadIdx.x) + 448) / 9) * 7)) + (ry_outer_outer * 7)) + ((((int)threadIdx.x) + 7) % 9)) - 8)] : 0.000000e+00f);
-      }
-      kernel_shared[((int)threadIdx.x)] = kernel[((((((((int)blockIdx.x) * 294912) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3))];
-      kernel_shared[(((int)threadIdx.x) + 224)] = kernel[((((((((int)blockIdx.x) * 294912) + (((((int)threadIdx.x) + 224) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
-      kernel_shared[(((int)threadIdx.x) + 448)] = kernel[((((((((int)blockIdx.x) * 294912) + (((((int)threadIdx.x) + 448) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
-      kernel_shared[(((int)threadIdx.x) + 672)] = kernel[(((((((((int)blockIdx.x) * 294912) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 129024)];
-      kernel_shared[(((int)threadIdx.x) + 896)] = kernel[((((((((int)blockIdx.x) * 294912) + (((((int)threadIdx.x) + 896) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 2) % 3))];
-      kernel_shared[(((int)threadIdx.x) + 1120)] = kernel[((((((((int)blockIdx.x) * 294912) + (((((int)threadIdx.x) + 1120) / 24) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + ((((int)threadIdx.x) + 1) % 3))];
-      if (((int)threadIdx.x) &lt; 192) {
-        kernel_shared[(((int)threadIdx.x) + 1344)] = kernel[(((((((((int)blockIdx.x) * 294912) + ((((int)threadIdx.x) / 24) * 4608)) + (rc_outer_outer * 72)) + (((((int)threadIdx.x) % 24) / 3) * 9)) + (ry_outer_outer * 3)) + (((int)threadIdx.x) % 3)) + 258048)];
-      }
-      __syncthreads();
-      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((((int)threadIdx.x) % 7) * 9)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
-      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[((((int)threadIdx.x) % 7) * 9)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 768)]));
-      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
-      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 769)]));
-      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
-      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 770)]));
-      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
-      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 63)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 771)]));
-      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
-      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 772)]));
-      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
-      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 773)]));
-      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
-      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 126)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 774)]));
-      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
-      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 775)]));
-      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
-      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 776)]));
-      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
-      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 189)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 777)]));
-      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
-      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 778)]));
-      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
-      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 779)]));
-      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
-      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 252)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 780)]));
-      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
-      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 781)]));
-      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
-      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 782)]));
-      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 315)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
-      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 315)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 783)]));
-      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 316)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
-      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 316)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 784)]));
-      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 317)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
-      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 317)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 785)]));
-      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 378)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
-      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 378)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 786)]));
-      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 379)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
-      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 379)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 787)]));
-      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 380)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
-      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 380)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 788)]));
-      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 441)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
-      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 441)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 789)]));
-      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 442)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
-      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 442)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 790)]));
-      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 443)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
-      conv2d_nchw[7] = (conv2d_nchw[7] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 443)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 791)]));
-      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 1)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
-      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 1)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 768)]));
-      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
-      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 769)]));
-      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 3)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
-      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 3)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 770)]));
-      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
-      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 64)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 771)]));
-      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
-      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 772)]));
-      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 66)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
-      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 66)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 773)]));
-      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
-      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 127)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 774)]));
-      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
-      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 775)]));
-      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 129)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
-      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 129)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 776)]));
-      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
-      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 190)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 777)]));
-      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
-      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 778)]));
-      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 192)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
-      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 192)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 779)]));
-      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
-      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 253)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 780)]));
-      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
-      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 781)]));
-      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 255)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
-      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 255)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 782)]));
-      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 316)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
-      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 316)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 783)]));
-      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 317)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
-      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 317)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 784)]));
-      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 318)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
-      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 318)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 785)]));
-      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 379)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
-      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 379)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 786)]));
-      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 380)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
-      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 380)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 787)]));
-      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 381)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
-      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 381)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 788)]));
-      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 442)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
-      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 442)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 789)]));
-      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 443)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
-      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 443)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 790)]));
-      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 444)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
-      conv2d_nchw[8] = (conv2d_nchw[8] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 444)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 791)]));
-      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 2)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
-      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 2)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 768)]));
-      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 3)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
-      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 3)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 769)]));
-      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 4)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
-      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 4)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 770)]));
-      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
-      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 65)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 771)]));
-      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 66)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
-      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 66)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 772)]));
-      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 67)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
-      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 67)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 773)]));
-      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
-      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 128)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 774)]));
-      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 129)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
-      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 129)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 775)]));
-      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 130)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
-      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 130)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 776)]));
-      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
-      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 191)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 777)]));
-      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 192)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
-      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 192)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 778)]));
-      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 193)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
-      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 193)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 779)]));
-      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
-      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 254)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 780)]));
-      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 255)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
-      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 255)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 781)]));
-      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 256)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
-      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 256)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 782)]));
-      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 317)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
-      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 317)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 783)]));
-      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 318)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
-      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 318)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 784)]));
-      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 319)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
-      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 319)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 785)]));
-      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 380)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
-      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 380)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 786)]));
-      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 381)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
-      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 381)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 787)]));
-      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 382)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
-      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 382)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 788)]));
-      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 443)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
-      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 443)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 789)]));
-      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 444)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
-      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 444)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 790)]));
-      conv2d_nchw[2] = (conv2d_nchw[2] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 445)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
-      conv2d_nchw[9] = (conv2d_nchw[9] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 445)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 791)]));
-      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 3)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
-      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 3)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 768)]));
-      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 4)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
-      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 4)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 769)]));
-      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 5)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
-      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 5)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 770)]));
-      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 66)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
-      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 66)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 771)]));
-      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 67)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
-      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 67)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 772)]));
-      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 68)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
-      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 68)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 773)]));
-      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 129)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
-      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 129)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 774)]));
-      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 130)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
-      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 130)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 775)]));
-      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 131)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
-      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 131)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 776)]));
-      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 192)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
-      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 192)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 777)]));
-      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 193)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
-      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 193)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 778)]));
-      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 194)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
-      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 194)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 779)]));
-      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 255)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
-      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 255)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 780)]));
-      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 256)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
-      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 256)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 781)]));
-      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 257)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
-      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 257)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 782)]));
-      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 318)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
-      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 318)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 783)]));
-      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 319)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
-      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 319)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 784)]));
-      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 320)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
-      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 320)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 785)]));
-      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 381)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
-      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 381)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 786)]));
-      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 382)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
-      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 382)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 787)]));
-      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 383)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
-      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 383)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 788)]));
-      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 444)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
-      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 444)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 789)]));
-      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 445)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
-      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 445)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 790)]));
-      conv2d_nchw[3] = (conv2d_nchw[3] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 446)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
-      conv2d_nchw[10] = (conv2d_nchw[10] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 446)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 791)]));
-      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 4)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
-      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 4)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 768)]));
-      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 5)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
-      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 5)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 769)]));
-      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 6)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
-      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 6)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 770)]));
-      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 67)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
-      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 67)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 771)]));
-      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 68)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
-      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 68)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 772)]));
-      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 69)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
-      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 69)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 773)]));
-      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 130)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
-      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 130)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 774)]));
-      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 131)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
-      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 131)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 775)]));
-      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 132)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
-      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 132)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 776)]));
-      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 193)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
-      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 193)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 777)]));
-      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 194)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
-      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 194)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 778)]));
-      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 195)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
-      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 195)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 779)]));
-      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 256)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
-      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 256)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 780)]));
-      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 257)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
-      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 257)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 781)]));
-      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 258)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
-      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 258)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 782)]));
-      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 319)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
-      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 319)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 783)]));
-      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 320)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
-      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 320)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 784)]));
-      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 321)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
-      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 321)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 785)]));
-      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 382)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
-      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 382)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 786)]));
-      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 383)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
-      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 383)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 787)]));
-      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 384)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
-      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 384)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 788)]));
-      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 445)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
-      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 445)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 789)]));
-      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 446)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
-      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 446)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 790)]));
-      conv2d_nchw[4] = (conv2d_nchw[4] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 447)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
-      conv2d_nchw[11] = (conv2d_nchw[11] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 447)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 791)]));
-      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 5)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
-      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 5)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 768)]));
-      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 6)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
-      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 6)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 769)]));
-      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 7)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
-      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 7)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 770)]));
-      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 68)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
-      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 68)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 771)]));
-      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 69)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
-      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 69)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 772)]));
-      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 70)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
-      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 70)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 773)]));
-      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 131)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
-      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 131)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 774)]));
-      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 132)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
-      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 132)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 775)]));
-      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 133)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
-      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 133)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 776)]));
-      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 194)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
-      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 194)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 777)]));
-      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 195)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
-      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 195)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 778)]));
-      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 196)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
-      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 196)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 779)]));
-      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 257)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
-      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 257)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 780)]));
-      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 258)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
-      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 258)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 781)]));
-      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 259)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
-      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 259)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 782)]));
-      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 320)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
-      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 320)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 783)]));
-      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 321)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
-      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 321)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 784)]));
-      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 322)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
-      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 322)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 785)]));
-      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 383)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
-      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 383)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 786)]));
-      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 384)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
-      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 384)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 787)]));
-      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 385)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
-      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 385)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 788)]));
-      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 446)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
-      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 446)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 789)]));
-      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 447)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
-      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 447)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 790)]));
-      conv2d_nchw[5] = (conv2d_nchw[5] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 448)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
-      conv2d_nchw[12] = (conv2d_nchw[12] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 448)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 791)]));
-      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 6)] * kernel_shared[((((int)threadIdx.x) / 7) * 24)]));
-      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 6)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 768)]));
-      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 7)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 1)]));
-      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 7)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 769)]));
-      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 8)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 2)]));
-      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 8)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 770)]));
-      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 69)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 3)]));
-      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 69)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 771)]));
-      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 70)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 4)]));
-      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 70)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 772)]));
-      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 71)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 5)]));
-      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 71)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 773)]));
-      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 132)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 6)]));
-      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 132)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 774)]));
-      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 133)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 7)]));
-      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 133)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 775)]));
-      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 134)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 8)]));
-      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 134)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 776)]));
-      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 195)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 9)]));
-      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 195)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 777)]));
-      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 196)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 10)]));
-      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 196)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 778)]));
-      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 197)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 11)]));
-      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 197)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 779)]));
-      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 258)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 12)]));
-      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 258)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 780)]));
-      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 259)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 13)]));
-      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 259)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 781)]));
-      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 260)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 14)]));
-      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 260)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 782)]));
-      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 321)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 15)]));
-      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 321)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 783)]));
-      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 322)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 16)]));
-      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 322)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 784)]));
-      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 323)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 17)]));
-      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 323)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 785)]));
-      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 384)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 18)]));
-      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 384)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 786)]));
-      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 385)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 19)]));
-      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 385)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 787)]));
-      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 386)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 20)]));
-      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 386)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 788)]));
-      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 447)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 21)]));
-      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 447)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 789)]));
-      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 448)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 22)]));
-      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 448)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 790)]));
-      conv2d_nchw[6] = (conv2d_nchw[6] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 449)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 23)]));
-      conv2d_nchw[13] = (conv2d_nchw[13] + (pad_temp_shared[(((((int)threadIdx.x) % 7) * 9) + 449)] * kernel_shared[(((((int)threadIdx.x) / 7) * 24) + 791)]));
+    __syncthreads();
+    if (((int)threadIdx.x) &lt; 36) {
+      pad_temp_shared[(((int)threadIdx.x) * 6)] = (((((1 &lt;= ((((((int)threadIdx.x) * 2) % 9) / 3) + (((int)blockIdx.x) % 7))) &amp;&amp; (((((((int)threadIdx.x) * 2) % 9) / 3) + (((int)blockIdx.x) % 7)) &lt; 8)) &amp;&amp; (1 &lt; ((((int)threadIdx.x) * 6) % 9))) &amp;&amp; (((((int)threadIdx.x) * 6) % 9) &lt; 8)) ? data[((((((rc_outer_outer * 392) + (((((int)threadIdx.x) * 2) / 9) * 49)) + ((((((int)threadIdx.x) * 2) % 9) / 3) * 7)) + ((((int)blockIdx.x) % 7) * 7)) + ((((int)threadId [...]
+    }
+    if (((int)threadIdx.x) &lt; 36) {
+      pad_temp_shared[((((int)threadIdx.x) * 6) + 1)] = (((((1 &lt;= ((((((int)threadIdx.x) * 2) % 9) / 3) + (((int)blockIdx.x) % 7))) &amp;&amp; (((((((int)threadIdx.x) * 2) % 9) / 3) + (((int)blockIdx.x) % 7)) &lt; 8)) &amp;&amp; (1 &lt;= (((((int)threadIdx.x) * 6) + 1) % 9))) &amp;&amp; ((((((int)threadIdx.x) * 6) + 1) % 9) &lt; 8)) ? data[((((((rc_outer_outer * 392) + (((((int)threadIdx.x) * 2) / 9) * 49)) + ((((((int)threadIdx.x) * 2) % 9) / 3) * 7)) + ((((int)blockIdx.x) % 7) * 7)) [...]
+    }
+    if (((int)threadIdx.x) &lt; 36) {
+      pad_temp_shared[((((int)threadIdx.x) * 6) + 2)] = (((((1 &lt;= ((((((int)threadIdx.x) * 2) % 9) / 3) + (((int)blockIdx.x) % 7))) &amp;&amp; (((((((int)threadIdx.x) * 2) % 9) / 3) + (((int)blockIdx.x) % 7)) &lt; 8)) &amp;&amp; (1 &lt; (((((int)threadIdx.x) * 6) + 2) % 9))) &amp;&amp; ((((((int)threadIdx.x) * 6) + 2) % 9) &lt; 8)) ? data[((((((rc_outer_outer * 392) + (((((int)threadIdx.x) * 2) / 9) * 49)) + ((((((int)threadIdx.x) * 2) % 9) / 3) * 7)) + ((((int)blockIdx.x) % 7) * 7))  [...]
+    }
+    if (((int)threadIdx.x) &lt; 36) {
+      pad_temp_shared[((((int)threadIdx.x) * 6) + 3)] = (((((1 &lt;= (((((((int)threadIdx.x) * 2) + 1) % 9) / 3) + (((int)blockIdx.x) % 7))) &amp;&amp; ((((((((int)threadIdx.x) * 2) + 1) % 9) / 3) + (((int)blockIdx.x) % 7)) &lt; 8)) &amp;&amp; (1 &lt; (((((int)threadIdx.x) * 6) + 3) % 9))) &amp;&amp; ((((((int)threadIdx.x) * 6) + 3) % 9) &lt; 8)) ? data[((((((rc_outer_outer * 392) + ((((((int)threadIdx.x) * 2) + 1) / 9) * 49)) + (((((((int)threadIdx.x) * 2) + 1) % 9) / 3) * 7)) + ((((int [...]
+    }
+    if (((int)threadIdx.x) &lt; 36) {
+      pad_temp_shared[((((int)threadIdx.x) * 6) + 4)] = (((((1 &lt;= (((((((int)threadIdx.x) * 2) + 1) % 9) / 3) + (((int)blockIdx.x) % 7))) &amp;&amp; ((((((((int)threadIdx.x) * 2) + 1) % 9) / 3) + (((int)blockIdx.x) % 7)) &lt; 8)) &amp;&amp; (1 &lt;= (((((int)threadIdx.x) * 6) + 4) % 9))) &amp;&amp; ((((((int)threadIdx.x) * 6) + 4) % 9) &lt; 8)) ? data[((((((rc_outer_outer * 392) + ((((((int)threadIdx.x) * 2) + 1) / 9) * 49)) + (((((((int)threadIdx.x) * 2) + 1) % 9) / 3) * 7)) + ((((in [...]
+    }
+    if (((int)threadIdx.x) &lt; 36) {
+      pad_temp_shared[((((int)threadIdx.x) * 6) + 5)] = (((((1 &lt;= (((((((int)threadIdx.x) * 2) + 1) % 9) / 3) + (((int)blockIdx.x) % 7))) &amp;&amp; ((((((((int)threadIdx.x) * 2) + 1) % 9) / 3) + (((int)blockIdx.x) % 7)) &lt; 8)) &amp;&amp; (1 &lt; (((((int)threadIdx.x) * 6) + 5) % 9))) &amp;&amp; ((((((int)threadIdx.x) * 6) + 5) % 9) &lt; 8)) ? data[((((((rc_outer_outer * 392) + ((((((int)threadIdx.x) * 2) + 1) / 9) * 49)) + (((((((int)threadIdx.x) * 2) + 1) % 9) / 3) * 7)) + ((((int [...]
+    }
+    kernel_shared[((int)threadIdx.x)] = kernel[(((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) % 72))];
+    kernel_shared[(((int)threadIdx.x) + 224)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 224) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+    kernel_shared[(((int)threadIdx.x) + 448)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 448) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+    kernel_shared[(((int)threadIdx.x) + 672)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 672) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) / 3) + 8) % 24) * 3)) + (((int)threadIdx.x) % 3))];
+    kernel_shared[(((int)threadIdx.x) + 896)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 896) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 32) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+    kernel_shared[(((int)threadIdx.x) + 1120)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 1120) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 40) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+    kernel_shared[(((int)threadIdx.x) + 1344)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 1344) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) / 3) + 16) % 24) * 3)) + (((int)threadIdx.x) % 3))];
+    kernel_shared[(((int)threadIdx.x) + 1568)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 1568) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 56) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+    kernel_shared[(((int)threadIdx.x) + 1792)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 1792) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 64) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+    kernel_shared[(((int)threadIdx.x) + 2016)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) % 72)) + 129024)];
+    kernel_shared[(((int)threadIdx.x) + 2240)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 2240) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+    kernel_shared[(((int)threadIdx.x) + 2464)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 2464) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+    kernel_shared[(((int)threadIdx.x) + 2688)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 2688) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) / 3) + 8) % 24) * 3)) + (((int)threadIdx.x) % 3))];
+    kernel_shared[(((int)threadIdx.x) + 2912)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 2912) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 32) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+    kernel_shared[(((int)threadIdx.x) + 3136)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 3136) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 40) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+    kernel_shared[(((int)threadIdx.x) + 3360)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 3360) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) / 3) + 16) % 24) * 3)) + (((int)threadIdx.x) % 3))];
+    kernel_shared[(((int)threadIdx.x) + 3584)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 3584) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 56) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+    kernel_shared[(((int)threadIdx.x) + 3808)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 3808) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 64) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+    kernel_shared[(((int)threadIdx.x) + 4032)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + ((((int)threadIdx.x) / 72) * 4608)) + (rc_outer_outer * 72)) + (((int)threadIdx.x) % 72)) + 258048)];
+    kernel_shared[(((int)threadIdx.x) + 4256)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 4256) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 8) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 2) % 3))];
+    if (((int)threadIdx.x) &lt; 128) {
+      kernel_shared[(((int)threadIdx.x) + 4480)] = kernel[((((((((int)blockIdx.x) / 7) * 294912) + (((((int)threadIdx.x) + 4480) / 72) * 4608)) + (rc_outer_outer * 72)) + ((((((int)threadIdx.x) + 16) % 72) / 3) * 3)) + ((((int)threadIdx.x) + 1) % 3))];
+    }
+    __syncthreads();
+    for (int rc_outer_inner = 0; rc_outer_inner &lt; 8; ++rc_outer_inner) {
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[((rc_outer_inner * 27) + (((int)threadIdx.x) % 7))] * kernel_shared[(((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9))]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[((rc_outer_inner * 27) + (((int)threadIdx.x) % 7))] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 2304)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 1)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 1)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 2305)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 2)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 2)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 2306)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 3)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 9)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 2307)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 4)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 10)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 2308)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 5)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 11)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 2309)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 6)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 18)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 2310)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 7)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 19)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 2311)]));
+      conv2d_nchw[0] = (conv2d_nchw[0] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 8)]));
+      conv2d_nchw[1] = (conv2d_nchw[1] + (pad_temp_shared[(((rc_outer_inner * 27) + (((int)threadIdx.x) % 7)) + 20)] * kernel_shared[((((((int)threadIdx.x) / 7) * 72) + (rc_outer_inner * 9)) + 2312)]));
     }
   }
-  for (int i3_inner = 0; i3_inner &lt; 7; ++i3_inner) {
-    compute[(((((int)blockIdx.x) * 3136) + (((int)threadIdx.x) * 7)) + i3_inner)] = max((conv2d_nchw[i3_inner] + bias[((((int)blockIdx.x) * 64) + (((int)threadIdx.x) / 7))]), 0.000000e+00f);
-    compute[((((((int)blockIdx.x) * 3136) + (((int)threadIdx.x) * 7)) + i3_inner) + 1568)] = max((conv2d_nchw[(i3_inner + 7)] + bias[(((((int)blockIdx.x) * 64) + (((int)threadIdx.x) / 7)) + 32)]), 0.000000e+00f);
-  }
+  compute[(((((((int)blockIdx.x) / 7) * 3136) + ((((int)threadIdx.x) / 7) * 49)) + ((((int)blockIdx.x) % 7) * 7)) + (((int)threadIdx.x) % 7))] = max((conv2d_nchw[0] + bias[(((((int)blockIdx.x) / 7) * 64) + (((int)threadIdx.x) / 7))]), 0.000000e+00f);
+  compute[((((((((int)blockIdx.x) / 7) * 3136) + ((((int)threadIdx.x) / 7) * 49)) + ((((int)blockIdx.x) % 7) * 7)) + (((int)threadIdx.x) % 7)) + 1568)] = max((conv2d_nchw[1] + bias[((((((int)blockIdx.x) / 7) * 64) + (((int)threadIdx.x) / 7)) + 32)]), 0.000000e+00f);
 }
 </pre></div>
 </div>
@@ -1451,7 +848,7 @@ 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  38.343 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 5 minutes  29.208 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 4fcb6a4360..65dbbd0059 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_cuda.html
@@ -916,7 +916,7 @@ 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)
-   7.9053       7.9052       7.9161       7.8946       0.0088
+   7.8731       7.8745       7.8789       7.8659       0.0054
 </pre></div>
 </div>
 </div>
@@ -938,7 +938,7 @@ 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  1.132 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  1.747 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 1fb80698fd..846c247af4 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_network_x86.html
@@ -935,7 +935,7 @@ 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)
-  755.9445     755.6418     757.1292     755.0625      0.8704
+  752.6776     753.1605     753.6184     751.2538      1.0239
 </pre></div>
 </div>
 </div>
@@ -957,7 +957,7 @@ 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  30.980 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  31.583 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 3c3a5aaabf..bba0cf75d7 100644
--- a/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
+++ b/docs/how_to/tune_with_autoscheduler/tune_sparse_x86.html
@@ -633,121 +633,30 @@ layout transformation, parallelization, vectorization, unrolling, and operator f
              placeholder_4: Buffer(placeholder_14: Pointer(float32), float32, [128, 512], []),
              compute: Buffer(compute_2: Pointer(float32), float32, [128, 512], [])}
   buffer_map = {placeholder_5: placeholder, placeholder_6: placeholder_1, placeholder_7: placeholder_2, placeholder_8: placeholder_3, placeholder_9: placeholder_4, compute_1: compute} {
-  for (i0.outer.i1.outer.fused: int32, 0, 64) &quot;parallel&quot; {
-    allocate(compute_3: Pointer(global float32), float32, [1024]), storage_scope = global {
-      for (i.outer.inner: int32, 0, 16) {
-        for (nb_j.inner: int32, 0, 2) {
-          let cse_var_2: int32 = ((i.outer.inner*64) + (nb_j.inner*16))
-          let cse_var_1: int32 = ((floormod(i0.outer.i1.outer.fused, 16)*2) + nb_j.inner)
-           {
-            compute_4: Buffer(compute_3, float32, [1024], [])[cse_var_2] = 0f32
-            compute_4[(cse_var_2 + 1)] = 0f32
-            compute_4[(cse_var_2 + 2)] = 0f32
-            compute_4[(cse_var_2 + 3)] = 0f32
-            compute_4[(cse_var_2 + 4)] = 0f32
-            compute_4[(cse_var_2 + 5)] = 0f32
-            compute_4[(cse_var_2 + 6)] = 0f32
-            compute_4[(cse_var_2 + 7)] = 0f32
-            compute_4[(cse_var_2 + 8)] = 0f32
-            compute_4[(cse_var_2 + 9)] = 0f32
-            compute_4[(cse_var_2 + 10)] = 0f32
-            compute_4[(cse_var_2 + 11)] = 0f32
-            compute_4[(cse_var_2 + 12)] = 0f32
-            compute_4[(cse_var_2 + 13)] = 0f32
-            compute_4[(cse_var_2 + 14)] = 0f32
-            compute_4[(cse_var_2 + 15)] = 0f32
-            compute_4[(cse_var_2 + 32)] = 0f32
-            compute_4[(cse_var_2 + 33)] = 0f32
-            compute_4[(cse_var_2 + 34)] = 0f32
-            compute_4[(cse_var_2 + 35)] = 0f32
-            compute_4[(cse_var_2 + 36)] = 0f32
-            compute_4[(cse_var_2 + 37)] = 0f32
-            compute_4[(cse_var_2 + 38)] = 0f32
-            compute_4[(cse_var_2 + 39)] = 0f32
-            compute_4[(cse_var_2 + 40)] = 0f32
-            compute_4[(cse_var_2 + 41)] = 0f32
-            compute_4[(cse_var_2 + 42)] = 0f32
-            compute_4[(cse_var_2 + 43)] = 0f32
-            compute_4[(cse_var_2 + 44)] = 0f32
-            compute_4[(cse_var_2 + 45)] = 0f32
-            compute_4[(cse_var_2 + 46)] = 0f32
-            compute_4[(cse_var_2 + 47)] = 0f32
-            for (elem_idx: int32, 0, (placeholder_15: Buffer(placeholder_13, int32, [33], [])[(cse_var_1 + 1)] - placeholder_15[cse_var_1])) {
-              let cse_var_35: int32 = (elem_idx*16)
-              let cse_var_34: int32 = (cse_var_2 + 9)
-              let cse_var_33: int32 = (cse_var_2 + 8)
-              let cse_var_32: int32 = (cse_var_2 + 7)
-              let cse_var_31: int32 = (cse_var_2 + 6)
-              let cse_var_30: int32 = (cse_var_2 + 5)
-              let cse_var_29: int32 = (cse_var_2 + 47)
-              let cse_var_28: int32 = (cse_var_2 + 46)
-              let cse_var_27: int32 = (cse_var_2 + 45)
-              let cse_var_26: int32 = (cse_var_2 + 44)
-              let cse_var_25: int32 = (cse_var_2 + 43)
-              let cse_var_24: int32 = (cse_var_2 + 42)
-              let cse_var_23: int32 = (cse_var_2 + 41)
-              let cse_var_22: int32 = (cse_var_2 + 40)
-              let cse_var_21: int32 = (cse_var_2 + 4)
-              let cse_var_20: int32 = (cse_var_2 + 39)
-              let cse_var_19: int32 = (cse_var_2 + 38)
-              let cse_var_18: int32 = (cse_var_2 + 37)
-              let cse_var_17: int32 = (cse_var_2 + 36)
-              let cse_var_16: int32 = (cse_var_2 + 35)
-              let cse_var_15: int32 = (cse_var_2 + 34)
-              let cse_var_14: int32 = (cse_var_2 + 33)
-              let cse_var_13: int32 = (cse_var_2 + 32)
-              let cse_var_12: int32 = (cse_var_2 + 3)
-              let cse_var_11: int32 = (cse_var_2 + 2)
-              let cse_var_10: int32 = (cse_var_2 + 15)
-              let cse_var_9: int32 = (cse_var_2 + 14)
-              let cse_var_8: int32 = (cse_var_2 + 13)
-              let cse_var_7: int32 = (cse_var_2 + 12)
-              let cse_var_6: int32 = (cse_var_2 + 11)
-              let cse_var_5: int32 = (cse_var_2 + 10)
-              let cse_var_4: int32 = (cse_var_2 + 1)
-              let cse_var_3: int32 = ((floordiv(i0.outer.i1.outer.fused, 16)*8192) + (i.outer.inner*512))
-               {
-                compute_4[cse_var_2] = (compute_4[cse_var_2] + (placeholder_16: Buffer(placeholder_11, float32, [78656], [])[((placeholder_15[cse_var_1]*16) + cse_var_35)]*max(placeholder_17: Buffer(placeholder_10, float32, [32768], [])[(cse_var_3 + placeholder_18: Buffer(placeholder_12, int32, [4916], [])[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                compute_4[cse_var_4] = (compute_4[cse_var_4] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 1)]*max(placeholder_17[(cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                compute_4[cse_var_11] = (compute_4[cse_var_11] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 2)]*max(placeholder_17[(cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                compute_4[cse_var_12] = (compute_4[cse_var_12] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 3)]*max(placeholder_17[(cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                compute_4[cse_var_21] = (compute_4[cse_var_21] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 4)]*max(placeholder_17[(cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                compute_4[cse_var_30] = (compute_4[cse_var_30] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 5)]*max(placeholder_17[(cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                compute_4[cse_var_31] = (compute_4[cse_var_31] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 6)]*max(placeholder_17[(cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                compute_4[cse_var_32] = (compute_4[cse_var_32] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 7)]*max(placeholder_17[(cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                compute_4[cse_var_33] = (compute_4[cse_var_33] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 8)]*max(placeholder_17[(cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                compute_4[cse_var_34] = (compute_4[cse_var_34] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 9)]*max(placeholder_17[(cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                compute_4[cse_var_5] = (compute_4[cse_var_5] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 10)]*max(placeholder_17[(cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                compute_4[cse_var_6] = (compute_4[cse_var_6] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 11)]*max(placeholder_17[(cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                compute_4[cse_var_7] = (compute_4[cse_var_7] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 12)]*max(placeholder_17[(cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                compute_4[cse_var_8] = (compute_4[cse_var_8] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 13)]*max(placeholder_17[(cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                compute_4[cse_var_9] = (compute_4[cse_var_9] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 14)]*max(placeholder_17[(cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                compute_4[cse_var_10] = (compute_4[cse_var_10] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 15)]*max(placeholder_17[(cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)])], 0f32)))
-                compute_4[cse_var_13] = (compute_4[cse_var_13] + (placeholder_16[((placeholder_15[cse_var_1]*16) + cse_var_35)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
-                compute_4[cse_var_14] = (compute_4[cse_var_14] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 1)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
-                compute_4[cse_var_15] = (compute_4[cse_var_15] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 2)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
-                compute_4[cse_var_16] = (compute_4[cse_var_16] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 3)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
-                compute_4[cse_var_17] = (compute_4[cse_var_17] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 4)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
-                compute_4[cse_var_18] = (compute_4[cse_var_18] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 5)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
-                compute_4[cse_var_19] = (compute_4[cse_var_19] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 6)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
-                compute_4[cse_var_20] = (compute_4[cse_var_20] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 7)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
-                compute_4[cse_var_22] = (compute_4[cse_var_22] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 8)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
-                compute_4[cse_var_23] = (compute_4[cse_var_23] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 9)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
-                compute_4[cse_var_24] = (compute_4[cse_var_24] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 10)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
-                compute_4[cse_var_25] = (compute_4[cse_var_25] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 11)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
-                compute_4[cse_var_26] = (compute_4[cse_var_26] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 12)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
-                compute_4[cse_var_27] = (compute_4[cse_var_27] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 13)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
-                compute_4[cse_var_28] = (compute_4[cse_var_28] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 14)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
-                compute_4[cse_var_29] = (compute_4[cse_var_29] + (placeholder_16[(((placeholder_15[cse_var_1]*16) + cse_var_35) + 15)]*max(placeholder_17[((cse_var_3 + placeholder_18[(placeholder_15[cse_var_1] + elem_idx)]) + 256)], 0f32)))
+  for (i0.outer.i1.outer.fused: int32, 0, 128) &quot;parallel&quot; {
+    allocate(compute_3: Pointer(global float32), float32, [512]), storage_scope = global {
+      for (i.outer.inner: int32, 0, 2) {
+        for (i.inner.init: int32, 0, 16) {
+          for (j.init: int32, 0, 16) {
+            compute_4: Buffer(compute_3, float32, [512], [])[(((i.outer.inner*256) + (i.inner.init*16)) + j.init)] = 0f32
+          }
+        }
+        for (elem_idx: int32, 0, let cse_var_1: int32 = floormod(i0.outer.i1.outer.fused, 32) in (placeholder_15: Buffer(placeholder_13, int32, [33], [])[(cse_var_1 + 1)] - placeholder_15[cse_var_1])) {
+          for (i.inner: int32, 0, 16) {
+            for (j: int32, 0, 16) {
+              let cse_var_2: int32 = floormod(i0.outer.i1.outer.fused, 32)
+              if @tir.likely((elem_idx &lt; (placeholder_15[(cse_var_2 + 1)] - placeholder_15[cse_var_2])), dtype=bool) {
+                let cse_var_3: int32 = (((i.outer.inner*256) + (i.inner*16)) + j)
+                compute_4[cse_var_3] = (compute_4[cse_var_3] + (placeholder_16: Buffer(placeholder_11, float32, [78656], [])[(((placeholder_15[cse_var_2]*16) + (elem_idx*16)) + j)]*max(placeholder_17: Buffer(placeholder_10, float32, [32768], [])[((((floordiv(i0.outer.i1.outer.fused, 32)*8192) + (i.outer.inner*4096)) + (i.inner*256)) + placeholder_18: Buffer(placeholder_12, int32, [4916], [])[(placeholder_15[cse_var_2] + elem_idx)])], 0f32)))
               }
             }
           }
         }
       }
       for (i0.inner: int32, 0, 32) {
-        for (i1.inner: int32, 0, 32) {
-          let cse_var_36: int32 = ((((floordiv(i0.outer.i1.outer.fused, 16)*16384) + (i0.inner*512)) + (floormod(i0.outer.i1.outer.fused, 16)*32)) + i1.inner)
-          compute_5: Buffer(compute_2, float32, [65536], [])[cse_var_36] = max((compute_4[((i0.inner*32) + i1.inner)] + placeholder_19: Buffer(placeholder_14, float32, [65536], [])[cse_var_36]), 0f32)
+        for (i1.inner: int32, 0, 16) {
+          let cse_var_4: int32 = ((((floordiv(i0.outer.i1.outer.fused, 32)*16384) + (i0.inner*512)) + (floormod(i0.outer.i1.outer.fused, 32)*16)) + i1.inner)
+          compute_5: Buffer(compute_2, float32, [65536], [])[cse_var_4] = max((compute_4[((i0.inner*16) + i1.inner)] + placeholder_19: Buffer(placeholder_14, float32, [65536], [])[cse_var_4]), 0f32)
         }
       }
     }
@@ -786,7 +695,7 @@ 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: 3.507 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 1.669 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 1dace7f044..01b88ea685 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>00:33.745</strong> total execution time for <strong>how_to_tune_with_autotvm</strong> files:</p>
+<p><strong>00:43.740</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>00:33.711</p></td>
+<td><p>00:43.708</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>00:00.019</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>
@@ -361,7 +361,7 @@
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="tune_relay_arm.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-arm-py"><span class="std std-ref">Auto-tuning a Convolutional Network for ARM CPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_arm.py</span></code>)</p></td>
-<td><p>00:00.004</p></td>
+<td><p>00:00.005</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="tune_relay_mobile_gpu.html#sphx-glr-how-to-tune-with-autotvm-tune-relay-mobile-gpu-py"><span class="std std-ref">Auto-tuning a Convolutional Network for Mobile GPU</span></a> (<code class="docutils literal notranslate"><span class="pre">tune_relay_mobile_gpu.py</span></code>)</p></td>
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 f140544cc3..2d2a4b4575 100644
--- a/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
+++ b/docs/how_to/tune_with_autotvm/tune_conv2d_cuda.html
@@ -690,7 +690,7 @@ Traceback (most recent call last):
   File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 56, in tvm._ffi._cy3.core.tvm_callback
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 875, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 2, 8, 4]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 32, 16]), (&#39;tile_ry&#39;, [-1, 3, 1]), (&#39;tile_rx&#39;, [-1, 1, 1]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 0)],None,3818882
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 2, 1, 256]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 1, 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;, 1)],None,7841897
 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 592, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
@@ -813,9 +813,8 @@ Traceback (most recent call last):
   File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 56, in tvm._ffi._cy3.core.tvm_callback
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 875, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 4, 2, 8]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 1, 32]), (&#39;tile_ry&#39;, [-1, 1, 3]), (&#39;tile_rx&#39;, [-1, 3, 1]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 1)],None,9821605
-No: 3   GFLOPS: 36.11/36.11     result: MeasureResult(costs=(0.006411471315789473,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.768928050994873, timestamp=1673555737.4988983)        [(&#39;tile_f&#39;, [-1, 1, 1, 2]), (&#39;tile_y&#39;, [-1, 7, 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;, 0), (&#39;unroll_explicit&#39;, 0)],None,77715
-No: 4   GFLOPS: 0.00/36.11      result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 2, 256, 1]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 8, 2]), (&#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,1788653
+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 592, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 544, in _build_func_common
@@ -937,8 +936,9 @@ Traceback (most recent call last):
   File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 56, in tvm._ffi._cy3.core.tvm_callback
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 875, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 64, 4, 2]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 7, 1, 1]), (&#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,4291838
-No: 5   GFLOPS: 0.00/36.11      result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 8, 2, 8]), (&#39;tile_y&#39;, [-1, 7, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 1, 512]), (&#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,8128046
+No: 4   GFLOPS: 825.44/825.44   result: MeasureResult(costs=(0.00028045692893401017,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.2746119499206543, timestamp=1673569917.1928902)     [(&#39;tile_f&#39;, [-1, 1, 16, 1]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 8, 1]), (&#39;tile_ry&#39;, [-1, 3, 1]), (&#39;tile_rx&#39;, [-1, 1, 1]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 1)],None,8919274
+No: 5   GFLOPS: 0.00/825.44     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 592, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 544, in _build_func_common
@@ -1060,8 +1060,8 @@ Traceback (most recent call last):
   File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 56, in tvm._ffi._cy3.core.tvm_callback
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 875, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 8, 1, 64]), (&#39;tile_y&#39;, [-1, 7, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 8, 2]), (&#39;tile_ry&#39;, [-1, 1, 1]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 0), (&#39;unroll_explicit&#39;, 0)],None,1207783
-No: 6   GFLOPS: 0.00/36.11      result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 4, 4, 8]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 2, 16]), (&#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,2835951
+No: 6   GFLOPS: 0.00/825.44     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 592, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 544, in _build_func_common
@@ -1183,8 +1183,8 @@ Traceback (most recent call last):
   File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 56, in tvm._ffi._cy3.core.tvm_callback
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 875, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 1, 32, 16]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 2, 256]), (&#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,7156344
-No: 7   GFLOPS: 0.00/36.11      result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 64, 1, 4]), (&#39;tile_y&#39;, [-1, 1, 7, 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, 3]), (&#39;auto_unroll_max_step&#39;, 512), (&#39;unroll_explicit&#39;, 0)],None,2920386
+No: 7   GFLOPS: 0.00/825.44     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 592, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 544, in _build_func_common
@@ -1306,11 +1306,8 @@ Traceback (most recent call last):
   File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 56, in tvm._ffi._cy3.core.tvm_callback
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 875, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 4, 4, 4]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 7, 1, 1]), (&#39;tile_rc&#39;, [-1, 2, 128]), (&#39;tile_ry&#39;, [-1, 1, 1]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 512), (&#39;unroll_explicit&#39;, 0)],None,3081657
-No: 8   GFLOPS: 443.05/443.05   result: MeasureResult(costs=(0.0005225226458333333,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.2629878520965576, timestamp=1673555740.8875287)      [(&#39;tile_f&#39;, [-1, 2, 8, 2]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 8, 4]), (&#39;tile_ry&#39;, [-1, 1, 1]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 512), (&#39;unroll_explicit&#39;, 0)],None,2983720
-No: 9   GFLOPS: 1.01/443.05     result: MeasureResult(costs=(0.22923324575,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.9031102657318115, timestamp=1673555745.067556)       [(&#39;tile_f&#39;, [-1, 1, 2, 256]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 2, 1]), (&#39;tile_ry&#39;, [-1, 3, 1]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 0), (&#39;unroll_explicit&#39;, 0)],None,1359598
-No: 10  GFLOPS: 366.19/443.05   result: MeasureResult(costs=(0.0006321807421383648,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.3297412395477295, timestamp=1673555745.7973425)      [(&#39;tile_f&#39;, [-1, 2, 8, 2]), (&#39;tile_y&#39;, [-1, 1, 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,8562480
-No: 11  GFLOPS: 0.00/443.05     result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 8, 16, 2]), (&#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, 1, 3]), (&#39;tile_rx&#39;, [-1, 3, 1]), (&#39;auto_unroll_max_step&#39;, 0), (&#39;unroll_explicit&#39;, 0)],None,1149588
+No: 8   GFLOPS: 0.00/825.44     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 592, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 544, in _build_func_common
@@ -1432,8 +1429,8 @@ Traceback (most recent call last):
   File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 56, in tvm._ffi._cy3.core.tvm_callback
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 875, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 16, 1, 1]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 2, 256]), (&#39;tile_ry&#39;, [-1, 1, 1]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 512), (&#39;unroll_explicit&#39;, 0)],None,3091004
-No: 12  GFLOPS: 0.00/443.05     result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 4, 8, 2]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 16, 16]), (&#39;tile_ry&#39;, [-1, 1, 3]), (&#39;tile_rx&#39;, [-1, 1, 1]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 1)],None,9236341
+No: 9   GFLOPS: 0.00/825.44     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 592, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 544, in _build_func_common
@@ -1555,10 +1552,10 @@ 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 875, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 4, 1, 128]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 64, 8]), (&#39;tile_ry&#39;, [-1, 3, 1]), (&#39;tile_rx&#39;, [-1, 1, 1]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 1)],None,9022412
-No: 13  GFLOPS: 323.70/443.05   result: MeasureResult(costs=(0.0007151712485875707,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.3592932224273682, timestamp=1673555748.9529142)      [(&#39;tile_f&#39;, [-1, 1, 4, 2]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 1, 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,7393832
-No: 14  GFLOPS: 0.83/443.05     result: MeasureResult(costs=(0.279625122,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.254995346069336, timestamp=1673555752.9961503) [(&#39;tile_f&#39;, [-1, 1, 4, 128]), (&#39;tile_y&#39;, [-1, 7, 1, 1]), (&#39;tile_x&#39;, [-1, 7, 1, 1]), (&#39;tile_rc&#39;, [-1, 2, 2]), (&#39;tile_ry&#39;, [-1, 1, 1]), (&#39;tile_rx&#39;, [-1, 3, 1]), (&#39;auto_unroll_max_step&#39;, 512), (&#39;unroll_explicit&#39;, 0)],None,2363235
-No: 15  GFLOPS: 0.00/443.05     result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 32, 4, 2]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 512, 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;, 1)],None,5260717
+No: 10  GFLOPS: 72.60/825.44    result: MeasureResult(costs=(0.003188784342857143,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.266326665878296, timestamp=1673569922.8719883)        [(&#39;tile_f&#39;, [-1, 2, 2, 4]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 4, 4]), (&#39;tile_ry&#39;, [-1, 1, 1]), (&#39;tile_rx&#39;, [-1, 3, 1]), (&#39;auto_unroll_max_step&#39;, 512), (&#39;unroll_explicit&#39;, 1)],None,7627509
+No: 11  GFLOPS: 369.71/825.44   result: MeasureResult(costs=(0.0006261778949416343,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.3138256072998047, timestamp=1673569923.858065)       [(&#39;tile_f&#39;, [-1, 8, 2, 1]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 2, 2]), (&#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;, 1)],None,9721153
+No: 12  GFLOPS: 0.00/825.44     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 592, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 544, in _build_func_common
@@ -1680,132 +1677,162 @@ Traceback (most recent call last):
   File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 56, in tvm._ffi._cy3.core.tvm_callback
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 875, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 2, 2, 2]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 32, 16]), (&#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,139545
-No: 16  GFLOPS: 11.83/443.05    result: MeasureResult(costs=(0.019574101666666666,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.9008331298828125, timestamp=1673555753.8789542)       [(&#39;tile_f&#39;, [-1, 2, 2, 32]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 2, 4]), (&#39;tile_ry&#39;, [-1, 1, 3]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 0)],None,5106391
-No: 17  GFLOPS: 0.00/443.05     result: Traceback (most recent call last):
-  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 592, in __call__
-    func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
-  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 544, in _build_func_common
-    func = build(s, args, target_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)
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 64, 4, 2]), (&#39;tile_y&#39;, [-1, 7, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 8, 1]), (&#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,400698
+No: 13  GFLOPS: 0.00/825.44     result: Traceback (most recent call last):
+  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 742, 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 706, 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)
   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/./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/./base.pxi&quot;, line 181, in tvm._ffi._cy3.core.CHECK_CALL
 tvm._ffi.base.TVMError: Traceback (most recent call last):
-  24: TVMFuncCall
+  4: 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:1730
-  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:1670
-  19: run&lt;&gt;
-        at ../include/tvm/runtime/packed_func.h:1630
-  18: run&lt;tvm::runtime::TVMMovableArgValueWithContext_&gt;
-        at ../include/tvm/runtime/packed_func.h:1630
-  17: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
-        at ../include/tvm/runtime/packed_func.h:1630
-  16: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
-        at ../include/tvm/runtime/packed_func.h:1630
-  15: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
-        at ../include/tvm/runtime/packed_func.h:1630
-  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:1645
-  13: operator()
-        at ../src/driver/driver_api.cc:395
-  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:381
-  11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array&lt;tvm::transform::Pass, void&gt;)
-        at ../src/driver/driver_api.cc:276
-  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:454
-  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:1749
-  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:1693
-  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:1617
-  2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
+  3: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
         at ../include/tvm/runtime/packed_func.h:1217
-  1: Call
-        at ../include/tvm/runtime/packed_func.h:1213
-  0: operator()
-        at ../src/runtime/c_runtime_api.cc:534
-  File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 56, in tvm._ffi._cy3.core.tvm_callback
-  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 875, in verify_pass
-    raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel
+  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):
-  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:1730
-  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:1670
-  19: run&lt;&gt;
-        at ../include/tvm/runtime/packed_func.h:1630
-  18: run&lt;tvm::runtime::TVMMovableArgValueWithContext_&gt;
-        at ../include/tvm/runtime/packed_func.h:1630
-  17: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
-        at ../include/tvm/runtime/packed_func.h:1630
-  16: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
-        at ../include/tvm/runtime/packed_func.h:1630
-  15: run&lt;tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_, tvm::runtime::TVMMovableArgValueWithContext_&gt;
-        at ../include/tvm/runtime/packed_func.h:1630
-  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:1645
-  13: operator()
-        at ../src/driver/driver_api.cc:395
-  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:381
-  11: tvm::LowerWithPassList(tvm::IRModule, tvm::runtime::Array&lt;tvm::transform::Pass, void&gt;)
-        at ../src/driver/driver_api.cc:276
-  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:454
-  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:1749
-  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:1693
-  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
+  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 706, 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 746, 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: 0x00007f839f345fa2
+  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
         at ../include/tvm/runtime/packed_func.h:1617
   2: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const
         at ../include/tvm/runtime/packed_func.h:1217
   1: Call
         at ../include/tvm/runtime/packed_func.h:1213
   0: operator()
-        at ../src/runtime/c_runtime_api.cc:534
-  File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 56, in tvm._ffi._cy3.core.tvm_callback
-  File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 875, in verify_pass
-    raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 2, 16, 4]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 128, 2]), (&#39;tile_ry&#39;, [-1, 1, 1]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 512), (&#39;unroll_explicit&#39;, 1)],None,8194467
-No: 18  GFLOPS: 0.00/443.05     result: Traceback (most recent call last):
+        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
+
+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, 8, 2, 8]), (&#39;tile_y&#39;, [-1, 7, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 4, 16]), (&#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;, 1)],None,5356926
+No: 14  GFLOPS: 40.92/825.44    result: MeasureResult(costs=(0.005657900444444444,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.81038236618042, timestamp=1673569933.718674)  [(&#39;tile_f&#39;, [-1, 1, 4, 1]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 16, 4]), (&#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,7053879
+No: 15  GFLOPS: 0.00/825.44     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 592, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 544, in _build_func_common
@@ -1927,8 +1954,10 @@ 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 875, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 32, 1, 1]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 7, 1, 1]), (&#39;tile_rc&#39;, [-1, 64, 1]), (&#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,7960265
-No: 19  GFLOPS: 0.00/443.05     result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 8, 8, 8]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 1, 4]), (&#39;tile_ry&#39;, [-1, 3, 1]), (&#39;tile_rx&#39;, [-1, 3, 1]), (&#39;auto_unroll_max_step&#39;, 0), (&#39;unroll_explicit&#39;, 1)],None,6069077
+No: 16  GFLOPS: 14.39/825.44    result: MeasureResult(costs=(0.01609306,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.153165578842163, timestamp=1673569934.541408)   [(&#39;tile_f&#39;, [-1, 2, 2, 1]), (&#39;tile_y&#39;, [-1, 1, 1, 7]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 2, 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;, 1)],None,8642271
+No: 17  GFLOPS: 78.30/825.44    result: MeasureResult(costs=(0.002956405617647059,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.077221393585205, timestamp=1673569939.8245046)        [(&#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, 64, 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,4282530
+No: 18  GFLOPS: 0.00/825.44     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 592, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 544, in _build_func_common
@@ -2050,8 +2079,9 @@ Traceback (most recent call last):
   File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 56, in tvm._ffi._cy3.core.tvm_callback
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 875, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 32, 2, 1]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 128, 2]), (&#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,2773335
-No: 20  GFLOPS: 0.00/443.05     result: Traceback (most recent call last):
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 4, 16, 1]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 32, 8]), (&#39;tile_ry&#39;, [-1, 1, 3]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 1)],None,10375676
+No: 19  GFLOPS: 144.58/825.44   result: MeasureResult(costs=(0.0016012034600000002,), error_no=MeasureErrorNo.NO_ERROR, all_cost=5.168120622634888, timestamp=1673569940.8427498)       [(&#39;tile_f&#39;, [-1, 4, 1, 1]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 64, 1]), (&#39;tile_ry&#39;, [-1, 3, 1]), (&#39;tile_rx&#39;, [-1, 1, 1]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 1)],None,8928482
+No: 20  GFLOPS: 0.00/825.44     result: Traceback (most recent call last):
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 592, in __call__
     func, arg_info = _build_func_common(measure_input, self.runtime, **kwargs)
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 544, in _build_func_common
@@ -2173,7 +2203,7 @@ Traceback (most recent call last):
   File &quot;tvm/_ffi/_cython/./packed_func.pxi&quot;, line 56, in tvm._ffi._cy3.core.tvm_callback
   File &quot;/workspace/python/tvm/autotvm/measure/measure_methods.py&quot;, line 875, in verify_pass
     raise InstantiationError(&quot;Skipped because of invalid gpu kernel&quot;)
-tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 128, 2, 1]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 2, 64]), (&#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,357737
+tvm.autotvm.task.space.InstantiationError: Skipped because of invalid gpu kernel        [(&#39;tile_f&#39;, [-1, 8, 1, 16]), (&#39;tile_y&#39;, [-1, 1, 1, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 1]), (&#39;tile_rc&#39;, [-1, 512, 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,9905447
 </pre></div>
 </div>
 <p>Finally we can inspect the best config from log file, check correctness,
@@ -2212,9 +2242,9 @@ 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, 2, 8, 2]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 7, 1]), (&#39;tile_rc&#39;, [-1, 8, 4]), (&#39;tile_ry&#39;, [-1, 1, 1]), (&#39;tile_rx&#39;, [-1, 1, 3]), (&#39;auto_unroll_max_step&#39;, 512), (&#39;unroll_explicit&#39;, 0)],None,2983720
+[(&#39;tile_f&#39;, [-1, 1, 16, 1]), (&#39;tile_y&#39;, [-1, 1, 7, 1]), (&#39;tile_x&#39;, [-1, 1, 1, 7]), (&#39;tile_rc&#39;, [-1, 8, 1]), (&#39;tile_ry&#39;, [-1, 3, 1]), (&#39;tile_rx&#39;, [-1, 1, 1]), (&#39;auto_unroll_max_step&#39;, 1500), (&#39;unroll_explicit&#39;, 1)],None,8919274
 Finish loading 20 records
-Time cost of this operator: 0.000654
+Time cost of this operator: 0.000570
 </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_aot.html b/docs/how_to/work_with_microtvm/micro_aot.html
index 9ad69f2386..7ba2497dd5 100644
--- a/docs/how_to/work_with_microtvm/micro_aot.html
+++ b/docs/how_to/work_with_microtvm/micro_aot.html
@@ -246,7 +246,6 @@
 <li class="toctree-l3 current"><a class="current reference internal" href="#">microTVM Host-Driven AoT</a><ul>
 <li class="toctree-l4"><a class="reference internal" href="#install-microtvm-python-dependencies">Install microTVM Python dependencies</a></li>
 <li class="toctree-l4"><a class="reference internal" href="#install-zephyr">Install Zephyr</a></li>
-<li class="toctree-l4"><a class="reference internal" href="#install-cmsis-nn">Install CMSIS-NN</a></li>
 <li class="toctree-l4"><a class="reference internal" href="#import-python-dependencies">Import Python dependencies</a></li>
 <li class="toctree-l4"><a class="reference internal" href="#import-a-tflite-model">Import a TFLite model</a></li>
 <li class="toctree-l4"><a class="reference internal" href="#defining-the-target">Defining the target</a></li>
@@ -411,7 +410,7 @@ pip install <span class="nv">pyserial</span><span class="o">==</span><span class
 <span class="c1"># By default, this tutorial runs on x86 CPU using TVM&#39;s C runtime. If you would like</span>
 <span class="c1"># to run on real Zephyr hardware, you must export the `TVM_MICRO_USE_HW` environment</span>
 <span class="c1"># variable. Otherwise (if you are using the C runtime), you can skip installing</span>
-<span class="c1"># Zephyr and CMSIS-NN. It takes ~20 minutes to install both of them.</span>
+<span class="c1"># Zephyr. It takes ~20 minutes to install Zephyr.</span>
 <a href="https://docs.python.org/3/library/functions.html#bool" title="builtins.bool" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">use_physical_hw</span></a> <span class="o">=</span> <span class="nb">bool</span><span class="p">(</span><a href="https://docs.python.org/3/library/os.html#os.getenv" title="os.getenv" class="sphx-glr-backref-module-os sphx-glr-backref-type-py-function"><span class="n">os</span><span class="o [...]
 </pre></div>
 </div>
@@ -449,22 +448,6 @@ python3 -m pip install -r <span class="s2">&quot;</span><span class="si">${</spa
 </div>
 </div></blockquote>
 </div>
-<div class="section" id="install-cmsis-nn">
-<h2>Install CMSIS-NN<a class="headerlink" href="#install-cmsis-nn" title="Permalink to this headline">¶</a></h2>
-<blockquote>
-<div><div class="highlight-bash notranslate"><div class="highlight"><pre><span></span>%%shell
-<span class="nv">CMSIS_SHA</span><span class="o">=</span><span class="s2">&quot;51263182d16c92649a48144ba56c0945f9fce60e&quot;</span>
-<span class="nv">CMSIS_URL</span><span class="o">=</span><span class="s2">&quot;http://github.com/ARM-software/CMSIS_5/archive/</span><span class="si">${</span><span class="nv">CMSIS_SHA</span><span class="si">}</span><span class="s2">.tar.gz&quot;</span>
-<span class="nb">export</span> <span class="nv">CMSIS_PATH</span><span class="o">=</span>/content/cmsis
-<span class="nv">DOWNLOAD_PATH</span><span class="o">=</span><span class="s2">&quot;/content/</span><span class="si">${</span><span class="nv">CMSIS_SHA</span><span class="si">}</span><span class="s2">.tar.gz&quot;</span>
-mkdir <span class="si">${</span><span class="nv">CMSIS_PATH</span><span class="si">}</span>
-wget <span class="si">${</span><span class="nv">CMSIS_URL</span><span class="si">}</span> -O <span class="s2">&quot;</span><span class="si">${</span><span class="nv">DOWNLOAD_PATH</span><span class="si">}</span><span class="s2">&quot;</span>
-tar -xf <span class="s2">&quot;</span><span class="si">${</span><span class="nv">DOWNLOAD_PATH</span><span class="si">}</span><span class="s2">&quot;</span> -C <span class="si">${</span><span class="nv">CMSIS_PATH</span><span class="si">}</span> --strip-components<span class="o">=</span><span class="m">1</span>
-rm <span class="si">${</span><span class="nv">DOWNLOAD_PATH</span><span class="si">}</span>
-</pre></div>
-</div>
-</div></blockquote>
-</div>
 <div class="section" id="import-python-dependencies">
 <h2>Import Python dependencies<a class="headerlink" href="#import-python-dependencies" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
@@ -562,7 +545,6 @@ respectively.</p>
         <span class="s2">&quot;board&quot;</span><span class="p">:</span> <span class="n">BOARD</span><span class="p">,</span>
         <span class="s2">&quot;serial_number&quot;</span><span class="p">:</span> <span class="n">SERIAL</span><span class="p">,</span>
         <span class="s2">&quot;config_main_stack_size&quot;</span><span class="p">:</span> <span class="mi">4096</span><span class="p">,</span>
-        <span class="s2">&quot;cmsis_path&quot;</span><span class="p">:</span> <a href="https://docs.python.org/3/library/os.html#os.getenv" title="os.getenv" class="sphx-glr-backref-module-os sphx-glr-backref-type-py-function"><span class="n">os</span><span class="o">.</span><span class="n">getenv</span></a><span class="p">(</span><span class="s2">&quot;CMSIS_PATH&quot;</span><span class="p">,</span> <span class="n">default</span><span class="o">=</span><span class="s2">&quot;/content/c [...]
         <span class="s2">&quot;zephyr_base&quot;</span><span class="p">:</span> <a href="https://docs.python.org/3/library/os.html#os.getenv" title="os.getenv" class="sphx-glr-backref-module-os sphx-glr-backref-type-py-function"><span class="n">os</span><span class="o">.</span><span class="n">getenv</span></a><span class="p">(</span><span class="s2">&quot;ZEPHYR_BASE&quot;</span><span class="p">,</span> <span class="n">default</span><span class="o">=</span><span class="s2">&quot;/content [...]
     <span class="p">}</span>
 
diff --git a/docs/how_to/work_with_microtvm/micro_autotune.html b/docs/how_to/work_with_microtvm/micro_autotune.html
index 4bf55c0929..c5e5682fea 100644
--- a/docs/how_to/work_with_microtvm/micro_autotune.html
+++ b/docs/how_to/work_with_microtvm/micro_autotune.html
@@ -247,7 +247,6 @@
 <li class="toctree-l3 current"><a class="current reference internal" href="#">Autotuning with microTVM</a><ul>
 <li class="toctree-l4"><a class="reference internal" href="#install-microtvm-python-dependencies">Install microTVM Python dependencies</a></li>
 <li class="toctree-l4"><a class="reference internal" href="#install-zephyr">Install Zephyr</a></li>
-<li class="toctree-l4"><a class="reference internal" href="#install-cmsis-nn">Install CMSIS-NN</a></li>
 <li class="toctree-l4"><a class="reference internal" href="#import-python-dependencies">Import Python dependencies</a></li>
 </ul>
 </li>
@@ -397,7 +396,7 @@ pip install <span class="nv">pyserial</span><span class="o">==</span><span class
 </pre></div>
 </div>
 </div></blockquote>
-<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="c1"># You can skip the following two sections (installing Zephyr and CMSIS-NN) if the following flag is False.</span>
+<div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="c1"># You can skip the following section (installing Zephyr) if the following flag is False.</span>
 <span class="c1"># Installing Zephyr takes ~20 min.</span>
 <span class="kn">import</span> <span class="nn">os</span>
 
@@ -438,22 +437,6 @@ python3 -m pip install -r <span class="s2">&quot;</span><span class="si">${</spa
 </div>
 </div></blockquote>
 </div>
-<div class="section" id="install-cmsis-nn">
-<h2>Install CMSIS-NN<a class="headerlink" href="#install-cmsis-nn" title="Permalink to this headline">¶</a></h2>
-<blockquote>
-<div><div class="highlight-bash notranslate"><div class="highlight"><pre><span></span>%%shell
-<span class="nv">CMSIS_SHA</span><span class="o">=</span><span class="s2">&quot;51263182d16c92649a48144ba56c0945f9fce60e&quot;</span>
-<span class="nv">CMSIS_URL</span><span class="o">=</span><span class="s2">&quot;http://github.com/ARM-software/CMSIS_5/archive/</span><span class="si">${</span><span class="nv">CMSIS_SHA</span><span class="si">}</span><span class="s2">.tar.gz&quot;</span>
-<span class="nb">export</span> <span class="nv">CMSIS_PATH</span><span class="o">=</span>/content/cmsis
-<span class="nv">DOWNLOAD_PATH</span><span class="o">=</span><span class="s2">&quot;/content/</span><span class="si">${</span><span class="nv">CMSIS_SHA</span><span class="si">}</span><span class="s2">.tar.gz&quot;</span>
-mkdir <span class="si">${</span><span class="nv">CMSIS_PATH</span><span class="si">}</span>
-wget <span class="si">${</span><span class="nv">CMSIS_URL</span><span class="si">}</span> -O <span class="s2">&quot;</span><span class="si">${</span><span class="nv">DOWNLOAD_PATH</span><span class="si">}</span><span class="s2">&quot;</span>
-tar -xf <span class="s2">&quot;</span><span class="si">${</span><span class="nv">DOWNLOAD_PATH</span><span class="si">}</span><span class="s2">&quot;</span> -C <span class="si">${</span><span class="nv">CMSIS_PATH</span><span class="si">}</span> --strip-components<span class="o">=</span><span class="m">1</span>
-rm <span class="si">${</span><span class="nv">DOWNLOAD_PATH</span><span class="si">}</span>
-</pre></div>
-</div>
-</div></blockquote>
-</div>
 <div class="section" id="import-python-dependencies">
 <h2>Import Python dependencies<a class="headerlink" href="#import-python-dependencies" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">json</span>
@@ -663,10 +646,10 @@ 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  310.8     98.721   (1, 2, 10, 10, 3)  2       1        [310.8]
-tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       3.045     0.967    (1, 6, 10, 10)     1       1        [3.045]
-tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.981     0.312    (1, 1, 10, 10, 3)  1       1        [0.981]
-Total_time                                    -                                             314.826   -        -                  -       -        -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc  tvmgen_default_fused_nn_contrib_conv2d_NCHWc  311.0     98.714   (1, 2, 10, 10, 3)  2       1        [311.0]
+tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       3.067     0.974    (1, 6, 10, 10)     1       1        [3.067]
+tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.984     0.312    (1, 1, 10, 10, 3)  1       1        [0.984]
+Total_time                                    -                                             315.051   -        -                  -       -        -
 </pre></div>
 </div>
 </div>
@@ -718,10 +701,10 @@ 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  134.2     97.923   (1, 6, 10, 10, 1)  2       1        [134.2]
-tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       1.86      1.357    (1, 6, 10, 10)     1       1        [1.86]
-tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.986     0.72     (1, 1, 10, 10, 3)  1       1        [0.986]
-Total_time                                    -                                             137.046   -        -                  -       -        -
+tvmgen_default_fused_nn_contrib_conv2d_NCHWc  tvmgen_default_fused_nn_contrib_conv2d_NCHWc  100.1     97.309   (1, 6, 10, 10, 1)  2       1        [100.1]
+tvmgen_default_fused_layout_transform_1       tvmgen_default_fused_layout_transform_1       1.794     1.744    (1, 6, 10, 10)     1       1        [1.794]
+tvmgen_default_fused_layout_transform         tvmgen_default_fused_layout_transform         0.975     0.948    (1, 1, 10, 10, 3)  1       1        [0.975]
+Total_time                                    -                                             102.869   -        -                  -       -        -
 </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 5f09653e72..24973a05fd 100644
--- a/docs/how_to/work_with_microtvm/micro_pytorch.html
+++ b/docs/how_to/work_with_microtvm/micro_pytorch.html
@@ -453,8 +453,8 @@ 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]
- 61%|######    | 2.09M/3.42M [00:00&lt;00:00, 16.6MB/s]
-100%|##########| 3.42M/3.42M [00:00&lt;00:00, 26.1MB/s]
+ 61%|######    | 2.09M/3.42M [00:00&lt;00:00, 20.7MB/s]
+100%|##########| 3.42M/3.42M [00:00&lt;00:00, 32.5MB/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.
@@ -578,7 +578,7 @@ 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  2.083 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  3.519 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_tflite.html b/docs/how_to/work_with_microtvm/micro_tflite.html
index b8ec5aad3f..195c704ce2 100644
--- a/docs/how_to/work_with_microtvm/micro_tflite.html
+++ b/docs/how_to/work_with_microtvm/micro_tflite.html
@@ -251,7 +251,6 @@
 <li class="toctree-l3 current"><a class="current reference internal" href="#">microTVM with TFLite Models</a><ul>
 <li class="toctree-l4"><a class="reference internal" href="#install-microtvm-python-dependencies">Install microTVM Python dependencies</a></li>
 <li class="toctree-l4"><a class="reference internal" href="#install-zephyr">Install Zephyr</a></li>
-<li class="toctree-l4"><a class="reference internal" href="#install-cmsis-nn">Install CMSIS-NN</a></li>
 <li class="toctree-l4"><a class="reference internal" href="#import-python-dependencies">Import Python dependencies</a></li>
 <li class="toctree-l4"><a class="reference internal" href="#defining-the-target">Defining the target</a></li>
 </ul>
@@ -402,7 +401,7 @@ pip install <span class="nv">pyserial</span><span class="o">==</span><span class
 <span class="c1"># By default, this tutorial runs on x86 CPU using TVM&#39;s C runtime. If you would like</span>
 <span class="c1"># to run on real Zephyr hardware, you must export the `TVM_MICRO_USE_HW` environment</span>
 <span class="c1"># variable. Otherwise (if you are using the C runtime), you can skip installing</span>
-<span class="c1"># Zephyr and CMSIS-NN. It takes ~20 minutes to install both of them.</span>
+<span class="c1"># Zephyr. It takes ~20 minutes to install Zephyr.</span>
 <a href="https://docs.python.org/3/library/functions.html#bool" title="builtins.bool" class="sphx-glr-backref-module-builtins sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">use_physical_hw</span></a> <span class="o">=</span> <span class="nb">bool</span><span class="p">(</span><a href="https://docs.python.org/3/library/os.html#os.getenv" title="os.getenv" class="sphx-glr-backref-module-os sphx-glr-backref-type-py-function"><span class="n">os</span><span class="o [...]
 </pre></div>
 </div>
@@ -440,22 +439,6 @@ python3 -m pip install -r <span class="s2">&quot;</span><span class="si">${</spa
 </div>
 </div></blockquote>
 </div>
-<div class="section" id="install-cmsis-nn">
-<h2>Install CMSIS-NN<a class="headerlink" href="#install-cmsis-nn" title="Permalink to this headline">¶</a></h2>
-<blockquote>
-<div><div class="highlight-bash notranslate"><div class="highlight"><pre><span></span>%%shell
-<span class="nv">CMSIS_SHA</span><span class="o">=</span><span class="s2">&quot;51263182d16c92649a48144ba56c0945f9fce60e&quot;</span>
-<span class="nv">CMSIS_URL</span><span class="o">=</span><span class="s2">&quot;http://github.com/ARM-software/CMSIS_5/archive/</span><span class="si">${</span><span class="nv">CMSIS_SHA</span><span class="si">}</span><span class="s2">.tar.gz&quot;</span>
-<span class="nb">export</span> <span class="nv">CMSIS_PATH</span><span class="o">=</span>/content/cmsis
-<span class="nv">DOWNLOAD_PATH</span><span class="o">=</span><span class="s2">&quot;/content/</span><span class="si">${</span><span class="nv">CMSIS_SHA</span><span class="si">}</span><span class="s2">.tar.gz&quot;</span>
-mkdir <span class="si">${</span><span class="nv">CMSIS_PATH</span><span class="si">}</span>
-wget <span class="si">${</span><span class="nv">CMSIS_URL</span><span class="si">}</span> -O <span class="s2">&quot;</span><span class="si">${</span><span class="nv">DOWNLOAD_PATH</span><span class="si">}</span><span class="s2">&quot;</span>
-tar -xf <span class="s2">&quot;</span><span class="si">${</span><span class="nv">DOWNLOAD_PATH</span><span class="si">}</span><span class="s2">&quot;</span> -C <span class="si">${</span><span class="nv">CMSIS_PATH</span><span class="si">}</span> --strip-components<span class="o">=</span><span class="m">1</span>
-rm <span class="si">${</span><span class="nv">DOWNLOAD_PATH</span><span class="si">}</span>
-</pre></div>
-</div>
-</div></blockquote>
-</div>
 <div class="section" id="import-python-dependencies">
 <h2>Import Python dependencies<a class="headerlink" href="#import-python-dependencies" title="Permalink to this headline">¶</a></h2>
 <div class="highlight-default notranslate"><div class="highlight"><pre><span></span><span class="kn">import</span> <span class="nn">json</span>
@@ -623,7 +606,6 @@ QEMU VM based on BOARD. In the example below the x86 arch is selected and a x86
         <span class="s2">&quot;board&quot;</span><span class="p">:</span> <span class="n">BOARD</span><span class="p">,</span>
         <span class="s2">&quot;serial_number&quot;</span><span class="p">:</span> <span class="n">SERIAL</span><span class="p">,</span>
         <span class="s2">&quot;config_main_stack_size&quot;</span><span class="p">:</span> <span class="mi">4096</span><span class="p">,</span>
-        <span class="s2">&quot;cmsis_path&quot;</span><span class="p">:</span> <a href="https://docs.python.org/3/library/os.html#os.getenv" title="os.getenv" class="sphx-glr-backref-module-os sphx-glr-backref-type-py-function"><span class="n">os</span><span class="o">.</span><span class="n">getenv</span></a><span class="p">(</span><span class="s2">&quot;CMSIS_PATH&quot;</span><span class="p">,</span> <span class="n">default</span><span class="o">=</span><span class="s2">&quot;/content/c [...]
         <span class="s2">&quot;zephyr_base&quot;</span><span class="p">:</span> <a href="https://docs.python.org/3/library/os.html#os.getenv" title="os.getenv" class="sphx-glr-backref-module-os sphx-glr-backref-type-py-function"><span class="n">os</span><span class="o">.</span><span class="n">getenv</span></a><span class="p">(</span><span class="s2">&quot;ZEPHYR_BASE&quot;</span><span class="p">,</span> <span class="n">default</span><span class="o">=</span><span class="s2">&quot;/content [...]
     <span class="p">}</span>
 
diff --git a/docs/how_to/work_with_microtvm/micro_train.html b/docs/how_to/work_with_microtvm/micro_train.html
index 7a308663e2..3cbe0ba497 100644
--- a/docs/how_to/work_with_microtvm/micro_train.html
+++ b/docs/how_to/work_with_microtvm/micro_train.html
@@ -523,7 +523,7 @@ 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/tmpbyatuwte/images/random&#39;
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>&#39;/tmp/tmpdg8c_6sp/images/random&#39;
 </pre></div>
 </div>
 </div>
@@ -583,8 +583,8 @@ 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], [1.0, 0.0], [1.0, 0.0], [1.0, 0.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [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/tmpbyatuwte/images/target contains 8144 images
-/tmp/tmpbyatuwte/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], [0.0, 1.0], [1.0, 0.0], [1.0, 0.0], [1.0, 0.0], [1.0, 0.0], [0.0, 1.0], [0.0, 1.0], [0.0, 1.0], [1.0, 0.0]" class = "sphx-glr-single-img"/><div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>/tmp/tmpdg8c_6sp/images/target contains 8144 images
+/tmp/tmpdg8c_6sp/images/random contains 5000 images
 </pre></div>
 </div>
 </div>
@@ -696,13 +696,13 @@ 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 - 46s - loss: 0.2104 - accuracy: 0.9268 - val_loss: 0.1214 - val_accuracy: 0.9585 - 46s/epoch - 141ms/step
+328/328 - 47s - loss: 0.2108 - accuracy: 0.9265 - val_loss: 0.1686 - val_accuracy: 0.9400 - 47s/epoch - 143ms/step
 Epoch 2/3
-328/328 - 42s - loss: 0.0854 - accuracy: 0.9678 - val_loss: 0.1452 - val_accuracy: 0.9551 - 42s/epoch - 129ms/step
+328/328 - 43s - loss: 0.0950 - accuracy: 0.9641 - val_loss: 0.1165 - val_accuracy: 0.9588 - 43s/epoch - 132ms/step
 Epoch 3/3
-328/328 - 43s - loss: 0.0654 - accuracy: 0.9766 - val_loss: 0.1017 - val_accuracy: 0.9634 - 43s/epoch - 131ms/step
+328/328 - 43s - loss: 0.0598 - accuracy: 0.9775 - val_loss: 0.1400 - val_accuracy: 0.9573 - 43s/epoch - 131ms/step
 
-&lt;keras.callbacks.History object at 0x7f7004188410&gt;
+&lt;keras.callbacks.History object at 0x7fcd419a44d0&gt;
 </pre></div>
 </div>
 </div>
@@ -962,7 +962,7 @@ 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  24.588 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 4 minutes  29.829 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 2d123cf8fd..c13429d908 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>06:28.891</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
+<p><strong>06:36.662</strong> total execution time for <strong>how_to_work_with_microtvm</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 83%" />
@@ -349,23 +349,23 @@
 </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>04:24.588</p></td>
+<td><p>04:29.829</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>01:02.083</p></td>
+<td><p>01:03.519</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>00:50.692</p></td>
+<td><p>00:51.606</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>00:07.742</p></td>
+<td><p>00:07.890</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>00:03.786</p></td>
+<td><p>00:03.819</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="micro_tvmc.html#sphx-glr-how-to-work-with-microtvm-micro-tvmc-py"><span class="std std-ref">Executing a Tiny Model with TVMC Micro</span></a> (<code class="docutils literal notranslate"><span class="pre">micro_tvmc.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 87d92a3165..c17e955c80 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>00:40.975</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
+<p><strong>00:44.127</strong> total execution time for <strong>how_to_work_with_relay</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 84%" />
@@ -349,15 +349,15 @@
 </colgroup>
 <tbody>
 <tr class="row-odd"><td><p><a class="reference internal" href="using_pipeline_executor.html#sphx-glr-how-to-work-with-relay-using-pipeline-executor-py"><span class="std std-ref">Using Pipeline Executor in Relay</span></a> (<code class="docutils literal notranslate"><span class="pre">using_pipeline_executor.py</span></code>)</p></td>
-<td><p>00:31.826</p></td>
+<td><p>00:32.385</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>00:07.588</p></td>
+<td><p>00:10.269</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><td><p><a class="reference internal" href="build_gcn.html#sphx-glr-how-to-work-with-relay-build-gcn-py"><span class="std std-ref">Building a Graph Convolutional Network</span></a> (<code class="docutils literal notranslate"><span class="pre">build_gcn.py</span></code>)</p></td>
-<td><p>00:01.555</p></td>
+<td><p>00:01.467</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="using_relay_viz.html#sphx-glr-how-to-work-with-relay-using-relay-viz-py"><span class="std std-ref">Use Relay Visualizer to Visualize Relay</span></a> (<code class="docutils literal notranslate"><span class="pre">using_relay_viz.py</span></code>)</p></td>
diff --git a/docs/how_to/work_with_schedules/intrin_math.html b/docs/how_to/work_with_schedules/intrin_math.html
index bd582ac58d..59f46d667b 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 0x7f6e32240950&gt;
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>&lt;function my_cuda_math_rule at 0x7fcce21f5680&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 2d79467a4e..739021eeca 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>00:07.381</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
+<p><strong>00:07.467</strong> total execution time for <strong>how_to_work_with_schedules</strong> files:</p>
 <table class="docutils align-default">
 <colgroup>
 <col style="width: 83%" />
@@ -349,7 +349,7 @@
 </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>00:04.854</p></td>
+<td><p>00:04.936</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>
@@ -357,15 +357,15 @@
 <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>00:00.581</p></td>
+<td><p>00:00.579</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>00:00.559</p></td>
+<td><p>00:00.564</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>00:00.113</p></td>
+<td><p>00:00.114</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>
diff --git a/docs/how_to/work_with_schedules/tensorize.html b/docs/how_to/work_with_schedules/tensorize.html
index 5fe9ce2529..fdedd38540 100644
--- a/docs/how_to/work_with_schedules/tensorize.html
+++ b/docs/how_to/work_with_schedules/tensorize.html
@@ -587,7 +587,7 @@ The importing needs to happen before the tensorized GEMV being executed.</p>
              B: Buffer(B_2: Pointer(float32), float32, [512, 64], []),
              C: Buffer(C_2: Pointer(float32), float32, [1024, 512], [])}
   buffer_map = {A_1: A, B_1: B, C_1: C} {
-  attr [IterVar(i: int32, (nullptr), &quot;DataPar&quot;, &quot;&quot;)] &quot;pragma_import_llvm&quot; = &quot;; ModuleID = &#39;/tmp/tmpz_jf9bov/input0.cc&#39;\nsource_filename = \&quot;/tmp/tmpz_jf9bov/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/tmp5m3vhvsv/input0.cc&#39;\nsource_filename = \&quot;/tmp/tmp5m3vhvsv/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 cd8b73699d..6f0fedaa35 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 2e761c3ca6..e7a3ba7806 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/be2580317/web/src/rpc_server.ts#L43">rpc_server.ts:43</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/rpc_server.ts#L43">rpc_server.ts:43</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/rpc_server.ts#L42">rpc_server.ts:42</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/rpc_server.ts#L63">rpc_server.ts:63</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/rpc_server.ts#L49">rpc_server.ts:49</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/rpc_server.ts#L57">rpc_server.ts:57</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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 274de55f9f..8f8af2a952 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/be2580317/web/src/memory.ts#L223">memory.ts:223</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L208">memory.ts:208</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L312">memory.ts:312</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L284">memory.ts:284</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L388">memory.ts:388</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L376">memory.ts:376</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L267">memory.ts:267</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L243">memory.ts:243</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L321">memory.ts:321</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L252">memory.ts:252</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L359">memory.ts:359</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L342">memory.ts:342</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L350">memory.ts:350</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L326">memory.ts:326</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L363">memory.ts:363</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L346">memory.ts:346</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L334">memory.ts:334</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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 5b37f4199b..4757e88729 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/be2580317/web/src/runtime.ts#L262">runtime.ts:262</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L260">runtime.ts:260</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L258">runtime.ts:258</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L262">runtime.ts:262</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L279">runtime.ts:279</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L270">runtime.ts:270</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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 55f9c61205..10d35c31d9 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/be2580317/web/src/runtime.ts#L202">runtime.ts:202</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L200">runtime.ts:200</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L198">runtime.ts:198</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L223">runtime.ts:223</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L230">runtime.ts:230</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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 983c326290..e010c32af9 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/be2580317/web/src/environment.ts#L86">environment.ts:86</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/environment.ts#L70">environment.ts:70</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/environment.ts#L69">environment.ts:69</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/environment.ts#L78">environment.ts:78</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/environment.ts#L84">environment.ts:84</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/environment.ts#L105">environment.ts:105</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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 d7f3868b01..81005279a9 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/be2580317/web/src/runtime.ts#L49">runtime.ts:49</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L46">runtime.ts:46</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L45">runtime.ts:45</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L44">runtime.ts:44</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L47">runtime.ts:47</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L76">runtime.ts:76</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L66">runtime.ts:66</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L84">runtime.ts:84</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L95">runtime.ts:95</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L72">runtime.ts:72</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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 7e3e220ec7..4af3914873 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/be2580317/web/src/runtime.ts#L583">runtime.ts:583</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L579">runtime.ts:579</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L654">runtime.ts:654</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L597">runtime.ts:597</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L631">runtime.ts:631</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L644">runtime.ts:644</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L621">runtime.ts:621</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L609">runtime.ts:609</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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 7da0fdaace..6f6454d72c 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/be2580317/web/src/runtime.ts#L692">runtime.ts:692</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L684">runtime.ts:684</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L683">runtime.ts:683</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L932">runtime.ts:932</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L994">runtime.ts:994</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L924">runtime.ts:924</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L732">runtime.ts:732</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L952">runtime.ts:952</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L816">runtime.ts:816</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L1033">runtime.ts:1033</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L846">runtime.ts:846</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L750">runtime.ts:750</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L1013">runtime.ts:1013</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L789">runtime.ts:789</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L914">runtime.ts:914</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L1145">runtime.ts:1145</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L740">runtime.ts:740</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L868">runtime.ts:868</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L857">runtime.ts:857</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L940">runtime.ts:940</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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 156b0532fc..8758c12a99 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/be2580317/web/src/memory.ts#L40">memory.ts:40</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L32">memory.ts:32</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L33">memory.ts:33</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L154">memory.ts:154</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L90">memory.ts:90</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L97">memory.ts:97</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L74">memory.ts:74</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L81">memory.ts:81</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L104">memory.ts:104</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L132">memory.ts:132</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L145">memory.ts:145</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L60">memory.ts:60</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L67">memory.ts:67</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L53">memory.ts:53</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L114">memory.ts:114</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L124">memory.ts:124</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/memory.ts#L175">memory.ts:175</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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 7192022910..2e6d89eb02 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/be2580317/web/src/runtime.ts#L504">runtime.ts:504</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L502">runtime.ts:502</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L516">runtime.ts:516</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L530">runtime.ts:530</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L561">runtime.ts:561</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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 55e3f52a74..bef7791188 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/be2580317/web/src/runtime.ts#L304">runtime.ts:304</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L297">runtime.ts:297</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L293">runtime.ts:293</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L289">runtime.ts:289</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L291">runtime.ts:291</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L295">runtime.ts:295</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L370">runtime.ts:370</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L414">runtime.ts:414</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L355">runtime.ts:355</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L474">runtime.ts:474</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L443">runtime.ts:443</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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 a454142af4..9cf03c2eca 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/be2580317/web/src/runtime.ts#L158">runtime.ts:158</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L157">runtime.ts:157</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L165">runtime.ts:165</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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 8ce1dd4aaa..1b9deb7cc4 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/be2580317/web/src/rpc_server.ts#L92">rpc_server.ts:92</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/rpc_server.ts#L82">rpc_server.ts:82</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/rpc_server.ts#L78">rpc_server.ts:78</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/rpc_server.ts#L81">rpc_server.ts:81</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/rpc_server.ts#L79">rpc_server.ts:79</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/rpc_server.ts#L80">rpc_server.ts:80</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/rpc_server.ts#L77">rpc_server.ts:77</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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 2eabc17fa3..4a855ad2b9 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/be2580317/web/src/runtime.ts#L145">runtime.ts:145</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L145">runtime.ts:145</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L143">runtime.ts:143</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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 811a776406..6a4ba9132b 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/be2580317/web/src/webgpu.ts#L57">webgpu.ts:57</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/webgpu.ts#L50">webgpu.ts:50</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/webgpu.ts#L51">webgpu.ts:51</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/webgpu.ts#L84">webgpu.ts:84</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/webgpu.ts#L172">webgpu.ts:172</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/web/src/webgpu.ts#L172">webgpu.ts:172</a></li>
 								</ul>
 							</aside>
 							<div class="tsd-comment tsd-typography">
@@ -238,7 +238,7 @@
 						<li class="tsd-description">
 							<aside class="tsd-sources">
 								<ul>
-									<li>Defined in <a href="https://github.com/apache/tvm/blob/be2580317/web/src/webgpu.ts#L67">webgpu.ts:67</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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 dedda7b86a..4fcb2f2efa 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/be2580317/web/src/ctypes.ts#L220">ctypes.ts:220</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L216">ctypes.ts:216</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L214">ctypes.ts:214</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L218">ctypes.ts:218</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L226">ctypes.ts:226</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L221">ctypes.ts:221</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L219">ctypes.ts:219</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L223">ctypes.ts:223</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L227">ctypes.ts:227</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L222">ctypes.ts:222</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L228">ctypes.ts:228</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L217">ctypes.ts:217</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L224">ctypes.ts:224</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L225">ctypes.ts:225</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L215">ctypes.ts:215</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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 64c1b9e75e..6170b223aa 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/be2580317/web/src/runtime.ts#L676">runtime.ts:676</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L675">runtime.ts:675</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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 06795bf1f5..2bf49f0739 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/be2580317/web/src/runtime.ts#L242">runtime.ts:242</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L240">runtime.ts:240</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L243">runtime.ts:243</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L241">runtime.ts:241</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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 dd53167453..9012ecd094 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/be2580317/web/src/rpc_server.ts#L27">rpc_server.ts:27</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/rpc_server.ts#L28">rpc_server.ts:28</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/rpc_server.ts#L29">rpc_server.ts:29</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/rpc_server.ts#L32">rpc_server.ts:32</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/rpc_server.ts#L31">rpc_server.ts:31</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/rpc_server.ts#L30">rpc_server.ts:30</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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 6448bae2c0..0d9011333f 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/be2580317/web/src/ctypes.ts#L206">ctypes.ts:206</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L207">ctypes.ts:207</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L203">ctypes.ts:203</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L204">ctypes.ts:204</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L201">ctypes.ts:201</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L202">ctypes.ts:202</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L205">ctypes.ts:205</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L200">ctypes.ts:200</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L199">ctypes.ts:199</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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 f0efd16e97..760cbaf701 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/be2580317/web/src/ctypes.ts#L112">ctypes.ts:112</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L128">ctypes.ts:128</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L144">ctypes.ts:144</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L136">ctypes.ts:136</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L121">ctypes.ts:121</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L160">ctypes.ts:160</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L77">ctypes.ts:77</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L83">ctypes.ts:83</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L67">ctypes.ts:67</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L57">ctypes.ts:57</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L100">ctypes.ts:100</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L88">ctypes.ts:88</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L94">ctypes.ts:94</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L34">ctypes.ts:34</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L52">ctypes.ts:52</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L42">ctypes.ts:42</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L48">ctypes.ts:48</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L150">ctypes.ts:150</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L167">ctypes.ts:167</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L170">ctypes.ts:170</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L187">ctypes.ts:187</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L179">ctypes.ts:179</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L193">ctypes.ts:193</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/webgpu.ts#L25">webgpu.ts:25</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L36">runtime.ts:36</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L25">ctypes.ts:25</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/ctypes.ts#L28">ctypes.ts:28</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/rpc_server.ts#L36">rpc_server.ts:36</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/support.ts#L25">support.ts:25</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/support.ts#L39">support.ts:39</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/support.ts#L52">support.ts:52</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/compact.ts#L38">compact.ts:38</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/webgpu.ts#L30">webgpu.ts:30</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/environment.ts#L32">environment.ts:32</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/compact.ts#L24">compact.ts:24</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L1367">runtime.ts:1367</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/support.ts#L62">support.ts:62</a></li>
+									<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L246">runtime.ts:246</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L247">runtime.ts:247</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L248">runtime.ts:248</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L249">runtime.ts:249</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L250">runtime.ts:250</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L175">runtime.ts:175</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L176">runtime.ts:176</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L180">runtime.ts:180</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L177">runtime.ts:177</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L178">runtime.ts:178</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L179">runtime.ts:179</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L183">runtime.ts:183</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L186">runtime.ts:186</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L184">runtime.ts:184</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L185">runtime.ts:185</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L189">runtime.ts:189</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L187">runtime.ts:187</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L188">runtime.ts:188</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/runtime.ts#L190">runtime.ts:190</a></li>
+								<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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 8358699e2b..618eb81930 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/be2580317/web/src/types.ts#L52">types.ts:52</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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 fe516ae76a..9e656cbb86 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/be2580317/web/src/webgpu.ts#L41">webgpu.ts:41</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/webgpu.ts#L42">webgpu.ts:42</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/webgpu.ts#L40">webgpu.ts:40</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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 d1d5e3ff81..e5947b7bde 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/be2580317/web/src/types.ts#L34">types.ts:34</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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/be2580317/web/src/types.ts#L39">types.ts:39</a></li>
+							<li>Defined in <a href="https://github.com/apache/tvm/blob/bd3d93b6a/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 85207b715d..2cd8e92df0 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 15a1ffd5a1..7ac8aff3cd 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>00:25.777</strong> total execution time for <strong>topic_vta_tutorials_autotvm</strong> files:</p>
+<p><strong>00:26.209</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>00:25.770</p></td>
+<td><p>00:26.202</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 0400a6922b..7438815ea7 100644
--- a/docs/topic/vta/tutorials/frontend/deploy_classification.html
+++ b/docs/topic/vta/tutorials/frontend/deploy_classification.html
@@ -583,7 +583,7 @@ 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 28.91s!
+resnet18_v1 inference graph built in 29.13s!
 </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 15d96ea0ab..bb280cdfe3 100644
--- a/docs/topic/vta/tutorials/frontend/deploy_detection.html
+++ b/docs/topic/vta/tutorials/frontend/deploy_detection.html
@@ -601,7 +601,7 @@ 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 19.59s!
+yolov3-tiny inference graph built in 19.60s!
 </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 c85ace0b95..9ad1196683 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>01:32.127</strong> total execution time for <strong>topic_vta_tutorials_frontend</strong> files:</p>
+<p><strong>01:32.450</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>00:46.327</p></td>
+<td><p>00:46.469</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>00:45.800</p></td>
+<td><p>00:45.981</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 3d48acbc69..48e3e362f0 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>00:03.127</strong> total execution time for <strong>topic_vta_tutorials_optimize</strong> files:</p>
+<p><strong>00:03.308</strong> total execution time for <strong>topic_vta_tutorials_optimize</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="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>00:02.671</p></td>
+<td><p>00:02.838</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-even"><td><p><a class="reference internal" href="matrix_multiply_opt.html#sphx-glr-topic-vta-tutorials-optimize-matrix-multiply-opt-py"><span class="std std-ref">Matrix Multiply Blocking</span></a> (<code class="docutils literal notranslate"><span class="pre">matrix_multiply_opt.py</span></code>)</p></td>
-<td><p>00:00.457</p></td>
+<td><p>00:00.470</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 </tbody>
diff --git a/docs/topic/vta/tutorials/sg_execution_times.html b/docs/topic/vta/tutorials/sg_execution_times.html
index 3d29d8b6f2..1c9c0a2d31 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>00:00.843</strong> total execution time for <strong>topic_vta_tutorials</strong> files:</p>
+<p><strong>00:00.819</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>00:00.455</p></td>
+<td><p>00:00.434</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>00:00.387</p></td>
+<td><p>00:00.385</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 ec7941938f..f217252293 100644
--- a/docs/tutorial/auto_scheduler_matmul_x86.html
+++ b/docs/tutorial/auto_scheduler_matmul_x86.html
@@ -492,6 +492,9 @@ 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>*E
+</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: 95.280 ms
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Execution time of this operator: 97.624 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  16.612 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 1 minutes  20.637 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 a85e4ed956..9c8c0c0fd2 100644
--- a/docs/tutorial/autotvm_matmul_x86.html
+++ b/docs/tutorial/autotvm_matmul_x86.html
@@ -680,16 +680,16 @@ 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: 1.18/1.18       result: MeasureResult(costs=(0.22698483580000003,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.9494411945343018, timestamp=1673554357.5490327)        [(&#39;tile_y&#39;, [-1, 16]), (&#39;tile_x&#39;, [-1, 1])],None,4
-No: 2   GFLOPS: 12.32/12.32     result: MeasureResult(costs=(0.0217813936,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5843517780303955, timestamp=1673554358.9106257)       [(&#39;tile_y&#39;, [-1, 256]), (&#39;tile_x&#39;, [-1, 512])],None,98
-No: 3   GFLOPS: 9.90/12.32      result: MeasureResult(costs=(0.027110872399999996,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.671090841293335, timestamp=1673554359.6068351)        [(&#39;tile_y&#39;, [-1, 1]), (&#39;tile_x&#39;, [-1, 256])],None,80
-No: 4   GFLOPS: 2.04/12.32      result: MeasureResult(costs=(0.131435967,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.3466086387634277, timestamp=1673554362.718498) [(&#39;tile_y&#39;, [-1, 8]), (&#39;tile_x&#39;, [-1, 2])],None,13
-No: 5   GFLOPS: 9.15/12.32      result: MeasureResult(costs=(0.0293520404,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.7791948318481445, timestamp=1673554363.6109908)       [(&#39;tile_y&#39;, [-1, 16]), (&#39;tile_x&#39;, [-1, 32])],None,54
-No: 6   GFLOPS: 8.79/12.32      result: MeasureResult(costs=(0.0305245594,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.7350471019744873, timestamp=1673554365.1322844)       [(&#39;tile_y&#39;, [-1, 2]), (&#39;tile_x&#39;, [-1, 64])],None,61
-No: 7   GFLOPS: 11.79/12.32     result: MeasureResult(costs=(0.0227621998,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6264462471008301, timestamp=1673554365.7575874)       [(&#39;tile_y&#39;, [-1, 32]), (&#39;tile_x&#39;, [-1, 256])],None,85
-No: 8   GFLOPS: 10.55/12.32     result: MeasureResult(costs=(0.025449613399999998,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.700911283493042, timestamp=1673554366.4198508)        [(&#39;tile_y&#39;, [-1, 512]), (&#39;tile_x&#39;, [-1, 64])],None,69
-No: 9   GFLOPS: 1.62/12.32      result: MeasureResult(costs=(0.1657385202,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.8665077686309814, timestamp=1673554369.4074454)       [(&#39;tile_y&#39;, [-1, 2]), (&#39;tile_x&#39;, [-1, 2])],None,11
-No: 10  GFLOPS: 3.58/12.32      result: MeasureResult(costs=(0.0750699752,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.434582233428955, timestamp=1673554370.872784) [(&#39;tile_y&#39;, [-1, 16]), (&#39;tile_x&#39;, [-1, 8])],None,34
+No: 1   GFLOPS: 3.68/3.68       result: MeasureResult(costs=(0.0730355736,), error_no=MeasureErrorNo.NO_ERROR, all_cost=1.420715570449829, timestamp=1673568505.9202921)        [(&#39;tile_y&#39;, [-1, 128]), (&#39;tile_x&#39;, [-1, 16])],None,47
+No: 2   GFLOPS: 12.29/12.29     result: MeasureResult(costs=(0.0218502902,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6094512939453125, timestamp=1673568506.527237)        [(&#39;tile_y&#39;, [-1, 8]), (&#39;tile_x&#39;, [-1, 256])],None,83
+No: 3   GFLOPS: 7.71/12.29      result: MeasureResult(costs=(0.0348013332,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.8950433731079102, timestamp=1673568508.1103191)       [(&#39;tile_y&#39;, [-1, 1]), (&#39;tile_x&#39;, [-1, 128])],None,70
+No: 4   GFLOPS: 1.29/12.29      result: MeasureResult(costs=(0.2072951306,), error_no=MeasureErrorNo.NO_ERROR, all_cost=3.529980421066284, timestamp=1673568512.4440742)        [(&#39;tile_y&#39;, [-1, 1]), (&#39;tile_x&#39;, [-1, 2])],None,10
+No: 5   GFLOPS: 13.08/13.08     result: MeasureResult(costs=(0.0205211554,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.5846850872039795, timestamp=1673568513.1419537)       [(&#39;tile_y&#39;, [-1, 64]), (&#39;tile_x&#39;, [-1, 512])],None,96
+No: 6   GFLOPS: 1.88/13.08      result: MeasureResult(costs=(0.142679848,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.5142734050750732, timestamp=1673568516.44907)  [(&#39;tile_y&#39;, [-1, 512]), (&#39;tile_x&#39;, [-1, 4])],None,29
+No: 7   GFLOPS: 0.94/13.08      result: MeasureResult(costs=(0.285880802,), error_no=MeasureErrorNo.NO_ERROR, all_cost=4.803407192230225, timestamp=1673568521.2701445) [(&#39;tile_y&#39;, [-1, 32]), (&#39;tile_x&#39;, [-1, 2])],None,15
+No: 8   GFLOPS: 1.89/13.08      result: MeasureResult(costs=(0.141984141,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.5025081634521484, timestamp=1673568523.7981641)        [(&#39;tile_y&#39;, [-1, 1]), (&#39;tile_x&#39;, [-1, 8])],None,30
+No: 9   GFLOPS: 2.05/13.08      result: MeasureResult(costs=(0.130725027,), error_no=MeasureErrorNo.NO_ERROR, all_cost=2.294708251953125, timestamp=1673568526.2706785) [(&#39;tile_y&#39;, [-1, 256]), (&#39;tile_x&#39;, [-1, 4])],None,28
+No: 10  GFLOPS: 11.74/13.08     result: MeasureResult(costs=(0.0228681012,), error_no=MeasureErrorNo.NO_ERROR, all_cost=0.6322391033172607, timestamp=1673568526.8919046)       [(&#39;tile_y&#39;, [-1, 64]), (&#39;tile_x&#39;, [-1, 32])],None,56
 </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 02fd579671..23435f2df0 100644
--- a/docs/tutorial/autotvm_relay_x86.html
+++ b/docs/tutorial/autotvm_relay_x86.html
@@ -558,7 +558,7 @@ 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;: 505.9607368499974, &#39;median&#39;: 506.24863114999243, &#39;std&#39;: 1.0552857114823602}
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>{&#39;mean&#39;: 512.5318330900063, &#39;median&#39;: 512.4503830500089, &#39;std&#39;: 1.829093237816327}
 </pre></div>
 </div>
 </div>
@@ -710,177 +710,179 @@ 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:   22.18/  22.18 GFLOPS | Progress: (4/20) | 8.82 s
-[Task  1/25]  Current/Best:   13.47/  22.18 GFLOPS | Progress: (8/20) | 12.00 s
-[Task  1/25]  Current/Best:   19.22/  22.18 GFLOPS | Progress: (12/20) | 15.26 s
-[Task  1/25]  Current/Best:   23.51/  23.51 GFLOPS | Progress: (16/20) | 18.13 s
-[Task  1/25]  Current/Best:    3.38/  23.51 GFLOPS | Progress: (20/20) | 21.79 s Done.
+[Task  1/25]  Current/Best:   18.99/  18.99 GFLOPS | Progress: (4/20) | 7.50 s
+[Task  1/25]  Current/Best:   14.97/  18.99 GFLOPS | Progress: (8/20) | 11.73 s
+[Task  1/25]  Current/Best:   19.09/  19.09 GFLOPS | Progress: (12/20) | 14.33 s
+[Task  1/25]  Current/Best:   12.91/  22.42 GFLOPS | Progress: (16/20) | 16.37 s
+[Task  1/25]  Current/Best:   11.11/  22.42 GFLOPS | Progress: (20/20) | 19.49 s Done.
 
 [Task  2/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task  2/25]  Current/Best:   12.14/  16.55 GFLOPS | Progress: (4/20) | 3.59 s
-[Task  2/25]  Current/Best:   11.63/  22.11 GFLOPS | Progress: (8/20) | 5.07 s
-[Task  2/25]  Current/Best:   16.72/  22.11 GFLOPS | Progress: (12/20) | 6.53 s
-[Task  2/25]  Current/Best:   20.96/  22.11 GFLOPS | Progress: (16/20) | 8.89 s
-[Task  2/25]  Current/Best:   10.74/  22.11 GFLOPS | Progress: (20/20) | 10.86 s Done.
+[Task  2/25]  Current/Best:   16.87/  17.11 GFLOPS | Progress: (4/20) | 3.96 s
+[Task  2/25]  Current/Best:    5.77/  17.97 GFLOPS | Progress: (8/20) | 5.48 s
+[Task  2/25]  Current/Best:   17.79/  17.97 GFLOPS | Progress: (12/20) | 7.11 s
+[Task  2/25]  Current/Best:   19.56/  20.75 GFLOPS | Progress: (16/20) | 8.55 s
+[Task  2/25]  Current/Best:    3.34/  20.75 GFLOPS | Progress: (20/20) | 10.01 s Done.
 
 [Task  3/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task  3/25]  Current/Best:   18.92/  18.92 GFLOPS | Progress: (4/20) | 3.92 s
-[Task  3/25]  Current/Best:    9.03/  18.92 GFLOPS | Progress: (8/20) | 6.27 s
-[Task  3/25]  Current/Best:    1.63/  18.92 GFLOPS | Progress: (12/20) | 9.98 s
-[Task  3/25]  Current/Best:    3.17/  18.92 GFLOPS | Progress: (16/20) | 14.25 s
-[Task  3/25]  Current/Best:    5.63/  18.92 GFLOPS | Progress: (20/20) | 16.97 s Done.
+[Task  3/25]  Current/Best:   13.70/  23.08 GFLOPS | Progress: (4/20) | 3.73 s
+[Task  3/25]  Current/Best:    7.54/  23.08 GFLOPS | Progress: (8/20) | 6.78 s
+[Task  3/25]  Current/Best:    9.82/  23.08 GFLOPS | Progress: (12/20) | 10.22 s
+[Task  3/25]  Current/Best:    9.19/  23.08 GFLOPS | Progress: (16/20) | 12.50 s
+[Task  3/25]  Current/Best:   12.74/  23.08 GFLOPS | Progress: (20/20) | 15.01 s Done.
 
 [Task  4/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task  4/25]  Current/Best:    8.10/  12.09 GFLOPS | Progress: (4/20) | 6.44 s
-[Task  4/25]  Current/Best:   16.94/  16.94 GFLOPS | Progress: (8/20) | 8.46 s
-[Task  4/25]  Current/Best:    9.85/  16.94 GFLOPS | Progress: (12/20) | 14.77 s
-[Task  4/25]  Current/Best:   11.83/  21.54 GFLOPS | Progress: (16/20) | 16.68 s
-[Task  4/25]  Current/Best:   13.20/  21.54 GFLOPS | Progress: (20/20) | 18.85 s Done.
+[Task  4/25]  Current/Best:   12.66/  18.10 GFLOPS | Progress: (4/20) | 4.54 s
+[Task  4/25]  Current/Best:   11.99/  18.10 GFLOPS | Progress: (8/20) | 10.14 s
+[Task  4/25]  Current/Best:    6.42/  18.10 GFLOPS | Progress: (12/20) | 12.24 s
+[Task  4/25]  Current/Best:    8.34/  18.10 GFLOPS | Progress: (16/20) | 14.48 s
+[Task  4/25]  Current/Best:   16.66/  18.10 GFLOPS | Progress: (20/20) | 18.45 s Done.
 
 [Task  5/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task  5/25]  Current/Best:   14.11/  16.26 GFLOPS | Progress: (4/20) | 3.51 s
-[Task  5/25]  Current/Best:   11.25/  20.76 GFLOPS | Progress: (8/20) | 5.61 s
-[Task  5/25]  Current/Best:    5.33/  20.76 GFLOPS | Progress: (12/20) | 8.01 s
-[Task  5/25]  Current/Best:   18.35/  20.76 GFLOPS | Progress: (16/20) | 10.22 s
-[Task  5/25]  Current/Best:   16.04/  20.76 GFLOPS | Progress: (20/20) | 12.68 s Done.
+[Task  5/25]  Current/Best:   15.51/  17.45 GFLOPS | Progress: (4/20) | 4.13 s
+[Task  5/25]  Current/Best:    8.46/  17.45 GFLOPS | Progress: (8/20) | 6.38 s
+[Task  5/25]  Current/Best:   18.06/  18.06 GFLOPS | Progress: (12/20) | 8.72 s
+[Task  5/25]  Current/Best:    3.07/  20.14 GFLOPS | Progress: (16/20) | 10.80 s
+[Task  5/25]  Current/Best:    4.86/  20.14 GFLOPS | Progress: (20/20) | 14.36 s Done.
 
 [Task  6/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task  6/25]  Current/Best:   17.88/  17.88 GFLOPS | Progress: (4/20) | 4.55 s
-[Task  6/25]  Current/Best:   13.05/  17.88 GFLOPS | Progress: (8/20) | 8.21 s
-[Task  6/25]  Current/Best:    9.49/  18.40 GFLOPS | Progress: (12/20) | 10.65 s
-[Task  6/25]  Current/Best:   18.05/  18.40 GFLOPS | Progress: (16/20) | 13.67 s
-[Task  6/25]  Current/Best:   11.04/  18.40 GFLOPS | Progress: (20/20) | 16.50 s Done.
+[Task  6/25]  Current/Best:   14.64/  17.80 GFLOPS | Progress: (4/20) | 4.66 s
+[Task  6/25]  Current/Best:   12.91/  22.51 GFLOPS | Progress: (8/20) | 7.25 s
+[Task  6/25]  Current/Best:   11.60/  22.51 GFLOPS | Progress: (12/20) | 11.12 s
+[Task  6/25]  Current/Best:   13.77/  22.51 GFLOPS | Progress: (16/20) | 13.38 s
+[Task  6/25]  Current/Best:    4.65/  22.51 GFLOPS | Progress: (20/20) | 17.60 s Done.
 
 [Task  7/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task  7/25]  Current/Best:    7.62/  18.65 GFLOPS | Progress: (4/20) | 5.36 s
-[Task  7/25]  Current/Best:   19.22/  19.22 GFLOPS | Progress: (8/20) | 7.52 s
-[Task  7/25]  Current/Best:   20.00/  20.00 GFLOPS | Progress: (12/20) | 9.70 s
-[Task  7/25]  Current/Best:   17.74/  20.00 GFLOPS | Progress: (16/20) | 12.03 s
-[Task  7/25]  Current/Best:   15.30/  20.00 GFLOPS | Progress: (20/20) | 14.41 s Done.
+[Task  7/25]  Current/Best:   13.58/  20.78 GFLOPS | Progress: (4/20) | 4.43 s
+[Task  7/25]  Current/Best:   19.29/  20.78 GFLOPS | Progress: (8/20) | 6.70 s
+[Task  7/25]  Current/Best:   16.13/  20.78 GFLOPS | Progress: (12/20) | 9.08 s
+[Task  7/25]  Current/Best:   18.73/  20.78 GFLOPS | Progress: (16/20) | 11.39 s
+[Task  7/25]  Current/Best:    6.22/  20.78 GFLOPS | Progress: (20/20) | 13.80 s Done.
 
 [Task  8/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task  8/25]  Current/Best:   10.25/  16.61 GFLOPS | Progress: (4/20) | 5.86 s
-[Task  8/25]  Current/Best:   11.58/  16.61 GFLOPS | Progress: (8/20) | 11.19 s
-[Task  8/25]  Current/Best:    7.79/  16.61 GFLOPS | Progress: (12/20) | 20.85 s
-[Task  8/25]  Current/Best:    9.29/  16.61 GFLOPS | Progress: (16/20) | 29.58 s
-[Task  8/25]  Current/Best:   13.11/  16.61 GFLOPS | Progress: (20/20) | 32.76 s Done.
+[Task  8/25]  Current/Best:   10.10/  19.90 GFLOPS | Progress: (4/20) | 7.59 s
+[Task  8/25]  Current/Best:   11.43/  19.90 GFLOPS | Progress: (8/20) | 10.37 s
+[Task  8/25]  Current/Best:    9.24/  20.66 GFLOPS | Progress: (12/20) | 21.22 s
+[Task  8/25]  Current/Best:    5.16/  20.66 GFLOPS | Progress: (16/20) | 25.70 s
+[Task  8/25]  Current/Best:   11.94/  20.66 GFLOPS | Progress: (20/20) | 28.56 s Done.
 
 [Task  9/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task  9/25]  Current/Best:   12.94/  12.94 GFLOPS | Progress: (4/20) | 5.07 s
-[Task  9/25]  Current/Best:   18.42/  18.42 GFLOPS | Progress: (8/20) | 7.76 s
-[Task  9/25]  Current/Best:   14.00/  18.42 GFLOPS | Progress: (12/20) | 18.81 s
-[Task  9/25]  Current/Best:    9.97/  18.42 GFLOPS | Progress: (16/20) | 22.42 s
-[Task  9/25]  Current/Best:    7.47/  18.42 GFLOPS | Progress: (20/20) | 24.20 s
+[Task  9/25]  Current/Best:   18.82/  18.82 GFLOPS | Progress: (4/20) | 3.72 s
+[Task  9/25]  Current/Best:   11.40/  18.82 GFLOPS | Progress: (8/20) | 11.05 s
+[Task  9/25]  Current/Best:   12.33/  18.82 GFLOPS | Progress: (12/20) | 13.71 s
+[Task  9/25]  Current/Best:    8.19/  18.82 GFLOPS | Progress: (16/20) | 17.19 s
+[Task  9/25]  Current/Best:   17.06/  18.82 GFLOPS | Progress: (20/20) | 19.19 s Done.
+
 [Task 10/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 10/25]  Current/Best:   15.69/  20.68 GFLOPS | Progress: (4/20) | 3.82 s
-[Task 10/25]  Current/Best:   12.88/  20.68 GFLOPS | Progress: (8/20) | 5.70 s
-[Task 10/25]  Current/Best:   16.44/  20.68 GFLOPS | Progress: (12/20) | 7.80 s
-[Task 10/25]  Current/Best:    9.40/  20.68 GFLOPS | Progress: (16/20) | 9.90 s
-[Task 10/25]  Current/Best:   11.94/  20.68 GFLOPS | Progress: (20/20) | 12.26 s Done.
+[Task 10/25]  Current/Best:   14.24/  15.97 GFLOPS | Progress: (4/20) | 3.50 s
+[Task 10/25]  Current/Best:    6.23/  15.97 GFLOPS | Progress: (8/20) | 6.29 s
+[Task 10/25]  Current/Best:    9.59/  17.28 GFLOPS | Progress: (12/20) | 8.93 s
+[Task 10/25]  Current/Best:    3.95/  17.28 GFLOPS | Progress: (16/20) | 11.30 s
+[Task 10/25]  Current/Best:   20.11/  20.11 GFLOPS | Progress: (20/20) | 13.67 s Done.
 
 [Task 11/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 11/25]  Current/Best:   18.09/  20.97 GFLOPS | Progress: (4/20) | 4.24 s
-[Task 11/25]  Current/Best:   21.82/  21.82 GFLOPS | Progress: (8/20) | 6.61 s
-[Task 11/25]  Current/Best:   19.04/  21.82 GFLOPS | Progress: (12/20) | 9.21 s
-[Task 11/25]  Current/Best:   11.93/  21.82 GFLOPS | Progress: (16/20) | 12.53 s
-[Task 11/25]  Current/Best:   22.68/  22.68 GFLOPS | Progress: (20/20) | 15.73 s Done.
+[Task 11/25]  Current/Best:   18.42/  18.42 GFLOPS | Progress: (4/20) | 4.05 s
+[Task 11/25]  Current/Best:   21.61/  23.89 GFLOPS | Progress: (8/20) | 6.37 s
+[Task 11/25]  Current/Best:   18.11/  23.89 GFLOPS | Progress: (12/20) | 8.58 s
+[Task 11/25]  Current/Best:   12.34/  23.89 GFLOPS | Progress: (16/20) | 10.91 s
+[Task 11/25]  Current/Best:   12.19/  23.89 GFLOPS | Progress: (20/20) | 13.78 s Done.
 
 [Task 12/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 12/25]  Current/Best:   13.99/  17.74 GFLOPS | Progress: (4/20) | 5.12 s
-[Task 12/25]  Current/Best:    6.63/  17.74 GFLOPS | Progress: (8/20) | 11.38 s
-[Task 12/25]  Current/Best:    4.96/  17.74 GFLOPS | Progress: (12/20) | 13.79 s
-[Task 12/25]  Current/Best:    4.47/  18.45 GFLOPS | Progress: (16/20) | 18.19 s
-[Task 12/25]  Current/Best:   14.83/  18.70 GFLOPS | Progress: (20/20) | 22.38 s Done.
+[Task 12/25]  Current/Best:   14.94/  14.94 GFLOPS | Progress: (4/20) | 4.52 s
+[Task 12/25]  Current/Best:    2.97/  15.49 GFLOPS | Progress: (8/20) | 7.26 s
+[Task 12/25]  Current/Best:   11.94/  21.12 GFLOPS | Progress: (12/20) | 9.50 s
+[Task 12/25]  Current/Best:    4.84/  21.12 GFLOPS | Progress: (16/20) | 12.84 s
+[Task 12/25]  Current/Best:   11.86/  21.12 GFLOPS | Progress: (20/20) | 15.73 s Done.
 
 [Task 13/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 13/25]  Current/Best:   11.65/  22.81 GFLOPS | Progress: (4/20) | 4.95 s
-[Task 13/25]  Current/Best:   12.58/  22.81 GFLOPS | Progress: (8/20) | 8.17 s
-[Task 13/25]  Current/Best:   20.06/  22.81 GFLOPS | Progress: (12/20) | 10.42 s
-[Task 13/25]  Current/Best:    9.65/  22.81 GFLOPS | Progress: (16/20) | 13.27 s
-[Task 13/25]  Current/Best:   16.39/  22.81 GFLOPS | Progress: (20/20) | 15.45 s Done.
+[Task 13/25]  Current/Best:   11.53/  20.90 GFLOPS | Progress: (4/20) | 5.20 s
+[Task 13/25]  Current/Best:   14.12/  20.90 GFLOPS | Progress: (8/20) | 9.15 s
+[Task 13/25]  Current/Best:   17.57/  20.90 GFLOPS | Progress: (12/20) | 12.37 s
+[Task 13/25]  Current/Best:   10.26/  20.90 GFLOPS | Progress: (16/20) | 15.33 s
+[Task 13/25]  Current/Best:    1.57/  20.90 GFLOPS | Progress: (20/20) | 19.26 s Done.
 
 [Task 14/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 14/25]  Current/Best:   15.45/  15.59 GFLOPS | Progress: (4/20) | 4.57 s
-[Task 14/25]  Current/Best:   13.40/  15.59 GFLOPS | Progress: (8/20) | 7.00 s
-[Task 14/25]  Current/Best:    4.76/  15.59 GFLOPS | Progress: (12/20) | 9.51 s
-[Task 14/25]  Current/Best:   15.78/  15.78 GFLOPS | Progress: (16/20) | 12.48 s
-[Task 14/25]  Current/Best:   14.04/  15.78 GFLOPS | Progress: (20/20) | 15.88 s
-[Task 15/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 15/25]  Current/Best:   10.21/  21.19 GFLOPS | Progress: (4/20) | 3.86 s
-[Task 15/25]  Current/Best:    6.03/  21.19 GFLOPS | Progress: (8/20) | 5.80 s
-[Task 15/25]  Current/Best:   22.19/  22.19 GFLOPS | Progress: (12/20) | 7.96 s
-[Task 15/25]  Current/Best:    9.38/  22.19 GFLOPS | Progress: (16/20) | 10.24 s Done.
- Done.
+[Task 14/25]  Current/Best:    2.99/   6.21 GFLOPS | Progress: (4/20) | 5.80 s
+[Task 14/25]  Current/Best:   14.21/  15.80 GFLOPS | Progress: (8/20) | 7.89 s
+[Task 14/25]  Current/Best:   13.34/  15.80 GFLOPS | Progress: (12/20) | 14.85 s
+[Task 14/25]  Current/Best:   14.44/  15.80 GFLOPS | Progress: (16/20) | 18.94 s
+[Task 14/25]  Current/Best:    4.42/  15.80 GFLOPS | Progress: (20/20) | 21.41 s Done.
 
-[Task 15/25]  Current/Best:   18.45/  22.19 GFLOPS | Progress: (20/20) | 12.46 s Done.
+[Task 15/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
+[Task 15/25]  Current/Best:    7.87/  14.65 GFLOPS | Progress: (4/20) | 4.48 s
+[Task 15/25]  Current/Best:   18.79/  20.68 GFLOPS | Progress: (8/20) | 6.07 s
+[Task 15/25]  Current/Best:   11.74/  20.68 GFLOPS | Progress: (12/20) | 10.44 s
+[Task 15/25]  Current/Best:   20.43/  20.68 GFLOPS | Progress: (16/20) | 16.39 s
+[Task 15/25]  Current/Best:    3.01/  20.68 GFLOPS | Progress: (20/20) | 18.37 s Done.
 
 [Task 16/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 16/25]  Current/Best:   14.17/  20.16 GFLOPS | Progress: (4/20) | 4.32 s
-[Task 16/25]  Current/Best:   18.56/  20.16 GFLOPS | Progress: (8/20) | 6.92 s
-[Task 16/25]  Current/Best:   12.78/  20.16 GFLOPS | Progress: (12/20) | 9.67 s
-[Task 16/25]  Current/Best:   16.64/  20.16 GFLOPS | Progress: (16/20) | 11.17 s
-[Task 16/25]  Current/Best:   15.49/  20.16 GFLOPS | Progress: (20/20) | 13.40 s Done.
+[Task 16/25]  Current/Best:    9.90/  13.66 GFLOPS | Progress: (4/20) | 4.15 s
+[Task 16/25]  Current/Best:   13.27/  13.66 GFLOPS | Progress: (8/20) | 6.53 s
+[Task 16/25]  Current/Best:   16.19/  17.85 GFLOPS | Progress: (12/20) | 8.43 s
+[Task 16/25]  Current/Best:   12.24/  17.85 GFLOPS | Progress: (16/20) | 11.05 s
+[Task 16/25]  Current/Best:   14.57/  17.85 GFLOPS | Progress: (20/20) | 13.84 s Done.
 
 [Task 17/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 17/25]  Current/Best:   11.49/  23.62 GFLOPS | Progress: (4/20) | 4.51 s
-[Task 17/25]  Current/Best:   10.41/  23.62 GFLOPS | Progress: (8/20) | 6.72 s
-[Task 17/25]  Current/Best:    8.97/  23.62 GFLOPS | Progress: (12/20) | 9.30 s
-[Task 17/25]  Current/Best:    7.08/  23.62 GFLOPS | Progress: (16/20) | 12.37 s
-[Task 17/25]  Current/Best:   17.45/  23.62 GFLOPS | Progress: (20/20) | 15.15 s Done.
+[Task 17/25]  Current/Best:   14.15/  19.09 GFLOPS | Progress: (4/20) | 4.42 s
+[Task 17/25]  Current/Best:    3.10/  19.09 GFLOPS | Progress: (8/20) | 8.62 s
+[Task 17/25]  Current/Best:   21.86/  21.86 GFLOPS | Progress: (12/20) | 11.16 s
+[Task 17/25]  Current/Best:   11.29/  21.86 GFLOPS | Progress: (16/20) | 14.36 s
+[Task 17/25]  Current/Best:   12.50/  21.86 GFLOPS | Progress: (20/20) | 17.27 s Done.
 
 [Task 18/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 18/25]  Current/Best:    9.14/  16.07 GFLOPS | Progress: (4/20) | 7.36 s
-[Task 18/25]  Current/Best:    3.08/  16.07 GFLOPS | Progress: (8/20) | 10.87 s
-[Task 18/25]  Current/Best:   10.27/  18.23 GFLOPS | Progress: (12/20) | 16.16 s
-[Task 18/25]  Current/Best:   12.73/  18.23 GFLOPS | Progress: (16/20) | 21.78 s
-[Task 18/25]  Current/Best:   10.68/  18.23 GFLOPS | Progress: (20/20) | 24.11 s Done.
+[Task 18/25]  Current/Best:   21.60/  21.60 GFLOPS | Progress: (4/20) | 3.56 s
+[Task 18/25]  Current/Best:   19.15/  21.60 GFLOPS | Progress: (8/20) | 6.05 s
+[Task 18/25]  Current/Best:   18.63/  21.60 GFLOPS | Progress: (12/20) | 10.00 s
+[Task 18/25]  Current/Best:    9.16/  21.60 GFLOPS | Progress: (16/20) | 15.87 s
+[Task 18/25]  Current/Best:   16.30/  21.60 GFLOPS | Progress: (20/20) | 17.83 s Done.
 
 [Task 19/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 19/25]  Current/Best:   15.02/  18.39 GFLOPS | Progress: (4/20) | 6.23 s
-[Task 19/25]  Current/Best:   10.02/  18.39 GFLOPS | Progress: (8/20) | 9.65 s
-[Task 19/25]  Current/Best:    4.19/  18.85 GFLOPS | Progress: (12/20) | 13.30 s
-[Task 19/25]  Current/Best:   22.20/  22.20 GFLOPS | Progress: (16/20) | 16.95 s
-[Task 19/25]  Current/Best:   11.44/  22.20 GFLOPS | Progress: (20/20) | 20.29 s Done.
+[Task 19/25]  Current/Best:    5.35/  19.37 GFLOPS | Progress: (4/20) | 4.78 s
+[Task 19/25]  Current/Best:   23.04/  23.04 GFLOPS | Progress: (8/20) | 8.27 s
+[Task 19/25]  Current/Best:    3.08/  23.04 GFLOPS | Progress: (12/20) | 11.26 s
+[Task 19/25]  Current/Best:   18.84/  23.04 GFLOPS | Progress: (16/20) | 15.97 s
+[Task 19/25]  Current/Best:   19.81/  23.04 GFLOPS | Progress: (20/20) | 19.20 s Done.
 
 [Task 20/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 20/25]  Current/Best:   11.40/  13.88 GFLOPS | Progress: (4/20) | 5.26 s
-[Task 20/25]  Current/Best:   12.76/  15.84 GFLOPS | Progress: (8/20) | 7.53 s
-[Task 20/25]  Current/Best:   11.24/  16.41 GFLOPS | Progress: (12/20) | 9.65 s
-[Task 20/25]  Current/Best:   18.54/  18.54 GFLOPS | Progress: (16/20) | 12.10 s
-[Task 20/25]  Current/Best:   16.74/  18.54 GFLOPS | Progress: (20/20) | 15.29 s
+[Task 20/25]  Current/Best:   12.14/  22.53 GFLOPS | Progress: (4/20) | 3.89 s
+[Task 20/25]  Current/Best:   12.72/  22.53 GFLOPS | Progress: (8/20) | 6.54 s
+[Task 20/25]  Current/Best:   14.48/  22.53 GFLOPS | Progress: (12/20) | 9.19 s
+[Task 20/25]  Current/Best:   13.78/  22.53 GFLOPS | Progress: (16/20) | 13.30 s
+[Task 20/25]  Current/Best:   14.14/  22.53 GFLOPS | Progress: (20/20) | 19.77 s
 [Task 21/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 21/25]  Current/Best:   12.70/  15.40 GFLOPS | Progress: (4/20) | 4.31 s Done.
+[Task 21/25]  Current/Best:   16.36/  21.86 GFLOPS | Progress: (4/20) | 4.96 s Done.
+
+[Task 21/25]  Current/Best:    5.34/  21.86 GFLOPS | Progress: (8/20) | 6.64 s
+[Task 21/25]  Current/Best:    5.25/  21.86 GFLOPS | Progress: (12/20) | 10.04 s
+[Task 21/25]  Current/Best:   15.39/  21.86 GFLOPS | Progress: (16/20) | 11.79 s
+[Task 21/25]  Current/Best:    9.50/  21.86 GFLOPS | Progress: (20/20) | 15.04 s Done.
 
-[Task 21/25]  Current/Best:   20.95/  20.95 GFLOPS | Progress: (8/20) | 5.32 s
-[Task 21/25]  Current/Best:    2.28/  20.95 GFLOPS | Progress: (12/20) | 7.92 s
-[Task 21/25]  Current/Best:   17.84/  20.95 GFLOPS | Progress: (16/20) | 10.98 s
-[Task 21/25]  Current/Best:   10.74/  20.95 GFLOPS | Progress: (20/20) | 13.73 s
 [Task 22/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 22/25]  Current/Best:   12.02/  12.77 GFLOPS | Progress: (4/20) | 3.71 s
-[Task 22/25]  Current/Best:   10.33/  19.31 GFLOPS | Progress: (8/20) | 6.60 s
-[Task 22/25]  Current/Best:   15.37/  19.31 GFLOPS | Progress: (12/20) | 8.37 s
-[Task 22/25]  Current/Best:    6.91/  20.63 GFLOPS | Progress: (16/20) | 10.75 s
-[Task 22/25]  Current/Best:   15.80/  20.63 GFLOPS | Progress: (20/20) | 12.47 s Done.
+[Task 22/25]  Current/Best:   17.80/  18.58 GFLOPS | Progress: (4/20) | 4.00 s
+[Task 22/25]  Current/Best:   10.84/  18.58 GFLOPS | Progress: (8/20) | 6.42 s
+[Task 22/25]  Current/Best:   14.81/  18.58 GFLOPS | Progress: (12/20) | 8.67 s
+[Task 22/25]  Current/Best:   17.91/  18.58 GFLOPS | Progress: (16/20) | 10.47 s
+[Task 22/25]  Current/Best:   12.37/  18.58 GFLOPS | Progress: (20/20) | 13.46 s Done.
 
 [Task 23/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 23/25]  Current/Best:   10.46/  20.90 GFLOPS | Progress: (4/20) | 4.71 s
-[Task 23/25]  Current/Best:   19.69/  20.90 GFLOPS | Progress: (8/20) | 12.68 s
-[Task 23/25]  Current/Best:    3.08/  22.58 GFLOPS | Progress: (12/20) | 16.74 s
-[Task 23/25]  Current/Best:    6.17/  22.58 GFLOPS | Progress: (16/20) | 21.47 s
-[Task 23/25]  Current/Best:   23.04/  23.04 GFLOPS | Progress: (20/20) | 26.96 s Done.
+[Task 23/25]  Current/Best:    6.11/  11.95 GFLOPS | Progress: (4/20) | 7.70 s
+[Task 23/25]  Current/Best:   18.42/  18.42 GFLOPS | Progress: (8/20) | 11.12 s
+[Task 23/25]  Current/Best:   11.10/  19.30 GFLOPS | Progress: (12/20) | 13.87 s
+[Task 23/25]  Current/Best:   12.18/  23.83 GFLOPS | Progress: (16/20) | 16.29 s
+[Task 23/25]  Current/Best:   22.51/  23.83 GFLOPS | Progress: (20/20) | 18.84 s Done.
 
 [Task 24/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 24/25]  Current/Best:    8.31/   8.31 GFLOPS | Progress: (4/20) | 12.77 s
-[Task 24/25]  Current/Best:    5.96/   8.31 GFLOPS | Progress: (8/20) | 23.72 s
-[Task 24/25]  Current/Best:    8.88/   8.88 GFLOPS | Progress: (12/20) | 35.23 s
-[Task 24/25]  Current/Best:    1.37/  10.26 GFLOPS | Progress: (16/20) | 47.10 s
-[Task 24/25]  Current/Best:   10.11/  10.26 GFLOPS | Progress: (20/20) | 58.04 s
-[Task 25/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s
-[Task 25/25]  Current/Best:    3.46/   4.13 GFLOPS | Progress: (4/20) | 12.76 s
-[Task 25/25]  Current/Best:    8.56/   8.56 GFLOPS | Progress: (8/20) | 24.50 s
-[Task 25/25]  Current/Best:    5.64/   8.56 GFLOPS | Progress: (12/20) | 35.47 s
-[Task 25/25]  Current/Best:    2.99/   9.39 GFLOPS | Progress: (16/20) | 46.42 s
-[Task 25/25]  Current/Best:    7.84/   9.39 GFLOPS | Progress: (20/20) | 48.59 s Done.
+[Task 24/25]  Current/Best:    3.15/   6.49 GFLOPS | Progress: (4/20) | 9.85 s
+[Task 24/25]  Current/Best:    4.23/   8.54 GFLOPS | Progress: (8/20) | 20.80 s
+[Task 24/25]  Current/Best:    9.34/   9.34 GFLOPS | Progress: (12/20) | 31.93 s
+[Task 24/25]  Current/Best:    2.04/   9.34 GFLOPS | Progress: (16/20) | 43.83 s
+[Task 24/25]  Current/Best:    2.99/   9.34 GFLOPS | Progress: (20/20) | 54.51 s
+[Task 25/25]  Current/Best:    0.00/   0.00 GFLOPS | Progress: (0/20) | 0.00 s Done.
+
+[Task 25/25]  Current/Best:    1.53/   8.39 GFLOPS | Progress: (4/20) | 4.33 s
+[Task 25/25]  Current/Best:    3.71/   8.99 GFLOPS | Progress: (8/20) | 15.30 s
+[Task 25/25]  Current/Best:    8.83/   9.14 GFLOPS | Progress: (12/20) | 27.08 s
+[Task 25/25]  Current/Best:    3.00/   9.14 GFLOPS | Progress: (16/20) | 37.75 s
+[Task 25/25]  Current/Best:    7.87/   9.69 GFLOPS | Progress: (20/20) | 48.67 s
 </pre></div>
 </div>
 <p>The output from this tuning process will look something like this:</p>
@@ -942,7 +944,7 @@ model using optimized operators to speed up our computations.</p>
 </pre></div>
 </div>
 <div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>class=&#39;n02123045 tabby, tabby cat&#39; with probability=0.621104
-class=&#39;n02123159 tiger cat&#39; with probability=0.356378
+class=&#39;n02123159 tiger cat&#39; with probability=0.356379
 class=&#39;n02124075 Egyptian cat&#39; with probability=0.019712
 class=&#39;n02129604 tiger, Panthera tigris&#39; with probability=0.001215
 class=&#39;n04040759 radiator&#39; with probability=0.000262
@@ -979,8 +981,8 @@ 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;: 419.9832372800074, &#39;median&#39;: 419.7767239500081, &#39;std&#39;: 1.768500945522016}
-unoptimized: {&#39;mean&#39;: 505.9607368499974, &#39;median&#39;: 506.24863114999243, &#39;std&#39;: 1.0552857114823602}
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>optimized: {&#39;mean&#39;: 410.4258411699948, &#39;median&#39;: 408.519481050007, &#39;std&#39;: 3.4550342127633344}
+unoptimized: {&#39;mean&#39;: 512.5318330900063, &#39;median&#39;: 512.4503830500089, &#39;std&#39;: 1.829093237816327}
 </pre></div>
 </div>
 </div>
@@ -994,7 +996,7 @@ 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  58.606 seconds)</p>
+<p class="sphx-glr-timing"><strong>Total running time of the script:</strong> ( 11 minutes  45.042 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 e4387ecd49..1a8b8cfbf3 100644
--- a/docs/tutorial/cross_compilation_and_rpc.html
+++ b/docs/tutorial/cross_compilation_and_rpc.html
@@ -538,7 +538,7 @@ 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>1.282e-07 secs/op
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>1.317e-07 secs/op
 </pre></div>
 </div>
 </div>
diff --git a/docs/tutorial/intro_topi.html b/docs/tutorial/intro_topi.html
index 80d8ee544d..aa255dbceb 100644
--- a/docs/tutorial/intro_topi.html
+++ b/docs/tutorial/intro_topi.html
@@ -495,7 +495,7 @@ 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, 0x1cdf7900)), stage(b, placeholder(b, 0x229ecbf0)), 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=[ [...]
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>[stage(a, placeholder(a, 0x1b0c3470)), stage(b, placeholder(b, 0x1fa44170)), 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=[ [...]
 </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 b3f2d6f9a5..7cfbd62269 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>15:10.668</strong> total execution time for <strong>tutorial</strong> files:</p>
+<p><strong>15:06.103</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>11:58.606</p></td>
+<td><p>11:45.042</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>01:16.612</p></td>
+<td><p>01:20.637</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>00:58.752</p></td>
+<td><p>00:58.450</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>00:33.702</p></td>
+<td><p>00:33.661</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>00:21.427</p></td>
+<td><p>00:26.705</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>00:00.817</p></td>
+<td><p>00:00.820</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>00:00.598</p></td>
+<td><p>00:00.608</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>00:00.153</p></td>
+<td><p>00:00.179</p></td>
 <td><p>0.0 MB</p></td>
 </tr>
 <tr class="row-odd"><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>
diff --git a/docs/tutorial/tensor_expr_get_started.html b/docs/tutorial/tensor_expr_get_started.html
index a583d0b11d..0f21fe54f6 100644
--- a/docs/tutorial/tensor_expr_get_started.html
+++ b/docs/tutorial/tensor_expr_get_started.html
@@ -549,7 +549,7 @@ helper function to run a profile of the TVM generated code.</p>
 <span class="n">evaluate_addition</span><span class="p">(</span><span class="n">fadd</span><span class="p">,</span> <a href="../reference/api/python/target.html#tvm.target.Target" title="tvm.target.Target" class="sphx-glr-backref-module-tvm-target sphx-glr-backref-type-py-class sphx-glr-backref-instance"><span class="n">tgt</span></a><span class="p">,</span> <span class="s2">&quot;naive&quot;</span><span class="p">,</span> <a href="https://docs.python.org/3/library/stdtypes.html#list" ti [...]
 </pre></div>
 </div>
-<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.000007
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.000008
 naive: 0.000007
 </pre></div>
 </div>
@@ -669,10 +669,10 @@ 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    7.031929999357089e-06                    1.0
-   naive    6.6723000000000005e-06    0.9488575683503719
-parallel    7.009799999999999e-06      0.996852926670329
-  vector             2.46551e-05       3.506164026412969
+   numpy    7.642879995728436e-06                    1.0
+   naive    6.6804000000000004e-06     0.874068414489517
+parallel    6.9625000000000015e-06    0.9109785844984227
+  vector             2.47534e-05      3.2387529326424778
 </pre></div>
 </div>
 <div class="admonition-code-specialization admonition">
@@ -988,7 +988,7 @@ 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: 0.018194
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>Numpy running time: 0.018027
 </pre></div>
 </div>
 <p>Now we write a basic matrix multiplication using TVM TE and verify that it
@@ -1029,7 +1029,7 @@ 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: 3.261212
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>none: 3.213367
 </pre></div>
 </div>
 <p>Let’s take a look at the intermediate representation of the operator and
@@ -1093,7 +1093,7 @@ 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: 0.294277
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>blocking: 0.301631
 </pre></div>
 </div>
 <p>By reordering the computation to take advantage of caching, you should see a
@@ -1151,7 +1151,7 @@ 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: 0.328194
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>vectorization: 0.332175
 @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, [1024, 1024], []),
@@ -1205,7 +1205,7 @@ 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: 0.116499
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>loop permutation: 0.117709
 @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, [1024, 1024], []),
@@ -1280,7 +1280,7 @@ 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: 0.109260
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>array packing: 0.109681
 @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, [1024, 1024], []),
@@ -1353,7 +1353,7 @@ 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: 0.110412
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>block caching: 0.110728
 @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, [1024, 1024], []),
@@ -1419,7 +1419,7 @@ 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: 0.145131
+<div class="sphx-glr-script-out highlight-none notranslate"><div class="highlight"><pre><span></span>parallelization: 0.146554
 @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, [1024, 1024], []),
@@ -1480,13 +1480,13 @@ 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             3.261211896                     1.0
-        blocking            0.2942766486     0.09023536586535254
-   vectorization            0.3281936381      0.1006354841592912
-loop permutation            0.1164992826     0.03572269644388664
-   array packing            0.1092597523     0.03350280686575786
-   block caching     0.11041168139999999     0.03385602804142353
- parallelization            0.1451309414     0.04450215012952964
+            none              3.21336669                     1.0
+        blocking     0.30163059130000003     0.09386746686541399
+   vectorization            0.3321748137     0.10337283159551268
+loop permutation     0.11770911390000001     0.03663108672480824
+   array packing            0.1096807973     0.03413267388416228
+   block caching     0.11072812300000001    0.034458601735241116
+ parallelization            0.1465540376      0.0456076295481858
 </pre></div>
 </div>
 <p>Note that the outputs on the web page reflect the running times on a